HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Marshall, B. Articles by Mellor, A. Search for Related Content PubMed PubMed Citation Articles by Marshall, B. Articles by Mellor, A.

Alternative Splicing and Hypermutation of a Nonproductively Rearranged TCR -Chain in a T Cell Hybridoma¹

Brendan Marshall^2,*, Ruth Schulz, Min Zhou^* and Andrew Mellor^*

^* Institute of Molecular Medicine and Genetics, Program in Molecular Immunology, Medical College of Georgia, Augusta, GA 30912; and Division of Molecular Immunology, National Institute for Medical Research, Mill Hill, London, United Kingdom

	Abstract

Top Abstract Introduction Materials and Methods Results Discussion References

 
Like Ig genes, TCR genes are formed by somatic rearrangements of noncontiguous genomic V, J, and C regions. Unlike Ig genes, somatic hypermutation of TCR V regions is an infrequent event. We describe the occurrence of spontaneous hypermutation in a nonproductively rearranged TCR -chain gene in a clonal T cell hybridoma that had lost its productively rearranged -chain. The mutating hybridoma was eventually supplanted in culture by a nonmutating variant that had restored an open reading frame in the nonproductively rearranged TCR -chain through the use of cryptic splice sites in the V region. Evidence is presented for the presence of cDNA reverse transcripts of the TCR -chain within the hybridoma, suggesting a role for reverse transcriptase in the generation of mutations.

	Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

 
The somatic rearrangement of Ig and TCR genes leads to both in-frame (productive) and out-of-frame (nonproductive) rearrangements. A nonproductive rearrangement is not necessarily disastrous for a T or B lymphocyte, as a second rearrangement may take place on the homologous chromosome. If productive, this results in the cell transcribing both productive and nonproductive rearrangements, although only the former is used for assembling functional Igs or TCRs. Nevertheless, the nonproductive rearrangement continues to be transcribed, although a number of studies indicate that the steady-state level of mRNA from nonproductively rearranged Ig and TCR genes is substantially lower than that of productively rearranged genes (1, 2, 3). Furthermore, nonproductive rearrangements of Ig genes also accumulate mutations during the process of somatic hypermutation (SH).³ This has proven useful in identifying the preferred target sequences of the mutational machinery without complications caused by the effects of Ag-mediated selection observed in productive rearrangements.

Despite the fact that TCRs and Igs are assembled by similar mechanisms, T cells are not generally considered to undergo hypermutation due to the stringent requirements of TCR-MHC molecule interaction. Thymic selection of T cell repertoires is designed to eliminate T cells with receptors that exhibit either strong or weak interactions with their MHC class I or II ligands (4, 5). Any changes in the amino acid sequence of the TCR due to mutation subsequent to selection, it is argued, could have potentially negative consequences in terms of autoimmunity. This reasoning is supported by several studies that have searched for but failed to find evidence of SH in T cells (6, 7, 8, 9).

In general, understanding the process of SH has been hindered by the absence of suitable cell lines that undergo hypermutation in vitro and that are amenable to experimentation. Several B cell lines have been described that undergo hypermutation spontaneously, including a human follicular B cell lymphoma (10) and the differentiated murine B cell line 18.81 (11, 12). Identifying the external stimuli responsible for hypermutation has only recently become possible with the triggering of nonspontaneous hypermutation in vitro by surface Ig cross-linking and coculture with helper T cells of a Burkitt’s lymphoma cell line (13). SH triggered in vivo has also been shown to be potentiated in vitro by provision of activated T cells (14).

The mechanism of Ig gene SH remains to be understood. Models that have been proposed can be grouped into two general categories depending on whether they invoke gene conversion or transcription to explain SH. Support for the former derives from studies on transgenic mice in which sequence transfers have been observed between distinct tandem VDJ regions of the same Ab-H chain transgene (15). Gene conversion has also been shown to be responsible for Ab diversity in chicken and rabbit (16, 17). Transcriptional models are supported by transgenic experiments in which transcriptional promoters have been inserted at different regions within the Ig locus and shown not only to initiate the production of new transcripts but also to direct hypermutation to the transcribed regions (18, 19). Transcription is suggested to bring about mutation by targeting error-prone DNA repair to transcriptional pause sites.

To the best of our knowledge, only two reports exist of hypermutation in T cells. It was initially reported that hypermutation of TCR V-chains but not Vß-chains occurred in splenic germinal center T cells of hapten-immunized mice (20). Features characteristic of Ig gene mutation were also noted, including DNA strand polarity and base substitution bias. There was also a substantial bias for hypermutation of nonproductive -chains compared with productive -chains, indicating a possible bias against surface expression of mutated TCR V-chains. More recently, it has also been reported that hypermutation of the TCR Vß-chain occurred in splenic white pulps of HIV-1-positive patients (21). In the present study we have analyzed the TCR V sequences present in a clonal murine T cell hybridoma that was initially identified as being autoreactive and CD8⁺. Our initial intention was to identify and clone both the V- and Vß-chains of the TCR to construct a transgenic mouse that carried a productively rearranged, autoreactive TCR in the germline. However, during the in vitro growth of the hybridoma, it became apparent that the hybridoma had lost surface expression of the TCR, as is not uncommon in cells of this type. Therefore, we investigated both the type and sequence of V- and Vß-chains present within the hybridoma to confirm that a productively rearranged - or ß-chain had been lost from the cell. To our surprise, we observed substantial sequence variation within the single, nonproductive V sequence identified. We show that this is due to hypermutation at the V locus. Subsequently, we observed the emergence of a nonmutating variant that had reopened the reading frame of the previously nonproductive -chain through cryptic splice site usage.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

 
IA3 hybridoma production and culture

Splenocytes from CD2-Kb mice were stimulated with irradiated (20 Gy) CBK splenocytes in 24-well plates. The CBK strain expresses the H2Kb transgene under the control of its own promoter, while CD2-Kb mice express the H2Kb transgene under the control of the CD2 minigene cassette principally on cells of the lymphoid lineage (22). Responder cells were stimulated approximately every 7 days, and on the third stimulation, irradiated CBK cells were added to responders in media containing recombinant IL-2. After 3–4 days, dead stimulator cells were removed by Ficoll separation, subsequent to which live cells were washed twice in Iscove’s modified Dulbecco’s medium (IMDM) with 2% FCS. Live cells were then added to the BW5147 TCRß-negative thymoma, and fusion was conducted according to standard procedures. Following 24 h in IMDM with 10% FCS, selection was performed in 1x hypoxanthine/aminopterin/thymidine (HAT) medium (Life Technologies, Grand Island, NY). Hybridomas were transferred to 50-ml flasks and passaged twice in IMDM with 10% FCS and 1x HAT. IA3 was selected as a CD3⁺CD4^- hybridoma by FACS analysis.

FACS analysis

IA3 was washed in sterile PBS containing 5% FCS and stained using filter-sterilized mAbs. Sorting was performed on a FACStar^Plus (Becton Dickinson, Mountain View, CA).

5'-rapid amplification of cDNA ends (5'-RACE)

Total cellular RNA was isolated from IA3 using RNA-STAT 60 (Tel-Test B; Tel-Test, Friendswood, TX), and 1 µg was used in a 5'-RACE reaction performed with a 5'-RACE kit (Life Technologies) and the following reverse primers: C, 5'-TGGCGTTGGTCTCTTTGAAG-3' (CAR1), and Cß, 5'-CCAGAAGGTAGCAGAGACCC-3' (CBR1) for the first-strand cDNA synthesis step. cDNA synthesis was performed at 42°C for 50 min. For the subsequent PCR step, the following reverse primers were used: C, 5'-GCAGGTGAAGCTTGTCTGG-3' (CAR2), and Cß, 5'-CTTGGGTGGAGTCACATTTCTC-3' (CBR2). Amplification was performed under the following conditions: 94°C for a 3-min initial denaturation, followed by 35 cycles of 94°C for 30 s, 55°C for 1 min, and 72°C for 1 min. The final RACE product was generated by performing a nested PCR using 1 µl of the previous PCR reaction as a template together with the following reverse primers: C, 5'-TCCAGATTCCATGGTTTTCGG-3' (CAR3), and Cß, 5'-ATTTCTCAGATCCTCTAG-3' (CBR3). The amplification conditions were the following: 94°C for a 3-min initial denaturation, followed by 35 cycles of 94°C for 30 s, 62°C for 1 min, and 72°C for 1 min. RACE products were cloned into pGEM-T (Promega, Madison, WI), and individual clones were sequenced using fluorescent dye terminator technology (Applied Biosystems, Foster City, CA).

Polymerase chain reaction

Rearranged TCR -chains were amplified from IA3 genomic DNA using the following primers: 5'-CAACCACACAAGCACCATG-3' (A1; forward) and 5'-CCACCAGCTGCGTCCCATCAC-3' (A2; reverse). The conditions were 94°C for a 4-min initial denaturation, followed by 35 cycles of 94°C for 30 s, 58°C for 1 min, and 72°C for 1 min. Rearranged TCR ß-chains were amplified from IA3 genomic DNA using the following primers: 5-'TCTGCCCTCAATCTGCCATG-3' (B1; forward) and 5'-CGCACCAAAGTACAAGGTG-3' (B2; reverse). The conditions were 94°C for a 3-min initial denaturation, followed by 35 cycles of 94°C for 30 s, 58°C for 1 min, and 72°C for 1 min.

The first exon of the C region was amplified using the following primers: 5'-CCCAGAACCTGCTGTGTAC-3' (C1; forward) and 5'-TGAACTGGGGTAGGTGGC-3' (C2; reverse). The conditions were 94°C for a 4-min initial denaturation, followed by 35 cycles of 94°C for 30 s, 58°C for 1 min, and 72°C for 1 min.

Exons 2–4 and the intervening two introns of the mouse -actin gene were amplified using the following primers: 5'-CCATGTGCGACGAAGACG-3' (forward) and 5'-GAATCCAACACGATGCCG-3' (reverse). The conditions were 94°C for a 3-min initial denaturation, followed by 35 cycles of 94°C for 30 s, 55°C for 1 min, and 72°C for 1 min, to produce a PCR product of 710 bp.

RT-PCR of V10J24 mRNA was performed with an Access RT-PCR kit (Promega) for 40 cycles using primers A1 (forward) and CAR3 (reverse) with the amplification conditions of 48°C for 45 min (1 cycle) and 94°C for 2 min (1 cycle), followed by 40 cycles of 94°C for 30 s, 58°C for 1 min, and 68°C for 2 min. A final step of 68°C for 5 min completed the reaction. PCR products were cloned into the pGEM-T vector (Promega).

For experiments in which VJC -chains or VDJC ß-chains were amplified directly from genomic DNA without reverse transcriptase, primer sets A1-CAR1 and B1-CBR1, respectively, were used. Amplification conditions for A1-CAR1 were the following: 94°C for 3 min (1 cycle), followed by 35 cycles of 94°C for 30 s, 58°C for 1 min, and 72°C for 1 min. For the B1-CBR1 primer pair, amplification conditions were 94°C for 3 min, followed by 35 cycles of 94°C for 30 s, 60°C for 1 min, and 72°C for 1 min.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

 
Characterization of hybridoma IA3

The T cell hybridoma IA3 was isolated during a screen for autoreactive, H2-Kb-specific, T cells in the spleens of H2-Kb transgenic mice. These mice (known as CD2-Kb-3) contain the H2-Kb structural gene under the control of the CD2 minigene cassette and express the transgene principally on cells of the lymphoid lineage, although some myeloid expression is seen (22). IA3 was isolated following fusion of H2-Kb-reactive CD2-Kb3 splenocytes with the TCR-negative BW5147 thymoma. Flow cytometric analysis indicated that IA3 was initially ßTCR⁺, CD3⁺, CD4^-, and CD8⁺ and expressed the H2-Kb transgene, although a significant fraction of cells had already lost CD8 expression. Further staining with Vß-specific Abs indicated that a Vß13 chain was expressed on the cell surface. IA3 subsequently lost both CD8 and ßTCR expression following in vitro passage, and PCR sequence analysis using a panel of TCR class-specific primers indicated that this was due to the absence of a productively rearranged TCR-chain. As the original hybridoma was ßTCR⁺, we concluded that a productively rearranged TCR -chain had been lost from the cell, probably due to chromosome losses that occurred during the return to diploidy from the original tetraploid state of the hybridoma. A nonproductive V10J24 rearrangement was detected as the only non-BW5147 -chain, together with a productive Vß13Jß2.5 rearrangement.

5'-RACE identifies multiple mutations in the V10J24 chain

To isolate all sequences transcribed from rearranged TCR genes and to confirm that no productive -chain rearrangements were present, we performed 5'-RACE simultaneously on both the TCR - and ß-chains (as internal control) using 3' primers specific for both the C and Cß regions of the TCR. RACE products were cloned, and 42 randomly picked clones were analyzed by restriction digestion. This indicated the presence of two groups of clones, which were subsequently identified by sequencing as either 1) nonproductive V10J24, with a premature stop codon at the beginning of the C region immediately following the J region due to an out-of-frame rearrangement, or 2) nonproductive BW5147 fusion partner transcripts. The V10 sequence did not correspond to any currently published V10 sequence. Surprisingly, sequencing revealed that considerable variation was present among 15 randomly selected V10J24 clones. No such sequence variation was observed in the V region in an equal number of productively rearranged Vß13Jß2.5 control clones. In particular, a base substitution frequency 16 times greater was observed in the V clones compared with the Vß clones, and mutations were observed in both the V and J regions.

Mutations are present in the rearranged -chain but not in the ß-chain

As the 5'-RACE protocol includes a reverse transcriptase step that is particularly susceptible to error, we amplified all rearranged TCR- and ß-chains directly from IA3 genomic DNA using Pfu polymerase and primers situated upstream of and including the ATG start codon (forward) and within the VJ region (reverse). In 23 randomly selected clones, we observed an approximately 40-fold increase in base substitution in the V region of the V10J24 rearrangement compared with the V region of the Vß13Jß2.5 rearrangement (Fig. 1), with a frequency of 5.4 x 10^-3 per base pair in the -chain (40 mutations in 7452 bp), while an equal number of ß-chain clones gave a frequency of 1.4 x 10^-4 per base pair. The -chain frequency was also significantly above that observed in exons 2–3 (including the intervening intron) of the -actin gene (8.3 x 10^-5 per base pair in 12 kb of sequence), which was amplified as a control (not shown) and thus likely to reflect the background misincorporation rate due to polymerase error. As the -chain mutations present in more than one clone may be due to a single mutational event, we recalculated the mutation frequency, calling each mutation present in more than one clone as only a single mutational event. The mutation frequency (2.5 x 10^-3 per base pair) was still 15-fold greater than that observed in the Vß region and 25-fold greater than that observed in the -actin gene.

View larger version (21K): [in this window] [in a new window]   FIGURE 1. Sequences of 23 randomly selected clones from the IA3 TCR V-chain genomic PCR amplification. PCR amplification products of the IA3 TCR V-J region were cloned into pGEM-T vectors, and clones were selected at random for sequencing. Regions of sequence difference between the V10J24 sequence and sequences from the clones are indicated by the appropriate letter. Identical sequences are indicated by a dot, while those mutations that could have arisen by gene conversion are indicated by an asterisk. Clones are numbered down the left column. The position of each mutated nucleotide relative to the ATG start codon (with +1 being the A from ATG) is indicated. A summary of mutations found in the intronic region of the V gene is found in Table I.

The -chain mutations are thus unlikely to be due to PCR artifacts. This conclusion is supported by the observation that the mutant sequences are genealogically related such that it is possible to trace the ancestry of all mutations in every clone (Fig. 2). This is consistent with their derivation from a single common ancestral sequence and not with errors introduced at random during PCR amplification nor with amplification of multiple genomic V loci (see below). The small number of ß-chain mutations were each detected once only and could not be genealogically structured. Interestingly, one mutation (TTTGTT) appears to have occurred independently on two occasions.

View larger version (21K): [in this window] [in a new window]   FIGURE 2. Family tree of the IA3 mutant TCR V sequences. The sequences presented here have been arranged to show the likely stepwise derivation of individual clones from the putative ancestral V10J24 rearrangement. Numbers of individual clones, corresponding to the numbering in Fig. 1, are indicated in parentheses at the appropriate part of the tree. Thus, clone 20 is derived from clones 18, 21, and 22. Asterisks indicate the same mutation occurring independently.

We noted that the location of several of the base substitutions coincided exactly with the sites of single- or double-base allelic sequence differences, embedded in overall regions of identity, among members of the V10 and V11 families. This suggested that sequence interchange between the V10J24 rearrangement and related V genes has given rise to some of the mutated sequences described here. For example, the linked double-base substitution within the CCTGCA sequence at position +84 to +87 appears to have been caused by recombination with the CCCGCG sequence of V10.2, V10.4, and V10.8 (23). Similarly, AGG at position +9 has been replaced with AGC present in several V10s, and TCT at position +213 has been replaced by TCA present in several V11 members. The latter occurs in a run of sequence identity between recombination partners of only 7 bp (i.e., V10, CTTCTGG; V11, CTTCAGG). The CAG sequence (position +225) has been changed to CAT present in V10.2. GAA at position +261 of all V10s has been replaced with GAG present in some V11s. Finally, ACA at position +217 has been replaced by GCA of V10.5. Overall, there is a preference for transitions over transversions (2:1).

An alternative explanation for these sequence transfers from other V10 and V11 members is the formation of PCR hybrids. This occurs during amplification of multigene families when less than full-length products of one PCR cycle are elongated on a different template in subsequent cycles, giving the appearance of gene conversion. However, none of the putative gene conversion products described above can be explained on the basis of elongation of incompletely synthesized products. Homology with donor sequences is confined to a small region around the area of sequence transfer (as little as 7 bp) and does not extend further along the template. This is more consistent with a crossover event than with elongation of less than full-length products. Secondly, to obtain an estimate of the likely error rate due to hybrid PCR products when amplifying TCR -chains, we amplified a nonproductive V3J21 rearrangement from a cloned T cell line. Of the 20 sequences analyzed, none showed any evidence of being PCR hybrids or having undergone any form of recombination. Furthermore, the mutation frequency was similar to that observed in both the Vß- and -actin genes of IA3. Finally, approximately half of the mutations in the IA3 V-chain do not have an obvious donor sequence and appear to have originated by another mechanism.

Mutations are also present in the V intron and C region but at lower frequency

We also noted the presence of mutations in the V intron and the first exon of the C region. However, the frequency and type of mutation were somewhat different from those observed in the V exons. As stated above, the V exons exhibited both transitions and transversions with an overall substitution frequency of 5.4 x 10^-3/bp or 2.5 x 10^-3/bp depending on the method of calculation, whereas the V intron and the first exon of the C region exhibited a lower mutation frequency that was very similar in each of these regions. The 181-bp intron contained six mutations in 4163 bp or 1.5 x 10^-3/bp (Table I). However, four of these mutations (at the splice donor site) are likely due to a single event. Taking this into account, we calculate a frequency of 8.3 x 10^-4/bp. This is 5-fold greater than observed in the Vß13 region and 8-fold greater than observed in the -actin gene. We also amplified the first exon of the C gene, where we observed six mutations in 4500 bp or 1.3 x 10^-3/bp, a frequency similar to that observed in the V intron. Furthermore, all mutations in the V intron and the C exon (12 of 12) were transitions.

View this table: [in this window] [in a new window]   Table I. Mutations in TCR V intron and C region

To conclusively demonstrate that mutation was indeed occurring in the V-chain of IA3, we subloned the hybridoma. If the mutations shown in Figs. 2 and 3 are truly present in IA3 cells, it should be possible, by subcloning the hybridoma, to isolate sublines in which every cell in the subclone contains one or more of the mutations shown in Figs. 2 and 3. Alternatively, if the mutations are merely in vitro PCR artifacts, then all subclones should have the same sequence, with only the occasional rare variant due to PCR error present in each subclone. IA3 was subcloned into 96-well microtiter dishes, and 50 randomly chosen clones were selected for analysis. Of the 50 subclones, 1 contained the double CCTCCC and GCAGCG mutation at nucleotides +84 and +87, the AGGAGC mutation at +9, and the GACGAA mutation at +303 typical of clones 1 and 2 (Fig. 1). Four of the 50 subclones contained both the AGGAGC at +9 and the TTTGTT at +163 typical of clones 12, 14, and 15. Ten of the 50 clones contained the TTGTTC at +278 typical of clones 4, 5, 8, 9, and 10, while 2 of the 50 contained the CTTCCT change at +32 typical of clones 18, 21, and 22. Four of the subclones contained the parental sequence, while a further 29 clones contained mutations not shown in Fig. 1. We therefore conclude that the TCR-chain of IA3 was accumulating mutations at rates substantially higher than normal. One possible explanation for some of the -chain sequence variants observed might be the rearrangement of multiple members of the V germline family with the same J24 segment to produce a series of similar sequences that appeared to be due to mutation. Although we considered this possibility unlikely due to the identical VJ joint sequences in all clones, we nevertheless amplified all the germline V10 family members from the CBA strain of mice from which IA3 was derived. None of the germline sequences could account for the variants presented in Fig. 1. Thus, the significantly elevated mutation rate of the TCR region compared with the Vß- and -actin regions, the genealogical relatedness of the mutated sequences, and the identification of the likely origin of some of these mutations led us to conclude that mutations were introduced into the nonproductively rearranged V10J24 chain during growth of the IA3 hybridoma.

View larger version (66K): [in this window] [in a new window]   FIGURE 3. Amplification of rearranged V-J-C region of TCR - and ß-chains from IA3 genomic DNA. IA3 or CBA genomic DNA was used as a template for the amplification of - and ß-chains of the TCR. For the amplification of -chains, primers were located in the 5'-untranslated region and the first exon of the C region, corresponding to primers A1 and CAR1, respectively (see Materials and Methods). For ß-chain amplification, primers were likewise situated in the 5'-untranslated region and within the Cß region corresponding to primers B1 and CBR1, respectively. Lanes 1 and 2, -chain V-J-C region amplification using IA3 DNA. Lanes 3–6, -chain V-J-C region amplification using CBA mouse DNA. Lanes 7 and 8, ß-chain V-DJ-C region amplification using IA3 DNA. Lanes 9 and 10, negative controls (no DNA). Lane 11, blank; Lane 12, m.w. standards.

Both the frequency and type (transitions only) of mutations observed in the V intron and C exon are very similar to those that are produced by reverse transcriptase error during the copying of RNA to DNA, as the enzyme contains no proofreading function. If TCR mRNA was being backcopied to cDNA in IA3 as part of the mutagenic process, we reasoned that it should be possible to identify cDNA copies of TCR -chain mRNA. We therefore attempted to amplify from IA3 DNA the VJC region of the TCR -chain using forward and reverse primers in the 5'-untranslated region and C region, respectively. Amplification would be possible only if fully spliced VJC mRNA had been reverse transcribed into cDNA, as under the PCR conditions used, amplification of the original rearranged V10J24-C region would not be possible due to the large J-C intron in rearranged -chains. This is spliced out only during mRNA maturation. Therefore, only spliced mRNAs that have been reverse transcribed into cDNA should be amplified. Using the primers situated in the V10 5'-untranslated region and within the first exon of the C region, we amplified TCR-chains from IA3 DNA. As controls, we also attempted to amplify the productively rearranged but nonmutated Vß13 chain from IA3 cells using primers situated in the V and C regions; to guard against the possibility that any amplified V10J24 sequences may have already been present in the CBA mouse genome, we also used CBA genomic DNA in place of IA3 DNA but with the same -chain primers. As seen in Fig. 3, a TCR-chain product of the correct size was amplified from IA3 DNA only. No TCRß-chain could be amplified, and the -chain sequences were not present in CBA genomic DNA. Furthermore, no primers from other V families gave PCR products when IA3 DNA was used as a template, indicating that the phenomenon is confined to the nonproductively rearranged V10. Cloning and sequencing indicated that the amplified product was the nonproductive V10J24 chain and lacked both V region intron and the large J-C intron. The presence within the genome of -chain sequences that lack introns provides circumstantial evidence for the synthesis of TCR cDNA by reverse transcription following splicing of mRNA.

An IA3 variant that does not hypermutate but produces a productive V-chain through the use of cryptic splice sites predominates following prolonged growth in vitro

Following 3 mo in continuous culture, we reamplified the V-J region of the IA3 V-chain and were surprised to see that only a single V sequence was present in the IA3 population. This corresponded to the sequence of the original, unmutated V10 rearrangement present in the parental IA3 line. Furthermore, there was no evidence of ongoing hypermutation of the -chain nor of reverse transcriptase activity, as we had observed previously. Instead, we noted the presence of multiple alternative transcripts when the VJC region of the V-chain was amplified from mRNA using RT-PCR (Fig. 4).

View larger version (45K): [in this window] [in a new window]   FIGURE 4. RT-PCR of IA3 TCR -chain mRNA following prolonged in vitro growth of the hybridoma. mRNA was isolated from IA3 cells and amplified by RT-PCR using primers located in the 5' upstream region (A1) and in the C region (CAR3). Lanes 1 and 2, with reverse transcriptase; Lanes 3 and 4, without reverse transcriptase. Lane 5, DNA size standards. FS, full-size transcript. 7bp, 100bp, and 107bp indicate the positions of transcripts that are deleted by 7, 100, and 107 bp, respectively. The band immediately below the full-size band is a result of nonspecific amplification and is not TCR related.

View larger version (37K): [in this window] [in a new window]   FIGURE 5. Location of cryptic splice sites within V10 exons. The intronic sequence is shown in lowercase letters with splice donor and acceptor sites underlined. Cryptic sites within exons are boxed. The 107-bp deletion observed in IA3 mRNA, which reopens the V10 reading frame, is shown by a broken line.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The occurrence of mutations within both the C region and the V intron leads us to believe that this process, at least as it occurs in IA3, is not related to modifying the interaction of the TCR with its ligand such as occurs in hypermutation of Ig genes in B cells. That is, the mutation process is not related to Ag recognition per se. While hypermutation of B cells results in the production of higher affinity Abs, such a result in autoreactive T cells would be undesirable. Recently, it has been reported that TCR Vß-chains undergo mutation in HIV-1-positive patients (21). Together with the present report, this provides support for the original description of this phenomenon and indicates that hypermutation of the TCR can indeed occur, given the necessary conditions. The detection of SH coincident with the loss of a productively rearranged TCR -chain and the cessation of SH coincident with the re-emergence of an open reading frame in the -chain suggests that these events may be connected. We hypothesize that the loss of a productively rearranged -chain triggered the process of mutation in the nonproductive rearrangement in an attempt at re-expressing a functional -chain and TCR. Loss of a functional TCR from a mature T cell would be expected to lead to apoptosis and cell death, and tumor cells, such as that used to produce the hybridoma, would be expected to try to avert this. Thus, IA3 may have acquired certain characteristics of genetic instability for which tumor cells are well known. The BW 5147 thymoma that was used as the fusion partner in this experiment is commonly used in T cell hybridoma production, and we are not aware of similar situations having been reported. Nevertheless, the results presented here may be relevant to the understanding of tumorigenic mutation mechanisms. A possible common thread that unites the reports of TCR hypermutation could be the need to avert the impending demise of the cell. Continuous HIV-1 stimulation of the immune system, leading to cell death, could be the trigger that activates the mutation process in HIV-1-positive individuals.

The emergence of a nonmutating IA3 variant that expresses an -chain with no premature termination codons but with a 107-bp internal deletion is consistent with the idea that the production of a TCR -chain with no internal stop codons is advantageous for the hybridoma in terms of its in vitro growth. We have not yet established whether this leads to the reappearance of some form of TCRß dimer on the cell surface. It does indicate that the nonmutating variant was at a selective growth advantage in cell culture. Alternative splicing of recombined VJC -chains, due to cryptic splice site usage, has been previously reported in normal blood cells and thymocytes (24) and also in lymphocytes with unusual VD- JC- hybrid recombinants. In the latter case, alternative splicing functioned to prevent expression of these apparently aberrant recombination products (25). The splicing machinery of T lymphocytes may thus be sensitive to the reading frame of rearranged - and ß-chains, as has been suggested previously (12, 26).

While hypermutation in IA3 may not be triggered by antigenic stimuli, there may be mechanistic similarities between the process responsible for hypermutation in IA3 and that which occurs in germinal center B cells. V region mutations appear to be due in part to recombination, with similar sequences present within other V10 and 11 members. However, the intronic J and C mutations are of a type and occur at a frequency reminiscent of those introduced by reverse transcriptase. Furthermore, the amplification of VJC -chains lacking introns, in the absence of added reverse transcriptase, suggests that cDNA copies of the -chain already exist in IA3 cells and that reverse transcription and retroposition may be occurring in this hybridoma. Reverse transcription of V10J24 transcripts may have occurred both before and after splicing, following which gene conversion or recombination with other V genes may have taken place. This could explain the presence of mutations within introns, albeit at a lower frequency than in exons. We were unable to amplify ß-chains using a similar protocol, suggesting that only -chains are being reverse transcribed in this hybridoma.

Regardless of the in vivo biological significance of our results, they indicate, at the very least, the existence of a targeting mechanism that can result in increased mutation rates specifically in the TCR -chain. They also point to a possible involvement of reverse transcriptase in the mutation process. The nature of such a targeting mechanism remains conjectural and without precedent. As reverse transcriptase requires a primer to initiate cDNA synthesis, it would be necessary to propose specific priming of -chains and not ß-chains. As mutation is not occurring in cells that express -chains with open reading frames due to alternative splicing, it is possible that the translational machinery of the cell plays a part in signaling. The likeliest source of primer would be the RNA molecule itself, which may be able to use its 3'-untranslated region as a primer for reverse transcriptase when folded back in some form of secondary structure. If confirmed, this would represent a novel means of mutagenesis. It has previously been suggested that reverse transcriptase could be involved in hypermutation of Ig genes in the form of a DNARNADNA loop (27). Although so far lacking clear experimental support, such a general mutation mechanism remains an attractive possibility, as both recombination and reverse transcriptase-mediated mechanisms are distinguished by the fact that they can occur in the absence of cell division or DNA synthesis. This provides a means whereby a cell that is inhibited or prevented from dividing or is threatened with imminent demise may introduce large numbers of mutations into one or more genes, perhaps providing a mammalian equivalent of the stationary-phase mutation observed in prokaryotes and some eukaryotes (28, 29). Gene conversion over short runs of sequence similarity has been shown to be the mechanism responsible for generating Ab diversity in both rabbits (17) and chickens (16, 30, 31), and evidence points to a similar situation in certain instances in the mouse (32, 33, 34), although it is not known whether reverse transcriptase plays a part in this process. It has recently been suggested that reverse transcriptase may prime within the J-C intron of Ig genes and transcribe cDNA through the J and V regions of Ig genes up to the transcriptional start site at the 5' end of the Ig pre-mRNA, thus confining hypermutation to the VJ region and intron and ensuring that the C region is left unmutated (35). The single report of TCR hypermutation likewise indicated that the C region was devoid of mutations. We speculate that hypermutation may be able to be targeted to all or part of these genes depending on whether the purpose is affinity maturation or production of an open reading frame. The use of hybridomas may allow the further investigation of this unusual phenomenon.

	Acknowledgments

 
We thank Carolyn Leithner for sequencing and Drs. Moshe Sadofsky and Leszek Ignatowicz for critical reading of the manuscript.

	Footnotes

 
¹ This work was supported by grants from the Medical College of Georgia Research Institute (to A.M.) and from the Medical College of Georgia Department of Medicine.

² Address correspondence and reprint requests to Dr. Brendan Marshall, Institute of Molecular Medicine and Genetics, Medical College of Georgia, 1120 15th Street, Augusta, GA 30912-3175. E-mail address:

³ Abbreviations used in this paper: SH, somatic hypermutation; IMDM, Iscove’s modified Dulbecco’s medium; RACE, rapid amplification of cDNA ends.

Received for publication June 22, 1998. Accepted for publication September 25, 1998.

	References

Top Abstract Introduction Materials and Methods Results Discussion References

 

1. Connor, A., E. Wiersma, M. J. Shulman. 1994. On the linkage between RNA processing and RNA translatability. J. Biol. Chem. 269:25178.[Abstract/Free Full Text]
2. Carter, M. S., S. Li, M. F. Wilkinson. 1996. A splicing-dependent regulatory mechanism that detects translation signals. EMBO J. 15:5965.[Medline]
3. Aoufouchi, S., J. Yelamos, C. Milstein. 1996. Nonsense mutations inhibit RNA splicing in a cell-free system: recognition of mutant codon is independent of protein synthesis. Cell 85:415.[Medline]
4. Zinkernagel, R., G. N. Callahan, A. Althage, S. Cooper, P. A. Klein, J. Klein. 1978. On the thymus in the differentiation of H-2 self recognition by T cells: evidence for dual recognition?. J. Exp. Med. 147:882.[Abstract/Free Full Text]
5. Marrack, P., D. Lo, R. Brinster, R. Palmiter, L. Burkly, R. H. Flavell, J. Kappler. 1988. The effects of thymus environment on T cell development and tolerance. Cell 53:627.[Medline]
6. Chien, Y., N. R. J. Gascoigne, J. Kavaler, N. E. Lee, M. M. Davis. 1984. Somatic recombination in a murine T-cell receptor gene. Nature 309:322.[Medline]
7. Behlke, M. A., D. G. Spinella, H. S. Chou, W. Sha, D. L. Hartl, D. Y. Loh. 1985. T cell receptor ß-chain expression: dependence on relatively few variable region genes. Science 229:566.[Abstract/Free Full Text]
8. Yoshikai, Y., N. Kimura, B. Toyonaga, T. W. Mak. 1986. Sequences and repertoire of human T cell receptor chain variable region genes in mature T lymphocytes. J. Exp. Med. 164:90.[Abstract/Free Full Text]
9. Concannon, P., L. A. Pickering, P. Kung, L. Hood. 1986. Diversity and structure of human T-cell receptor ß chain variable region genes. Proc. Natl. Acad. Sci. USA 83:6598.[Abstract/Free Full Text]
10. Wu, H., E. Pelkonen, S. Knuutila, M. Kaartinen. 1995. A human follicular lymphoma B cell line hypermutates its functional immunoglobulin genes in vitro. Eur. J. Immunol. 25:3263.[Medline]
11. Green, N. S., J. L. Rabinowitz, M. Zhu, B. J. Kobrin, M. D. Scharff. 1995. Immunoglobulin variable region hypermutation in hybrids derived from a pre-B and a myeloma cell line. Proc. Natl. Acad. Sci. USA 92:6304.[Abstract/Free Full Text]
12. Zhu, M., J. L. Rabinowitz, N. S. Green, B. J. Kobrin, M. D. Scharff. 1995. A well differentiated B-cell line is permissive for somatic mutation of a transfected immunoglobulin heavy-chain gene. Proc. Natl. Acad. Sci. USA 92:2810.[Abstract/Free Full Text]
13. Denepoux, S., D. Razanajaona, D. Blanchard, G. Meffre, J. D. Capra, J. Banchereau, S. Lebecque. 1997. Induction of somatic mutation in a human B cell line in vitro. Immunity 6:35.[Medline]
14. Kallberg, E., S. Jainandunsing, D. Gray, T. Leanderson. 1996. Somatic mutation of immunoglobulin V genes in vitro. Science 271:1285.[Abstract]
15. Selsing, E., B. Xu, D. Sigurdarddottir. 1996. Gene conversion and homologous recombination in murine B cells. Semin. Immunol. 8:151.[Medline]
16. Reynaud, C. A., V. Anquez, H. Grimal, J. C. Weill. 1987. A hyperconversion mechanism generates the chicken light chain preimmune repertoire. Cell 48:379.[Medline]
17. Knight, K. L., R. S. Becker. 1990. Molecular basis of the allelic inheritance of rabbit immunoglobulin VH allotypes: implications for the generation of antibody diversity. Cell 60:963.[Medline]
18. Peters, A., U. Storb. 1996. Somatic hypermutation of immunoglobulin genes is linked to transcription initiation. Immunity 4:57.[Medline]
19. Storb, U., A. Peters, E. Klotz, B. Rogerson, J. Hackett. 1996. The mechanism of somatic hypermutation studied with transgenic and transfected target genes. Semin. Immunol. 8:131.[Medline]
20. Zheng, B., W. Xue, G. Kelsoe. 1994. Locus-specific somatic hypermutation in germinal centre T cells. Nature 372:556.[Medline]
21. Cheynier, R., S. Henrichwark, S. Wain-Hobson. 1998. Somatic hypermutation of the T cell receptor Vß gene in microdissected splenic white pulps from HIV-1-positive patients. Eur. J. Immunol. 28:1604.[Medline]
22. Sponaas, A. M., P. D. Tomlinson, R. Schulz, J. Antoniou, P. Ize-Iyamu, A. M. Schmitt-Verhulst, A. L. Mellor. 1994. Tolerance induction by elimination of self-reactive thymocytes. Int. Immunol. 6:1593.[Abstract/Free Full Text]
23. Arden, B., S. P. Clark, D. Kabelitz, T. W. Mak. 1995. Mouse T-cell receptor variable gene families. Immunogenetics 42:501.[Medline]
24. Mahotka, C., T. E. Hansen-Hagge, C. R. Bartram. 1995. Alternative splicing of T cell receptor chain transcripts containing V 1 or V14 elements. Leukemia 9:1700.[Medline]
25. Hansen-Hagge, T. E., C. Mahotka, R. E. Panzer-Grumayer, H. J. Reuter, K. Schwarz, J. J. van Dongen, C. R. Bartram. 1995. Alternative splicing restricts translation of rearrangements at the T cell receptor / locus. Blood 85:1888.[Abstract/Free Full Text]
26. Li, S., D. Leonard, M. F. Wilkinson. 1997. T cell receptor (TCR) minigene expression regulated by nonsense codons: a nuclear associated mechanism. J. Exp. Med. 185:985.[Abstract/Free Full Text]
27. Steele, E. J., J. W. Pollard. 1987. Hypothesis: somatic hypermutation by gene conversion via the error prone DNARNADNA information loop. Mol. Immunol. 24:667.[Medline]
28. Cairns, J., J. Overbaugh, S. Miller. 1988. On the origin of mutants. Nature 335:142.[Medline]
29. Hall, B. G.. 1992. Selection induced mutations occur in yeast. Proc. Natl. Acad. Sci. USA 89:4300.[Abstract/Free Full Text]
30. Reynaud, C. A., A. Dahan, V. Anquez, J. C. Weill. 1989. Somatic hyperconversion diversifies the single VH gene of the chicken with a high incidence in the D region. Cell 59:171.[Medline]
31. Thompson, C. B., P. E. Neiman. 1987. Somatic diversification of the chicken immunoglobulin light chain gene is limited to the rearranged variable gene segment. Cell 48:369.[Medline]
32. David, V., N. L. Folk, N. Maizels. 1992. Germ line variable regions that match hypermutated sequences in genes encoding murine anti-hapten antibodies. Genetics 132:799.[Abstract]
33. Giusti, A. M., T. Manser. 1994. Somatic generation of hybrid antibody H chain genes in transgenic mice via interchromosomal gene conversion. J. Exp. Med. 179:235.[Abstract/Free Full Text]
34. Xu, B., E. Selsing. 1994. Analysis of sequence transfers resembling gene conversion in a mouse antibody transgene. Science 265:1590.[Abstract/Free Full Text]
35. Blanden, R. V., H. S. Rothenfluh, P. Zylstra, G. F. Weiller, E. J. Steele. 1998. The signature of hypermutation appears to be written into the germline IgV segment repertoire. Immunol. Rev. 162:117.[Medline]

This article has been cited by other articles:

				H. Chen, S. Kshirsagar, I. Jensen, K. Lau, R. Covarrubias, S. F. Schluter, and J. J. Marchalonis Characterization of arrangement and expression of the T cell receptor {gamma} locus in the sandbar shark PNAS, May 26, 2009; 106(21): 8591 - 8596. [Abstract] [Full Text] [PDF]

				A. Roers, M. L. Hansmann, K. Rajewsky, and R. Kuppers Single-Cell PCR Analysis of T Helper Cells in Human Lymph Node Germinal Centers Am. J. Pathol., March 1, 2000; 156(3): 1067 - 1071. [Abstract] [Full Text] [PDF]

				M. Muschen, K. Rajewsky, A. Brauninger, A. S. Baur, J. J. Oudejans, A. Roers, M.-L. Hansmann, and R. Kuppers Rare Occurrence of Classical Hodgkin's Disease as a T Cell Lymphoma J. Exp. Med., January 17, 2000; 191(2): 387 - 394. [Abstract] [Full Text] [PDF]

				S. Winandy, L. Wu, J.-H. Wang, and K. Georgopoulos Pre-T Cell Receptor (TCR) and TCR-controlled Checkpoints in T Cell Differentiation Are Set by Ikaros J. Exp. Med., October 18, 1999; 190(8): 1039 - 1048. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Marshall, B. Articles by Mellor, A. Search for Related Content PubMed PubMed Citation Articles by Marshall, B. Articles by Mellor, A.