Characteristics, causes and functional consequences of brain inflammation in diabetes and hypertension

Diabetes and its complications, such as hypertension and diabetic cardiomyopathy, increase the risk of morbidity and mortality in human populations. Despite decades of research, the full aetiology of these complications still remains unclear. Strong evidence suggests there is dysfunction in the autonomic neurons system, including abnormal activity in the sympathetic nervous system and baroreflex impairment, which may contribute to the development of cardiovascular complications in diabetes; however, the mechanisms behind these abnormalities in diabetes are not well understood. Autonomic nuclei within the central nervous system (CNS) that are involved in the control of sympathetic nerve activity, the sensitivity of the baroreflex and the cardiovascular system are the paraventricular nucleus of the hypothalamus (PVN), the rostral ventrolateral medulla (RVLM) and the nucleus tractus solitarius (NTS). Inflammation and oxidative stress in these centres have been identified to contribute to the hyperactivity of the sympathetic nerves and baroreflex dysfunction in other diseases, but whether this true in diabetes is not clear. Microglia and astrocytes are the resident immune inflammatory cells within the CNS. Once they become activated in response to various stimuli, they can release pro-inflammatory molecules and reactive oxygen species; however, the role of these cells in the development of diabetic cardiovascular complications remains unclear. Drinking 1% NaCl has been shown to have biphasic effects on the development of diabetic complications in animals, but whether these effects are mediated via influencing brain inflammation has not been investigated. In addition, little is known about the influence of antioxidants and high fat feeding on microglial activation in central autonomic centres in diabetic animals. Thus, the aim of this thesis was to identify whether inflammation occurs in central autonomic centres in different species and models of diabetes and to explore the consequences of neuroinflammation in these animals. In chapter 3, the effect of the inhibition of microglia in the PVN on blood pressure and heart rate in long-term (8 weeks) STZ-diabetic rats was investigated. Microglia and astrocytes were activated in the PVN in STZ diabetic rats. Pro-inflammatory cytokines released from both activated microglia and astrocytes have been implicated in spinal neuronal hyperexcitability. This suggested that the activation of microglia and astrocytes may be important for mediating inflammation in the PVN in diabetic rats, which may cause neuronal hyperactivity and then lead to increased sympathetic activity. Minocycline treatment inhibited PVN microglial activation but not the activation of astrocytes, suggesting that microglial activation was not responsible for astrocyte activation; however, in this study, the consequence of the inhibition of microglial activation could not be tested for several reasons, including morbidity and the lack of hypertension in STZ diabetic animals. Therefore, an alternative approach to this study was justified. In most studies on long-term (6-8 weeks) STZ diabetic rats, unchanged or lowered blood pressure has been observed, which is consistent with the data presented in chapter 3. Drinking 1% NaCl, which may prevent dehydration, causes hypertension in diabetic rats within 2 weeks after STZ treatment, but how this occurs and whether brain inflammation contributes to this hypertension has not been investigated. Therefore, in chapter 4, we investigated the effects of 1% NaCl intake on blood pressure, baroreflex sensitivity and inflammation in the PVN, NTS and RVLM in 2-week STZ diabetic rats. In addition, we investigated whether the inhibition of microglia in the PVN can prevent hypertension in STZ diabetic rats given 1% NaCl. Diabetic rats given saline exhibited hypertension, dysfunction of the bradycardic baroreflex and signs of normalised blood volume in comparison with the control rats and the diabetic rats that drank water. Diabetic rats that drank 1% NaCl also showed increased microglial activation in the PVN, NTS and RVLM. The inhibition of activated microglia in the PVN via administering ICV minocycline prevented the hypertension seen in diabetic rats given 1% NaCl, strongly suggesting that microglial activation plays an important role in the generation of hypertension in these animals. Despite this, the possibility that increased blood volume and/or baroreflex dysfunction are mechanisms by which 1% NaCl intake induces hypertension in STZ diabetic rats cannot be ruled out. While it is clear from these studies that 1% NaCl intake can reduce baroreflex sensitivity in 2-week STZ diabetic rats, another study reported that the prolonged drinking of 1% NaCl causes improvement in ex vivo cardiac function in 6-week STZ diabetic rats. Whether treatment with 1% NaCl for longer periods produces similar beneficial effects on the baroreflex sensitivity is not known. Therefore, in chapter 5, we aimed to investigate the effects of 1% NaCl on the baroreflex sensitivity and to determine whether 1% NaCl intake influences inflammation cardiovascular centres in longer term (6 weeks) STZ diabetic rats. The diabetic rats showed dysfunction in the barorflex sensitivity. This dysfunction was associated with increased microglial activation in the NTS and PVN. Drinking 1% NaCl for 6 weeks restored the function of bradycardic baroreflex and also reduced the activation of microglia and neurons in the NTS in these animals. The data suggest that microglial activation in the NTS may be responsible for the baroreflex dysfunction seen in 6-week STZ diabetic animals. The data also suggest that drinking 1% NaCl can prevent cardiovascular complications through a reduction in microglial activation in longer term STZ diabetic rats. In addition to hypertension and the baroreflex dysfunction, cardiomyopathy is a common form of diabetic complication and can occur independently of hypertension in diabetes. There is evidence that inflammation in the PVN is a potential contributing factor to the development of cardiomyopathy in other diseases. In chapters 3 and 5, evidence is provided that microglial cells are activated in the PVN in 6-8 STZ diabetic rats, but the role of these cells in diabetic cardiomyopathy has not been investigated. Therefore, in chapter 6, we aimed to investigate the structural and functional parameters of the left ventricle in diabetic rats at 6 weeks following STZ injection and to determine whether the inhibition of microglial activation in the PVN could reverse any of the changes observed. Six-week STZ diabetic rats showed clear left ventricular dysfunction, including elevated end diastolic pressure, an increased internal diameter in the systole and diastole and a decreased E/A ratio when compared with the control rats. These animals also displayed marked activation of microglia and neurons in the PVN. Inhibition of microglial activation via administering ICV minocycline reduced the PVN neuronal activity and significantly normalised the left ventricular function. This study suggests that microglial activation in the PVN leads to PVN neuronal excitation in STZ diabetic rats. The data also confirm our proposal that microglial activation in the PVN plays a critical role in the pathogenesis of diabetic complications. Drinking 1% NaCl for 6 weeks in STZ diabetic rats reduces cardiovascular complications, but the exact mechanism is still unclear. The data from chapter 6 indicated that microglial activation in the PVN contributes to cardiac dysfunction in STZ diabetic rats; however, whether the beneficial effects of the prolonged drinking of 1% NaCl on the cardiac function are mediated by reducing the inflammatory response is not known. Therefore, in chapter 7, we investigated the effects of 1% NaCl on the cardiac function in vivo and whether 1% NaCl intake influences microglial activation in 6-week STZ diabetic rats. Drinking 1% NaCl restored elevated end diastolic pressure but not the other parameters of the left ventricle in these animals. The 1% NaCl intake also reduced the activation of microglia and neurons in the NTS and PVN when compared to the STZ diabetic rats given tap water. When the drug minocycline was accompanied by prolonged 1% NaCl intake, the STZ diabetic rats showed a further improvement in cardiac performance and a reduction in microglial and neuronal activation in PVN compared with STZ diabetic rats given saline alone. These results indicated that changes in cardiac function are paralleled by the level of microglial activation in the PVN observed in diabetic animals. Because the mouse is the most suitable animal model for genetic manipulations, it was important to examine whether microglia and neurons are also activated in the PVN in STZ diabetic mice as well as the time period of any activation. Hydrogen sulphide (H2S) has been shown to inhibit microglial activation in vitro experiments, but the effect of the systemic infusion of H2S on PVN function in STZ mice has not been investigated. Therefore, in chapter 8, we examined the dose and time dependence of microglial and neuronal activation and the effects of H2S in STZ treated mice at 7 weeks. The microglia were activated in the PVN in STZ diabetic mice at 16 weeks after STZ injection but not at the 7-week time period compared to the control. In diabetic mice treated with a low dose of STZ, the microglia in the PVN were not activated at the 10-week time period; however, the PVN neuron activation was observed in STZ diabetic mice at all time periods as well as in diabetic mice treated with a low dose of STZ, suggesting neuronal activation precedes microglial activity. H2S treatment attenuated the increase in neuronal activation in the PVN and prevented dehydration at 7 weeks following STZ injections. These findings suggest that H2S treatment may play an important role in regulating kidney and PVN neuron function in diabetes. A high fat diet is a major contributing factor in the pathogenesis of type 2 diabetes, which is the most common form of diabetes in humans. Whether PVN inflammation occurs in models of type II diabetes requires investigation. Thus, in chapter 8, we also examined the PVN inflammation in mice treated with a low dose of STZ and high fat feeding. No significant difference was observed in the percentage of microglia activated in the PVN in mice treated with STZ and high fat feeding when compared to the control; however, the neuronal activation was significantly increased in the PVN in these mice. The addition of a high fat diet to the STZ treatment attenuated the increase in neuronal activation in the PVN. These findings suggest that the activation of microglia in the PVN does not occur in all animal models of diabetes. In conclusion, microglial activation occurs in the PVN in long-term STZ-induced diabetic rats and mice but not in a mouse model of type II diabetes. Although the cause is not fully known, microglial activation in the PVN plays an important role in the pathogenesis of diabetic complications. While 1% NaCl prevents dehydration in short and long-term STZ diabetic rats, it has a biphasic effect on diabetic complications. This effect was mediated, at least in part, by changes in neuroinflammation. The beneficial effects of hydrogen sulphide on PVN function may be mediated via reducing neuroinflammation and/or oxidative stress in diabetes. Thus, new therapeutic approaches aimed to target neuroinflammation may be of clinical importance in preventing diabetic cardiovascular complications in humans