Pregestational diabetes increases fetoplacental vascular resistance in rats

Placenta 63 (2018) 32e38

Contents lists available at ScienceDirect

Placenta journal homepage: www.elsevier.com/locate/placenta

Pregestational diabetes increases fetoplacental vascular resistance in rats Olga Vajnerova a, *, Petr Kafka a, b, Tereza Kratzerova a, Karel Chalupsky a, Vaclav Hampl a a b

Department of Physiology, Second Faculty of Medicine, Charles University, Prague, Czech Republic Department of Anesthesiology and Intensive Care Medicine, Kralovske Vinohrady University Hospital, Prague, Czech Republic

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 January 2017 Received in revised form 2 January 2018 Accepted 4 January 2018

Introduction: Diabetes is a well-known risk factor in pregnancy. Because maternal diabetes involves oxidative stress that is also induced by chronic hypoxia and can alter vascular function, we sought to determine the effects of chronic maternal hyperglycemia on the fetoplacental vasculature in rats and to compare it with the effects of chronic hypoxia. Methods: Diabetes was induced in female rats by a streptozotocin injection at a neonatal age. When these animals reached adulthood, their hyperglycemia was confirmed and they were inseminated. Half of them were exposed to hypoxia (10% O2) for the last week before the delivery. One day before the expected date of delivery, one of their placentae was isolated and perfused. Results: Fetoplacental vascular resistance was increased equally by experimental diabetes, chronic hypoxia, and their combination. Fetoplacental perfusion pressure-flow analysis suggested increased resistance in the small vessels in chronic hypoxia and in larger vessels in diabetes. Fetal plasma nitrotyrosine levels, measured as a marker of peroxynitrite (reaction product of superoxide and nitric oxide), mirrored the differences in fetoplacental resistance, suggesting a causative role. Fetoplacental vasoconstrictor reactivity to acute hypoxic stimuli was reduced similarly in all groups. Fasudil, a strong vasodilator agent, reduced fetoplacental vascular resistance similarly in all groups, suggesting that for the observed differences among the groups, the changes in vascular morphology were more important than variances in vascular tone. Discussion: Maternal diabetes increases fetoplacental vascular resistance to a similar extent as chronic hypoxia. These stimuli are not additive. Changes in vascular tone are not responsible for these effects. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Placenta Diabetes Chronic hypoxia Oxidative stress Fetoplacental vessels

1. Introduction Chronic in utero hypoxia results in a sustained elevation of vascular resistance on the fetal side of the rat placenta [1]. The hypoxic state of the placenta could be a result of reduced availability of oxygen to the mother (e.g. maternal high altitude exposure or chronic respiratory disease of the mother) or could be a consequence of reduced uteroplacental blood flow (e.g. in preeclampsia or maternal heart failure). It is generally assumed that the hypoxic increase in fetoplacental vascular resistance leads to placental hypoperfusion and fetal undernutrition, and thus is a major factor in the pathogenesis of intrauterine growth restriction

* Corresponding author. Department of Physiology, Second Faculty of Medicine, Charles University, Plzenska 130/221, 150 00, Prague 5, Czech Republic. E-mail address: [email protected] (O. Vajnerova). https://doi.org/10.1016/j.placenta.2018.01.008 0143-4004/© 2018 Elsevier Ltd. All rights reserved.

in situations like persistent maternal hypoxemia or chronic maternal vascular disorders [2]. The possible mechanisms responsible for the chronic hypoxic elevation of fetoplacental vascular resistance have not been elucidated, but based on many similarities between adult pulmonary circulation and fetoplacental vasculature [3] may involve oxidant injury of the vascular wall by oxygen radicals known to participate in the chronic hypoxic pulmonary hypertension [4e6]. Diabetes is one of the most common complications during pregnancy [7]. It is well known to affect fetal development [8e10]. Similarly to hypoxia, diabetes also induces oxidative stress. In particular, it causes overproduction of superoxide in the mitochondria of endothelial and myocardial cells (for review, see e.g. [11]). There is evidence that in diabetic pregnancies the fetus experiences chronic hypoxia [12,13]. The effects of maternal diabetes on fetoplacental vascular resistance have so far been little studied. The first aim of the

O. Vajnerova et al. / Placenta 63 (2018) 32e38

33

present study, therefore, was to test the hypothesis that maternal diabetes causes an elevation of vascular resistance on the fetal side of the placenta. We also assumed that the relationships between diabetes, hypoxia, and vascular resistance in the fetal side of the placenta could be highlighted by combining the two factors (experimental diabetes and chronic maternal hypoxia) together.

reservoir at a constant flow rate of 1 ml/min. At these baseline conditions, the common perfusate was gassed with a normoxic mixture of 21% O2 and 5% CO2 in nitrogen. Umbilical vein was punctured to permit an easy outflow of the perfusate. After rapidly establishing the perfusion, all fetuses were sacrificed with thiopental overdose.

2. Materials and methods

2.3. Measurements

All procedures were in accordance with the European Guidelines on Laboratory Animal Care and were approved by the Animal Care Committee of the Second Faculty of Medicine, Charles University in Prague.

After the dual perfusion had been established, we measured fetoplacental perfusion pressure using a PowerLab data acquisition system (ADInstruments, Spechbach, Germany); the perfusion pressure at the maternal side was also monitored. After the preparation was stabilized and perfused at baseline conditions for at least 15 min, the perfusion pressure-flow relationship was determined by measuring fetoplacental perfusion pressure while the perfusion flow rate was continuously increased from 0 up to 2 ml/ min during 120 s (P/Q ramp). The flow rate was then returned to the baseline value of 1 ml/ min and the acute hypoxic vasoconstrictor reactivity was measured by changing the gas mixture bubbling the perfusate to anoxic one (95% N2 þ 5% CO2) until the fetoplacental perfusion pressure stabilized at a new level (20 min). After that, a Rho-kinase inhibitor fasudil (LC Laboratories, Woburn, Massachusetts, USA) was added to the perfusate (10 mM) in order to assess the vasoconstrictor component of the vascular resistance [17]. At the end of the perfusion, the mother was sacrificed by thiopental overdose. The perfused placenta, the other, non-perfused placentae from the same mother and all fetuses were weighed. Blood samples were collected from the mothers' tails at the beginning of each experiment to measure hematocrit (by microcapillary tube centrifugation), glycemia, and plasma 3nitrotyrosine. Plasma 3-nitrotyrosine was measured as an indicator of nitrating species, especially peroxynitrite [18], by inhibition ELISA (using antibodies prepared in our laboratory) as described previously [19]. Fetal blood was collected from the fetal body after decapitation and used to measure glycemia and plasma 3nitrotyrosine.

2.1. Experimental groups All animals were fed ad libitum with free access to water and kept in regular light cycles of 12/12 h light/dark. Four groups of laboratory rats were used: one group had experimental diabetes (D, n ¼ 9), another was exposed to chronic hypoxia (H, n ¼ 9), the third had both diabetes and was exposed to hypoxia (DH, n ¼ 9), and the fourth group were healthy control rats living in room air (C, n ¼ 10). To induce experimental diabetes we used the model of beta cell regeneration after neonatal streptozotocin (STZ) injection [14]. Pregnant Wistar albino rats were obtained from commercial breeding colony (Velaz, Prague, Czech Republic). Pups were delivered spontaneously and nursed by their mothers. A single subcutaneous injection of STZ (Santa Cruz Biotechnology, Dallas, Texas, USA) in a dose of 100 mg/kg in 20 mM sodium citrate buffer solution (pH 4.5; Sigma-Aldrich, Munich, Germany) was administered on day 2 or 3 of postnatal life. Control animals received an equal volume of solvent. To confirm the diabetic state, the blood glucose concentration was monitored with a glucose analyzer (FreeStyleOptium, Abbott Diabetes Care Ltd., Maidenhead, UK) in the period between the 6th and 10th postnatal week. Blood was obtained by cutting off the very tip of the tail and squeezing it gently. Since, in line with reported values [15], the average glycemia in our control rats was 6.3 mmol/l (range 4.7e8.0) we included into the diabetic groups only animals with blood glucose concentration in excess of 8 mmol/l at least once during this period. When adult (11th week of life), the diabetic and non-diabetic female rats were mated overnight with non-diabetic males. Then, when pregnant, they were randomly divided into two subgroups: the normoxic subgroup spent the whole gravidity in the atmospheric air, while the chronic hypoxic subgroup spent the last 7 days of pregnancy in a hypoxic normobaric chamber (10% O2) [16]. Experiments with perfused placenta were performed one day before the expected date of delivery (term ¼ 21 days). We did not notice alterations in the length of gestation.

2.4. Data analysis The results were analyzed statistically using the Prism software version 7 (GraphPad Software, Inc., CA, USA). The groups were compared with two-way ANOVA (one factor normoxia vs. chronic hypoxia, second factor healthy vs. diabetes). The effects of acute hypoxic challenge and of fasudil were evaluated using two-way repeated measures ANOVA. In all cases, post hoc analysis was performed using Fisher's LSD test. P < .05 was considered significant. Linear regression was used to evaluate the P/Q ramp data. The results are presented as means ± SEM.

2.2. Perfused placenta preparation 3. Results We used the model of isolated, dually perfused rat placenta as previously reported [1,17]. Briefly, pregnant rats, anesthetized with sodium thiopental (Valeant, Prague, Czech Republic, 50 mg/kg i.p.), were placed into a bath of Ringer solution kept at 37  C. After lower midline laparotomy the maternal side of selected placenta (uterine artery) was cannulated using a 24-gauge catheter and perfused at 1 ml/min with Krebs solution, whereas the uterine vein was ligated behind the perfused placenta and carefully cut to allow free drainage. The uterus was then opened on the antimesometrial side and the fetus was exposed. The umbilical artery of one of the placentae (chosen based on absence of visually apparent abnormalities and technical accessibility) was cannulated and perfused with the same perfusate as the maternal side from a common

Mother rats subjected to chronic hypoxia (H and DH groups) had lower body weight than normoxic groups, whereas experimental diabetes alone did not significantly affect maternal weight (Table 1). The body weight of the fetuses was not affected by either hypoxia alone or diabetes alone, but the combination of diabetes plus hypoxia resulted in significantly smaller fetuses (Table 1). The weights of the placentae did not differ among the groups, so the placental-fetal ratio was higher in the DH group than in the normoglycemic groups that did not differ one from another (Table 1). In all groups, the placental weight after the perfusion did not differ from the weight of placentae that were not perfused (Table 1), indicating that the perfusion protocol did not result in gross edema.

34

O. Vajnerova et al. / Placenta 63 (2018) 32e38

Table 1 Characteristics of experimental groups. Group C Body weight of mothers (g) 348 ± 10 (10) Fetal weight (all fetuses, g) 4.6 ± 0.1 (40) Wet weight of non-perfused placentae (mg) 533 ± 12 (30) Wet weight of placentae after perfusion (mg) 521 ± 35 (10) Placental-fetal weight ratio (perfused placentae and their fetuses) 0.112 ± 0.008 (9) Litter size 12.8 ± 1.0 Hematocrit (% packed cell volume) 42.1 ± 1.2 (9) Maternal blood glucose concentration (mmol/l) 6 - 10 week of age 6.5 ± 0.4 (10) 16 - 20 week of age (i. e. day of experiment) 5.2 ± 0.2 (10)þþ Fetal blood glucose concentration (mmol/l) 6.7 ± 0.6 (5) Values are means ± SEM, in parentheses are ns. * Significantly different from C group (*P < .05, **P < .01,

***

P < .001), #from H group (#P < .05,

As expected, mothers' hematocrit was increased similarly in both hypoxic groups but was not affected by diabetes (Table 1). There were no significant differences in the litter size among the groups (Table 1). Blood glucose concentration of rats in the D and DH groups measured before gravidity ranged from 8.3 to 25.1 mmol/l, which was significantly higher than in the non-diabetic groups (Table 1). In half of the diabetic animals, the glycemia was 10 mmol/l or more. At the beginning of placenta preparation, the blood glucose concentration of the mothers was still elevated in the D group as compared to both non-diabetic groups (that did not differ one from another), but it was not in the DH group (Table 1). In general, fetal glycaemia was somewhat higher than in mothers (except in the group H, where it was about the same) and followed the same intergroup differences as in the mothers. Thus, compared to controls, fetal blood glucose concentration was elevated in the D group but not in the H and DH groups (Table 1). Fetoplacental perfusion pressure during constant base-line flow rate (directly reflecting vascular resistance) under normoxic conditions did not differ among the H, D and DH groups and was higher in all of them than in the control group (Fig. 1A). A more detailed analysis of the resistive properties of the fetoplacental circulation by the P/Q measurement confirmed higher resistance in H, D and HD groups than in controls (Fig. 2A). All P/Q data fit well into a linear regression model (linear regression coefficient >0.985). While chronic hypoxia without diabetes caused a parallel shift of the P/Q line towards higher pressures (higher pressure-axis

##

H

D **þþ

346 ± 10 (9) 4.5 ± 0.1 (34) 537 ± 22 (25) 574 ± 30 (9) 0.132 ± 0.009 (9) 11.8 ± 1.2 44.0 ± 1.0 (9)# 12.1 ± 1.8 (9)**## 12.0 ± 3.0 (9)**## 15.9 ± 2.6 (8)**##

311 ± 7 (9) 4.2 ± 0.1 (35) 565 ± 19 (26) 518 ± 38 (9) 0.124 ± 0.004 (9) 11.9 ± 1.2 48.8 ± 1.5 (9)**þ 6.2 ± 0.3 (9)þþ 4.8 ± 0.2 (9)þþ 4.6 ± 0.2 (4)þþ

P < .01), þfrom D group (þP < .05,

DH ##

þþ

P < .01,

þþþ

310 ± 7 (9)**þþ 3.7 ± 0.2 (34)***þþþ 525 ± 16 (25) 543 ± 20 (9) 0.155 ± 0.013 (9)**# 11.0 ± 1.1 49.9 ± 1.7 (9)***þþ 13.6 ± 2.3 (9)**## 6.4 ± 1.1 (9)þ 8.4 ± 1.8 (4)þ

P < .001).

intercept in the H group than in controls, unchanged slope), experimental diabetes without hypoxia increased the steepness of the P/Q line (higher slope in the D group than in controls, unchanged pressure-axis intercept). When hypoxia was combined with diabetes (group DH), the slope was similarly elevated as in the D group. The pressure-axis intercept in the DH group was higher than in C group and did not differ from the H group (while that of the D group did) (Fig. 2B and C). In all groups, acute hypoxic challenge (during constant flow rate) elicited a significant increase in perfusion pressure (Fig. 1B). The rel-ative magnitude of this acute hypoxic fetoplacental vasoconstriction (HFPV), expressed as % of the base-line perfusion pressure in normoxia, was reduced similarly by chronic hypoxia (þ8.7 ± 1.8% of resting baseline), experimental diabetes (þ10.5 ± 2.8%), and their combination (þ4.8 ± 1.5) as compared to normoxic healthy controls (þ21.0 ± 2.6%). HFPV was completely reversed by the Rho-kinase inhibitor, fasudil, in all groups. In fact, in response to Rho-kinase inhibition, the perfusion pressure fell even bellow the resting base-line value (by about 13% in all groups) (Fig. 1C). Fasudil treatment did not remove the differences in perfusion pressure between the groups that were observed at base-line (Fig. 1A). Perfusion pressure after fasudil was 86.2 ± 1.4% of that at the beginning of perfusion in C, 86.7 ± 2.3% in D, 87.5 ± 1.4% in H and 82.7 ± 3.1% in DH groups. Maternal plasma 3-nitrotyrosine concentrations were not significantly affected by either chronic hypoxia or experimental diabetes alone. Their combination, however, resulted in

Fig. 1. Chronic hypoxia, experimental diabetes and their combination elevate fetoplacental perfusion pressure in normoxia, acute hypoxia, and in the presence of a strong vasodilator. Fetoplacental perfusion pressure under constant-flow conditions in normoxia (panel A), in acute hypoxia (panel B), and after fasudil was added during hypoxic perfusion (panel C) in control (group C) and diabetic (group D) placentae, in placentae exposed to chronic hypoxia (group H) and in diabetic placentae exposed to chronic hypoxia (DH). In parentheses are n. The effects of acute hypoxia and of fasudil were significant in all groups (by two-way repeated measures ANOVA with Fisher's LSD test). * significantly different from the control (C) group (*P < .05, **P < .01, ***P < .001, ****P < .0001) by two-way ANOVA with Fisher's LSD test (other differences among the groups not significant).

O. Vajnerova et al. / Placenta 63 (2018) 32e38

35

Fig. 2. Chronic hypoxia, experimental diabetes and their combination alter pressure-flow characteristics of the fetoplacental vasculature. Pressure-flow (P/Q) plot of all experimental groups in normoxia. Lines in panel A are linear regressions for each group (denoted as in Fig. 1). Panels B and C show, respectively, the slope and intercept with the pressure axis of the P/Q lines. n ¼ 9/group except in the group C, where n ¼ 10. * significantly different from the control (C) group (*P < .05, **P < .01), ##significantly different from the diabetic (D) group (P < .01) by two-way ANOVA with Fisher's LSD test (other differences among the groups not significant).

significantly elevated maternal plasma 3-nitrotyrosine levels (Fig. 3A). In contrast, fetal plasma 3-nitrotyrosine concentrations were increased (about equally) in all experimental groups (H, D, DH) in comparison with controls (Fig. 3B).

4. Discussion The main finding of this study is that relatively mild experimental diabetes of the mother causes consistent significant elevation of vascular resistance in the fetal side of the placenta in rats (Figs. 1A and 2A). This increase is of about the same magnitude as that induced by chronic hypoxic exposure of the dam, yet the combination of the two insults (diabetes and hypoxia) does not have additive effect on the fetoplacental vascular resistance. Nevertheless, a more detailed characterization of the resistive properties of the fetoplacental vascular bed by P/Q analysis revealed that the mechanisms by which diabetes/hyperglycemia and chronic hypoxia elevate fetoplacental vascular resistance differ (Fig. 2). It is well established that maternal diabetes - both pre-existing

and gestational - is a risk factor in pregnancy [for review, see e.g. 7]. While most focus so far has been directed to metabolic aspects (including fetal size), birth defects, and obstetrical complications, the role of changes in fetoplacental vascular resistance has been little studied. Nevertheless, changes in fetoplacental vascular resistance directly affect fetoplacental blood flow and thus fetal growth, development, and wellbeing. We have shown previously that chronic hypoxic exposure of the mother induces elevation of vascular resistance on the fetal side of the placenta [1]. In this respect, the fetoplacental vessels resemble adult pulmonary circulation (where chronic hypoxia causes pulmonary hypertension) and differ from all other vessels in the body (where hypoxia reduces resistance or has little effect). There is evidence from both animal models and human patients that maternal diabetes is associated with fetal hypoxemia [12,13]. It might thus be assumed that lowered O2 content in the fetoplacental blood contributes to the elevated fetoplacental vascular resistance in diabetes. However, the differing effects of diabetes and hypoxia on the P/Q lines (Fig. 2) seem to question this possibility. Our animal model is one of untreated diabetes. In medical

Fig. 3. Chronic hypoxia, experimental diabetes and their combination are associated with elevated fetal plasma nitrotyrosine concentration in the fetus, but not in the mother. In the mothers, plasma 3-nitrotyrosine is elevated only by the combination of experimental diabetes and chronic hypoxia (A). In the fetus, diabetes, chronic hypoxia and their combination increase plasma 3-nitrotyrosine similarly (B). Group abbreviations as in Fig. 1, in parentheses are n. * significantly different from control (C) group (*P < .05, ***P < .001), ##significantly different from the chronically hypoxic group (P < .01), by two-way ANOVA with Fisher's LSD test (other differences among the groups are not significant).

36

O. Vajnerova et al. / Placenta 63 (2018) 32e38

practice, maternal diabetes is almost always treated with the aim to achieve blood glucose levels similar to normal. However, in reality, some degree of hyperglycemia is frequently present in diabetic pregnant women [20,21]. At the same time, the elevation of glycemia was not large in our diabetic animals (great majority had glycaemia under 15 mmol/l) and virtually absent after 1 week of hypoxia (this is most likely due to reduced food intake that is symptomatic for the first several days of hypoxia, and thus much less insulin needed to maintain glycemia [22]). Thus, our results appear clinically relevant. The increase in the fetoplacental vascular resistance induced by diabetes is expected to reduce fetal blood flow through the placenta (unless fetal cardiac output is also elevated). This would limit nutrients and oxygen supply to the fetus and thus fetal growth. However, this effect is likely to be to some extent offset or even overcome by the higher glucose offer to the fetus in the placenta. This may explain why fetal macrosomia is a common feature of diabetes in humans and why, in our study, the pups of diabetic mothers were not smaller than controls. Nevertheless, any reduction in fetoplacental blood flow also decreases oxygen delivery to the fetus (and removal of metabolites), possibly contributing to fetal morbidity - a well-known problem in maternal diabetes in pregnancy. The observed effects of diabetes and hypoxia are statistically significant, however, they may not seem numerically large. For example, the difference in baseline perfusion pressure between controls and all other groups was about 10 mmHg (Fig. 1A). In the systemic circulation, this would not be a big effect. However, the fetoplacental circulation is a low-resistance, low-pressure circuit [3], and the ~10 mmHg difference represents a 40% (in the H group) to 55% (in the DH group) increase over the control group. That would correspond to about one third reduction in fetoplacental blood flow in vivo, clearly a physiologically very significant effect. While the perfusion pressure during the constant-flow conditions directly reflects vascular resistance, the pressure-flow relationship analysis provides more detailed characterization of vascular resistance and its components. In this study, chronic hypoxia shifted the P/Q lines up in parallel with the control group (i. e. significantly increased the intercept with the pressure axis), while experimental diabetes increased the slope of the P/Q lines without changing the intercept (Fig. 2B and C). The pressure axis intercept appears to represent the critical closing pressure of peripheral collapsible vessels, i. e. the pressure needed to overcome the sum of the vascular tone and external pressure that causes the vessels to close completely at low intravascular pressures [25]. The location of the critical closing pressure in the fetoplacental circulation is likely to be in the very small peripheral arterioles (capable of tone and yet collapsible). The slope of the P/Q line corresponds to the resistance of the open vascular channels. The increase in the pressure axis intercept without a concomitant increase in the P/Q slope in the H group thus can be interpreted as a selective increase in resistance in the locus responsible for the critical closing pressure (probably small arterioles, as discussed above). This corresponds with the observations that acute hypoxia causes vasoconstriction in the small arterioles both in the fetal placenta [26] and in the analogous situation in the postnatal lung [27], even though the mechanism of the effect of chronic hypoxia seems to include more than just the extension of acute hypoxia [28]. On the other hand, the increased P/Q slope without a change in the intercept in the diabetic group can be interpreted as an elevation of resistance of other vessels than the small arterioles. It could be the larger arteries, and/or the venous side of the fetoplacental vascular bed. However, a major venous contribution seems unlikely because it could be expected to aggravate the outflow from the placenta and thus promote edema formation that was not

detected in our experiments. In terms of the P/Q relationship, chronic hypoxia and experimental diabetes appear complementary - the DH group shares the elevated slope with the D group and elevated intercept with the H group (Fig. 2B and C). This further supports the conclusion that they act in different segments of the vasculature and/or by different mechanisms. In principle, any changes in vascular resistance can be brought about by changes in vascular tone and/or in vascular morphology. To assess their relative role in diabetes, we eliminated the former by administering a strong vasodilator, fasudil. In all groups, fasudil not only completely reversed the acute HFPV (as expected), but it also set the perfusion pressure below the baseline value, indicating the presence of resting tone in fetoplacental vessels (Fig. 1C). This contrasts with the common notion that normal fetoplacental vessels lack resting tone [26,30]. It is likely that vasodilators used in previous studies (sodium nitroprusside, nifedipine) act through mechanisms that are not sufficient for a complete smooth muscle relaxation in the perfused placenta preparation. Nevertheless, fasudil did not remove the differences among the groups, suggesting that chronic hypoxia and experimental diabetes exert their effect on the fetoplacental vessels not through altered vascular tone. Reduced total and mean cross-sectional area of villous vessels was described in human diabetes [31]. Since oxidant injury plays an important role in the consequences of diabetes [reviewed e.g. in 11], it is likely that it may participate in the mechanism of the observed fetoplacental vascular changes. To evaluate this possibility, we measured maternal and fetal plasma nitrotyrosine, a marker of the presence of peroxynitrite that is a product of a fast reaction between superoxide and nitric oxide. The differences among the groups in fetal plasma nitrotyrosine paralleled the differences in fetoplacental vascular resistance, i.e. the nitrotyrosine levels were similarly elevated by chronic hypoxia and experimental diabetes and their combination (Fig. 3). This is compatible with the possibility that vascular wall injury caused by oxygen radicals and/or nitric oxide and its derivates contributes to the observed increase in fetoplacental vascular resistance in diabetes and chronic hypoxia. However, our analysis of the P/Q data suggests that experimental diabetes and chronic hypoxia act at different loci. Therefore, the increased fetal plasma levels of nitrotyrosine or its precursors are unlikely to be the primary cause of the elevated resistance. Rather, they seem to amplify the effect of some other, primary causes acting differently on smaller (in chronic hypoxia) and larger (in diabetes) fetoplacental vessels. Judging from the analogy with similar processes in the pulmonary circulation, mast cell [32e34] and macrophage [35,36] activation and related alterations in connective tissue of vascular wall [37e40] can be involved. Interestingly, maternal plasma nitrotyrosine concentrations were affected by our experimental interventions quite differently than the fetal levels. Apparently, the mothers were able to maintain normal plasma nitrotyrosine during both diabetes and chronic hypoxia. This defense was overwhelmed only when the two adverse stimuli (diabetes and chronic hypoxia) acted simultaneously - only then the maternal plasma nitrotyrosine was elevated. This finding partly contradicts our previous report that plasma nitrotyrosine is increased in adult rats within the first few days of chronic hypoxia [41]. It is possible that this could be due to increased antioxidant defense in pregnancy. Quite a few studies have found increased plasma nitrotyrosine in adults with diabetes, although unaltered levels were also reported [reviewed in 42]. Our data indicate that the elevated plasma nitrotyrosine levels in the fetus in diabetes, chronic hypoxia, and their combination are not a passive consequence of maternal levels, but rather are independently regulated and can contribute to the functional changes of

O. Vajnerova et al. / Placenta 63 (2018) 32e38

the fetoplacental vasculature. By an unknown mechanism, high glucose concentration in the perfusate elevates fetoplacental vascular resistance [43]. In our experiment, the placentae from both diabetic and control pregnancies were perfused with zero glucose perfusate, and the vasoconstrictor response to acute hypoxia was reduced. Thus, the differences that we have observed among the groups were caused by the chronic conditions before the actual measurements rather than by the acute effect of glucose during the perfusion. We observed a similar reduction in fetoplacental vasoconstrictor reactivity to acute hypoxic stimuli by diabetes, chronic hypoxia, and their combination (Fig. 1B). While the mechanism of hypoxic vasoconstriction is little characterized in the placenta, in the lung the only other organ where hypoxia causes vasoconstriction - it has been shown that increased glucose during acute hypoxic vasoconstriction reduces the response [44] and vice versa - glucose-free perfusion potentiates it [45]. Pulmonary vasoconstrictor reactivity to a pharmacological stimulus (U-46619) is reduced in rats with experimental diabetes regardless of glucose concentration in the perfusate [29]. In the fetoplacental vasculature, there is a report of reduced both vasoconstrictor and vasodilator reactivity in human diabetes [46]. Thus, our study is in line with these previous findings and extends them to the specific, non-pharmacological stimulus of acute hypoxia. Also, since we have found increased fetal plasma nitrotyrosine levels in the groups whose fetoplacental vascular reactivity to acute hypoxia was decreased (H, D, and DH), our study is in agreement with that of Kossenjans et al. [46] implying increased peroxynitrite might be the mechanism of reduced fetoplacental vascular reactivity. In contrast to the present study, we have found HFPV to be increased by chronic hypoxia in our previous report [1]. One possible explanation of this discrepancy could be the repeated injections of angiotensin II prior to hypoxic challenges in the previous study that were not done in the present study. The pre-existing level of vascular tone is a well-known factor in the magnitude of vasoreactivity in the lungs [28] and seems to play a role in the fetoplacental vessels as well [26]. There is considerable literature on the effects of diabetes on the uterine arteries - the maternal side of the placenta [for review, see e.g. 47]. As far as the vasculature of the fetal side of the placenta is concerned, the available data are more scarce and conflicting. Some reports show reduced umbilical blood flow in diabetic animals [48,49]. In humans, umbilical blood flow is more likely to be reduced by diabetes when glycemia is poorly controlled as compared to well-controlled blood glucose levels [50e52]. On the other hand, there are studies where abnormal fetoplacental blood flow was observed only in a minority of human diabetic pregnancies [53,54]. However, umbilical blood flow reflects both the vascular resistance of the placenta and the umbilicus. The latter was not assessed in the present study. One limitation of the present study is that it does not consider the possibility of gender variability. There is good evidence that there are sex differences in placental vascular function [55e57], as well as in the vascular effects of diabetes [58e60]. We have shown previously quite marked effect of gender on the consequences of perinatal hypoxia on pulmonary vessels [61]. However, the scatter of values in most of our measurements is relatively small, suggesting that sex differences might not be large (with the possible exception in the case of acute hypoxic reactivity). Nevertheless, this issue may deserve further exploration. Grants The study was supported by the Grant Agency of the Czech Republic 13-01710S and by COST LD 14 068.

37

Disclosures No conflicts of interest, financial or otherwise, are declared by all authors. Acknowledgements The authors thank prof. Jan Herget for the invaluable help in experiment planning and further discussions. They also thank Ms.  for excellent technical assistance. Andrea Trnkova References , K. Venclíkova , A. Trnkov [1] V. Jakoubek, J. Bíbova a, J. Herget, V. Hampl, Chronic hypoxia increases fetoplacental vascular resistance and vasoconstrictor reactivity in the rat, Am. J. Physiol. Heart Circ. Physiol. 294 (2008) H1638eH1644. [2] D. Brodsky, H. Christou, Current concepts in intrauterine growth restriction, J. Intensive Care Med. 19 (2004) 307e319. [3] V. Hampl, V. Jakoubek, Regulation of fetoplacental vascular bed by hypoxia, Physiol. Res. 58 (2009) S87eS93. [4] J.Q. Liu, I.N. Zelko, E.M. Erbynn, J.S.K. Sham, R.J. Folz, Hypoxic pulmonary hypertension: role of superoxide and NADPH oxidase (gp91phox), Am. J. Physiol. Lung Cell. Mol. Physiol. 290 (2006) L2eL10. , V. Hampl, J. Herget, N-ace[5] V. Lachmanov a, O. Hnili ckov a, V. Povýsilova tylcysteine inhibits hypoxic pulmonary hypertension most effectively in the initial phase of chronic hypoxia, Life Sci. 77 (2005) 175e182. [6] M. Chovanec, J. Novotn a, J. Wilhelm, V. Hampl, M. Vízek, J. Herget, Hypercapnia attenuates hypoxic pulmonary hypertension by inhibiting lung radical injury, Physiol. Res. 58 (2009) S79eS85. [7] C.L. DeSisto, S.Y. Kim, A.J. Sharma, Prevalence estimates of gestational diabetes mellitus in the United States, Pregnancy Risk Assessment Monitoring System (PRAMS), 2007-2010, Prev. Chronic Dis. 11 (2014) E104. [8] A. Vambergue, I. Fajardy, Consequences of gestational and pregestational diabetes on placental function and birth weight, World J. Diabetes 2 (2011) 196e203. [9] J.L. Mills, Malformations in infants of diabetic mothers, Birth Defects Res. A Clin. Mol. Teratol. 88 (2010) 769e778. [10] I.F. Casson, C.A. Clarke, C.V. Howard, O. McKendrick, S. Pennycook, P.O.D. Pharoah, M.J. Platt, M. Stanisstreet, D. van Velszen, S. Walkinshaw, Outcomes of pregnancy in insulin dependent diabetic women: results of a five year population cohort study, BMJ 315 (1997) 275e278. [11] F. Giacco, M. Brownlee, Oxidative stress and diabetic complications, Circ. Res. 107 (2010) 1058e1070. [12] D.R. Salvesen, J.M. Brudenell, R.J. Snijders, R.M. Ireland, K.H. Nicolaides, Fetal plasma erythropoietin in pregnancies complicated by maternal diabetes mellitus, Am. J. Obstet. Gynecol. 168 (1993) 88e94. [13] E. Taricco, T. Radaelli, G. Rossi, M.S. Nobile de Santis, G.P. Bulfamante, L. Avagliano, I. Cetin, Effects of gestational diabetes on fetal oxygen and glucose levels in vivo, BJOG 116 (2009) 1729e1735. [14] Y.K. Sinzato, G.T. Volpato, I.L. Iessi, A. Bueno, M. Calderon Ide, M.V. Rudge, D.C. Damasceno, Neonatally induced mild diabetes in rats and its effect on maternal, placental, and fetal parameters, Exp. Diabetes Res. 2012 (2012), 108163. [15] R. Rasch, Control of blood glucose levels in the streptozotocin diabetic rat using a long-acting heat-treated insulin, Diabetologia 16 (1979) 185e190. [16] V. Hampl, J. Herget, Perinatal hypoxia increases hypoxic pulmonary vasoconstriction in adult rats recovering from chronic exposure to hypoxia, Am. Rev. Respir. Dis. 142 (1990) 619e624. [17] P. Kafka, O. Vajnerov a, J. Herget, V. Hampl, Rho-kinase inhibition attenuates acute hypoxic fetoplacental vasoconstriction in the rat, Physiol. Res. 61 (2012) S43eS48. [18] B. Halliwell, What nitrates tyrosine? Is nitrotyrosine specific as a biomarker of peroxynitrite formation in vivo? FEBS Lett. 411 (1997) 157e160. rkov sek, D. Mikova , M. Vízek, Hyperoxia attenuated nitro[19] B. Fisa a, R. Vyta tyrosine concentration in the lung tissue of rats with experimental pneumonia, Physiol. Res. 53 (2004) 487e492. [20] G.L. Nielsen, M. Moller, H.T. Sorensen, HbA1c in early diabetic pregnancy and pregnancy outcomes: a Danish population-based cohort study of 573 pregnancies in women with type 1 diabetes, Diabetes Care 29 (2006) 2612e2616. [21] T. Ulmannova, D. Bartaskova, I. Spalova, P. Zoban, Z. Vesely, K. Stechova, Maternal BMI and HDL as predictors of pregnancy outcome in women with type 1 diabetes, J. Matern. Fetal Neonatal Med. 27 (2014) 1580e1583. [22] E.K. Weir, H.L. Reeve, J.M.C. Huang, E. Michelakis, D.P. Nelson, V. Hampl, S.L. Archer, Anorexic agents aminorex, fenfluramine, and dexfenfluramine inhibit potassium current in rat pulmonary vascular smooth muscle and cause pulmonary vasoconstriction, Circulation 94 (1996) 2216e2220. mek, J. Valenta, [25] J. Herget, V. Hampl, Pulmonary circulation, in: B.B. Sra F. Klimes (Eds.), Biomechanics of the Cardiovascular System, Czech Technical University Press, Prague, 1995, pp. 327e336. , Z. Stran a k, X. Wu, E.D. Michelakis, K. Hashimoto, [26] V. Hampl, J. Bíbova S.L. Archer, Hypoxic fetoplacental vasoconstriction in humans is mediated by

38

[27]

[28] [29]

[30]

[31]

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39] [40]

[41]

[42] [43]

O. Vajnerova et al. / Placenta 63 (2018) 32e38 potassium channel inhibition, Am. J. Physiol. Heart Circ. Physiol. 283 (2002) H2440eH2449. C.A. Dawson, J.H. Linehan, T.A. Bronikowski, D.A. Rickaby, Pulmonary microvascular hemodynamics: occlusion methods, in: J.A. Will, C.A. Dawson, E.K. Weir, C.K. Buckner (Eds.), The Pulmonary Circulation in Health and Disease, Academic Press, Orlando, 1987, pp. 175e197. V. Hampl, J. Herget, Role of nitric oxide in the pathogenesis of chronic pulmonary hypertension, Physiol. Rev. 80 (2000) 1337e1372. R.D. Russ, B.W. Tobin, Alteration of segmental vascular resistance in the pulmonary circulation of diabetic rats, Can. J. Physiol. Pharmacol. 74 (1996) 1010e1015. V. Jakoubek, J. Bíbov a, V. Hampl, Voltage-gated calcium channels mediate hypoxic vasoconstriction in the human placenta, Placenta 27 (2006) 1030e1033. I.M. Calderon, D.C. Damasceno, R.L. Amorin, R.A. Costa, M.A. Brasil, M.V. Rudge, Morphometric study of placental villi and vessels in women with mild hyperglycemia or gestational or overt diabetes, Diabetes Res. Clin. Pract. 78 (2007) 65e71.  asova , H. Maxova , V. Hampl, M. Vízek, V. Povýsilova , J. Novotn A. Ban a, O. Vajnerov a, O. Hnili ckov a, J. Herget, Prevention of mast cell degranulation by disodium cromoglycate attenuates the development of hypoxic pulmonary hypertension in rats exposed to chronic hypoxia, Respiration 76 (2008) 102e107. A. Tucker, I.F. McMurtry, A.F. Alexander, J.T. Reeves, R.F. Grover, Lung mast cell density and distribution in chronically hypoxic animals, J. Appl. Physiol. 42 (1977) 174e178. sek, V. Lachmanov dova , J. Novotn L. Vajner, R. Vyta a, J. Uhlík, V. Konra a, V. Hampl, J. Herget, Acute and chronic hypoxia as well as 7-day recovery from chronic hypoxia affects the distribution of pulmonary mast cells and their MMP-13 expression in rats, Int. J. Exp. Pathol. 87 (2006) 383e391.   sek, O. Vajnerov , M. Vízek, V. Hampl, M. Zaloudíkov a, R. Vyta a, O. Hnili ckova J. Herget, Depletion of alveolar macrophages attenuates hypoxic pulmonary hypertension but not hypoxia-induced increase in serum concentration of MCP-1, Physiol. Res. 65 (2016) 763e768. M.G. Frid, J.A. Brunetti, D.L. Burke, T.C. Carpenter, N.J. Davie, J.T. Reeves, M.T. Roedersheimer, N. van Rooijen, K.R. Stenmark, Hypoxia-induced pulmonary vascular remodeling requires recruitment of circulating mesenchymal precursors of a monocyte/macrophage lineage, Am. J. Pathol. 168 (2006) 659e669. , J. Herget, Exposure to chronic hypoxia induces qualitative changes J. Novotna of collagen in the walls of peripheral pulmonary arteries, Life Sci. 62 (1998) 1e12. , J. Bíbov  kova , V. Hampl, MetalJ. Herget, J. Novotna a, V. Povýsilov a, M. Van loproteinase inhibition by Batimastat attenuates pulmonary hypertension in chronically hypoxic rats, Am. J. Physiol. Lung Cell. Mol. Physiol. 285 (2003) L199eL208. M. Rabinovitch, Molecular pathogenesis of pulmonary arterial hypertension, J. Clin. Invest. 122 (2012) 4306e4313. G.J. Poiani, C.A. Tozzi, S.E. Yohn, R.A. Pierce, S.A. Belsky, R.A. Berg, S.Y. Yu, S.B. Deak, D.J. Riley, Collagen and elastin metabolism in hypertensive pulmonary arteries of rats, Circ. Res. 66 (1990) 968e978. , J. Tkaczyk, H. Maxov D. Hodyc, E. Johnson, A. Skoumalova a, M. Vízek, J. Herget, Reactive oxygen species production in the early and later stage of chronic ventilatory hypoxia, Physiol. Res. 61 (2012) 145e151. X.L. Wang, D.L. Rainwater, A. Leone, M.C. Mahaney, Effects of diabetes on plasma nitrotyrosine levels, Diabet. Med. 21 (2004) 577e580. J.B. Roth, J.A. Thorp, S.M. Palmer, P.C. Brath, S.W. Walsh, S.S. Crandell, Response of placental vasculature to high glucose levels in the isolated human

placental cotyledon, Am. J. Obstet. Gynecol. 163 (1990) 1828e1830. [44] S.S. Rounds, I.F. McMurtry, J.T. Reeves, Glucose metabolism accelerates the decline of hypoxic vasoconstriction in rat lungs, Respir. Physiol. 44 (1981) 239e249. [45] H.S. Stanbrook, I.F. McMurtry, Inhibition of glycolysis potentiates hypoxic vasoconstriction in rat lungs, J. Appl. Physiol. 55 (1983) 1467e1473. [46] W. Kossenjans, A. Eis, R. Sahay, D. Brockman, L. Myatt, Role of peroxynitrite in altered fetal-placental vascular reactivity in diabetes or preeclampsia, Am. J. Physiol. Heart Circ. Physiol. 278 (2000) H1311eH1319. [47] L. Leach, A. Taylor, F. Sciota, Vascular dysfunction in the diabetic placenta: causes and consequences, J. Anat. 215 (2009) 69e76. [48] U.J. Eriksson, L. Jansson, Diabetes in pregnancy: decreased placental blood flow and disturbed fetal development in the rat, Pediatr. Res. 18 (1984) 735e738. [49] S.S. Crandell, D.J. Fisher, F.H.J. Morriss, Effects of ovine maternal hyperglycemia on fetal regional blood flows and metabolism, Am. J. Physiol. Endocrinol. Metab. 249 (1985) E454eE460. [50] L. Bracero, H. Schulman, A. Fleischer, G. Farmakides, B. Rochelson, Umbilical artery velocimetry in diabetes and pregnancy, Obstet. Gynecol. 68 (1986) 654e658. [51] L. Nylund, N.O. Lunell, R. Lewander, B. Persson, B. Sarby, Uteroplacental blood flow in diabetic pregnancy: measurements with indium 113m and a computer-linked gamma camera, Am. J. Obstet. Gynecol. 144 (1982) 298e302. [52] D.R. Salvesen, M.T. Higueras, C.A. Mansur, J. Freeman, J.M. Brudenell, K.H. Nicolaides, Placental and fetal Doppler velocimetry in pregnancies complicated by maternal diabetes mellitus, Am. J. Obstet. Gynecol. 168 (1993) 645e652. [53] M. Pietryga, J. Brazert, E. Wender-Ozegowska, M. Dubiel, S. Gudmundsson, Placental Doppler velocimetry in gestational diabetes mellitus, J. Perinat. Med. 34 (2006) 108e110. [54] G.M. Fadda, D. D'Antona, G. Ambrosini, P.L. Cherchi, G.B. Nardelli, G. Capobianco, S. Dessole, Placental and fetal pulsatility indices in gestational diabetes mellitus, J. Reprod. Med. 46 (2001) 365e370. [55] N.B. Ojeda, D. Grigore, L.L. Yanes, R. Iliescu, E.B. Robertson, H. Zhang, B.T. Alexander, Testosterone contributes to marked elevations in mean arterial pressure in adult male intrauterine growth restricted offspring, Am. J. Physiol. Regul. Integr. Comp. Physiol. 292 (2007) R758eR763. [56] T. Prior, M. Wild, E. Mullins, P. Bennett, S. Kumar, Sex specific differences in fetal middle cerebral artery and umbilical venous Doppler, PLoS One 8 (2013), e56933. [57] A. Ghidini, C.M. Salafia, Gender differences of placental dysfunction in severe prematurity, Br. J. Obstet. Gynaecol. 112 (2005) 140e144. [58] J.G. Regensteiner, S. Golden, A.G. Huebschmann, E. Barrett-Connor, A.Y. Chang, D. Chyun, C.S. Fox, C. Kim, N. Mehta, J.F. Reckelhoff, J.E. Reusch, K.M. Rexrode, A.E. Sumner, F.K. Welty, N.K. Wenger, B. Anton, L. American Heart Association Diabetes Committee of the Council on, C.o.E. Cardiometabolic Health, C.o.F.G. Prevention, B. Translational, H. Council on, Sex differences in the cardiovascular consequences of diabetes mellitus: a scientific statement from the American Heart Association, Circulation 132 (2015) 2424e2447. [59] C. Maric-Bilkan, Sex differences in micro- and macro-vascular complications of diabetes mellitus, Clin. Sci. 131 (2017) 833e846. [60] A. Kautzky-Willer, J. Harreiter, G. Pacini, Sex and gender differences in risk, pathophysiology and complications of type 2 diabetes mellitus, Endocr. Rev. 37 (2016) 278e316. , I. Ost a dalov , J. Herget, Gender differences [61] V. Hampl, J. Bíbova a, V. Povýsilova in the long-term effects of perinatal hypoxia on the pulmonary circulation in rats, Am. J. Physiol. Lung Cell. Mol. Physiol. 285 (2003) L386eL392. 0