Molecular analysis of an unstable genomic region at chromosome band 11q23 reveals a disruption of the gene encoding the α2 subunit of platelet-activating factor acetylhydrolase (Pafah1a2) in human lymphoma

Article metrics

Abstract

A region of 150 kb has been analysed around a previously isolated, lymphoma associated, translocation breakpoint located at chromosome band 11q23. This balanced and reciprocal translocation, t(11;14)(q32;q23), has been shown to result in the fusion between chromosome 11 specific sequence and the switch γ4 region of the IGH locus. The LPC gene, encoding a novel proprotein convertase belonging to the furin family, has been identified in this region. In order to characterize further the region surrounding the translocation, we have determined the detailed structure of LPC. Here we show that LPC consists of at least 16 exons covering 25 kb, and that there is a partial duplication, involving mobile genetic elements and containing LPC exons 13 – 17 in a tail – tail configuration at 65 kb downstream. Since the chromosomal breakpoint lay between these two structures, the intervening region was further analysed and shown to contain at least two unrelated genes. The previously known SM22 gene was localized close to the 3′ tail of LPC. Furthermore, we identified the gene encoding the α2 subunit of platelet-activating factor acetylhydrolase (Pafah1a2) at the chromosomal breakpoint. The position of another previously identified breakpoint was also located to within the first intron of this gene. Altogether, our results give evidence of a genomic instability of this area of 11q23 and show that Pafah1a2 and not LPC is the gene disrupted by the translocation, suggesting that deregulated Pafah1a2 may have a role in lymphomagenesis.

Introduction

Human chromosome 11q23 is a pathologically important region, particularly in haematopoietic malignancies where it can be found rearranged with many different chromosomal regions in a variety of different leukaemia subtypes (Young, 1992). Cytogenetically visible alterations to 11q23 have been also noted in a proportion of malignant lymphoma, including the t(11;14)(q23;q32) translocation which is a rare but recurring event (Mitelman et al., 1997). The molecular cloning of such a t(11;14) translocation from a non-Hodgkin's lymphoma of a particular subgroup (MLCLS: Mediastinal Large Cell Lymphoma with Sclerosis) lead to the identification of a new gene at band 11q23 (Meerabux et al., 1994, 1996), distinct from both the RCK gene previously identified in a lymphoma cell line (Akao et al., 1993) and from the HTRX/MLL/ALL1 gene which is the target of multiple chromosomal translocations in acute leukemias (Bernard and Berger, 1995). This gene, termed LPC (also known as PC7/PC8/SPC7) belongs to the family of proprotein convertases (PCs) which are serine proteases, structurally related to bacterial subtilisins and yeast kexin (Seidah et al., 1994). PCs are implicated in the processing of numerous protein precursors to produce mature forms of regulatory proteins such as growth factors, hormones and viral glycoproteins. Notably, LPC is thought to be one of the proteases responsible for in vivo activation of HIV envelope glycoproteins gp160 and gp140 (Hallenberger et al., 1997). Further characterization of the LPC locus revealed an inverted partial duplication of the 3′ end of the gene, proximal to the breakpoint on chromosome 11. Within this region, we identified two different genes, one of which was the human homolog of the α2 subunit of bovine platelet-activating factor (PAF) acetylhydrolase (Pafah1a2). PAF is a biologically active phospholipid involved in diverse physiological events such as inflammation and anaphylaxis (Bazan, 1995). It is inactivated by a specific enzyme, PAF acetylhydrolase (PAF-AH) (Blank et al., 1981), the cytosolic isoform Ib being a heterotrimer comprising a regulatory subunit of 45K, and two catalytic subunits of 30K and 29K, previously referred as α, β and γ, respectively (Hattori et al., 1995). It has been shown recently that this intact PAF-AH (Ib) molecule is an unusual G-protein-like trimer (Ho et al., 1997). Homology to the structurally related G-protein subunits has led to a new nomenclature for the α, β and γ subunits which have been renamed β, α2 and α1, respectively. This paper describes a comprehensive map of approximately 150 kb of genomic DNA from 11q23, containing mobile genetic elements and at least three unrelated genes including LPC and the gene encoding the α2 subunit of PAF acetylhydrolase. Our results strongly suggest that this area of chromosome 11 is an unstable genomic region, prone to rearrangements. Moreover, we show here that Pafah1a2 is the target of the translocation, leading to the loss of its first non-coding exon and juxtaposition to the immunoglobulin heavy chain locus. The potential role of this gene in lymphomagenesis is discussed.

Results

Structure of genomic LPC

The molecular cloning of the breakpoint of a t(11;14)(q23;q32) translocation led to the identification of the LPC gene, a previously unknown member of the subtilisin proprotein convertase gene family (Meerabux et al., 1996). In order to establish clearly the consequences of the translocation and to facilitate the search for new rearrangements in the human locus, the genomic region surrounding the breakpoint has been mapped and partially sequenced. A probe located at the 3′ end of the LPC cDNA was used to screen a P1 genomic library. Three genomic P1 clones (K1680 O1669 and B2221) were isolated and mapped by FISH to chromosome 11q23, in agreement with the previously determined location of the LPC gene (Meerabux et al., 1996) (not shown). Hybridization experiments with LPC cDNA probes revealed that two of the genomic clones, K1680 and O1669, contained all the coding exons of the gene (not shown) whereas the clone B2221 appeared to contain only a portion of the 3′ part of LPC and hybridized with a different pattern. We were able to determine the position of introns as well as exon – intron boundaries of the LPC gene, by PCR amplification from the P1 clones K1680 and O1669 using cDNA primers and comparing the size of the PCR products against that predicted from the cDNA sequence. All products which were longer than predicted were expected to contain intronic sequences, and were subsequently cloned and sequenced. The sequence data were in agreement with the PCR analysis, with a few exceptions where the sequencing revealed the presence of two introns between the PCR primers. Similar results were obtained by amplifying from genomic DNA with the same primers. These results show that human LPC is composed of at least 16 exons, of relatively short size (119 bp on average, except for exons 3 (480 bp) and 17 (1271 bp)). Parallel analysis of the murine LPC gene (Ebihara et al., submitted) indicated the existence of a 1st non-coding exon which we have so far been unable to identify. For this reason the exons are numbered from 2 – 17 (Figure 1a). The catalytic domain, common to all the PCs and important for the processing of proproteins, is encoded by exons 4 – 11. Table 1 summarized the exact sizes and positions of exons and introns of LPC, as well as boundary sequences. All the splice junctions exhibit a high similarity to the consensus sequences for donor and acceptor sites. Restriction enzyme digestion and Southern hybridization analysis of K1680 were used to construct a precise map of the genomic LPC region which covers about 25 kb (Figure 1a).

Figure 1
figure1

Genomic map of the 11q23 LPC locus. Scales are indicated in kb. (a) Genomic organization of the LPC and SM22 genes. The positions of the 16 exons of LPC and the five exons of SM22 as predicted by XGRAIL analysis are indicated (solid boxes, translated areas; open boxes, untranslated regions). The position of the probes 21-36 and 22-T2 are indicated by dark lines. The map shows the positions of the BamHI (B) restriction sites and some of the XbaI (Xb) and HindIII (H) restriction sites. (b) Restriction map of this region of 11q23 showing BamHI (B), NotI (N) and SacII (Sc) restriction sites. Transcriptional orientation of the different genes is indicated by arrows. The P1 clones are represented by hatched bars. (c) Genomic organization of the Pafah1a2 locus showing the position of BamHI (B), SacII (Sc) and some HindIII (H) and XbaI (Xb) restriction sites. The positions of the Pafah1a2 exons and the duplicated exons of LPC (solid boxes, translated areas; open boxes, untranslated regions) are shown. The position of the t(11;14) breakpoint is indicated by an arrow. The L1 retrotransposon is represented by a grey bar

Table 1 Table 1

The 3′ end of LPC is duplicated

Using a probe located at the 5′ end of exon 17 (probe 21 – 36, see Figure 1a), the third P1 clone (B2221) showed a different hybridization pattern from K1680 when digested with a variety of restriction enzymes (Figure 2a). Furthermore, only probes covering LPC exons 13 – 17 gave a signal with B2221 (not shown), suggesting that it did not contain a full copy of LPC. To determine whether these configurations were derived from two different alleles of the same gene or from a duplication event, we performed Southern analysis with the probe 21 – 36 on a YAC clone (911f02) known to include this region. As shown on Figure 2b, a hybridization pattern was obtained which is a composite of that obtained with K1680 and B2221 when hybridized separately. The fact that both configurations were present on the same YAC clone suggested that part of LPC could be duplicated. Since YACs have a propensity to undergo recombination, we used the 22-T2 probe (located just downstream of the 21 – 36 probe, and covering the 3′ part of LPC exon 17, see Figure 1) for Southern blot analysis on 14 unrelated human genomic DNAs. In all cases the composite pattern detected with the YAC clone was obtained, confirming that the 3′ end of LPC is duplicated. The results from three representative genomic DNA samples are shown in Figure 3. Restriction mapping analysis showed that B2221 and K1680 overlap by 40 kb (see Figure 1b), and that LPC exons 13 – 17 are duplicated, in a tail – tail configuration with the expressed gene. In order to compare the duplicated regions, we entirely sequenced the 3.3 kb covering exons 13 – 17 in the expressed gene and in the partial duplication (GenBank accession numbers AF057710 and AF057709, respectively). We found 35 point mutations as well as a 10 bp insertion in the partial duplication (not shown). Interestingly, the majority of the mutations occurred in the regions corresponding to the introns of the actual gene.

Figure 2
figure2

A composite pattern for the 3′ end of LPC. Southern blots of DNA from the P1 clones K1680 and B2221 (a) and YAC 911f02 (b) were hybridized with the probe 21-36, derived from the 5′ part of LPC exon 17. The restriction enzymes used were B: BamHI; E: EcoRI; H: HindIII. The sizes of the different bands are indicated in kb on both sides of the figures

Figure 3
figure3

Part of LPC is duplicated. A Southern blot of human genomic DNAs was hybridized with the probe 22-T2, covering the 3′ part of LPC exon (i.e. 200 bp of LPC duplicated 17th exon). Numbers 1, 2 and 3 indicate three (over 14) unrelated human DNAs, digested with BamHI. The sizes of the different bands are indicated in kb on the right side of the figure

The duplicated region of 11q23 contains mobile genetic elements such as Alu repeats and L1 retrotransposons

A number of Alu repeats telomeric to the duplicated exon 17 were identified and have replaced the 3′ end of this last exon (Figure 4). Further sequence analysis also demonstrated the presence of a long interspersed nuclear element (LINE or L1) telomeric to this Alu sequence (Figure 4). L1s are an abundant class of mammalian retrotransposons lacking long terminal repeats, which are estimated to comprise 5% of nuclear DNA (Feng et al., 1996; Kazazian Jr and Moran, 1998; Sassaman et al., 1997). The full-length L1 consensus contains two intact open reading frames (ORF1 and ORF2) encoding a nucleic acid binding protein and a protein with endonuclease (EN) and reverse transcriptase (RT) activities, respectively (Feng et al., 1996; Kazazian Jr and Moran, 1998; Sassaman et al., 1997) (Figure 4). As for most L1 repeats, the LPC line repeat is truncated at the 5′ end, leading to a 3.6 kb element, sharing about 90 – 92% nucleotide identity with the human L1 consensus. As shown in Figure 4, this truncated copy is flanked by a perfect 18 bp target site duplication.

Figure 4
figure4

Diagrammatic representation of the organization of the LPC LINE1 locus. A schematic of a human L1 consensus is represented on the top of the figure (from Sassaman et al., 1997). ORF1 and ORF2 are indicated by a light grey box and a dark grey box, respectively. The 5′ and 3′ untranslated regions (UTR) are indicated by striped boxes and the poly(A) tail by An. The approximate position of the endonuclease (EN), reverse transcriptase (RT) and cysteine-rich motifs (C) in ORF2 are indicated. The 5′ truncated LPC L1 element is represented below, with part of ORF2 and its poly(A) tail. Flanking repeats of this LINE1 are indicated by dotted circles, with the precise sequence of the target site duplication framed by a white box below. Alu repeats are indicated by a hatched line. The duplicated 17th exon of LPC and the 1st exon of Pafah1a2 are shown as well as the position of the t(11;14) translocation breakpoint indicated by an arrow. The scale is indicated in bp

Existence of genes between LPC and its duplicated counterpart

The translocation breakpoint previously described (Meerabux et al., 1994) occurred between LPC and its duplicated counterpart (Figure 1b), and we therefore analysed further the intervening region. Analysis of a GenBank cosmid (U73638) which covers the 3′ end of the LPC gene, revealed the presence of the gene SM22, which encodes a calponin-related protein, specifically expressed in adult smooth muscle (Li et al., 1996a). SM22 which terminates 250 bp from 17th exon of LPC, is composed of five exons covering about 6 kb (see Figure 1a). The gene is transcribed in the opposite orientation from LPC. Further analysis of the intervening region also revealed the presence of a sequence with 91% identity with the bovine Pafah1a2 gene, located telomeric to the duplicated part of LPC (Figures 1c and 4). To facilitate the analysis of this gene, we cloned and sequenced the human Pafah1a2 cDNA, extending by 400 bp the sequence recently published (Adachi et al., 1997). By using the same method as for LPC we determined the genomic structure of the gene which is summarized in Table 2. The human Pafah1a2 consists of six exons, spread over approximately 20 kb, with the coding region starting in exon 2 (Figure 1c). The first non-coding exon of Pafah1a2 is located 540 bp from the LINE1 3′ end (Figure 4). The translocation breakpoint previously described occurs in the Pafah1a2 first intron, with the predicted consequence of placing the coding exons under the control of immunoglobulin heavy chain (IgH) regulatory elements (Meerabux et al., 1994). As recently shown (Adachi et al., 1997), we found by Northern analysis that the Pafah1a2 gene is ubiquitously expressed (not shown).

Table 2 Table 2

Discussion

Band q23 of human chromosome 11 is frequently involved in haematopoietic disorders by chromosomal rearrangements (Mitelman et al., 1997). In order to facilitate the search of additional mutations in this region, which may play a role in human pathogenesis, we have determined the genomic organization of about 150 kb of DNA surrounding a t(11;14)(q23;q32) translocation breakpoint, associated with a human high-grade lymphoma. The molecular cloning of this translocation led us previously to the identification of the LPC gene, a novel member of the proprotein convertase family (Meerabux et al., 1996).

Human LPC consists of at least 16 exons covering 25 kb. The general organization of the gene is nearly identical to that of the murine gene (17 exons spread over 25 kb) and the short average length of the exons is also conserved between both species (Ebihara et al., submitted). Interestingly, the positions of exon boundaries appear to be very similar to those found for the murine LPC, but are quite different from those of other members of the PCs (Ebihara et al., submitted). This suggests that LPC could belong to a separate subgroup in this family. We also showed that the gene SM22, encoding a protein specific for smooth muscle cells, was located just downstream of LPC in the opposite orientation. Our results correlate with the recently described cytogenetic localization and structure of the human gene (Camoretti-Mercado et al., 1998). Moreover, the position and the orientation of this gene are also conserved in the murine locus (Ebihara et al., submitted).

Furthermore, we showed that LPC is followed at 65 kb by a partial duplication, containing LPC exons 13 – 17 in a tail – tail configuration. Sequence analysis of the duplicated region showed that the nucleotide divergence from the real gene is 1.07%. Based on an estimated mutation rate of nuclear DNA of about 0.14 – 0.21% per million years (MY) (Li et al., 1996b), we can envisage that the duplication event occurred around 2.5 and 3.3 MY ago. Since the divergence between primates and rodents occurred at least 65 MY ago, these results are consistent with the fact that the murine LPC locus does not seem to contain such a duplication (M Ebihara, personal results). Such constitutional partial duplications have been found for other genes such as BRCA1 and PKD1, which are implicated in breast cancer and in polycystic-kidney disease, respectively (Brown et al., 1996; Consortium TEPKD, 1994).

Detailed analysis of the organization of DNA sequences in the proximal part of the translocation breakpoint region reveals a complex picture. Various mobile genetic elements, including Alu sequences and LINE1 retrotransposons were found telomeric to the duplication. Alu repeats are thought to be critical mediators of gene duplication. For example, the partial duplication of MLL has been shown to occur from an Alu-mediated homologous recombination (Schichman et al., 1994). This mechanism appears to play a role in some cases of gene rearrangement in somatic tissues (Strout et al., 1998). Insertion of LINE1 elements have been previously associated with human diseases, including hemophilia A and colorectal cancer, by disrupting the genes encoding factor VIII or APC, respectively (Kazazian Jr et al., 1988; Miki et al., 1992). However, since ORF1 is deleted and ORF2 is inactivated by several mis-sense mutations, the LPC line repeat does not belong to the active class of L1 recently described (Sassaman et al., 1997). This truncated copy is flanked by a perfect 18 bp target site duplications, suggesting it is the result of a retrotransposition event. Furthermore, the 5′ end of this flanking repeat begins with the TTGAAA sequence (see Figure 3), which is defined as a target signal for elements with 3′ poly(A) tails and which is the consensus sequence cleaved by L1 EN (Feng et al., 1996). Our results strongly favour the recent model of preferred integration sites for L1 elements in the genome (Feng et al., 1996; Jurka, 1997).

It is not unexpected to find Alu repeats and L1 close to each other since Alu elements are mobilized by L1 at a very high frequency, L1 acting on Alu to instigate their retrotransposition (Boeke, 1997; Kazazian Jr and Moran, 1998). To explain the origin of a gene duplication, several mechanisms have been proposed including RNA-mediated transposition or unequal but homologous crossing-over between repeated sequences (Maeda and Smithies, 1986). The conservation of the exon/intron structure in the partial duplication suggests that it arose from the integration of genomic material rather than cDNA copies. Furthermore, we have evidence that there is another 5′ truncated LINE1 on the B2221 clone, centromeric to the duplication. So, even if the exact sequence of events that led to the present duplication is not clear, our results raise the possibility of pairing and unequal crossing-over between L1 and/or Alu elements.

Altogether, our results give evidence of a genomic instability in this area of chromosome 11q23, emphasized by the presence of mobile genetic elements (LINE1, Alu) and the duplicative inversion of LPC. The organization of DNA in tandemly duplicated form is known to be unstable and to promote rearrangements. Moreover, Alu and LINE1 have often been found at or surrounding translocations breakpoints (Kurahashi et al., 1998; Toriello et al., 1996; Von Lindern et al., 1992). Notably in the t(6;9) translocation associated with acute myeloid leukemia (AML) the presence of L1 in the breakpoint region was postulated to play a role in the translocation process (Von Lindern et al., 1992). In our case, the significance and consequences of the unusual instability of this area of chromosome 11q23 need to be explored. Further work may provide explanations for understanding some of the mechanisms which lead to chromosomal translocations associated with human malignancies.

Our analysis of this locus maps the human Pafah1a2 gene to 11q23, in accordance with its recently FISH-described cytogenetic localization (Moro et al., 1998). A particularly significant finding is that the translocation breakpoint occurred in the first intron of Pafah1a2. This recombination event does not affect the coding sequences of the gene but removes the first non-coding exon and places the remaining exons under the control of IgH regulatory elements (Meerabux et al., 1994). Chromosome 14 breaks are the most frequent aberrations in lymphoma. Interestingly, in this translocation, the break on chromosome 14 occurs in the switch region, a few bases away from that of the RCK lymphoma cell line (Akao et al., 1993; Meerabux et al., 1994). Furthermore, such a translocation is similar to the t(8;14) which deregulates the c-myc gene in human Burkitt's lymphoma. Using transgenic animal models, the deregulation of c-myc by loss of the 1st exon and juxtaposition to the IgH enhancer sequence has proven to be oncogenic (Adams et al., 1985). Other cases of recurring chromosomal breakpoints have been described in the first intron of a number of genes which become deregulated by their transposition next to regulatory sequences of the IgH gene. This includes the t(14;19) translocation implicating bcl-3 in chronic B-cell leukemia (CLL) (Crossen et al., 1993), and the t(3;14) translocation deregulating bcl-6 in diffuse large cell lymphomas (DLCL) (Baron et al., 1993; Ye et al., 1995). By analogy with these cases, we may expect that this t(11;14)(q23;q32) translocation could lead to deregulation of Pafah1a2 in lymphoid cells. We previously identified another lymphoma with a rearrangement in the same region (Meerabux et al., 1996), suggesting that disruption to Pafah1a2 may be a recurrent event in a proportion of non-Hodgkin's lymphomas. Although the LPC gene is relatively distant from the breakpoint it remains possible that some long range effect on its expression could be an additional consequence of this event. Also the role of the partially duplicated LPC sequence remains uncertain in this context. However, the fact that two breakpoints have been found within the 1st intron of the Pafah1a2 gene strongly suggests that it, rather than LPC, is the primary genetic target in these events. Unfortunately, lack of available tissue has limited our ability to look at the consequences of the translocation. Clearly, much additional work is needed to elucidate how translocation involving Pafah1a2 on 11q23 could contribute to malignancy. Particularly, it would be very useful to screen for new patients with MLCLS to determine whether the Pafah1a2 expression is deregulated in leukemic cells.

Platelet-activating factor (PAF) is a potent phospholipid mediator of inflammation synthesized by many cell types, including macrophages, platelets, basophils, eosinophils and endothelial cells on appropriate stimulation (Bazan, 1995; Hattori et al., 1995). It mediates a broad spectrum of biological activities such as hypotension, smooth muscle contraction and an increase in vascular permeability (Hattori et al., 1995). Moreover, PAF is known to have potent regulatory activities in vitro on numerous blood cell types, including T and B lymphocytes, monocytes and neutrophils (Denizot et al., 1995b). These actions of PAF are mediated mainly through specific cell surface receptors, although accumulation of intracellular PAF may also influence cell function. PAF is degraded by the specific enzyme, PAF acetylhydrolase (PAF-AH) (Blank et al., 1981; Hattori et al., 1995). The physiological function of PAF-AH has not been established yet, but several hypotheses have been proposed. PAF-AH abolishes the inflammatory effects of PAF on leucocytes and the vasculature. In addition, it has been speculated that PAF acetylhydrolase may scavenge oxidized phospholipids produced inside or outside the cells during oxidative stress and block the generation of modified low-density lipoprotein (LDL) (Bazan, 1995). The cytosolic enzyme (Ib) exists as a heterotrimer (α1/α2β) (Hattori et al., 1994, 1995; Ho et al., 1997), whose cellular concentration and activities are regulated by the level of expression of each subunit. The α2 and α1 subunits are also known to exist as homodimers but little is known about their role in this form (Ho et al., 1997). Thus it may be speculated that if there is a deregulation of Pafah1a2 expression in lymphoid cells, it could have a profound effect on the PAF-AH Ib form present in the cytosol and that it could affect the processing and levels of biologically important molecules such as PAF and/or LDL. It is interesting to note that PAF itself has been directly implicated in proliferation, differentiation and transformation of various cell types (Bennett and Birnboim, 1997; Bennett et al., 1993; Roth et al., 1996; Shimada et al., 1998) and that decreased levels of PAF have been reported in blood and bone marrow cells of patients with lymphoid malignancies (Denizot et al., 1995a,b).

The gene for the β subunit (Pafah1b1 or LIS1), located at 17p13, has been cloned independently as the gene involved in Miller – Dieker lissencephaly (MDS), a human brain malformation manifesting as a smooth cerebral surface and abnormal neuronal migration (Hattori et al., 1994; Reiner et al., 1993). Pafah1b1 has been postulated to interact with tubulin, suppressing microtubule dynamics (Sapir et al., 1997) and has recently been shown to be crucial for neuronal migration (Hirotsune et al., 1998). Interestingly, the 50 kb first intron of this gene is interrupted by translocations or interstitial deletions associated with lissencephaly (Chong et al., 1997; Kurahashi et al., 1998). In one case recently described, sequence analysis showed that the breakpoint area was rich in Alu and L1 elements (Kurahashi et al., 1998). The gene encoding the α1 subunit, which shows 62.4% identity at the amino acid level with the α2 subunit (Adachi et al., 1997), has been located to the long arm of chromosome 19 (Moro et al., 1998). Further studies will be required to determine whether alterations to the component parts of PAF acetylhydrolase might be a recurring feature of human high-grade lymphoma and possibly other malignancies.

Materials and methods

P1 clones and YAC

The P1 clones K1680, O1669 and B2221, were isolated from a human male lymphoblastoid cell line library obtained from the Reference library – ICRF (Lehrach, 1990), using part of LPC exon 17 as a probe (probes 21 – 36, Figure 1a). The exact references of each clone are as follows: K1680: ICRFP700K1680Q06; O1669: ICRFP700O1669Q06; B2221: ICRFP700B2221Q06. The Yeast Artificial Chromosome (YAC) 911fO2 was obtained from the CEPH YAC library.

Fluorescence in situ hybridization

Probe DNA was labelled by nick translation using biotin-14-dATP (Gibco – BRL Bio-nick kit), and purified over a Sephadex G50 column. Pre-association of the probe DNA (40 ng/μl) plus unlabelled human Cot-1 DNA (0.4 μg/μl) was at 37°C for 3 h. Chromosome preparations were pre-treated with RNase (100 μg/ml) for 60 min at 37°C, washed three times in 2×SSC, pH 7.0 and dehydrated through an ethanol series. Chromosomal DNA was denatured in 70% formamide, 2×SSC at 70 – 75°C for 3 – 5 min then dehydrated through a cold alcohol series. The preannealed probe mixture (in 50% formamide, 2×SSC, 10% dextran sulphate) was applied to the denatured chromosomes on slides and hybridized overnight at 37°C. Following hybridization, sequential washing was performed to remove unbound probe (3×5 min washes in 50% formamide, 2×SSC, pH 7, followed by 3×5 min washes in 2×SSC, pH 7, both at 42°C). Protein binding sites were blocked using 4×SSC, 3% bovine serum albumin (BSA). Bound probe was detected following incubation with (1) 5 μg/ml fluorescein isothiocyanate-conjugated avidin D – cell sorter grade (FITC-avidin DCS, Vector Laboratories; (2) 5 μg/ml biotinylated goat anti-avidin antibody (Vector Laboratories); and (3) FITC-avidin DCS. All incubations were at 37°C for 30 min. Detection reagents were diluted in 4×SSC, 0.05% Triton X-100, 3% BSA. Three washes of 3 min each were performed between layers using 4×SSC, 0.05% Triton X100. Chromosomes were counterstained using propodium iodide and mounted in Citifluor AF1 medium (Citifluor Ltd).

Determination of LPC and Pafah1a2 intron-exon structure

Introns were localized by amplification from the genomic P1 clones using combinations of primers from the LPC or Pafah1a2 cDNA sequences (listed below) and comparison of the PCR product size with that predicted from the cDNA. PCR conditions were as follows: Taq: 1 U/50 μl (Promega), buffer as supplied, Mg2+: 1.5 mM, dNTPs: 100 μM each, oligos: 0.4 μM each. Reactions were cycled at 94°C for 7 min, then 25 times at 94°C for 30 s, 58 or 60°C for 45 s and 72°C for 4 min. All the products expected to contain intronic sequences, were purified with the Gene Clean kit (BIO 101), cloned in the T/A pCRTM 2.1 vector (Invitrogen), as indicated by the supplier and subsequently sequenced to determine exon/intron boundaries. Primers from LPC cDNA or Pafah1a2 cDNA are given in their sense orientation. The second number indicates the use of the antisense oligonucleotide.

LPC primers

These primers have been used in the following combinations:

Pafah1a2 primers

These primers have been used in the following combinations:

Preparation and restriction mapping of DNA

YAC and P1 clone DNA was extracted using Qiagen tips (Qiagen) or PEG precipitation (Sambrook et al., 1989). Restriction enzyme analysis was performed with BamHI, HindIII, NotI, SacII and XbaI in single and double digests (Promega) in accordance with the supplier's instructions. Digested DNA was separated by electrophoresis through 0.5 – 1% agarose and this DNA visualized by staining with ethidium bromide. DNA was transferred by Southern blotting to Hybond N+ membranes (Amersham) for subsequent hybridization with probes from the LPC and Pafah1a2 genes.

DNA sequencing

Between 200 and 500 ng of cloned PCR products was initially sequenced with M13 Reverse and Forward primers using DyeDeoxy Terminator Cycle Sequencing (Applied Biosystems), by cycling 25 times at 96°C for 30 s, 50°C for 15 s and 60°C for 4 min). A primer-walking strategy was then adopted for sequencing large PCR fragments. Sequencing products were extracted once with chloroform and were run on an Applied Biosystems 373A DNA sequencer. In order to determine the boundaries of large introns, direct sequencing of the P1 clones was performed, using the Thermo Sequenase radiolabelled terminator cycle sequencing kit (Amersham) with α-33P-ddNTPs. All the results were analysed using the Sequence Navigator program and contigs were generated with the DNASTAR program Seqman II (Lasergene). The sequences of the various sequencing primers used are available on request. All the relevant information reported in this study have been submitted to GenBank.

Computer analysis

Additional gene sequences were localized on to our P1-based map of this region of chromosome 11 and analysed as follows. A GenBank (http://www.ncbi.nlm.nih.gov) cosmid sequence (accession number U73638) was found to contain the 3′ end of LPC starting about 1 kb upstream of LPC seventh exon. XGRAIL analysis (http://compbio.ornl.gov) of the cosmid sequence predicted the presence of five exons downstream of LPC. These putative exons showed identity with SM22 gene sequences in the Genbank database when it was searched using the BLAST program.

Gene bank accession numbers

AF057709: Duplicated exons 13 – 17 of LPC, LINE1 element and Pafah1a2 first exon. AF057710: LPC from exons 13 – 17, normal part.

References

  1. Adachi H, Tsujimoto M, Hattori M, Arai H and Inoue K. . 1997 Biochem. Biophys. Res. Com. 233: 10–13.

  2. Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S, Palmiter RD and Brinster RL. . 1985 Nature 318: 533–538.

  3. Akao Y, Tsujimoto Y, Seto M, Imai T, Bergeron D, Berbeau B and Otsuki Y. . 1993 Genes Chrom. Cancer 8: 167–171.

  4. Baron BW, Nucifora G, McCabe N, Espinosa RD, Le Beau MM and McKeithan TW. . 1993 Proc. Natl. Acad. Sci. USA 90: 5262–5266.

  5. Bazan NG. . 1995 Nature 374: 501–502.

  6. Bennett SA and Birnboim HC. . 1997 Mol. Carcinog. 20: 366–375.

  7. Bennett SAJ, Leite LCC and Birnboim HC. . 1993 Carcinogenesis 14: 1289–1296.

  8. Bernard OA and Berger R . (1995) Genes Chrom. Cancer 13: 75–85.

  9. Blank ML, Cress EA, Whittle T and Snyder F. . 1981 Life Sci. 29: 769–775.

  10. Boeke JD. . 1997 Nature Genet. 16: 6–7.

  11. Brown MA, Xu CF, Nicolai H, Griffiths B, Chambers JA, Black D and Solomon E. . 1996 Oncogene 12: 2507–2513.

  12. Camoretti-Mercado B, Forsythe SM, LeBeau MM, Espinosa RR, Vieira JE, Halayko AJ, Willadsen S, Kurtz B, Ober C, Evans GA, Thweatt R, Shapiro S, Niu Q, Qin Y, Padrid PA and Solway J. . 1998 Genomics 49: 452–457.

  13. Chong SS, Pack SD, Roschke AV, Tanigami A, Carrozzo R, Smith AC, Dobyns WB and Ledbetter DH. . 1997 Human Mol. Genet. 6: 147–155.

  14. Consortium TEPKD. . 1994 Cell 77: 881–894.

  15. Crossen PE, Kennedy MA, Heaton DC and Morrison MJ. . 1993 Genes Chrom. Cancer 8: 60–62.

  16. Denizot Y, Trimoreau F, Duouis F, Verger C and Praloran V. . 1995a Blood 85: 2992–2993.

  17. Denizot Y, Trimoreau F, Duouis F, Verger C and Praloran V. . 1995b Leukemia 9: 1982–1983.

  18. Feng Q, Moran JV, Kazazian Jr HH and Boeke JD. . 1996 Cell 87: 905–916.

  19. Hallenberger S, Moulard M, Sordel M, Klenk HD and Garten W. . 1997 J. Virol. 71: 1036–1045.

  20. Hattori M, Adachi H, Aoki J, Tsujimoto M, Arai H and Inoue K. . 1995 J. Biol. Chem. 270: 31345–31352.

  21. Hattori M, Adachi H, Tsujimoto M, Arai H and Inoue K. . 1994 Nature 370: 216–218.

  22. Hirotsune S, Fleck MW, Gambello MJ, Bix GJ, Chen A, Clark GD, Ledbetter DH, McBain CJ and Wynshaw-Boris A. . 1998 Nature Genet. 19: 333–339.

  23. Ho YS, Swenson L, Derewenda U, Serre L, Wei Y, Dauter Z, Hattori M, Adachi T, Aoki J, Arai H, Inoue K and Derewenda ZS. . 1997 Nature 385: 89–93.

  24. Jurka J. . 1997 Proc. Natl. Acad. Sci. USA 94: 1872–1877.

  25. Kazazian Jr HH and Moran JV. . 1998 Nature Genet. 19: 19–24.

  26. Kazazian Jr HH, Wong C, Youssoufian H, Scott AF, Phillips DG and Antonarakis SE. . 1988 Nature 332: 164–166.

  27. Kurahashi H, Sakamoto M, Ono J, Honda A, Okada S and Nakamura Y. . 1998 Hum. Genet. 103: 189–192.

  28. Lehrach H. . (1990) In: Genome analysis volume1. Davies KE and Tilghman SM (eds).. Cold Spring Harbor Laboratory Press: Cold Spring Harbor pp.39–81.

  29. Li L, Miano JM, Cserjesi P and Olson EN. . 1996a Circ. Res. 78: 188–195.

  30. Li WH, Ellsworth DL, Krushkal J, Chang BHJ and Hewett-Emmett D. . 1996b Mol. Phylogenet. Evol. 5: 182–187.

  31. Maeda N and Smithies O. . 1986 Ann. Rev. Genet. 20: 81–108.

  32. Meerabux J, Yaspo ML, Roebroek AJ, Van de Ven WJ, Lister TA and Young BD. . 1996 Cancer Res. 56: 448–451.

  33. Meerabux JM, Cotter FE, Kearney L, Nizetic D, Dhut S, Gibbons B, Lister TA and Young BD. . 1994 Oncogene 9: 893–898.

  34. Miki Y, Nishisho I, Horii A, Miyoshi Y, Utsunomiya J, Kinzler KW, Vogelstein B and Nakamura Y. . 1992 Cancer Res. 52: 643–645.

  35. Mitelman F, Mertens F and Johansson B. . 1997 Nature Genet. 15: 417–474.

  36. Moro F, Arrigo G, Fogli A, Bernard L and Carrozzo R. . 1998 Genomics 51: 157–159.

  37. Reiner O, Carrozzo R, Shen Y, Wehnert M, Faustinella F, Dobyns WB, Caskey CT and Ledbetter DH. . 1993 Nature 364: 717–721.

  38. Roth M, Nauck M, Yousefi S, Tamm M, Blaser K, Perruchoud AP and Simon HU. . 1996 J. Exp. Med. 184: 191–201.

  39. Sambrook J, Fritsch EF and Maniatis T. . 1989 Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor.

  40. Sapir T, Elbaum M and Reiner O. . 1997 EMBO J. 16: 6977–6984.

  41. Sassaman DM, Dombroski BA, Moran JV, Kimberland ML, Naas TP, DeBerardinis RJ, Gabriel A, Swergold GD and Kazazian Jr HH. . 1997 Nature Genet. 16: 37–43.

  42. Schichman SA, Caligiuri MA, Strout MP, Carter SL, Gu Y, Canaani E, Bloomfield CD and Croce CM. . 1994 Cancer Res. 54: 4277–4280.

  43. Seidah NG, Chretien M and Day R. . 1994 Biochimie 76: 197–209.

  44. Shimada A, Ota Y, Sugiyama Y, Sato S, Kume K, Shimizu T and Inoue S. . 1998 J. Invest. Dermatol. 110: 889–893.

  45. Strout MP, Marcucci G, Bloomfield CD and Caligiuri MA. . 1998 Proc. Natl. Acad. Sci. USA 95: 2390–2395.

  46. Toriello HV, Glover TW, Takahara K, Byers PH, Miller DE, Higgins JV and Greenspan DS. . 1996 Nature Genet. 13: 361–365.

  47. Von Lindern M, Breems D, van Baal S, Adriaansen H and Grosveld G. . 1992 Genes Chrom. Cancer 5: 227–234.

  48. Ye BH, Chaganti S, Chang CC, Niu H, Corradini P, Chaganti RS and Dalla-Favera R. . 1995 EMBO J. 14: 6209–6217.

  49. Young BD. . 1992 Bailliere's Clin. Haematol. 5: 881–895.

Download references

Acknowledgements

This work is dedicated to Alain Lecointe, Botanist (1943 – 1998). We thank Christine Courtes and Emmanuel Douzery for assistance in the sequencing, and Debra Lillington and Michael Neat for the FISH analysis. We are grateful to Armelle Degeorges, Emmanuel Douzery and Ian Robbins for helpful discussions. NL was supported by a fellowship from the Association de la Recherche contre le Cancer. AH was supported by a grant from the Commission of European Communities, contract number F14P-CT-95-008. This work was supported in part by the Concerted Action, contract number CA-11.

Author information

Correspondence to Bryan D Young.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • 11q23
  • mediastinal large cell lymphoma with sclerosis (MLCLS)
  • lymphoma proprotein convertase (LPC)
  • α2 subunit of platelet activating factor acetylhydrolase (Pafah1a2)
  • gene duplication
  • LINE1 element

Further reading