3.1 Changes in diabetic parameters were observed in T1D and T2D mice compared to controls.
After one week of STZ injection in both T1D and T2D groups, mice had fasting blood glucose levels higher than 200 mg/dl and were considered diabetic (Fig. 1B). They remained diabetic till the end of the study, ascertained by fasting blood glucose levels at the end of the 18th week (Fig. 1B). During the time-course of the study, food and water intake measured were increased in T1D and T2D groups compared to controls indicating polyphagia and polydipsia characteristic of the diabetic condition but these changes were insignificant (S. Figure 1B, C). However, the body weight was slightly decreased in T1D compared to controls, while T2D mice displayed increased body weight (S. Figure 1A). An oral glucose tolerance test (OGTT) was also performed at the end of the study, i.e. 18th week, to assess these animals' systemic clearance of glucose. The area under the curve (AUC) was significantly elevated in T1D and T2D mice compared to controls indicating impaired glucose tolerance (Fig. 1D, E). Also, a significant decrease was observed in serum insulin levels in Type 1 Diabetic mice and Type 2 Diabetic mice (Fig. 1C). But the decrease was more prominent in Type 1 Diabetic mice (Fig. 1C). These results indicate that glucose homeostasis was impaired in T1D and T2D animals w.r.t. Controls.
3.2 Behavioral impairments were observed in T1D and T2D mice compared to controls assessed by Morris Water Maze, Novel object recognition and Y Maze.
Specific behavioural tasks were employed to measure spatial learning and memory and recognition memory in all the groups. Morris water maze was used to study spatial learning and memory behaviour. After 18 weeks, in the acquisition trial, T1D and T2D groups had significantly increased escape time (latency) and the distance to find the platform compared to controls (Fig. 2A, B, C). Further, in the probe trial, T1D and T2D mice had a significantly lower number of entries and time spent w.r.t. controls in the quadrant containing platform (Fig. 2D, E, F). The data indicate that spatial learning and memory were significantly impaired in both T1D and T2D groups.
Further, object recognition memory was assessed using a Novel object recognition test. It was found that during the testing phase, the preference index for novel objects significantly decreased in T1D and T2D animals compared to controls (Fig. 2H). Furthermore, the discrimination index was significantly lowered in the T1D and T2D groups w.r.t control group (Fig. 2I). These findings indicated diminished object recognition memory in T1D and T2D mice.
Y Maze was used to assess spatial and recognition memory based on their ability to explore novel environments. In the testing phase, T1D and T2D mice exhibited significantly fewer entries and time spent compared to controls in the novel arm that had otherwise been closed during the training phase (Fig. 2K, L). The data indicated that spatial and recognition memory was reduced in T1D and T2D mice compared to control counterparts.
Overall, the data indicate significant cognitive deficits were observed in T1D and T2D mice compared to controls. Furthermore, no significant intergroup changes in behaviour were observed between T1D vs T2D groups.
3.3 Aberrant dendritic spine density in association with loss of neuronal proteins was observed in the cortex and hippocampus of T1D and T2D mice.
Dendritic spines seed the formation of synapses reinforcing the neuronal networks, which play an important role in memory formation. Therefore, we performed Golgi-cox staining to measure the density of dendritic spines in the hippocampus region of diabetic mice. Dendritic spine density was significantly reduced across the axonal length in T1D and T2D mice compared to the vehicle-treated mice (Fig. 3A, B). Decreased dendritic spine density could be the reason for altered synaptic regulation that impaired cognitive function observed in behaviour studies in T1D and T2D mice. Further, we investigated the protein and gene expression of neuronal proteins, i.e. pre and post-synaptic markers viz SYP and PSD-95 as well as BDNF. BDNF plays a major role in the activation of neurotransmission and promotes memory functions. It was found that protein expression of PSD-95 was diminished in the cortex and hippocampus of T1D (56%, 52%) and T2D (37%, 40%) mice (Fig. 3C, D). Also, the expression of pro-BDNF was significantly lowered in both brain regions by nearly 50% in both T1D and T2D versus controls (Fig. 3C, E). SYP expression was also found to be significantly diminished in the cortex and hippocampus of T1D (42%, 43%) and T2D mice (34%, 30%) (Fig. 3C, F). Although we observed that the reduction was more profound in the T1D group compared to T2D mice, the change between T1D and T2D was not significant. Consistently, gene expression of SYP, PSD-95 and BDNF was also found to be significantly lowered in brain tissues in T1D and T2D groups. w.r.t. Control group (Fig. 8).
3.4 H3 and H4 acetylation regulate neuronal gene expression in the cortex and hippocampus of T1D and T2D mice.
To study the epigenetic basis of observed effects, we performed histone modification profiling in the brain tissues of all the groups. H3 profiling revealed that acetylated H3 (K9/14) was found to be significantly lowered in T1D (35%, 32%) and T2D (33%, 30%) brain regions (cortex and hippocampus) versus controls (Fig. 4A, B). However, no significant change was observed in H3acK27, H3meK27 and H3me2K27 (Fig. 4A, C, E, F). Interestingly, we observed a significant increase in H3me3K27 levels in the cortex and hippocampus of T1D (27%, 32%) and T2D (31%, 42%) group w.r.t. controls (Fig. 4A, D). We also profiled H4 acetylation and observed that neither H4acK5 nor H4acK8 levels significantly differed from controls versus diseased brain tissues (Fig. 4G, H, I). However, H4acK12 levels were found to be significantly diminished in T1D and T2D animal brain regions by nearly 20% (Fig. 4J). Next, ChIP was performed to study the association of modified histones (H3 ac (K9/14), H4acK12) with the promoter of Psd-95, Bdnf (pII, pIV), Syp, c-Fos genes that may account for diminished levels of these markers in T1D and T2D brain regions (Fig. 5). Lowered enrichment was observed with H3 ac (K9/14) to the promoters of Psd-95, Bdnf (pII, pIV), Syp, c-Fos in the brain samples of T1D and T2D mice (Fig. 5A). Furthermore, the enrichment using H4acK12 antibody at promoter sequences of Bdnf pIV, c-Fos was also lowered in the diabetic brain (Fig. 5B). Together, these results suggest that H3 and H4 acetylation regulates the expression of neuronal genes (Psd-95, Bdnf (pII, pIV), Syp, c-Fos) involved in synaptic plasticity and memory.
3.5 Elevated Class I HDAC enzymes were observed in the cortex and hippocampus of T1D and T2D mice.
We evaluated HDAC and HAT activities that may be responsible for lowering H3 and H4 acetylation. Strikingly, we found that HDAC activity was significantly increased in brain regions of T1D and T2D animals w.r.t. controls (Fig. 6A). However, we found a non-significant decrease in HAT activity of the cortex and hippocampus of diabetic mice when compared to controls (Fig. 6B). Further, the protein expression analysis of Class I HDACs (HDAC-1, HDAC-2 and HDAC-3) revealed that HDAC-2 and HDAC-3 were found to be significantly increased (by nearly 2 fold) in the cortex and hippocampus of T1D and T2D mice (Fig. 6C, E, F). However, no significant change was observed in the relative protein expression of HDAC-1 in the diseased versus the control group (Fig. 6D). Confirming these findings, mRNA expression of HDAC-2 and HDAC-3 has also been found to be significantly increased, while HDAC-1 remain unchanged in T1D and T2D brain tissues (Fig. 8). Thus it can be inferred that increased HDAC-2 and HDAC-3 levels might corroborate with increased HDAC activity in brain regions of diabetic mice.
3.6 Differential changes in Class II HDAC were observed in the cortex and hippocampus of T1D and T2D mice.
We performed the gene and protein expression analysis of Class II HDACs. Class II HDACs are mainly comprised of HDACs 4, 5, 6, and 7. Protein and mRNA expression of HDAC 5, 6, 7 were not significantly affected in the cortex and hippocampus of diseased animals versus controls (Fig. 7, 8). However, HDAC-4 was found to be significantly elevated in brain regions of T2D w.r.t. Controls (Fig. 7A, B). No significant change was observed in HDAC4 protein expression in T1D w.r.t. Control.