Resistance to the sympathoexcitatory effects of insulin and leptin in late pregnant rats

Abstract

Key points: Pregnancy increases sympathetic nerve activity (SNA), although the mechanisms responsible for this remain unknown. We tested whether insulin or leptin, two sympathoexcitatory hormones increased during pregnancy, contribute to this. Transport of insulin across the blood-brain barrier in some brain regions, and into the cerebrospinal fluid (CSF), was increased, although brain insulin degradation was also increased. As a result, brain and CSF insulin levels were not different between pregnant and non-pregnant rats. The sympathoexcitatory responses to insulin and leptin were abolished in pregnant rats. Blockade of arcuate nucleus insulin receptors did not lower SNA in pregnant or non-pregnant rats. Collectively, these data suggest that pregnancy renders the brain resistant to the sympathoexcitatory effects of insulin and leptin, and that these hormones do not mediate pregnancy-induced sympathoexcitation. Increased muscle SNA stimulates glucose uptake. Therefore, during pregnancy, peripheral insulin resistance coupled with blunted insulin- and leptin-induced sympathoexcitation ensures adequate delivery of glucose to the fetus.

Abstract: Pregnancy increases basal sympathetic nerve activity (SNA), although the mechanism responsible for this remains unknown. Insulin and leptin are two sympathoexcitatory hormones that increase during pregnancy, yet, pregnancy impairs central insulin- and leptin-induced signalling. Therefore, to test whether insulin or leptin contribute to basal sympathoexcitation or, instead, whether pregnancy induces resistance to the sympathoexcitatory effects of insulin and leptin, we investigated α-chloralose anaesthetized late pregnant rats, which exhibited increases in lumbar SNA (LSNA), splanchnic SNA and heart rate (HR) compared to non-pregnant animals. In pregnant rats, transport of insulin into cerebrospinal fluid and across the blood-brain barrier in some brain regions increased, although brain insulin degradation was also increased; brain and cerebrospinal fluid insulin levels were not different between pregnant and non-pregnant rats. Although i.c.v. insulin increased LSNA and HR and baroreflex control of LSNA and HR in non-pregnant rats, these effects were abolished in pregnant rats. In parallel, pregnancy completely prevented the actions of leptin with respect to increasing lumbar, splanchnic and renal SNA, as well as baroreflex control of SNA. Blockade of insulin receptors (with S961) in the arcuate nucleus, the site of action of insulin, did not decrease LSNA in pregnant rats, despite blocking the effects of exogenous insulin. Thus, pregnancy is associated with central resistance to insulin and leptin, and these hormones are not responsible for the increased basal SNA of pregnancy. Because increases in LSNA to skeletal muscle stimulates glucose uptake, blunted insulin- and leptin-induced sympathoexcitation reinforces systemic insulin resistance, thereby increasing the delivery of glucose to the fetus.

Keywords: BBB insulin transport; S961; arcuate; insulin resistance; leptin; pregnancy; sympathetic nerve activity.

Figure 1.. Insulin transport into brain.

The rate of transport across the blood-brain barrier of radioactive insulin after correction for vascular space using radioactive albumin was determined by the multiple-time regression analysis method. The transport rate into cortex (lower left panel; n=12, pregnant; n=13, nonpregnant) was significantly increased (*: P<0.05). However, significant differences for whole brain (upper left panel; P=0.096; n=12, pregnant; n=13, nonpregnant) olfactory bulb (upper right panel; n=12; pregnant; n=12, nonpregnant) and for cerebellum (lower right panel; n=12, pregnant; n=13, nonpregnant) were not observed.

Figure 2.. Vascular space, insulin transport into the CSF, and brain degradation of insulin in pregnant and nonpregnant rats.

A. The vascular space, as measured with radioactive albumin, was increased in pregnant compared to nonpregnant rats in whole brain (upper panel; n=14, pregnant; n=14, nonpregnant) and in specific brain regions (lower panel; n=14, pregnant; n=14 nonpregnant). B. The CSF/serum ratio for radioactive insulin (n=10, pregnant; n=8, nonpregnant), but not for radioactive albumin (n=8, pregnant; n=8, nonpregnant), was increased in pregnancy. The uptake of radioactive insulin was about 50 times greater than that of radioactive albumin consistent with its saturable transport across the BBB. C. Brain degradation of insulin increased during pregnancy. Grey circles indicate pregnant animals (n=14) and black circles indicate nonpregnant (n=14) animals. *: P<0.05; ***: P<0.001.

Figure 3.. The sympathoexcitatory responses to insulin are abolished in urethane-anesthetized pregnant rats.

In nonpregnant rats (left column), icv insulin increased LSNA, baroreflex control of LSNA by increasing the baroreflex maximum and gain, HR, and baroreflex control of HR by increasing the maximum and gain. In contrast, insulin only increased the gain of baroreflex control of HR in pregnant rats. *: P<0.05 within group; †: P<0.05 pregnant compared to nonpregnant rats.

Figure 4.. The sympathoexcitatory responses to insulin are abolished in α-chloralose-anesthetized pregnant rats.

In nonpregnant rats, icv insulin increased baseline and baroreflex control of LSNA and HR. In contrast, icv insulin had no effects in pregnant rats. *: P<0.05 within group; †: P<0.05 pregnant compared to nonpregnant rats.

Figure 5.. Blockade of ArcN InsR does not lower LSNA in pregnant or nonpregnant rats.

A. ArcN nanoinjections of insulin increase LSNA in nonpregnant rats, and this response was completely blocked by prior injection of the insulin antagonist, S961. B. Nanoinjections of aCSF or S961 do not significantly alter LSNA, HR, or MAP in pregnant (P) and nonpregnant (NP) rats. Basal values of MAP (in mmHg) were: 110±5 (NP+S961), 103±8 (NP+aCSF), 80±8 (P+S961), and 71±7 (P+aCSF). Basal values of HR (in bpm) were: 329±10 (NP+S961), 325±17 (NP+aCSF), 460±16 (P+S961), and 453±18 (P+aCSF). C. Histological maps illustrating ArcN injection sites for the experiments shown in A. and B. Note that all pregnant rats that received S961 (n=4) first received aCSF and that 4 of 6 nonpregnant rats that received S961 first received aCSF.

Figure 6.. The sympathoexcitatory responses to leptin are abolished in α-chloralose-anesthetized pregnant rats.

In nonpregnant rats, icv leptin increased baseline and baroreflex control of LSNA, SSNA, RSNA, and HR. These effects were not evident in pregnant rats. *: P<0.05 within group; †: P<0.05 pregnant compared to nonpregnant rats.