Sex-biased islet β cell dysfunction is caused by the MODY MAFA S64F variant by inducing premature aging and senescence in males

Abstract

A heterozygous missense mutation of the islet β cell-enriched MAFA transcription factor (p.Ser64Phe [S64F]) is found in patients with adult-onset β cell dysfunction (diabetes or insulinomatosis), with men more prone to diabetes than women. This mutation engenders increased stability to the unstable MAFA protein. Here, we develop a S64F MafA mouse model to determine how β cell function is affected and find sex-dependent phenotypes. Heterozygous mutant males (MafA^S64F/+) display impaired glucose tolerance, while females are slightly hypoglycemic with improved blood glucose clearance. Only MafA^S64F/+ males show transiently higher MafA protein levels preceding glucose intolerance and sex-dependent changes to genes involved in Ca²⁺ signaling, DNA damage, aging, and senescence. MAFA^S64F production in male human β cells also accelerate cellular senescence and increase senescence-associated secretory proteins compared to cells expressing MAFA^WT. These results implicate a conserved mechanism of accelerated islet aging and senescence in promoting diabetes in MAFA^S64F carriers in a sex-biased manner.

Keywords: MAFA; beta cell; cellular senescence; diabetes; islet biology; sexual dimorphism.

Figure 1.. Male but not female MafA^S64F/+ mice become glucose intolerant between 4 and 5 weeks of age

(A) Fasted male and female animals underwent intraperitoneal glucose tolerance tests at 4 and 5 weeks of age. Male heterozygous (termed Het) MafA^S64F/+ mice (red line) had improved glucose clearance at 4 weeks but become glucose intolerant by 5 weeks. Female MafA^S64F/+ mice (teal line) had significantly lower fasting blood glucose levels and improved glucose clearance at both time points. (B) High glucose-stimulated insulin secretion in isolated islets was impaired in male MafA^S64F/+ samples at 5 weeks. Islets were incubated with 4.6 mM (low, LG) or 16.8 mM (high, HG) glucose for 1 h. Secreted insulin was normalized to DNA content. (C) Islet insulin content trended lower in MafA^S64F/+ males and was significantly decreased in females. Levels were normalized to DNA content. (D) Male (left panel) and female (right panel) islet β cell area was reduced at 7 weeks in MafA^S64F/+ mice. The area was calculated by dividing the total insulin⁺ area by the total pancreatic area (eosin staining) in pancreas sections obtained every 50 μm multiplied by 100 to obtain percentage (%). (E) Indicative of reduced islet area, the combined islet β and α cell area was significantly reduced in male and female MafA^S64F/+ mice at 7 weeks. (A–C) Two-tailed Student t test; *p < 0.05; **p < 0.01; ***p < 0.001. (D and E) Two-way ANOVA; *p < 0.05; **p < 0.01; ***p < 0.005.

Figure 2.. MAFA protein levels were transiently upregulated at 4 weeks in male MafA^S64F/+ islets

(A) Immunostaining over the course of 3, 4, 5, 6, and 7 weeks revealed that MAFA protein staining intensity only increased at 4 weeks of age in male MafA^S64F/+ (Het) islets. (B) In contrast, MAFA staining intensity was unchanged between female MafA^S64F/+ and WT islets. (C) MafA mRNA levels were significantly reduced at 4 and 5 weeks in both male and female MafA^S64F/+ islet samples. Fold change shown relative to male WT islets. Two-tailed Student t test; *p < 0.05; scale bars, 50 μm.

Figure 3.. Male and female MafA^S64F/+ islet cells regulate a common and distinct set of genes associated with Ca²⁺ and K⁺ channel activity

(A) The Venn diagram illustrates the total number of RNA-seq-identified genes differentially up- or down-regulated between 5-week-old WT and MafA^S64F/+ islets. (B) Gene Ontology (GO): molecular function analysis (p < 0.05) of the 391 genes commonly up- or down-regulated in MafA^S64F/+ (Het) islets revealed alterations in multiple ion channel activity pathways. (C) Heatmaps showing channel gene expression changes common between male and female MafA^S64F/+ islets. (D) Heatmap of Ca²⁺ signaling pathway genes uniquely increased in male MafA^S64F/+ islets identified by Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis (also see Figure 5A). False discovery rate (FDR) < 0.05.

Figure 4.. Glucose-induced Ca²⁺ oscillations and KCl-induced Ca²⁺ responses are altered in MafA^S64F/+ islets

(A) Representative Fura2 traces show loss of the β cell glucose-induced Ca²⁺ oscillations (11 mM glucose [G]) in 5-week-old male MafA^S64F/+ (Het) islets (red line, “non-responders”). Female MafA^S64F/+ islets had 2 different functionally responsive islet populations (“responders,” teal line; non-responders, green line). (B) Quantitation of male (left; 2-tailed Student t test; ***p < 0.001) and female (right; 1-way ANOVA; ****p < 0.0001) islets showed reduced cytoplasmic Ca²⁺ following stimulation with 11 mM glucose. The average peak amplitude was quantitated by dividing the first peak ΔCa²⁺ after 11 mM glucose application by the baseline response at 5 mM glucose. (C) The Ca²⁺ response to 30 mM KCl is reduced in male MafA^S64F/+ islets but increased in female MafA^S64F/+ islets. Representative traces for this experiment are shown in Figure S8. Two-tailed Student t test; **p < 0.01; ***p < 0.001.

Figure 5.. Male MafA^S64F/+ islets display increased expression of aging, DDR, SASP, and cytokine pathway gene signatures

(A) KEGG analysis of the 1,842 genes specifically upregulated in 5-week-old male MafA^S64F/+ (Het) islets. (B–E) Heatmaps reveal male MafA^S64F/+ islets have increased expression of pathway genes associated with (B) aging, (C) DDR, (D) SASP, and (E) cytokine-cytokine receptor interactions. FDR < 0.05. (F) qRT-PCR confirmation of pathway gene changes in 5-week-old male MafA^S64F/+ islets (solid bars). Gene expression was unchanged at 4-week-old male MafA^S64F/+ islets, with the exception of Cdkn1a (p21) upregulation (hash-marked bars). Two-tailed Student t test; *p < 0.05; **p < 0.01.

Figure 6.. DNA damage and senescence markers are increased in dysfunctional male MafA^S64F/+ islets

(A–C) gH2AX staining (A) and 53BP1 (B), markers of DNA double-strand breaks, and p21, a cell-cycle inhibitor (C), were present in male MafA^S64F/+ (Het) islets. Male MafA^S64F/+ islets had a significant increase in the proportion of islets with >10% p21⁺ cells. (D) Male MafA^S64F/+ islets showed reduced LaminB1 protein (left) and mRNA (right). (E and F) SA-β-gal was not produced in 4-week-old male MafA^S64F/+ islets, but was detected in MafA^S64F/+ males by 7 weeks. (E) The SA-β-gal in 7-week-old MafA^S64F/+ islets was of similar intensity to 10- to 12-month-old WT male mouse islets. (F) The proportion of SA-β-gal⁺ islets was quantitated for each sample (n = 3–4). Two-tailed Student t test; **p < 0.01; scale bar, 50 μm.

Figure 7.. Human β cells expressing MAFA^S64F demonstrate accelerated senescence and a species-specific SASP signature

(A) SA-β-gal staining on EndoC-βH2 cells transduced to express WT MAFA or MAFA^S64F and stained for MAFA (red). Scale bar, 100 μm. (B) Inset areas outlined in boxes in (A) are magnified in the left panel. Scale bar, 25 μm. β-gal⁺ area was significantly increased in MAFA^S64F cells (right panel). (C) Human expression of β cell-enriched proteins and senescence associated proteins identified in male MafA^S64F/+ mice. Two-tailed Student t test; *p < 0.05; **p < 0.01. (D) Immunostaining for senescence markers p21 (green) was increased and LaminB1 (red) decreased in MAFA^S64F-expressing EndoC-βH2 cells. (E) CM from WT MAFA or MAFA^S64F-expressing EndoC-βH2 cells were collected, purified and added to EndoC-βH2 cells cultured for 72 h. SASP genes upregulated in male MafA^S64F/+ mice were not identified in EndoC-βH2 cells. However, novel, species-specific secretory senescence-associated factors were identified in this context. N.D., not detectable. One-way ANOVA; *p < 0.05; **p < 0.01; ****p < 0.0001. (F) Schematic of temporal senescence and aging responses in MafA^S64F-expressing male β cells. In male MafA^S64F/+, transiently high MafA protein triggers molecular insults to induce premature senescence marked by cell-cycle arrest, senescence staining (blue cytoplasmic shade), and initiation of a senescence-associated secretory phenotype (SASP, pink secretory factors), resulting in impaired GSIS. Late (progressive) senescence propagates this phenotype with SASP amplification and diversification.