*2.2. Micro-Hemorrhage in Fetal Brains*

Using H&E staining, red blood cells can be visualized by their red staining in tissues. Figure 2A shows the location of brain slices used to quantify micro-bleeds. Figure 2B,C show representative micro-bleeds observed in the parenchyma (2B, cortex) and lateral and third ventricles (2C).

**Figure 2.** Placental ischemia exposure leads to increased number of micro-bleeds in brains of exposed fetuses. ( **A**) Schematic of the regions where coronal sections were collected. (**B**) Representative images of fetal brains showing micro-bleeds in the cortex and ( **C**) ventricles. Number of micro-bleeds in the (**D**) anterior (**E**) posterior slices of fetal brains. Points represent average micro-bleeds from 1–2 pups per dam (n = 5 dams per group). Mean ± SEM is also depicted. Data were analyzed using unpaired *t*-test and *p*-values are indicated.

We counted the number of micro-bleeds and found significantly higher numbers of micro-hemorrhages in the brains of fetuses exposed to placental ischemia in the posterior sections, and a trend for increased micro-bleeds in the anterior slices of the brain. In the anterior slices, RUPP-exposed fetuses had 3.4 ± 0.8 bleeds compared to the sham-exposed (1.8 ± 0.4 bleeds; *p* = 0.056; Figure 2D). In the posterior slices, sham-exposed fetuses had 4 ± 1 while RUPP-exposed fetuses had 7 ± 1 bleeds (*p* = 0.026, Figure 2E). Thus, exposure to placental ischemia almost doubled the incidence of fetal brain micro-bleeds *in utero*.

### *2.3. Inflammatory Profile in Fetal Brains*

Due to placental ischemia inducing increased maternal, circulating and placental inflammatory cytokines [20,21,27], we hypothesized that the developing fetal brain may mirror the maternal pro-inflammatory environment. We therefore measured the levels of cytokines/chemokines in fetal

brains exposed to normal pregnancy and placental ischemia. Out of 27 cytokines and chemokines, seven were undetectable or observed in only 1–2 fetal brains per group: EGF, G-CSF, GM-CSF, GRO/KC, IL-2, IL-5, and IL-13. We divided the remaining 20 cytokines/chemokines into pro-inflammatory/ cytotoxic, anti-inflammatory, and chemokines/growth factors [28]. The pro-inflammatory cytokines IL-1β, IL-6, and IL-18 increased significantly in fetal brains from placental ischemia-exposed pregnancies (Figure 3).

**Figure 3.** Placental ischemia leads to a shift towards a pro-inflammatory status in brains of exposed fetuses. A rat multiplex kit array of 27 cytokine/chemokine was used. Values were normalized to protein concentration. IL-1β, IL-6, and IL-18 increased significantly in fetal brains exposed to placental ischemia. Blue points represent extrapolated values (one value below the lowest detectable value and normalized to protein concentration). Not shown are: IL-2, IL-5, and IL-13. Values for individual rat fetuses (*n* = 5 fetuses per group) are shown along with the mean ± SEM. Only one fetal brain was collected per dam for cytokines/chemokines.

There was a trend toward increased anti-inflammatory cytokines, IL-4 and IL-10. Lastly, the chemokines/growth factors eotaxin (CCL11), LIX/CXCL5, and MIP-2/CXCL2 increased significantly in fetal brains exposed to placental ischemia compared to sham-exposed (Figure 4).

These data demonstrate that placental ischemia exposure induces a pro-inflammatory environment in fetal brains *in utero*. Whether the cerebral inflammatory profile persists in the postnatal period is unknown. Thus, future studies will determine whether the fetal cerebral pro-inflammatory status is unique to the *in utero* environment or whether it persists postnatally.

 Because micro-hemorrhages are associated with a pro-inflammatory environment [29], we performed correlations to identify whether any factors were strongly associated with the number of micro-bleeds observed. Comparing fetuses from the same dam, we found that cerebral tissue IL-6 levels were positively associated with the number of micro-bleeds detected (Figure 5A; *r* = 0.673; *p* = 0.017). Surprisingly, although there were no differences in brain water content between the groups, fetal brain water content was negatively associated with the number of micro-bleeds (Figure 5B; r = −0.672; *p* = 0.017).


**Figure 4.** Changes in chemokines/growth factors in fetal brains. Placental ischemia exposure increased eotaxin, LIX, and MIP2 levels. Not shown are: EGF, G-CSF, GM-CSF, and GRO/KC. Blue points represent extrapolated values (one value below lowest detectable value and normalized to protein concentration). Values for individual fetuses (*n* = 5 fetuses per group) are shown along with the mean ± SEM. Data were analyzed using unpaired *t*-tests, and *p*-values are indicated.

**Figure 5.** Correlation between number of micro-bleeds and IL-6 levels or brain water content. The number of micro-bleeds is (**A**) positively associated with fetal brain IL-6 levels and (**B**) negatively associated with fetal brain water content. Fetuses from same dam were used for the association analysis. Values for individual rats (*n* = 5 fetuses per group) are shown. Relationships between factors were analyzed using the Pearson correlation.

### *2.4. Microglia Changes in Fetal Brains*

Microglia are key producers of cytokines in the brain, so we assessed changes in microglial density and morphology in the brains of exposed fetuses. As reviewed in [30], microglia migrate from the ventricles and meninges during development, making the sub-ventricular zone (SVZ) the ideal region to quantify changes in microglial density. Representative images of a brain section and the third ventricle from sham and placental ischemia-exposed fetuses are shown in Figure 6A, B. We found fewer Iba1 + cells in the SVZ of the third ventricle (40 ± 3 in shams versus 19 ± 4; *p* = 0.003), and this reduction was observed both in the open (23 ± 2 in sham versus 13 ± 4; *p* = 0.036) and closed (21 ± 1 in shams versus 9 ± 2; *p* = 0.005) portion of the third ventricle (Figure 6C). We further characterized the microglia based on morphology into primitive ramified or amoeboid microglia [31] (Figure 6D) and counted these separately. We found a significant decrease in primitive ramified microglia in fetuses exposed to RUPP (13 ± 4 versus 28 ± 2 in Sham, *p* = 0.007; Figure 6D) and no change in the number of amoeboid microglia (13 ± 4 in Sham versus 5 ± 2 in RUPP, *p* = 0.089). We found no significant difference in cortical plate thickness (a potential indicator of neuronal density) between the fetuses exposed to normal or placental ischemic pregnancies (Figure 6E,F).

**Figure 6.** Changes in microglia at the proliferative areas (sub-ventricular zones—SVZ) of the 3rd Ventricle. (**A**) Representative image showing, at the brain-level, the section where the microglial analysis was conducted. Scale bar = 100 μm. (**B**) Representative images of Iba1 staining in the open region and closed region of the 3rd ventricle. Red—Iba1+ cells; Blue—DAPI+ nuclei. Scale bar represents 50 μm. (**C**) Decreased number of microglia in the SVZ of pups exposed to placental ischemia. (**D**) Examples of different microglia types observed (scale bar = 20 μm) and quantification of primitive ramified and amoeboid microglia in SVZ. (**E**) Representative images showing the different cortical layers in the embryonic day (E19) pup brain. Scale bar represents 100 μm. (**F**) Quantification of cortical plate thickness in fetuses exposed to sham or placental ischemia. Values from 1–2 pups were averaged per dam (*n* = 4–5 per group) and shown along with the mean ± SEM. Data were analyzed using unpaired *t*-tests, and *p*-values are indicated. MZ—marginal zone, CP—cortical plate, SP—subcortical plate, SVZ—sub-ventricular zone, VZ—ventricular zone, and LV—lateral ventricle.
