Allele-specific hybridization of lipoprotein lipase and factor-V Leiden missense mutations with direct label alkaline phosphatase-conjugated oligonucleotide probes

Genetic Analysis: BiomolecularEngineering 13 (1996) 59-65

ELSEVIER

NENETIC ALYSIS BlomolecularEngineering

Allele-specific hybridization of lipoprotein lipase and factor-V Leiden missense mutations with direct label alkaline phosphatase-conjugated oligonucleotide probes Calvin P.H. Vary*, Marybeth Carmody, Renee LeBlanc, Tim Hayes, Clark Rundell, Len Keilson The Maine Medical Center Research Institute, 125 John Roberts Road, South Portland, M E 04106, USA

Received 31 January 1996; accepted 6 May 1996

Abstract

Direct label alkaline phosphatase (AP) conjugated oligonucleotide probes (AP-DNA) were prepared to assess their utility for allele-specific detection of si~agle base substitutions. Oligonucleotide conjugates were designed to detect point mutations in the genes for lipoprotein lipase (LPL) and coagulation factor-V (FV). Genomic DNA samples, including ones known to harbor point mutations in the genes for LPL and FV, were prepared from whole blood and subjected to polymerase chain reaction (PCR). PCR products were analyzed by Southern hybridization with the allele-specific AP-DNA probes and restriction endonuclease analysis. Thermal profiles for hybridization indicate optimal allele-specific selectivity was achieved with temperatures ranging from 45°C to 55°C at a total Na + concentration of 150 mM. Under these conditions the base changes studied were easily discriminated with allele specific hybridization signals in excess of 200:1 as estimated by scanning densitometry. Complete concordance was observed between hybridization and restriction analyses for 175 LPL and 201 FV clinical and reference samples. The total time for analysis of the PCR products was le,;s than 2 h with a dot blot hybridization protocol. Keywords: Direct label alkaline phosphatase conjugated oligonucleotide probes; Lipoprotein lipase; Coagulation factor-V; Point

mutations; Polymerase chain reaction

1. Introduction

Alkaline phosphatase labeled D N A probes (APDNA) offer significant advantages over other D N A probe systems [1-3]. They provide an alternative to the use of radioisotopically labeled D N A for identification of PCR products [4-6], detection of single copy genomic sequences [7], and detection of infectious disease agents [8-10] following Southern hybridization. APD N A probes may be used at concentrations sufficient to achieve complete hybridization in 5 - 3 0 min. The alkaline phosphatase label is easily detected using a

* Corresponding author. 1050-3862/96/$15.00 © 1996 Elsevier Science B.V. All rights reserved P H S1050-3862(95)00149-0

variety of colorimetric [1] and chemiluminescent substrates [2,7]. Finally, this class of reagents possesses a shelf life of years. A P - D N A probes have not, to our knowledge, been adequately characterized for determination of single point mutations. Simple oligonucleotides have been well documented for use in determination of point mutations characteristic of heritable genetic disorders [11-14] and viral drug resistance [15]. These assays have employed radioisotopic labels or, more recently, horseradish peroxidase [11] or electroluminescent [13] nonradioisotopic labels. Here we describe direct label alkaline phosphatase-conjugated oligonucleotide reagents for allele-specific determination of LPL and FV Leiden mutations following PCR.

60

C.P.H. Vary et al. / Genetic Analysis: Biomolecular Engineering 13 (1996) 59-65

1.1. Lipoprotein lipase deficiency

In the French-Canadian population of Quebec, LPL deficiency is prevalent; approximately 1 in 5000 are affected with the disorder [16] while as many as 1:40 are heterozygous [16 21]. Though common in individuals of French-Canadian descent, LPL mutations are also seen in other populations with different allele frequencies for the LPL 188 and 207 mutations [18]. LPL plays a primary role in triglyceride metabolism by hydrolyzing the core triglycerides of circulating chylomicrons and very low density lipoprotein. LPL also plays a role in lipoprotein catabolism [22]. Homozygous patients with LPL deficiency exhibit symptoms during childhood including abdominal pain, inflammation of the pancreas, enlargement of the liver and/or spleen, eruptive xanthomas, and lipemia retinalis [23]. Heterozygous patients are said to be asymptomatic, but have been found to have a form of familial hypertriglyceridemia, characterized by high triglyceride levels in conjunction with low cholesterol levels. Heterozygotes are thought to have a greater risk of coronary artery disease [23]. Research is currently being done involving heterozygotes because it is thought that heterozygotes with aberrations in insulin and/or estrogen levels develop symptoms much like those observed in homozygotes [23-25]. The LPL gene consists of 10 exons spanning approximately 30 kb at position p22 on chromosome 8. The LPL gene has been sequenced in its entirety and was found to be highly conserved among mammals [18]. Two major mutations, affecting LPL exon 5 codons 188 and 207, account for 95% of the mutant LPL alleles in the French-Canadian population [18,21]. Each of these mutations is a single base pair transition. Mutation 188 (M-188) is a substitution of glutamate for glycine ( G ~ A transition), while mutation 207 (M-207) is a substitution of leucine for proline (C ~ T transition). These mutations are not thought to directly involve the catalytic site, but they are thought to change the three-dimensional structure of the protein and therefore affect the stability, dimerization, apo CII activation and/or sub strate interaction of the protein [18,19]. 1.2. Factor V based familial thrombosis

Resistance to activated protein C (APC) appears to be the most prevalent cause of hereditary venous thrombosis [26]. The prevalence of APC resistance in the European population is estimated to be between 3% and 7% [27-31]. The incidence of APC resistance is as high as 30-50% in patients investigated for venous thrombosis [28-30]. A single missense mutation in the gene for coagulation Factor V has been found to be the major cause of APC resistance [27]. The mutation ( G ~ A transition) at nucleotide 1691 of the Factor V

gene causes the loss of an arginine residue and replacement with a glutamine residue at amino acid residue 506 [27]. This substitution alters the proteolytic cleavage target in Factor Va and results in APC resistant Factor Va. Restriction enzyme digestion of PCR products followed by gel electrophoresis [17] and, alternatively, allele-specific hybridization analysis of PCR products with radioisotopically labeled probes [19,21] have been used to detect LPL mutations in PCR products derived from clinical isolates. Similarly, nonradioisotopic detection of FV via biotin and streptavidin-alkaline phosphatase has been used [27] but requires more steps than direct label DNA probes. Due to the prevalence of these point mutations and their importance in health care management we sought to evaluate AP-DNA probes as simple, rapid, specific and nonradioisotopic reagents for determination of the carrier status of individuals thought to be at risk for disease as the result of these mutations.

2. Materials and methods 2.1. D N A probe conjugation

Oligonucleotides were conjugated with alkaline phosphatase using a procedure related to that of Jablonski et al. [1-3]. 3'-Aminoalkyl-derivatized oligonucleotides (100 nmol) in 0.01 M bicine buffer, pH 8.2 containing 0.5 M NaC1, were reacted with succinimidyl 4-(Nmaleimidomethyl)cyclohexane-l-carboxylate (1 gmol) for 10 min at room temperature. The maleimidoyloligonucleotide was purified using a NAP-10 column (Pharmacia Inc.) equilibrated with 1 M NaC1, 0.1 M phosphate buffer. Alkaline phosphatase (71 nmol, BoehringerMannheim) was reacted with N-succinimidyl 3-(2pyridyldithio)propionate (350 nmol, Sigma Inc., St Louis, MO) in 1 M NaC1, 0.1 M phosphate (pH 7.5) for 30 min at room temperature. Modified alkaline phosphatase preparations were desalted using a NAP10 column (Pharmacia Inc., Upsala, Sweden) equilibrated with 1 M NaC1, 0.1 M phosphate buffer (pH 7.5). Following reduction of the pyridyldithiol linkage with dithiothreitol (500 nmol), the thiolated alkaline phosphatase was desalted on a NAP-25 column as described above. The maleimidoyl-oligonucleotide and the thiolated alkaline phosphatase were combined and allowed to react for 16 h at room temperature. Conjugates were blocked with 100 nmol N-ethylmaleimide for 30 min at room temperature. Uncoupled oligonucleotide was removed by extensive ultrafiltration of the conjugate in an Centricon-100 ultrafiltration device (Amicon Corp., Beverly, MA).

C.P.H. Vary et al. / Genetic Analysis: Biomolecular Engineering 13 (1996) 59-65

M

+

+

. . . .

blbl

b!'+

+

. . . .

61

bl bl

816

M +

+

-

bl bl

203G 163S 1018

- __

M

+

2036

- -

163S

- -

1018

- -

+

bl

bl

$16/506

S 16/S06

394 344 298

394

-__

344

298 220~ 200

220.

Fig. 1. AP-DNA probe analysis of the two most common LPL point mutations. LPL 188/207 PCR products were fractionated on a 6% polyacrylamide gel and the DNA visualized following staining with ethidium bromide. Top panel: Lane M, 1 kb sizing ladder. Lanes 2 and 3 represent samples that are heterozygous for the LPL188 mutation, and a double heterozygote for the LPL188/207 point mutations, respectively. Lanes 4-7, PCR product derived from a genomic DNA samples known to be negative for both LPL 188 and 207 point mutations. Lanes 8 and 9 represent PCR reagent blank samples. All samples were run in duplicate and, following electrotransfer, the membrane was divided in half (lower panel) and hybridized with the M-188 AP-DNA probe (left panel) and M-207 AP-DNA probe (right panel). 2.2. Detection o f L P L gene m u t a t i o n s

D N A was isolated from whole blood samples using previously described procedures [32]. D N A concentrations in samples were determined by ultra-violet absorbance at 260 nm. L P L exon 5 primer sequences were as follows: Sense primer; 5 ' - T T C C C T T T T A A G G C C T C G A T - 3 ' and anti:sense primer; 5'-AAGTCC T C T C T C T G C A A T C A C - 3 ' [19]. The P C R reactions contained 10 m M Tris-HC1, 50 m M KC1, 200 ~tM of each dNTPs, 0.8 pM of each primer, 1.5 m M MgCI2, 2.5 U Amplitaq D N A Polymerase (Perkin Elmer Inc., Norwalk, CT), and 200-5;00 ng of D N A template in a final volume of 50 ~1. P C P was conducted in a PE-9600 (Perkin Elmer Inc., Norwalk, CT) thermal cycler. Thermal cycling parameters were 94°C for 30 s, 50°C for 30 s, and 72°C for 30 s, for a total of 25 cycles. For Southern hybridization, amplification products were separated by electropho~resis on 6% nondenaturing polyacrylamide gels run at approximately 35 m A for about 1.5 h. Gels were stained with ethidium bromide and products were visualized under UV light. The P C R products on the gel were then electrotransferred to a

Sureblot T M (Oncor Inc., Gaithersburg, M D ) hybridization membrane at 80 V for 0.5 h. F o r dot blot analysis, 1 - 3 ~tl of P C R product was applied to the nylon hybridization membrane. All hybridization membranes were subsequently treated with 0.2 N N a O H for 5 min at room temperature. M-188 hybridization membranes were prehybridized at 50°C for 30 min in Blocking Buffer (0.2% I-Block Reagent, 1 × PBS, 0.5% SDS). L P L mutation 188 was detected with the M-188 probe, 5 ' - C A C C A G A G A G T C C C C T - A P . Hybridization membranes were hybridized with A P - D N A at a concentration of 0.1-1 ng/ml in Blocking Buffer solution for 30 min at 50°C. Following hybridization, membranes were washed three times with Washing Buffer (1 x PBS, 0.5% SDS) at the hybridization temperature of 50°C. After washing, the membranes were rinsed with D E A buffer (10 m M diethanolamine (DEA), and 1 m M MgC12 p H = 10) at r o o m temperature for 5 min. The membranes were then treated with 1:200 CSPD (Tropix Inc., Bedford, MA) in D E A buffer for 5 min at room temperature. The membranes were blotted to remove excess reagent, wrapped in Saran Wrap, and exposed to film for 5 - 6 0 min at r o o m temperature.

C.P.H. Vary et al. / Genetic Analysis: Biomolecular Engineering 13 (1996) 59-65

62

A M12

B 3 4567

C 1 2

3

4

5

6

7

1 2

3

4

5

6

7

516/50~ 298 Z20/20¢

Fig. 2. LPL hybridization stringency study. Panel A: LPL PCR products were fractionated on a 6% polyacrylamide gel as above, and the DNA visualized following staining with ethidium bromide. Panel A, the lanes are from the left: M, 1 kb ladder; Lane 1, PCR product derived from a genomic DNA sample known to be positive for LPL 188 point mutation. Lanes 2-5, PCR product derived from genomic DNA samples known to be negative for LPL 188 point mutation. Lanes 6 and 7, PCR reagent blanks. Panel B: hybridization results at 50°C with M-188 probe; Panel C: hybridization results at 45°C with M-188 probe.

LPL mutation 207 was detected with the M-207 AP-DNA probe, 5'-CATTTACCTGAATGGAG-AP [21]. M-207 membranes were prehybridized at 48°C with Blocking Buffer for 30 min. They were then hybridized and washed at 48°C. All other manipulations were as described above for the M-188 probe.

2.3. Detection of factor V-Leiden gene mutations PCR amplification and hybridization studies were conducted as described above. PCR primers and probe were identical to those described by Bertina et al. [27]. The sense primer sequence was 5'-GAGAGACATCGCCTCTGGGCTA. The antisense primer was 5'-TGTTATCACACTGGTGCTAA. The probe was the AP conjugate of 5'-TGGACAGGCGAGGAATACA. Hybridization and washes were conducted as above at 50°C. All other methods were as described for lipoprotein lipase (above).

2.4. Restriction analysis of LPL and FV PCR products Aliquots (15 ~tl) of the FV Leiden [27] and LPL 207 PCR reactions were digested overnight at 37°C with 5 U of Mnll (New England Biolabs, Beverly, MA). LPL 188 PCR products were digested as above with 4 U Avail [18]. The reactions were stopped by the addition of gel loading buffer. FV Leiden digests were separated on an 8% polyacrylamide gel and visualized with ethidium bromide. LPL digests were separated on 10% acrylamide gels and also visualized with ethidium bromide.

2.5. Scanning densitometry Image analysis was conducted using the PDI image analysis software (PDI Inc., Huntington Sta., NY). Images of Southern blots were captured using a flatbed densitometric scanner interfaced to a SUN SPARC T M workstation (Sun Microsystems, Mountain View, CA). Following background subtraction, the total optical density in a lane was measured and expressed as OD × mm 2.

3. Results

The successful use of AP-DNA conjugates requires careful control of the thermal conditions throughout the course of hybridization and washing of the Southern or dot blot membranes. The upper duplicate panels in Fig. 1 show PCR products corresponding to, from the left, a known 188 heterozygote, a known 188/207 double heterozygote followed by four known normal PCR products and two reagent negative controls. The lower left hand and right hand panels correspond to Southern hybridization of the duplicate membranes with the M-188 and M-207 AP-DNA probes, respectively. The selectivity of the probes for both the 188 and 207 mutations is apparent. Fig. 2 illustrates the loss of specificity with the M-188 probe incurred by decreasing the temperature from 50°C to 45°C. The probe for the FV Leiden mutation behaved similarly. Comparison of the pattern of hybridization in the right hand panel of Fig. 3 with the known positive and negative PCR products visualized in the left hand

C.P.H. Vary et aL / Genetic Analysis: Biomolecular Engineering 13 (1996) 59-65

M M 12

1

2

3

63

4

34 2036 1635 1.018

Slt61506 516/506

220 ~..__ 200

394 344 298

220,, 200

Fig. 3. PCR analysis of Factor V mutant and wild type samples. Left panel: Factor V Leiden positive and negative derived PCR products were fractionated on a 6% polyacrylamide gel and the DNA visualized following staining with ethidium bromide. Left panel, the lanes are, from the left: M, 1 kb ladder; Lane 1, PCR product derived from a genomic DNA sample known to be heterozygous for the FV point mutation. Lanes 2 and 3, PCR product derived fi'om genomic DNA samples known to be negative for the Factor V Leiden point mutation. Lane 4, PCR reagent blank. Right panel: Lane M, 1 kb ladder; Lane 1, known Factor V mutant. Lanes 2 and 3, wild type samples. Lane 4, reagent blank. Hybridization for the Factor V mutant probe was conducted at 51°C. panel shows that the FV :probe easily discriminates the FV positive from two negative samples. D o t blot hybridization of 20 known samples of known FV genotype, shown in Fig. 4, agreed completely with Southern hybridization and restriction analyses. This result indicates that the probe functions reliably and equivalently in the dot blot format. Autoradiographic images of Southern hybridization membranes, including those shown in Figs. 1-3, were subjected to scanning densitometry in order to determine the optical densities of the hybridization signals. The positive to negative ratios, as shown in Table 1, were all in excess of 200. This response indicates that the selectivity of the probes for the mutant alleles is robust. O f 175 individuals studied for possible L P L deficiency markers, depicted in Table 2, 10 were positive for M-188 mutation and two were positive for both M-207 and M-188 mutations. Restriction analysis confirmed all hybridization results. O f the 10 hybridization positive samples, seven (4.0%) were M-188 heterozygotes, two (1.1%) were M-188/207 double heterozygotes, and one was an M-188 homozygote. Analysis of 201 patients studied for the FV mutation, including a confirmation panel of 33 known samples, is shown in Table 3. Restriction analysis of all 42 hy-

bridization positive samples, including 17 positive and 16 negative samples from the confirmation panel, indicated complete concordance of hybridization results with restriction analysis. O f these, 39 were heterozygous, and three were homozygous for the factor V-Leiden mutation.

4. Discussion

In this study, we have described the use of A P - D N A probes to detect single base substitutions. The probes were direct enzyme-labeled oligonucleotides composed of calf intestinal AP covalently linked to the 3' terminus of a 16-18 base long oligonucleotide. A P - D N A probes have molecular masses of about 145 k D a and are approximately 97% protein by weight. These types of probes are known to hybridize well with target nucleic acids, and possess thermal annealing temperatures as much as 10°C below that of their unmodified oligonucleotide counterparts [1] but little else is known of their biochemical behavior. Given the demonstrated utility of A P - D N A probes in other applications, it was of interest to characterize their behavior and potential utility as allele specific probes. In addition to the examples presented here, we have applied these probes to mutations

64

C.P.H. Vary et al. / Genetic Analysis: Biomolecular Engineering 13 (1996) 59 65

tk

B

1

2

3

4

5

6

7

8

9

10

11

12

15

16

17

13 14

19

20

1

2

7

18

8

13 14

+

19

20

3

4

5

9

10

11

15

16 17

6

12

18

+

Fig. 4. Dot blot analysis of Factor V mutant and wild type samples. Twenty Factor V samples and an FV mutant positive control were analyzed by dot blot analysis as described in Section 2. Samples 2, 4, 6, 8, 9, 11, 17 and 18 were positive by Southern hybridization analysis and heterozygous for the FV mutation by restriction endonuclease analysis (data not shown). All other samples were negative by the same criteria. Panel A, hybridization and wash temperatures were 50°C. Panel B, hybridization and wash temperatures were 45°C. Table 1 Peak optical density measurements of autoradiograpic exposures Probe

Average peak O.D. positive

Average peak O.D. negative

Positive/negative

LPL 188 LPL 207 Factor V

0.833 (n = 6) 0.92 (n = 1) 1.06 (n = 6)

0.002 (n = 18) 0.001 (n = 3) 0.004 (n = 7)

417 920 265

in the p53 tumor suppressor gene, the DCC (deleted in colon cancer) gene, human P450 aromatase gene and to HIV-1 drug resistance mutations. In all cases the probes behaved as expected with synthetic targets or PCR products. In the case of the lipoprotein lipase mutations, we have, in a limited number of samples, found that the 188 LPL mutation appears to be predominant over the LPL 207 mutation in the Maine population. This is consistent with the majority of our samples being of other than French-Canadian descent. This observation Table 2 AP-DNA hybridization and restriction analysis for lipoprotein lipase mutations

is of significance for the application of this technology to the Maine State population and, therefore, will require further substantiation with a larger number of samples from many other areas of the State of Maine. Recent studies involving the FV mutation have suggested a relationship between activated protein c resistance and arterial thrombosis [33] and coronary artery disease [34]. The ability to rapidly assess both the LPL and FV genotype is likely to be of increasing importance as more information is acquired regarding the relationship of genotype to phenotypic expression. Table 3 AP-DNA hybridization and restriction analysis for factor V-Leiden ASO hybridization (n =

Restriction analysis

201) ASO hybridization (n = 175)

Positive Negative

10 (5.7%) 165 (94.3%)

Restriction analysis Normal 188/normal

188/207

188/188

7 (4.0%) ND

2 (1.1%) ND

1 (0.6%) ND

Positive

42 (20.9%)

0/42

Negative

159 (79.1%)

16/16

Heterozygous

Homozygous

39/42 (19.4%) 0/16

3/42 (1.5%) 0/16

C.P.H. Vary et al. / Genetic Analysis: Biomolecular Engineering 13 (1996) 59-65

The results of both the Southern and dot hybridization procedures outlined here have been found, in our hands, to be much easier to interpret than restriction enzyme digest results. The dot hybridization approach is much faster, easier and more cost effective than restriction analysis since it is not only nonradioisotopic, but a dot blot procedure eliminates the need for restriction enzymes, gel electrophoresis and gel transfer. Dot blot procedures for LPL and factor V mutations, performed exactly as described here for Southern hybridization, are sufficiently robust that they are routinely used for the :rapid determination of these mutations in our clinical laboratory.

Acknowledgements The authors are gratefully indebted to, and would like to thank the following for their generous help: Janet Bayleran, Ph.D. for DNA samples from individuals of Maine French-Canadian descent for an LPL negative control population; Paul Foster, M.D. of the Blood Center of Southeastern Wisconsin, Karl Voelkerding, M.D. of the University of Wisconsin and Marjorie Boyd, M.D. generously provided factor-V samples candidate blood samples. Richard Press, M.D., Ph.D. of the Oregon Health Science University, Portland OR, provided facto~r V samples and their restriction analytical status.

References [1] Jablonski E, Moomaw EW, Tullis RH, Ruth JL. Nucleic Acids Res 1986; 14:6115 6128. [2] Carmody MW, Vary CPH. BioTechniques 1994; 16: 1044-1050. [3] Reyes RA, Cockerell GL. Nucleic Acids Res 1993; 21(23): 5532-5533. [4] Mullis KB, Faloona FA. Methods Enzymol 1989; 155:335 350. [5] Mullis KB, Faloona FA, S,zharf SJ, Saiki RK, Horn GT, Erlich HA. Cold Spring Harbor gymp Quant Biol 1986; 51: 263-273. [6] Saiki RK, Gelfand DH, Stoffel S, et al. Science 1988; 239: 487-491. [7] Cate RL, Ehrenfels CW, Wysk M, et al. Genet Anal Tech Appl 1991; 8(3): 102-106. [8] Conway B, Adler KE, Be,zhtel LJ, Kaplan JC, Hirsch MS. J Acquir Immune Defic Syndr 1990; 3(11): 1059-1064.

65

[9] Panda CS, Riley LW, Kumari SN, Khanna KK, Prakash K. J Clin Microbiol 1990; 28(9): 2122-2124. [10] Olive DM, Johny M, Sethi SK. J Clin Microbiol 1990; 28(7): 1565 1569. [11] Saiki RK, Chang C, Levenson CH, et al. New Engl J Med 1988; 319(9): 537-541. [12] Hansen TS, Petersen NE, Iitia A, et al. Clin Chem 1995; 41(3): 413-418. [13] Iitia A, Mikola M, Gregersen N, et al. BioTechniques 1994; 17(3): 232-237. [14] Prior TW, Papp AC, Snyder PJ, et al. Clin Chem 1990; 36(12): 2113-2117. [15] Kaye S, Loveday C, Tedder RS. J Med Virol 1992; 37:241 246. [16] Gagne C, Brun LD, Julien P, Moorjani S, Lupien PJ. Can Med Assoc J 1989; 140: 405-410. [17] Auwerx JH, Babirak SP, Hokanson JE, Stahnke G, Will H, Deeb SS, Brunzell JD. Am J Hum Genet 1990; 46(3): 470-477. [18] Monsalve MV, Henderson H, Roederer G, Julien P, Deeb S, Kastelein JJP, Peritz L, Devlin R, Bruin T, Murthy MR, Cagne C, Davignon J, Lupien PJ, Brunzell JD, Hayden MR. J Clin Invest 1990; 86: 728-734. [19] Normand T, Bergeron J, Fernandez-Margallo T, Bharucha A, Ven Muthy MR, Julien P, Gagne C, Dionne C, De Braekeleer M, Ma R, Hayden MR, Lupien PJ. Hum Genet 1992; 89: 671 - 674. [20] Bergeron J, Normand T, Bharucha A, Ven Murthy MR, Jullen P, Gagne C, Dionne C, De Braekeleer M, Brun D, Hayden MR, Lupien PJ. Clin Genet 1992; 41: 206-210. [21] Ma Yuanhong, Henderson HE, Ven Murthy MR, Roederer C, Monsalve MV, Clarke LA, Normand R, Julien P, Gagne C, Lambert M, Davignon J, Lupien PJ, Brunzell J, Hayden MR. New Engl J Med 1991; 324: 1761-1766. [22] Santamarina-Fojo S, Dugi KA. Lipidology 1994; 5: 117-125. [23] Wilson DE, Emi M, lverius PH, Hata A, Wu LL, Hillas E, Williams RR, Lalouel JM. J Clin Invest 1990; 86: 735-750. [24] Brun LD, Gagne C, Rosseau C, Moorjani S, Lupien P. Cancer 1986; 57: 2123-2126. [25] Keilsen L, Vary CPH, Sprecher DL, Renfrew R. Ann Intern Med 1996, in press. [26] Zoller B, Dahlback B. Lancet 1994; 343: 1536-1538. [27] Bertina RM, Koeleman BPC, Koster T, Rosendaal FR, Dirven R J, de Ronde H, van der Velden PA, Reitsma PH. Nature 1994; 369: 64-67. [28] Svensson PJ, Dahlback, B. New Engl J Med 1994; 330:517 522. [29] Griffin JH, Evatt C, Wideman C, Fernande JA. Blood 1993; 82: 1989 1993. [30] Koster T, Rosendaal TR, de Ronde H, Brier E, Vandenbrouke JP, Bertina RM. Lancet 1993; 342: 1503-1506. [31] Faoni EM, Franchi F, Asti D, Sacchi E, Bernardi F, Manucci PM. Thromb Haemost 1993; 70: 1067-1071. [32] John SWM, Weitzner G, Rozen R, Scriver CR. Nucleic Acids Res 1991; 19: 408. [33] Ma DD, Aboud MR, Williams BG, Isbister JP. Aust New Zealand J Med 1995; 25(2): 151-154. [34] Marz W, Seydewitz H, Winkelmann B, Chen M, Nauck M, Witt 1. Lancet 1995; 345(8948): 526-527.