Neonatal inflammatory pain and systemic inflammatory responses as possible environmental factors in the development of autism spectrum disorder of juvenile rats

Abstract

Background: Autism spectrum disorder (ASD) affects many children and juveniles. The pathogenesis of ASD is not well understood. Environmental factors may play important roles in the development of ASD. We examined a possible relationship of inflammatory pain in neonates and the development of ASD in juveniles.

Methods: Acute inflammation pain was induced by 5 % formalin (5 μl/day) subcutaneous injection into two hindpaws of postnatal day 3 to 5 (P3-P5) rat pups. Western blot, immunohistochemical, and behavioral examinations were performed at different time points after the insult.

Results: Formalin injection caused acute and chronic inflammatory responses including transient local edema, increased levels of inflammatory cytokines, TNF-α, and IL-1β in the blood as well as in the brain, and increased microglia in the brain. One day after the pain insult, there was significant cell death in the cortex and hippocampus. Two weeks later, although the hindpaw local reaction subsided, impaired axonal growth and demyelization were seen in the brain of P21 juvenile rats. The number of bromodeoxyuridine (BrdU) and doublecortin (DCX) double-positive cells in the hippocampal dentate gyrus of P21 rats was significantly lower than that in controls, indicating reduced neurogenesis. In the P21 rat's brain of the formalin group, the expression of autism-related gene neurexin 1 (NRXN1), fragile X mental retardation 1 (FMR1), and oxytocin was significantly downregulated, consistent with the gene alteration in ASD. Juvenile rats in the formalin group showed hyperalgesia, repetitive behaviors, abnormal locomotion, sleep disorder, and distinct deficits in social memory and social activities. These alterations in neuroinflammatory reactions, gene expression, and behaviors were more evident in male than in female rats. Importantly, an anti-inflammation treatment using indomethacin (10 mg/kg, i.p.) at the time of formalin injections suppressed inflammatory responses and neuronal cell death and prevented alterations in ASD-related genes and the development of abnormal behaviors.

Conclusions: These novel observations indicate that severe inflammatory pain in neonates and persistent inflammatory reactions may predispose premature infants to development delays and psychiatric disorders including ASD. The prevention of pain stimuli and prompt treatments of inflammation during development appear vitally important in disrupting possible evolution of ASD syndromes.

Keywords: Autism spectrum disorder; Cell death; FMR1; Inflammatory pain; NRXN1; Oxytocin; Social behavior.

Fig. 1

Neonatal inflammatory pain increased inflammatory cytokines in the blood and cortex. Quantitative RT-PCR (qRT-PCR) analysis was performed to measure inflammatory factors in blood samples and cortical tissues. a The levels of TNF-α, IL-1β, and IL-6 in the blood 1 day after the 3-day formalin injections, TNF-α and IL-1β significantly increased in both male and female rats; indomethacin (Indo) co-applied with formalin blocked the inflammatory reaction. b TNF-α, IL-1β, and IL-6 levels in the cortex 1 day after formalin injection. Only the TNF-α level was significantly enhanced in the male brain. Indomethacin (Indo) showed inhibitory effect on TNF-α expression. c Inflammatory factors were measured in blood samples 16 days after formalin injections (P21 rats). TNF-α and IL-1β increased in male but not in female rats in the formalin group, which were blocked by indomethacin (Indo). d Increased levels of TNF-α and IL-1β persisted in the male brain cortex 16 days after formalin injections. Indomethacin (Indo) blocked these increases. *P < 0.05 vs. control, ^# P < 0.05 vs. formalin group, ANOVA plus Bonferroni’s correction; n = 4 per group

Fig. 2

Neonatal peripheral inflammatory pain-induced microglia activation and neuronal cell death in the cortex and hippocampus. Microglia cells and neuronal cell death were measured using Iba-1 and TUNEL staining, respectively, in the cortex and hippocampus. a, b Immunohistochemical staining was performed 1 day after the last formalin injection. Iba-1-positive cells (red) were identified as microglia cells in the cortex. Nuclei were stained with Hoechst 33342 (blue). Different shapes of Iba-1-positive microglia cells at resting state (control) and activation state (formalin) are shown in the insets of enlarged magnification. Scale bars = 100 μM. b The ratio of Iba-1+ cells against total Hoechst + cells. Formalin injections significantly increased activated microglia while indomethacin (Indo) attenuated this event. *P < 0.05 vs. control, ^# P < 0.05 vs. formalin, ANOVA plus Bonferroni’s correction; n = 6–8 per group. c TUNEL staining (green) and Hoechst 33342 (blue) were used to evaluate cell death in the cortex and hippocampus. Arrows point to some TUNEL-positive green color cells. Scale bars = 100 μm. d, e Quantified data of TUNEL-positive cells/field in the cortex (d) and hippocampus (e). *P < 0.05 vs. control; n = 6–7 per group. f TUNEL-positive cells/field in the hippocampus in saline control and formalin groups. A marked increase in TUNEL-positive cells was seen in CA2. *P < 0.05 vs. CA1 or CA3, ANOVA plus Bonferroni’s correction; n = 6–7 per group

Fig. 3

Neonatal inflammatory pain affected axonal development in the juvenile brain. Immunohistochemical staining was performed to determine whether peripheral inflammatory pain in neonates could affect axonal growth and myelination at P21. a Double-labeled images of NeuN (red) and neurofilament (NF; green) in the cortex of the control or formalin group. Nuclei were stained with Hoechst 33342 (blue). Scale bars = 20 μm. b–f Axon diameter was estimated by measuring the distance perpendicular to the center of the maximum diameter of the axon profile. Measured axons were categorized as small (<0.35 μm), medium (0.35–0.69 μm), large (0.7–1.4 μm), and extra-large (>1.4 μm) groups. *P < 0.05 vs. control; **P < 0.01 vs. control; ***P < 0.001 vs. control; n = 6 per group. g Images of MBP (red) in the cortex. Nuclei were stained with Hoechst 33342 (blue). Scale bars = 20 μm. h Juvenile rats subjected to neonatal inflammatory pain showed a significant decrease of MBP expression compared to control rats. *P < 0.05 vs. control; n = 6 per group

Fig. 4

Neonatal inflammatory pain decreased neurogenesis in the juvenile brain. Immunofluorescent double labeling for DCX and BrdU was performed to examine neurogenesis in P21 rats. a–d Double-labeled images of BrdU (red) and DCX (green) in the dentate gyrus of the control (a, b) or formalin (c, d) group. Nuclei were stained with Hoechst 33342 (blue). b, d 3-D confocal images of BrdU (red) and DCX (green) in the dentate gyrus of the control and formalin groups. Scale bars = 50 μm. e, f Summarized data of the total numbers of DCX-positive and DCX/BrdU-positive cells in three sections of each animal. There were fewer BrdU-positive (e) and DCX/BrdU-positive cells (f) in the formalin group compared to the control group. ***P < 0.001 vs. control; n = 6 per group

Fig. 5

Neonatal peripheral inflammatory pain altered ASD-related genes in juvenile rats. Western blotting analysis was performed to determine the expression of ASD-related genes in the cortex of P21 rats subjected to saline and formalin injections at neonatal stage. a Representative Western blot bands of NRXN1, NLGN3, FMR1, AUTS2, oxytocin, and oxytocin receptor of male and female rats. b, d, and f Optic density of NRXN1 (b), FMR1 (d), and oxytocin (f) in the cortex of male rats. *P < 0.05 vs. control; n = 12–14 per group. c, e, and g Optic density of NRXN1 (c), FMR1 (e), and oxytocin (g) in the cortex of female rats. *P < 0.05 vs. control; n = 12–14 per group. h–k The expression of NRXN1, FMR1, and oxytocin in control and formalin groups of male sex (P21). The reductions induced by the formalin insult were all blocked by the anti-inflammatory treatment of indomethacin (Indo) co-applied with formalin. *P < 0.05 vs. control, ^# P < 0.05 vs. formalin group, ANOVA plus Bonferroni’s correction; n = 4 per group

Fig. 6

Neonatal peripheral inflammatory pain diminished the oxytocin receptor expression in the hippocampus of the juvenile brain. Immunohistochemical staining measured the expression and distribution of the oxytocin receptor in the CA1, CA2, and CA3 regions of the hippocampus. P21 rats were subjected to formalin injections or saline injections at P3–P5 age. a Oxytocin receptor-positive cells (green) and NeuN-positive cells (red, shown as purple due to overlay with blue of Hoechst 33342 nuclei staining) in the hippocampus from the control and formalin groups. The first row in each group shows oxytocin receptor only, and the next row shows the overlay images of all three immunostainings. Scale bars = 50 μM. b, c Quantified data from experiments in a. The numbers of oxytocin receptor/NeuN double-positive cells per survey area under × 20 magnification. These double-labeled cells in both CA1 and CA2 were significantly less in male rats in the formalin group comparing to male controls (b). A trend of reduced oxytocin receptor level was also seen in CA3. In female rats of the formalin group, a significant reduction of oxytocin-positive neurons was only seen in the CA1 region, although there was a trend of reduction in CA2 (c). *P < 0.05 vs. control, ***P < 0.001 vs. control; n = 6–7 per group

Fig. 7

Neonatal inflammatory pain caused increased pain sensation, self-repetitive behaviors, and sleep disorder in juvenile rats. Hot-plate and repetitive behavior tests were performed to examine pain sensation and ASD-related activities in P21 juvenile rats. a In the hot-plate test, rats were placed individually on a hot plate of 55 °C. The latency time to jump was measured. Rats in the formalin group were more sensitive to the hot stimulation, showing a reduced latency to jump compared to controls. **P < 0.01 vs. control; n = 9–11 per group. b–f Six-hour monitoring of the HomeCage system for other repetitive behaviors and sleep pattern. HomeCage monitoring system monitored behavioral patterns of control and formalin-injected rats in a natural environment without human intervention. Rats in the formalin group showed significantly increases in repetitive grooming (b), spontaneous jumping (c), and muscle twitching (d) behaviors at P21. These animals showed increased frequencies of getting awake and trying to sleep compared to control rats during the same monitoring time (e, f). *P < 0.05 vs. control, **P < 0.01 vs. control; n = 16 per group

Fig. 8

Neonatal inflammatory pain-induced delayed impairments in social activities and the blocking effects of the anti-inflammatory treatment. Social behavioral tests were carried out to evaluate long-term adverse impacts of the early inflammatory pain on social interactions in juvenile rats. The indomethacin (Indo) treatment was applied to delineate whether inflammation was a crucial mediator in the social behavior changes. a In the three chamber test, male rats, not female rats, in the formalin group, spent less time with the novel stranger and preferred to hide in an empty chamber compared to the control rats. Indomethacin (Indo) co-applied with formalin completely blocked the development of this abnormal behavior. *P < 0.05 vs. control; n = 16 in control and formalin group. n = 5 in indomethacin group. b In the social novelty test, since male rats in the formalin group preferred to stay with a familiar rat, their time with a novel stranger was significantly less than the time spent by controls. The indomethacin (Indo) treatment blocked the social abnormality. *P < 0.05 vs. control; n = 16 in control and formalin group. n = 5 in indomethacin group. c–e In the social interaction test, male rats in the formalin group showed reduced time in social sniffing and social following, although no change was seen with social grooming. Indomethacin (Indo) reversed the behavioral changes. Female rats which received formalin did not show significant deficits in these tests. *P < 0.05 vs. control, ^# P < 0.05 vs. formalin; n = 16 in control and formalin group. n = 5 in indomethacin group. f A summarized data analysis of all social behavioral tests for male and female rats. The total social time was significantly less with male rats in the formalin group, and the phenotype was prevented by the anti-inflammation treatment. Again, female rats were much more resistant to the formalin insult, showing no social behavior deficits at the juvenile age. *P < 0.05 vs. control, ^# P < 0.05 vs. formalin; n = 16 in control and formalin group. n = 5 in indomethacin group. ANOVA plus Bonferroni’s analysis was applied to all comparisons in this experiment