A repeated element in the regulatory region of the MNK gene and its deletion in a patient with occipital horn syndrome
A repeated element in the regulatory region of the MNK gene and its deletion in a patient with occipital horn syndrome Barbara Levinson1, Rebecca Conant1,2, Rhonda Schnur4, Soma Das3, Seymour Packman3 and Jane Gitschier1,2,3,*
1Howard Hughes Medical Institute, 2Department of Medicine and 3Department of Pediatrics, University of California, San Francisco, CA 94143, USA and 4Children's Hospital of Philadelphia, Philadelphia, PA, USA
Received May 30, 1996;Revised and Accepted August 22, 1996DDBJ/EMBL/GenBank accession no. U59314
Occipital horn syndrome (OHS), an X-linked connective tissue disorder, has recently been shown to result from mutations in the Menkes disease gene (MNK), which encodes a copper-transporting ATPase. By Southern analysis we detected a small deletion in a region 5' to the MNK gene in one patient with OHS. Genomic clones from an unaffected individual were isolated and sequenced, revealing three tandem 98 bp repeats situated upstream of the reported transcription start site, and analysis of the patient's DNA showed a deletion of one of the repeats. The deletion is likely to be responsible for the disease in this patient, as it was not observed in 110 unaffected individuals analyzed, and no other mutation in the patient was detected by RT-PCR and chemical cleavage mismatch analysis or by cDNA sequence analysis. The deletion is associated with a dramatic decrease in expression of a chloramphenicol acetyltransferase reporter gene, implicating the repeat sequences in regulation of MNK expression, although a quantitative analysis of MNK mRNA from a cell line derived from the patient shows no detectable reduction. Other experiments revealed no effect on the site of transcription initiation, termination or on splicing.
The X-linked disorders occipital horn syndrome (OHS) and Menkes disease profoundly differ in their clinical presentation but are both the result of mutations in a gene encoding a copper-transporting P-type ATPase (MNK) (1 ). Menkes disease results from mutations that cause debilitating effects on the MNK protein, such as deletions, nonsense mutations, frameshifts and mutations in the invariant splice junction sequences (GT and AG) (2 ). These mutations are associated with the classic, severe Menkes phenotype, which includes neurological degeneration and death in early childhood, resulting from a global deficiency of copper (1 ). In contrast, the three reported OHS mutations and one mild Menkes mutation lie adjacent to the invariant splice sequences in the flanking conserved residues which form the splice donor and acceptor sites (3 ,4 ). These defects reduce the efficiency of normal splicing, resulting in very reduced levels of otherwise normal mRNA. OHS is a mild connective tissue disorder characterized by lax skin, bony abnormalities, hernias and bladder diverticulae, stemming from deficiency of the cupro- enzyme lysyl oxidase, with little or no neurological impairment (5 ,6 ). Thus we hypothesize that in OHS, lysyl oxidase may be more susceptible than other cupro-enzymes to reduced amounts of copper consequent to reduced levels of the copper-transporting ATPase (3 ).
Another mechanism which could lead to a decrease in expression of the MNK protein involves mutations affecting promoter or enhancer activities. Such mutations have been described for other genetic disorders, e.g. [beta]-thalassemia, hemophilia B and retinoblastoma (7 ). The regulatory elements for the MNK copper- transporting ATPase gene have not been defined. We report the cloning and sequencing of the 5' upstream region of the MNK gene and the finding of a small deletion in this 5' sequence in a patient with OHS. Our analysis of this deletion has led to the discovery of three novel tandem 98 bp repeat sequences, which are probably necessary for the proper expression of the MNK gene in humans.
A thin, 32-year old man was noted to have loose finger joints, a hyperextensible back, prominent conjunctival vasculature, prominent veins and redundant skin folds on the plantar surface of the feet, loose `velvety' skin, easy bruisability limited to his feet, easy fatiguability and frequent exercise-induced cramping of his legs and feet. The patient has had repeated bladder diverticulae and an inguinal hernia. X-rays revealed a small occipital horn, but no abnormalities in the clavicles. The plasma ceruloplasmin and copper levels were low and low-normal, respectively. The patient has a healthy five year old son and three healthy brothers, all of them have several healthy children. A sister, with `vein trouble' in her legs which led to a unilateral amputation, has two sons, one having bilateral hernias in his teens and the other having very loose skin and hyperextensibility. The family members were not available for study.
Cell cultures of skin biopsies (line #RS-93.1) and Epstein-Barr virus-transformed lymphoblastoid cells (#93-104) were prepared from the patient as described. Copper accumulation, as measured by atomic absorption (8 ), was assayed in lymphoblastoid cells, and the patient's cells exhibited a 4- to 6-fold increase over the control cell lines. Based on these findings, the patient was given the diagnosis of OHS. This patient has not been reported previously.
Southern blots of the OHS patient's genomic DNA digested with eight different restriction enzymes were probed sequentially with three MNK cDNA probes encompassing the open-reading frame (9 ). No changes were found with probes from the central or 3' portions of the cDNA (probe 7.2.2 or [beta]8.1 respectively; data not shown); however, with a probe (8.2.4) encompassing the 5' region, the 3.5 kb BglII and the 7.0 kb PvuII bands were found to be slightly smaller in the patient. Rehybridization of the same blot with a 155 bp probe which includes only the 5'-untranslated region and the first 19 bp of the open-reading frame showed the same alterations (Fig. 1 ). These findings suggested a small deletion in the 5' region of the MNK gene in this patient.
A genomic clone from a X-chromosomal specific library was isolated by screening with the 155 bp 5' probe and found to contain the 3.5 kb BglII fragment. The BglII fragment was sub-cloned and partial sequence was obtained as described in Materials and Methods and as shown in Figure 2 . The sequence includes 1336 bp 5' to the reported transcription start site, 124 bp of exon 1, all of which is untranslated, and 352 bp of the intron 3' to exon 1.
PCR amplification of the patient's DNA (see Materials and Methods) revealed an ~100 bp deletion of the upstream sequence by gel analysis. Sequencing of the PCR products further showed that precisely one of the three 98 bp tandem repeats is deleted in the patient's DNA (Fig. 2 ), probably resulting from homologous recombination between misaligned repeats. The exact boundaries of the deletion could not be determined, due to the similarity of the repeats, but the few sequence differences allow the exchanges to be positioned between positions -521 and -403.
Two strategies were used to assess whether this small deletion was the cause of the patient's disease. First, DNA samples from 110 unaffected individuals, of both sexes, were analyzed by PCR and none were found to have this deletion. This result suggests that the deletion is not a polymorphism, although the possibility that it is a rare neutral variant cannot be ruled out. Second, the patient's MNK cDNA, prepared by RT-PCR, was screened for other mutations in two ways. The complete coding region was sequenced, and no nucleotide changes were found. In addition, amplified segments were screened for mutations by chemical cleavage as described previously (2 ), and no changes of PCR-product size, indicative of splice site mutations, were observed. Thus, our evidence suggests that this mutation is found only in the patient with OHS, and that the patient has only this mutation in the regions of the gene assayed.
To determine whether the deletion of one of the three repeats could affect gene expression, a series of reporter gene constructs were made and tested. Portions of the 5' region of the MNK gene were PCR-amplified from patient and control cell lines and were inserted in front of a chloramphenicol acetyltransferase (CAT) coding sequence and CAT protein was measured by an ELISA assay (Fig. 3 A). Almost identical levels of CAT were produced from plasmids containing either two or three repeats when the constructs extend from just before the repeats through exon 1 (Fig. 3 B). In contrast, no CAT is produced in a construct including an additional 242 bases 5' to the repeats if only two repeats are present, whereas constructs with the three repeats yielded ~75% of CAT activity in the other constructs. These findings suggest that the presence of three, as opposed to two, repeats is probably needed for proper expression of the MNK gene in the context of a larger region of genomic DNA. The formal interpretation of this finding is that the more upstream sequences exert a negative effect on transcription which can be relieved only if three, but not two, of the repeats are present.
Studies were done to assess the effects of this deletion on expression of the MNK gene in cell lines derived from the patient. A northern blot containing poly(A)+ RNA isolated from both fibroblast and lymphoblastoid cell lines from the patient and a control individual was hybridized to 5' and central probes (8.2.4 and 7.2.2, respectively) as described in Materials and Methods. Although no striking difference in either size or abundance was observed between control and patient's MNK transcript, a more quantitative assessment was undertaken. A densitometric analysis of serial dilutions of total RNA on a northern blot hybridized to MNK and [beta]-actin control probes also confirmed the presence of comparable levels of MNK transcripts in both normal and patient's cells (see Materials and Methods, data not shown).
Thus a series of experiments were performed to investigate whether a more subtle change exists in the MNK transcript of the patient. First, we considered that the deletion could affect either the start site of transcription or the splicing of the initial exons. (Only the initial exons were considered, since aberrations in splicing of the coding exons had already been ruled out by the RT-PCR experiments described above.) A 5' RACE procedure (19 ), initiated by a primer in exon 5,was used to address both questions. Multiple overlapping transcription initiation sites were observed within a 65 bp region in both the control and patient samples, as shown in Figure 2 , and we do not regard the differences in these to be significant. In addition, an alternatively spliced, untranslated exon (exon 1A, Fig. 3 ), located between published exons 1 and 2 (Dierick,H., Glover,T. and Mercer,J., pers. comm.), was found to be incorporated in 25% of transcripts in both normal and patient samples as determined by cloning and sequence analysis. Thus, by RACE no significant difference could be found between patient and normal samples in either the site of transcription initiation or in the pattern of alternative splicing of 5' exons.
Second, we considered whether the deletion could somehow lead to aberrant termination of transcription and polyadenylation in the MNK transcript of the patient. A 3' RACE protocol (19 ) was used to detect possible differences in transcript length or polyadenylation sites. No difference between the 3'-ends in the patient and control samples was observed in this experiment, but it did lead to an unexpected finding. Previously it was noted (20 ) that the MNK transcript lacks the canonical polyadenylation signal within 15-40 bases of the poly(A) tail but one does exist 462 bp 5' to the end of the transcript (9 ). Using primers 5' to this site, we found that 90% of the cloned RACE molecules, in both control and patient samples, end 19 bp from the U in this AAUAAA motif and the poly(A) tract is added at the most commonly reported CA dinucleotide. The remaining 10% of the clones end at the previously published site. The sequence downstream from the AAUAAA site also consists of a very U- and G-rich region, which is putatively involved in the assembly of the pre-mRNA cleavage complex (21 ).
We report the sequence of the region 5' of the MNK gene and describe three 98 bp tandem repeats which contain a number of potential binding sites for known transcription factors. We demonstrate that the upstream region will promote transcription, as assayed by expression of a CAT reporter gene. We found that one of the three repeats is deleted in a patient with OHS, and that loss of this repeat abolishes expression if sufficient flanking genomic DNA is included in the reporter gene assay. These findings, together with the apparent lack of any other mutations in the MNK gene of the patient, suggest that OHS in this patient is due to a deletion of one of the 98 bp repeat sequences.
Surprisingly, the 98 bp deletion does not result in a reduction in MNK expression although there is a profound effect on reporter gene synthesis. Both fibroblast and lymphoblastoid cell lines have levels of MNK mRNA comparable with that of control cell lines and equivalent in size. In a series of extensive experiments, we have investigated whether transcription initiation, termination, or splicing are aberrant in the patient's cells and have found no significant changes from the normal control. To date we have no explanation for this intriguing inconsistency. We must consider that the presence of MNK RNA in the patient's cultured cells may not accurately mirror what is happening in vivo. For example, one of the three potential MREs is deleted in the patient's DNA, and thus his gene may not have the normal response to metals.
Interestingly, another patient with OHS has been reported to have a deletion of sequences upstream of the transcription start site (Tumer,Z., Tonneson,T. and Horn,N., pers. comm.). By Southern blot analysis, the patient's deletion is clearly different from that described herein. In addition this patient has low MNK RNA by northern blot (B.Levinson, unpublished results). Future analysis of this patient may also lead to new insights into MNK regulation.
These findings provide the foundation for future investigations of the regulatory elements in the MNK gene. For example, the effects of the 98 bp deletion on genomic sequence can be examined in more detail. Is the specific 98 bp repeat sequence required to alleviate repression of gene expression, or does the sequence merely provide the proper spacing? Are the putative binding sites for transcription factors actually utilized? Is there an effect of copper on transcription and is it mediated through the potential MRE sites identified?
Previously, we reported that the MNK gene is widely expressed, with the transcript present in brain, muscle, lung, heart and spleen (9 ).This nearly ubiquitous expression is consistent with the presence of a CpG island, which is often found in the 5' region of `housekeeping' genes. A notable exception to this pattern of expression of MNK is the liver, in which no MNK transcript is found. Thus, there must be elements in the regulatory region of the MNK gene which specifically turn off expression in the liver. Parallel studies on the regulatory region of the WD gene (associated with Wilson disease), which is expressed primarily in the liver, should lead to an interesting comparison of the regulation of these two copper-transporting ATPase genes.
Four MNK cDNA probes were used in this study to screen Southern, northern and genomic libraries. The 155 bp 5' probe consists of the MNK cDNA sequence from nucleotides 9 to 164 (9 ) which includes most of the 5' UTR and 19 nt of the open reading frame. Probe 8.2.4 spans nucleotides 48-1905, probe 7.2.2 spans nucleotides 1151-2936, andprobe [beta]8.1 spans nucleotides 3239-5081.
Phage clones (100 000) from an X-specific, human genomic library (ATCC cat. #5770) were screened with the nick-translated 32P-labeled 155 bp probe. Two positive phage clones were isolated, plaque purified and DNA was prepared. A Southern blot of the BglII digested phage DNA was screened with the 155 bp probe and the positive 3.5 kb fragment was gel purified and cloned into a BamHI digested Bluescript KS (Stratagene) vector and partially sequenced using USB Sequenase 2.
Five different amounts of total RNA derived from patient and control lymphoblastoid cell lines were run on a northern blot. The blot was hybridized to 32P-labeled [beta]-actin and MNK (7.2.2) probes and the bands were scanned and quantitated using an Alpha Innotech IS-1000 Digital Imaging System. The integrated densities of each band were plotted and the ratios of MNK to actin mRNA were derived from the slopes of the lines. The densities of the bands were linear with concentration of RNA.
To determine the MNK transcription start sites, poly(A)+ RNA was isolated from 107 fibroblast and lymphoblastoid cells using the Pharmacia QuickPrep Micro RNA Purification Kit. Poly(A)+ RNA (1 [mu]g) was reverse-transcribed using the Invitrogen cDNA kit with a gene-specific, anti-sense primer GSP-RT (5'-CATTAGTTGAAGTCAGAAGCG-3') in exon 5 (20 ,22 ). The first strand was purified away from excess primers using a Qiagen PCR purification kit. A poly(A) tail was added to the first-strand cDNA product and amplified with GSP1 antisense (5'-CGATGAAGGCTGAGCTATTAC-3') and Primer Qo and Qt as described by Froman (19 ). The product from this PCR reaction was diluted 1:20, and 1 [mu]l was amplified with QI and nested GSP2 antisense (5'-CTACCAACGGCTCATTCGTG-3'). The final product was shot-gun cloned using the Invitrogen TA cloning kit. The clones were sequenced with USB Sequenase 2.
The 3' end of the human MNK transcript was isolated by RT-PCR, as described for 5' RACE, using an oligo-dT primer for reverse transcription of the mRNA. The first-strand product was amplified using a GSP1 sense (5'-TTCACTCAGCAACCATGCCC-3') and the Qo primer (19 ). The PCR product was diluted as above and amplified with a nested GSP-2 sense (5'-GCCGGTTCAGACTCACCT-3') and QI primers. The PCR product was cloned and sequenced as for the 5'RACE.
Poly(A)+ RNA, isolated from the patients cultured fibroblasts, was reverse-transcribed using the Invitrogen cDNA kit and random primers. The resulting cDNA was amplified in four domains using a series oligonucleotide primers as described by Das et al. (2 ). These PCR products were cloned into a Invitrogen TA cloning vector and sequenced using an ABI prism TM 377 DNA sequencer.
The CAT reporter constructs were prepared by inserting upstream sequences amplified from normal and OHS patient DNA using primers with PstI and HindIII restriction sites (sense primers 5'-CTGTAAAGCTTTTGCATACCGGAATCTTC-3' and 5'-TCT- GAAGCTTACATCTGGAATTAGGCTA-3', antisense primer 5'-AACTCTGCAGACCGTGGCAGCGA-3') into the HindIII /PstI sites of the pCAT Basic expression vector (Promega). Each construct was prepared twice, once with Perkin-Elmer Amplitaq and once with a proofreading Taq polymerase (Ultima, Perkin-Elmer). The resulting colonies were screened directly by PCR to confirm presence of the correct insert, and the sequence was verified by DNA sequencing.
A293 cell lines were grown and maintained in DMEM + 10% fetal calf serum. 24 h before transfection, the cells were plated at a density of 1 * 105 cells per 35 mm plate. Each plasmid was transfected using Lipofectamine (GibcoBRL) into duplicate plates with a [beta]-galactosidase ([beta]-gal) plasmid cotransfected to control for transfection efficiency. Three control transfections involved either a no DNA negative control, a positive control of pCAT control plasmid (with CAT and a promoter), or a pCAT basic without an insert. The GibcoBRL lipofection protocol for 293 cells was followed, using 1 [mu]g CAT plasmid DNA +- 0.2 [mu]g [beta]-gal plasmid DNA. Each experiment was repeated three times.
The plates were harvested 48 h after transfection and the cells incubated for 15 min at room temperature in 300 [mu]l [beta]-gal reporter lysis buffer (Promega). Cell lysates were spun at 14 000 r.p.m. for 3 min and the supernatant was used for assays of the amount of CAT produced, [beta]-gal activity and total protein. Lysate (100 [mu]l) from each plate was used in duplicate in the CAT ELISA kit (BMB) assay. [beta]-Gal activity was determined by the Promega protocol using 50 [mu]l of lysate. Protein determinations were done using the Biorad reagent. The CAT values were standardized by dividing by their respective [beta]-gal values and corrected to 100 [mu]g protein for each sample. Each transfection experiment was done twice using the two independently derived constructs.
We thank Tom Glover, Herman Dierick and Julian Mercer for communicating unpublished results on the 5' alternatively spliced exon; Zenip Tumer, Tonne Tonneson and Nina Horn for information concerning another OHS patient with a possible 5' rearrangement; Michael Frohman for advice and primers for the 5' and 3' RACE experiments; Eric Fodor for analysis of the 5' regulatory region; Martha Gunthorpe for sequencing the patient's cDNA and Hernan Consenco for technical assistance.
1 Danks, D.M. (1989) In Scriver,C., Beaudet,A., Sly,W. and Valle,D. (eds), The Metabolic Basis of Inherited Disease. McGraw-Hill, Inc., New York, NY, pp. 1411-1432.
2 Das, S., Levinson, B., Whitney, S., Vulpe, C., Packman, S. and Gitschier, J. (1994) Diverse mutations in patients in Menkes disease often lead to exon skipping. Am. J. Hum. Genet., 55, 883-889.MEDLINE Abstract
3 Das, S., Levinson, B., Vulpe, C., Whitney, S., Gitschier, J. and Packman, S. (1995) Similar splicing mutations of the Menkes/Mottled copper-transporting ATPase gene in occipital horn syndrome and the blotchy mouse. Am. J. Hum. Genet., 56, 570-576.MEDLINE Abstract
4 Kaler, S.G., Gallo, L.K., Proud, V.K., Percy, A.K., Mark, Y., Segal, N.A., Goldstein, D.S., Holmes, C.S. and Gahl, W.A. (1994) Occipital horn syndrome and a mild Menkes phenotype associated with splice site mutations at the MNK locus. Nature Genet., 8, 195-202.MEDLINE Abstract
5 Lazoff, S.G., Rybak, J.J., B.R., P. and Lazzatti, L. (1975) Skeletal dysplasia, occipital horns, diarrhea and obstructive uropathy-a new hereditary syndrome. Birth Defects: Original Article Series, 11, 71-74.
6 Byers, P.H., Siegel, R.C., Holbrook, K.A., Naryanan, A.S., Bornstein, P. and Hall, J.G. (1980) X-linked cutis laxa. N. Engl. J. Med., 303, 61-65.MEDLINE Abstract
7 Cooper, D. (1993) Human gene mutations affecting RNA processing and translation. Ann. Med., 25, 11-17.MEDLINE Abstract
8 Packman, S., Chin, P. and O'Toole, C. (1984) Copper utilization in cultured skin fibroblasts of the mottled mouse, an animal model for Menkes' kinky hair syndrome. J. Inher. Metab. Dis., 7, 168-170.
9 Vulpe, C., Levinson, B., Whitney, S., Packman, S. and Gitschier, J. (1993) Isolation of a candidate gene for Menkes disease and evidence that it encodes a copper-transporting ATPase. Nature Genet., 3, 7-13.MEDLINE Abstract
10 Bird, A.P. (1986) CpG-rich islands and the function of DNA methylation. Nature, 321, 209-213.MEDLINE Abstract
11 Larsen, F., Gundersen, B., Lopez, R. and Prydz, H. (1992) CpG islands as gene markers in the human genome. Genomics, 13, 1095-1107.MEDLINE Abstract
12 Kadonaga, J.T., Jones, K.A. and Tjian, R. (1986) Promoter-specific activation of RNA polymerase II transcription by Sp1. Trends Genet., 11, 20-23.
13 Montminy, M.R. and Bilezikjian, L.M. (1987) Binding of a nuclear protein to the cyclic-AMP response element of the somatostatin gene. Nature, 328, 175-178.MEDLINE Abstract
14 Lee, W., Mitchell, P. and Tjian, R. (1987) Purified transcription factor AP-1 interacts with TPA-inducible enhancer elements. Cell, 49, 741-752.MEDLINE Abstract
15 Reisman, D. and Rotter, V. (1993) The helix-loop-helix containing transcription factor USF binds to and transactivates the promoter of the p53 tumor suppressor gene. Nucleic Acid Res., 21, 345-350.
16 Jones, K.A., Kadonaga, J.T., Rosenfeld, P.J., Kelly, T.J. and Tjian, R. (1987) A cellular DNA-binding protein that activates eukaryotic transcription and DNA replication. Cell, 48, 79-89.MEDLINE Abstract
17 Stuart, G.W., Searle, P.F. and Palmiter, R.D. (1985) Identification of multiple metal regulatory elements in mouse metallothionein-I promoter by assaying synthetic sequences. Nature, 317, 828-831.MEDLINE Abstract
18 Cousins, R.J. (1994) Metal elements and gene expression. Annu. Rev. Nutrition, 14, 449-69.
19 Frohman, M. A. (1994) On Beyond Classic RACE (Rapid Amplification of cDNA Ends) in PCR Methods and Applications. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY, pp.S40-S57.
20 Dierick, H.A., Ambrosini, L., Spencer, J., Glover, T.W. and Mercer, J.F.B. (1995) Molecular structure of the Menkes disease gene (ATP7A). Genomics, 28, 462-469.MEDLINE Abstract
21 Keller, W. (1995) No end yet to messenger RNA 3' processing! Cell, 81, 829-832.MEDLINE Abstract
22 Tumer, Z., Vural, B., Tonnesen, T., Chelly, J., Monaco, A.P. and Horn, N. (1995) Characterization of the exon structure of the Menkes disease gene using vectorette PCR. Genomics, 26, 437-442.MEDLINE Abstract
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:28:51 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
