Diabetes induces changes in ILK, PINCH and components of related pathways in the spinal cord of rats

Abstract

Recent work suggests that diabetes affects processing of peripheral, spinal and supraspinal signals in the spinal cord. However, there is little evidence for spinal cord lesions that would account for alterations in behavioral responses induced by experimental diabetes. Therefore, we assessed the expression of proteins that might affect neuronal cytoskeletal stability and thus promote dendritic and synaptic reorganization in diabetic rats. Expression of ILK, PINCH, PI3K, GSK-3beta, tau, MAP2, synaptophysin and drebrin in the lumbar spinal cord of non-diabetic and streptozotocin-diabetic rats was assessed by Western-blot analysis and immunocytochemistry after 8 and 20weeks of diabetes. The impact of diabetes on the proteins studied was duration-dependent with changes observed after 20 but not 8weeks of diabetes. ILK and PINCH proteins levels were significantly decreased and both colocalized to neurons and oligodendrocytes. PI3K protein levels were also significantly decreased, while GSK-3beta activity tended to be increased. Phosphorylation of tau and MAP2A/B protein expression were significantly increased, and expression of synaptophysin and drebrin were reduced in diabetic rats. Decreased ILK and PINCH as well as alterations of components of related signaling pathways are associated with tau hyperphosphorylation, MAP2 overexpression and reduction of synaptic proteins in the spinal cord of diabetic rats, suggesting that ILK and PINCH contribute to stabilization of axonal and dendritic structures. However, these changes are not likely the cause of altered behavioral responses in diabetic rats that occur after short-term diabetes, but may contribute to structural changes occurring in long-term diabetes.

Fig. 1

Protein levels of PINCH and ILK in lumbar spinal cord of non-diabetic control (C) and 20-week diabetic (D) rats. A: Intensity of bands corresponding to PINCH 70kDa normalized to the intensity of bands corresponding to actin. B: Intensity of bands corresponding to ILK normalized to the intensity of bands corresponding to actin. Data are represented as mean + SEM, n=6, **p<0.01, ***p<0.001 using unpaired, two-tailed t-test.

Fig. 2

Immunoreactivity of PINCH in lumbar spinal cord. PINCH is coexpressed with NeuN in dorsal horn gray matter (A, B, C) and with APC in white matter (D, E, F), localizing PINCH in spinal cord to neurons and oligodendrocytes. Bar = 40μm. Inset in A: low-power view of the dorsal horn immunostained for PINCH, with the approximate location of higher-power images delimited by the white square.

Fig. 3

Immunoreactivity of ILK in lumbar spinal cord. ILK is present in dorsal horn gray matter (A) and in white matter (D). Similarly, PINCH immunoreactivity is present in dorsal horn gray matter (B) and in white matter (E). Note the coexpression of ILK and PINCH in both gray matter (C) and white matter (F). Bar = 40μm. Inset in A: low-power view of the dorsal horn immunostained for ILK, with the approximate location of higher-power images delimited by the white square.

Fig. 4

Protein levels of PI3K and phosphorylation levels of GSK3 in lumbar spinal cord of non-diabetic control (C) and 20-week diabetic (D) rats. A: Intensity of bands corresponding to PI3K normalized to the intensity of bands corresponding to actin. B: Intensity of bands corresponding to phosphorylated GSK3 at ser9 (inactivating site) and at Tyr 216 (activating site) normalized to the intensity of bands corresponding to total GSK3. Bands corresponding to total GSK3 were normalized to bands corresponding to actin. Data are represented as mean + SEM, n=6, **p<0.01 using unpaired, two-tailed t-test.

Fig. 5

Phosphorylation levels of tau and protein level of MAP2A/B in lumbar spinal cord of non-diabetic control (C) and 20-week diabetic (D) groups. A: Intensity of bands corresponding to phosphorylated tau (threonine 231) normalized to the intensity of bands corresponding to total tau. B: Intensity of bands corresponding to MAP2A/B normalized to the intensity of bands corresponding to actin. Data are represented as mean + SEM, n=6, *p<0.05 using unpaired, two-tailed t-test.

Fig. 6

Protein levels of synaptophysin and drebrin in lumbar spinal cord of non-diabetic control (C) and 20-week diabetic (D) groups. A: Intensity of bands corresponding to synaptophysin normalized to the intensity of bands corresponding to actin. B: Intensity of bands corresponding to drebrin normalized to the intensity of bands corresponding to actin. Data are represented as mean + SEM, n=6, *p<0.05 using unpaired, two-tailed t-test.