Page 0: Page 1: Placenta xxx (2010) 1e6 Contents lists available at ScienceDirect Placenta journal homepage: www.elsevier.com/locate/placenta A sequence variation in the promoter of the placental alkaline phosphatase gene (ALPP) is associated with allele-specific expression in human term placenta L. Bellazi a, S. Germond a, C. Dupont a, I. Brun-Heath a, A. Taillandier b, P. De Mazancourt a, M.N. Dieudonné a, E. Mornet a, b, * a Unité de Pathologie Cellulaire et Génétique, UPRES-EA2493, Faculté de Médecine Paris Ile de France Ouest PRES Universud Paris, Université de Versailles-Saint Quentin en Yvelines, 78035 Versailles, France b Laboratoire SESEP, Centre Hospitalier de Versailles, Bâtiment EFS, 2 rue Jean-Louis Forain, 78150 Le Chesnay, France a r t i c l e i n f o Article history: Accepted 22 June 2010 Keywords: Placental alkaline phosphatase RNA expression Gene polymorphism Enzymatic activity Transcription factor a b s t r a c t Placental alkaline phosphatase (PLAP), encoded by the ALPP gene, is produced by the fetal side of the placenta. This enzyme displays strong genetic variability. Some of the variants were reported to be associated with pathology of pregnancy. We show here that the two most common ALPP allelic variants, Pl1 and Pl2, differ in mRNA expression level. This differential expression was independent of the parental origin and probably results from linkage disequilibrium with the sequence variation rs2014683G>A in the ALPP gene promoter that was shown to have allele-specific binding patterns to placental nuclear proteins. The possible role of this allelic-specific expression in placenta-related pathology is discussed. Ó 2010 Elsevier Ltd. All rights reserved. 1. Introduction Placental alkaline phosphatase (PLAP) is a member of a multigene family that is in human composed of four isoenzymes. Placental, germ-cell (GCAP) and intestinal (IAP) alkaline phosphatases are tissue-specific enzymes encoded by clustered genes located on 2q34-q37 whereas the tissue non-specific alkaline phosphatase (TNAP), a ubiquitous isoform, is controlled by a gene located on 1p36-1p34. These enzymes hydrolyze phosphate ester at alkaline pH. PLAP is produced by the fetal side of the placenta. It is anchored, in a homodimeric form [1], to apical and basal plasma membrane of syncytiotrophoblasts and, to a lesser extent, at the surface of cytotrophoblasts [2]. Its expression increases steadily throughout pregnancy until term [3,4]. PLAP is also found released in maternal serum from the second trimester of gestation [5]. Despite many studies suggesting a role in feto-maternal metabolism and placental development [6e8], the specific role of PLAP is still unclear. * Corresponding author. Laboratoire SESEP, Centre Hospitalier de Versailles, Bâtiment EFS, 2 rue Jean-Louis Forain, 78150 Le Chesnay, France. Tel.: þ33 1 39 63 80 13; fax: þ33 1 39 63 80 12. E-mail address: emornet@labosesep.org (E. Mornet). 0143-4004/$ e see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.placenta.2010.06.012 The ALPP gene, coding this enzyme, displays strong genetic variability [9]. Indeed, three alleles Pl1, Pl2 and Pl3, have been identified, giving rise to six common electrophoretic phenotypes [10] characterized by different enzyme activities [11,12]. Pl2 allele differs from Pl1 for one nucleotide substitution in the coding sequence c.692G>C (p.R209P) (rs1048988) and Pl3 differs from Pl1 for seven nucleotide substitutions in the coding sequence and three silent substitutions [13]. In addition, 18 rare variants have been reported, representing less than 2% of allele frequencies [14]. Among these rare variants the D allele, which ranges from 0.8% to 1.4% of the PLAP alleles [15,16], is associated with spontaneous abortions [17e19]. Moreover, quantitative variations of circulating alkaline phosphatase concentrations during pregnancy could be associated with premature birth [20e22], low birth weight [23] and pre-eclampsia [24]. However, the exact molecular mechanisms responsible for these biological effects are not elucidated and no mutation in the ALPP gene is known to be responsible for any monogenic disease, pregnancy related or not. The association between some of these PLAP variants and pathologies occurring during pregnancy could be caused either by variations of enzymatic properties of the variants or by variations in PLAP expression levels. Wennberg et al. [17] showed that properties of the rare D allozyme of PLAP are significantly different from those of the common PLAP allozymes encoded by Pl1 and Pl2 alleles, suggesting that it could explain the apparent negative selective pressure of the PLAP D allele. However no study was performed at Please cite this article in press as: Bellazi L, et al., A sequence variation in the promoter of the placental alkaline phosphatase gene (<ce:itali..., Placenta (2010), doi:10.1016/j.placenta.2010.06.012 Page 2: 2 L. Bellazi et al. / Placenta xxx (2010) 1e6 at room temperature with DIG-end-labeled probes (À384A: 50 -GGGTGCCAGAGCAGCAAAG-30 , À384G: 50 -GGGTGCCAGGGCAGCAAAG-30 , oct2A: 50 -GTACGGAGTATCCAGCTCCGTAGCATGCAAATCCTCTGG-30 ) according to the manufacturer’s instructions (DIG Gel Shift Kit, 2nd Generation, Roche). The resulting DNAeprotein complexes were separated from the unbound probe by electrophoresis on a native 6% polyacrylamide gel in 0.5Â Tris/borate/EDTA buffer. The specificity of DNAeprotein binding was confirmed in competition experiments which were carried out as described above with the addition of 200Â unlabeled specific À384A or G or nonspecific oct2A probes. Quantification was realized with the Bio-1D software (Vilber Lourmat, Marne la Vallée, France). All values were expressed as means Æ SEM of the five separate experiments and statistical analysis was performed using the paired t-test. 2.7. Enzymatic activity study A full-length Pl1 cDNA of the ALPP was obtained by RT-PCR and cloned in the pcDNA3.1 plasmid (Invitrogen) by standard molecular biology methods. Pl2 and D variants were obtained directly from the Pl1 cloned cDNA by site-directed mutagenesis. The mutated cDNA was fully sequenced to make sure that the target mutation (c.692G>C and c.1352G>A for Pl2 and D, respectively) was inserted without other mutations. Pl1, Pl2 or D plasmids were transiently transfected in COS-7 cells for 48 h as previously described [26]. The plasmid pcDNA3.1/His/LacZ containing the b-galactosidase gene was used as a positive control of transfection. Total alkaline phosphatase activity was measured with a COBAS Integra 800 automate (Roche) and was weighted with b-galactosidase activity. The results were expressed in percent of Pl1 cDNA activity taken as reference. For each variant, experiments were repeated at least four times independently. All values were expressed as means Æ SEM of the separate experiments and statistical analysis was performed using the one way ANOVA test. the allelic mRNA level. Here, we have measured the allele-specific mRNA expression of ALPP gene in term placentas of heterozygous fetuses for the common exonic single nucleotide polymorphism (SNP) c.692G>C that distinguish Pl1 and Pl3 (both G allele) from Pl2 (C allele). We observed a differential expression between the 2 alleles, showed that this differential expression was independent of the parental origin and probably results from linkage disequilibrium (LD) with sequence variation in the ALPP gene promoter. Method 2.1. Samples This study was approved by the local ethical committee (CCPPRB) and informed consent was obtained from each donor before clinical sampling. Third trimester human placentas (n ¼ 34, 38e41 weeks of gestation), maternal venous blood and fetal umbilical blood were obtained from unrelated patients (mean age: 32.1 Æ 4.2 years) with clinically normal single pregnancies. 2.2. Genotyping The ALPP gene was amplified using the forward primer 50 -GCTCATACTCCATGCCCA-30 and the reverse primer 50 -GTGGGAGTGGTCGGCAGTG-30 and the amplified DNA was digested with the Nco I enzyme (New England Biolabs). The c.692G>C substitution creates an additional Nco I restriction site that indicates the presence of the C allele. 2.3. Determination of allelic expression Purification of total RNA and reverse transcription were performed according to standard procedures. PCRs were performed using the forward primer 50 -FAMGGGCAACTTCCAGACCATT-30 and the reverse primer 50 -GATTCTTCCCGTCCAGCCT-30 with the annealing temperature at 58  C for a limited number of 28 cycles in order to be within the linear range of amplification. Because the primers used here cannot distinguish ALPP cDNA and its strongly homologous gene ALPPL coding the germcell alkaline phosphatase, the PCR product (386 bp) was co-digested with the Nco I and Sph I enzymes (New England Biolabs). The digestion with Sph I generates a 149bp labeled fragment in presence of ALPPL transcript whereas the digestion with Nco I allows to identify the G and C alleles of ALPP (c.692G>C), generating 386- and 315bp labeled fragment respectively. Quantification of the digested purified product was realized by GeneScan analysis with the GeneScan ROX-500 as internal size standard. Peak sizes were determined using the GeneScan Analysis 3.7 software on an ABI310 platform (Applied Biosystem). All values were expressed as means Æ SEM of the 12 separate experiments and statistical analysis was performed using the paired t-test. 2.4. Promoter sequencing and cloning PCR of genomic DNA was performed using the forward primer 50 -AGCAGGAGGTAGAAGGAACT-30 and the reverse primer 50 -CGCCCAGGAAGATGATGAG-30 with the annealing temperature at 57  C. Sequencing was performed using Big Dye Terminator chemistry (Applied Biosystems). The proximal promoter sequence (À899 to À1 bp) and a part of the first exon (þ1 to þ162) were cloned into the pGL3-basic vector (Promega) containing the luciferase reporter gene (Luc). This construct contains G at rs2014683. The ALPP(A) (À899/þ162)-Luc was obtained directly from the ALPP(G) (À899/þ162)-Luc construct by site-directed mutagenesis. The mutated fragment was fully sequenced to make sure that the target mutation (rs2014683G>A) was inserted without other mutations. 2.5. Luciferase reporter assay Cells of the human extravillous trophoblast cell line HTR-8/Svneo were plated and transfected in 12-well plates with a cell density of 1.5 Â 105 and transiently transfected by ALPP(G) (À899/þ162)-Luc, ALPP(A) (À899/þ162)-Luc or control pGL3 vectors using lipofectamine reagent (Invitrogen). The plasmid pcDNA3.1/His/LacZ containing the b-galactosidase gene was cotransfected with the ALPP(À899/þ162)Luc vector to monitor transfection efficiency. Luciferase assay was performed using the Luciferase Assay System (Promega) according to manufacturer’s instructions. For each construct, experiments were repeated five times independently. All values were expressed as means Æ SEM of the five separate experiments and statistical analysis was performed using the paired t-test. 2.6. Electrophoretic mobility shift assay (EMSA) Nuclear extracts were isolated from HTR-8/SVneo cells as described by Dieudonné et al. [25]. Ten micrograms nuclear protein extract was incubated for 15 min 3. Results 3.1. Expression study Firstly, we determined the genotype of thirty four fetuses from cord blood genomic DNA for the c.692G>C (rs1048988) polymorphism of the ALPP gene. We found 20 homozygotes for the G allele, 2 homozygotes for the C allele and 12 heterozygotes. The deduced frequencies of G and C alleles were 0.76 and 0.24, respectively. Secondly, the relative fetal G/C allele-specific expression, in the placental tissues corresponding to the 12 heterozygous fetuses, was measured by semi quantitative RT-PCR. G/G and C/C homozygous samples were used as digestion controls (data not shown). The results showed that each allele of the ALPP gene was expressed in term placenta but a differential allelic level of expression was observed (Fig. 1). In placenta samples 1e10, the C allele was significantly less expressed than the G allele, with a mean variation of approximately 30% (Fig. 1C). Two other samples, identified as number 11 and number 12, did not show significant difference in relative C/G expression levels (<20%). In sample 11, sequencing showed that the G allele corresponds to a Pl3 variant instead of Pl1 in other samples. We also verified that none of the G alleles corresponded to the rare D variant (data not shown). In case 12, although repeated experiments, we observed that the G allele was less expressed than the C allele. 3.2. Parental origin study As many genes are totally or partially imprinted in placenta, we investigated if the differential expression observed above was related to the parental origin or was allele-specific. In order to determine the parental origin of fetal alleles, we genotyped the c.692G>C polymorphism from maternal genomic DNA using the same method as previously described. The low expressed C allele was found to be from maternal origin in samples 1e5 and from paternal origin in samples 6e7. In other samples (8e12), the parental origin could not be determined because of absence of available maternal sample or because maternal and fetal genotypes were identical. Finally, our results show that the differential Please cite this article in press as: Bellazi L, et al., A sequence variation in the promoter of the placental alkaline phosphatase gene (<ce:itali..., Placenta (2010), doi:10.1016/j.placenta.2010.06.012 Page 3: L. Bellazi et al. / Placenta xxx (2010) 1e6 3 Fig. 1. Allele-specific mRNA expression of ALPP gene in term placenta from 12 subjects with heterozygous C/G genotype assayed by RFLP and GeneScan analysis as described in Materials and methods. An example of GeneScan analysis is given Fig. 1A (sample 9). In each case, the relative expression of C/G allele was calculated from GeneScan peak sizes with G þ C alleles (Fig. 1B) or G allele (Fig. 1C) as 100% to adjust the amount between samples. Bars means Æ SEM of the data. **p < 0.01, paired t-test. expression between alleles was independent of the parental origin, leading to the hypothesis that this differential expression was allele-specific and could be caused by cis DNA elements. 3.3. Promoter study In order to understand how the differential allelic expression of ALPP is regulated, we have investigated the functional SNP(s) in the promoter that could explain the difference in relative transcript abundance in heterozygotes. We focused on SNPs, referenced in the Genomatix Database (http://www.genomatix.de/), potentially affecting transcription factor binding sites in the ALPP promoter. Then, we genotyped fetal DNA for seven SNPs located upstream from the start transcription site (Fig. 2a). We observed that samples 1e10 showing significant differential allelic expression of ALPP were also heterozygous at the SNP rs2014683G>A polymorphism (Table 1) whereas samples 11 and 12 were homozygous. This suggests LD between this SNP and c.692G>C polymorphism. Fig. 2. 2a- Schematic representation of human ALPP proximal promoter. The position of SNPs potentially affecting transcription factors biding sites within this part of the gene is indicated. The arrow depicts the transcription initiation site (þ1). 2b- Expression of Luc reporter gene under the control of the ALPP proximal promoter with A or G at rs2014683. Plasmid constructs with the Luc gene under the control of the ALPP(G) (À899/þ162) or ALPP(A) (À899/þ162) promoter were tested for Luc activity by transfection into HTR-8/ SVneo cells. The relative Luc activity was calculated with ALPP(A) promoter used as 100%. These data were derived from triplicate assays, representative of 5 independent experiments. Bars means Æ SEM of the data. *p < 0.05, paired t-test. Please cite this article in press as: Bellazi L, et al., A sequence variation in the promoter of the placental alkaline phosphatase gene (<ce:itali..., Placenta (2010), doi:10.1016/j.placenta.2010.06.012 Page 4: 4 L. Bellazi et al. / Placenta xxx (2010) 1e6 Table 1 Genotypes of the 12 fetal blood samples at 7 SNPs in the ALPP gene The SNPs are ordered according to their position on the ALPP gene (see Fig. 2). Fetal sample 1 2 3 4 5 6 7 8 9 10 11 12 rs2741312 T>C C/C C/T C/C C/T C/T T/T C/C C/T C/T C/C C/C T/T rs3762524 T>C C/T C/T C/T T/T T/T T/T T/T C/T T/T T/T T/T T/T rs3762525 T>C T/T T/T T/T T/T T/T T/T T/T T/T T/T T/T T/C T/T rs3762526 A>C A/A A/A A/A A/A A/A A/A A/A A/A A/A A/A A/C A/A rs2014683 G>A A/G A/G A/G A/G A/G A/G A/G A/G A/G A/G G/G G/G rs61282047 T>C T/T T/T T/T T/T T/T T/T T/T T/T T/T T/T T/T T/T rs16828965 C>T C/C C/C C/C C/C C/C C/C C/C C/C C/C C/C C/C T/T Moreover, by comparison with maternal genotypes, the four samples that were informative for the determination of haplotypes (rs2014683-c.692) showed always GeC and AeG associations. For the other SNPs, no association was found between the distribution of genotypes and the differential expression level. To test the transcriptional effect of the rs2014683 SNP in placental cells, we created an ALPP promoter (À899/þ162)-luciferase fusion gene containing either A or G at rs2014683, and transiently transfected HTR-8/SVneo trophoblast cells. These cells, although non-tumorigenic and metastatic, exhibit phenotypic properties of extravillous cytotrophoblasts including the expression of placental alkaline phosphatase [27]. Results show that the ALPP(A) (À899/þ162)-Luc construct exhibits a significant higher transcriptional activity than the ALPP(G) (À899/þ162)-Luc construct (Fig. 2b). Beta-galactosidase activity measures indicate no difference in transfection efficiency (data not shown). To confirm the critical role of the polymorphism rs2014683 in regulation of ALPP expression, DNAeprotein interactions by EMSA were realized. Nuclear extracts from HTR-8/SVneo cells were probed for their ability to interact in vitro with synthetic À384A or G oligonucleotide sequences. As shown in Fig. 3A, À384A and À384G probes have different binding patterns to HTR-8/SVneo nuclear proteins. These DNAeprotein complexes disappear when these experiments were repeated in the presence of 200-fold excess of specific unlabeled oligonucleotide (Fig. 3A). Moreover, we observe a significant decrease in À384G DNA binding activity compared with À384A (35%) (Fig. 3B). These in vitro results suggest that the rs2014683 polymorphism could be responsible of the in vivo differential allelic expression observed in A/G heterozygotes. 3.4. In vitro activity study Earlier studies demonstrated that ALPP variants showed differential enzyme activities in placental extracts [11,12]. However, the published data were contrasting and the differential allelic expression that we showed here certainly modulated the activity results. Therefore, we measured the enzyme activity of Pl1, Pl2 and D variant in COS-7 cells transfected by plasmid constructs containing ALPP cDNA with a constitutive promoter. COS-7 cells do not express endogenous alkaline phosphatase. As shown in Fig. 4,the activities of Pl2 and D variant were significantly higher than Pl1 variant (4100 U/L for Pl1 vs 5300 U/L and 5000 U/L for Pl2 and D, respectively; p < 0.01). However, the difference of activity observed between the Pl2 and D variants was not statistically significant. Fig. 3. rs2014683G>A dependent differential DNA binding of the ALPP promoter in HTR-8/Svneo cells. Nuclear extracts were prepared and incubated with DIG-labeled probes as described in Materials and methods. A: The figure is representative of five independent experiments. B: Densitometric analysis of EMSA experiments. Results are means Æ SEM of five separate experiments. **p < 0.01. Fig. 4. In vitro alkaline phosphatase activity of Pl1, Pl2 and the rare D variant of ALPP gene after transfection in COS-7 cells. Alkaline phosphatase activity was normalized with b-galactosidase activity as described in Materials and methods. COS-7 cells only transfected with pcDNA3.1/His/LacZ were used as negative control of alkaline phosphatase activity. In each case, experiments were repeated at least four times independently. Bars means Æ SEM of the data. (a) Pl2 or D vs Pl1, (b) Pl2 vs D. **p < 0.01; ns: non-significant; one way ANOVA test. Please cite this article in press as: Bellazi L, et al., A sequence variation in the promoter of the placental alkaline phosphatase gene (<ce:itali..., Placenta (2010), doi:10.1016/j.placenta.2010.06.012 Page 5: L. Bellazi et al. / Placenta xxx (2010) 1e6 5 mRN A level rs 2014683 c.692 c.1352 PL AP activity D Pl1 Pl2 A A G G G G C A A Fig. 5. Hypothetic model for counterbalancing of mRNA expression and enzymatic activity in the two common variants Pl1 and Pl2 and the rare D variant of ALPP gene. Arrows indicate relative increase or decrease of expression or activity, depending of the alleles harbored at the rs204683 locus in the promoter, the c.692G>C locus distinguishing Pl1 and Pl2 variants, and the c.1352G>A locus distinguishing D variant from others. 4. Discussion Because PLAP is subject to high sequence variation and because some variants were reported to be associated with pathologies, it was of interest to study at the mRNA level the expression of the two most common ALPP allelic variants in human placenta. Interestingly we found that the Pl2 variant (C allele) is significantly less expressed than Pl1 (G allele) with a mean variation of approximately 30%, and that this difference of expression is probably due to a sequence variation in the ALPP promoter. Indeed our results suggest that sequence variation rs2014683 in the promoter, in LD with c.692G>C (p.R209P) acts by modulating the mRNA level of ALPP. According to the Genomatix database the A>G substitution seems to delete a binding site for the ubiquitous transcription factor p53, but supershift analysis using p53 antibody did not allow to confirm the implication of this transcription factor (not shown). The relatively low level of variation in allelic expression (approximately 30%) is however consistent with a sequence variation affecting no placental-specific transcription factors. Indeed, a variation affecting a transcription factor specific for placenta would be expected to produce much more variation. In addition, we observed that only the rs2014683 SNP was in LD with c.692G>C, while the other SNPs did not show any significant association. This could reflect a positive selection of the haplotype composed of A at rs2014683 and G at c.692 among the various possible haplotypes, corroborating the possible role of this SNP in the modulation of ALPP mRNA level. However, we cannot rule out that the mRNA level difference between the two alleles is due to another sequence variation of the 50 region, not yet characterized. Although repeated experiments, the mRNA levels of G allele at c.692 were lower than those of C allele in placental sample 12. Sequencing showed that, similarly to placental samples 1e10, this allele corresponded to the Pl1 variant. Interestingly, we noticed that the G allele in this placental tissue was associated with the G allele at rs2014683 instead of A in other placentas. Together with our EMSA experiments, this corroborates the hypothesis of the possible role of rs2014683 in the transcription level of ALPP. By contrast with the D variant for which kinetic studies were performed [17], it remains difficult to link the difference of expression between Pl1 and Pl2 alleles with pathologies such as spontaneous abortion, low birth weight or pre-eclampsia. Indeed, Pl1 and Pl2 are quite frequent in human populations (see the SNP database at http://www.ncbi.nlm.nih.gov/snp?term¼alpp), while D seems relatively rare, corroborating a possible negative selective pressure of this allele. Since Pl1 share with D the c.692G>C (p. R209P) variation, it would be interesting to determine if c.692G>C on D allele is also in LD with rs2014683G>A. Then, the pathologic effect of D variant could result from both its suboptimal enzymatic function due to c.1352A>G (p.E429G) and from the high mRNA level due to the promoter sequence variation. The high mRNA level could increase the negative effect of the E429G variation (Fig. 5). Corroborating this hypothesis, we found that the plasmid containing the variant D expressed more alkaline phosphatase activity than Pl1 when transfected in COS cells. This may be due to the effect of G429 that increases the overall accessibility to the active site when compared to E429 [17]. Similarly, P209 carried by Pl2, and located on the surface of the calcium binding site, could increase the enzymatic activity by modifying the regional polarity of this region. So, by contrast with Pl1 and Pl2 where transcriptional and enzymatic activities seem to be counterbalanced, the variant D shows both high mRNA level and high enzyme activity, in addition to the negative effect of the E429G substitution. This may account for the association of this variant with pathology. In conclusion we have shown here that Pl1 and Pl2 variants display differential mRNA expression that is linked to a sequence variation in the ALPP promoter and may account for associations with pregnancy related diseases. Acknowledgments The authors thank the staff of the maternity of the Centre Hospitalier Intercommunal of Poissy-Saint Germain for management of the patients and for providing the placentas and blood samples. The human extravillous trophoblast cell line HTR-8/Svneo was kindly provided by Dr Nadia Alfaidy (CEA Grenoble France) in agreement with Dr Charles Graham. References [1] Holmgren PA, Stigbrand T. Purification and partial characterization of two genetic variants of placental alkaline phosphatase. Biochem Genet 1976;14:777e89. [2] Leitner K, Szlauer R, Ellinger I, Ellinger A, Zimmer KP, Fuchs R. Placental alkaline phosphatase expression at the apical and basal plasma membrane in term villous trophoblasts. J Histochem Cytochem 2001;49:1155e64. [3] Fishman L, Miyayama H, Driscoll SG, Fishman WH. Developmental phasespecific alkaline phosphatase isoenzymes of human placenta and their occurrence in human cancer. Cancer Res 1976;36:2268e73. [4] Okamoto T, Seo H, Mano H, Furuhashi M, Goto S, Tomoda Y, et al. Expression of human placenta alkaline phosphatase in placenta during pregnancy. Placenta 1990;11:319e27. [5] Fishman WH, Ghosh NK, Inglis NR, Green S. Quantitation of the placental isoenzyme of alkaline phosphatase in pregnancy sera. Enzymologia 1968;34:317e21. [6] She QB, Mukherjee JJ, Chung T, Kiss Z. Placental alkaline phosphatase, insulin, and adenine nucleotides or adenosine synergistically promote long-term survival of serum-starved mouse embryo and human fetus fibroblasts. Cell Signal 2000;12:659e65. [7] She QB, Mukherjee JJ, Huang JS, Crilly KS, Kiss Z. Growth factor-like effects of placental alkaline phosphatase in human fetus and mouse embryo fibroblasts. FEBS Lett 2000;469:163e7. [8] Makiya R, Stigbrand T. Placental alkaline phosphatase as the placental IgG receptor. Clin Chem 1992;38:2543e5. [9] Robson EB, Harris H. Further studies on the genetics of placental alkaline phosphatase. Ann Hum Genet 1967;30:219e32. [10] Robson EB, Harris H. Genetics of the alkaline phosphatase polymorphism of the human placenta. Nature 1965;207:1257e9. Please cite this article in press as: Bellazi L, et al., A sequence variation in the promoter of the placental alkaline phosphatase gene (<ce:itali..., Placenta (2010), doi:10.1016/j.placenta.2010.06.012 Page 6: 6 L. Bellazi et al. / Placenta xxx (2010) 1e6 [20] Holmgren PA, Stigbrand T, Damber MG, von Schoultz B. Serum levels of placental alkaline phosphatase in high-risk pregnancies. Obstet Gynecol 1979;54:631e4. [21] Meyer RE, Thompson SJ, Addy CL, Garrison CZ, Best RG. Maternal serum placental alkaline phosphatase level and risk for preterm delivery. Am J Obstet Gynecol 1995;173:181e6. [22] Moawad AH, Goldenberg RL, Mercer B, Meis PJ, Iams JD, Das A, et al. The preterm prediction study: the value of serum alkaline phosphatase, alphafetoprotein, plasma corticotropin-releasing hormone, and other serum markers for the prediction of spontaneous preterm birth. Am J Obstet Gynecol 2002;186:990e6. [23] Brock DJ, Barron L. Measurement of placental alkaline phosphatase in maternal plasma as an indicator of subsequent low birthweight outcome. Br J Obstet Gynaecol 1988;95:79e83. [24] Fox H, Agrafojo-Blanco A. Scanning electron microscopy of the human placenta in normal and abnormal pregnancies. Eur J Obstet Gynecol Reprod Biol 1974;4:45e50. [25] Dieudonne MN, Leneveu MC, Giudicelli Y, Pecquery R. Evidence for functional estrogen receptors alpha and beta in human adipose cells: regional specificities and regulation by estrogens. Am J Physiol, Cell Physiol 2004;286: C655e61. [26] Fauvert D, Brun-Heath I, Lia-Baldini AS, Bellazi L, Taillandier A, Serre JL, et al. Mild forms of hypophosphatasia mostly result from dominant negative effect of severe alleles or from compound heterozygosity for severe and moderate alleles. BMC Med Genet 2009;10:51. [27] Graham CH, Hawley TS, Hawley RG, MacDougall JR, Kerbel RS, Khoo N, et al. Establishment and characterization of first trimester human trophoblast cells with extended lifespan. Exp Cell Res 1993;206:204e11. [11] Lucarelli P, Scacchi R, Corbo RM, Benincasa A, Palmarino R. Human placental alkaline phosphatase electrophoretic alleles: quantitative studies. Am J Hum Genet 1982;34:331e6. [12] Beckman G. Placental alkaline phosphatase, relation between phenotype and enzyme activity. Hum Hered 1970;20:74e80. [13] Henthorn PS, Knoll BJ, Raducha M, Rothblum KN, Slaughter C, Weiss M, et al. Products of two common alleles at the locus for human placental alkaline phosphatase differ by seven amino acids. Proc Natl Acad Sci USA 1986;83:5597e601. [14] Millan J. Mammalian alkaline phosphatases: from biology to applications in medicine and biotechnologies. Weinheim: Wiley-VCH; 2006. [15] Boyer SH. Alkaline phosphatase in human sera and placentae: starch gel electrophoresis reveals many phosphatase components including a polymorphism in placentae. Science 1961;134:1002e4. [16] Beckman G, Beckman L, Wennberg C, Sikstrom C, Millan JL. PstI restriction fragment length polymorphism of the human intestinal alkaline phosphatase gene. Hum Hered 1994;44:175e7. [17] Wennberg C, Kozlenkov A, Di Mauro S, Frohlander N, Beckman L, Hoylaerts MF, et al. Structure, genomic DNA typing, and kinetic characterization of the D allozyme of placental alkaline phosphatase (PLAP/ALPP). Hum Mutat 2002;19:258e67. [18] Wennberg C, Kivela A, Holmgren PA. Placental and germ cell alkaline phosphatase RFLPs and haplotypes associated with spontaneous abortion. Hum Hered 1995;45:272e7. [19] Beckman G, Beckman L, Magnusson SS. Placental alkaline phosphatase phenotypes and pre-natal selection. Evidence from studies of spontaneous and induced abortions. Hum Hered 1972;22:473e80. Please cite this article in press as: Bellazi L, et al., A sequence variation in the promoter of the placental alkaline phosphatase gene (<ce:itali..., Placenta (2010), doi:10.1016/j.placenta.2010.06.012 Page 7: