Genetic and microscopic assessment of the human chemotherapy-exposed placenta reveals possible pathways contributive to fetal growth restriction.

INTRODUCTION: Fetal growth restriction (FGR) carries an increased risk of perinatal mortality and morbidity. A major cause of FGR is placental insufficiency. After in utero chemotherapy-exposure, an increased incidence of FGR has been reported. In a prospective cohort study we aimed to explore which pathways may contribute to chemotherapy-associated FGR. METHODS: Placental biopsies were collected from 25 cancer patients treated with chemotherapy during pregnancy, and from 66 control patients. Differentially expressed pathways between chemotherapy-exposed patients and controls were examined by whole transcriptome shotgun sequencing (WTSS) and Ingenuity Pathway Analysis (IPA). Immunohistochemical studies for 8-OHdG and eNOS (oxidative DNA damage), proliferation (PCNA) and apoptosis (Cleaved Caspase 3) were performed. The expression level of eNOS, PCNA and IGFBP6 was verified by real-time quantitative Reverse Transcription Polymerase Chain Reaction (RT-qPCR). RESULTS: Most differential expressed genes between chemotherapy-exposed patients and controls were related to growth, developmental processes, and radical scavenging networks. The duration of chemotherapy exposure had an additional impact on the expression of genes related to the superoxide radicals degeneration network. Immunohistochemical analyses showed a significantly increased expression of 8-OHdG (P = 0.003) and a decreased expression of eNOS (P=0.015) in the syncytiotrophoblast of the placenta of cancer patients. A decreased expression of PCNA was detected by immunohistochemistry as RT-qPCR (NS). CONCLUSION: Chemotherapy exposure during pregnancy results in an increase of oxidative DNA damage and might impact the placental cellular growth and development, resulting in an increased incidence of FGR in this specific population. Further large prospective cohort studies and longitudinal statistical analyses are needed.

Placental effects of systemic tumour necrosis factor-α in an animal model of gestational diabetes mellitus.

BACKGROUND: Gestational diabetes mellitus (GDM) may adversely affect fetoplacental interaction. Numerous reports demonstrate that GDM women have increased circulating tumour necrosis factor-α (TNF), a pro-apoptotic peptide. OBJECTIVE: To examine whether implantation site apoptosis is increased by exogenous TNF in mice heterozygous for a defective leptin receptor (db/+), a GDM animal model. STUDY DESIGN: Implantation sites were studied at gestational day (gd)18.5 in 3 groups: saline-treated wild-type (wt) and db/+ mice, and TNF-treated db/+ mice. Saline or TNF (total dose 4 µg) was administered by miniosmotic pump from gd11.5. Immunostaining for cleaved caspase-3, PAS and cytokeratin was performed for quantification of apoptotic cells, uterine natural killer (uNK) cells, and trophoblast invasion, respectively. The mRNA expression of TNF and TNF-induced apoptotic markers in placenta and mesometrial triangle (MT) was measured by quantitative RT-PCR. RESULTS: The implantation sites from saline-treated wt and db/+ mice showed comparable numbers of apoptotic cells and uNK cells. Compared with the saline-treated groups, TNF-treated db/+ dams had less fetuses; the placental labyrinth and trophospongium contained more apoptotic cells; and the MT contained a higher total number of uNK cells including more cells intensely stained for cleaved caspase-3 as well as cells with negative staining. Trophoblast invasion was shallower in db/+ than in wt mice (14% and 30% of total invasion into MT, respectively) but this was not affected by TNF. The mRNA expression of TNF and apoptotic markers was comparable in the 3 groups. CONCLUSIONS: TNF treatment in db/+ mice raises the number of apoptotic cells in the placenta, and appears to increase the retention of uNK cells in the MT. Db/+ mice demonstrate shallower trophoblast invasion which is unaffected by exogenous TNF.

Transplacental transfer of anthracyclines, vinblastine, and 4-hydroxy-cyclophosphamide in a baboon model.

OBJECTIVE: The paucity of data on the fetal effects of prenatal exposure to chemotherapy prompted us to study transplacental transport of chemotherapeutic agents. METHODS: Fluorouracil-epirubicin-cyclophosphamide (FEC) and doxorubicin-bleomycin-vinblastine-dacarbazine (ABVD) were administered to pregnant baboons. At predefined time points over the first 25 h after drug administration, fetal and maternal blood samples, amniotic fluid (AF), urine, fetal and maternal tissues, and cerebrospinal fluid (CSF) were collected. High-performance liquid chromatography (HPLC) and liquid chromatography-mass spectrometry (LC-MS) were used for bioanalysis of doxorubicin, epirubicin, vinblastine, and cyclophosphamide. RESULTS: In nine baboons, at a median gestational age of 139 days (range, 93-169), FEC 100% (n = 2), FEC 200% (n=1), ABVD 100% (n = 5), and ABVD 200% (n = 1) were administered. The obtained ratios of fetal/maternal drug concentration in the different simultaneously collected samples were used as a measure for transplacental transfer. Fetal plasma concentrations of doxorubicin and epirubicin averaged 7.5 ± 3.2% (n = 6) and 4.0 ± 1.6% (n = 8) of maternal concentrations, respectively. Fetal tissues contained 6.3 ± 7.9% and 8.7 ± 8.1% of maternal tissue concentrations for doxorubicin and epirubicin, respectively. Vinblastine concentrations in fetal plasma averaged 18.5 ± 15.5% (n=9) of maternal concentrations. Anthracyclines and vinblastine were neither detectable in maternal nor in fetal brain/CSF. 4-Hydroxy-cyclophosphamide concentrations in fetal plasma and CSF averaged 25.1 ± 6.3% (n = 3) and 63.0% (n = 1) of the maternal concentrations, respectively. CONCLUSION: This study shows limited fetal exposure after maternal administration of doxorubicin, epirubicin, vinblastine, and 4-hydroxy-cyclophosphamide.

Upregulation of Wilms' tumour gene 1 (WT1) in uterine sarcomas.

AIM: Overexpression of Wilms' tumour gene (WT1) has been proven in several tumours. Previous research of our group on the cell cycle of uterine leiomyosarcoma (LMS) and carcinosarcoma (CS) suggested a possible role for WT1. We therefore intended to further explore the expression pattern of WT1 in uterine sarcomas. METHODS: 27 CS, 38 LMS, 15 endometrial stromal sarcomas (ESS) and seven undifferentiated sarcomas (US) were collected. WT1 expression was evaluated by immunohistochemistry (IHC) in 87 samples, by RT-PCR (m-RNA expression) in 23 random selected samples and by Western blotting in 12 samples, separating cytoplasmic and nuclear proteins. A pilot study to detect mutations (exons 7-10) was performed on eight samples. RESULTS: IHC showed WT1 positivity in 12/27 CS, 29/38 LMS, 7/15 ESS and 4/7 US. All-but-one sample had a positive RT-PCR. All Western blottings were positive with more cytoplasmic expression in 9/12 cases. No mutations were found. CONCLUSIONS: WT1 is overexpressed in uterine sarcomas. Since increased levels of mRNA determine the biological role, WT1 might contribute to uterine sarcoma tumour biology.

Pancreatic islet transplantation in diabetic pregnant rats prevents acquired malformation of the ventromedial hypothalamic nucleus in their offspring.

Exposure to a diabetic intrauterine environment leads to diabetogenic disturbances throughout later life in rats. This is accompanied by a fetally acquired dysplasia of the ventromedial hypothalamic nucleus (VMN) which is decisively involved in the regulation of metabolism. We investigated whether malformation of the VMN is preventable by normalization of gestational hyperglycaemia. Correction of hyperglycaemia in pregnant streptozotocin-diabetic rats was achieved by pancreatic islet transplantation. The number of neurons in the VMN was significantly reduced in adult offspring of non-treated, sham-transplanted mother rats (P<0.05), but did not differ between offspring of islet-transplanted mother rats and offspring of control mothers. In conclusion, prevention of VMN malformation in offspring of islet-transplanted diabetic mothers might be co-responsible for normalization of their glucose homeostasis during life.

PROLACTIN-deficiency in adult offspring of diabetic mothers.

Maternal diabetes induces fetal alterations, resulting in lasting consequences for the glucose tolerance of the offspring over several generations. In our experimental rat model, circulating prolactin, oestradiol, progesterone and corticosterone levels, known to influence insulin secretion and action, are determined in plasma of female adult offspring of mildly and severely diabetic mothers. Prolactin and progesterone levels are equally low in both groups as compared to controls, stressing the involvement of the CNS in the transgeneration effect; oestradiol and corticosterone levels are normal. No correlation is found between these hormonal alterations and the known differences in glucose tolerance.

Effects of long-term diabetes and/or high-dose 17 beta-estradiol on bone formation, bone mineral density, and strength in ovariectomized rats.

Long-term diabetes in female rats preserves the bone mineral density (BMD) but impairs the strength of the femur. In this study, we have compared the effects of diabetes and high-dose 17 beta-estradiol (E2), two conditions of low bone formation, in ovariectomized (ovx) rats. Spontaneously diabetic BB rats were ovx 0-3 days after onset, and nondiabetic ovx littermates were used as controls; the rats were either untreated or treated with E2 (30 micrograms/day, subcutaneously), for 6 or 12 weeks (n = 9 in each of the eight groups). Analysis included: plasma 1,25-dihydroxyvitamin D3, insulin-like growth factor-I (IGF-I), and osteocalcin concentrations; histomorphometry of the proximal tibial metaphysis (PTM); and DXA and biomechanical testing of the femur. Both E2 treatment and diabetes markedly lowered plasma IGF-I and osteocalcin concentrations, as well as dynamic morphometric parameters of bone formation in the PTM. Plasma IGF-I and osteocalcin were correlated (R2 = 0.55; p < 0.0001). E2 treatment in both control and diabetic ovx rats increased the trabecular bone volume in the PTM and the BMD in the metaphysis of the distal femur; there was no difference between control and diabetic rats, however. The diaphyseal area and BMC were decreased in E2-treated or/and diabetic ovx rats, but the diaphyseal BMD remained unchanged compared with untreated ovx rats. The biomechanical properties of the whole femur (strength, angular deformation, and stiffness) were decreased in E2-treated and diabetic E2-treated ovx rats after 12 weeks. The data indicate that in situations of chronic low bone formation, whole bone strength does not reflect total BMD but correlates better with bone size and bone mineral content measurements.

Effects of recombinant human growth hormone and insulin-like growth factor-I, with or without 17 beta-estradiol, on bone and mineral homeostasis of aged ovariectomized rats.

This study aimed to evaluate whether recombinant human growth hormone (rhGH) or insulin-like growth factor-I (rhIGF-I) can reverse or prevent further bone loss in aged osteopenic ovariectomized (OVX) rats and to compare their effects with those of 17 beta-estradiol (E2). Twelve-month-old rats were OVX, remained untreated for 8 weeks, and subsequently received daily subcutaneous (SC) injections of rhGH (75 micrograms/day), rhIGF-I (250 micrograms/day), E2 (1.5 micrograms/day), and their respective combinations during 8 weeks, and were then compared with sham-operated, pretreatment OVX, and saline-treated OVX rats. A single sc injection of rhGH resulted in peak hGH concentrations after 90 minutes, with a half-life of 124 minutes; the highest plasma IGF-I concentrations were reached 45 minutes after rhIGF-I injection (+57% vs. baseline) with a gradual decline thereafter. Measurements included: biochemical parameters of bone remodeling (plasma osteocalcin and urinary pyridinolines); histomorphometry of proximal tibial metaphysis; DXA of femur; biomechanical analysis of femur and fifth lumbar vertebra (L5); plasma 1,25-dihydroxyvitamin D3 (1,25(OH)2D3), and calbindin-D9K in duodenal mucosa. Whereas all E2-treated OVX rats had much suppressed bone remodeling, rhGH or rhIGF-I had no effect on any biochemical or histomorphometrical parameter of remodeling. The bone mineral density (BMD) at the distal femoral metaphysis as well as parameters of strength at L5 were maintained at pretreatment values in OVX rats treated with E2, GH, or IGF-I, but not in saline-treated OVX rats; their effects were not additive, however. Trabecular bone volume in the tibial metaphysis was also higher in rats treated with these agents than in saline-treated rats, but this was more apparent at the primary than at the secondary spongiosa, suggesting that their mechanism of action is on primary spongiosa formation or breakdown. E2 alone was ineffective to augment the BMD at the femoral diaphysis; however, the diaphyseal BMD was 12-14% higher (p < 0.01) after 8 weeks of GH treatment than in pretreatment or saline-treated OVX rats and sham-operated rats, while IGF-I was less effective than GH, GH or IGF-I treatment had no effect on plasma 1,25(OH)2D3 or duodenal calbindin-D9K concentrations, but the combination of GH or IGF-I with E2 potentiated the effect of E2 to stimulate calbindin-D9K concentrations and urinary calcium excretion, indicating "hyperabsorption hypercalciuria." In conclusion, the administration of rhGH and rhIGF-I, like that of E2, into aged OVX rats prevents further loss of bone mass and strength at sites containing trabecular bone. In addition, rhGH increases cortical bone mass above pretreatment values.

Effects of IGF-I, alone and in combination with 17 beta-oestradiol, on bone remodelling parameters in ovariectomised rats.

We studied the effects of recombinant human IGF-I (250 micrograms/day) and/or 17 beta-oestradiol (E2; 5 or 50 micrograms/kg/day), s.c., for 28 days, on plasma IGF-I and 1,25-(OH)2 vitamin D3 concentrations and on biochemical indices of bone remodelling (plasma osteocalcin, urinary pyridinolines) in 3 month-old rats that had been ovariectomised (OVX) 6 weeks before. Ten weeks after ovariectomy, plasma 1,25-(OH)2D3 was increased, but IGF-I and bone remodelling indices were within the normal range. RhIGF-I administration to OVX rats increased both IGF-I and 1,25-(OH)2D3 concentrations, as well as plasma osteocalcin and urinary pyridinoline excretion. By contrast, E2 decreased plasma IGF-I and 1,25-(OH)2D3 and the markers of bone remodelling. The combination of rhIGF-I and E2 resulted in intermediate effects. Multiple regression analysis showed that IGF-I correlated with plasma osteocalcin, and also with the urinary excretion of pyridinolines. The data shows that rhIGF-I stimulates bone remodelling in growing OVX rats, and that circulating IGF-I is a determinant of bone remodelling in vivo.

1,25-Dihydroxyvitamin D3 and osteocalcin in maternal and fetal guinea pigs.

Maternal and fetal 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) and osteocalcin were measured in guinea pigs, to examine their potential use as animal models for fetal bone development and calcium homeostasis. Measurements were performed on days 42, 57 and 63 of gestation. Maternal serum total 1,25(OH)2D3 concentrations were increased only at the end of gestation (day 63). However, because the vitamin D binding protein (DBP) and albumin levels were decreased by 35-50% from day 42 onwards, the unbound 1,25(OH)2D3, calculated as the 1,25(OH)2D3/DBP molar ratio, was increased before day 63. Osteocalcin concentrations during gestation were 50-54% of levels found in nongravid animals. Fetal serum total 1,25(OH)2D3 concentrations were 20% of those in maternal guinea pigs. Since DBP levels were only 9-15% of maternal levels, the unbound 1,25(OH)2D3 was consistently higher in fetuses, from day 42 onwards. There was a rise in total and unbound 1,25(OH)2D3 between days 57 and 63 of fetal life. Osteocalcin concentrations were higher in fetal than in adult guinea pigs, and reached peak values on day 57 (1023 micrograms/l, i.e. 4.2 times higher than in adult female guinea pigs). Fetuses of guinea pigs that had received a restricted food supply for 14 days (days 49-63) had normal 1,25(OH)2D3 concentrations, but decreased osteocalcin concentrations compared with normal fetuses. The data obtained in fetal guinea pigs are comparable with those found in human fetuses, and suggest that the guinea pig may be a suitable model for studies on fetal bone and mineral development.

Influence of a vitamin E-deficient diet on prostacyclin production by mesometrial triangles and aortic rings from nondiabetic and diabetic pregnant rats.

We investigated the effect of a vitamin E-deficient diet on the synthesis of prostacyclin in nondiabetic and diabetic pregnant rats. In both groups the diet reduced fetal growth and prostacyclin production by mesometrial triangles; it also increased serum lipid peroxides, which are known to inhibit prostacyclin synthesis. Independently of the diet, mesometrial triangles of diabetic rats produced more prostacyclin than those of control rats, whereas an opposite tendency was found in the aortic rings. Our model seems useful for fundamental and pharmacologic studies of the vascular pathology of pregnancy.