**1. Introduction**

Insulin-mediated glucose disposal widely varies in its sensitivity across populations [1], and depending on the level of compensatory hyperinsulinemia, resistance to insulin can or cannot be overcome. This insensitivity can lead to glucose intolerance, high-plasma triglyceride levels, low high-density lipoprotein cholesterol (HDL-C) concentrations, and hypertension [2]. If not overcome, it will lead to type 2 diabetes (T2D) development. Simultaneously, this collection of abnormalities has been linked with a significantly increased risk of cardiovascular diseases (CVD) [2]. The proposed association was formally established in the report of the adult treatment panel III of the National Cholesterol Educational Program. Formerly known as syndrome X, insulin resistance (IR) syndrome represents the insensitivity of the peripheral tissues (e.g., muscle, liver, adipose tissue) to the effects of insulin. IR is defined as a state wherein normal insulin concentrations evoke a less-than-normal biological response [3]. Although not a disease per se, it may be understood as a condition that increases the likelihood of developing a cluster of abnormalities, such as glucose intolerance, dyslipidemia, endothelial dysfunction, hemodynamic changes, increased testosterone secretion, and sleep-disordered breathing [4]. Additionally, it does not necessarily lead to, but instead increases the risk of clinical syndromes like CVD, essential hypertension, polycystic ovary syndrome, nonalcoholic fatty liver disease, and certain forms of cancers [2]. Metabolic syndrome (MS) is a collection of cardiometabolic risk factors. It is often characterized by IR that may provide a link between physical inactivity and MS development [5]. Most importantly, IR and impaired insulin secretion play a crucial role in the pathogenesis of T2D [6]. Although the potential causative connection has been shown for certain pollutants and IR development, the role of environmental chemicals in IR pathogenesis and the molecular mechanisms contributing to its development have not been fully elucidated ye<sup>t</sup> [7].

Over fifty years ago, there was evidence suggesting that certain inorganic elements may alter glucose utilization in target tissues via sulfhydryl modification [8]. Since the 1970s, there has been a growing body of evidence that supports the involvement of these elements in various metabolic disorders associated with impaired β-cell function [9–12]. Several inorganic elements may influence the proper regulation of insulin/glucose homeostasis. They can be divided into two categories—hyperglycemic and hypoglycemic—based on their effects on insulin production or insulin action at target tissues [13]. The categorization of different elements is displayed in Table 1.

**Table 1.** Categorization of different elements as hyper- or hypoglycemic [13].


Our interest is focused on the actions of cadmium (Cd) and its ability to alter numerous cell and organ systems. This toxic metal is characterized by a high soil-to-plant transfer rate, which makes the dietary exposure to this metal inevitable and a matter of grea<sup>t</sup> public health concern [14,15]. Another important source of exposure of the general population to Cd is smoking, as shown by elevated Cd levels in the smokers' blood [16]. Once inside the organism, Cd has a long biological half-life, with estimates reaching 45 years for humans [17]. Whole-blood and urinary Cd concentrations are widely accepted markers of Cd exposure and accumulation [14,18]. Long-term exposure to Cd has been associated with various conditions, including various renal syndromes, osteoporosis and osteomalacia, CVD, and different types of cancer [14,15,19–25]. Its endocrine-disrupting properties have also been shown, suggesting its possible effects on estrogenic activity [26–29], alterations in semen and the testis [30–32], and a role in thyroid disorders [33–35]. The mode of toxic Cd actions in the organism have been extensively investigated, but still not entirely elucidated, mainly because they may change with the dose and the detailed health status of the exposed subjects. Recent reviews by Ðuki´c-Cosi´ ´ c et al. [36] and Wallace et al. [37] have summarized the most critical mechanisms of Cd toxicity: changes in gene expression and DNA repair, interference with autophagy and apoptosis pathways, oxidative stress induction, interaction with bioelements, and epigenetic modifications. These mechanisms underlie the possible role of Cd as a metabolic disruptor. Its direct pancreatotoxic actions are buttressed by the Cd's ability to accumulate in the pancreas, as shown in many human studies [22,38–40]. Similar results have been obtained in animal studies as well [41,42], with a dose-dependent accumulation pattern observed

in rats [22]. In vitro studies have shown not only dose- but also a time-dependent accumulation of Cd in insulin-producing β-cells [43]. Furthermore, having in mind Cd's deleterious e ffects on the kidneys [14] and the role of kidneys in glucose homeostasis, which is accomplished through the processes of gluconeogenesis, glucose filtration, glucose reabsorption, and glucose consumption [44], it could be presumed that Cd's e ffects in the kidneys do contribute to IR development to a certain point. One of the first reports of Cd's ability to promote the development of diabetes appeared nearly four decades ago, when Merali and Singhal reported that neonatal exposure to Cd resulted in IR and diabetes development in rats [45]. The present review aims to provide an overview of the potential role of Cd exposure in IR collected in human, animal, and cell studies, focusing mainly on those conducted in the last two decades. Furthermore, the review will also briefly discuss the existence of a threshold for this e ffect.

#### **2. Insulin Resistance and Cadmium: Human Studies**

Human studies investigating the link between Cd exposure and IR are limited and have yielded somewhat conflicting data. The first association between Cd content, impaired fasting glucose (IFG), and diabetes was suggested by Schwartz et al. [46], who analyzed the data of the third National Health and Nutrition Examination Survey (NHANES III). This large, cross-sectional study revealed a significant, dose-dependent association between Cd urinary levels and IFG/diabetes prevalence, regardless of the source of Cd exposure. Another study analyzing NHANES participants for the years 2005 through 2010 aged ≥40 years revealed a complex, non-linear association between higher Cd levels and prediabetes state. Since this association varied across smoking groups and age, the authors suggested a complex relationship between Cd exposure, age, smoking habits, and prediabetes odds. Nevertheless, since no di fferences in the Homeostatic Model Assessment for IR (HOMA-IR) were observed across the exposure quintiles, the authors marked changes in IR as an unlikely cause of Cd effects on glucose levels [47]. The relationship between Cd exposure and T2D occurrence was confirmed in the study comparing the levels of Cd in various biological samples (blood, urine, and scalp hair) of patients having T2D (age range 31–60) with the levels in control subjects. Significantly higher levels of Cd were observed in scalp hair samples from patients compared to control individuals, along with a similar trend in observed values obtained from blood and urine samples [48]. Studies that followed tried to establish possible mechanisms of Cd in disturbing glucose metabolism. Pizzino et al. [49] investigated glycemic control, oxidative stress markers, and urinary Cd levels from 111 males (aged 12–14 years) living in polluted areas of Sicily and control age-matched population of 60 males living 28–45 km from the contaminated site. The results revealed altered glycemic control in adolescents that was associated with higher Cd levels. Altered glycemic control was demonstrated by the robust correlation between Cd and the homeostatic model assessment of HOMA-IR, along with markers of disturbed oxidative status. The authors identified oxidative stress disturbance to play a role in Cd-induced IR [49]. Apart from oxidative stress induction, Cd's ability to induce inflammation was also investigated. In a case-control, cross-sectional study, including 120 healthy controls and 105 systemic lupus erythematosus (SLE) patients, the relationship between various trace elements with SLE diagnosis, disease activity, and IR was assessed [50]. Serum levels of Cd were higher in patients with IR. Cd's ability to impair insulin sensitivity was connected to the positive association between Cd and the C-reactive protein (CRP). Namely, CRP as an inflammatory marker was shown to have a role in the development of diabetes [51].

Conflicting results from multiple studies have complicated the interpretation of Cd-mediated hyperglycemia. Jacquet et al. provided a comprehensive review of these conflicting reports in a 2016 review, where they categorized the e ffects as having "associations", "no association", or "potentiation" in diabetes [52]. Examples of this data variability is reflected in a study of Scandinavian Caucasian women, aged 64, which showed conflicting data with the previously mentioned studies [53]. Two thousand five hundred and ninety-five women were screened with oral glucose tolerance tests to identify subjects with T2D, impaired glucose tolerance (IGT), and normal glucose tolerance (NGT), and samples were

randomly chosen from each group. Cd concentration was measured in blood and urine samples, while the HOMA-IR calculations assessed the acute insulin response. A follow-up examination was also performed. Both cross-sectional and prospective studies showed no association between Cd exposure and increased risk of T2D, impaired insulin secretion, or insulin sensitivity [53]. The discrepancies in these results with previously published data were explained by di fferences in the occurrence of T2D risk factors in investigated groups, such as age, obesity, smoking, lifestyle, and ethnic predispositions. Furthermore, all women studied were aged 64+, which does not inform on the behavior of other populations. Interestingly, however, Wu et al. [54], in their PRISMA-compliant systematic review and meta-analysis based on 11 cohort/cross-sectional studies included in the meta-analysis, determined that high Cd exposure may not be a risk factor for diabetes development. Moreover, Anetor et al. [55] found significantly lower Cd blood levels from diabetic patients when compared to controls. This conflicted finding was partly attributed to the observed higher Zn levels in the same group of patients and the relatively small sample size (65 participants). Indeed, Cd belongs to the same group of elements as zinc, and the number of common biological targets of the two metals abound [56–60]. The role of zinc in insulin regulation has been extensively examined [61–63]. Zinc associates with insulin in exocytosis granules, and a significant amount of zinc is subsequently released into the extracellular space. Zinc can act as an autocrine mediator, a ffecting the activity of surrounding β-cells [61]. Defects in the transporters, such as SLC30A (ZNT) for the Zn provision for insulin secretion or SLC39A (ZIP) for replenishment, result in reduced intracellular zinc and a reduction in functional insulin release [62]. Changes in functional insulin release and alterations in zinc homeostasis may thus combine to contribute to glucose intolerance and IR [63].

Swaddiwudhipong and associates conducted a series of studies in Cd-exposed adults from Mae Sot District, Tak Province, in northwestern Thailand. This region was contaminated by the Cd-rich waste of Zn mines, and the population was Cd-exposed by the consumption of rice and other crops irrigated by downstream water. No association between urinary Cd levels and an increase in diabetes prevalence and risk were found [64,65]. The follow-up examination conducted on 436 persons who had urinary Cd levels >5 μg/g creatinine revealed a significant increase in the prevalence of diabetes compared to baseline [66].

The question of the role of kidneys in Cd-induced IR has been raised earlier. The kidney is responsible for up to 20% of all glucose production, and these figures are even higher in diabetic conditions [44]. Moreover, IR represents an early metabolic alteration in chronic kidney disease (CKD) patients, with the skeletal muscle representing the primary site of IR [67]. On the other hand, a recent study in a group of 395 subjects from low- and high-Cd exposure areas demonstrated that glomerular filtration rate could be linked to Cd exposure and tubular toxicity [68]. This linkage was shown to act in both dose and toxicity severity-dependent manners. The association of Cd with CKD was recently highlighted in a review by Satarug [14], which addressed the connection between Cd dietary intake and its e ffects on kidneys. However, animal studies have shown that the Cd e ffect on fasting blood glucose elevation is evident before signs of renal dysfunction are overt [69]. It is, nevertheless, highly plausible that Cd acts synergistically with chronic hyperglycemia seen in diabetic nephropathy. For example, research on 65 participants, consisting of 45 T2D and 20 healthy individuals, revealed the association between higher Cd levels in the poor glycemic control group [55]. Thus, Cd should certainly be considered as the agen<sup>t</sup> of high importance in the progression of diabetes-related kidney disease, and the toxic e ffects of Cd in the kidney certainly further contributes to the role of Cd in IR.

Studies conducted in human subjects have shown conflicting data on the role of Cd in IR development. The obtained results depend on many factors, and the actual prevalence of diabetes in the study population seems to have an important impact on them. Although questionable due to ethical reasons, prospective studies investigating the Cd levels, especially low-level exposure, before the presentation of pathologies/toxicity is warranted to establish the causality bases for this association.

#### **3. Insulin Resistance and Cadmium: Animal Studies**

For the last five decades, studies in animals have suggested that both acute and chronic Cd exposure can a ffect glucose metabolism and synthesis regulation, and alter insulin secretion [70,71]. Intraperitoneal administration of a single dose of Cd-acetate to mice (2.0–6.0 mg/kg body weight (b.w.) and rats (0.84 mg/kg b.w.) was shown to cause a significant increase of blood glucose [70,72]. In addition, feeding animals chow containing increasing amounts of Cd (0–200 ppm CdCl2) for 30 days resulted in a significant dose-dependent elevation of blood glucose levels [73]. The ability of the pancreas to accumulate Cd has been demonstrated in multiple animal studies. Chronic oral administration of Cd for 60 days (100 mg/L) in rats resulted in the accumulation of the metal in the pancreas and a significant decrease in serum insulin levels, followed by a reduction in insulin gene expression [74]. Similar results were obtained after a single oral exposure to a high dose of Cd in rats, where the pancreas accumulated this toxic metal [22]. Studies directed at investigating the impact of environmentally relevant doses of Cd showed that Cd exhibited gender-specificity in glucose metabolism disruption, with females being more sensitive [75]. The same authors demonstrated that low-level maternal exposure to Cd influences glucose homeostasis in o ffspring and increases the risk of o ffspring developing T2D later in life [76].

Although di fferent animal experiments indicate impaired glucose metabolism, insulin secretion, and tissue resistance to insulin, the exact mechanisms of these e ffects of Cd are still hardly known. Under physiological conditions, the maintenance of glucose homeostasis depends on a coordinated process of balancing circulating glucose levels and the release of insulin by the pancreatic β-cells. In the post-absorptive state, 75% of glucose uptake occurs in the insulin-independent tissues, mainly in the liver and brain tissue. In comparison, the remaining glucose uptake (25%) occurs in insulin-dependent tissues, as well as muscle and adipose tissue [77]. To some extent, β-cells can compensate hyperglycemic states by elevating the secretion of insulin at the expense of the likelihood of IR development in multiple peripheral tissues [78].

Based on the studies in animal models of Cd exposure, Cd can produce a direct e ffect on the pancreas and a ffect glucose transport in insulin-independent and insulin-dependent tissues. Numerous animal studies have demonstrated that Cd exposure influences glucose metabolism by directly a ffecting pancreas morphology and β-cell function, resulting in cellular damage. Furthermore, oxidative damage, as a known phenomenon important in diabetes development, may occur upon Cd accumulation [52].

A hyperglycemic state in animals involves the increased activity of the glycogenolysis pathway and stimulation of enzymes associated with the gluconeogenesis pathway [73]. Apart from the Cd effects on gluconeogenesis, Cd can influence insulin via di fferent patterns. Lei et al. [74] demonstrated that subcutaneously administered cadmium (0.5, 1.0, and 2.0 mg/kg b.w.) decreased insulin gene expression in exposed rats. This investigation suggests that Cd can influence the biosynthesis of insulin but has no e ffects on its release. Additionally, the same study revealed pancreatic dysfunction occurring earlier than kidney dysfunction following Cd administration.

Apart from the direct Cd e ffect on pancreatic β-cells, exposure to this metal a ffects glucose transport in other tissues, all potential sites of Cd toxicity. In these tissues, insulin exhibits di fferent e ffects: in skeletal muscle, it promotes glucose utilization and storage by increasing glucose transport and net glycogen synthesis; in the liver, it activates glycogen synthesis, increases lipogenic gene expression, and decreases gluconeogenic gene expression, whereas, in white adipocyte tissue (WAT), it suppresses lipolysis and increases glucose transport and lipogenesis [79,80]. Glucose uptake in di fferent animal cell types is mediated by a family of intrinsic membrane proteins (products of the *GLUT*/*SCL2A* genes) that facilitate glucose transport through membranes. GLUT4, the insulin-responsive glucose transporter, is selectively produced in muscle and adipose cells, while GLUT2 occurs in hepatic cells [81,82].

Generally, in skeletal muscle, Cd-induced IR suppresses glucose utilization and storage by decreasing glucose transport and net glycogen synthesis [79,80]. One-month administration of CdCl2 (50 mg/L) with drinking water in male rats resulted in a significant decrease in plasma insulin-like growth factor 1 (IGF-I) and insulin-like growth factor binding protein-3 (IGFBP-3) levels [83], factors known to

be altered in IR. Insulin resistance in the muscle of rats chronically exposed to Cd has been associated with the reduction of glycogen synthesis. Studies with rodents exposed to Cd have shown decreased GLUT4 expression, which could partly explain the reduction in glycogen synthesis [79].

In the liver, insulin activates glycogen synthesis, increases lipogenic gene expression, and decreases gluconeogenic gene expression. Cadmium-induced IR in the liver leads to a significant increase in hepatic GLUT2, carbohydrate regulatory element-binding protein, glucokinase, and pyruvate kinase mRNA [84]. Zhang et al. [85] have demonstrated that animals exposed to Cd developed IR as the result of the activation of lipogenic proteins, leading to a significant increase in serum glucose and free fatty acids.

Finally, in adipose tissue, especially white adipocytes, insulin suppresses lipolysis and increases glucose transport and lipogenesis [80]. Thus, the adipose IR is the inability of insulin to activate adipose glucose transport, promote lipid uptake, and suppress lipolysis [86]. Several mechanisms were suggested to contribute to the adverse Cd effects on adipose tissue pathophysiology and subsequently increased IR. Han et al. [79] demonstrated that subacute administration of Cd (subcutaneously. 2 mg/kg daily for four days) produces impaired glucose tolerance (IGT) in rats, which was associated with a dose-dependent reduction in GLUT4 protein and GLUT4 mRNA levels in adipocytes. Furthermore, the IR state in adipose tissue favors lipolytic pathways, resulting in an elevation of free fatty acids (FFA), which additionally contributes to impaired insulin secretion when released in the plasma.

#### **4. Insulin Resistance and Cadmium: Cellular Studies**

### *4.1. Non-Pancreatic Cells*

Many in vitro studies have used non-pancreatic cell lines, such as adipocytes or ovarian/granulosa cells, that are typically involved with insulin action or glucose utilization [87]. In adipocytes directly obtained from Wistar rats, exposure to 5 μM Cd increased the cellular metabolism of glucose, similar to zinc, but did not increase glucose uptake, which is the opposite of zinc's action [88]. The primary glucose transporter stimulated by insulin is GLUT4, and an early study suggests that Cd exposure decreases the activity of the GLUT1 transporter [89]. Exposure to Cd also reduces critical cellular mediators involved in the differentiation and normal function of adipocytes. Decreased leptin, adiponectin, and resistin alter cellular ability to normally process lipids, potentially leading to IR [90,91]. Proteins that interact with the motif cytosine–cytosine–adenosine–adenosine–thymidine (CCAAT) are referred to as "CCAAT-enhancer-binding proteins" and are a target of Cd action in adipocytes. There are six CCAAT-enhancer-binding proteins involved in normal adipogenesis, including β and δ, which are activated in early adipocyte differentiation, and α, which is upregulated in the later stages of adipogenesis. Early in adipogenesis, β and δ stimulate peroxisome proliferator–activator receptor-γ (PPARγ). Exposure to Cd has been shown to inhibit the production of both CCAAT-binding enhancer proteins and PPARγ, leading to IR and an increase in adipogenesis [91,92]. In adipocytes, exposure to Cd alters the cellular functions involved with lipid metabolism and IR development, leading to obesity or diabetes.
