Lower Incidence of M2/ANXA5 Carriage in Recurrent Pregnancy Loss Patients With Elevated Lipoprotein(a) Levels

Abstract

This study compared the incidence of M2/ANXA5 haplotype carriage, a documented repeated miscarriage risk factor, in patient groups with normal and elevated lipoprotein(a) (Lp(a)) levels. A total of 138 women with ≥2 consecutive, idiopathic recurrent miscarriages, categorized in patients with elevated (≥30 mg/dL, n = 44) and normal Lp(a) level (<30 mg/dL, n = 94) were recruited at the recurrent pregnancy loss (RPL) clinic of Munich Großhadern University Hospital. A total of 500 fertile women served as controls. All patients were genotyped for ANXA5 promoter haplotypes, genetic frequencies were compared, and odds ratios (ORs) and relative risks of M2 carriers were calculated. Women with M2 haplotype had an almost 2 times higher relative risk of RPL (OR 2.6, 95% confidence interval 1.5-4.6, P = .001) than fertile controls. Furthermore, risk rises to 2.47 in patients having normal Lp(a) levels (OR 3.2, 95% confidence interval 1.7-5.9, P = .001), whereas women with high Lp(a) levels exhibit notably lower apparent RPL risk of 1.39 (OR 1.4, 95% confidence interval 0.5-4.1, P = .659).

Keywords

annexin A5 recurrent miscarriage recurrent pregnancy loss lipoprotein(a)

Introduction

Thrombophilia predispositions are a recognized factor for miscarriages and placental obstetric complications. Significant share of these conditions are hereditary and occur due to genetic alterations in the availability or affinity of blood coagulation factors and auxiliary proteins involved in the thromboplastin complex.¹ Reduced expression of the ANXA5 gene product, potent placental anticoagulant, has been documented in series of obstetric pathologies, antiphospholipid syndrome,² preeclampsia (PE),^3,4 fetal growth restriction (FGR),^4,5 and nonpregnant women with previous recurrent pregnancy loss (RPL).⁶ A promoter haplotype of ANXA5 identified in 2007 was shown to drive down gene expression.^7,8 This haplotype-designated M2 has been documented as RPL predisposition factor in populations of Central Europe^7,9 and recently in the Japanese population.¹⁰ The M2 consists of four consecutive nucleotide exchanges in the proximal core promoter region of ANXA5 gene and a subset of this haplotype harboring only the 2 “central” SNPs, SNP2 and SNP3, and defined as M1 haplotype,⁷ does not appear to be associated with any obstetric complications from the previous studies. A recent study demonstrated fetal loss, FGR, and placental thrombosis in homo- and heterozygous animals of an AnxA5 loss of function murine model.¹¹

A direct interaction of ANXA5 with lipoprotein (a) (Lp(a)) leads to attenuated membrane binding of the former, upon physiologically relevant high Lp(a) concentrations of 10 to 40 mg/dL.¹² High Lp(a) is a well-established risk factor for coronary disease and stroke.^13,14 Pathologically elevated Lp(a) has been associated with PE,¹⁵ FGR,¹⁶ and RPL.^17,18

Since both proteins, ANXA5 and Lp(a), are involved in the formation of the thromboplastin complex and high circulating concentrations of Lp(a) would effectively reduce the anticoagulation potential of ANXA5, we aimed to evaluate the distributions of M2/ANXA5 carriers in patients with RPL having normal and elevated Lp(a) levels.

Patients and Methods

Study Population

The present study complied with the ethical guidelines of all institutions involved. Furthermore, it was approved by the review board of the Ludwig-Maximilians University of Munich (institutional review board), and informed consent was obtained from all patients examined.

A total of 138 women were prospectively recruited between June 2011 and July 2012 at the Recurrent Pregnancy Loss Clinic of the Division of Gynecological Endocrinology and Reproductive Medicine. All women were screened negative for any potential cause of their recurrent miscarriage as described previously.^19,20 Lipoprotein(a) levels of women were screened at least 6 weeks after the last pregnancy had ended. According to these measured levels, study patients were grouped in patients with elevated Lp(a) levels ≥30 mg/dL (subgroup 1) or normal Lp(a) levels <30 mg/dL (subgroup 2) according to the general definition.²¹

DNA was extracted from white blood cells, using QIAmp DNA blood mini kit (Qiagen Hilden, Germany) and stored in 100 µL aliquots at −20°C for further analyses.

A previously recruited control group consisted of 500 fertile women from the registry of the Institute of Human Genetics, UKM Muenster, and genetic backgrounds of patients and control groups were confirmed comparable, as previously described.⁷

Genotyping and Statistical Analysis

Extracted DNA was genotyped by direct amplicon Sanger sequencing as previously described.⁷ Differences of M2 carriage in women with elevated or normal Lp(a) levels were assessed using the 2-tailed Fisher exact test. Statistical significance was interpreted at odds ratios (ORs) of 2 (±0.25) and above, characteristic for the RPL populations genotyped so far.^7,9,10 The significance level of the analysis was set at P < .05.

Results

In all, 44 (32%) women had elevated (median: 60 mg/dL, min–max: 30-141 mg/dL) and 94 (68%) had normal Lp(a) levels (median 12 mg/dL, min-max: 1-29 mg/dL). Patient subgroups did not differ in terms of common clinical or specific obstetric history (Table 1).

Table 1.

Relevant Clinical Data of the Study Population.^a

	Subgroup 1 (n = 44)	Subgroup 2 (n = 94)
Age, mean (min–max)	32 (22-45)	35 (23-45)	ns
Pregnancies, n, mean (min–max)	3 (2-7)	3 (2-10)	ns
Deliveries n, mean (min–max)	0 (0-2)	0 (0-3)	ns
Miscarriages, n, mean (min–max)	3 (2-6)	3 (2-10)	ns
Early RPL, ≤12 weeks of gestation	37 (84%)	86 (91.5%)
Late RPL, 12.-20 weeks of gestation	0	1 (1%)
Early and late RPL	7 (16%)	7 (7.5%)
Primary RPL	28 (64%)	64 (68%)
Secondary RPL	16 (36%)	30 (32%)
Number of miscarriages
2	20 (45.5%)	39 (41.5%)
3	13 (29.5%)	37 (39.4%)
4	6 (13.6%)	7 (7.4%)
≥5	5 (11.4%)	11 (11.7%)

Abbreviations: n, number; RPL, recurrent pregnancy loss; ns, not significant; Lp(a), lipoprotein(a); WHO, World Health Organization.

^aSubgroup 1: women with RPL and high Lp(a) levels and subgroup 2: women with RPL and normal Lp(a) levels. Primary RPL: no infant born either dead or alive after the 20th completed week of gestation and/or weighing more than 500 g before the series of miscarriages (WHO).²² Secondary RPL: at least 1 infant born either dead or alive after the 20th completed week of gestation or weighing more than 500 g before the series of miscarriages (WHO).²²

The role of M2/ANXA5 as a risk factor for idiopathic RPL was verified, comparing the 138 women with recurrent miscarriages to the control group (Table 2). The study group slightly deviated from Hardy-Weinberg equilibrium (Markov chain Monte Carlo [MCMC] P = .1951), mainly due to an excess of homozygous carriers of both haplotypes. Allelic frequencies (AFs) of M2 were found twice as high in the study group (0.105) compared to the control group (0.051). Consequently, women with M2/ANXA5 appeared to have a 1.98 higher risk of RPL than noncarriers, compared to fertile controls (OR 2.6, 95% confidence interval 1.5-4.6, P = .001).

Table 2.

Genotype Frequencies of ANXA5 Gene Promoter Haplotypes in the Patient Group With Subgroups and in the Fertile Controls Group.

Genotype	RPL Patient Group		Subgroup 1		Subgroup 2		Fertile Controls
Genotype	O	E^a	O	E	O	E	O	E
N/N	93 (67.4%)	91.6 (66.4%)	32 (72.7%)	30.2 (68.6%)	61 (64.9%)	61.4 (65.3%)	356 (71.2%)	343.6 (68.7%)
N/M1	17 (12.3%)	18 (13%)	5 (11.4%)	7.6 (17.3%)	12 (12.8%)	10.6 (11.3%)	87 (17.4%)	99.5 (20%)
M1/M1	2 (1.4%)	0.9 (0.7%)	2 (4.5%)	0.4 (0.9%)	0	0.4 (0.4%)	16 (3.2%)	7.2 (1.3%)
N/M2, M1/M2 ^b	23 (16.7%)	26 (18.8%)	4 (9.1%)	5.6 (12.7%)	19 (20.2%)	20.3 (21.6%)	31 (6.2%)	48.4 (10%)
M2/M2	3 (2.2%)	1.5 (1.1%)	1 (2.3%)	0.2 (0.5%)	2 (2.1%)	1.3 (1.4%)	10 (2%)	1.4 (0%)
Total	138	138	44	44	94	94	500	500

Abbreviations: RPL, recurrent pregnancy loss; Lp(a), lipoprotein(a); O, observed.

^aE, genotype frequency expected at Hardy-Weinberg equilibrium, calculated with the Genepop package.

^bGenotype M1/M2 was observed only in 1 patient and 1 fertile control participant. Subgroup 1: women with RPL and high Lp(a) levels and subgroup 2: women with RPL and normal Lp(a) levels. Genotype M1/M2 was observed only in 1 patient with normal Lp(a).

Analyzing the distribution of M2 carriers in women with elevated (1) versus normal (2) Lp(a) levels, we found subgroup 1 was not in Hardy-Weinberg equilibrium (MCMC P = .0231), whereas patients with normal Lp(a) of subgroup 2 were in Hardy-Weinberg equilibrium for ANXA5 haplotypes (MCMC P = .8504).

Allelic frequency of M2 in elevated Lp(a) level women was 0.068, being close to the AF determined in controls (0.051) with resulting relative risk of 1.39 (OR 1.4, 95% confidence interval 0.5-4.1, P = .659) compared to fertile women. In contrast, those with normal Lp(a) levels had a relative risk of 2.47 for RPL (OR 3.2, 95% confidence interval 1.7-5.9, P = .000) with an M2 AF of 0.122 (Table 2). There is a greater apparent risk of 1.24 for M2 carriers of the second subgroup when performing intragroup comparison of patients with RPL (OR 2.2, 95% confidence interval 0.8-6.4, P = .059, Fisher's exact test). The subgroup analysis of women with RPL and elevated versus normal Lp(a) levels indicates that M2 carriers are substantially rarer in the subgroup of patients with higher Lp(a).

Discussion

There was an enrichment trend of M2 carriers in the patient subgroup with normal Lp(a) levels. In contrast, among patients with RPL having elevated Lp(a), M2 carriers seemed underrepresented, with genetic frequency very close to this of the fertile controls and of the general population.⁷ This underrepresentation could be hardly evaluated as statistical trend because of the low number of patients with RPL having elevated Lp(a) levels. On the other hand, the distribution of M2 carriers in this particular subgroup is in strong deviation of Hardy-Weinberg equilibrium that reflects upon the whole patient group. In contrast, the M2 distribution in the second subgroup, with normal Lp(a), is in almost perfect Hardy-Weinberg equilibrium. The observed underrepresentation of M2 heterozygotes and overrepresentation of M2 homozygotes in subgroup 1 would lead to the conclusion of a negative ascertainment bias if generally the homozygotes count in both subgroups was not that small. For a pilot study, no statistically justified conclusion can be made on such effect at present because of the small sample size, although the genetic distribution even at this scale is evidently correct. In a study patient population, one would expect distribution of M2 carriers that is similar to the fertile controls (positive ascertainment bias under the “fertility” criterion) if a competitive genetic interaction model is in play for this particular group, meaning negative ascertainment bias due to a second risk factor (sic, high Lp(a)). We believe this to be indeed the case for this particular subgroup, and our hypothesis is supported by biochemical evidence of Lp(a)/ANXA5 direct interaction in physiologically relevant concentrations,¹² which effectively reduces the ANXA5 anticoagulation potential. This would in consequence yield a prothrombotic phenotype. This initial finding has to be confirmed in larger patient groups, in order to gain on statistical significance and possibly be instrumental in the diagnostic workup of patients with RPL.

Footnotes

Acknowledgments

The authors would like to thank Ursula Antkowiak for the competent technical assistance. The authors acknowledge the continuous support of IZKF Muenster.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Bogdanova

Markoff

. Hereditary thrombophilic risk factors for recurrent pregnancy loss. J Community Genet. 2010;1(2):47–53.

Rand

Guller

. Reduction of annexin-V (placental anticoagulant protein-I) on placental villi of women with antiphospholipid antibodies and recurrent spontaneous abortion. Am J Obstet Gynecol. 1994;171(6):1566–1572.

Shu

Sugimura

Kanayama

Kobayashi

Terao

. Immunohistochemical study of annexin V expression in placentae of preeclampsia. Gynecol Obstet Invest. 2000;49(1):17–23.

Chinni

Tiscia

Colaizzo

Vergura

Margaglione

Grandone

. Annexin V expression in human placenta is influenced by the carriership of the common haplotype M2. Fertil Steril. 2009;91(3):940–942.

Sifakis

Soufla

Koukoura

. Decreased annexin A5 mRNA placental expression in pregnancies complicated by fetal growth restriction. Thromb Res. 2010;125(4):326–331.

Rand

Arslan

. Reduction of circulating annexin A5 levels and resistance to annexin A5 anticoagulant activity in women with recurrent spontaneous pregnancy losses. Am J Obstet Gynecol. 2006;194(1):182–188.

Bogdanova

Horst

Chlystun

. A common haplotype of the annexin A5 (ANXA5) gene promoter is associated with recurrent pregnancy loss. Hum Mol Genet. 2007;16(5):573–578.

Markoff

Gerdes

Feldner

Bogdanova

Gerke

Grandone

. Reduced allele specific annexin A5 mRNA levels in placentas carrying the M2/ANXA5 allele. Placenta. 2010;31(10):937–940.

Tiscia

Colaizzo

Chinni

. Haplotype M2 in the annexin A5 (ANXA5) gene and the occurrence of obstetric complications. Thromb Haemost. 2009;102(2):309–313.

10.

Miyamura

Nishizawa

Ota

. Polymorphisms in the annexin A5 gene promoter in Japanese women with recurrent pregnancy loss. Mol Hum Reprod. 2011;17(7):447–452.

11.

Ueki

Mizushina

Laoharatchatathanin

. Loss of maternal annexin A5 increases the likelihood of placental platelet thrombosis and foetal loss. Sci Rep. 2012;2:827.

12.

Yang

Chen

. Effect of lipoprotein (a) on annexin A5 binding to cell membrane. Clin Chim Acta. 2010;411(23-24):1915–1919.

13.

Stein

Rosenson

. Lipoprotein Lp(a) excess and coronary heart disease. Arch Intern Med. 1997;157(11):1170–1176.

14.

Hobbs

White

. Lipoprotein(a): intrigues and insights. Curr Opin Lipidol. 1999;10(3):225–236.

15.

Van Pampus

Koopmann

Wolf

Büller

Prins

van den Ende

. Lipoprotein(a) concentrations in women with a history of severe preeclampsia. Thromb Haemost. 1999;82(1):10–13.

16.

Berg

Roald

Sande

. High Lp(a) lipoprotein level in maternal serum may interfere with placental circulation and cause fetal growth retardation. Clin Genet. 1994;46(1 spec no):52–56.

17.

Szczepański

Bauer

Gardas

Duchiński

. Antiphospholipid antibodies and lipoprotein (a) in women with recurrent fetal loss. Int J Gynaecol Obstet. 1998;61(1):39–44.

18.

Nowak-Göttl

Sonntag

Junker

Cirkel

von Eckardstein

. Evaluation of lipoprotein(a) and genetic prothrombotic risk factors in patients with recurrent foetal loss. Thromb Haemost. 2000;83(2):350–351.

19.

Rogenhofer

Ochsenkühn

Von Schönfeldt

Assef

Thaler

. Anti-trophoblast antibodies associated with recurrent miscarriages. Fertil Steril. 2012;97(2):361–366.

20.

Makris

Tomsu

Tuckerman

Laird

. Recurrent miscarriage: aetiology, management and prognosis. Hum Reprod Update. 2002;8(5):601–612.

21.

Enas

Chacko

Senthilkumar

Puthumana

Mohan

. Elevated lipoprotein(a)—a genetic factor for premature vascular disease in people with and without standard risk factors: a review. Dis Mon. 2006;52(1):5–50.

22.

Rowe

Comhaire

Hargreave

Mellows

, eds. WHO Manual for the Standardized Investigation and Diagnosis of the Infertile Couple. World Health Organization, Cambridge University Press: Cambridge; 1993.