Prolactin Receptors Are Critical to the Adaptation of Islets to Pregnancy

During pregnancy there is an enhanced need for insulin to accommodate the growing fetal compartment as well as the substantial increase in insulin resistance. Failure of the islets to adapt to this increased demand for insulin leads to gestational diabetes. The report by Huang et al. (1), using an in vivo model of prolactin (PRL) receptor deficiency, provides an important validation of the hypothesis that β-cell PRL receptors are central to mechanisms whereby islets adapt to pregnancy (Fig. 1). This study showed not only that PRL receptor deficiency leads to gestational diabetes, but the genetic phenotype of the mother influences the outcome of islet development in the fetus. During pregnancy, islet mass increases, and insulin secretion becomes increasingly sensitive to glucose stimulation. The figure highlights cellular events that occur in islet β-cells as they adapt to pregnancy. These were determined by comparing the results of lactogen-treated islets with those observed in islets during pregnancy. This has led to the hypothesis that β-cell PRL receptors have a central role in the adaptation of islets to pregnancy. The paper by Huang et al. (1 ) adds validity to this model by demonstrating that a consequence of PRL receptor deficiency is that it leads to a condition of gestational diabetes in which the islets fail to adapt to the increased need for insulin. Glut-2, Glucose transporter-2; Jak2, Janus kinase 2; PI3K, phosphatidylinositol 3-kinase. That islets undergo changes during pregnancy was observed as early as 1930 (2), when Cramer concluded that islets are a very plastic tissue that undergo considerable changes, especially during pregnancy. The changes he observed included increased secretory activity, islet growth, and neogenesis. Direct evidence for functional changes in islets did not occur until the development of a sensitive insulin assay in the 1960s, when it was reported that there was a progressive increase in both fasting and glucose-stimulated insulin secretion throughout the course of human pregnancy (3). This and subsequent research led to the characterization of pregnancy as a condition of elevated serum insulin levels, slightly lower blood glucose levels, and peripheral insulin resistance. The short-term regulation of insulin secretion is achieved by elevating blood glucose. However, if this were the only mechanism available for increasing insulin secretion during pregnancy, there would be a need for persistent hyperglycemia, a condition detrimental both to the mother and the developing fetus. Thus, when there is an increased need for insulin over a prolonged period of time, such as occurs in pregnancy, islets must undergo adaptive changes. The outcome of the up-regulation must be such that there is enhanced insulin secretion at normal glucose levels. There are three basic mechanisms whereby one can increase insulin secretion at normal glucose concentrations: increase the islet β-cell mass, increase the sensitivity of insulin secretion in response to glucose, or both. Evidence that there is an increase islet/β-cell mass comes from a number of studies (4, 5, 6, 7). These studies indicate that there is an approximate 2-fold increase in islet mass. DNA content per islet also indicates an increase in islet mass, and both morphometric and DNA to protein ratio methods indicate β-cell hypertrophy as well as β-cell hyperplasia (8, 9, 10, 11, 12). Evidence for increased β-cell mitosis comes from tritiated thymidine and bromodeoxyuridine labeling of islets during pregnancy (4, 13, 14). Overall, these reports indicate that islet mass doubles, and the increase is a consequence of new β-cell formation as well as an increase in β-cell size. There is less information on whether islet neogenesis occurs during pregnancy. One report failed to detect evidence for islet neogenesis during pregnancy but did show that another condition of persistent hyperlactogenemia may result in islet neogenesis (15). There is also evidence for an increase in glucose-sensitive insulin secretion. During rodent pregnancy, fasting serum insulin levels are increased, and glucose levels are decreased (8, 16, 17), similar to that observed in human pregnancy. Green and Taylor (9) showed a leftward shift of the glucose-stimulated insulin secretion response curve in islets isolated from pregnant rats. In a similar study, we used perfused pancreas preparations to examine insulin secretion throughout rat pregnancy. A leftward shift in glucose-responsive insulin secretion as well as above-threshold insulin release was first noted on d 10 and peaked by d 15 (4). The threshold for glucose-stimulated insulin secretion decreased from 5.7–3.3 mm by d 15 pregnancy. The lowering of the threshold for glucose-stimulated insulin secretion is a key feature of islets as they adapt to pregnancy. It is through this maneuver that a large increase in insulin secretion can be achieved at fasting blood glucose levels, as is seen during pregnancy. The most important changes in islet adaptation to pregnancy are enhanced insulin secretion and enhanced β-cell mass. Candidate hormones must be shown to induce these changes. Although a variety of hormones increase during pregnancy, only PRL and placental lactogen, which acts through PRL receptors, are capable of inducing the changes that occur in islets during pregnancy. A number of in vivo and in vitro studies comparing the effect of lactogenic hormones on islets with the effects of pregnancy on islets have been previously summarized (18). These studies were done using species homologous hormones or heterologous hormones with known receptor specificity (19, 20, 21). In all instances lactogens (PRL or placental lactogen) induce the same changes in islets as those observed during pregnancy, and led to the hypothesis that β-cell PRL receptors are central to mechanisms whereby islets adapt to pregnancy (18). Importantly, the onset of placental lactogen secretion occurs at the same time that the earliest changes in β-cell division and insulin secretion are detected (4). These studies were done with rodents raising the question about their relevance to humans. This question is particularly pertinent because human placental lactogen has 85% sequence homology with human GH (21). However, human placental lactogen binds poorly to the human GH receptor but with high affinity to the human PRL receptor. Although limited, studies on human islets indicate that lactogen treatment increases insulin secretion and islet cell proliferation (4, 22). Thus, it is quite likely that in humans as well as rodents, it is lactogenic hormones that are responsible for up-regulating islets during pregnancy. In support of the lactogen hypothesis is that molecular changes in β-cells, expected to be responsible for those observed during pregnancy, appear to be regulated by PRL receptor signal transduction pathways. Within islets only β-cells possess PRL receptors (23, 24, 25, 26). PRL binding to its receptor leads to activation of Janus kinase, which in turn phosphorylates signal transducer and activator of transcription (STAT) 5, resulting in its translocation to the nucleus where it binds to target genes at specific binding sites (27). This pathway has been demonstrated and its activation by PRL characterized in islet β-cells (23, 28, 29, 30, 31). For PRL receptors to regulate β-cell proliferation, its signal transduction events must be involved in the regulation of cell cycle control mechanisms. For recent reviews on cell cycle control in β-cells, see Refs. 32 and 33 . Evidence indicates that regulation of the D cyclins is not only important in β-cell growth but appears to be targets for regulation by the PRL receptor by way of STAT5 (34, 35, 36, 37). Lactogen stimulation increases cyclin D expression. Constitutive active STAT5 increases replication and cyclin D expression, and overexpression of a dominant negative STAT5 prevents lactogen-induced replication. Recently, a role for menin in adaptation of islets to pregnancy was reported (38). This study showed that during pregnancy, levels of menin and cell cycle inhibitors, p18 and p27, decline in maternal islets. This sequence of events is STAT5 dependent, where activated STAT5 leads to increased expression of Bcl6, and this in turn represses Men 1 expression followed by a decrease in p18 and p27. These molecular events were studied during pregnancy as well as in islets treated with lactogens in vitro. Thus, several lines of evidence support the hypothesis that islet β-cell expansion during pregnancy is regulated by activation of the PRL receptor. There is also evidence that PRL receptor activation leads to increased glucose-stimulated insulin secretion with a decreased glucose stimulation threshold. Glucose metabolism is central to glucose-regulated insulin secretion, and glucokinase is the critical glucose sensor. Although a number of treatments can lead to a lower glucose stimulation threshold (39), importantly, PRL treatment has been shown in both in vitro and in vivo studies to lead to enhanced insulin secretion, characteristic of pregnancy (12, 19, 20, 40, 41). It appears that the key factor in this increase in glucose sensitivity is an increase in glucokinase and, consequently, an increase in glucose metabolism. Glucokinase protein, enzyme activity, and mRNA are increased during pregnancy, and to a similar extent, by PRL treatment of islets in vitro. The extent of glucokinase increase is that which is expected to account for the decrease in the glucose stimulation threshold observed during pregnancy. The upstream glucokinase promoter contains multiple STAT5 binding sequences, and evidence supports a role for their involvement in the regulation of glucokinase expression by PRL. To understand the physiological roles of PRL receptors, a PRL receptor-deficient mouse model was developed (42). Islet β-cell mass, insulin content, and insulin secretion are reduced in these mice, indicating a role for PRL receptors in the normal development and function of islets. However, female mice are sterile, so the effect of PRL receptor deficiency was not studied during pregnancy. STAT5 knockout mice have been studied in which the deletion is targeted to β-cells (43). These mice show mild glucose intolerance with age and possibly during pregnancy. Whether other islet STATs such as STAT1 and STAT3 compensate for the deleted STAT5 was not investigated nor was any information provided on islet mass in these animals. Although most efforts have focused on STAT5 mediation of PRL receptor signal transduction, studies also indicate that phosphatidylinositol 3-kinase and MAPK pathways are also involved. Both in vitro experiments and examination of islets during pregnancy indicate that there is increased activity in these pathways as a consequence of lactogen activation (44, 45). The aforementioned reports provide a large body of evidence supporting the hypothesis that PRL receptors are central to the adaptation of islets to pregnancy. However, this hypothesis has not previously been tested to determine whether the absence of PRL receptors would result in a failure of islets to adapt to the increased demand for insulin. The studies by Huang et al. (1) fulfill this important need. In this study, heterozygous null PRL receptor mice were studied during pregnancy. There was impaired glucose clearance, decreased glucose-stimulated insulin release, higher nonfasted blood glucose, and lower insulin levels. In addition, the increase in islet density, β-cell number, and mass that occurred in normal mice during pregnancy failed to occur in the heterozygous null mice. These observations are the first in vivo demonstration that PRL receptors are required for normal islet function during pregnancy. What remains to be determined is a more complete delineation of the PRL receptor pathways, their relative importance, and possible redundancies. It will also be important to determine whether a PRL receptor knockout restricted to β-cells will have the same pregnancy outcome as observed in these experiments. The early paper by Green and Taylor (46) showed that restricting caloric intake in pregnant animals to that of controls prevented the changes in insulin secretion. PRL treatment results in increased feeding behavior (47) and weight gain (48). Thus, it may be that PRL receptors both in the islet and the brain contribute to islet adaptation to pregnancy. An added feature to the Huang et al. (1) study was the observation that the genetic background of the mother significantly impacted the severity of the phenotype in the offspring, importantly demonstrating that the fetal environment as well as genetics determine the outcome of developing islets. Prolactin; signal transducer and activator of transcription.