We thank Karatas et al for their letter highlighting the potential contribution of the autonomic nervous system (ANS) in mediating the effects of metformin in a model of maternal obesity. Pregnancy, characterized by maternal adaptation to and support of fetal growth, is accompanied by significant ANS changes. ANS activity changes with the progression of gestation, with an increase in vagal activity and lower sympathetic activity during the first trimester followed by lower vagal activity and increased sympathetic activity during the third trimester. Various conditions that accompany pregnancy can also affect ANS activity. As highlighted by Karatas et al, obesity/overfeeding and metabolic syndrome correlate with sympathetic overactivity. Together, these factors, along with the pregnant state, create a complex system/model to analyze.

Although obesity-enhanced sympathetic nervous system (SNS) activity has been linked to hyperinsulinemia and insulin resistance, the exact relationship between obesity/metabolic syndrome and the sympathetic activity is not completely understood. Sympathetic activity regulates metabolic responses and in turn, insulin, fatty acids, and other metabolic mediators regulate SNS activity. While metformin does not affect muscle-sympathetic nervous activity it may reduce SNS overactivity by improving insulin sensitivity, dyslipidemia, and/or other unknown factors that influence SNS.

In our study, a high-fat/high-sugar (HCAL) diet fed to female rats significantly increased prepregnancy weight and plasma leptin levels and in the pregnant dams, the HCAL diet increased maternal plasma high-density lipoprotein, triglycerides, cholesterol:high-density lipoprotein ratios, insulin, and leptin levels in the absence of maternal inflammation. Interestingly, metformin had no effect on these parameters in obese dams. By contrast, maternal HCAL feeding increased fetal plasma chemokine (C-C motif) ligand 2 and tumor necrosis factor-α and placental tumor necrosis factor-α and maternal metformin administration reversed these effects. Precisely how metformin counter-regulates HCAL-induced fetal inflammation is not known, but our data implicate nuclear factor κ B activation.

In the absence of metformin-mediated effects on maternal outcomes, our results do not support investigating metformin-mediated regulation of ANS during obesity-related maternal inflammation. However, our results do not rule out potential metformin-mediated control of fetal inflammatory (and possibly other) responses via ANS, particularly knowing that metformin crosses the placenta. These measurements may prove challenging. While several studies have measured maternal ANS activity via direct and indirect measurements, few studies have calculated fetal ANS activity. Fetal magnetocardiography signals are too weak to precisely evaluate fetal ANS <28 weeks’ gestation (in normal fetuses). One alternative might be to measure ANS activity in neonates. However, unless uniform technologies and protocols (controlling for factors affecting ANS, using consistent measurement times, etc) are employed, the results may be difficult to interpret and unable to discriminate between cause vs effect and pathologic vs physiologic responses.