The endocrine pancreas of Sur1 knockout mice : optical and ultra-structural study and relationship with congenital hyperinsulinism

In congenital hyperinsulinism of infancy (CHI), the lack of KATP channel activity in β cells, due to mutations in genes coding for KATP channel subunits SUR1 or Kir6.2, induces a permanent insulin secretion and severe hypoglycaemia. By contrast, KATP deficient mice knockout for Sur1 gene; Sur1-/- mice do not present similar glucose homeostasis. The first goal of my thesis was to investigate whether Sur1 gene inactivation has an impact on the morphology of Sur1-/- mouse islets as observed in human CHI. Our investigations using pancreases from young, adult and old WT and Sur1-/- mice revealed that in young Sur1-/- mice, the total endocrine tissue and β cell proportion were higher, while δ cell proportion was lower compared to WT mice. In old Sur1-/- mouse islets, α cells were mostly located in the centre contrary to WT islets and their proportion was significantly increased. Glut2 protein and mRNA levels were lower in old Sur1-/- islets. These modifications and especially the α cell location between β cells and the low content of Glut2, that are not described in CHI, might reflect an adaptative mechanism to maintain near-normal glucose homeostasis in Sur1-/- mice. Since the trafficking of pancreatic KATP channels to the plasma membrane critically depends on masking the endoplasmic reticulum (ER) retention signals of SUR1 and Kir6.2 subunits upon their correct assembly in functional complex, the second aim of this thesis was to assess the effect of the absence of SUR1 on the intra-cell trafficking of Kir6.2 subunits in Sur1-/- islet cells. Our study revealed that unlike in WT islet cells where Kir6.2 was mainly expressed on secretory granules and at the plasma membrane, Kir6.2 was primarily localized in the rough ER (RER) of Sur1-/- islet cells similarly to the ER chaperone calreticulin. This ER retention of Kir6.2 was attested by a significant increase in the RER length and Kir6.2-density in Sur1-/- vs. WT islet cells. However, despite Kir6.2 retention in the RER, Xbp1 mRNA splicing, the mRNA levels of pre-proinsulin and the ER stress-response genes were similar in WT and Sur1-/- islets. In contrast, the mRNA levels of the anti-oxidant enzymes were significantly up-regulated in Sur1-/- islets. We demonstrated thus, that the absence of SUR1 subunits results in a sequestration of Kir6.2 in the RER of Sur1-/- islet cells associated with an increase in RER length and mild oxidative stress without activation of the classical ER stress response. In conclusion, this work gives morphological evidence that Sur1-/- mice, originally produced as a model of human diffuse CHI, are actually not mimicking this pathology. The lack of SUR1 subunits and functional KATP channels in mice is far less dramatic than in human and produces a different morphological phenotype.(SBIM 3) -- UCL, 201