Eur. J. Immunol. 1991. 21: 1821-1827

Walter Pargent, Alfons Meindl, Rainer Thiebe, Sabme Mitzel and Hans G. Zachau Institut fiir Physiologische Chemie der Universitat Munchen, Miinchen

0 regions of the human immunoglobulin x locus

The human immunoglobulin at locus. Characterization of the duplicated 0 regions* Two large regions of the human immunoglobulin x locus, the so-called 0 regions, have been characterized on cosmid and phage clones.The two regions are very similar but not identical duplicates belonging to the C, proximal (p) and the distal (d) copies of the x locus. The Op and Od regions comprise contigs of 90 and 120 kb, respectively, and contain 20 V, genes and pseudogenes which have been sequenced. Three pairs of V, genes were found to be practically identical in the duplicates while allotypic differences, at least for two of the genes, are considerable.The similarities between the duplicate genes may be related to the fact that the two copies of the x locus are arranged in a palindrome-like fashion with the 5’ sides of the 0 regions pointing towards each other (CxJxB Lp Ap Op Od Ad Ld). This may have contributed to equalizing the sequences. Beyond Op and Od no further V, genes were found within about 80 kb. Instead, repetitive DNA sequences have been localized there, the structures of which suggest that they may have been involved in the evolution of the V, gene-containingregions. TheV, pseudogene containing W regions, that had been transposed in evolution from the short to the long arm of chromosome 2 by a pericentric inversion, may have been derived from the Oregions according to structural homologies between defined sections of the 0 and W regions.

1 Introduction The human Ig x locus has been studied by cloning and sequencing for the past 10years. After early work by Bentley and Rabbitts (e.g. [l, 21) we isolated systematically from cosmid and phage A. libraries V, gene-containing clones that were mapped and characterized in detail. From the cosmid and phage clones contigs were assembled which were termed the A, B, L, 0 and W regions. While the structural data on the A, B, L and W regions were published in some detail [3-61 the 0 regions have been described only at an early state of analysis [7]. Large parts of the human Ig x locus occur in the genome in duplicated form. The duplication was first detected for one specific V, gene region [2] and was later proven for the whole A and L regions [3,5]. The B and J,C, regions are present only in one copy per haploid genome [4] and the W regions [6] were shown to be transposed and be no longer part of the x locus. A duplication of the A and L regions was first surmised when for various parts of these regions two sets of very similar but not identical cosmid clones were isolated. It was then shown by hybridization experiments with DNA from a number of unrelated individuals that the differences are not allelic but duplication related. Also for the 0 regions two sets of cosmid clones had been found and

[I 93921

*

1821

This work was supported by Bundesministerium fiir Forschung und Technologie, Center Grant 0316200A, and Fonds der Chemischen Industrie.

Correspondence: Hans G. Zachau, Institut fiir Physiologische Chemie, Schillerstr. 44, D-8000 Munchen 2, FRG Abbreuiatioas: 0, A, L, B regions: V, gene-containing contigs of the human x locus PFGE: Pulsed-field gel electrophoresis 0 VCH Verlagsgesellschaft mbH, D-6940 Weinheim, 1991

it was postulated that they belong to the duplicate copies of the x locus [7] but further analyses were necessary to prove the existence of duplicated 0 regions. The duplication of large parts of the x locus was corroborated by pulsed-field gel electrophoresis (PFGE) experiments [S-lO].TheV, gene-containingcontigs were assigned to the C,-proximal and distal copies of the locus and consequently named Ap, Lp, Op and Ad, Ld, Od. The 0 regions turned out to be located at the 5’ sides of the two copies that lie 800 kb apart in the genome and are arranged in opposite 5 ’ , 3’ polarities ([9, 101; Fig. 1in [ll]). In the present report the main structural characteristics of the Op and Od regions are described, while in the accompanying report [ l l ] polymorphisms of the 0 regions are discussed together with those of the other regions of the x locus.

2 Materials and methods 2.1 Enzymes and reagents Restriction nucleases, Klenow fragment of DNA polymerase,T4 DNA ligase, deoxynucleotidesand dideoxy-nucleotides were from Boehringer Mannheim (Mannheim, FRG) and T7 DNA polymerase (sequenase) from USB, Cleveland, OH. [35S]dATPaS, [a-”P]dATP and [a-32P]dCTP were obtained from New England Nuclear, Boston, MA. The organic and inorganic reagents came from Merck, Darmstadt, FRG.

2.2 Recombinant DNA libraries, hybridization cos 143x0s 237 were isolated from a cosmid library of placenta St DNA [7].The cos 607clones are derived from a cosmid library of cell line GM6U7 DNA [12]. cos 850 and cos 851 stem from an MspI library of placenta N DNA in pHC79 [13]. The phage clones h3 and A.7 are from the 0014-2980/91/0808-1821$3.50+ .25/0

Eur. J. Immunol. 1991.21: 1821-1827

W.Pargent, A. Meindl, R. Thiebe et al.

1822

3 Results

library of Lawn et al. [14,15], h316-h318 from a library of Eco RI-digested placenta N DNA in charon 40 phage (C. Huber, unpublished).V, gene probes were the same as in [3]. Blot hybridizations were performed as described in [3] and also by the alkali and phosphate methods [16, 171.

3.1 Restriction maps of the 0 regions

The clones representing the 0 regions have been isolated by screening cosmid and phage h libraries with V, gene probes and with specificallyconstructed “walking” probes. The current picture of the 0 regions is shown in Fig. 1and some subclones and detailed maps that have been essential for the construction of the contigs are compiled in Fig. 2. What has been previously called the Oa and Ob regions [7] is now described on the basis of the PFGE experiments [8,10] as the distal and proximal regions Od and Op, respectively. The two cosmid clones, cos 143 and cos 146, that had previously been thought to be part of the Od copy (Fig. 6 in [7]) could now, after detailed restriction mapping with the help of new subclones, be assigned to the Op copy. This way the gap between cos 216 and cos 147 was closed and the Op contig extended beyond the point of artifactual ligation in cos 147 [7]. Also the linking between both the proximal and the distal copies of 0 and A which was indicated already in [3] now rests on several cosmid clones.

2.3 Sequencing

The genes 0 1 , 0 2 and 0 4 were sequenced from cos 151[7], 011from both cos 216 and cos 164 [7] as well as fromphage ?3(“V3a”; [15]), 012 from cos 164 and phage h3 (“V3b”; [15]), 014 from cos 143 (Fig. 1). Subclones were prepared in M13 phages [18] or pUC18 [19] and the sequencing was carried out by standard procedures. The repetitive sequences (Fig. 3) were determined in pUC19 and Bluescript SK- clones in one of the strands by the method of [19]. The latter sequences and the sequencing strategy are described in [20]. All sequenceshave been submitted to the EMBL Data Library.The accessionnumbers are as follows: forthegene sequencesX59311-X59318: 0 1 , 0 2 , 0 4 , 0 1 1 , 012, 014, V3A, and V3B, respectively; and for the repetitive sequences X59319-X59326. cOs151

K

Od

Y

Y

011

. O+P% I.

I

I

,I

021

I

-

-

031

albb-2

I

BamHl Bglll Kpnl Smrl Clal Xhol Pvul Sat1 Nrul Hlndlll

I

I

I I I

I

I

Y

!11111I

YI I I l l II

A

I 1 I V I

I I B

I I

09

I

II

II I IIII

II

I I I1 I I I I B

I

I l l

I I

II

I

I

I I

II

I

I

TI I I 1

I

I 1

1

I

A

10 I

~0.146

I

20 I I

I

30

I

-

BarnHl BgLll Kpnl Smal CLal

//

COD237

x7 corlb’l

A310 C0’216cor164 COD850 COD851

I

COD154

corlb3

I

I

I

Sfll

co.217 corn203 tot207 cor214 _IA

Ad

//

Xhol Pvul sat1 Nrul H lndll I Sphl EcoRl SniBl Sfll

I

II

I

13

I

I I

0

1.1

Y

07m

061

:: O q O l O , A11

Y

I

m

.I

//

i -//

1

plS1-b

A

I

Sphl

EcoRl SnaBl

I

I

I

I I I II I I I Ill

cos607/7 o r to814b2

c

--

- -

-

-

-

-//

x3

1

10

1

1

50

1

1

60

I

I

70

I

I

80 I

I

90 I

I

100 I I

110 , I

,

120 I

/

130 kb /

Figure 1. Restriction maps of the duplicated regions Od and Op of the human Ig x locus. Positions of restriction sites that are identical in both regions or occur in sections for which there is no cloned counterpart are indicated by small vertical bars. Sites that are located only in one of the regions are marked by arrowheads pointing to the respective sites; open arrowheads are used for sites localized only on cosmid clones, full arrowheads for sites that had been characterized as duplication-differentiatingpolymorphisms by studying individual DNA [ll]. Hind III, Sph I and Eco RI are mapped only in part of the 0 regions [20]. Some details on the mapping of cos 607/7, cos 607/30, and 113 are shown in Fig. 2. Artificially ligated parts of cos 147 are indicated by a dashed line [7].V, genes are shown as boxes. All subclones hybridize to both copies.The following subclones are indicated: 111144-2 [7]; p151-4, 1.7-kb BamHI in pUC19; p217-13, 1.7-kb StuI in Bluescript SK-; m217-1 [7]; p207-6, 0.8-kb BamHILHindIII in pUC19 (subsubclone of p207-5, Fig. 6); m143-1 [7]; m146-5,6, 0.4- and 0.25-kb Hpa 11-Hind11 fragments in M13mplO; 111146-8, 0.4-kb Pvu 11-StuI in M13mplO.

Eur. J. Immunol. 1991. 21: 1821-1827

0 regions of the human immunoglobulin x locus

1823

Figure 2. Polymorphic sites in the 0 regions.

I I

I I II I

012

p207-6 45 /

/

ml44-2

50

p151-4

55

I

m146-5.6 100

//x1-/+//

I

0

m143-1

[kbi

115

tion sites are included that help to define polymorphisms.The Hind 111site in pos. 45.5 is absent from the d copy of most individuals. The Eco RI site in pos. 52.5 of Op is present only in some individuals, in cos 607/30 and in h3. The Bgl I1 site at pos. 57.2 of o p is absent in some individuals and in cos 607/30.The Bam HI site at pos. 70 of Od is absent in some individuals.The Rsa I site of Od is absent in some individuals and in cos 60717.-

restriction sites as Xho I and the absence of common ones as Bgl I1 and Hind III.We exploited the latter feature to search for V, genes in genomic blots of the still uncloned 5' regions. In PFGE experiments an 80-kb Hind III fragment was detected with the subclone p217-13 (Fig. 3) which extends from the HindIII site at position 36 into the uncloned regions.This 80-kb fragment did not hybridize to V,I or V,II probes indicating that another approx. 45 kb beyond the 5' end of the contigs is free of V, genes [20]. VJII genes would have hybridized with the V,I probe under the relaxed conditions employed. The unique clone p217-13 hybridizes in PFGE experiments to both copies of the x locus indicating that the conclusions about the absence of V, genes may hold for both the Od and Op extension.

The various cosmid and phage clones were assigned to Od or Op mainly with the help of duplication-differentiating polymorphisms [ll]. These polymorphisms are detected in blot hybridizations by certain subclones (Fig. 1 and 2) which are compiled together with other subclonesinTable 1 of the accompanying report [ll].

3.2 Repetitive sequences at the 5' sides of the 0 regions

The 5' sides of the Oregions are characterized by the presence of repetitive sequences which rendered all attempts to extend the locus in this direction unsuccessful up to now. It was also not yet possible to decide whether cos 217, i.e. the clone extending furthest to the 5' side, is part of the Od or Op region since no restriction site difference between the two regions was detected in this part of the locus, that would have allowed an unequivocal 3.3 The V, genes of the 0 regions assignment of the clone. The first such duplication differentiating polymorphism is the Hind 111site at position 45 The genes of the Od region are called 01-010, the ones of Op 011-018 (Fig. 1). The duplicates of 0 9 and 010 had which is present only in the Op copy (Fig. 1). been found first on a cosmid clone extendingthe Ap region In unsuccessful attempts to find a "walking" clone cos 217 and are, therefore, called A15 and A16 [3]. All 20 V, genes was fully subcloned (Fig. 3). All clones from the end to have been sequenced. The gene pairs 01/011, 02/012, about position 33 yielded several bands and/or a smear on 04/014, and 08/018 are potentially functional. The first blot hybridization with human DNA digests.The reason for three pairs are described here while the 08/018 pair is part this became clear when some of the clones were sequenced. of another report (M. Scott et al., in preparation). The Numerous tandem repeats of TGGAA were found in both other six V, gene pairs of the 0 regions represent pseudostrands. These repeats are known also as motifs in human genes and will be dealt with together with other pseudosatellite DNA (e.g. [21]). Also regions hybridizingwith Alu gene sequences from the x locus (K. F. Schable et al., in preparation). It should be mentioned here, however, that and LINE 1probes were detected. all V, genes and pseudogenes are oriented within the The sequence peculiarities at the 5' sides of the 0 regions contigs in the same transcriptional polarity and that all result also in the clustered occurrence of less common genes on Od are extremely similar to their duplicates on

-

-LINE

pJ17-k

p217-7

~217-6

'

I

p207-1'

-2 '

N u I

p217-5 -3

'

H H

I

-4

p217-13 n217-1

~217-8 1

I

I

I

I

&Q

30

20

10

0

I

I

I

ILbl

1

Figure 3. The 5' part of the 0 regions. Subclones are indicated as thin bars, LINE 1and Alu hybridizationsas fat bars.The LINE 1clones pCDlla and b [32] were subcloned in M13mpll; and the BLUR 8 clone [33] was used for detecting Alu elements.The clones were kindly donated by Drs. M. Singer and F! L. Deininger, respectively. Hybridizations of LINE and Alu probes were under relaxed conditions (washing with 100 m~ sodium phosphate). Open and filled triangles indicate clusters of TGGAA and TI'CCA sequences, respectively, located on the same strand.

1824

W. Pargent, A. Meindl, R. Thiebe et al.

Eur. J. Immunol. 1991.21: 1821-1827

allelic differences are considerable in this part of the genome in contrast to the absence of differences between the duplicated genes 01 and 011 of individual St. The details of subcloning and sequencing the 01/011 regions are described in [22].

A similar situation was found for the genes 0 2 and 012 which belong to subgroupI. They were found to be identical in 1009 bp of the coding region and the flanks except for one difference near the promoter region.The 012 gene region from phage h3 (V3b), on the other hand, differed in 18 of 1028 bp from 01 and in 17 of 1009 bp from 012 of the respective cosmid clones. Because of a stop codon in the leader region the V3b gene is a pseudogene. This is the second case in the x locus where a gene in one individual is a pseudogene in another one; the other example is in the L region [5].The cluster of five differences in the CDR3 region is remarkable; since it is unlikely that it arose by five independent somatic mutations other mechanisms including gene conversion-like events have to be considered. The rearranged V, genes of the cell lines Nalm-6 (11.5-kb allele in [23]) and KOBKlOl (Ch2 allele in [24]) were found to be identical to the 012 gene in the coding and the upstream regions except for three differences (605 bp each compared in total); the differences are in positions distinct from the single 02/012 difference. Also the rearranged V, gene of the cell line Walker [25] seems to be derived from 012. The nine differences between the two genes (total comparable sequence 608 bp) may be the result of somatic Figure 4. Restriction maps and sequencing strategies of three V, gene pairs and two allelicV, genes of the 0 regions. The exons of mutations, but allelic differences cannot be excluded. 0 2 is the V, genes are indicated by solid boxes. Subclones used for unlikely to be the parent of the Walker gene for two sequencingare shown below the maps of the respective genes.The reasons: in the position of the one 02/012 difference the horizontal arrows indicate the direction of sequencing; continuous Walker gene has the 012 sequence and the cell line Walker lines mark the sequenced part of the subclones, those not shown in does not contain a reciprocal recombination product or full length are indicated (41).(A) The vertical arrow marks the 5’ signal joint (H. G. Klobeck, personal communication) as end of fragment V3a. The Pst I site exists only in gene V3a, while would be expected if it had been generated by inversion the downstream Sph I site is absent inV3a. (B) The Hind11 site is absent in V3b. (C) A clone sequenced using a synthetic oligonu- from the 0 2 gene which is part of the distal copy [9]. The cleotide complementary to an internal sequence is marked by an rearranged V, gene of the cell line Daudi is probably also derived from the 012 gene. Its sequence differs consideropen circle. ably in the coding region, i.e. by about lo%, from the sequences of 021012 and the Walker gene, but 442 bp of Op. The restriction maps of the gene regions and the the upstream and leader regions and 5 kb of the restriction sequences of the threeV, gene pairs 01/011,02/012 and maps are identical for the gene regions of the cell lines 04/014 are shown in Fig. 4 and 5, respectively. Daudi and Walker [25]. The Daudi and the Walker genes show the 012 sequence at the one position of difference The V, genes 01 and 011 belong to subgroup II.They between 0 2 and 012. Both rearranged V, genes are more turned out to be identical in the coding regions and to differ closely related to the 012 gene from the cosmid clone than by 1bp near the promoter regions. In previous studies to its allele of the h3 phage. duplicatev, genes had been found to be very similar but not identical [3,5]. Therefore, the gene 011 was sequenced The V,I genes 0 4 and 014 were found to be identical in from two different clones (derived from DNA of the same 959 bp of coding and flanking sequences except for 1bp of individual) and the sequenceswere found to be identical to difference downstream of the recombination signals. They each other and in the coding region also to the one of 01. are potentially functional genes if the presence of Arg and Part of the gene 011is also contained in phage h3 that had Gly at the positions of the invariant Glu and Cys (no. 37 and been partially mapped previously [15].The phage clone was 88; [26]) does not render them pseudogenes. derived from the library of [14] and, thus, from the DNA of a different individual than our cosmid clones.The gene 011 from phage U,called V3a, was sequenced from position 55 3.4 Similarities between parts of the 0 and W regions onwards; the upstream part of the gene is not present in the h clone. When the sequence of the 823 bp of the 011 gene It was noticed that subclones from the 01/02 region such as and its 3’ flank that were sequenced on DNA from both p207-6 (Fig. 1) hybridize even under stringent conditionsto sources were compared, 13 nucleotide differences were parts of the triplicated W regions: the W regions had been detected, 3 of them in the coding region. Apparently, the transposed in evolution from the site of the x locus on the

Eur. J. Immunol. 1991. 21: 1821-1827

0 regions of the human immunoglobulin x locus

1825

Figure 5. Sequences of three V, gene pairs and two allelicV, genes of the Oregions. The 011 gene difTers from 01 at pos. 70; 012 differs from 0 2 at pos. 64 (C instead of T in both cases); 0 1 4 has a G in pos.1012 (not shown) as compared to a C in 04. A vertical arrow indicates the 5’ end of the V3a fragment. A deletion in the upstream region of the V3b gene is indicated by points under the 0 2 sequence. DNA sequences underlined are the decanucleotide box 1341 and the V-J joining signal sequences. Invariant amino acids ([26]; in the leader region own sequence comparisons) are underlined in the formal translation products. A line above amino acid symbols indicates that an invariant amino acid has been replaced by another amino acid.

1826

Eur. J. Immunol. 1991.21: 1821-1827

W. Pargent, A. Meindl, R. Thiebe et al.

-n151-3

~207-7

**

Figure 6. Characterization of homologous sections of the W and 0 regions. The maps were drawn as superpositions of the triplicated W regions (I, I[ and III in Fig. 1 of [6]) and the duplicated 0 regions. The hybridization probes a-i were isolated from the respective subclones.They are indicated as bars and their hybridization targets as dotted lines. c, d and e yield signals with hybridization factors 0.3-0.7 [31]at stringent conditions(washing with 40 mM sodiumphosphate),while the probes a, b, f and g hybridized with factors0.1-0.5 at relaxed conditions (washing with 100 mM sodium phosphate) and probes h and i did not yield cross-hybridization signals. Subclones 111165-4 and pWuV-1 are described in [7] and [25], respectively, the others in [20].

short arm of chromosome2 to the long arm of this While the allelic variation in the structure of the x locus will chromosome [6]. Other cross-hybridizations of 0 region be discussed mainly in the accompanying report [ll] the clones, e.g. to the Aregions or the transposed Chr22 allelic differences in the 0 gene sequences should be regions [27] were much weaker and were seen only under mentioned here. These differences vary for different gene relaxed conditions. The hybridization of seven different regions: they amount to about 1.7% for the 011/012 probes from the region between 01/011 and 02/012 to regions as determined in the germ-line DNA of two parts of the W region were determined in a semiquantita- individuals and 0.9% and 2.5% for L region genes [5]. tive way (Fig. 6). It turned out that the similarities as Rearranged V, genes of human cell lines are usually derived detected by hybridization are not restricted to a small from individuals whose germ-line gene sequences are defined sequence as for instance in the case of the so-called unknown. Therefore, as a matter of principle, one cannot “homox” cross-hybridization [28] but extend over much of distinguish between allotypic differences and differences the intergenic regions on the 0 and the W sides. Also the caused by the somatic hypermutation process. The three similarity between the 01/011 and the W8 [29] gene examples of rearranged 012 genes are grossly different with sequences is considerable and the similarity in the intron respect to the extent of divergence: 0.5% for the Nalmd regions is especially remarkable [20]. Since gene regions in and “Ch2” genes [23,24], 1.5% for the Walker and about the same contig are frequently more closely related to each 10% for the Daudi gene [25].It seems safe to assume that other than to distant ones we postulate that the W regions most of the sequence differences in the latter two genes are may have originated in evolution from the 5’ side of an caused by somatic mutations. 0 region precursor. The sequence of events would have involved the duplication of a gene region, a pericentric The repetitive sequences on the 5’ sides of the 0 regions inversion and subsequent amplification steps; for the prevented us so far from extending the contigs in this direction. Attempts to overcome this barrier by cloning in discussion of the latter steps see [6]. yeast artificial chromosomes [30] are in progress in our laboratory. For the time being, however,we assume that the 4 Discussion genes 01 and 011 mark the 5’ ends of V, gene-containing The p and d copies of the 0 regions seem to be more similar regions of the x locus. For one, no V, gene-specific to each other than the duplicated copies of the A and L hybridization was detected within about 80 kb 5’ to regions. This is seen for instance in the sequences of the V, 01/011. Secondly, the results of semiquantitativeV, gene genes.While the three pairs of gene regions of Op and Od hybridizations accordingto which most of theV, genes have described in this report are practically identical, corre- already been cloned from genomic DNA and only very few sponding pairs of the Lp and Ld differ by 0.5%-3.5% [5]. genes are still missing [31].Various mechanistic possibilities An analogous situation occurs with the restriction site including “slipped mispairing” and unequal chromatid differences: while Op and Od differ in 8 of 133 six-cutter exchange can be envisioned to explain how the repetitive sites, 15 of 144 sites differ between Ap and Ad [3]. This sequences,which are in part palindromically arranged, may corresponds to a divergence of 1.0% and 1.7%, respective- have acted in the evolution of the x locus [20]. These ly. Because of the high homology of Op and Od particular mechanismsmay also have contributed to the transposition emphasis had to be laid upon finding restriction sites and of the W regions [6] and perhaps also of the Chr22 regions subclones which allow to differentiate the duplicated [27] which are more or less homologous to the 01/02 copies. Such duplication-differentiatingpolymorphisms are region bordering the stretch of repetitive DNA. The mainly described in the accompanying report [ll]. With structural information on the 5’ adjacent region of 01/011 their help most of the cosmid and phage h clones could be and the transposed regions is not yet sufficient,however, to propose a single convincing evolutionary scheme. unequivocally assigned to either Op or Od.

Eur. J. Immunol. 1991. 21: 1821-1827

The homology between Op and Od extends far into the uncloned region as judged from the map of rare cutter resriction sites constructed from PFGE experiments [9,10]. The distance between the 5’ sides of Op and Od is estimated to be about 800 kb. If the homology between the two halves of this region is confirmed in further structural studies the x locus may be considered a megabase-sized palindrome. Already the present information on the high homology between the cloned copies of the 0, A and L regions argues in favor of a palindromic structure which may promote fairly frequent recombination events taking place between the two halves. Received March 20, 1991.

5 References 1 Bentley, D. L. and Rat-itts, T. , Nature 1980. 288: 730. 2 Bentley, D. L. and Rabbitts, T. H., Cell 1983. 32: 181. 3 Straubinger, B., Huber, E., Lorenz, W., Osterholzer, E., Pargent, W., Pech, M., Pohlenz, H.-D., Zimmer, F.-J. and Zachau, H. G., J. Mol. Bwl. 1988. 199: 23. 4 Lorenz, W., Schable, K. F., Thiebe, R., Stavnezer, J. and Zachau, H. G., Mol. Immunol. 1988. 25: 479. 5 Pech, M., Smola, H., Pohlenz, H.-D., Straubinger,B., Gerl, R. and Zachau, H. G., J. Mol. Biol. 1985. 183: 291. 6 Zimmer, E-J., Hameister, H., Schek, H. and Zachau, H. G., EMBO J. 1990. 9: 1535. 7 Pohlenz, H.-D., Straubinger, B., Thiebe, R., Pech, M., Zimmer, E-J. and Zachau, H. G., J. Mol. Biol. 1987. 193: 241. 8 Lorenz,W., Straubinger, B. and Zachau, H. G., Nucleic Acids Res. 1987. 15: 9667. 9 Weichhold, G. M., Klobeck, H.-G., Ohnheiser, R., Combriato, G. and Zachau, H. G., Nature 1990. 347: 90. 10 Weichhold, G. M., PhD Thesis 1991, Fakultat fiir Chemie und Pharmazie, Universitat Miinchen. 11 Pargent, W., Schlble, K. E and Zachau, H. G., Eur. J. Immunol. 1991. 21: 1829. 12 Klobeck, H.-G., Zimmer, E-J., Combriato, G. and Zachau, H. G.. Nucleic Acids Res. 1987. 15: 9655.

0 regions of the human immunoglobulin x locus

1827

13 Lorenz,W., PhD Thesis 1989, Fakultat fiir Biologie, Universitat Munchen. 14 Lawn, R. M., Fritsch, E. E, Parker, R. C., Blake, G. and Maniatis, T., Cell 1978. 1.5: 11.57. 15 Jaenichen, H. R., PhD Thesis 1984, Fakultat fiir Biologie, Universitat Miinchen. 16 Reed, K. C. and Mann, D. A., Nucleic Acids Res. 1985. 13: 7207. 17 Church, G. M. and Gilbert, W., Proc. Natl. Acad. Sci. USA 1984. 81: 1991. 18 Messing, J., Methodr Enzymol. 1983. 101: 20. 19 Chen, E.Y. and Seeburg, P. H., DNA 1985. 4: 165. 20 Pargent,W., PhD Thesis 1991, Fakultat fur Biologie, Universitat Miinchen. 21 Hollis, M. and Hindley, J., Nucleic Acids Res. 1988. 16: 363. 22 Meindl, A., PhD Thesis 1989, Fakultat fiir Biologie, Universitat Miinchen. 23 Graninger,W. B., Goldman, P. L., Morton, C. C., OBrien, S. J. and Korsmeyer, S. J., J. Exp. Med. 1988. 167: 488. 24 Kato, S. ,%chibana, K. ,%kayama, N., Kataoka, H. ,Yoshida, C. and Takano, T., Gene 1991. 97: 239. 2.5 Klobeck, H.-G., Combriato, G. and Zachau, H. G., Nucleic Acids Res. 1984. 12: 6995. 26 Kabat, E. A., Wu, T. T., Reid-Miller, M., Perry, H. M. and Gottesman, K. S., Sequences of Proteins of lmunological Interest, 4th Edn., NIH Publication, Bethesda 1987, p. 41. 27 Lotscher, E., Grzeschik, K.-H., Bauer, H. G., Pohlenz, H.-D., Straubinger, B. and Zachau, H. G., Nature 1986. 320: 456. 28 Klobeck, H.-G., Combriato, G. and Zachau, H. G., Bid. Chem. Hoppe-Seyler 1989. 370: 1007. 29 Zimmer, F.-J., Huber, C., Quenzel, M., Schek, H., Stiller, C., Thiebe, R. and Zachau, H. G., Biol. Chem. Hoppe-Seyler 1990. 371: 283. 30 Burke, D.T., Carle, F. G. and Olson, M.V., Science 1987.236: 806. 31 Meindl, A., Klobeck, H.-G., Ohnheiser, R. and Zachau, H. G., Eur. J. Immunol. 1990. 20: 1855. 32 Skowronski, J., Fanning, T. G. and Singer, M. F., Mol. Cell. Biol. 1988. 8: 138.5. 33 Deininger, P. L., Jolly, D. J., Rubin, C. M., Friedmann,T. and Schmid, C.W., J. Mol. Biol. 1981. 151: 17. 34 Falkner, F. G. and Zachau, H. G., Nature 1984. 310: 71.