Fetal Brain-Derived Exosomal miRNAs from Maternal Blood: Potential Diagnostic Biomarkers for Fetal Alcohol Spectrum Disorders (FASDs)

Abstract

Fetal alcohol spectrum disorders (FASDs) are leading causes of neurodevelopmental disability but cannot be diagnosed early in utero. Because several microRNAs (miRNAs) are implicated in other neurological and neurodevelopmental disorders, the effects of EtOH exposure on the expression of these miRNAs and their target genes and pathways were assessed. In women who drank alcohol (EtOH) during pregnancy and non-drinking controls, matched individually for fetal sex and gestational age, the levels of miRNAs in fetal brain-derived exosomes (FB-Es) isolated from the mothers' serum correlated well with the contents of the corresponding fetal brain tissues obtained after voluntary pregnancy termination. In six EtOH-exposed cases and six matched controls, the levels of fetal brain and maternal serum miRNAs were quantified on the array by qRT-PCR. In FB-Es from 10 EtOH-exposed cases and 10 controls, selected miRNAs were quantified by ddPCR. Protein levels were quantified by ELISA. There were significant EtOH-associated reductions in the expression of several miRNAs, including miR-9 and its downstream neuronal targets BDNF, REST, Synapsin, and Sonic hedgehog. In 20 paired cases, reductions in FB-E miR-9 levels correlated strongly with reductions in fetal eye diameter, a prominent feature of FASDs. Thus, FB-E miR-9 levels might serve as a biomarker to predict FASDs in at-risk fetuses.

Keywords: brain development; exosomes; fetal alcohol syndrome; fetal eye; miR-9.

Figure 1

Effects of EtOH exposure on the normal reductions in miRNA expressions in fetal brain and maternal plasma during the course of gestation. Non-EtOH-exposed human fetal brain tissues from the first and second trimesters were compared with unexposed controls with respect to the expression of 84 miRNAs, using Real-Time qRT-PCR. The results in fold-changes are displayed as heat maps. To assess the normal developmental changes, in (A), only unexposed controls were measured and expressed as fold-changes relative to their expression levels at the earliest-studied time point, 11.3 weeks GA, which is not shown because, by definition, the values equal 1. Expression of the selected miRNAs generally decreased (green) at later GAs in maternal blood (top). The pattern was less clear in the fetal brain (bottom), with expression levels at first increasing (red) but then decreasing toward the 11.3 values (black). N = 6 for each group. In (B), each box represents the average change in expression for a specific miRNA in EtOH-exposed cases relative to their unexposed individually matched controls. N = 6 paired cases for each group, including the 11.3-week time point in (A). In the second trimester, EtOH exposure was associated with a significant upregulation (red) of target miRNAs in the maternal blood (top) but a downregulation (green) in the fetal brain (bottom). (C) The position of each studied miRNA and housekeeping control gene in the 96-well array is shown (all controls and SNORDs for normalization were positioned in row H from H1 to H12). hsa: human, Homo sapiens; miR-9-3p vs. miR-9-5p: The 5p strand is present in the forward (5′-3′) sense, while the 3p strand is the reverse sense complementary strand.

Figure 1

Effects of EtOH exposure on the normal reductions in miRNA expressions in fetal brain and maternal plasma during the course of gestation. Non-EtOH-exposed human fetal brain tissues from the first and second trimesters were compared with unexposed controls with respect to the expression of 84 miRNAs, using Real-Time qRT-PCR. The results in fold-changes are displayed as heat maps. To assess the normal developmental changes, in (A), only unexposed controls were measured and expressed as fold-changes relative to their expression levels at the earliest-studied time point, 11.3 weeks GA, which is not shown because, by definition, the values equal 1. Expression of the selected miRNAs generally decreased (green) at later GAs in maternal blood (top). The pattern was less clear in the fetal brain (bottom), with expression levels at first increasing (red) but then decreasing toward the 11.3 values (black). N = 6 for each group. In (B), each box represents the average change in expression for a specific miRNA in EtOH-exposed cases relative to their unexposed individually matched controls. N = 6 paired cases for each group, including the 11.3-week time point in (A). In the second trimester, EtOH exposure was associated with a significant upregulation (red) of target miRNAs in the maternal blood (top) but a downregulation (green) in the fetal brain (bottom). (C) The position of each studied miRNA and housekeeping control gene in the 96-well array is shown (all controls and SNORDs for normalization were positioned in row H from H1 to H12). hsa: human, Homo sapiens; miR-9-3p vs. miR-9-5p: The 5p strand is present in the forward (5′-3′) sense, while the 3p strand is the reverse sense complementary strand.

Figure 2

Exposure to EtOH is associated with changes in the expression of eight neurological disorder-related miRNAs. Expressions of candidate miRNAs from Table 4 were assayed by qRT-PCR on samples of maternal serum (A) and fetal brain homogenates (B) from 12 pregnancies and their fetal GA- and sex-matched controls. Results were expressed as fold change between each EtOH-exposed case and its control. An overall upregulation in miRNA expression was seen in the second trimester compared to the first in maternal serum but not in the fetal brain. The expression of miR-509 was almost unaltered, so it is not included in this graph.

Figure 3

Exposure to EtOH is associated with reduced expression of miR-9 and miR-132 in FB-Es. (A) Levels of miRNA were measured by qRT-PCR in FB-Es of 20 EtOH-exposed maternal plasmas and compared with 20 individually matched unexposed controls, 10 matched pairs from the first trimester and 10 from the second. Expressions of miR-9 (A) and miR-132 (B) were normalized to SNORD, according to miScript Qiagen recommendations. Each assay was performed in triplicate and averaged. Changes in EtOH-exposed cases were expressed in comparison with their fetal sex- and GA-matched controls, measured relative to SNORD, and then their averages were assigned a value of 1. Error bars indicate the standard deviations for the 20 averages per group. Significance levels are for the comparison between all EtOH-exposed and all unexposed controls, based on ANOVA. * p < 0.05, ** p < 0.01.

Figure 4

Prenatal exposure to EtOH is associated with inhibition of miR-9 expression in FB-Es. Absolute quantification of miR-9 in FB-Es isolated from the blood of 10 EtOH-exposed pregnant women and their individually matched controls was determined by measuring RNA copy numbers using ddPCR. Five ng of exosomal RNA (including miRNAs) was used in the reaction. Each assay was performed in triplicate. Control FB-Es contained 200–1200 copies of miR-9 per μL, while in EtOH-exposed FB-Es, 180–820 copies were found. In each case, the EtOH-exposed sample contained fewer copies of miR-9 than its GA-and sex-matched control (*** p = 0.006).

Figure 5

Potential targets of miRNA-9 in double-negative feedback loop pathways observed in FB-Es. The cargo of FB-Es from 20 pregnant women who had consumed EtOH and their individually matched controls were analyzed by ELISA and quantitative Western blot analysis (qWestern; (A–C)). Relative downregulation was observed quantitatively by qWestern for Synapsin (A), REST (B), and Shh (C) and by ELISA for BDNF (D). * p < 0.05, ** p < 0.01. (E). Proposed double-negative feedback loop pathways for miR-9 expression. In the brain, miR-9 inhibits transcription of the transcription factor REST, which in turn inhibits the expression of miR-9. REST also inhibits the translation of BDNF, which in turn inhibits the expression of miR-9, as well as upregulating Synapsin and reciprocally upregulating Shh. Activation of the Shh pathway induces an increase in BDNF expression and results in neuroprotection to oxidative stress.

Figure 6

Reductions in FB-E miR-9 levels correlate with the reductions in eye diameter in a larger sample of fetuses exposed to EtOH. To enhance the potential practicality of FB-E miR-9 levels as a biomarker to predict FASDs, FB-Es were isolated from the blood of a larger group of mothers who consumed alcohol during pregnancy and their unexposed controls, matched only for GA from 9–23 weeks, disregarding other control factors, such as fetal sex. FB-E miR-9 levels were measured by ddPCR. Eye diameters were measured in histological sections of the fetuses. (A). Copy numbers of miR-9 were reduced by almost 37.5% in EtOH-exposed cases. (B). Each individual FB-E miR-9 level from the 40 EtOH-exposed fetuses is graphed next to its GA-matched control and arrayed in order of GA. In all but the two youngest cases, the EtOH-exposed levels were higher than those of their GA-matched controls. Note also that the control levels of miR-9 rose dramatically early in the second trimester, whereas the miR-9 levels of the EtOH-exposed group did not show this increase. The assay was performed in triplicate. The dashed green line separates the first-trimester cases (on the left) from the second-trimester cases (on the right). (C). The correlation between the reduction in eye size (difference between EtOH-exposed fetus and its paired control) and the reduction in exosomal miR-9 levels is presented as a scatter plot. First-trimester (9 to 14 weeks GA) and second-trimester pregnancies (14.1 to 23 weeks GA) are graphed together. *** p << 0.001 (p = 0.000000464825), based on Spearman’s correlation with exact two-tailed critical p values. Data in (D) are presented in %.