Aquaporin 1 gene deletion affects the amniotic fluid volume and composition as well as the expression of other aquaporin water channels in placenta and fetal membranes

Accepted Manuscript Aquaporin 1 gene deletion affects the amniotic fluid volume and composition as well as the expression of other aquaporin water channels in placenta and fetal membranes

Hui Luo, Ailan Xie, Ying Hua, Jing Wang, Yanyun Liu, Xueqiong Zhu PII: DOI: Reference:

S0009-8981(18)30161-X doi:10.1016/j.cca.2018.04.001 CCA 15135

To appear in:

Clinica Chimica Acta

Received date: Revised date: Accepted date:

15 March 2018 31 March 2018 3 April 2018

Please cite this article as: Hui Luo, Ailan Xie, Ying Hua, Jing Wang, Yanyun Liu, Xueqiong Zhu , Aquaporin 1 gene deletion affects the amniotic fluid volume and composition as well as the expression of other aquaporin water channels in placenta and fetal membranes. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Cca(2018), doi:10.1016/j.cca.2018.04.001

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Aquaporin 1 Gene Deletion Affects the Amniotic Fluid Volume and Composition as well as The Expression of Other Aquaporin Water Channels in Placenta and Fetal Membranes Hui Luo, Ailan Xie, Ying Hua, Jing Wang, Yanyun Liu, Xueqiong Zhu* Department of Obstetrics and Gynecology, the Second Affiliated Hospital of Wenzhou Medical University,

PT

Wenzhou 325027, Zhejiang Province, China.

RI

*Corresponding Author: The Second Affiliated Hospital of Wenzhou Medical University, Wenzhou,

AC C

EP T

ED

MA

NU

SC

Zhejiang Province, China; Tel.: 86-13906640759. Email addresses: [email protected].

ACCEPTED MANUSCRIPT Abstrac Objective To explore the role of aquaporin 1 (AQP1) in regulation of amniotic fluid volume and composition. To investigate the effects of AQP1 gene knockout on expression of other aquaporin water channels (AQP3,

PT

AQP8 and AQP9) in placentas and fetal membranes.

RI

Methods

SC

Mice were sacrificed at 9.5, 13.5 and 16.5 gestational day (GD). Amniotic fluid volume, osmolality and composition, fetal membranes, placental and fetal weights as well as placenta areas were recorded in AQP1

NU

homozygote conceptus group, heterozygote conceptus group and wild-type group, respectively. The

MA

expression of AQP1, AQP3, AQP8 and AQP9 mRNA and protein in placenta and fetal membranes were examined by quantitative Reverse Transcription-Polymerase Chain Reaction (RT-PCR) and Western

ED

blotting.

EP T

Results

AQP1 homozygote conceptus had a greater volume of amniotic fluid, lower osmolality and calcium concentration than their wild-type counterparts at 16.5 GD. There was no significant difference in

AC C

expression of AQP1, AQP3, AQP8 and AQP9 in placentas among three groups. While expression of AQP8 was increased at 13.5 and 16.5 GD in fetal membranes, the expression of AQP9 was significantly decreased in fetal membranes in AQP1 homozygote group compared with AQP1 heterozygote and wild-type groups. Conclusion AQP1 may play an important role in the homeostasis of maternal- fetal fluid at late gestation days. The mechanism of mutual compensation among AQPs gene needs further investigation.

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

PT

Keywords: Aquaporin; Amniotic fluid; Fetal membrane; Placenta.

ACCEPTED MANUSCRIPT 1. Introduction During pregnancy, normal amniotic fluid quality and quantity is critical for fetal growth and development [1]. Abnormalities in amniotic fluid volume (polyhydramnios or oligohydramnios) are associated with fetal developmental deficiencies and increased perinatal mortality and morbidity [2-4].

PT

Water flow across placenta and fetal membranes progressively increases throughout gestation. Amniotic fluid is largely a product of fetal urine and lung fluid. On the other hand, amniotic fluid is swallowed,

RI

reabsorbed into the fetal circulation, and ultimately transferred to maternal circulation. At the end of

SC

gestation, up to 400 mL/day is transferred from the amniotic cavity [5]. As described below, transfer by

NU

fetal membrane provides another pathway regulating the amniotic fluid volume [6]. Both amniotic membrane reabsorption and placental transfer of water from fetal to maternal

MA

circulation [3] are mediated by transmembrane water channels which are formed by aquaporin proteins (AQPs) [7]. In mammals, at least thirteen AQPs (AQP 0-12) have been identified in different tissues.

ED

Specifically, as reported previously by our and other laboratories, expression of four AQPs (AQP 1, 3, 8, 9)

EP T

have been detected in placentas and fetal membranes at third gestation stage [2-3, 7-8]. Several studies explored the relationship between the expression of AQPs in human fetal membranes

AC C

and amniotic fluid volume disorders. Mann et al. [9] found that Aqp1 knockout mice had a greater volume of amniotic fluid and lower osmolality than wild-type and heterozygote controls and speculated that Aqp1 deficiency might be a cause of polyhydramnios. However, in human, AQP1 expression was found to be substantially increased in the amnion in pregnancies complicated by idiopathic polyhydramnios [10]. It was postulated that the upregulation of AQP1 might be a compensatory response in the process of idiopathic polyhydramnios. In a previous study [11], we found that the expression of AQP8 and AQP9 was significantly increased in fetal membranes in the idiopathic polyhydramnios group. Intriguingly, their expression in the placenta was significantly decreased. Hence, the expression and exact role(s) of different

ACCEPTED MANUSCRIPT AQPs in placentas and fetal membranes in the regulation of water flux remain unclear. The Aqp gene knockout mice provided a useful model for studying the physiological role(s) and functional compensation of AQPs. In order to determine how Aqp1 gene deletion affects other Aqp genes’ expression, Li et al. [12] used this model and quantitative Reverse Transcription-Polymerase Chain

PT

Reaction (RT-PCR) method to measure the expression of other AQPs in mouse gallbladder. They found that

RI

the expression of AQP8 mRNA was increased mildly in gallbladder of Aqp1 null mice compared to

SC

wild-type controls. Ma et al. [13] found that Aqp3 null mice had a greater fluid consumption and lower urine osmolality than wild-type mice in the renal cortex, and Aqp3 gene deletion led to a decreased AQP2

NU

protein expression but had little effect on AQP1 or AQP4 protein expression in the same tissue. Expression

MA

change of AQPs in placenta and fetal membranes after AQP1 gene deletion has not been systematically determined. In this study, we bred the Aqp1 heterozygous (Aqp1+/-) mice to produce the Aqp1 homozygotes

ED

deletion (Aqp1-/-), heterozygotes (Aqp1+/-) and wild type (Aqp1+/+) fetuses to explore the relationship

EP T

between the AQP1 expression and amniotic fluid volume as well as composition changes at different gestation days. In addition, the alteration of AQPs (AQP1, AQP3, AQP8, AQP9) expression in mouse placentas and fetal membranes was also investigated to determine the effects of Aqp1 gene deletion on the

AC C

expression of other aquaporin water channels (AQP3, AQP8 and AQP9) in placentas and fetal membranes.

2. Materials and methods 2.1. Mouse model

Aqp1 heterozygous mice (Aqp1+/-) originally generated by Ma et al. [14] were provided by Professor Yuanlin Song from Zhongshan Affiliated Hospital of Fudan University [15]. The genotype of mice was verified by Polymerase Chain Reaction using three primers: 5’-ACT CAG TGG CTA ACA ACA AAC

ACCEPTED MANUSCRIPT AGG, 5’-AAG TCA ACC TCT GCT CAG CTG GG and 5’-CTC TAT GGC TTC TGA GGC GGA AAG. The null allele and wild-type allele generated PCR products of 400 bp and 500 bp, respectively. All mice were housed under standard lighting (12 h light and 12 h darkness) and temperature (23±1 ℃), with free access to a nutritionally balance diet and tap water. Experimental procedures were approved by the Animal

PT

Use Committee of Wenzhou Medical University following the Institutional Guidelines for the Care and

RI

Use of Laboratory Animals.

SC

Male and female Aqp1 heterozygous (Aqp1+/-) mice were mated, and copulation plug was observed at the gestational day 0.5 (0.5 GD). The mice delivered normally at term (19 GD to 21 GD). Fetuses were

NU

collected and divided into three subgroups based on their genotypes and examined at 9.5 GD, 13.5 GD and

MA

16.5 GD.

2.2. Measurements of amniotic fluid, embryo, fetal membrane and placenta weights and placental areas

ED

The pregnant mice were anesthetized and a cesarean section was performed at 9.5 GD, 13.5 GD and

EP T

16.5 GD, respectively. Individual gestational sacs were carefully removed from the uterus and weighted, and the fluid was drained into an Eppendorf tube and the volumes were measured. The embryo, fetal membrane and placenta were separated carefully and weighted respectively. The placental diameter was

AC C

recorded and the discoid area was calculated using the formula: S=π (1/2  diameter)2 . The tissue of each embryo was removed to detect genotype by PCR as mentioned above. Fetal membranes and placentas were rinsed thoroughly in phosphate-buffered saline to remove excess blood. Tissues were snap frozen and stored as quickly at -80 ℃ until use. 2.3. Determination of amniotic fluid osmolality and electrolytes Amniotic fluid was centrifuged for 2 min at 3,000 g and the supernatant was collected. Amniotic fluid osmolality was measured by cryoscopic method using the Automatic Freezing Point Osmometer (FM-8P)

ACCEPTED MANUSCRIPT machine. The Biuret method was used to detect total protein and glucose in amniotic fluid, and enzymic method was used to determine urea and creatinine concentrations. Electrolyte concentration (Na+, K +, Cl-) was determined by electrode method and Ca 2+ concentration was determined using o-cresolphthalein complexone on an Automatic Biochemical Analyzer (ADVIA2400).

PT

2.4. RNA isolation, classic and quantitative RT-PCR

RI

Total RNA was extracted from the placenta and fetal membrane using Trizol reagent according to the

SC

manufacturer’s instructions (Invitrogen, Carlsbad, CA, USA). RNA quality and quantity were determined with spectrophotometry at an absorbance of 260 nm (NanoDrop 2000, Thermo Scientific, Wilmington, DE,

NU

USA). Reverse transcription was performed in a 20 μL reaction with 1 μg total RNA using the quantitative

MA

RT-PCR kit (Thermo Scientific, Wilmington, DE, USA). Quantitative RT-PCR was performed using a Light cycler 480 (Roche, Meylan, France) and SYBR®Green Master Mix (TransStart TipTop, China).

ED

Multiplex PCR was performed in 10 μL reactions containing 0.5 μL of AQPs or GAPDH TaqMan primers (Table 1) and probes, 5 μL of TaqMan Universal Master Mix, 1 μL of the RT-PCR reaction product, and 3.5

EP T

μL of H2 O. The reaction was performed with initial denaturation at 95 C for 10 min, followed by 45 cycles of 95 C for 15 s and 60 °C for 1 min. Quantitative PCR was performed in triplicates. Comparative

AC C

threshold (Ct) cycle method was used for data analysis. The expression of AQPs was evaluated using arithmetic formula: Expression =2−△△Ct for relative changes, where △Ct = Ct (AQPs) – Ct (GAPDH) and △△Ct

= △Ct (heterozygous or homozygous) – △Ct (wild-type).

2.5. Western Blotting Placenta and fetal membrane tissues were homogenized in Radio Immunoprecipitation assay buffer containing protease inhibitor Phenylmethylsulfonyl Fluoride and centrifuged at 12,000 g for 10 min at 4 C. Insoluble debris were removed and protein concentration of supernatant was measured with

ACCEPTED MANUSCRIPT spectrophotometry. 50 μg of protein in gel loading buffer was denatured at 100 C for 10 min and loaded to each well and separated in 12% Sodium Dodecyl Sulfate-polyacrylamide gel electrophoresis. After electrophoresis, proteins were blotted onto Polyvinylidene Fluoride membranes. Antibodies (ɑ-Tubulin 1:2,000 (Beyotime Biotechnology, China); AQP1 1:1,000 (ab168387, Abcam, Cambridge, USA); AQP3

PT

1:500 (ab125219, Abcam, Cambridge, USA); AQP8 1:500 (SC-14984, Santa Cruz, CA, USA); and AQP9

RI

1:500 (ab105148, Abcam, Cambridge, USA), were diluted with tris-buffered saline. Membranes were

SC

incubated with primary antibody at 4℃ overnight. Peroxidase-conjugated secondary antibody was applied at 1:5,000 (Biosharp, Heifei, China) for 2 h at room temperature. After extensive washing, protein band

NU

visualization was carried out using an Enhanced Chemiluminescence kit. All experiments were performed

MA

in triplicates. 2.6. Statistical analyses

ED

Statistical analyses were conducted using the SPSS 19.0 software. Data with normal distribution was

EP T

analyzed with one-way analysis of variance (ANOVA) followed by least significant difference (LSD) post hoc tests. Otherwise, the non- normal distribution data was analyzed with Kruskal-Wallis rank test and an extension of the Mann-Whitney U test. P < 0.05 was considered statistically significant. All values were

3. Results

AC C

reported as the mean ± SD.

Genotype analysis showed a Mendelian distribution of embryo s. Ten litters including 18 Aqp1 knockout homozygotes, 44 heterozygotes and 22 wild-type fetuses were studied at 9.5 GD. Moreover, 21 homozygotes, 49 heterozygotes and 24 wild-type fetuses were collected from 10 litters at 13.5 GD, and 23 homozygotes, 34 heterozygotes and 15 wild-type fetuses were collected from 10 litters at 16.5 GD.

ACCEPTED MANUSCRIPT 3.1. Amniotic fluid volumes in three groups at different gestation stages It is difficult to measure amniotic fluid volume or dissect placenta in gestational sac at 9.5 GD. No significant difference was observed in the amount of amniotic fluid among the Aqp1 homozygote, heterozygote and wild-type groups at 13.5 GD (Table 2, P > 0.05). However, the amount of amniotic fluid

PT

was significantly increased in Aqp1 homozygote embryos than their wild-type counterparts at 16.5 GD

RI

(Table 3, P < 0.05).

SC

3.2. Amniotic fluid osmolality and composition

Amniotic fluid osmolality and concentrations of total protein, glucose, urea, creatinine, Na +, K +, Cl-,

NU

Ca2+ showed no significant difference among the three groups at 13.5 GD (Table 2, P > 0.05).

MA

At 16.5 GD, no significant difference was observed in the concentrations of total protein, urea, creatinine, glucose, Na+, K+ and Cl- among three groups (Table 3). However, Aqp1 homozygote embryos

ED

had lower osmolality in amniotic fluid than that wild-type group. The concentration of Ca2+ was lower in

EP T

homozygote and heterozygote groups than wild-type group (Table 3, P < 0.05). 3.3. Fetus, fetal membrane and placenta weights and placental areas Fetus, fetal membrane, and placenta weights and placental areas were increased from 13.5 GD to 16.5

AC C

GD. No significant difference in these parameters was observed among the three groups either at 13.5 GD or 16.5 GD (Table 2 and 3, P > 0.05). 3.4. Expression of AQPs mRNA and protein in the placenta at 13.5 GD and 16.5 GD The quantitative RT-PCR and Western blotting results confirmed that four members of the aquaporin family (AQP1, AQP3, AQP8 and AQP9) mRNA and protein were detectable in placentas of the three groups at 13.5 GD and 16.5 GD. The results are shown in Fig. 1 and Fig. 2. Compared with wild-type group, there was no significant change in APQ3, AQP8 and AQP9 mRNA

ACCEPTED MANUSCRIPT and protein levels in placenta in homozygotes and heterozygotes at 13.5 GD and 16.5 GD. In heterozygotes, since mouse placentas have the labyrinth structure composed of tangled maternal and fetal blood vessels [16], AQP1 expression is also detected in placentas as shown in Fig. 1A and Fig. 1C, Fig. 2A and Fig. 2C. There was no significant difference in AQP1 expression in placentas of the three groups.

PT

3.5. Expression of AQPs mRNA and protein in the fetal membrane at 13.5 GD and 16.5 GD

RI

Using quantitative RT-PCR and Western blotting, AQP3, AQP8 and AQP9 mRNA and protein

SC

expression in fetal membranes were determined in the three groups. Consistent with the genotype, no expression of AQP1 mRNA and protein expression was observed in Aqp1 gene deletion homozygotes.

NU

There was no difference in AQP3 mRNA and protein expression in fetal membranes among three groups at

MA

13.5 GD and 16.5 GD. However, AQP9 mRNA and protein expression in fetal membranes was lower in Aqp1 homozygote group than heterozygote and wild-type groups at 13.5 GD and 16.5 GD. In contrary, the

ED

expression of AQP8 in fetal membrane in Aqp1 deletion homozygote group was higher than that in

4. Discussion

EP T

heterozygote and wild-type groups (Fig. 1B and Fig. 1D, Fig. 2B and Fig. 2D).

AC C

The present study demonstrated dynamic changes in amniotic fluid volume and compositions as well as Aqps gene expression in placenta and fetal membranes during the second and late gestation following Aqp1 gene deletion. Compared with wild phenotypes, amount of amniotic fluid was increased, osmolality was lower, and the concentration of Ca2+ was decreased in Aqp1 homozygote group at 16.5 GD. No significant change in APQ3, AQP8 and AQP9 mRNA and protein levels in placenta was observed after Aqp1 gene deletion at 13.5 GD and 16.5 GD. Whereas in fetal membranes the expression AQP9 mRNA and protein in fetal membranes was decreased, AQP8 expression was increased in Aqp1 deletion

ACCEPTED MANUSCRIPT homozygote group. Our finding suggested that there was no significant difference in placenta AQP1 expression between Aqp1 deletion homozygote, heterozygote, and wild-type groups. As placenta was a fetomaternal organ consisting of both fetal and maternal tissue [17], it was largely comprised of fetal blood vessels and

PT

trophoblasts, making up the chorionic villi covered by the chorion frondosum (plate). W hat’s more,

RI

placental villi were immerse in maternal blood delivered from the uterine spiral arteries, and the placenta

SC

was anchored to the endometrium via maternal decidual stromal cells, leading to the same expression levels of AQP1 in placenta between the three groups [18].

NU

In previous study, AQP1 mRNA and protein were present in amnion and chorion throughout human

MA

gestation. It was reported that AQP1 presented the highest levels of expression during the first trimester and globally tended to decrease after 11 week of gestation (WG). AQP1 mRNA quantity in 11 and 21 WG

ED

was significantly increased in total membranes, as well as amnion and chorion. Moreover, the protein

EP T

expression of AQP1 also peaked at 11 and 26 WG, instead of at 21 WG [19]. These results indicated that AQP1 was potentially involved in the regulation of amniotic fluid homeostasis throughout gestation. Zheng et al. [20] mated homozygous Aqp1 knockout mice and reported that fetuses exhibited

AC C

polyhydramnios. The number of embryos decreased along the advance of gestational age and the fetal weight of Aqp1 knockout mice was significantly lower than wild-type. The Aqp1 knockout mice displayed degenerating placentas as evidenced by altered blood vessel structure and increased syncytiotrophoblast nodules. In Aqp8-knockout mice, Sha et al. [21] showed an increase in embryo numbers, heavier placenta and fetal weight and an increase in the amount of amniotic fluid. These results strongly suggest that the AQP1 and AQP8 may carry out diverged function in the regulation of amniotic fluid volume and embryo development. Our results confirmed that loss of AQP1 expression in fetal membranes was related to

ACCEPTED MANUSCRIPT increased amniotic fluid volume and reduced Ca2+ concentration. Mola et al. [22] suggested that AQPs might influence swelling kinetics and mediate calcium signaling after hypotonic stimulus, they speculated that AQP-dependent local swelling could trigger stretch-activated calcium channels at the leading edge with a resulting enhanced calcium influx and enhanced motility. It could partly explain these phenomena.

PT

Further studies are needed to understand the mechanism by which AQP1 modulate the calcium

RI

concentration. Moreover, this effect was manifested in late gestation (16.5 GD) rather than middle (13.5

SC

GD) stage of the pregnancy. This timeline aligned well with that of the occurrence of polyhydramnios in human beings in the late gestation [23].

NU

Up to date, little is known about the mechanisms by which AQPs may regulate the amniotic fluid

MA

volume. Indeed, renal movement of water through the AQP1 channel in fetus may contribute to the homeostasis of amniotic fluid [14] [24-25]. In addition, the heterozygote pregnant mice may have

ED

decreased expression of AQP1 in many organs including kidney and gastrointestinal system, which may

EP T

also have an impact on the water balance in plasma and tissues [26]. Compensatory changes in fetus as well as maternal side may further complicate the situation. Our observation on the increased AQP8 expression in fetal membranes after Aqp1 deletion could be explained by the compensation assumption.

AC C

The decreased AQP9 expression in fetal membranes after Aqp1 deletion seemed suggest that this water channel could play a different role as compared to AQP8. We speculated that the different responses by different Aqps genes after Aqp1 gene deletion may be due to the classification and function of aquaporins. AQP1 and AQP8 were classified as classical AQPs selectively permeable to water [27], while AQP9 was ranged as aquaglyceroporins, which allow the transport of water and non-polar solutes, reactive oxygen species, gases and metalloids [28-30]. Further studies such as Aqp9 gene knockout is required to specifically address this issue.

ACCEPTED MANUSCRIPT In summary, our findings suggest that AQP1 in fetal membranes might be involved in the regulation of amniotic fluid amount, osmolality, and concentration of Ca2+, causing an increased amniotic fluid amount. In addition, loss of AQP1 expression in fetal membranes resulted in the downregulation of AQP9 expression and upregulation of AQP8 expression. Further studies should be conducted to clarify the

PT

mechanisms of mutual compensation in AQPs expression and functions of AQPs in placentas and fetal

SC

RI

membranes.

Acknowledgements

NU

This work was supported by grants from Zhejiang Science and Technology Agency (grant number

MA

2015C37096), National Nature Science Fund of China (grant number 81601319) and Project of Wenzhou

AC C

EP T

ED

Science and Technology (grant number 2016Y0551).

ACCEPTED MANUSCRIPT References [1]. S. Tomoda, R.A. Brace, L.D. Longo, Amniotic fluid volume regulation: basal volumes and responses to fluid infusion or withdrawal in sheep, Am. J. Physiol. 252(1987) R380-387. [2]. H. Liu, Z. Zheng, E.M. Wintour, Aquaporins and fetal fluid balance, Placenta. 29(2008) 840-847.

PT

[3]. M.H. Beall, J.P. Van Den Wijngaard, M.J. Van Gemert, M.G. Ross, Regulation of amniotic fluid volume, Placenta. 28(2007)824-832.

RI

[4]. C.R. Harman, Amniotic fluid abnormalities, Perinatol. Semin. 32(2008) 288-294.

SC

[5]. S.E. Mann, M.J. Nijland, M.G. Ross, Mathematic modeling of human amniotic fluid dynamics, Am. J.

NU

Obstet. Gynecol. 175(1996) 937-944.

[6]. W.M. Gilbert, E. Eby-Wilkens, A.F. Tarantal, The missing link in rhesus monkey amniotic fluid

MA

volume regulation: intramembranous absorption, Obstet. Gynecol. 89(1997) 462-465. [7]. Y. Zhang, S. Ding, Q. Shen, J. Wu, X. Zhu, The expression and regulation of aquaporins in placenta

ED

and fetal membranes, Front. Biosci (Landmark Ed). 17(2012):2371-2382.

EP T

[8]. Y. Hua, W. Jiang, W. Zhang, Q. Shen, M. Chen, X. Zhu, Expression and significance of aquaporins during pregnancy, Front. Biosci (Landmark Ed). 18(2013)1373-1383.

AC C

[9]. S.E. Mann, E.A. Ricke, E.A. Torres, R.N. Taylor, A novel model of polyhydramnios: amniotic fluid volume is increased in aquaporin 1 knockout mice, Am. J. Obstet. Gynecol. 192(2005)2041-2044. [10]. S.E. Mann, N. Dvorak, H. Gilbert, R.N. Taylor, Steady-state levels of aquaporin 1 mRNA expression are increased in idiopathic polyhydramnios, Am. J. Obstet. Gynecol. 194(2006)884-887. [11]. X. Zhu, S. Jiang, Y. Hu, X. Zheng, S. Zou, Y. Wang, The expression of aquaporin 8 and aquaporin 9 in fetal membranes and placenta in term pregnancies complicated by idiopathic polyhydramnios, Early. Hum. Dev. 86(2010)657-663. [12]. L. Li, H. Zhang, T. Ma, A.S. Verkman, Very high aquaporin-1 facilitated water permeability in mouse

ACCEPTED MANUSCRIPT gallbladder, Am. J. Physiol. Gastrointest. Liver. Physiol. 296(2009)G816-822. [13]. T. Ma, Y. Song, B. Yang, A. Gillespie, E.J. Carlson, CJ. Epstein, Nephrogenic diabetes insipidus in mice lacking aquaporin-3 water channels, Proc. Natl. Acad. Sci. U.S.A. 97(2000)4386-4391. [14]. T. Ma, B. Yang, A. Gillespie, E.J. Carlson, C.J. Epstein, A.S. Verkman, Severely impaired urinary

PT

concentrating ability in transgenic mice lacking aquaporin-1 water channels, J. Biol. Chem.

RI

273(1998)4296-4299.

SC

[15]. H. Ge, H. Zhu, N. Xu, D. Zhang, J. Ou, G. Wang, Increased lung ischemia-reperfusion injury in Aquaporin 1-null mice is mediated via decreased hypoxia- inducible factor 2alpha stability, Am. J. Respir.

NU

Cell. Mol. Biol. 54(2016)882-891.

MA

[16]. M.T. Ratsep, A.M. Felker, V.R. Kay, L. Tolusso, A.P. Hofmann, B.A. Croy, Uterine natural killer cells: supervisors of vasculature construction in early decidua basalis, Reproduction. 149(2015)R91-102.

ED

[17]. O. Parolini, F. Alviano, I. Bergwerf, D. Boraschi, C. De Bari, P. De Waele, Toward cell therapy using

EP T

placenta-derived cells: disease mechanisms, cell biology, preclinical studies, and regulatory aspects at the round table. Stem. Cells. Dev. 19(2010)143-154. [18]. RA. Pelekanos, VS. Sardesai, K. Futrega, WB. Lott, M. Kuhn, MR. Doran, Isolation and Expansion

AC C

of Mesenchymal Stem/Stromal Cells Derived from Human Placenta Tissue. J. Vis. Exp. 112(2016)54204. [19]. C. Prat, L. Blanchon, V. Borel, D. Gallot, A. Herbet, D. Bouvier, G. Marceau, V. Sapin, Ontogeny of aquaporins in human fetal membranes. Biol. Reprod. 86(2012)48. [20]. Z. Zheng, H. Liu, M. Beall, T. Ma, R. Hao, M.G. Ross, Role of aquaporin 1 in fetal fluid homeostasis, J. Matern. Fetal. Neonatal. Med. 27(2014)505-510. [21]. X.Y. Sha, Z.F. Xiong, H.S. Liu, Z. Zheng, T.H. Ma, Pregnant phenotype in aquaporin 8-deficient mice, Acta. Pharmacol. Sin. 32(2011)840-844.

ACCEPTED MANUSCRIPT [22]. MG. Mola, A. Sparaneo, CD. Gargano, DC. Spray, M. Svelto, A.Frigeri, E. Scemes, GP. Nicchia. The speed of swelling kinetics modulates cell volume regulation and calcium signaling in astrocytes: A different point of view on the role of aquaporins. Glia. 64(2016)139-154. [23].R.A. Brace, Physiology of amniotic fluid volume regulation, Clin. Obstet. Gynecol. 40(1997)280-289.

PT

[24]. C.L. Chou, M.A. Knepper, A.N. Hoek, D. Brown, B. Yang, T. Ma, Reduced water permeability and

RI

altered ultrastructure in thin descending limb of Henle in aquaporin-1 null mice, J. Clin. Invest.

SC

103(1999)491-496.

[25]. C. Bondy, E. Chin, B.L. Smith, G.M. Preston, P. Agre, Developmental gene expression and tissue

NU

distribution of the CHIP28 water-channel protein, Proc. Natl. Acad. Sci. U. S. A. 90(1993)4500-4504.

MA

[26]. X.Y. Sha, Z.F. Xiong, H.S. Liu, X.D. Di, T.H. Ma, Maternal- fetal fluid balance and aquaporins: from molecule to physiology, Acta. Pharmacol. Sin. 32(2011)716-720.

ED

[27]. K. Ishibashi, S. Kondo, S. Hara, Y. Morishita, The evolutionary aspects of aquaporin family, Am. J.

EP T

Physiol. Regul. Integr. Comp. Physiol. 300(2011)R566-576. [28]. A. Madeira, TF. Moura, G. Soveral, Aquaglyceroporins: implications in adipose biology and obesity, Cell. Mol. Life. Sci. 72(2015)759-771.

AC C

[29]. M. Hara-Chikuma, AS. Verkman, Physiological roles of glycerol-transporting aquaporins: the aquaglyceroporins, Cell. Mol. Life. Sci. 63(2006)1386-1392. [30]. R. Mukhopadhyay, H. Bhattacharjee, BP. Rosen, Aquaglyceroporins: generalized metalloid channels. Biochim. Biophys. Acta. 1840(2014)1583-1591.

ACCEPTED MANUSCRIPT Figure legends Fig. 1. Expression levels of AQPs mRNA in placentas and fetal me mbranes of Aqp1 gene knockout homozygote, heterozygote and wild-type groups at 13.5 GD and 16.5 GD. (A) Relative AQP1, AQP3, AQP8 and AQP9 mRNA expression in placenta at 13.5 GD. No significant difference in AQP1, AQP3,

PT

AQP8 and AQP9 mRNA expression was observed among three groups. (B) Relative, AQP3, AQP8 and

RI

AQP9 mRNA expression in fetal membranes in three groups at 13.5 GD. Compared with Aqp1

SC

heterozygote and wild-type groups, while expression of AQP8 mRNA in fetal membrane was significantly increased, expression of AQP9 mRNA in fetal membranes was decreased in Aqp1 homozygote group. *P <

NU

0.05 indicates significant difference compared with the Aqp1 wild-type group by LSD post hoc test after a

MA

one-way analysis of variance. Horizontal lines show the means ± SD. (C) Relative AQP1, AQP3, AQP8 and AQP9 mRNA expression in placentas at 16.5 GD. No significant difference in AQP1, AQP3, AQP8

ED

and AQP9 mRNA expression was observed among the three groups. (D) Relative AQP3, AQP8 and AQP9

EP T

mRNA expression in fetal membranes in three groups at 16.5 GD. Compared with Aqp1 heterozygote and wild-type groups, while expression of AQP8 mRNA in fetal membranes significantly increased, the expression of AQP9 mRNA was decreased in Aqp1 homozygote group, *P < 0.05 indicates significant

AC C

difference compared with the Aqp1 wild-type group by LSD post hoc test after a one-way analysis of variance. Horizontal lines show the means ± SD. Fig. 2. Expression of AQPs protein in the placentas and fetal membranes in the Aqp1 gene knockout homozygote, heterozygote and wild-type groups at 13.5 GD and 16.5 GD. (A) AQP1, AQP3, AQP8 and AQP9 protein expression in placenta at 13.5 GD. No significant difference of AQP1, AQP3, AQP8 and AQP9 protein expression was observed among the three groups. (B) AQP3, AQP8 and AQP9 protein expression in fetal membrane at 13.5 GD. Compared with Aqp1 heterozygote and wild-type groups, while

ACCEPTED MANUSCRIPT expression of AQP8 protein was significantly increased, the expression of AQP9 protein in fetal membranes was decreased in Aqp1 homozygote group. (C) AQP1, AQP3, AQP8 and AQP9 protein expression in placenta in three groups at 16.5 GD. No significant difference of AQP1, AQP3, AQP8 and AQP9 protein expression was observed among three groups. (D) AQP3, AQP8 and AQP9 protein

PT

expression in fetal membranes at 16.5 GD. Compared with Aqp1 heterozygote and wild-type groups, while

RI

expression of AQP8 protein was significantly increased, the expression of AQP9 protein in fetal membrane

AC C

EP T

ED

MA

NU

SC

was decreased in Aqp1 homozygote group.

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

Fig. 1

PT

ACCEPTED MANUSCRIPT

AC C

EP T

ED

MA

NU

SC

RI

Fig. 2

ACCEPTED MANUSCRIPT Table 1. Primers of AQPs and GAPDH for murine placentas and fetal membranes. Forward primer (5’-3’)

Reverse primer (3’-5’)

AQP1

AGCAGCGACTTCACAGAC

CTATTTGGGCTTCATCTCC

AQP3

ATTGTCTCCCCACTCCTG

TCACATTCTCTTCCTCGG

AQP8

TAAGCCCCATTCTCCATT

AGTAGCCAGCCATCACAG

AQP9

CTTCCACCATCCTTCCAC

TGAGCAATAGAGCCACATC

GAPDH

AAGAAGGTGGTGAAGCAGG

GAAGGTGGAAGAGTGGGAGT

PT

Gene

AC C

EP T

ED

MA

NU

SC

RI

AQP, aquaporin; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

ACCEPTED MANUSCRIPT Table 2. Changes in Amniotic fluid, osmolality, composition concentration and fetus, fetal membrane and placenta weights as well as placenta areas among the AQP1 homozygote, heterozygote and wild-type groups at 13.5 GD. AQP1(-/-)

AQP1(+/-)

AQP1(+/+)

Amniotic fluid (g)

131.7±20.8

127.8±21.7

140.6±19.4

Osmolality (mOsm/L)

293.3±16.9

287.3±5.5

297.2±19.4

Total protein (g/L)

2.9±0.46

2.8±0.6

Glucose (g/L)

2.7±0.7

2.8±0.9

Urea (mmol/L)

8.6±0.3

Creatinine (mmol/L)

9.8±0.2

Na+ (mmol/L)

134.7±13.2

K+ (mmol/L)

7.5±1.7

Cl- (mmol/L)

Placenta weight (mg) Placental area (mm2 )

AC C

Fetal membrane weight (mg)

RI

SC

2.41±0.5

9.8±0.7

10.3±0.6

137.7±6.8

136.1±10.3

6.5±0.6

6.6±0.8

100.4±12.5

100.5±8.1

98.1±9.9

1.12±0.2

1.4±0.2

1.5±0.2

133.2±31.2

135.6±32.6

146.3±30.6

15.2±3.0

16.2±4.1

17.2±3.4

70.4±20.9

68.9±16.8

72.7±9.6

0.44±0.08

0.45±0.08

0.42±0.06

MA

NU

7.6±0.9

ED

Fetus weight (mg)

3.8±1.8

8.2±0.8

EP T

Ca2+ (mmol/L)

PT

Parameters

Data were compared by ANOVA, Values are reported as mean±SD. No significant difference in these parameters was observed among the AQP1 homozygote, heterozygote and wild-type groups at 13.5 GD.

ACCEPTED MANUSCRIPT Table 3. Changes in Amniotic fluid, osmolality, composition concentration and fetus, fetal membrane and placenta weights as well as placenta areas among the AQP1 homozygote, heterozygote and wild-type groups at 16.5 GD. +/+

AQP1( )

Amniotic fluid (g)

207.4±22*

195.9±18.1

192.8±19.5

Osmolality (mOsm/L)

291.7±13.8*

308.7±4.8

324.7±6.8

Total protein (g/L)

2.5±0.5

2.03±0.3

2.85±0.5

Glucose (g/L)

1.63±0.6

1.68±0.6

1.63±0.5

Urea (mmol/L)

7.2±0.3

8.2±0.8

7.6±0.9

Creatinine (mmol/L)

15.3±6.1

11.9±1.3

13.1±2.7

145.2±4.2

145.5±4.6

6.5±0.6

6.6±0.8

111.4±4.2

109.2±6.3

109.01±2.1

Ca (mmol/L)

1.26±0.3*

1.21±0.2*

1.84±0.73

Fetus weight (mg)

654.1±115.8

621.8±96.8

629.9±104.5

Fetal membrane weight (mg)

31.2±4.3

29.8±6.7

33.1±7.2

81.4±10.0

78.8±6.9

85.3±11.3

0.46±0.07

0.47±0.05

0.49±0.06

145.5±5.8

+

K (mmol/L)

7.9±1.8

-

Cl (mmol/L)

ED

2+

2

EP T

Placenta weight (mg) Placental area (mm )

NU

Na (mmol/L)

SC

+

PT

AQP1( )

RI

+/-

AQP1( )

MA

-/-

Parameters

AC C

Data were compared by ANOVA, Values are reported as mean±SD. Compared with their AQP1 wild type counterparts, AQP1 homozygote conceptus had a greater volume of amniotic fluid, lower osmolality and calcium concentration at 16.5 GD. “*” indicates significant statistical difference compared with the AQP1 wild-type groups, P<0.05.

ACCEPTED MANUSCRIPT Highlights 

AQP1 knockout led to increased volume, decreased osmolality and calcium concentration in amniotic fluid. AQP3, AQP8 and AQP9 expression in placenta is not affected by AQP1 gene knockout.



AQP8 expression in fetal membrane is increased after AQP1 knockout.



AQP9 expression in fetal membranes is significantly decreased after AQP1 knockout.

AC C

EP T

ED

MA

NU

SC

RI

PT

