Sex differences in diabetes and kidney disease: mechanisms and consequences

Abstract

Type 1 and type 2 diabetes, along with their accompanying hyperglycemia, are associated with a multitude of comorbidities including the development of diabetic kidney disease. Although the hallmarks of these metabolic disorders have been well characterized in population and animal studies, it is becoming increasingly apparent that diabetes manifests itself differently in men and women. This review summarizes the recent diabetic literature with a focus on known sex differences in clinical and preclinical studies. It explores the physiological differences of glucose handling and the development of diabetes between men and women. This review also uncovers potential mechanisms for these differences, honing in on the vital role that sex hormone signaling plays in the progression of diabetes and renal complications.

DIABETES AND DIABETIC KIDNEY DISEASE

Every year, millions of individuals are diagnosed with diabetes or prediabetes, which is most often marked by hyperglycemia and elevated hemoglobin A₁C levels (15, 63). In type 1 diabetes (T1D), or insulin-dependent diabetes, this disease originates as an autoimmune disorder where pancreatic β-cells lose their ability to produce insulin. In contrast, type 2 diabetes (T2D), or insulin-independent diabetes, is associated with obesity and insulin resistance in peripheral tissues. Despite the mechanistic differences, both T1D and T2D present with a multitude of comorbidities, including diabetic retinopathy, neuropathy, and nephropathy.

Diabetic nephropathy, or diabetic kidney disease, is the leading cause of end-stage renal disease (ESRD), with more than half of all cases attributed to these metabolic disorders (7). The kidney is responsible for filtering all blood glucose, and under conditions of hyperglycemia, the increased filtered load has been shown to induce oxidative damage and induce injury to the delicate vasculature and surrounding tubules (10). Nonetheless, the kidney has remarkable flexibility when it comes to glucose handling; even large fluctuations in blood glucose levels can be adequately controlled by the renal proximal tubule such that glycosuria is typically observed only under conditions of significant hyperglycemia (40, 41, 57).

Because of the rising prevalence, both T1D and T2D are extremely well studied; there are a number of excellent reviews on the development of diabetic kidney disease (61, 63, 64) and on the role that the kidney can play in combating hyperglycemia (44, 65, 70, 72). However, historically, the hallmarks of this disease have only been studied in male subjects, and yet it has become increasingly apparent that there are significant sex differences when it comes to renal glucose handling and the prevalence and progression of diabetes. The focus of this review is on these diabetic sex differences, both with respect to the disease manifestation as well as the proposed mechanisms for the disparities between the sexes.

RENAL GLUCOSE HANDLING

The kidney plays a major role in the regulation of blood glucose levels. Upon filtration at the glomerulus, all glucose is reabsorbed from the lumen of the kidney within the proximal tubule via two Na⁺-glucose cotransporters (SGLT1 and SGLT2). The net result is that little to no glucose is lost in the final urine. When loss of insulin production or sensitivity lead to increased circulating blood glucose levels, the task of handling this excess glucose load falls on the kidney. There is no clear consensus in the literature regarding the expression or activity of these transporters in the setting of T1D or T2D (57). Nonetheless, when filtered blood glucose levels exceed the transport maximum of SGLT1 and SGLT2 (as in cases of diabetes), glycosuria will occur. Historically, glycosuria has been viewed as a hallmark of uncontrolled diabetes, but with the development of SGLT inhibitors, this pathophysiological condition is now used as a mechanism to lower blood glucose (38, 41, 43, 57). As SGLT inhibitors gain in popularity, it is important to understand that male and female populations rely on these transporters to different extents. Recently, there has been a focus on characterizing the sex differences of all renal transporters across the nephron (66). Most studies that have examined SGLT1 and SGLT2 knockout (KO) mice have been performed only in male mice. However, higher expressions of SGLT1 and SGLT2 have been observed in female rats compared with male rats under standard conditions (4, 52, 53). Compared with male rats, female rats preferentially use these transporters over other channels/exchangers for Na⁺ reabsorption in the proximal tubule (31). The increased reliance on these glucose transporters in female animals could be linked to a higher incidence of glomerular hyperfiltration given the accompanied increase in Na⁺ reabsorption. However, whether or not these differences extend to humans is still unclear. Contrary to the studies performed in mice and rats, it does not appear that there are any sex differences in SGLT1 and SGLT2 expression in kidneys from men or women (67). Sex differences regarding activity of these transporters have yet to be explored. The sex-independent nature of SGLT1 and SGLT2 may explain why sex differences have not been reported in the effectiveness of the SGLT2 inhibitors on the market. There are, however, greater adverse side effects reported in women taking a SGLT2 inhibitor. These include higher incidence of urinary tract infections, genital infections, and diabetic ketoacidosis. In general, women with T2D are less likely to adhere to their drug treatment regimen compared with men (24), and, given these reported side effects, it is possible that this will be observed for SGLT2 inhibitors as well.

SEX DIFFERENCES IN DIABETIC PREVALENCE AND RENAL COMPLICATIONS

The prevalence of both T1D and T2D is rapidly increasing, and recent population studies have detected key differences between the sexes (Fig. 1). On the whole, it appears that women are less prone to developing both metabolic disorders, although diabetic comorbidities including cardiovascular disease and end-stage kidney disease (ESKD) are more likely to occur in women than in men.

Fig. 1.

Diabetic and renal sex differences. Recent reports have noted differences in the prevalence and severity of diabetes and diabetic kidney disease (DKD) between men (A), premenopausal women (B), and postmenopausal women (C). Overall, men have been found to have an increased incidence of type 2 diabetes (T2D) and equal prevalence of type 1 diabetes (T1D) compared with premenopausal women. Postmenopausal females, however, have an increased risk of developing DKD and end-stage kidney disease (ESKD) and glomerular hyperfiltration. These changes have been linked, in part, to the differences in gene expression and sex hormone signaling within the kidney that contribute to the regulation of blood pressure (BP) and glomerular filtration rate (GFR) (kidney insets). ER, estrogen receptor; ESRD, end-stage renal disease; GPER, G protein-coupled estrogen receptor; HFD, high-fat diet; OAT, organic anion; OCT, organic cation; SGLT, Na⁺-glucose cotransporter; E, estrogen.

T2D: Prevalence

The rise of T2D in Western societies has been attributed, in part, to the ongoing obesity epidemic. On the whole, adult women are more likely to be obese, but men have a higher risk of developing T2D (14). Although counterintuitive, this can be explained by the types of adipose tissue that accumulates during weight gain. A high body mass index more strongly correlates with subcutaneous adipose tissue (8). However, it is an increase in visceral adipose tissue that is linked to the development of T2D (16). This may be explained by the differences in expression of androgen (high) and estrogen (low) receptors in this type of fat. As discussed below, changes in circulating levels of sex hormones have been linked to the development of diabetes, and this tracks with the expression of the receptors in visceral fat. No clear sex differences in visceral adiposity have emerged between men and women. Men have been found to be less physically active, have higher rates of alcohol consumption, and have a higher daily energy intake compared with women (2), and these risk factors may predispose them to the development of T2D. In a recent cross-sectional and prospective study, there was a 61% higher chance of a T2D patient being a man as opposed to a woman (2). This finding was mirrored in other studies examining European and North American prevalence of diabetes and impaired fasting blood glucose with a clear tilt toward male prevalence (1, 15, 29). Although this male dominance is clear for adults of European decent, there are some discrepancies in other populations. The incidence of T2D in those of Asian descent has been only weakly associated with male prevalence (2, 32), and although there is an alarmingly linear increase in the rate of T2D among youths, this increase is seen in both men and women (42). As with all population studies, it is clear that genetic and lifestyle factors lead to differences even within similar populations.

T1D: Prevalence

T1D is considered an autoimmune disease and usually asserts itself in the younger population, with most diagnoses occurring before the age of 20 yr old. Although there are clear ethnic and population differences in prevalence (35), whether sex plays a role is slightly more muddled. One recent study examining the rate of diagnosis from 2002 to 2012 found that the increase in diagnosis was attributed more strongly to men than women (42), and an older study also identified a slight association with male predominance (5). Conversely, there are several other reports showing that both men and women are equally affected by this disease (35). In particular, the SEARCH for Diabetes in Youth study that was initiated in 2000 by the Centers for Disease Control and Prevention and the National Institute of Diabetes and Digestive and Kidney Disease found that although the prevalence of T1D rose significantly from 2001 to 2009, there was no clear sex difference (19). Overall, these studies are at odds with the notion that women are more prone to developing other autoimmune disorders such as rheumatoid arthritis and lupus (46). This inconsistency may be attributed to the age of onset occurring during the prepubescent years before the rise in circulating sex hormone levels (see below) (35).

Diabetic Complications

Although diabetes, especially T2D, asserts itself in a male-dominated fashion, how this disease manifests itself is also governed by sex. In particular, studies have found that women who do develop diabetes are at a heightened risk of developing chronic kidney disease later in life. When the progression of diabetic kidney disease was tracked over time, diabetic women were found to be more susceptible to changes in glomerular filtration rate (GFR), with obesity and vascular stiffness positively correlating with an overall decline in GFR (7, 13, 58). In early stages of T1D, women have been found to develop glomerular hyperfiltration at a significantly higher rate than men (7, 34). This increased filtration rate has been shown to worsen the progression of ESKD by speeding up damage to the filtration barrier and likely contributes to the female-dominated renal disease. It should be noted that diabetic renal complications manifest more than 10 yr later in women than in men (9). Thus, although women are more prone to developing ESKD, the overall progression of the disease is slowed. In fact, this induction of ESKD generally coincides with the woman entering her postmenopausal period, which has been linked to a worsened outcome (Fig. 1). These observations no doubt identify a major potential mechanism for these differences and are explored in more detail below.

POTENTIAL MECHANISMS: THE ROLE OF SEX HORMONES

Perhaps the first indication that sex hormones contribute to the development of diabetes came when population studies were performed on postmenopausal women. Unlike in premenopausal women who are protected from the development of diabetes, this protection all but disappears in postmenopausal women, whose circulating levels of endogenous estradiol are greatly diminished (13, 37, 50). This strongly implicates sex hormones and signaling in the protection or risk factors for the development and progression of diabetes.

Estrogen and Estrogen Receptor Signaling

Circulating estrogens, converted from androgens by aromatase, initiate estrogen receptor (ER) signaling in a variety of tissues inducing diverse cellular processes. Clinical and preclinical studies have identified links between plasma estrogen levels and diabetic susceptibility as well as diabetic kidney disease.

In animal models, T1D can be induced by the injection of streptozotocin (STZ), an alkylating agent that causes pancreatic β-cell apoptosis leading to the classical presentation of diabetes, including hyperglycemia and impaired glucose tolerance (62). It has long been understood that female mice and rats are much less susceptible to the induction of T1D via STZ (47, 51), but this protection is abolished when estrogen and ER signaling are downregulated. On the whole animal level, knockout of ERα or one of its splice variants or the prevention of estrogen generation in aromatase KO mice have been shown to increase pancreatic β-cell apoptosis in female mice, leading to hyperglycemia, glomerular hyperfiltration, and proteinuria (22, 30). Additionally, ovariectomized female animals have been found to be more susceptible to induction of both T1D and T2D (20). These data imply that circulating estrogen, produced by the sex gonads, and ER signaling have protective roles when it comes to the development of diabetes.

The protective effects of estrogen may be due to direct effects on the kidney or indirect effects of circulating estrogen (28). Estrogen has been shown to have both pro- and anti-inflammatory effects, and ERs are present on immune cells (both antigen presenting cells and T cells). These receptors have been shown to modulate the immune response, which may contribute to the protection of the kidney during periods of hyperglycemia (15). Renal cells themselves express both full-length estrogen ERα and ERβ as well as several splice variants, and estrogen has been shown to play a renoprotective role when it comes to the development of diabetic kidney disease (Fig. 1) (22, 33, 36). ER signaling (ERα) has been found to have a direct effect on gene expression within the juxtamedullary region of the cortex in both mice and rats (23), suggesting that the renoprotection may be due to direct activation of ERα in the kidney. Thus, the protective effects of estrogen may be through direct renal ERα signaling or through indirect extrarenal signaling pathways.

Although classical steroid hormone receptors (ERα and ERβ) are likely responsible for estrogen-mediated gene expression changes in the kidney, recently, a membrane-bound G protein-coupled receptor (GPCR) has also been found to be activated by estrogen and estrogen analogs (71). G protein-coupled estrogen receptor (GPER or Gpr30) is ubiquitously expressed throughout the body and may add another layer of complexity to estrogen signaling. Whole animal female GPER KO mice present with elevated plasma glucose levels and decreased insulin responsiveness (39), suggesting a link to the development of diabetes. Furthermore, renal and vasculature-expressed GPERs have been shown to contribute to blood pressure and GFR regulation (Fig. 1) (11). Clearly, further study into the renal-specific functions of GPER is required to understand if it may also be involved in the progression, or protection, of diabetes and diabetic kidney disease.

Although estrogen does appear to play a protective role in the development of diabetes, it may also be associated with the increased prevalence of ESKD in the female population. Estradiol has been shown to upregulate renal angiotensin receptor type I, leading to increased efferent arteriole resistance. This can partially mediate glomerular hyperfiltration observed in the early stages of diabetes (7). Thus, this hormone and its downstream signaling likely have both protective and injurious roles in the kidney and elsewhere.

Testosterone and Sex Hormone-Binding Globulin

There is a clear sex dimorphism when it comes to testosterone and the development of diabetes. For all of the studies that show that estrogen and ER signaling provides renal and systemic protection in the development of T2D in women, the opposite is true for testosterone. Women with higher plasma testosterone levels have been found to be at a heightened risk of developing T2D, and lower levels correlate with a lower perceived risk (12). Conversely, in men, higher levels of testosterone are associated with a lower risk, whereas hypoandrogenism is seen as a risk factor. This may be due, in part, to the protective role that testosterone appears to play on the pancreas. In situations of glucotoxicity, higher levels of testosterone significantly decrease apoptosis in the pancreatic β-cells by acting as an antagonist for angiotensin II type 1 receptor (27). Androgens have also been linked to the development of hypertension. Given the strong connection between diabetes and cardiovascular disease (including hypertension), testosterone levels may also influence the development of hypertension and hypertension-induced renal injury (49). Recently, androgens were found to modulate the renal vascular response to angiotensin II (60). This may contribute to the development of diabetes-induced hypertension, especially in women who are more prone to the development of these diabetic complications.

Recent studies have also honed in on the role that sex hormone-binding globulin (SHBG) may have in diabetes. As a circulating protein, it binds to both testosterone and estrogen to control the bioavailability of these sex hormones. Given that it has a stronger affinity for testosterone, it is often viewed as a way to control the levels of free testosterone in circulation. In women, elevated levels of SHBG were associated with a decreased risk of developing diabetes, implying that its control of testosterone has a direct effect on the risk of developing diabetes (12). This was also observed in men; slightly lower levels of SHBG correlated with an increased risk of T2D (12). Unbound SHBG has also been shown to bind to membrane-bound receptors on testosterone-target tissues where it can act to antagonize or agonize the effects of sex hormones, and this may suggest that SHBG levels are an independent risk factor outside of controlling the bioavailability of sex hormones (45). Furthermore, polymorphisms within SHBG have been identified that may correlate with an increased or decreased incidence of T2D (12).

Whereas most studies examining sex hormones have focused on T2D, there is no clear link between testosterone levels and the development of T1D (6, 25). In preclinical T1D studies, however, testosterone has been shown to accentuate hyperglycemia in STZ-treated rats and predispose animal models to the development of T1D and T2D (47).

Although circulating levels of sex hormones undoubtedly contribute to the development and progression of diabetes in both men and women, the absolute levels of these hormones may be only half the story. Recent studies have identified that the balance of these hormones (presented as a ratio of testosterone to estrogen) may be the more appropriate risk factor. Alterations in this ratio in both men and women and in animal models show that an increase in testosterone:estrogen in female subjects and a decrease testosterone:estrogen in male subjects is linked to the development of diabetic kidney disease and adipocyte function (54, 69). Thus, a balance is likely the key to limiting the development of diabetes in both men and women.

POTENTIAL MECHANISMS: UNDERSTUDIED GENES

Over the years, numerous animal models have been used to study diabetes and the development of diabetic kidney disease (26). For T2D, these models include an ad libitum high-fat diet feeding (42–60% kcal from fat for >8 wk), the obese Zucker rat, and the leptin/leptin receptor-deficient ob/ob and db/db mice. STZ is commonly used to induce T1D in C57BL/6J or DBA/2J mice and rats, and T1D also manifests in Akita mice that harbor a mutation in the insulin 2 gene, leading to reduced insulin secretion and a decrease in pancreatic β-cells. Although the mode of action for these animal models is diverse, a common theme is that the female animals are less susceptible to the development of either T1D or T2D unless they are further challenged (i.e., high-fat diet feeding or castration) (3, 17, 21, 30, 48, 59, 68). Although these models make it difficult to study the disease progression in the female population (recall that women are actually more prone to the development of diabetic kidney disease), they are useful for identifying underlying differences between male and female subjects and their susceptibility to diabetes.

Numerous sex differences have been observed in the expression and function of renal transporters (4, 52, 53, 66). Included in this list is the family of organic anion and cation transporters that are expressed throughout the kidney, where they contribute to the secretion of endogenous and exogenous drugs including many diabetic pharmaceuticals (3). In lean Zucker rats (nondiabetic), female rats exhibit increased gene expression of both organic anion and cation transporters compared with both their lean and obese male counterparts. However, when these female rats are fed a high-fat diet to induce obesity and renal damage, the expression of these transporters is decreased (3). These changes may contribute to the severity of the disease progression in the diabetic female population and its ability (or inability) to effectively secrete drugs and other endogenous compounds.

Along the same lines, another study using a diet-induced obesity model to mimic early stages of T2D found that more than 1,000 genes are differentially expressed in the renal cortex between obese and control male mice (18). In followup TaqMan quantitative PCR analysis, they determined that the vast majority of these gene expression changes were only observed in the male mice even when female mice were maintained on the diet for longer to match the weights of the male mice. Indeed, many of these identified genes had not been previously linked to diabetes or diabetic kidney disease. Of note, many of the downregulated genes were found to be associated with the spliceosome, whereas both upregulated and downregulated genes were associated with a variety of metabolic pathways (18). Although it remains to be determined whether these gene expression changes are being driven by sex hormone signaling, this study unveiled a plethora of genes that may contribute to the sex differences observed in patients with T2D.

Finally, in our own work, we have unveiled sex differences with regard to the function of a renal olfactory receptor, Olfr1393 (55). Olfr1393 has been previously found to be expressed in the renal proximal tubule, where it contributes to glucose handling possibly as a regulator of the SGLTs. Under normal, nondiabetic conditions, Olfr1393 KO mice present with glycosuria, with a stronger phenotype in female animals compared with male animals (55). However, when these mice were challenged with a high-fat diet to induce early stages of T2D, male KO mice exhibited a significant improvement in glucose tolerance, whereas both male and female KO mice had an attenuation in glomerular hyperfiltration (56). Similar findings have been observed in Olfr1393 KO animals that were challenged with STZ to induce T1D. Collectively, these findings indicate that Olfr1393 can contribute to the progression of both T1D and T2D, with a stronger effect in male mice. Nonetheless, the increased glycosuria in female Olfr1393 KO animals, coupled with the attenuation in diabetic hyperfiltration, indicates that Olfr1393 plays a strong role in renal function in the female population. Studies in susceptible diabetic female Olfr1393 KO mice are necessary to truly understand the role of this receptor in both male and female populations.

THE FUTURE OF DIABETES RESEARCH

Diabetes is a multisystem disease. It is clear that sex and circulating sex hormones are variables when it comes to understanding its development and progression. With this knowledge comes a plethora of new findings that have unveiled novel signaling pathways and changes in protein activity and gene expression that are all contributing to these differences. As is the case for the expression differences for SGLT1 and SGLT2, it will be important to further investigate the novel signaling pathways to see if the studies performed in animal models can be translated to diabetic men and women. From a clinical standpoint, there is a clear discrepancy between men and women in regard to the side effects of diabetic drugs (see Ref. 24 for a comprehensive list). This may stem from the lack of preclinical studies that have been performed in both sexes. In particular, SGLT1 and SGLT2 KO mouse models have been mainly characterized in male mice. Expanding these studies to be performed in both male and female animals (and in pregnant female subjects) will go a long way toward informing the scientific community about the physiological and pathophysiological differences between men and women.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Disease Grant K01-DK-106400 (to B. D. Shepard).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

B.D.S. prepared figures; B.D.S. drafted manuscript; B.D.S. edited and revised manuscript; B.D.S. approved final version of manuscript.