The JI PBL Intereron Source
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Sun, J.
Right arrow Articles by Nishi, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sun, J.
Right arrow Articles by Nishi, Y.
The Journal of Immunology, 2001, 167: 5775-5785.
Copyright © 2001 by The American Association of Immunologists

Molecular Evolution of Catalytic Antibodies in Autoimmune Mice1

Jialin Sun, Naoko Takahashi2, Hiroyuki Kakinuma3 and Yoshisuke Nishi4

Laboratory of Life Science and Biomolecular Engineering, Japan Tobacco, Yokohama, Kanagawa, Japan


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Catalytic Abs (catAbs) preferentially evolved in autoimmune MRL/MPJ-lpr/lpr (MRL/lpr) mice upon immunization with the phosphonate transition-state analogue (TSA), but this did not happen in normal BALB/c mice. The majority of the catAbs from MRL/lpr mice were from several independent clones of the same family. Most of them had a lysine at position 95 in the heavy chain (H95), which is at the junctional region. This residue, which interacts with the phosphonate moiety of the TSA and presumably is involved in the catalytic activity, was not changed even after expansive evolution following multiple mutations. By contrast, the majority that arose from BALB/c mice were the non-catAbs, which were quite different in the sequence from the catAbs from MRL/lpr mice, but they were clonally related to one another, so most of them were originated from a single clone. In the MRL/lpr mice, the catalytic subsets that existed in the initial repertoire were effectively captured by the phosphonyl oxygens in the TSA by interacting with the lysine at H95. In the BALB/c mice, however, another noncatalytic subset with only the binding capability directed to a moiety other than the phosphonate moiety was alternatively evolved, because of the lowest abundance or elimination of the catalytic subsets.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Directed in vivo selection based on the complementarity to the transition-state analogue (TSA)5 for a given reaction (1, 2) has made it possible to efficiently produce catalytic Abs (catAbs). Phosphonate derivatives were the first and most successful haptens to elicit tailor-made esterolytic catAbs. They show good approximation of the transition state of substrates for ester hydrolysis because of their tetrahedral geometry and anionic character (3, 4). Since then, a variety of catAbs has been generated based on haptenic TSAs (5).

CatAbs are potentially selected from an enormously diverse repertoire that continuously evolves and comprises the limitless potential of a given individual to produce different Abs following antigenic stimuli (6, 7). Nevertheless, only a small portion of repertoire subsets that bind haptenic TSAs expresses catalytic activity. One of the reasons of lower recovery could be that the virtual limit of shape space, whose diversity is represented by the complementarity-determining regions (CDRs), was within the existing repertoire of 5 x 108 mature B cells in the peripheral lymph nodes. Expanding the shape space beyond the existing repertoire, in which the possible combinations of Ab repertoire are in the range of 1011–1012 (8), could allow us to recruit more efficiently potential candidates for catalysts with desired rate enhancements and/or unique reactions. One of the most promising ways of expanding the shape space could be to select candidate Abs by a phage-displayed system, in which heavy and light chain gene shuffling with subsequent randomized mutagenesis is possible. It is underway and has reached some success, although it was possible only within the existing repertoire (9, 10). Finding a way to expand the repertoire beyond the existing repertoire is of critical importance.

Autoimmunity caused by the breakage of self-tolerance, which resulted in a high incidence of abnormal autoreactive Abs, gives us a chance to investigate an unusual Ab repertoire (11, 12). Such a trial has been conducted by testing the ability of several autoimmune-prone mouse strains to elicit esterolytic catalysts (13). The occurrence of catAbs is dramatically higher in autoimmune mouse strains, such as MRL/MPJ-lpr/lpr (MRL/lpr) and SJL, than it is in the conventionally used normal mouse strains (13). This study was motivated by previous findings of naturally occurring catAbs in patients with autoimmune diseases, such as asthma (14) or systemic lupus erythematosus (SLE) (15). Using a different phosphonate hapten, we have confirmed that catAbs could be recovered at higher incidence from the MRL/lpr mice than from the conventionally used normal mouse strains (16, 17).

Taken together, these findings strongly suggest that catAbs exist at a high incidence in the autoimmune repertoire, and that these catAbs are not included in the repertoire of conventionally used normal strains. In our effort to explore the immunological evolution of Ab catalysts in an expanded shape space, we have tried to elucidate the evolution of catAbs in the autoimmune repertoire based on the sequence analysis. In the present work, we have shown that catAbs were uniquely obtained from a subset of the repertoire of an autoimmune mouse strain, but not from the repertoire of a normal mouse strain.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Preparations of chemicals and cloning of mAbs

TSA1 (1, R = OH) and the haptenic TSA1, or a protein conjugate form of TSA1 (1, R = NH-keyhole limpet hemocyanin (NH-KLH) or NH-BSA), TSA2 (2), dansylated substrate (4), and its cleaved product (5) were synthesized, as previously described (18) (Fig. 1Go). The compound 3, resembling the structure of the cleaved product 5, or product analogue, was also synthesized (data not shown). Immunization of the KLH-conjugated haptenic TSA1 to the two mouse strains, MRL/lpr and BALB/c, screening the mAbs with the BSA-conjugated haptenic TSA1 for ELISA and TSA2, and purification of the mAbs were done as previously described (18). For each strain, two mice were used, each of which received two or three injections. In this work, MS5 and MS6 were designated as Abs from the pooled hybridomas from two MLR/lpr mice. The MS5 mouse received two injections, and the MS6 mouse received three injections. TSA2 and compound 3 were used for measuring affinities of the Abs. The substrate 4 and the cleaved product 5 were used for measuring catalytic activities. Isotypes of mAbs were predetermined with a mouse mAb isotyping kit (Boehringer Mannheim, Indianapolis, IN).



View larger version (8K):
[in this window]
[in a new window]
  		FIGURE 1. Chemical structures of TSAs, substrate, and cleaved products. 1, A hapten used for immunization (R = NH-KLH), ELISA screening (R = NH-BSA), and measuring the affinities of the Abs (R = OH or TSA1); 2 (TSA2), used for ELISA screening and measuring the affinities of the Abs; 3, the analogue of the cleaved product (5) used for measuring the affinities of the Abs; 4, the substrate; 5, used for the assay of catalytic activities; and 6, another cleaved product.

 
Cloning and sequencing of cDNAs from hybridomas

Total RNA was purified from the hybridomas using a Sepasol RNA kit (Nacalai Tesque, Kyoto, Japan), and cDNA was generated from mRNA using a first-strand cDNA synthesis kit (Amersham Pharmacia Biotech, Buckinghamshire, U.K.). Primers for cloning Fab were designed as shown below. In the following sequences, S = G or C, W = A or T, K = G or T, M = A or C, R = A or G, Y = C or T, and B = not A. The restriction enzymes used for cloning are shown in parentheses, and their sites are underlined. The sequences of the Ig genes are double underlined. 5' primers (VH-fwd) for the heavy chain V regions: VH-fwd-1, TCCCCCGGGCCCCAGGTSMARCTGCAGCAGYCTGGT (SmaI); VH-fwd-2, TCCCCCGGGCCCGARGTGMAGCTGGTGGARTCTG(SmaI); VH-fwd-3, TCCCCCGGGCCCGARGTGAAGCTGGAKGAGWCTG(SmaI); and VH-fwd-4, TCCCCCGGGCCCGATGTRCAGCTTCAGGAGTCRGGACCT (SmaI). The 3' primers (IG-rev) for the heavy chain V regions: IgG2a-rev, GCCACTAGTTTATGGAGGACAGGGGTTGATTGT (SpeI); IgG1-rev, GCCACTAGTTTATATGCAAGGCTTACAACCACA(SpeI). The 5' (V{kappa}-fwd) and 3' (C{kappa}-rev) primers for the {kappa}-chains: V{kappa}-fwd-1, GCCGAGCTCCCTGCAGGCGAYATTGTGATGACBCAGKCTKCA (SacI); C{kappa}-rev, CCCAAGCTTCTAGAGTTAACACTCATTCCTGTTGAAGCTCTT (HindIII). The 5' (V{lambda}-fwd) and 3' (C{lambda}-rev) primers for the {lambda}-chains: V{lambda}-fwd, GCCGAGCTCCCTGCAGGCCAGGCTGTTGTGACTCAGGAATCTGC (SseI); C{lambda}-rev, CCCAAGCTTCTAGAGTTAACASTCWGCASGRGACARACTCTTCTCCAC (HindIII).

PCR amplification was performed for 30 cycles at 94°C for 20 s, 55°C for 30 s, and 72°C for 60 s using Taq DNA polymerase. The heavy and light chain genes were inserted in a pARA7 expression vector (19). The PCR products of the heavy chain genes were digested with SmaI and SpeI and inserted between a PvuII site near the XhoI site and a SpeI site. The PCR products of the light chain genes were digested with SacI and HindIII and inserted between a SacI site near the NcoI and a HindIII site near the XbaI site. The plasmid DNA was purified with a Qiagen plasmid miniprep kit (Santa Clarita, CA). The amplified PCR products of the {lambda}-chains were directly sequenced. DNA sequences were determined by a DNA sequencer (model 373; Applied Biosystems, Foster City, CA) using a Taq dye deoxy terminator cycle sequencing kit (Applied Biosystems). Sequences were analyzed using Genetyx-Mac software (Software Development, Tokyo, Japan; version 10).

Measurement of catalytic activities and kinetic parameters

We classified all the purified mAbs as catalytic or noncatalytic, as described previously (16, 17, 18). Reaction rates were determined using HPLC by measuring the cleaved product 4 (18). The kinetic parameters, kcat and Km, were determined by fitting the data to a Michaelis-Menten equation with the program KALEDAGRAPH (Synergy Software, Reading, PA), and kuncat was determined by an initial rate analysis and by extrapolating to a zero buffer concentration (16, 17, 18).

Measurement of binding parameters

Dissociation constants were measured for two inhibitors (TSA1 and TSA2) that were used for an ELISA assay and for the compound 3, which resembles the structure of the cleaved product 4, as previously described (16, 17, 18).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
To define the 3' primer sequences of V regions in conjunction with the C regions, we determined the isotypes of the H and L chains. In the vast majority of the catAbs derived from MRL/lpr mice, the isotypes were IgG2a, and the remaining were IgG1 (Table IGo). In the group of IgG2a, the L chains were all {kappa}. In the non-catAbs from MRL/lpr mice, the isotypes were either IgG1 or IgG2a for the H chain, and {kappa} or {lambda} for the L chain (Table IGo). In BALB/c, the isotypes of the H chain were all IgG1, the L chain of the catAbs was {kappa}, and the L chain of the non-catAbs was either {lambda} or {kappa} (Table IGo).


View this table:
[in this window]
[in a new window]
  		Table I. Utilization of gene segments of Abs derived from MRL/lpr and BALB/c mice

 
Utilization of the V gene segments of the H and L chain genes in the catAb and non-catAb repertoire

Based on the isotyping as described above, we used the 5' and 3' heavy chain primers in conjunction with a C{gamma}2 or a C{gamma}1 sequence, and the 5' and 3' L chain primers in conjunction with a C{kappa} or a C{lambda} sequence for PCR amplification of cDNAs. With the sequence information, we defined each Ig gene segment for the MRL/lpr mice (Fig. 2Go) and the BALB/c mice (Fig. 3Go). The results are also summarized in Table IGo.



View larger version (58K):
[in this window]
[in a new window]
  		FIGURE 2. Comparison of amino acid sequences of a H chain V region and a L chain V region of the catAbs and non-catAbs derived from MRL/lpr mice. A, VHGN2-M is a VH germinal sequence of MRL/lpr mice (22 ) belonging to the V186.2 gene of the J558 family. MS con. denotes the MRL/lpr H chain V consensus sequence commonly conserved in the group of Abs (MS6-164 to MS5-290). B, {kappa}24A is a VL germinal sequence of BALB/c mice (GenBank database, accession AAA39052) belonging to the {kappa}24 family. MS con. denotes the MRL/lpr L chain V consensus sequence commonly conserved in the group of Abs from MS6-164 to MS5-290. Asterisks denote a catAb. The numbering of amino acid residues and the alignments of the D, JH, and JL segments follow the criteria of Kabat et al. (20 ).

 


View larger version (47K):
[in this window]
[in a new window]
  		FIGURE 3. Comparison of amino acid sequences of a H chain V region and a L chain V region of the catAbs and non-catAbs derived from BALB/c mice. A, 8-1-12-B is a VH sequence derived from a BALB/c Ig-null immature B cell line (23 ) belonging to the M315 family. BS con. denotes the BALB/c H chain V consensus sequence commonly conserved in the group of Abs (BS6-29 to BS5-18). B, {lambda}2 is a VL germinal sequence of BALB/c mice (24 ). BS con. denotes the BALB/c L chain V consensus sequence commonly conserved in the group of Abs (BS6-29 to BS5-18). Asterisks denote a catAb. The numbering of amino acid residues and the alignments of the D, JH, and JL segments followed the criteria of Kabat et al. (20 ). The codon usage of Ser at L32, Thr at L51, and Leu at L90 in the BS cons. was TCT, ACC, and CTA, respectively, whereas that of Ser (boxed) at L32 in the BS6-17 and Thr (boxed) at L51 and Leu (boxed) at L90 in the BS6-16 was TCA, ACT, and CTG, respectively.

 
In the sequence study, we found two distinct features in the utilization of the DNA segments, specifically both in the group of the catAbs derived from MRL/lpr mice, and of the non-catAbs derived from BALB/c mice. In the group of the catAbs from MRL/lpr mice, there was a major subgroup, in which the utilization of the Ig gene segments was homologously conserved (subgroup M). In this subgroup, which included MS6-164, MS6-191, MS6-192, MS6-126, MS5-255, MS5-346, MS5-393, MS5-233, MS5-200, MS5-9, and MS6-12, the VH genes belonged to V186.2 of the J558 family with 93% homology, and all the VL belonged to V{kappa}24 with 98% identity. In this group, most of the D segments were DSP2.3-4, and the rest were DQ52 (MS6-12) and DFL16.1 (MS5-9), and all the JH were JH3, although the JL were J{kappa}4 or J{kappa}5. Two of the catAbs, MS6-51 and MS5-298, were different from the subgroup M in the utilization of the gene segments. The VH gene of MS6-51 belonged to MOPC104E of the J558 family, and that of MS5-298 belonged to VHOx2 of the Q52 family, both of which had V{kappa}9 as the VL. In these cases, the D segments were DQ52 and DFL16.1, the JH was JH3, and the JL was either J{kappa}4 or J{kappa}5.

In the non-catAbs derived from MRL/lpr mice, four V genes, including V186.2, MOPC104E, VHOx2, and vhsm7-13 of the SM7 family, were utilized for the VH, and V{kappa}1, V{kappa}9, and V{kappa}24 for the VL. In this group, the D segment was DFL16.1, DSP2.7, or DSP2.9 with JH3 for the JH, and J{kappa}4 or J{kappa}5 for the JL.

In both the catAbs and non-catAbs that were derived from BALB/c mice, the utilization of the VH gene segments was completely different from the utilization of the Abs derived from MRL/lpr mice. In more than half of the non-catAbs from BALB/c, both the heavy and light chain gene segments were completely identical to one another. In this subgroup (subgroup B), which included BS6-29, BS6-16, BS6-17, BS6-35, BS6-40, BS6-37, BS6-14, and BS5-18, the VH genes belonged to M315 of the 36-60 family, the VL were all V{lambda}2, the D segments were all DQ52, the JH were all JH3, and the JL were all J{lambda}2.

Somatic hypermutations and recombinations with N addition in the catAb repertoire

Detailed analysis of recombination and mutation events by comparing the genomic consensus sequences of each segment can reveal the clonal relations of the Abs and the evolving process of the catAbs in the original Ab repertoire. Accordingly, we compared the sequences of the heavy chain V regions (H1 to H94) with the reported genomic sequences. The D and JH and the sequences of the N addition in the regions of the complementarity-determining region of the H chain (H-CDR)3 were separately classified and defined. In the {kappa}- and {lambda}-L chains, we compared the sequences of the V regions (L1 to L95) with the reported genomic sequences. The region of the JL was separately classified.

One germline sequence (VHGN2-M) of the V186.2-related germline genes completely matched the consensus sequence of the V regions of the subgroup M of the catAbs derived from MRL/lpr mice (Fig. 2GoA). This germline gene was found from the MRL/lpr mice as a nephritogenic autoantibody (22). In the subgroup M, somatic mutations, identified by comparing each sequence with the consensus sequence, were scattered throughout the H chain segments, whereas there were a few in the L chain genes, the number being 4.5 per a H chain and 2 per a L chain. In the H chain genes, mutations were more frequently found in the CDR2 region than in the other regions.

The H-CDR3 regions varied in length and sequence components (Fig. 2GoA). Utilization of the D and JH segments and the sequences of the N addition showed that the subgroup M was divided into three subsets, whose amino acid sequences were closely related to one another (Fig. 2GoA). The first subset, including MS6-164, MS6-191, MS6-192, and MS6-126, in addition to having similar D and the same JH segments, also had almost identical sequences of the N addition at the VH-D junction and the D-JH junction. The second subset (MS5-255 and MS5-346) and the third subset (MS5-393 and MS5-233) had completely identical sequences of the N addition at the VH-D, and the same JH segments. In addition, the second subset had similar D segments, and the third subset had the same D segments.

In the light chain genes, no germline sequences completely matched with the consensus sequence of the subgroup M. Subsequently, we compared the sequences with {kappa}24A, the genomic sequence that is the most homologous to the consensus sequence of the subgroup M (Fig. 2GoB). The sequences of the subgroup M all had Pro at L9, Val at L11, Pro at L15, Glu at L17, Val at L19, Asn at L28, Arg at L39, Arg at L50, Met at L51, and His at L91. We tentatively considered these conserved residues to be nonmutated (polymorphism). They are probably new germline residues that have not yet been reported. Leu at L89 in the CDR3 is one of consensus residues, but it seems to be a mutation, rather than a new germline residue, because its conservation was incomplete in the subgroup M. Other residues scattered in the sequences were counted as mutational events.

The overall results showed that the catAbs in the subgroup M utilized the H and L chain V gene segments of the same family. The subgroup M could be further separated into at least three subsets, each of which was distinguished by the different utilizations of the D and JH segments and by the sequence variability of the N addition in the H-CDR3 region.

Somatic hypermutations and recombinations with N addition in the non-catAb repertoire

In the heavy chain genes, none of the germline sequences completely matched the consensus sequence in the subgroup B (BS6-29, BS6-16, BS6-17, BS6-35, BS6-40, BS6-37, BS6-14, and BS6-18). Subsequently, we compared the sequences that were most homologous with the sequences of subgroup B, 8-1-12-B (23), which is the M315-related gene derived from a BALB/c Ig null immature B cell line (Fig. 3GoA). Of all the residues that were different from the 8-1-12-B sequence, only one residue (Arg at H52) was shared by all the sequences of the subgroup B. We tentatively excluded these residues to be a mutational event, but consider a polymorphism (new germline sequences) that has not yet been reported. The other consensus residues (Phe at H53, Thr at H56, Ser at H79, Leu at H81) found in the subgroup B could be derived from mutational events, because they were not completely conserved among Abs in the subgroup B. Phe at H53, Thr at H56, and Ser at H79 could be derived from single point mutations, whereas Leu at H81 could be derived from double mutational events. Ala (GCT), Gly (GGT), and Asn (AAT) at H32, which were observed in several Abs, could be derived from Asp (GAT) at H32 in the germline sequence, if one point mutation occurred. Ser (AGT) at H32 of BS6-14 might be due to a rare double-mutational event. Even after neglecting these consensus residues, the same mutations were found in the same positions of the CDRs and the frames in a subset of the subgroup B of the non-catAbs (BS6-29, BS6-16, BS6-17, and BS6-35), and were also found in another subset in the subgroup B (BS6-37 and BS6-40). For example, changes from Asp to Gly at H32 in the CDR1, from Asn to Asp at H43 in the frame 2, and from Tyr to Phe at H91 in the frame 3 were commonly found in BS6-16, BS6-17, and BS6-35. In BS6-40 and BS6-37, a change from Thr to Pro at H73 in the frame 3 was commonly observed.

As shown in Fig. 3GoA, the H-CDR3 regions, which were characteristic in length, very short, with only three amino acid residues, were almost completely identical among the subgroup B, with the exception of BS5-18, which had no N addition events found at either the VH-D junction or the D-JH junction.

In the light chains, we compared the sequences of the V regions with the genomic sequence of the {lambda}-L chain (V{lambda}2) (Fig. 3GoB). We found two consensus residues of Ser at H32 and Ile at L85. We defined these changes as point mutations, as they were not completely conserved in the subgroup B. Three differences in the codon usage for the same amino acid, which were found at L32, L51, and L90, were derived from a mutational event. The codon for Ser at L32 was TCA in BS6-17, and TCT in the consensus sequence. The codon for Thr at L51 was ACT in BS6-17 and ACC in the consensus sequence. The codon for Leu at L90 was CTG in BS6-16 and CTA in the consensus sequence.

The overall results showed that, in most of the non-catAbs recovered from BALB/c mice, utilization of the heavy and light chain V gene segments was completely different from that in most of the catAbs recovered from MRL/lpr mice. In the subgroup B, the recovered non-catAbs utilized the same H and L chain V gene segments. In the subgroup B, there were at least two subsets with sequence similarity. These similarities indicated that the non-catAbs in these subsets were clonally related to one another. This was not found to be the case with the catAbs derived from MRL/lpr mice. This means that these Abs were derived from one clone.

Binding and catalytic characters of the catAb and non-catAb repertoire

The hapten used in this experiment is composed of three chemical moieties: a phosphonate, a benzyl group, and a phenetyl group (Fig. 1Go). TSA1, which was used for measuring affinity toward the hapten, has all components. Of these components, the phosphonate moiety was introduced to elicit a catalytic pocket for ester hydrolysis. A phenetyl group was introduced to create a binding pocket for the substrate by a hydrophobic interaction. TSA2, which does not have this phenetyl group, was used for the binding assay to estimate the contribution of the affinity between the phenyl group and the surrounding residues in the hydrophobic binding pocket. The product analogue (3), which has a side chain benzyl group, was also used for the binding assay to estimate how a side chain benzyl group contributes to affinity for the surrounding residues in the binding pocket.

The catAbs in the subgroup M showed a strong affinity toward TSA1 (Kd = 10-10–10-7 M) (Table IIGo). The affinity of these catAbs for TSA2 was lower, but proportional to their affinity toward TSA1 (Kd = 10-4–10-2 M) (Table IIGo). All of the catAbs in the subgroup M showed affinities for product analogue (3), but the affinities were much lower than those of the other Abs (Kd > 2 mM) (Table IIGo). In the subgroup M, the catalytic activity expressed as a rate enhancement, kcat/kunca, were within the range of 104–101 (Table IIGo). In general, the higher the catalytic activity, the stronger the affinity toward TSA1 and TSA2.


View this table:
[in this window]
[in a new window]
  		Table II. Binding and catalytic parameters of Abs derived from MRL/lpr and BALB/c mice

 
The non-catAbs in the subgroup B showed even a stronger affinity toward TSA1 than the catAbs in the subgroup M (Kd = 10-11–10-8 M) (Table IIGo). The affinity of these non-catAbs for TSA2 was lower than, but proportional to their affinity for TSA1 (Kd = 10-5–10-2 M) (Table IIGo). So, these values were again stronger than the values of the catAbs in the subgroup M. The affinities of all the subgroup B for product analogue (3) (10-5–10-3 M) were stronger than those of the catAbs in the subgroup M.

In contrast to the MRL/lpr mice, the BALB/c mice had only two catAbs, which showed the catalytic activities of 100–200 kcat/kuncat and no sequence similarities to those of the catAbs from MRL/lpr, or the non-catAbs from BALB/c mice.

In summary, the cat Abs in the subgroup M bound both TSA1 and TSA2, whereas they failed to bind the product (3). However, it was not the case with the non-catAbs in the subgroup B. These Abs showed positive binding toward all three compounds, TSA1, TSA2, and the product (3), with different levels of affinity.

Key residues in the catAb repertoire

In the amino acid sequences, we surveyed the consensus residues, especially in the CDRs commonly found in the catAbs, but not found in the non-catAbs. Such residues should play an important role in catalysis. If the catAbs stabilize the tetrahedral conformation of the transition state of the substrate, such residues could be basic and/or hydrophilic amino acids. In the H chain V regions, a Lys that might be involved in a hydrogen bonding with a phosphonate moiety of the hapten was almost completely conserved at H95 in the H-CDR3 of the catAbs (Fig. 2GoA). But it was not conserved in the non-catAbs, such as MS5-389 and MS5-290, even though they were derived from MRL/lpr mice and utilized the same V and D segments as the catAb, such as MS5-9 in the subgroup M. In the subgroup M, there was only one exception (MS5-200), in which a hydrophilic Ser residue instead of a Lys residue was found at H95 (Fig. 2GoA). Of the subgroup M, only two catAbs, MS6-51 and MS5-298, did not utilize Lys at H95. MS6-51 had a Ser residue at this position, and MS5-298 had another basic amino acid, a His. Asp at H100 and Gly at H100a were also conserved among the catAbs in the subgroup M that had relatively higher catalytic activities.

In the survey of the L chain CDRs, four candidates of basic amino acid residues were identified that could form a hydrogen bond with a phosphonate moiety of the hapten (Fig. 2GoB). They were Arg at L24 and His at L27d in the CDR1, Arg at L50 in the CDR2, and His at L91 in the CDR3. They were completely conserved among the catAbs of the subgroup M and two other catAbs from MRL/lpr. All these residues were conserved in only one of the non-catAb (MS5-389). Of the subgroup M, Arg at L24 was conserved among the non-catAbs from MRL/lpr. The other three residues, His at L27d in the CDR1, Arg at L50 in the CDR2, and His at L91 in the CDR3, are candidates that might be involved in a hydrogen bond with a phosphonate moiety of the hapten.

Key residues in the non-catAb repertoire

Neither basic nor the hydrophilic amino acids were placed at H95 in the Abs derived from BALB/c mice, with the exception of Tyr at H95 in the non-catAbs, such as BS6-12 and BS6-38 (Fig. 3GoA). The striking feature found in the non-catAbs from BALB/c mice was that all the non-catAbs in the subgroup B had a common short hydrophobic motif in the H-CDR3, Leu-Gly-Pro. This was quite different from the case of the H-CDR3 of the catAbs in the subgroup M, which had relatively long and variable lengths with an abundance of hydrophilic amino acid residues. The reading frame (RF) of the DH of the catAbs of the subgroup M utilized a very common RF3, whereas that of the DH, Leu-Gly-Pro, found in the non-catAbs of the subgroup B, was a rare RF1.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The preceding results show that, even when identical haptens, immunization regimes, and screening protocols are used, the Ab repertoire selected from autoimmune disease-prone MRL/lpr mice was quite different from that selected from the normal counterpart, BALB/c mice. In the Abs selected from MRL/lpr mice, catAbs were selectively recovered, whereas in the Abs selected from BALB/c mice, the recovery of catAbs was much lower than the MRL/lpr, and non-catAbs were predominantly recovered. Based on the recovery of the catAbs, we roughly estimated that ~3% of the hybridoma supernatants from the MRL/lpr mice could be catalytic, whereas it was one-fifth of the recovery from the MRL/lpr mice in the supernatants from the BALB/c mice.

The difference in the immune response might be related to the difference in such type of clonal selections. A phylogenic diagram can demonstrate the clonal relations within the two subgroups, subgroup M and subgroup B, respectively.

Before doing so, the detailed nucleotide sequence alignments of the H-CDR region of the catAbs of the subgroup M were required to know the utilization of the N additions, D regions, and JH regions (Fig. 4Go). Based on these sequences, we developed a phylogenic diagram for the clonal relations found in the catAbs in the subgroup M (Fig. 5Go). This phylogenic diagram suggests that the obtained catAbs were derived from separate clones (oligoclonal), as they utilized different D segments and/or different N-additional sequences, even though they utilized the same heavy chain (V186.2 of J558) and light chain gene (V{kappa}24) families. This diagram shows that there could be at least five independent clonal lines, each of which utilized the same D segment and the N-additional sequences. This is what we thought at first. Even though, however, the junctional sequences at the N addition corresponding to the sequences at H95 were completely conserved with respect to a Lys residue. The subsequent somatic hypermutations, which seem to be directed for the affinity maturation toward the haptenic TSA, can be seen in each offshoot of the phylogenic tree. Even after the somatic hypermutations, which are given in each offshoot of the phylogenic tree, the lysine at H95 was completely conserved among various clones.



View larger version (29K):
[in this window]
[in a new window]
  		FIGURE 4. Comparison of nucleotide sequences of H-CDR3 regions covering the V-D-J junctional sequences of the catAbs and non-catAbs derived from MRL/lpr mice. The alignments of the VH, D, and JH segments follow the criteria of Kabat et al. (20 ). RSS, recombination signal sequence. Asterisks denote a catAb.

 


View larger version (34K):
[in this window]
[in a new window]
  		FIGURE 5. Clonal relations of the catAbs of the subgroup M from MRL/lpr mice. Squares and circles on the offshoot arrows show amino acid substitutions due to mutations with their positions in the H and L chain genes. In the number X/Y in the Ab code, X represents the affinity (Kd, nM) to 1 (TSA1, R = OH), and Y represents the rate enhancementin kcat/kuncat.

 
In contrast, most of the non-catAbs in the subgroup B seemed to come out from a single clone that could be a germline and utilized the M315 (8-1-12-B) for the heavy chain gene and V{lambda}2 for the light chain gene (Fig. 6Go). This means that the clonally related Abs, which originated from a single clone, were preferentially adapted to dominate the population upon immunization of the hapten. This speculation is supported by the finding that the recovered Abs were derived from a repertoire in the different stages of the normal process of the affinity maturation. The greater the affinity toward the haptenic TSA, the greater was the number of mutations that accumulated in the sequences.



View larger version (23K):
[in this window]
[in a new window]
  		FIGURE 6. Clonal relations of the non-catAbs of the subgroup B from BALB/c mice. The circles on the offshoot arrows show amino acid substitutions due to mutations with their positions in the heavy and light chain genes. The number on the Ab code represents the affinity (Kd, nM) to 1 (TSA1, R = OH).

 
The difference in the Ab evolution observed between these mouse strains is also reflected in their three-dimensional structures, both with/without TSA haptens. There are several preceded literatures of the three-dimensional structure of the catAbs useful for better understanding the interacting residues for catalysis. For example, in the structure of the CNJ206 complexed with the TSA ligand (25), the ligand is oriented with a p-nitrophenyl group at the bottom of the apolar pocket, where hydrophobic and van der Waals interactions predominate, and its phosphonate group is located at the mouth of the cavity with specific hydrogen bonds (26). In the structure of 17E8, an aryl leaving group buried deeply at the bottom of an apolar binding pocket and the hydrogen bonds from the backbone NH and the side chains of two cationic residues provided complementary electrostatic/hydrogen-bonding interactions with the phosphonyl oxygens (27). In the case of 48G7, the apolar binding pocket for the aryl moiety of the hapten was buried, and the phosphonate group was held in place near the mouth of the pocket through specific electrostatic and hydrogen-bonding interactions with the backbone or the side chain of the residues (28). In this case, a second pocket for the alkyl side chain of the hapten was also present. In the case of 43C9, the crystal structure obtained after three-dimensional modeling revealed an unexpected extended {beta}-sheet that created a deep Ag binding site with a hydrophobic pocket that enclosed the p-nitrophenyl group (29, 30). It differed from structures in other reports in that it has amidase activity with two-step reactions, formation of a covalent acyl-Ab intermediate, followed by deacylation.

In essence, the overall themes commonly seen in these cases are: 1) the aryl moiety of the hapten is deeply buried in the pocket through hydrophobic interactions, and 2) the phosphonate group is held in place near the mouth of the pocket through specific electrostatic and hydrogen-bonding interactions. In the residues, which interact with the phosphonyl oxygens, the basic amino acids, such as a His (H35) for CNJ206, 17E8, 48G7, and 43C9, and an Arg (L96) for 17E8, 48G7, and 43C9, were commonly seen. In fact, when the former three Abs were overlaid, the phosphonate groups occupied similar locations (31). The most important variations for catalysis are the residues that form stabilizing contacts with the phosphonyl oxygens of the TSA, because esterolytic catalysis by the catAbs proceeds by facilitating a direct hydroxide attack on the scissile carbonyl of the substrate by stabilizing the oxyanion intermediate and flanking transition state through specific hydrogen bonds and/or electrostatic interactions.

The structure and orientation of the moieties used in this study, the alkyl ester (4) and its haptenic TSA1, are essentially similar to the reported aryl esters and their TSAs (31). Therefore, the fundamental theme of the catAbs in these reports could be applied to our catAbs. Based on the above assumption and the results on the affinities toward TSAs and the product analogue, we proposed a model structure of the Abs (Fig. 7Go). The structure suggests that the Abs essentially accommodate the three major pockets (or interacting residues), which interact with the phenyl moiety, the side chain benzyl moiety, and the phosphonyl group, respectively. The affinity studies suggested that in the catAbs of the subgroup M, there could be two major pockets for binding: a hydrophobic pocket for the phenetyl group, which could be buried deeply in the cleft, and interacting residues for the phosphonyl group, which is placed at the mouth of the hydrophobic pocket and which is a key for catalysis. A Lys at H95, which could be a key residue for interaction with the phosphonyl group to form a hydrogen bond, which creates a catalytic pocket for the substrate, because it is a basic amino acid and is almost commonly conserved among the catAbs from MRL/lpr mice, is not conserved in the non-catAbs from BALB/c mice. This speculation is supported by the site-directed mutagenesis data of the catAb (manuscript in preparation). Taken together, the fundamental theme of the catAbs could be applied to our catAbs in the subgroup M, but the interacting residues (Lys at H95 and His at L91) that constitute the catalytic pocket are quite different from the reported esterase catAbs (His at H35 and Arg at L96).



View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 7. Schematic illustrations of the binding modes of catAbs from MRL/lpr mice and of non-catAbs from BALB/c mice to TSAs and product analogue. + and +++, Indicate pockets that are positive and strongly positive, respectively, with respect to binding of the moieties, such as phosphonate, phenetyl, or benzyl. –, Indicates a pocket that does not bind with the moieties.

 
In the non-catAbs obtained from BALB/c mice, however, there could be two hydrophobic pockets: one for the phenyl group and another that accommodates the benzyl side chain. The presence of a rare reading frame RF1 for the H-CDR3 and the rare {lambda}2 light chain genes utilized in these Abs, which were reported to facilitate the hydrophobic interactions (21, 32), may constitute the second hydrophobic pocket for the benzyl moiety. The residues resulting from these changes could result in a stronger binding to the TSA, but they might also result in no residue(s) that interacts with the phosphonyl group. In these models, the catalytic activities of the Abs may be exerted after the surrounding residues interact with the phosphonyl group. But this interaction may force the phosphonyl group to have an inappropriate orientation, so that it cannot interact with the surrounding residues. As a result, the Abs would no longer be catalytic.

What is the source of the catAbs? Could the catAbs be derived from an abnormal repertoire that specifically exists in or is enriched in MRL/lpr mice? Autoreactive Abs, such as anti-nuclear Abs, Abs to ribonucleoproteins, ssDNAs, or dsDNAs, which are normally eliminated, but have a high incidence in autoimmune strains such as MLR/lpr mice (33, 34), seem like the most likely source of the catAbs. As a consequence, such mouse strains develop multisystem autoimmunity, including a lupus-like glomerulonephritis and arthritis (35, 36).

Thus, the amino acid sequences support the assumption that catAbs are derived from a repertoire of Abs to DNAs. However, this may not be necessarily the case in this study for the following reasons. First, the autoimmune strain NZB x NZW, which develops SLE, also produces Abs to DNAs that play a demonstrable role in the pathogenesis of disease in this strain (37, 38, 39), but this strain did not produce a higher incidence of catAbs as compared with the normal counterpart (13). Second, the catAbs from MRL/lpr mice obtained in this study, and those reported by Tawfik et al. (13) did not show a positive cross-react with either the bovine ssDNAs or dsDNAs (data not shown). Alternatively, the catAbs derived from MRL/lpr mice could be derived from the repertoire of the other autoreactive Abs to the molecules. Such molecules could be ones like flexible phosphodiester polymers such as RNA, teichoic acid, phosphotyrosine, phosphocholine, or other molecules whose distribution of phosphates (or equivalent negatively charged epitopes) conforms with the available contact with the combining sites of the Abs (40, 41, 42, 43, 44). The thorough survey on the data base did not show any identical sequences to our catAbs, but we found only one Ab that showed the highest similarity to one of our catAbs in the heavy chain gene sequence (GenBank accession no. L48662). It was an anti-DNA Ab, which was derived from C3H lpr, and had a Lys at H95, although the amino acid sequence of the H-CDR3 was quite different from our catAbs.

The lpr gene encodes the cell surface receptor molecule Fas, and its ligand FasL is encoded by the gld gene (45, 46, 47, 48). MRL/lpr mice are characterized by a defect in Fas expression. As a result, they develop an autoimmune syndrome associated with massive lymphoaccumulation and excessive production of many of the autoreactive Ab specificities associated with the human disease SLE. Although Fas-FasL has an essential role in the tolerance to self Ag, its exact function is not clear. Immunization of prediseased lpr mice with conventional hapten-carrier conjugates results in germinal center reactions that are essentially normal (49, 50). Experiments with transgene-encoded B cell receptors have shown that the central tolerance mechanisms are intact in lpr mice (51, 52). However, Fas in combination with CD40 seems to play a role in peripheral tolerance of the B cells that had bound self Ags chronically, and had thus became anergic (53). Its most important role seems to prevent the nonspecific activation of such anergic (or indifferent) autoreactive B cells that normally circulate in the periphery (54, 55).

If the above story is the case, MRL/lpr mice defective in Fas-FasL-mediated apoptosis were forced to nonspecifically accumulate the abnormal B cells that were normally inactivated and eliminated. In such a repertoire, there was an ancestral subset for esterolytic catalysts. Upon immunization and ELISA screening with the haptenic TSA, such a repertoire of B cells with diversified gene sequences was predominantly captured after it evolved with somatic mutations, but without changing the key residue Lys at the junctional H95, which interacted with the phosphonyl oxygen. It is noteworthy that the Abs with the key residue Lys at the junctional H95 already existed in the initial repertoire of MRL/lpr mice. In normal BALB/c mice, however, that would not happen, because of the lowest abundance (~1% in the data base; 20) or elimination of the autoreactive B cell repertoire, in which a subset of B cells with the residue in the binding site of the Abs complementing to the phosphonyl oxygen was included. Instead, another subset with hydrophobic residues in the binding site of the Abs was selectively captured to specifically bind to the haptenic TSA.


   Acknowledgments
 
We thank Kiyoe Inoue and Naoko Murakami for their technical assistance. We also thank Fumiyo Satito, Noriko Niikura, Keiko Shindo, and Fukumi Kuratomi for their assistance, Katsumi Hamada (Immuno-Biological Laboratories, Gunma, Japan) for preparation of a series of mAbs, and Dr. Yoshiki Yamasaki (Pharmaceutical Frontiers Research Laboratories, Japan Tobacco, Yokohama, Japan) for discussion and valuable comments.


   Footnotes
 
1 This work was supported by the New Energy and Industrial Technology Development Organization as a research and development project of the Industrial Science and Technology Frontier Program. Back

2 Current address: Plant Science Center, Laboratory for Remediation Research, The Institute of Physical and Chemical Research (Japan), Hirosawa 2-1, Wako, Saitama 351-0198, Japan. Back

3 Current address: Medical Research Laboratories, Taisho Pharmaceutical Co., Ltd., Yoshinocho 1-403, Omiya, Saitama 330-8530, Japan. Back

4 Address correspondence and reprint requests to Dr. Yoshisuke Nishi, Laboratory of Life Science and Biomolecular Engineering, Japan Tobacco Inc., 6-2 Umegaoka, Aoba-ku, Yokohama, Kanagawa, 227-8512, Japan. E-mail address: yoshisuke.nishi{at}ims.jti.co.jp Back

5 Abbreviations used in this paper: TSA, transtition-state analogue; catAb, catalytic Ab; CDR, complementarity-determining region; FasL, Fas ligand; H-CDR, complementarity-determining region of the H chain; KLH, keyhole limpet hemocyanin; RF, reading frame; SLE, systemic lupus erythematosus. Back

Received for publication June 13, 2001. Accepted for publication August 16, 2001.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Pauling, L.. 1948. Chemical achievement and hope for the future. Am. Sci. 36:51.
  2. Jenks, W. P.. 1969. Catalysis in Chemistry and Enzymology 288. McGraw-Hill, New York.
  3. Tramantano, A., K. D. Janda, R. A. Lerner. 1986. Catalytic antibodies. Science 234:1566.[Abstract/Free Full Text]
  4. Pollack, S. J., J. W. Jacobs, P. G. Schultz. 1986. Selective chemical catalysis by an antibody. Science 234:1570.[Abstract/Free Full Text]
  5. Blackburn, M., A. Data, H. Denham, Jr P. Wentworth. 1998. Catalytic antibodies. Adv. Phys. Org. Chem. 3:249.
  6. Lerner, R. A., S. J. Benkovic, P. G. Schultz. 1991. At the crossroads of chemistry and immunology: catalytic antibodies. Science 252:659.[Abstract/Free Full Text]
  7. Benkovic, S. J.. 1992. Catalytic antibodies. Annu. Rev. Biochem. 61:29.[Medline]
  8. Davis, M. M., P. J. Bjorkman. 1988. T-cell antigen receptor genes and T-cell recognition. Nature 334:395.[Medline]
  9. Baca, M., T. S. Scanlan, R. C. Stephenson, J. A. Wells. 1997. Phage display of a catalytic antibody to optimize affinity for transition-state analog binding. Proc. Natl. Acad. Sci. USA 94:10063.[Abstract/Free Full Text]
  10. Takahashi, N., H. Kakinuma, L. Liu, Y. Nishi, I. Fuji. 2001. In vitro abzyme evolution: optimization of antibody recognition for catalysis. Nat. Biotechnol. 19:563.[Medline]
  11. Alarcón-Riquelme, M. E., C. Fernández. 1995. CDR3 regions in the preimmune VH B cell repertoire of lpr mice. Clin. Exp. Immunol. 101:73.[Medline]
  12. Erikson, J., L. Mandik, A. Bui, A. Eaton, H. Noorchashm, K. T. Nguyen, J. H. Roark. 1998. Self-reactive B cells in nonautoimmune and autoimmune mice. Immunol. Res. 17:49.[Medline]
  13. Tawfik, D. S., R. Chap, B. S. Green, M. Sela, Z. Eshhar. 1995. Unexpectedly high occurrence of catalytic antibodies in MRL/lpr and SJL mice immunized with a transition-state analog: is there a linkage to autoimmunity?. Proc. Natl. Acad. Sci. USA 92:2145.[Abstract/Free Full Text]
  14. Paul, S., D. J. Volle, C. M. Beach, D. R. Johnson, M. J. Powell, R. J. Massey. 1989. Catalytic hydrolysis of vasoactive intestinal peptide by human autoantibodies. Science 244:1158.[Abstract/Free Full Text]
  15. Shuster, A. M., G. V. Gololobov, O. A. Kvashuk, A. E. Bogomolova, I. V. Smirnov, A. G. Gabibov. 1992. DNA hydrolyzing autoantibodies. Science 256:665.[Abstract/Free Full Text]
  16. Takahashi, N., H. Kakinuma, K. Hamada, K. Shimazaki, K. Takahashi, S. Niihata, Y. Aoki, H. Matsushita, Y. Nishi. 1999. Efficient screening for catalytic antibodies using a short transition-state analog and detailed characterization of selected antibodies. Eur. J. Biochem. 261:108.[Medline]
  17. Takahashi, N., H. Kakinuma, K. Hamada, K. Shimazaki, Y. Yamasaki, M. Matsushita, Y. Nishi. 2000. Improved generation of catalytic antibodies by MRL/MPJ-lpr/lpr autoimmune mice. J. Immunol. Methods 235:113.[Medline]
  18. Kakinuma, H., K. Shimazaki, N. Takahashi, K. Takahashi, S. Niihata, Y. Aoki, K. Hamada, H. Matsushita, Y. Nishi. 1999. Comparison of phosphonate transition state analogs for inducing catalytic antibodies and evaluation of key structural factors by an ab initio study. Tetrahedron 55:2559.
  19. Miyashita, H., T. Hara, R. Tanimura, S. Fukuyama, C. Cagnon, A. Kohara, I. Fujii. 1997. Site-directed mutagenesis of active site contact residues in a hydrolytic abzyme: evidence for an essential histidine involved in transition state stabilization. J. Mol. Biol. 267:1247.[Medline]
  20. Kabat, E. A., T. T. Wu, H. M. Perry, K. S. Gottesman, C. Foeller. 1991. Sequences of Proteins of Immunological Interest National Technical Information Service (NTIS), National Institutes of Health, Bethesda, MD.
  21. 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.[Medline]
  22. Ono, M., T. Yamamoto, M. Nose. 1995. A new germline VH gene encoding a mouse nephritogenic autoantibody. Immunogenetics 42:426.[Medline]
  23. Komori, T., H. Sugiyama, S. Kishimoto. 1989. A novel VHDJH to JH joining that induces H chain production in an Ig-null immature B cell line. J. Immunol. 143:1040.[Abstract]
  24. Tonegawa, S., A. M. Maxam, R. Tizard, O. Bernard, W. Gilbert. 1978. Sequence of a mouse germ-line gene for a variable region of an immunoglobulin light chain. Proc. Natl. Acad. Sci. USA 75:1485.[Abstract/Free Full Text]
  25. Zemel, R., D. G. Schindler, D. S. Tawfik, Z. Eshhar, B. S. Green. 1994. Differences in the biochemical properties of esterolytic antibodies correlate with structural diversity. Mol. Immunol. 31:127.[Medline]
  26. Charbonnier, J.-B., E. Carpenter, B. Gigant, B. Golinelli-Pimpaneau, Z. Eshhar, B. S. Green, M. Knossow. 1995. Crystal structure of the complex of a catalytic antibody Fab fragment with a transition state analog: structural similarities in esterase-like catalytic antibodies. Proc. Natl. Acad. Sci. USA 92:11721.[Abstract/Free Full Text]
  27. Zhou, G. W., J. Guo, W. Huang, R. J. Fletterick, T. S. Scanlan. 1994. Crystal structure of a catalytic antibody with a serine protease active site. Science 265:1059.[Abstract/Free Full Text]
  28. Patten, P. A., N. S. Gray, P. L. Yang, C. B. Marks, G. J. Wedemayer, J. J. Boniface, R. C. Stevens, P. G. Schult. 1996. The immunological evolution of catalysis. Science 271:1086.[Abstract]
  29. Roberts, V. A., J. Stewart, S. J. Benkovic, E. D. Getzoff. 1994. Catalytic antibody model and mutagenesis implicate arginine in transition-state stabilization. J. Mol. Biol. 235:1098.[Medline]
  30. Thayer, M. M., E. H. Olender, A. S. Arvai, C. K. Koike, I. L. Canestrelli, J. D. Stewart, S. J. Benkovic, E. D. Getzoff, V. A. Roberts. 1999. Structural basis for amide hydrolysis catalyzed by the 43C9 antibody. J. Mol. Biol. 291:329.[Medline]
  31. MacBeath, G., D. Hilvert. 1996. Hydrolytic antibodies: variations on a theme. Chem. Biol. 3:433.[Medline]
  32. Azuma, T., V. Igras, E. B. Reilly, H. N. Eisen. 1984. Diversity at the variable-joining region boundary of {lambda} light chains has a pronounced effect on immunoglobulin ligand-binding activity. Proc. Natl. Acad. Sci. USA 81:6139.[Abstract/Free Full Text]
  33. Shlomchik, M. J., A. H. Aucoi, D. S. Pisetsky, M. G. Weigert. 1987. Structure and function of anti-DNA autoantibodies derived from a single autoimmune mouse. Proc. Natl. Acad. Sci. USA 84:9150.[Abstract/Free Full Text]
  34. Radic, M. Z., M. Weigert. 1994. Genetic and structural evidence for antigen selection of anti-DNA antibodies. Annu. Rev. Immunol. 12:487.[Medline]
  35. Kelley, V. E., J. B. Roths. 1985. Interaction of mutant lpr gene with background strain influences renal disease. Clin. Immunol. Immunopathol. 37:220.[Medline]
  36. Hang, L., A. N. Theofilopoulos, F. J. Dixon. 1982. A spontaneous rheumatoid arthritis-like disease in MRL/l mice. J. Exp. Med. 155:1690.[Abstract/Free Full Text]
  37. Shlomchik, M., M. Mascelli, H. Shan, M. Z. Radic, D. Pisetsky, A. Marshak-Rothstein, M. Weigert. 1990. Anti-DNA antibodies from autoimmune mice arise by clonal expansion and somatic mutation. J. Exp. Med. 171:265.[Abstract/Free Full Text]
  38. Tillman, D. M., N. T. Jou, R. J. Hill, T. N. Marion. 1992. Both IgM and IgG anti-DNA antibodies are the products of clonally selective B cell stimulation in (NZB x NZW)F1 mice. J. Exp. Med. 176:761.[Abstract/Free Full Text]
  39. Krishnan, M. R., N. T. Jou, T. N. Marion. 1996. Correlation between the amino acid position of arginine in VH-CDR3 and specificity for native DNA among autoimmune antibodies. J. Immunol. 157:2430.[Abstract]
  40. Wicken, A. J., K. W. Knox. 1975. Lipoteichoic acids: a new class of bacterial antigen. Science 187:1161.[Free Full Text]
  41. Kabat, E. A., K. G. Nickerson, J. Liao, L. Grossbard, E. F. Osserman, E. Glickman, L. Chess, J. B. Robbins, R. Schneerson, Y. H. Yang. 1986. A human monoclonal macroglobulin with specificity for {alpha} (2->8)-linked poly-N-acetyl neuraminic acid, the capsular polysaccharide of group B meningococci and Escherichia coli K1, which crossreacts with polynucleotides and with denatured DNA. J. Exp. Med. 164:642.[Abstract/Free Full Text]
  42. Eilat, D., D. M. Webster, A. R. Rees. 1988. V region sequences of anti-DNA and anti-RNA autoantibodies from NZB/NZW F1 mice. J. Immunol. 141:1745.[Abstract]
  43. Lafer, E. M., J. Rauch, Jr C. Andrzejewski, D. Mudd, B. Furie, B. Furie, R. S. Schwartz, B. D. Stollar. 1981. Polyspecific monoclonal lupus autoantibodies reactive with both polynucleotides and phospholipids. J. Exp. Med. 153:897.[Abstract/Free Full Text]
  44. Minota, S., W. N. Jarjour, N. Suzuki, Y. Nojima, R. A. Roubery, T. Miura, A. Yamada, T. Hosoya, F. Takaku, J. B. Winfield. 1991. Autoantibodies to nucleolin in systemic lupus erythematosus and other diseases. J. Immunol. 146:2249.[Abstract]
  45. Watanabe-Fukunaga, R., C. I. Brannan, N. G. Copeland, N. A. Jenkins, S. Nagata. 1992. Lymphoproliferation disorder in mice explained by defects in Fas antigen that mediates apoptosis. Nature 356:314.[Medline]
  46. Adachi, M., R. Watanabe-Fukunaga, S. Nagata. 1993. Aberrant transcription caused by the insertion of an early transposable element in an intron of the Fas antigen gene of lpr mice. Proc. Natl. Acad. Sci. USA 90:1756.[Abstract/Free Full Text]
  47. Takahashi, T., M. Tanaka, C. I. Brannan, N. A. Jenkins, N. G. Copeland, T. Suda, S. Nagata. 1994. Generalized lymphoproliferative disease in mice, caused by a point mutation in the Fas ligand. Cell 76:969.[Medline]
  48. Lynch, D. H., M. L. Watson, M. R. Alderson, P. R. Baum, R. E. Miller, T. Tough, M. Gibson, T. Davis-Smith, C. A. Smith, K. Hunter. 1994. The mouse Fas-ligand gene is mutated in gld mice and is part of a TNF family gene cluster. Immunity 1:131.[Medline]
  49. Very, D. L., D. J. Panka, D. Weissman, L. Wysocki, A. Marshak-Rothstein. 1993. Lack of connectivity between the induced and autoimmune repertoires of lpr/lpr mice. Immunology 80:518.[Medline]
  50. Smith, K. G. C., G. J. V. Nossal, D. M. Tarlinton. 1995. FAS is highly expressed in the germinal center but is not required for regulation of the B-cell response to antigen. Proc. Natl. Acad. Sci. USA 92:11628.[Abstract/Free Full Text]
  51. Rathmell, J. C., C. C. Goodnow. 1994. Effects of the lpr mutation on elimination and inactivation of self-reactive B cells. J. Immunol. 153:2831.[Abstract]
  52. Rubio, C. F., J. Kench, D. M. Russell, R. Yawger, D. Nemazee. 1996. Analysis of central B cell tolerance in autoimmune-prone MRL/lpr mice bearing autoantibody transgenes. J. Immunol. 157:65.[Abstract]
  53. Rathmell, J. C., S. E. Townsend, J. C. Xu, R. A. Flavell, C. C. Goodnow. 1996. Expansion or elimination of B cells in vivo: dual roles for CD40- and Fas (CD95)-ligands modulated by the B cell antigen receptor. Cell 87:319.[Medline]
  54. Lagresle, C., C. Bella, P. T. Daniel, P. H. Kramer, T. Defrance. 1995. Regulation of germinal center B cell differentiation: role of the human APO-1/Fas (CD95) molecule. J. Immunol. 154:5746.[Abstract]
  55. Ozdemirli, M., M. El-Khatib, L. C. Foote, J. K. Wang, A. Marshak-Rothstein, T. Rothstein, S. T. Ju. 1996. Fas (CD95)/Fas ligand interactions regulate antigen-specific, major histocompatibility complex-restricted T/B cell proliferative responses. Eur. J. Immunol. 26:415.[Medline]



This article has been cited by other articles:


Home page
 J BiochemHome page
Y. Nishi, N. Yamamoto, K. Shimazaki, N. Takahashi-Ando, H. Kakinuma, S. Jialin, S. N. Ruzheinikov, T. A. Muranova, D. W. Rice, and Y. Kajihara
Mechanistic Analysis of the Phosphonate Transition-state Analogue-derived Catalytic and Non-catalytic Antibody
J. Biochem., October 1, 2007; 142(4): 421 - 433.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
S. Paul, S. Planque, Y.-X. Zhou, H. Taguchi, G. Bhatia, S. Karle, C. Hanson, and Y. Nishiyama
Specific HIV gp120-cleaving Antibodies Induced by Covalently Reactive Analog of gp120
J. Biol. Chem., May 23, 2003; 278(22): 20429 - 20435.
[Abstract] [Full Text] [PDF]

Home page
 FASEB J.Home page
Y. BANGALE, S. KARLE, S. PLANQUE, Y.-X. ZHOU, H. TAGUCHI, Y. NISHIYAMA, L. LI, R. KALAGA, and S. PAUL
VIPase autoantibodies in Fas-defective mice and patients with autoimmune disease
FASEB J, April 1, 2003; 17(6): 628 - 635.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar</