Postnatal neurodevelopmental expression and glutamate-dependent regulation of the ZNF804A rodent homologue

Abstract

The zinc finger protein ZNF804A rs1344706 variant is a replicated genome-wide significant risk variant for schizophrenia and bipolar disorder. While its association with altered brain structure and cognition in patients and healthy risk allele carriers is well documented, the characteristics and function of the gene in the brain remains poorly understood. Here, we used in situ hybridization to determine mRNA expression levels of the ZNF804A rodent homologue, Zfp804a, across multiple postnatal neurodevelopmental time points in the rat brain. We found changes in Zfp804a expression in the rat hippocampus, frontal cortex, and thalamus across postnatal neurodevelopment. Zfp804a mRNA peaked at postnatal day (P) 21 in hippocampal CA1 and DG regions and was highest in the lower cortical layers of frontal cortex at P1, possibly highlighting a role in developmental migration. Using immunofluorescence, we found that Zfp804a mRNA and ZFP804A co-localized with neurons and not astrocytes. In primary cultured cortical neurons, we found that Zfp804a expression was significantly increased when neurons were exposed to glutamate [20μM], but this increase was blocked by the N-methyl-d-aspartate receptor (NMDAR) antagonist MK-801. Expression of Comt, Pde4b, and Drd2, genes previously shown to be regulated by ZNF804A overexpression, was also significantly changed in an NMDA-dependent manner. Our results describe, for the first time, the unique postnatal neurodevelopmental expression of Zfp804a in the rodent brain and demonstrate that glutamate potentially plays an important role in the regulation of this psychiatric susceptibility gene. These are critical steps toward understanding the biological function of ZNF804A in the mammalian brain.

Keywords: Gene expression; In situ hybridization; Neurodevelopment; Schizophrenia; ZNF804A; Zfp804a.

Fig. 1

Zfp804a mRNA is widely expressed throughout the rat brain at multiple postnatal time points. Example in situ hybridization autoradiographic images are shown at five different postnatal days (P) 1, 7, 21, 49, and 86. Panels show representative images of sagittal (A) and coronal (B) and sections from the rat brain. (C) Representative images from P86 brains are shown for non-specific hybridization, sense oligonucleotide, and scrambled oligonucleotide. Scale bars: 4.0 mm.

Fig. 2

Zfp804a mRNA decreases in frontal cortex during postnatal life. (A) Representative autoradiographic images of sagittal and coronal sections through the rat frontal cortex. Note the dramatic difference in transcript levels between the lower and upper layers of frontal cortex. (B) Image densitometry measures of the lower and upper layers of cortex show a pattern of decreasing Zfp804a mRNA levels across postnatal life with P1 being significantly elevated over other time points (Tukey's HSD, P<0.05). Asterisks indicate significant differences: *p<0.05, **, p<0.005. Scale bars: 3.0 mm.

Fig. 3

Postnatal expression of Zfp804a mRNA varies in hippocampal areas across postnatal life. (A) Representative images of coronal sections showing the hippocampus at different postnatal days. (B) Top, Zfp804a gene expression within area CA1 peaks at P21, which is significantly higher than other time points (Tukey's HSD, P<0.005). Middle, gene expression is fairly low and stable throughout postnatal life in area CA3. Bottom, within the DG, Zfp804a gene expression is significantly lower at P7 compared to P21 (Tukey's HSD, P<0.05), but does not differ from other time points. (C) Representative examples of total and non-specific silver grain labeling within CA1. Dark silver grains show cellular localization of Zfp804a mRNA within thionin-stained hippocampal neurons. Insets show magnified view of labeled neurons. (D) Silver grain density for CA1 and DG confirmed the pattern of elevated mRNA at P21 in both hippocampal regions. (Tukey's HSD, * P<0.001, ** P<0.0005) (E) A subset of smaller, thionin-positive cells that are morphologically consistent with glial cells, show a marked absence of silver grain labeling (red arrows) in all three hipppocampal regions. All scale bars: 10 μm.

Fig. 4. Immunofluorescence of ZFP804A in neurons

Expression of ZFP804A in frontal cortex and hippocampal neurons. (A) Representative immunofluorescence images show ZFP804A expression in neurons labeled with NeuN antibody in the frontal cortex and DG of hippocampus of P21 rats. (B) ZFP804A expression was not observed in GFAP-positive astrocytes. Scale bars: 50 μm.

Fig. 5

Zfp804a mRNA in MGN and TRN of thalamus. (A) Top, representative images of sagittal sections through the MGN. Bottom, image densitometry measures within the MGN through postnatal life show that Zfp804a expression in P86 adult rat is significantly lower than earlier P7 and P21 ages (Tukey's HSD, P<0.05). (B) Top, representative images of sagittal sections through the TRN. Bottom, image densitometry measures within the TRN show no significant changes across postnatal life. Scale bars: 4.0 mm.

Fig. 6

Glutamate exposure upregulates Zfp804a, Pde4b, and Comt expression but decreases levels of Drd2 expression. Plots show quantitative PCR results performed on cultures of rat primary cortical neurons collected at embryonic day 19. All results normalized to the reference gene Hprt1. (A) Cortical neurons exposed to glutamate [20μM] show increased Zfp804a expression. The NMDA receptor antagonist MK-801 blocked this increase. Glutamate exposure led to a four-fold increase in the expression of Pde4b (B) and a two-fold increase in Comt (C) levels in cortical neurons (Student's t-test, . Both increases were blocked by MK-801. (D) Levels of Drd2 were significantly decreased by glutamate exposure, but not in the presence of MK-801. Mean and SEM of three experiments is shown; Mann-Whitney, *P<0.05, **, P<0.01.