FBS patients develop different patterns of dysglycemia, ranging from fasting hypoglycemia, postprandial hyperglycemia, glucose intolerance to diabetes mellitus [3]. Compound heterozygous or homozygous mutations in SLC2A2 are the only identified gene mutations implicated with FBS. However, the pattern of dysglycemia does not correlate with the mutation type [3]. The mechanisms involved in the development of dysglycemia in FBS patients is not well known. We hypothesized that classic disturbances in GLUT2 structure and/or function is associated with dysglycemia in FBS.
Glucose uptake activity in peripheral blood mononuclear cells (PBMCs) is vital and plays a key role in inflammation and immune response [23–25]. GLUT2 proteins were not expected to be found in blood cells, and they were not present in the plasma using mass spectrometry (https://www.proteinatlas.org/ENSG00000163581-SLC2A2/blood). Fu et al. reported a difference in the expression and immune action of GLUT1, 3 and 5 in resting and activated human macrophages, monocytes, and lymphocytes [26]. In addition, Palmer et al. reported that the expression of GLUT1 is significantly increased in proinflammatory monocytes from HIV + persons in comparison to HIV- controls [27]. A recent study showed a significant increase in the expression of GLUT4 in the PBMCs extracted from athletes compared to sedentary participants [28]. Moreover, Haas et al. proved using a mouse model an increase in the expression level of GLUTs (1,2,3 and 4) in CD4 + cells activated with CXCL100 [29].
We were interested in examining the expression of GLUT2 in human PBMCs to study the impact of SLC2A2 mutations on glucose uptake activity. We successfully detected for the first time the expression of GLUT2 in human PBMCs using qRT-PCR. Then, we studied the expression of GLUT2 in different cell types of human PBMCs (T lymphocytes (CD4 + and CD8+), B Lymphocytes (CD19+), and monocytes (CD14+)). We found a very low expression of GLUT2 in different cell types of PBMCs extracted from the healthy control (no dysglycemia) (Fig. 5). However, the expression of GLUT2 was upregulated in all activated PBMCs cell types of a healthy control who recently received the COVID19 vaccine.
Based on the results that we have generated from patient’s PBMCs, we suggest that the exonic mutation in GLUT2 affects its glucose transport activity and that this explains the dysglycemia observed in this FBS patient. However, it is still unclear if and how the intronic mutation in GLUT2 impacts on glucose metabolism. In an effort to address this question, we used the Nanostring miRNA panel v3b to investigate if the intronic mutation affected the expression of miRNAs correlated with dysglycemia. The molecular analyses showed that the miRNA expression profiles of the patient were more similar to that of the healthy control than to that of the mother (Supplementary Fig. 5). We identified 123 miRNAs that were expressed specifically in the patient, and the subsequent IPA analysis identified 118 mapped miRNAs (Fig. 10). The highest 30 counts of difference in the expression in the patient in comparison to the control were presented, and 14 miRNAs (miR-199a, miR-25-3p, miR-93-5p, miR-19b-3p, miR-107, miR-24-3p. miR-18a-5p, miR-125b-5p, miR-324-5p, miR-331-3p, miR-144-3p, let-7e-5p, hsa-miR-29a-3p, and hsa-miR-143-3p) were correlated with type 1DM (Fig. 11).
Interestingly, we found the expression of hsa-miR-29a-3p was significantly increased in the patient in comparison to the healthy control. Aghaei et al. reviewed the miR-29 family and its association with insulin secretion and identified three separate mechanisms; either by direct targeting of pancreatic p85α or Stx-1a to influence insulin signaling or fusion of insulin granule with the membrane, respectively, or by targeting hepatic p85α to activate gluconeogenesis [30].
Furthermore, one study demonstrated that miR-29a inhibits glucose-stimulated insulin secretion (GSIS) and cell proliferation in MIN6 cells via a negative effect on Cdc42/β-Catenin signaling [31]. It has also been suggested that miR-29a inhibits GSIS by targeting syntaxin-1 and Mct1, as well as insulin signaling by targeting INSIG1, CAV2, PIK3R1 [32, 33]. Furthermore, Zhou et al. reported that miR-29a is also implicated in insulin resistance by decreasing ATP production, GLUT4 expression, and glucose uptake through targeting PPARδ [34]. Hromadnikova et al. suggested that hsa-miR-29a-3p could play a role in heart disease and diabetes mellitus [35]. In addition, a recent study showed that miR-29 is involved in inflammation and diabetes mellitus through the down regulation of TRAF3 [36]. Interestingly, we found that the expression of CAV2, SLC16A1, PIK3R1, and SLC2A4 were downregulated in the patient carrying the intronic GLUT2 mutation in comparison to control (Fig. 12). TargetScan (prediction of miRNAs target website) predicted an interaction between miR-29a-3b and position 1341–1347 in the human 3'UTR of SLC2A2 (ENST00000314251.3). However, miRecords did not predict any interaction.
In addition, two other miRNAs, hsa-miR-144-3p and hsa-let-7e-5p, were also significantly increased in the patient in comparison to the healthy control. Shen et al. showed hsa-miR-144-3p is associated with adipogenesis by promoting C/EBPα activity [37]. Demirsoy et al. reported that the expression of hsa-let-7e-5p was significantly downregulated in patients with T2D after receiving metformin therapy [38]. TargetScan proposed an interaction between has-let-7e-5p and position 878–884 in the 3’UTR of SLC2A2. However, miRecords did not predict any interaction.
Moreover, we found that another set of 11 miRNAs (miR-199a, miR-25-3p, miR-93-5p, miR-19b-3p, miR-107, miR-24-3p. miR-18a-5p, miR-125b-5p, miR-324-5p, miR-331-3p miR-199a, miR-25-3p, miR-93-5p, miR-19b-3p, miR-107, miR-24-3p. miR-18a-5p, miR-125b-5p, miR-324-5p, miR-331-3p, and hsa-miR-143-3p) were overexpressed in the patient in comparison to the control. Jordan et al. reported that miR-143 impairs the capability of insulin to stimulate AKT activation and glucose homeostasis by downregulation of oxysterol-binding-protein-related protein 8 (ORP8) [39]. Guo et al. reported that the expression of miR-324-5p was elevated in patients with hyperlipidemia and hyperglycemia due to suppression of ROCK1 [40]. Yu et al. reported that overexpression of miR-125b-5p improves the function of pancreatic β-cell through suppression of DACT1 [41]. Tavano et al. reported reported that miR-18a-5p was overexpressed in both pancreatic cancer and non-pancreatic cancer with a recent onset of diabetes in comparison to healthy controls [42]. Xu et al. reported that overexpression of miR-125a-5p enhanced hepatic glucose and lipid metabolism (decreasing lipid and glucose levels and increasing glycogen storage) in type 2 diabetes through inhibition of STAT3 expression [43]. Overall, our miRNA analysis clearly demonstrates that the patient carrying the intronic GLUT2 mutation display a different miRNA profile compared to healthy controls. Interestingly, many of the overexpressed miRNAs have been linked to cardiometabolic disease. However, if and how the mutation in GLUT2 is linked to the dysregulated expression of these miRNAs remains to be explored.
The limitations of this study is that we only studied 2 mutations and thus cannot rule out other possible mechanisms of dysglycemia in FBS patients and further functional analysis are required to prove causality of increase in expression of miR-29a-3p in the patient with intronic SLC2A2 mutation.
In conclusion, our study confirms that homozygous SLC2A2 mutations are involved in the development of dysglycemia in FBS either by a direct effect on GLUT2 expression and/or activity or by an indirect effect on other molecules involved in glucose homeostasis.