ABSTRACT TRIVEDI, SHWETA. Host cytokines and immune ...

ABSTRACT

TRIVEDI, SHWETA. Host cytokines and immune responses in pregnancy

associated transmission of arrested hookworm larvae (Under the direction of

Dr. Prema Arasu)

Over one billion people worldwide are infected with the hookworms, Necator and

Ancylostoma spp. Upon entry into the host, infective larvae (third stage L3 which are

free-living and non-feeding) typically mature into blood-feeding adults in the small

intestines. An important aspect of the life cycle for A. duodenale (humans) and A.

caninum (dogs) is the propensity for L3 to undergo a temporary state of developmental

arrest in the host. In female hosts, these tissue-arrested L3 reactivate during pregnancy

and are transmitted to the neonates through milk. During pregnancy transforming growth

factor (TGF)-β is apparently upregulated in host tissues including the mammary gland.

Studies from the free-living nematode Caenorhabditis elegans show that TGF-β and

insulin-like signaling pathways regulate larval arrest and resumption of development.

Similar signaling pathways are proposed in the pregnancy-associated reactivation of

arrested Ancylostoma larvae. We have previously used an in vitro assay to demonstrate

that recombinant human TGF-β can stimulate a feeding response in tissue-arrested A.

caninum L3 larvae. We speculate that host factors like TGF-β and pregnancy hormones

such as estrogen and prolactin signal arrested L3 larvae to resume development. To

facilitate analyses of mechanisms of reactivation and transmission in vivo, we have

utilized a mouse model of A. caninum infection; mice serve as an excellent model

because infective L3 do not develop into adults but migrate to different somatic tissues

and arrest, later reactivating during the periparturient period to transmit through milk.

Skeletal muscle and mammary gland are the major tissues of interest during this process

of arrest, reactivation and transmission. We investigated TGF-β1, TGF-β2 and IGF-1

serum and transcript cytokine profiles during late pregnancy, early lactation and mid-

lactation in mice infected with A. caninum to correlate their levels with the

transmammary transmission of the larvae to the nursing pups. An in vitro co-culture

system was also developed in an attempt to mimic in vivo conditions for assessing the

effects of TGF-β and, estrogen and prolactin on larval reactivation. A. caninum L3 were

co-incubated with primary skeletal muscle and mammary epithelial cells in a Transwell®

setup and larval reactivation was measured utilizing the in vitro feeding assay.

Additionally, the immune responses during concurrent pregnancy and helminthic

infection were assessed given that both conditions are known to be biased towards a T

helper (Th)-2 type of response. Serum and transcript levels of IFN-γ (representative of

the Th1 arm of the immune response) and IL-4 (for Th2) were measured in skeletal

muscle, mammary gland and spleen during pregnancy and A. caninum infection in the

mouse. These findings which are based upon serum and transcript levels suggest that

host-derived TGF- β1 and IGF-1 may play roles in the reactivation and transmission of

arrested A. caninum larvae; levels of TGF- β2 did not however, show a correlation with

the timepoints of pregnancy and lactation associated with larval reactivation and transfer.

Also, a Th2-like response characterized by elevation in IL-4 transcript levels was

observed in skeletal muscle while a mixed Th1/Th2 profile was observed in mammary

gland when comparing the different permutations of infection with A. caninum versus

pregnancy/lactation in BALB/c mice.

ii

Dedication

To my parents: Dr. Amresh Kumar and Mrs. Mridula Trivedi Thanks to you, I never needed to search for other role models in life.

iii

BIOGRAPHY

Shweta Trivedi was born on 10th June, 1975 in Moradabad, Uttar Pradesh, India. She

finished her primary schooling until high school from Campus School, Pantnagar in

1993. It was her ambition to become a veterinary surgeon like her father. She joined the

College of Veterinary Medicine at Gobind Ballabh Pant University of Agriculture and

Technology, Pantnagar in 1993 and graduated with a degree in Bachelor of Veterinary

Science and Animal Husbandry in 1998. She successfully competed in a national exam

for Junior Research Fellowship awarded by the Indian Council of Agricultural Research

and joined Indian Veterinary Research Institute, Izatnagar in 1998. For her Master’s

work, she characterized and tested the short-term culture filtrate proteins from

Mycobacterium bovis as potential diagnostic reagents for tuberculosis testing. After

completing her postgraduate degree in Master’s of Veterinary Immunology in 2000, she

got admission in the Immunology Program at College of Veterinary Medicine, North

Carolina State University and in the Microbiology department at University of

Tennessee, Knoxville. She joined the Immunology program at CVM, NCSU in 2001

where she worked on her PhD under the direction of Dr Prema Arasu. The major focus of

her graduate work was on host-parasite interactions involved in arrest and reactivation in

canine hookworm, Ancylostoma caninum. She will join the National Institute of Allergy

and Infectious Diseases at Rockville, Maryland as a visiting fellow in the lab of Dr.

Andrea Keane-Myers studying the role of T regulatory cell in the development of allergic

diseases.

iv

ACKNOWLEDGEMENTS

I express sincere gratitude to my major advisor, Dr Prema Arasu, for giving me

the opportunity to pursue my cherished desire of getting a higher education in the United

States. Her support, thoughtful comments and guidance along the way helped in timely

completion of this project. I am very thankful to my advisory committee members, Dr

Scott Laster, Dr Bill Miller and Dr Paul Mozdziak for their encouragement and valuable

suggestions through the course of my dissertation research. I would like to specially

thank Dr Miller for being an excellent mentor during my training in `Preparing the

Professoriate’ program. I gratefully acknowledge my graduate program coordinator, Dr

Wayne Tompkins for constantly pushing me to think critically in Immunology Journal

Club.

I thank the past and present members of Arasu lab, Cortney Cowan, Tori Freitas,

and Rita Simoes for being wonderful colleagues as well as friends to me. I am extremely

indebted to Dr. Susan Lankford for introducing me to the real-time PCR technology. I

heartily thank Dr. Barb Sherry for an unending supply of mice for cell culture studies. I

would also like to thank Derek Coombs for helping me with statistical analysis. I greatly

appreciate Paula Delong, Mary Jane, Toni Grenther and LAR staff who took excellent

care of beagle dogs and my experimental mice. I always enjoyed scientific deliberations

with my friend Dr. Kristina Howard, who has been very supportive of me all these years.

Additionally, I would like to thank all my friends who have been there for me through

thick and thin. Big thanks to Gregg Cowan for printing my dissertation. Finally, I feel

blessed to have such a caring and loving life-partner in Siddhartha who has always

encouraged me to keep moving forward and helped me realize my dreams.

v

TABLE OF CONTENTS

List of Tables……………………………………………………………………….. vii

List of Figures………………………………………………………………………. viii

1. Introduction ……………………………………………………………………… 1

2. Literature Review…………………………………………………………………… 5

2.1 Life cycle of Ancylostoma caninum………………………………………… 5 2.2 Developmental Arrest and Reactivation in Hookworms……………………. 7 2.3 Host Response to Arrest and Reactivation of Parasites…………………….. 17 2.4 Helminthic infection and Host Immune Responses………………………… 19 2.5 Immune responses during pregnancy………………………………………..24 3. Transcript and serum levels of TGF-β and IGF-1 during pregnancy and Ancylostoma caninum infection in BALB/c mice

3.1 Introduction………………………………………………………………… 31 3.2 Materials and Methods…………………………………………………….. 34 3.3 Results……………………………………………………………………… 42

3.4 Discussion………………………………………………………………….. 48 4. Development of an in vitro co-culture system to study the effects of TGF-β and pregnancy hormones on reactivation of hookworm larvae

4.1 Introduction………………………………………… ……………………… 79 4.2 Materials and Methods……………………………………………………… 81 4.3 Results………………………………………………………………………. 87 4.4 Discussion…………………………………………………………………... 89

5. IL-4 and IFN-γ serum protein and transcript levels during pregnancy and Ancylostoma caninum infection in BALB/c mice

5.1 Introduction………………………………………… ……………………... 110 5.2 Materials and Methods……………………………………………………... 113 5.3 Results…………………………………………………………………….... 118 5.4 Discussion………………………………………………………………….. 121

6. Evaluation of endogenous reference genes for real-time PCR quantification of gene expression in Ancylostoma caninum……………………………………….... 131

vi

7. References………………………………………………………………………….. 146 8. Appendix ………………………………………………………………………........ 162

vii

LIST OF TABLES

Page

Table 3.1 Primer sequences used for cytokine and reference gene transcript

quantification by real-time RT-PCR……………………………………………………..55

Table 5.1 Primer sequences used for cytokine and reference gene transcript

quantification by real-time RT-PCR……………………………………………………125

viii

LIST OF FIGURES

Figure 2.1 Life cycle of Ancylostoma caninum in the host and environment……… 29

Figure 2.2 Model of larval arrest and reactivation…………………………………. 30

Figure 3.1 Experimental design…………………………………………………….. 56

Figure 3.2 Comparison of total larval burden in unbred versus bred BALB/c mice

infected at times corresponding to day 19 gestation, day 1 and day 10 of postpartum

lactation……………………………………………………………………………… 57

Figure 3.3 Comparison of serum TGF-β1 levels in UN, IN, UB and IB mouse. …… 58

Figure 3.4 Comparison of serum TGF-β2 levels in UN, IN, UB and IB mouse. …… 59

Figure 3.5 Comparison of serum TGF-β1 levels in UN, IN, UB and IB mouse …… 60

Figure 3.6a-c TGF-β1 transcript levels in skeletal muscle of IN, UB and IB mice… 61

Figure 3.6d-f TGF-β1 transcript levels in mammary gland of IN, UB and IB mice .. 62

Figure 3.6g-i TGF-β1 transcript levels in spleen of IN, UB and IB mice. …………... 63

Figure 3.7a-c TGF-β2 transcript levels in skeletal muscle of IN, UB and IB mice. … 64

Figure 3.7d-f TGF-β2 transcript levels in mammary gland of IN, UB and IB mice. ... 65

Figure 3.7g-i TGF-β2 transcript levels in spleen of IN, UB and IB mice.…………... 66

Figure 3.8a-c IGF-1 transcript levels in skeletal muscle of IN, UB and IB mice. …… 67

Figure 3.8d-f IGF-1 transcript levels in mammary gland of IN, UB and IB mice. ….. 68

Figure 3.8g-i IGF-1 transcript levels in spleen of IN, UB and IB mice. ……………. 69

Figure 3.9a, b, c Comparison of TGF-β1 transcript levels in skeletal muscle of IN (a),

UB (b) and IB (c) mice over time……………………………………………………... 70

ix

Figure 3.9d, e, f Comparison of TGF-β1 transcript levels in mammary gland of IN (d),

UB (e) and IB (f) mice over time……………………………………………………. 71

Figure 3.9g, h, i Comparison of TGF-β1 transcript levels in spleen of IN (g), UB (h) and

IB (i) mice over time…………………………………………………………………. 72

Figure 3.10a, b, c Comparison of TGF-β2 transcript levels in skeletal muscle of IN (a),

UB (b) and IB (c) mice over time……………………………………………………. 73

Figure 3.10d, e, f Comparison of TGF-β2 transcript levels in mammary gland of IN (d),

UB (e) and IB (f) mice over time…………………………………………………….. 74

Figure 3.10g, h, i Comparison of TGF-β2 transcript levels in spleen of IN (g), UB (h)

and IB (i) mice over time……………………………………………………………… 75

Figure 3.11a, b, c Comparison of IGF-1 transcript levels in skeletal muscle of IN (a), UB

(b) and IB (c) mice over time………………………………………………………….. 76

Figure 3.11d, e, f Comparison of IGF-1 transcript levels in mammary gland of IN (d),

UB (e) and IB (f) mice over time………………………………………………………. 77

Figure 3.11g, h, i Comparison of IGF-1 transcript levels in spleen of IN (g), UB (h) and

IB (i) mice over time………………………………………………………………….. 78

Figure 4.1a-f Immunohistochemistry slides of skeletal muscle and mammary epithelial

cells…………………………………………………………………………………….. 93

Figure 4.2a Effect on percent feeding response of L3 larvae co-cultured in the

presence or absence of primary skeletal muscle cells………………………………….. 95

Figure 4.2b Effect of serum on percent feeding response of L3 larvae co-cultured

with primary skeletal muscle cells. …………………………………………………… 96

Figure 4.2c Effect of serum on percent feeding response of L3 larvae co-cultured

x

with primary mammary epithelial cells. ………………………………………………. 97

Figure 4.3a Effect of TGF-β2 on feeding response of L3 larvae co-cultured

with primary skeletal muscle cells for 24 h. …………………………………………. 98

Figure 4.3b Effect of TGF-β2 on feeding response of L3 larvae co-cultured

with primary skeletal muscle cells for 48 h.. …………………………………………. 99

Figure 4.3c Effect of TGF-β2 on feeding response of L3 larvae co-cultured

with primary skeletal muscle cells for 72 h.. …………………………………………. 100

Figure 4.4a Effect of TGF-β2 on feeding response of L3 larvae co-cultured

with primary mammary epithelial cells for 24 h. ……………………………………… 101

Figure 4.4b Effect of TGF-β2 on feeding response of L3 larvae co-cultured


Figure 4.4c Effect of TGF-β2 on feeding response of L3 larvae co-cultured


Figure 4.5a Effect of pregnancy-associated hormones on feeding response of L3 larvae

co-cultured with primary skeletal muscle cells for 24 h………………………………. 104

Figure 4.5b Effect of pregnancy-associated hormones on feeding response of L3 larvae

co-cultured with primary skeletal muscle cells for 48 h……………………………… 105

Figure 4.5c Effect of pregnancy-associated hormones on feeding response of L3

larvae co-cultured with primary skeletal muscle cells for 72 h………………………. 106

Figure 4.6a Effect of pregnancy-associated hormones on feeding response of L3 larvae

co-cultured with primary mammary epithelial cells for 24 h………………………... 107

Figure 4.6b Effect of pregnancy-associated hormones on feeding response of L3 larvae

co-cultured with primary mammary epithelial cells for 48 h………………………… 108

xi

Figure 4.6c Effect of pregnancy-associated hormones on feeding response of L3 larvae

co-cultured with primary mammary epithelial cells for 72 h………………………… 109

Figure 5.1 Comparison of total larval burden in unbred versus bred BALB/c mice

infected at times corresponding to day 19 gestation, day 1 and day 10 of postpartum

lactation……………………………………………………………………………… 126

Figure 5.2a. IL-4 transcript levels in skeletal muscle of infected bred mice……….. 127

Figure 5.2b. IL-4 transcript levels in mammary gland of infected unbred mice..….. 128

Figure 5.2c. IL-4 transcript levels in mammary gland of infected bred mice………. 129

Figure 5.3. IFN-γ transcript levels in mammary gland of infected unbred mice. ….. 130

Figure 6.1 Expression levels of candidate reference genes across different

developmental stages of A. caninum. ……………………………………………….. 143

Figure 6.2 Expression levels of candidate reference genes in two different strains

of A. caninum………………………………………………………………………… 144

Figure 6.3 Effect of combination of serum treatment and strain of A. caninum

on expression levels of candidate reference genes…………………………………… 145

1

1. Introduction

The World Health Organization estimates that over one billion people in most tropical

and subtropical regions of the world are infected with blood-feeding intestinal hookworms,

predominantly Necator americanus and Ancylostoma duodenale (Chan, et al., 1997). In

addition, approximately 44 million women are infected with hookworms during pregnancy

with the potential for transmammary transmission of A. duodenale to their nursing infants

(Bundy, et al., 1995; Schad and Page, 1982). Studies have correlated retarded physical and

cognitive development with early childhood parasitic diseases (Watkins and Pollitt, 1997;

Drake, et al., 2000; Dickson, et al., 2000). Hookworms are also prevalent in other hosts; A.

caninum is a major parasite of dogs in the U.S. and other parts of the world (Blaghburn, et

al., 1996) and can cause moderate to severe iron deficiency anemia, hypoproteinemia and

bloody diarrhea that can be fatal to puppies and immunosuppressed dogs (Georgi and Georgi,

1991). A. caninum is also a zoonosis to humans due to skin penetration by the soil-dwelling

infective larvae resulting in cutaneous larva migrans (Miller, et al., 1991) and eosinophilic

enteritis due to the potential, albeit rare, of larval development to the adult stage in the human

gut (Prociv and Croese, 1990; Croese et al., 1994).

The lifecycle of nematodes is relatively simple involving development from egg,

through different larval stages to reach adulthood. With parasitic nematodes, the later stages

of development have an obligate requirement of a host. Interestingly, A. duodenale and A.

caninum, like a number of other parasitic nematodes, also have the capacity to infect a host,

abort their normal maturation pathway and undergo developmental arrest in somatic tissues.

Tissue-arrested larvae are metabolically quiescent and resilient to host immune responses as

well as the chemotherapeutic agents typically used to eliminate the intestinal adult stage

2

(Lee, et al., 1975; Schad, 1991; Arasu, 1998). This reservoir of larvae can however respond

to host signals and reactivate during pregnancy or stress (Stone and Girardeau, 1968). Once

reactivated, the third stage larvae (L3) either resume development to become egg-producing

adults in the intestines or are transmitted to the newborns via milk. Vertical transmission of

infection is a relatively common route of passage of parasites to a new generation of hosts

but little is understood about the molecular or immuno-physiological mechanisms facilitating

the process in the host/pathogen relationship. While a generalized immunosuppressive state

is associated with gestation, previous studies have shown no correlation between the immune

responses and reactivation/transmission of A. caninum larvae (Arasu and Heller, 1999). In

separate studies, the hormonal fluxes during pregnancy have been implicated to directly or

indirectly mediate the reactivation of the arrested larval population (Stoye and Krause, 1976);

in the proposed 'indirect' role, estrogen and prolactin are known to regulate the expression of

various developmental and immunomodulatory cytokines such as transforming growth factor

beta (TGF-β) and insulin-like growth factor (IGF). Recent studies have shown that these

mammalian signaling molecules are also present and critical to the development of

nematodes (Crook, et al., 2005; Brand and Hawdon, 2004).

The free-living nematode, Caenorhabditis elegans, has been used as a model to draw

parallels for understanding the signaling mechanisms responsible for the phenomena of arrest

and reactivation in parasitic nematodes (Hotez, et al., 1993). At least three signaling

pathways, the TGF-β pathway, insulin-like pathway and a cGMP pathway, have been

implicated to control the dauer or arrested form in C. elegans (Riddle and Albert, 1998). C.

elegans and hookworms fall in Clade V of Phylum Nematoda (Blaxter, 1998). Given their

close phylogenetic relationship, it is likely that signaling pathways involved in regulating

3

development might be commonly shared. We and others have hypothesized that tissue-

arrested L3 larvae of A. caninum could receive signals from the host resulting in similar

signaling through these known pathways in C. elegans eventually leading to reactivation

(Hawdon and Schad, 1991).

One objective of this study was to use an in vivo system to examine the host's

circulating and tissue-specific profile of expression of TGF-β and IGF during different

phases of pregnancy and lactation in the presence and absence of A. caninum infection, and

to correlate these levels with larval transmission. The second objective of this study was to

establish an in vitro system to examine the direct effects of TGF-β and IGF on A. caninum

larvae cultivated in the environment of muscle (site of larval arrest) versus mammary cells

(site of larval transmission).

Using mice, an experimental model for studies of tissue-arrest, reactivation and

transmammary transmission of A. caninum larvae was previously established (Arasu and

Kwak 1998). Mice serve as a paratenic host for A. caninum in that the L3 larvae enter the

host (by ingestion or skin penetration) and persist in various somatic tissues but are unable to

mature into adults (Arasu and Kwak, 1999). These tissue-arrested L3 larvae however, display

pregnancy-associated transmammary transmission of infection to the nursing pups providing

an encapsulated model for studies of this process of vertical transmission.

For the in vitro analyses, primary skeletal and mammary epithelial murine cell

cultures were established and the conditions were defined for propagation of A. caninum L3s.

The co-culture system was used to examine the effect of normal dog serum and recombinant

mammalian TGF-β as well as the indirect effects of estrogen and prolactin on larval

reactivation using an established feeding/reactivation assay.

4

The third objective of this study was to examine the immunological effects of A.

caninum infection on the viability of pregnancy. Pregnancy and helminthic infection are

both known to evoke Th2-biased immune responses (Wegmann, et al., 1993; Pearce and

Reiner, 1995). However, there are no reports to date on the concurrent effects of pregnancy

and helminthic infection or the inflammatory responses associated with larval arrest and

reactivation.

The fourth aspect of this thesis concerns a relatively new and powerful analytical tool.

Real time polymerase chain reaction (PCR) analysis is extensively used, including in the

above described studies, to compare gene expression levels between different treatments and

experimental groups. The validity of relative comparisons is however highly dependent on

the use of a suitable endogenous reference gene for normalization of the measured transcript

levels. This section describes the validation of six different commonly used endogenous

reference genes in real time PCR analyses of gene expression in different developmental

stages, strains and treatments of A. caninum. These studies also serve as a template for

validation studies in any given experimental system of relative expression using real time

PCR analyses.

5

2. Literature Review

Despite their health significance and global importance, little is known about how

hookworm parasites interact with their hosts to establish a chronic infection. The review will

outline the biology, normal lifecycle and public health significance of hookworm infection.

Additionally, hookworms undergo developmental arrest during one of their environmental

life-stages (as free-living infective larvae) as well as inside the host (hypobiosis). Hypobiotic

hookworm larvae can reactivate in response to host signals received during pregnancy /

lactation and resume their development resulting in the transmammary transmission of

infection to the newborns. This phenomenon of arrest and reactivation has been well studied

in Caenorhabditis elegans, a free-living nematode in which distinct signaling pathways have

been discovered. The immuno-physiological changes occurring during pregnancy have also

been implicated in facilitating the reactivation process and is also addressed below, followed

by a discussion of the host immune responses during acute and chronic parasitic infections

and during pregnancy.

2.1 Life cycle of Ancylostoma caninum

Ancylostoma caninum is one of the two most common parasitic infections of

domestic dogs (Blaghburn, et al., 1996) and is a good model for studies on human hookworm

disease. The life cycle of A. caninum is similar to that for A. duodenale and begins with

shedding of embryonated eggs in the feces of an infected host (Kassai, 1999). The hatched

first-stage larva develops through two molts to the infective third-stage (L3), which is still

ensheathed in the outer cuticle from the second larval stage. By positioning itself on top of a

grass blade or other objects, this non-feeding L3 maximizes its chances of penetrating the

6

skin of a host. Infective L3s can persist in the environment under warm moist conditions for

extended periods of time (~6 months or more). While infection by skin penetration is

prevalent, larvae can also be orally ingested. After entering the host through skin the

maturing fourth stage larvae find their way to the lungs via blood vessels or lymphatics

where they are coughed up and swallowed. They become blood feeding adults in the

intestines by 14-21 days post-infection. In contrast, larvae that enter by ingestion may either

follow the above mentioned path or bypass the lungs and mature directly in the intestines to

become sexually mature male and female worms (Figure 1). While attached to the intestinal

mucosa, adult worms secrete a range of molecules including proteases and anti-coagulants to

prevent clotting in order to receive a blood meal (Williamson, et al., 2003; Stassens, et al.,

1996). A single female worm can produce about 10,000 eggs per day resulting in large

numbers of infective larvae in the environment; by contrast, only 50-100 adult worms are

sufficient to cause life threatening anemia in a neonatal puppy (Loukas, et al., 2005).

The currently available approach for controlling hookworm infection in humans or

animals is targeted at elimination of the intestinal adult stages of the parasite with various

classes of anthelmintics such as the macrolides (e.g. ivermectin) and benzimidazoles

(Bungiro and Cappello, 2004). However, the requirement of repeated doses, the development

of drug-resistant strains and the risk of reinfection are the main limitations of

chemotherapeutic intervention (Quinnell et al., 1993; Hotez and Pritchard, 1995). Besides the

use of anthelminthics, other factors like sanitation, use of footwear and health education can

prevent or reduce the burden of infection (Burke and Roberson, 1979a,b) as occurred in the

hookworm eradication era of the 1920s-40s in many parts of the southern U.S. (Ciesielski et

7

al., 1992). Efforts are also underway to develop an effective vaccine against hookworms

(Brooker, et al., 2005; Goud, et al., 2005).

2.2 Developmental Arrest and Reactivation in Hookworms

An interesting aspect of the A. caninum and A. duodenale life cycles is the capacity of

the parasite to undergo developmental arrest at two stages (Figure 2). The first one occurs

when the non-feeding third larval stage is in the environment waiting to infect a suitable host.

Unless it infects the host, further development cannot occur as it only reactivates upon

receiving host cues and resumes development within the host. The second stage occurs when

the infecting third stage larvae enter a host; instead of continuing to develop into a blood-

feeding adult in the intestines, a portion of the entering larvae remain within the somatic

tissues as developmentally arrested L3s; these L3s have been found throughout the body in

different organs and especially in skeletal muscle (Lee, et al., 1975; Schad, 1990). The

developmentally arrested larvae can persist in the tissues for long periods of time. They

display reduced susceptibility to drugs that effectively eliminate adult stage infections

(Schad, 1991; Arasu, 1998). These arrested larvae possess the capacity to reactivate and

resume normal development thereby serving as a reservoir of infection in the same host i.e.

being able to reactivate and re-establish an active infection (Schad and Page, 1982).

Secondly, in a female host during pregnancy/lactation, the arrested larvae apparently receive

specific cues to reactivate and get transmitted through the mammary glands to the newborn

pups via milk (Stone and Smith, 1975; Burke and Roberson, 1985).

Vertical transmission has been also observed in other helminths including Toxocara

canis, the other major intestinal parasite of dogs (Shoop, 1991; Swerczek et al., 1971). The

8

mechanisms by which developmentally arrested parasitic stages are cued to reactivate during

pregnancy and transmit to a new generation of hosts, either via milk or the uterus (as in the

case of Toxocara), are unknown but serve as a major route of transmission of infection.

Previous studies have shown that host humoral responses (Arasu and Heller, 1999) and the

generalized immunosuppression associated with pregnancy are not directly involved in the

process of parasite reactivation.

2.2.1 Developmental arrest/reactivation and signaling pathways in Caenorhabditis

elegans

Caenorhabditis elegans, the free-living soil nematode, is a valuable model system for

parasitic nematodes of both plants and animals. C. elegans has a relatively short life cycle

(~4 days) which also involves development through four larval stages (L1-L4) to the adult

stage. Normal development and hermaphroditic reproduction occurs in the presence of

adequate food (soil bacteria), optimal temperatures and low population density (Riddle and

Albert, 1998). However, when the population density or temperatures are very high and food

resources are limited, an alternate developmental pathway occurs resulting in a non-feeding

dauer larva which can persist for several weeks without feeding (but also with the capacity

for motility, albeit only when necessary e.g. prodding). With the return of favorable

environmental and food conditions, the dauer larva begins feeding and resumes development

to an adult. C. elegans perceives the environmental signals via chemosensory neurons located

in its head, namely ASI, ASJ, ADF and ASG (Riddle and Albert, 1998). From extensive

genetic and molecular studies, at least three signaling pathways have been shown to regulate

9

the entry and exit from the dauer stage, namely, a TGF-β-like, an insulin-like, and a cGMP-

regulated pathway (Massague, et al., 2000; Braeckmann, et al., 2001; Birnby, et al., 2000).

TGF-β pathway

The TGF-β pathway in C. elegans comprises a series of genes including a ligand

encoding daf-7, and the type I and type II receptors (daf-1 and daf-4 respectively). By

analogy to TGF-β signaling in the mammalian system, it was hypothesized that daf-7 binds

to and activates the receptor daf-1/daf-4 complex. Genetic analysis has shown that daf-1 acts

as the type I receptor in the signaling cascade and has been localized to the amphidial

chemosensory neurons in the head of the worm (Gunther, et al., 2000). The second receptor

is encoded by daf-4 and shows 30% amino acid identity with the conserved kinase domain in

daf-1, and 40% and 34% identity to the mouse activin type II receptor and human type II

TGF-beta receptor, respectively (Estevez, et al., 1993). It was also shown by the authors that

DAF-4 binds to the mammalian bone morphogenetic protein BMP-2 and its expression at the

L1 stage is enough to rescue the dauer constitutive phenotype (Estevez, et al., 1993). Since

DAF-4 can act in a cell non-autonomous way (Inoue and Thomas, 2000), it has been

suggested that DAF-7 might be binding to DAF-4 on neurons to generate a secondary signal

that acts to promote differentiation of tissue involved in development to the adult stage (Beall

and Pearce, 2002).

A second TGF-β pathway, involving the dbl-1 TGF-β like ligand, and the receptors

sma-6 (another Type I receptor) and daf-4, has been shown to be involved in the regulation

of body size and male tail formation (Patterson and Padgett, 2000). Other genes with

10

homology to the TGF-β ligands have been identified in C. elegans but have not been

characterized.

Insulin-like growth factor pathway

The insulin-like daf-2 receptor associated pathway acts synergistically with the

pathway activated by the daf-7 TGF-β-type signal (Ogg, et al., 1997) in dauer

arrest/reactivation. There are 38 insulin-like proteins that have been identified in C. elegans

and can act as potential ligands (Pierce, et al., 2001; Li, et al., 2003). Of these, DAF-28 is

expressed in the ASI and ASJ chemosensory neurons and its expression is downregulated by

dauer-inducing conditions (Li, et al., 2003). These findings suggest that DAF-28 is the

functional ligand for DAF-2 and necessary for prevention of dauer formation. The

transmembrane DAF-2 receptor is a tyrosine kinase that facilitates the formation of a dauer

under favorable conditions (Kimura, et al., 1997). Upon ligand binding, DAF-2 gets

autophosphorylated and activates AGE-1, a phosphatidylinositol-3-OH kinase (Morris, et al.,

1996; Malone et al., 1996). Subsequently, activated AGE-1 produces secondary messengers

phosphatidylinositol bis and tris- phosphate phospholipids, PIP2 and PIP3. These secondary

messengers activate other kinases like Protein kinase B (PKB/Akt) and 3-phosphoinositide-

dependent kinase (PDK). Two PKB/Akt kinases, AKT-1 and AKT-2 have been found to

mediate this signaling in C. elegans (Paradis and Ruvkun, 1998). Later, PDK-1 was

identified and shown to be involved in DAF-2 signaling (Paradis, et al., 1999). These

activated kinases phosphorylate DAF-16, a forkhead transcription factor, which presumably

regulates long-lived dauer larva (McElwee, et al., 2003).

11

Cyclic GMP signaling pathway

The cGMP pathway includes a membrane guanylyl cyclase that binds an undefined

ligand and catalyses the synthesis of the secondary messenger cGMP from GTP. ‘Loss-of-

function’ mutants of the membrane guanylyl cyclase constitutively enter the dauer stage and

have been shown to be rescued by membrane permeable analogues of cGMP (Birnby, et al.,

2000). It appears that the cGMP signaling pathway might be involved in dauer development

upstream of daf-7 pathway. There is evidence that a guanylyl cyclase, DAF-11, is necessary

to stimulate expression of DAF-7 (Murakami, et al., 2001). Expression of daf-11 cDNA by

cell specific promoters suggests that daf-11 acts cell autonomously in ASI chemosensory

neurons for daf-7 expression (Murakami, et al., 2001).

2.2.2 Signaling Pathways in Parasitic Nematodes

Over the years evidence has emerged that similar signaling pathways might be

present in parasitic helminths for the purpose of regulating the resumption of development

upon receiving the appropriate signals from the host. Components of both the TGF-β and

insulin-like pathways have been investigated in several different parasites to understand how

they may regulate the entry into and exit from the arrested state. In parasitic nematodes

components of the TGF-β signaling pathway were first identified in the filarial nematodes,

Brugia malayi and Brugia pahangi, which cause lymphatic filariasis and elephantiasis in

humans and domestic pets like cats and dogs.

The first type I TGF-beta receptor, Bp-trk-1, from a parasitic helminth was cloned

from Brugia using degenerate primers (Gomez-Escobar, et al., 1997). EST sequencing efforts

of the Washington University Nematode Sequencing Project subsequently showed the

12

presence of TGF-β receptor homologs in both animal and plant parasitic nematodes

Identification of the Type II receptor daf-4-like gene has been more difficult; to date it has

been identified only by random EST sequence analyses of Strongyloides stercoralis

(GenBank Accession number BE029357).

Two genes, tgh-1 and tgh-2, encoding ligands from the TGF-beta superfamily

have been also identified in Brugia malayi (Gomez-Escobar, et al., 1998; Gomez-Escobar, et

al., 2000). The tgh-1 gene shows greatest homology to the bone morphogenetic protein

subfamily of BMP1 and has been shown to be maximally expressed at the first (L1 to L2)

and second (L2 to L3) molts, and is completely absent in the stages associated with

developmental arrest, i.e. microfilaria (similar to an L1) (Gomez-Escobar, et al., 1998).

Expression of tgh-1 at a time when the parasites are maturing and molting and its homology

to BMP subfamily is suggestive of a role in growth and development of the parasite. The

second ligand encoding gene, tgh-2, was identified the B. malayi EST database (Washington

University Nematode Sequencing Project) and is most similar to the C. elegans daf-7 gene.

Expression of tgh-2 is greatest in microfilariae (L1 larvae) and also in adult male and female

parasites (Gomez-Escobar, et al., 2000), therefore, coinciding with developmental arrest as

well as terminal development of the parasite.

Components of the TGF-β signaling pathway have also been identified in other

classes of helminthic parasites including Schistosoma mansoni which is a trematode or

flatworm. The type I TGF-β receptor, Smrk-1, shares up to 58% homology with the

conserved kinase domain of other type I TGF-β receptors but is considered a divergent

member of this family as it has an atypical GS domain that is involved in regulation of type I

receptor kinase activity (Davies, et al., 1998). A chimeric receptor containing the

13

extracellular domain of SmRK1 joined to the intracellular domain of the human type I TGF-β

receptor. The chimeric receptor bound radiolabeled TGF-β and activated a luciferase reporter

gene in response to both TGF-β1 and TGF-β3 but not BMP7 (Beall and Pearce, 2001). This

study suggests that a host ligand may directly stimulate parasite-associated receptors of the

TGF-β pathway. Recently, a type II TGF-β receptor was isolated from S. mansoni and was

found to be most closely related to the Activin type II receptor family (Forrester, et al.,

2004).

In 2001, an insulin-like homolog was cloned and sequenced in Strongyloides

stercoralis (GenBank Accession number BG224639) but a daf-2 like insulin receptor has not

been identified. Regardless, a forkhead transcription factor gene, fktf-1, proposed to be

orthologous to the C. elegans dauer-regulatory gene daf-16 was discovered in S. stercoralis.

Discovery of fktf-1 indicates the presence of an insulin-like signaling pathway in S.

stercoralis similar to that known to regulate dauer development in C. elegans (Massey, et al.,

2003). Ablation of neuron pairs ASF and ASI in S. stercoralis larvae, an intestinal nematode

of humans and dogs, caused the larvae to develop directly into the dauer-like infective resting

stage and prevented development of the soil-dwelling adult worms (Ashton, et al., 1998).

Loss of the sensory neurons presumably interfered with generation of signals needed for

normal development from a larva to an adult.

2.2.3 Signaling Pathways in Hookworms

The hookworm parasites and C. elegans fall within the same clade in the phylum

Nematoda ie Clade V (Blaxter, 1998). It is therefore plausible to speculate that many

developmental control mechanisms are shared between A. caninum and C. elegans. Recently,

14

two A. caninum TGF-β-like ligands, Ac-dbl-1 and Ac-daf-7 were cloned and characterized

(Freitas and Arasu, 2005). Ac-dbl-1 showed 60% amino acid identity to the C. elegans dbl-1

while Ac-daf-7 showed 46% amino acid identity to C. elegans daf-7. Since C. elegans daf-7

mutants are constitutive dauers, Rajan (1998) hypothesized that parasitic nematodes behave

as if they were daf-7 mutants to ensure developmental arrest at the L3 stage. He proposed

that these parasites then utilize the daf-7 gene product from their host to reenter the

developmental pathway.

In support of this hypothesis, an in vitro assay was used to show that physiological

concentrations of recombinant mammalian TGF-β isoforms, TGF-β1 and TGF-β2, had

significant stimulatory effects on tissue-arrested larvae resulting in the resumption of a

feeding response; additionally, the stimulatory effect of normal dog serum could be blocked

by preincubation with anti-TGF-β antibodies (Arasu, 2001). These in vitro

feeding/reactivation analyses also showed that the pregnancy/lactation associated hormones

estrogen and prolactin did not have a stimulatory effect on infective or tissue-arrested A.

caninum larvae (Arasu, 2001); insulin was also shown to not have a direct effect but as

mentioned above, TGF- β isoforms 1 as well as 2 had a direct effect on the feeding or

reactivation response of tissue-arrested larvae.

To further investigate the role of insulin signaling in hookworm larval activation, the

phosphatidylinositol-3-OH kinase inhibitor LY294002 was tested for its effect on in vitro

activation using the resumption of feeding as a marker for activation. LY294002 prevented

feeding in A. caninum infective larvae stimulated with host serum filtrate and a glutathione-

analogue, the muscarinic agonist arecoline, or the cell permeable cGMP-analogue 8-bromo-

cGMP (Brand and Hawdon, 2004). Similar results were seen with the congeneric hookworm

15

A. ceylanicum. These data suggest that an insulin-signaling pathway mediates activation in

hookworm larvae, as in C. elegans, and that the phosphatidylinositol-3-OH kinase inhibitor

acts downstream of the cGMP and muscarinic signaling steps in the pathway. In A. caninum,

LY294002 had no effect on the release of excretory/secretory products associated with

activation, suggesting that the secretory pathway diverges from the activation pathway

upstream of the phosphatidylinositol-3-OH kinase step (Brand and Hawdon, 2004). These

results provide additional support for the insulin-signaling pathway as one of the primary

pathway for activation to parasitism in hookworm larvae.

The in vitro feeding assay has been used in several studies above as a direct

method to evaluate the activation status of larval stages with feeding being equated to

resumption of development (Hawdon and Schad, 1990, 1992; Hawdon, et al., 1993). Over

the years many factors such as 10% normal canine serum (Hawdon and Schad, 1990),

reduced glutathione (Hawdon and Schad, 1992), muscarinic antagonists (Tissenbaum, et al.,

2000) and cyclic GMP (Hawdon and Datu, 2003) have been shown to stimulate similar

feeding behavior in A. caninum iL3. Using flourescein isothiocyanate (FITC)-labeled bovine

serum albumin it was shown that in the presence of 10% normal canine serum A. caninum

iL3 displayed significant feeding behavior (a classic C. elegans dauer reactivation response)

at 37ºC in presence of 5% CO2 (Hawdon and Schad, 1990). Reduced glutathione stimulated

larval feeding in greater than 90% of the iL3 population, in a specific and concentration-

dependent manner with highest at 5-10 mM, and reaching a plateau at 25-50 mM (Hawdon

and Schad, 1992). Using increasing concentrations of oxotremorine and other muscarinic

agonists, it was shown that muscarinic pathway regulates the recovery of A. caninum

developmental arrest (Tissenbaum, et al., 2000). Recently, a membrane permeable analogue

16

of cyclic GMP, 8-bromo-cyclic GMP, was tested for its ability to stimulate feeding in iL3.

Populations of iL3 reached maximum feeding at 3.5-5.0 mM and secreted Ancylostoma

secreted protein 1 suggesting reactivation (Hawdon and Datu, 2003).

2.2.4 TGF-β and insulin in the mammalian system

Different isoforms of mammalian TGF-β are closely related structurally and

functionally but have diverse tissue-specific signaling effects on growth and differentiation.

TGF- β1 isoform predominates in many tissue locations including skeletal muscle which is

the preferred site of arrest for the A. caninum larvae (Lee, et al., 1975). It is known to be

released by major inflammatory cells like eosinophils and mast cells that respond to parasitic

infections (Wong, et al., 1991; Baumgartner, et al., 1996). TGF- β1 has also been suggested

to play a role in the regeneration of skeletal muscle tissue (Husmann, et al., 1996). TGF- β2

is the dominant isoform in the uterus and mammary gland and has been shown to be

specifically upregulated by estrogen and prolactin, the hormones associated with

pregnancy/lactation (Cheng, et al., 1993; Schneider, et al., 1996). Plasma levels of TGF- β2

show a transient increase during the peri-parturient period as compared to levels during

normal non-pregnant state (Schneider, et al., 1996). IGF-1 has been shown to be present in

myoepithelial cells of mammary gland of non-pregnant, pregnant and lactating rats. IGF-1

synthesized in mammary gland is reported to play a role in development of this organ during

pregnancy as mammary epithelial cells also express specific IGF-1 receptors (Marcotty, et

al., 1994). During lactation, IGF-1 can be transferred from serum into milk (Marcotty, et al.,

1994).

17

2.3 Host Response to Arrest and Reactivation of Parasites

During developmental arrest or hypobiosis parasites persist inside the host in a

quiescent manner and reactivate in response to unknown stimuli. Upon reactivation the

parasite resumes development thereby leading to reinfection of the host, spread into the

environment and transmission into the newborn (Schad and Page, 1982; Schad, 1990; Shoop,

1991). It has been well documented in A. caninum infection of dogs that developmentally

arrested L3 larvae reactivate during pregnancy and are transmitted to the neonatal puppies

via suckling of milk from infected dams (Stone and Girardeau, 1968; Stoye, 1973; Burke and

Roberson, 1985). A similar phenomenon of arrest and reactivation has been suggested for A.

duodenale infection of humans (Arasu and Kwak, 1999). Reactivation of arrested larvae has

been hypothesized to occur, directly or indirectly, due to influence of maternal hormones

during pregnancy and lactation (Stone and Smith, 1973; Stoye and Krause, 1976). It was

shown that repeated exogenous administration of pregnancy-associated hormones, such as

estrogen, progesterone and prolactin, to ovariectomized post-partum A. caninum infected

dogs harboring tissue arrested L3 led to gradual resurgence of L3 larvae in milk by induction

with oxytocin for milk let down (Stoye and Krause, 1976).

Previous work has shown that expression and secretion of TGF-β2 is tightly regulated

by estrogen and prolactin which are critical factors in the tissue-specific regulation of the

local production of TGF-β2 in the mammary gland and uterine tissues (Schneider, et al.,

1996). Furthermore, elevated levels of TGF-β2 were detected in late pregnant maternal

plasmas (> 100 pM), and in the milk (> 500 pM) during early lactation (Schneider, et al.,

1996) whereas normal endogenous levels of TGF-β1 and TGF-β2 in sera of mouse are 125

18

ng/ml and 5 ng/ml (Emax Immunoassay kit, Promega, Madison, WI). These findings suggest

that host-derived TGF-β might play a role in reactivation of tissue-arrested larvae.

Furthermore, studies in rats and pigs have demonstrated that IGF-1 is also

upregulated during pregnancy (Marcotty, et al., 1994; Lee, et al., 1993). In rats, maternal

serum IGF-I concentration rose during the first half of pregnancy while in the second half of

pregnancy, the mean serum IGF-I concentration fell sharply from 1140 +/- 150 ng/ml at

seven days of pregnancy to 470 +/- 85 ng/ml at 20 days. Using RNase protection assay it was

determined that from the onset of pregnancy to term, IGF-I gene expression in the mammary

gland diminished (Marcotty, et al., 1994). Similarly in mammary tissue of pregnant pigs,

steady-state levels of the mRNAs encoding IGF-I, IGF-II and type-I IGF receptor as well as

the levels of the membrane-associated type-II IGF receptor were higher during the early

phase of mammogenesis (< or = day 45) than during the subsequent stages of mammary

development. Mammary IGF-I, IGF-II and type-I receptor mRNAs were expressed at their

lowest levels around day 90 of pregnancy (20-40% of those for day 30 of pregnancy) (Lee, et

al., 1993).

To understand the underlying mechanisms of tissue-arrest and subsequent

reactivation, a murine model of infection was developed (Stoye and Krause, 1976; Arasu and

Kwak, 1999). The mouse serves as a paratenic host for A. caninum in that infective larva

does not develop into mature adults but instead distributes throughout the body persisting for

extended periods in a developmentally arrested state (Lee, et al., 1975). These tissue-arrested

larvae do however have the capacity to reactivate and display the same transmammary

transmission as seen in dogs and humans. In previous studies, 2-4 % (Steffe and Stoye, 1984)

19

and up to 8% (Arasu and Kwak, 1999) of tissue-arrested larvae in the dam reactivated and

got transmitted to the nursing pups during post-partum lactation.

From an immunological perspective, BALB/c and C57BL/6 mice are known to

typically display divergent immune responses to infection with BALB/c displaying a Th2

biased response and C57BL/6 mice predominantly produce Th1 cytokines (Brenner, et al.,

1994; Nabors, et al., 1995; Honore, et al., 1998). However, when these two strains were

compared there was no difference in tissue larval burden or in numbers transferred to pups

(Arasu and Kwak, 1999). Initial comparisons of BALB/c versus C57BL/6 mice showed that

both the strains mounted strong Th2 biased IgG1 and IgE antibody responses to A. caninum

infection (Arasu and Heller, 1998) and that the immune response was not directly correlated

with the phenomenon of larval reactivation/transmission.

2.4 Helminthic infection and Host Immune Responses

Mosmann and others defined the two distinct CD4+ T cell subsets by differential

secretion of cytokines (Mosmann et al., 1986), which has revolutionized the understanding of

regulatory mechanisms underlying resistance and susceptibility to helminth infection. The T

helper type 1 (Th1) cells produce type 1 cytokines IFN-γ, lymphotoxin, and IL-2 stimulating

immunoglobulin (Ig) G2a production and cell mediated effector responses. The T helper 2

(Th2) secrete type 2 cytokines IL-4, IL-5, IL-6, IL-9 and IL-13 and promote mastocytosis,

eosinophilia, and the production of IgE and IgG1 (Mosmann and Sad, 1996; O` Garra, 1998).

Traditionally, immune responses to extracellular helminth parasites have been considered to

be Th2 in nature (Pearce and Reiner, 1995). This inference is based primarily on animal

models of helminthic infections. Human studies also suggest that Th2 cytokines dominate the

20

immune responses seen in chronic, longstanding helminth infections (King and Nutman,

1991). Where helminths inhabit a tissue environment, the situation is considerably more

complex. One explanation for this is that, although Th1 responses may be more effective at

parasite clearance, they are also more likely to cause more pathology (Pearce, et al., 1996).

The skew to a Th2 response may reflect the classic compromise of minimizing host

pathology at the expense of accepting some level of continuing infection.

2.4.1 Immune Responses to Parasitic Helminths

Recently an in vitro system was used to study the early immune responses to infective

L3 from Brugia malayi by co-culturing with peripheral blood mononuclear cells (PBMCs)

from previously unexposed individuals. After 24 h of culture, the frequency of T cells

expressing Th1 cytokines (IFN-γ, TNF-α) was significantly increased in comparison to ones

expressing Th2 cytokines (IL-4, IL-5, IL-10) (Babu and Nutman, 2003). This data suggests

that the initial primary immune response to infective L3s of B. malayi is not predominantly

Th2 but rather dominated by a proinflammatory Th1 response. This finding in conjunction

with similar findings from study of immune responses in the schistosome trematodes reveals

an emerging pattern of dominant Th1 responses during the early phase of helminthic

infections (Pearce, et al., 1991). The main adaptive immune response against Schistosomes is

mediated by CD4 T cells (Hernandez, et al., 1997a). An initial pro-inflammatory Th1-

polarized response lasts around five weeks post-infection, at which point granulomatous

inflammation gets underway. However, within the next one or two weeks, granuloma

formation rises amid a dramatic change in the cytokine environment which become

dominated by anti-inflammatory Th2-type cytokine (Pearce, et al., 1991). IL-4 along with T-

21

cell co-stimulatory systems (B7-CD28, CD40-CD40L) contribute to this conversion (Brunet,

et al., 1997; Hernandez, et al., 1999; MacDonald, et al., 2002). The evolving cellular

response is gradually accompanied by abundant production of mainly non-complement fixing

IgG and IgE antibodies (Hernandez, et al., 1997b). A Th1 to Th2 conversion is vital for the

host because it has been demonstrated that failure to convert is associated with a lethal

disease characterized by severe hepatic inflammation with hepatocellular injury and necrosis

(Hernandez, et al., 1999; MacDonald, et al., 2002).

A spectrum of responses develops against Trichuris muris, whipworm nematode

infection in genetically different strains of mice. Cellular responses range from a strong Th2

response associated with worm expulsion (BALB/c mouse), to a mixed Th1 and Th2

response and delayed expulsion (C57BL/6 mouse), to finally a Th1 response resulting in a

chronic infection (AKR mouse) (Deschoolmeester, et al., 2002; Anderson, 2000). Additional

studies demonstrated that blockade of the IL-4 receptor in the C57BL/6 resistant mouse

strain resulted in the production of a Th1 response with the development of a chronic

unresolving infection (Else, et al., 1994). Conversely, administration of IL-4 to susceptible

mouse strain BALB/c resulted in the expansion of Th2 response and clearance of infection

(Else, et al., 1994). The T. muris system provides clear evidence that it is the interplay

between host response and parasite survival strategies that lead to the observed infection

levels. It is clear that a host genotype that leads to development of a strong and dominant Th2

response infection will lead to resistance (Grencis, 2001).

22

2.4.2 Immune Responses to Hookworms

The complexity of the hookworm life cycle offers numerous opportunities for the

parasite and host to interact at the molecular level. Extensive antibody responses are mounted

against larval and adult hookworms, but their effect on the parasite remains unclear.

Furthermore, it is difficult to distinguish between anti-larval and anti-adult responses, given

that L3s and adults share many antigens (Carr and Pritchard, 1987; Behnke J, 1991). In many

respects, hookworms are typical gastrointestinal nematodes in the types of immune responses

they generate in their definitive hosts (hosts in which the life cycle is completed i.e. infective

larvae can mature to reach the adult stage). Antibody isotypes and subclasses as well as

cellular responses loosely fit within the framework of a Th2 immune response. The common

features of helminth-induced Th2 responses have long been noted as IgE production,

eosinophilia and mastocytosis. While much of the IgE response in helminth infection is not

directed against the parasite, the detection of IgE antibodies against Necator hookworm L3s

proved to be highly specific and sensitive in diagnosing infections and, furthermore, IgE was

observed to be the least cross-reactive isotype (Ganguly et al., 1988; Pritchard and Walsh,

1995).

Humoral responses (antibody production) to hookworms have been well documented;

little is known about the role of adaptive T cell responses (Loukas and Prociv, 2001). While

hookworm infection exhibits some of the hallmarks of a Th2 response (IgE and local and

systemic eosinophilia), the immune responses clearly fail to protect most infected

individuals. Recently, several studies have described T helper cell responses and

susceptibility to hookworm infection (Quinnell et al., 2004; Pit, et al., 2000 and 2001).

Observations from endemic regions in China and Brazil have shown profound cellular

23

hyporesponsiveness induced by chronic hookworm infection (Loukas et al., 2005). In a re-

infection study in Papua New Guinea, cytokine and proliferative responses to Necator were

measured. Most subjects produced detectable Th1 (IFN-γ) and Th2 (IL-4 and IL-5) cytokines

in response to crude adult worm extract before anthelminthic treatment. Pre-treatment IFN-γ

responses were negatively associated with hookworm burden and increased significantly

after anthelminthic treatment (Quinnell et al., 2004). In a separate study, peripheral blood

mononuclear cells (PBMCs) from N. americanus infected school children, who had recently

received chemotherapy, had reduced proliferative capacity against the phytohemagglutinin

mitogen and adult worm antigen extract compared to controls (Geiger et al., 2004). These

individuals also produced higher levels of IL-10 and lower levels of both Th1 (IL-12 and

IFN- γ) and Th2 (IL-5 and IL-13) cytokines. Such mixed Thl-type and Th2-type immune

responsiveness associated with persisting gastrointestinal parasitic nematodes may reflect a

state of infection where a permissive Th1-type cytokine profile favors parasite persistence

and the chronicity of infection (Pit, et al., 2001).

Over the years immune responses to hookworms have been well-studied in animal

models as well. The Syrian Golden hamster (Mesocricetus auratus) has been used to model

infections with the Ancylostoma ceylanicum hookworm (Garside and Behnke, et al., 1989).

To determine the impact of A. ceylanicum hookworm infection on host cellular responses,

cytokine production and lymphoproliferation were measured (Mendez, et al., 2005). Initial

larval infection with 100 third-stage A. ceylanicum larvae resulted predominantly in Th1

responses characterized by upregulation of IL-2, IFN-γ and TNF-α mRNA levels which

occurred during larval migration and continued up to 14 days postinfection or prepatency

(period before production of eggs). Subsequently, development of larvae into egg-laying

24

adult hookworms or patency coincided with a switch to Th2 predominant responses with a

marked increase in IL-4 and IL-10 production. This switch also concurred with reduced host

lymphoproliferative responses to hookworm antigens (Mendez, et al, 2005).

2.5 Immune responses during pregnancy

It is well-understood that pregnancy is associated with changes in local and peripheral

immune responses which appear necessary for a successful implantation of a semiallogeneic

graft, which is the fetus. Tom Wegmann and others suggested that there was a bidirectional

cytokine interaction between the maternal immune system and reproductive system during

pregnancy and, it appeared that successful pregnancy was a Th2 phenomenon (Wegmann, et

al., 1993). Based on different studies it has been a generally accepted idea that Th1 cytokines

involved in cellular responses are deleterious to pregnancy whereas Th2 cytokines involved

in humoral responses are protective for the fetus (Wegmann, et al., 1993; Nieuwenhoven, et

al., 2002). However, in the light of evidence from studies showing the requirement of IFN-γ

for implantation of blastocyst this would be an oversimplification (Ashkar and Croy, 2001).

Uterine natural killer (uNK) cells were shown to produce 90% of pregnancy-induced uterine

IFN-γ. Implantation sites in uNK cell-deficient and IFN-γ signal-disrupted mice displayed

anomalies in decidua and its spiral arteries which could adversely affect the establishment of

pregnancy (Ashkar and Croy, 2001).

It has also been shown in humans that during pregnancy the ratio of production of

Th1 and Th2 cytokines from peripheral lymphocytes and natural killer cells was decreased

(Nieuwenhoven, et al., 2002). It was previously demonstrated in rats that peripheral

lymphocytes, monocytes and granulocytes shown an activated phenotype in the last week of

25

pregnancy (Faas, et al., 2000). A recent study evaluated the species difference between

humans and rats for lymphocyte cytokine production during pregnancy (Faas, et al., 2005).

The study revealed that during human pregnancy, the percentage of lymphocytes producing

IFN-γ was decreased but the percentage of IL-4 producing lymphocytes was not affected. In

contrast, the rat immune system adapted to pregnancy by decreasing the total number of

various lymphocytic populations but not by affecting the percentage of IFN-γ and IL-4

producing lymphocytes (Faas, et al., 2005).

2.5.1 Effect of Pregnancy Hormones on Immune Responses

There is a large body of data implicating the effect of pregnancy levels of estrogen on

the Th1/Th2 cytokines. Previously published data suggests that female hormones may act to

bias T cells towards a Th2 phenotype (Krishnan, et al., 1996a and b). Support for estrogen’s

role in enhancement of Th2 responses is provided by studies in experimentally-induced

allergic encephalomyelitis (EAE) and collagen induced arthritis (CIA) models; which are

believed to be instructive models for multiple sclerosis (Kim, et al., 1999; Gilmore, et al.,

1997). Treatment of EAE mice with estrogen led to reduction of IFN-γ- dependent anti-

myelin basic protein IgG2a and increased production of IL-10 (Kim, et al., 1999). In the CIA

model, estradiol caused a reduction in IFN-γ expression whereas Th2 cytokine production

was not increased (Gilmore, et al., 1997). Clinical signs in both the diseases improved after

estrogen treatment.

Besides estrogen, progesterone is also likely to contribute to pregnancy being a Th2-

dominant effect (Kidd, 2003). When mice made arthritic with Borrelia burgdorferi, the

causative agent for Lyme disease, were either impregnated or injected with progesterone,

26

arthritic signs ameliorated and IL-4 was markedly increased (Moro, et al., 2001). Joachim

and others showed that dydrogesterone (6-dehydro-retroprogesterone), a progesterone

derivative, dramatically increased the percentage of IL-4 positive decidual immune cells in

stressed mice. Their data suggested that dydrogesterone abrogated stress-triggered abortion

by inducing a Th2 biased local immune response (Joachim, et al., 2003).

Very little data is available to implicate prolactin, the pituitary-derived hormone

responsible for milk letdown after parturition, as a factor affecting Th1/Th2 responses during

pregnancy. It is well known that prolactin has an immunoregulatory function and receptors

for prolactin are present on T and B lymphocytes (Gunes and Mastro, 1997). Prolactin

production increases 10-fold at full-term pregnancy and 30-fold during suckling in the post-

partum period (Ostensen, 1999). Prolactin has also been found elevated in Systemic Lupus

Erythematosus (SLE) patients of both sexes and correlated to disease activity in several

studies (Ostensen, 1999).

2.5.2 Effect of Parasitic Infection on Pregnancy

Relatively few studies have been focused on the effect of infection concurrent with

pregnancy. Infection with apicomplexan parasite Plasmodium falciparum, which causes

malaria, has been recognized to have adverse effects on pregnancy (Brabin and Brabin,

1992). Pregnant women suffer higher incidences of infection, more severe pathology and

higher mortality than any other group of the population (Menendez, 1995). In a report on

cytokine concentrations in placentas collected from women delivering in urban hospitals in

malaria-holoendemic or nonendemic areas of Kenya, normal placentas displayed a bias

toward Th2 cytokines; Th1 cytokines IFN-γ and IL-2 were absent in placentas not exposed to

27

malaria but present in a large proportion of placentas from the holoendemic areas. TNF-α and

TGF-β concentrations were significantly higher in placentas from the holoendemic area

(Fried, et al., 1998). Other studies have also suggested that malaria infection induces a

potentially harmful proinflammatory response in the placenta with a significantly increased

mRNA expression of IL-1β, IL-8, and TNFα (Moormann, et al., 1999).

The protozoal parasite, Neospora caninum, causes neuromuscular disease, abortion,

stillbirth and congenital infection in livestock and companion animals (Dubey and Lindsay,

1996). Reactivation of latent N. caninum is often associated with pregnancy (Quinn, et al.,

2002a). Numerous studies in mice and cattle have shown that immunity to N. caninum

infection involves a predominantly Th1-type immune response as would be expected for

single-celled parasites (Khan, et al, 1997; Lunden, et al., 1998; Tanaka, et al., 2000).

However, a variety of studies on the immune system during pregnancy along with a

concurrent N. caninum infection have demonstrated that immunity during pregnancy is

biased towards a Th2 type response and away from a Th1 response (Quinn, et al., 2002b;

Kano, et al., 2005). Spleen cells from both infected/non-pregnant and infected/pregnant mice

produced IFN-γ, TNF-α and IL-12 (Th1) and IL-10 (Th2); however the levels of Th1 (IFN-γ,

TNF-α and IL-12) cytokines were significantly lower. Infected/pregnant mice exclusively

produced higher levels of IL-4 and it appeared to be responsible for decline in Th1 cytokine

production (Quinn, et al., 2004). Another study conducted in BALB/c mice to examine the

relationship between occurrence of vertical transmission and Th1 /Th2 type of immune

responses suggested that mice infected during pregnancy may acquire a weaker immune

response against N. caninum than mice infected before pregnancy (Kano, et al., 2005).

28

Results also suggested that mice infected during pregnancy may show an enhanced Th2

immune response during recurrence of infection (Kano, et al., 2005).

Leishmania major infection in mice is an excellent model for illustrating the

importance of Th1 response for the control of intracellular protozoan infections (Louis, et al.,

1998; Scott and Farrell, 1998). C3H and C57BL/6 strains of mice mount a strong Th1

response to L. major, which controls parasite multiplication, whereas BALB/c mount a

strong Th2 response and are incapable of resolving the infection (Sadick, et al., 1986;

Heinzel, et al., 1989; Locksley and Louis, 1992). The combination of pregnancy and L. major

infection in resistant mice (C57BL/6) can be predicted to have two outcomes. Firstly,

pregnancy may compromise resistance to L. major infection and secondly, a dominant Th1

response to the parasite may in turn compromise pregnancy (Krishnan, et al., 1996a;

Krishnan, et al., 1996b). In the first scenario where pregnancy was suggested to compromise

resistance to infection, parasite burden was increased when compared with non-pregnant

infected mice. This was seen in association with enhanced expression of cytokines such as

IL-4, IL-5 and IL-10 as well as reduced production of IFN-γ by lymph node and spleen cells

(Krishnan, et al., 1996a). In the second scenario where a Th1 response was implicated for

compromising pregnancy, frequency of viable pregnancies in infected mice was much less

than in pregnant non-infected mice. This corresponded with a relatively low placental

production of IL-4 and IL-10, and an increase in IFN-γ and TNF-α production by placental

cell (Krishnan, et al., 1996b). It was suggested that increased implantation failure or fetal

resorption was probably due to beneficial anti-parasite Th1 responses which were adversely

affecting the pregnancy outcome (Krishnan, et al., 1996b).

29

Figure 2.1 Life cycle of Ancylostoma caninum in the host and environment.

Host

Environment

Larval arrest

EEgggg -- LL11//LL22 -- LL33 -- LL44 -- AAdduulltt ((mmaallee aanndd ffeemmaallee))

30

Figure 2.2 Model of larval arrest and reactivation.

Environment

Egg L1 L2 L3

Pre-parasitic

L3 L4 Adult

Parasitic

L3 Reactivation To newborn Arrested

Host

31

3. Transcript and serum levels of TGF-β and IGF-1 during pregnancy and Ancylostoma

caninum infection in BALB/c mice

3.1 Introduction

Ancylostoma caninum is a blood feeding intestinal nematode found in dogs and is one

of the major parasitic infections in the US pet population (Blaghburn, et al., 1996). It is

closely related to human hookworms, A. duodenale and Necator americanus, which are

estimated to infect over 1.2 billion people worldwide (De Silva, et al., 2003). Like several

other parasitic nematodes, A. caninum also has the capacity to undergo developmental arrest

within the infected host (Schad, 1979, 1990; Schad and Page, 1982). Developmental arrest or

hypobiosis is a strategy used by parasites to evade the host responses, persist for long periods

and subsequently reactivate upon receiving appropriate but unknown stimuli for resuming

development. Reactivation of this latent reservoir of infection can lead to reinfection as well

as transmission of relevant parasitic stages to the surrounding environment or to the neonates

(Shoop, 1991). It has been well documented in dogs that developmentally arrested third-stage

larvae of Ancylostoma sp. reactivate during pregnancy and get transmammarily transmitted to

the suckling newborn puppies via milk (Stone and Girardeau, 1968; Burke and Roberson,

1985). A similar phenomena has been suggested for A. duodenale infection in humans and is

of considerable importance as of the 1.2 billion people infected worldwide with hookworm

disease (A. duodenale and N. americanus), 44 million are pregnant women (Wang, 1988;

Schad, 1991; Bundy, et al., 1995). The reactivation of arrested larvae during pregnancy has

long been hypothesized to be under the influence of hormonal fluctuations associated with

different stages of pregnancy and lactation (Stoye and Krause, 1976). Experiments in which

32

pregnancy associated hormones- estrogen, progesterone and prolactin- were administered

exogenously into ovariectomized, post-partum bitches, a gradual resurgence of L3 stage

larvae were observed in oxytocin-induced milk (Stoye and Krause, 1976).

Caenorhabditis elegans, a well-studied free-living nematode, undergoes similar

arrested development in response to environmental cues like food, temperature and

pheromones. Extensive studies in C. elegans have identified genes that control entry into and

exit from the arrested larval state (Georgi, et al., 1990; Estevez, et al., 1993; Ren, et al.,

1996). One of these genes is daf-7 which has sequence homology to members of the TGF-β

superfamily and is expressed on one of the chemosensory neurons located in the amphidial

region at the anterior end of the worm. Environmental cues such as food exposure triggers

daf-7 expression and correlate with larval recovery from the arrested state (Ren, et al., 1996).

Recently, two TGF-β-like genes, Ac-dbl-1 and Ac-daf-7, were cloned and characterized from

A. caninum suggesting the existence of similar signaling machinery (Freitas and Arasu,

2005). Additionally an in vitro assay was used to show that at physiological concentrations,

recombinant mammalian TGF-β isoforms, TGF-β1 and TGF-β2, had significant stimulatory

effects on tissue-arrested larvae which could be blocked by preincubation with anti-TGF-β

antibodies (Arasu, 2001). However, in vitro analyses also showed that estrogen, prolactin and

insulin did not have a direct effect on the feeding/reactivation response of infective or tissue-

arrested A. caninum larvae (Arasu, 2001). Other studies have shown that expression and

secretion of TGF-β2 is regulated by estrogen and prolactin which are critical factors in the

tissue-specific regulation of the local production of TGF-β2 in the mammary gland and

uterine tissues (Schneider, et al., 1996). These findings suggest host-derived TGF-β might

33

play a role in reactivation of tissue-arrested larvae during pregnancy resulting in

transmammary transmission of infection to the newborn mice puppies.

Another signaling mechanism that has been shown to regulate dauer formation in C.

elegans is daf-2 which is an insulin-like receptor gene (Kimura, et al., 1997). IGF-1 which is

also upregulated during pregnancy might mediate a separate signaling pathway in larval

reactivation (Lee, et al., 1993). Since phosphatidylinositol-3-OH kinase (PI3K) is component

of the insulin-like signaling pathway, role of insulin-like signaling in hookworm larval

activation was investigated in vitro using the PI3K inhibitor, LY294002, and resumption of

feeding was considered as a marker for activation (Brand and Hawdon, 2004). Muscarinic

agonists like arecoline have been shown to initiate dauer recovery in C. elegans and A.

caninum (Tissenbaum, et al., 2000). LY294002 prevented feeding in A. caninum infective

larvae stimulated with host serum filtrate, a glutathione-analogue and the muscarinic agonist

arecoline. These results suggest that an insulin-like signaling pathway mediates the

reactivation of hookworm larvae similar to C. elegans. However, it has been shown that

direct stimulation of A. caninum larvae with physiological concentrations of insulin did not

elicit a stimulatory response (Arasu, 1999).

A third signaling pathway in C. elegans for dauer arrest and reactivation involves

cGMP. The cGMP pathway includes a membrane associated guanylyl cyclase and loss-of-

function mutants of this guanylyl cyclase enter into the dauer stage constitutively (Birnby, et

al., 2000). The cell permeable cGMP-analogue, 8-bromo-cGMP, has been shown to mediate

recovery of C. elegans dauers as well as A. caninum L3 larvae from arrested state to feeding

state (Birnby, et al., 2000; Hawdon and Datu, 2003; Brad and Hawdon, 2004). Similar results

were seen with the congeneric hookworm A. ceylanicum (Brand and Hawdon, 2004). Taken

34

together, these studies suggest that several signaling pathways may be similarly relevant

during post-partum lactation in the reactivation and transmission of hookworm larvae.

To facilitate in vivo analyses of mechanisms mediating the pregnancy-induced

reactivation and transmammary transmission, our lab has defined the mouse as an

experimental model of infection. Mice serve as normal paratenic hosts for A. caninum as it

has been previously shown that the larvae distribute and arrest in different tissues throughout

the body without maturing into the adult intestinal stage (Nichols, 1956; Lee, et al., 1975;

Arasu and Kwak, 1999). Furthermore, we along with others have shown that similar to dogs,

pregnancy in mice triggers the reactivation and transmammary transmission of larvae to the

neonatal pups (Steffe and Stoye, 1984; Burke and Roberson, 1985; Arasu and Kwak, 1999).

In this study we investigated TGF-β1, TGF-β2 and IGF-1 cytokine profiles during late

pregnancy, early lactation and mid-lactation in the mouse model of A. caninum infection to

correlate their levels with the transmammary transmission of the larvae to the nursing pups.

We measured the changes in serum levels of TGF-β1, TGF-β2 and IGF-1 as well as mRNA

transcript levels in skeletal muscle (site of arrest), mammary gland (site of transmammary

transmission) and spleen as the indicator organ of immune responses to the parasitic

infection.

3.2 Materials and Methods

3.2.1 Larval Parasite cultures

Infective A. caninum L3 were harvested from charcoal co-cultures of feces from male

laboratory beagle dogs (Marshal Farms, North Rose, NY) orally infected at 8-10 weeks of

age with 150 L3 as previously described (Burke and Roberson, 1979; Arasu, 1997). The use

35

of dogs in this study was approved by the Institutional Animal Care and Use Committee at

North Carolina State University. Feces containing A. caninum eggs (range of 500-3000

eggs/g depending on stage of infection) were mixed with activated charcoal and deionized

water in Pyrex® baking trays and placed in a humidified incubator, in the dark, at 25ºC for 7-

10 days. The L3 were harvested utilizing a modified Baermann funnel assembly (Garcia and

Ash, 1979), repeatedly washed in phosphate-buffered saline (PBS) medicated with 20 mg/L

gentamicin (Sigma Chemical Co., Missouri) and 20mg/L lincomycin (Sigma Chemical Co.,

Missouri) and maintained at room temperature for 5-7 days prior to use.

3.2.2 Mice breeding

Nine to ten-week-old female BALB/c mice were purchased from Charles River

Laboratories, and maintained at the Laboratory Animal Resources facility in accordance with

Institutional Animal Care and Use Committee guidelines. Mice were mated at a ratio of 2

females: 1male for a period of 7 days and examined twice daily for the presence of a vaginal

impregnation plug. Observation of a plug was designated as day 0 of the pregnancy. On day

5 post-impregnation (observation of vaginal plug), the female mice were injected sub-

cutaneously in the dorsal cervical interscapular region with PBS (control) or with 1000 larvae

in PBS. Mice were killed by decapitation at three different time points: time point 1

corresponding to day 19 of gestation or day 14 post-infection (pi), and time points 2 and 3

corresponding to day 1 and 10 of postpartum lactation or day 15-16 pi and day 25-26 pi,

respectively (Figure 3.1). The tissue samples collected immediately after killing were

sections of the gastrocnemius skeletal muscle, mammary gland, and spleen as well as blood.

Approximately 200mg of muscle, mammary tissue or spleen was stored in RNA-STAT 60

36

(Tel-Test, Texas) at -80ºC for RNA extraction. From the blood, serum was collected after

incubation at 37ºC for 1 h, 4ºC for 4 h and subsequently stored at -20ºC until further use. The

female mice were grouped as infected or not infected and bred or not bred (n = 5 mice per

group). The resulting four groups were uninfected/unbred (UN or normal controls),

infected/unbred (IN), uninfected/bred (UB) and infected/bred (IB).

3.2.3 Harvesting tissue L3

For harvesting tissue-arrested larvae, carcasses were skinned, minced, wrapped in

eight layers of cheesecloth and incubated in a modified Baermann assembly for 4 h at 37ºC

in medicated PBS (Arasu, 1998). Migrating larvae that collected at the bottom of the cups

were repeatedly washed with medicated PBS and gently spun at 2000 rpm for 10 minutes.

Larval numbers in dams versus each litter of pups were counted with the aid of a dissecting

microscope.

3.2.4 TGF-β1 and –β2 ELISA

Biologically active TGF-β1 and β2 levels were determined using the TGF-β1 and β2

Emax® Immunoassay system (Promega, Madison, WI), a sandwich enzyme linked

immunosorbent assay (ELISA) kit. The assays were performed by coating flat-bottom 96-

well ELISA plates (Corning Costar® Acton, MA) with 100 µl of TGF- β1 and TGF- β2 coat

antibody dissolved in carbonate coating buffer. The plates were incubated at 4ºC overnight.

Non-specific binding sites were blocked using 270 µl of TGF- β1 and TGF –β2 blocking

buffer at 37ºC for 35 minutes. Serum samples diluted 1:4 using PBS and for each 50 µl of

diluted sample, 1µl of 1N HCl was added and the pH was verified to be 3.0 or lower. Serum

37

samples were mixed, incubated for 15-20 minutes at room temperature and neutralized by

adding 1µl of 1N NaOH; the pH was verified to be approximately 7.6. According to the

manufacturer, TGF-β1 standard curve was linear between 15.6 and 1000 pg/ml of the TGF-

β1 standard whereas the TGF-β2 standard curve was linear between 32 and 1000 pg/ml. The

TGF-β1 and TGF-β2 standard were supplied at the concentration of 1µg/ml and were diluted

1:1000 in TGF-β sample buffer. For each plate (100µl/well), six 1:2 serial dilutions of the

standards (starting with 1 ng/ml) and acidified serum samples (starting with 1:20) were

analyzed. The TGF-β1 and TGF –β2 controls were set at 100 pg/ml and 500 pg/ml for each

plate. The 96-well plate was incubated on a plate shaker at 500 rpm for 1.5 hours at room

temperature. After washing five times with TBST (20 mM Tris-HCl, pH 7.6, 150 mM NaCl,

0.05% Tween-20) wash buffer, anti-TGF-β1 and TGF-β2 polyclonal antibody were

respectively added at 1:1000 and 1:2000 dilutions and incubated with shaking for an

additional 2 hours at room temperature. The plates were washed five times with TBST buffer

followed by 100 ul of the TGF-β/horseradish peroxidase (HRP) conjugates diluted 1:100

with sample buffer. The plates were incubated with shaking for 2 hours at room temperature

and subsequently washed five times at the end of the incubation. For color development

100µl of 3, 3', 5, 5'-Tetramethylbenzidine (TMB) One was added to each well using a

multichannel pipettor and the plates were incubated for 15 minutes without shaking. The

color reaction was stopped by adding 100µl of 1N HCl in the same order in which substrate

was added to the wells. The blue color changed to yellow as the pH decreased. The

absorbance was recorded at 450nm on a plate reader within 30 minutes of stopping the

reaction.

38

3.2.5 IGF-1 Enzyme ImmunoAssay (EIA)

IGF-1 levels in mouse sera were determined using the Active® Mouse/Rat IGF-1 EIA

kit based on a competitive binding enzyme immunoassay (Diagnostic Systems Laboratory,

Inc., Webster, Texas). The kit contains microtitration strips coated with rabbit anti-goat

gamma globulin immobilized on the inside of each well. Pretreatment of samples involved

addition of 140 µl of the provided Sample Buffer I to 10 µl of mouse serum and incubation at

room temperature (~25ºC) for 30 minutes. Subsequently, 150 µl of Sample Buffer II was

added and mixed thoroughly. The mouse/rat IGF-1 standards were used at 0, 150, 300, 650,

1250 and 3600 ng/ml. The two controls, Level I and Level II, contained low (600 ± 175

ng/ml) and high (1100 ± 325 ng/ml) levels of mouse/rat IGF-1 in a protein-based buffer,

respectively. In the assay, 50 µl of standards, controls and pretreated sera samples were

incubated on an orbital microplate shaker at 500 rpm for 1 h at room temperature with 100 µl

of biotin-labeled mouse/rat IGF-1 and 100 µl of goat anti-mouse/rat IGF-1 antiserum in

microtitration wells allowing the unlabeled and biotin-labeled antigens to compete for the

limited number of anti-mouse/rat IGF-1 binding sites. Wells were washed five times with the

wash solution and incubated with streptavidin-horseradish peroxidase (HRPO), which binds

to the biotinylated mouse/rat IGF-1. The unbound streptavidin-HRPO was washed, followed

by incubation with 100 µl of substrate TMB chromogen solution. An acidic stopping solution

was then added to each well with a multichannel pipettor and the plate was shaken by hand

for 5-10 seconds. The degree of enzymatic turnover of the substrate was determined by dual

wavelength absorbance measurement at 450 with background wavelength correction set at

600 or 620 nm.

39

3.2.6 RNA extraction and cDNA synthesis

Total RNA from all the tissues (skeletal muscle, mammary gland and spleen) was

extracted using RNA-STAT (Tel-Test, Inc., Friendswood, TX) according to Chomczynski

and Sacchi method (Chomczynski and Sacchi, 1987). Residual genomic DNA was removed

from RNA by treating with DNase (20U per 100 µg of RNA). RNA quantity and quality was

checked by spectrophotometric measurements at 260 and 280 nm (Pharmacia) and by

analyzing 1µg RNA for rRNA bands on a 1% ethidium bromide agarose gel.

For reverse transcription, first strand cDNA was synthesized by adding 10µg of total

RNA to a 40µl reaction mix containing a final concentration of 2.5nM dNTPs, 0.1M,

dithiothreitol, 1ug Oligo d (T) primer (Promega, Madison, WI), 24U RNase inhibitor and

incubation with 200U Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) at 37°C

for 1 hour. To confirm absence of contaminating genomic DNA, RT negative reactions with

5µg total RNA were setup. RT-plus (cDNA)/minus reactions were subjected to RT-PCR

(reverse transcription-polymerase chain reaction) using GAPDH primers as it was abundantly

expressed in all tissues of interest. PCR reactions were run on a 2% agarose gels to check for

product formation.

3.2.7 Primer design

Primer sets for genes of interest, TGF-β1, -β2, IGF-1, IL-4 and IFN-γ, and reference

gene 60S acidic ribosomal protein were designed from mouse-specific Genbank sequences

listed in Table 3.1 using Primer 3 software (Rozen and Skaletzky, 2000) (http://www-

genome.-wi.mit.edu/cgi-bin/primer/primer3_www.cgi). In order to minimize primer-dimer

formation, the maximum self-complementarity was 6 and the maximum 3’ self-

40

complementarity was 0. The targets amplified by primer pairs were characterized using the

Mfold program (SantaLucia, 1998) (http://bioinfo.math.rpi.edu/~mfold/dna/form1.cgi) for

predicting the nature of any secondary structures which may form at the site of primer

binding. Primer pairs that bound at the site of predicted loop were discarded. Primer sets

were synthesized by Integrated DNA Technologies (Coralville, IA) and primers were

reconstituted at100 pM/ul in nuclease-free water prior to use.

3.2.8 Real-time PCR

The real-time PCR reactions were carried out using the iCyclerTM iQ PCR detection

system (Bio-Rad Laboratories, Hercules, CA, USA). In each 25µl reaction, 12.5µl of iQTM

SYBR green supermix (Bio-Rad) was added to 300nM of each primer along with 250ng of

cDNA. PCR amplification was performed in duplicate for each sample using the following

cycle conditions: 3 min at 95°C followed by 45 repeats of 1 min at 95°C, 30s at 55°C and 30s

at 72°C. Temperature optimization was carried out for all the primer sets to be amplified

simultaneously. Annealing temperatures were tested from the 50-65ºC range; all the primer

sets amplified optimally at 55ºC. A melt curve analysis step was included at the end of cycles

to check for primer-dimer and non-specific product formation. Efficiency of the PCR

reactions was derived by doing a standard curve of 10-fold serially diluted mouse spleen

cDNA and was consistently in the 95-98% range. Non-template controls were used to detect

any genomic DNA contamination and amplified products were also examined on a 2%

agarose gel to verify that the amplified products were of the expected sizes. Raw Ct values

were analyzed using Relative Expression Software Tool-384 (REST-384) to generate a fold

increase or decrease in the transcript levels (Pfaffl, et al., 2002). The 60S acidic ribosomal

41

protein was used as the endogenous reference gene for normalizing transcript levels among

tissues of interest. 60S ribosomal protein was previously shown to be the least variable from

comparative analyses of various genes including β-actin, cyclophilin and glyceraldehyde-3-

phosphate dehydrogenase (GAPDH) all of which were tested as potential candidate reference

genes similar to the ones described in Chapter 6.

3.2.9 Statistics

Data reported for larval burden and serum levels of cytokines as mean ± standard

deviation was analyzed by two-way ANOVA using the General Linear Models (GLM)

procedure in SAS (Cary, North Carolina) and p< 0.05 was considered significant. For

adjustment of multiple comparisons, Tukey`s test was used to detect significant differences

between uninfected controls, infected, not bred and bred groups of mice. For comparing

variation in transcript levels of cytokine genes, Pair-Wise Fixed Reallocation Randomization

Test© in the REST-384 was used (Pfaffl, et al., 2002). The mathematical model used to

compute the relative expression ratio of a target gene relies on its real-time PCR efficiencies

(E) and the threshold cycle difference (∆Ct) of an unknown sample versus a control

(∆Ct control - sample). The target gene expression is normalized to a reference gene (ref) in the

Equation I mentioned below:

Ratio = (E target) ∆Ct target (control - sample) / (E ref)

∆Ct ref (control - sample)

In our experimental setup control was the uninfected/unbred (UN) mice and samples were

the infected/unbred, uninfected bred (UB) and infected/bred (IB). For evaluating fold

changes in transcript levels, analyses were also done by comparing IN, UB and IB group

42

each to the UN at respective timepoints. For example, Equation II mentioned below shows

calculations for changes in transcript levels of IN when compared to UN:

Ratio = (E target) ∆Ct target (UN - IN) / (E ref)

∆Ct ref (UN - IN)

For evaluating fold upregulation or downregulation of transcripts between timepoints, early-

lactation (timepoint II) and mid-lactation (timepoint III) were compared to late-gestation

(timepoint I). For example, Equation III mentioned below shows calculations for changes in

transcript levels of early lactation (timepoint II) when compared to late-gestation (timepoint

I):

Ratio = (E target) ∆Ct target (timepoint I - timepoint II) / (E ref)

∆Ct ref (timepoint I - timepoint II)

3.3 Results

3.3.1 Assessment of larval burden

For comparing larval distribution in dams, eight week old female BALB/c mice were

subcutaneously injected with 1000 A. caninum L3 larvae. A comparison of larval burden

during the course of A. caninum infection was made between infected/not bred (IN) and

infected/bred (IB) mice at the three different timepoints corresponding to late gestation and

early to mid lactation (Figure 3.2). The total number of larvae recovered from IN mice was

not significantly different from the IB mice at late gestation or mid lactation. However, there

was a significant difference (p< 0.005) in the larval burden of IN mice (368 ± 64) and IB

mice (249 ± 31) at timepoint II, during the initial 24 hours of post-partum lactation.

To correlate the level of transmammary transmission from dam to pups, larval counts

were also assessed in pup litters from IB dams: 5 ± 4 and 9 ± 2 larvae were respectively

obtained during the timepoints for early and mid lactation (for timepoints II and III,

43

respectively). The larval burden in the feti (timepoint I) was not assessed as previous studies

have shown that there is no in utero transmission of A. caninum larvae from dam to fetus

during pregnancy (Steffe and Stoye, 1984; Arasu and Kwak, 1998).

3.3.2 Serum levels of TGF-β1, TGF-β2 and IGF-1 during A. caninum infection and breeding

TGF- β1 serum levels

To compare the TGF- β1 levels during different phases of breeding, sera was

collected at the three time points indicated above spanning gestation and post-partum

lactation. TGF-β1 levels were assessed in uninfected/not bred (UN) controls and compared to

the levels in infected/not bred (IN), uninfected/bred (UB) and infected/bred (IB) group of

mice. At timepoint I (late gestation), the circulating levels of TGF-β1 were not significantly

different between UN, IN and UB groups; however, there was a significantly higher level in

IB mice when compared with normal controls (p< 0.0001) (Figure 3.3a). At timepoint II,

TGF-β1 levels were significantly lower among bred groups, UB (p< 0.0001) and IB (p<

0.05), when compared to normal (Figure 3.3b). At timepoint III, all three IN, UB and IB

groups were significantly different from controls; TGF-β1 levels were significantly elevated

in IN mice (p< 0.0001) whereas the levels were significantly lower in both UB (p< 0.0001)

and IB (p< 0.0001) (Figure 3.3c). In figure 3.3d, comparison of TGF-β1 levels among

different mouse groups was done over 3 timepoints of interest (I, II and III). In summary,

TGF-β1 dropped significantly at timepoints II and III in the UB and IB groups as compared

to controls suggesting that TGF- β1 levels decline after parturition and remains so until mid-

lactation. Also, the significant increase in TGF-β1 levels at timepoint III in the IN groups

suggests a time-dependent role for TGF-β1 during A. caninum infection.

44

TGF- β2 serum levels

Circulating TGF- β2 levels remained unchanged and did not vary significantly

between IN, UB and IB when compared to the UN controls at timepoint I (Figure 3.4a). At

timepoint II, TGF-β2 levels decreased significantly in UB (p≤ 0.0001) and IB (p≤ 0.0001)

when compared to the UN mice. However, TGF-β2 levels did not vary significantly in the IN

mice compared to the UN mice (Figure 3.4b). At timepoint III, significantly lower TGF- β2

levels were observed in all the groups (IN, UB, IB) when compared to UN mice (p< 0.0001)

(Figure 3.4c). In figure 3.4d, comparison of TGF-β2 levels among different mouse groups

was done over 3 timepoints of interest (I, II and III). Therefore, an overall significant drop in

TGF-β2 levels was observed at timepoints II and III in the UB and IB groups when compared

to UN mice corroborating with previous findings where higher TGF-β2 levels were

documented during pregnancy and significantly lower levels after parturition and beginning

of lactation (Schneider, et al., 1996).

IGF-1 serum levels

Serum IGF-1 levels decreased significantly in IN (p< 0.0001) and UB (p< 0.0001)

compared to UN at timepoint I (Figure 3.5a). IGF-1 levels in all groups were similar at

timepoint II (Figure 3.5b). At timepoint III, IGF-1 levels were significantly lower in IN (p<

0.0001), UB (p< 0.0001) and IB (p< 0.0001) groups when compared to UN mice (Figure

3.5c). In figure 3.5d, comparison of IGF-1 levels among different mouse groups was done

over 3 timepoints of interest (I, II and III). In summary, a significant drop in IGF-1 levels in

the IN group was observed at two weeks and continued to remained so at the later timepoint

45

of A. caninum infection (timepoint III). Similarly, serum IGF-1 levels decreased below

normal levels at the end of gestation (timepoint I) and during mid-lactation (timepoint III) in

bred and infected/bred groups.

3.3.3 Transcript levels of TGF-β1, TGF-β2 and IGF-1 in muscle, mammary gland and spleen

TGF-β1 transcript levels

TGF-β1 transcripts levels were evaluated during the course of pregnancy and

infection with A. caninum using real-time PCR. REST software was used to analyze the fold

changes in expression of transcript levels (Pfaffl, et al., 2002) in skeletal muscle, mammary

tissue and spleen. Transcript levels of TGF-β1 were evaluated using equation II and III

(Materials and Methods). Data presented in Figures 3.6 a-i mentioned in the discussion

section of this chapter was calculated by comparing IN, UB and IB with UN controls

(Equation II). Data presented in Figures 3.9 a-i was calculated by comparing timepoint II and

III with timepoint I (Equation III) as discussed in this section in more detail. In skeletal

muscle, TGF-β1 transcripts levels were significantly downregulated at timepoint III in IN (p<

0.001), UB (p< 0.01) and IB (p< 0.05) groups with 37, 13 and 22-fold decrease compared to

timepoint I (Figure 3.9a, 3.9b and 3.9c). In mammary gland, TGF-β1 transcripts levels were

significantly upregulated by 11-fold at timepoint III in the IN (p< 0.05) group compared to

timepoint I (Figure 3.9d). A significant upregulation by 29 and 17-fold in TGF-β1 transcripts

levels was observed in the mammary gland of UB (p< 0.001) and IB (p< 0.01) mice

respectively during early lactation i.e. timepoint II (Figure 3.9e and 3.9f). In spleen, TGF-β1

transcripts levels were significantly downregulated at timepoint II in IN (p< 0.05) (Figure

3.9g). However, TGF-β1 transcripts levels were not significantly different in UB and IB

46

groups in spleen (Figure 3.9h and 3.9i). In summary, an overall downregulation of TGF-β1

transcripts was observed in skeletal muscle 24 days into A. caninum infection in the IN group

and during mid-lactation in UB and IB groups. However, in mammary gland significant

upregulation of TGF-β1 transcripts was detected 24 days into A. caninum infection in IN

group and during early lactation in UB and IB groups.

TGF-β2 transcripts levels

TGF-β2 transcripts levels were evaluated during the course of A. caninum infection

and different stages of pregnancy and lactation in skeletal muscle, mammary gland and

spleen. Previous studies have documented the upregulation of TGF-β2 in mammary gland

during late pregnancy and this level of expression is maintained during the lactational phase

(Schneider, et al., 1996). Transcript levels of TGF-β2 were evaluated using equation II and

III (Materials and Methods). Data presented in Figures 3.7 a-i mentioned in the discussion


(Equation II). Data presented in Figures 3.10 a-i was calculated by comparing timepoint II

and III with timepoint I (Equation III) as discussed in this section in more detail. In skeletal

muscle, TGF-β2 transcripts levels were significantly downregulated at timepoint III in IN (p<

0.001), UB (p< 0.001) and IB (p< 0.001) groups with 18, 5 and 61-fold decrease compared to

timepoint I (Figure 3.10a, 3.10b and 3.10c). In mammary gland, there was a significant

downregulation of TGF-β2 transcripts in the IN groups at timepoint I (p< 0.01), II (p< 0.001)

and III (p< 0.01) (Figure 3.10d). During early lactation, there was a significant upregulation

of TGF-β2 transcripts in mammary gland with an 85 -fold increase in the UB group (p< 0.05)

at timepoint II (Figure 3.10e). There was no significant change in TGF-β2 transcripts in

47

mammary gland of IB group (Figure 3.10f). In spleen, TGF-β2 transcripts were significantly

upregulated with a 35 -fold increase at timepoint I (p< 0.01) in the IN group (Figure 3.10g).

Similarly, at timepoint I (p< 0.05) significant upregulation with a 37 -fold increase was also

observed in the UB group (Figure 3.10h) but no significant changes were observed in spleen

of IB group (Figure 3.10i). In summary, TGF-β2 transcripts levels were significantly

downregulated in the skeletal muscle at 24 days of A. caninum infection in the IN group and

mid-lactation in the UB and IB bred groups. In mammary gland, there was a significant

downregulation in TGF-β2 transcripts levels at all the timepoints of A. caninum infection in

the IN group. A significant upregulation observed in TGF-β2 transcripts during early

lactation in the UB group was similar to previously published studies done in mammary

gland (Schneider, et al., 1996). However, no upregulation of TGF-β2 transcripts was

observed in the IB group at any of the timepoints during pregnancy or lactation.

IGF-1 transcripts levels

IGF-I mRNA content in mammary glands has been shown to be upregulated during

pregnancy (Lee, et al., 1993). Transcript levels of IGF-1 were evaluated using equation II and

III (Materials and Methods). Data presented in Figures 3.8 a-i mentioned in the discussion


(Equation II). Data presented in Figures 3.11 a-i was calculated by comparing timepoint II

and III with timepoint I (Equation III) as discussed in this section in more detail. In skeletal

muscle, IGF-1 transcript levels were significantly upregulated at timepoint III (p< 0.05) in

the IN groups compared to timepoint I (Figure 3.11a). No significant changes were observed

in skeletal muscle of UB group (Figure 3.11b). IGF-1 transcript levels were also significantly

48

upregulated in skeletal muscle at timepoint II in the IB group compared to timepoint I (p<

0.05) (Figure 3.11c). In mammary gland, IGF-1 transcript levels were significantly

downregulated at the timepoints I (p< 0.01) and II (p< 0.001) but upregulated at timepoint III

(p< 0.01) in the IN groups (Figure 3.11d). No significant changes were observed in IGF-1

transcript levels in mammary gland of UB group (Figure 3.11e). However, in the IB group

IGF-1 transcript levels were significantly upregulated in the mammary gland at timepoint II

(p< 0.01) and timepoint III (p< 0.01) with a 9 and 112- fold increase compared to timepoint

I(Figure 3.11f). IGF-1 transcript levels did not vary significantly in spleen of IN, UB and IB

mice at the timepoints of interest (Figures 3.11g, 3.11h and 3.11h). In summary, IGF-1

transcript levels were upregulated in skeletal muscle and mammary gland at 24 days of A.

caninum infection. IGF-1 transcript was upregulated in skeletal muscle and mammary gland

of IB mice during early to mid-lactation.

3.4 Discussion

Mice serve as an excellent model to study host-parasite interactions of A. caninum as

they are paratenic hosts and encapsulate the phenomena of arrest and pregnancy-associated

reactivation leading to subsequent transmammary transmission as observed in dog and

human Ancylostoma hookworm infection. As reported previously, infection of mice with A.

caninum results in the gradual migration of L3s throughout the body particularly to the

skeletal muscles (Arasu and Kwak, 1999) with stable distribution by 10-14 days post-

infection (Lee, et al., 1975; Steffe and Stoye, 1984). Due to previously observed low

breeding efficiencies with early infection (Arasu and Kwak, 1999), we followed the infection

protocol of sub-cutaneous injection of 1000 L3 at 5 days post-impregnation which allows for

49

larval migration/distribution during the remaining time of the 19-21 day gestational period.

In the studies reported here, the total number of tissue larvae harvested was not significantly

different at day 19 of gestation between bred and unbred groups of infected mice which is

consistent with the absence of pre-partum transmission of infection. However, there were

significant differences observed in the total number of larvae between bred and unbred

infected groups at the timepoint corresponding to day 1 of lactation suggesting that

transmammary transmission starts within the first 24 hours of suckling. At day 10 of

lactation, even though the total number of larvae was lower in bred versus unbred infected

mice, no significant difference was observed. Previous studies have shown that in

experimentally infected dogs, only a subpopulation of arrested larvae are transmitted during

lactation leaving a residual population (re-entering a phase of arrest) and available for

transmission in following pregnancies (Stoye, 1973).

Host factors that facilitate the transition of arrested larval stages to a reactivated state

for transmammary transmission during hookworm infection have not yet been clearly

defined. Though the precise nature of these host factors remains unknown, it appears to

involve specific components of serum (Hawdon and Schad, 1990, 1991, 1992). Parallels have

also been drawn between the arrested dauer larval stage of C. elegans and the third-stage

larvae from parasitic nematodes (Hotez, et al., 1993; Rajan, 1998). TGF-β and insulin-like

signaling have been shown to regulate dauer formation in C. elegans (Georgi, et al., 1990;

Estevez, et al., 1993; Ren, et al., 1996; Kimura, et al., 1997). Studies utilizing an in vitro

feeding assay have shown that both TGF-β1 and TGF-β2 are able to stimulate tissue-arrested

A. caninum larvae (Arasu, 1999). In the host, TGF-β1 appears to be required for

regeneration of skeletal muscle tissue and is produced by eosinophils and mast cells

50

associated with the inflammatory response to parasites (Wong, et al., 1991; Baumgartner, et

al., 1996; Maizels, et al., 1993; Cooper, et al., 1991). It is also an important cytokine

detectable at high levels in feto-placental tissues during pregnancy (Thompson, et al., 1989).

In these studies, TGF- β1 serum levels were significantly higher in the infected bred IB group

(Fig. 3.3a) group at day 19 of gestation relative to all other groups, and immediately after

parturition relative to the uninfected bred group (Fig. 3.3b), suggesting that infection as well

as pregnancy/lactation may play a role in sustained production of TGF- β1 and a potential

role in reactivation of arrested larvae in muscle sites. Differences in circulating level of TGF-

β1 were found in UN control mouse during 3 timepoints (Fig. 3.3d). At present, we are

unable to explain such changes in serum levels. However, it is possible that TGF- β1levels

fluctuate during stress which can be caused by loud noise or change in environment in the

housing facility. With infection, an increase in TGF- β1 levels was seen at timepoint III in the

IN group which could suggest an increase in cells secreting TGF- β1 that respond to infection

such as eosinophils and mast cells (Figure 3.3d). Comparison of mRNA levels revealed that

TGF- β1 transcripts were significantly downregulated in skeletal muscle at ~24 days of A.

caninum infection in both IN and IB groups (timepoint III) which would be consistent with

the period of stabilized dispersion and diminished migration of the larvae; however, a similar

decline was also seen in uninfected bred UB mice (Fig. 3.9b). When UB was compared to

UN, TGF- β1 transcripts were significantly downregulated at timepoint III in skeletal muscle

(Figure 3.6b). In mammary gland, TGF- β1 mRNA levels were significantly elevated during

early lactation but eventually declined during mid-lactation in both uninfected and infected

bred groups of mice (Fig. 3.9e and 3.9f). When UB and IB were compared to UN, TGF- β1

transcripts showed a significant downregulation it was not as dramatic as observed during IN

51

and UN comparison during timepoint II (Fig. 3.6d, e and f). Also, during timepoint III there

was a significant upregulation in TGF-β1 transcript when IN was compared to UN (Fig.

3.6d). These findings suggest that dramatic suppression of TGF-β1 transcript seen during

early infection is overcome by ~24d of infection and pregnancy appears to offset such

suppression as observed by less dramatic downregulation in UB and IB groups (Fig. 3.6 e

and f). Similarly, a significant downregulation is also observed in spleen of IN mice during

timepoint II (Fig. 3.6g).

In these studies, the infected bred IB group of mice is of greatest interest when

assessing the fluctuations in TGF-β isoforms as this group encompasses the phenomena of

larval arrest and transmission in arrested A. caninum larvae. We expected TGF-β1 levels to

be elevated during the peri-parturient period as this is the timepoint when arrested larvae are

assumed to `awaken’ to resume migration. Serum levels were indeed elevated in the IB group

of mice compared to all other groups at late gestation (Fig. 3.3a). However, TGF- β1

transcript levels in skeletal muscle were generally not elevated (Figs. 3.9a, 3.9b and 3.9c)

except for showing a slight increase at timepoint II (day 16-17 pi/day 1 lactation) of the

infected groups, IN and IB (Fig. 3.9a and 3.9c). Taken together with previous in vitro

feeding/reactivation results on tissue-arrested larvae (Arasu, 1999), these results suggest that

TGF- β1 may potentially be playing a role in larval reactivation in the in vivo situation but

would require further confirmatory analyses.

With TGF β2, elevated levels were detected in late pregnant maternal plasma in the

rat model which declined immediately after parturition (Schneider, et al., 1996). In contrast

to TGF-β1, TGF-β2 levels in the UN groups remained constant over time. Our results from

the mouse show a similar trend with no significant differences in serum levels amongst any

52

of the groups at late gestation followed by a significant decrease in UB and IB groups

relative to UN subsequent to parturition (Fig. 3.4a). TGF-β2 levels declined sharply in the

IN group at timepoint III during which larvae are expected to undergo arrest (Figure 3.4d).

This decline suggests the possibility that initiation of arrest might lead of secretion of a TGF-

β2 receptor or cells making TGF- β2 in response to A. caninum infection get inhibited. With

the TGF-β2 transcripts, levels were significantly elevated in the mammary gland and uterine

tissues during late gestation and lactation in rats (Schneider, et al., 1996). Our findings

corroborate these studies as a similarly significant upregulation was observed in TGF-β2

transcripts in mammary gland during early lactation in the uninfected bred group (Fig.

3.10e). TGF-β2 transcript levels were significantly downregulated in skeletal muscle at

timepoint III ie mid-lactation in UB mice when compared to UN controls which correlate

with decline in circulating levels also (Fig. 3.7b and 3.4d). However, it should also be noted

that numerous studies have shown that mRNA transcript levels are not always reflective of

protein levels (Anderson and Seilhamer, 1997; Gygi, et al., 1999). Hence, there was a

significant upregulation of TGF-β2 transcript in the in mammary gland of UB group at

timepoint II coinciding with sustained expression of TGF- β2 during early lactation whereas

serum levels had started to decline sharply (Fig. 3.7e). In the IB group relative to other

groups, we had accordingly expected TGF-β2 transcript levels to be elevated in mammary

gland during the lactational period as reflective of high localized levels of the expressed

cytokine in order to sustain a state of larval reactivation and facilitate transmission of larvae

via milk. Serum levels of TGF- β2 were not remarkable (as mentioned above) and consistent

with the well-documented tissue-specific localization of TGF- β2 (Cheng, et al., 1993;

Schneider, et al., 1996). Transcript levels of TGF- β2 were also not significantly different in

53

the IB group. From these observations, there does not appear to be a correlation between

host-derived TGF-β2 and larval reactivation.

In studies from rats and pigs, IGF-1 mRNA and protein levels are differentially

regulated during pregnancy (Marcotty, et al., 1994; Lee, et al., 1993); in rats, maternal serum

IGF-I levels were high during the first half of pregnancy and fell sharply during the second

half of pregnancy (Marcotty, et al., 1994). Using RNase protection assay, IGF-I gene

expression in the mammary gland diminished during the course of pregnancy (Marcotty, et

al., 1994). In our studies, serum IGF-1 levels in IB mice were not significantly different from

normal UN mice during the peri-parturient period (timepoints I and II; Figures 3.5a, 3.5b) but

showed a gradual decrease from ~800 pg/ml at late gestation to ~100 pg/ml at day 10

lactation. Infection at timepoint II ie early lactation had an effect on IGF-1 serum levels as

observed by an increase in IN and IB groups (Figure 3.5d). No significant differences were

observed in IGF-1 levels in skeletal muscle of any group of mice when compared with the

UN (Fig. 3.8a, b, c). However, there was a significant upregulation in IGF-1 transcript in the

IB group at parturition when comparing day 19 gestation and day 1 lactation timepoints in

both skeletal muscle (Fig. 3.11c) and mammary gland (Fig. 3.11f). These in vivo analyses of

IGF-1 levels show a potential correlation between infection and pregnancy/lactation with a

potential role in the reactivation/transmission of arrested larvae. In previous in vitro studies

to identify factors associated with signaling and reactivation of tissue-arrested A. caninum

larvae, insulin did not appear to have a direct effect on the feeding response (Arasu, 1999);

IGF-1 was not assessed in these studies. However, the IGF-1 signaling inhibitor LY294002

was shown to prevent a feeding response in A. caninum infective larvae stimulated with dog

54

serum (Hawdon and Datu, 2003) supporting the hypothesis that an insulin-like signaling

pathway involving IGF-1 may be involved in larval reactivation during pregnancy.

In order to directly implicate the involvement of host-derived TGF-β and IGF-1

signaling in larval reactivation and transmission, binding kinetics studies between the

parasite encoded TGF- β receptors with mammalian TGF-β should be performed as have

been done with C. elegans and the parasitic fluke, Schistosoma (Ren et al., 1996; Davies and

Pearce, 1999). A. caninum TGF- β like receptor sequences as well as two TGF-β like

ligands have been cloned in our laboratory (Freitas and Arasu, 2005; data not shown).

It is well known that anthelminthic drugs are inefficacious in eradicating tissue-arrested

larval stages (Burke and Roberson, 1983; Arasu, 1998). Therefore, elucidating the host

responses during different phases of pregnancy and lactation during a concurrent A. caninum

infection may help in a better understanding of the phenomenon of tissue-arrest, reactivation

and subsequent transmammary transmission of the parasite. Effective control strategies can

be aimed at triggering the reactivation of arrested larvae followed by anthelminthic therapy.

55

Table 3.1 Primer sequences used for cytokine and reference gene transcript quantification by real-time RT-PCR

Primer pairs Gene Accession # Forward primer Reverse primer Product size 1 GAPDH BC098095 ccactacatggtctacatggttc ctcgctcctggaagatg 122 2 60S ribosomal protein BC011291 gattcgggatatgctgttgg aaagcctggaagaaggaggt 132 3 TGF-β1 NM_011577 tgcttcagctccacagagaa tggttgtagagggcaaggac 182 4 TGF-β2 NM_009367 aatgacaacgatgacgacca gatgcagactaacgccttcc 198 5 IGF-1 AF440690 tggatgctcttcagttcgtg tctgagtcttgggcatgtca 225

56

Mouse Groups evaluated at: Time Points Groups

I = day 19 gestation/14d pi UN = Uninfected / not bred II = day 1 lactation/15-16d pi IN = Infected / not bred

III = day 10 Lactation/25-26d pi UB = Uninfected / bred IB = Infected / bred

0 5 19-21 1 10 Days

I II III

Breed Mice

Late Gestation

Infect Mice

Early Lactation

Mid Lactation

Timepoints

Figure 3.1 Experimental design. Nine to ten week old BALB/c female mice were mated and injected with 1000 A. caninum L3 larvae on day 5 post-impregnation. Mice were sacrificed at three different timepoints during pregnancy and lactation and divided into four groups based on infection and breeding. Serum, skeletal muscle, mammary gland and spleen were collected from each timepoint in each group for ELISA and real-time PCR analyses.

Pregnancy Lactation

57

Larval counts

050

100150200250300350400450

IN IB IN IB IN IB

I II III

Num

ber

of la

rvae

Figure 3.2 Comparison of total larval burden in unbred versus bred BALB/c mice infected at times corresponding to day 19 gestation, day 1 and day 10 of postpartum lactation. Mice were infected with 1000 A. caninum L3 larvae subcutaneously. n = 5 mice per group. * indicates significant difference at p< 0.005 between unbred and bred groups.

*

58

Figure 3.3 Comparison of serum TGF-β1 levels in UN, IN, UB and IB mouse. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). a. TGF-β1 levels at Timepoint I * indicates significant difference at p< 0.0001 between uninfected unbred and infected bred groups. b. TGF-β1 levels at Timepoint II * indicates significant difference at p< 0.0001 between uninfected unbred and uninfected bred groups. ** indicates significant difference at p< 0.05 between uninfected unbred and infected bred groups. c. TGF-β1 levels at Timepoint III * indicates significant difference at p< 0.0001 between uninfected unbred and infected unbred, uninfected bred and infected bred groups. d. Comparison of TGF-β1 levels in UN, IN, UB and IB groups over time

0

200

400

600

800

1000

1200

I II III

TGF-

bet

a1 p

g/m

l

UN

IN

UB

IB

Figure 3.3a Figure 3.3b

Figure 3.3c Figure 3.3d

0

200

400

600

800

1000

1200

I II III

TGF-

bet

a1 p

g/m

l

UN

IN

UB

IB



*

***

*

* *

59

Figure 3.4 Comparison of serum TGF-β2 levels in UN, IN, UB and IB mouse. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). a. TGF-β2 levels at Timepoint I b. TGF-β2 levels at Timepoint II * indicates significant difference at p< 0.0001 between uninfected unbred and uninfected bred and between uninfected unbred and infected bred groups. c. TGF-β2 levels at Timepoint III * indicates significant difference at p< 0.0001 between uninfected unbred and infected unbred, uninfected bred and infected bred groups. d. Comparison of TGF-β2 levels in UN, IN, UB and IB groups over time

0

200

400

600

800

1000

1200

I II III

TG

F-b

eta2

pg

/ml

UN

IN

UB

IB



0

200

400

600

800

1000

1200

I II III

TG

F-b

eta2

pg

/ml

UN

IN

UB

IB



* *

* * *

60

Figure 3.5 Comparison of serum IGF-1 levels in UN, IN, UB and IB mouse. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). a. IGF-1 levels at Timepoint I * indicates significant difference at p< 0.0001 between uninfected unbred and infected unbred and uninfected bred groups. b. IGF-1 levels at Timepoint II c. IGF-1 levels at Timepoint III * indicates significant difference at p< 0.0001 between uninfected unbred and infected unbred, uninfected bred and infected bred groups. d. Comparison of IGF-1 levels in UN, IN, UB and IB groups over time

0

200

400

600

800

1000

1200

I II III

IGF

-1 p

g/m

l UN

IN

UB

IB



*

*

**

*

61

-35

-30

-25

-20

-15

-10

-5

0

5

I II III

Fol

d c

hang

e- T

GF-

ββ ββ1

mR

NA

-35

-30

-25

-20

-15

-10

-5

0

5

I II III

Fol

d ch

ange

- TG

F-ββ ββ1

mR

NA

-35

-30

-25

-20

-15

-10

-5

0

5

I II III

Fold

cha

nge-

TG

F- ββ ββ

1 m

RN

A

Figure 3.6a, b, c TGF-β1 transcript levels in skeletal muscle of IN (a), UB (b) and IB (c) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.6a No significant differences were found between UN and IN at given timepoints.

Figure 3.6b * indicates significant difference at p< 0.05 between UN and UB at timepoint III.

Figure 3.6c No significant differences were found between UN and IB at given timepoints.

*

62

-5000

-2500

-50-40-30-20-10

010

I II III

Fold

cha

nge

- T

GF-

ββ ββ1

mR

NA

-20

-10

0

10

20

I II III

Fold

cha

ng

e- T

GF

- ββ ββ1

mR

NA

-20

-10

0

10

20

I II III

Fol

d c

han

ge-

TG

F- ββ ββ

1 m

RN

A

Figure 3.6d, e, f TGF-β1 transcript levels in mammary gland of IN (d), UB (e) and IB (f) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.6e * indicates significant difference at p< 0.01 between UN and UB at timepoint II.

Figure 3.6d * indicates significant difference at p< 0.01 between UN and IN at timepoint I, II and III.

Figure 3.6f * indicates significant difference at p< 0.01 between UN and IB at timepoint II.

*

*

*

*

*

63

-5500-4500-3500

-50

-25

0

25

50

I II III

Fold

cha

nge

- TG

F-ββ ββ

1 m

RN

A

-200-175-150-125-100-75-50-25

0255075

100

I II III

Fold

ch

ange

- TG

F-ββ ββ1

mR

NA

-200-175-150-125-100-75-50-25

0255075

100

I II III

Fo

ld c

han

ge-

TG

F- ββ ββ

1 m

RN

A

Figure 3.6g, h, i TGF-β1 transcript levels in spleen of IN (g), UB (h) and IB (i) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.6g * indicates significant difference at p< 0.05 between UN and IN at timepoint II.

Figure 3.6h No significant differences were found between UN and UB at given timepoints

Figure 3.6i No significant differences were found between UN and IB at given timepoints

*

64

-10

-5

0

5

10

I II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-10

-5

0

5

10

I II III

Fold

cha

nge

- T

GF-

ββ ββ2

mR

NA

-10

-5

0

5

10

I II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

Figure 3.7a, b, c TGF-β2 transcript levels in skeletal muscle of IN (a), UB (b) and IB (c) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.7a No significant differences were found between UN and IN at given timepoints

Figure 3.7b * indicates significant difference at p< 0.005 between UN and UB at timepoint III.


*

65

-2000

-1000

-10

0

10

I II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-10

-5

0

5

10

I II III

Fold

cha

nge

- T

GF-

ββ ββ2

mR

NA

-10

-5

0

5

10

I II III

Fol

d ch

ange

- T

GF-

ββ ββ2

mR

NA

Figure 3.7d, e, f TGF-β2 transcript levels in mammary gland of IN (d), UB (e) and IB (f) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.7d * indicates significant difference at p< 0.01 between UN and IN at timepoints I, II and III.

Figure 3.7e * indicates significant difference at p< 0.001 between UN and UB at timepoint II.

Figure 3.7f No significant differences were found between UN and IB at given timepoints

*

*

*

*

66

-50

-25

0

25

50

75

I II III

Fo

ld c

han

ge-

TG

F- ββ ββ

2 m

RN

A

-50

-25

0

25

50

75

I II III

Fol

d ch

ange

- T

GF-

ββ ββ2

mR

NA

0

100

200

300

400

I II III

Fol

d c

han

ge-

TG

F-ββ ββ2

mR

NA

Figure 3.7g, h, i TGF-β2 transcript levels in spleen of IN (g), UB (h) and IB (i) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.7g * indicates significant difference at p< 0.005 between UN and IN at timepoint I.

Figure 3.7h * indicates significant difference at p< 0.05 between UN and UB at timepoint I.

Figure 3.7i No significant differences were found between UN and IB at given timepoints.

*

*

67

-20

-10

0

10

I II III

Fold

cha

nge-

IGF-

1 m

RN

A

-20

-10

0

10

I II III

Fo

ld c

han

ge-

IG

F-1

mR

NA

-20

-10

0

10

I II III

Fo

ld c

han

ge-

IG

F-1

mR

NA

Figure 3.8a, b, c IGF-1 transcript levels in skeletal muscle of IN (a), UB (b) and IB (c) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.8a No significant differences were found between UN and IN at given timepoints.

Figure 3.8b No significant differences were found between UN and UB at given timepoints.


68

-125

-100

-75

-50

-25

0

25

I II III

Fold

cha

nge-

IGF-

1 m

RN

A

-125

-100

-75

-50

-25

0

25

I II III

Fo

ld c

han

ge-

IGF

-1 m

RN

A

-125

-100

-75

-50

-25

0

25

I II III

Fol

d ch

ange

- IG

F-1

mR

NA

Figure 3.8d, e, f IGF-1 transcript levels in mammary gland of IN (d), UB (e) and IB (f) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.8d * indicates significant difference at p< 0.01 between UN and IN at timepoints I and II.

Figure 3.8e No significant differences were found between UN and UB at given timepoints.

Figure 3.8f * indicates significant difference at p< 0.01 between UN and IB at timepoint II.

*

*

*

69

-700-600-500-400-300-200-100

0100200

I II III

Fold

ch

ange

- IG

F-1

mR

NA

-700-600-500-400-300-200-100

0100200

I II III

Fold

ch

ang

e- IG

F-1

mR

NA

-700-600-500-400-300-200-100

0100200

I II III

Fold

ch

ange

- IG

F-1

mR

NA

Figure 3.8g, h, i IGF-1 transcript levels in spleen of IN (g), UB (h) and IB (i) mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts in IN, UB and IB groups, comparisons were made relative to UN controls.

Figure 3.8g No significant differences were found between UN and IN at given timepoints.

Figure 3.8h No significant differences were found between UN and UB at given timepoints.

Figure 3.8i No significant differences were found between UN and IB at given timepoints.

70

-40-35-30-25-20-15-10-505

II III

Fold

cha

nge-

TG

Fββ ββ1

mR

NA

-40-35-30-25-20-15-10-505

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

-40-35-30-25-20-15-10-505

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

Figure 3.9a, b, c Comparison of TGF-β1 transcript levels in skeletal muscle of IN (a), UB (b) and IB (c) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.9a * indicates significant difference at p< 0.001 between timepoint I and timepoint III.

Figure 3.9b * indicates significant difference at p< 0.01 between timepoint I and timepoint III.

Figure 3.9c * indicates significant difference at p< 0.05 between timepoint I and timepoint III.

*

*

*

71

0

10

20

30

40

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

0

10

20

30

40

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

-10

0

10

20

30

40

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

Figure 3.9d, e, f Comparison of TGF-β1 transcript levels in mammary gland of IN (d), UB (e) and IB (f) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.9d * indicates significant difference at p< 0.05 between timepoint I and timepoint III.

Figure 3.9e * indicates significant difference at p< 0.001 between timepoint I and timepoint II.

Figure 3.9f * indicates significant difference at p< 0.01 between timepoint I and timepoint II.

*

*

*

72

-5500-4500-3500

-50

-25

0

25

50

II III

Fol

d c

ha

nge

- TG

F-ββ ββ

1 m

RN

A

-10

0

10

20

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

-10

0

10

20

II III

Fold

cha

nge-

TG

F-ββ ββ

1 m

RN

A

Figure 3.9g, h, i Comparison of TGF-β1 transcript levels in spleen of IN (g), UB (h) and IB (i) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.9g * indicates significant difference at p< 0.05 between timepoint I and timepoint II.

Figure 3.9h No significant differences were found between timepoints.

Figure 3.9i No significant differences were found between timepoints

*

73

-70

-60

-50

-40

-30

-20

-10

0

II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-70

-60

-50

-40

-30

-20

-10

0

II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-70

-60

-50

-40

-30

-20

-10

0

II III

Fold

cha

nge-

TG

Fββ ββ

2 m

RN

A

Figure 3.10a, b, c Comparison of TGF-β2 transcript levels in skeletal muscle of IN (a), UB (b) and IB (c) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.10c * indicates significant difference at p< 0.001 between timepoint I and timepoint III.

Figure 3.10b * indicates significant difference at p< 0.001 between timepoint I and timepoint III.

Figure 3.10a * indicates significant difference at p< 0.001 between timepoint I and timepoint III.

*

*

*

74

-75

-50

-25

0

25

50

75

100

125

II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-75

-50

-25

050

75

100

125

II III

Fol

d ch

ang

e- T

GF

-ββ ββ2

mR

NA

-75

-50

-25

0

II III

Fol

d ch

ange

- TG

F-ββ ββ

2 m

RN

A

Figure 3.10d, e, f Comparison of TGF-β2 transcript levels in mammary gland of IN (d), UB (e) and IB (f) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.10e * indicates significant difference at p< 0.05 between timepoint I and timepoint II.

Figure 3.10d * indicates significant difference at p< 0.001 between timepoint I and timepoint III.

Figure 3.10f No significant differences were found between timepoints

*

*

75

-10

0

10

20

30

40

50

60

70

II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-10

0

10

20

30

40

50

60

70

II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

-10

0

10

20

30

40

50

60

70

II III

Fold

cha

nge-

TG

F-ββ ββ

2 m

RN

A

Figure 3.10g, h, i Comparison of TGF-β2 transcript levels in spleen of IN (g), UB (h) and IB (i) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.10g No significant differences were found between timepoints

Figure 3.10h No significant differences were found between timepoints

Figure 3.10i No significant differences were found between timepoints

76

-25

-20

-15

-10

-5

0

5

10

15

II III

Fold

cha

nge-

IGF-

1 m

RN

A

-15

-10

-5

0

5

10

15

20

25

II III

Fold

cha

nge-

IGF-

1 m

RN

A

-15

-10

-5

0

5

10

15

20

25

II III

Fold

cha

nge-

IGF-

1 m

RN

A

Figure 3.11a, b, c Comparison of IGF-1 transcript levels in skeletal muscle of IN (a), UB (b) and IB (c) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.11c * indicates significant difference at p< 0.05 between timepoint I and timepoint II.

Figure 3.11b No significant differences were found between timepoints.

Figure 3.11a * indicates significant difference at p< 0.05 between timepoint I and timepoint III. *

*

77

-125

-100

-75

-50

-25

0

25

II III

Fold

cha

nge-

IGF-

1 m

RN

A

-50

-25

0

25

50

75

100

125

150

II III

Fold

cha

nge-

IGF-

1 m

RN

A

-50

-25

0

25

50

75

100

125

150

II III

Fold

cha

nge-

IGF-

1 m

RN

A

Figure 3.11d, e, f Comparison of IGF-1 transcript levels in mammary gland of IN (d), UB (e) and IB (f) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.11e No significant differences were found between timepoints.

Figure 3.11d * indicates significant difference at p< 0.01 between timepoint I and timepoint II ** indicates significant difference at p< 0.001 between timepoint I and timepoint III.

Figure 3.11f * indicates significant difference at p< 0.01 between timepoint I and timepoint II, and timepoint I and timepoint III.

*

**

*

*

78

-45-40-35-30-25-20-15-10-505

II III

Fold

cha

nge-

IGF-

1 m

RN

A

-45-40-35-30-25-20-15-10-505

II III

Fold

cha

nge-

IGF-

1 m

RN

A

05

1015202530354045

II III

Fold

cha

nge-

IG

F-1

mR

NA

Figure 3.11g, h, i Comparison of IGF-1 transcript levels in spleen of IN (g), UB (h) and IB (i) mice over time. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. For evaluating fold upregulation or downregulation of transcripts between timepoints, timepoint II and timepoint III were compared to timepoint I using equation III.

Figure 3.11h No significant differences were found between timepoints.

Figure 3.11g No significant differences were found between timepoints.

Figure 3.11i No significant differences were found between timepoints.

79

4. Development of an in vitro co-culture system to study the effects of TGF-β and

pregnancy hormones on reactivation of hookworm larvae

4.1 Introduction

After infecting a host parasitic nematodes may display developmental arrest where

larvae have the option to resume further development upon receiving appropriate stimuli

from the host. In hookworms, reactivation of the arrested larval population has a tremendous

biological and economic significance as it results in active intestinal infection as well as

transmammary transmission of the parasite (Schad, 1990). Molecular signals mediating such

resumption of development have been well-studied in developmentally arrested dauer stage

of the free-living soil nematode Caenorhabditis elegans (Riddle and Alberts, 1998). Genetic

and molecular analyses in C. elegans have shown that transforming growth factor (TGF)-β

and insulin-like growth factor (IGF)-1 signaling pathways mediate neuroendocrine responses

leading into dauer formation as well as exit from the dauer stage (Georgi, et al., 1990;

Estevez, et al., 1993; Ren, et al., 1996; Kimura, et al., 1997). Parallels have been drawn from

these signaling pathways in studying the mechanisms of larval arrest and reactivation in

hookworms (Hawdon and Schad, 1993). It has also been speculated that hormonal changes

occurring during pregnancy potentially provide a stimulus for larval reactivation. Exogenous

administration of estrogen and prolactin in A. caninum infected dogs resulted in resurgence

of larvae in milk (Stoye and Krause, 1976). Since estrogen and prolactin have been shown to

upregulate TGF-β2 isoform in uterus and mammary gland (Schneider et al., 1996), it is likely

that a combination of hormones and cytokines in a timely manner causes the reactivation of

the latent larval reservoir for re-infection and transmission.

80

Cell culture systems have been previously used in studies on helminth biology

mainly, to culture the infective-stage L3 larvae to promote molting into the fourth stage in

Brugia (Smith, et al., 2000), develop a cell line derived from viable L3 to evaluate antigen-

specific humoral immune response in Haemonchus (Coyne and Brake, 2001), and nematode

establishment, molting and reproduction in intestinal epithelial cells by Trichinella

(ManWarren, et al., 1997; Gagliardo, et al., 2002). One of the primary goals of in vitro

culture systems is to mimic the in vivo environment, thus providing the parasite with host-

like conditions.

In vivo, the reactivation of arrested larvae leads to resumption of development with

growth and morphological changes which require energy and therefore, intake of food. In C.

elegans, dauer larvae start to feed within 2-3 hours of exposure to bacterial food and exit

from the arrested stage (Ren, et al., 1996). In hookworms, a simple in vitro assay was

developed in which larval reactivation results in activation of the feeding response and the

ingestion of flourescein-labeled albumin leading to fluorescent intestinal tracts in the larvae

that can be easily scored under a fluorescent microscope (Hawdon, et al., 1993; Kumar and

Pritchard, 1994). Besides hookworms, the in vitro feeding assay has been used as a direct

method for measuring the reactivation of larvae in other parasitic nematodes also as it is

speculated that resumption of feeding is associated with resumption of development

(Hawdon, et al., 1992, 1993; Gamble and Mansfield, 1996). It was used in the present study

to evaluate the candidate cytokines and hormones that may be responsible for triggering

reactivation of A. caninum larvae. It appears that the amphids which contain axon endings of

chemosensory neurons at the anterior end of the larvae are exposed to the environment and

are therefore, critical for signaling during entry into and exit from the arrested stage

81

(Bargmann and Horvitz, 1991; Ashton, et al., 1998, 1999). Using immunofluorescence

studies, it was recently shown that anti-TGH-2 (A. caninum TGF-β isoform) antibodies

identified native protein in the anterior region of the larvae resembling areas corresponding

to the chemosensory region of the amphidial pores in C. elegans (Freitas and Arasu, 2005).

With the objective to compliment our in vivo studies on the influence of pregnancy

hormones and cytokines associated with reactivation of tissue-arrested hookworm larvae, an

in vitro cell culture system was developed. We focused on establishing primary skeletal and

mammary epithelial cell cultures as upon reactivation larvae migrate from skeletal muscle,

which is the favored site of arrest, to mammary gland leading to transmammary transmission

to the newborn. We successfully established both types of primary cell cultures in our

laboratory and co-cultured these cells with infective A. caninum L3 larvae. Previous in vitro

feeding/reactivation assays had shown that A. caninum infective L3 were not susceptible to

the effects of TGF-β while ‘tissue arrested’ L3s recovered from the carcasses of infected

mice were stimulated to resume feeding (albeit with no evidence of morphological change or

development to the 4th larval stage; Arasu, 2001). This experimental setup was designed to

pre-condition or transform infective L3 larvae by the mammalian cell co-culture to mimic

‘tissue-arrested’ larvae and to then test the hypothesis that mammalian host-derived TGF-β

and pregnancy hormones result in reactivation and resumption of development.

4.2 Materials and Methods 4.2.1 Primary skeletal muscle cell culture

For starting a primary culture of skeletal muscle, 4-5 of <1 day-old neonatal Cr:NIH (S)

mouse pups (National Cancer Institute, Frederick, MD) were killed. After skinning,

82

hindlimbs and forelimbs were collected in Hank`s Balanced Salt Solution (HBSS)

(Mediatech, Herndon, VA) and minced finely with scissors. The minced muscle mass was

subjected to digestion using an enzymatic solution of 0.17% Trypsin (T-8128, Sigma-

Aldrich, St. Louis, Missouri) and 0.085% Collagenase (C-6885, Sigma-Aldrich, St. Louis,

Missouri) in a beaker incubated at 37ºC for 25 minutes to obtain a free cell suspension . After

incubation the contents in the beaker were transferred into a 15 ml sterile disposable tube and

centrifuged at 2500 rpm for 5 minutes. Supernatant was removed by a Pasteur pipette and the

digested muscle mass along with cells were resuspended in 5 ml of complete Dulbecco`s

Modified Eagles Medium (cDMEM) containing 5% Fetal Bovine Serum (FBS) (Mediatech,

Herndon, VA) and 1% antibiotic/antimycotic solution (Mediatech, Herndon, VA) to

neutralize Trypsin. The muscle mass was washed 3 times as mentioned above and then

filtered through a NytexTM (TETKO, Kansas City, MO) membrane filter into a fresh tube.

The cell suspension was diluted to 5 ml of cDMEM and cell concentration was determined

with a hemocytometer. Cells were plated at a density of 100,000 cells per well in a 24-well

cell culture plate (Costar). The monolayer typically became 75-80% confluent after 72 hours.

4.2.2 Primary mammary epithelial cell culture

Timed pregnant full-term Cr:NIH (S) female mice (National Cancer Institute, Frederick,

MD) were killed by cervical dislocation. Mammary glands were excised, collected in HBSS

and minced into a fine paste with scissors and scalpel. The minced mammary gland tissue

was subjected to digestion using an enzymatic solution of 0.35% Collagenase Type 3 (C-

6885, Sigma-Aldrich, St. Louis, Missouri) and 0.65% Hyaluronidase (H-3506, Sigma-

Aldrich, St. Louis, Missouri) in a beaker shaking at 200 rpm at 37ºC for 2 hours for a free

83

cell suspension. After incubation the mammary gland was mixed well with the help of a 10

ml pipette and filtered through a Nytex membrane into a 15 ml sterile disposable tube and

centrifuged at 1500 rpm for 5 minutes. The supernatant containing milk, fat globules and

tissue debris was removed carefully without disturbing the cell pellet. The cell pellet was

resuspended in cDMEM and washed three times at 2500 rpm for 5 minutes each at room

temperature. After the final washing the cell pellet was resuspended in cDMEM, filtered

again through the Nytex membrane and collected in 5 ml cDMEM for cell enumeration using

a hemocytometer. Cell were plated at a density of 50,000 cells per well in a 24-well Collagen

I -coated cell-culture plate (Greiner Bio-One, Kremsmünster, Austria). The monolayer

typically became 75-80% confluent around 72 hours.

4.2.3 Immunohistochemistry

Immunohistochemistry was performed on primary skeletal muscle cells and

mammary epithelial cells to confirm the presence/prevalence of the respective cells of

interest. For preparing a cytospin, cells were detached using 50 µl trypsin/EDTA at 37ºC for

4 minutes and suspended in trypsin inhibitor (DMEM with 10% FBS). Cells were dissociated

by pipetting up and down ten times. After adding cytospin fluid, the cell suspension was

attached to the glass slides by centrifuging at 1000 rpm for 1 minute in Shandon cytospin-3

(Labtics, Helsinborg, Sweden). Cytospins were fixed using 2% ice-cold paraformaldehyde

for 10 minutes and rinsed well with PBS. Histostain®-SP kit (Zymed, San Francisco, CA)

was used for immuno-histological staining. Peroxidase quenching solution consisting of 30%

hydrogen peroxide and 9 parts of absolute methanol was used to flood the whole slide for 10

minutes. After rinsing the slide with PBS, one drop of serum blocking solution was applied

84

per cytospin for 10 minutes. Anti-rabbit c-met (Receptor tyrosine kinase) antibody (Zymed,

San Francisco, CA) was used at a 1:100 dilution for staining skeletal muscle cells overnight

at 4ºC. Anti-rabbit cytokeratin antibody (Novocastra, NewCastele, UK) was used at 1:50

dilution for staining mammary epithelial cells for 1 hour and incubated in a moist chamber.

After rinsing the slides with PBS, 100 µl of biotinylated secondary antibody was added to

each slide and incubated for 10 minutes. After the incubation step, the rinsing step was

repeated to remove any remaining biotinylated secondary antibody. Two drops or 100 µl of

enzyme-conjugate solution was added to each slide for 10 minutes and rinsed off with PBS.

Two drops or 100 µl of substrate-chromogen mixture was added to each slide and incubated

for 15 minutes and rinsed well with distilled water. The slide were counterstained with two

drops or 100 µl of hematoxylin for 3 minutes and washed under tap water. Two drops of

mounting solution was put on the slides before applying the coverslip over the cell pellet.

The slide was observed under microscope for the development of a red color indicating

positive staining. For negative controls, primary antibody was not added and C2C12 cells

served as a positive control for skeletal muscle cells whereas HC11 cells served as a positive

control for mammary epithelial cells.

4.2.4 L3 larvae

Infective A. caninum L3 were harvested from charcoal co-cultures of feces from male

laboratory beagle dogs (Marshal Farms, North Rose, NY) orally infected at 8-10 weeks of

age with 150 L3 as previously described (Burke and Roberson, 1979a; Arasu, 1998). The use

of dogs in this study was approved by the Institutional Animal Care and Use Committee at

North Carolina State University. Feces containing A. caninum eggs (range of 500-3000

85

eggs/g depending on stage of infection) were mixed with activated charcoal and deionized

water in Pyrex® baking trays and placed in a humidified incubator, in the dark, at 25ºC for 7-

10 days. The L3 were harvested utilizing a modified Baermann assembly (Garcia and Ash,

1979), repeatedly washed in phosphate-buffered saline (PBS) medicated with 20 mg/L

gentamicin (Sigma Chemical Co., Missouri) and 20mg/L lincomycin (Sigma Chemical Co.,

Missouri) and maintained at room temperature for 5-7 days prior to use.

To prepare the larvae for co-culture, infective L3 were exsheathed using 0.1% Clorox

bleach (Oakland, CA) to 5000 L3 larvae suspended in 10 ml of PBS in a 50 ml

polypropylene tube and kept on a shaker (Fisher Scientific, Pittsburgh, PA) at 150 pm for 5

minutes. After filling the tube with excess PBS up to 45 ml, the larvae were washed at 1000

rpm for 3 minutes by centrifugation. The washing step was repeated 4-5 times prior to use.

4.2.5 Primary cells and Larval Co-culture

The monolayer of cells was washed 3 times with 750 µl of cDMEM to remove media

and residual serum as serum can activate a feeding response in the L3 larvae. DMEM-Serum

Replacement II (SRII) (Sigma Aldrich, St. Louis, Missouri) was prepared the same way as

cDMEM but by replacing serum with 1% SRII. After adding 1000 µl of DMEM-SRII to each

well, Transwell© polycarbonate membrane with 3.0 µm pore size (Costar, Corning Inc.,

Corning, NY) were placed into each well. The Transwell© membranes split the cell culture

wells into a lower well and an upper well. Negative controls were setup in triplicate by

adding 5% SRII to the final volume of 1.25 ml. Similarly, positive controls were setup in

86

triplicate by adding 5% FBS to the same final volume. To each well, 1000 exsheathed larvae

were added for setting up the co-culture.

4.2.6 In vitro Feeding Assay

Briefly, 250 exsheathed L3 larvae harvested from the co-culture system were washed

at 1000 rpm for 15 seconds three times with PBS prior to use. Larvae were placed in 96-well

microtiter plate and each treatment or control was assayed in triplicate. To each well 100 µl

of 5 mg/ml flourescein-isothiocyanate-conjugated bovine serum albumin (FITC-BSA;

Sigma-Aldrich, St. Louis, Missouri) was added and the plate was incubated at 37ºC, 5% CO2

for 1.5-2 hours (Hawdon and Schad, 1990). The larvae were washed three times with PBS to

remove residual FITC prior to examination by UV microscopy at 400X magnification. At

least 50-100 larvae were enumerated for fluorescent intestinal tracts as positive (reactivated)

or negative (arrested). Percentage feeding was determined by 100 X (number positive)/

(number positive + number negative).

For each experiment, the positive control was normal dog serum and the negative

control was cDMEM. Test reagents included TGF-β1 and TGF-β2 (recombinant human;

R&D Systems, Inc., Minneapolis, Minnesota), Estrogen (Sigma-Aldrich, St. Louis, Missouri)

and Prolactin (mouse; National Hormone and Pituitary Program, Rockville, Maryland). For

functional activation, TGF-β1 and TGF-β2 were acidified in 4 mM HCl, 0.1% BSA at a

stock concentration of 1 µg/ml prior to use, control wells were similarly treated with

HCl/BSA. Estrogen was dissolved in sterile distilled water at stock concentration of 25

mg/ml and prolactin was dissolved in PBS at a stock concentration of 1 µg/100 µl. All

experiments were repeated two or more times.

87

4.3 Results

4.3.1 Staining of primary cell cultures

Primary skeletal muscle cells and primary mammary epithelial cells showed positive

staining with anti-c-met antibody and anti-cytokeratin antibody respectively, using the

Histostain kit (Zymed laboratories, San Francisco, CA). After evaluating the stained cells

under the microscope it was found that more than 80% of the cells in the primary skeletal

muscle cell culture were of skeletal muscle origin (Figure 4.1a, 4.1b, 4.1c). The other major

cell type was fibroblast which did not stain with anti-c-met antibody. Based on staining with

anti-cytokeratin, more than 90% of the cells in the primary culture were mammary epithelial

cells (Figure 4.1d, 4.1e, 4.1f). Negative controls stained blue indicating absence of antigen

recognition by the primary antibody. Positive controls (C2C12, skeletal muscle cells and

HC11, mammary epithelial cells) for respective cultures stained and were used for

comparison when identifying positive staining in test cells.

4.3.2 Effect of serum on larval feeding

Preparasitic A. caninum L3 larvae were co-incubated in the presence or absence of

primary skeletal muscle cells to test the effect of DMEM media supplemented with 5% SRII

instead of serum. No differences were observed in the basal levels of reactivation between

the larvae that were co-incubated with cells or in the absence of cells (Figure 4.2a). Normal

dog serum had previously been shown to stimulate feeding behavior in infective A. caninum

L3 larvae (Schad, 1991; Arasu, 2001). Based on these findings, 5% normal dog serum was

used as positive control and absence of serum served as a negative control. As shown in Fig.

4.2a, L3s co-cultured with primary skeletal muscle cells showed a significant

88

reactivation/feeding response when exposed to serum (Figure 4.2b). Similar results were

observed with L3 obtained from co-cultures with mammary epithelial cells (Figure 4.2c).

4.3.3 Effect of TGF-β on larval feeding

TGF-β has been shown to be exhibit a dose-dependent stimulatory effect on the

tissue-arrested larvae that were harvested from the carcass of infected BALB/c mice but not

on infective, pre-parasitic L3 (Arasu, 2001). In these analyses, TGF-β had no stimulatory

effect on the A. caninum L3 larvae co-incubated with primary skeletal muscle cells for 24 h,

48 h and 72 h; the background levels of percent reactivated L3 larvae were similar to the

negative control (Figure 4.3a, 4.3b, 4.3c, respectively). Similar results were obtained with co-

incubation of A. caninum larvae with primary mammary epithelial cells (Figure 4.4a, 4.4b

and 4.4c). The experiment was repeated three times.

4.3.4 Effect of pregnancy hormones on larval feeding

In a previously published report, pregnancy-associated hormones, estrogen and

prolactin, had no effect on the tissue-arrested L3 or infective larvae at physiological (0.1-1

µg/ml) or at high concentrations of 1 mg/ml (Arasu, 2001). To determine if hormones might

have a modulatory effect on the mammalian cells with subsequent excretion of stimulatory

cytokines such as TGF-beta, estrogen (100 ng/ml) and/or prolactin (100 ng/ml) were added to

the co-cultures for 24 h, 48 h and 72 h (Figure 4.5a, 4.5b and 4.5c). There was no effect on

the percentage feeding response of L3 larvae co-cultured with primary skeletal muscle cells

as compared to that of serum alone (positive control). Similarly, no effect of hormones was

observed on the L3 larvae co-cultured with primary mammary epithelial cells (Figure 4.6a,

89

4.6b and 4.6c). Supernatants were not tested to evaluate if TGF-β was released by exposure

of the cells to the hormones.

4.4 Discussion

Mice have been used as normal paratenic hosts for in vivo studies on A. caninum

larval arrest and reactivation (Arasu and Kwak, 1999). Similar to the dog,

pregnancy/lactation in the mouse triggers the transmammary transmission of larvae to the

nursing neonates (Steffe and Stoye, 1984; Arasu and Kwak, 1999). It is known from studies

in C. elegans that TGF-β signaling is one of the pathways involved in arrest and reactivation

(Georgi, et al., 1990; Ren, et al., 1996). In mammals, TGF-β2 is apparently upregulated in

uterine and mammary tissue during pregnancy (Schneider, e al., 1996). It was therefore

hypothesized that A. caninum larvae have evolved to utilize signaling via host-derived TGF-β

resulting in reactivation, migration through somatic tissues and subsequent transmission to

the newborn. In a previously published report, physiological concentrations of TGF-β

stimulated a feeding response which was comparable to the stimulatory effect induced by

10% normal dog serum on tissue-arrested A. caninum L3 larvae (Arasu, 2001).

Pregnancy-associated hormones such as estrogen and prolactin have been implicated

in larval reactivation and transmission. When estrogen and prolactin were administered

exogenously to post-partum dogs with latent (arrested) infections of A. caninum, a resurgence

of larvae was seen in their milk if let-down was induced by administration of oxytocin (Stoye

and Krause, 1976). However, hormones did not have any direct stimulatory effect on tissue-

arrested larvae as mentioned above. There might be indirect effects of hormonal fluctuations

observed during pregnancy. TGF-β2 tissue mRNA and serum levels were elevated during

90

late pregnancy (Schneider, et al., 1996). It was further shown that TGF-β2 mRNA levels

were elevated in the same tissues (uterine and mammary gland) as a result of exogenous

estrogen treatments in ovariectomized rats (Das, et al., 1992; Cheng, et al., 1993; Schneider,

et al., 1996). Taken together, these findings suggest that host-derived cytokine and hormonal

cues may be involved in specific reactivation of tissue-arrested A. caninum larval stages

during pregnancy and lactation.

In an attempt to test our hypothesis that host factors like cytokines and hormones

stimulate reactivation of tissue-arrested A. caninum larvae, we developed a primary

cell/larvae co-culture system. Our major focus was to use the primary skeletal muscle and

mammary epithelial cell cultures for co-incubation with pre-parasitic A. caninum L3 to

mimic tissue-arrest outside the body of the host. As mentioned before, skeletal muscle is the

favored site of larval arrest and mammary gland is the site where the larvae get transmitted to

the neonates via suckling. Larvae were co-incubated with cells for 24 hours before adding the

cytokines (TGF-β1 and TGF-β2) and hormones (estrogen and prolactin) of interest. We did

not observe any significant feeding behavior in the co-cultured L3 larvae that were treated

with TGF-β isoforms and hormones. One of the reasons for such observations could be the

inability of the co-culture setup to induce tissue-arrest as it occurs inside a host.

In order to closely match the in vivo conditions, a variety of host factors and

molecules have been added to parasitic cultures in different studies. Certain factors like

ascorbic acid are critical in the morphogenesis and development of parasitic helminths like

Brugia (Rajan, et al., 2003). Similarly, a simple reproducible method has been developed for

protozoa i.e. short-term ex-vivo Plasmodium vivax culture in which glucose, ascorbic acid,

thiamine, hypoxanthine, and 50% human AB+ serum are added to the standard P. falciparum

91

in vitro culture medium (Chotivanich, et al., 2001). This simple method of culturing P. vivax

ex vivo is suitable for anti-malarial susceptibility and immunoparasitology studies. It is

therefore possible that besides the supplementation of media with serum replacement media

II (Sigma Aldrich), we would have to add other factors to simulate in vivo conditions closely.

Since host molecules important in A. caninum larval arrest are not known, it would laborious

and time-consuming to test candidate reagents. Transwell® filters keep the L3 larvae

physically separate from the cell monolayer. It is possible that the larvae have to be in close

physical contact with the cells to receive signals for undergoing arrest. However, it is very

difficult to do so in case of A. caninum L3 larvae as they are constantly moving thereby,

disrupting the cell monolayer within hours of physical contact.

We have proposed that larval reactivation can be assessed in independent ways

besides the in vitro co-culture setup. Firstly, PCR amplification can be used to evaluate the

upregulation of Ac-daf-7, which is the C. elegans daf-7 homologue recently cloned and

characterized in A. caninum, in reactivated versus arrested L3 larvae. However, recent studies

by Freitas and Arasu (2005) show that expression of daf-7 is maximal at the arrested

infective L3 as well as reactivated (serum-stimulated) L3 larval stages which differs from

that of Ce-daf-7 expression and may be unique to parasitic nematodes that have an obligate

requirement to undergo developmental arrest (Freitas and Arasu, 2005). Secondly,

reactivation status can be tested by using protein binding studies i.e. whether Ac-DAF-7 can

bind to and stimulate signaling from mammalian TGF-β receptors. Similar studies in

Schistosoma mansoni have shown that a chimeric receptor containing the extracellular

domain of type I TGF-β receptor, SmRK1, that is joined to the intracellular domain of the

human type I TGF-β receptor, could activate a luciferase reporter gene in response to TGF-β

92

(Beall and Pearce, 2001). The efforts to express Ac-DAF-7 are under way in our lab. Similar

studies of B. malayi have suggested an immunomodulatory role for recombinant Bm-TGH-2

which can bind to and stimulate signaling from mammalian TGF-β receptors (Gomez-

Escobar et al., 2000). Further studies are required to examine the host-parasite interaction in

A. caninum infection utilizing ligand and receptor interaction studies. Finally, larval

reactivation status can be checked by utilizing antibodies to Ancylostoma secreted protein

(ASP). It has been previously demonstrated that preparasitic L3 larvae stimulated by serum

start releasing excretory-secretory products such as ASPs which appear to facilitate in larval

penetration and migration (Hawdon, et al., 1996b). A better understanding of the mechanisms

associated with larval arrest and reactivation in A. caninum can contribute to development of

chemotherapeutic and vaccine-based strategies for elimination of latent infections.

93

A B

C

Figure 4.1 Immunohistochemistry. Cytospins of primary skeletal muscle were stained with c-met antibody overnight. A, C2C12 skeletal muscle cell line, positive control; B, no primary antibody, negative control; C, skeletal muscle cells stained positive with c-met antibody (20X).

94

D E

F

Figure 4.1 Immunohistochemistry. Cytospins of primary mammary epithelial cells were stained with anti-cytokeratin antibody overnight. A, HC11 mammary epithelial cell line, positive control; B, no primary antibody, negative control; C, mammary epithelial cells stained positive with anti-cytokeratin antibody (20X).

95

No cells Cells0

10

20

30

% F

eedi

ng L

3 la

rvae

Figure 4.2a Effect on percent feeding response of L3 larvae co-cultured in the presence or absence of primary skeletal muscle cells. L3 larvae were co-incubated in DMEM supplemented with 5% Serum Replacement media II in Transwells©.

96

Media ctrl 5% Serum0

10

20

30

40

50

60

% F

eedi

ng L

3 La

rvae

Figure 4.2b Effect of serum on percent feeding response of L3 larvae co-cultured with primary skeletal muscle cells. L3 larvae were co-incubated in DMEM supplemented with 5% Serum Replacement media II as negative control and larvae were stimulated with 5% normal dog serum as positive control. The asterisk indicates significant difference at p< 0.05.

*

97

Media ctrl 5% Serum0

10

20

30

40

50

60

% F

eedi

ng L

3 La

rvae

Figure 4.2c Effect of serum on percent feeding response of L3 larvae co-cultured with primary mammary epithelial cells. L3 larvae were co-incubated in DMEM supplemented with 5% Serum Replacement media II as negative control and larvae were stimulated with 5% normal dog serum as positive control. The asterisk indicates significant difference at p< 0.05.

*

98

Media ctrl 5% Serum TGF-ββββ20

25

50

75

% F

eedi

ng L

3 La

rvae

Figure 4.3a Effect of TGF-β2 on feeding response of L3 larvae. L3 larvae were co-cultured with primary skeletal muscle cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and TGF-β2 (10 ng/ml) for 24 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

99


10

20

30

40

50

60

70

80

% F

eedi

ng L

3 La

rvae

Figure 4.3b Effect of TGF-β2 on feeding response of L3 larvae. L3 larvae were co-cultured with primary skeletal muscle cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and TGF-β2 (10 ng/ml) for 48 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

100


102030405060708090

% F

eedi

ng L

3 La

rvae

Figure 4.3c Effect of TGF-β2 on feeding response of L3 larvae. L3 larvae were co-cultured with primary skeletal muscle cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and TGF-β2 (10 ng/ml) for 72 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

101


10

20

30

40

50

60

70

% F

eedi

ng L

3 La

rvae

Figure 4.4a Effect of TGF-β2 on feeding response of L3 larvae. L3 larvae were co-cultured with primary mammary epithelial cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and TGF-β2 (10 ng/ml) for 24 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

102


102030405060708090

% F

eedi

ng L

3 La

rvae

Figure 4.4b Effect of TGF-β2 on feeding response of L3 larvae. L3 larvae were co-cultured with primary mammary epithelial cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and TGF-β2 (10 ng/ml) for 48 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

103


10

20

30

40

50

60

70

80

% F

eedi

ng L

3 La

rvae

Figure 4.4c Effect of TGF-β2 on feeding response of L3 larvae. L3 larvae were co-cultured with primary mammary epithelial cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and TGF-β2 (10 ng/ml) for 72 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

104

Media

ctrl

5% S

erum

Estro

gen

Prolac

tin

Estro

gen/P

rolac

tin0

10

20

30

40

50

60

70

% F

eedi

ng L

3 la

rvae

Figure 4.5a Effect of pregnancy-associated hormones on feeding response of L3 larvae. L3 larvae were co-cultured with primary skeletal muscle cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and estrogen (100 ng/ml), prolactin (100 ng/ml) and estrogen/prolactin for 24 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

105

Media

ctrl

5% S

erum

Estro

gen

Prolac

tin

Estro

gen/P

rolac

tin0

10

20

30

40

50

60

70

% F

eedi

ng L

3 La

rvae

Figure 4.5b Effect of pregnancy-associated hormones on feeding response of L3 larvae. L3 larvae were co-cultured with primary skeletal muscle cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and estrogen (100 ng/ml), prolactin (100 ng/ml) and estrogen/prolactin for 48 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

106

Media

ctrl

5% S

erum

Estro

gen

Prolac

tin

Estro

gen/P

rolac

tin0

10

20

30

40

50

60

70

80

% F

eedi

ng L

3 La

rvae

Figure 4.5c Effect of pregnancy-associated hormones on feeding response of L3 larvae. L3 larvae were co-cultured with primary skeletal muscle cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and estrogen (100 ng/ml), prolactin (100 ng/ml) and estrogen/prolactin for 72 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

107

Media

ctrl

5% S

erum

Estro

gen

Prolac

tin

Estro

gen/P

rolac

tin0

10

20

30

40

50

60

70

80

% F

eedi

ng L

3 La

rvae

Figure 4.6a Effect of pregnancy-associated hormones on feeding response of L3 larvae. L3 larvae were co-cultured with primary mammary epithelial cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and estrogen (100 ng/ml), prolactin (100 ng/ml) and estrogen/prolactin for 24 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

108

Figure 4.6b Effect of pregnancy-associated hormones on feeding response of L3 larvae. L3 larvae were co-cultured with primary mammary epithelial cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and estrogen (100 ng/ml), prolactin (100 ng/ml) and estrogen/prolactin for 48 h in triplicate. The asterisk indicates significant difference at p< 0.05.

Media

ctrl

5% S

erum

Estro

gen

Prolac

tin

Estro

gen/P

rolac

tin

0

10

20

30

40

50

60

70

80

% F

eedi

ng L

3 La

rvae

*

109

Media

ctrl

5% S

erum

Estro

gen

Prolac

tin

Estro

gen/P

rolac

tin0

10

20

30

40

50

60

70

80

% F

eedi

ng L

3 La

rvae

Figure 4.6c Effect of pregnancy-associated hormones on feeding response of L3 larvae. L3 larvae were co-cultured with primary mammary epithelial cells in DMEM supplemented with 5% Serum Replacement media II as negative control, larvae were stimulated with 5% normal dog serum as positive control and estrogen (100 ng/ml), prolactin (100 ng/ml) and estrogen/prolactin for 72 h in triplicate. The asterisk indicates significant difference at p< 0.05.

*

110

5. IL-4 and IFN-γ serum protein and transcript levels during pregnancy and

Ancylostoma caninum infection in BALB/c mice

5.1 Introduction

More than one billion people in the tropics and sub-tropics are known to be infected

with hookworms (Chan, 1997). Thus, hookworm infection remains a major contributor to

iron-deficiency anemia as adult parasites feed on blood from the intestines. Since hookworms

are known to exhibit latency and reactivation, they interact with their hosts at an intimate

level and have been reported to 'counter' the complex immune responses generated against

them (Behnke, 1991; Loukas and Prociv, 2001). Immune responses to hookworms have been

studied in both humans and experimental animal hosts in attempts to understand host

protection as well as parasite survival particularly with regard to the development of

hookworm vaccines. Eosinophilia, mastocytosis and IgE production are the most prominent

immune alterations observed during a hookworm infection (Loukas and Prociv, 2001). In

many ways, hookworms are typical gastrointestinal nematodes based on types of immune

responses generated in definitive hosts. Humoral and cellular responses fit loosely within the

framework of a T helper 2 (Th2) type of response (Loukas, et al., 2005). Humans infected

with Necator hookworms produce high levels of parasite-specific and total IgE (Pritchard

and Walsh, 1995; Pritchard, et al., 1995) which is accompanied by peripheral and local

eosinophilia (White, et al., 1986; Maxwell, et al., 1987). While humoral responses to

hookworms have been well documented; little was known about the role of adaptive T cell

response until recently (Quinnell et al., 2004; Pit, et al., 2000 and 2001).

111

There is a notable lack of suitable animal models for human hookworm infection, and

extrapolating from immunological models of an abnormal host can be unreliable. N.

americanus matures in hamsters; however, there is wide variability in the number of L3 that

develop into intestinal adult worms (Rose and Behnke, 1990; Jian, et al., 2003). Similarly, A.

duodenale has been shown to develop in dogs, but only with exogenous administration of

steroids (Leiby, et al., 1987). Syrian Golden hamsters (Mesocricetus auratus) have also been

used to model infections with Ancylostoma ceylanicum (Garside, et al., 1990). In a recent

study, impact of A. ceylanicum hookworm infection on host cellular responses, cytokine

production and lymphoproliferation were measured (Mendez, et al., 2005). Initial larval

infection with 100 third-stage A. ceylanicum larvae resulted predominantly in Th1 responses

(upregulation of proinflammatory cytokines such as IFN-γ and TNF-α) which occurred

during larval migration and continued up to 14 days post-infection or prepatency.

Subsequently, development of larvae into egg-laying adult hookworms or patency coincided

with a switch to Th2 predominant responses with a marked increase in IL-4 and IL-10

production. This switch also concurred with reduced host lymphoproliferative responses to

hookworm antigens (Mendez, et al, 2005).

A. caninum infection in dogs serves as a good model for understanding the host-

parasite interactions occurring during human hookworm infection. However, canines are not

a realistic option due to high animal costs as well as ethical concerns with euthanasia for

these types of studies. Our lab has utilized the mouse as a circumscribed model for

specifically assessing pregnancy-associated transmammary transmission of A. caninum

infection in the neonate (Arasu and Heller, 1999; Arasu and Kwak, 1999). The murine model

was used to compare the responses of infected versus uninfected animals that were either

112

bred or not bred to ascertain whether suppression of A. caninum-specific antibody responses

during pregnancy facilitated the reactivation and transmammary transfer of hookworm

larvae. Initial comparisons of genetically divergent BALB/c versus C57BL/6 mice showed

that both the strains mounted strong Th2 biased IgG1 and IgE antibody responses to A.

caninum infection (Arasu and Heller, 1999). It was also confirmed that larval transfer to the

mouse pups occurred during the post-partum lactational period as there was only one

previously published report using mice to assess the transmammary transmission of A.

caninum larvae (Steffe and Stoye, 1984). In dams, levels of total and antigen-specific IgG1

and total IgE (Th2 response) were highly correlated with parasite burden (Arasu and Heller,

1999). During most phases of pregnancy and lactation, infected dams had lower total IgG1,

IgG2a and IgE levels as compared to unbred mice at comparable times post-infection which

supported the established dogma of a generalized immunosuppression associated with

pregnancy (Arasu and Heller, 1999). However, correlative studies showed that the parasite-

specific antibody responses did not play a major role in the pregnancy-associated

transmammary transmission of A. caninum larvae (Arasu and Heller, 1999). These studies do

not rule out the possibility of underlying fluctuations in the levels of Th1 and Th2 cytokines

associated with pregnancy and infection that may be involved in the process of larval

reactivation and transmission.

We report here the immunological profile of IL-4 and IFN-γ mRNA transcripts

during pregnancy and A. caninum infection in skeletal muscle (the favored site of latent

infection), mammary gland (where the larvae get transmitted) and spleen (major site to

monitor immune responses during parasitic infections). Interestingly, in our model of

breeding and infection we found that during early lactation IL-4 transcript levels predominate

113

over IFN-γ in skeletal muscle but were concurrently downregulated in mammary gland in

infected bred mice. We also found that both IL-4 and IFN-γ transcripts were upregulated at

14 days post-infection in mammary gland of infected unbred mice which support the findings

that acute helminthic infections are often associated with mixed Th1/Th2 responses. We were

unable to correlate mRNA transcript levels with serum levels as IL-4 was undetectable in the

sera of majority of the mice.

5.2 Materials and Methods

5.2.1 Mice breeding and infection

Nine to ten-week-old female BALB/c mice were purchased from Charles River

Laboratories, and maintained at the Laboratory Animal Resources facility in accordance with

Institutional Animal Care and Use Committee guidelines. Mice were mated at a ratio of 2

females: 1male for a period of 7 days and examined twice daily for the presence of a vaginal

impregnation plug. Observation of a plug was designated as day 0 of the pregnancy. On day

5 post-impregnation (observation of vaginal plug), the female mice were injected sub-

cutaneously in the dorsal cervical interscapular region with 100 ul PBS (control) or with

1000 larvae in PBS. Mice were killed by decapitation at three different time points: time

point I corresponding to day 19 of gestation or day 14 post-infection (pi), and time points II

and III corresponding to day 1 and 10 of postpartum lactation or day 15-16 pi and day 25-26

pi, respectively. The tissue samples collected were sections of the gastrocnemius skeletal

muscle, mammary gland, and spleen as well as blood. Approximately 200mg of each tissue

sample (except blood) was stored in RNA-STAT 60 (Tel-Test, Texas) at -80ºC for RNA

extraction. Serum was collected after incubating whole blood at 37ºC for 1 h and

114

subsequently at 4ºC for 4 h and subsequently stored at -20ºC until further use. The female

mice were grouped as infected or not infected and bred or not bred (n = 5 per group). The

resulting four groups were uninfected/ unbred (UN or normal controls), infected/ unbred

(IN), uninfected/ bred (UB) and infected/ bred (IB).

5.2.2 Harvesting tissue L3

For harvesting tissue-arrested larvae, carcasses were skinned, minced, wrapped in

eight layers of cheesecloth and incubated in a modified Baermann assembly for 4 h at 37ºC

in PBS medicated with 20 mg/L gentamicin (Sigma Chemical Co., Missouri) and 20mg/L

lincomycin (Sigma Chemical Co., Missouri). Migrating larvae that collected at the bottom of

the cups were repeatedly washed at 2000 rpm for 10 minutes with medicated PBS. Larval

numbers in dams versus each litter of pups were counted with the aid of a dissecting

microscope.

5.2.3 Mouse IL-4 ELISA

IL-4 levels were determined using a Mouse IL-4 ELISA kit (Pierce Endogen,

Rockford, IL) which is based on the principle of an indirect ELISA. The assays were

performed using anti-mouse IL-4 pre-coated 96-well strip plates. The standards (0, 15, 75

and 375 pg/ml) were prepared according to the manufacturer’s protocol. Using a

multichannel pipettor 50 µl of plate reagent was added to each well before adding 50 µl of

reconstituted standards and test sera samples from different groups of mice. The plate was

covered with an adhesive tape and incubated for 3 hours at 37ºC in a humidified incubator.

After washing five times with wash buffer, 100 µl prediluted conjugate reagent was added to

115

each well and the plate was incubated for 1 hour 37ºC in a humidified incubator. After

repeating the washing step five times, 100 µl of TMB substrate solution was added to each

well. Enzymatic color reaction was allowed to develop at room temperature in the dark for 30

minutes. The substrate reaction yielded a blue solution that turned yellow when Stop Solution

containing hydrochloric acid was added. Absorbance was measured on an ELISA plate

reader set at 450 nm and 550 nm. The standard curve was used to determine IL-4 in pg/ml in

the unknown samples.

5.2.4 RNA extraction and cDNA synthesis

Total RNA from all the tissues (skeletal muscle, mammary gland and spleen) was

extracted using RNA-STAT (Tel-Test, Inc., Friendswood, TX) according to Chomczynski

and Sacchi (1987). Residual genomic DNA was removed from RNA by treating with DNase

(20U per 100 µg of RNA). RNA quantity and quality was checked by spectrophotometric

measurements at 260 and 280 nm (Pharmacia) and by analyzing 1µg RNA for rRNA bands

and integrity on a 1% ethidium bromide agarose gel.

For reverse transcription, first strand cDNA was synthesized by adding 10µg of total

RNA to a 40µl reaction mix containing a final concentration of 2.5nM dNTPs, 0.1M

dithiothreitol, 1 µg Oligo d (T) primer (Promega, Madison, WI), 24U RNase inhibitor and

incubation with 200U Superscript II reverse transcriptase (Invitrogen, Carlsbad, CA) at 37°C

for 1 hour. To confirm absence of contaminating genomic DNA, RT negative reactions with

5µg total RNA were setup. RT-plus (cDNA)/minus reactions were subjected to RT-PCR

(reverse transcription-polymerase chain reaction) using GAPDH primers. PCR reactions

were run on a 2% agarose gels to check for product formation and DNA contamination. The

116

thermal cycler program was one cycle of 94ºC for 5 min, 40 cycles of 94ºC for 1 min, 56ºC

for 1 min and 72ºC for 1 min, followed by one cycle of 72ºC for 1 min.

5.2.5 Primer design

Primer sets for genes of interest, IL-4 and IFN-γ and reference gene 60S acidic

ribosomal protein were designed from mouse-specific Genbank sequences listed in Table 1

using Primer 3 software (Rozen and Skaletzky, 2000) (http://www-genome.-wi.mit.edu/cgi-

bin/primer/primer3_www.cgi). In order to minimize primer-dimer formation, the maximum

self-complementarity was 6 and the maximum 3’ self-complementarity was 0. The targets

amplified by primer pairs were characterized using the Mfold program (SantaLucia, 1998)

(http://bioinfo.math.rpi.edu/~mfold/dna/form1.cgi) for predicting the nature of any secondary

structures which may form at the site of primer binding. Primer pairs that bound at the site of

a predicted loop were discarded. Primer sets were synthesized by Integrated DNA

Technologies (Coralville, IA) and primers were reconstituted at100 pM/ul in nuclease-free

water prior to use.

5.2.6 Real-time PCR

The real-time PCR reactions were carried out using the iCyclerTM iQ PCR detection

system (Bio-Rad Laboratories, Hercules, CA, USA). In each 25µl reaction, 12.5µl of iQTM

SYBR green supermix (Bio-Rad) was added to 300nM of each primer along with 250ng of

cDNA. PCR amplification was performed in duplicate for each sample using the following

cycle conditions: 3 min at 95°C followed by 45 repeats of 1 min at 95°C, 30s at 55°C and 30s

at 72°C. Temperature optimization was carried out for all the primer sets to be amplified

117

simultaneously. Annealing temperatures were tested from the 50-65ºC range; all the primer

sets amplified optimally at 55ºC. A melt curve analysis step was included at the end of cycles

to check for primer-dimer and non-specific product formation. Efficiency of the PCR

reactions was derived by doing a standard curve of 10-fold serially diluted mouse spleen

cDNA and was consistently in the 95-98% range. Non-template controls were used to detect

any genomic DNA contamination and amplified products were also examined on a 2%

agarose gel to verify that the amplified products were of the expected sizes. Raw Ct values

were analyzed using Relative Expression Software Tool-384 (REST-384) to generate a fold

increase or decrease in the transcript levels (Pfaffl, et al., 2002). The 60S acidic ribosomal

protein was used as the endogenous reference gene for normalizing transcript levels among

tissues of interest. 60S ribosomal protein was previously shown to be the least variable from

comparative analyses of various genes including β-actin, cyclophilin and glyceraldehyde-3-

phosphate dehydrogenase (GAPDH) all of which were tested as potential candidate reference

genes (using methodology as described in Chapter 6 of this dissertation).

5.2.7 Statistical analysis

Data reported for larval burden as mean ± standard deviation was analyzed by two-

way ANOVA using the General Linear Models (GLM) procedure in SAS (Cary, North

Carolina) and p<0.05 was considered significant. For comparing variation in transcript levels

of cytokine genes, Pair-Wise Fixed Reallocation Randomization Test© in the REST-384 was

used (Pfaffl, et al., 2002). The mathematical model used to compute the relative expression

ratio of a target gene relies on its real-time PCR efficiencies (E) and the threshold cycle

118

difference (∆Ct) of an unknown sample versus a control (∆Ct control - sample). The target gene

expression is normalized to a reference gene (ref) in the equation mentioned below:





upregulation or downregulation of transcripts between timepoints, early-lactation (timepoint

II) and mid-lactation (timepoint III) were compared to late-gestation (timepoint I) expression

of which was relative to control (UN). For evaluating fold changes in transcript levels,

analyses were also done by comparing IN, UB and IB group each to the UN at respective

timepoints (See Appendix II).

5.3 Results

5.3.1 Breeding efficiency and larval burden assessment

Breeding efficiency, measured by dividing bred mice by total number of mice mated,

was used to evaluate the effect of A. caninum infection on the outcome of pregnancy. A total

of 82 female BALB/c mice were infected with 1000 A. caninum larvae post-impregnation, of

which 16 gave birth to live pups. Thus, breeding efficiency was 20% in the infected/bred (IB)

mice. Compared to the IB group, uninfected/bred (UB) mice had breeding efficiency of 45%

as 22 of 49 resulted in live litters. These results suggest that A. caninum infection has an

adverse effect on pregnancy, as shown by lower breeding efficiencies.

To compare larval distribution in dams, eight week old female BALB/c mice were

subcutaneously injected with 1000 A. caninum L3 larvae. A comparison of larval burden

during the course of A. caninum infection was made between infected/not bred (IN) and IB

119

mice at the three different timepoints corresponding to late gestation and early to mid

lactation (Figure 5.1). The total number of larvae recovered from IN mice was not

significantly different from the IB mice at late gestation or mid lactation. However, there was

a significant difference (p< 0.005) in the larval burden of IN mice (368 ± 64) and IB mice

(249 ± 31) at timepoint II, immediately postpartum.

To correlate the level of transmammary transmission from dam to pups, larval counts

were also assessed in litters from IB dams: 5 ± 4 and 9 ± 2 larvae were respectively obtained

during the timepoints for early and mid lactation, for timepoints II and III, respectively. The

larval burden in the feti (timepoint I) was not assessed as previous studies have shown that

there is no in utero transmission of A. caninum larvae from dam to fetus during pregnancy

(Steffe and Stoye, 1984; Arasu and Kwak, 1999).

5.3.2 Serum IL-4 levels

To compare the IL-4 levels during different phases of breeding, serum was collected

at late gestation (timepoint I), early lactation (timepoint II) and mid-lactation (timepoint III).

IL-4 levels were assessed in uninfected/not bred (UN) controls and compared to IL-4 levels

from IN, UB and IB mice at timepoints I, II and III. IL- 4 was not detectable in serum

samples from any group except for two individual mice in the IN group. Mouse # 4 and

mouse #5 had serum IL-4 levels of 210 pg/ml and 302 pg/ml respectively at timepoint III,

which is the last timepoint included in this study. Serum levels of IFN-γ could not evaluated

due to insufficient quantities of serum remaining as 50µl/well was required for the IFN-γ

ELISA (Pierce Endogen, Rockford, IL).

120

5.3.3 Transcript levels of IL-4 and IFN-γ in skeletal muscle and mammary gland

IL-4 mRNA transcripts levels were evaluated during the course of A. caninum

infection at different stages of pregnancy and lactation in skeletal muscle, mammary gland

and spleen using real-time PCR. REST software was used to analyze the fold changes in

expression of transcripts (Pfaffl, et al., 2002). Skeletal muscle was assessed as it is the

favored site of arrest for L3 larvae.

In skeletal muscle, IL-4 transcript levels were significantly upregulated during early

lactation at timepoint II (p< 0.05) in the IB group compared to the timepoint I (Figure 5.2a).

No significant changes in IL-4 levels were observed skeletal muscle of IN and UB groups

(Figures 9.3a, 9.3b in Appendix II). In mammary gland, IL-4 transcript levels were

significantly upregulated at timepoint I (p< 0.005) corresponding to two weeks (timepoint II

for IN group) of A. caninum infection in the IN group when compared to UN (Figure 5.2b).

However, IL-4 transcript levels were significantly downregulated at timepoint II (p< 0.05) in

the mammary gland of IB group as compared to timepoint I (Figure 5.2c). No significant

changes were observed in IL-4 transcripts in the UB group (Figure 9.3c in Appendix II). In

the spleen, no significant changes in expression of IL-4 were identified in the IN, UB or IB

groups (Figures 9.3d, 9.3e, 9.3f in Appendix II). In summary, IL-4 was predominantly

expressed in skeletal muscle but downregulated in mammary gland of the IB group.

IFN-γ transcript levels were also evaluated during the course of A. caninum infection

and different stages of pregnancy and lactation in skeletal muscle, mammary gland and

spleen using real-time PCR. No significant changes in IFN-γ transcripts were observed in any

of the groups in skeletal muscle (Figures 9.4a, 9.4b, 9.4c in Appendix II). In mammary gland,

IFN-γ transcript levels were significantly upregulated at timepoint I (p< 0.05) in the IN group

121

compared to the UN controls (Figure 5.3). This is consistent with the recent reports that

during the early phase of hookworm infection Th1 responses predominate over Th2. No

significant changes in IFN-γ transcripts were observed in mammary gland of UB and IB

groups. Similarly, no significant differences were noted in IFN-γ transcript levels in skeletal

muscle or spleen of the IN, UB and IB groups at the three timepoints of interest (Figures

9.4d-h in Appendix II).

Discussion

Hookworm infection has been globally recognized as one of the most common

persistent infections of humans, with majority of cases occurring in the tropics and subtropics

(de Silva, et al., 2003). Clinical symptoms of hookworm infection include anemia,

eosinophilic gastroenteritis, weight loss and retardation of growth. Latency and reactivation

is a common phenomenon in the life cycle of the hookworm which leads to chronicity of

infection that is well documented in dogs (Stoye and Krause, 1976). Developmentally

arrested third stage larvae (L3) of Ancylostoma species of hookworms have the capacity to

reactivate and mobilize during pregnancy which leads to transmission of L3 to the

immunologically naïve offspring via milk (Stone and Smith, 1973; Schad, 1979). We have

developed a mouse model of A. caninum infection that includes arrest and reactivation to

study the immunological changes occurring inside somatic tissues where larvae arrest and

migrate after reactivation, i.e. skeletal muscle and mammary gland, respectively. Classically,

a helminth infection generates a Th2 dominant response in the host defined by the production

of interleukin-4 (IL-4), IL-5, IL-9, IL-10 and IL-13 and consequently the development of

122

strong IgE, eosinophil and mast cell responses. Similarly, pregnancy has also been suggested

to model the parameters of a Th2 response (Wegmann, et al., 1993).

In this report we show that infected mice have a breeding efficiency of 20% as

compared to 45% in normal uninfected mice. It was also observed that some female mice had

a post-implantation vaginal plug but did not become pregnant. In those cases scarring in the

uterus was observed (personal observation, S. Trivedi), which suggested that fetal resorption

might have occurred. Immune responses generated against parasitic infections have been

implicated to have deleterious effects on pregnancy ultimately leading into its termination.

Studies in the protozoal parasite, Leishmania major, suggest that although a Th1 response is

protective against the parasitic infection, it can adversely affect pregnancy outcome. Th1

cytokines may be deleterious not only for placental maintenance but also for preimplantation

events (Krishnan, et al., 1996b).

Since little is known about cytokine patterns associated with hookworm infection;

especially at the site of larval arrest (skeletal muscle) and the site to which larvae migrate

after resuming development (mammary gland), we chose to evaluate two characteristic

cytokines which promote Th1 (IFN-γ) and Th2 (IL-4) cell expansion, respectively. We

measured IL-4 levels in serum because it was not feasible to isolate peripheral blood

mononuclear cells (PBMCs) or T cells from a relatively large sampling group of small

animals We utilized serum samples collected for measuring TGF-β and IGF-1 levels (S.

Trivedi Dissertation, Chapter 3) to detect IL-4 levels from the different groups. However, we

were unable to detect IL-4 levels in the majority of the serum samples except for two A.

caninum infected mice. Cytokine production is measured directly by either isolating PBMCs

123

or specific immune cells such as T helper (CD4+) and T cytotoxic (CD8+), and comparing

cytokine production between basal and mitogen-stimulated levels.

Since larvae arrest more commonly in skeletal muscle one would expect an ensuing

inflammatory response against the parasite. Using histochemistry we have previously

demonstrated a zone of eosinophils surrounding the larvae in the skeletal muscle (Arasu and

Kwak, 1999). It is likely that once the larvae have reactivated during pregnancy, a similar

inflammation would be caused by the traversing of the larvae during migration from skeletal

muscle to other parts of the body. Using real-time RT-PCR, we found that IL-4 was

prominently expressed in skeletal muscle (Figure 5.2a) but downregulated in mammary gland

of the infected bred mice immediately after parturition i.e. the early lactation (timepoint II)

(Figure 5.2c). Establishment of a Th2 response by an increase in IL-4 production has been

reported to favor worm expulsion in other gastrointestinal helminths (Finkelmann, et al.,

1997).

We also found that both IL-4 and IFN-γ transcripts were upregulated at two weeks

post-infection with A. caninum L3 (timepoint I) in mammary gland of infected mice (Figure

5.2b and 5.3). This mixed Th1/Th2 response, which could be caused by larvae traversing

throughout the body before arrest, has also been shown to occur in filarial nematodes.

Immune responses to larval stages during acute infections in residents of filarial-endemic

areas, travelers and transmigrants from a non-endemic to an endemic have been found to be

associated with a mixed Th1/Th2 cytokine profile (Klion, et al., 1991; Elson, et al., 1995;

Cooper, et al., 2001; Henry, et al., 2001). Observations from human subjects in endemic

regions in China and Brazil have shown profound cellular hyporesponsiveness induced by

chronic hookworm infection (Loukas et al., 2005). In a re-infection study in Papua New

124

Guinea, cytokine and proliferative responses to Necator were measured. Most subjects

produced detectable Th1 (IFN-γ) and Th2 (IL-4 and IL-5) cytokines in response to crude

adult worm antigen extract before anthelminthic treatment. Pre-treatment IFN-γ responses

were negatively associated with hookworm burden and increased significantly after

anthelminthic treatment (Quinnell et al., 2004). In a separate study, peripheral blood

mononuclear cells (PBMCs) from N. americanus infected school children, who had recently

received chemotherapy, had reduced proliferative capacity against phytohemagglutinin and

adult worm antigen extract compared to controls (Geiger et al., 2004). These individuals also

produced higher levels of IL-10 and lower levels of both Th1 (IL-12 and IFN-γ) and Th2 (IL-

5 and IL-13) cytokines. Such mixed Th1-type and Th2-type immune responsiveness

associated with chronic gastrointestinal parasitic nematode infections may reflect a state of

infection where a permissive Th1-type cytokine profile favors parasite persistence (Pit, et al.,

2001). These reports suggest that the mixed cytokine profile observed in mammary gland of

A. caninum infected mice may represent a typical cytokine profile associated with hookworm

infection. In conclusion, our results indicate that a Th2-like response characterized by

elevation in IL-4 levels predominates at the site of larval arrest in the infection and

pregnancy model. However, a mixed Th1/Th2 profile is observed in mammary gland of the

A. caninum infection mouse model.

125

Table 5.1 Primer sequences used for cytokine and reference gene transcript quantification by real-time RT-PCR

Primer pairs Gene Accession # Forward primer Reverse primer Product size 1 GAPDH BC098095 ccactacatggtctacatggttc ctcgctcctggaagatg 122 2 60S ribosomal protein BC011291 gattcgggatatgctgttgg aaagcctggaagaaggaggt 132 3 IL-4 NM_021283 cctcacagcaacgaagaaca atcgaaaagcccgaaagagt 155 4 IFN-γ K00083 gctttgcagctcttcctcat gtcaccatccttttgccagt 162

126

Larval counts

050

100150200250300350400450

IN IB IN IB IN IB

I II III

Num

ber

of la

rvae

Figure 5.1 Comparison of total larval burden in unbred versus bred BALB/c mice infected at times corresponding to day 19 gestation, day 1 and day 10 of postpartum lactation. Mice were infected with 1000 A. caninum L3 larvae subcutaneously. n = 5 mice per group. * indicates significant difference at p< 0.005 between unbred and bred groups.

*

127

0102030405060708090

100

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 5.2a IL-4 transcript levels in skeletal muscle of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. * indicates significant difference at p< 0.05 between uninfected unbred and infected bred groups.

*

128

-10

0

10

20

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 5.2b IL-4 transcript levels in mammary gland of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. * indicates significant difference at p< 0.005 between uninfected unbred and infected unbred groups.

*

129

-20

-15

-10

-5

0

5

10

15

20

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 5.2c IL-4 transcript levels in mammary gland of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. * indicates significant difference at p< 0.05 between uninfected unbred and infected bred groups.

*

130

-10

-5

0

5

10

15

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 5.3 IFN-γ transcript levels in mammary gland of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. * indicates significant difference at p<0.05 between uninfected unbred and infected unbred groups.

*

131

6. Evaluation of endogenous reference genes for real-time PCR quantification of

gene expression in Ancylostoma caninum

Shweta Trivedi, Prema Arasu*

Department of Molecular Biomedical Sciences, North Carolina State University, 4700

Hillsborough Street, Raleigh, NC 27606, USA.

Keywords: Ancylostoma hookworms, real-time PCR, validation, reference gene

*Corresponding author:

Prema Arasu

NC State University, 4700 Hillsborough Street, Raleigh, NC 27606, USA.

Tel: 1-919-513-6530; Fax: 1-919-513-6465; Email: [email protected]

132

Real-time PCR is a powerful technique for analyzing gene expression especially in parasitic

systems where samples are typically limited. As compared to conventional PCR, it provides a

dynamic range of absolute and relative quantification, greater precision, technical sensitivity,

and lower risk of sample contamination [1]. In relative quantification, expression levels are

normalized to endogenous reference genes or housekeeping genes which must be carefully

selected since high variability has been noted with commonly used genes such as actin and

glyceraldehyde-3-phosphate dehydrogenase [2-4]. The aim of this report was to identify a

reference gene(s) that showed minimal variability for gene expression analyses across

various life stages, strains and experimental conditions of the intestinal blood feeding

hookworm, Ancylostoma caninum. A. caninum is a major parasite of dogs and can cause

moderate to severe iron deficiency anemia, hypoproteinemia and bloody diarrhea that can be

fatal to puppies and immunosuppressed dogs [5]. A. caninum can also contribute to

cutaneous larva migrans and eosinophilic enteritis in man [6]. Related hookworms in tropical

and subtropical regions of the world also cause malnutrition and anemia in humans; in fact

children infected as infants show retarded cognitive and physical development [7]. For this

study, egg, larval stages 1 and 2 (L1/L2 combined), and infective larval stage 3 (iL3) were

derived from an A. caninum strain from a naturally infected dog in North Carolina (strain

‘N’) that has been passaged in laboratory Beagles since 2000 [ 8]. Adult worms were

recovered from the intestines of infected dogs that had been euthanized by the local animal

shelter and from each batch, intact male and female worms were separately picked and stored

at -70C. For inter-strain comparisons, A. caninum iL3s were obtained from T. Nolan,

University of Pennsylvania, from a strain that has been propagated since the early 1970s by

infecting steroid immuno-suppressed dogs (strain ‘P’). For the studies reported here, both A.

133

caninum strains were maintained in non-steroid treated Beagles. To study the effect of

treatment, L3s from N and P strains were incubated in the absence or presence of normal dog

serum which stimulates the infective pre-parasitic larvae to resume feeding, thereby

simulating the early phases of transition to the parasitic mode of development [9]. Total

RNA was extracted from the parasites using RNA Stat-60 as per manufacturer’s instructions

(TelTest, Friendswood, TX) and residual genomic DNA was removed by DNase I treatment.

Reverse transcription was done using oligo-d(T) priming and RT-PCR was performed with

intron-spanning actin primers to confirm the absence of contaminating genomic DNA (data

not shown). Based on previous reports [10 -12], genes for actin, 18S rRNA, 60S acidic

ribosomal protein, cAMP-dependent protein kinase A (cAMP), β-tubulin and RNA

polymerase II subunit (RNA pol II) were selected as candidate references from the A.

caninum database. While 18S rRNA does not have a polyA tail, it was included in the

analyses because sufficient internal priming by oligo d(T) can apparently occur [13]. The 5’

and 3’ gene-specific primers (presented 5’ to 3’) designed using the Primer3 software [ 14]

were respectively: Actin, GenBank Accession Number BQ667158, GAT CTG GCA CCA

CAC CTT CT and TCT CTG TTG CTC TTG GGG TTC (99 bp product), 18S rRNA,

BI773318, GCC CTC CAA TAG ATC CTC GT and CGC GCA AAT TAC CCA CTC (125

bp), 60S acidic ribosomal protein, BF250585, GTC GGA ATC GTC GGA AAG TA and

GTC TTG TTG CAT TTC GAG CA (167 bp), cAMP, U15983, ATG GGA GAA TCC AGC

AGA and TCC AAA ATC TTC ATG GCA AA (138 bp), β-tubulin, AF077870, CTG TTG

TCC CCT CAC CAA AG and TTT CAA GGT TCG GAA GCA AA (146 bp), and RNA pol

II, AW588389, TCT TGG TAC TCG TGC GCT TC and AGC GGA TCC GTC TCT CCT T

(75 bp). Primer design with the Primer 3 software was based on melting temperatures in the

134

57-63ºC range; actual optimization by test runs in the 50-65ºC range showed optimal

amplification at 55ºC. Conventional RT-PCR confirmed that none of the primer sets

amplified products from dog or E. coli (data not shown). The real-time PCR reactions were

carried out simultaneously on all candidate genes using the iCyclerTM iQ PCR detection

system (Bio-Rad Laboratories, Hercules, CA). Amplified product was detected with the

SYBR Green I DNA binding dye which has been shown to provide similar sensitivity and

reproducibility as sequence-specific fluorescent probes [15, 16]. In each 25µl reaction,

12.5µl of iQTM SYBR green supermix (Bio-Rad) was added to 300nM of each primer with

250ng of cDNA. PCR amplification was performed in duplicate for each sample using the

following cycle conditions: 3 min at 95°C followed by 45 repeats of 1 min at 95°C, 30s at

55°C and 30s at 72°C. A melt curve analysis step, included at the end of each run, verified

the absence of primer-dimers and non-specific products. Efficiency of the reactions, derived

from a standard curve of 10-fold serially diluted iL3 cDNA, was consistently in the 95-98%

range. For variation across stages, cDNAs from egg, L1/L2, iL3, and adult worms were

assessed and for variation among N and P strains, iL3 cDNA was used. The effect of serum

treatment was tested with iL3 and ssL3 cDNA from both strains. Non-template controls were

used to verify the absence of genomic DNA contamination. Each cDNA was analyzed in

duplicate and each experiment was repeated twice. The amplified products were also

examined on a 2% agarose gel for verification of the expected sizes (data not shown). In

real-time PCR, the threshold cycle number (Ct) at which the amplification-associated

fluorescence exceeds a specific threshold level of detection (background noise) is inversely

correlated with the amount of nucleic acid present in the sample [17]; Ct values were

therefore used to compare the variability in transcript levels of the reference genes and are

135

displayed as "box and whisker" plots with medians (as lines), 25th percentile to the 75th

percentile (as boxes) and ranges (as whiskers). Determination of coefficients of variation

(CV) and one-way ANOVA were performed using the Prism 4.0 software (GraphPad

Software Inc., San Diego, CA). For stage-specific analyses across egg, L1/L2, iL3, and

adult male and female worms, actin and cAMP showed the highest variability with CV of

27.4% and 16.4%, respectively, as compared to CV of 12.4%, 6.6% and 6.6% for the 60S, β-

tubulin and RNA pol II genes (Figure 6.1). The highest transcript levels were noted for 60S

(median Ct = 15.3) followed by β-tubulin (median Ct = 17.6) and RNA pol II (median Ct =

19.1). For comparison between the N and P geographical strains (Figure 6.2), actin, β-

tubulin and RNA pol II showed highest variability with CVs > 10% as compared to CVs of

3.5% and 1.5% for 60S and cAMP respectively. As with the analyses across stages, the 60S

transcript level was the most abundant with median Ct = 15.9. Finally, to assess the effect of

serum treatment on transcript levels of the reference genes, evaluations were done with iL3

and ssL3 from the N as well as the P strain of A. caninum. With the N strain, all reference

genes had CVs < 5% except for actin (CV of 17.6%); with the P strain, CVs were <7% for all

genes except for actin and 18S with CVs of 13.2% and 17% respectively. From a combined

analysis of the threshold cycle values from amplifications with iL3 and ssL3 cDNAs from

both N and P strains (Figure 6.3), the 60S (median Ct =15.9) and cAMP (median Ct =21)

genes emerged as the most suitable references with the lowest CVs of 4.3% and 3.4%,

respectively, as compared to CV > 10% for all the other genes. From these results, RNA pol

II and β-tubulin are suitable reference genes for analysis of transcript differences across

developmental stages of A. caninum but are highly unsuitable for comparisons between

different strains or treatment effects. Similarly, cAMP-dependent protein kinase A would be

136

an appropriate reference for less abundantly expressed genes in analyses of strains and

treatments but showed a high CV in stage-specific analyses. Overall, the 60S acidic

ribosomal protein was the best reference gene for analyses across stages, strains and

treatments especially for genes that are expressed at equally high abundance [4, 11].

Interestingly, the 18S rRNA non-polyadenylated transcript which is not expected to be

reproducibly reverse transcribed by oligo-d(T) primers showed lower CVs than the more

commonly used actin (Figure 6.1) or β-tubulin (Figures 6.2, 6.3) cytoskeletal proteins.

While this might suggest using the 18S rRNA as a reference gene for templates generated by

reverse transcription with random primers, randomly primed cDNA has been reported to

overestimate mRNA copy number in real time PCR analyses by up to 19-fold [18]. In a

recent report of gene expression in adult male and female worms of the Brugia malayi filarial

nematode, actin 2B, histone H3 and NADH dehydrogenase subunit I were all shown to be at

approximately equivalent transcript levels resulting in the use of actin-2B as the reference

gene for real-time PCR analyses [21]. In this study with A. caninum adult worms, Ct levels

for actin-1 were 29+/-1 and 16+/-0.3 for adult male and females, respectively, indicating a

wide discrepancy and its unsuitability as a reference gene. In conclusion, this study

confirms that reference genes should always be validated for real time gene expression

analyses and that the 60S acidic ribosomal protein is a suitable choice for studies on A.

caninum and related nematodes.

137

Acknowledgements

This work was supported by a grant from the National Institutes of Health (AI42908). We

thank Tori C. Freitas for helpful comments on the manuscript and Maria Correa for

assistance with the statistical analyses.

References

[1] Bustin, SA. Quantification of mRNA using real-time reverse transcription PCR

(RT-PCR): trends and problems. J Mol Endocrinol 2002; 29(1): 23-39.

[2] Gao B, Allen R, Davis EL, Baum TJ, and Hussey RS. Molecular characterization and

developmental expression of a cellulose-binding protein gene in the soybean cyst

nematode Heterodera glycines.Int J Parasitol. 2004;34 (12):1377-83.

[3] Schmittgen TD, Zakrajeck. B. Effect of experimental treatment on housekeeping gene

expression: validation by real-time, quantitative RT-PCR. J Biochem Biophys Methods

2002;46(1-2): 69-81.

[4] Thellin O, Zorzi W, Lakaye B, De Borman B, Coumans B, Hennen G, Grisar T, Igout

A, Heinen E. Housekeeping genes as internal standards: use and limits. J Biotechnol

1999;75(2-3): 291-5. [5] Georgi JR, Georgi, JR. Helminths. In: Georgi, J.R., Georgi, M.E.

(Eds.), Parasitology for Veterinarians, vol. V. WB Saunders Company,

Philadelphia, 1991;pp. 171–177.

[6] Croese J, Loukas A, Opdebeeck J, Prociv P. Occult enteric infection

by Ancylostoma caninum: a previously unrecognised zoonosis. Gastroenterology

1994;106, 3–12.

138

[7] Lozoff B, Jimenez E, Wolf AW. Long-term developmental outcome of infants with iron

deficiency. New Engl. J. Med. 1991 Sep 5;235, 687-694.

[8]Arasu P, Kwak D. Developmental arrest and pregnancy-induced transmammary

transmission of Ancylostoma caninum larvae in the murine model. Parasitol. 1999

Oct;85(5):779-84.

[9] Hawdon JM, Jones BF, Hoffman DR, Hotez PJ. Cloning and characterization of

Ancylostoma-secreted protein. A novel protein associated with the transition to parasitism

by infective hookworm larvae. J. Biol. Chem. 1996 Mar 22;271, 6672-8.

[10] Stepek G, Houston KM, Goodridge HS, Devaney E, Harnett W. Stage-specific and

species-specific differences in the production of the mRNA and protein for the filarial

nematode secreted product, ES-62. Parasitology. 2004 Jan;128(Pt.1): 91-98.

[11] Dheda K, Huggett JF, Bustin SA, Johnson MA, Rook G, Zumla A. Validation of

housekeeping genes for normalizing RNA expression in real-time PCR. Biotechniques

2004 Jul;37(1):112-4, 116, 118-9.

[12] Aboobaker AA, Blaxter, ML. Use of RNA interference to investigate gene function in

the human filarial nematode parasite Brugia malayi. Mol Biochem Parasitol 2003

Jun;129(1):41-51.

[13] Perrin R, Lange H, Grienenberger JM, Gagliardi D. AtmtPNPase is required for

multiple aspects of the 18S rRNA metabolism in Arabidopsis thaliana mitochondria.

Nucleic Acids Res. 2004 Sep 30;32(17):5174-82.

139

[14] Rosen S and Skaletsky HJ. Primer3 on the WWW for general users and for

biologist programmers. In: Krawetz S, Misener S (eds) Bioinformatics Methods and

Protocols: Methods in Molecular Biology. Humana Press, Totowa, NJ. 2000. pp 365-

386.

[15] Simon A, Labalette P, Ordinaire I, Frealle E, Dei-Cas E, Camus D and Delhaes L. Use

of fluorescence resonance energy transfer hybridization probes to evaluate quantitative

real-time PCR for diagnosis of ocular toxoplasmosis. J Clin Microbiol. 2004

Aug;42(8):3681-5.

[16] Maeda H, Fujimoto C, Haruki Y, Maeda T, Kokeguchi S, Petelin M, Arai H, Tanimoto

I, Nishimura F, Takashiba S. Quantitative real-time PCR using TaqMan and SYBR

Green for Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis, Prevotella

intermedia, tetQ gene and total bacteria. FEMS Immunol Med Microbiol. 2003 Oct

24;39(1):81-6.

[17] Walker NJ. Tech.Sight. A technique whose time has come. Science. 2002

296(5567):557-9.

[18] Zhang G and Byrne CD. Differential priming of RNA templates during cDNA

synthesis markedly affects both accuracy and reproducibility of quantitative

competitive reverse-transcriptase PCR. Biochem J. 1999 Jan 15; 337 ( Pt 2):231-41.

[19] Nolan TJ, Bhopale VM, Megyeri Z, Schad GA. Cryopreservation of human

hookworms. J Parasitol. 1994 Aug;80(4):648-50.

[20] Arasu P. In vitro reactivation of Ancylostoma caninum tissue-arrested third-stage

larvae by transforming growth factor-beta. J Parasitol 2001; 87: 733-8.

140

[21] Li BW, Rush AC, Tan J, Weil GJ. Quantitative analysis of gender-regulated

transcripts in the filarial nematode Brugia malayi by real-time RT-PCR. Mol Biochem

Parasitol. 2004 Oct;137(2):329-37

[22] Hawdon JM, Schad GA. Serum-stimulated feeding in vitro by third-stage infective

larvae of the canine hookworm Ancylostoma caninum.

J Parasitol. 1990 Jun;76(3):394-8.

141

Figure Legends

Figure 6.1 Expression levels of candidate reference genes across different developmental

stages of A. caninum. Box plots for each reference gene represent a compilation of the

threshold cycles (Ct) for cDNAs from egg, L1/L2, iL3, and adult male and female worms (n

= 20 from duplicate analyses of each cDNA in 2 independent experiments). Eggs were

harvested via sucrose flotation as previous described [19]. L1/L2 stage larvae were obtained

by hatching eggs on agar plates for 24-36 hrs, sucrose centrifugation to remove remaining

eggs and repeated washing of the larvae in medicated PBS before freezing at -70C. L3

stage larvae were harvested from fecal/charcoal co-cultures [20] and adult worms were

recovered from the intestines of infected dogs at necropsy.

Figure 6.2 Expression levels of candidate reference genes in two different strains of A.

caninum. The ‘N’ strain was derived from a naturally infected dog in North Carolina and the

‘P’ strain was originally derived from Maryland and has been propagated in laboratory dogs

for more than 25 years. Box plots for each reference gene include threshold cycles (Ct) for

cDNAs from infective iL3 (n = 8).

Figure 6.3 Effect of combination of serum treatment and strain of A. caninum on expression

levels of candidate reference genes. Box plots for each reference gene include threshold

cycles (Ct) for cDNAs from infective L3 (n = 4) and serum-stimulated L3 (n = 4). Infective

iL3 from N and P strains of A. caninum were incubated for 20-24 hrs at 37C, 5% CO2 with or

without 5% normal dog serum; larvae were then incubated with an equal volume of 5 mg/ml

142

flourescein isothiocyanate-labeled bovine serum albumin (Sigma). Serum-stimulated larvae

(80-90% positive) were scored for reactivation by examination for fluorescent intestinal

tracts using UV microscopy [22].

143

Figure 6.1

144

Figure 6.2

145

Figure 6.3

146

7. References Anderson, L., and J. Seilhamer. (1997). A comparison of selected mRNA and protein abundances in human liver. Electrophoresis. 18(3-4):533-7. Anderson, R. C. (2000). Nematode Parasites of Vertebrates: Their development and Transmission, edn 2, CABI Publishing, CAB International, Wallingford Oxon, UK. Arasu, P. (1998). Anthelmintic efficacy against tissue-arrested larvae of Ancylostoma caninum in murine hosts. J. Parasitol. 84(6):1263-7. Arasu, P., and A. Heller. (1999). Antibody responses in pregnancy-induced transmammary transmission of Ancylostoma caninum hookworm larvae. Vet. Immunol. Immunopathol. 70(3-4):289-98. Arasu, P. (2001). In vitro reactivation of Ancylostoma caninum tissue-arrested third-stage larvae by transforming growth factor-beta. J. Parasitol. 87(4):733-8. Arasu, P., and D. Kwak. (1999). Developmental arrest and pregnancy-induced transmammary transmission of Ancylostoma caninum larvae in the murine model. J. Parasitol. 85(5):779-84. Ashkar, A. A., and B. K. Croy. (2001). Functions of uterine natural killer cells are mediated by interferon gamma production during murine pregnancy. Semin. Immunol. 13(4):235-41. Ashton, F. T., V. M. Bhopale, D. Holt, G. Smith, and G. A. Schad. (1998). Developmental switching in the parasitic nematode Strongyloides stercoralis is controlled by the ASF and ASI amphidial neurons. J. Parasitol. 84(4):691-5. Ashton, F. T., J. Li, and G. A. Schad. (1999). Chemo- and thermosensory neurons: structure and function in animal parasitic nematodes. Vet. Parasitol. 84(3-4):297-316. Babu, S., and T. B. Nutman. (2003). Proinflammatory cytokines dominate the early immune response to filarial parasites. J. Immunol. 171(12):6723-32. Banerjee, D., O. Prakash, and M. G. Deo. (1970). Studies on the early stage of infection of Ancylostoma caninum in mice. Indian J. Med. Res. 58(10):1321-7. Bargmann, C. I., and H. R. Horowitz. (1991). Control of larval development by chemosensory neurons in Caenorhabditis elegans. Science. 251(4998):1243-6. Baumgartner, R. A., V. A. Derama, and M. A. Beaven. (1996). Constitutive and inducible mechanisms for synthesis and release of cytokines in immune cell lines. J. Immunol. 157(9):4087-93.

147

Beall, M. J., and E. J. Pearce. (2001). Human transforming growth factor-beta activates a receptor serine/threonine kinase from the intravascular parasite Schistosoma mansoni. J. Biol. Chem. 276(34):31613-9. Beall, M. J., and E. J. Pearce. (2002). "Transforming growth factor-beta and insulin-like signalling pathways in parasitic helminths." Int. J. Parasitol. Apr;32(4):399-404. Bhopale, V. M., E. K. Kupprion, F. T. Ashton, R. Boston, and G. A. Schad. (2001). Ancylostoma caninum: the finger cell neurons mediate thermotactic behavior by infective larvae of the dog hookworm. Exp. Parasitol. 97(2):70-6. Behnke, J. (1991). Human parasitic diseases: hookworm infections. In: Immunology (HM. Gilles and P. Ball, eds), pp. 93-155. Amsterdam: Elsevier. Birnby, D.A., E. M. Link, J. J. Vowels, H. Tian, P. L. Colacurcio, and J. H. Thomas. (2000). A transmembrane guanylyl cyclase (DAF-11) and Hsp90 (DAF-21) regulate a common set of chemosensory behaviors in Caenorhabditis elegans. Genetics. 155(1):85-104. Blagburn, G.L., D. S. Lindsay, J. L. Vaughan, N. S. Rippey, J. C. Wright, R. C. Lynn, W.G. Kelch, G.C., Ritchie, and D.I. Hepler. (1996). Prevalence of canine parasites based on fecal flotation examination. Comp. Cont. Edu. Pract. Veter. 18:483-489. Blaxter, M. (1998). Caenorhabditis elegans is a nematode. Science. 282(5396):2041-6. Boag, P. R., J. C. Parsons, P. J. Presidente T. W. Spithill, and J. L. Sexton. (2003). Characterisation of humoral immune responses in dogs vaccinated with irradiated Ancylostoma caninum. Vet. Immunol. Immunopathol. 92(1-2):87-94. Brabin, L., and B. Brabin. (1992). Parasitic infections in women and their consequences. Adv. Parasitol. 31:1-81. Braeckman, B. P., K. Houthoofd, and J. R.Vanfleteren. (2001). Insulin-like signaling, metabolism, stress resistance and aging in Caenorhabditis elegans. Mech. Ageing Dev. 122(7):673-93. Brand, A., and J. Hawdon. (2004). Phosphoinositide-3-OH-kinase inhibitor LY294002 prevents activation of Ancylostoma caninum and Ancylostoma ceylanicum third-stage infective larvae. Int. J. Parasitol. 34(8):909-14. Brenner, G. J., C. Nohan, and J. A. Moynihan. (1994). Similar immune response to nonlethal infection with herpes simplex virus-1 in sensitive (BALB/c) and resistant (C57BL/6) strains of mice. Cell. Immunol. 157(2):510-24. Brooker, S., J. M. Bethony, L. C. Rodrigues, N. Alexander, S. M. Geiger, and P. J. Hotez. (2005). Epidemiologic, immunologic and practical considerations in developing and evaluating a human hookworm vaccine. Expert Rev. Vaccines. 4(1):35-50.

148

Brunet, L. R., F. Finkelman, A. W. Cheever, M. A. Kopf, and E. J. Pearce. (1997). IL-4 protects against TNF-alpha-mediated cachexia and death during acute schistosomiasis. J. Immunol. 159(2):777-85. Bundy, D. A., M. S. Chan, and L. Savioli. (1995). Hookworm infection in pregnancy. Trans. R. Soc. Trop. Med. 89(5):521-2. Bungiro, R., and M. Cappello. (2004). Hookworm infection: new developments and prospects for control. Curr. Opin. Infect. Dis. 17(5):421-6. Burke, T. M., and E. L. Roberson. (1979a). Critical studies of fenbendazole suspension (10%) against naturally occurring helminth infections in dogs. Am. J. Vet. Res. 39(11):1799-1801. Burke, T. M., and E. L. Roberson. (1979b). Use of fenbendazole suspension (10%) against experimental infections of Toxocara canis and Ancylostoma caninum in beagle pups. Am. J. Vet. Res. 40(4):552-54. Burke, T. M., and E. L. Roberson. (1985). Prenatal and lactational transmission of Toxocara canis and Ancylostoma caninum: experimental infection of the bitch before pregnancy. Int. J. Parasitol. 15(1):71-5. Carr, A., and D. I. Pritchard. (1987). Antigen expression during development of the human hookworm, Necator americanus. Parasite Immunol. 9:219-234. Chan, M. S. (1997). The global burden of intestinal nematode infections--fifty years on. Parasitol. Today. 13(11):438-43. Cheng, H. L., S. L.Schneider, C. M. Kane, S. O. Gollnick, C. Grande, D. Thompson, E. Pietrzak, and T. B. Tomasi. (1993). TGF-beta 2 gene and protein expression in maternal and fetal tissues at various stages of murine development. J. Reprod. Immunol. 25(2):133-48. Chomczynski, P., and N. Sacchi. (1987). Single-step method of RNA isolation by acid guanidinium isothiocyanate-phenol-chloroform extraction. Anal. Biochem. 162(1): 156-59 Chotivanich, K., K. Sialmut, R. Udomsangpetch, K. A. Stepniewska, S. Pukrittayakamee, S. Looareesuwan, and N. J. White. (2001). Ex-vivo short-term culture and developmental assessment of Plasmodium vivax. Trans. R. Soc. Trop. Med. Hyg. 95(6):677-80. Ciesielski, S. D., J. R. Seed, J. C. Ortiz, and J. Metts. (1992). Intestinal parasites among North Carolina migrant farm workers. Am. J. Public Health. 82(9):1258-62. Cooper, P. J., T. Mancero, M. Espinel, C. Sandoval, R. Lovato, R. H. Guderian, and T. B. Nutman. (2001). Early human infection with Onchocerca volvulus is associated with an enhanced parasite-specific cellular immune response. J. Infect. Dis. 183(11):1662-8.

149

Coyne, C. P., and D. Brake. (2001). Characterisation of Haemonchus contortus-derived cell populations propagated in vitro in a tissue culture environment and their potential to induce protective immunity in sheep. Int. J. Parasitol. 31(4):359-76. Croese, J., A. Loukas, J. Opdebeeck, and P. Prociv. (1994). Occult enteric infection by Ancylostoma caninum: a previously unrecognized zoonosis. Gastroenterology. 106(1):3-12. Crompton, D.W. (1989). Hookworm disease: Current status and new directions. Parasitol. Today. 5: 1-2 Crompton, D. W. (2000). The public health importance of hookworm disease. Parasitology. 121 Suppl: S: 39-50. Crook, M., F. J. Thompson, W. N. Grant, and M. E. Viney. (2005). daf-7 and the development of Strongyloides ratti and Parastrongyloides trichosuri." Mol. Biochem. Parasitol. 139(2):213-23. Das, S. K., K. C. Flanders, G. K. Andrews, and S. K. Dey. (1992). Expression of transforming growth factor-beta isoforms (beta 2 and beta 3) in the mouse uterus: analysis of the periimplantation period and effects of ovarian steroids. Endocrinology. 130(6):3459-66 Davies, S. J., C. B. Shoemaker, and E. J. Pearce. (1998). A divergent member of the transforming growth factor beta receptor family from Schistosoma mansoni is expressed on the parasite surface membrane. J. Biol. Chem. 273(18):11234-40. De Clercq, D., M. Sacko, J. Behnke, F. Gilbert, P. Dorny, and J. Vercruysse. (1997). Failure of mebendazole in treatment of human hookworm infections in the southern region of Mali. Am. J. Trop. Med. Hyg. 57(1):25-30. de Silva, N. R., S. Brooker, P. J. Hotez, A. Montresor, D. Engels, and L. Savioli. (2003). Soil-transmitted helminth infections: updating the global picture. Trends Parasitol. 19(12):547-51. Deschoolmeester, M. L., and K. J. Else. (2002). Cytokine and chemokine responses underlying acute and chronic Trichuris muris infection. Int. Rev. Immunol. 21(4-5):439-67. Dickson, R., S. Awasthi, P. Williamson, C. Demellweek, and P. Garner. (2000). Effects of treatment for intestinal helminth infection on growth and cognitive performance in children: systematic review of randomised trials. B.M.J. 320(7251):1697-701. Drake, L. J., M. C. H. Jukes, R. J. Sternberg, and D. A. P.Bundy. (2000). Geohelminth Infections (Ascariasis, Trichuriasis and Hookworm): cognitive and developmental impacts. Seminars in Pediatric Infectious diseases. 11, 245-251. Dubey, J. P., and D. Lindsay. (1996). A review of Neospora caninum and neosporosis. Vet. Parasitol. 67(1-2):1-59.

150

Else, K. J., F. D. Finkelman, C. R. Maliszewski, and R. K.Grencis. (1994). Cytokine-mediated regulation of chronic intestinal helminth infection. J. Exp. Med. 179(1):347-51. Elson, L. H., M. Calvopina, W. Paredes, E. Araujo, J. E. Bradley, R. H. Guderian, and T. B. Nutman. (1995). Immunity to onchocerciasis: putative immune persons produce a Th1-like response to Onchocerca volvulus. J. Infect. Dis. 171(3):652-8. Estevez, M., L. Attisano, J. L. Wrana, P. S. Albert, J. Massague, and D. L. Riddle. (1993). The daf-4 gene encodes a bone morphogenetic protein receptor controlling C. elegans dauer larva development. Nature. 365(6447):644-9. Faas, M. M., A. Bouman, A. L. Veenstra van Nieuwenhoven, G. van der Schaaf, H. Moes,M. J. Heineman, and P. de Vos. (2005). Species differences in the effect of pregnancy on lymphocyte cytokine production between human and rat. J. Leukoc. Biol. (July Epub). Faas, M. M., G. A. Schuiling, E. A. Linton, I. L. Sargent, and C. W. Redman. (2000). Activation of peripheral leukocytes in rat pregnancy and experimental preeclampsia. Am. J. Obstet. Gynecol. 182(2):351-7. Finkelman, F. D., T. Shea-Donohue, J. Goldhill, C. A.Sullivan, S. C. Morris, K. B. Madden, W. C. Gause, and J. F. Urban Jr. (1997). Cytokine regulation of host defense against parasitic gastrointestinal nematodes: lessons from studies with rodent models. Annu. Rev. Immunol. 15:505-33. Forrester, S. G., P. W. Warfel, and E. J. Pearce. (2004). Tegumental expression of a novel type II receptor serine/threonine kinase (SmRK2) in Schistosoma mansoni. Mol. Biochem. Parasitol. 136(2):149-56.

Freitas, T. C., and P. Arasu (2005). Cloning and characterization of genes encoding two Transforming Growth Factor-beta-like ligands from the hookworm, Ancylostoma caninum. Int. J. Parasitol. (In Press).

Fried, M., R. O. Muga, A. O. Misore, and P. E. Duffy. (1998). Malaria elicits type 1 cytokines in the human placenta: IFN-gamma and TNF-alpha associated with pregnancy outcomes. J. Immunol. 160(5):2523-30. Gagliardo, L. F., C. McVay, and J. A. Appleton. (2002). Molting, ecdysis, and reproduction of Trichinella spiralis are supported in vitro by intestinal epithelial cells. Infect. Immun. 70(4):1853-9. Gamble, H. R., and L. S. Mansfield. (1996). Characterization of excretory-secretory products from larval stages of Haemonchus contortus cultured in vitro. Vet. Parasitol. 62(3-4):291-305.

151

Ganguly, N. K., R. C. Mahajan, R. Sehgal, P. Shetty, and J. B. Dilawari. (1988). Role of specific immunoglobulin E to excretory-secretory antigen in diagnosis and prognosis of hookworm infection. J. Clin. Microbiol. 26(4):739-42. Garcia, L. S. and L.R. Ash. (1979). Diagnostic parasitology: clinical laboratory manual. C. V. Mosby Co., St. Louis, Missouri, 46p. Garside, P., and J. M. Behnke. (1989). Ancylostoma ceylanicum in the hamster: observations on the host-parasite relationship during primary infection. Parasitology. 98 Pt 2:283-9. Gause, W. C., J. F. Urban Jr, and M. J. Stadecker. (2003). The immune response to parasitic helminths: insights from murine models. Trends Immunol. 24(5):269-77. Geiger, S. M., C. L. Messara, J. Bethony, P. T. Soboslay, and R. Correa-Oliveira. (2004). Cellular responses and cytokine production in post-treatment hookworm patients from an endemic area in Brazil. Clin. Exp. Immunol. 136(2):334-40. Georgi, J. R., and M. E. Georgi. (1991). Helminths. In: Georgi JR, Georgi ME, editors. Parasitology for veterinarians, vol. V. Philadelphia: WB Saunders Company; 1991. p. 171-7. Georgi, L. L., P. S. Albert, and D. L. Riddle. (1990). daf-1, a C. elegans gene controlling dauer larva development, encodes a novel receptor protein kinase. Cell. 61: 635-645. Gilmore, W., L. P. Weiner, and J. Correale. (1997). Effect of estradiol on cytokine secretion by proteolipid protein-specific T cell clones isolated from multiple sclerosis patients and normal control subjects. J. Immunol. 158(1):446-51. Gomez-Escobar, N., A. van den Biggelaar, and R. M. Maizels. (1997). A member of the TGF-beta receptor gene family in the parasitic nematode Brugia pahangi. Gene.199(1-2):101-9. Gomez-Escobar, N., E. Lewis, and R. M. Maizels. (1998). A novel member of the transforming growth factor-beta (TGF-beta) superfamily from the filarial nematodes Brugia malayi and B. pahangi. Exp. Parasitol. 88(3):200-9. Gomez-Escobar, N., W. F. Gregory, and R. M. Maizels. (2000). Identification of tgh-2, a filarial nematode homolog of Caenorhabditis elegans daf-7 and human transforming growth factor beta, expressed in microfilarial and adult stages of Brugia malayi. Infect. Immun. 68(11):6402-10. Goud, G. N., M. Bottazzi, B. Zhan, S. Mendez, V. Deumic, J. Plieskatt, S. Liu, Y. Wang, L. Bueno, R. Fujiwara, A. Samuel, S. Y. Ahn, M. Solanki, O. A. Asojo, J. Wang, J. M. Bethony, A. Loukas, M. Roy, and P. J. Hotez. (2005). Expression of the Necator americanus hookworm larval antigen Na-ASP-2 in Pichia pastoris and purification of the recombinant protein for use in human clinical trials. Vaccine (September Epub).

152

Granzer, M., and W. Haas. (1991). Host-finding and host recognition of infective Ancylostoma caninum larvae. Int. J. Parasitol. 21(4):429-40. Grencis, R. K. (2001). Cytokine regulation of resistance and susceptibility to intestinal nematode infection - from host to parasite. Vet. Parasitol. 100(1-2):45-50. Gunes, H., and A. Maestro. (1997). Prolactin receptor gene expression in rat splenocytes and thymocytes during oestrous cycle, pregnancy and lactation. Cell Prolif. 30(5):219-35. Gunther, C.V., L. L. Georgi, and D. L. Riddle. (2000). A Caenorhabditis elegans type I TGF beta receptor can function in the absence of type II kinase to promote larval development. Development. 127(15):3337-47. Gygi, S. P., Y. Rochon, B. R. Franza, and R. Aebersold. (1999). Correlation between protein and mRNA abundance in yeast. Mol. Cell Biol. 19(3):1720-30. Hawdon, J. M., and G. A. Schad. (1990). Serum-stimulated feeding in vitro by third-stage infective larvae of the canine hookworm Ancylostoma caninum. J. Parasitol. 76(3):394-8. Hawdon, J. M., and G. A. Schad. (1991). "Developmental adaptations in nematodes. In Parasitism-conflict or co-existence. C. Toft (ed.) Oxford University Press, Oxford, U.K., p274-298." Hawdon, J. M., and G. A. Schad. (1992). Ancylostoma caninum: reduced glutathione stimulates feeding by third-stage infective larvae. Exp. Parasitol. 75(1):40-6. Hawdon, J. M., S. W. Volk, D. I. Pritchard, and G. A. Schad. (1992). Resumption of feeding in vitro by hookworm third-stage larvae: a comparative study. J. Parasitol. 78(6): 1036-40. Hawdon, J. M., and G. A. Schad. (1993). Ancylostoma caninum: glutathione stimulates feeding in third-stage larvae by a sulfhydryl-independent mechanism. Exp. Parasitol. 77(4):489-91. Hawdon, J. M., S. W. Volk, R. Rose, D. I. Pritchard, J. M. Behnke, and G. A. Schad. (1993). Observations on the feeding behaviour of parasitic third-stage hookworm larvae. Parasitology. 106 (Pt 2):163-9. Hawdon, J. M., and S. A. Johnston. (1996a). Hookworms in the Americas: An alternative to trans-Pacific contact. Parasitol. Today. 12(2):72-4. Hawdon, J. M., B. F. Jones, D. R. Hoffman, and P. J. Hotez. (1996b). Cloning and characterization of Ancylostoma-secreted protein. A novel protein associated with the transition to parasitism by infective hookworm larvae. J. Biol. Chem. 271(12):6672-8. Hawdon, J. M., and B. Datu. (2003). The second messenger cyclic GMP mediates activation in Ancylostoma caninum infective larvae. Int. J. Parasitol. 33(8):787-93.

153

Heinzel, F. P., M. D. Sadick, B. J. Holaday , R. L. Coffman, and R. M. Locksley. (1989). Reciprocal expression of interferon gamma or interleukin 4 during the resolution or progression of murine leishmaniasis. Evidence for expansion of distinct helper T cell subsets. J. Exp. Med. 169(1):59-72. Henry, N. L., M. Law, T. B. Nutman, and A. D. Klion. (2001). Onchocerciasis in a nonendemic population: clinical and immunologic assessment before treatment and at the time of presumed cure. J. Infect. Dis.183(3):512-6. Hernandez, H. J., Y. Wang, and M. J. Stadecker. (1997). In infection with Schistosoma mansoni, B cells are required for T helper type 2 cell responses but not for granuloma formation. J. Immunol. 158(10):4832-7. Hernandez, H. J., Y. Wang, N. Tzellas, and M. J. Stadecker. (1997). Expression of class II, but not class I, major histocompatibility complex molecules is required for granuloma formation in infection with Schistosoma mansoni. Eur. J.Immunol. 27(5):1170-6. Hernandez, H. J., A. H. Sharpe, and M. J. Stadecker. (1999). Experimental murine schistosomiasis in the absence of B7 costimulatory molecules: reversal of elicited T cell cytokine profile and partial inhibition of egg granuloma formation. J. Immunol. 162(5):2884-9. Honore, S., Y. J. Garin, A. Sulahian, J. P. Gangneux, and F. Derouin. (1998). Influence of the host and parasite strain in a mouse model of visceral Leishmania infantum infection. FEMS Immunol. Med. Microbiol. 21(3):231-9. Hotez, P., J. Hawdon, and G. A. Schad. (1993). Hookworm larval infectivity, arrest and amphiparatenesis: the Caenorhabditis elegans Daf-c paradigm. Parasitol. Today. 9(1): 23-6. Hotez, P. J., S. Brooker, J. M. Bethony, M. E. Bottazzi, A. Loukas, and S. Xiao. (2004). Hookworm infection. N. Engl. J. Med. 351(8):799-807. Hotez, P. J., and D. I. Pritchard. (1995). Hookworm infection. Sci Am. 272(6):68-74. Husmann, I., L. Soulet, J. Gautron, I. Martelly, and D. Barritault. (1996). Growth factors in skeletal muscle regeneration. Cytokine Growth Factor Rev. 7(3):249-58. Inoue, T., and J. H. Thomas. (2000). Targets of TGF-beta signaling in Caenorhabditis elegans dauer formation. Dev. Biol. 217(1):192-204. Jian, X., L. Sen, Q. Hui-Qin, R. Hai-Nan, L. Tie-Hua, X. Hai-Chou, P. J. Hotez ,and X. Shu-Hua. (2003). Necator americanus: maintenance through one hundred generations in golden hamsters (Mesocricetus auratus). I. Host sex-associated differences in hookworm burden and fecundity." Exp. Parasitol. 104(1-2):62-6.

154

Joachim, R., A. C. Zenclussen, B. Polgar, A. J. Douglas, S. Fest, M. Knackstedt, B. F. Klapp, and P. C. Arck. (2003). The progesterone derivative dydrogesterone abrogates murine stress-triggered abortion by inducing a Th2 biased local immune response. Steroids. 68(10-13):931-40. Kano, R., Y. Masukata, Y. Omata, Y. Kobayashi, R. Maeda, and A. Saito. (2005). Relationship between type 1/type 2 immune responses and occurrence of vertical transmission in BALB/c mice infected with Neospora caninum. Vet. Parasitol. 129(1-2):159-64. Kassai, T. (1999). Ancylostomatidosis: Hookworm disease. In Veterinary Helminthology. Kassai T Eds Butterworth Heinmann, Oxford. Khan, I. A., J. D. Schwartzaman, S. Fonseka, and L. H. Kasper. (1997). Neospora caninum: role for immune cytokines in host immunity. Exp. Parasitol. 85(1):24-34. Kidd, P. (2003). Th1/Th2 balance: the hypothesis, its limitations, and implications for health and disease. Altern. Med. Rev. 8(3):223-46. Kim, S., and R. R. Voskuhl. (1999). Decreased IL-12 production underlies the decreased ability of male lymph node cells to induce experimental autoimmune encephalomyelitis. J. Immunol. 162(9):5561-8. Kimura, K. D., H. A. Kissenbaum, Y. Liu, and G. Ruvkun. (1997). daf-2, an insulin receptor-like gene that regulates longevity and diapause in Caenorhabditis elegans. Science. 277(5328):942-6. King, C. L., and T. B. Nutman. (1991). Regulation of the immune response in lymphatic filariasis and onchocerciasis. Immunol. Today. 12(3):A54-8. Klion, A. D., A. Massougbodji, B. C. Sadeler, E. A. Ottesen, and T. B.Nutman. (1991). Loiasis in endemic and nonendemic populations: immunologically mediated differences in clinical presentation. J. Infect.Dis. 163(6):1318-25. Krishnan, L., L. J. Guilbert, A. S. Russell, T. G. Wegmann, T. R. Mosmann, and M. Belosevic. (1996a). Pregnancy impairs resistance of C57BL/6 mice to Leishmania major infection and causes decreased antigen-specific IFN-gamma response and increased production of T helper 2 cytokines. J. Immunol. 156(2):644-52. Krishnan, L., L. J. Guilbert, T. G. Wegmann, M. Belosevic, and T. R. Mosmann. (1996b). T helper 1 response against Leishmania major in pregnant C57BL/6 mice increases implantation failure and fetal resorptions. Correlation with increased IFN-gamma and TNF and reduced IL-10 production by placental cells. J. Immunol. 156(2):653-62. Kumar, S., and D. I. Pritchard. (1994). Apparent feeding behaviour of ensheathed third-stage infective larvae of human hookworms. Int. J. Parasitol. 24(1):133-6.

155

Landmann, J. K., and P. Prociv. (2003). Experimental human infection with the dog hookworm, Ancylostoma caninum. Med. J. Aust. 178(2):69-71. Lee, C.Y., F. W. Bazer, and F. A. Simmen. (1993). Expression of components of the insulin-like growth factor system in pig mammary glands and serum during pregnancy and pseudopregnancy: effects of oestrogen. J, Endocrinol. 137(3):473-83. Lee, K. T., M. D. Little, and P. C. Beaver PC. (1975). Intracellular (muscle-fiber) habitat of Ancylostoma caninum in some mammalian hosts. J. Parasitol. 61(4):585-98. Leiby, D. A., H. M. el Nagger, and G. A. Schad. (1987). Thirty generations of Ancylostoma duodenale in laboratory-reared beagles. J. Parasitol. 73(4):844-8. Li, W., S. G. Kennedy, and G. Ruvkun. (2003). daf-28 encodes a C. elegans insulin superfamily member that is regulated by environmental cues and acts in the DAF-2 signaling pathway. Genes Dev. 17(7):844-58. Locksley, R. M., and J. A. Louis. (1992). Immunology of leishmaniasis. Curr. Opin. Immunol. 4(4):413-8. Louis, J., H. Himmelrich, C. Parra-Lopez, F. Tacchini-Cottier, and P. Launois. (1998). Regulation of protective immunity against Leishmania major in mice. Curr. Opin. Immunol. 10(4):459-64. Loukas, A., S. L. Constant, and J. M. Bethony. (2005). Immunobiology of hookworm infection. FEMS Immunol. Med. Microbiol. 43(2):115-24. Loukas, A., and P. Prociv. (2001). Immune responses in hookworm infections. Clin. Microbiol. Rev. 14(4):689-703. Lunden, A., J. Mark, S. W. Maley, and E. A. Innes. (1998). Cellular immune responses in cattle experimentally infected with Neospora caninum. Parasite Immunol. 20(11):519-26. MacDonald, A. S., and E. J. Pearce. (2002). Cutting edge: polarized Th cell response induction by transferred antigen-pulsed dendritic cells is dependent on IL-4 or IL-12 production by recipient cells. J. Immunol. 168(7):3127-30. Maizels, R. M., D. A. Bundy, M. E. Selkirk, D. F. Smith, and R. M. Andersen. (1993). Immunological modulation and evasion by helminth parasites in human populations. Nature. 365(6449):797-805. Malone, E. A., T. Inoue, and J. H. Thomas. (1996). Genetic analysis of the roles of daf-28 and age-1 in regulating Caenorhabditis elegans dauer formation. Genetics. 143(3):1193-205.

156

ManWarren, T., L. F. Gagliardo, J. Geyer, C. McVay, S. Pearce-Kelling, and J. Appleton. (1997). Invasion of intestinal epithelia in vitro by the parasitic nematode Trichinella spiralis. Infect. Immun. 65(11):4806-12. Marcotty, C., F. Frankenne, S. Meuris, and G. Hennen. (1994). Immunolocalization and expression of insulin-like growth factor I (IGF-I) in the mammary gland during rat gestation and lactation. Mol. Cell Endocrinol. 99(2):237-43. Massague, J., S. W. Blain, and R. S. Low. (2000). TGF-beta signaling in growth control, cancer, and heritable disorders. Cell. 103(2):295-309. Massey, H. C. Jr., M. Nishi, K. Chaudhary, N. Pakpour, and J. B. Lok . (2003). Structure and developmental expression of Strongyloides stercoralis fktf-1, a proposed ortholog of daf-16 in Caenorhabditis elegans. Int. J. Parasitol. 33(13):1537-44. Maxwell, C., R. Hussain, T. B. Nutman, R. W. Poindexter, M. D. Little, G. A. Schad, and E. A. Ottesen. (1987). The clinical and immunologic responses of normal human volunteers to low dose hookworm (Necator americanus) infection. Am. J. Trop. Med. Hyg. 37(1):126-34. McElwee, J., K. Bubb, and J. H. Thomas. (2003). Transcriptional outputs of the Caenorhabditis elegans forkhead protein DAF-16. Aging Cell. 2(2):111-21. Mendez, S., J. Valenzuela, W. Wu, and P. J. Hotez. (2005). Host cytokine production, lymphoproliferation, and antibody responses during the course of Ancylostoma ceylanicum infection in the Golden Syrian hamster. Infect. Immun. 73(6):3402-7. Menendez, C. (1995). "Malaria during pregnancy: a priority area of malaria research and control." Parasitol Today. 11(5):178-83. Miller, A C., J. Walker, R. Jaworski, W. de Launey, and R. Paver. (1991). Hookworm folliculitis. Arch. Dermatol. 127(4):547-9. Moormann, A. M., A. D. Sullivan, R. A. Rochford, S. W. Chensue, P. J. Bock, T. Nyirenda, and S. R. Meshnick. (1999). Malaria and pregnancy: placental cytokine expression and its relationship to intrauterine growth retardation. J. Infect. Dis. 180(6):1987-93. Moro, M. H., J. Bjornson, E. V. Marietta, E. K. Hofmeister, J. J. Germer, E. Bruinsma, C. S. David, and D. H. Persing. (2001). Gestational attenuation of Lyme arthritis is mediated by progesterone and IL-4. J. Immunol. 166(12):7404-9. Morris, J. Z, H. A. Tissenbaum, and G. Ruvkun. (1996). A phosphatidylinositol-3-OH kinase family member regulating longevity and diapause in Caenorhabditis elegans. Nature. 382(6591):536-9.

157

Mosmann, T. R., H. Cherwinski, M. W. Bond, M. A. Giedlin, and R. L. Coffman. (1986). Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J. Immunol. 136(7):2348-57. Mosmann, T. R, S. Sad. (1996). The expanding universe of T-cell subsets: Th1, Th2 and more. Immunol. Today. 17(3):138-46. Murakami, M., M. Koga, and Y. Ohshima. (2001). DAF-7/TGF-beta expression required for the normal larval development in C. elegans is controlled by a presumed guanylyl cyclase DAF-11. Mech. Dev. 109(1):27-35. Nabors, G. S., T. Nolan, W. Croop, J. Li, and J. P. Farrell. (1995). The influence of the site of parasite inoculation on the development of Th1 and Th2 type immune responses in (BALB/c x C57BL/6) F1 mice infected with Leishmania major. Parasite Immunol. 17(11):569-79. Nichols, R. L. (1956). The etiology of visceral larva migrans. II. Comparative larval morphology of Ascaris lumbricoides, Necator americanus, Strongyloides stercoralis and Ancylostoma caninum. J. Parasitol. 42: 363-99. Norris, D. E. (1971). The migratory behavior of the infective-stage larvae of Ancylostoma braziliense and Ancylostoma tubaeforme in rodent paratenic hosts. J. Parasitol. 57(5):998-1009. O’Gara, A. (1998). Cytokines induce the development of functionally heterogeneous T helper cell subsets. Immunity. 8(3):275-83. Ogg, S., S. Paradis, S. Gottlieb, G. I. Patterson, L. Lee, H. A. Tissenbaum, and G. Ruvkun. (1997). The Fork head transcription factor DAF-16 transduces insulin-like metabolic and longevity signals in C. elegans. Nature. 389(6654):994-9. Ostensen, M. (1999). Sex hormones and pregnancy in rheumatoid arthritis and systemic lupus erythematosus. Ann. N. Y. Acad. Sci. 876:131-43; discussion 144. Paradis, S., M. Ailion, A. Toker, J. H. Thomas, and G. Ruvkun. (1999). A PDK1 homolog is necessary and sufficient to transduce AGE-1 PI3 kinase signals that regulate diapause in Caenorhabditis elegans. Genes Dev. 13(11):1438-52. Paradis, S., and G. Ruvkun. (1998). Caenorhabditis elegans Akt/PKB transduces insulin receptor-like signals from AGE-1 PI3 kinase to the DAF-16 transcription factor. Genes Dev. 12(16):2488-98. Patterson, G. I., and R. W. Padgett. (2000). "TGF beta-related pathways. Roles in Caenorhabditis elegans development." Trends Genet. 16(1):27-33.

158

Pearce, E. J., P. Caspar, J. M. Grzych, F. A. Lewis, and A. Sher. (1991). Downregulation of Th1 cytokine production accompanies induction of Th2 responses by a parasitic helminth, Schistosoma mansoni. J. Exp. Med. 173(1):159-66. Pearce, E. J., and S. L. Reiner. (1995). Induction of Th2 responses in infectious diseases. Curr. Opin. Immunol. 7(4):497-504. Pearce, E. J., J. P. Vasconselos, L. R. Brunet, and E. A. Sabin. (1996). IL-4 in schistosomiasis. Exp. Parasitol. 84(2):295-9. Pfaffl, M. W., G. W. Horgan and L. Dempfle. (2002). Realtive Expression Software Tool (REST) for group-wise comparison and statistical analysis of relative expression results in real-time PCR. Nucleic Acids Res. 30(9): e36 Pierce, S. B., M. Costa, R. Wisotzkey, S. Devadhar, S. A. Homburger, A. R. Buchman, K. C. Ferguson, J. Heller, D. M. Platt, A. A. Pasquinelli, L. X. Liu, S. K. Doberstein, and Pit, D. S., A. M. Polderman, S. Baeta, H. Schulz-Key, and P. T. Soboslay. (2001). Parasite-specific antibody and cellular immune responses in human infected with Necator americanus and Oesophagostomum bifurcum. Parasitol. Res. 87(9):722-9. Pit, D. S., A. M. Polderman, H. Schulz-Key, and P. T. Soboslay. (2000). Prenatal immune priming with helminth infections: parasite-specific cellular reactivity and Th1 and Th2 cytokine responses in neonates. Allergy. 55(8):732-9. Pritchard, D. I., and E. A. Walsh. (1995). The specificity of the human IgE response to Necator americanus. Parasite Immunol. 17(11):605-7. Pritchard, D. I., R. J. Quinnell, and E. A. Walsh. (1995). Immunity in humans to Necator americanus: IgE, parasite weight and fecundity Parasite Immunol. 17(2):71-5. Prociv, P., and J. Croese. (1990). Human eosinophilic enteritis caused by dog hookworm Ancylostoma caninum. Lancet. 335(8701):1299-302. Quinn, H. E., C. M. Miller, P. A. Windsor, and J. T.Ellis. (2002a). Characterization of an outbred pregnant mouse model of Neospora caninum infection. J. Parasitol. 88(4):691-6. Quinn, H. E., J. T. Ellis, and N. C. Smith. (2002b). Neospora caninum: a cause of immune-mediated failure of pregnancy? Trends Parasitol. 18(9):391-4. Quinn, H. E., C. M. Miller, and J. T.Ellis. (2004).The cell-mediated immune response to Neospora caninum during pregnancy in the mouse is associated with a bias towards production of interleukin-4. Int. J. Parasitol. 34(6):723-32. Quinnell, R. J., D. I. Pritchard, A. Raiko, A. P. Brown, and M. A. Shaw. (2004). Immune responses in human necatoriasis: association between interleukin-5 responses and resistance to reinfection. J. Infect. Dis. 190(3):430-8.

159

Quinnell, R. J., A. F. Slater, P. Tighe, E. A. Walsh, A. E. Keymer, and D. I. Pritchard. (1993). Reinfection with hookworm after chemotherapy in Papua New Guinea. Parasitology. 106 ( Pt 4):379-85. Rajan, T. V. (1998). A hypothesis for the tissue specificity of nematode parasites. Exp. Parasitol. 89(1):140-2. Rajan, T. V., N. Paciorkowski, I. Kalajzic, and C. McGuiness. (2003). Ascorbic acid is a requirement for the morphogenesis of the human filarial parasite Brugia malayi. J Parasitol. 89(4):868-70. Ren P, C. S. Lim, R. Johnsen, P. S. Albert, D. Pilgrim, and D. L. Riddle. (1996). Control of C. elegans larval development by neuronal expression of a TGF-beta homolog. Science. 274 (5291):1389-91. Riddle, D., and P. Albert. (1998). Regulation of dauer larva development. In “C. elegans II.” Ridle D, Blumental T, Meyer BJ, and Priess JR. Eds. Cold Spring Harbor Laboratory Press. Rose R. A., and J. Behnke (1990). "Necator americanus in the DSN hamster: density-dependent expulsion of adult worms during primary infection." Parasitology. 100 Pt 3:469-78. Rozen, S., and H. Skaletsky (2000). Primer3 on the WWW for general users and for biologist programmers. Methods Mol. Biol. 132:365-86. Ruvkun, G. (2001). Regulation of DAF-2 receptor signaling by human insulin and ins-1, a member of the unusually large and diverse C. elegans insulin gene family. Genes Dev. 15(6):672-86. Sadick, M. D., R. M. Locksley, C. Tubbs, and H. V. Raff. (1986). Murine cutaneous leishmaniasis: resistance correlates with the capacity to generate interferon-gamma in response to Leishmania antigens in vitro. J. Immunol. 136(2):655-61. SantaLucia, J. Jr. (1998). A unified view of polymer, dumbbell, and oligonucleotide DNA nearest-neighbor thermodynamics. Proc. Natl. Acad. Sc. USA. 95(4):1460-5. Schad, G. A. (1979). Ancylostoma duodenale: Maintenance through six generations in helminth-naïve pups. Exp. Parasitol. 47, 246-53. Schad, G. A., and M. R. Page. (1982). Ancylostoma caninum: adult worm removal, corticosteroid treatment, and resumed development of arrested larvae in dogs. Exp. Parasitol. 54(3):303-9.

160

Schad, G. A. (1990). Hypobiosis and related phenomena in hookworm infection. In “Hookworm Disease: Current Status and New Direction” Schad GA and Warren KS, Eds. Taylor and Francis, London. Schad, G. A. (1991). Presidential address. Hooked on hookworm: 25 years of attachment. J. Parasitol. 77(2):176-86. Schneider, S. L., S. O. Gollnick, C. Grande, J. E. Pazik, and T. B. Tomasi. (1996). Differential regulation of TGF-beta 2 by hormones in rat uterus and mammary gland. J. Reprod. Immunol. 32(2):125-44. Sciacca, J., W. M. Forbes, F. T. Ashton, E. Lombardini, H. R. Gamble, and G. A. Schad. (2002). Response to carbon dioxide by the infective larvae of three species of parasitic nematodes. Parasitol. Int. 51(1):53-62. Scott, P., and J. P. Farrell. (1998). Experimental cutaneous leishmaniasis: induction and regulation of T cells following infection of mice with Leishmania major. Chem. Immunol. 70:60-80. Shoop, W. L. (1990). Vertical transmission of helminths: Hypobiosis and amphiparatenesis. Parasitol. Today. Feb;7(2):51-4. Smith, H. L., N. Paciorkowski, S. Babu, and T. V. Rajan. (2000). Development of a serum-free system for the in vitro cultivation of Brugia malayi infective-stage larvae. Exp. Parasitol. 95(4):253-64. Stassens, P., P. W. Bergum, Y. Gansemans, L. Jespers, Y. Laroche, S. Huang, S. Maki, J. Messens, M. Lauwereys, M. Cappello, P. J. Hotez, I. Lasters, and G. P. Vlasuk. (1996). Anticoagulant repertoire of the hookworm Ancylostoma caninum. Proc. Natl. Acad. Sci. USA. (93(5):2149-54. Steffe, G. A., and M. Stoye. (1984). [Behavior of galactogen-transmitted larvae of parasitic nematodes in the mouse]. Zentralbl Veterinarmed B. 31(4):248-60. Stone, W., and F. Smith. (1973). Infection of mammalian hosts by milk-borne nematode larvae: a review. Exp. Parasitol. 34(2):306-12. Stone, W. M., and M. R.Girardeau. (1968). Transmammary passage of Ancylostoma caninum larvae in dogs. J. Parasit. 54: 426-429. Stoye, M. (1973). [Studies on the possibility of prenatal and galactogenic infections with Ancylostoma caninum Ercolani 1859 (Ancylostomidae) in the dog]. Zentralbl Veterinarmed B. 20(1):1-39. Stoye, M., and J. Krause. (1976). [Reactivation of inhibited larvae of Ancylostoma caninum. Effect of estradiol and progesterone]. Zentralbl Veterinarmed B. 23(10):822-39.

161

Swerczek, T. W., S. W. Nielsen, and C. F. Helmboldt. (1971). Transmammary passage of Toxocara cati in the cat. Am. J. Vet. Res. 32, 89-92. Tanaka, T., T. Hamada, N. Inoue, H. Nagasawa, K. Fujisaki, N. Suzuki, and T. Mikami. (2000). The role of CD4(+) or CD8(+) T cells in the protective immune response of BALB/c mice to Neospora caninum infection. Vet. Parasitol. 90(3):183-91. Thompson, N. L., K. C. Flanders, J. M. Smith, L. R. Ellingsworth, A. R. Roberts and M. B. Sporn. (1989). Expression of transforming growth factor-β1 in specific cells and tissues of adult and neonatal mice. J. Cell. Biol. 108:661-669. Tissenbaum, H.A., J. Handon, M. Perregaux, P. Hotez, L. Guarente, and G. Ruvkun. (2000). A common muscarinic pathway for diapause recovery in the distantly related nematode species Caenorhabditis elegans and Ancylostoma caninum. Proc. Natl. Acad. Sci. USA. 97(1): 460-5. Traub, R. J., I. D. Robertson, P. Irwin, N. Mencke, and R. C. Thompson. (2004). Application of a species-specific PCR-RFLP to identify Ancylostoma eggs directly from canine faeces. Vet. Parasitol. 123(3-4):245-55. Veenstra van Nieuwenhoven, A. L., A. Bowman, H. Moes, M. J. Heineman, L. F. de Leij, J. Santema, and M. M. Faas. (2002). Cytokine production in natural killer cells and lymphocytes in pregnant women compared with women in the follicular phase of the ovarian cycle. Fertil. Steril. 77(5):1032-7. Wang, C. I. (1988). Parasitic diarrheas in China. Parasitol. Today. 4: 284-87. Watkins, W. E., and E. Pollitt. (1997). Stupidity or Worms: do intestinal worms impair mental performance? Psychological Bulletin 121: 171-191. Wegmann, T. G., H. Lin, L. Guilbert, and T. R. Mosmann. (1993). Bidirectional cytokine interactions in the maternal-fetal relationship: is successful pregnancy a TH2 phenomenon? Immunol. Today. 14(7):53-6. White CJ, C. Maxwell, and J. I. Gallin. (1986). Changes in the structural and functional properties of human eosinophils during experimental hookworm infection. J Infect Dis. 154(5):778-83. Williamson, A. L., P. J. Brindley, D. P. Knox, P. J. Hotez, and A. Loukas. (2003). Digestive proteases of blood-feeding nematodes. Trends Parasitol. 19(9):417-23. Wong, D. T., A. Elovic, K. Matossian, N. Nagura, J. McBride, M. Y. Chou, J. R.Gordon, T. H. Rand, S. J. Galli, and P. F. Weller. (1991). Eosinophils from patients with blood eosinophilia express transforming growth factor beta 1. Blood. 78(10):2702-7.

162

8. APPENDIX

For comparing variation in transcript levels of cytokine genes, Pair-Wise Fixed

Reallocation Randomization Test© in the REST-384 was used (Pfaffl, et al., 2002).The

mathematical model used to compute the relative expression ratio of a target gene relies on

its real-time PCR efficiencies (E) and the threshold cycle difference (∆Ct) of an unknown

sample versus a control

(∆Ct control - sample). The target gene expression is normalized to a reference gene (ref) in the

equation mentioned below:





upregulation or downregulation of transcripts between timepoints, early-lactation (timepoint

II) and mid-lactation (timepoint III) were compared to late-gestation (timepoint I) expression

of which was relative to control (UN). For fold upregulation or downregulation of TGF-β

isoforms and IGF-1 in IN, UB and IB groups, comparisons were also made relative to UN

controls during early-lactation (timepoint II) and mid-lactation (timepoint III) were compared

to late-gestation (timepoint I).

163

Figure legends Figure 8.1a IL-4 transcript levels in skeletal muscle of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1b IL-4 transcript levels in skeletal muscle of uninfected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1c IL-4 transcript levels in skeletal muscle of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1d IL-4 transcript levels in mammary gland of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1e IL-4 transcript levels in mammary gland of uninfected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1f IL-4 transcript levels in mammary gland of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1g IL-4 transcript levels in spleen of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1h IL-4 transcript levels in spleen of uninfected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early

164

lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.1i IL-4 transcript levels in spleen of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2a IFN-γ transcript levels in skeletal muscle of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2b IFN-γ transcript levels in skeletal muscle of uninfected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2c IFN-γ transcript levels in skeletal muscle of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from skeletal muscle collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2d IFN-γ transcript levels in mammary gland of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2e IFN-γ transcript levels in mammary gland of uninfected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2f IFN-γ transcript levels in mammary gland of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from mammary gland collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN.

165

Figure 8.2g IFN-γ transcript levels in spleen of infected unbred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2h IFN-γ transcript levels in spleen of uninfected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.2i IFN-γ transcript levels in spleen of infected bred mice. Mice were sacrificed at three different timepoints during pregnancy and lactation (I= late gestation, II= early lactation and III= mid-lactation). RNA was extracted from spleen collected from each timepoint and cDNA was synthesized for real-time PCR analyses. Comparisons were made relative to UN. Figure 8.3a IL-4 transcript levels in skeletal muscle of infected unbred mice. Comparisons were made relative to timepoint I. Figure 8.3b IL-4 transcript levels in skeletal muscle of uninfected bred mice. Comparisons were made relative to timepoint I. Figure 8.3c IL-4 transcript levels in mammary gland of uninfected bred mice. Comparisons were made relative to timepoint I. Figure 8.3d IL-4 transcript levels in spleen of infected unbred mice. Comparisons were made relative to timepoint I. Figure 8.3e IL-4 transcript levels in spleen muscle of uninfected bred mice. Comparisons were made relative to timepoint I. Figure 8.3f IL-4 transcript levels in spleen muscle of infected bred mice. Comparisons were made relative to timepoint I. Figure 8.4a IFN-γ transcript levels in skeletal muscle of infected unbred mice. Comparisons were made relative to timepoint I. Figure 8.4b IFN-γ transcript levels in skeletal muscle of uninfected bred mice. Comparisons were made relative to timepoint I. Figure 8.4c IFN-γ transcript levels in skeletal muscle of infected bred mice. Comparisons were made relative to timepoint I.

166

Figure 8.4d IFN-γ transcript levels in mammary gland of uninfected unbred mice. Comparisons were made relative to timepoint I. Figure 8.4e IFN-γ transcript levels in mammary gland of uninfected bred mice. Comparisons were made relative to timepoint I. Figure 8.4f IFN-γ transcript levels in mammary gland of infected bred mice. Comparisons were made relative to timepoint I. Figure 8.4g IFN-γ transcript levels in spleen of infected unbred mice. Comparisons were made relative to timepoint I. Figure 8.4h IFN-γ transcript levels in spleen of uninfected bred mice. Comparisons were made relative to timepoint I. Figure 8.4i IFN-γ transcript levels in spleen of infected bred mice. Comparisons were made relative to timepoint I.

167

0

10

20

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1a

-5

0

5

10

15

20

25

30

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1b

0102030405060708090

100

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1c

168

-10

0

10

20

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1d

-5

0

5

10

15

20

25

30

35

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1e

-20

-15

-10

-5

0

5

10

15

20

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1f

169

-10

0

10

20

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1g

-5

0

5

10

15

20

25

30

35

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.1h

-5

-4

-3

-2

-1

0

1

2

3

I II III

Fold

cha

nge

- TG

F-ββ ββ

1 m

RN

A

Figure 8.1i

170

-10

-5

0

5

10

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.2a

-10

-5

0

5

10

I II III

Fold

cha

nge-

IFN

- γγ γγ m

RN

A

Figure 8.2b

-5

0

5

10

15

20

25

Fold

cha

ge-

IFN

-γγ γγ m

RN

A

Figure 8.2c

171

-10

-5

0

5

10

15

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.2d

-10-505

101520253035

I II III

Fold

cha

nge-

IFN

- γγ γγ m

RN

A

Figure 8.2e

-10

0

10

20

30

I II III

Fold

cha

nge-

IFN

mR

NA

Figure 8.2f

172

-60

-50

-40

-30

-20

-10

0

10

I I I I I I

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.2g

-5

0

5

10

15

20

25

30

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.2h

-10

0

10

20

30

I II III

Fold

cha

nge

IFN

-γγ γγ m

RN

A

Figure 8.2i

173

-5-4-3-2-101234

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.3a

-4

-3

-2

-1

0

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.3b

0

1

2

3

4

5

6

7

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.3c

174

-10.0

-7.5

-5.0

-2.5

0.0

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.3d

0

10

20

30

40

50

60

70

I II III

Fold

cha

nge-

IL-4

mR

NA

Figure 8.3e

-10

-5

0

5

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.3f

175

-20

-10

0

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4a

-20

-10

0

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4b

-7.5

-5.0

-2.5

0.0

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4c

176

-10

-5

0

5

10

15

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4d

0

1

2

3

4

5

6

I II III

Fold

cha

nge-

IFN

- γγ γγ m

RN

A

Figure 8.4e

0.00.51.01.52.02.53.03.54.04.5

I II III

Fold

cha

nge-

IFN

- γγ γγ m

RN

A

Figure 8.4f

177

-20

-10

0

10

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4g

0.0

2.5

5.0

7.5

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4h

-10

-5

0

5

I II III

Fold

cha

nge-

IFN

-γγ γγ m

RN

A

Figure 8.4i

ABSTRACT TRIVEDI, SHWETA. Host cytokines and immune ...

Documents