Taniyama et al. (1) have recently reported that angiotensin II (Ang II) in vitro decreases insulin receptor substrate-1 protein levels via Src, phosphoinositide-dependent kinase-1, and reactive oxygen species–mediated phosphorylation of Ser307. This leads to the targeting of insulin receptor substrate-1 for proteasome-dependent degradation, which then impairs insulin signaling. These findings provide a rationale for understanding the molecular basis of the positive effect of Ang II type 1 receptor antagonists on insulin resistance.
The relationship between Ang II and insulin signaling shown in vitro leads us to assess whether this is operative also in vivo in humans. We analyzed a cohort of patients with Bartter’s/Gitelman’s syndrome (BS/GS), which attract much attention for persistent normo-/hypotension despite biochemical and hormonal abnormalities typical of hypertension. BS/GS, caused by gene defects in specific kidney transporters and ion channels, presents hypokalemia, sodium depletion, activation of the renin-angiotensin-aldosterone system, and increased levels of Ang II, yet normo-/hypotension, reduced peripheral resistance, and hyporesponsiveness to pressors (2, 3). BS/GS is a good human model to explore the mechanisms responsible for Ang II signaling (2, 4). In BS/GS specifically, the short-term Ang II signaling is blunted (increased regulator of G-protein signaling-2 , reduced Gαq expression [6, 7], and reduced related downstream cellular events [6, 8, 9]), while the NO system is upregulated (2, 10–12). The long-term signaling of Ang II, which modulates the cell redox state to promote cardiovascular remodeling and atherosclerosis, is also altered in BS/GS (13, 14). In addition, the RhoA/Rho kinase (ROK) pathway, which is activated by Ang II and shown to affect the Akt–phosphatidylinositol 3-kinase pathway (15), which, in turn, is involved in glucose transport and metabolism (16), is downregulated in BS/GS (17, 18). Thus BS/GS’s molecular and biochemical characteristics make it an attractive model to explore whether high Ang II is indeed affecting glucose homeostasis also in vivo in humans.
Six patients with BS/GS (1 with BS, 5 with GS) and 10 normotensive healthy subjects underwent oral glucose tolerance tests to determine not only the glucose tolerance but also, using the oral glucose insulin sensitivity index (19), the glucose clearance as a function of insulin concentration.
All patients showed a normal oral glucose tolerance test at baseline and at 120 min (5.0 ± 0.5 vs. 5.30 ± 0.8 and 6.41 ± 1.9 vs. 5.06 ± 1.2, respectively). Insulin at baseline was significantly reduced compared with control subjects (22.5 ± 9.8 vs. 57.0 ± 26.3, P = 0.008), while it was not different at 120 min (198.3 ± 136.0 vs. 190.4 ± 142.7). Oral glucose insulin sensitivity was markedly higher in BS/GS (694.6 ± 103.6 vs. 446.5 ± 48.04 ml · min−1 · m−2, P = 0.00001).
These results point toward a reduced insulin resistance in BS/GS, therefore not only confirming the blunted nature of Ang II signaling in BS/GS but also supporting in vivo in humans the link between insulin signaling, glucose metabolism, and Ang II signaling demonstrated in vitro (1).
The possible involvement of the RhoA/ROK pathway in glucose metabolism is one more piece of indirect evidence. Ang II, via stimulation of RhoA/ROK activity, inhibits insulin signaling through the inhibition of phosphatidylinositol 3-kinase and its downstream Akt pathway (16), induction of oxidative stress, decreased NO production, increased myosin light-chain activation, vasoconstriction, and reduced glucose transport (16). On the contrary, RhoA/ROK inhibition activates the Akt pathway leading to cardiovascular protection via activation of endothelial NO synthase (15, 20). BS/GS shows an upregulation of the NO system (10–12) and downregulation of RhoA/ROK activity (17, 18), which may induce Akt pathway supported by BS/GS increased expression of hemeoxygenase-1 (13), which is under Akt control (21).
Finally, the increased expression of p22phox mRNA and oxidative stress as well as the increased expression of p66shc we have shown in type 2 diabetic subjects (22, 23) are opposed to the findings in BS/GS (13), in which preliminary results also show reduced p66shc expression (E.P., L.A.C., personal observation).
In conclusion, our data in BS/GS represent the first direct confirmation in humans of the Ang II/glucose metabolism relationship, thereby supporting the positive effect of blocking the renin-angiotensin-aldosterone system not only for blood pressure control but also glucose tolerance, diabetes, and atherogenesis.