Isolator piglets infected with porcine reproductive and respiratory syndrome virus (PRRSV), which is related to the lactate dehydrogenase-elevating virus of mice, develop severe hypergammaglobulinemia, lymph node adenopathy, and autoimmune disease. Many of the polyclonally activated B cell clones bear hydrophobic H chain CDR3s (HCDR3s) and are disseminated to most lymphoid tissues. We show in this study that B cells with identical hydrophobic HCDR3s are expressed with all major isotypes in PRRSV-infected piglets (PIPs), explaining why PRRSV-induced hypergammaglobulinemia is seen in all major isotypes. Up to one-third of randomly selected VDJ clones from the respiratory tract of PIPs have hydrophobic HCDR3s exclusively bearing VDJ rearrangements with CDR1, CDR2, and nearly intact DH segments in germline configuration. These HCDR3s are long and DHA and DHB are exclusively used in reading frame 3. A minimal tripeptide motif containing three hydrophobic amino acids (Leu, Val, and Ile) or any two plus alanine is common to this hydrophobic patch. We propose that PRRSV infection causes generalized Ag-independent B cell activation and hypergammaglobulinemia with biased expansion of a subpopulation of the preimmune repertoire with hydrophobic binding sites that normally disappears during Ag-driven repertoire diversification. Elevated Ig levels in PIP cannot be explained as antiviral Abs; some Igs can account for autoantibodies to dsDNA and Golgi, whereas those with hydrophobic binding sites may account for the Ig aggregates seen in PIPs and lactate dehydrogenase-elevating virus-infected mice. This diversion from normal repertoire development may explain the delayed immune response to PRRSV.

Porcine reproductive and respiratory syndrome (PRRS)3 is a world pandemic causing major economic loss to the swine industry (1). The etiological agent of this disease, the PRRS virus (PRRSV), is an enveloped, positive, single-stranded RNA virus that is a member of the Arteriviridae family in the order Nidovirales (2, 3). PRRSV can persist for 150 days (4) whereas swine influenza virus (SIV) infection, which also produces lung lesions in newborn piglets, is resolved within 3 weeks. Using the isolator piglet model we have shown that PRRSV causes immune dysregulation, a phenomenon not observed with SIV. This is manifest by serious lymphoid hyperplasia, hypergammaglobulinemia, and autoimmunity (see Fig. 1 and Ref. 5). More recently, we have shown that this syndrome is also associated with the apparent dissemination of certain B cell clones throughout the body (see Fig. 2) that preferentially bear BCRs with hydrophobic H chain CDR3 (HCDR3) regions (6). Skewing of hydropathicity profiles toward B cells with hydrophobic binding sites is unusual in mice and humans (7, 8, 9) and was not observed in littermates infected with SIV, sham controls, young pigs infected with helminthic parasites, or in newborn piglets (6). However, that study did not randomly sample B cell clones, so the size of the subpopulation could not estimated and the source of the expanded clones with regard to either VH gene usage or isotype expression was not identified. In the current study we addressed two issues. First, we characterized expanded clones bearing HCDR3s of the same length that were expressed with IgM, IgG, and IgA in respiratory tissues of PRRSV-infected isolator piglets (PIPs), those of SIV-infected piglets (IIPs), and those of sham-inoculated littermates. We show that in PIPs the same parent B cell clones express IgG, IgM, and IgA transcripts. Second, we show that one-third of randomly selected VDJ clones using VHA, VHZ, and VH“other” (55–70% of total repertoire: Ref. 10) in PIPs are comprised of those with VDJs in which CDR1, CDR2, and DHA or DHB are in germline configuration (nonmutated) and bear long hydrophobic HCDR3s. This subpopulation is also represented in the preimmune repertoire but disappears after Ag exposure. Although the mechanism for selectively expanding this segment of the preimmune repertoire remains unknown, we believe that this subversion of normal repertoire development may explain the delay in the development of effective PRRSV-specific adaptive immunity.

FIGURE 1.

Ig levels in plasma and BAL at three time points/periods postinfection (dpi) in isolator piglet littermates infected with PRRSV, SIV, PCV-2, and sham inoculant. Error bars denote SD. A split y-axis is needed to accommodate both high and low values. Asterisks indicate Ig levels in PIP that are significantly higher than in all other treatment groups (p > 0.05–0.001, depending on comparison). Differences among SIV, PCV-2, and sham were not statistically evaluated. Levels in BAL are expressed relative to total protein as a partial correction for transudation.

FIGURE 1.

Ig levels in plasma and BAL at three time points/periods postinfection (dpi) in isolator piglet littermates infected with PRRSV, SIV, PCV-2, and sham inoculant. Error bars denote SD. A split y-axis is needed to accommodate both high and low values. Asterisks indicate Ig levels in PIP that are significantly higher than in all other treatment groups (p > 0.05–0.001, depending on comparison). Differences among SIV, PCV-2, and sham were not statistically evaluated. Levels in BAL are expressed relative to total protein as a partial correction for transudation.

Close modal
FIGURE 2.

A, Spectratypic analysis of HCDR3s expressed with IgG from PIPs and various controls. B, Scanograms (profiles) of the spectratypyes in A. BLN, tracheal bronchial lymph node; BM, bone marrow; IPP, ileal Peyer’s patch; JPP, jejunal Peyer’s patch; Lg, lung; MLN, mesenteric lymph node; Spl, spleen. Prominent shared same -ength HCDR3s in tissues are connected with vertical lines.

FIGURE 2.

A, Spectratypic analysis of HCDR3s expressed with IgG from PIPs and various controls. B, Scanograms (profiles) of the spectratypyes in A. BLN, tracheal bronchial lymph node; BM, bone marrow; IPP, ileal Peyer’s patch; JPP, jejunal Peyer’s patch; Lg, lung; MLN, mesenteric lymph node; Spl, spleen. Prominent shared same -ength HCDR3s in tissues are connected with vertical lines.

Close modal

All animal studies were conducted at the National Animal Disease Center (NADC) in Ames, IA. Piglets were recovered by closed hysterectomy from 113-day gravid outbred swine, placed in groups of four in rigid tub isolators, and reared on ESPLac (PetAg) as previously described (5, 11). All studies were approved by the Animal Care and Use Committee of the NADC. Groups of 3–5 piglets from the same litter were maintained germfree and treated one of three ways: 1) inoculated intranasally with 104 TCID50 PRRSV (where TCID50 is 50% tissue culture-infective dose); 2) inoculated intranasally with 104 TCID50 SIV; or 3) given a sham inoculum. All inoculations took place on day 7 (0 days postinoculation (dpi)). Animals were maintained for up to 46 days postinoculation (dpi) but most were necropsied at 10 and 25 dpi. SIV-infected (Ia30 strain) animals develop pronounced lung lesions at 10 dpi, but these are resolved on or before 25 dpi. These animals mount virus-neutralizing Ab responses (12) and are considered to have developed sterilizing immunity. The isolator piglet model is reviewed elsewhere (13, 14). A number of piglets were euthanized at delivery to obtain baseline data on the preimmune repertoire of newborns. Mesenteric lymph nodes and PBMCs from parasite-infected conventional (PIC) young pigs were kindly provided by Dr. J. Urban, Jr, U. S. Department of Agriculture, Agricultural Research Service, Beltsville, MD.

Isolator piglets were euthanized with pentobarbital (Sleepaway; Fort Dodge Laboratories). At necropsy, lungs were lavaged with 50 ml of saline. Because both PRRSV and SIV are respiratory pathogens, focus was on the lung and associated lymph nodes. This bronchoalveolar lavage (BAL) was centrifuged at 1000 × g for 15 min to recover the cellular function and the supernatant was later used for Ig and protein analysis (Fig. 1). In PIPs, 2.1 × 107 B cells were routinely recovered from BAL and 5.9 × 106 from the BAL of IIP. The BAL cell pellet was suspended in saline for flow cytometric studies and the remainder was stored in TRIzol at −20°C. Blood collected in anti-coagulation tubes was processed as previously described for the recovery of PBMCs and plasma (5, 15). The plasma was used for Ig determinations (Fig. 1) and was cultured for virus as described previously (12). Also at necropsy, various solid lymphoid tissues such as tracheal bronchial lymph nodes were also collected and frozen in liquid nitrogen.

VDJs were cloned to test two separate hypotheses: first to test whether in the same tissue the same clones were expressed with IgM, IgG, and IgA, and second to determine the frequency of VDJ clones bearing hydrophobic HCDR3s through random sampling. For both studies solid tissues preserved at −70°C and BAL cells in TRIzol were processed for the preparation of total RNA and subsequently cDNA (5, 16). In BAL from PIPs the cDNA used in PCR represented 4 × 106 B cells. VDJs expressed on IgM, IgA, and IgG transcripts were recovered using appropriate primer sets as described previously (17, 18, 19). These VDJs were then cloned, individual clones were grown out in microtiter wells, and their plasmid DNA was purified and then immobilized on nylon membranes. The VH usage of each clone was determined by sequential hybridization with probes specific for seven VH genes that account for >95% of the preimmune repertoire (7). “Other” VH means a VH-containing clone that does not hybridize with CDR-specific probes for the seven VH genes; mostly because of somatic hypermutation (SHM) (7). Newborn piglets, which express a nondiversified preimmune repertoire, and PIC pigs, which give strong Th2 responses, were included for comparison.

Only a single JH (20), seven VH genes, and two DH segments comprise 80% of the swine preimmune repertoire (7, 16). Because of this limited combinatorial diversity, the BCR repertoire is overwhelmingly determined by HCDR3 diversity (21). Diversity in HCDR3 lengths is a quasi-clone marker, so spectratypic analysis (HCDR3 length analysis) provides a quasi-clonotypic analysis (6, 22). The HCDR3 segments from VDJs expressed with IgA, IgM, and IgG from the same tissue sample from the same animal were recovered by PCR and the products were spectratyped side-by-side so that the profiles for IgA, IgM, and IgG could be compared. Same-length HCDR3 polynucleotides (pnt) shared among isotypes were recovered from polyacrylamide spectratyping gel segments using a scalpel and placed into 100 μl of dH2O (6). These were incubated overnight to permit diffusion of the DNA and the fluid phase was targeted for 15 cycles of PCR with the primer set used for the original amplification of HCDR3. These HCDR3s were cloned and sequenced.

In a previous study (6) we analyzed only HCDR3s from clones expanded in PRRS. In this study we randomly selected approximately equal numbers of VDJ clones using VHA, VHZ, and “other” VH genes that were expressed with IgM, IgA, and IgG (Table I). These three gene categories account for 55–75% of all VH usage in PIPs. DNA from selected clones was amplified, products of the expected length were verified by agarose gel electrophoresis, and the gel segments were recovered and cloned into pCR2 TOPO (Invitrogen Life Technologies). Plasmid DNA was sequenced using the four-color ABI PRISM DNA analyzer (Applied Biosystems). Nucleotide sequences were analyzed for VH and DH usage and SHM using the Omiga program (Accelrys). The deduced amino acid sequence of the HCDR3 region, extending from but not including the framework 3 cysteine at position 104, down to the 5′ region of JH, but not including the invariant tryptophan that starts framework 4, were analyzed as described by Kyte and Doolittle (23) using the principles described by Eisenberg (24) as applied to HCDR3 sequences by Ivanov et al. (9). Sequence data were used to compute a hydropathicity index (H.I.). Based on inspection of sequences and calculations of H.I., it became clear that HCDR3s dominated by L,V, I, M, and A and without R, K, and D would have H.I. > 5.0, whereas those dominated by R, D, K, Y, and H would be strongly hydrophilic (H.I. = −0.4 to −0.1). Sequences containing mixtures of these amino acids and those of neutral charge belong to region II (0.0–0.3). HCDR3 sequences from randomly selected VDJ clones of known VH and isotype usage from PIPs and IIPs were compared with HCDR3s of randomly selected VDJ clones from newborn piglets and PIC young pigs (6, 7).

Table I.

The distribution of VDJ clones selected for studya

PRRSV-Infected Piglets
Isotype UsedVH GeneTotal Clones
VHAVHZOther
IgA 22 
IgG 21 
IgM 11 27 
PRRSV-Infected Piglets
Isotype UsedVH GeneTotal Clones
VHAVHZOther
IgA 22 
IgG 21 
IgM 11 27 
SIV-Infected Piglets
Isotype UsedVH GeneTotal Clones
VHAVHZOther
IgA 24 
IgG 10 25 
IgM 21 
SIV-Infected Piglets
Isotype UsedVH GeneTotal Clones
VHAVHZOther
IgA 24 
IgG 10 25 
IgM 21 
a

Sequence analyses revealed that, in the case of PRRS, all clones picked as “other” and with hydrophobic HCDR3s were VHA clones that, for unexplained reasons, did not originally hybridize for both VHA-specific CDR1 and CDR2 probes.

IgM, IgA, and IgG levels in blood plasma and BAL were determined by sandwich ELISA as previously described (25).

Simple mean differences were compared by two-sided Student’s t analysis using the stats program of Prism (GraphPad Software). Statistical analyses of HCDR3 distribution among hydropathicity regions was done with the help of Dr. K. Chaloner (Head of the Department of Biostatistics, University of Iowa College of Public Health, Iowa City, IA) using the method of Holm (26), and exact tests were implemented by the R Development Core Team of Vienna Austria (www.R-project.org).

Fig. 1 shows that plasma IgM, IgG, and IgA are elevated >10-fold in PIPs compared with those infected with SIV and porcine circovirus-2 (PCV-2). Thus, only PRRSV induces the hypergammaglobulinemia we have previously described (5), indicating that it is specific to PRRSV and not a generic effect of the viral infection of isolator piglets. The kinetics of the hypergammaglobulinemia differs among isotypes. IgM levels are elevated at 10 dpi but serum IgG and IgA levels peak at 25 dpi. Significantly elevated levels are also present in the BAL fluid of PIPs, although peak levels of all Igs are maintained longer in BAL (e.g., 46 dpi).

Fig. 2 demonstrates that the Gaussian-like polyclonal HCDR3 length profile of bone marrow shows evidence of selection in the mesenteric lymph nodes of conventional control pigs. In PIPs, the quasi-oligoclonal IgG spectratype is shared by nearly all tissues including those remote from the site of infection (Fig. 2 and Ref. 6). Fig. 3, left panel, shows that an oligoclonal pattern shared among isotypes in PIPs begins to appear at 10 dpi but does not spread to all isotypes until 25 dpi. At that time IgG, IgM, and IgA display nearly identical oligoclonal HCDR3 length profiles. This pattern of shared same-length HCDR3s is sometimes seen out to 46 dpi (Fig. 3, lower left panel). When isotype HCDR3 profiles from IIPs and-sham inoculated controls were studied the pattern is also quasi-oligoclonal but, with few exceptions, IgG, IgM, and IgA do not display the same profile as that seen in PIPs (Fig. 3 B).

FIGURE 3.

Left panel, Scan profiles of HCDR3s for IgG (G), IgM (M), and IgA (A) from the same tissue of PIPs at various times postinfection (dpi). Right panel, Scan profiles for different isotypes from sham-inoculated and SIV-infected (IIPs) littermates at various times postinfection. Same-length HCDR3s present in all isotypes are connected by vertical lines. BAL, BAL cells; BLN, tracheal bronchial lymph node cells. The 60-pnt HCDR3 length recovered for cloning and sequencing (Table II) is indicated by a small vertical bar and star.

FIGURE 3.

Left panel, Scan profiles of HCDR3s for IgG (G), IgM (M), and IgA (A) from the same tissue of PIPs at various times postinfection (dpi). Right panel, Scan profiles for different isotypes from sham-inoculated and SIV-infected (IIPs) littermates at various times postinfection. Same-length HCDR3s present in all isotypes are connected by vertical lines. BAL, BAL cells; BLN, tracheal bronchial lymph node cells. The 60-pnt HCDR3 length recovered for cloning and sequencing (Table II) is indicated by a small vertical bar and star.

Close modal

Sharing the same profile among spectratypes generated from different isotypes only suggests shared clonality, because many different B cell clones can have HCDR3s of the same length. This assumption was tested by the recovery of prominent HCDR3s of 60 pnt in length that were shared by all three isotypes (Fig. 3, left panel; vertical bar and star). These were cloned and sequenced as previously described (6). We also choose same-length HCDR3s that were prominent and shared among isotypes in IIPs (42 pnt) and those that fit the same criterion for sham-inoculated animals (36 pnt). The results of analyzing a dozen clones of each isotype for each treatment group are summarized in Table II. Those recovered from PIPs differ from controls (IIPs and sham) in several ways. Most obvious is the use of DHA in reading frame (RF) 3 (65% for PIPs vs 15% for IIP and sham). Overall use of DHA RF3 plus DHB RF3 was significantly greater in PIPs (74 ± 11% vs 33 ± 9%; p < 0.01). Usage of RF3 in PIPs was heavily skewed to favor DHA 8:1 over DHB. Although the majority of DH usage was RF3 in PIPs, the majority of VDJs in IIFs and sham controls used DHA and DHB in RF2 (data not shown).

Table II.

Characteristic of clones from same length HCDR3s expressed with IgM, IgG, and IgA

PIPs
CharacteristicIgMIgGIgA
Identical clonesa 1/12 5/13 + 2/13 + 2/13 3/13 + 4/13 
Unique clones 11 
DHA in RF3 
 Length of germlineb 7.4 ± 1.9 7.2 ± 2.7 6.4 ± 1.9 
DHB in RF3 
 Length of germlineb NA NA 5.0 ± 0 
DHA RF3 + DHB RF3 64% 72% 87% 
Shared with other isotypes 
PIPs
CharacteristicIgMIgGIgA
Identical clonesa 1/12 5/13 + 2/13 + 2/13 3/13 + 4/13 
Unique clones 11 
DHA in RF3 
 Length of germlineb 7.4 ± 1.9 7.2 ± 2.7 6.4 ± 1.9 
DHB in RF3 
 Length of germlineb NA NA 5.0 ± 0 
DHA RF3 + DHB RF3 64% 72% 87% 
Shared with other isotypes 
IIPs
CharacteristicIgMIgGIgA
Identical clonesa 3/7 3/8 2/14 
Unique clones 13 
DHA in RF3 
Length of germlineb NA 5.0 ± 0 7.3 ± 2.1 
DHB in RF3 
Length of germlineb 5.50.7 6.0 5.6 ± 0.6 
DHA RF3+ DHB RF3 40% 33% 46% 
Shared with other isotypes 
IIPs
CharacteristicIgMIgGIgA
Identical clonesa 3/7 3/8 2/14 
Unique clones 13 
DHA in RF3 
Length of germlineb NA 5.0 ± 0 7.3 ± 2.1 
DHB in RF3 
Length of germlineb 5.50.7 6.0 5.6 ± 0.6 
DHA RF3+ DHB RF3 40% 33% 46% 
Shared with other isotypes 
Sham
CharacteristicIgMIgGIgA
Identical clonesa 3/14 + 4/14 + 2/14 + 3/14 6/15 5/9 
Unique clones 10 
DHA in RF3 
Length of germlineb 6.0 ± 1.4 5.0 ± 0 NAc 
DHB in RF3 
Length of germlineb NAc 4.0 ± 1.4 5.0 
DHA RF3 + DHB RF3 33% 30% 20% 
Shared with other isotypes 
Sham
CharacteristicIgMIgGIgA
Identical clonesa 3/14 + 4/14 + 2/14 + 3/14 6/15 5/9 
Unique clones 10 
DHA in RF3 
Length of germlineb 6.0 ± 1.4 5.0 ± 0 NAc 
DHB in RF3 
Length of germlineb NAc 4.0 ± 1.4 5.0 
DHA RF3 + DHB RF3 33% 30% 20% 
Shared with other isotypes 
a

The proportion of identical clones to the total; e.g., in the case of IgG, 5/13 share one sequence, 2/13 share another, and still 2/13 share another sequence. Clones that differ by one nucleotide are considered identical. Of the total, only 2.6 % had a one-nucleotide change.

b

DHA in RF3 encodes 11 amino acids and DHB in RF3 encodes eight amino acids. In PIP, 64% of DHA is full length. Too few DHA RF 3 clones were recovered from IIP and sham to allow a comparison.

c

Not applicable since there are no clones.

A high proportion of duplicate clones were recovered in IgA and IgG transcripts from PIPs and from all isotypes in IIPs and shams (Table II). Notably, half of the unique IgA and IgG clones in PIPs were common to both, and some of these were also found in IgM transcripts. By contrast, this was not seen in IIPs and only once in sham animals. We interpret this to mean that: 1) 30–50% of same-length HCDR3s from all treatment groups come from the same B cell clone; but 2) only in PIPs are ∼30% of these shared among isotypes.

One-third of randomly selected VDJ clones from newborns (preimmune repertoire) comprise hydropathicity region I (H.I. = 0.5–0.8) whereas after Ag exposure, e.g., PIC young pigs, there is nearly complete loss of region I and a significant increase in region III (Fig. 4,A). Approximately 40% of VDJ clones randomly selected from PIP also appear in region I, which is significantly greater than the percentage for randomly selected IIP clones (Fig. 4,B). The profile for IIP resembles that of PIC pigs. Assuming that the number of B cell clones corresponds to the amount of secreted Ig, Fig. 4 C demonstrates that IgG with hydrophobic HCDR3 in PIPs accounts for more IgG than the total concentration of IgG in antigenized PIC young pigs. Predicted levels of IgG carrying HCDR3 fitting to region II in PIPs are 2-fold higher that in PIC pigs.

FIGURE 4.

A, Hydropathicity profiles for randomly selected VDJ clones from newborn piglets and PIC young pigs. B, Hydropathicity profile of 70 randomly selected VDJ clones from PIP and IIP littermates. Hydropathicity profiles are divided into region I (hydrophobic; 0.5–0.8), II (neutral to slightly hydrophobic; 0.0–0.35) and III (hydrophilic; −0.4 to −0.1). In the legend the regions in which distributions between the two groups significantly differ are shown in the associated boxes. The stacked bar graphs indicate what proportion of the VDJ sequences use “other” VH, meaning they are diversified either by SHM or the use of seldom-used VH genes. C, The predicted plasma IgG concentration in PIP and IIP calculated from the total IgG concentration at 25 dpi (Fig. 1) and compared with newborn and 4-mo-old PICs. IgG concentrations are presented to the right of the boxes in the legend on the left (NB, newborn). The calculation assumes that the proportion of secreted IgG corresponds to the proportion of B cells with HCDR3s that fit to regions I, II, and III. Error bars = SD. Because the mean IgG concentration for newborns is 0.017 ± 0.006 mg/ml, the error bars cannot be seen. Asterisks indicate that the IgG levels in the PIPs are significantly higher than in the other groups (p < 0.02).

FIGURE 4.

A, Hydropathicity profiles for randomly selected VDJ clones from newborn piglets and PIC young pigs. B, Hydropathicity profile of 70 randomly selected VDJ clones from PIP and IIP littermates. Hydropathicity profiles are divided into region I (hydrophobic; 0.5–0.8), II (neutral to slightly hydrophobic; 0.0–0.35) and III (hydrophilic; −0.4 to −0.1). In the legend the regions in which distributions between the two groups significantly differ are shown in the associated boxes. The stacked bar graphs indicate what proportion of the VDJ sequences use “other” VH, meaning they are diversified either by SHM or the use of seldom-used VH genes. C, The predicted plasma IgG concentration in PIP and IIP calculated from the total IgG concentration at 25 dpi (Fig. 1) and compared with newborn and 4-mo-old PICs. IgG concentrations are presented to the right of the boxes in the legend on the left (NB, newborn). The calculation assumes that the proportion of secreted IgG corresponds to the proportion of B cells with HCDR3s that fit to regions I, II, and III. Error bars = SD. Because the mean IgG concentration for newborns is 0.017 ± 0.006 mg/ml, the error bars cannot be seen. Asterisks indicate that the IgG levels in the PIPs are significantly higher than in the other groups (p < 0.02).

Close modal

Half of the sequences in region II for both PIP (14/32) and IIP (17/39) use “other” VH and, of these, half are somatically hypermutated (data not shown). Region II encompasses the predominant hydropathicity region for HCDR3s from all of the piglets studied (Fig. 4), which is the same for mice and humans (7, 8, 9). Use of “other” VH is a feature of repertoire diversification and in young PIC pigs it comprises 80% (10), whereas only 15% of the preimmune repertoire of newborn uses “other” VH (10, 16, 21). Therefore, both IIP and PIP have undergone repertoire diversification but mostly in region II.

Sixty percent of randomly selected IgM clones from PIPs have hydrophobic HCDR3s (region I; H.I. = 0.5–0.8), which is significantly greater than that in IIP littermates (Fig. 5,A). The hydropathicity profile of HCDR3 recovered from IgG transcripts of PIPs was also significantly shifted to the hydrophobic region (38%) in comparison to that of IIPs (8%; Fig. 5,B), although this shift was not seen in the IgA transcripts from PIP (Fig. 5 C).

FIGURE 5.

The hydropathicity profile expressed as bar graphs of HCDR3s from randomly selected VDJ clones in relationship to Ig isotype for PIP and IIP littermates. Hydropathicity regions I, II, and III are as indicated in Fig. 4. Regions that differ significantly in distribution are indicated in the boxes.

FIGURE 5.

The hydropathicity profile expressed as bar graphs of HCDR3s from randomly selected VDJ clones in relationship to Ig isotype for PIP and IIP littermates. Hydropathicity regions I, II, and III are as indicated in Fig. 4. Regions that differ significantly in distribution are indicated in the boxes.

Close modal

VH usage in PIPs failed to indicate a preferential VH usage associated with this infection (6, 27), thus failing to provide evidence for a conventional B cell superantigen (BSAg) effect (28, 29). In all newborn piglets, VHA expressed with IgM and IgG comprises 25–35% of total VH usage (Table III), but “other VH” contributes up to 45% in PIPs (6, 27). More recent studies have shown that VHZ accounts for 10% of all VH usage (Table III) and that its usage doubles in PIPs (10–20%) while VHB usage sharply declines (Ref. 27 and J. E. Butler, P. Weber, and N. Wertz, unpublished observations). Therefore, we chose to focus on the three VH categories that account for 55–70% of the total repertoire in PIPs: VHA, VHZ, and “other VH”. Fig. 6 shows that the hydrophobic HCDR3 population (region I) expressed with VHZ is significantly over-represented (57%) and that only one “other” VH clone appears in this category. This is a nonmutated VHX (IgG 49E4; Table IV). Although hydrophobic HCDR3s are also expressed with VHA, the proportional usage of VHA does not differ from its usage in IIPs.

Table III.

The protein sequence and H.I. of CDRs of the major VH genes comprising the preimmune repertoire in swine

VH GenePercentage Usage (%)aSequence (H.I.)Combined H.I.
CDR1CDR2
VHA 30 STYIN (0.03) AISTSGC (0.33) 0.204 
VHB 18 DNAFX (−0.03) AIASSDYDG (0.11) 0.067 
VHC SYEIS (0.03) GIYSSGS (0.15) 0.100 
VHE SYAVS (0.39) GIDSGSYIG (0.21) 0.271 
VHF SYGVG (0.23) SIGSGSYIG (0.28) 0.278 
VHY SYEIS (0.03) AISTSGC (0.48) 0.296 
VHZ 10 SYAVS (0.39) GIYSSGS (0.15) 0.250 
Other 21 NAb NAb NAb 
VH GenePercentage Usage (%)aSequence (H.I.)Combined H.I.
CDR1CDR2
VHA 30 STYIN (0.03) AISTSGC (0.33) 0.204 
VHB 18 DNAFX (−0.03) AIASSDYDG (0.11) 0.067 
VHC SYEIS (0.03) GIYSSGS (0.15) 0.100 
VHE SYAVS (0.39) GIDSGSYIG (0.21) 0.271 
VHF SYGVG (0.23) SIGSGSYIG (0.28) 0.278 
VHY SYEIS (0.03) AISTSGC (0.48) 0.296 
VHZ 10 SYAVS (0.39) GIYSSGS (0.15) 0.250 
Other 21 NAb NAb NAb 
a

Mean values for newborn piglets.

b

Not applicable.

FIGURE 6.

The hydropathicity profiles expressed as bar graphs of HCDR3s in randomly selected VDJ clones presented in relationship to three major VH genes or VH gene groups in PIP and IIP littermates. Hydropathicity regions I, II, and III are as indicated in Fig. 4 and regions that differ significantly in distribution are in indicated in the box.

FIGURE 6.

The hydropathicity profiles expressed as bar graphs of HCDR3s in randomly selected VDJ clones presented in relationship to three major VH genes or VH gene groups in PIP and IIP littermates. Hydropathicity regions I, II, and III are as indicated in Fig. 4 and regions that differ significantly in distribution are in indicated in the box.

Close modal
Table IV.

Characteristics of 25 VDJs in region I with hydrophobic HCDR3s from PIPs

CloneaVH GeneH.I.DH UsageAmino Acid Sequenceb
DHRFcLengthFR3dDHJH
      IAIAMVLVAIV  
IgG A4 VHA 0.59 7/11 CAR DF IAMVLVA AYYFP MDLW 
IgG 49E4 VHX 0.77 7/11 CAR G IAMVLVA MDLW 
IgM F5 VHA 0.50 7/11 CAR GYL AIAMVLV GARS TDLW 
IgM A6 VHA 0.66 8/11 CAR GPL IAIAMVLV PVKYA MDLW 
IgG B6 VHZ 0.62 8/11 CAR GYG IAMVLVAI GPN MDLW 
IgA B9 VHZ 0.62 7/11 CIR DPR IAIAMVL L MDLW 
IgA C10 VHZ 0.61 6/11 CVR DPR IAIAMVL L MDLW 
IgA C7 VHZ 0.67 6/11 CAR GLR IAMVLV VYA MDLW 
IgM D3 VHZ 0.50 8/11 CAI GRIR IAIAMVLV PRVRRLSWYP MDLW 
IgM D4 VHZ 0.57 7/11 CAR DPR IAIAMVL L MDLW 
IgM D6 VHZ 0.47 5/11 CGG VLVAI GHYYYA MDLW 
IgG E4 VHZ 0.57 7/11 CAR DPR IAIAMVL L MDLW 
IgM E5 VHZ 0.49 9/11 CAR GGA IAMVLVAI RGHYYA MDLW 
IgM G1 VHA 0.68 10/11 CAI GRIR IAIAMVLVAMV PRYA MDLW 
IgA G9 VHA 0.75 4/11 CAT GLV VLVA PT MDLW 
IgG F10 VHA 0.49 5/11 CAR GFRVAKGL VLVAI QYA MDLW 
IgM C6 VHA 0.80 3/11 CAT LVA MM MDLW 
IgM 59H1 VHA 0.48 9/11 CAR GG AIAMVLVAI RGHYYA MDLW 
         
      TIAVAIAV  
IgM C4 VHA 0.58 7/8 CAR DW IAVAIAV PSIYYA MDLW 
IgG F3 VHA 0.63 7/8 CAR DW IAVAIAV PIIYYA MDLW 
IgM C1 VHZ 0.61 7/8 CAR D IAVAIAV FYYA MDLW 
IgG E1 VHA 0.60 7/8 CAQ IAVAIAV TGRW VDLW 
IgM G3 VHZ 0.55 3/8 CAR R IAV TSGMFLVFYA MDLW 
IgA A11 VHA 0.66 6/8 CAR K AVAIAV IYA MDLW 
IgM H1 VHA 0.45 6/8 CAR G AVAIAV TSGGKVIGAPH MDLW 
CloneaVH GeneH.I.DH UsageAmino Acid Sequenceb
DHRFcLengthFR3dDHJH
      IAIAMVLVAIV  
IgG A4 VHA 0.59 7/11 CAR DF IAMVLVA AYYFP MDLW 
IgG 49E4 VHX 0.77 7/11 CAR G IAMVLVA MDLW 
IgM F5 VHA 0.50 7/11 CAR GYL AIAMVLV GARS TDLW 
IgM A6 VHA 0.66 8/11 CAR GPL IAIAMVLV PVKYA MDLW 
IgG B6 VHZ 0.62 8/11 CAR GYG IAMVLVAI GPN MDLW 
IgA B9 VHZ 0.62 7/11 CIR DPR IAIAMVL L MDLW 
IgA C10 VHZ 0.61 6/11 CVR DPR IAIAMVL L MDLW 
IgA C7 VHZ 0.67 6/11 CAR GLR IAMVLV VYA MDLW 
IgM D3 VHZ 0.50 8/11 CAI GRIR IAIAMVLV PRVRRLSWYP MDLW 
IgM D4 VHZ 0.57 7/11 CAR DPR IAIAMVL L MDLW 
IgM D6 VHZ 0.47 5/11 CGG VLVAI GHYYYA MDLW 
IgG E4 VHZ 0.57 7/11 CAR DPR IAIAMVL L MDLW 
IgM E5 VHZ 0.49 9/11 CAR GGA IAMVLVAI RGHYYA MDLW 
IgM G1 VHA 0.68 10/11 CAI GRIR IAIAMVLVAMV PRYA MDLW 
IgA G9 VHA 0.75 4/11 CAT GLV VLVA PT MDLW 
IgG F10 VHA 0.49 5/11 CAR GFRVAKGL VLVAI QYA MDLW 
IgM C6 VHA 0.80 3/11 CAT LVA MM MDLW 
IgM 59H1 VHA 0.48 9/11 CAR GG AIAMVLVAI RGHYYA MDLW 
         
      TIAVAIAV  
IgM C4 VHA 0.58 7/8 CAR DW IAVAIAV PSIYYA MDLW 
IgG F3 VHA 0.63 7/8 CAR DW IAVAIAV PIIYYA MDLW 
IgM C1 VHZ 0.61 7/8 CAR D IAVAIAV FYYA MDLW 
IgG E1 VHA 0.60 7/8 CAQ IAVAIAV TGRW VDLW 
IgM G3 VHZ 0.55 3/8 CAR R IAV TSGMFLVFYA MDLW 
IgA A11 VHA 0.66 6/8 CAR K AVAIAV IYA MDLW 
IgM H1 VHA 0.45 6/8 CAR G AVAIAV TSGGKVIGAPH MDLW 
a

GenBank accession numbers for the clones are EU267244–EU267268.

b

HCDR3 region is separated into distinguishable parts. The regions 5′ and 3′ of DH do not distinguish among n-region additions, palindromic additions, or somatic hypermutation of DH and JH gene segments. The mean HCDR3 length for these sequences is 55.0 ± 11.5 pnt, which is significantly longer (p < 0.01) than the mean of 41.2 ± 9.2 for all pre-immune HCDR3 (21 ).

c

Germlines DHA (A) and DHB (B) in RF3.

d

Framework 3.

A total of 25 VDJ clones account for all the hydrophobic HCDR3s in region I (0.5–0.8) in PIPs (Fig. 4,B; Table IV). When the CDR1 and CDR2 regions of all 25 VDJ clones with hydrophobic HCDR3s were examined, one used an “other” VH (see above) and all but three were in germline configuration, but these carried silent mutations (IgM F5, IgG F10, IgG F3).

Furthermore, ∼70% of the DH regions remained in germline configuration although truncated to some degree. More than 70% of these VDJs use DHA and more than half are expressed with IgM, consistent with the pattern observed when sequences were displayed according to isotype (Fig. 5 A). It is noteworthy that there are no hydrophobic HCDR3s that were not encoded by RF3 of DHA or DHB, indicating either that: 1) any hydrophobic sequence cannot substitute for those encoded especially by DHA; or 2) hydrophobic HCDR3s not encoded by RF3 of DHA or DHB are so rare in infected isolator piglets that the chance of their recovery from 70 sequences is remote. Although several related tripeptide motifs, Leu-Val-Ala, Ile-Ala-Val, Val-Ala-Ile, Val-Leu-Leu, Ile-Ala-Ile, and Val-Leu-Val, are encoded by both DHA and DHB, these might need to be surrounded by a larger hydrophobic patch. For example, the IgM G3 clone uses only IAV of DHB, but there is a very hydrophobic MFLVF region that is closely associated. The IgM C6 clone also uses only three amino acids (LVA) of DHA but is adjacent to two hydrophobic methionines. It is noteworthy that the mean HCDR3 length (55.0 ± 11.5) of these sequences from region I is significantly longer (p < 0.01) than the mean HCDR3 length of the preimmune repertoire (41.5 ± 9.3; Ref. 21).

VHZ usage is increased in PRRSV infections (J. E. Butler, P. Weber, and N. Wertz, unpublished observations) and significantly over-represented in region I (Fig. 6,B). This is correlated with the relatively high hydrophobicity of CDR1 of VHZ (Table III). However, if the hydropathicity of CDR1 and/or CDR2 is important, clones expressing VHF, VHY, and perhaps VHE should also have shown increased usage in PIPs. Although this is not seen, the increase in VHA at 10 dpi is counteracted by a decrease in VHB usage (27), which is the most hydrophilic VH gene of the porcine preimmune repertoire. (Table III). Thus, the hydrophobicity of at least CDR1 may contribute to VH selection, although the hydrophobicity of HCDR3 appears to be the major determinant in selective expansion.

The data presented extend in five ways our initial observation concerning the expansion, dissemination, and differentiation of B cells in PIPs with hydrophobic HCDR3s (6). First, we show that the hypergammaglobulinemia in PIP is not a generic effect of the viral exposure of isolator piglets to any virus and that elevated Ig levels are also seen at the site of infection (BAL; Fig. 1).

Second, we show that by 25 dpi all major isotypes in PIPs show selective expansion of HCDR3s derived from the same parent B cell clones (Fig. 3, left panel, and Table II). This explains why the hypergammaglobulinemia in PIPs affects all major isotypes in both blood and BAL (Fig. 1). Because fetal piglets undergo class switch recombination without SHM halfway through gestation in the absence of environmental Ag, possibly stochastically (18, 30, 31), the association of a germline (preimmune) repertoire with IgG and IgA as well as IgM is not surprising.

Third, we show that 30–40% of randomly selected VDJ clones from PIP belong to hydropathicity region I, which nearly disappears in PIC pigs and IIP littermates (Fig. 4). A similar shift is also seen in all mouse B cell subsets after the early transitional (T1) stage (9). The disappearance of region I following Ag exposure may result from death by neglect, because these HCDR3s have been considered least optimal for paratopes recognizing pathogens (32). HCDR3s in region I express a nonmutated germline VH repertoire, retain >70% of their germline DHA and DHB sequences, all of which are expressed in RF3, and these HCDR3s are significantly longer than the mean length of the preimmune repertoire (Table IV). The higher incidence of IgM clones with BCRs bearing nondiversified hydrophobic HCDR3s (Fig. 5,A) is reminiscent of the preimmune repertoire in which IgM dominates (18). Additional support for the preimmune nature of the expanded subpopulation comes from flow cytometric studies of B cells from the BAL of PIPs, showing that a very high proportion are CD2+ compared with IIP littermates; CD2 in swine is a marker for undifferentiated B cells (33). The shift away from B cells with hydrophobic BCRs after Ag exposure correlates with an increase in use of “other” VH genes from 15 to 85%, in PIC pigs (10) “Other” VH usage also occurs in PIP and IIP but only in region II (Fig. 4 B).

Fourth, because approximately one-third of the B cells from BAL in PIPs belong to region I (Fig. 4,B) and many tissues share the same spectratype (Fig. 2), it suggests that the same B cell subpopulation is disseminated to all lymphoid tissues (Fig. 2 and Ref. 6). A 4-wk-old piglet has 3 × 1011 lymphocytes (34) although isolator piglets may have half that number, of which 16% of the PBMCs are B cells (35). If one-third of all B cells in PRRS comprises the expanded preimmune repertoire (Table IV), it suggests that there may be 8 × 109 cells with hydrophobic, germline-encoded BCRs in PIP. The number of such cells approaches the number of malignant B cells in the bone marrow of myeloma patients (N. Rosenthal (Department of Pathology, University of Iowa, Iowa City, IA), unpublished observations). Furthermore there would be more serum IgG with hydrophobic binding sites (15 mg/ml) than the total IgG concentration in PIC young pigs (Fig. 4 C).

Finally, this study better defines the motif in the BCR of expanded B cells in PIPs. The minimal motif appears to be a tripeptide rich in isoleucine, valine, and leucine often in association with an alanine. A pentapeptide motif (AMVLV) is encoded by DHA, but the tripeptide is common to DHA and DHB (Table IV). Adjacent hydrophobic amino acids may also contribute to a hydrophobic patch, which may explain the preference for DHA and VHZ with its hydrophobic CDR1 (Table III and Fig. 6). In addition, the significantly longer HCDR3s in region I of PIP (p < 0.01) compared with the mean HCDR3 length in the preimmune repertoire (Table IV and Ref. 21) would increase the size of the patch. The longer HCDR3 is not the result of an age-related increase in terminal deoxynucleotidyltransferase (TdT) activity, because TdT is active at the first time of VDJ rearrangement at 20 days of gestation in the yolk sac (36) and there is no change in mean HCDR3 length throughout fetal life (21). Hence, HCDR3 length, hydrophobicity and even VH usage may contribute to the formation of a hydrophobic patch.

PRRSV is a member of the Arteriviridae that includes equine arterivirus, simian hemorrhagic fever virus, and lactate dehydrogenase-elevating virus (LDV) of mice. The latter is the best studied and is persistent for life (37). The appearance of inefficient neutralizing Abs is delayed, appearing 1–2 mo after infection (38). LDV produces polyclonal B cell activation (39, 40) that results in the formation of immune complexes and autoantibodies (41, 42). These have been reported not to contain anti-LDV Abs but appear to be complexes of autoantibodies and self-antigens (43) and hydrophobic aggregates that spontaneously adsorb to polystyrene (44). Thus, generalized polyclonal B cell activation characteristic of PRRS and LDV could generate autoantibodies to dsDNA, Golgi glycoproteins, and other autoantigens. The immune response to PRRSV is also characterized by a delay in responsiveness (45, 46), a surprisingly weak inflammatory response in the lung (47, 48), and the need for 150 days to clear the infection (4). This delay is believed to increase susceptibility to secondary infections that may also interfere with the effectiveness of vaccines (49, 50, 51). Thus, there are numerous features of the response to LDV that parallel those to PRRSV, although studies at the molecular level on LDV Abs like those described in this article have not been reported. LDV is a persistent infection that causes no harm, whereas PRRSV is responsible for a world pandemic in swine (1).

The phenomenon we describe for PIP contains a nearly untestable but reasonable assumption, namely that one-third of the 10- to 100-fold increase in Igs, i.e., 15 mg/ml plasma IgG (Fig. 1), is derived from B cells bearing hydrophobic motifs (Fig. 4,C). Because the response is polyclonal, individual Igs cannot be recovered for sequencing to test this assumption unless they can be transformed and cloned. Accepting this assumption, the elevated level of IgG with hydrophobic binding sites could contain specific Abs to a hydrophobic epitope of a T cell-independent (TI) Ag from the viral envelope (52) or the amphipathic helix of the M protein (53). These may have been overlooked because commercial and experimental ELISAs depend on hydrophilic glycoproteins or nucleocapsid proteins (see below) and because ELISAs are biased against hydrophobic Abs (54). Finding such a high level of specific Abs in viral infections would be quite extraordinary. Initially we showed that <1 of the total IgG recognized PRRSV Ags using the HerdChek assay (IDEXX Laboratories). Dot plot assays using inactivated whole virus gave similar results (27). There is some evidence for autoantibodies to hydrophobic self-antigens, e.g., those to TGFβ in lpr mice (55) and perhaps in LDV (44). Sixty percent of IgGs in hypergammaglobulinemia plasma (25 mg/ml IgG and double the concentration in PIP; Fig. 4,C) have HCDR3s in hydropathicity region II with a charged binding site. The autoantibodies to Golgi proteins, dsDNA, and others we observed (5) would most likely be accounted for by these Igs rather than by those with hydrophobic HCDR3s (region I; Fig. 4). This assumption is based on studies showing that autoantibodies are typically rich in arginine and lysine and would distribute to region II (28, 56, 57, 58). It is generally recognized that: 1) the “natural” preimmune repertoire overexpresses autoreactive Abs (59, 60, 61, 62); and 2) polyreactivity is associated with HCDR3 (63, 64, 65). The Abs to PRRSV glycoproteins and the highly immunogenic nucleocapsid protein in PIP (Refs. 66, 67, 68, 69, 70, 71, 72 ; M. Murtaugh, unpublished observations) are most likely to have HCDR3s that fit to region II, which is where we observed that repertoire diversification had occurred (Fig. 4 B).

The predicted magnitude of the hydrophobic Ab response (15 mg/ml; Fig. 4 C), combined with the 25 mg/ml IgG predicted to have HCDR3s that fit to region II, is 100-fold higher than the estimated antiviral response, leaving the remainder to be accounted for by autoantibodies or by nonspecific Igs resulting from polyclonal B cell activation. Such high levels of autoantibodies have never been described, certainly not those specific to hydrophobic Ags. Therefore, what drives the polyclonal activation in PIPs and in LDV-infected mice? Well-defined BSAgs target the framework of certain VH gene families, e.g., VH3 (73, 74, 75, 76). All VH genes of swine belong to the VH3 family, their framework regions (except for VHB) are identical (10, 16, 77), and there is no (6, 27) or only weak evidence for the increased usage of a particular VH gene in PIP, e.g., VHZ (J. E. Butler, P. Weber, and N. Wertz, unpublished observations). Perhaps a BSAg targets all porcine B cells but with bias for a hydrophobic motif in HCDR3. Because there is no evidence for a hydrophobic BSAg, the hydrophobic patch may not be the target. Rather, the hydrophobicity of the binding site may alter the BCR conformation, resulting in increased exposure of a BSAg binding site in a framework region. Although BSAgs usually cause deletion or suppression of targeted B cells (78, 79, 80), some BSAgs like staphylococcal enterotoxins promote survival (75, 76). A putative BSAg of the latter type could then trigger Ag- and T cell-independent proliferation and differentiation through a second signal involving TLR3 or TLR7 in the manner described by Marshak-Rothstein and others (81, 82, 83). The BSAgs need not come from PRRSV but may be a consequence of viral infections that activates endogenous BSAgs encoded by endogenous retroviruses as in EBV (84). PRRSV is not known to infect B cells but enters permissive macrophages through CD163 (85), and infectivity depends on the interaction of viral RNA with CD151 (86). Hence, such infected macrophages might present both viral RNA and an endogenous BSAg to B cells. The putative BSAg hypothesis is consistent with observations that the preimmune repertoire, such as that encoded in marginal zone and B-1 cells, is preferentially susceptible (78, 87, 88). Interestingly, conventionally reared PIPs show a reduced degree of immune dysregulation compared with isolator PIPs (27), perhaps because a much smaller proportion of nondiversified B cells remain. This may also explain why adult mice infected with LDV show a reduced level of immune dysregulation compared with isolator piglets (43).

This study raises evolutionary and biotechnical issues concerning BCRs with hydrophobic HCDR3s. If these are deselected because there are no hydrophobic Ags (32, 89), why do mammals continue to generate them as part of their preimmune repertoire? Is it possible that there are important hydrophobic epitopes that immunologists have overlooked because of the entrenched bias against hydrophobic Abs in ELISAs?

The transition from fetal life to weaning traverses a “critical window” in immunological development (13, 14) in which the neonate must: 1) juggle the protective and/or suppressive effects of passive immunity; 2) respond to the pathogen-associated molecular patterns of gut flora that stimulate development of the adaptive immune system; 3) refine the innate protective preadaptive (natural) Ab repertoire to one that is less cross-reactive, less autoreactive, and with refined specificity to pathogens; and 4) simultaneously develop tolerance to nonpathogenic foreign Ags. Failure to establish homeostasis during this process results in immune dysregulation. This “window” is when pathogens and BSAgs may have their greatest impact (90). We believe PRRSV is a pathogen that uses polyclonal B cell activation through a virus-derived or virus-stimulated BSAg (84) to subvert the normal development of adaptive immunity. This distraction causes a delay in the appearance of a protective antiviral response and, consequently, resolution of the infection. This allows more time for viral replication and subsequently a greater chance for the infection to spread to other swine.

The authors have no financial conflict of interest.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1

This work was supported by Grant 05-174 from the National Pork Board and U.S. Department of Agriculture Agricultural Research Service Cooperative Agreement 58-3625-4-155.

3

Abbreviations used in this paper: PRRS, porcine reproductive and respiratory syndrome; BAL, bronchoalveolar lavage; BSAg, B cell superantigen; dpi, days postinoculation; HCDR3, H chain CDR3; H.I., hydropathicity index; IIP, SIV-infected piglets; LDV, lactate dehydrogenase-elevating virus; PCV-2, porcine circovirus-2; PIC, parasite-infected conventional pigs; PIP, PRRSV-infected piglets; pnt, polynucleotide; PRRSV, PRRS virus; RF, reading frame; SHM, somatic hypermutation; SIV, swine influenza virus.

1
Neumann, E. J., J. B. Kliebenstein, C. D. Johnson, E. J. Mabry, J. W. Bush, A. H. Seitzinger, A. L. Green, J. J. Zimmerman.
2005
. Assessment of the economic impact of porcine reproductive and respiratory syndrome on swine production in the United States.
J. Am. Vet. Med. Assoc.
227
:
385
-392.
2
Benfield, D. A., E. Nelson, J. E. Collins, L. Harris, S. M. Goyal, D. Robison, W. T. Christianson, R. B. Morrison, D. Gorcyca, D. Chladek.
1992
. Characterization of swine infertility and respiratory syndrome (SIRS) virus (isolate ATCC VR-2332).
J. Vet. Diagn. Invest.
4
:
127
-133.
3
Conzelmann, K. K., N. Visser, P. Van Woensel, H. J. Thiel.
1993
. Molecular characterization of porcine reproductive and respiratory syndrome virus, a member of the arterivirus group.
Virology
193
:
329
-339.
4
Allende, R., W. W. Laegieid, G. F. Kutish, J. A. Galesta, R. W. Wills, F. A. Ososrio.
2000
. Porcine reproductive and respiratory syndrome virus: description of persistence in individual pigs upon experimental infection.
J. Virol.
74
:
10834
-10837.
5
Lemke, C. D., J. S. Haynes, R. Spaete, D. Adolphson, A. Vorwald, K. Lager, J. E. Butler.
2003
. Lymphoid hyperplasia resulting in immune dysregulation is caused by PRRSV infection in pigs.
J. Immunol.
172
:
1916
-1925.
6
Butler, J. E., C. D. Lemke, P. Weber, M. Sinkora, K. D. Lager.
2007
. Antibody repertoire development in fetal and neonatal piglets: XIX. Undiversified B cells with hydrophobic HCDR3s preferentially proliferate in PRRS.
J. Immunol.
178
:
6320
-6331.
7
Ippolito, G. C., R. L. Schelonka, M. Zemlin, C. Zemlin, Y. Zhaung, G. L. Gartland, L. Nitschke, J. Pelkonen, K. Rajewsky, H. W. Schroeder, Jr.
2006
. Forced enrichment for hydrophobic amino acids in immunoglobulin CDR-H3 impairs splenic B cell development but not antibody production.
J. Exp. Med.
203
:
1567
-1578.
8
Ivanov, I., J. Link, G. C. Ippolito, H. W. Schroeder, Jr.
2002
. Constraints on the hydropathicity and sequence composition of HCDR3 are conserved across evolution. M. Zanetti, Jr, and J. D. Capra, Jr, eds.
The Antibodies
43
-67. Taylor and Francis, London.
9
Schelonka, R. L., J. Tamer, Y. Zhuang, G. I. Gartland, M. Zemlin, H. W. Schroeder, Jr.
2007
. Categorical selection of the antibody repertoire in splenic B cells.
Eur. J. Immunol.
37
:
1010
-1021.
10
Butler, J. E., P. Weber, N. Wertz.
2006
. Antibody repertoire development in fetal and neonatal pigs: XIII. “Hybrid VH genes” and the pre-immune repertoire re-visited.
J. Immunol.
177
:
5459
-5470.
11
Miniatis, P., D. Joh.
1978
. Gnotobiotic pigs derivision and rearing.
Can. J. Comp. Med.
42
:
428
-437.
12
Vincent, A. L., K. M. Lager, W. Ma, P. Lekcharoensuk, M. R. Gramer, C. Loiacona, J. A. Richt.
2006
. Evaluation of hemagglutinin subtype 1 swine influenza viruses from the United States.
Vet. Microbiol.
118
:
212
-222.
13
Butler, J. E., M. Sinkora.
2007
. The isolator piglet: a model for studying the development of adaptive immunity.
Immunol. Res.
39
:
33
-51.
14
Butler, J. E., J. Sun, N. Wertz, M. Sinkora.
2006
. Antibody repertoire development in swine.
Dev. Comp. Immunol.
30
:
199
-221.
15
Butler, J. E., P. Weber, M. Sinkora, D. Baker, A. Schoenherr, B. Mayer, D. Francis.
2002
. Antibody repertoire development in fetal and neonatal piglets: VIII. Colonization is required for newborn piglets to make serum antibodies to T-dependent and type 2 T-independent antigens.
J. Immunol.
169
:
6822
-6830.
16
Sun, J., C. Hayward, R. Shinde, R. Christenson, S. P. Ford, J. E. Butler.
1998
. Antibody repertoire development in fetal and neonatal piglets: I. Four VH genes account for 80% of VH usage during 84 days of fetal life.
J. Immunol.
161
:
5070
-5078.
17
McAleer, J., P. Weber, J. Sun, J. E. Butler.
2005
. Antibody repertoire development in fetal and neonatal piglets: XI. The thymic B cell repertoire develops independently from that in blood and mesenteric lymph nodes.
Immunology
114
:
171
-183.
18
Butler, J. E., P. Sun, J. Weber, S. P. Ford, Z. Rehakova, J. Sinkora, K. Lager.
2001
. Antibody repertoire development in fetal and neonatal piglets: IV. Switch recombination, primarily in fetal thymus occurs independent of environmental antigen and is only weakly associated with repertoire diversification.
J. Immunol.
167
:
3239
-3249.
19
Navarro, P., R. Christenson, P. Weber, M. Rothschild, G. Ekhard, J. Lemky, J. E. Butler. Porcine IgA allotypes are not equally transcribed or expressed in heterozygous swine.
Mol. Immunol.
37
:
653
-664.
20
Butler, J. E., J. Sun, P. Navarro.
1996
. The swine immunoglobulin heavy chain locus has a single JH and no identifiable IgD.
Int. Immunol.
8
:
1897
-1904.
21
Butler, J. E., P. Weber, M. Sinkora, J. Sun, S. J. Ford, R. Christenson.
2000
. Antibody repertoire development in fetal and neonatal piglets: II. Characterization of heavy chain CDR3 diversity in the developing fetus.
J. Immunol.
165
:
6999
-7011.
22
Bonefant, C., I. Vallee, J. Sun, G. Thibault, J. M. Guillaumin, Y. Lebranchu, P. Bardos, J. E. Butler, H. Watier.
2002
. Analysis of the human CD4 T lymphocyte proliferation induced by retrovirally-infected porcine lymphoblastoid B cell lines.
Zenotransplantation
9
:
1
-13.
23
Kyte, J., R. F. Doolittle.
1982
. A simple method for displaying the hydropathic character of a protein.
J. Mol. Biol.
157
:
105
-132.
24
Eisenberg, D..
1984
. Three-dimensional structure of membrane and surface proteins.
Annu. Rev. Biochem.
53
:
595
-623.
25
Butler, J. E., J. Sun, P. Weber, D. Francis.
2000
. Antibody repertoire development in fetal and neonatal piglets: III. Colonization of the gastrointestinal tracts results in preferential diversification of the pre-immune mucosal B-cell repertoire.
Immunology
100
:
119
-130.
26
Holm, S..
1979
. A simple sequentially rejective multiple test procedure.
Scand. J. Statist.
6
:
65
-70.
27
Lemke, C. D..
2006
.
PRRSV infection of neonatal piglets induces immune dysregulation and modulation. Doctoral Dissertation
University of Iowa, Iowa City, IA.
28
Barbas, S. M., P H. J. Yang, G. Silverman, D. R. Barton.
1995
. Human autoantibody recognition of DNA.
Proc. Natl. Acad. Sci. USA
92
:
2529
-2533.
29
Silverman, G. J., J. V. Nayak, A. La Cava.
1997
. B cell superantigens: molecular and cellular implications.
Int. Rev. Immunol.
14
:
259
-290.
30
Butler, J. E., N. Wertz.
2006
. Antibody repertoire development in fetal and neonatal pigs: XVII. IgG subclass transcription in newborns revisited with emphasis on new IgG3.
J. Immunol.
177
:
5480
-5489.
31
Deenick, E. K., J. Hasbold, P. D. Hodgkins.
1999
. Switching to IgG3, IgG2b, and IgA is division linked and independent revealing a stochastic framework for describing differentiation.
J. Immunol.
163
:
4707
-4717.
32
Raaphorst, F. M., C. S. Raman, B. T. Nall, J. M. Teale.
1997
. Molecular mechanisms governing reading frame choice of immunoglobulin diversity genes.
Immunol. Today
18
:
37
-43.
33
Sinkora, M., J. Sinkorova, and J. E. Butler. 2007. CD2 and CD21 expression can be used to describe maturation pathways for porcine B cells. In 8th International Veterinary Immunology Symposium, August 15–19. Ouro Preto, Brazil. Abstract AP129, p. 95.
34
Pabst, R., F. Trepel.
1975
. Quantitative evaluation of the total number and distribution of lymphocytes in young pigs.
Blut
31
:
77
-86.
35
Pabst, R., E. Kaupp, F. Trepel.
1077
. Relative and absolute numbers of E- and EAC-rosette-forming cells in lymphoid organs.
Blut
34
:
210
-210.
36
Sinkora, M., J. Sun, J. Sinkorova, R. K. Christenson, S. P. Ford, J. E. Butler.
2003
. Antibody repertoire development in fetal and neonatal piglets, VI. B cell lymphogenesis occurs in multiple sites with differences in the frequency of in-frame rearrangements.
J. Immunol.
170
:
1781
-1788.
37
Plagemann, P. G. W., V. Moenning.
1992
. Lactate dehydrogenase elevating virus, equine arteritis virus, and Simian hemorrhagic fever virus, a new group of positive strand RNA viruses.
Adv. Virus Res.
41
:
99
-192.
38
Cafruny, W. A., P. G. W. Plagemann.
1982
. Immune response to lactate dehydrogenase elevating virus: isolation of infectious virus-immunoglobulin complexes and quantitation of specific anti-viral immunoglobulin G response in wild-type and nude mice.
Infect. Immun.
37
:
1001
-1006.
39
Li, X., B. Hu, J. Harty, C. Even, P. G. W. Plagemann.
1990
. Polyclonal B cell activation of IgG2a and IgG2b production by infection of mice with lactate dehydrogenase-elevating virus is partly dependent on CD4+ lymphocytes.
Viral Immunol.
3
:
273
-287.
40
Bradley, D. S., J. Broen, W. A. Cafruny.
1991
. Infection of SCID mice with lactate dehydrogenase-elevating virus stimulates B cell activation.
Viral Immunol.
4
:
59
-70.
41
Cafruny, W. A., D. E. Hovinen.
1988
. Infection of mice with lactate dehydrogenase-elevating virus leads to stimulation of autoantibodies.
J. Gen. Virol.
69
:
723
-729.
42
Rowland, R. R. R., C. Even, G. W. Anderson, Z. Chen, B. Hu, P. G. W. Plagemann.
1994
. Neonatal infection of mice with lactate dehydrogenase-elevating virus results in suppression of humoral anti-viral immune response but does not alter the course of viremia or the polyclonal activation of B cells and immune complex formation.
J. Gen. Virol.
75
:
1071
-1081.
43
Hu, B., C. Even, P. G. W. Plagemann.
1992
. Immune complexes bind ELISA plates not coated with antigen in mice infected with lactate dehydrogenase-elevating virus: relationship to IgG2a and IgG2b specific polyclonal activation of B cells.
Viral Immunol.
5
:
27
-38.
44
Zitterkopf, N. L., Q. A. Jones, D. S Bradley, K. Durick, R. R. Rowland, P. G. Plagemann, W. A. Cafruny.
2003
. Hydrophobic IgG-containing immune complexes in the plasma of autoimmune MRL/lpr mice, lactate dehydrogenase-elevating virus-infected mice, and pigs: association with transforming growth factor-β and pH-dependent amplification.
Viral Immunol.
16
:
511
-523.
45
Bautista, E. M., T. W. Molitor.
1997
. Cell-mediated immunity to porcine reproductive and respiratory syndrome virus in swine.
Viral Immunol.
10
:
83
-94.
46
Lopez, O. J., L. Fuertes, N. Domenech, B. Alvarez, A. Ezquerra, J. Pominquea, J. M. Castro, F. Alovso.
1999
. Analysis of cellular immune response in pigs recovered from porcine reproductive and respiratory syndrome infection.
Virus Res.
64
:
33
-42.
47
Van Reeth, K., H. Nauwynck.
2000
. Proinflammatory cytokines and viral respiratory disease in pigs.
Vet. Res.
31
:
187
-213.
48
Wills, R. W., A. R. Doster, J. A. Galeota, J. H. Sur, F. A. Osorio.
2004
. Duration of infection and proportion of pigs persistently infected with porcine reproductive and respiratory syndrome virus.
J. Clin. Microbiol.
41
:
58
-62.
49
DeBruin, M. G. M., J. N. Samson, J. J. M. Voermans, E. M. A. van Rooij, Y. E. DeVisser, A. T. J. Bianchi.
2000
. Effects of a porcine reproductive and respiratory syndrome virus infection on the development of the immune response against pseudorabies virus.
Vet. Immunol. Immunopath.
76
:
125
-135.
50
Suradhat, S., S. Kesdangsakonwut, W. Sada, S. Buranapraditkum, S. Wongsawang, R. Thanawongnuwech.
2006
. Negative impact of porcine reproductive and respiratory syndrome virus infection on the efficacy of classical swine fever vaccine.
Vaccine
24
:
2634
-2642.
51
Thacker, E. L..
2006
. Lung inflammatory responses.
Vet. Res.
37
:
469
-486.
52
Lee, C., D. Yoo.
2006
. The small envelope protein of porcine reproductive and respiratory syndrome virus possesses ion channel protein-like properties.
Virolgy
355
:
30
-43.
53
Chen, Z., L. Kuo, R. R. R. Rowland, C. Even, K. S. Faaberg, P. G. W. Plagemann.
1993
. Sequences of 3′ end of genome and of 5′ end of open reading frame 1a of lactate dehydrogenase-elevating virus and common junction motifs between 5′ leader and bodies of sugenomic in RNAs.
J. Gen. Virol.
74
:
643
-660.
54
Butler, J. E..
1991
. Perspectives, configurations, and principles. J. E. Butler, Jr, ed.
Immunochemistry of Solid-Phase Immunoassay
3
-26. CRC Press, Boca Raton, FL.
55
Rowley, D. A., E. T. Becken, R. M. Stach.
1995
. Autoantibodies produced spontaneously by young 1pr mice carry transforming growth factor β and suppress cytotoxic T lymphocyte responses.
J. Exp. Med.
181
:
1875
-1880.
56
Marion, T. N., M. R. Krishnan, D. D. Desai, N.-T. Jou, D. M. Tillman.
1997
. Monoclonal anti-DNA antibodies: structure, specificity and biology.
Methods
11
:
3
-11.
57
Shlomchik, M. J., A. H. Aucoin, D. S. Pistesky, M. G. Weigert.
1987
. Structure and function of anti-DNA autoantibodies derived from a single autoimmune mouse.
Proc. Natl. Acad. Sci. USA
84
:
9150
-9154.
58
Wloch, M. K., S. H. Clark, G. S. Gilkeson.
1997
. Influence of VH CDR3 arginine and light chain pairing on DNA reactivity of a bacterial DNA induced anti-DNA antibody from a BALB/c mouse.
J. Immunol.
159
:
6083
-6090.
59
Cukrowska, B., J. Sinkora, Z. Rehakova, M. Sinkora, I. Splichal, L. Tukova, S. Avrameas, A. Saalmueller, R. Barto-Ciorbarus, H. Tlaskalova-Hogenova. Isotype and antibody specificity of spontaneously formed immunoglobulins in pig fetuses and germ-free piglets: production of CD5 B cells.
Immunology
88
:
611
-617.
60
Dighiero, G. P., P. Lymberi, D. Holmberg, I. Lundquist, A. Coutinho, S. Avrameas.
1985
. High frequency of natural autoantibodies in normal mice.
J. Immunol.
134
:
765
-771.
61
Alt, F. W., K. Blackwell, G. D. Yancopoulos.
1987
. Development of the primary antibody repertoire.
Science
238
:
1079
-1087.
62
Schroeder, H. W., Jr, J. L. Hillson, R. M. Perlmutter.
1987
. Early restriction of the human antibody repertoire.
Science
238
:
791
-793.
63
Chen, C., M. P. Stenzel-Poore, M. R. Rittenberg.
1991
. Natural auto- and polyreactive antibodies differing from antigen-induced antibodies in the H chain CDR3.
J. Immunol.
147
:
2359
-2367.
64
Crouzier, R., T. Martin, J. L. Pasquali.
1995
. Heavy chain variable region light chain variable region and heavy chain CDR3 influences on the mon- and polyreactivity and on the affinity of human monoclonal rheumatoid factors.
J. Immunol.
154
:
4526
65
Ichiyoshi, Y., P. Casali.
1994
. Analysis of the structural correlates for antibody polyreactivity by multiple reassortments of chimeric human immunoglobulin heavy chain and light chain VH segments.
J. Exp. Med.
180
:
885
-895.
66
Labarque, G. G., H. J. Nauwynck, K. Van Reeth, M. B. Pensaert.
2000
. Effect of cellular changes and onset of humoral immunity on the replication of porcine reproductive and respiratory syndrome virus in the lungs of pigs.
J. Gen. Virol.
81
:
1327
-1334.
67
Loemba, H. D., S. Mounir, H. Mardassi, D. Archambault, S. Dea.
1996
. Kinetics of humoral immune response to the major structural proteins of the porcine reproductive and respiratory syndrome virus.
Arch. Virol.
141
:
751
-761.
68
Vezina, S. A., H. Loemba, M. Fournier, S. Dea, D. Archambault.
1996
. Antibody production and blastogenic response in pigs experimentally infected with porcine reproductive and respiratory syndrome virus.
Can. J. Vet. Res.
60
:
94
-99.
69
Lopez, O. J., F. A. Osorio.
2004
. Role of neutralizing antibodies in PRRSV protective immunity.
Vet. Immunol. Immunopath.
102
:
155
-163.
70
Gonin, P., B. Pirzadeh, C. A. Gagnon, S. Dea.
1999
. Serum neutralization of porcine reproductive and respiratory syndrome virus correlates with antibody response to the GP5 major envelope glycoprotein.
J. Vet. Diagn. Invest.
11
:
20
-26.
71
Yang, L., K.-J. Yoon, Y. Li, J.-H. Lee, J. J. Zimmerman, M. L. Frey, K. M. Harmon, K. B. Platt.
1999
. Antigenic and genetic variations of the 15kD nucleocapsid protein of porcine reproductive and respiratory syndrome virus isolates.
Arch. Virol.
144
:
525
-546.
72
Plana-Duran, J., I. Climent, J. Sarraseca, A. Urniza, E. Cortes, C. Vela, J. I. Casal.
1997
. Bacculovirus expression of proteins of porcine reproductive and respiratory syndrome virus strain Olot/91: involvement of ORF3 and ORF5 proteins in protection.
Virus Genes
14
:
19
-29.
73
Silverman, G. J..
1997
. B cell superantigens.
Immunol. Today
18
:
379
-386.
74
Zouali, M..
1995
. B cell superantigens: implications for selection of the human antibody repertoire.
Immunol. Today
16
:
399
-405.
75
Domiati-Saad, R. J., F. Attrep, H. P. Brezinzchek, A. H. Cherrie, D. R. Karp, P. E. Lipsky.
1996
. Staphylococcal endotoxin D functions as a human B cell superantigen by rescuing VH4-expressing B cells from apoptosis.
J. Immunol.
156
:
3608
-3620.
76
Domiati-Saad, R., P. E. Lipsky.
1998
. Staphylococcal enterotoxin A induces survival of VH3-expressing human B cells by binding to the VH region with low affinity.
J. Immunol.
161
:
1257
-1266.
77
Sun, J., J. E. Butler.
1996
. Molecular characteristics of VDJ transcripts from a newborn piglet.
Immunology
88
:
331
-339.
78
Silverman, G. J., J. V. Nayak, K. Warnatz, F. F. Hajjar, S. Gary, H. Tighe, J. E. Curtiss.
1998
. The dual phases of the response to neonatal exposure to a VH family-restricted staphylococcal B cell superantigen.
J. Immunol.
161
:
5720
-5732.
79
Silverman, G. J., C. S. Goodyear.
2006
. Confounding B cell defenses: lessons from a staphylococcal superantigen.
Nat. Rev. Immunol.
6
:
465
-475.
80
Goodyear, C. S., M. Corr, F. Sugiyama, D. L. Boyle, G. J. Silverman.
2007
. Cutting edge: Bim is required for superantigen-mediated B cell death.
J. Immunol.
178
:
2636
-2640.
81
Marshak-Rothstein, A., I. R. Rifkin.
2007
. Immunologically active autoantigens: the role of Toll-like receptors in development of chronic inflammatory disease.
Annu. Rev. Immunol.
25
:
419
-441.
82
Alexopoulos, L., A. Czopik Hoft, R. Medzhitov, R. A. Flavell.
2001
. Recognition of double-stranded RNA and activation of NK-κB by Toll-like receptor 3.
Nature
413
:
732
-738.
83
Busconi, L., C. M. Lau, A. S. Tabor, M. B. Uccellini, Z. Ruhe, S. Akira, G. A. Viglianti, I. R. Riflkin, A. Marshak-Rothstein.
2006
. DNA and RNA autoantigens as autoadjuvants.
J. Endotoxin Res.
12
:
379
-384.
84
Hsiao, F. C., M. Lin, A. Tai, G. Chen, B. T. Huber.
2006
. Cutting edge: Epstein-Barr virus transactivates the HERV-K18 superantigen by docking to the human complement receptor 2 (CD21) on primary B cells.
J. Immunol.
177
:
2056
-2060.
85
Clavert, J. G., D. E. Slade, S. L. Shield, R. Jolie, R. M. Mannan, R. G. Ankenbauer, S. K. Welch.
2007
. CD163 expression confers susceptibility to porcine reproductive and respiratory syndrome viruses.
J. Virol.
81
:
7371
-7379.
86
Shanmukhappa, K., J. K. Kim, S. Kapil.
2007
. Role of CD151, a tetraspanin, in porcine reproductive and respiratory syndrome virus infection.
Virol. J.
4
:
62
87
Dammers, P. M., A. Visser, E. R. Popa, P. Nieuwenhuis, F. G. M. Kroese.
2000
. Most marginal zone B cells in rats express germline encoded IgVH genes and are ligand selected.
J. Immunol.
165
:
6156
-6169.
88
Viau, M., M. Zouali.
2005
. Effect of the B cell superantigen protein A of S. aureus on the early lupus disease of (NZB × NZB) F1 mice.
Mol. Immunol.
42
:
849
-855.
89
Briles, D. E., M. Nahm, K. Schroer, J. Davie, P. Baker, J. K. Kearney, R. Barletta.
1981
. Anti-phosphocholine antibodies found in normal mouse serum are protective against intravenous infection with type 3 Streptococcus pneumoniae.
J. Exp. Med.
153
:
691
-705.
90
Viau, M., B. Cholly, L. Bjorck, M. Zouali.
2004
. Down-modulation of the antigen receptor by a superantigen in human B cells.
Immunol. Lett.
92
:
91
-96.