In the present work, we obtained several novel findings using a baboon model of pregnancy. First, we established that in vitro pressurized cerebral arteries harvested from near-term fetuses have an ability to develop tone and to respond to a non-specific contractile agent (high KCl) in a matter that is similar to responses of maternal arteries. With regard to modifications of a cerebral artery diameter by toxicologically relevant levels of alcohol, we documented that the up-regulation of fetal cerebral artery CB2 receptor function is limited to mid-pregnancy. This up-regulation was not observed at near-term. Lastly, we documented that ethanol modification of a fetal cerebral artery diameter depends on the gestational age of the fetuses. In particular, CB receptor-mediated dilation of fetal cerebral arteries by ethanol that was shown to take place during mid-gestation [8], was not present near-term, and the contribution of CB receptors to ethanol effect on a cerebral artery diameter was lost.

The fact that near-term fetal cerebral arteries of baboon exhibited consistent autoregulation (myogenic tone) and constriction in response to KCl similar to those of adult maternal arteries (Figure 1) was quite surprising, as perinatal cerebral circulation in other species, including humans, often exhibits higher inter-individual variability and vulnerability to stressors when compared to that of adult individuals [25]. Higher variability in fetal organ responses to pharmacological modulators is attributed to the complex course of developmental curves of proteins and signaling molecules that control myogenic tone in cerebral arteries [26,27,28,29]. Interestingly, we detected a qualitative difference in KCl-induced constriction of fetal arteries when compared to maternal ones. Indeed, in maternal arteries, the KCl effect was often biphasic, with subtle dilation preceding a development of constriction (Figure 1C). The similar biphasic effect of 60 mM KCl was observed in other species [18,19] and could be attributed to the vasodilatory effect of low KCl concentrations that was reported in coronary and cerebral arteries [30]. In a pressurized artery technique, experimental solutions are introduced into an artery perfusion chamber gradually; it takes several minutes to substitute physiologic saline with high KCl solution. Thus, the initial phase of KCl effect, i.e., vasodilation, could be explained by low KCl targeting inward rectifier K⁺ channels [30]. The lack of biphasic responses to KCl presence in near-term fetal arteries suggests the lack of inward rectifier K⁺ channel involvement into an artery diameter regulation at this gestational stage.

With regard to the eCB system, it has been widely recognized as a critical player in fetal development [31,32]. Recently, we have detected up-regulation of vasodilatory responses to anandamide when probed in vitro at mid-pregnancy following three repeated episodes of maternal alcohol intoxication [8]. The current study aimed to determine whether up-regulation of vasodilatory responses of fetal cerebral arteries to anandamide was long lasting and could be observed near-term following repeated episodes of maternal alcohol intoxication during the mid-pregnancy (Material and Methods). We did not detect statistically significant differences in anandamide-driven changes of a cerebral artery diameter between near-term fetal and maternal arteries (Figure 2). This result suggests that the alcohol-triggered modification in cerebral artery response to anandamide that we observed previously [8] is transient. Notably, the transient nature of alcohol-driven re-shaping of eCB system components has been recently reported in male rat entorhinal cortex and hippocampus [33]. In contrast, another recent study has documented life-long changes in arterial function following prenatal (fetal) alcohol exposure [7]. In particular, fetal alcohol exposure alters age-dependent changes in blood velocity within carotid artery and impairs post-stroke behavioral recovery in adult mice [7]. Thus, besides transient changes in fetal artery eCB system, alcohol is expected to target other signaling pathways that result in permanent changes in arterial function. The mechanisms that underlie the transient nature of alcohol-driven re-shaping of eCB system components remain unknown and may range from de-sensitization of relevant alcohol targets to the development of compensatory mechanisms that negate alcohol presence.

At the level of fetal cerebral circulation, alcohol at physiologically relevant levels (<300–400 mg/dL; [13, 34]) was consistently reported to increase fetal cerebral perfusion or, accordingly, drop fetal cerebral artery Doppler indices of cerebral blood flow velocity [5,6,10,11]. Both outcomes are consistent with dilation of fetal cerebral arteries. Indeed, we have recently shown that ethanol at concentration that is reached in amniotic fluid during alcohol intoxication in pregnant baboons, dilated in vitro pressurized cerebral arteries harvested from fetuses at the end of the second trimester equivalent [8]. In the present work, however, fetal cerebral artery dilation by ethanol vanished later in gestation, as ethanol-driven changes in a fetal cerebral artery diameter near-term (Figure 4A-B) were significantly smaller than those changes in fetal cerebral arteries harvested at the end of the mid-pregnancy (Figure 4B insert; [8]). In addition, there was no apparent dilation by ethanol in adult maternal cerebral arteries (Figure 4B). The lack of ethanol-induced dilation during the perinatal period and adulthood is consistent with other reports that describe ethanol-induced constriction of adult cerebral arteries harvested from rodent species and dogs [18,19, 35].

Reasons for the age-dependent effect of ethanol remain speculative. However, the most straightforward explanation would be the loss of relevant ethanol targets that were present during mid-pregnancy and/or emerging of new targets during pre-term stage of development. Indeed, ethanol-induced dilation of fetal cerebral arteries during mid-pregnancy was totally blunted by a mixture of CB1 and CB2 receptor blockers. Moreover, ethanol-induced dilation was turned into a measurable constriction when CB receptors were blocked [8]. To some extent, similar phenomenon (i.e. ethanol-induced constriction in presence of CB receptor blockers) was observed in several artery segments in the current work (Figure 4D). This phenomenon may be explained by the involvement of calcium-/voltage-gated potassium channel of large conductance (BK) in artery diameter regulation by ethanol. Indeed, in adult cerebral arteries, ethanol-induced constriction was shown to arise from ethanol-induced inhibition of BK channels [18]. More recently, ryanodine receptors (RyR) were also identified as highly-sensitive proteins that control cerebral myocyte contractility [37]. Notably, RyR function was shown to be similar in ovine fetal and adult cerebral arteries [27], while BK channels evolve during development [29,38]. Thus, BK channels may emerge as a relevant ethanol target at late gestational stage. The idea that eCB system within a fetal cerebral artery could lose functional significance after birth seems plausible considering that age-dependent modifications in the eCB system have been documented in several reports [39,40].

We would like to note that our study has several limitations. These limitations stem from the experimental model – non-human primates. Undoubtedly, non-human primates (including baboons) offer an invaluable advantage over other widely used laboratory animals for studies on PAE and fetal cerebral artery (reviewed by [9]). Indeed, baboon reproductive physiology and obstetrics closely resemble those of human [41]. Moreover, non-human primate overall physiological and behavioral similarities with humans make them an attractive model for alcohol researchers [42]. However, the availability of primates for experimental work is limited. Thus, sample sizes in our work are very small, and this limitation raises the issue of false-positive as well as false-negative discoveries [17]. However, several pieces of our current data point at the lack of long-lasting effect of PAE during mid-pregnancy on fetal cerebral artery eCB system. It is unlikely that such consistency could be reached with false discoveries. Another limitation is the baboons’ heterogeneous medical histories and environmental exposures. Although this heterogeneity closely resembles the diversity of human population, it also introduces variability in experimental outcomes. Thus, our study is likely underpowered to detect subtle differences between the groups.