Conservation of Babesia bovis Small Heat Shock Protein ...Conservation of Babesia bovis Small Heat Shock Protein (Hsp20) among Strains and Deﬁnition of T Helper Cell Epitopes Recognized

INFECTION AND IMMUNITY, Feb. 2004, p. 1096–1106 Vol. 72, No. 20019-9567/04/$08.00�0 DOI: 10.1128/IAI.72.2.1096–1106.2004Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Conservation of Babesia bovis Small Heat Shock Protein (Hsp20) amongStrains and Definition of T Helper Cell Epitopes Recognized

by Cattle with Diverse Major HistocompatibilityComplex Class II Haplotypes

Junzo Norimine,1 Juan Mosqueda,1† Guy H. Palmer,1 Harris A. Lewin,2 and Wendy C. Brown1*Department of Veterinary Microbiology and Pathology, Washington State University, Pullman, Washington 99164,1

and Department of Animal Sciences, University of Illinois, Urbana-Champaign, Illinois 618012

Received 15 October 2003/Returned for modification 10 November 2003/Accepted 19 November 2003

Babesia bovis small heat shock protein (Hsp20) is recognized by CD4� T lymphocytes from cattle that haverecovered from infection and are immune to challenge. This candidate vaccine antigen is related to a protectiveantigen of Toxoplasma gondii, Hsp30/bag1, and both are members of the �-crystallin family of proteins that canserve as molecular chaperones. In the present study, immunofluorescence microscopy determined that Hsp20is expressed intracellularly in all merozoites. Importantly, Hsp20 is also expressed by tick larval stages,including sporozoites, so that natural tick-transmitted infection could boost a vaccine-induced response. Thepredicted amino acid sequence of Hsp20 from merozoites is completely conserved among different B. bovisstrains. To define the location of CD4� T-cell epitopes for inclusion in a multiepitope peptide or minigenevaccine construct, truncated recombinant Hsp20 proteins and overlapping peptides were tested for their abilityto stimulate T cells from immune cattle. Both amino-terminal (amino acids [aa] 1 to 105) and carboxy-terminal (aa 48 to 177) regions were immunogenic for the majority of cattle in the study, stimulating strongproliferation and IFN-� production. T-cell lines from all individuals with distinct DRB3 haplotypes respondedto aa 11 to 62 of Hsp20, which contained one or more immunodominant epitopes for each animal. One epitope,DEQTGLPIKS (aa 17 to 26), was identified by T-cell clones. The presence of strain-conserved T helper cellepitopes in aa 11 to 62 of the ubiquitously expressed Hsp20 that are presented by major histocompatibilitycomplex class II molecules represented broadly in the Holstein breed supports the inclusion of this region invaccine constructs to be tested in cattle.

Babesiosis in cattle is caused primarily by infection with theBoophilus tick-transmitted protozoan parasites Babesia bovisand B. bigemina. While B. bigemina infection is comparativelymild, infection with B. bovis is typically acute and characterizedby severe anemia, cachexia, cerebral dysfunction, and pulmo-nary edema (60). Cattle that do recover from B. bovis infection,either naturally or through chemotherapeutic intervention, areresistant to developing clinical disease on subsequent chal-lenge, indicating the feasibility of achieving protective immu-nity against this disease by vaccination. The mechanisms ofprotective immunity involve both innate and adaptive immuneresponses (12, 25, 36). Inhibition of the growth of B. bovis invitro is partially dependent on nitric oxide (NO) produced bygamma interferon (IFN-�)-activated macrophages (51, 52). Inaddition, IFN-� produced by Babesia-specific CD4� T helper(Th) cells is associated with enhanced synthesis of the opso-nizing immunoglobulin G2 (IgG2) antibody by B lymphocytes(8). Thus, identifying B. bovis antigens and their epitopes thatelicit IFN-�-producing effector/memory CD4� T-cell re-sponses in immune cattle is a rational strategy for developingan effective subunit or nucleic acid vaccine against babesiosis.

However, single-antigen vaccines have previously failed to pro-vide significant protection against challenge, even though highantibody titers and strong Th1 responses were elicited (28, 44).Effective vaccines against this complex organism will probablyrequire inclusion of multiple B- and T-lymphocyte epitopes rep-resenting different antigens and perhaps different parasite stages.

Leading candidate B. bovis antigens for vaccine develop-menthaveincludedrhoptry-associatedprotein1(RAP-1),mero-zoite surface antigen 1 (MSA-1), 12D3, and spherical-body pro-tein 1 (SBP1), formerly called Bv80 or Bb-1 (9, 12, 14, 16, 27, 28,56, 59). Mapping of CD4� T-cell epitopes on RAP-1, SBP-1,and 12D3 and characterization of the cytokine responses byantigen-specific Th cells were also carried out with the ultimategoal of constructing a multiepitope vaccine (14, 16, 43).

To identify additional immunostimulatory antigens, size-fractionated merozoite proteins were tested for stimulation ofTh-cell responses from B. bovis-immune cattle with differentgenetic backgrounds (7, 12, 53). A novel 20-kDa protein wasidentified that stimulated recall responses from immune cattlewith different major histocompatibility complex (MHC) class IIhaplotypes, and CD4� T-cell clones specific for this proteinwere also shown to express high levels of IFN-� (13). The B.bovis 20-kDa protein is a homologue of mammalian �-crystal-lin and is similar to other small heat shock proteins (Hsps)identified in plants (34); it is designated Hsp20. The smallHsp/�-crystallin protein family has significant homology amongspecies within the same genus but limited homology between

* Corresponding author. Mailing address: Department of Veteri-nary Microbiology and Pathology, Washington State University, Pull-man, WA 99164-7040. Phone: (509) 335-6067. Fax: (509) 335-8529.E-mail: [email protected].

† Present address: Centro Nacional de Investigaciones en Parasito-logia Veterinaria, INIFAP, Morelos, Mexico 62500.

1096

on Decem

ber 28, 2020 by guesthttp://iai.asm

.org/D

ownloaded from

http://iai.asm.org/

genera (34). The only known related protein in protozoal par-asites is Toxoplasma gondii Hsp30/bag1, which is selectivelyexpressed by the bradyzoite stage (5). In mice, immunizationwith Hsp30/bag1 enhanced protective immunity against chal-lenge (38, 42).

Although the biological function of small Hsps is not wellunderstood, some members of the �-crystallin family functionas molecular chaperones (29, 31). Consistent with a chapero-nin function, B. bovis Hsp20 was reportedly expressed in anapical pattern on 5 to 10 and 100% of merozoites in twoseparate studies (45, 48) and was present in both membrane-and organelle-enriched and cytosolic merozoite fractions (13).Furthermore, small Hsps are highly conserved within a genus(34), and at least one B-cell epitope and one T-cell epitopefrom B. bovis Hsp20 (Mexico strain) were shown to be con-served among geographically distant strains of B. bovis and theMexico strain of B. bigemina (13).

The primary goal of this study was to identify Hsp20 Th-cellepitopes recognized by immunized cattle with diverse MHCclass II haplotypes to enable the selection of epitopes with abroad spectrum of recognition in the Holstein population. Weshow that Hsp20 is completely conserved among otherwiseantigenically variant strains of B. bovis and that it containsmultiple conserved T-lymphocyte epitopes. In contrast, there islittle sequence identity to bovine �-crystallin. Furthermore, theBabesia antigen is expressed in both merozoite and tick larvalstages, so that immunization with Hsp20 or Hsp20 epitopescould stimulate immunity against tick-transmitted sporozoitesas well as blood-stage merozoites. These results identify Hsp20as a candidate protein for inclusion in a multiepitope vaccinefor babesiosis.

MATERIALS AND METHODS

B. bovis-infected Boophilus microplus tick larvae and induction of sporozoitedevelopment. Uninfected and B. bovis (Mexico strain)-infected Boophilus micro-plus tick larvae were obtained as described previously (40). To stimulate thedevelopment of B. bovis sporozoites, infected larvae were fed on an uninfectedcalf for 60 h by using skin patches (17). After this period, the larvae wereremoved and incubated at 37°C for an additional 12 h. Uninfected larvae wereobtained by using the same procedure with ticks from the same colony, exceptthat the adult ticks were fed on an uninfected calf. Temperature and humidityconditions were the same for uninfected and infected adult ticks and larvae.

Extraction of tick larval RNA and RT-PCR analysis. Infected Boophilus mi-croplus larvae were homogenized in a mortar, and total RNA was extracted usingTRIzol reagent (Invitrogen). RNA samples were treated with DNase by usingthe DNA-free kit (Ambion) and with the addition of RNase inhibitor (Roche).RNA was reverse transcribed and processed using a commercial reverse tran-scription-PCR (RT-PCR) kit (One Step SuperScript RT-PCR, GIBCO BRL).For RT and predenaturation, samples were incubated at 50°C for 30 min andthen at 94°C for 2 min. For PCR amplification, the forward and reverse primersfor Hsp20 were 5�-ATGTCGTGTATTATGAGGTG and 5�-GGCCTTGGCGTCAATCTGAA, respectively (13). RNA and DNA from B. bovis Mo7 Mexicostrain-infected erythrocytes were used as positive controls. To control for DNAcontamination, the PCR amplification was performed without reverse transcrip-tase. RNA extracted from uninfected larvae was used as a negative control. Asa control for the presence of tick RNA in the uninfected-larva samples, primersamplifying a 400-bp fragment from the Boophilus microplus Bm86 gene wereused (47). Amplicons were cloned and sequenced to confirm the identity of thetranscripts. The RT-PCR products were cloned into the pCR 4-TOPO plasmidvector by using the TOPO TA cloning kit (Invitrogen). Clones were sequencedin both directions by using the Prism Ready Reaction Dye Deoxy Terminatorcycle-sequencing kit and analyzed with the ABI Prism 373 genetic analyzer(Applied Biosystems). The sequences were compared with published sequencesby using Nucleotide BLAST (http://www.ncbi.nlm.nih.gov/blast/).

Immunoblot analysis of Hsp20 expression in merozoites and tick larvae.

Merozoites were obtained from in vitro cultures of the Mo7 Mexico clone ofB. bovis as described previously (39). Cultures containing free merozoites werecentrifuged twice at 400 � g for 10 min at 4°C to pellet the erythrocytes andintracellular parasites. The supernatant containing free merozoites was centri-fuged at 958 � g for 30 min, and the merozoites were resuspended in lysis bufferwith proteinase inhibitors. Approximately 300 infected, fed larvae were ground ina mortar containing 1.5 ml of lysis buffer with proteinase inhibitors. The mac-erate was centrifuged at 70 � g for 5 min, and the supernatant was collected.Merozoites (4 �l) or tick larva extract (40 �l) was boiled in sample buffercontaining 2% (wt/vol) sodium dodecyl sulfate and 2.5% (wt/vol) �-mercapto-ethanol and loaded on precast 10% acrylamide gels (Bio-Rad). Proteins weretransferred to a nitrocellulose membrane and incubated for 1 h with a 1:500dilution of B. bovis Hsp20-specific mouse antibody (13) in Tris-buffered saline(pH 7.6)–0.2% casein. Bound antibody was detected using an alkaline phos-phatase-labeled goat anti-mouse IgG (Applied Biosystems) at a 1:10,000 dilutionfollowed by enhanced chemiluminescence using the Western-Star System re-agents (Applied Biosystems). Uninfected Boophilus microplus larvae and unin-fected bovine erythrocytes were used as negative control antigens.

Analysis of Hsp20 expression in merozoites by immunofluorescence. Smearsof cultured merozoites were made using Probe-On slides (Fisher), air dried for2 h, and fixed in methanol for 5 min. The smears were rinsed in 125 mM Trisbuffer containing 0.05% Triton X-100 and were blocked at 37°C for 10 min withthis buffer containing 5% goat serum. B. bovis Hsp20-specific mouse serum (13)or rabbit anti-RAP-1 serum (54) was used at a final dilution of 1:100 andincubated at 37°C for 15 min. The bound primary antibodies were then labeledwith either fluorescein isothiocyanate (FITC)-conjugated goat anti-mouse IgG ortetramethylrhodamine isothiocyanate-conjugated goat anti-rabbit IgG (JacksonImmunoresearch Laboratories), and 4�,6-diamidino-2-phenylindole (DAPI;Sigma) was added to stain the nuclear DNA. The slides were then incubated for20 min at 37°C, blotted, and rinsed in distilled water 10 times between steps andthen 3 times with a final wash for 1 min. The smears were transferred tocoverslips in glycerol and DABCO (Sigma) and analyzed using confocal fluores-cence microscopy. Four images of two separate smears were taken using phasecontrast with sets of filters for fluorescein, rhodamine, and DAPI. As negativecontrols, merozoites were incubated with the two secondary antibodies only orwith a mouse antiserum against recombinant Anaplasma marginale major surfaceprotein 2 (MSP2) operon-associated gene 3 (OpAG3) (35) plus a rabbit anti-serum against recombinant A. marginale MSP1 (39).

B. bovis strains, antigen preparation, and peptides. For lymphocyte prolifer-ation assays, B. bovis and B. bigemina merozoite antigens were prepared from thecultured Mexico strain by homogenization of merozoites with a French pressurecell (SLM Instruments) and ultracentrifugation to obtain a fraction enriched incellular membranes and organelles (CM) (6). Membranes from uninfected redblood cells (URBC) were similarly prepared and used as a negative controlantigen for CM. Construction and expression of recombinant B. bovis Hsp20proteins were described previously (13). Briefly, full-length hsp20 cDNA encod-ing amino acids (aa) 1 to 177 or cDNAs encoding the N-terminal aa 1 to 105(Hsp20 NT) or C-terminal aa 48 to 177 (Hsp20 CT) were cloned into thepTrcHis2 expression vector by using the pTrcHis2 TOPO TA cloning kit (In-vitrogen). The recombinant protein was purified on a Ni2� column by usingProBond resin (Invitrogen) as described previously (43), dialyzed extensivelyagainst phosphate-buffered saline (pH 7.2), and quantified using a micro-bicin-choninic acid protein reagent kit (Pierce). Recombinant A. marginale salivarygland variant 1 (SGV1) or MSP5 proteins expressed in the same vector andpurified in the same manner were used as negative control proteins for prolif-eration assays. Peptides spanning B. bovis Hsp20 were synthesized by GerhardtMunske (Laboratory for Biotechnology and Bioanalysis I, Washington StateUniversity, Pullman). A peptide from A. marginale MSP2 (11) was used as anegative control peptide for proliferation assays. Recombinant proteins andpeptides are listed in Table 1.

Sequence analysis of the B. bovis hsp20 gene. The GenBank accession numberfor B. bovis hsp20 DNA from the Mexico strain is AF331455. PCR was used toamplify genomic hsp20 DNAs from the Australian strains G36 and vaccine strainT (kindly provided by Terry McElwain, Washington State University), Texasstrain BoT24, and Argentina strains S2P and R1A by using the same primers asdescribed for infected tick larvae. The amplicons were cloned into the pCR2.1TOPO TA cloning vector (Invitrogen) and sequenced. A BLAST search of theSwiss-Prot and GenBank databases was performed to determine sequence ho-mology to bovine �-crystallin.

Experimental cattle. Brahman-Angus cross cow C97 was infected with theMexico strain of B. bovis as described previously (6) and was the source ofHsp20-specific T-cell clones (13). The class II alleles of cow C97 obtained bycloning and sequencing (43) are DRB3*3001/*4501 and DQA1*0301/*0202. The

VOL. 72, 2004 B. BOVIS Hsp20 T-CELL EPITOPES 1097

on Decem


.org/D

ownloaded from

http://iai.asm.org/

DRB3 haplotype of cow C97 determined previously by restriction fragmentlength polymorphism (RFLP)-PCR analysis of exon 2 is 15/34. Four Holsteinsteer calves aged 5 to 6 months and weighing 200 to 250 kg were also typed forMHC class II DRB3 by RFLP-PCR analysis of exon 2 (58). The nomenclatureof the alleles is described on the bovine leukocyte antigen (BoLA) nomenclaturewebsite (http://www2.ri.bbsrc.ac.uk.bola/). This analysis revealed that the calveshave heterozygous and distinct DRB3 haplotypes, as shown in Table 2 (seebelow). The calves received four subcutaneous inoculations, at 3-week intervals,of 20 �g of recombinant Hsp20 emulsified in 1 ml of RIBI adjuvant (catalog no.R-730; RIBI Immunochem Research, Inc., Hamilton Mont., now Corixa, Seattle,Wash.) consisting of monophosphoryl lipid A, trehalose dimycolate, and cell wallskeleton. For the first immunization, 10 �g (0.5 ml) of human interleukin-12IL-12 (kindly provided by Genetics Institute, Cambridge, Mass.) was also inoc-ulated into the same injection site immediately after the recombinant Hsp20.Peripheral blood mononuclear cells (PBMC) and serum samples were collectedbefore each immunization for analysis of lymphocyte proliferation and antigen-specific antibody production.

T-lymphocyte cell lines and proliferation assays. Generation and culture ofHsp20-specific CD4� T-cell clones from B. bovis-infected immune cow C97 havebeen described previously (13). Short-term T-cell lines were established fromcow C97 and Hsp20-immunized calves by stimulating 4 � 106 PBMC in 24-wellplates (Costar) containing 1.5 ml of complete RPMI 1640 medium (6) with 10 �gof B. bovis CM antigen per ml for 1 week and resting them without antigen for1 week before use unless indicated otherwise (13). T-cell proliferation assayswere performed for 3 days in duplicate or triplicate wells of round-bottom96-well plates at 37°C in a humidified atmosphere of 5% CO2 in air. Then 3 �104 T cells and 2 � 105 irradiated PBMC as a source of autologous antigen-presenting cells (APC) per well were assayed with 0.1 to 25 �g of various antigensper ml in a total volume of 100 �l of complete RPMI 1640 medium. Proliferativeresponses were measured by radiolabeling cells with 0.25 �Ci of [3H]thymidine(Dupont New England Nuclear) for the last 18 h of culture.

To determine if antigen recognition by the CD4� T-cell clones was restrictedby DRB3 or DQ molecules, 2 � 105 autologous APC were incubated in 96-wellplates for 1 h with 4 �g of either monoclonal antibody (MAb) ILA-21 (anti-DR);(18) or MAb TH22A (anti-DQ) (20) per ml prior to the addition of T cells.Isotype-matched control IgG2a MAb Colis205 was used as a negative control. AllMAbs were obtained from Washington State University Monoclonal AntibodyCenter and purified by affinity chromatography to protein G by using an Equil-ibrate Hi Trap Protein G column (Pharmacia Biotech) as recommended by themanufacturer. Student’s one-tailed t test was used to determine the statisticalsignificance of levels of T-cell proliferation by using different antigens or MAb.

IFN-� ELISA. Supernatants (50 �l) were collected from triplicate wells ofproliferation assay mixtures and pooled before being labeled with [3H]thymidine.The level of IFN-� in the supernatants diluted 1:4 to 1:20 was determined byenzyme-linked immunosorbent assay (BOVIGAM; CSL Ltd., Parkville, Victoria,Australia) and compared with a standard curve obtained with a supernatant froma Mycobacterium bovis purified protein derivative-specific Th-lymphocyte clonethat contained 440 U of IFN-�/ml (previously determined by the neutralization

of vesicular stomatitis virus) (8). In the assay, 1 U corresponds to approximately1.7 ng of IFN-� (4).

Sequence analysis of T-cell receptor � and � chains. Sequencing of the T-cellreceptor � (TCR-� and TCR-� chains of Hsp20-specific Th-cell clones wasperformed as described previously (22). Briefly, Hsp20-specific CD4� T-cellclones were cultured for 7 days with antigen and APC as described previously(13), washed, and then cultured for 1 week with complete RPMI 1640 mediumcontaining 10% T-cell growth factor without antigen or APC to expand theCD4� T cells and to eliminate contamination with RNA derived from APC.Total RNA was extracted using the TRIzol reagent. The first-strand DNA wassynthesized from 1 to 3 �g of total RNA by using oligo(dT) and was precipitatedwith spermine. The pellet was resuspended with 68 �l of H2O, and a poly(dG)tail sequence was introduced in a 100-�l reaction mixture containing 1 mMdGTP (Perkin-Elmer Cetus), 30 U of deoxynucleotidyltransferase (TdT) (In-vitrogen), and 5� TdT buffer (Invitrogen) and incubated for 1 h at 37°C. Toamplify the TCR cDNAs, PCR was performed using a mixture of the ANpolyCprimer (5�-GCATGCGCGCGGCCGCGGAGGCCCCCCCCCCCCCC-3�) andthe AN primer (5�-GCATGCGCGCGGCCGCGGAGGCC-3�) at a ratio of 1:9as a forward primer and 5�-GAGCCGCAGCGTCATGAGCAGATTA-3� or5�-AGCACAGCGTACAGGGTGGCCTTCC-3�, for � and � chains, respec-tively, as reverse primers. The first five cycles of amplification involved annealingat 50°C, and the subsequent 30 cycles involved annealing at 55°C. The ampliconswere cloned into pCR2.1-TOPO TA cloning vector (Invitrogen) and sequenced.

RESULTS

Conservation of Hsp20 among strains of B. bovis. Previousstudies using proliferation assays with T-cell clones and immu-noblotting with either MAb 23/28.57 (45), which was subse-quently found to recognize Hsp20, or an Hsp20 peptide-spe-cific mouse serum revealed conservation of B. bovis Hsp20epitopes among strains from several different geographical lo-cations (13). To determine the level of amino acid sequenceconservation in the full-length proteins, hsp20 genomic DNAobtained from several parasite strains was sequenced. Thepredicted amino acid sequences of the encoded mature pro-teins were identical in parasite strains from Mexico, Texas,Argentina, and Australia (data not shown). When the B. bovishsp20 sequence was used to BLAST search the Swiss-Prot andGenBank databases, the best matched bovine protein was �A-crystallin (accession no. P02470). As described previously forthe �-crystallin family of proteins (13), there was a significantsequence alignment of Hsp20 with bovine �A-crystallin in theC-region, consisting of 25% identity of the 89 aa compared inthis region and the conserved GXLXXXXP motif (reference13 and data not shown). However, this amino acid identity wasscattered throughout the sequence and did not compriseblocks of identical sequence (data not shown).

TABLE 1. Recombinant B. bovis Hsp20 proteins andthe overlapping peptides

Protein orpeptide

Amino acidposition Amino acid sequence

Hsp20 1–177 AAK11624a

Hsp20 NT 1–105 AAK11624a

Hsp20 CT 48–177 AAK11624a

P1 11–40 DQEVIIDEQTGLPIKSHDYSEKPSVIYKPSP1-1 11–25 DQEVIIDEQTGLPIKSP1-2 23–37 PIKSHDYSEKPSVIYP2 26–55 SHDYSEKPSVIYKPSTTVPQNTLLEIPPPKP3 41–62 TTVPQNTLLEIPPPKELENPITP4 53–82 PPKELENPITFNPTVDTFFDADNNKLVLLMP5 73–102 ADNNKLVLLMELPGFSSTDINVECGWGELIP6 93–122 NVECGWGELIISGPRNKDELYEKFGNNLDIP7 113–142 YEKFGNNLDIHIRERKVGYFYRRFKLPNNAP8 133–162 YRRFKLPNNAIDKSISVGYSNGILDIRIECP9 153–177 NGILDIRIECSQFSEMRRVQIDAKAMSP2 P1 NAb MSAVSNRKLPLGGVLMALVAAVAPIHSLLA

a GenBank accession number.b NA, not applicable.

TABLE 2. Proliferative responses of CD4� T-cell lines fromB. bovis Hsp20-immunized calves to B. bovis and B. bigemina

AntigenProliferation (cpm) of short-term T-cell linesa from calf b:

50 (8/1) 56 (10/7) 57 (22/11) 58 (24/12)

Medium 1,576 � 210 1,666 � 867 463 � 128 113 � 10URBC 953 � 136 1,465 � 82 436 � 139 95 � 17B. bovis CM 62,952 � 658 32,215 � 1,195 11,942 � 1,094 24,922 � 1,024B. bigemina

CM60,068 � 1,285 27,339 � 1,683 10,987 � 758 22,854 � 1,073

a Short term T-cell lines were established by stimulation with B. bovis CM for1 week and resting for 1 week. Lymphocytes were cultured for 3 days with 25 �gof the indicated antigens per ml, radiolabeled, and counted. Results are ex-pressed as the mean counts per minute (cpm) of [3H]thymidine incorporation �1 SD. Significant proliferation is indicated in bold type.

b DRB3 haplotypes determined by RFLP-PCR are indicated for each calf.

1098 NORIMINE ET AL. INFECT. IMMUN.

on Decem


.org/D

ownloaded from

http://iai.asm.org/

Cellular localization of B. bovis Hsp20 in merozoites andexpression in larval tick stages. Conflicting results on the per-centage of merozoites that expressed Hsp20 (45, 48) led us tocompare the expression of this protein with that of surface-exposed proteins MSA-1 and RAP-1 by using a multicolorconfocal immunofluorescence technique (39). In contrast toMSA-1 and RAP-1, Hsp20 was never visualized on the surfaceof live merozoites (data not shown). However, in fixed mero-zoites, Hsp20 was expressed in 100% of the parasites, exhibit-ing diffuse fluorescence (Fig. 1A, panel b) that also colocalizedwith RAP-1 in the apical complex (panels a and d). In parasitesthat resembled the trophozoite stage, characterized by a largersize and a single intraerythrocytic organism, Hsp20 was visu-alized as one or two bands traversing the parasite (Fig. 1B,panels a and b). Labeling was not observed when parasiteswere incubated with a combination of a mouse antiserum againstA. marginale MSP2 OpAG3 plus rabbit antiserum against re-combinant A. marginale MSP1 (data not shown). Analysis ofHsp20 expression by RT-PCR (Fig. 2A) and immunoblotting(Fig. 2B) demonstrated that Hsp20 transcripts and protein areexpressed in tick larval stages as well as blood-stage merozo-ites. Similar studies performed with B. bigemina have also con-firmed the expression of Hsp20 in sporozoites (J. Mosqueda,unpublished observations).

CD4� T-lymphocyte responses in Hsp20-immunized calves.Previous studies demonstrated recognition of B. bovis Hsp20by cattle that had recovered from B. bovis infection (13). Toidentify B. bovis Hsp20 CD4� T-cell epitopes for cattle withdifferent genetic backgrounds, four calves with completely dif-ferent DRB3 haplotypes were immunized with recombinant B.bovis Hsp20 and short-term T-cell lines were tested with dif-

ferent antigens and overlapping peptides shown in Table 1. Allfour calves responded to Hsp20, B. bovis, and B. bigemina CMantigens, demonstrating recognition of epitopes on the nativeB. bovis Hsp20 that are conserved in B. bigemina (Table 2).Furthermore, all vaccinated calves responded to both the N-terminal and C-terminal regions, and animal 58 preferentiallyresponded to the C-terminal region (Fig. 3A to D). IFN-�production by antigen-stimulated T-cell lines mirrored that ofproliferation and reached high levels in three animals (Fig. 3Eto H). Using overlapping peptides spanning aa 11 to 177, wefound that each peptide stimulated the proliferation of T cellsfrom at least one animal and that peptide P2 was recognized byall four calves (Table 3). Peptide P1 also stimulated T-cell linesfrom three calves. Based on peptide recognition, at least fourT-cell epitopes were recognized by three calves and at least twoepitopes were recognized by the other calf. The P2 peptide hasat least one T-cell epitope recognized by all calves (Table 3).These results show that Hsp20 has multiple T-cell epitopes,and that the region of aa 11 to 62, represented by overlappingpeptides P1 to P3, stimulates cells from cattle with four differ-ent MHC class II haplotypes.

CD4� T-cell epitope mapping on B. bovis Hsp20 with Th-cellclones. It was also of interest to identify epitopes recognizedby Hsp20-specific T cells from infected animals. T-cell linesfrom B. bovis-infected and immune cow C97, that respondedstrongly to Hsp20 (13), were tested for lymphocyte prolifera-tion in response to recombinant full-length Hsp20, the N-terminal region of Hsp20, and the C-terminal region of Hsp20.In three assays, independently derived T-cell lines from thisanimal responded only to the N-terminal region of Hsp20 butnot to the C-terminal region, suggesting that the epitope(s)

FIG. 1. Localization of B. bovis Hsp20 protein in B. bovis merozoites by using immunofluorescence microscopy. Smears of B. bovis-infectederythrocytes were incubated with mouse anti-B. bovis Hsp20 peptide-specific antiserum and rabbit anti-RAP-1 serum. The bound primary anti-bodies were then labeled with secondary antibodies, a goat anti-mouse IgG conjugated with FITC (green) and a goat anti-rabbit IgG conjugatedwith rhodamine (red). To visualize the nuclei, DNA was labeled with DAPI (blue). (A) Merozoites; (B) trophozoites. Each image was recordedunder phase contrast with the following fluorescence channels: FITC (b), DAPI (c), rhodamine (d), and combined channels for all three stains (a).


on Decem


.org/D

ownloaded from

http://iai.asm.org/

resided between aa 1 and 47 (Table 4). To further define these,overlapping peptides P1 to P3 spanning aa 11 to 62 were testedwith the cell lines, and only peptide P1 (aa 11 to 40) stimulateda proliferative response (Table 4). Finally, peptides P1-1 andP1-2, which spanned peptide P1, were tested, and only peptideP1-1 (aa 11 to 25) was stimulatory for the cell lines (Table 4).To further define the epitope(s) recognized by animal C97,four CD4� T-cell clones specific for B. bovis Hsp20, previouslyshown to proliferate and produce IFN-� on stimulation withboth soluble cytosolic and CM antigens of B. bovis (13; W. C.Brown, unpublished observations), were also tested with re-combinant proteins and peptides. All T-cell clones respondedto B. bovis CM, Hsp20, and Hsp20 NT antigen (aa 1 to 105) butnot Hsp20 CT antigen (aa 48 to 177), suggesting that theepitope(s) is located between aa 1 and 47 (Fig. 4). Interest-ingly, clones 4C2 and 4C9 did not recognize B. bigemina CMwhereas clones 3B11 and 3G5 did recognize B. bigemina CM(representative clones are shown in Fig. 4A and 4B, respec-tively). Using synthetic peptides listed in Tables 1 and 5, wemapped the minimal epitopes recognized by these T-cellclones to a 10-amino acid sequence, DEQTGLPIKS. Deletionof either the N-terminal aspartic acid residue or the C-terminalserine residue abolished proliferation, confirming that this wasthe minimal sequence required for the T-cell response (Table5). To explain the differential response to B. bigemina by thetwo sets of T-cell clones, clones were tested with the homolo-

gous peptide derived from B. bigemina Hsp20, DEQTGLPVKN (BbgP1–4) (Table 5, experiment 2). As expected, clones4C2 and 4C9 responded only to B. bovis-derived peptide(BboP1–4), while clones 3B11 and 3G5 responded to both B.bovis- and bigemina-derived peptides. Two amino acid residuesat positions 8 and 10 of the peptide derived from B. bovis differfrom those at the same positions in the peptide from B. bigem-ina (boldface type in amino acid sequences). To determinewhich amino acid substitution was critical for loss of recogni-tion by clones 4C2 and 4C9, two other altered peptides, P1–4Nand P1–4V, were tested (Table 5, experiment 2). Clones 4C2and 4C9 were unable to recognize the peptide if the isoleucineresidue at position 8 was replaced by a valine residue, whereasthe response was still present when the serine residue at posi-tion 10 was replaced by an asparagine residue. As expected,clones 3B11 and 3G5 still responded to peptides with thesesubstitutions at either position 8 or position 10.

The finding that the two sets of T-cell clones recognize thesame epitope on B. bovis Hsp20 with different fine specificitiessuggested that either different MHC class II molecules present-ed the peptides to the respective pairs of clones or that TCRusage by the clones may be different. To attempt to determinewhether different MHC molecules are involved in presentingthe epitope, blocking studies using anti-DR and anti-DQ MAbswere performed. However, all clones were apparently re-stricted by the DQ molecule (Fig. 5 and data not shown).

TCR-� and TCR-� chain sequences used by CD4� T-cellclones. Because we could not determine whether different classII molecules presented the epitope to the different T-cellclones, the TCR-� and TCR-� chains from clones 4C2, 4C9,3B11, and 3G5 were sequenced. Clones 4C2 and 4C9 hadidentical sequences in both the � and � chains throughoutvariable (V), diversity (D), joining (J), and constant (C) re-gions, suggesting that these two clones were originally derivedfrom the same cell (data not shown). Similarly, clones 3B11and 3G5 also had completely identical TCR-� and TCR-�chains, indicating that these two clones also originated fromthe same cell (data not shown). However, the TCR-� andTCR-� chain sequences differed between the pairs of clones(clones 3G5 and 4C9 are shown in Fig. 6). Interestingly, inspite of their similar responses to the same minimal epitope,the TCR sequences of clones 3G5 and 4C9 did not have strongidentity in their V region of either the � or � chain (32.6 and34.8% identity, respectively). However, these clones use thesame J segment in the � chain, and so a large part of comple-mentarity-determining region 3 (CDR3) is identical, consistentwith CDR3 of the TCR-� chain playing a major role in deter-mining epitope specificity. In contrast, the �-chain CDRs didnot have appreciable sequence similarity and the CDR3s wereof different lengths, which may explain the different epitopefine specificities of the two sets of clones.

DISCUSSION

We have focused on discovering B. bovis antigens and theirepitopes that elicit a type 1 memory CD4� T-cell response inprotectedcattleaspotentialcandidatesforinclusioninamultiepi-tope vaccine construct (12). B. bovis Hsp20 was identified as animmunostimulatory antigen present in low-molecular-weightprotein fractions of merozoites that induced proliferation and

FIG. 2. Expression of B. bovis Hsp20 in Boophilus microplus ticklarvae. (A) B. bovis hsp20 transcripts were amplified by RT-PCR fromB. bovis merozoites (lane 2), B. bovis-infected larvae (lane 3), B. bovis-infected larvae without RT (lane 4), and uninfected larvae (lane 5).Larva-derived bm86 transcript was also amplified by RT-PCR fromthe same sample of uninfected larvae (lane 6). As a positive control,B. bovis hsp20, which contains a 119-bp intron, was amplified fromgenomic DNA by PCR (lane 1). (B) Immunoblot analysis of B. bovisHsp20 in the following antigens at approximately 10 �g of protein perlane: URBC (lane 1), B. bovis CM (lane 2), uninfected tick larvae (lane3), and B. bovis-infected tick larvae (lane 4). B. bovis Hsp20 wasdetected with a 1:200 dilution of mouse anti-B. bovis Hsp20 peptide-specific antiserum.


on Decem


.org/D

ownloaded from

http://iai.asm.org/

IFN-� production by B. bovis-specific T-cell lines and clones (7,13, 53). The present study has determined that Hsp20 is anintracellular protein that is diffusely expressed in the merozoiteand associated with the apical complex and has additionally

demonstrated the expression of B. bovis Hsp20 in tick stages.Furthermore, a homologue of Hsp20 is present in B. bigemina(13) and there is complete amino acid sequence conservationof Hsp20 among B. bovis strains. This would indicate the like-

FIG. 3. Proliferative responses and IFN-� production by T-cell lines from Hsp20-immunized calves. Short-term T-cell lines from calves 50 (Aand E), 56 (B and F), 57 (C and G), and 58 (D and H) were stimulated with 1 (white bars), 5 (striped bars), or 25 (black bars) �g of URBC, B.bovis CM, B. bigemina CM, recombinant B. bovis Hsp20 (full length, aa 1 to 177), Hsp20 NT (aa 1 to 105), and Hsp20 CT (aa 48 to 177) antigensper ml, and proliferation (A to D) and IFN-� production (E to H) were determined. Recombinant A. marginale SGV-1 protein expressed in thesame vector was used as a negative control antigen. Results are presented as the mean counts per minute � 1 standard deviation (SD) of triplicatecultures of T cells stimulated with antigen for 3 days. To measure IFN-� production, supernatants (50 �l) were collected from triplicate wells ofthe proliferation assay mixture before being labeled. IFN-� production was measured in pooled supernatants by ELISA, and the results areexpressed as mean nanograms per milliliter � 1 SD of duplicate wells in the ELISA.


on Decem


.org/D

ownloaded from

http://iai.asm.org/

lihood of inducing a cross-reactive immune response. The se-quence conservation and ubiquitous expression of Hsp20 inmerozoite and tick larval stages indicate the potential use ofthis protein in a vaccine construct that would stimulate immu-nity against sporozoites and merozoites and provided the ra-tionale for identification of CD4� T-cell epitopes recognizedby cattle with a diverse repertoire of MHC class II haplotypes.Although there was homology to bovine �A-crystallin, includ-ing conservation of an �-crystallin signature motif in the C-region, the amino acid identity was limited to amino acidsscattered throughout the sequence in this region and therewere no blocks of identical sequence. Therefore, it is unlikelythat immunization with Hsp20 would stimulate a cross-reactiveresponse to the bovine �A-crystallin.

The ubiquitous expression and intracellular localization ofHsp20 are consistent with its potential role as a molecular chap-erone. In some parasites that resembled trophozoites, Hsp20was visualized as one or two bands traversing the parasite.Although this staining pattern resembles that of microtubules(3), further analysis is required to confirm such an association.

The colocalization of Hsp20 with RAP-1 in the merozoiteapical complex is also consistent with a previously reportedpolar pattern of expression by indirect immunofluorescenceanalysis using Hsp20-specific MAb 23/28.57 (45). We also de-tected Hsp20 in CM and soluble cytosolic fractions by T-cellproliferation and immunoblotting assays (13; also see above).In other organisms, small Hsps are soluble and can function asmolecular chaperones in cases where association with otherproteins acts to stabilize them and permit correct folding (31).

CD4� T-cell lines from immunized calves responded withcomparable levels of proliferation when recombinant Hsp20and native B. bovis antigens were compared. This indicates thatprocessing and epitope presentation of the native and recom-binant antigens are similar and that immunization with recom-

FIG. 4. Proliferative response of B. bovis Hsp20-specific CD4� T-cell clones. Clones 4C9 (A) and 3G5 (B) were stimulated with 1 (stippledbars), 5 (striped bars), or 25 (black bars) �g of URBC, B. bovis CM, B. bigemina CM, recombinant B. bovis Hsp20 (full length, aa 1 to 177), Hsp20NT (aa 1 to 105), Hsp20 CT (aa 48 to 177), and control A. marginale MSP-5 antigens per ml. Results are presented as the mean counts per minute� 1 SD of duplicate cultures of T cells stimulated with antigen for 3 days.

TABLE 3. Proliferative responses of CD4� T-cell lines fromB. bovis Hsp20-immunized calves to Hsp20 peptides

PeptideProliferation (SI) by T-cell lines from calf a:

50 56 57 58

P1 103.1 1.7 52.9 4.1P2 108.0 50.6 57.9 8.7P3 117.0 1.1 78.8 27.1P4 1.4 4.9 1.6 2.4P5 2.3 1.9 14.7 1.3P6 104.2 1.4 1.5 5.1P7 65.3 1.7 2.8 145.7P8 6.5 1.7 4.4 2.4P9 1.5 1.4 2.7 68.1

a PBMC were stimulated for 1 week with B. bovis CM and rested for 1 weekbefore being tested. Results were obtained for the optimal response to 1 or 10 �gof each peptide per ml. SI indicates stimulation indices (mean counts per minuteof the response to Hsp20 peptide/mean of the response to control A. marginaleMSP2 peptide). Significant responses (SI � 3.0) are indicated in bold type. Theseresponses were significant when individual counts per minute were compared bythe one-tailed Student t test (P 0.005).

TABLE 4. Proliferative responses of CD4� T cells fromB. bovis-infected immune cow C97 to protein antigens

Antigen Proliferation (cpm)of T cellsa

Native proteinURBC........................................................................... 211 � 101B. bovis CM .................................................................79,270 � 1,184

Recombinant proteinHsp20............................................................................74,009 � 2,261Hsp20 NT.....................................................................85,673 � 10,169Hsp20 CT..................................................................... 476 � 119MSP5 ............................................................................ 163 � 23

PeptideP1 ..................................................................................82,691 � 665P2 .................................................................................. 177 � 31P3 .................................................................................. 224 � 41P1-1 ...............................................................................85,689 � 5,222P1-2 ............................................................................... 292 � 96P MSP2......................................................................... 100 � 7

a Cell lines were established by stimulation with B. bovis CM for 1 week,resting for 1 week, stimulation with Hsp20 for 1 week, and resting for 1 week.Lymphocytes were cultured for 3 days with 25 �g of native or recombinantprotein per ml and 10 �g of peptide per ml, radiolabeled, and counted. Resultsare expressed as the mean counts per minute (cpm) of [3H]thymidine incorpo-ration � 1 SD. Significant proliferation is indicated in bold type.


on Decem


.org/D

ownloaded from

http://iai.asm.org/

binant Hsp20 elicits a population of memory T cells that couldbe activated on infection. Moreover, high levels of IFN-� wereproduced ex vivo by three calves in response to these proteins,possibly as a result of using IL-12 as an adjuvant (57, 61).

The most immunostimulatory region of B. bovis Hsp20 forTh cells from the cattle used in this study is the N-terminalregion spanning aa 11 to 62. This region contains one or moreepitopes that stimulated T cells from all five cattle in the study,and at least two epitopes are present which stimulated T cellsfrom three calves. Based on the frequencies of DRB3 alleles inHolstein cattle (50; H. A. Lewin, unpublished observations)and their association with DQ alleles due to strong linkagephase disequilibrium (1), it is estimated that at least 70% ofHolstein or Friesian cattle in a given herd would have at leastone of the DRB3-DQ haplotypes evaluated in this study.Therefore, aa 11 to 62 of Hsp20 should be broadly recognizedby these breeds and may be useful in a multiepitope vaccineconstruct against B. bovis.

Four B. bovis Hsp20-specific CD4� T-cell clones from im-mune cow C97 recognized a minimal epitope in peptide P1 (aa11 to 40), consisting of 10 aa, DEQTGLPIKS (peptide BboP1–4). Interestingly, further analyses revealed that two of theseclones (3B11 and 3G5), but not clones 4C2 and 4C9, alsorecognized B. bigemina Hsp20 protein and its correspondingpeptide, DEQTGLPVKN (BbgP1–4), which has amino acidchanges at positions 8 and 10 (shown in bold type). The iso-leucine residue at position 8 is critical for recognition by the B.bovis Hsp20-specific clones 4C2 and 4C9. None of the clonesresponded to 9-mers consisting of either DEQTGLPIK orEQTGLPIKS. However, replacement of the serine residue atposition 10 of BboP1-4 with an asparagine residue was toler-ated by all four clones, indicating that the amino acid at posi-tion 10 does not determine antigen specificity per se but may

play a role in stabilizing peptide binding. Together, these re-sults indicate that the aspartic acid residue at position 1 is thefirst MHC class II anchor residue and the serine residue atposition 10 lies outside of the MHC anchor residues in thisepitope. Flanking as well as anchor residues are important forimmunogenicity, since in certain MHC haplotypes Th-cell epi-topes can be modulated by altering the amino acids flankingthe minimal core determinant (15, 41).

The TCRs of the two sets of T-cell clones were sequenced toclarify how the amino acid differences in the C terminus ofminimal epitopes BboP1-4 and BbgP1-4 were recognized.Clone pairs 4C2/4C9 and 3B11/3G5 use different V segmentswith little similarity in the CDR1 and CDR2 regions of eitherthe � or � chain. However, the finding that the main similarityin the TCR CDRs is within the �-chain CDR3 contributed bythe J region suggests that this region plays a major role indetermining the shared specificity against the conserved Nterminus of the minimal P1-4 epitopes in B. bovis and bigeminaHsp20. In support of this, analysis of the crystal structure of theTCR-peptide-MHC class II complex has demonstrated thatrecognition of the N-terminal part of an antigenic peptide isdominated by the TCR V� domain (49). On the other hand,sequence differences in the other regions of the TCRs, spe-

FIG. 5. MHC DQ restriction of Hsp20-specific CD4� T-cell clones.T-cell clones 4C2 (A) and 3B11 (B) were stimulated with 10 �g of B.bovis CM per ml and autologous APC that were precultured for 1 hwith no MAb (none) or with 4 �g of MAb specific for DR�, MAbspecific for DQ�, or isotype-matched MAb Colis205D (control) per mlfor 3 days. Results are presented as the mean counts per minute � 1SD of duplicate cultures of T cells stimulated with antigen. Asterisksindicate that the response is significantly lower than the response in thepresence of isotype control MAb (P 0.05). Results are representativeof at least two experiments performed with each clone.

TABLE 5. Identification of the minimal CD4� T-cell epitoperecognized by C97 Th-cell clones

Peptide Amino acid sequenceaSIe

4C9f 3G5f

Expt 1P1 DQEVIIDEQTGLPIKSHDYSEKPSVIYKPS 20.7 51.8P1-1 DQEVIIDEQTGLPIKS 31.2 52.9P1-2 PIKSHDYSEKPSVIY 1.4 1.1P1-3 EQTGLPIKS 0.9 1.1P1-4 DEQTGLPIKS 5.4 18.5P1-5 IDEQTGLPIKS 25.5 37.6P1-5S IDEQTGLPIK 1.4 0.2P1-5KS IDEQTGLPI 1.1 0.4

Expt 2BboP1-4 DEQTGLPIKS 10.3 34.8BbgP1-4b DEQTGLPVKN 0.8 15.0P1-4Nc DEQTGLPIKN 5.7 18.3P1-4Vd DEQTGLPVKS 0.8 41.2

a The minimal T-cell epitope that stimulated CD4� T cell clones is underlined.b Amino acid sequence of B. bigemina Hsp20 corresponding to P1-4. Amino

acid residues in bold type indicate differences between the B. bovis and B.bigemina Hsp20 epitopes.

c Replacement of the serine residue with an asparagine residue (bold type).d Replacement of the isoleucine residue with a valine residue (bold type).e SI, stimulation index (mean counts per minute of cells stimulated with 1 �g

of Hsp20 peptide per ml/mean counts per minute of control MSP2 peptide). SIvalues of �3.0 are considered positive and are shown in bold type.

f Hsp20-specific CD4� T-cell clones described previously (13).


on Decem


.org/D

ownloaded from

http://iai.asm.org/

cifically in the length of the �-chain CDR3, are probablyresponsible for the different recognition patterns of the twosets of clones (19, 37). Although the epitope-specific responseby both sets of clones is apparently restricted by a DQ mole-cule, it is possible that different DQ molecules present theBboP1-4 and BbgP1-4 epitopes, since these T-cell clones werederived from a cow with heterozygous DQ alleles. Further-more, cross-haplotype pairing of the DQ � and � chains canoccur, increasing the number of functional DQ heterodimersthat can present antigen (10, 24).

Induction of protective immunity against intraerythrocyticprotozoan parasites by immunization has been extensivelystudied and has proven to be challenging (26). We and othershave found that use of a single vaccine antigen, such as thehighly immunogenic B. bovis RAP-1 or MSA-1 protein, is notsufficient for stimulating protective immunity against B. bovisinfection (28, 45). On the other hand, multiepitope vaccineshave been shown to induce a protective level of specific anti-body and cellular immunity against experimental models ofmalaria and to stimulate T-lymphocyte responses in Plasmo-

dium falciparum-vaccinated or exposed humans (21, 23, 32, 33,46). As we have demonstrated in the present and recent stud-ies, B. bovis Hsp20 is conserved among B. bovis strains and B.bigemina and is immunogenic for T cells in cattle with differentgenetic backgrounds. The present study provides new informa-tion on the immunogenic epitopes of B. bovis Hsp20 that arerecognized broadly in the Holstein breed and that can be usedin designing a multiepitope vaccine against highly pathogenicB. bovis strains.

ACKNOWLEDGMENTS

We thank Kim Kegerreis, Shelley Whidbee, Deb Alperin, andEmma Karel for excellent technical assistance; Colleen Olmstead forBoLA DRB3 typing; Steve Hines and Will Goff for providing B. bovisparasites; Terry McElwain and Shawn Berens for providing B. bovisDNA samples; Lance Perryman for providing hybridoma cells produc-ing MAb 23/28.57; and Genetics Institute for providing human IL-12.

This research was supported by National Institutes of Health grantR01-A130136.

FIG. 6. Comparison of the TCR-� and TCR-� chain sequences between B. bovis Hsp20-specific CD4� T-cell clones with different finespecificities. The amino acid sequence alignment of TCR-� chains (A) and TCR-� chains (B) from clones 4C9 and 3G5 is shown. Leader (L),variable (V), joining (J), constant (C), and CDR1-3 regions are indicated by arrows or bars. The diversity (D) region is not indicated because ofthe ambiguous boundary. Boundaries of L, V, J, and C regions and CDRs were predicted based on data from references 2, 30, and 55. Conservedcysteine residues in the V region are indicated by asterisks.


on Decem


.org/D

ownloaded from

http://iai.asm.org/

REFERENCES

1. Amills, M., V. Ramiya, J. Norimine, and H. A. Lewin. 1998. The majorhistocompatibility complex of ruminants. Rev. Sci. Tech. Off. Int. Epizoot.1:108–120.

2. Arden, B., S. P. Clark, D. Kabelitz, and T. W. Mak. 1995. Human T-cellreceptor variable gene segment families. Immunogenetics 42:455–500.

3. Bannister, L. H., J. M. Hopkins, R. E. Fowler, S. Krishna, and G. H.Mitchell. 2000. A brief illustrated guide to the ultrastructure of Plasmodiumfalciparum asexual blood stages. Immunol. Today 16:427–433.

4. Beyer, J. C., R. W. Stich, W. C. Brown, and W. P. Cheevers. 1998. Cloningand expression of caprine interferon-gamma. Gene 210:103–108.

5. Bohne, W., U. Gross, D. J. P. Ferguson, and J. Heesemann. 1995. Cloningand characterization of a bradyzoite-specifically expressed gene (hsp30/bag1)of Toxoplasma gondii, related to genes encoding small heat-shock proteins ofplants. Mol. Microbiol. 16:1221–1230.

6. Brown, W. C., K. S. Logan, G. G. Wagner, and C. L. Tetzlaff. 1991. Cell-mediated immune responses to Babesia bovis antigens in cattle following in-fection with tick-derived or cultured parasites. Infect. Immun. 59:2418–2426.

7. Brown, W. C., K. S. Logan, S. Zhao, D. K. Bergman, and A. C. Rice-Ficht.1995. Identification of Babesia bovis merozoite antigens separated by con-tinuous-flow electrophoresis that stimulate proliferation of helper T cellclones derived from B. bovis-immune cattle. Infect. Immun. 63:106–116.

8. Brown, W. C., T. F. McElwain, G. H. Palmer, S. E. Chantler, and D. M.Estes. 1999. Bovine CD4� T-lymphocyte clones specific for rhoptry-associ-ated protein-1 (RAP-1) of Babesia bigemina stimulate enhanced immuno-globulin G1 (IgG1) and IgG2 synthesis. Infect. Immun. 67:155–164.

9. Brown, W. C., T. F. McElwain, B. J. Ruef, C. E. Suarez, V. Shkap, C. G.Chitko-McKown, A. C. Rice-Ficht, and G. H. Palmer. 1996. Babesia bovisrhoptry-associated protein-1 is immunodominant for T helper cells of im-mune cattle and contains T cell epitopes conserved among geographicallydistant B. bovis strains. Infect. Immun. 64:3341–3350.

10. Brown, W. C., T. C. McGuire, W. Mwangi, K. A. Kegerreis, H. Macmillan,H. A. Lewin, and G. H. Palmer. 2002. Major histocompatibility complex classII DR-restricted memory CD4� T lymphocytes recognize conserved immu-nodominant epitopes of Anaplasma marginale major surface protein 1a.Infect. Immun. 70:5521–5532.

11. Brown, W. C., T. C. McGuire, D. Zhu, H. A. Lewin, J. Sosnow, and G. H.Palmer. 2001. Highly conserved regions of the immunodominant majorsurface protein 2 (MSP2) of the ehrlichial pathogen Anaplasma marginaleare rich in naturally derived CD4� T lymphocyte epitopes that elicit strongrecall responses. J. Immunol. 166:1114–1124.

12. Brown, W. C., and G. H. Palmer. 1999. Designing blood stage-stage-vaccinesagainst Babesia bovis and B. bigemina. Parasitol. Today 15:275–281.

13. Brown, W. C., B. J. Ruef, J. Norimine, K. A. Kegerreis, C. E. Suarez, P. G.Conley, R. W. Stich, K. H. Carson, and A. C. Rice-Ficht. 2001. A novel20-kilodalton protein conserved in Babesia bovis and B. bigemina stimulatesmemory CD4� T lymphocyte responses in B. bovis-immune cattle. Mol.Biochem. Parasitol. 119:97–109.

14. Brown, W. C., S. Zhao, V. M. Woods, C. A. Tripp, C. L. Tetzlaff, V. T.Heussler, D. A. Dobbelaere, and A. C. Rice-Ficht. 1993. Identification of twoTh1 cell epitopes on the Babesia bovis-encoded 77-kilodalton merozoiteprotein (Bb-1) by use of truncated recombinant fusion proteins. Infect.Immun. 61:236–244.

15. Carson, R. T., K. M. Vignali, D. L. Woodland, and D. A. A. Vignali. 1997. Tcell receptor recognition of MHC class II-bound peptide flanking residuesenchances immunogenicity and results in altered TCR V region usage. Im-munity 7:387–399.

16. Court, R. A. K. Sitte, J. P. Opdebeeck, and I. A. East. 1998. Mapping T cellepitopes of Babesia bovis antigen 12D3: implications for vaccine design.Parasite Immunol. 10:1–8.

17. Dagliesh, R. J., and N. P. Stewart. 1982. Some effects of time, temperatureand feeding on infection rates with Babesia bovis and Babesia bigemina inBoophilus microplus larvae. Int. J. Parasitol. 12:323–326.

18. Davies, C. J., I. Joosten, D. L. Andersson, M. A. Arriens, D. Bernoco, B.Bissumbhar, G. Byrns, M. J. T. van Eijk, B. Kristensen, H. A. Lewin, S.Mikko, A. L. G. Morgan, N. E. Muggli-Cockett, P. R. Nilsson, R. A. Oliver,C. A. Park, J. J. van der Poel, M. Polli, R. L. Spooner, and J. A. Stewart.1994. Polymorphism of bovine MHC class II genes. Joint report of the FifthInternational Bovine Lymphocyte Antigen (BoLA) Workshop. Eur. J. Im-munogenet. 21:259–289.

19. Davis, M. M., and M. G. McHeyzer-Williams. 1995. Antigen-specific devel-opment of primary and memory T cells in vivo. Science 268:106–111.

20. Dutia, B. M., L. MacCarthy-Morrough, E. J. Glass, R. L. Spooner, and J.Hopkins. 1995. Discrimination between major histocompatibility complexclass II DQ and DR locus products in cattle. Anim. Genet. 26:111–114.

21. Doolan, D. L., S. L. Hoffman, S. Southwood, P. A. Wentworth, J. Sidney,R. W. Chesnut, E. Keough, E. Appella, T. B. Nutman, A. A. Lal, D. M.Gordon, A. Oloo, and A. Sette. 1997. Degenerate cytotoxic T cell epitopesfrom P. falciparum restricted by multiple HLA-A and HLB-B supertypealleles. Immunity 7:97–112.

22. Elwyn, Y. L., J. F. Elliott, S. Cwirla, L. L. Lanier, and M. M. Davis. 1989.

Polymerase chain reaction with single-sided specificity: analysis of T cellreceptor � chain. Science 24:217–220.

23. Enders, B., E. Hundt, and B. Knapp. 1992. Strategies for the development ofan antimalarial vaccine. Vaccine 10:920–927.

24. Glass, E. J., R. A. Oliver, and G. C. Russell. 2000. Duplicated DQ haplotypesincrease the complexity of restriction element usage in cattle. J. Immunol.165:134–138.

25. Goff, W. L., W. C. Johnson, S. M. Parish, G. M. Barrington, W. Tuo, andR. A. Valdez. 2001. The age-related immunity in cattle to Babesia bovisinfection involves the rapid induction of interleukin-12, interferon-�, andinducible nitric oxide synthase mRNA expression in the spleen. ParasiteImmunol. 23:463–471.

26. Good, M. F. 2001. Towards a blood-stage vaccine for malaria: are we fol-lowing all the leads? Nat. Rev. Immunol. 1:117–125.

27. Hines, S. A., G. H. Palmer, W. C. Brown, C. E. Suarez, O. Vidotto, T. F.McElwain, and A. C. Rice-Ficht. 1995. Genetic and antigenic characteriza-tion of Babesia bovis merozoite spherical body protein Bb-1. Mol. Biochem.Parasitol. 69:149–159.

28. Hines, S. A., G. H. Palmer, D. P. Jasmer, W. L. Goff, and T. F. McElwain.1995. Immunization of cattle with recombinant Babesia bovis merozoitesurface antigen-1. Infect. Immun. 63:349–352.

29. Horwitz, J. 1992. Alpha-crystallin can function as a molecular chaperone.Proc. Natl. Acad. Sci. USA 89:10449–10453.

30. Ishiguro, N., A. Tanaka, and M. Shinagawa. 1990. Sequence analysis ofbovine T-cell receptor � chain. Immunogenetics 31:57–60.

31. Jakob, U., M. Gaestel, K. Engel, and J. Buchner. 1993. Small heat-shockproteins are molecular chaperones. J. Biol. Chem. 268:1517–1520.

32. Khusmith, S., Y. Charoenvit, S. Kumar, M. Sedegah, R. L. Beaudoin, andS. L. Hoffman. 1991. Protection against malaria by vaccination with sporo-zoite surface protein 2 plus CS protein. Science 252:715–718.

33. Knapp, B, E. Hundt, B. Enders, and H. A. Kupper. 1992. Protection of Aotusmonkeys from malaria infection by immunization with recombinant hybridproteins. Infect. Immun. 60:2397–2401.

34. Lindquist, S., and E. A. Craig. 1998. The heat-shock proteins. Annu. Rev.Genet. 22:631–77.

35. Lohr, C. V., K. A. Brayton, V. Shkap, R. T. Molad, A. F. Barbet, W. C. Brown,and G. H. Palmer. 2002. Expression of Anaplasma marginale major surfaceprotein 2 operon-associated proteins during mammalian and arthropod in-fection. Infect. Immun. 70:6005–6012.

36. Mahoney, D. F. 1986. Studies on the protection of cattle against Babesia bovisinfection, p. 539–541. In W. I. Morrison (ed.), The ruminant immune system inhealth and disease. Cambridge University Press, Cambridge, United Kingdom.

37. McHeyzer-Williams, M. G., J. D. Altman, and M. M. Davis. 1996. Enumer-ation and characterization of memory cells in the TH compartment. Immu-nol. Rev. 150:5–21.

38. Mohamed, R. M., F. Aosai, M. Chen, H.-S. Mun, K. Norose, U. S. Belal, L.-X.Piao, and A. Yano. 2003. Induction of protective immunity by DNA vacci-nation with Toxoplasma gondii HSP70, HSP30, and SAG1 genes. Vaccine21:2852–2861.

39. Mosqueda, J., T. F. McElwain, and G. H. Palmer. 2002. Babesia bovis surfaceantigen 2 proteins are expressed on the merozoite and sporozoite surface,and specific antibodies inhibit attachment and invasion of erythrocytes. In-fect. Immun. 70:6448–6455.

40. Mosqueda, J., T. F. McElwain, D. Stiller, and G. H. Palmer. 2002. Babesiabovis merozoite surface antigen 1 and rhoptry-associated protein 1 are ex-pressed in sporozoites and specific antibodies inhibit sporozoite attachmentto erythrocytes. Infect. Immun. 70:1599–1603.

41. Moudgil, K. D., E. E. Sercarz, and I. S. Grewal. 1998. Modulation of theimmunogenicity of antigenic determinants by their flanking residues. Immu-nol. Today 19:217–220.

42. Mun, H.-S., F. Aosai, and A. Yano. 1999. Role of Toxoplasma gondii Hsp70and Toxoplasma gondii Hsp30/bag1 in antibody formation and prophylacticimmunity in mice experimentally infected with Toxoplasma gondii. Micro-biol. Immunol. 43:471–479.

43. Norimine, J., C. E. Suarez, T. F. McElwain, M. Florin-Christensen, andW. C. Brown. 2002. Immunodominant epitopes in Babesia bovis rhoptry-associated protein 1 that elicit memory CD4� T lymphocyte responses in B.bovis-immune individuals are located in the amino-terminal domain. Infect.Immun. 70:2039–2048.

44. Norimine, J., J. Mosqueda, C. Suarez, G. H. Palmer, T. F. McElwain, G.Mbassa, and W. C. Brown. 2003. Stimulation of T helper cell gamma inter-feron and immunoglobulin G responses specific for Babesia bovis rhoptry-associated protein 1 (RAP-1) or a RAP-1 protein lacking the carboxy-terminal repeat region is insufficient to provide protective immunity againstvirulent B. bovis challenge. Infect. Immun. 71:5021–5032.

45. Palmer, G. H., T. F. McElwain, L. E. Perryman, W. C. Davis, D. R. Reduker,D. P. Jasmer, V. Shkap, E. Pipano, W. L. Goff, and T. C. McGuire. 1991.Strain variation of Babesia bovis merozoite surface-exposed epitopes. Infect.Immun. 59:3340–3342.

46. Patarroyo, M. E., P. Romero, M. L. Torres, P. Clavijo, A. Moreno, A.Maritines, R. Rodrıguez, F. Buzman, and E. Cabezas. 1987. Induction of


on Decem


.org/D

ownloaded from

http://iai.asm.org/

protective immunity against experimental infection with malaria using syn-thetic peptides. Nature 328:629–632.

47. Rand, K. N., T. Moore, A. Sriskantha, K. Spring, R. Tellam, P. Willadsen,and G. S. Cobon. 1989. Cloning and expression of a protective antigen fromthe cattle tick Boophilus microplus. Proc. Natl. Acad. Sci. USA 86:9657–9661.

48. Reduker, D. W., D. P. Jasmer, W. L. Goff, L. E. Perryman, W. C. Davis, andT. C. McGuire. 1989. A recombinant surface protein in Babesia bovis elicitsbovine antibodies that react with live merozoites. Mol. Biochem. Parasitol.35:239–247.

49. Reinherz, E. L., K. Tan, L. Tang, P. Kern, J.-H. Liu, Y. Xiong, R. E. Hussey,A. Smolyar, B. Hare, R. Zhang, A. Joachimiak, H.-C. Chang, G. Wagner, andJ.-H. Wang. 1999. The crystal structure of a T cell receptor in complex withpeptide and MHC class II. Science 286:1913–1921.

50. Sharif, S., B. A. Mallard, B. N. Wilkie, J. M. Sargeant, H. M. Scott, J. C. M.Dekkers, and K. E. Leslie. 1998. Associations of the bovine major histocom-patibility complex DRB3 (BoLA-DRB3) alleles with occurrence of diseaseand milk somatic cell score in Canadian dairy cattle. Anim. Genet. 29:185–193.

51. Shoda, L. K. M., J. Florin-Christensen, M. Florin-Christensen, G. H.Palmer, and W. C. Brown. 2000. Babesia bovis-stimulated macrophages ex-press interleukin-1�, interleukin-12, tumor necrosis factor-alpha, and nitricoxide, and inhibit parasite replication in vitro. Infect. Immun. 68:5139–5145.

52. Stich, R. W., L. K. M. Shoda, M. Dreewes, B. Adler, T. W. Jungi, and W. C.Brown. 1998. Stimulation of nitric oxide production in macrophages byBabesia bovis. Infect. Immun. 66:4130–4136.

53. Stich, R. W., A. C. Rice-Ficht, and W. C. Brown. 1999. Babesia bovis: com-mon protein fractions recognized by oligoclonal B. bovis-specific CD4� T celllines from genetically diverse cattle. Exp. Parasitol. 91:40–51.

54. Suarez, C. E., T. F. McElwain, E. B. Stephens, V. S. Misha, and G. H.

Palmer. 1991. Sequence conservation among merozoite apical complex pro-teins of Babesia bovis, Babesia bigemina and other apicomplexa. Mol. Bio-chem. Parasitol. 49:329–332.

55. Tanaka, A., N. Ishiguro, and M. Shinagawa. 1990. Sequence and diversity ofbovine T-cell receptor �-chain genes. Immunogenetics 32:263–271.

56. Tetzlaff, C. L., A. C. Rice-Ficht, V. M. Woods, and W. C. Brown. 1992.Induction of proliferative responses of T cells from Babesia bovis-immunecattle with a recombinant 77-kilodalton merozoite protein (Bb-1). Infect.Immun. 60:644–652.

57. Tuo, W., G. H. Palmer, T. C. McGuire, D. Zhu, and W. C. Brown. 2000.Interleukin-12 as an adjuvant promotes immunoglobulin G and type 1 cyto-kine recall responses to major surface protein 2 of the ehrlichial pathogenAnaplasma marginale. Infect. Immun. 68:270–80.

58. van Eijk, M. J. T., J. A. Stewart-Haynes, and H. A. Lewin. 1992. Extensivepolymorphism of the BoLA-DRB3 gene distinguished by PCR-RFLP. Anim.Genet. 23:483–496.

59. Wright, I. G., R. Casu, M. A. Commins, B. P. Dalrymple, K. R. Gale, B. V.Goodger, P. W. Riddles, D. J. Waltisbuhl, I. Abetz, D. A. Berrie, Y. Bowles,C. Dimmock, T. Hayes, H. Kalnins, G. Leatch, R. McCrae, P. E. Montague,I. T. Nisbet, F. Parrodi, J. M. Peters, P. C. Scheiwe, W. Smith, K. Rode-Bramanis, and M. A. White. 1992. The development of a recombinantBabesia vaccine. Vet. Parasitol. 44:3–13.

60. Wright, I. G., B. V. Goodger, and I. A. Clark. 1988. Immunopathophysiologyof Babesia bovis and Plasmodium falciparum infections. Parasitol. Today4:214–218.

61. Zhang, Y., G. H. Palmer, J. R. Abbott, C. J. Howard, J. C. Hope, and W. C.Brown. 2003. CpG ODN 2006 and IL-12 are comparable for priming Th1lymphocyte and IgG responses in cattle immunized with a rickettsial outermembrane protein in alum. Vaccine 21:3307–3318.

Editor: W. A. Petri, Jr.


on Decem


.org/D

ownloaded from

http://iai.asm.org/

Conservation of Babesia bovis Small Heat Shock Protein ...Conservation of Babesia bovis Small Heat Shock Protein (Hsp20) among Strains and Deﬁnition of T Helper Cell Epitopes Recognized

Documents