Appetite control : genetic, immunological and neurobiological aspects

Abstract

The overall goal of this thesis is to develop a better understanding of the brain’s regulation of appetite, i.e. the hypothalamic regulation of hunger and satiety signals. The correct regulation of appetite is crucial, potentially involved both in traditional eating disorders, such as anorexia nervosa and bulimia nervosa, as well as overweight, obesity and failure to thrive in infants and children. Anorexia/cachexia is also a severe and frequent complication to several common disorders such as cancer, HIV, renal failure and Alzheimer’s disease and is one of the most important factors determining both quality of life and mortality in these conditions. In this thesis, the anx/anx mouse is used as a tool to develop a better understanding of the hypothalamic regulation of food intake in an anorectic condition.

Previous studies on the anx/anx mouse hypothalamus showed several abnormalities in the expression of neurotransmitters and neuropeptides in the orexigenic NPY/AgRP and the anorexigenic POMC/CART system. In order to elucidate when this abnormal phenotype first appears, we studied the development of the NPY/AgRP system in Paper I. The development of the NPY/AgRP system in the anx/anx mice is normal until P12. From this day, a gradual decrease in fiber density is seen with the weakest expression at P21, results that clearly show an impaired development of the NPY/AgRP system in the anx/anx mouse. A second aim for Paper I was to study the underlying mechanisms for this impaired development. We therefore analyzed the expression of markers of activated microglia and detected a gradual increase in these markers in the same areas and at the same time points as was seen for the gradual decrease in density of AgRP fibers.

The impaired development of the AgRP system and the overlapping activation of microglia in the hypothalamus of the anx/anx mouse indicate an inflammatory/degenerative process. This was further investigated in Paper II, where we detected expression of MHC class I mRNA and protein in the arcuate nucleus, both in glial cells and in neurons. We also found that the neurons expressing the MHC class I subunit are ‘silenced’ or have a very low activity. Further evidence for neurodegeneration was seen in Paper II, as we detected a significantly increased number of apoptotic cells in several hypothalamic areas of the anx/anx mouse, as well as a double-labeling of NPY and active caspase 6, a marker for axonal degeneration, in neuronal fibers. Altogether, these results point to a neurodegenerative process in the hypothalamus of the anx/anx mouse.

In Paper III, we continued the search for the anx gene/mutation to find the causes leading to the anorectic phenotype. Through a microarray study and the following pathway analysis the first indications were given that the anorectic phenotype in the anx/anx mouse is related to mitochondrial dysfunction and oxidative stress. We also detected that the anx mutation leads to decreased expression of the Ndufaf1 gene and protein. This gene, located within the anx interval, encodes an assembly factor for mitochondrial complex I. We show that the downregulation of Ndufaf1 is associated with the anx-allele and not due to a secondary effect of the starvation. We could also see that the downregulation of the Ndufaf1 gene leads to a less completely assembled mitochondrial complex I and the accumulation of sub-complexes, as well as increased levels of reactive oxygen species, in the anx/anx hypothalamus. Based on these results, we concluded that the anorexia and premature death of the anx/anx mouse is related to mitochondrial dysfunction and oxidative stress.

In Paper V, we evaluated whether the activity in the hypothalamus is attenuated in the anx/anx mouse as a consequence of the defect mitochondrial complex I. We could see that hypothalamic glucose uptake in the fasted state was reduced in the anx/anx mouse. Further, the anx/anx hypothalamus had elevated levels of one of the glucose transporters, GLUT4 and a key metabolic molecule, AMPK. However, the hypothalamic activation state of AMPK was significantly decreased. Finally, during metabolic stress, levels of both AMP and IMP (both breakdown products of ATP) were decreased, while ATP levels were increased in the anx/anx hypothalamus. Together these results indicate that the anx/anx mouse has a reduced hypothalamic metabolism. This may contribute to the anorectic behavior of this mouse i.e. its inability to regulate food intake in response to the energy status.

In Paper IV, we evaluated whether the inflammation and the mitochondrial dysfunction in the hypothalamus of the anx/anx mouse also could be detected in the endocrine pancreas, and if glucose homeostasis is disturbed. We found a strong downregulation of the Ndufaf1 gene, paralleled by a reduced mitochondrial complex I activity in isolated anx/anx islets. In addition, there was an increased macrophage infiltration in anx/anx islets, indicating inflammation. Moreover, elevated levels of free fatty acids were seen in anx/anx serum. In contrast, isolated islets from anx/anx mice cultured in the absence of free fatty acids did not show any inflammation. Also, an intraperitoneal injection of glucose to the anx/anx mouse revealed a marked glucose intolerance associated with reduced insulin release. However, the insulin release from isolated anx/anx islets was increased after stimulation either with glucose or KCl. The conclusion of Paper IV is that the anx/anx endocrine pancreas display marked reduction in insulin release that correlates with the increased serum levels of free fatty acids, and that the accompanying in vivo inflammation may lead to inhibition of insulin secretion.