Mutation hotspots in the PHKA2 gene in X-linked liver glycogenosis due to phosphorylase kinase deficiency with atypical activity in blood cells (XLG2)
Mutation hotspots in the PHKA2 gene in X-linked liver glycogenosis due to phosphorylase kinase deficiency with atypical activity in blood cells (XLG2)Barbara Burwinkel, Yoon S. Shin1, Henk D. Bakker2, Johann Deutsch3, María José Lozano4, Irène Maire5 and Manfred W. Kilimann*
Institut für Physiologische Chemie, Medizinische Fakultät, Ruhr-Universität Bochum, D-44780 Bochum, Germany, 1Stoffwechselzentrum, Dr. v. Haunersches Kinderspital der LMU München, D-80337 München, Germany, 2Emma Kinderziekenhuis, Universiteit van Amsterdam, NL-1105 AZ Amsterdam, The Netherlands, 3Universitätsklinik für Kinder- und Jugendheilkunde, A-8036 Graz, Austria, 4Dpto. de Pediatría, Hospital Universitario M. de Valdecilla, Universidad de Cantabria, E-39011 Santander, Spain and 5Centre d'Étude des Maladies Métaboliques, Hôpital Debrousse, F-69322 Lyon Cedex 5, France
Received February 19, 1996; Revised and Accepted February 26, 1996
In five cases of X-linked liver glycogenosis subtype 2 (XLG2), we have identified mutations in the gene encoding the liver isoform of the phosphorylase kinase [alpha] subunit (PHKA2). XLG2 is a rare variant of X-linked phosphorylase kinase (Phk) deficiency of the liver. Whereas in the more common form of X-linked hepatic Phk deficiency, XLG1, the enzyme's activity is decreased both in liver and in blood cells, Phk activity in XLG2 is low in liver but normal or even enhanced in blood cells. Although missense, nonsense and splice-site mutations in the PHKA2 gene were recently identified in several cases of XLG1, no mutations have yet been described for XLG2 and a molecular explanation for the peculiar biochemical phenotype of XLG2 has been lacking. All mutations found in the present study result in non-conservative amino acid replacements of residues that are absolutely conserved between the [alpha]L, [alpha]M and [beta] subunits of Phk [H132P, H132Y, R186H (twice) and D299G]. Strikingly, in two pairs of cases the mutations affect the same codon. These results demonstrate that: (i) XLG2 is caused by mutations in PHKA2 and is therefore allelic with XLG1; and (ii) XLG2 mutations appear to cluster in limited sequence regions or even individual codons.
Glycogen storage diseases are caused by deficiencies of various enzymes directly involved in glycogen metabolism or indirectly linked to it (glucose-6-phosphatase, enzymes of glycolysis). One quarter of all cases of glycenosis are due to deficiency of phosphorylase kinase (Phk), a protein kinase regulating the activity of the glycogen-degrading enzyme, glycogen phosphorylase (1 ). Several subtypes of Phk deficiency are known which differ in their mode of inheritance and in the tissues affected. This heterogeneity is ascribed to the structural complexity of Phk. This enzyme consists of four subunits, ([alpha][beta][gamma][delta])4, each of which has several isoforms or splicing variants (2 ). The most frequently occurring subtype of Phk deficiency (~75%) is X-linked liver glycogenosis (XLG). Infants affected by XLG typically present with hepatomegaly, growth retardation and muscle weakness. Unlike other glycogenoses, XLG is generally a benign condition. Patients improve with age and are often asymptomatic as adults (1 ,3 ). Accurate diagnosis is therefore also of prognostic interest.
In XLG, Phk activity is low in liver but normal in muscle. In most cases, Phk activity is also decreased in blood cells (XLG, subtype 1), and diagnosis is normally based on measurement of Phk activity in erythrocytes without the need for a liver biopsy (3 ,4 ). However, in recent years a number of cases of hepatic Phk deficiency have been observed in which Phk activity was untypically high in erythrocytes or leukocytes, necessitating liver biopsy to establish the diagnosis. In some of these cases, family history suggests an X-linked mode of inheritance (XLG, subtype 2) (5 -10 ).
Both the XLG1 disease gene and the structural gene of the liver isoform of the Phk [alpha] subunit (PHKA2) have been mapped to the distal short arm of the X chromosome (11 ,12 ). Recently, several mutations were described in PHKA2 which document that it is indeed the XLG1 gene (13 -15 ). In four families with XLG2, this trait was also mapped by linkage analysis to the chromosomal region that contains the PHKA2 gene, suggesting that XLG1 and XLG2 are allelic (10 ). To test whether this is indeed the case, and to resolve the molecular basis of the peculiar biochemical phenotype of XLG2, we have searched for, and identified, mutations in the PHKA2 gene in five male cases of hepatic Phk deficiency with atypical activity in blood cells (two of them with evidence of X-linked inheritance).
RNA was purified from frozen whole blood samples or cultivated fibroblasts, and the complete PHKA2 coding sequence amplified by RT-PCR in seven overlapping segments and analyzed by direct sequencing (Table 1 ). Previous PHKA2 mutation analysis was either based on genomic DNA and thus limited to the downstream 30% of the PHKA2 gene that has been characterized to date (13 ), or it was based on RNA obtained from cultured lymphocytes (14 ). We demonstrate here that RNA from whole blood is a readily accessible source for the analysis of mutations in the complete PHKA2 coding sequence.
We initially suspected that the differential manifestation, in XLG2, of Phk deficiency in liver and blood cells might be caused by mutations in sequence regions that are expressed in a correspondingly tissue-specific way, e.g. by differential splicing. Analysis of the only PHKA2 sequence region known to be subject to differential splicing, the downstream half of the multiphosphorylation domain (16 ), did not lead to the identification of mutations. Moreover, the amplification of the complete PHKA2 coding sequence by RT-PCR from both human liver and blood RNA, showed that both sequences were identical with the known PHKA2 cDNA sequence (14 ) over their entire length. We also performed a more specific search for differential splicing in two candidate regions, namely: (i) the upstream half of the multiphosphorylation domain; and (ii) the region that is co-linear with the so-called [alpha]'-region of the muscle isoform of the [alpha] subunit, a calmodulin-binding sequence whose differential deletion produces a splicing variant of [alpha]m characteristic for heart and slow muscle (17 ). RT-PCR with primer pairs flanking these two regions was performed on RNAs from multiple human and rabbit tissues but yielded no indications of differential splicing (data not shown, see Materials and Methods for experimental details).
Finally, the complete PHKA2 coding sequences of five male XLG2 cases, three sporadic and two with evidence of X-linked inheritance, were screened for mutations. In all five cases we found missense mutations leading to the non-conservative replacement of amino acid residues that are highly conserved in the [alpha]L, [alpha]m and [beta] subunits (Fig. 1 ). These were the only sequence abnormalities found in the probands' PHKA2 coding sequences. Remarkably, the mutations in these five apparently unrelated families are restricted to only three codons that cluster in a region of particularly high sequence conservation (Figs 1 , 2 ; ref. 12 ). Where additional family members were analyzed, the respective mutations were never detected in unaffected male relatives whereas they were always found in obligate carriers and affected relatives (Fig. 3 ). All mutations were confirmed from at least two independent RNA preparations and RT-PCR reactions by multiple sequencing runs and (in some cases) restriction analyses. Two additional, isolated male cases initially diagnosed for XLG2 were found to have normal PHKA2 sequences and are currently being analyzed for mutations in other Phk genes.
. Primers employed for PCR (I-VII) and sequencing (S)
PCR/Seq.
Primer
Position/Direction
Sequence
I
Le [alpha] 10
-119>
5'-ATC CCA AGA ACC GAC TAA GG-3'
I
Le [alpha] 23
315<
5'-AGT GTG TTT GAA CTT CTC CAC-3'
II
Le [alpha] 12
216>
5'-CAA GGC CTA CGA GCT GGA G-3'
II
Le [alpha] 22
1039<
5'-GAA CAG CAT CAC CAC TGA AG-3'
III
Le [alpha] 26
798>
5'-GGC CTT TGC AGT GGA AGA TG-3'
III
Le [alpha] 13
1376<
5'-CCG TGT TTC CTC AAT AAG TC-3'
IV
Le [alpha] 6
1250>
5'-TCC TTG CCG CTG GTG AAA TC-3'
S
Le [alpha] 20
1679>
5'-CAC ATC CAC CTT CCC CAT CA-3'
IV
Le [alpha] 4*
2409<
5'-AAC GGT GAC CCC GTG CTG T-3'
V
Le [alpha] 15
2319>
5'-CTA CAG GAC CAA GCA GAC AT-3'
V
Le [alpha] 16
2763<
5'-TGA TCA GTC CAA TCC GGA GT-3'
VI
Le [alpha] 5*
2687>
5'-TTT ACC TGG CCA TGT ACG TCA-3'
S
LePCR[alpha]2*
2909>
5'-TGC GTC CTA TCC ACT CCT C -3'
VI
LePCR[alpha]1*
3260<
5'-CTC TGG TAG AAG CCC ACG G-3'
VII
Le [alpha] 17
3151>
5'-TCA GAC TCG GGA GGA CAT CA-3'
S
Le [alpha] 19
3489>
5'-TGG ACC AGA TCG TGC AGA TG-3'
S
Le [alpha] 27
3575<
5'-TGG TCT TTC TCC AGG GTG TC-3'
VII
Le [alpha] 18
3990<
5'-AGA GTC CGT GAG ACC AGA TG-3'
mut.
Le [alpha] 28
268>
5'-CAG TGC ATG ATG AGA CAG GT-3'
diff. splic.
Le [alpha] 3*
1850>
5'-CCT TCC TGG ATC CAG ACT GT-3'
*Rabbit PhK-[alpha]L primers, positions refer to the rabbit PhK-[alpha]L-sequence.The position is given by the 5'-terminal nucleotide of each primer. Arrowheads indicate direction (forward or reverse).
Cases 1 (ref. 9 ), 2 (ref. 7 ), 4 (ref. 8 , member of group 2 with Phk deficiency) and 5 (ref. 10 , individual IV-3 of family 2) have been described previously, case 3 is unpublished. They all presented as infants with hepatomegaly, growth retardation and elevated triglycerides and transaminases, some also with muscle hypotonia and delayed motoric development.
Two candidate regions were amplified: the upstream half of the multiphosphorylation domain using primers Le[alpha]5 and LePCR[alpha]1 (= PCR VI, Table 1 ); and the region corresponding to the [alpha]M' region using primers Le [alpha]3 and Le [alpha]4 (Table 1 ) with the same PCR conditions as in PCR IV. For the first region, RT-PCR was carried out on RNAs from human liver, muscle, heart, testis and lymphocytes, and from rabbit liver, white muscle, red muscle, heart, intestine, brain, kidney, lung, uterus, testis and spleen; the resultant PCR products were cut with SphI. The second region was amplified from RNAs from human liver, muscle and lymphocytes, and from rabbit liver, white muscle, red muscle, heart, brain, kidney, lung, uterus and testis; PCR products were cut with RsaI. SphI and RsaI sites are within sequences that were suspected to be subject to differential splicing. However, only PCR products and restriction fragments of the sizes corresponding to the known sequences were obtained.
Total RNA was isolated according to Chomczynski and Sacchi (23 ) from cultured fibroblasts (patient 4 and relatives), whole blood samples (all others) or human and rabbit tissues (for investigation of differential splicing). EDTA-blood was quick-frozen after drawing, stored at -70oC, and shipped on dry-ice. Typically, a preparation from 1-2 ml blood yielded 10-20 [mu]g RNA. This quantity was usually sufficient for complete RT-PCR analysis. First-strand cDNA synthesis from 2-10 [mu]g of total RNA, primed with oligo-dT (Gibco BRL), was carried out with MMLV-reverse transcriptase (Gibco BRL) according to the manufacturer's instructions in a volume of 50 [mu]l.
The coding sequence of the [alpha]L subunit mRNA was amplified in seven overlapping fragments using primers listed in Table 1 . PCR reactions were carried out in 100 [mu]l volumes containing 0.3-1 [mu]l of template (first-strand cDNA reaction), 200 [mu]M of each dNTP and 600 ng of each primer. Enzyme and buffer were either 2.5 U Taq DNA Polymerase (Perkin Elmer) in 50 mM KCl and 10 mM Tris-HCl (pH 8.3) or 0.5 U Goldstar DNA Polymerase (Eurogentec) in 20 mM (NH4)2SO4, 0.01% (w/v) Tween-20, 75 mM Tris-HCl (pH 9.0). The MgCl2 concentration was 1.5 mM in all PCRs except PCR IV which required 3.0 mM MgCl2. Amplification was carried out in 45 cycles. After denaturation for 60 s at 96oC, annealing (60 s) followed a step-down temperature protocol: in the first 5 cycles by 2oC, in the following 10 cycles by 4oC, and in the final 30 cycles by 6oC below the calculated melting temperatures (Tm) of the primers (2oC for each A/T and 4oC for each G/C residue). Elongation was at 72oC for 90-120 s depending on product length. Primers LePCR[alpha]1, LePCR[alpha]2 and Le[alpha]3-Le[alpha]5 are complementary to the rabbit [alpha]L sequence. They contain 1-2 mismatches to the human sequence which, however, do not interfere with PCR or sequencing.
The PCR products were purified by electrophoresis in 1-2% low-melting-point agarose gels and sequenced directly with the ABI DyeDeoxy Terminator Cycle Sequencing kit, using the PCR primers and additional, internal sequencing primers (S) (Table 1 ). Bands were excised from the gel, melted and diluted with sterile H2O to a final template concentration of 50-100 ng/[mu]l. For each sequencing reaction, 9.5 [mu]l of molten gel slice was added to 9.5 [mu]l DyeDeoxy Terminator Premix and 1.0 [mu]l of primer (10 pmol/[mu]l) and overlayed with oil. After an initial denaturation at 96oC for 3 min, 30 sequencing cycles were performed as follows: 30 s denaturing at 96oC; 20 s annealing at 60oC or at Tm minus 2oC, whichever is lower; 4 min elongation at 60oC. Dye terminators were removed by phenol/chloroform extraction following the manufacturer's protocol.
PCR VI encompasses the multiphosphorylation site subject to differential splicing, resulting in heterogeneous amplification products. In order to separate the full-length PCR-product (containing regions A, B and C; ref. 16 ) from heteroduplexes formed with products lacking region B or regions A and B, the PCR product was digested with ScrF1. This enzyme recognizes two restriction sites, one 6 nt from the end of the PCR-product and a second site within region B. The resulting bands of 381 and 187 nt length were isolated by gel electrophoresis and sequenced with primers LePCR[alpha]2 and LePCR[alpha]1, respectively.
The mutations of cases 1, 2 and 4 create restriction sites. The region containing both sites was amplified with primers Le[alpha]28 and Le[alpha]22 (Table 1 ). ApaI or Bsp120 I leave the resulting PCR product of 771 nt uncut in the normal sequence but produce fragments of 647 and 124 nt with the P1 mutation. NlaIII produces fragments of 9, 24, 47, 303 and 388 nt in the normal sequence, whereas the P2/P4 mutation results in the additional cleavage of the latter fragment into pieces of 105 and 283 nt.
We thank Ms I. Schmitt and Dr O. Riess (Bochum) for human RNA samples, and Dr M.-T. Zabot (Lyon) for fibroblast cell culture. This work was supported by the Deutsche Forschungsgemeinschaft through a project grant and a Heisenberg fellowship to M.W.K., and by the Fonds der Chemischen Industrie.
1 Chen, Y.T. and Burchell, A. (1995) Glycogen storage diseases. In: Scriver, C.R., Beaudet, A.L., Sly, W.S. and Valle, D. (eds.) The Metabolic and Molecular Bases of Inherited Disease, Vol. I (7th edn), McGraw-Hill, New York, pp. 935-965.
2 Schneider,A., Davidson,J.J., Wüllrich,A. and Kilimann,M.W. (1993) Phosphorylase kinase deficiency in I-strain mice is associated with a frameshift mutation in the [alpha] subunit muscle isoform. Nature Genet. 5, 381-385.MEDLINE Abstract
3 Willems, P.J., Gerver, W.J.M., Berger, R. and Fernandes, J. (1990) The natural history of liver glycogenosis due to phosphorylase kinase deficiency: a longitudinal study of 41 patients. Eur. J. Pediatr.149, 268-271.MEDLINE Abstract
4 Shin, Y.S. (1990) Diagnosis of glycogen storage disease. J. Inher. Metab. Dis.13, 419-434.MEDLINE Abstract
5 Bashan, N., Potashnik, R., Ehrlich, T. and Moses, S.W. (1987) Phosphorylase kinase in leukocytes and erythrocytes of a patient with glycogen storage disease type IX. J. Inher. Metab. Dis.10, 119-127.MEDLINE Abstract
6 Alvarado, L.J.F., Gasca-Centebo, E. and Grier, R.E. (1988) Hepatic phosphorylase b kinase deficiency with normal enzyme activity in leukocytes. J. Pediatr.113, 865-867.
7 Bakker, H.D., Taminiau, J.A.J.M., Van Den Berg, J.E.T. and Berger, R. (1991) Hepatic phosphorylase b kinase deficiency with normal enzyme activity in leukocytes and erythrocytes. J. Inher. Metab. Dis.14, 269-270.MEDLINE Abstract
8 Maire, I., Baussan, C., Moatti, N., Mathieu, M. and Lemonnier, A. (1991) Biochemical diagnosis of hepatic glycogen storage diseases: 20 years French experience. Clin. Biochem.24, 169-178.MEDLINE Abstract
9 Lozano, M.J., Benlloch, T., Garcia, L.V. and Garcia Fuentes, M. (1994) Hepatic phosphorylase b kinase deficiency with normal enzyme activity in erythrocytes and muscle. J. Inher. Metab. Dis.17, 116-117.MEDLINE Abstract
10 Hendrickx, J., Coucke, P., Hors-Cayla, M.-C., Smit, G.P.A., Shin, Y.S., Deutsch, J., Smeitink, J., Berger, R., Lee, P., Fernandes, J. and Willems, P.J. (1994) Localization of a new type of X-linked liver glycogenosis to the chromosomal region Xp22 containing the liver [alpha]-subunit of phosphorylase kinase (PHKA2). Genomics21, 620-625.MEDLINE Abstract
11 Willems, P.J., Hendrickx, J., Van der Auwera, B.J., Vits, L., Raeymaekers, P., Coucke, P.J., Van den Bergh, I., Berger, R., Smit, G.P.A., Van Broeckhoven, C., Kilimann, M.W., Van Elsen, A.F. and Fernandes, J.F. (1991) Mapping of the gene for X-linked liver glycogenosis due to phosphorylase kinase deficiency to human chromosome region Xp22. Genomics9, 565-569.MEDLINE Abstract
12 Davidson, J.J., Özçelik, T., Hamacher, C., Willems, P.J., Francke, U. and Kilimann, M.W. (1992) cDNA cloning of a liver isoform of the phosphorylase kinase [alpha] subunit and mapping of the gene to Xp22.2-p22.1, the region of human X-linked liver glycogenosis. Proc. Natl Acad. Sci. USA89, 2096-2100.MEDLINE Abstract
13 Hendrickx, J., Coucke, P., Dams, E., Lee, P., Odievre, M., Corbeel, L., Fernandes, J.F. and Willems, P.J. (1995) Mutations in the phosphorylase kinase gene PHKA2 are responsible for X-linked liver glycogen storage disease. Hum. Mol. Genet.4, 77-83.MEDLINE Abstract
14 Van den Berg, I.E.T., van Beurden, E.A.C.M., Malingre, H.E.M., Ploos van Amstel, H.K., Poll-The, B.T., Smeitink, J.A.M., Lamers, W.H. and Berger, R. (1995) X-linked liver phosphorylase kinase deficiency is associated with mutations in the human liver phosphorylase kinase [alpha] subunit. Am. J. Hum. Genet.56, 381-387.
15 Hirono, H., Hayasaka, K., Sato, W., Takahashi, T. and Takada, G. (1995) Isolation of cDNA encoding the human liver phosphorylase kinase alpha subunit (PHKA2) and identification of a missense mutation of the PHKA2 gene in a family with liver phosphorylase kinase deficiency. Biochem. Mol. Biol. Int.36, 505-511.MEDLINE Abstract
16 Wüllrich, A., Hamacher, C., Schneider, A. and Kilimann, M.W. (1993) The multiphosphorylation domain of the phosphorylase kinase [alpha]M and [alpha]L subunits is a hotspot of differential mRNA processing and of molecular evolution. J. Biol. Chem.268, 23208-23214.MEDLINE Abstract
17 Harmann, B., Zander, N.F. and Kilimann, M.W. (1991) Isoform diversity of phosphorylase kinase [alpha] and [beta] subunits generated by alternative RNA splicing. J. Biol. Chem. 266, 15631-15637.MEDLINE Abstract
18 Newgard, C.B., Littman, D.R., van Genderen, C., Smith, M. and Fletterick, R.J. (1988) Human brain glycogen phosphorylase: Cloning, sequence analysis, chromosomal mapping, tissue expression, and comparison with the human liver and muscle isozymes. J. Biol. Chem.263, 3850-3857.MEDLINE Abstract
19 Dahan, N., Baussan, C., Moatti, N. and Lemonnier, A. (1988) Use of platelets, mononuclear and polymorphonuclear cells in the diagnosis of glycogen storage disease type VI. J. Inher. Metab. Dis.11, 253-260.MEDLINE Abstract
20 Cohen, P. (1973) The subunit structure of rabbit-skeletal-muscle phosphorylase kinase, and the molecular basis of its activation reactions. Eur. J. Biochem. 34, 1-14.MEDLINE Abstract
21 Cooper, D.N. and Krawczak, M. (1990) The mutational spectrum of single base-pair substitutions causing human genetic disease: patterns and predictions. Hum. Genet.85, 55-74.MEDLINE Abstract
22 Kilimann, M.W., Zander, N.F., Kuhn, C.C., Crabb, J.W., Meyer, H.E. and Heilmeyer, L.M.G.Jr. (1988) The [alpha] and [beta] subunits of phosphorylase kinase are homologous: cDNA cloning and primary structure of the [beta] subunit. Proc. Natl Acad. Sci. USA85, 9381-9385.MEDLINE Abstract
23 Chomczynski, P. and Sacchi, N. (1987) Single-step method of RNA isolation by acidic guanidinium thiocyanate-phenol-chloroform extraction. Anal. Biochem. 162, 156-159.MEDLINE Abstract
*To whom correspondence should be addressed
This page is maintained by OUP admin. Last updated Thu Oct 31 15:23:49 GMT 1996. Part of the OUP Journals World Wide Web service.Copyright Oxford University Press, 1996
