GENOMICS l&1093-1096 (1991)

SHORT COMMUNICATION

Evidence That Descendants of Three Founders Constitute about 25% of Hemophilia B in the United States

RHETT P. KETTERLING,* CYNTHIA D. K. BOTTEMA,* JOHN A. PHILLIPS Ill,t AND STEVE 5. SOMMER*-’

*Department of Biochemistry and Molecular Biology, Mayo Clinic/Foundation, Rochester, Minnesota 55905; and tDivision of Genetics, Department of Pediatrics, Vanderbilt University School of Medicine, Nashville, Tennessee 37232

Received January 2, 1991; revised March 20, 1991

In our sample of 160 consecutive Caucasian hemophil- iacs, I4 (9%) had a G --+ A transition at bp 10,430 that substitutes serine for glycine 60 in the first EGF domain of the factor IX molecule. Each of these hemophiliacs had clinically mild disease. Haplotype data and familial pedi- grees indicate that 12 of these hemophiliacs are likely to be related to a common ancestor. The 13th and 14th patients possess different haplotypes and thus represent indepen- dent origins of the mutation. In addition, we have screened these 160 hemophiliacs for the previously reported muta- tions resulting from founder effects at Des@’ + Thr and ThrsBB + Met. Herein we report an additional nine hemo- philiacs with the mutant Thr*” allele and five additional hemophiliacs with the mutant MetW allele. Haplotype data for these 14 hemophiliacs support the original founder ef- fect hypotheses for these mutations. In total, the above three mutations are found in 44 of the 160 seemingly unre- lated Caucasian hemophiliacs (28%). The sample includes patients from all regions of the United States and Ontario, Canada. Descendants of these three founders comprise ap- proximately two-thirds of the mlssense mutations found in our sample of Caucasian hemophiliacs with clinically mild diseased. Q 1991 Academic Press, Inc.

Hemophilia B is an X-linked disease characterized hy a dysfunctional factor IX protein which affects ap- proximately 1 in 30,000 males. Many mutations in patients with hemophilia B have been delineated (re- viewed in Giannelli et al., 1990). We have previously reported two common mutations in moderate/mild he- mophilia B within our United States and Canadian Caucasian population (Bottema et al., 1990; Ketter- ling et al, in press). One of these, isoleucine 397 + threonine (T + C transition at base 31,311), has been observed in at least 25 families in the United States,

1 To whom correspondence should be addressed.

Canada, and France. The haplotype is identical in the 20 families for which it has been determined, indicat- ing a common ancestor for most if not all the affected patients (Bottema et al., 1990; Thompson et aZ., 1990). The second common mutation is threonine 296 + methionine (C --) T transition at base 31,008). This mutation occurs at a CpG dinucleotide. Transitions at CpG dinucleotides are elevated 24-fold in the factor IX gene (Koeberl et aZ., 1990; Green et aZ., 1990). Nev- ertheless, large pedigrees and haplotype analysis strongly suggest a founder effect in the Amish and Protestant populations dating to at least 1789 (Ket- terling et al., in press).

We have elucidated a third founder effect in mild hemophilia B within our Caucasian population of pri- marily Northern and Western European descent. Utilizing the direct sequencing method of GAWTS (Koeberl et al., 1990), we previously delineated four transitions at a CpG dinucleotide that results in Gly@’ + Ser. Herein, using sequencing and PASA (PCR amplification of specific alleles; Sommer et aZ., 1989), we report the Serao mutation in an additional 10 hemophiliacs for a total of 14 of the 160 consecu- tive Caucasian patients that we have examined (9%). All 14 hemophiliacs exhibited factor IX coagulant ac- tivities from 5 to 22% and appeared clinically mild (criteria of Eyster et aZ., 1980). Factor IX therapy was required only for surgery, dental extractions, and ma- jor trauma. No spontaneous hemorrhages into joints or other tissues were observed. In cases in which ex- tensive pedigrees were available, the disease was fa- milial.

Glycine 60 is conserved in the related serine pro- teases (factors VII, X, and protein C) as well as in the three available factor IX sequences from cow, dog, and mouse. This conservation suggests that glycine 60 is important for factor IX function and we con- clude that the substitution by serine is the cause of disease in these hemophiliacs.

1093 o&w-7543/91 $3.00 Copyright 0 1991 by Academic Press, Inc.

All rights of reproduction in any form reserved.

1094 SHORT COMMUNICATION

TABLE 1

Patients with the Se@’ Mutation

Family Hinf I”

Haplotype

XmnI TaSI HhuI FIX:C from

referring center FIX:C* one lab

FIXzAg* one lab

Specific activity

Clinical severity

HB3 HB4 HB7 HB57 HB99 HB103 HB117 HB133 HB140 HB146 HB155 HB160 HB176 HB184

III III III III III III III III III III

II III III III

- - + 14% 19% 35% 0.54 Mild - - + 16% 17% 40% 0.43 Mild - - + 12% 17% 35% 0.49 Mild - - + 10% - - - Mild? - - + 17% - - - Mild - - + 12% 8% 32% 0.25 Mild - - + 5% - - - Mild? - - - 10% - - - Mild? - - + 15% - - - Mild? - - + 22% 10% 23% 0.43 Mild - - + 9% 12% 20% 0.60 Mild - - + 15% 5% 20% 0.25 Mild - - + 16% 17% - - Mild - - + 9% 7% - - Mild

a HB155 possessed the rare A,B type II Hinf I allele as described by Sarkar (Ref. (10)). At least four alleles exist in this repetitive sequence of alternating purines and pyrimidines. The alleles differ in the number of 26 bp (A) and 24 bp (B) repeats. Allele I has the sequence AB, allele II has A,B, allele III has AzBz , and allele IV has A,B, . Alleles I and III are the most common.

* Factor IX activity (FIXC) was determined by a one-stage clotting assay and factor IX antigen (FIX:Ag) was determined by an ELISA assay as previously described (Ref. (1)).

Haplotype analysis was performed for each hemo- philiac utilizing the Hinf I, XmnI, TuqI, and HhaI polymorphisms (Koeberl et al., 1990; Bottema et CL, 1990; Ketterling et aZ., in press). Twelve of the 14 he- mophiliacs had the haplotype [HinfI (+), XmnI (-), TaqI (-), and HhaI (+)] (see Table 1). Since this haplotype is present in only 9% of the Caucasian pop- ulation (Ketterling et al., in press), it is extremely un- likely that this mutation would recur independently in these 12 patients. Based on the population preva- lence of 9% for this haplotype, the probability that all 12 mutations would have arisen independently in this haplotype is miniscule (P < 10-l’). The probability of even three independent occurrences in this haplotype is also low (P < 0.01). Thus, it appears that a founder effect is reasonable for most, if not all, of the patients with this haplotype.

Two hemophiliacs had a different haplotype, indi- cating independent origins of these mutations. Pa- tient HB155 has a rare HinfI allele. Direct sequenc- ing revealed the type II (A,B) allele which is present in only 1% of the Caucasian population (Sarkar et al., 1991). The second patient, HB133, has a different haplotype at the HhaI allele (-). Thus, the Ser@’ mu- tation occurred independently at least three times in our patient population. This is not surprising since the mutation is at a CpG dinucleotide, a known muta- tional hotspot in the factor IX gene.

In addition to the 14 hemophiliacs listed above, the SerSO mutation has been previously reported by others

in the United States and Europe (Denton et al., 1988; Poort et al., 1989; Green et al., 1990). However, the lack of published haplotype data precludes an assess- ment of whether founder effect also contributes to these mutations.

Two other mutations are common in our sample of 160. In addition to the 10 Thr3” mutations previously reported by our laboratory (Bottema et al., 1990), we report 9 more bringing the total to 19/160 (12%). We also have detected 5 more patients with the Met= mutation, bringing the total to 11/160 (7%) (Ketter- ling et uZ.; in press). Haplotype analysis for these 14 newly reported hemophiliacs support the founder-ef- fect hypotheses for these mutations (Table 2).

When the SerGO, Thr3s7, and Met? common muta- tions are combined, they account for approximately 28% of all the mutations seen in our sample of Cauca- sian hemophiliacs of European descent. No other common mutations were found in this population. In a population-based sample [our first 60 patients (Koeberl et al., 1990)], an identical fraction of pa- tients (17/60 = 0.28) had one of these three mutations that results in mild disease. These data suggest that our current sample, while no longer population-based, is not biased against the ascertainment of individuals with mild disease.

Patients with the Se#O, Met’=, and Thtig7 muta- tions reside in at least 12, 9, and 12 states, respec- tively. The states represented include California, Alaska, Minnesota, Florida, Iowa, Vermont, Virginia,

SHORT COMMUNICATION 1095

TABLE 2

Common Factor IX Mutations in 160 Caucasian Hemophiliacs of European Descent

Mutation

GlyBO + Ser

Thp + Met

Iless + Thr

Probands with Mutation % of

Family Haplotype” haplotype at CpG? patients

HB3, 4, 7, 57, 99, 103, 117, 140, 146, 160, 176, 184 A 12 Yes 9% HB133 B 1 HB155 C 1

HB19, 21, 22, 84,89, 126, 129, 138, 148, 167 D 10 Yes 7% HB70 E 1

HBlO, 11, 12, 14, 16, 18, 34,40, 46, 50,63,67, 69, E 19 No 12%

93,94,113,119,123,125

Total 28%

o For each haplotype, the polymorphisms are listed in the following order: Hinf I, XtnnI, TaqI, and HhaI (see Table 1). Haplotype A = III, - -, +; B = III, -, -, -; C = II, -, -, +; D = I, -, -, -; E = I, -, -, +. >

Tennessee, Oklahoma, Illinois, North Dakota, Ohio, Indiana, Wisconsin, Georgia, Maryland, New York, Texas, and Pennsylvania. The SerGO allele also was found in a patient from Ontario, Canada, and the Thr3g7 allele was also found in patients from Ireland, Germany, and Ontario, Canada (Giannelli et oz., 1990). Thus, these mutations are widespread among Caucasians. However, they were not found in 23 non- Caucasian hemophiliacs that we have analyzed.

and Charles Sexauer, M.D., of the Children’s Hospital, Oklahoma City, Oklahoma; Betty Schmalz, R.N., Bonnie Koenig, R.N., and Virginia Michels, M.D., from the Mayo Hemophilia Center, Roch- ester, Minnesota; and Kim Maness and Melinda Cohen of Vander- bilt University, Nashville, Tennessee, for providing blood samples, pedigrees, and lab values. This work was supported by NIH HL39762.

What fraction of mild disease is due to Se$O, Met’%, or Thr3s7? Clinical criteria (Eyster et al., 1980) and factor IX coagulation activities performed at the Mayo Clinic indicate that all the individuals with the Ser60 or Metzw mutations have mild disease and that the individuals with the Thr?’ mutation have mild to borderline mild/moderate disease (Ketterling et aZ., in press; Bottema et al., 1990). However, factor IX coagu- lant activities from patient records were often inaccu- rate, erroneously suggesting severe or moderately severe disease. Of the patients for which clinical in- formation and/or factor IX coagulant activities per- formed in the same laboratory (Mayo Clinic) were available, 33% (39/118) had mild disease, a fraction close to that found in population-based studies (Eys- ter et at., 1980; Larsson et at., 1982). Of the 37 mild hemophiliacs with missense mutations, 24 (65%) have one of the three above mutations. Thus, a sub- stantial fraction of all mild hemophilia B in the United States is due to founder effect.

1.

2.

3.

4.

5.

ACKNOWLEDGMENTS 6.

We thank Gobinda Sarkar, Ph.D., Dwight D. Koeberl, M.D., Ph.D., Setsuko Ii, M.D., Tristi Muir, Amy Grozbach, Hong-Sup Yoon, M.D., and Chary1 Dutton for assistance with haplotype anal- ysis and Beth Plumhoff and Todd Daniels for factor IX coagulant and factor IX antigen assays. We also thank Mary Johnson for excellent secretarial assistance. We thank Joyce Shissler, R.N.,

7.

REFERENCES

Bcr-rrn~~, C. D. K., KOJZBERL, D. D., KE?TERLING, R. P., BOWIE, E. J. W., TAYLOR, S. A. M., LILLICRAP, D., SHAPIRO, A., GILCHRIST, G., AND SOMMER, S. S. (1990). A past muta- tion of isoleucine 337 is now a common cause of moderate/ mild hemophilia B. Br. J. Haematol. 75: 212-216.

DENTON, D. H., FOWLKES, D. M., LORD, S. T., AND REISNER, H. M. (1988). Hemophilia B Durham: A mutation in the first EGF-like domain of factor IX that is characterized by poly- merase chain reaction. Blood 72: 1407-1411.

EYSTER, M. E., LEWIS, J. H., SHAPIRO, S. S., GILL, F., KA- JANI, M., PRAGER, D., DJERASSI, I., RICE, S., LUSCH, C., AND KELLER, A. (1980). The Pennsylvania hemophilia program 1973-1978. Am. J. Hemutol. 9: 277-286.

GIANNELLI, F., GREEN, P. M., HIGH, K. A., LOZIER, J. N., LILLICRAP, D. P., LUDWIG, M., OLEK, K., REITSMA, P. H., GOOSSENS, M., YOSHIOKA, A., SOMMER, S., AND BROWNLEE, G. G. (1990). Haemophilia B: Database of point mutations and short additions and deletions. Nucleic Acids Res. 18: 4053-4059.

GREEN, D. M., MONTANDON, A. J., BENTLEY, D. R., LJIJNG, R., NILSSON, I. R., AND GIANNELLI, F. (1990). The incidence and distribution of CpG + TpG transitions in the coagula- tion factor IX gene. A fresh look at CpG mutational hotspots. Nucleic Acids Res. 18: 3227-3231.

KE~RLING, R. P., B~EMA, C. D. K., KOJZBERL, D. D., II, S., AND SOMMER, S. S. T= + M, a common mutation causing mild hemophilia B in the Amish and others: Evidence for founder effect, variability in factor IX coagulants and rapid carrier detection. Hum. Genet., in press.

KOEBERL, D. D., BOTTEMA, C. D. K., KE?TERLING, R. P., BRIDGE, P. J., LILLICRAP, D. P., AND SOMMER, S. S. (1990). Mutations causing hemophilia B: Direct estimate of the un-

1096 SHORT COMMUNICATION

derlying rates of spontaneous germ-line transitions, trans- versions, and deletions in a human gene. Am. J. Hum. Genet. 47: 202-217.

8. LARSSON, S. A., NILSSON, I. M., AND BLOMBACK, M. (1982). Current status of Swedish hemophiliacs. Acta Med. Stand. 212: 195-200.

9. POORT, S. R., BR&, E., BERTINA, R. M., AND REITSMA, P. H. (1989). A Dutch pedigree with mild hemophilia B with a missense mutation in the first EGF domain (factor IX adsn Niouwo,,,ta,). Nuckic Acids Res. 17: 5869.

10. SARKAR, G., PAYNTON, C., AND SOMMER, S. S. (1991). Seg-

ments containing alternatingpurine andpyrimidine dinucleo- tides: Patterns of polymorphism in humans and prevalence throughout phylogeny. Nucleic Acids Res. 19: 631-636.

11. SOMMER, S. S., CASSADY, J. D., SOBELL, J. L., ANDBO?TEMA, C. D. K. (1989). A novel method for detecting point muta- tions or polymorphisms and its application to population screening for carriers of phenylketonuria. Mayo Clin. Proc. 64: 1361-1372.

12. THOMPSON, A. R., BAIAJ, S. P., CKEN, S-H., AND IMAcGIL- LTVRAY, R. T. A. (1990). ‘Founder’ effect in different families with haemophilia B mutation. l&wet 2: 418.