Ethnicity-related differences in mitochondrial regulation by insulin stimulation in diabetes

Abstract

Mitochondrial dysfunction has long been implicated in the development of insulin resistance, which is a hallmark of type 2 diabetes. However, recent studies reveal ethnicity-related differences in mitochondrial processes, underscoring the need for nuance in studying mitochondrial dysfunction and insulin sensitivity. Furthermore, the higher prevalence of type 2 diabetes among African Americans and individuals of African descent has brought attention to the role of ethnicity in disease susceptibility. In this review, which covers existing literature, genetic studies, and clinical data, we aim to elucidate the complex relationship between mitochondrial alterations and insulin stimulation by considering how mitochondrial dynamics, contact sites, pathways, and metabolomics may be differentially regulated across ethnicities, through mechanisms such as single nucleotide polymorphisms (SNPs). In addition to achieving a better understanding of insulin stimulation, future studies identifying novel regulators of mitochondrial structure and function could provide valuable insights into ethnicity-dependent insulin signaling and personalized care.

1. Introduction

Type 2 diabetes mellitus (T2DM) has emerged as a significant global public health challenge, with an escalating prevalence worldwide [1]. As of 2021, an estimated 537 million adults were living with diabetes, accounting for around 10% of the global adult population, with this number projected to rise to 783 million by 2045 if current trends continue [2], with a staggering 80% of these cases clustered in low- and middle-income countries [1]. Notably, regions such as China and India have witnessed a sharp rise in T2DM prevalence despite relatively lower obesity rates [3]. This paradoxical trend may be attributed to unique factors among Asians, including a higher proportion of body fat mass, elevated abdominal obesity, and reduced muscle mass, contributing to their heightened susceptibility to T2DM [4].

Although often treatable, diabetes is a significant public health issue due to its various complications, including cardiovascular disease, stroke, kidney disease, retinopathy, and neuropathy [5]. Environmental influences, particularly poor nutrition during gestation and early life combined with excessive calorie intake later on, are pivotal drivers of the T2DM epidemic, especially in populations undergoing rapid nutritional transitions characterized by shifting dietary patterns and decreased physical activity [1]. The economic burden of the disease is also substantial, with global diabetes-related healthcare expenditures estimated to be $966 billion in 2021, which is a 316% increase over the previous 15 years [2,6]. In the United States, healthcare costs are 2.5 times higher for people with diabetes than for people without, accounting for $1 of every $4 spent on healthcare [7]. While T2DM prevalence exhibits a slight male predilection, epidemiological studies have identified various behavioral, lifestyle, and biological risk factors [1]. Increased adiposity, as indicated by elevated body mass index (BMI), stands out as the foremost risk factor for T2DM [8]. Sedentary lifestyle, prolonged sitting, inadequate sleep duration, and smoking further exacerbate T2DM risk [9,10,11,12].

Although genetic predisposition plays a role, the surge in T2DM is predominantly linked to the obesity epidemic [13]. Obesity and sedentary lifestyles contribute to insulin resistance, exacerbating genetic predispositions [14]. Insulin is an endocrine peptide hormone that responds to nutrient availability [15] and is secreted by β-cells in the pancreas as preproinsulin [16]. Insulin resistance typically precedes the onset of type 2 diabetes mellitus (T2DM) by several years [17,18,19,20]. Various abnormalities in insulin signaling and mitochondrial function contribute to insulin resistance in muscle [17,21,22], resulting in impaired insulin receptor phosphorylation leading to reduced GLUT4 translocation to the plasma membrane and decreased glucose uptake [21,22]. In the liver and kidneys, insulin resistance (combined with other factors) leads to increased gluconeogenesis and fasting hyperglycemia [23,24,25,26]. In the vascular endothelium glucose delivery is reduced due to impaired vasodilation [27,28]. Adipose tissue insulin resistance leads to dysregulated adipokines (anti-inflammatory hormones) secretion, contributing to chronic inflammation and systemic metabolic dysfunction, visible in elevated pro-inflammatory cytokines such as C-reactive protein (CRP), interleukin-6 (IL-6), and tumor necrosis factor (TNF) [29,30,31]. Finally, in the brain and pancreatic β-cells dysregulated appetite control, altered glucose sensing, and impaired insulin secretion further complicate metabolic regulation [17,31].

2. Differences in prevalence of type 2 diabetes across ethnicities.

A complex interplay between ethnicity and metabolic health, variations in adipokine profiles, insulin resistance, anthropometric indices and genetic predispositions together influence risk factors for type 2 diabetes (T2D) [32,33]. Adipokine profiles vary markedly among different ethnic groups, indicating distinct physiological responses to adiposity and dietary factors [34]. South Asians exhibit lower adiponectin levels, an adipokine associated with insulin sensitivity and anti-inflammatory properties, compared to Europeans, even when matched for adiposity. Additionally, South Asians display higher leptin levels, which regulate energy balance and appetite, despite similar adiposity levels, potentially indicating differences in leptin sensitivity or adipose tissue function. Aboriginal people also demonstrate higher leptin levels, suggesting differences in adipose tissue metabolism compared to Europeans. In contrast, Chinese individuals exhibit lower adiponectin levels compared to Europeans, implying ethnic-specific variations in adipokine regulation and metabolism [34].

The influence of diet on adipokine levels varies across ethnicities, with higher consumption of high glycemic index foods leading to larger decreases in adiponectin levels among South Asians and Aboriginal people compared to Europeans and Chinese [35]. This differential response suggests varying susceptibility to dietary factors contributing to metabolic disparities among different ethnic groups. South Asians further exhibit a greater increase in insulin resistance with decreasing adiponectin levels, indicating a heightened vulnerability to metabolic dysfunction in response to dietary factors within this population [35]. In a study of 3 major ethnic populations in Singapore Asian Indians exhibit the highest insulin resistance levels, followed by Malays and Chinese [36].

Insulin resistance shows distinct patterns across ethnicities. Lower adiponectin and higher leptin levels independently contribute to increased insulin resistance across all ethnic groups. However, South Asians and Aboriginal people display a significantly greater increase in insulin resistance per unit decrease in adiponectin compared to Chinese and Europeans, highlighting ethnic-specific differences in the pathophysiology of insulin resistance. Anthropometric metrics, such as body mass index (BMI), waist circumference (WC), and waist-to-hip ratio (WHR), also vary among ethnic groups and correlate differentially with metabolic risk [35]. South Asians, for instance, display higher levels of abdominal obesity compared to Europeans and Chinese, particularly among women, even when BMI is similar [35]. Genetic predispositions play a significant role in shaping metabolic health disparities among ethnic groups. Studies in Korean cohorts reveal positive correlations between polygenic risk scores related to oxidative stress and the risk of developing T2D, particularly among middle-aged and elderly individuals with normal body weight profiles [37].

In the United States, African Americans are 60% more likely to be diagnosed with diabetes than White individuals, in all age groups [38,39]. Despite a lower prevalence of metabolic syndrome, African Americans exhibit greater insulin resistance, attributed to impaired metabolic action of insulin. Factors like obesity, particularly abdominal adiposity, play a significant role in T2DM risk, with variations in body fat distribution influencing insulin resistance differently among African Americans and White Americans. African American women in particular display lower insulin sensitivity [40], even among lean African American women, particularly in skeletal muscle, compared to White American women, suggesting genetic predispositions influencing insulin action [40]. The exact mechanisms driving this disparity remain poorly understood, but increased skeletal muscle volume may contribute to greater insulin resistance in African American women [39]. Skeletal muscle fiber distribution, lower mitochondrial content and respiration are likely contributing to this lower insulin sensitivity [40]. Despite these differences, African American women demonstrate intact metabolic flexibility, contrasting with the typical inflexibility seen in insulin-resistant individuals [40]. Furthermore, in African American children with comparable or higher vegetable and fruit intake than their White American peers, early signs of T2DM, such as hyperinsulinemia and insulin resistance, are observed independently of dietary factors [38], even though up to 25% of children diagnosed with T2DM do not exhibit obesity [41]. Although this difference may reflect disparities in socioeconomic status [42], rates of insulinemia, obesity, and family history of diabetes are higher in African American children even after accounting for lifestyle differences [43]. Furthermore, mesenchymal stem cells (MSCs) derived from African American and Caucasian infants show distinctive metabolic profiles present from birth [44]. African American MSCs exhibit greater glucose partitioning toward non-oxidative glucose metabolism (NOGM) and a persistent preference for glycolysis when compared to Caucasian MSCs [44], but no inherent decrements in fatty acid oxidation, which echoes findings in adult African American individuals. Those results are further exacerbated by variations in insulin sensitivity when African, Caucasian, and East Asians are compared [45]. Africans exhibit lower insulin sensitivity, but higher insulin response compared to Caucasians, despite having less visceral fat. Conversely, East Asians demonstrate limited innate capacity for insulin secretion [45].

Despite this higher T2D prevalence, African Americans are underrepresented in clinical trials (>5% participation) relative to their disease burden, which limits the generalizability of research findings and hampers efforts to address the specific healthcare needs of this population [46,47]. Furthermore, as insulin stimulation regulates both mitochondria morphology and function [48,49,50], mitochondria may contribute to the different outcomes of diabetes among different ethnicities [51].

3. Roles of insulin signaling

Insulin orchestrates its physiological effects by binding to the insulin receptor, a transmembrane receptor belonging to the tyrosine kinase receptor family [52]. Upon insulin binding, the receptor undergoes conformational changes, enhancing the kinase activity of the receptor [53]. Subsequently, intracellular substrate proteins known as insulin-responsive substrates (IRS) become tyrosine phosphorylated [54]. IRS proteins initiate downstream signaling cascades, ultimately culminating in the translocation of glucose transporter 4 (GLUT4) to the cell membrane, facilitating glucose uptake [52]. Insulin receptor activates two main pathways: the phosphatidylinositol-3-kinase (PI3K)-protein kinase B (Akt or PKB) pathway and the Ras–Raf mitogen-activated protein kinase (MAPK) pathway [53,54]. While the PI3K pathway predominantly regulates metabolic effects, the MAPK pathway governs gene expression, and cell growth and differentiation through crosstalk with other pathways.

Activation of the insulin receptor triggers phosphorylation of IRS proteins at sites allowing the interaction with PI3K [55] (Figure 1A). Subsequently, PI3K phosphorylates phosphatidylinositol (4,5) bisphosphate (PIP2), generating phosphatidylinositol (3,4,5) trisphosphate (PIP3). Akt, a critical downstream effector of PI3K, translocates from the cytosol to the plasma membrane, where it becomes phosphorylated [56]. Akt activation, facilitated by 3-phosphoinositide-dependent protein kinases-1 and −2 (PDK1, PDK2), modulates cell growth, proliferation, and the cell cycle by targeting substrates such as Gsk3β or FoxOs [57]. Akt-mediated inhibition of Gsk3β promotes glycogen synthesis by activating glycogen synthase [58].

Figure 1. Key mediative factors between insulin and mitochondria.

A. Pathway depicting phosphatidylinositol-3-kinase (PI3K)-protein kinase B (Akt). Split end arrows represent poorly explicated effects of genetic variations on insulin receptor substrate 1. Following insulin binding to receptors, the PI3K/Akt pathway can lead to mitochondrial uncoupling in insulin resistance or mitochondrial biogenesis at baseline. B. Additional studies must focus on how mitochondrial DNA (mtDNA) may be altered by single nucleotide polymorphisms (SNPs). Of relevance, mtDNA damage is correlated with insulin resistance while a higher mtDNA copy number is associated with insulin sensitivity. C. While reactive oxygen species (ROS) has been linked to insulin resistance, it is unclear if it plays a mediating role between mitochondria and ethnicity-dependent differences.

FoxO transcription factors, another target of Akt, regulate diverse physiological functions including cell proliferation, apoptosis, and metabolism [59]. FoxO proteins shuttle between the cytoplasm and nucleus depending on their phosphorylation state [60]. Upon insulin stimulation, Akt translocates to the nucleus, where it phosphorylates FoxO, facilitating its interaction with 14-3-3 protein [61]. This interaction promotes the nuclear export of FoxO, leading to the inhibition of target gene expression and promoting cell survival [61]. The PI3K/Akt pathway also inhibits negative regulators of mammalian target of rapamycin (mTOR), a kinase that regulates cell growth and metabolism in response to various stimuli [62]. mTOR exists in two distinct complexes, mTOR complex 1 (mTORC1) and mTOR complex 2 (mTORC2), each with unique signaling properties [63]. Additionally, mTOR regulates insulin signaling by directly phosphorylating the insulin receptor, thereby modulating its internalization [63,64].

Mitochondria undergo fusion and fission events, essential for maintaining mitochondrial homeostasis in response to various stresses [65,66]. Fusion aids alleviation of stress by merging contents of partially damaged mitochondria through complementation. In highly metabolically active cells such as muscle cells, mitochondria fuse to form elongated networks connecting oxygen-rich peripheral regions with oxygen-poor interiors of muscle fibers. This facilitates membrane potential transmission along mitochondrial filaments, dissipating energy and enabling adenosine triphosphate (ATP) production at diverse cellular sites [66]. Meanwhile, fission facilitates mitochondrial biogenesis, aiding cell survival under oxidative stress and ensuring quality control by eliminating damaged mitochondria through autophagy [66,67]. Those processes involve guanosine triphosphatases (GTPases), which regulate the fusion and division of the lipid bilayers surrounding mitochondria [68]. Mitochondrial fusion primarily relies on three GTPases: Mitofusin 1 and 2 (MFN1 and MFN2), and optic atrophy 1 (OPA1) [69]. While MFN1 and MFN2 mediate outer mitochondrial membrane (OMM) fusion, OPA1 facilitates inner mitochondrial membrane (IMM) fusion [67]. Interactions between these proteins are crucial, with Mfn1 and Mfn2 tethering adjacent mitochondria, while OPA1 maintains mitochondrial cristae morphology [67].

The health of mitochondria and their interaction with insulin can be assessed in experiments involving the Akt-mTOR-NFκB-Opa-1 signaling pathway [49], which demonstrate that insulin stimulation at 0.5, 3, or 6 hours induces mitochondrial fusion both in vitro and in vivo in mice. Insulin-induced activation of the Akt-mTOR-NFκB-Opa-1 pathway increases basal and maximum membrane potential capacities, resulting in elevated oxygen consumption and ATP production [49], targeting mitochondria through the oxidation of pyruvate [70,71]. Prolonged insulin resistance typically causes mitochondrial dysfunction [48], leading to mitochondrial fragmentation in moderate cases and cell death in extreme cases [72]. Individuals with diabetes exhibit increased mitochondrial fragmentation and/or degeneration in skeletal muscle, heart, liver, pancreas, CNS and PNS, when compared to healthy counterparts [73]. Studies on cristae and the MICOS complex remain limited, which is especially pertinent since OPA1 interacts with the MICOS complex to regulate mitochondrial dynamics [74], however, overexpression of Mitofilin, a key component of the MICOS complex, in mice can serve a protective role against diabetes [75].

Recent studies show that mitochondrial energy production and oxidative phosphorylation affect bioenergetic capacity in sarcopenia by altering mTOR, ERRα, and NRF1 signaling pathways across ethnicities [75], suggesting that age-related loss of mitochondrial metabolism in skeletal muscle is common across ethnicities. While studies examining tissue-specific activation of the mTOR pathway by insulin signaling in people of different ethnicities are currently lacking [76,77], but some studies show that PI3K/mTOR inhibitors have stronger effects in Asians when compared with Caucasians [78]. Thus, although mTOR pathway induction of insulin signaling to mitochondria may be independent of ethnicity in skeletal muscle, further investigation of this pathway in other tissue types and across a wider range of ethnicities is warranted.

4. Genetics of type 2 diabetes mellitus.

While lifestyle factors such as diet and exercise play crucial roles in T2D development, genetics also significantly contribute to an individual’s susceptibility to the disease. Over the years, extensive research has uncovered numerous genetic variants and mutations associated with T2D, shedding light on its complex etiology but it was the advent of Genome-wide association studies (GWAS), made possible by the Human Genome and International HapMap Projects, which revolutionized the field. GWAS enabled the discovery of numerous gene variants with small individual effects on diseases, primarily through single nucleotide polymorphisms (SNPs). SNPs are single nucleotide variations occurring roughly every 300 base pairs in the genome. The HapMap project initially mapped 3.9 million SNPs across different ethnic groups, elucidating patterns of genetic inheritance, quantified through linkage disequilibrium (LD). LD measures the likelihood of nearby SNPs co-inheriting, influenced by recombination rates. This correlation structure enables use of a subset of SNPs, called “tag” SNPs, in genotyping arrays, capturing the majority of common genomic variation, followed by imputation, based on LD to infer non genotyped variants from the genotyped subset [79].

One of the earliest identified genetic factors in T2D is the P12A polymorphism in the peroxisome proliferator-activated receptor gamma (PPAR-gamma) gene. Altshuler et al. [80] confirmed its association with T2D, revealing that the more common proline allele was linked to an increased risk of diabetes. This finding has significant implications, considering the relatively high frequency of the proline allele in the general population, contributing to a sizable portion of the overall T2D burden. Up to this day PPARG SNP remains the most significant T2D risk factor in the general population. Beyond single-gene mutations, severe insulin resistance, a hallmark of T2D, can arise from complex genetic interactions. Savage et al. [81] described a family with severe insulin resistance, characterized by acanthosis nigricans, elevated fasting plasma insulin levels, and hypertension. Mutation screening in affected family members identified heterozygous frameshift mutations in the PPARG and protein phosphatase 1 regulatory subunit 3A (PPP1R3A) genes, indicating double heterozygosity at separate loci. Further unraveling the genetic architecture of T2D, associations with genes involved in insulin signaling pathways have been elucidated. Mammarella et al. [82] found a strong association between the insulin receptor substrate-2 (IRS2) G1057D polymorphism and T2D risk in Italian patients, suggesting a protective effect against the disease. Additionally, genes involved in adipokine signaling, such as adiponectin (ADIPOQ), have been implicated in T2D pathogenesis. Variation in the ADIPOQ gene on chromosome 3q27 has been associated with T2D risk, underscoring the role of adipose tissue function in insulin sensitivity and glucose homeostasis.

Mitochondrial DNA (mtDNA) variants have also emerged as potential contributors to T2D susceptibility. Poulton et al. [83,84] observed a positive correlation between the common mtDNA variant T16189C and blood fasting insulin levels, suggesting a role in insulin metabolism. Subsequent case-control studies revealed a significant association between the T16189C variant and T2D, particularly in individuals with a family history of diabetes from the father’s side. While mitochondrial variants may influence metabolic parameters, their direct contribution to T2D susceptibility appears to be modest, as evidenced by studies examining their association with birth weight and diabetes status. Expanding the scope of mitochondrial genetics, Fuku et al. [85] conducted a comprehensive association study involving over 2,900 Japanese and Korean individuals to identify mitochondrial haplogroups associated with T2D susceptibility. They found that the mitochondrial group N9a was significantly associated with resistance against T2D, suggesting a potential protective role of certain mitochondrial haplotypes against the development of diabetes.

In addition to mitochondrial genetics, mutations in specific genes have been implicated in T2D susceptibility. For instance, Shimajiri et al. [86] identified the connection of R121W mutation in the PAX4 gene among Japanese individuals with T2D. Clinical observations revealed that mutation carriers often had a family history of diabetes or impaired glucose tolerance, with some progressing to an insulin-dependent state without autoimmune involvement, suggesting a unique disease trajectory. Similarly, Maeda et al. [87] conducted a gene-based association study implicating variations within the TFAP2B gene in T2D risk in Japanese and U.K. populations. Moreover, mutations in genes like ABCC8 and WFS1 have been linked to T2D pathogenesis, highlighting the genetic complexity underlying the disease. Babenko et al. [88] identified heterozygous missense mutations in the ABCC8 gene in patients with neonatal diabetes, suggesting a potential link to monogenic forms of T2D. Sandhu et al. [89] revealed strongly associated SNPs within the WFS1 gene, despite their intronic location and lack of apparent biological function, underscoring the intricate genetic mechanisms at play in T2D susceptibility.

Inflammatory pathways have also been implicated in T2D risk, with genetic variants in genes like IL6 influencing disease susceptibility. Mohlig et al. [90] found that the IL6 C174G SNP altered the correlation between BMI and IL6 levels, with certain genotypes exhibiting a notably stronger increase in IL6 levels with higher BMI, predisposing obese individuals to a higher risk of T2D. Conversely, Illig et al. [91] found no significant associations between IL6 SNPs and T2D risk but observed elevated levels of monocyte chemoattractant protein-1 (MCP1) in carriers of protective genotypes, suggesting an indirect effect on the innate immune system.

Interestingly, the link between the immune system and T2D progression showed ethnicity-based differences in novel SNPs, connected to immune cell function, identified by Shoily et al [92] in TNFA G308A allele, leading to excessive TNFA production, IL6 G174C allele associated with higher IL6 levels and eNOS G894T allele reduces basal NO levels, associated with hypertension, dyslipidemia, and endothelial dysfunction are more prevalent in European populations but significantly reduced in frequency in African populations. These variants affect the AGE-RAGE pathway, oxidative stress, inflammation and insulin resistance [92].

Furthermore, genetic variants in genes like KCNJ15 and MTNR1B have been linked to T2D risk. Okamoto et al. [93] identified a significant SNP within KCNJ15, particularly affecting lean individuals, while Bonnefond et al. [94] uncovered rare variants in MTNR1B associated with the disease, shedding light on potential molecular mechanisms underlying T2D predisposition.

Recent advancements in genetic research have led to the discovery of novel genetic loci associated with T2D susceptibility. The SIGMA Type 2 Diabetes Consortium [95] identified a novel locus spanning SLC16A11 and SLC16A13 in Latin American populations, with genetic variants likely inherited from Neanderthals potentially influencing triacylglycerol metabolism. This highlights the importance of considering genetic ancestry and evolutionary history in understanding T2D susceptibility across diverse populations.

Finally, the role of specific genes like STARD10 in T2D pathogenesis has been investigated. Carrat et al. [96] revealed decreased STARD10 mRNA levels in carriers of T2D risk alleles, suggesting its involvement in diabetes development. Functional studies in beta cell-selective knockout of Stard10 in mice further supported the role of STARD10 in glucose-stimulated insulin secretion, emphasizing its potential as a therapeutic target for T2D.

5. Effects of insulin on mitochondrial quality control

Mitochondrial pathways can influence insulin signaling and inversely insulin also impacts mitochondrial function [71,97,98]. A biomarker preceding insulin resistance is mitochondrial dysfunction, which is marked by reduced oxidative capacity and increased oxidative stress [99]. The canonical pathway through which this occurs is the mTOR pathway, which stimulates mitochondrial uncoupling and promotes insulin resistance [49,71]. However, both alterations in insulin-like-growth-factor 1 and insulin receptors can modulate mitochondrial function [71]. Of relevance, genetic variants in the insulin receptor substrate 1, which predict insulin resistance, differ based on ethnicity [100]. Similarly, levels of IGFs display race differences independent of other factors [101,102]. This suggests that ethnicity-related differences in insulin pathways lead to changes in mitochondrial functions, although several aspects of mitochondrial functional dynamics remain poorly understood.

An essential regulator of mitochondrial function is mitochondrial DNA (mtDNA), with nucleoid organization and mutation both affecting oxidative phosphorylation and metabolic pathways [103,104] (Figure 1B). mtDNA often preserves differences in ethnicity-related ancestry better than nuclear DNA [105]. mtDNA damage, as detected using plasma mtDNA, serves as a biomarker of insulin resistance [106]. Similarly, mtDNA copy number is positively associated with insulin sensitivity [107]. However, it is unclear whether heterogeneity in mtDNA depending on ethnic ancestry affects these pathways. Furthermore, considering the heritable aspects of insulin resistance [108], it is possible that resistance is conferred through generations partially due to highly preserved mtDNA.

Insulin resistance caused by a high-fat diet increases reactive oxygen species (ROS) emission and reduces antioxidant enzyme activity in the brain and skeletal tissue, potentially leading to cognitive decline [109] (Figure 1C). Black individuals show increased ROS in the myometrium and leiomyoma [110]. Additional studies could investigate whether build-up of ROS differs across ethnicities, which would shed light on whether insulin resistance causes these differences. ROS may also affect hyperglycemia, as elevated oxidative stress occurs concomitantly with hyperglycemia [111]. On the other hand, hyperglycemia increases insulin resistance and reduces the metabolic clearance rate [111]. The insulin clearance rate, which differs among ethnicities, can often be explained by insulin sensitivity and secretion [112]. However, it remains unclear whether people of different ethnicities show alterations in pathways that affect the insulin clearance rate depending on oxidative stress and insulin sensitivity.

6. Effects of insulin on mitochondrial morphology

Studies in cardiomyocytes show that insulin stimulation increases OPA1 expression and promotes mitochondrial fusion [49,113], suggesting direct effect on mitochondrial morphology. Loss of MFN2 expression in the murine liver impairs insulin signaling by modulating ER stress [114]. Conversely, MFN1 deficiency may protect against insulin resistance in the murine liver [115]. Loss of both murine MFN1 and MFN2 decreases plasma insulin concomitantly with reduced mitochondrial length, suggesting synergistic effects despite their contrary roles when knocked out individually [116]. MFN1 and MFN2 clustering occurs at mitochondrial-peroxisome contact sites [117], which may mediate insulin resistance by impacting lipid toxicity [118,119].

Potential ethnicity-related variations in these fusion proteins, however, remain poorly understood. MFN2 mutations occur at different rates among different Asian populations [120]. Also, different nucleotide substitutions associated with neuropathy are reported among different Eastern European populations [121]. MFN2 has unique exon-determined isoforms that can have differential abundances based on genotype [122]. As a potential mechanism underlying ethnicity-related differences in MFN2-dependent insulin stimulation, increased expression of microRNA (miR)-106b downregulates MFN2 and causes insulin resistance in C2C12 myotubes, whereas its inhibition improves MFN2-dependent insulin sensitivity [123]. The expression of some miRNAs differs across ethnicities [124]. For example, the prognostic role of miR-106b in colorectal cancer is greater in Asian populations than in European populations [125]. It is unclear whether there are ethnicity-related differences in OPA1 regulation that could result in different mitochondrial substructure. Notably, however, one study suggests the differential import of OPA1 polymorphisms between Caucasians and Asians [126]. Specifically, Guo et al. found that while OPA1 polymorphisms were significantly associated with risk for normal tension glaucoma in Caucasians, this was not the case for the Asian cohort [126]. Similarly, another study conducted by Titilayo et al. found that different ethnic groups of Nigerians display heterogenic, although not significant, OPA1 polymorphic band sizes [127]. Therefore, future research should prioritize investigating the impact of insulin stimulation on OPA1 and other regulators of mitochondrial dynamics in diverse ethnic populations.

In addition to regulating fusion, insulin also influences mitochondrial fission by promoting phosphorylation of DRP1, a key fission protein, thereby facilitating the maintenance of a balanced mitochondrial network [128,129]. Loss of DRP1 in pancreatic β-cells causes highly diverse mitochondrial morphology and impairs insulin secretion [130]. Notably, the expression of DRP1 in breast cancer differs across ethnicities [131]. MFF, MID49, and MID51 recruit DRP1 to the outer membrane, with the latter acting as DRP1 receptors, whereas MFF and FIS1 independently interact with DRP1 to determine the post-fission mitochondrial fate [132–134]. However, the biological significance of various forms of DRP1-mediated fission is still poorly understood. As a result, there is a gap in the literature regarding whether ethnicity influences the type of mitochondrial fission regulated by DRP1 and thereby explains how ancestry influences insulin-dependent mitochondria effects.

Collectively, previous findings indicate that mitochondrial fission acts as a pro-insulin stimulation factor [135], whereas insulin stimulation promotes mitochondrial fusion [49]. However, most studies only examined mitochondrial elongation or fragmentation, whereas a more detailed investigation of mitochondrial 3D structure may offer more information about the functional capacity of mitochondria [136]. Of studies utilizing 3D, it was shown that MFN2 silencing alters mitochondrial 3D morphology and impairs insulin response [137]. Another study using 4Pi microscopy found that mitochondria in β-cells form a highly networked reticulum, which breaks down in Goto Kakizaki rats with T2D in ways otherwise not observable by 2D techniques [138]. Notably, different ethnicities may possess unique metabolites that could potentially impact the 3D shape and function of mitochondria [136]: such as mimicking a recently reported loss of DRP1 in pancreatic β-cells, causing clustered and looped mitochondrial morphology [130,139].

7. Role of MERCs in insulin resistance

MERCs are regions of contact between mitochondria and ER located within a range of 10–30 nm [140]. MERCs are responsible for many cellular processes, including mitochondrial quality control, lipid metabolism, calcium homeostasis, and ER stress [141,142]. MERCs are associated with ER stress and the unfolded protein response [143–145]. The two main pathways involving MERCs include PERK and IRE1α [141]. PERK interacts with MFN2, which helps eliminate weakened MERCs and protect mitochondria against ROS damage [145], therefore linking MERCs to insulin regulation [145–148].

In mouse and human primary hepatocytes, loss of the biochemical fraction of MERCs, known as mitochondria-associated membranes (MAMs), is associated with alterations in insulin signaling [149,150]. By contrast, an increase in MERCs improves the action of insulin [149,150]. However, obesity causes MAM accumulation in a pyruvate dehydrogenase lipoamide kinase isozyme 4 (PDK4)-dependent manner, resulting in impaired insulin signaling [151], and a deficiency in PDK4 prevents MERC formation and associated calcium overload [152]. Furthermore, PDK4 also affects mitochondrial fission, as increased PDK4 can induce muscle atrophy, potentially in a MERC-dependent manner [153]. MERCs are also influenced by the mTOR-Akt pathway, with enhanced formation following insulin stimulation, suggesting OPA1 involvement [154]. However, MERCs variations by ethnicity remain understudied. For example, mutations in TMEM63C, which is involved in regulating MERCs and a genetic risk factor in hereditary spastic paraplegias [155], have only been identified in families from Oman, Iran and North Africa [155]. Whereas in animal models parental exercise induces epigenetic modulation of PDK4 [156], and PDK4 is elevated in humans with T2D, during fasting and in humans and animals on a high fat diet [157], a study in Korean population of 5 known PDK4 SNPs showed no significant association between PDK4 polymorphism and T2D or metabolic syndrome [158]. An outstanding question is whether insulin stimulation affects MERC morphology in an MFN-dependent manner, as it involves MFN2 and OPA1 [49], or if there are alternative mechanisms at play. This is particularly relevant as recent findings indicate the existence of different types of fission [159], and OPA1 is known to cleave into distinct isoforms with varying roles in insulin stimulation [113,159]. Therefore, certain ethnicities could exhibit a greater prevalence of specific isoforms of OPA1, or insulin stimulation could favor the selection of particular OPA1 isoforms. The involvement of these isoforms in the modulation of different MERC structures is similarly unclear. Beyond these standard mitochondrial fusion proteins, other proteins involved in MERC formation remain understudied. For example, MERCs can form through the binding of vesicle-associated membrane protein-associated protein B (VAPB), an ER protein, to tyrosine phosphatase-interacting protein 51 (PTPIP51), a mitochondrial outer membrane protein [160]. Notably, MERCs formed in these VAPB-PTPIP51 interactions have inhibitory roles in autophagy [160]. Thus, various regulators of MERCs could give rise to MERCs with distinct primary purposes and regulatory roles. Currently, however, there are few studies investigating ethnicity-dependent regulation of mitochondrial protein-mediated formation of contact sites. Specifically, it is unclear whether there is ethnic variation in forms of MERCs that have different roles in disease states influenced by insulin resistance or specific types of diabetes. Moreover, the potential existence of alternative non-canonical pathways through which MERCs form in different ethnicities remains unexplored.

Another outstanding question is how calcium regulation, which is MERC-dependent [140,141], may be affected by insulin stimulation. Recent studies highlight that alterations in calcium signaling may result in increased longevity in people of Han Chinese ancestry relative to people of European ancestry, although this could also be a result of dietary differences [161]. Another study noted high ethnicity-related variability in pathways involved in T2D, whereas calcium signaling pathways were conserved across ethnicities [51]. However, given a lack of studies exploring calcium signaling across ethnicities, there remains a gap in the literature. One potential avenue of exploration is whether MERCs are impacted by ethnicity-related differences in calcium signaling. For instance, changes in reticulum-mitochondria calcium exchange and miscoupling in cardiomyocytes are observed in T2D, and targeting dysfunction in MAM-dependent calcium transfer may prevent alterations in insulin resistance [162]. Similarly, hepatic insulin resistance can be caused by a lack of calcium exchange between ER and mitochondria [163]. Thus, MERCs in calcium pathways can respond to insulin stimulation, alter insulin signaling, and contribute to insulin resistance. Therefore, ethnicity or associated epigenetic factors that impact calcium signaling [161,164,165] may manifest as alterations in MERCs and associated insulin resistance.

8. Exploration of novel mitochondrial dynamic players

A promising direction of future research involves investigating novel regulators of mitochondrial function that have less defined roles in insulin stimulation and may vary by ethnicity. For example, solute carrier family 25 member 46 (SLC25A46) is a recently discovered outer mitochondrial membrane protein implicated in mitochondrial dynamics [166]. As a prominent effector in neurological diseases, SLC24A46 interacts with several other mitochondrial proteins, including MFN2, OPA1, and members of the MICOS complex [167]. Although its precise role is still unclear, SLC25A46 contributes to the regulation of mitochondrial fusion and promotes the fission process [168]. In insulin-stimulating cells, SLC25A46 expression does not affect overall insulin secretion but rather affects glucose-stimulated insulin secretion and sensitivity to lipotoxicity [168]. Recent murine studies further suggest that SLC25A46 is involved in the regulation of insulin signaling [169]. Whereas some genetic association studies indicate that SLC25A46 expression or mutations in the SLC25A46 gene are ubiquitous across ethnicities [170,171], few direct comparisons across non-European ethnicities have been performed. Thus, investigating the interplay among SLC25A46, insulin signaling, and mitochondrial morphology could provide valuable insights into the complex regulation of mitochondrial dynamics across ethnicities.

Regulators that affect mitochondrial fission are similarly underexplored. Fission 1 (FIS1), a small mitochondrial outer membrane protein, facilitates mitochondrial fission by interacting with DRP1 [172]. Notably, inhibition of FIS1 in cybrid cells has analogous effects to inhibition of DRP1, with both reducing insulin resistance by modulating IRS1-Akt pathways [172]. These effects were replicated in C2C12 cells, showing that FIS1 overexpression is a driver of mitochondrial fission alongside DRP1 in impairing insulin sensitivity [127]. Similar to Fis1, an alternative docking site for DRP1 is mitochondrial fission factor (Mff), another outer mitochondrial membrane protein that participates in mitochondrial fission and may contribute to DRP1-dependent alterations in insulin secretion [129]. Notably, Mff regulates mitochondrial fission and insulin resistance by interacting with sphingolipid C16:0 ceramide produced by ceramide synthase 6 (CerS6) [173]. Additionally, mitochondrial dynamics proteins of 49 and 51 kDa (MiD49 and MiD51) are mitochondrial proteins that regulate mitochondrial fission [174]. Thus, as fission is implicated in the process of insulin stimulation, these proteins may be valuable targets. Although its roles in insulin stimulation remain unclear, lipid infusion reduces insulin sensitivity alongside increased expression of DRP1 and MiD49 independently of changes in fusion proteins, leading to increased fission and decreased mitochondrial membrane potential [175]. In addition, expression of MiD49 and DRP1 is increased in T2D patients [176]. Notably, MiD49 and MiD51 are involved in the epigenetic modification of mitochondrial dynamics [177] and are epigenetically modified by miRNA-34a-3p in cancer [117]. Recently a novel, inherited, unique mutation in a MIEF2 gene (encoding MID49) was reported in an individual of Jewish ancestry [174]. Although research on miRNA-34a-3p remains limited, one study shows that there are differences in blood plasma levels of miRNA-34a-3p between people from Beijing, China, and Sheffield, UK [177]. Therefore, these associated regulators of DRP1 deserve more attention, as they may differentially impact mitochondrial structure and, consequently, insulin resistance across ethnicities.

In addition to the roles of MFN2 in insulin stimulation, mutations affecting MFN2 and SLC25A46 are also associated with Charcot-Marie-Tooth Type 2 neuropathy [179]. In this disease, imbalanced mitochondrial fusion events caused by increased MFN2 expression result in abnormal cristae morphology [179]. Another factor involved in this hereditary peripheral neuropathy is ganglioside-induced differentiation-associated protein 1 (GDAP1), of which mutations are more likely to affect mitochondrial mutations through alteration of the fission process, which is linked to poorer outcomes [178]. As an outer membrane protein, GDAP1 is mainly defined by its roles in mitochondrial dynamics and calcium signaling [180], although it is also involved in the modulation of lysosome-mitochondria contact sites, which participate in the regulation of autophagy [181,182]. Beyond this, GDAP1 mutations have different effects on mitochondria depending on their mode of inheritance [183]. Although studies of MFN2 remain limited, one of the most common mutations in North African individuals impacts GDAP1 [184]. Thus, ethnicity-dependent mitochondrial dysfunction may be due to the genetic inheritance of mutated mitochondria, and GDAP1 may serve as a genetic factor influencing insulin stimulation, although additional research is needed. In addition, AMP-activated protein kinase (AMPK) targets GDAP1, which impairs GDAP1 function and increases insulin AMPK activity [185]. Insulin diminishes AMPK activity through an AKT pathway in a murine model [186], suggesting that insulin increases GDAP1 levels. Although the link between insulin and autophagy remains poorly elucidated [187], future research on relationships between GDAP1 and insulin would provide insight into the pathways through which insulin alters autophagy and mitochondrial function.

Another future avenue for research is understanding alterations in optic atrophy in genetically heritable diseases. A recent case report shows that a genetically inherited mutation in SIRT3 causes optic neuropathy due to OPA1 targeting [188]. However, little is known about the roles of other optic atrophy proteins. For example, optic atrophy 3 (OPA3) localizes to the inner mitochondrial membrane and is implicated in mitochondrial dynamics, as mutations in OPA3 associated with hereditary optic neuropathies affect mitochondrial morphology [189]. OPA3 and OPA1 may have converse roles, with loss of OPA3 causing greater mitochondrial elongation [189]. Given that OPA1 is affected by insulin stimulation and optic atrophy proteins can be genetically inherited, these proteins may affect ancestry-dependent congenital alterations in insulin stimulation.

Furthermore, future research could focus on the mitochondrial contact site and cristae organizing system (MICOS) complex, which resides within the inner mitochondrial membrane and cristae [190,191]. MICOS proteins in the cristae enhance mitochondrial function and contribute to the structural remodeling of mitochondria [191,192]. Similarly, whereas insulin stimulation may control OPA1-mediated fusion, it may also regulate the MICOS complex and cristae tightening. Yet, the potential genetic associations of the MICOS complex remain unclear. A case report shows that two siblings with a neurodegenerative disorder had similar mutations in MICOS complex subunits that led to mitochondrial dysfunction [193]. In addition, disrupted-in-schizophrenia-1 (DISC1), which resides in the MICOS complex [194], was found to have genetic variants with roles in hereditary psychosis [195]. Together, these findings suggest that the MICOS complex is genetically regulated, but it is unclear whether MICOS complex expression differs across ethnicities and whether it affects insulin stimulation.

9. Conclusion and future perspectives

Ethnicity-related differences arising from genetic, environmental, and lifestyle factors are increasingly understood to influence mitochondrial function and glucose metabolism. However, it remains unclear which genetic factors may contribute to ethnicity-related differences in insulin-dependent mitochondrial dysfunction. Indeed, in seeking to study genetic factors, historical oppression and exploitation of minority groups certainly must remain in ethical considerations. Furthermore, the equation of ethnicity to socially constructed factors including race can portend research being done without prioritization of social equity. Yet, studying genetic differences among ethnic groups offers the advancement of the still nascent field of personalized medicine targeting mitochondrial. Thus, the broader impact of explicating the relationship between mitochondria, insulin signaling, and ethnicity, especially as studied in an ethical manner with a consilient approach, offers the opportunity for tremendous improvements in understanding of ethnicity-dependent regulators and, ultimately, quality of care.

Genetic variations in mitochondrial function, insulin signaling, glucose metabolism, and oxidative stress responses may result in differential risks among ethnicities and warrant further study. A shift toward considering how ethnicity alone affects whole-body insulin signaling and specific insulin signaling pathways. Although mitochondria should remain a key focus, these investigations must go beyond simple consideration of dysfunction [196] and include studies on mitochondrial quality control mechanisms, as those may provide deeper insight into the interface between insulin and ethnicity.

Studies of differential insulin signaling in mitochondria across ethnicities are still limited. In future studies, it is necessary to consider other contact sites beyond MERCs. For instance, insulin impacts lipid droplets, which can also create contact sites with mitochondria and potentially influence mitochondrial quality in conditions with altered insulin signaling [197]. As the MICOS complex is implicated in the functional capacity of mitochondria, future studies could investigate whether insulin modulates mitochondrial structure and function by altering the MICOS complex. Similarly, whereas OPA1 and other common fusion proteins have been the focus of study due to their established role in insulin stimulation, other mitochondrial proteins (e.g., SLC25A46) might also be differentially involved in insulin-dependent mitochondria function across ethnicities. Therefore, more knowledge on mitochondrial dynamics and regulators, as well as the influence of these factors on 3D mitochondrial morphology [136], is crucial for understanding potential differences among ethnicities that may impact mitochondrial insulin stimulation.

Whereas variations in SNPs and truncated proteins are observed among different ethnicities [198], the degree to which these factors contribute to differences in insulin signaling observed between ethnicities remains unclear. Although genome-wide association studies show that many genes involved in T2D pathways are highly conserved across ethnicities [199], their heritability and impact on diabetes risk within an ethnically diverse population is poorly understood. Recent research shows that certain SNPs increase the risk of gestational diabetes mellitus in women by altering insulin sensitivity [200]. Similarly, mutations resulting in truncated proteins lead to insulin-resistant diabetes [201]. Genome-wide association studies could improve future clinical practice by uncovering genetic factors contributing to T2D [202]. As mtDNA has been recognized as an underutilized alternative that better retains ancestral factors [105], future studies should interrogate whether genetic factors may be better viewed through mtDNA. This leads to the important question of whether mitochondria are inherited in different ways in different ethnicities. As mitochondria can be genetically regulated, a promising future avenue of research could investigate whether mtDNA is an overlying genetic factor determining insulin regulation [203].

Data Availability Statement

Data sharing not applicable to this article as no datasets were generated or analysed during the current study.