Maternal immune activation with toll-like receptor 7 agonist during mid-gestation alters juvenile and adult developmental milestones and behavior

Abstract

Infections during pregnancy are associated with increased risk for adult neuropsychiatric disease, such as major depressive disorder, schizophrenia, and autism spectrum disorder. In mouse models of maternal immune activation (MIA), different toll-like receptors (TLRs) are stimulated to initiate inflammatory responses in mother and fetus. The goal of this study was to determine sex-dependent aspects of MIA using a TLR7/8 agonist, Resiquimod (RQ), on neurodevelopment. RQ was administered to timed-pregnant mice on embryonic day (E) 12.5. At E15, maternal/fetal plasma cytokines were measured by enzyme-linked immunosorbent assay (ELISA). Maternal cytokines interleukin (IL)-6 and IL-10 were higher while tumor necrosis factor (TNF)-α and IL-17 were lower in pregnant dams exposed to RQ. Fetal cytokines (E15) were altered at the same timepoint with fetal plasma IL-6 and IL-17 greater after RQ compared to vehicle, while IL-10 and TNF-α were higher in male fetuses but not female. Other timed-pregnant dams were allowed to give birth. MIA with RQ did not alter the female to male ratio of offspring born per litter. Body weights were reduced significantly in both sexes at birth, and over the next 5 weeks. Offspring from RQ-injected mothers opened their eyes 5 days later than controls. Similarly, female offspring from RQ-injected mothers exhibited pubertal delay based on vaginal opening 2-3 days later than control females. On the behavioral side, juvenile and adult male and female MIA offspring exhibited less social-like behavior in a social interaction test. Anhedonia-like behavior was greater in MIA adult female mice. This study provides support for sex-dependent influences of fetal antecedents for altered brain development and behavioral outputs that could be indicative of increased susceptibility for adult disorders through immune mechanisms. Future studies are needed to determine neural cellular and molecular mechanisms for such programming effects.

Keywords: behavior; depression; immune; pregnancy; stress.

Figure 1.. Resiquimod exposure during mid-gestation altered maternal serum cytokine profiles 60-hours post-injection.

RQ increased (A) IL-6 and decreased (B) IL-17A and (C) TNFα. RQ exposure further increased (D) IL-10. **p < 0.01, ***P<0.001, ****p < 0.0001. RQ = Resiquimod. VEH = Vehicle. Error bars represented in +/− SEM. N = 7–9/group.

Figure 2.. Resiquimod (RQ) exposure during mid-gestation (GD12.5) altered fetal serum cytokine levels 60-hours post-injection.

(A) IL-6 and (B) IL-17A were increased in RQ male and female fetuses, while (C) TNFα was decreased in RQ males. (D) IL-10 was only increased in RQ male, but not female fetuses. RQ = Resiquimod. VEH = Vehicle. GD = Gestation Day. *p < 0.05, **p < 0.01, ***p < 0.001, ****p<0.0001. Error bars represented in +/− SEM. N = 7–9/group.

Figure 3.. Exposure to maternal immune activation led to decreased body weights

(A) in males at P0, 7, 21, and 28 indicated by (*) p < 0.05, *** < 0.001 vs RQ and females at P7, 21, and 28 indicated by (+) p < 0.05 vs. RQ. (B) RQ exposure did not alter number of male:female offspring ratio in litters. P = Postnatal day. RQ = Resiquimod. VEH = Vehicle. Error bars represented in +/− SEM. N = 8–12/group.

Figure 4.. Developmental milestones were delayed in offspring exposed to maternal Resiquimod injection.

While (A) Surface righting reflex was unchanged, (B) fetal RQ exposure caused slower eye openings in male and female offspring. (C) Maternal RQ treatment also delayed vaginal openings and (D) first day of estrous following vaginal opening in females, indicative of delayed pubertal development. ***p < 0.001, ****p < 0.0001. RQ = Resiquimod. VEH = Vehicle. Error bars represented in +/− SEM. N = 11–15/group.

Figure 5.. Social- and anxiety-like behavioral phenotypes in juvenile offspring were assessed in social interaction and open field tests.

Social-like behavior was lowered in juvenile males and females prenatally exposed to RQ, with (A) less time spent with stimulus, (B) reduced number of visits to stimulus mouse, and (C) normalized time spent with stimulus in 10-minute social interaction test. Anxiety-like behavior was not altered by RQ treatment in open field assay with no changes in (D) time spent in the center, (E) number of entries into center ring, or (F) total distance traveled. ***p < 0.001, ****< 0.0001. RQ = Resiquimod. VEH = Vehicle. Error bars represented in +/− SEM. N = 8–10/group.

Figure 6.. Social- and anxiety-like behavioral phenotypes were assessed in adult offspring in social interaction and open field tests.

Social-like behavior was lower in adult males and females prenatally exposed to RQ, with (A) less time spent with stimulus, (B) no change in number of visits to stimulus mouse, and (C) less normalized time was spent with stimulus. Anxiety-like behavior was reduced by RQ treatment in 10-minute open field assay by (D) time spent in the center ring in a male-dependent manner. No change was seen in (E) number of entries into center ring or (F) total distance traveled in open field test. *p < 0.05, ** < 0.01, **** < 0.0001. RQ = Resiquimod. VEH = Vehicle. Error bars represented in +/− SEM. N = 8–10/group.

Figure 7.. Sucrose preference test revealed more depressive-like behavior in female mice exposed to maternal immune activation.

RQ females had decreased normalized sucrose intake compared to VEH females. No change in males. **p < 0.01. RQ = Resiquimod. VEH = Vehicle. Error bars represented in +/− SEM. N = 6–9/group.