A histological and functional study on hippocampal formation of normal and diabetic rats

Abstract

Background: The hippocampus is a key brain area for many forms of learning and memory and is particularly sensitive to changes in glucose homeostasis.

Aim of the work: To investigate in experimentally induced type 1 and 2 diabetes mellitus in rat model the effect of diabetes mellitus on cognitive functions and related markers of hippocampal synaptic plasticity, and the possible impact of blocking N-methyl-D-aspartic acid (NMDA) receptors by memantine.

Materials and methods: Seven rat groups were included: non-diabetic control and non-diabetic receiving memantine; type-1 diabetic groups - untreated, treated with insulin alone and treated with insulin and memantine; and type 2 diabetic groups - untreated and memantine treated. Cognitive functions were assessed by the Morris Water Maze and passive avoidance test. Biochemical analysis was done for serum glucose, serum insulin and insulin resistance. Routine histological examination was done, together with immunohistochemistry for detection of the hippocampal learning and memory plasticity marker, namely activity regulated cytoskeletal-associated protein (Arc), and the astrocytes reactivity marker, namely glial fibrillary acidic protein (GFAP).

Results: Both type 1 and 2 untreated diabetic groups showed significantly impaired cognitive performance compared to the non-diabetic group. Treating the type 1 diabetic group with insulin alone significantly improved cognitive performance, but significantly decreased GFAP and Arc compared to the untreated type 1 group. In addition, the type 2 diabetic groups showed a significant decrease in hippocampus GFAP and Arc compared to the non-diabetic groups. Blocking NMDA receptors by memantine significantly increased cognitive performance, GFAP and Arc in the type 1 insulin-memantine group compared to the type 1-insulin group and significantly increased Arc in the type 2-memantine group compared to the untreated type 2 diabetic group. The non-diabetic group receiving memantine was, however, significantly adversely affected.

Conclusion: Cognitive functions are impaired in both types of diabetes mellitus and can be improved by blockage of NMDA receptors which may spark a future therapeutic role for these receptors in diabetes-associated cognitive dysfunction.

Figure 1.. Escape latency recorded during passive avoidance test.

Latency to enter the dark compartment in type 1 diabetic ( A) and type 2 diabetic ( B) subgroups compared to non-diabetic subgroups. *: significant compared to non-diabetic group, +: significant compared to non-diabetic-memantine group, #: significant compared to untreated type 1 diabetic group, @: significant compared to type 1-insulin group, $: significant compared to untreated type 2 diabetic group at p<0.05.

Figure 2.. Latency to reach the hidden platform during training blocks of Morris water maze in the different groups.

A: escape latency to hidden platform in type 1 diabetic subgroups compared to non-diabetic groups. B: escape latency to hidden platform in type 2 subgroups compared to non-diabetic groups. Data are presented as mean±SD (n=6/group).

Figure 3.. Distance travelled to reach the hidden platform during training blocks of Morris water maze in the different groups.

A: distance travelled to hidden platform by type 1 diabetic subgroups compared to non-diabetic groups. B: distance travelled to hidden platform by type 2 subgroups compared to non-diabetic groups. Data are presented as mean±SD (n=6/group).

Figure 4.. Proximity during training blocks of Morris water maze in the different groups.

Proximity (% time spent within 40 cm of the platform) in type 1 ( A) and in type 2 ( B) diabetic subgroups compared to non-diabetic groups. Data are presented as mean±SD (n=6/group).

Figure 5.. Proximity during immediate and 24 h probe trials in Morris water maze of the different groups.

Proximity (% time spent within 40 cm to where the platform was previously present) in type 1 ( A) and in type 2 ( B) diabetic subgroups compared to non-diabetic group during immediate (open bars) and 24 h (filled bars) probe trials. *: significant compared to non-diabetic group, +: significant compared to non-diabetic-memantine group, #: significant compared to untreated type 1 diabetic group, @: significant compared to type 1-insulin group, $: significant compared to untreated type 2 diabetic group at p<0.05. Data are presented as mean±SD (n=6/group).

Figure 6.. Section from non-diabetic control showing the different areas of the hippocampal formation where the hippocampus proper is formed of the Cornu Ammonis (CA) as CA1, CA2, CA3 & CA4 regions, and is continued as subiculum (S).

Dentate gyrus (DG) is seen surrounding CA4 by its upper & lower limbs. Note lateral ventricle (LV) related to CA1 & CA2. M denotes molecular layer inside concavity of CA and of DG. (H & E ×40).

Figure 7.

( a): Section from non-diabetic control showing 5–6 compact layers of small pyramidal cells of CA1 region, most with vesicular nuclei; ( b): shows few layers of large pyramidal cells in CA3 region, also with vesicular nuclei (arrows). Molecular layer (ML) shows many glial cells (*) among neuronal processes. ( c): shows layers of compact granular cells with dark nuclei in dentate gyrus G. Molecular layer shows glial cells (*) as well as pyramidal cells (↑) (H & E ×400).

Figure 8.

Section from non-diabetic group receiving memantine alone showing: ( a): decreased thickness of layer of small pyramidal cells of CA1 to reach 2 layers in some areas (↑); and ( b): more marked affection of large pyramidal cells of CA3 where areas are devoid of cells (↑). Cells have vesicular nuclei. ( c): granular cells show marked retraction of processes with vacuolations, and molecular layer (ML) shows enlarged neurons (n) and enlarged glial cells (*). (H & E ×400).

Figure 9.

The untreated type 1 diabetic group shows ( a): disorganization and areas of cell loss of small pyramidal cells; some having pale nuclei and others dark. Note also clumping of neuronal processes. ( b): marked shrinkage in size of large pyramidal cells, affecting outer layer more, with darkened nuclei (↑). ( c): Granular cell layers also showed marked vacuolations. Molecular layer (ML) shows marked enlargement of neurons (n) and of glial cells (*).

Figure 10.

The type 1 diabetic group treated with insulin only shows ( a): preservation of small pyramidal cells; but with ( b): marked apoptosis of large pyramidal cells (↑) and ( c): marked disorganization, vacuolation (V) and decreased population of granular cells. Molecular layer (M) mostly shows normal cells & fibres. (H & E ×400).

Figure 11.

The type 1 group treated with insulin and memantine shows also ( a): preservation of small pyramidal cells of CA1 while; ( b): some of large pyramidal cells of CA3 show apoptosis (↑) with some clumping of neuronal fibrils (f) ( c): Granular cells show less vacuolation, & molecular layer shows normal size of cells, with widened capillaries ( c). (H & E ×400).

Figure 12.. The untreated type 2 diabetic group shows changes similar to DM1.

These include: ( a): many darkened nuclei of small pyramidal layer (arrowhead) mainly in deep layer, with vacuolation (v) and clumping of processes. ( b): layer of large pyramidal cells shows disorganization with many apoptotic cells (↑). ( c): granular layer shows some cell loss (arrowhead) and increased glial cells with no change in their size (*) while ( d) molecular layer shows apoptotic cells & (H & E ×400).

Figure 13.

The type 2 diabetic group treated with memantine alone shows some protective effect for the drug in the form of: ( a): preservation of small pyramidal cells except for deepest layer (↑) with clumping of neuronal fibrils (f); but with ( b): shrinkage and darkening of many large pyramidal cells and ( c): clumping & disorganization of granular cells with dilated vessels (v) and normal glial cells (*) in molecular layer. (H & E ×400).

Figure 14.. % area in the hippocampal field of glial fibrillatory acidic protein (GFAP) and Arc in the different studied groups assessed by immunohistochemistry.

*: significant compared to non-diabetic group, +: significant compared to non-diabetic-memantine group, #: significant compared to untreated type 1 diabetic group, @: significant compared to type 1-insulin group, $: significant compared to untreated type 2 diabetic group at p<0.05. Data are presented as mean±SD (n=6/group).

Figure 15.. Anti-GFAP immunostaining x400.

( a) Non-diabetic control group. ( b) Non-diabetic memantine group. ( c) Type 1 DM untreated. ( d) Type 1 DM + insulin. ( e) Type 1 DM + insulin + memantine. ( f) Type 2 DM untreated. ( g) Type 2 DM + memantine. Note increased staining in b and c.

Figure 16.. Anti-Arc immunostaining x400.

( a) Non-diabetic Control group. ( b) Non-diabetic memantine group. ( c) Type 1 DM untreated. ( d) Type 1 DM + insulin. ( e) Type 1 DM + insulin + memantine. ( f) Type 2 DM untreated. ( g) Type 2 DM + memantine. Note patchy disorganized staining in d, and to a lesser extent in e.