Differential expression of U2AF35 in the arthritic joint of avian reovirus-infected chicks

www.elsevier.com/locate/vetimm

Veterinary Immunology and Immunopathology 114 (2006) 49–60

Differential expression of U2AF35 in the arthritic joint of avian

reovirus-infected chicks

Yi-Hsin Fan a, Chun-Da Lin a, Shiow-Her Chiou a,*, Kuan-Chih Chow b,*,Yung-Sheng Chi a, Long-Huw Lee c, Jui-Hung Shien c, Happy K. Shieh c

a Graduate Institute of Veterinary Microbiology, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 40227, Taiwan, ROCb Graduate Institute of Biomedical Sciences, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 40227, Taiwan, ROC

c Department of Veterinary Medicine, National Chung Hsing University, 250 Kuo Kuang Road, Taichung 40227, Taiwan, ROC

Received 21 April 2006; accepted 5 July 2006

Abstract

To identify cell types and genes that are differentially expressed during immunopathogenesis of avian reovirus (ARV)-induced

viral arthritis (VA), we inoculated arthrotropic strain S1133 of ARV into 1-day-old broilers, and examined tissue histology as well as

RNA expression at different days post-inoculation (PI). Using immunohistochemical staining, we detected many CD68 expressing

macrophages in and around the blood vessels of the arthritic joints. By RT-PCR, we found that expression of matrix

metalloproteinase-2 (MMP-2) and bone morphogenetic protein-2 (BMP-2) was induced earlier in footpads and hock joints of

ARV-infected chickens. By employing suppression subtractive hybridization (SSH) technique and RT-PCR, we further identified

that small subunit of U2 snRNP auxiliary factor (U2AF35 or U2AF1) mRNAwas differentially induced in the joint of ARV-infected

chickens. By in situ hybridization (ISH), mRNA signals of U2AF35 and BMP-2 were located in chondrocytes within/near the

epiphyseal plate and secondary center of ossification, and in epidermal cells and dermal fibroblast-like cells of arthritic joints. In

addition, U2AF35 mRNA was expressed in the inflammatory infiltrates of the bone marrow of ARV-infected arthritic joints, while

MMP-2 was mainly detected in chondrocytes. Interestingly, among U2AF35, MMP-2, and BMP-2 that were differentially expressed

in the joint of ARV-infected chickens, only U2AF35 induction correlated well with arthritic manifestation. Because U2AF35 may

assist in mRNA splicing of proinflammatory chemokines and cytokines, our results indicated that U2AF35 induction might play an

immunopathological role in ARV-induced arthritis. This study has first associated U2AF35 to viral arthritis.

# 2006 Elsevier B.V. All rights reserved.

Keywords: Avian reovirus (ARV); Bone morphogenetic protein-2 (BMP-2); Matrix metalloproteinase-2 (MMP-2); Small subunit of U2 snRNP

auxiliary factor (U2AF35 or U2AF1); Viral arthritis (VA)

1. Introduction

Avian reovirus (ARV) infections have been asso-

ciated with viral arthritis (VA) (also known as

tenosynovitis), enteric diseases and other pathological

* Corresponding authors. Tel.: +886 4 2285 1343;

fax: +886 4 2285 9270.

E-mail addresses: [email protected] (S.-H. Chiou),

[email protected] (K.-C. Chow).

0165-2427/$ – see front matter # 2006 Elsevier B.V. All rights reserved.

doi:10.1016/j.vetimm.2006.07.003

conditions such as myocarditis, and pericarditis in

chickens and turkeys (Clark et al., 1990; Robertson and

Wilcox, 1986; van der Heide et al., 1981; Walker et al.,

1972). Among these diseases, the causative role of ARV

has only been established in VA (Robertson and Wilcox,

1986). Young chicks are especially susceptible to ARV

infection, and resistance to infection increases with age

(Rosenberger and Olson, 1991).

The apparent lesions of ARV-induced VA include

gross swelling and inflammatory infiltration of the

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–6050

tibiotarsal and hock joints of infected chickens (Glass

et al., 1973; Kerr and Olson, 1969; Pertile et al., 1996).

In addition to VA lesions developed in the joints, ARV

can lead to a transient reduction of immune responses in

chickens (Montgomery et al., 1985; Neelima et al.,

2003; Pertile et al., 1995; Springer et al., 1983), which

may predispose chickens to secondary infections

(Kibenge et al., 1982). Interestingly, arthritic lesions

and temporal immunosuppression developed in ARV-

infected chickens resemble symptoms that are exhibited

in patients with rheumatoid arthritis (RA) (Harris, 1990;

Verwilghen et al., 1990). RA is a chronic inflammatory

joint disease mediated by activated macrophages, T

lymphocytes, and B lymphocytes. Macrophages, in

particular, play a pivotal role during RA pathogenesis

by expressing major histocompatibility complex

(MHC) class II molecules, proinflammatory cytokines,

and matrix degrading enzymes (Kinne et al., 2000). In

fact, Pertile et al. (1996) have studied in detail the

pattern of lymphocyte (CD4- and CD8-positive T

lymphocytes, B lymphocytes/plasma cells) infiltration

in ARV-induced VA and found it to be similar to that of

RA. Apart from the well-characterized lymphocytes,

they have also described the infiltration of Ia (class II

MHC antigen)-positive non-lymphocyte cell popula-

tions during acute phase arthritis (Pertile et al., 1996).

However, it remains to be determined whether these

non-lymphocyte cells are macrophages.

Studies in RA and other human arthritis have shown

that matrix metalloproteinases (MMPs) are major

catabolic enzymes participating in the destruction of

cartilage and bone matrix during arthritis pathogenesis

(Giannelli et al., 2004; Vincenti and Brinckerhoff,

2002). As a counterbalance, anabolic factors such as

bone morphogenetic proteins (BMPs) are induced to

repair damage in arthritis (Nakase et al., 2003).

However, it is still not clear whether MMPs and BMPs

are involved in ARV-induced VA as well.

In this study we employed immunohistochemistry,

suppression subtractive hybridization (SSH) technique

(Diatchenko et al., 1996), and reverse transcription-

polymerase chain reaction (RT-PCR) to recognize the

identity of the CD68-positive macrophage infiltrates

and to identify the gene(s) that may participate in

immunopathogenesis of ARV-induced viral arthritis.

2. Materials and methods

2.1. Virus and virus titration

The arthrotropic virulent strain S1133 of ARV (van der

Heide and Kalbac, 1975), originally obtained from

Vineland Laboratories (Vineland, NJ, USA), is a

challenge strain used for the vaccine evaluation. The

virus was propagated briefly in chicken embryo fibrob-

lasts (CEF). Virus titration was performed as described

(Robertson and Wilcox, 1984), and virus titer was

expressed as 50% tissue culture infectious dose (TCID50).

2.2. Experimental chickens

Sixty specific pathogen free (SPF) broiler chickens (1-

day-old) were used in this study. Each chicken was

inoculated either with 103 TCID50/50 ml of S1133 strain

ARV (ARV-infected group) or 50 ml of phosphate

buffered saline (PBS) (control group) into each footpad.

ARV-infected and PBS-inoculated chickens were housed

in separate isolation units with biologically filtered air. At

12 h, 1–4, 6, 8, 10, 12, 14, 16, 18, 21, 28, and 35 days after

inoculation, two chickens, respectively from the ARV-

infected group and the control group were sacrificed. The

experiment was carried out following the guideline of the

Institutional Animal Care and Usage Committee,

National Chung Hsing University, Taiwan.

2.3. Enzyme-linked immunosorbent assay (ELISA)

Serum was collected from each chicken before

sacrifice. A commercial ELISA kit (avian reovirus

antibody test kit, Idexx Laboratories, Westbrook, ME,

USA) was used to determine the titer of anti-ARV

antibody.

2.4. Histology and immunohistochemistry

Lung, heart, liver, spleen, kidney, bursa of Fabricius,

proventriculus, duodenum, jejunum, ileum, footpads,

and hock joints were dissected from each chicken and

cut into half. One half of the tissue was frozen in liquid

nitrogen and stored at �70 8C before RNA extraction.

The other half of the tissue was fixed with 10% PBS

buffered formalin, embedded in paraffin wax, sectioned

at 5 mm, and then mounted on glass slides. Two slides

for each tissue were stained with hematoxylin and eosin.

For immunohistochemistry, slides were incubated

with anti-CD68 monoclonal antibody (KP1) (Dako,

Carpenteria, CA, USA) by using the labeled strepta-

vidin–biotin (LSAB) method as described previously

(Chow et al., 2004; Peng et al., 1999).

2.5. RNA extraction

Frozen tissues of hock joints and footpads were

removed from�70 8C and ground under liquid nitrogen

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–60 51

in a mortar. Total RNA was extracted by using Trizol

reagent (Invitrogen, Rockville, MD, USA) and RNA

concentration was determined by spectrophotometry.

2.6. Reverse transcription-polymerase chain

reaction (RT-PCR)

RT-PCR was performed in a 50 ml reaction volume

containing RT-PCR buffer, dNTPs (0.2 mM each), RNA

template (30 ng of total RNA), sense primer (20 pmol),

antisense primer (20 pmol), Taq DNA polymerase

(1 U), and AMV reverse transcriptase (1.5 U). RT

reaction was carried out at 42 8C for 45 min. After a

denaturation step at 94 8C for 3 min, 35 PCR cycles

were performed (denaturation at 94 8C for 1 min,

primer annealing at 52 8C for 1 min, extension at 72 8Cfor 1 min), followed by a final extension at 72 8C for

10 min. Primer sequences are listed in Table 1. RT-PCR

products were resolved by agarose gel electrophoresis.

Representative RT-PCR product was sequenced to

confirm the identity of each gene.

2.7. Suppression subtractive hybridizaition (SSH)

Equal amounts of total RNA extracted respectively

from hock joints and footpads of 2, 3, 4, 6, 8, 10 days

post-inoculated chickens were pooled together. Mes-

senger RNA (mRNA) was isolated from the pooled

RNA using a Chemagic mRNA T kit (Chemagen,

Baesweiler, Germany). About 5 mg of mRNA isolated

from each group (ARV-infected or control group) of

chickens were annealed with oligo dT primer and

reverse transcribed into cDNA by using reagents from a

commercial kit (SuperScriptTM

plasmid system with

GatewayTM

technology for cDNA synthesis and cloning,

Table 1

Primer sequences used in RT-PCR

Gene Amplicon (bp) Sense or antisense Primer seque

S1 (ARV) 941 Sense 50-GATACTG

Antisense 50-CACCTTA

S4 (ARV) 290 Sense 50-GCAGCCA

Antisense 50-ACATGGT

MMP-2 381 Sense 50-GCACCGT

Antisense 50-TGGCGTG

BMP-2 512 Sense 50-GTTGGTG

Antisense 50-GTTCTGT

U2AF35 416 Sense 50-AGACGTG

Antisense 50-ACTGGCG

b-actin 295 Sense 50-GTACCAC

Antisense 50-CTCGTCT

Invitrogen). After Rsa I digestion, cDNAs were

subjected to SSH using a PCR-select cDNA subtraction

kit (BD Clontech, Palo Alto, CA, USA). Briefly, Rsa I

digested-cDNA of ARV-infected group was divided into

two populations, ligated to specific adaptors, and

hybridized separately with excessive Rsa I digested-

cDNA of the control group. The two populations were

finally hybridized with each other and underwent two

rounds of PCR amplification. The resultant PCR

products were subcloned into pCRII vector, using a

TA cloning kit (Invitrogen).

2.8. In situ hybridization (ISH)

Based on a random primed DNA labeling method,

RT-PCR amplified ARV S1, MMP-2, BMP-2, and

U2AF35 DNA fragments were incorporated respectively

with fluorescein isothiocyanate (FITC) using a Fluor-

escein-High Prime kit (Roche, Mannheim, Germany).

ISH was performed as described previously (Chiou

et al., 2005). Briefly, formalin-fixed sections were

dewaxed and incubated with 20 mg/ml of proteinase K

for 15 min at room temperature, and rinsed with

distilled deionized water. DNA probe in hybridization

buffer (200 ng/ml of FITC-labeled DNA fragment, 50%

formamide, 6� SSC, 0.25% non-fat dry milk) was

added onto sections, denatured at 70 8C for 10 min, and

hybridized at 37 8C in a moist chamber overnight.

Sections were washed twice at room temperature for

5 min in wash buffer containing 2� SSC and 0.25%

Brij35, and once at 42 8C for 10 min in wash buffer

containing 0.1� SSC and 0.25% Brij35. Hybridized

products were incubated with alkaline phosphatase-

conjugated anti-FITC polyclonal antibodies (Dako) and

visualized by chromogen NBT/BCIP (Sigma, St. Louis,

nce Reference

TCATTGACTTCGA-30 Shapouri et al. (1995)

GGTGTCGATGCC-30

TGGACAACACCG-30 Chiu and Lee (1997)

GTGCGTAGTTGATTAGATCCGAG-30

CACCAATCATTAA-30 Aimes et al. (1994)

ACATCACTCCAG-30

CATCATAATGCAAG-30 Francis et al. (1994)

CGGCACACAGCAA-30

AGTTGTACGGGCG-30 Pacheco et al. (2004)

CTCCTCCAATACA-30

AATGTACCCTGGCA-30 Kost et al. (1983)

TGTTTTATGCGCA-30

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–6052

Fig. 1. The titers of anti-ARV antibody at different days post-inocu-

lation (PI). Each chicken was inoculated with either S1133 strain ARV

(ARV-infected group, &) or phosphate buffered saline (control group,

*) into both footpads. Anti-ARV antibody, as determined by ELISA,

was first detected at day 6 PI. The titer of antibody increased steadily

after day 8 PI and finally reached 10854 (13.4 log2) at day 35 PI in

ARV-infected chickens. Anti-ARV antibody remained undetectable in

the control chickens throughout the experimental period. Each data

point represents an average result obtained from two chickens.

MO, USA). Positive staining (purple blue precipitate)

was identified by a light microscope.

3. Results

3.1. Virus inoculation and clinical manifestation

A total of 60 SPF broiler chickens (1-day-old) were

inoculated either with S1133 strain ARV (30 for the

ARV-infected group) or PBS (30 for the control group)

into each footpad. At 12 h, 1–4, 6, 8, 10, 12, 14, 16, 18,

21, 28, and 35 days after inoculation, two chickens

respectively from the ARV-infected group and the

control group were sacrificed and analyzed. ARV-

infected chickens manifested primarily with gross

swelling in the footpad (inoculation site) from day 3

to 6 post-inoculation (PI), and in the hock joint from day

4 to 8 PI. Between days 4 and 6 PI, most ARV-infected

chickens showed severe lameness. The infected chick-

ens gradually recovered macroscopically from day 8 to

12 PI. In contrast, chickens inoculated with PBS (the

control group) remained healthy throughout the

experimental period (to day 35 PI).

3.2. Humoral immune response in ARV-infected

chickens

To measure host humoral immune response, serum

was collected from each chicken, and titer of antibody

to ARV was determined by ELISA. Each data point

represents an average antibody titer of two chickens.

Anti-ARV antibody was detected at about 6 days after

ARV infection. The titer of antibody increased steadily

from day 8 to 35 PI. At 35th day (5 weeks) PI, anti-ARV

antibody reached an average titer of 10,854 (13.4 log2)

in infected chickens (Fig. 1). In contrast, anti-ARV

antibody was undetectable in the control chickens

throughout the experimental period (Fig. 1).

3.3. Histology, and immunohistochemistry

Histological staining revealed inflammatory cells in

synovium, dermis, bone marrow, and blood vessels of

footpads and hock joints of ARV-infected chickens

between day 1 and 8 PI. At day 1 PI, endothelial lining

of blood vessel in the secondary center of ossification

was seriously destructed, and many infiltrates leaked

out from blood vessel (Fig. 2B and D). At day 6 PI,

macrophage-like infiltrates were observed near the

secondary center of ossification in ARV-infected joints

(Fig. 2F). Using monoclonal antibody KP1 that is

specific for CD68, we performed immunohistochem-

istry to characterize these inflammatory infiltrates.

Based on the strong immunostaining of CD68 as well as

the macrophage-like morphology, we confirmed that

many of the inflammatory infiltrates were CD68-

positive macrophages. Fig. 3 showed the CD68 staining

on the footpad section at day 4 PI. CD68-expresssing

macrophages were abundant in blood vessel of the

ARV-infected chicken during arthritic manifestation

(Fig. 3B).

3.4. Persistent presence of ARV in the joints of

ARV-infected chickens

Using primers specific for viral S4 gene, we

performed RT-PCR to analyze the tissue distribution

of ARV S1133 strain. The expected RT-PCR product of

S4 gene region was 290 bp. At 12 h PI, S4 gene of the

virus was first detected in hock joints, footpads, liver,

and spleen. At day 1 PI, the virus continued to spread in

lung, heart, kidney, and proventriculus. It became

widespread in all tissues tested, including hock joints,

footpads, liver, spleen, lung, heart, kidney, proventri-

culus, bursa of fabricius, duodenum, jejunum, and

ileum, from day 2 to 6 PI (Fig. 4). However, the virus

gradually diminished in most tissues except for in

footpads and hock joints from day 8 to 16 PI, and was

exclusively confined to footpads and hock joints after

day 18 PI (Fig. 4). The RT-PCR product was sequenced

and the identity of S4 gene was confirmed. ARV signal

was undetectable in the control chickens (data not

shown).

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–60 53

Fig. 2. Histological examinations of chicken hock joints at day 1 and 6 PI. (A) In the control chicken, blood vessels in the secondary center of

ossification of the hock joint were well developed at day 1 PI (2-day-old) (original magnification 100�). (B) In the ARV-infected chicken, the

endothelial linings of the blood vessels in the secondary center of ossification of the hock joint were seriously destructed at day 1 PI (original

magnification 100�). (C) Higher magnified view of area highlighted in A (original magnification 200�). (D) Higher magnified view of area

highlighted in B. Many inflammatory infiltrates leaked out from the blood vessel (original magnification 200�). (E) Junction between chondrocytes

and the secondary center of ossification of the hock joint from a control chicken at day 6 PI (original magnification 200�). (F) Many inflammatory

cells (denoted by,5) were in the blood vessel. Macrophage-like infiltrates (indicated by, ) were observed near the secondary center of ossification

in the ARV-infected joint at day 6 PI (original magnification 200�).

3.5. Early induction of MMP-2, and BMP-2 in the

joints of ARV-infected chickens

Among few avian MMPs and BMPs that are

identified, chicken MMP-2 and BMP-2 are well

characterized with the protein coding sequence avail-

able (Aimes et al., 1994; Francis et al., 1994). Both

MMP-2 and BMP-2 may serve as entries for us to

explore molecular mechanism of ARV-induced arthritis

in chickens. To investigate whether MMP-2 may be

involved in ARV-induced arthritis, we performed RT-

PCR and analyzed the gene expression of chicken

MMP-2. The expected RT-PCR product of MMP-2 was

381 bp. Chicken b-actin mRNA, with a RT-PCR

product of 295 bp, was measured in parallel. Twelve

hours PI, cellular MMP-2 mRNA was induced in

footpads and hock joints of ARV-infected chickens

(1.5-day old chicks). As a comparison, MMP-2 was

expressed at 1 day PI in the control chickens (2-day old)

(Fig. 5A).

Likewise, RT-PCR was employed to determine the

differential gene expression of BMP-2 in ARV-

infected chickens. The expected RT-PCR product of

BMP-2 was 512 bp. Twelve hours PI, BMP-2 was

induced in footpads and hock joints of the infected

chickens (1.5-day old). On the other hand, in the

control group, BMP-2 was expressed 2 days PI (3-day

old) (Fig. 5B).

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–6054

Fig. 3. Immunohistochemical staining of CD68 in chicken footpads at day 4 PI. (A) No CD68 signal was detected in the footpad section of the

control chicken. (B) Many CD68-enriched macrophages were detected in the blood vessel of the footpad of the ARV-infected chicken. (C) Human

peripheral monocyte, a precursor of macrophage, was stained as a positive control.

For both MMP-2 and BMP-2 expressions, similar

results were observed in left and right footpads and hock

joints. Results presented in this study were duplicated in

two chickens.

3.6. Searching for genes that are differentially

expressed in the arthritic joints of ARV-infected

chickens

SSH technique (Diatchenko et al., 1996), which has

been applied successfully to identify differentially

Fig. 4. RT-PCR detection of ARV S4 gene expression. ARV S4 gene region

after footpad inoculation of 1-day-old broiler chicks with ARV S1133 strain. S

loaded from left to right in the consecutive order. The virus was first detected

the virus was widespread in all tissues tested. After that, the virus gradually di

hock joints after day 18 PI.

regulated genes in RA and osteoarthritis (OA) (Justen

et al., 2000), was employed in this study to explore any

cellular genes that may be involved in ARV-induced

arthritis. Each group (the ARV-infected or control

group) of cDNA was transcribed from pooled mRNAs,

containing total transcripts of footpads and hock joints

from 2–4, 6, 8, to 10 days post-inoculated chickens. By

SSH, we obtained PCR products that may reflect

upregulated genes of the arthritic joints in ARV-infected

chickens. The resultant PCR products were subcloned

into pCRII vector and sequenced. We identified two

(290 bp) was detected by RT-PCR in various tissues at different times

amples of 12 h, day 1–4, 6, 8, 10, 12, 14, 16, 18, 21, 28, and 35 PI were

at 12 h PI in hock joints, footpads, liver, and spleen. From day 2 to 6 PI,

minished in most tissues, and was exclusively confined to footpads and

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–60 55

Fig. 5. Differential gene expression of MMP-2 and BMP-2 in ARV-infected chickens. Expressions of MMP-2 (A), and BMP-2 (B) in footpad and

hock joint of each chicken were determined by RT-PCR. The expected RT-PCR products of MMP-2 and BMP-2 were 381 and 512 bp, respectively.

Samples collected at 12 h, day 1–4, 6, 8, 10, 12, 14, 16, 18, 21, 28, and 35 PI were loaded from left to right in the consecutive order. MMP-2 mRNA

was detected at 12 h PI in the ARV-infected chickens, and at day 1 PI in the control chickens. BMP-2 mRNA was detected at 12 h PI in the ARV-

infected chickens, and at day 2 PI in the control chickens, respectively. Chicken b-actin mRNA was measured as a mRNA control. Representative

data of b-actin expression in the hock joint of the control chickens are shown.

gene sequences including chicken small 35-kDa subunit

of the U2 snRNP auxiliary factor (U2AF35 or U2AF1)

(Pacheco et al., 2004) and ferritin H chain protein

(Stevens et al., 1987). By RT-PCR, we further verified

the differential expression of U2AF35 in the joint tissues

of ARV-infected chickens. The expected RT-PCR

product of U2AF35 was 415 bp. U2AF35 mRNA was

induced in the arthritic footpads from 12 h to day 6 PI,

and in the arthritic hock joints from day 1 to 8 PI

(Fig. 6). Ferritin H chain mRNA, on the other hand, was

constitutively expressed in the joints of ARV-infected

and control chickens (data not shown).

Fig. 6. Induction of U2AF35 expression in ARV-infected chickens. Different

RT-PCR product of U2AF35 was 415 bp. Samples collected at 12 h, day 1–4,

order. U2AF35 mRNAwas not expressed in the control chickens throughout th

from 12 h to day 6 PI, and the in hock joints from day 1 to 8 PI in the ARV i

hock joint are shown.

3.7. Distribution of ARV, MMP-2, BMP-2, and

U2AF35 mRNAs in the joints of ARV-infected

chickens

ISH was performed to determine ARV distribution in

the joints of ARV-infected chickens. Using S1 gene of

ARVas a probe, we localized the viral expression in the

epidermal cells, fibroblast-like cells in the dermis, and

chondrocytes within/near the epiphyseal plate and near

the secondary center of ossification of ARV-infected

joints (Fig. 7). Viral S1 gene was first detected in

footpads at 12 h PI, later in hock joints at day 1 PI, and

ial gene expression of U2AF35 was verified by RT-PCR. The expected

6, 8, 10, 12, and 14 PI were loaded from left to right in the consecutive

e experimental period (A). U2AF35 mRNAwas induced in the footpads

nfected chickens (B). Representative data of b-actin expression in the

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–6056

Fig. 7. Detection of ARV expression in chicken hock joints by in situ hybridization (ISH). Using ARV S1 DNA fragment as a probe, the viral

expression in the chicken hock joint was determined by ISH. (B) Viral expression was localized in epidermal cells and fibroblast-like cells in the

dermis of the ARV-infected chicken at day 3 PI. (D) Viral expression was localized in chondrocytes of the hock joint in the ARV-infected chicken at

day 3 PI. (A and C) ISH results from the control chicken are shown in parallel.

remained detectable to day 12 PI in both footpads and

hock joints. However, expression levels of viral S1 gene

gradually decreased from day 8, and the signal was

undetectable by ISH after day 14 PI.

Using RT-PCR products of MMP-2, BMP-2, and

U2AF35 as probes, cells expressing respective genes

were localized by ISH. MMP-2 mRNA was mainly

detected in chondrocytes within/near the epiphyseal

plate and near the secondary center of ossification

(Fig. 8B). Gene expressions of BMP-2 (Fig. 8D and F)

and U2AF35 (Fig. 9B and D) were detected in the

chondrocytes around the epiphyseal plate and second-

ary center of ossification, in epidermal cells, and in

dermal fibroblast-like cells of arthritic joints. In

addition, U2AF35 mRNA was expressed in some

inflammatory cells distributed in the bone marrow of

arthritic joints (Fig. 9F). The expression periods of

MMP-2, BMP-2 and U2AF35 observed by ISH (data not

shown) were equivalent to that detected by RT-PCR

(Figs. 5 and 6).

4. Discussion

In this study, we employed SSH technique and RT-

PCR to identify the upregulation of U2AF35 mRNA in

the footpads and hock joints of arthrotropic ARV-

infected chickens. Our RT-PCR results illustrated that,

in contrast to the continuous presence of ARV in the

joint throughout the experimental period (Fig. 4),

U2AF35 mRNA was only expressed during arthritic

manifestation period (Fig. 6) when tissues were

infiltrated with inflammatory cells. U2AF35 mRNA,

as determined by ISH, coincided well with ARV signal

in epidermal cells, dermal fibroblast-like cells, and

chondrocytes around the epiphyseal plate and second-

ary center of ossification. U2AF35 is the small 35-kDa

subunit (also known as U2AF1) of U2AF splicing

factor. It interacts with U2AF65, the large 65-kDa

subunit (also known as U2AF2) of U2AF, and binds to

the AG dinucleotide at the 30 splice site (Guth et al.,

2001) of pre-mRNAs. While binding of U2AF65 to the

pyrimidine (Py) tract is a prerequisite for assisting U2

snRNP to catalyze splicing reaction, U2AF35 is only

essential for splicing of a subset of AG-dependent pre-

mRNAs (Guth et al., 2001; Zamore and Green, 1991).

Remarkably, a recent study on Shigella infection has

revealed the involvement of U2AF35 in the expression

of some proinflammatory chemokines and cytokines,

including interleukin-8 (IL-8), RANTES (regulated on

activation, normal T cell expressed and secreted),

granulocyte-macrophage colony-stimulating factor

(GM-CSF), and interleukin-1b (IL-1b) (Okuda et al.,

2005). Based on this activity, it is feasible that U2AF35

induced in ARV-infected cells (epidermal cells, dermal

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–60 57

Fig. 8. ISH examination of MMP-2 and BMP-2 gene expressions in chicken hock joints. Using MMP-2 and BMP-2 DNA fragments as probes, cells

expressing MMP-2 and BMP-2 were determined by ISH, respectively. (B) At 12 h PI, MMP-2 mRNA was detected by ISH in chondrocytes of the

epiphyseal plate in the hock joint of the ARV-infected chicken. (D) At day 1 PI, BMP-2 mRNA was expressed in the chondrocytes around the

secondary center of ossification in the hock joint of the ARV-infected chicken. (F) BMP-2 mRNA was detected in the epidermal cells of the ARV-

infected chicken at day 1 PI. (A, C, and E) ISH results from the control chicken were shown in parallel.

fibroblast-like cells, and chondrocytes) might facilitate

proinflammatory chemokines and cytokines production,

which in turn could attract immune cells to migrate or

infiltrate toward the infected tissues. These inflamma-

tory infiltrates, like those appeared in bone marrow of

the arthritic joints (Fig. 9F), may express U2AF35 and

result in more chemokines and cytokines production. In

fact, enhanced expressions of chemokines have been

observed in chondrocytes of OA and RA (Borzi et al.,

1999) human patients, or in synovial fluid of dogs with

OA (de Bruin et al., 2005). It would be interesting to

know whether U2AF35 may be involved in the arthritic

development of OA and RA, like the ARV-induced

arthritis reported here. If so, U2AF35 may serve as a

potential target for anti-arthritic drug development.

Previous study by Pertile et al. (1996) has shown that

subtypes and activation levels of infiltrated lymphocytes

in ARV-induced VA are similar to that observed in RA. In

addition, they have noticed the infiltration of class II

MHC antigen-positive non-lymphocytes during acute

phase arthritis (Pertile et al., 1996). In this study, we used

anti-CD68 monoclonal antibody KP1 to investigate

whether macrophages, which are known to express

abundant class II MHC molecule, may be involved in

ARV-induced VA. Besides monocytes/macrophages,

CD68 expression has been detected (by KP1 stain) in

some primary granules of neutrophils (Saito et al., 1991)

as well as in fibroblasts (Kunisch et al., 2004).

Nonetheless, based on results from Pertile et al.

(1996), we could rule out cells that do not express class

II MHC antigen. Therefore, in this study, the strong

immunostaining of CD68 as well as macrophage-like

morphology of the infiltrates suggest that CD68-enriched

macrophages are actively involved in the ARV-mediated

arthritic manifestation (Fig. 3B), which is also the major

feature detected in RA (Kinne et al., 2000). Taken

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–6058

Fig. 9. ISH examination of U2AF35 gene expression in chicken hock joints. Using U2AF35 DNA fragment as a probe, cells expressing U2AF35 were

localized by ISH. U2AF35 mRNAwas localized in chondrocytes around the epiphyseal plate at day 1 PI (B), in epidermal cells at day 2 PI (D), and in

bone marrow at day 2 PI (F) in the hock joint of the ARV-infected chicken. ISH results from the control chicken were shown in parallel in (A, C and

E), respectively.

together, ARV-induced VA and RA indeed share a similar

pattern of inflammatory infiltration. It is worth pursuing

in the future to investigate whether the two diseases may

at certain point have similar pathogenesis.

By using the highly sensitive RT-PCR, we demon-

strated that ARV S1133 was widely distributed in all

tissues from day 2 to 6 PI, gradually diminished in most

non-joint tissues from day 8 PI, and became exclusively

confined to footpads and hock joints after day 18 PI. In

contrast to the persistent presence of ARV in the joints,

viral expression patterns in some non-joint tissues (e.g.,

heart) were rather inconsistent among infected chick-

ens. Our RT-PCR results corroborated those of earlier

immunohistochemical studies (Jones and Kibenge,

1984; Jones et al., 1989; al Afaleq and Jones, 1990),

in that the arthrotropic ARV mainly persists in joint

tissues, despite of its ability to spread to many non-joint

tissues in a small portion of infected chickens during

early stage of infection. Our ISH study, nonetheless,

revealed decreasing viral expressions in the joints after

day 8 PI (data not shown). It is likely that, as the titers of

anti-ARV antibody increased steadily after day 8 PI

(Fig. 1), ARV was gradually cleared from non-joint

tissues and was reduced to a minimum level in the joint

eventually. This may explain our histological observa-

tion that cellular immune response in the joint, as

reflected by the presence of inflammatory infiltrates,

subsided after day 8 PI. However, the continuous low-

level expression of ARV in the joints may provide

sufficient antigen to elicit anti-ARV antibody through-

out the experimental period (Fig. 1).

Interestingly, our results showed that ARV, MMP-2,

and BMP-2 genes were coexpressed in chondrocytes of

ARV-infected joints, when chickens were 1.5 day old

(12 h PI). Expression of MMP-2 and BMP-2 in the

control group, however, was not detected until chickens

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–60 59

were 2-day old and 3-day old, respectively. Based on the

collagenase activity of chicken MMP-2 (Aimes and

Quigley, 1995), the earlier induction of MMP-2 is likely

to cause matrix damage around the viral-infected

chondrocytes of 1-day-old chicks. Conversely, chicken

BMP-2 is capable of promoting cartilage extracellular

matrix production (Roark and Greer, 1994). Similar to

human arthritis (Nakase et al., 2003), BMP-2 may be

induced here to counterbalance MMP activity and to

repair damage. MMP family members are enzymes that

decompose extracellular matrix (Nagase and Woessner,

1999). Among many MMP members that have been

detected in the RA synovium, MMP-1, MMP-3, and

MMP-13 are particularly important for joint destruction

(Hegemann et al., 2003; Vincenti and Brinckerhoff,

2002; Westhoff et al., 1999). On the other hand, BMPs

are members of the TGF-beta (TGF-b) superfamily.

Apart from playing important roles in almost all

processes of skeletal morphogenesis (Wan and Cao,

2005), some BMPs, such as BMP-2, BMP-6, and BMP-

7, have been identified in arthritic tissues as well (Gortz

et al., 2004; Nakase et al., 2003). These BMPs are

thought to be involved in cartilage and bone repair in

human arthritis (Nakase et al., 2003). Therefore, it is

worth pursuing in the future to investigate whether

chicken MMP-1, MMP-3, MMP-13, BMP-6, BMP-7, or

any chicken homologs of other MMP and BMP

members may participate in ARV-induced arthritis.

In conclusions, among U2AF35, MMP-2, and BMP-2

that were differentially expressed in the joint of ARV-

infected chickens, only U2AF35 expression correlated

well with the arthritic development. Because U2AF35

may assist in the mRNA splicing of some proinflam-

matory chemokines and cytokines, we propose that

U2AF35 induction may play an immunopathological

role in ARV-induced arthritis. This study has first

associated U2AF35 to viral arthritis.

Acknowledgments

We would like to thank Mr. Chun-Hao Lin for his

technical assistance. This study was supported in part by

grants from the National Science Council (NSC87-

2313-B-005-070) and Council of Agriculture, Execu-

tive Yuan (88-AS-2.2-BQ-01-11 and 89-AS-2.2-BQ-

62), Taiwan, ROC.

References

Aimes, R.T., French, D.L., Quigley, J.P., 1994. Cloning of a 72 kDa

matrix metalloproteinase (gelatinase) from chicken embryo fibro-

blasts using gene family PCR: expression of the gelatinase

increases upon malignant transformation. Biochem. J. 300,

729–736.

Aimes, R.T., Quigley, J.P., 1995. Matrix metalloproteinase-2 is an

interstitial collagenase. J. Biol. Chem. 270, 5872–5876.

al Afaleq, A.I., Jones, R.C., 1990. Localisation of avian reovirus in the

hock joints of chicks after entry through broken skin. Res. Vet. Sci.

48, 381–382.

Borzi, R.M., Mazzetti, I., Macor, S., Silvestri, T., Bassi, A., Cattini, L.,

Facchini, A., 1999. Flow cytometric analysis of intracellular

chemokines in chondrocytes in vivo: constitutive expression

and enhancement in osteoarthritis and rheumatoid arthritis. FEBS

Lett. 455, 238–242.

Chiou, S.-H., Chow, K.-C., Yang, C.-H., Chiang, S.-F., Lin, C.-H.,

2005. Discovery of Epstein-Barr virus (EBV)-encoded RNA

signal and EBV nuclear antigen leader protein DNA sequence

in pet dogs. J. Gen. Virol. 86, 899–905.

Chiu, C.J., Lee, L.H., 1997. Cloning and nucleotide sequencing of the

S4 genome of avian reovirus S1133. Arch. Virol. 142, 2515–2520.

Chow, K.-C., Hsiao, C.-H., Lin, T.-Y., Chen, C.-L., Chiou, S.-H., 2004.

Detection of severe acute respiratory syndrome-associated corona-

virus in pneumocytes of the lung. Am. J. Clin. Pathol. 121, 574–580.

Clark, F.D., Ni, Y., Collisson, E.W., 1990. Characterization of avian

reovirus strain-specific polymorphisms. Avian Dis. 34, 304–314.

de Bruin, T., de Rooster, H., van Bree, H., Cox, E., 2005. Interleukin-8

mRNA expression in synovial fluid of canine stifle joints with

osteoarthritis. Vet. Immunol. Immunopathol. 108, 387–397.

Diatchenko, L., Lau, Y.F., Campbell, A.P., Chenchik, A., Moqadam,

F., Huang, B., Lukyanov, S., Lukyanov, K., Gurskaya, N., Sver-

dlov, E.D., Siebert, P.D., 1996. Suppression subtractive hybridiza-

tion: a method for generating differentially regulated or tissue-

specific cDNA probes and libraries. Proc. Natl. Acad. Sci. U.S.A.

93, 6025–6030.

Francis, P.H., Richardson, M.K., Brickell, P.M., Tickle, C., 1994.

Bone morphogenetic proteins and a signalling pathway that con-

trols patterning in the developing chick limb. Development 120,

209–218.

Giannelli, G., Erriquez, R., Iannone, F., Marinosci, F., Lapadula, G.,

Antonaci, S., 2004. MMP-2, MMP-9, TIMP-1 and TIMP-2 levels

in patients with rheumatoid arthritis and psoriatic arthritis. Clin.

Exp. Rheumatol. 22, 335–338.

Glass, S.E., Naqi, S.A., Hall, C.F., Kerr, K.M., 1973. Isolation and

characterization of a virus associated with arthritis of chickens.

Avian Dis. 17, 415–424.

Gortz, B., Hayer, S., Redlich, K., Zwerina, J., Tohidast-Akrad, M.,

Tuerk, B., Hartmann, C., Kollias, G., Steiner, G., Smolen, J.S.,

Schett, G., 2004. Arthritis induces lymphocytic bone marrow

inflammation and endosteal bone formation. J. Bone Miner.

Res. 19, 990–998.

Guth, S., Tange, T.O., Kellenberger, E., Valcarcel, J., 2001. Dual

function for U2AF(35) in AG-dependent pre-mRNA splicing.

Mol. Cell. Biol. 21, 7673–7681.

Harris, E.D., 1990. Rheumatoid arthritis: pathophysiology and impli-

cations for therapy. N. Engl. J. Med. 322, 1277–1289.

Hegemann, N., Wondimu, A., Ullrich, K., Schmidt, M.F., 2003.

Synovial MMP-3 and TIMP-1 levels and their correlation with

cytokine expression in canine rheumatoid arthritis. Vet. Immunol.

Immunopathol. 91, 199–204.

Jones, R.C., Islam, M.R., Kelly, D.F., 1989. Early pathogenesis of

experimental reovirus infection in chickens. Avian Pathol. 18,

239–253.

Jones, R.C., Kibenge, F.S.B., 1984. Reovirus-induced tenosynovitis in

chickens: the effect of breed. Avian Pathol. 13, 511–528.

Y.-H Fan et al. / Veterinary Immunology and Immunopathology 114 (2006) 49–6060

Justen, H.P., Grunewald, E., Totzke, G., Gouni-Berthold, I., Sachinidis,

A., Wessinghage, D., Vetter, H., Schulze-Osthoff, K., Ko, Y., 2000.

Differential gene expression in synovium of rheumatoid arthritis and

osteoarthritis. Mol. Cell Biol. Res. Commun. 3, 165–172.

Kerr, K.M., Olson, N.O., 1969. Pathology of chickens experimentally

inoculated or contact-infected with an arthritis-producing virus.

Avian Dis. 13, 729–745.

Kibenge, F.S.B., Robertson, M.D., Wilcox, G.E., 1982. Staphylococ-

cus aureus isolated from poultry in Australia. II. Epidemiology of

strains associated with tenosynovitis. Vet. Microbiol. 7, 485–491.

Kinne, R.W., Brauer, R., Stuhlmuller, B., Palombo-Kinne, E.,

Burmester, G.R., 2000. Macrophages in rheumatoid arthritis.

Arthritis Res. 2, 189–202.

Kost, T.A., Theodorakis, N., Hughes, S.H., 1983. The nucleotide

sequence of the chick cytoplasmic beta-actin gene. Nucleic Acids

Res. 11, 8287–8301.

Kunisch, K., Fuhrmann, R., Roth, A., Winter, R., Lungershausen, W.,

Kinne, R.W., 2004. Macrophage specificity of three anti-CD68

monoclonal antibodies (KP1, EBM11, and PGM1) widely used for

immunohistochemistry and flow cytometry. Ann. Rheum. Dis. 63,

774–784.

Montgomery, R.D., Villegas, P., Dowe, D.L., Brown, J., 1985. Effect of

avian reovirus on lymphoid organ weights. Avian Dis. 29, 552–560.

Nagase, H., Woessner Jr., J.F., 1999. Matrix metalloproteinases. J.

Biol. Chem. 274, 21491–21494.

Nakase, T., Miyaji, T., Tomita, T., Kaneko, M., Kuriyama, K., Myoui,

A., Sugamoto, K., Ochi, T., Yoshikawa, H., 2003. Localization of

bone morphogenetic protein-2 in human osteoarthritic cartilage

and osteophyte. Osteoarthritis Cartilage 11, 278–284.

Neelima, S., Ram, G.C., Kataria, J.M., Goswami, T.K., 2003. Avian

reovirus induces an inhibitory effect on lymphoproliferation in

chickens. Vet. Res. Commun. 27, 73–85.

Okuda, J., Toyotome, T., Kataoka, N., Ohno, M., Abe, H., Shimura, Y.,

Seyedarabi, A., Pickersgill, R., Sasakawa, C., 2005. Shigella

effector IpaH9.8 binds to a splicing factor U2AF(35) to modulate

host immune responses. Biochem. Biophys. Res. Commun. 333,

531–539.

Pacheco, T.R., Gomes, A.Q., Barbosa-Morais, N.L., Benes, V.,

Ansorge, W., Wollerton, M., Smith, C.W., Valcarcel, J., Carmo-

Fonseca, M., 2004. Diversity of vertebrate splicing factor U2AF35:

identification of alternatively spliced U2AF1 mRNAs. J. Biol.

Chem. 279, 27039–27049.

Peng, C.T., Chow, K.C., Chang, W.C., Tsai, C.H., Lin, T.Y., Lin, S.S.,

Chiu, C.F., 1999. Expression of Fas ligand in Langerhans’ cell

histiocytosis: A case report of a boy with multisystem involve-

ment. Am. J. Hematol. 61, 256–261.

Pertile, T.L., Sharma, J.M., Walser, M.M., 1995. Reovirus infection in

chickens primes splenic adherent macrophages to produce nitric

oxide in response to T cell-produced factors. Cell. Immunol. 164,

207–216.

Pertile, T.L., Walser, M.M., Sharma, J.M., Shivers, J.L., 1996. Immu-

nohistochemical detection of lymphocyte subpopulations in the

tarsal joints of chickens with experimental viral arthritis. Vet.

Pathol. 33, 303–310.

Roark, E.F., Greer, K., 1994. Transforming growth factor-beta and

bone morphogenetic protein-2 act by distinct mechanisms to

promote chick limb cartilage differentiation in vitro. Dev. Dyn.

200, 103–116.

Robertson, M.D., Wilcox, G.E., 1984. Serological characteristics of

avian reoviruses of Australian origin. Avian Pathol. 13, 585–594.

Robertson, M.D., Wilcox, G.E., 1986. Avian reovirus. Vet. Bull. 56,

726–733.

Rosenberger, J.K., Olson, N.O., 1991. Reovirus infections. Dis. Poult.

9, 639–674.

Saito, N., Pulford, K.A., Breton-Gorius, J., Masse, J.M., Mason, D.Y.,

Cramer, E.M., 1991. Ultrastructural localization of the CD68

macrophage-associated antigen in human blood neutrophils and

monocytes. Am. J. Pathol. 139, 1053–1059.

Shapouri, M.R.S., Kane, M., Letarte, M., Bergeron, J., Arella, M.,

Silim, A., 1995. Cloning, sequencing, and expression of the S1

gene of avian reovirus. J. Gen. Virol. 76, 1515–1520.

Springer, W.T., Olson, N.O., Kerr, K.M., Fabacher, C.J., 1983.

Response of specific pathogen free chicks to concomitant infec-

tion of reovirus (WVU 2937) and infectious bursal disease virus.

Avian Dis. 27, 911–917.

Stevens, P.W., Dodgson, J.B., Engel, J.D., 1987. Structure and expres-

sion of the chicken ferritin H-subunit gene. Mol. Cell. Biol. 7,

1751–1758.

van der Heide, L., Kalbac, M., 1975. Infectious tenosynovitis (viral

arthritis): characterization of a Connecticut viral isolant as a

reovirus and evidence of viral egg transmission by reovirus-

infected broiler breeders. Avian Dis. 19, 683–688.

van der Heide, L., Lutticken, D., Horzinek, M., 1981. Isolation of

avian reovirus as a possible etiologic agent of osteoporosis

(‘‘brittle bone disease’’; ‘‘femoral head necrosis’’) in broiler

chickens. Avian Dis. 25, 847–856.

Verwilghen, J., Vertessen, S., Stevens, E.A., Dequeker, J., Ceuppens,

J.L., 1990. Depressed T-cell reactivity to recall antigens in rheu-

matoid arthritis. J. Clin. Immunol. 10, 90–98.

Vincenti, M.P., Brinckerhoff, C.E., 2002. Transcriptional regulation of

collagenase (MMP-1, MMP-13) genes in arthritis: integration of

complex signalling pathways for the recruitment of gene-specific

transcription factors. Arthritis Res. 4, 157–164.

Walker, E.R., Friedman, M.H., Olson, N.O., 1972. Electron micro-

scopic study of an avian reovirus that causes arthritis. J. Ultra-

struct. Res. 41, 67–79.

Wan, M., Cao, X., 2005. BMP signaling in skeletal development.

Biochem. Biophys. Res. Commun. 328, 651–657.

Westhoff, C.S., Freudiger, D., Petrow, P., Seyfert, C., Zacher, J.,

Kriegsmann, J., Pap, T., Gay, S., Stiehl, P., Gromnica-Ihle, E.,

Wernicke, D., 1999. Characterization of collagenase 3 (matrix

metalloproteinase 13) messenger RNA expression in the synovial

membrane and synovial fibroblasts of patients with rheumatoid

arthritis. Arthritis Rheum. 42, 1517–1527.

Zamore, P.D., Green, M.R., 1991. Biochemical characterization

of U2 snRNP auxiliary factor: an essential pre-mRNA splicing

factor with a novel intranuclear distribution. EMBO J. 10, 207–

214.