The involvement of the trans-generational effect in the high incidence of the hydatidiform mole in Africa

(1)

The involvement of the trans-generational effect in the high incidence of the hydatidiform mole in Africa

P. Coullin

^a^,^b^,^*

, A.L. Diatta

^c

, H. Boufettal

^d

, J. Feingold

^e

, E. Leguern

^e^,^f^,^g

, J.J. Candelier

^a^,^b^,^*

aINSERM U 972, H^opital P. Brousse, B^atiment Lavoisier, 14 Avenue P. V. Couturier, 94800 Villejuif, France

bUniversite Paris XI, Paris Sud, Orsay, France

cLaboratoire de cytogenetique et service d'obstetrique, CHU Le Dantec, Dakar, Senegal

dService de Gynecologie-Obstetrique, CHU Ibn Rochd, Casablanca, Morocco

eAP-HP, Departement de genetique et cytogenetique, Federation de Genetique, H^opital de la Pitie-Salp^etriere, F-75013 Paris, France

fINSERM, CRicm (U975), H^opital de la Pitie-Salp^etriere, F-75013 Paris, France

gUPMC Universite Paris 06, F-75005 Paris, France

a r t i c l e i n f o

Article history:

Accepted 31 October 2014

Keywords:

Placental development Hydatidiform mole Nutrition Oocyte Trophoblast

a b s t r a c t

Introduction: While the incidence of various chromosomal anomalies observed, including triploid partial moles is independent of the socio-economic level, higher incidences of complete hydatidiform mole

“CHM”is generally associated with under developed areas. Moreover, studies have shown that some nutritional deﬁciencies are related to the abnormal development of oocytes and placenta. In Senegal and Morocco, the annual seasonal cycle contains one period with food shortages and the incidence of complete moles is signiﬁcant. Accordingly, accurate statistical analyses have been performed in these two countries.

Methods:Each month during a one year period, we investigated the occurrence of normal conceptions, molar conceptions and the conception of the future patients in Senegal and Morocco. The comparisons of the conception dates for these three types of conception were analyzed using the Chi-squared test.

Results:94% of the patients were conceived just prior to the period in the year with food shortages.

Consequently, the development of the female embryos occurred under nutritional constraints, which negatively affect the recruitment of the vital factors required for the normal synthesis of DNA, proteins and placental differentiation.

Discussions:A nutritional deficiency in the mother at conception of their daughter (future patient) is implicated in the higher incidence of CHM in their daughters'filiation. These nutritional deficiencies during thefirst weeks of pregnancy will have repercussions on the normal development of the oocytes.

Accordingly, these developmental impairments take place during the embryonic life of the future mothers of complete moles and not during the conception of the moles themselves.

1. Introduction

The hydatidiform mole belongs to the pathologies of gestation in which the trophoblast proliferates in an anarchistic manner and where the chorionic villi become hydropic (Fig. 1). It can be associated with embryonic tissues (partial hydatidiform mole: PHM) or with a complete lack of these tissues (complete hydatidiform mole:

CHM). This abnormal proliferation of the trophoblastic cells could

be a result of nutritional deﬁciency, particularly vitamin A[1e3]

and the folates[4,5], an expression disorder of the variant hyper- glycosylated hCG[6], or a defect in the methylation of imprinted genes [7]. These hypotheses can be modulated by the age and ethnic origin of the patient[8,9]. From a genetic viewpoint, hydatidiform moles are the only pathologies that can be of androgenetic origin. However, their genotype is variable according to the kar- yotype of the oocyte and the mode of zygote formation during the fertilization. This zygote can be diploid biparental, diploid androgenetic monospermic, diploid androgenetic dispermic, triploid diandric dispermic, triploid digynic, tetraploid triandric, aneuploid or mosaic. The situation can be further complicated as we cannot assign the speciﬁc mole type to a corresponding genotype. But, the

*Corresponding authors. INSERM U 972, H^opital P. Brousse, B^atiment Lavoisier, 14 Avenue P. V. Couturier, 94800 Villejuif, France.

E-mail address:jean-jacques.candelier@u-psud.fr(J.J. Candelier).

Contents lists available atScienceDirect

Placenta

j o u r n a l h o m e p a g e :w w w . e l s e v ie r . c o m / l o c a t e / p l a c e n t a

Placenta 36 (2015) 48e51

(2)

great majority of CHM are diploid androgenetic, the other genotypes reported here are a minority of cases[10].

When the mole is formed, in particular if it is a CHM can become invasive or give rise to a gestational choriocarcinoma[11,12]. The frequency of these molar pathologies can vary from less than 1/

1000 pregnancies in western countries to over 1/400 in developing and undeveloped countries [13e15]. The risk of developing a choriocarcinoma after a complete mole is approximately 1000 times more likely than after any pregnancy event[16]. The development of these molar pathologies remains difﬁcult to understand.

Their progression is difﬁcult to evaluate, even after the evacuation of molar tissue[3,17].

These data and the fact that: (1) these CHM are sporadic and not recurrent, (2) in the literature some CHM contain small amounts of DNA of maternal origin[18], (3) some CHM arising from dispermic fertilization[9], strengthens our hypothesis leading us to consider that the starting point of these molar pathologies is not a genetic mutation but rather the inaptitude of the oocyte to undergo normal embryonic development.

We studied the monthly data from normal and molar conceptions as well as the dates of conception of women who will subsequently bear a complete mole (these women will be referred to here as the patient) in two areas, Senegal and Morocco. In these two countries we observed bioclimatic cycles with one period of transition during which food substances rich in certain vitamins were reduced and there was a high incidence of moles. Moreover, the women who will subsequently generate a CHM primarily developed their ovaries during this period of transition of the seasonal cycle (period of the year with signiﬁcant food shortages), whereas the normal and molar conceptions date were not associated with the seasonal cycle.

2. Materials and methods 2.1. Study groups

The patients that had a CHM were recruited in the hospitals in Dakar in Senegal and Casablanca in Morocco. These two countries present agricultural cycles associated with seasonal cycles. In all cases, echographic, anatomical and histological in- terpretations of uterine pathological tissues were carried out. Biochemistry of the bHCG was also interpreted. The data was obtained from the registers of these hospital departments, particularly by analysing the dates of the last menstruations of the patients. The most frequent age of these moles was three months, because the women presented several clinical signs at these dates (pains, bleedings) causing them to consult. We cannot be certain in these cohorts which CHM case was a CHM of androgenic origin. However, the literature shows that the presence of a CHM of another origin, in particular diploid bi-parental, is lower than 10%. Accordingly Jacobs and Couillin in Hawaii and Senegal had respectively observed 1/19 in each case [19,20]. To perform the statistical studies we investigated the calendar distribution of:

- normal conceptions - molar conceptions

- conception of the patients (their mothers pregnancy)

2.2. Data analysis (Fig. 2)

To compare the number of conceptions during the climatic and agricultural cycles it was necessary to collect the data of these conceptions each month over one year. Consequently, we were able to analyze the conceptions of patients between the years 1950e1990 for Senegal and the years 1968e1988 for Morocco. We studied 123 conception dates of the patients in Senegal and 34 in Morocco. These patients conceived 126 mol in 2004/2005 in Senegal and 34 mol were conceived in 2007 in Morocco. The normal reference (normal conception) corresponded to the two years where we obtained a large number of reliable data, the years 1981 for Senegal and 1985 for Morocco.

The comparison of the conception dates each month during one year for the normal fetuses, the hydatidiform moles and the patients were analyzed using the Chi-squared test and the conﬁdence intervals were deﬁned. As a control we studied the calendar distribution of the conception (between 2006 and 2008) of 219 patients (with future molar pregnancy) born in France between 1952 and 1989.

3. Results

The results are presented inFigs. 3 and 4. In Senegal (Fig. 3) we observed a signiﬁcant reduction in normal conceptions (curve A) during the period preceding the harvest period (food shortage period). The number of molar conceptions progresses at the same time as that of the normal conceptions (curve B). On the other hand, the period of the year during which the patients were conceived oscillates around a peak, which is present just before the food shortage period (curve C,c²¼20.66;p<0.05).

In Morocco (Fig. 4) the number of normal conceptions presents few variations during the year, except around May/June, which represents the beginning of the food shortage period, where a decrease was observed (curve A). The number of molar conceptions (curve B) does not present any differences during the year compared to the normal conceptions. Conversely 94% of the patients were conceived during the AprileJuly period with a peak Fig. 1.Second trimester CHM after blood removal. (1) Macroscopic observation: note the absence of embryo and the grape-like appearance of the chorionic villi (arrows).

(2) Histology withb-catenin staining of hydropic villi. Arrows show the cytotrophoblast (scale 50mm).

Fig. 2.Types and number of samples. Kinetics of the nutritional deficit related to oogenesis of the patients. In green, showing the food shortage effects on the embryonic development of the patients (girl) in the uterus of their mother, when these mothers undergo these nutritional difficulties during thefirst three months of their pregnancy.

P. Coullin et al. / Placenta 36 (2015) 48e51 49

(3)

during June (curve C,c²¼72.33;p<0.001). The conceptions of 219 French patients do not show any signiﬁcant differences associated with one season (data not shown).

4. Discussion

The various studies state that the androgenetic is the largely dominating etiology for CHM: in Hawaii[19], in Senegal[20], in

England[21]. Consequently, CHM of other origin are unlikely to be present in our panel (<10%).

Throughout the year, specific areas (with rural populations) in Senegal and Morocco are significantly affected by a problematic season of limited nutritional resources (food) where the sustenance of pregnant women from lower socio-economic classes is a significant public health problem. In these two countries the normal and molar conceptions depend only on the general incidence of conceptions dates during the year. Overall, it was found that women conceive less during the period of food shortages or right before such periods. This is particularly evident in Senegal (Fig. 3).

On the other hand, the conception dates of the patients are not randomly distributed. It was noticed that the patients were conceived in a speciﬁc time period when comparing the normal or molar conceptions. This period corresponds to the time points preceding the harvest in Senegal, and during the warm and dry season in Morocco, i.e. when there is a signiﬁcant shortage of food.

Accordingly, the development of the female embryo is carried out under nutritional constraints that could deteriorate the recruitment of the necessary elements for the synthesis of the DNA, the regulation of its expression and the proteins produced. The critical time point during pregnancy is located between 18 and 21 days after conception. During this period it is necessary to obtain nutritional precursors such as vitamin A, folic acid (vitamin B9) and histidine[4]. Moreover, the reproductive cells of the female embryo will begin their migration towards the genital ridges at approximately the 5th week of gestation and the differentiation of the oogonia will be complete at approximately the 12th week. The nutritional deﬁcit during theﬁrst three months of the pregnancy has been shown to have repercussions on the differentiation of the oogonia [22e24]. The food shortage period also affects the methylation of imprinted genes [22]. This incorrect imprinting could block the effective maturation of the oocyte and/or zygote, and could be responsible for the instability of the chromosomes of maternal origin[26].

This immaturity of the oocyte could be at the origin of (1) the damaged maternal pronucleus, which would result in an inefﬁcient initiation of cell division when compared to the paternal pronucleus and would involve its destruction by asynchronous division (2) an abnormal cortical reaction which would support the greatest frequency of triploidies by dispermy. These two hypotheses may explain (separately or collectively) the formation of CHM; either the fertilization of an empty oocyte by one or two spermatozoa (a historical model) or the dispermic fertilization followed by diploidization (a recent model[25]).

Furthermore, there have been various studies highlighting specific epidemiological observations that further support our conclusions. A study in Hawaii (defined as a weak risk region) shows that the migrant women with a high-risk of CHM were mainly those who were born in their country of origin (the Philippines, defined as a high risk region) and whose migration was recent. On the other hand, Japanese women (defined as a high risk region) whose migration to Hawaii was much older were no longer at risk[19]. This situation was found in post-war Japan, where between 1974 and 2000 the incidence of hydatidiform moles significantly decreased from 2.79/1000 live births in 1976 to 1.61 in 1997. The distinction between the CHM and the PHM showed that this regression is not related to the PHM (as the incidence remain constant) but to the CHM (the incidences decreased from 1.71/1000 live births in 1985 to 0.49 in 2000). This occurrence was simulta- neous with the improvement of socio-economic conditions and therefore female gametogenesis gradually normalized[27]. Partial and complete moles, which could seem to have a very similar molar phenotype, present significant differences in their epidemiology.

PHM (the majority of which possess the two parental genotypes) Fig. 3.Monthly conception rates (A: normal pregnancies, B: molar pregnancies, C:

patients). In Senegal: the number of normal conceptions per month in 1981 (12043 cases in all) are indicated by black dots (curve A; lefty-axis); the number of hydatidiform molar conceptions per month in the years 2004e2005 (126 cases in all) are indicated by black squares (curve B; righty-axis); the number of conceptions per month of patients with a molar pregnancy in the years 2004e2005, who were born between 1950 and 1990 (123 cases in all) are indicated by triangles (curve C; righty- axis); the harvest period (from the end of October to early February) and the food shortage period (from mid-June to end of October) are respectively indicated by a hatched and dotted tone rectangle.

Fig. 4.Monthly conception rates (A: normal pregnancies, B: molar pregnancies, C:

patients). In Morocco: the number of normal conceptions in 1985 (10576 cases in total) are indicated by black dots (curve A; lefty-axis); the number of conceptions per month of patients with a molar pregnancy in 2007, who were born between 1968 and 1988 (34 cases in total), are indicated by triangles (curve C; righty-axis); the conception rates of these moles are illustrated by curve B (righty-axis); the warm and dry period (from mid-May to mid-September) and the food shortage period (from June to October) are respectively indicated by a dotted tone rectangle.

(4)

are mainly the result of chronic anomalies of the gametogenesis/

fertilization process in the human species, while CHM (the majority of which have an androgenetic genotype) as our data suggests; are further dependent on the external environment.

Conﬂict of interest

The authors have no relevant interests to declare.

Authors' roles

P.C. declares that I conceived the idea of the study and its design, A.L.D., H.B. and P.C. collected the samplings, J.F. performed the statistical analyses, E.L. contributed intellectually to the study, J.J.C.

wrote the manuscript. We have seen and approved theﬁnal version.

Funding

This project was supported by the INSERM grant.

Acknowledgments

We acknowledge the anatomy and pathology laboratory of Bic^etre Hospital for their help in performing the histological cuts; in particular: Claire Soulier, Marie-Jo Redon and Martine Prsle. We are endebted to Zara Hannoun for her corrections of the English text.

References

[1] Craighill MC, Cramer DW. Epidemiology of complete molar pregnancy.

J Reprod Med 1984;29:784e7.

[2] Berkowitz RS, Cramer DW, Bernstein MR, Cassells S, Driscoll SG, Goldstein DP.

Risk factors for complete molar pregnancy from a case-control study. Am J Obstet Gynecol 1985;152:1016e20.

[3] Andrijono A, Muhilal M. Prevention of post-mole malignant trophoblastic disease with vitamin A. Asian Pac J Cancer Prev 2010;11:567e70.

[4] Reynolds SR. Hydatidiform mole: a vascular congenital anomaly. Obstet Gynecol 1976;47:244e50.

[5] Kokanali MK, Oztürkkan D, Unsal N, M€oroy P, Güng€or T, Mollamahmutoglu L.

Plasma homocysteine, vitamin B12 and folate levels in hydatidiform moles and histopathological subtypes. Arch Gynecol Obstet 2008;278:531e4.

[6] Cole LA. New discoveries on the biology and detection of human chorionic gonadotropin. Reprod Biol Endocrinol 2009;7(8):1e37.

[7] Judson H, Hayward BE, Sheridan E, Bonthron DT. A global disorder of imprinting in the human female germ line. Nature 2002;416:539e42.

[8] Neuber M, Rehder H, Zuther C, Lettau R, Schwinger E. Polyploidies in abortion material decrease with maternal age. Hum Genet 1993;91:563e6.

[9] Buckley JD. The epidemiology of molar pregnancy and choriocarcinoma. Clin Obstet Gynecol 1984;27:153e9.

[10] Slim R, Wallace EP. NLRP7 and the genetics of hydatidiform moles: recent advances and new challenges. Front Immunol 2013;4:1e12.

[11] Candelier JJ, Frappart L, Diatta AL, Yadaden T, Cisse ML, Afoutou JM, et al.

Differential expression of E-cadherin,b-catenin, and Lewis x between invasive hydatidiform moles and post-molar choriocarcinomas. Virchows Arch 2013;462:653e63.

[12] Candelier JJ, Frappart L, Yadaden T, Poaty H, Picard JY, Prevot S, et al. Altered p16 and Bcl-2 expression reﬂects pathologic development in hydatidiform moles and choriocarcinoma. Pathol Oncol Res 2013;19:217e27.

[13] Song HZ, Wu PC. Hydatidiform mole in China: a preliminary survey of incidence on more than three million women. Bull World Health Organ 1987;65:

507e11.

[14] Altieri A, Franceschi S, Ferlay J, Smith J, La Vecchia C. Epidemiology and aetiology of gestational trophoblastic diseases. Lancet Oncol 2003;4:670e8.

[15] Parazzini F, Ricci E, Cipriani S, Bulfoni G, Mangili G, Chiaffarino F. Temporal trends in the frequency of hydatidiform mole in Lombardy, northern Italy, 1996-2008. Int J Gynecol Cancer 2012;22:318e22.

[16] Lurain JR. Gestational trophoblastic disease I: epidemiology, pathology, clinical presentation and diagnosis of gestational trophoblastic disease, and management of hydatidiform mole. Am J Obstet Gynecol 2010;203:531e9.

[17] Sivanesaratnam V. Management of gestational trophoblastic disease in developing countries. Best Pract Res Clin Obstet Gynaecol 2003;17:925e42.

[18] Fisher RA, Nucci MR, Thaker HM, Weremowicz S, Genest DR, Castrillon DH.

Complete hydatidiform mole retaining a chromosome 11 of maternal origin:

molecular genetic analysis of a case. Mod Pathol 2004;17:1155e60.

[19] Jacobs PA, Hunt PA, Matsuura JS, Wilson CC, Szulman AE. Complete and partial hydatidiform mole in Hawaii: cytogenetics, morphology and epidemiology. Br J Obstet Gynaecol 1982;89:258e66.

[20] Couillin P, Afoutou JM, Faye O, Ravise N, Correa P, Boue A. Androgenetic origin of African complete hydatidiform moles demonstrated by HLA markers. Hum Genet 1985;71:113e6.

[21] Savage P, Williams J, Wong SL, Short D, Casalboni S, Catalano K, et al. The demographics of molar pregnancies in England and Wales from 2000-2009.

J Reprod Med 2010;55:341e5.

[22] Bowles J, Koopman P. Sex determination in mammalian germ cells: extrinsic versus intrinsic factors. Reproduction 2010;139:943e58.

[23] Clagett-Dame M, Knutson D. Vitamin A in reproduction and development.

Nutrients 2011;3:385e428.

[24] Yokobayashi S, Liang CY, Kohler H, Nestorov P, Liu Z, Vidal M, et al. PRC1 coordinates timing of sexual differentiation of female primordial germ cells.

Nature 2013;495:236e40.

[25] Golubovsky MD. Postzygotic diploidization of triploids as a source of unusual cases of mosaicism, chimerism and twinning. Hum Reprod 2003;18:236e42.

[26] Kawahara M, Wu Q, Takahashi N, Morita S, Yamada K, Ito M, et al. High-frequency generation of viable mice from engineered bi-maternal embryos. Nat Biotechnol 2007;25:1045e50.

[27] Matsui H, Iitsuka Y, Yamazawa K, Tanaka N, Seki K, Sekiya S. Changes in the incidence of molar pregnancies. A population-based study in Chiba prefecture and Japan between 1974 and 2000. Hum Reprod 2003;18:172e5.