of October 4, 2017.This information is current as

and the Preimmune Repertoire Revisited GenesHand Neonatal Piglets. XIII. Hybrid V

Antibody Repertoire Development in Fetal

John E. Butler, Patrick Weber and Nancy Wertz

http://www.jimmunol.org/content/177/8/5459doi: 10.4049/jimmunol.177.8.5459

2006; 177:5459-5470; ;J Immunol 

Referenceshttp://www.jimmunol.org/content/177/8/5459.full#ref-list-1

, 16 of which you can access for free at: cites 39 articlesThis article

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists All rights reserved.Copyright © 2006 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

Antibody Repertoire Development in Fetal and Neonatal Piglets.XIII. Hybrid VH Genes and the Preimmune Repertoire Revisited1

John E. Butler,2 Patrick Weber, and Nancy Wertz

The expressed porcine VH genes belong to the VH3 family (clan), four of which, VHA, VHB, VHC, and VHE, alone comprise �80%of the preimmune repertoire. However, so-called “hybrid” VH genes that use CDR1 of one VH gene and the CDR2 of another arefrequently encountered. We studied >3000 cloned VDJs and found that such hybrids can contribute up to 10% of the preimmunerepertoire. Based on the 1) recovery of hybrid VH genes from bacterial artificial chromosome clones, 2) frequency of occurrenceof certain hybrids in the preimmune repertoire, and 3) failure to recover equal numbers of reciprocal hybrids, we concluded thatsome chimeric genes are present in the genome and are not PCR artifacts. Two chimeric germline genes (VHZ and VHY), togetherwith VHF and the four genes mentioned above, constitute the major VH genes and these account for >95% of the preimmunerepertoire. Diversification of the preimmune IgG and IgM repertoires after environmental exposure was mainly due to somatichypermutation of major VH genes with no evidence of gene conversion. Somatic hypermutation was 3- to 10-fold higher in CDRsthan in framework regions, most were R mutations and transversions and transitions equally contributed. Data were used to 1)develop an index to quantify the degree of VH repertoire diversification and 2) establish a library of 29 putative porcine VH genes.One-third of these genes are chimeric genes and their sequences suggest that the porcine VH genome developed by duplication andsplicing from a small number of prototypic genes. The Journal of Immunology, 2006, 177: 5459–5470.

A decade ago, we reported that the preimmune VH reper-toire of swine was composed of �20 VH3 family genes(1). In fetal piglets, 4 of these VH genes (VHA, VHB,

VHC, and VHE) account for �80% of the total usage of rearrangedVH genes in DNA (2–4) and this preferential pattern of VH geneusage is also seen in transcripts from various tissues of young pigs(5). These observations are inconsistent with the multigenic para-digm established from studies in mice and humans. Rather, theyare more reminiscent of rabbits in which the most 3� VH gene(VH1) is used 90% of the time (6), of sheep in which five V� genesaccount for �70% of the repertoire (7), and chickens in whichsingle 3� VH and 3� V� genes account for the entire repertoire (8,9). In rabbits, age and antigenic exposure can result in usage ofgenes other than VH1 (10, 11). Although both chickens and therabbits use gene conversion (templated somatic mutation) to com-pensate for the restricted VH usage (12, 13), gene conversion hasnot been convincingly demonstrated in swine. However, so-calledhybrid VH genes are frequently encountered in swine. Hybrid VH

genes are chimeras containing the CDR1 region of one majorporcine VH gene and the CDR2 of another. [The major porcine VH

are the four genes that form �80% of the preimmune repertoire (2)plus VHF, VHY, and VHZ that when combined, comprise �95% of

the preimmune repertoire (see Table I). Other VH genes shown inTable I that have been recovered in full-length from DNA and/orbacterial artificial chromosomes (BACs)3 or repeatedly in tran-scripts, are called minor VH genes. The term hybrid and “chimera”are used interchangeably.] In vitro studies demonstrated that pre-sumably because of the low processivity of thermostable poly-merases (14, 15), up to 38% of such hybrids could be generated asPCR artifacts when a 1:1 ratio of two different porcine VH geneswere used as templates (16). However, the proportion was verysmall when the template ratio was large so VDJ chimeras gener-ated as PCR hybrids from most cDNA preparations should be rare.Moreover, the 15–20% of VHA/VHB hybrids predicted from our invitro studies has never been observed in cDNAs even though VHAand VHB transcripts together account for 40–60% of the preim-mune cDNA (Refs. 2, 4, 5; also, this report). In contrast, hybridsof VHE/VHC approaching such frequencies have been seen but notthe reciprocal VHC/VHE hybrid (this report). [VHE/VHC meansCDR1 is VHE and CDR2 is from VHC. This nomenclature is fur-ther abbreviated to simply E/C (see Table I).] We reasoned thatboth forms should be equally abundant if the hybrids are due to thelow processivity of Taq polymerase. These observations warrantedthe need to quantify the frequency of porcine VH gene hybrids sothe constituency and diversification of the preimmune repertoirecould be properly assessed. The piglet model is of great value indevelopmental studies because fetal piglets are not exposed toenvironmental Ag and maternal regulatory factors, and becausepostnatal exposure to such factors as normal gut flora, diet,pathogens, and maternal factors can be controlled postnatallyby the experimenter.

In most of our previous studies on VH usage in swine usingclonal hybridization (2, 3, 5) and those in humans using the samemethod (17), only single VH gene-specific probes were used. How-ever, in limited studies using dual probes for CDR1 and CDR2, weidentified hybrid clones, but because of the low frequency of theiroccurrence, they were included into a category labeled “other” (5).This policy was adopted not only because of their infrequent use butout of concern that such VDJ clones could be PCR artifacts (16). The

Department of Microbiology, University of Iowa, Iowa City, IA 52242

Received for publication March 20, 2006. Accepted for publication June 29, 2006.

The costs of publication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked advertisement in accordancewith 18 U.S.C. Section 1734 solely to indicate this fact.1 This work was supported by National Pork Board Grants 05-143 and 05-174 and byCooperative Agreement 58-3625-4-155 with the U.S. Department of Agriculture-Agricultural Research Service (Ames, IA).2 Address correspondence and reprint requests to Dr. John E. Butler, University ofIowa, Department of Microbiology, Interdisciplinary Immunology Program, 51 New-ton Road, Iowa City, IA 52242. E-mail address: john-butler@uiowa.edu3 Abbreviations used in this paper: SHM, somatic hypermutation; MLN, mesentericlymph node; BAC, bacterial artificial chromosome; CSR, class switch recombination;FR, framework; LB, Luria-Bertani; GF, germfree; RDI, repertoire diversificationindex.

The Journal of Immunology

Copyright © 2006 by The American Association of Immunologists, Inc. 0022-1767/06/$02.00

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

“other” category also includes VH genes that are no longer able tohybridize with CDR-specific probes because of somatic hypermuta-tion (SHM). In this study, we report on a systematic study to tabulateand characterize hybrid and other VH gene usage in transcripts fromthe mesenteric lymph nodes (MLN) of late-term fetal piglets, 5-wk-old colonized isolator piglets and conventional piglets heavily anti-genized for a Th2 response by parasitic infection. The MLN waschosen because it is influenced by both the normal flora that colonizesthe gut and infection with intestinal helminthes. We addressed theissue of hybrid/chimeric VH genes by 1) quantifying their occurrence

in �3000 VDJ clones and 2) determining whether they occur in thegenome by examining several porcine VH-containing BACs. Wealso sequenced �100 hybridizing and nonhybridizing VDJclones to assess the role of SHM in generating genes of theother category. By comparing the repertoire in IgM and IgGtranscripts, we were able to test the affect of class switch recom-bination (CSR) on the expressed VH Ab repertoire. A comparisonof VH usage in the preimmune repertoire with that in colonizedpiglets or those exposed to intestinal parasites, allowed an indexfor quantifying the degree of repertoire diversification to be

Table I. Porcine VH repertoire

DQ886392

DQ886393

DQ886394

DQ886395

a Proposed name after comparison to sequences in GenBank. � � Denotes major VH genes of the preimmune repertoire (see Fig. 4).b The single Arabic designator indicates the CDR region. The sequence for each are given below. Various VH genes share CDR sequences with other genes. Some share one

CDR but possess also a unique CDR, e.g. VHL shares CDR1 with VHE but has a unique CDR2 called “L.”c B � Butler Lab based on repeated recovery; K � Kim, from GenBank.d Sequences entered in GenBank as protein sequences by other investigators are not included since they cannot be identified without their nucleotide sequences.

5460 Ab REPERTOIRE DEVELOPMENT IN FETAL AND NEONATAL PIGLETS

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

devised. This index has proven valuable in studies on viral andbacterial infections of gnotobiotic isolator piglets. The study ex-tends our 1998 report by providing information on 29 VH genesand by showing that inclusion of three additional VH genes meansthat seven major genes account for �95% of the preimmunerepertoire. Evidence presented here indicate that chimeric/hybridVH genes are present in the genome, apparently as the consequenceof gene duplication and recombination during evolution.

Materials and MethodsSource of animal tissue

MLN were collected after euthanization of uninfected 110–114 fetal pig-lets and after 6 wk from young pigs infected with Ascaris suum or Tri-churis suis. Samples collected were by-products of colonization studieswith neonatal isolator piglets at South Dakota State University (Brookings,SD) and helminthic studies at the Agricultural Research Center (Beltsville,MD). Isolator piglets were colonized by monoassociation with Escherichiacoli G58-1 (18). Animal studies were approved by the Animal Care andUse Committees of the respective research institutes. Samples were frozenin liquid nitrogen and transported on dry ice to the University of Iowa. Allanimals were selectively outbred making them less heterozygous than hu-mans when typed for MHC (D. Smith, unpublished observation).

Preparation of cDNA

Frozen tissues were pulverized with a mortar and pestle and then solubi-lized in TRI-reagent (MRC). Total RNA was prepared according to themanufacturer’s recommendations and first strain cDNA was prepared aspreviously described (5, 19). Random hexamers as well as isotype-specificprimers were used, the latter to enrich the recovery of isotype-associatedVDJ transcripts (Fig. 1).

Cloning and identification by hybridization

IgM- and IgG-associated VDJs were recovered by two stage PCR. Usinga primer pair consisting of antisense IgM and the common 5� framework(FR)1 primer (Fig. 1) the first round IgM-associated VDJ products were

amplified. This PCR product was then separated on a 1.5% agarose gel andthe polynucleotide band corresponding to the correct size for the IgM prod-uct, recovered from the gel using a Costar Spin-x spin column and sub-jected to a second round PCR using 3� FR1 primer and antisense JH (Fig.1). IgG-bearing transcripts were recovered in a similar manner but using anantisense C�1 primer in the first round (Fig. 1).

The second round PCR product was cloned into pCR TOPO2.1 and grownin TOP TEN cells as described previously (5, 20). Briefly, positive colonieswere selected and transferred to the wells of 96-well microtiter plates (Costar;3799). Colonies were grown overnight at 37°C in Luria-Bertani (LB) ampi-cillin broth (LB-Amp), the plasmid DNA was recovered by alkaline lysis andtransferred to nylon membranes (Schleicher & Schuell Microscience).

The membranes were then sequentially hybridized with 32P-labeledprobes specific for the major VH genes of swine and finally a common FR2probe to determine the total number of VDJ-containing clones (Fig. 1;Refs. 2, 3, and 5). The definition of “other” VH genes (Figs. 2 and 3) refersto all clones that hybridize with the FR2 probe but that fail to hybridizewith both probes for CDR1 and CDR2 of the major VH genes (Fig. 1; Refs.2 and 3; see Table I). The “other” category therefore potentially containsfour different categories: 1) major genes that do not hybridize at all be-cause of SHM in the CDR1 and CDR2 probe binding regions; 2) cloneshybridizing with only one CDR probe because the probe binding region ofthe other CDR is mutated; 3) clones hybridizing with only one CDR probebecause the other CDR is not found in a major porcine VH genes; and 4)clones that only hybridize with FR2 because they are minor VH genes forwhich CDR probes were not prepared (see Table I). Distinguishing amongthese categories was based on clonal hybridization and sequence analysis.Hybrid genes or others with unique sequences that were repeatedly recov-ered were considered putative germline VH genes. Those infrequently en-countered that could not be readily identified as a mutant of a frequentlyencountered VH gene or a minor VH gene were considered to be the con-sequence of SHM or a PCR artifact.

Sequence analysis

One hundred five VDJ clones were sequenced and analyzed, including 81clones that hybridized with the FR2-specific probe but with only one or noprobes for CDR1 or CDR2. These clones were transferred to 17 � 100Falcon tubes and grown overnight in LB-Amp. Cultures were then pro-cessed using the Fast Plasmid Mini procedure (Eppendorf). Plasmid DNA

FIGURE 1. Major porcine VH genes andpolynucleotide probes and primers. Top, Nu-cleotide sequences of the seven major VH

genes of the porcine preimmune repertoire.CDR probe binding regions are boxed aswell as the region recognized by the pan-spe-cific FR2 probe. Dots indicate identical nu-cleotides using VHA as the reference VH

gene. VHY and VHZ are chimeric genes de-tected with the probes for CDR1 of VHC andCDR2 of VHA (VHY) and CDR1 of VHE andCDR2 of VH (VHZ). Bottom, Polynucleotideprimers used for the preparation of cDNAand for first and second round PCR.

5461The Journal of Immunology

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

concentration was determined by A260/280 spectrophotometry and sam-ples containing 400 ng of DNA were provided to the University of IowaCore Facility for analysis. Sequence analysis was done using the AppliedBiosystems capillary sequencer (ABI Prism) and data were analyzed usingthe Omiga program (Accelyr) as previously described (20).

VH genes and mutated VH genes were identified by reference to theknown sequence of repeatedly recovered porcine VH genes (Fig. 1; seeTable I). Sequence data were used to distinguish between major and mi-nor VH genes (see Table I) and hybrid combination that had not previouslybeen described. Sequence data were also used to determine the number ofmutations in probe binding regions needed to prevent hybridization. Thenumber of transition and transversion changes was recorded as the numberof R and S mutations. DH segments and CDR3 length was determined aspreviously described (4).

In some cases, VDJ clones hybridized with three CDR probes. Eight ofthese were sequenced and directly rehybridized or subcloned and rehybrid-ized. These studies were undertaken to determine the basis of occasional“triple probe hybridization.”

VH-containing BAC clones

BAC clones containing porcine VH genes were obtained in collaborationwith Drs. P. Chardon and F. Piumi (Institut National de la RechercheAgronomique, Joey-en-Josas, France). These were recovered by PCR usingprimers for 5� FR1 and 3� FR3. The various BAC clones were grown up inLB chloramphenical medium and processed using Qiagen reagents. Theprocess was done as a Maxiprep but using two times the normal amount ofreagents as specified by the supplier. Samples were then processed throughan isopropanol precipitation step, dissolved in water, and ethanol precipi-tated. BAC clones were digested with restriction enzymes and hybridizedin Southern blots with a pan-specific (FR2) probe, certain VH gene-specificprobes and a probe for the common leader sequence. The VH gene content

of BAC clones was further analyzed by PCR amplification using a primerset comprised of the 5� FR1 primer and the antisense primer for the con-served region of FR3. The resultant products were then cloned into pCRTOPO2.1, plated, and the plasmid DNA was transferred to nylon mem-branes as described above. Clone-bearing membranes were hybridized withvarious VH-specific probes as described above to determine the VH contentof the original BAC clone. A number of these were selected at random forsequence analysis. Because of concern for hybrid PCR products, specialattention was given to the identification of VH genes in the same BACclone that could provide donor sequence for possible PCR-inducedhybrid VDJs.

Statistical analyses

Advice was provided by Dr. K. Chaloner (Department of Biostatistics,University of Iowa, Iowa City, IA). Simple mean differences were com-pared by two-tailed Student’s t analysis.

ResultsDual VH gene probes reveal hybrid VDJ clones

Fig. 1 gives the complete sequences of the major porcine VH

genes that comprise the preimmune repertoire. Also shown are thesites where the 10 CDR1- and CDR2-specific probes and the pan-specific FR2 probe used for clonal hybridization bind their targets.Fig. 1 reviews that differences among porcine VH genes are almostexclusively confined to CDR regions which explains why onlyCDR sequences are reported in Table I (see later). Fig. 2 comparesthe VH usage by IgM and IgG transcripts recovered from the MLNof newborn and colonized isolator piglets. The colonized piglets

FIGURE 2. VH gene usage by newborn and E. coli-colonized piglets identified using single VH gene probes (�) or dual VH gene probes (f). The x-axisidentifies the various VH gene based on their CDR regions, e.g., A/A or A/F. Complete VH gene names can be obtained by reference to Table I. The veryshortest bars represent the recovery of one clone. �, Statistically different between single- and dual-probe assays.

5462 Ab REPERTOIRE DEVELOPMENT IN FETAL AND NEONATAL PIGLETS

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

were monoassociated with E. coli G58-1, a benign commensal.Blots containing up to 96 clones were sequentially hybridized withone probe each for VHA, VHB, VHC, VHE, VHF, and FR2 (single-probe system) or sequentially with all 10 VH-specific probes (dou-ble probe system) and the control FR2 probe (Fig. 1). These com-parisons show that while the overall profile of VH usage is similarin single- and dual-probe assays, statistically significant differ-ences are observed. Although the overwhelming use of VHA bynewborn IgG transcripts was seen with both assays, the single-probe assay most frequently overestimated the usage of VHC andVHE. This occurred because VHE/VHC chimeras account for�10% of total VH usage by IgG transcripts. However, the recip-rocal chimera (VHC/VHE) was absent. Among colonized piglets,twice as many hybrids were detected among IgM transcripts thanamong IgG transcripts, whereas more IgG transcripts are ac-counted for by “other” (see later).

In summary, Fig. 2 shows that 1) up to 15% of total VH usagein the MLN of newborn and colonized piglets is comprised ofchimeras of the major five VH genes (VHF is now included as amajor VH gene), 2) the chimeras VHE/VHC (VHZ) and VHC/VHA(VHY) alone account for up to 10% of the repertoire, and 3) theoverall profile of VH usage is the same for both dual- and single-probe assay systems.

Fidelity and reliability of clonal hybridization

VDJ transcripts should hybridize with one CDR1 and one CDR2probe if they contain a major, nonmutated germline VH gene. Onoccasion, a clone may hybridize with three different CDR probes.Tables IIA summarizes data on such clones. These data show thatof 80 triple hybridizers, the occurrence is lower in clones fromparasite-infected pigs. Because �5% or less of all clones belong to

the triple hybridizer category, their presence can be ignored unlessthis form of nonspecific hybridization was VH gene specific andmight therefore skew the results. Thus, nine such clones were se-quenced and each was shown to be a single VDJ sequence and nota double clone (Table IIB). Six dual hybridizers were sequencedto serve as controls (Table IIC). Triple hybridizing clones weregrown up, replated, and rehybridized. Half of these continued tohybridize with three probes. The sequences of VDJs that con-tinued to hybridize with three probes were analyzed in Omiga todetermine whether a spurious binding site could be found out-side the CDRs, but none could be found.

The reliability of clonal hybridization as a measure of the degreeof SHM was also tested using the sequences of 89 CDR1 regionsand 94 CDR2 regions (Table III). These data show that CDR-specific probes differ in their ability to tolerate mutations in theirCDRs. The CDR2 probe for VHB was most demanding in that only1 of 10 clones allowed even a single mutation in the probe bindingregion and the average mutation frequency for nonhybridizingVHB-2 clones was only 1.6, whereas the average for all nonhy-bridizing CDRs was 3.3. By contrast, the CDR2 probes for VHA,VHC, and the CDR1 probe for VHE tolerated up to three mutationsin some clones and could still hybridize. Although stringency differed,the overall data show that the average number of mutations inCDR regions of hybridizing clones was 0.7, whereas for nonhy-bridizing clones, it was 5-fold higher.

Proportional VH usage is similar among piglets in differenttreatment groups

A great deal of immunological research is based on the use ofgenetically defined strains of mice which raises the question ofwhether individual genetic differences in outbred populations such

Table II. Analysis of triple hybridizing VDJ clones

a All sequences submitted to GenBank.b Both probes are for CDR2. None of the five CDR1 probes (Fig. 1) hybridized.

5463The Journal of Immunology

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

as humans and swine, substantially effect conclusions drawn fromstudies of the type we describe. Fig. 3 shows the proportionalusage of the most frequently used major VH genes/categories andother, among three different treatment groups of piglets. The re-sults indicate that 1) the usage profile is similar in nearly everycase for both IgG and IgM transcripts, 2) all piglets use the samemajor VH genes in forming their preimmune repertoire, and 3)differences in proportional VH usage are the consequence of dif-ferent environmental exposure. VHB, VHC, and VHE/VHF usagedid not differ between colonized isolator piglets and newborns.Rather, colonization affected usage of VHA and “other.” The great-est individual variation in proportional usage was seen in VHA andVHB expressed with IgM among newborn piglets and VHE/VHFusage by IgG transcripts of newborns.

Parasite infection results in diversification of the preimmuneIgM and IgG repertoires

Conventionally reared piglets are subjected to greater environmen-tal exposure than are isolator piglets monoassociated with a singlecolonizer. In addition to a more complex gut flora, we used con-ventional piglets that were parasitized by intestinal helminthic par-asites that favor Th2 responses. Figs. 3 and 4 compare the VH

repertoire of IgM and IgG transcripts from the MLN of newborn,colonized isolator piglets and parasite-infected, conventional pig-lets using dual probe hybridization. Fig. 4 show that �95% of thenewborn preimmune IgM repertoire is formed from VHA, VHB,VHC, VHE, VHF, VHZ, and VHY. [We define the preimmune rep-ertoire as the one seen in fetal and newborn piglets that have never

FIGURE 3. Individual and group variation inproportional VH usage by IgM and IgG transcripts.Data are presented for the five most frequently usedVH genes and “other” in the MLN of newborns,mono-associated isolator piglets, and parasite-in-fected piglets. Each point represents a single animaland is based on 250–350 VDJ clones per point. Thehorizontal bar depicts the mean usage. Brackets con-nect groups that did not differ significantly. VHE andVHZ were grouped together in interest of space andbecause usage is often complementary, suggestingthey may be allelic VH genes. Note that the scale ofthe y-axis differs among panels.

5464 Ab REPERTOIRE DEVELOPMENT IN FETAL AND NEONATAL PIGLETS

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

experienced maternal Abs or environmental Ag including those ofcolonizing bacteria.] By contrast, hybridizing forms of these genesaccount for only 12% of the IgM repertoire of parasite-infectedanimals. Rather, �80% of the repertoire of parasite-infected ani-

mals is accounted for by “other” VH genes which is significantlygreater than for all other groups studied (Figs. 3 and 4). A similardifference between the preimmune IgG repertoire and that ofyoung parasite-infected pigs was seen except that 17% of the

FIGURE 4. Trends in VH gene usage in the MLNof newborn piglets, 5-wk-old GF piglets, E. coli-mono-associated (colonized) isolator piglets, andparasite-infected conventional piglets using dualprobe hybridization. Data are based on the numberof VDJ clones examined for each group (in paren-theses). “Other” includes all VDJ clones that did nothybridize with either CDR1 and CDR2 probes forthe major VH genes, but did hybridize with a pan-specific probe for FR2 (Figs. 1 and 5 and Table IV).Hyb includes all clones that hybridized with dualprobes, but for different CDRs than those that iden-tify VHA, VHB, VHC, VHE, VHF, VHZ, and VHY. Thevalue in brackets is the calculated value for the RDI(see Discussion). To reduce clutter, error bars areonly included for data on newborn piglets and sta-tistical outcomes are described rather than indicated.The mean values for VHA, VHB, and VHE/VHF usagein IgG transcripts were statistically lower in para-site-infected animals while the value for “other” wasstatistically higher than for other groups. In the caseof IgM transcripts, VHC usage was also significantlylower than in parasite-infected pigs.

Table III. Mutation analysis of hybridizing/nonhybridizing VDJ clones

a Number in parentheses indicates number of sequences without mutations.b �1 � CDR1 so VHA-1 � CDR1 of VHA; VHA-2 � CDR2 of VHA.c Mutmin � Minimum number of mutations per CDR region.d Mutmax � Maximum number of mutations found in probe-binding region.e Mutavg � Average number of mutations per CDR1 or CDR2.

5465The Journal of Immunology

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

repertoire in parasite-infected animals was still composed of hy-bridizing VHA. The overall IgM repertoire in colonized piglets wasnot statistically different from newborns (Fig. 4) but differences inVHA and other usage were significant (Fig. 3). Twenty percent ofthe IgM repertoire and 38% of the IgG repertoire in E. colicolonized piglets is classified as “other” as opposed to �80% forboth IgM and IgG repertoires in parasite-infected piglets. Themean value for “other” in parasite-infected animals was signifi-cantly higher (0.001 level) than “other” in all other groups (Figs.3 and 4). Age alone had almost no significant effect on the VH

repertoire of IgM as indicated by comparing 5-wk-old germfree(GF) animals with newborns. The IgG repertoire was significantlydiversified from that of newborns which may contain an agecomponent although animal variation (SE � 5.3) renders this effectinsignificant. The values in brackets are the repertoire diversificationindex (RDI). These values are explained in Discussion.

The nature of nonhybridizing VDJ clones

The method of clonal hybridization results in four categories ofVDJ clones: 1) clones hybridizing with two probes (CDR1 andCDR2); 2) clones hybridizing with only one CDR probe; 3) cloneshybridizing with no CDR probe; and 4) clones hybridizing withthree CDR probes (triple hybridizers). To qualify as a VDJ clone,all must hybridize with the pan-specific FR2 probe (Fig. 1). Thenature of clones from category 4) was presented in Table II anddiscussed above. These triple hybridizers appear to be the result ofspurious binding of a third CDR probe. Because they constitute 5%or less of all clones (Table IIA) it has a very small impact onstudies of VH usage and VH repertoire diversification.

Of greater importance is information about nonhybridizing orpartial hybridizing (only one CDR probe) VDJ clones. This cate-gory was formerly placed in the “other” category (Ref. 5; Figs.2–4). Table IV summarizes sequence data on 105 VDJ clones

including 81 that fit into categories 2) and 3) described above.Table IV shows that consistent with Table III, partial hybridizingor nonhybridizing VDJs have 2-fold more mutations than doublehybridizers. Furthermore, there is a nonsignificant trend for usageof DHB to decrease and for other DH usage (27%) to increase inpartial or nonhybridizing VDJ clones. Among those VDJ clonesthat do not completely hybridize, 90% of VH genes expressed withIgM and 80% of those expressed with IgG use major or minor VH

genes and not rare hybrids. Partial hybridizing or nonhybridizingVDJs use mutated major VH genes twice as frequently as minorVH. Use of hybrids not regarded as either major or minor genes(Table I) comprise a minority of “other” for both isotypes. Note-worthy from Table IV is that there is no difference between IgM-and IgG-associated VDJs, suggesting that CSR and repertoire di-versification are not linked.

Nonhybridizing transcripts from antigenized piglets havemutations that alter the Ab binding site

Fig. 5 summarizes sequence data on 105 VDJ clones. Data pre-sented show the frequency of SHM to be 5-fold lower in hybrid-izing than nonhybridizing VDJ clones. In the former, there is nodifference between the frequency of SHM in CDRs and FR regionsand no significant difference in mutation frequency in FR regionsbetween the two groups (Fig. 5). Thus, the higher mutation fre-quency in nonhybridizing clones is confined to CDR1 and CDR2.The terminal nine nucleotides of FR3 are known to be influencedby combinatorial joining, so the frequency of mutation in the ter-minal three codons of FR3 or in CDR3 cannot be distinguishedfrom nucleotide changes that result from junctional diversity. Ev-idence suggesting that changes in 3� FR3 are the result of junc-tional diversity is based on evidence that codon sequences in thisregion do not differ between hybridizing and nonhybridizingclones while major changes in CDR regions are observed. The

Table IV. Characterization of VDJ clones classified as “other”

a Total mutations include FRs and CDRs and are expressed as mutations per kilobase. See Fig. 5 for distribution of mutations.b Major VH genes are VHA, VHB, VHC, VHE, VHF, VHZ, and VHY. See Table I.c Minor VH genes are VH genes that have been recovered from DNA and/or BACs or repeatedly as full-length transcripts

(see Table I).d Three sequences could not be ascribed to any known VH gene and could be artifactual mosaics.e Mean DH region length. Values in parentheses are percent usage.f Proportion of clones using DHC or an unidentified DH gene.g Includes major and minor VH genes.

5466 Ab REPERTOIRE DEVELOPMENT IN FETAL AND NEONATAL PIGLETS

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

frequency of transitions and transversions in FR regions of bothhybridizing and nonhybridizing clones and CDRs of nonhybridiz-ing clones ranged from 6.5–14.7/kb and differences were not sig-nificant. The ratio of transitions (44.5 � 33.6) to transversions(50.8 � 38.7) in CDRs of nonhybridizing clones and in hybridiz-ing clones (14.2 � 19.6 to 10.2 � 14.6) was the same although thefrequency was, of course, 5-fold higher in nonhybridizing clones(Fig. 5).

Notable in the data presented in Fig. 5 is that FR2 is essentiallynonmutated in all sequences (statistics considered). This validatesour routine use of FR2 as a pan-specific marker for porcine VH

genes (Fig. 1; Refs. 2–5).

The porcine VH genome contains hybrid VH genes, suggesting amechanism for genomic diversification

Early studies in our laboratory revealed the presence of porcine VH

genes containing the CDR1 region of a well-known VH gene anda CDR2 region of another. Because these chimeras were usuallyrecovered by using PCR somewhere in the process, the possibilitythat they could be PCR-generated artifacts was examined in vitrostudies (16). These studies revealed that to recover a significantfrequency of hybrids, e.g., 38%, a high concentration of donortemplates, e.g., a 1:1 ratio of two templates, was needed. Becausesuch a contrived situation is an unlikely in vivo event, we consid-ered the alternative that some hybrid VH genes were alreadypresent in the germline. The ultimate answer of course rests on acomplete characterization of the porcine VH genome. Because theswine is not a priority species for basic research, ultimate resolu-tion may require decades. In lieu of such evidence, we analyzedseveral VH-containing BAC clones. Table V lists the occurrence ofVH genes in four different VH-containing BAC clones. Noteworthyis that the VHC/VHA combination was recovered from BAC 352,but VHC (C/C; a potential donor of CDR1 of VHC) was absent.Similarly, a VHG/VHE hybrid was recovered from BAC 663 and1084 but a donor sequence that could be a CDR1 of VHG was notrecovered. Furthermore, VHG/VHE could be recovered from BAC747 but neither potential parent (VHG or VHE) was recovered. Asimilar situation occurred for VHE/VHF in BAC 747 that containedneither VHE nor VHF that could serve as a donor for a PCR artifact.These data indicate that VHG/VHE (VHG, Table I), VHE/VHF (VHT,Table I), and VHC/VHA (VHY, Table I) are present as germlinegenes so their expression is most likely not a PCR artifact.

The porcine VH repertoire

Table I lists all the VH gene sequences that have been reportedfrom our own laboratory and those by Kim. The investigator col-umn of Table I indicate which research group (B or K) has iden-tified the sequence. It is presumed that Kim also observed themajor VH genes (indicated by an asterisk) but did not report thembecause they were already in GenBank. Noteworthy is that manyreported porcine VH genes are chimeras or hybrids as regard thesequences of their CDR1 and CDR2 regions. Because the FR re-

gions of the porcine VH3 family genes are virtually identical (Fig.1), only their CDR regions, designated by Arabic capitals are in-dicated. The actual sequences of these CDRs are given at the bot-tom of Table I. The data of Table I suggest that the porcine VH

repertoire developed by duplication and exchanges of CDR seg-ments from a small number of precursors. For example there areonly nine unique CDR1 sequences and 15 unique CDR2 sequencesamong 29 reported VH genes. Based on their occurrence in thepreimmune repertoire (Figs. 2 and 4), major VH genes are desig-nated with an asterisk.

DiscussionThe isolator piglet model has provided insight into in vivo aspectsof developmental immunology during the critical window of post-natal life (21, 22). The model has demonstrated its value in un-derstanding the role of colonization and pathogen-associated mo-lecular patterns in stimulating the development of adaptiveimmunity (18, 23) and the role of neonatal viruses on developmentof immune responses (24, 25). This has been in part possible be-cause of a user-friendly feature of swine in that they use only afew, well-defined VH genes to form their preimmune VH reper-toire. Because this repertoire changes with age and exposure toenvironmental factors (Figs. 3 and 4), it provides a means for po-tentially quantifying the degree of Ab repertoire development thatoccurs following different environmental stimuli. In this revisitedstudy of the porcine preimmune repertoire, we have addressed sev-eral issues associated porcine VH genes and the methods used forstudying their usage. These include the following: 1) individualvariation in VH usage; 2) the origin of chimeric VH genes; 3) thefidelity of clonal hybridization; 4) the impact of SHM on hybrid-ization; 5) the characterization of the other category of porcine VH

genes; 6) establishment of a useful index of repertoire diversifica-tion; and 7) estimation of repertoire size.

A danger in the use of outbred animals such as humans or swineis that universal patterns of gene expressed may be masked byindividual variation or differential treatment. Data presented in Fig.3 should belay such concerns for the piglet system as regards VH

usage. These show that all piglets in each treatment group use thesame major VH genes and in similar proportions and that changesin proportional usage are related to environmental influences. Thisobservation is not new. Sun et al. (2) showed that 89% of the VDJrearrangements in fetal piglets used VHA, VHB, VHC, and VHE.Each animal used the same VH genes according to the followingstatistics: VHA (39.2 � 7.1), VHB (7.4 � 0.8), VHC (19.6 � 2.5),VHE (22.2 � 5.1). This pattern of VH usage differs from the patternshown for transcripts in newborns only in the greater usage of VHBand the lower usage of VHE/VHZ. Accepting these differences inVH usage in DNA rearrangements vs transcription, the constancyof usage of a few major VH genes in all animals would be difficultto explain by a somatic gene rearrangement process or a somaticgene conversion progress. However, it may be influenced by the

Table V. Recovery of hybrid VH genes from BAC clonesa

a VH genes are designated A/A etc. to indicate the sequence of CDR1 and CDR2, respectively. See Table I.b Each have a potential parental donor for one CDR region in the same clone.c No potential parental donor for any CDR region is present in the same BAC clone.

5467The Journal of Immunology

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

more limited genetic diversity of swine compared with humans (D.Smith, Baylor University, Waco, TX, unpublished observation).

Our argument that VHZ, VHY, and certain other chimeric genesare not somatic or PCR recombinants, but are present in the germ-line, is based on several criteria. First, Table V shows that chimericgenes can be recovered by PCR from BAC clones that appear tolack donor genes that could produce artifactual hybrids because ofthe low processivity of Taq polymerase. Second, VHZ (E/C hy-brid) and VHY (C/A hybrid) were repeatedly found in the preim-mune repertoire but not their reciprocal combination even thoughthe donor genes, i.e., VHA, VHC, and VHE, are major componentsof the preimmune repertoire (Fig. 2). Assuming that hybrids aris-ing as PCR artifacts are random, the lack or unequal recovery ofreciprocal combination argues against the PCR artifact theory.Therefore, hybrids like VHZ and VHY are most likely germlinegenes. Whether all of the various chimeric combination identified(Fig. 2) and listed in Table I represents germline chimeras, cannotbe determined at this time.

Species like rabbit (10, 12) that also use only a few VH genes orchickens that use one VH gene (9, 13) use templated mutations

(gene conversion) to form their Ab repertoire. The chimeras wereport do not resemble the mosaic gene conversion products re-ported by others (8, 26, 27). Rather, entire CDR1 regions are re-placed, not segments of them as in the case of converted VH genes.Because all newborn piglets use the same seven VH genes in ap-proximately the same proportions, it is unlikely that the swinepreimmune repertoire develops by some somatic gene conversionprocess.

The clonal hybridization system we used in the analysis of�3000 cloned VDJs appears to be reliable in determining VH geneusage. A small number of clones hybridized with three CDRprobes (Table IIA) although all were single clones (Table IIB).Because these represent �5% all clones, this nonspecific hybrid-ization could not impact the overall conclusions.

We knew from limited sequence analysis that clones we iden-tified as “other” were either 1) mutated variants of the major fourVH genes that comprised the VH repertoire or 2) less frequentlyused VH genes (2, 3). Table III shows that VDJ clones of major VH

genes that have on average �0.7 mutations in their CDR probebinding regions are unlikely to hybridize and would therefore be

FIGURE 5. Mutation frequency(Mut/Kb) in VH gene segments ofnonhybridizing (top) and hybridizing(bottom) VDJ clones. The number inparentheses is the number of VDJ se-quences analyzed. The difference inmutation frequency between cloneshybridizing with no probe or onlyone was small so that the results wereaveraged together. The shaded por-tion of the stacked bar graph showsthe proportion of mutations that re-sult in changes in the amino acid se-quence (R mut). The terminal portionof CDR3 most likely includeschanges due to combinatorial joiningas well as SHM. The scattergram onthe extreme right conveys the degreeof individual variation that consti-tutes the mean (horizontal bar). Thepercent value above (81 or 71%) isthe proportion of R mutations. Thevalues above the bracketed bars indi-cate the frequency of transition (Ts)and transversions (Tv) per kilobase inthe appropriately bracketed CDR andFR regions. All sequences have beensubmitted to GenBank.

5468 Ab REPERTOIRE DEVELOPMENT IN FETAL AND NEONATAL PIGLETS

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

classified as “other.” Although there were differences amongprobes in the number of mutations tolerated, the mean number ofmutations in the probe binding region of major, nonhybridizationVH genes was �3.3, i.e., 5-fold higher than among those that hy-bridize. Sequence analysis also showed that the majority of non-hybridizing “other” VH genes were mutated major VH genes, notunknown hybrid VH genes (Table IV). The analyses of the 105VDJ sequences revealed that sequence differences between non-hybridizing and hybridizing clones were concentrated in CDR re-gions and in the 3� FR3 region. Because the frequency of changein 3� FR3 is not statistically different between hybridizing andnonhybridizing clones (Fig. 5) and because this portion of FR3 issubject to change through junctional diversity, variation in FR3 3�is most likely not the result of SHM. As generally reported forother species, nucleotide changes that were unequivocally the re-sult of SHM were more frequent in CDRs than FRs (Fig. 5). Al-though the frequency has been reported to be higher in CDR1 thanCDR2 (28), we observed no significant difference. Most codonchanges, even in hybridizing clones, resulted in changes in CDRprotein sequences so they may potentially alter the Ab binding site.We found no significant differences in the number of codonchanges resulting from transversions vs transitions among anycomparable region as had been reported by Betz et al. (28). How-ever, they studied a transgene that was not under selective pres-sure, whereas most of the highly mutated sequences we studiedcame from animals responding to a strong Th2 stimulus. Thehigher proportion of R:S substitutions especially in CDRs, is gen-erally well-known (29–31) and may not exclusively result fromselection because it is seen in fetal/neonatal lambs never exposedto environmental Ag (32).

Although SHM can relegate a clone to the category “other,”nonhybridizers also include minor VH genes, a trend toward use ofunidentified DH segments and a decreased use of the most JH prox-imal DH (DHB; Table IV). However, progressive nonhybridizationwith probes specific for the major VH genes primarily results fromSHM of major VH genes, but also occurs by use of minor VH

genes but least of all by use of previously unknown chimeras.Data presented confirm our earlier observation that H chain rep-

ertoire diversity was due to CDR3 diversity not combinatorial di-versity (4). Approximately 80% of all DH segments were eitherDHA or DHB, even among VDJ that used minor porcine VH genes(Table IV). This is no longer surprising because transgenic micewith only a single DH segment can generate a diverse repertoire(33) as can nurse sharks using only two segments like swine (34).

In an effort to assess the impact of viral infections, probiotics,colonization, diet, and maternal factors on the neonatal Ab reper-toire, we constructed a first generation RDI (Ref. 5). The originalequation has now been modified using data from �3000 clonedVDJs that use the seven major VH genes comprising �95% ofthe preimmune repertoire. The equation is derived by weightingthe effect of VH gene expression based on the proportional usageof the major VH genes in the preimmune repertoire and consider-ing how the usage of these genes is increased (numerator) or de-creased (denominator) after environmental exposure (see Fig. 4).

RDI �

(0.06 � usage VHZ)� 0.15 � usage “other”) (0.02 � usage VHY�

0.4 � usage VHA� � 0.1 � usage VHB � VHC�� 0.05 � usage VHE � VHF�

� 10

We tested this RDI equation on data collected from the MLN.Helminthic parasites stimulate strong Th2 responses, commensalbacteria stimulate Th1 responses (35) and both should impact theMLN (36–38). Using the RDI index, the preimmune IgM reper-

toire of newborn gives an RDI value of 2.1, whereas a value of 73is generated in parasite-infected pigs. An RDI of only 4.4 (mar-ginally significant) for 5-wk-old piglets monoassociated with E.coli G58-1 vs 2.3 for 5-wk-old GF controls was calculated. Thus,the preimmune IgM repertoire scarcely diversifies with time orwith monoassociation of isolator piglets in E. coli. When valuesfor the RDI were generated for the IgG repertoire, there is moreevidence for age-dependent (2.6–7.0) and colonization-dependent(2.6–9.2) diversification than in the IgM repertoire but differenceswere not significant. The small change in Ab repertoire in mono-associated isolator pigs compared parasite-infected piglets proba-bly reflects not only parasitic infection but also conventional rear-ing in which newborn piglets encounter �30 members of thenormal gut flora. However, the high level of repertoire diversifi-cation in IgM transcripts from parasite-infected piglets was sur-prising because current dogma suggests that most Ags are T de-pendent and should drive responses that involves switchrecombination and the preferential appearance of transcripts withSHM VDJ expressed with IgG or other switched isotypes (39).This suggests that in parasite-infected pigs, diversification, andSHM (Table IV, top) are not coupled to CSR. Mutated IgM se-quences have also been reported in humans (30, 40, 41).

Accepting that many if not all hybrid VH genes in swine areactually germline genes, it is now possible to update informationon the porcine VH repertoire. We adopted a system using Arabicletters with the anticipation that these will be eventually changedto the more standardized system of IGHV3-1, IGHV3-2, etc. (42),when their location in the genome has been established. During theinterim, the A-Z system has been used (Table I). All are membersof a single VH3 family (clan; i.e., IGHV3) and with the exceptionof VHB (Fig. 1), share identical FR regions that allow use of apan-specific FR2 probe for total VH detection (Refs. 1 and 43;Figs. 1 and 5). Therefore, only sequence information for CDRregions is given in Table I. This conservation of FR regions issomewhat greater than human VH3 genes (92 vs 96%). Accordingto this system, the VHE/VHC hybrid gene discussed in this report isdesignated porcine VHZ, hybrid VHA/VHB becomes VHV, and hy-brid VHE/VHF becomes VHT. Assuming that the majority of the 29VH genes (or alleles) listed in Table I are not artifacts of PCR, itsuggests that the porcine VH locus arose from duplications andrearrangements of a small number of prototypic genes that differonly in their CDR regions. This pattern is similar to the VH3 familyin humans in which numerous genes share CDR1 or CDR2 seg-ments (42).

The data provided resolve numerous biological and method-ological issues regarding the porcine VH repertoire. In the process,these studies provide an equation for quantifying the diversifica-tion of this repertoire in piglets exposed to the various environ-mental and maternal factors that impact the developing neonatalmammal. Such quantitative data has already been valuable in gain-ing a better understanding of the lymphoproliferative disordercaused by the porcine respiratory and reproductive syndrome virusand should be valuable in future studies with other neonatal patho-gens that may collectively serve as models for events occurring inhuman infants.

DisclosuresThe authors have no financial conflict of interest.

References1. Sun, J., I. Kacskovics, W. R. Brown, and J. E. Butler. 1994. Expressed swine VH

genes belong to a small VH gene family homologous to human VH III. J. Immu-nol. 153: 5618–5627.

2. Sun, J., C. Hayward, R. Shinde, R. Christenson, S. P. Ford, and J. E. Butler. 1998.Antibody repertoire development in fetal and neonatal piglets. I. Four VH genes

5469The Journal of Immunology

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from

account for 80% of VH usage during 84 days of fetal life. J. Immunol. 161:5070–5078.

3. Sun, J., M. Shey, and J. E. Butler. 1998. Determination of gene usage by differ-ential polymerase chain reaction product hybridization. Anal. Biochem. 260:71–79.

4. Butler, J. E., P. Weber, M. Sinkora, J. Sun, S. J. Ford, and R. Christenson. 2000.Antibody repertoire development in fetal and neonatal piglets. II. Characteriza-tion of heavy chain CDR3 diversity in the developing fetus. J. Immunol. 165:6999–7011.

5. McAleer, J., P. Weber, J. Sun, and J. E. Butler. 2005. Antibody repertoire de-velopment in fetal and neonatal piglets. XI. The thymic B cell repertoire developsindependently from that in blood and mesenteric lymph nodes. Immunology 114:171–183.

6. Knight, K. L., and R. S. Becker. 1990. Molecular basis of allelic inheritance ofrabbit immunoglobulin VH allotypes: implications for the generation of antibodydiversity. Cell 60: 963–970.

7. Jenne, C. N., L. J. Kennedy, P. McCullagh, and J. D. Reynolds. 2003. A newmodel of sheep Ig diversification: shifting the emphasis toward combinatorialmechanisms and away from hypermutation. J. Immunol. 170: 3739–3750.

8. McCormack, W. T., L. W. Tjoelker, and C. B. Thompson. 1991. Avian B-celldevelopment: generation of an immunoglobulin repertoire by gene conversion.Annu. Rev. Immunol. 9: 219–241.

9. Reynaud, C. A., A. Dahan, V. Anquez, and J. C. Weill. 1989. Somatic hyper-conversion diversifies the single VH gene of the chicken with a high incidence inthe D region. Cell 59: 171–183.

10. Knight, K. L. 1992. Restricted VH gene usage and generation of antibody diver-sity in rabbit. Annu. Rev. Immunol. 10: 593–616.

11. Pospisil, R., G. O. Young-Cooper, and R. G. Mage. 1996. Preferential expansionand survival of B-lymphocytes based on VH framework 1 and framework 3 ex-pression: positive selection in appendix of normal and VH mutant rabbits. Proc.Natl. Acad. Sci. USA 92: 6961–6965.

12. Becker, R. S., and K. L. Knight. 1990. Somatic diversification of immunoglobulinheavy chain V(D)J genes: evidence for somatic gene conversion in rabbit. Cell63: 987–997.

13. Ratcliff, M. J. H. 2006. Antibodies, immunoglobulin genes and the bursa offabricus in chicken B cell development. Dev. Comp. Immunol. 30: 101–118.

14. Meyerhans, A., J. P. Vartanian, and S. Wain-Hobson. 1990. DNA recombinationduring PCR. Nucleic Acids Res. 18: 1687–1691.

15. Johnson, K. A. 1993. Conformational coupling in DNA polymerase fidelity.Annu. Rev. Biochem. 62: 685–713.

16. Sinkora, M., J. Sun, and J. E. Butler. 2000. Antibody repertoire development infetal and neonatal piglets. V. VDJ gene chimeras resembling gene conversionproducts are generated at high frequency by PCR in vitro. Mol. Immunol. 37:1025–1034.

17. Kraj, P., S. P. Rao, A. M. Glas, R. R. Hardy, E. C. B. Milner, andL. E. Silverstein. 1997. The human heavy chain Ig V region gene repertoire isbiased at all stage of B cell ontogeny, including early pre-B cells. J. Immunol.158: 5824–5837.

18. Butler, J. E., P. Weber, M. Sinkora, D. Baker, A. Schoenherr, B. Mayer, andD. Francis. 2002. Antibody repertoire development in fetal and neonatal piglets.VIII. Colonization is required for newborn piglets to make serum antibodies toT-dependent and type 2 T-independent antigens. J. Immunol. 169: 6822–6830.

19. Butler, J. E., J. Sun. P. Weber, S. P. Ford, Z. Rehakova, J. Sinkora, and K. Lager.2001. Antibody repertoire development in fetal and neonatal piglets.IV. Switch recombination, primary in fetal thymus occurs independent of envi-ronmental antigen and is only weakly associated with repertoire diversification.J. Immunol. 167: 3239–3249.

20. Butler, J. E., N. Wertz, J. Sun, and R. Sacco. 2005. Characterization of theporcine V� repertoire in thymocytes versus peripheral T-cells. Immunology 114:184–193.

21. Butler, J. E., M. Sinkora, N. Wertz, W. Holtmeier, and C. D. Lemke. 2006.Development of the neonatal B- and T-cell repertoire in swine: implications forcomparative and veterinary immunology. Vet. Res. 37: 417–441.

22. Butler, J. E., J. Sun, N. Wertz, and M. Sinkora. 2006. Antibody repertoire de-velopment in swine. In Antibody Repertoire Development: Developmental andComparative Immunology, Vol. 30. J. E. Butler, ed., pp. 199–221.

23. Butler, J. E., D. Francis, J. Freeling, P. Weber, J. Sun, K. L. Nielsen, andA. M. Krieg. 2005. Antibody repertoire development in fetal and neonatal piglets.

IX. Three PAMPs act synergistically to allow germfree piglets to respond to TI-2and TD antigens. J. Immunol. 175: 6772–6785.

24. Lemke, C. D. 2006. PRRSV infection of neonatal piglets induces immune dys-regulation and modulation. Doctor of Philosophy thesis, University of Iowa, IowaCity, IA.

25. Lemke, C. D., J. S. Haynes, R. Spaete, D. Adolphson, A Vorwald, K. Lager, andJ. E. Butler. 2003. Lymphoid hyperplasia resulting in immune dysregulation iscaused by PRRSV infection in pigs. J. Immunol. 172: 1916–1925.

26. D’Avirro, N., D. Truong, B. Xu, and E. Selsing. 2005. Sequence transfer betweenvariable regions in a mouse transgene can occur by gene conversion. J. Immunol.175: 8133–8137.

27. Winstead, C. R., S.-K, Zhai, P. Sethupathi, and K. L. Knight. 1999. Antigeninduced somatic diversification of rabbit IgH genes: gene conversion and pointmutation. J. Immunol. 162: 6602–6612.

28. Betz, A. G., C. Rada, C. Pannell, C. Milstein, and M. S. Heuberger. 1993. Pas-senger transgenes reveal intrinsic specificity of the antibody hypermutation mech-anism: clustering, polarity and specific hot spots. Proc. Natl. Acad. Sci. USA 90:2385–2388.

29. Mehr, R., H. Edelman, D. Sehgal, and R. Mage. 2004. Analysis of mutationallineage trees from sites of primary and secondary gene diversification in rabbitsand chickens. J. Immunol. 172: 4790–4796.

30. Fais, F., F. Ghiotto, S. Hashimoto, M. Dono, A. Valetto, S. L. Allen, P. Shulman,V. P. Vineiguerra, K. Rai, M. Ferrarini, and N. Chiorazzi. 2000. ImmunoglobulinV gene sequence analysis distinguish two subsets of chronic lymphocytic leuke-mia B cells. In The Antibodies, Vol. 6. M. Zanetti and J. D. Capra, eds. AcademicPress, New York, pp. 123–150.

31. Shlomchik, M. J., A. H. Aucoin, D. Asetsky, and M. G. Weigert. 1987. Structureand function of anti-DNA autoantibodies derived from a single autoimmunemouse. Proc. Natl. Acad. Sci. USA 84: 9150–9154.

32. Reynaud, C. A., C. R. Mackay, R. G. Muller, and J. C. Weill. 1991. Somaticgeneration of diversity in a mammalian primary lymphoid organ: the sheep ilealPeyer’s patches. Cell 64: 995–105.

33. Schelonka, R. L., I. I. Ivanov, D. H. Jung, G. C. Ippolito, L. Nitschke, Y. Zhuang,G. L. Gartland, J. Pelkonen, F. W. Alt, K. Rajesky, and H. W. Schroeder, Jr.2005. A single DH gene segment creates its own unique CDR-H3 repertoire andis sufficient for B cell development and immune function. J. Immunol. 175:6624–6632.

34. Malecek, K., J. Brandman, J. E. Brodsky, Y. Ohata, M. F. Flajnik, and E. Hsu.2005. Somatic hypermutation and junctional diversification at Ig heavy chain lociin the nurse shark. J. Immunol. 175: 8105–8115.

35. Elliott, D. E., T. Setiawan, A. Metwali, A. Blum, J. F. Urban, Jr., andJ. V. Weinstock. 2004. Heligmosomoides polygyrus inhibits established colitis inIL-10-deficient mice. Eur. J. Immunol. 34: 2690–2698.

36. Liu, L. M., and G. G. MacPherson. 1993. Antigen acquisition by dendritic cells:intestinal dendritic cells acquire antigen administered orally and can prime naiveT cells in vivo. J. Exp. Med. 177: 129–1307.

37. Kelsall, B. L., F. Leon, L. E. Smythies, and P. D. Smith. 2004. Antigen handlingand presentation by mucosal dendritic cells and macrophages. In Mucosal Im-munology, 3rd Ed. J. Mestecky, M. E. Lamm, W. Strober, J. Bienenstock,J. R. McGhee, and L. Mayer, eds. Academic Press, New York, pp. 451–485.

38. Rothkotter, H. J., C. Krieski, N. N. Barman, and R. Pabst. 1999. B and also Tlymphocytes migrate via gut lymph to all lymphoid organs and the gut wall, butonly IgA cells accumulate in the lamina propria of the intestinal mucosal. Eur.J. Immunol. 29: 327–333.

39. Janeway, C. A., Jr., P. Travers, M. Walport, and M. J. Shlomchik, eds. 2005.Immunobiology, 6th Ed. Garland Press, New York, p. 821.

40. Klein, U., K. Rajewsky, and R. Kappers. 1998. Human immunoglobulin IgM

IgD peripheral blood B cells expressing the CD27 cell surface antigen carrysomatically mutated variable region genes: CD27 as a general marker for somat-ically mutated (memory) B cells. J. Exp. Med. 188: 1679–1689.

41. van Es, J. H., F. H. Meylng, and T. Logtenberg. 1992. High frequency of mutatedIgM molecules in the human adult blood B cell repertoire. Eur. J. Immunol. 22:2761–2764.

42. Lefranc, M.-P., and G. Lefranc. 2001. The Immunoglobulin Facts Book. Aca-demic Press, New York, pp. 457.

43. Sun, J., and J. E. Butler. 1996. Molecular characteristics of VDJ transcripts froma newborn piglet. Immunology 88: 331–339.

5470 Ab REPERTOIRE DEVELOPMENT IN FETAL AND NEONATAL PIGLETS

by guest on October 4, 2017

http://ww

w.jim

munol.org/

Dow

nloaded from