A. Output of insulin from pancreatic islet cells:
Pancreatic islets1, prepared by the collagenase procedure from the pancreas of 3-4 mice, were incubated in groups of 8 islets each for 90 min at 37°C in 1.0 ml of a Hepes- and bicarbonate-buffered medium containing 5 mg/ml bovine serum albumin and, as required, 11.1 mM D-glucose and the secretagogue under consideration. The insulin released in the medium was measured by radioimmunoassay. All results are expressed as mean values (± SEM), together with the number of separate determinations (n).
Table 1: Insulin release by pancreatic islets from SHIP2+/+ and SHIP2+/- mice:
Mice / SHIP2+/+ / SHIP2+/-Insulin output
(µU/islet per 90 min)
Basal / 32.2 ± 18.8 (28) / 48.7 ± 8.4 (48)
11.1 mM D-glucose / 63.1 ± 16.3 (28) / 69.1 ± 11.9 (48)
11.1 mM D-glucose +
1.4 mM theophylline / 156.5 ± 12.8 (27) / 146.4 ± 12.1 (49)
11.1 mM D-glucose +
1.0 mM carbamylcholine / 151.3 ± 12.1 (29) / 151.8 ± 11.9 (49)
Reference:
1. Malaisse-Lagae, F., and Malaisse W. J. Insulin release by pancreatic islets. In: Larner J., Pohl S. L. (Eds) Methods in Diabetes Research, vol 1, part B. Wiley, New York, pp 147-152 (1984).
B. Specificity of SHIP2 action in SHIP2+/- and SHIP2-/- mice:
The ubiquitously expressed SHIP2 was found to be phosphorylated in response to several growth factors (EGF, PDGF, IGF-1, NGF, ...), and insulin. Our data on SHIP2-/- and SHIP2+/- mice show that, in vivo, this 5-phosphatase is mainly involved in the control of the insulin signalling cascade. We summarize here our unpublished results suggesting that SHIP2 in vivo is not mainly implicated in systems other than insulin signalling:
1. Histologic examination of brain, lung, heart, thymus, stomach, pancreas, kidney, skin, bladder, small and large intestine, liver, muscle, and spleen did not reveal an obviously altered organization of these tissues (hypo-, hyper-, or dysplasia) nor obvious abnormalities in SHIP2-/- newborn mice. No tumor was found in adult heterozygous mice, and these mice have a normal lifespan. SHIP2+/- mice, when irradiated (200 rads for 129 x Balb/c, and 300 rads for 129 x C57BL/6) have no increased incidence of tumors, as compared to SHIP2+/+ mice.
Together, these results suggest that at least in vivo, SHIP2 is not a critical negative regulator of growth factors signalling that, when released from negative control, is implicated in the control of cell proliferation/apoptosis and tumor development.
2. Since SHIP2, in the brain, is mainly expressed in neurons of the cortical ventricular layer, this specific region was investigated in newborn SHIP2-/- mice. No difference was seen between SHIP2-/- and SHIP2+/+ in the organization of the different cortical layers. This result suggest that SHIP2 does not specifically control neuron division and migration in that region.
3. SHIP2 is also expressed in the hematopoietic system, and particularly in thymocytes. Thymocytes, splenic, bone marrow, and lymph node B and T cells as well as macrophages were analyzed by cytofluorometry (Facs) in SHIP2 -/- newborn mice and in Scid mice reconstituted with liver cells from SHIP2-/- embryos using anti-CD3, -CD4, -CD8, -CD25, -IgM, -IgD, -B220, -Gr1, -Mac1, -B7.1, -B7.2, -CD40L, -CD40, -CD28, and -PNA specific antibodies. No specific alteration (number of positive cells, intensity of antigen expression at the cell surface) was found in the different cell populations tested (see Figure 1 for thymocytes analysis) . As it is known that SHIP1 is coexpressed with SHIP2 in the cell populations tested, we hypothesized that SHIP1 could compensate for the absence of SHIP2. The production of double KO is in progress to analyze the effects of SHIP1+2 deficiency on T and B cell ontogeny, activation,...
4. Spontaneous apoptosis of thymocytes from SHIP2-/- newborn mice did not reveal a significant difference when compared with SHIP2+/+ mice (Figure 2). Unlike Pten, a 3-phosphatase, our results suggest that SHIP2 does not tighly control apoptosis in thymocytes.
5. To further investigate the potential role of SHIP2 in growth factor signalling control, MEF from SHIP2+/+, SHIP2+/- and SHIP2-/- mice were stimulated with 25 nM EGF. PKB and MAPK (Erk-2) activities were analyzed 2 and 5 min after stimulation. We did not observed any significant and reproductible effect of SHIP2 deletion on PKB and MAPK activities.
In conclusion, despite carefull phenotypic and biochemical characterization of SHIP2-/- and SHIP2+/- mice and cells, we are unable to confer a very important role for SHIP2 in vivo in the control of other signalling cascades beside insulin.