Several pathogenic microorganisms, ranging from viruses to protozoa, evade the mammalian immune response by periodically changing the molecular constituents on their surface, a phenomenon called antigenic variation(1). Two pathogens particularly adept at using this strategy are Borrelia hermsii, a prokaryotic spirochete that causes relapsing fever in the western United States and Canada(2), and African trypanosomes, eukaryotic protozoan parasites that cause sleeping sickness in Africa(3). B. hermsii is transmitted by ticks and African trypanosomes by tsetse flies. Both of these organisms circulate extracellularly in the bloodstreams of their mammalian hosts and keep one step ahead of their hosts' immune systems by periodically switching the major protein on their surface. Recent studies demonstrate that the molecular mechanisms responsible for antigenic variation in these two pathogens share many features. The first two sections of this review compare these features, and the third section illustrates that their mechanisms of immune evasion resemble in some ways the mechanisms utilized by the immune system to attack them.
Antigenic Variation in B. hermsii
Spirochetes constitute a separate phylum of eubacteria distinguished by their helical shape and the presence of multiple flagella that lie in the periplasmic region between the inner and outer membranes, rather than beyond the outer membrane(4, 5). The most extensively characterized spirochetes are B. hermsii and Borrelia burgdorferi, the agent responsible for Lyme disease. These species of Borrelia have a linear DNA molecule of about 1 million bp, 1as well as circular plasmid-like DNAs and several small linear DNA molecules of 10-200 kb that are called linear plasmids or minichromosomes(4). Furthermore, these organisms are polyploid with each cell containing between 10 and 20 copies of the large chromosome and each of the linear plasmids(6). The extreme ends, or telomeres, of both the large linear chromosomal DNA molecule and the linear plasmids are covalently closed hairpins(7, 8). The linear plasmids contain genes encoding the outer membrane lipoproteins, called the variable major protein (Vmp) in the case of B. hermsii and the outer surface protein (Osp) for B. burgdorferi(9, 10, 11). In B. burgdorferi, circular plasmids also can encode Osps, and as many as four different osp genes can be expressed simultaneously. In B. hermsii only one vmp at a time has been found to be active(12). Since antigenic variation is much better understood in B. hermsii than in B. burgdorferi, the vmps of B. hermsii will be the focus of this review.
At least 40 antigenically distinct serotypes can arise from a single cell of B. hermsii, each of which appears to be due to the expression of a different surface Vmp(12). New serotypes occur spontaneously within the bacterial population at a frequency of 10Graphic-10Graphic per generation time of about 6 h (13). The different Vmps whose genes have been sequenced fall into two size categories, one group of Vmps that have 360-370 amino acids and another group containing about 210 amino acids (14). No functional differences between these two groups have been reported. As a group, the genes for the smaller Vmps display 70-80% identity, with the most similarity occurring in the N and C termini coding regions and the least similarity within the middle region(15). Collectively, the small Vmp genes and the large Vmp genes have 40-50% identity over the length that they can be compared. Presumably, the unique antigenicities of the different Vmps are conferred by the segments exposed on the outer surface of the bacterium that differs in sequence and structure.
The vmps studied to date are located on linear plasmids of 28-32 kb(12, 16). In a given bacterium all vmps are transcriptionally silent except for one, which can be activated by at least two different mechanisms. In one activation mechanism, typified by the switch from B. hermsii serotype 7 to serotype 21, the silent vmp21 on one linear plasmid is duplicated, and the duplicated copy is transposed downstream of a promoter at a telomere-linked expression site on another linear plasmid(12, 16). The vmp7 originally in this expression site is displaced by the incoming vmp21 and disappears from the genome. The genetic information of vmp7 is not eliminated from the genome by the displacement, however, because it was itself duplicated from a silent vmp7 donor gene located on another plasmid. The boundaries of the duplicated regions usually are a short common sequence just upstream of the start codons of silent and expressed genes and a 200-bp conserved sequence located downstream of their stop codons(16, 17). This duplicative translocation is equivalent to a gene conversion event involving the unidirectional, nonreciprocal transfer of nucleotide sequences to a new site from a donor gene that remains unchanged after the transfer(18).
A second activation mechanism is associated with a frequently detected switch from serotype 7 to serotype 26(15). In this case, a pseudogene version of vmp26, called Graphicvmp26, is located immediately downstream of vmp7, both in the telomere-linked expression site of one linear plasmid (Fig. 1A), and in another linear plasmid containing the correspondingly linked donor genes. The silent version of Graphicvmp26 lacks the first two codons of the expressed vmp26 but has at this position the same 20 nucleotides that follow the first two codons of vmp7. In the switch from vmp7 to vmp26 the 20 nucleotides common to the two adjacent genes serve as sites for a homologous recombination event within the expression site that deletes the expressed vmp7 as a transient circular DNA segment and positions the first two codons of vmp7 in front of Graphicvmp26, creating a functional vmp26 in the same expression site. Thus, an intramolecular deletion juxtaposes coding regions derived from two genes, generating a functional composite gene, an event reminiscent of DNA rearrangements leading to functional antibody genes in the vertebrate immune system (see below).
Diagrams of the expression sites for the vmp genes of B. hermsii (A) and the vsg genes of African trypanosomes (B). In both panels rectangles represent genes, blackcircles are telomeres, redflags are promoters, and redhorizontalarrows are RNAs. A, in this example, taken from (19), the expression of B. hermsii vmp7 is initiated by a joint duplication of vmp7 and the downstream pseudogene Graphicvmp26 from one linear plasmid to the indicated expression site near the telomere of another linear plasmid. RNA synthesis of vmp7 extends from the promoter to a transcription termination site between vmp7 and Graphicvmp26. The subsequent intramolecular deletion of most of vmp7 places its first two codons in front of, and in frame with, the coding sequence of Graphicvmp26 and creates a functional vmp26 in the expression site. With time, templated point changes are donated to vmp26 from a cluster of other pseudo-vmps located upstream of the promoter for the expression site. B, VSG switching in African trypanosomes often is caused by the duplicative transposition of a vsg to a telomere-linked expression site and the concomitant deletion of the previous vsg at that site. The boundaries of the duplicated segment in the expression site are within several hundred 76-bp repeats located upstream and within the conserved C-terminal coding or 3′-untranslated regions located downstream. Sometimes the gene duplication is accompanied by the formation of a mosaic vsg and/or non-templated point changes in the VSG coding region. Transcription of the expression site is initiated at a promoter located 45-60 kb upstream of the vsg and extends through a cluster of expression site-associated genes (ESAG cluster) that are members of as many as seven different gene families. Two ESAG families encode transferrin binding proteins and adenylate cyclases (66, 67, 68). The functions of the other ESAGs are not known. The polycistronic pre-mRNA is processed by 5′ spliced leader addition and 3′ polyadenylation to generate individual mRNAs for the ESAG products and the VSG.
Variants of the Vmp26 amino acid sequence can be generated by post-switch mutations of the expressed vmp26 within its expression site via a mechanism thought to introduce mutations into other expressed vmp genes as well(19). This additional Vmp diversity appears to be caused by partial gene conversions that are templated from Vmp pseudogenes located directly upstream of the expression site and oriented in both directions (Fig. 1A). Sequence determinations of seven different expressed vmp26 gene sequences present in Vmp26 relapse populations following infections of mice with single cells of serotype 7 revealed that specific nucleotides within vmp26 were replaced with nucleotides within 10-20-bp blocks that appeared to be derived from the upstream pseudogenes(19). Furthermore, the experimental evidence suggested that the mutations occurred after vmp26 was placed in the expression site and that they appeared as early as 2-3 days after the switch. It is not known whether these mutations continue to accumulate asymptotically to the point where few if any bacteria contain an expressed vmp26 unchanged in sequence, but the rate of new mutations did appear to decline later in the infection. Remarkably, the mutated vmp26 genes occurred in Vmp26 populations succeeding serotype 7 but not in Vmp26 populations following serotype 17, indicating that the mutation phenomenon might be specific for the intramolecular deletion that accompanied the switch from expression of vmp7 to vmp26. A similar deletion is not associated with the switch from expression of vmp17 to vmp26; rather, that switch is accompanied by a duplicative transposition.
Thus, the two known mechanisms by which B. hermsii can activate a silent vmp are a gene conversion between linear plasmids and a gene deletion within a linear plasmid. Imposed on top of these two mechanisms is a third mechanism by which the diversity of a rearranged vmp can be increased via post-switch mutations templated from pseudo-vmps located on the same linear plasmid as the expression site.
Antigenic Variation in African Trypanosomes
Several African trypanosome species have been shown to evade their hosts' immune responses by undergoing antigenic variation(3), although most of the studies have been conducted on the Trpanosoma brucei subspecies or on Trpanosoma equiperdum. In the bloodstream of their mammalian hosts each trypanosome is coated with about 107 molecules of a glycolipid-anchored glycoprotein called the variant surface glycoprotein or VSG. Antigenic variation of VSGs has several of the features described above for the B. hermsii Vmps but with an apparent increase in complexity. At 3.7 × 107 bp, the African trypanosome haploid genome is 30-40 times larger than that of B. hermsii(20). The diploid organisms contain about 20 chromosomes, whose DNAs range from several hundred to several thousand kb, and 100 or more linear minichromosomes, whose DNAs are 50-150 kb (21). Within the trypanosome genome are as many as 1,000 different vsgs(22), some of which are closely related isogenes but only one of which is usually expressed at a time. The transcriptionally silent vsgs are scattered about all of the chromosomes, including the minichromosomes which appear to be repositories for unexpressed vsgs(23). As many as 20 different potential expression sites for these vsgs may exist, all of which are situated near a telomere(24). These potential expression sites appear to be located on most, if not all, of the chromosomes except the minichromosomes.
Mature VSGs possess about 450 amino acids after the removal of a signal peptide of 20-30 amino acids and a C-terminal hydrophobic tail of about 20 amino acids. The last 50 amino acids of mature VSGs are rich in cysteines and display some sequence similarities that permit VSGs to be classified into two or three groups(3). The C terminus itself is linked to a phosphatidylinositol anchor so these C-terminal 50 amino acids are likely to be in close contact with the membrane and not exposed on the surface. The three-dimensional structures of the N-terminal variable domains (about 380 amino acids) of two antigenically distinct VSGs have been determined by x-ray crystallography and found to be strikingly similar despite very little sequence similarity(25). This finding suggests that all VSGs may possess similar three-dimensional structures, an unexpected result given the variability in VSG sequences but perhaps not too surprising since all VSGs must pack tightly together on the surface. It is not known how many antigenically distinct VSGs potentially can be produced from a single trypanosome, but more than 100 serotypes have been detected in a single infection(26). The estimate of 1000 vsgs in the genome and demonstrations that new genes can be created (see below) suggest the maximum number is much higher.
The frequency at which a new VSG coat arises ranges from 10Graphic to 10Graphic per cell doubling time of 5-6 h in the blood(27). The appearance of new VSGs normally does not follow a preprogrammed order, although some VSGs do tend to appear early in an infection whereas others usually occur later(28, 29, 30, 31). There is no evidence that the immune system of the host induces the VSG switches, but it does contribute to an environment in which new, and temporarily unrecognized, serotypes can prosper, giving rise to a new wave of parasitemia.
The activation of a new vsg and formation of a new serotype often are associated with one of three types of gene rearrangements(3, 32, 33). The best studied rearrangement is the duplicative transposition of a silent, donor vsg from either an interior chromosomal location or a telomeric location to a telomere-linked expression site, displacing the vsg already at that site (34) (Fig. 1B). This gene conversion can be either an inter- or intrachromosomal event. In many cases it is mediated on the 5′ side by homologous recombination between a few copies of a 76-bp repeat upstream of the donor gene and hundreds of copies of this same repeat in the expression site. On the 3′ side of the duplication, homologous recombination often occurs between the segments of sequence similarities in the C-terminal coding regions or in the 3′-untranslated regions(35, 36). A variation on this gene conversion mechanism is a telomere conversion whereby one entire telomeric region including its expressed vsg is replaced with a duplicated copy of another telomeric region and its silent vsg, followed by activation of the duplicated silent gene(37, 38, 39). The third kind of rearrangement is telomere exchange in which two telomeres and their associated vsgs undergo a reciprocal exchange, activating one gene and inactivating the other(40). In addition to these DNA rearrangement events, still other vsgs that are already telomere-linked can be activated in situ without apparent DNA rearrangement via an unknown mechanism. Thus, to be expressed it is necessary, but not sufficient, for a vsg to be located near a telomere. Furthermore, since with very few exceptions (41, 42, 43) one and only one telomere-linked vsg is expressed at a time, other events that are not understood must activate one telomere-linked expression site and silence all of the others.
It has been suggested for both B. hermsii(19) and African trypanosomes (45) that nucleotide modifications might be involved in regulating the expression of the vmp or vsg once it has reached its respective telomere-linked expression site. B. hermsii has a dam methylation system (44) that possibly could methylate silent vmp genes(19), whereas trypanosomes have an unusual glycosylated hydroxymethyluracil at some positions within silent telomere-linked vsgs but not at the corresponding positions within actively transcribed telomere-linked vsgs(45). Proof that these modifications in either B. hermsii or trypanosomes regulate gene expression remains to be provided.
A dramatic outcome of some vsg gene conversions in trypanosomes is the formation of mosaic, or composite, vsgs that are derived from two or more donor vsgs(46, 47, 48). In some cases the newly created vsg is generated during the duplication event via multiple crossovers among related donor pseudogenes containing internal termination codons, resulting in a functional mosaic vsg with an open translation reading frame in the expression site(47, 48). In other cases, the newly duplicated vsg in the expression site contains internal sequences derived from unknown regions of the genome(46). In still other examples, the newly duplicated vsg clearly is derived from a single specific donor gene, but it has point mutations distributed primarily across the middle one-half of the coding region at a frequency of 1-3% of the base pairs(36, 49). In contrast to the templated vmp mutations of B. hermsii described above, these point changes do not appear to be templated from elsewhere in the genome. One trypanosome example has been reported in which the newly duplicated vsg is a mosaic of three closely related donor genes and also has three point changes that are not derived from the donor genes(48).
A general feature of Trypanosomatids is that many of their genes are transcribed into large polycistronic precursor RNAs, which are processed into individual mRNAs by internal cleavages followed by addition of a 39-nucleotide spliced leader to the 5′ ends and polyadenylation at the 3′ ends(32). This unusual property is exemplified by transcription of at least some telomere-linked vsg expression sites (Fig. 1B). The promoters for two specific vsg expression sites have been found to occur 60 and 45 kb, respectively, upstream of the vsg(50, 51, 52). Between the promoter and the vsg in each case is a series of expression site-associated genes (ESAGs) that are co-transcribed with the vsg and that are members of as many as seven different gene families. All vsg expression sites examined carefully contain at least one ESAG. None of the ESAGs encode other VSGs, although pseudogenes or partial genes for VSGs occasionally have been detected in the vicinity of an expression site(53). The steady state levels of the mRNAs for the ESAG products and the VSG differ by as much as 100-fold(54), indicating that expression of these co-transcribed genes is regulated at least in part by post-transcriptional events such as pre-mRNA processing and/or mRNA stability.
Similarities with the Mammalian and Avian Immune Systems
Many reviews have summarized the gene rearrangements associated with the creation of vertebrate immunoglobulins(55, 56). Briefly, in mammals individual genes responsible for the primary antibody repertoire are assembled by intrachromosomal DNA rearrangements that juxtapose a member of the variable region (V) genes adjacent to a member of the diversity (D) and/or joining (J) elements and a representative of the constant region gene family. The diversity of this repertoire is enhanced by the multiplicity of V genes and the D and J elements (combinatorial diversity), and imprecision in the duplex DNA joining process (junctional diversity). Several features of these general DNA rearrangements leading to antibody diversity are similar to those discussed above for B. hermsii and African trypanosomes. For example, the intrachromosomal deletion of sequences that often accompanies the juxtaposition of a V gene and a D or J element to create an open translation reading frame across their junction is reminiscent of the vmp7 and Graphicvmp26 fusion described above for B. hermsii. Likewise, the creation of functional mosaic vsgs in trypanosomes from two or more donor vsgs is similar to the multi-segmental nature of functional antibody genes.
However, the most striking similarity of these three gene systems (two responsible for antigen variability in a pathogen and one for antibody diversity in the host) is the fact that in each case there are built-in mechanisms to introduce point mutations that increase the number of their potential gene products to a virtually infinite number. In the case of immunoglobulin genes, these somatic mutations in the rearranged V gene, coupled with antigen selection, often drive maturation of the immune response to the production of antibodies with an improved affinity for the antigen(56). Nevertheless, as many as 75% of these mutations appear to have no effect on affinity, suggesting that they simply are part of an intrinsic mutational process(57, 58). One way to distinguish between somatic mutations selected for their increased antigen affinity and those that merely are carried along by “hitchhiking” is to examine mutations in associated passenger transgenes that have not been selected for their improved antigen binding(56). Another way is to analyze mutations in the flanking regions of antigen-selected V genes where the nucleotide changes are not expected to be selected for increased antigen binding(59, 60, 61). Both approaches have been used and generated rather similar results (70). In B. hermsii and trypanosomes there is no evidence that the mutations are selected for an improved property of their corresponding surface proteins.
Table 1 compares 71 somatic mutations detected in a passenger transgene within the immunoglobulin gene system (56, 57) and 74 non-templated mutations detected in three separate trypanosome vsg duplications from the same donor MVAT5 vsg gene(36). This comparison shows that in both antibody genes and trypanosome vsgs, the mutational process exhibits a strong bias in the coding strand for transitions over transversions and favors purine transitions over pyrimidine transitions. In addition, in both systems thymines in the coding strand have the least likelihood of undergoing a mutation (only 10% of the mutations for antibody genes and 5% for vsg genes). This strand-specific bias of the mutations almost surely provides a clue about the molecular mechanism responsible for the mutations, but unfortunately despite substantial effort, it is not clear what that mechanism is. Nevertheless, for both systems several models come to mind, including the possibilities that the mutations are (i) associated with the strand-specific process of transcription(62, 63), (ii) initiated by nicks at one end of the mutated region followed by subsequent error-prone 5′ to 3′ DNA repair (36, 56), or (iii) generated by the presence of nucleotide modifications(45).
An indication that the molecular mechanisms of the antibody and trypanosome vsg mutational systems are not completely identical is the distinctive difference in the mutations that occur from cytosine in the coding strand (redbox in Table 1). In the antibody system most detected cytosine mutations are transitions to thymines (80%), whereas only one of the 22 cytosine mutations in the vsg system is a transition to a thymine. The remaining 21 substitutions are transversions to adenines (68%) or guanines (27%). Thus, in the antibody system cytosine to thymine transitions are clearly favored, whereas in the vsg system they are not preferred. Another slightly less dramatic difference in the two systems is the adenine mutations. In the antibody system 6 of the 24 adenine mutations (25%) are transversions to thymine, whereas in the vsg system, none of the 18 mutated adenines convert to a thymine. Although the total number of mutations available for analysis in both systems is not large (71 and 74 for the antibody and vsg systems, respectively), the fact that in the antibody system the mutational proportions shown in Table 1 compare favorably with a compilation of more than 1000 naturally occurring base substitutions detected in rearranged V genes (56) suggests that these differences in the two systems indeed do reflect a difference in at least part of the molecular mechanism responsible for them.
As described above and in contrast to the non-templated base substitutions in the duplicated trypanosome MVAT5 vsg gene, the mutations in vmp26 of B. hermsii appear to be derived from upstream pseudo-vmps(19). Interestingly, an analogy to this situation exists in, not the mammalian immune system, but the immune system of birds(64, 65). In avian B cells a single V gene for the light chain immunoglobulin is activated by an intrachromosomal deletion that positions it beside a J element preceding a constant region gene. Upstream of this V gene lies a group of pseudo-V genes, oriented in either direction, that donate homologous, but non-identical, segments of 10-100 bp as replacements within the activated V gene, thereby diversifying the amino acid sequence of the encoded antibody. Thus, both B. hermsii and the avian immune system increase diversity by small partial gene conversions templated from upstream pseudogenes, whereas African trypanosomes and the mammalian immune system increase the diversity via non-templated mutations.
Previous SectionNext Section
I thank A. G. Barbour, A. D. Klion, and L. V. Kirchhoff for reading the manuscript.Footnotes
This article was first published at jbc The Journal of Biological Chemistry.
The abbreviations used are:
bp base pair(s)
Vmp variable major protein of B. hermsii
vmp Vmp gene
Osp outer surface protein
osp Osp gene
VSG variant surface glycoprotein of African trypanosomes
vsg VSG gene
ESAG expression site-associated gene
V variable region segment of vertebrate immunoglobulin genes.
1.↵Borst P.,Graves D. R.(1987) Science 235, 658–667
2.↵Barbour A. G.,Hayes S. F.(1986) Microbiol. Rev. 50, 381–400
3.↵Donelson J. E.,Rice-Ficht A. C. (1985) Microbiol. Rev. 49, 107–125
4.↵Barbour A. G.(1993) Trends Microbiol. 1, 236–239
5.↵Paster B. J., Dewhirst F. E., Weisburg W. G.,Tordoff L. A., Fraser G. J., Hespell R. B., Stanton T. B., Zablen L., Mandelco L., Woese C. R.(1991) J. Bacteriol. 173, 6101–6109
6.↵Kitten T.,Barbour A. G. (1992) Genetics 132, 311–324
7.↵Hinnebusch, Barbour A. G.(1992) J. Bacteriol. 174, 5251–5257
8.↵Ferdows M. S., Barbour A. G.(1989) Proc. Natl. Acad. Sci. U. S. A. 86, 5969–5973
9.↵Barbour A. G.,Tessier S. L.,Stoenner H. G.(1982) J. Exp. Med. 156, 1312–1324
10.↵Howe T. R.,Laquier F. W., Barbour A. G.(1986) Infect. Immun. 54, 207–212
11.↵Caporale D. A., Kocher T. D. (1994) Mol. Biol. Evol. 11, 51–64
12.↵Plasterk R. H.,Simon M. I., Barbour A. G.(1985) Nature 318, 257–26
13.↵Stoenner H. G., Dodd T., Larsen C. (1982) J. Exp. Med. 156, 1297–1311
14.↵Carter C. J., Bergstrom S., Norris S. J., Barbour A. G.(1994) Infect. Immun. 62, 2792–2799
15.↵Restrepo B. I., Carter C. J., Barbour A. G. (1994) Mol. Microbiol. 13, 287–299
16.↵Kitten T., Barbour A. G.(1990) Proc. Natl. Acad. Sci. U. S. A. 87, 6077–608
17.↵Barbour A. G.,Burman N.,Carter C. J., Kitten T., Bergstrom S.(1991) Mol. Microbiol. 5, 489–49
18.↵Fink G.,Petes T(1984) Nature 310, 728–729
19.↵Restrepo B. I. Barbour A. G. (1994) Cell 78, 867–876
20.↵ Borst P., Van der Ploeg L., Van Hoek J. F. M., Tas J.,James J (1982) Mol. Biochem. Parasitol. 6, 13–2
21.↵Van der Ploeg L. H., Smith C. L., Polvere R. I., Gottesdiener K. M.(1989) Nucleic Acids Res. 17, 3217–322
22.↵Van der Ploeg L. H. T.,Valerio D., de Lange T., Bernards A., Borst P., Grosveld F. G.
(1982) Nucleic Acids Res. 10, 5905–592
23.↵Weiden M., Osheim Y. N., Beyer A. L., Van der Ploeg L. H.(1991) Mol. Cell. Biol. 11, 3823–3834
24.↵Clayton C. E.(1988) Genet. Eng. 7, 1–56
25.↵Blum M. L., Down J. A., Gurnett A. M., Carrington M., Turner M. J., Wiley D. C (1993) Nature 362, 603–609
26.↵Capbern A., Giroud C., Baltz T., Mattern P.. (1977) Exp. Parasitol. 42, 6–13
27.↵Turner C. M., Barry J. D.(1989) Parasitol. Today 99, 67–75
28.↵De Gee A. L. W., Shah S. D.,Doyle J. J.(1981) Exp. Parasitol. 51, 392–39
29.↵Agur Z.,Abiri D.,Van der Ploeg L. H. T.(1989) Proc. Natl. Acad. Sci. U. S. A. 86, 9626–9630
30.↵Longacre S.,Eisen H.(1986) EMBO J. 5, 1057–1063
31.↵Miller E. N., Turner M. J.(1981) Parasitol. Today 82, 63–80
32.↵Hajduk S.,Adler B.,Bertrand K.,Fearon K., Hager K., Hancock K., Harris M., Le Blanc A., Moore R., Pollard V., Priest J., Wood Z (1992) Am. J. Med. Sci. 303, 258–2
33.↵Van der Ploeg L. H. T., Gottesdiener K., Lee M. G. (1992) Trends Genet. 8, 452–457
34.↵Hoeijmakers J. H., Frasch A. C., Bernards A.,Borst P.,Cross G. A (1980) Nature 284, 78–80
35.↵Van der Ploeg L. H. T., Liu A. Y. C.,Michels P. A. M.,de Lange T., Borst P., Majumder H. K.,Weber H., Veeneman G. H.,Van Boom J.(1982) Nucleic Acids Res. 10, 3591–3604
36.↵Lu Y., Alarcon C. M., Hall T., Reddy L. V., Donelson J. E (1994) Mol. Cell. Biol. 14, 3971–3980
37.↵Bernards A., de Lange T., Michels P. A., Liu A. Y., Huisman M. J., Borst P. (1984) Cell 36, 163–170
38.↵ Bernards A., Kooter J. M., Michels P. A. M., Moberts R. M. P., Borst P. (1986) Gene (Amst.) 42, 313–322
39.↵Borst P., Bernards A., Van der Ploeg L. H. T., Michels P. A. M., Liu A. Y. C., de Lange T., Sloof P., Veeneman G. H., Tromp M. C., Van Boom J. H. (1983) Genetic Rearrangement (Chater K. F., Cullis C. A., Hopwood D. A., Johnston A. A. W. B., Woolhouse H. W., eds) pp. 207–233, Croom & Helm
40.↵ Pays E.,Guyaux M., Aerts D.,Van Meirvenne N., Steinert M. (1985) Nature 316, 562–564
41.↵Baltz T., Giroud C. Baltz D., Roth C., Raibaud A., Eisen H.(1986) Nature 319, 602–604
42.↵Cornelissen A. W., Johnson P. J., Kooter J. M., Van der Ploeg L. H., Borst P. (1985) Cell 41, 825–832 C
43.↵ Esser K. M., Schoenbechler M. J. (1985) Science 229, 190–193
44.↵Hughes C. A. N.,Johnson R. C. (1990) J. Bacteriol. 172, 6602–6604
45.↵Gommers-Ampt J. H., Van Leeuwen F., de Beer A. L. J., Vliegenthart J. F. G., Dizdaroglu M., Kowalak J. A., Crain P. F., Borst P. (1993) Cell 75, 1129–1136
CrossRefMedline Search Google Scholar
46.↵Pays E.,Van Assel S., Laurent M., Darville M., Vervoort T., Van Meirvenne N., Steinert M. (1983) Cell 34, 371–381
47.↵Thon G.,Baltz T., Eisen H. (1989) Genes & Dev. 3, 1247–125 CrossRef Search Google Scholar
48.↵Kamper S. M.,Barbet A. F. (1992) Mol. Biochem. Parasitol. 53, 33–44 CrossRefMedline Search Google Scholar
49.↵Lu Y., Hall T., Gay L. S.,Donelson J. E. (1993) Cell 72, 397–406 CrossRefMedline Search Google Scholar
50.↵ Kooter J. M., van der Spek H. J., Wagter R., d'Oliveira C. E., van der Hoeven F., Johnson P. J., Borst P.(1987) Cell 51, 261–272 C
51.↵ Johnson P. J., Kooter J. M., Borst P. (1987) Cell 51, 273–281
52.↵ Pays E., Tebabi P., Pays A., Coquelet H., Revelard P., Salmon D., Steinert M. (1989) Cell 57, 835–845
53.↵Crozatier M.,Van der Ploeg L. H. T., Johnson P. J., Gommers-Ampt J., Borst P.(1990) Mol. Biochem. Parasitol. 42, 1–12
54.↵ Cully D. F., Ip H. S., Cross G. A. M. (1985) Cell 42, 173–182 CrossRefMedline Search Google Scholar
55.↵ Alberts B., Bray D., Lewis J., Raff M., Roberts K., Watson J. D. (1994) Molecular Biology of the Cell, pp. 1221–1227, Garland Publishing, Inc., New York
56.↵ Betz A. G.,Neuberger M. S., Milstein C. (1993) Immunol. Today 14, 405–411
57.↵ Betz A. G., Rada C., Pannell R., Milstein C., Neuberger M. S. (1993) Proc. Natl. Acad. Sci. U. S. A. 90, 2385–2388
58.↵ Berek C., Milstein C. (1987) Immunol. Rev. 96, 23–41
59.↵ Golding G. B., Gearhart P. J., Glickman B. W. (1987) Genetics 115, 169–176
60.↵Both G. W., Taylor L., Pollard J. W., Steele E. J. (1990) Mol. Cell. Biol. 10, 5187–5196
61.↵ Rogerson B., Hackett J. Jr.., Peters A., Haasch D., Storb U. (1991) EMBO J. 10, 4331–4341
62.↵ Lommel L., Hanawalt P. C (1993) Mol. Cell. Biol. 13, 970–976
63.↵ Selby C. P., Sancar A.(1993) Science 260, 53–58
64.↵ McCormack W. T., Thompson C. B. (1990) Genes & Dev. 4, 548–558
65.↵ Reynaud C.-A., Dahan A., Anquez V., Weill J.-C. (1989) Cell 59, 171–183
66.↵Steverding D., Stierhof Y-D., Chaudhri M., Ligtenberg M., Schell D., Beck-Sickinger A. G., Overath P. (1994) Eur. J. Cell Biol. 64, 78–87
67.↵ Chaudhri M., Steverding D., Kittelberger D., Tjia S., Overath P. (1994) Proc. Natl. Acad. Sci. U. S. A. 91, 6443–6447
68.↵ Paindavoine P., Rolin S., Van Assel S., Geuskens M., Jauniaux J-C., Dinsart C., Huet G. Pays E. (1992) Mol. Cell. Biol. 12, 1218–1225