SAMJ

ARTICLES

in the central Cape had high stillbirth, ENNO and LBW rates, which suggest both poor living conditions and inadequate Familial defective perinatal care. Of interest are the eastern Cape regions where the stillbirth rate is high in spite of relatively few LBW apolipoprotein-B is rare in infants. The reasons for this are uncertain, but intrapartum deaths due to hypoxia in well-grown infants may be hypercholesterolaemic common. The two regions with the highest PMRs were Lower South African Afrikaners, Orange River and Port Elizabeth-Uitenhage. The Lower Orange River region is an area that stretches over a great coloureds and Indians- distance with a low population density and only 6 889 deliveries. A probable explanation for the high PMR is the David C. Rubinsztein, Gerhard A. Coetzee, Deneys inaccessibility of health services in this region. In contrast R. van der Westhuyzen, Elzet Langenhoven, Port Elizabeth-Uitenhage, which is a densely populated Maritha J. Kotze region, has a large, unemployed community and many migrants seeking care from limited services. Having identified regions with many LBW infants and high The frequency of familial defective apolipoprotein B-100 PMRs, further studies are needed to document the causes (FOB) was assessed among hypercholesterolaemic and plan appropriate intervention policies. A shortcoming of Afrikaners, coloureds and Indians. Patients selected for the present study was that data were collected according to screening did not carry any of the founder or common the place of birth and not the place of residence. The results LOL-receptor mutations known to be associated with could therefore be influenced by patterns of migration and referral. Coding data according to home addresses would these groups. No FOB was detected and the mutation is be of greater value but more difficult to collect. All infants therefore a rare cause of hypercholesterolaemia in these weighing between 500 and 1 000 g should also be included South African populations. in future surveys. These additional data would increase the S Atr Med J 1995; 85: 355-357. mortality rates, especially the stillbirth rate. In conclusion, a survey of perinatal mortality within the planning regions of the Cape Province reveals wide Familial defective apolipoprotein B-100 (FOB) is geographical differences and a pattern expected in a characterised by the presence of LDL particles that carry developing country. In addition, the survey identifies regions apolipoprotein B-100 with the arg -glu mutation; these where particular attention must be paid to living conditions 3500 have a markedly lower affinity for the LOL receptor and are and perinatal health services. cleared abnormally slowly from the circulation.' Heterozygotes for this disorder present with We would like to thank the staff of all the hospitals and clinics in the Cape Province who helped with the collection of the data hypercholesterolaemia and clinical sequelae such as presented in this study. xanthomata with a frequency and severity similar to that found in heterozygotes with familial hypercholesterolaemia REFERENCES (FH), a disorder caused by LD.L-receptor mutations!-4 Some

1. Van Coerverden de Groat HA, Oavey OA, Howland RC. The Peninsula Maternity and individual cases, however, have been reported in which FOB Neonatal Service. S AIr Med J 1982; 61: 35-36. heterozygotesS and one homozygote· had relatively 2. New Health Regions for South Africa. Epidemiological Comments 1992; 19: 59-68. 3. World Health Organisation. Division of Family Medicine. The incidence of low birth moderate hypercholesterolaemia. The incidence of FOB is weight: a critical review of available information. World Health Stat 0 1980; 33(3): about 1/500 - 11700 in North America and Europe.3 Detailed 197-224. 4. World Health Organisation. Social and Biological Effects on Perinatal Mortality Vol. 1 haplotype analysis of mutant apo B alleles from many FOB (Report on an international comparative study sponsored by the World Health individuals suggests that this is a founder-type mutation that Organisation). Geneva: WHO. 1978. 7 5. Alberman E. Why are stillbirth and neonatal mortality rates continuing to fall? Br J occurred only once on an ancestral chromosome. The Obslet Gynaecol 1985; 92: 559-564. frequency of such a mutation in different populations is likely 6. World Health Organisation. World Health Statistics Annual. Vol. 1. Geneva: WHO, 1977. to be governed by founder effects and random genetic drift and should vary greatly. Indeed this mutation has not been Accepted 5 Apr 1994.

Medical Research Council/UCT Research Unit for the Cell Biology of Atherosclerosis, Department of Medical Biochemistry, University of Cape Town

David C. Rubinsztein. M.B. CH.B.• PH.D.

Gerhard A. Coetzee. PHD.

Deneys R. van der Westhuyzen. PHD.

Department of Human Genetics, University of Stellenbosch, Tygerberg, W. Cape

Elzet Langenhoven. M.SC.

Maritha J. Kotze. PHD.

SAMJ VO/lIme 85 No.5 May 1995 4 detected in Finns," Japanese9or Israelis.'o Consequently, in English and Afrikaner descent. ,22 It was impossible to . order to determine the impact of this mutation in South determine the geographical or genetic origin of the mutation Africa, its frequency was determined in . in these two cases. hypercholesterolaemic Afrikaners, coloureds and Indians Table I. Summary of South African lipid clinic patients screened with or without xanthomata. for FOB' Definite MH* 'Type 2a'11

t 20§ ­ Methods Afrikaners 23 Colouredst 8 8 All patients were unrelated and attended lipid clinics in Cape Indianst 6 16 Town, Durban and Johannesburg. Individuals were • All patients listed in this table were negative for FOB. . considered Afrikaner, coloured or Indian only if both parents t All Afrikaner and coloured patients were screened for the FH Afrikaner-l, -2 and -3 founder mutations prior to analysis for FDB. Coloured patIents were also screened _ belonged to that population group. The so-called coloured for FH Cape Town-2. Indian patients were screened for the FH Zambia mutation. . population of South Africa are of mixed racial descent and All patients positive for any of these LDL-receptor mutations were excluded from thiS have genes of indigenous Africans and of European and analysis. *Definite MH (monogenic hypercholesterolaemia) patients had total cholesterol Indonesian settlers. The Indian population is descended > 7,5 mmol/l or LDL-eholesterol > 5,2 mmol/l together with xanthomata or thickened from Indian immigrants who arrived in South Africa between Achilles tendons in themselves or a first-degree relative.. Type 2a patients had total cholesterol> 7,5 mmol/l or LDL-eholesterol > 5,2 ~mol/l 1860 and 1911. Blood was taken after informed consent had and triglycerides < 2,3 mmol/l without xanthomata in themselves Or a first-deg~ been obtained, and with ethical approval by the appropriate relative. institution. § Includes 13 patients screened by Graadt van Roggen et al." DNA isolation from blood was carried out as described by Talmud et al." FDB was screened for as described by The frequencies of FDB and FH in North America and Hansen et al. 12 and Tybjaerg-Hansen et al. 13 The FH Afrikaner-1, FH Afrikaner-3 and FH Cape Town-2 mutations Europe are both about 1/500,3-23 and the two diseases are were assayed as described previously."'" The FH Afrikaner-2 clinically almost indistinguishable in terms of the presence of mutation was detected by means of an amplification xanthomata.'" In such populations, one would expect refractory mutation system (ARMS) polymerase chain therefore to find equal numbers of FDB and FH alleles in reaction assay." All Indian patients were screened for the FH hypercholesterolaemic individuals with xanthomata from which FH patients with common or founder LDL-receptor Zambia mutation, as described by Rubinsztein et al. 17 mutations were excluded. Since we detected no such patients with FOB, this disorder is probably not as common in Afrikaners, coloureds and Indians as sporadic FH. Also, Results and discussion because no FOB was detected in patients with type 2A hyperlipidaemia without xanthomata, it is likely that the Hypercholesterolaemic South African Afrikaner, coloured and frequency of FOB in these populations is less than 1/500. Indian patients were selected for screening for FDB after The possibility exists that FOB does occur with a higher individuals who had anyone of the common or founder frequency in these groups, but does not cause LDL-receptor mutations known to be associated with these hypercholesterolaemia or xanthomatosis uniquely. We groups'4,'5,17.'9 had been eliminated (Table I). These individuals consider this possibility unlikely. were classified as having either definite monogenic Since FDB is a founder-type mutation,' its frequency is hypercholesterolaemia or type 2A hyperlipidaemia. likely to vary between different populations. To our individuals were considered to have definite monogenic knowledge, FDB has not been detected yet among Finns," hypercholesterolaemia, Le. either FH or FDB, if they had Japanese9or Israelis'o and no studies have been reported for total cholesterol concentrations greater than 7,5 mmol/I or Indians or black Africans. It is possible that this mutation LDL-cholesterol concentrations greater than 5,2 mmol/I, might have occurred after the events that gave rise to the together with xanthomata or thickened tendons in different racial groups. The absence of FDB in the Afrikaners themselves or in a first-degree relative. Cholesterol and coloureds could have arisen from a 'negative' founder­ concentrations above these limits are compatible with FH.'o effect that diluted the frequency of this gene in the settlers Three founder LDL-receptor mutations, FH Afrikaner-1, -2 relative to their parent European populations. We conclude and -3, account for apprOXimately 90% of the FH in that FOB is a rare cause of hypercholesterolaemia in these 14 Afrikaners. .2' These three mutations and the FH Cape South African populations. Town-2 mutation have been found in the coloured population (D. J. van der Westhuyzen and M. J. Kotze­ The South African Medical Research Council, Old Mutual, the unpublished data). Therefore, all patients with these Universities of Cape Town and Stellenbosch and the Stella and mutations were eliminated from the study. The only mutation Paul Loewenstein Trust are thanked for financial support. detected so far in more than one South African Indian family Ms S. Jones and Ms H. Karjiker provided excellent technical is FH-Zambia. 17 Similarly, all Indian patients with this assistance. Ors A. O. Marais, G. M. Berger, M. C. Rajput, mutation were not considered. None of the patients listed in F. Raal, H. Settel and K. Steyn are thanked for providing blood Table I was found to have FDB. Therefore, all patients with and clinical information on patients from their lipid clinics. definite monogenic hypercholesterolaemia are likely to have Or P. Talmud is thanked for providing FOB-positive control rare sporadic LDL-receptor mutations. FOB has been samples. detected in two unrelated families in South Africa of mixed

Volume 85 No.5 May 1995 SAMJ • -- SAMJ

ARTICLES

REFERENCES

1. lnnerarity TL. Mahtey RW, Weisgraber KH, et al. Familial defective apolipoprotein Recurrent LDL-receptor 8-100: a mutation of apolipoprotein B that causes hypercholesterolemia. J Lipid Res 1990; 31: 1337-1349. 2. Rauh G, Keller C, Karmann B, et al. Familial defective apolipoprotein 8-100: mutation causes familial clinical characteristics of 54 cases. Atherosclerosis 1992; 92: 233-241. 3. Tybjaerg-Hansen A, Humphries SE. Familial defective apolipoprotein 8-100: a single mutation that causes hyperchoJesteroJemia and premature coronary heart hypercholesterolaemia in disease. Atherosclerosis 1992; 96: 91-107. 4. Rubinsztein DC. Raal FJ, Settel He, Pilcher G, eoetzee GA, Van der Westhuyzen OR. Characterization of six patients who are double heterozygotes for familial South African coloureds hypercholesterolemia and familial defective apo B-l00. Arteriosclerosis and Thrombosis 1993; 13: 1076-1081- 5. Davignon J, Dufour R, Roy M, et aJ. Phenotypic heterogeneity associated with and Afrikaners defective apolipoprotein B-100 and occurrence of the familial hypercholesterolemia phenotype in the absence of an LDL-receptor defect within a Canadian kindred. EurJ Epidemio/1992; 8: 10-17. M. J. Kotze, E. Langenhoven, L. Theart, O. Loubser, 6. Marz W, Ruzicka C, Pohl T, Usadel KH, Gross W. Familial defective apolipoprotein B-l00: mild hypercholesterolemia without atherosclerosis in a A. Micklem, C. J. J. Oosthuizen homozygous patient. Lancet 1992; 340: 1362. 7. Ludwig EH, McCarthy BJ. Haplotype analysis of the human apolipoprotein B mutation associated with familial defective apolipoprotein B-l00. Am J Hum Genet 1990; 47: 712-720. Three low-density lipoprotein receptor (LDLR) gene 8. Hamalainen T, Palotie A, Aalto-Setala K, Kontula K, Tikkanen MJ. Absence of mutations were previously shown to cause familial familial defective apolipoprotein B-l00 in Finnish patients with elevated serum cholesterol. Atherosclerosis 1990; 82: 177-183. hypercholesterolaemia (FH) in up to 90% of affected 9. Hosking JL, Bais R, Roach PO, Thomas DW. Hypercholesterolemia due to familial defective apolipoprotein 8-100 in two Australian families (Correspondence). Med Afrikaners. Association of each mutation with a single J Aust 1991; 155: 572-573. 10. Friedlander Y. Dan EJ, Leitersdorf E. Absence of familial defective apolipoprotein chromosomal background provided molecular genetic B-l00 in Israeli patients with dominantly inherited hypercholesterolemia and in offspring with parental history of myocardial infarction. Hum Genet 1993; 91: evidence that the proposed 'founder gene effect' was 299-300. 11. Talmud P, Tybjaerg-Hansen A, Bhatnagar 0, et al. Rapid screening for specific responsible for the high prevalence of FH among white mutations in patients with clink::al diagnosis of familial hypercholesterolemia. Atherosclerosis 1991; 89: 137-141. Afrikaners. In this study we report the identification of the 12. Hansen PS, ROdiger N, Tybjaerg-Hansen A, Faergemann 0, Gregian N. Detection FH Afrikaner-2 (FH2) mutation, Val408 to Met, in the so­ of the apo-B-3500 mutation (glutamine for arginine) by gene amplification and cleavage with Msp 1- J Upid Res 1991; 32: 1229-i 233. called coloured population of South Africa, a people of 13. Tybjaerg-Hansen A. Gallagher J, Vincent J, et al. Familial defective apolipoprotein 8-100; detection in the United Kingdom and Scandinavia and clinical mixed ancestry, with rapid non-radioactive methods for characteristics of the cases. Atherosclerosis 1990; 80: 235-242. 14. Graadt van Roggen FJ, Van der Westhuyzen OR, Marais AD, Gevers W, Coetzee mutation detection. Haplotype analysis with GA. Low density lipoprotein founder mutations in Afrikaner familial hyperchalesterolemia patients: a comparison of two geographical areas. Hum polymorphisms on both sides of the FH2 mutation Genet 1991; 88: 204-208. 15. Henderson HE, Serger GM8, Marais AD. A new LDL receptor gene deletion in the indicated that the identical LDLR gene mutations found in South African population. Hum Genet 1988; 80: 371-374. 16. Kotze MJ, Langenhoven E, Theart L, Loubser 0, Micklem A, Oosthuizen CJJ. two different South African population groups were Recurrent LDL-receptor mutation causes familial hypercholesterolaemia in South caused by independent events at a potential CpG African coloureds and Afrikaners. S Atr Med J 1995; 85: 357-361 (this issue.) 17. Rubinsztein DC, Coetzee GA, Marais AD, Leitersdorf E, Settel HC, Van der mutational 'hot spot'. The allelic variation giving rise to the Westhuyzen OR. Identification and properties of the proline..-Ieucine mutant LDL receptor in South Africans of Indian origin. J Lipid Res 1992; 33: 1647-1655. different chromosomal backgrounds of the FH2 mutation 18. Leitersdorf E, Van der Westhuyzen OR, Coetzee GA, Hobbs HH. Two common low density lipoprotein gene mutations cause familial hypercholesterolemia in does not affect the properties of the abnormal LDLR Afrikaners. J Clin Invest 1989; 84: 954-961. 19. Kotze MJ, Langenhoven E, Warnich L, et al. The identification of two low-density protein product which causes FH in these subjects. This lipoprotein receptor gene mutations in South African hypercholesterolaemics. S Atr Med J 1989; 76: 399-401 . mutation is thus expected to cause the same severe form 20. Study Group, European Atherosclerosis Society. The recognition and of FH in affected coloureds as was previously management of hyperlipidaemia in adults: a policy statement of the European Atherosclerosis Society. Eur Heart J 1988; 9: 571 -600. demonstrated in Afrikaners. Detection of mutant LDLR 21. Kotze MJ, Langenhoven E, Warnich L, Du Plessis L, Retief AE. The molecular basis and diagnosis of familial hypercholesterolemia in South African Afrikaners. gene alleles in polymerase chain reaction products, Ann Hum Gene! 1991; 55: 115-121- 22. Rubinsztein DC. Monogenic hypercholesterolemia in Indians and familial directly after gel electrophoresis, now allows accurate defective apolipoprotein 8-100. PhD Thesis, University of Cape Town, 1993. 23. Goldstein JL, 8rown MS. Familial hypercholesterolemia. In: Scriver CR, Beaudet presymptomatic diagnosis of the FH2 mutation in FH AL, Sly WS. Valle 0, eds. The Metabolic Basis of Inherited Disease. 6th ed. New York: McGraw-HiII, 1989. patients from two different South African population

Accepted 18 Apr 1994. groups.

S Afr Med J 1995; 85: 357-361.

Department of Human Genetics, University of Stellenbosch, Tygerberg, W. Cape

M. J. Kotze. PHD.

E. Langenhoven, M.SC.

L Theart, M.SC.

O. Loubser, M.SC.

A. Micklem. B.SC. HONS

C, J. J. Oosthuizen, PH.D.

I SAMJ Volume 85 No. 5 May 1995 ~