GRAPHICAL ABSTRACT

Highlights

• Expression of Fam151a is reduced in pancreatic β-cells of T2D mice.

• Fam151a deletion in β-cells reduces glucose-stimulated insulin secretion.

• Fam151a knockout in mice leads to dysregulation of glucose tolerance and biphasic insulin secretion.

• Metabolomic analysis reveals that Fam151a regulates several key metabolic pathways.

INTRODUCTION

Diabetes mellitus is a chronic metabolic disorder defined by impaired glucose homeostasis. Type 2 diabetes mellitus (T2DM), which accounts for more than 90% of global diabetes cases (International Diabetes Federation, 2025), progresses via a well-established pathological sequence: peripheral insulin resistance initially triggers compensatory β-cell hyperinsulinemia, which ultimately leads to β-cell failure and progressive insulin deficiency [1-3]. Notably, in obese individuals, chronic exposure of pancreatic β-cells to glucolipotoxicity induces abnormal islet hyperplasia, activates integrated stress responses, and impairs insulin secretory function through oxidative stress and endoplasmic reticulum (ER) stress [4]. Concurrently, macrophage infiltration into pancreatic islets exacerbates T2DM progression via dual mechanisms: cell contact-dependent endocytosis of insulin and the release of pro-inflammatory cytokines (such as interleukin-1β and tumor necrosis factor-α) and metabolites (such as lactate, succinate, and nitric oxide) that damage β-cells [5-10].

Insulin synthesis and secretion are dynamically regulated by local hormonal and microenvironmental factors. Biosynthesis of insulin proceeds through pre-proinsulin and proinsulin, with transport from the rough ER to the trans-Golgi network for cleavage, processing, and packaging into mature secretory granules [11]. Canonical biphasic secretion relies on a metabolic-electrical cascade: glucose-adenosine triphosphate (ATP)/adenosine diphosphate (ADP) ratio-ATP-sensitive potassium (K_ATP) channel closure-membrane depolarization-Ca²⁺ influxinsulin granule exocytosis. Recent studies propose a novel redox spatiotemporal coupling model (Mito_Cat-Mito_Ox model), which proposes that during the triggering phase of insulin secretion, intracellular pyruvate is utilized by cell membrane-associated pyruvate kinase to generate adequate ATP for K_ATP channel closure, while oxidative phosphorylation remains relatively low. The subsequent amplifying phase depends on the phosphoenolpyruvate cycle to reset oxidative phosphorylation, which in turn generates a surge of ATP to support sustained secretion. The establishment of this model underscores the exquisite precision and inherent spatiotemporal complexity of insulin secretion [12-14]. Pancreatic endocrine cells exhibit marked heterogeneity in transcriptome profiles, protein expression patterns, and hormone secretion dynamics. Using single-cell RNA sequencing (scRNA-seq), researchers have identified multiple pancreatic β-cell subpopulations and characterized their distinct expression patterns under diverse pathological conditions [15-17].

Family with sequence similarity 151 member A (FAM151A) is a 68 kDa protein that harbors two evolutionarily conserved domains of unknown function 2181 (DUF2181) domains that exhibit phylogenetic preservation across diverse taxa. Notably, in Caenorhabditis elegans, the DUF2181-containing protein MNR-1 has been demonstrated to regulate neuronal function [18]. In contrast, in mice, FAM151B, which contains a single DUF2181 domain, is essential for the maintenance of retinal function, whereas FAM151A is not [19]. This observation reflects both functional conservation and divergence within the FAM151 family. However, the relationship between FAM151A and diabetes remains unclear.

In this study, we used scRNA-seq to reveal a significant downregulation of Fam151a in pancreatic β-cells in db/db mice. We thus characterized the potential functions of FAM151A in vitro and in vivo. At the cell level, deficiency of Fam151a resulted in retention of proinsulin in the cytoplasm and reduced insulin secretion. At the animal level, deletion of Fam151a in β-cells caused impaired glucose tolerance, attenuated biphasic glucose-stimulated insulin secretion (GSIS), and reduced potassium chloride (KCl)-stimulated insulin secretion (KSIS), while maintaining normal insulin sensitivity and pancreatic islet morphology. These results demonstrate for the first time that FAM151A plays an important role in regulating insulin secretion and the pathogenesis of T2DM.

METHODS

Cell culture

HeLa and HEK293T cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM) with 100 U/mL penicillin/streptomycin (P/S) and 10% fetal bovine serum (FBS). Rat INS-1E cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 with 100 U/mL P/S and 10% FBS. Mouse MIN6 cells were cultured in DMEM with 100 U/mL P/S and 15% FBS. All cell lines were cultured at 37°C in a humidified incubator with 5% CO₂. Transient transfection was performed with polyethyleneimine for HEK293T cells and jetPRIME transfection reagent (Polyplus, Illkirch-Graffenstaden, France; cat#101000046) for HeLa and MIN6 cells.

RNA extraction and real-time quantitative polymerase chain reaction

Total RNA was extracted from cells and mouse tissues using the TRIzol method. Equal amounts of RNA (1 μg) were subsequently reverse-transcribed into cDNA using the Fast Quant RT Kit (ABclonal, Woburn, MA, USA; cat#RK20400). Realtime quantitative polymerase chain reaction (qPCR) was performed on an ABI QuantStudio 6 system (Thermo Fisher Scientific, Waltham, MA, USA). The mRNA expression level of the target gene was normalized to the average value of the β-actin gene. The primers used in qPCR were as follows: 5ʹ-GATCATTGCTCCTCCTGAGC-3ʹ, and 5ʹ-ACTCCTGCT TGCTGATCCAC-3ʹ for β-actin, and 5ʹ-GCTGTGAGAACG ATGCAATTTG-3ʹ, and 5ʹ-TTGAACCCTTCTACGGTGA CA-3ʹ for Fam151a.

Immunoblotting

Total protein from cells and mouse tissues was lysed using radioimmunoprecipitation assay (RIPA) lysis buffer containing 10% proteinase inhibitor cocktail and phosphatase inhibitors (Sigma-Aldrich, St. Louis, MO, USA). After centrifugation at 12,000 ×g for 5 minutes at 4°C, the supernatant was analyzed using the bicinchoninic acid (BCA) kit (Beyotime, Shanghai, China; cat#P0012) to normalize for protein content. Samples were resolved by 10% or 15% sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to 0.22 or 0.45 μm polyvinylidene fluoride membranes. The membrane was then incubated in 3% bovine serum albumin (BSA) blocking buffer for 1 hour and with the primary antibody at 4°C overnight. Horseradish peroxidase (HRP)-conjugated secondary antibodies were incubated at room temperature (RT) for 2 hours. Antibodies used in immunoblotting were as follows: FLAG (Proteintech, Rosemont, IL, USA; cat#66008-4-1g), E-cadherin (Cell Signaling Technology, Danvers, MA, USA; cat# 3195S), glyceraldehyde 3-phosphate dehydrogenase (GAPDH; Proteintech, cat#60004-1-1g), INSULIN (Cell Signaling Technology, cat#8138S), ACTIN (Santa Cruz, cat#517582), FAM151A (Novus, cat#NBP2-13983-25ul), heat shock protein 90 (HSP90; Cell Signaling Technology, cat#4874S), P62 (Cell Signaling Technology, cat#5114S), light chain 3 (LC3; Proteintech, cat#14600-1-AP), activating transcription factor 6 (ATF6; Abcam, Waltham, MA, USA; cat#ab37149), heat shock protein family A members 5 (HSPF5 or binding immunoglobulin protein [BIP]; Cell Signaling Technology, cat#3177S), and DNA damage inducible transcript 3 (DDIT3 or C/EBP homologous protein [CHOP]; Proteintech, cat# 15204-1-AP). The images were captured using the Tanon-5200 visualization instrument (Tanon, Shanghai, China).

Lentivirus packaging and infection

Knockout and overexpression of the Fam151a gene were performed using a lentiviral packaging system to generate stable cell lines. The lentivirus was generated by transfecting two lentiviral packaging vectors (pMD2.G and psPAX2) and the core plasmid into HEK293T cells. Lentivirus supernatants were harvested through a 0.45 μm filter after 48 and 72 hours. To concentrate the virus, the supernatant was subjected to centrifugation at 4°C and 20,000 rpm for 2 hours. The supernatant was then carefully aspirated, and the viral pellet was resuspended in 1 mL of 1×phosphate-buffered saline (PBS). The suspended lentivirus was divided into four portions and stored at –80°C. INS-1E cells were seeded in a 6-cm dish and infected with the corresponding lentivirus. After 48 hours, transduced cells were screened by the addition of puromycin (1 μg/mL) for a further 48 hours. The medium was replaced with fresh medium every day.

CRISPR/Cas9 to delete Fam151a

The single guide RNA (sgRNA) sequence (5ʹ-GCTGTTGG CTGCGTGATACC-3ʹ targeting exon 1) used to delete the Fam151a gene was designed using the website: http://crispr.mit.edu. Then the sgRNA oligo duplex was inserted into lenti-CRISPRv2 and confirmed by DNA sequencing. Lentivirus packaging and infection were performed as described above. After INS-1E cells had been infected for 2 days, genomic DNA was extracted using a genome extraction kit (Tiangen, Beijing, China; cat#DP304-03) according to the manufacturer’s instructions and amplified by PCR to obtain the targeted region. The primers for knockout efficiency determination were as follows: F: 5ʹ-CCTGCTGTCTCTCTTCCGAC-3ʹ, R: 5ʹ-GTGG TCAGGTGTCCCTTACAG-3ʹ. In parallel, cells were lysed for immunoblotting to validate the knockout effect.

Plasmid construction

Fam151a cDNA was amplified from HEK293T cells and mouse kidney tissues uing high-fidelity KOD (Thermococcus kodakarensis) polymerase. Fam151a was cloned into pLVXm-Myc and pCDH-3xFlag plasmids, respectively. All plasmids were subjected to DNA sequencing and immunoblotting verification.

FAM151A subcellular localization

Cells were seeded on glass coverslips and allowed to adhere overnight. Transient transfection of HeLa and MIN6 cells was performed with empty vector or Fam151a overexpression plasmids on the following day. After 48 hours of cultivation, cells were fixed in 4% paraformaldehyde (PFA) for 10 minutes and permeabilized with 0.1% Triton X-100 in 1×PBS for another 10 minutes. After blocking with 3% BSA in 1×PBS for 1 hour at RT, cells were incubated with the primary antibody at 4°C overnight. Then, cells were incubated with fluorescein-labeled secondary antibodies for 1 hour, and 4ʹ,6-diamidino-2- phenylindole (DAPI) was used to stain the nuclei for 8 minutes at RT away from light. Between each step, samples were washed three times with 1×PBS for 5 minutes. Fluorescence images were captured and processed using a ZEISS LSM880 microscope (Oberkochen, Germany). The primary antibodies were as follows: FLAG (Proteintech, cat#66008-4-1g), MYC (Santa Cruz, Santa Cruz, CA, USA; cat#sc-40), CALNEXIN (Proteintech, cat#10427-2-AP), and CHOP (Servicebio, Wuhan, China; cat#GB115691). Fluorescein-labeled secondary antibodies were as follows: Alexa Fluor^TM 488 goat anti-mouse immunoglobulin G (IgG) (H+L) (Invitrogen, Waltham, MA, USA; cat#A11029) and Alexa Fluor^TM 546 goat anti-rabbit IgG (H+L) (Invitrogen, cat#A11035).

Membrane and cytosol protein extraction

HEK293T cells were maintained in 10-cm culture dishes and passaged at approximately 90% confluency. For transient transfection, cells were transfected with empty vector or Fam151a overexpression plasmids for each group. After 48 hours, cells were harvested and extracted using the Membrane and Cytosol Protein Extraction Kit (Beyotime, cat#P0033) according to the manufacturer’s instructions.

Measurement of total insulin

INS-1E cell pellets were lysed in buffer (20 mM Tris-HCl, pH 7.4, 150 mM NaCl, 1 mM ethylenediaminetetraacetic acid [EDTA], 1 mM ethylene glycol-bis-(β-aminoethyl ether-N,N, Nʹ,Nʹ-tetraacetic acid [EGTA], 1% Triton X-100) and subjected to three freeze-thaw cycles in liquid nitrogen. Isolated mouse islets were lysed in RIPA buffer supplemented with protease and phosphatase inhibitors, followed by sonication on ice. All lysates were centrifuged at 12,000×g for 5 minutes at 4°C. Supernatants were collected and measured using a commercial enzyme-linked immunosorbent (ELISA) assay kit (Ezassasy, cat#MS200) according to the manufacturer’s instructions. The total insulin content of cells and isolated islets was normalized to protein concentration and islet number, respectively.

GSIS for INS-1E cells

INS-1E cells cultured in 24-well plates were starved for 1 hour with Krebs-Ringer bicarbonate buffer (KRB buffer: 2.6 mM CaCl₂/2H₂O, 1.2 mM MgSO₄/7H₂O, 1.2 mM KH₂PO₄, 4.9 mM KCl, 98.5 mM NaCl, and 25.9 mM NaHCO₃, supplemented with 20 mM HEPES). The culture supernatant was aspirated and centrifuged at 4,000 rpm for 3 minutes. Insulin concentrations were measured using the ELISA kits mentioned above.

Transmission electron microscopy

INS-1E cells from control (CTRL) and Fam151a-knockout (FKO) groups and primary islets from Fam151a^WT and β-cellspecific Fam151a-knockout (Fam151a^βKO) mice were fixed in electron microscopy-grade fixative (2.5% glutaraldehyde in 0.1 M phosphate buffer). Subsequent processing, including postfixation, dehydration, embedding, sectioning, and staining, was performed by Servicebio. The distances were quantified using ImageJ software (National Institutes of Health, Bethesda, MD, USA).

Seahorse real-time cell metabolic analysis

As briefly described previously, INS-1E cells were seeded in XFe 96-well microplates (Agilent Technologies, Santa Clara, CA, USA) at a density of 8,000 cells per well and cultured for 24 hours [20]. For measurements of the oxygen consumption rate (OCR), cells were washed and maintained in unbuffered XF assay medium supplemented with 10 mM glucose, 2 mM glutamine, and 1 mM pyruvate (pH 7.4). Sequential injections of oligomycin (1 μM final concentration), a mixture of rotenone and antimycin A (RAA, 0.5 μM each), and carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone (FCCP; 2 mM final) were performed at specified time points in accordance with the manufacturer’s protocol. For assessment of the extracellular acidification rate (ECAR), cells were washed and incubated in unbuffered XF assay medium containing 2 mM glutamine (pH 7.4). Glucose (10 mM final), oligomycin (1 μM final), and 2-deoxyglucose (2-DG, 100 mM final) were injected sequentially at the indicated time points according to the manufacturer’s instructions. Data were normalized to cell number. All plates were normalized to protein content measured from the same cells after Seahorse analysis by BCA.

Mouse studies

All mice used in this study were on a C57BL/6J background and maintained under specific pathogen-free conditions. Animal experiments were conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee of the Shanghai Institute of Nutrition and Health, Chinese Academy of Sciences (CAS) (Approval No. SINH-2024-CY-1). db/db mice were purchased from GemPharmatech Company (Nanjing, China). High-fat/high-sucrose (HF/HS) isolated mouse islet samples were kind gifts from Huiyong Yin’s lab. Islet samples from high-fat diet (HFD) mice were isolated from mice following a 3-month HFD with 60% of kilocalories (Research Diets, New Brunswick, NJ, USA; cat#D12492). The Fam151a-flox and Ins2-Cre mice were purchased from Shanghai Model Organisms Center (Shanghai, China). The strategy for generating Fam151a^WT and Fam151a^βKO mice is illustrated in Supplementary Fig. 5A. Genotyping of mice expressing Fam151a alleles with a floxed site was performed with primers: F: 5ʹ-CTGTGTGCTGCCGTGATAGT-3ʹ, R: 5ʹ-ACATACAGGC AAACCCCACA-3ʹ. Genotyping of mice expressing Ins2-Cre was performed with primers: P1: 5ʹ-TCGATGCAACGAGT GATGAG-3ʹ, P2: 5ʹ-TCCATGAGTGAACGAACCTG-3ʹ, P3: 5ʹ-CAAATGTTGCTTGTCTGGTG-3ʹ, P4: GTCAGTCGAG TGCACAGTTT-3ʹ. Blood glucose levels were measured from tail vein blood using a glucometer after a 6-hour fast. Body composition (lean mass and fat mass) was analyzed by non-invasive quantitative magnetic resonance.

Glucose tolerance test, insulin tolerance test, and glucosestimulated insulin secretion

Mice were fasted for 16 hours (starting at 5:00 PM) for the glucose tolerance test (GTT) and GSIS and for 6 hours (starting at 9:00 AM) for the insulin tolerance test (ITT), with free access to autoclaved water. D-Glucose (2 g/kg) or insulin (1 U/kg) was injected intraperitoneally for GTT and ITT. Blood glucose levels at 0, 15, 30, 60, 90, and 120 minutes were measured sequentially after injection. For GSIS, mice were intragastrically administered D-glucose (3 g/kg). Blood glucose levels or serum insulin samples were obtained at 0, 15, and 30 minutes after gastric gavage. Insulin concentration was measured using the insulin ELISA kit.

Isolation of the mouse islet

Mice were euthanized with isoflurane, followed immediately by fresh collagenase P (Roche, Basel, Switzerland; cat#112138 57001) perfusion. Briefly, the pancreas was distended via bile duct infusion with collagenase P (1 mg/mL in Hanks’ balanced salt solution [HBSS]) and digested at 37°C for 15 minutes. After terminating the digestion, the fully digested pancreas was agitated vigorously for 30 seconds, followed by centrifugation at 400×g and 4°C for 3 minutes. The pellet was resuspended in HBSS and passed through a 40 μm cell strainer. Purified islets were then hand-picked three times in pre-cooled HBSS with 10% FBS under a stereo microscope. Islets were equilibrated in RPMI 1640 with 100 U/mL P/S and 10% FBS overnight at 37°C in a 5% CO₂ incubator.

Islet perifusion assay

Primary mouse islets were cultured overnight following isolation before the perifusion assay was performed as described [21]. For each assay, perifusion chambers were loaded with 150 islet equivalents. The islets were initially perfused with KRB buffer containing 2.8 mM glucose for 30 minutes for recovery. Subsequently, perifusion was performed with KRB buffer containing 2.8 mM glucose, 16.7 mM glucose, and 2.8 mM glucose for 25, 30, and 20 minutes, respectively, at a flow rate of 1 mL/min. Fractions were collected every minute. Islets were collected from the perifusion chamber to determine total protein content using the BCA protein assay after the assay. Insulin secretion at each time point was measured with the homogeneous time-resolved fluorescence (HTRF) insulin assay kit (Cisbio, Codolet, France; cat#62INSPEC) according to the manufacturer’s instructions and normalized to total protein content. To quantitatively assess biphasic insulin secretion, phases I and II of insulin secretion were determined according to phase II plateau formation. Subsequently, the phases were quantified by calculating the area under the curve for each respective phase.

Intracellular calcium imaging of primary islets

Primary murine islets, cultured overnight after isolation, were washed twice with KRB buffer containing 2 mM glucose. Then the islets were loaded with KRB buffer containing 5.6 μM Fluo-4 AM (Thermo Fisher Scientific, cat#F14201), 0.4‰ Pluronic F127 (Sigma-Aldrich, cat#P2443), and 2 mM glucose for 45 minutes. After staining, the islets were washed twice more with KRB buffer containing 2 mM glucose, seeded on a 35-mm glass-bottom dish (MatTek, Ashland, MA, USA), and cultured for an additional 15 minutes. Images were captured after the islets had settled in a humidified 37°C cell culture chamber with 5% CO₂, mounted on a Zeiss LSM 800 inverted confocal microscope. Continuous image acquisition was initiated at 10-second interval precisely 2 minutes after the introduction of glucose supplementation to attain a final concentration of 16.7 mM. This continuous imaging process persisted until the fluorescence signal reached a plateau. Fluorescence intensity dynamics were calculated using ImageJ software.

ATP content measurement

Primary murine islets were isolated from Fam151a^WT and Fam151a^βKO mice, cultured overnight, and washed twice with KRB buffer. The islets were incubated in KRB buffer containing 2.8 or 16.7 mM glucose for 1 hour. Then islets were lysed with the lysis buffer in the commercial ATP Assay Kit (Beyotime, cat# S0026), and ATP concentrations were measured following the manufacturer’s instructions, and normalized to the respective protein content.

Immunofluorescence staining

Pancreatic tissue was fixed in 4% PFA overnight. Paraffin-embedded and frozen sections were prepared by Servicebio. Paraffin sections were baked at 68°C for 30 minutes and then subjected to xylene-alcohol gradient dewaxing treatment. Antigen retrieval was subsequently performed with sodium citrate buffer (pH 6.0) in boiling water for 30 minute, followed by cooling to RT. For frozen sections, heat-mediated antigen retrieval was omitted if the primary antibody specifically recognized a native epitope. Sections were incubated with blocking buffer containing 5% normal goat serum and 0.1% Triton X-100 in 1×PBS at RT for 1 hour. The subsequent steps were similar to those used for cell immunofluorescence. The primary antibodies used were as follows: INSULIN (Proteintech, cat#66198-1-1g), GLUCAGON (Abcam, cat#92517), pancreatic and duodenal homeobox 1 (PDX1; Cell Signaling Technology, cat# 5679S), and Maf bZIP transcription factor A (MAFA; Cell Signaling Technology, cat#79737S). The quantitative indicators for the islets were assessed as previously reported [22].

Immunohistochemistry staining

The immunohistochemistry procedure was similar to the immunofluorescence protocol for paraffin-embedded sections. Sections were incubated with the corresponding HRP-conjugated secondary antibodies instead of fluorescent secondary antibodies. In addition, the sections were incubated with 3,3ʹ- diaminobenzidine (DAB) chromogen solution for 30 minutes and hematoxylin for 8 minutes at RT. Between each step, samples were washed three times with 1×PBS for 5 minutes. Images were captured using an Olympus BX51 microscope (Olympus Corporation, Tokyo, Japan) and analyzed with ImageJ software.

Metabolomics

INS-1E cells from the CTRL and FKO groups (n=3 per group) were washed with pre-cooled 1×PBS and harvested with extraction solution consisting of: 40: 40: 20(v/v/v) methanol: acetonitrile: water, 0.1% formic acid, and 15% ammonium bicarbonate. All reagents were of mass spectrometry grade. After two rounds of vortexing for 30 seconds, samples were subjected to centrifugation at 12,000×g and 4°C for 20 minutes. A minimum of 200 μL supernatant was required for liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.

scRNA-seq and cluster analysis

The analytical methods for scRNA-seq data are described previously [23,24]. Analysis of the scRNA-seq data was performed using the online software at https://singlecell.novelbrain.com/.

Statistical analysis

Data analysis was performed with GraphPad Prism 8 (Graph-Pad Software Inc., San Diego, CA, USA). All data are expressed as the mean±standard error of the mean. Statistical significance was evaluated using the unpaired Student’s t-test for two groups. For comparisons of three or more conditions under a single factor, one-way analysis of variance (ANOVA) with Tukey’s honestly significant difference (HSD) post hoc analysis was applied. For data involving two factors, two-way ANOVA with Tukey’s HSD post hoc analysis was used. Differences were considered significant at P<0.05, P<0.01, and P<0.001.

RESULTS

Fam151a expression was significantly downregulated in β-cells in T2DM mice and under chronic high-glucose conditions

To identify novel genes involved in the pathogenesis of T2DM, we performed scRNA-seq on isolated islets from 10-week-old male C57BL/6 mice (wild type [WT]), 10-week-old db/db mice (early db/db), and 14-week-old db/db mice (late db/db), as previously reported by our group (Fig. 1A) [24]. Based on the expression profiles of marker genes, we reclassified four major endocrine cell types (α, β, δ, and pancreatic polypeptide [PP] cells) into 10 transcriptionally distinct clusters. Among these, clusters 3, 4, 5, 6, 10, 12, and 15 were identified as β-cell subpopulations (Fig. 1B). Of particular interest, Fam151a expression was significantly enriched in clusters 3 and 12, indicating that its expression was upregulated in these two β-cell clusters (Fig. 1C). Further sub-clustering analysis based on expression of Ins1, Mafa, Slc2a2 (signature markers of mature β-cells), and Aldh1a3 (a marker gene of dedifferentiated β-cells) revealed that Fam151a expression was highest in mature β-cells and markedly decreased throughout diabetic progression in db/db mice (Fig. 1D and E) [25,26]. As Fam151a was detected only in β-cells, we analyzed the β-cell clusters in detail. There were 11 β-cell clusters in the scRNA-seq data; clusters 1, 2, and 3 were mainly from the WT mice, while clusters 6 to 11 were mainly from db/db mice (Supplementary Fig. 1). All 11 clusters expressed pan-β-cell markers, including Ins1, Ins2, and Iapp (Supplementary Fig. 2), confirming the β-cell identity of these clusters. Fam151a was highly expressed in clusters 1 and 2, which also had high expression of mature β-cell markers such as Mafa, Slc2a2, and Slc30a8 (Supplementary Fig. 2). As expected, most clusters of β-cells from db/db mice had high expression of dedifferentiation markers, including Aldh1a3 and Cd81 (Supplementary Fig. 2). Interestingly, markers of endocrine cells other than β-cells, such as Gcg, Sst, and Ppy, were highly expressed in clusters 3, 5, and 11 (Supplementary Fig. 2). Clusters 3 and 11 expressed acinar cell markers, while cluster 6 was enriched for Mt1 expression and cluster 9 was enriched for Cck expression (Supplementary Fig. 2). Overall, these analyses indicated that Fam151a was expressed only in mature β-cells from WT mice and not in β-cells from db/db mice.

Consistent with the results of scRNA-seq analysis, we observed significant downregulation of Fam151a in islets from both db/db and HF/HS-fed mice (Fig. 1F and G), as well as in HFD-fed mice (Fig. 1H). At the cellular level, prolonged high-glucose treatment induced a time-dependent reduction in Fam151a expression in INS-1E cells (Fig. 1I). These results, therefore, indicated that Fam151a expression is downregulated in β-cells under T2DM and high-glucose conditions.

FAM151A is an endoplasmic reticulum-resident single-pass transmembrane protein

Given the limited characterization of FAM151A, we initially focused on determining its subcellular localization. Bioinformatic analyses (https://services.healthtech.dtu.dk/services/TMHMM-2.0/) identified FAM151A as a single-pass transmembrane protein (Fig. 2A). To validate this prediction, we overexpressed Fam151a-3×Flag in HEK293T cells and isolated membrane and cytoplasmic fractions, which confirmed its location on cellular membranes (Fig. 2B). Immunofluorescence staining in HeLa cells transfected with either MYC- or FLAG-tagged Fam151a plasmid constructs demonstrated predominant localization to the ER (Fig. 2C). Consistently, this ER-specific distribution was observed in MIN6 pancreatic β-cells (Fig. 2D), confirming FAM151A as an ER-resident single-pass transmembrane protein.

Fam151a deletion in INS-1E cells leads to insulin granule retention and alteration in metabolic activity

To investigate the role of FAM151A in diabetes pathogenesis, we generated a stable FKO INS-1E cell model using clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9 technology, with knockout efficiency >50% confirmed by Western blot (Fig. 3A). Notably, intracellular insulin and proinsulin protein levels were significantly elevated in FKO cells (Fig. 3B), and total insulin content was markedly increased as measured by ELISA (Fig. 3C). Transmission electron microscopy (TEM) further revealed substantial insulin vesicle accumulation in FKO cells (Fig. 3D). Despite elevated insulin storage, FKO cells did not show detectable changes in autophagy, the unfolded protein response, or ER-mitochondria contact distance (Supplementary Fig. 3A-C). Furthermore, the ER localization of FAM151A itself remained unchanged after 24- hour thapsigargin treatment used to induce ER stress (Supplementary Fig. 3D). However, insulin secretion stimulated by 16.7 mM glucose and 30 mM KCl was significantly impaired in FKO cells, indicating a defect in regulated insulin vesicle exocytosis (Fig. 3E). Assessment of cellular energy metabolism revealed that Fam151a deficiency reduced maximal respiratory capacity in the OCR assay (Fig. 3F). Fam151a deletion also suppressed glycolysis, as indicated by decreased ECAR (Fig. 3G). To further elucidate the function of FAM151A, we also established a Fam151a-3×Flag-overexpressing (FOV) INS-1E cell line (Supplementary Fig. 4A). Contrary to the effect of Fam151a knockdown, Fam151a overexpression enhanced GSIS and KSIS (Supplementary Fig. 4B) and maximal respiratory capacity (Supplementary Fig. 4C). These results thus suggested that Fam151a deficiency disrupts insulin secretion, likely through defective insulin granule release, associated with reductions in maximal respiration and glycolysis. Conversely, Fam151a overexpression mitigates the defects in insulin secretion and metabolic changes.

Fam151a^βKO mice exhibit intracellular insulin accumulation and secretory dysfunction in β-cells

To elucidate the physiological role of FAM151A, we generated Fam151a^βKO by crossing Fam151a^flox/flox mice (Fam151a^WT) with Ins2-Cre transgenic lines (Supplementary Fig. 5A). Western blotting revealed high FAM151A protein expression in kidney, intestine, and liver tissues and validated its specific deletion in pancreatic β-cells in Fam151a^flox/βKO mice (Fig. 4A). At 8 weeks of age, Fam151a^βKO mice exhibited comparable body weight, fasting blood glucose, lean mass, and fat mass relative to Fam151a^WT mice (Fig. 4B and Supplementary Fig. 5B). Metabolic phenotyping demonstrated significantly impaired glucose tolerance during GTT at the 30- and 60-minute time points at 9 weeks of age, while ITT showed no differences in Fam151a^βKO mice (Fig. 4C and D). GSIS in Fam151a^βKO mice was also impaired in vivo at 11 weeks of age (Fig. 4E). Notably, as metabolic stress progressed, Fam151a^βKO mice exhibited progressive deterioration in both GTT and in vivo GSIS at 24 weeks of age (Supplementary Fig. 5C and D).

Islet perifusion experiments further demonstrated impaired biphasic insulin secretion in Fam151a^βKO mice with blunted responses during both the triggering and amplifying phases. Notably, KCl-induced depolarization, which bypasses upstream metabolic sensing and directly elicits Ca²⁺ entry, also elicited compromised insulin release (Fig. 4F). Complementing these findings, TEM revealed a significant increase in intracellular mature insulin granules in Fam151a^βKO cells, accompanied by a tendency toward a reduction in insulin vesicles near the plasma membrane (Fig. 4G and H). Under hyperglycemic stimulation, live-cell calcium imaging showed attenuated cytosolic Ca²⁺ transients in Fam151a^βKO islets (Fig. 4I), consistent with diminished metabolic-electrical coupling. Direct measurement revealed a reduction in ATP content in the islets of Fam151a^βKO mice (Fig. 4J). Notably, islets isolated from Fam151a^βKO mice exhibited increased intracellular insulin content, characterized by elevated protein levels of both proinsulin and insulin (Fig. 4K). Collectively, these results indicated that FAM151A is required for normal insulin secretion in pancreatic β-cells. Its loss leads to defective glucose- and KCl-stimulated insulin exocytosis and aberrant intracellular insulin accumulation, independent of peripheral insulin sensitivity.

Fam151a^βKO mice preserve normal islet morphology and identity

We next analyzed the morphology of the islets using pancreas sections. Immunohistochemical assessment revealed no significant differences in islet number or total islet area between Fam151a^βKO and Fam151a^WT mice (Fig. 5A). Immunofluorescent staining of pancreas sections revealed that the overall morphology of the islets as well as the mean fluorescence intensity of α-cells and β-cells within islets, was not altered by Fam151a deletion (Fig. 5B). Consistently, the expression levels of key β-cell identity marker genes Mafa and Pdx1 exhibited no notable alterations in the islets of Fam151a^βKO mice (Fig. 5C and D). Collectively, these results demonstrated that pancreatic islet morphology and β-cell identity remained unaltered under the experimental conditions, indicating that Fam151a deficiency does not affect pancreatic β-cell morphology, identity or survival.

Fam151a deletion in INS-1E cells disrupts pentose phosphate and purine metabolic pathways

Based on previous studies, multiple sequence alignment of amino acid sequences confirmed that FAM151A belongs to the glycerophosphodiester phosphodiesterase/PLC-like phosphodiesterases (GDPD/PLCD) superfamily, as its first DUF2181 domain is associated with phosphodiester bond hydrolysis [19, 27]. To explore the effects of FAM151A on pancreatic β-cells and its potential enzymatic functions, we conducted metabolomic analysis in INS-1E cells. Partial least squares discriminant analysis results showed significant differences caused by Fam151a knockout in INS-1E cells (FKO cells) (Fig. 6A). Volcano plot analysis revealed that FKO cells had significant alterations in metabolite levels. Among the downregulated metabolites were phosphoribosyl pyrophosphate (PRPP; 0.30-fold), inosine (0.39-fold), inosine monophosphate (IMP; 0.47-fold), and fructose-6-phosphate (0.21-fold). Conversely, metabolites such as 6-hydroxy-5-methoxyindole glucuronide (2.10-fold) and 2’’,4’’,6’’-triacetylglycitin (2.03-fold) were markedly upregulated (Fig. 6B). Pathway enrichment analysis of all differentially expressed metabolites identified the pentose phosphate pathway (PPP) and purine metabolism as the most significantly altered pathways in FKO cells (Fig. 6C). Consistent with the heatmap analysis, which highlighted all differentially expressed metabolites associated with these pathways in red and green (Fig. 6D-F), FKO cells exhibited significant metabolic perturbation. Specifically, despite unchanged absolute ATP levels, we observed a marked elevation in ADP (1.25-fold) and adenosine monophosphate (1.58-fold), thereby significantly reducing the ATP/ADP ratio (0.73-fold); this decrease likely contributes to suppression of insulin secretion. Most glycolytic intermediates were reduced by Fam151a knockout (Fig. 6D), suggesting inhibition of glycolysis in these cells. In addition, consistent with the observation that the PPP pathway and purine metabolism were altered in FKO cells, certain metabolites within these two pathways were also altered (Fig. 6E and F). Therefore, metabolomic profiling revealed that the loss of Fam151a compromises β-cell function, likely through concurrent disruptions in the ATP/ADP ratio, glycolysis, the PPP, and purine synthesis, highlighting the potential role of FAM151A as a key integrator of core metabolic pathways (Fig. 6G).

DISCUSSION

scRNA-seq has proven instrumental in addressing the significant heterogeneity of pancreatic islets, thereby revealing critical aspects of cellular diversity and pathological variation. Our study provides the first evidence that FAM151A is involved in the regulation of insulin secretion and plays a functional role during the pathogenesis of T2DM. We found that the expression of Fam151a is downregulated in the β-cells of T2DM mice and in high-glucose treated β-cells (Fig. 1). Studies at the cellular level demonstrated that FAM151A is an ER-resident single-pass transmembrane protein (Fig. 2). Deletion of Fam151a resulted in proinsulin accumulation and secretory vesicle retention in INS-1E cells (Fig. 3). Deletion of Fam151a specifically in β-cells of the mice revealed dysfunction in biphasic insulin secretion and glucose tolerance (Fig. 4). These results pinpoint a critical function of FAM151A in regulating insulin homeostasis and β-cell function. Interestingly, our findings are supported by previous observations that Fam151a expression was altered in islets when glucose homeostasis is compromised. Comprehensive RNA-seq analyses established Fam151a as a gene highly and preferentially expressed in pancreatic β-cells, with minimal expression in α- and δ-cells [28]. Strikingly, in islets from T2DM model mice, Fam151a ranks among the most significantly downregulated transcripts [29]. Islets from mice lacking Islet1 or Gprc5b (essential for β-cell maturation and insulin secretion) also exhibit decreased Fam151a expression [30,31]. In addition, Fam151a is differentially expressed in β-cells of obese SM/J mice (Jackson Laboratory, Bar Harbor, ME, USA) following 20- or 30-week HFD feeding and displays a significant negative correlation with serum insulin levels [32,33]. Collectively, these reports and our findings suggest an association between FAM151A and the functional integrity of pancreatic β-cells particularly under T2DM conditions. Our functional characterization revealed that Fam151a^βKO mice exhibit β-cell dysfunction, manifested as glucose intolerance, blunted GSIS and KSIS, dampened Ca²⁺ responses, and decreased ATP content under high-glucose conditions—despite preserved systemic insulin sensitivity (Fig. 4). Furthermore, Fam151a^βKO mice at an older age displayed a progressive worsening of glucose homeostasis shown as deteriorating glucose tolerance and impaired insulin secretory function (Supplementary Fig. 5C and D). However, Fam151a deletion had no effect on the size, morphology, or distribution pattern of the islets (Fig. 5A and B). In addition, the expression levels of key markers of mature β-cells, including Mafa and Pdx1 were not altered by Fam151a deletion (Fig. 5C and D). These observations suggest that FAM151A mainly modulates insulin secretion and cellular metabolism rather than transcriptional regulation, consistent with its ER localization. Fam151a deficiency appears to specifically compromise insulin secretory capacity likely by disrupting enzyme activity, metabolite transport or organelle function [34].

Our study of cellular energy metabolism assays (OCR and ECAR) revealed that Fam151a deficiency impairs maximal respiration and glycolysis in β-cells (Fig. 3F and G). Recent studies have also suggested that FAM151A may regulate the development and progression of colorectal cancer via mitochondrial fission [35]. Based on the observed alterations in cellular energy metabolism in Fam151a-deleted β-cells, one of potential mechanisms underlying the regulation of insulin secretion by FAM151A is the ATP/ADP ratio. Our metabolomics analysis revealed a decrease in the ATP/ADP ratio (0.73-fold) in Fam151a-deficient β-cells (Fig. 6F). Direct measurement also demonstrated a reduction in ATP content in the islets of Fam151a^βKO mice (Fig. 4J). It is well known that ADP can reopen K_ATP channels and reset the membrane potential in β-cells. The reduction in the ATP/ADP ratio caused by deletion of Fam151a could attenuate insulin secretion [36].

It is noteworthy that although our in vitro, ex vivo, and in vivo experiments consistently demonstrated that Fam151a deletion impairs GSIS in β-cells, the magnitude of this reduction differed across experimental systems. In vivo, glucose stimulation increased insulin secretion by approximately 20% in Fam151a^βKO mice, compared with nearly 300% in control littermates (Fig. 4E). By contrast, high-glucose induced insulin secretion was reduced by only around 30%–40% in in vitro and ex vivo assays (Figs. 3E and 4F). This discrepancy likely reflects context dependent and systemic factors unique to the in vivo setting, where islets mount a coordinated physiological response to elevated blood glucose levels.

Our metabolomic analysis in INS-1E cells identified certain key metabolites downregulated by Fam151a deficiency, including PRPP, inosine, and IMP, which are pivotal nodes in the PPP and de novo purine synthesis, this finding was confirmed by pathway enrichment analysis (Fig. 6B-E). Consistently, inosine has been reported to potentiate insulin secretion from mouse pancreatic β-cells via a dual mechanism involving ion channel and nucleoside phosphorylase activities, with a clear dose-dependent effect [37-39]. The level of IMP is governed by IMP dehydrogenase (IMPDH) acting as a molecular switch, thereby controlling GTP-dependent insulin secretion [40,41]. Although no studies have reported that PRPP directly stimulates insulin secretion, it functions as an essential precursor for IMP and inosine synthesis. As a key metabolite in the PPP, PRPP drives de novo purine synthesis, potentially influencing insulin secretion through its downstream metabolites [42]. Importantly, glucose stimulation elevates adenosylsuccinate (S-AMP), a purine pathway intermediate that promotes insulin exocytosis via sentrin/SUMO-specific protease 1 (SENP1), thereby establishing a functional coupling between PPP-de novo purine synthesis and insulin granule release [43].

Structural analysis places ER-localized FAM151A within the GDPD/PLCD enzyme superfamily [18,19], implying potential phosphodiesterase activity for glycerophosphodiester or nucleoside diphosphate substrates, thereby controlling the concentration of a rate-limiting precursor for the PPP or directly modulating ATP/ADP balance. The loss of this enzymatic activity also disrupts glycolytic flux, a conclusion supported by a concomitant decrease in all glycolytic metabolites upstream of 3-phosphoglycerate (Fig. 6D). Functionally, Fam151a^βKO mice exhibited a dysfunction in the stimulus-secretion coupling cascade: Fam151a deficiency in β-cells→metabolic impairment→reduced ATP/ADP ratio→attenuated Ca²⁺ transients and inadequate fueling of ATP-dependent exocytosis→defective insulin granule exocytosis in response to both metabolic (glucose) and non-metabolic (KCl) secretagogues (Fig. 6G).

However, a few limitations should be noted: (1) Although sequence conservation analysis classifies FAM151A as a member of the GDPD/PLCD enzyme family, its specific physiological substrates and phosphodiester bond hydrolysis activity require direct experimental validation. The specific functions of the two DUF2181 domains in FAM151A remain to be elucidated, as does whether restoration of FAM151A or the DUF2181 domain in diabetic animals alleviates diabetic phenotypes. It is also unknown whether FAM151B, a homologous family member, shares similar functional roles [18,19]. (2) Although comprehensive analyses of databases, such as the Human Protein Atlas, indicate low Fam151a expression levels in human pancreatic islets compared to mouse databases, and large-scale diabetes genetic databases (e.g., T2DM Knowledge Portal) currently lack direct associations or expression data for FAM151A in human T2DM cohorts, future research is therefore warranted to clarify the translational potential of modulating FAM151A for diabetes prevention and β-cell-targeted therapy and to enhance the clinical relevance of our findings [15]. (3) FAM151A is highly expressed in the kidney and localizes to the ER membrane. It has been reported that deficiency of aquaporin 11 (AQP11), a key membrane-associated protein (mainly located in the ER) in autosomal dominant polycystic kidney disease (ADPKD), leading to downregulation of Fam151a in the kidney [44]. Therefore, it is imperative to investigate the potential roles of FAM151A in the renal system.

Collectively, our study demonstrates FAM151A as a novel regulator of β-cell function and glucose homeostasis in T2DM mice. Integrated metabolic and metabolomic analyses demonstrate that loss of FAM151A disrupts key metabolic pathways including glycolysis, the PPP, purine synthesis pathway, and the ATP/ADP ratio in pancreatic β-cells (Fig. 6G). This metabolic impairment likely underlies the observed β-cell dysfunction and defective insulin granule secretion, ultimately leading to intracellular insulin accumulation in Fam151a-deleted β-cells. These findings position FAM151A as a potential target for therapeutic strategies aimed at preserving pancreatic β-cell function in T2DM.

SUPPLEMENTARY MATERIALS

NOTES

CONFLICTS OF INTEREST

No potential conflict of interest relevant to this article was reported.

AUTHOR CONTRIBUTIONS

Conception or design: J.Z., Y.C.

Acquisition, analysis, or interpretation of data: J.Z., C.L., S.Z. (Shuaishuai Zhu), J.W., X.Z., S.Z. (Shixuan Zhuo), Y.Z., Z.M.

Drafting the work or revising: all authors.

Final approval of the manuscript: all authors.

FUNDING

This study was funded by Chronic Non-Communicable Diseases- National Science and Technology Major Project of China (2404ZD0531300 to Yan Chen), National Natural Science Foundation of China (32230047 to Yan Chen), the National Key Research and Development Program of China (2023YFA 1801100 to Yan Chen), and Shanghai Municipal Science and Technology Major Project (to Yan Chen).

ACKNOWLEDGMENTS

The authors thank all the staff in the institutional animal facility of the Shanghai Institute of Nutrition and Health (SINH), Chinese Academy of Sciences (CAS), for the careful management of the experimental animals. We thank all the staff in Mass Spectrometry Analysis and Mass Spectrometry Imaging Technology Platform in SINH, CAS for the metabolomics experiments. We thank Hongxiang Gao from Institutional Center for shared Technologies and Facilities in SINH, CAS for assisting islet perifusion studies. We thank Professor Huiyong Yin from City University of Hong Kong and his student Huimin Jiang for the high-fat/high-sucrose islet samples.

DATA AVAILABILITY

Fig. 1.

Family with sequence similarity 151 member A (Fam151a) expression was downregulated in β-cells of type 2 diabetes mellitus mice and in high-glucose treated β-cells. (A) Workflow for single-cell RNA sequencing (scRNA-seq) analysis of mouse islets. Fresh pancreatic islets were isolated from 10-week-old male C57BL/6 mice (wild type [WT]), 10-week-old early db/db mice (early db/db), and 14-week-old db/db mice (late db/db). Islets from three mice per group were pooled for scRNA-seq analysis. (B) A total of 22,749 cells were projected into two-dimensional space using uniform manifold approximation and projection (UMAP) based on differential gene expression. Each cell cluster is annotated with its assigned cell type on the right. (C) Violin plot showing relative Fam151a expression levels across the cell clusters. (D) T-distributed stochastic neighbor embedding (t-SNE) visualization of β-cell subpopulations of three origins (left panel) and t-SNE visualization of Fam151a expression (right panel). (E) t-SNE plots for the markers of mature and dedifferentiated β-cells. (F, G) The mRNA levels of Fam151a in islets from control mice and db/db mice (F) and from mice fed a high-fat/high-sucrose diet (HF/HS) (G) (n=4 for each group). (H) The mRNA levels of Fam151a in islets from control mice and high-fat diet (HFD) mice (n=6 for each group). (I) The mRNA level of Fam151a in INS- 1E cells cultured under high-glucose (30 mM) conditions for 1 week (n=4 for each group). All data are presented as the mean±standard error of the mean. ^aP<0.05, ^bP<0.01, ^cP<0.001.

Fig. 2.

Family with sequence similarity 151 member A (FAM151A) is a single-pass transmembrane protein localized to the endoplasmic reticulum. (A) Transmembrane topology of FAM151A protein predicted using the transmembrane helix hidden Markov model (TMHMM). (B) Western blotting analysis of all cellular membrane and cytosolic fractions from HEK293T cells transfected with pCDH and pCDH-Fam151a-3xFlag for 48 hours. Quantitation results of relative Fam151a-3xFlag protein levels are shown in the right panel (shown as the mean±standard error of the mean). (C) Subcellular localization of FAM151A in HeLa cells. Representative immunofluorescence images of HeLa cells transfected with pLVXm-Myc-Fam151a or pCDH-Fam151a-3xFlag for 48 hours, respectively (scale bar=5 μm). The right panel displays the fluorescence intensity profile analysis for each channel. (D) Subcellular localization of FAM151A in MIN6 cells. Representative immunofluorescence images of MIN6 cells transfected with pCDH-Fam151a- 3xFlag for 48 hours (scale bar=2 μm). The right panel displays the fluorescence intensity profile analysis for each channel. FOV, Flag-overexpressing; GAPDH, glyceraldehyde 3-phosphate dehydrogenase; DAPI, 4ʹ,6-diamidino-2-phenylindole. ^aP<0.001.

Fig. 3.

Family with sequence similarity 151 member A (Fam151a) deletion in INS-1E cells leads to insulin granule retention and impairment in metabolic activity. (A) Western blotting analysis of control (CTRL) and Fam151a-knockout (FKO) in INS-1E cells. Quantitation of FAM151A protein levels normalized to β-actin is shown in the right panel. (B) Western blotting analysis of insulin and proinsulin in CTRL and FKO cells. Quantitation results of INSULIN/β-actin and PROINSULIN/β-actin are shown in the right panel. (C) Levels of total intracellular insulin content in CTRL and FKO cells as measured by enzyme-linked immunosorbent assay (n=4 for each group). (D) Representative transmission electron microscopy (TEM) image of the cells. Red arrows indicate insulin vesicles (scale bar=0.5 μm). (E) Glucose-stimulated insulin secretion and KCl-stimulated insulin secretion of the cells (n=4 for each group). (F) Oxygen consumption rate (OCR) results of the cells. (G) Extracellular acidification rate (ECAR) results of the cells. The bar charts in the right panels of (F) and (G) are quantitative representations derived from the left diagrams. All data are presented as mean±standard error of the mean. FCCP, carbonyl cyanide-4-(trifluoromethoxy)phenylhydrazone; ATP, adenosine triphosphate; 2-DG, 2-deoxyglucose. ^aP<0.05, ^bP<0.01, ^cP<0.001.

Fig. 4.

β-Cell-specific Fam151a-knockout (Fam151a^βKO) mice exhibit intracellular insulin accumulation and secretory dysfunction. (A) Western blotting analysis of Fam151a ablation in islets but not in other tissues. (B) Body weight and fasting blood glucose levels of the mice. (C) Glucose tolerance test (GTT) of the mice at the 9th week. Mice were fasted for 16 hours before the test (n=8 per group). Quantification of the area under the curve (AUC) is shown on the right. (D) Insulin tolerance test (ITT) of the mice at the 10th week. The mice were fasted for 6 hours before the test (n=8 per group). Blood glucose levels at each time point were normalized to the initial measurement (0 minute). AUC is shown on the right. (E) Glucose-stimulated insulin secretion (GSIS) of the mice at the 11th week; n=8 mice per group. (F) Perifusion analysis of dynamic GSIS (left panel) and biphasic insulin release levels (right panel) in islets from the mice at the 12th week. Each group of islets was pooled from at least three mice. (G) Representative transmission electron microscopic images of β-cells of the mice after treatment with 16.7 mM glucose for 1 hour. The red dashed line demarcates the cell periphery (scale bar=1 μm). (H) Quantification of different types of insulin granules (mature granules, immature granules and empty granules) per cell (left panel) and frequency distribution of the minimum distance from each vesicle to the plasma membrane (right panel). (I) Glucose-stimulated (16.7 mM) Ca²⁺ influx in islets from the mice (n=28 islets per group). Each group of islets was pooled from at least three mice. (J) Adenosine triphosphate (ATP) levels in islets from the mice under low-glucose (2.8 mM) or high-glucose (16.7 mM) treatments. (K) Western blotting analysis of insulin and proinsulin in isolated islets. Quantitation of insulin and proinsulin protein levels normalized to heat shock protein 90 (HSP90) is shown in the right panel. All data are presented as the mean±standard error of the mean. WT, wild type; ^aP<0.05, ^bP<0.01, ^cP<0.001.

Fig. 5.

β-Cell-specific Fam151a-knockout (Fam151a^βKO) mice preserve normal islet morphology and β-cell identity. (A) Immunohistochemical images of pancreatic sections from Fam151a^WT and Fam151a^βKO mice stained with anti-insulin antibody (scale bar=500 μm). The right panel shows the quantitative analysis of islet number and area per tissue section. (B) Immunofluorescence staining of pancreas sections of the mice for insulin (green), glucagon (red), and nuclei (4ʹ,6-diamidino-2-phenylindole [DAPI], blue) (scale bar=50 μm). The right panel shows the quantitative analysis of mean fluorescence intensity (MFI) of insulin and glucagon. (C, D) Immunofluorescence staining of pancreas sections for insulin (green), Mafa (red), Pdx1 (red), and nuclei (DAPI, blue), as indicated (scale bar=50 μm). Quantitative analysis of Mafa⁺/Ins⁺ and Pdx1⁺/Ins⁺ double-positive cells per islet is shown in the right panels.

Fig. 6.

Family with sequence similarity 151 member A (Fam151a) deletion in INS-1E cells disrupts pentose phosphate and purine metabolic pathways. (A) Partial least squares discriminant analysis (PLS-DA) score plot showing separation of metabolomic profiles between control (CTRL) and Fam151a-knockout (FKO) INS-1E cell groups as described in Fig. 3. Each point represents an individual sample (n=3 per group). (B) Volcano plot of differentially expressed metabolites. Metabolites with significant differential abundance between the two groups of cells are highlighted in red (upregulated) and blue (downregulated). (C) Pathway enrichment analysis derived from differentially expressed metabolites. (D, E, F) Heatmap showing the relative abundance of key metabolites in glycolysis, the pentose phosphate pathway, and purine metabolism respectively. The color intensity correlates with absolute metabolite levels, with red and green representing significantly upregulated and downregulated metabolites. (G) A diagram illustrating the metabolic and functional changes caused by Fam151a deletion in pancreatic β-cells. IMP, inosine monophosphate; FC, fold change; FDR, false discovery rate; GAP, glyceraldehyde-3-phosphate; DAP, dihydroxyacetone phosphate; 3-PGA, 3-phosphoglycerate; PEP, phosphoenolpyruvate; NADP, nicotinamide adenine dinucleotide phosphate; NADPH, nicotinamide adenine dinucleotide phosphate (reduced form); PRPP, phosphoribosyl pyrophosphate; GMP, guanosine monophosphate; ATP, adenosine triphosphate; GTP, guanosine triphosphate; ADP, adenosine diphosphate; GDP, guanosine diphosphate; CoA, coenzyme A; OAA, oxaloacetic acid; TCA, tricarboxylic acid cycle; α-KG, α-ketoglutarate; ER, endoplasmic reticulum.

REFERENCES

• 1. IDF global clinical practice recommendations for managing type 2 diabetes 2025. Diabetes Res Clin Pract 2025;224:112238.ArticlePubMed
• 2. Defronzo RA. Banting lecture: from the triumvirate to the ominous octet: a new paradigm for the treatment of type 2 diabetes mellitus. Diabetes 2009;58:773-95.PubMedPMC
• 3. Weir GC, Bonner-Weir S. Five stages of evolving beta-cell dysfunction during progression to diabetes. Diabetes 2004;53 Suppl 3:S16-21.ArticlePubMedPDF
• 4. Lytrivi M, Castell AL, Poitout V, Cnop M. Recent insights into mechanisms of β-cell lipo- and glucolipotoxicity in type 2 diabetes. J Mol Biol 2020;432:1514-34.ArticlePubMedPMC
• 5. Ying W, Lee YS, Dong Y, Seidman JS, Yang M, Isaac R, et al. Expansion of islet-resident macrophages leads to inflammation affecting β cell proliferation and function in obesity. Cell Metab 2019;29:457-74.e5.ArticlePubMedPMC
• 6. Donath MY, Boni-Schnetzler M, Ellingsgaard H, Halban PA, Ehses JA. Cytokine production by islets in health and diabetes: cellular origin, regulation and function. Trends Endocrinol Metab 2010;21:261-7.ArticlePubMed
• 7. Kim HE, Choi SE, Lee SJ, Lee JH, Lee YJ, Kang SS, et al. Tumour necrosis factor-alpha-induced glucose-stimulated insulin secretion inhibition in INS-1 cells is ascribed to a reduction of the glucose-stimulated Ca²⁺ influx. J Endocrinol 2008;198:549-60.PubMed
• 8. Chen L, Lin Y, Zhu X, Zhuo S, Li Z, Guo C, et al. MCT1-mediated lactate shuttle to mitochondria governs macrophage polarization and modulates glucose homeostasis by affecting β cells. Adv Sci (Weinh) 2025;12:e14760.ArticlePubMedPMCPDF
• 9. van Diepen JA, Robben JH, Hooiveld GJ, Carmone C, Alsady M, Boutens L, et al. SUCNR1-mediated chemotaxis of macrophages aggravates obesity-induced inflammation and diabetes. Diabetologia 2017;60:1304-13.ArticlePubMedPMCPDF
• 10. Lau A, Ramanathan S, Poussier P. Excessive production of nitric oxide by macrophages from DP-BB rats is secondary to the T-lymphopenic state of these animals. Diabetes 1998;47:197-205.ArticlePubMed
• 11. Vakilian M, Tahamtani Y, Ghaedi K. A review on insulin trafficking and exocytosis. Gene 2019;706:52-61.ArticlePubMed
• 12. Campbell JE, Newgard CB. Mechanisms controlling pancreatic islet cell function in insulin secretion. Nat Rev Mol Cell Biol 2021;22:142-58.ArticlePubMedPMCPDF
• 13. Merrins MJ, Corkey BE, Kibbey RG, Prentki M. Metabolic cycles and signals for insulin secretion. Cell Metab 2022;34:947-68.ArticlePubMedPMC
• 14. Lewandowski SL, Cardone RL, Foster HR, Ho T, Potapenko E, Poudel C, et al. Pyruvate kinase controls signal strength in the insulin secretory pathway. Cell Metab 2020;32:736-50.e5.ArticlePubMedPMC
• 15. Xin Y, Kim J, Okamoto H, Ni M, Wei Y, Adler C, et al. RNA sequencing of single human islet cells reveals type 2 diabetes genes. Cell Metab 2016;24:608-15.ArticlePubMed
• 16. Qiu WL, Zhang YW, Feng Y, Li LC, Yang L, Xu CR. Deciphering pancreatic islet β cell and α cell maturation pathways and characteristic features at the single-cell level. Cell Metab 2017;25:1194-205.e4.ArticlePubMed
• 17. Wang YJ, Kaestner KH. Single-cell RNA-seq of the pancreatic islets: a promise not yet fulfilled? Cell Metab 2019;29:539-44.ArticlePubMedPMC
• 18. Salzberg Y, Diaz-Balzac CA, Ramirez-Suarez NJ, Attreed M, Tecle E, Desbois M, et al. Skin-derived cues control arborization of sensory dendrites in Caenorhabditis elegans. Cell 2013;155:308-20.ArticlePubMedPMC
• 19. Findlay AS, McKie L, Keighren M, Clementson-Mobbs S, Sanchez-Pulido L, Wells S, et al. Fam151b, the mouse homologue of C.elegans menorin gene, is essential for retinal function. Sci Rep 2020;10:437.ArticlePubMedPMCPDF
• 20. Li C, Wu J, Zhao Y, Zhu J, Zhu X, Chen Y, et al. Identification of amino acids that regulate angiogenesis and alter pathogenesis of a mouse model of choroidal neovascularization. Nutrients 2025;17:3006.ArticlePubMedPMC
• 21. Qiu X, Wang RR, Wu QQ, Fu H, Zhu S, Chen W, et al. BAF60a-dependent chromatin remodeling preserves β cell function and contributes to the therapeutic benefits of GLP-1R agonists. J Clin Invest 2025;135:e177980.ArticlePubMedPMC
• 22. Li Z, Lin Y, Zhuo S, Chen J, Chen L, Wang S, et al. PAQR9 regulates glucose homeostasis in diabetic mice and modulates insulin secretion in β cells in vitro under stress conditions. Mol Cell Endocrinol 2023;575:112032.ArticlePubMed
• 23. Li C, Zhu J, Wei S, Ye X, Yang L, Wang Z, et al. Intermittent protein restriction improves glucose homeostasis in Zucker diabetic fatty rats and single-cell sequencing reveals distinct changes in β cells. J Nutr Biochem 2023;114:109275.ArticlePubMed
• 24. Wei S, Li C, Luo X, Yang L, Yu L, Wang Q, et al. Intermittent protein restriction protects islet β cells and improves glucose homeostasis in diabetic mice. Sci Bull (Beijing) 2022;67:733-47.ArticlePubMed
• 25. Ono Y, Kataoka K. MafA, NeuroD1, and HNF1β synergistically activate the Slc2a2 (Glut2) gene in β-cells. J Mol Endocrinol 2021;67:71-82.ArticlePubMed
• 26. Kim-Muller JY, Fan J, Kim YJ, Lee SA, Ishida E, Blaner WS, et al. Aldehyde dehydrogenase 1a3 defines a subset of failing pancreatic β cells in diabetic mice. Nat Commun 2016;7:12631.ArticlePubMedPMCPDF
• 27. Sonkar K, Ayyappan V, Tressler CM, Adelaja O, Cai R, Cheng M, et al. Focus on the glycerophosphocholine pathway in choline phospholipid metabolism of cancer. NMR Biomed 2019;32:e4112.ArticlePubMedPMCPDF
• 28. DiGruccio MR, Mawla AM, Donaldson CJ, Noguchi GM, Vaughan J, Cowing-Zitron C, et al. Comprehensive alpha, beta and delta cell transcriptomes reveal that ghrelin selectively activates delta cells and promotes somatostatin release from pancreatic islets. Mol Metab 2016;5:449-58.ArticlePubMedPMC
• 29. Ahmed F. Deciphering the gene regulatory network associated with anti-apoptosis in the pancreatic islets of type 2 diabetes mice using computational approaches. AIMS Bioeng 2023;10:111-40.Article
• 30. Wang T, Bonnavion R, Piesker J, Gunther S, Wettschureck N. GPRC5B preserves a mature β cell state in obesity by controlling MafA expression. JCI Insight 2025;10:e194115.ArticlePubMedPMC
• 31. Ediger BN, Du A, Liu J, Hunter CS, Walp ER, Schug J, et al. Islet-1 is essential for pancreatic β-cell function. Diabetes 2014;63:4206-17.ArticlePubMedPMCPDF
• 32. Miranda MA, Macias-Velasco JF, Schmidt H, Lawson HA. Integrated transcriptomics contrasts fatty acid metabolism with hypoxia response in β-cell subpopulations associated with glycemic control. BMC Genomics 2023;24:156.ArticlePubMedPMCPDF
• 33. Miranda MA, Carson C, St Pierre CL, Macias-Velasco JF, Hughes JW, Kunzmann M, et al. Spontaneous restoration of functional β-cell mass in obese SM/J mice. Physiol Rep 2020;8:e14573.ArticlePubMedPMCPDF
• 34. Zhang Z, Gao Y, Meng ZX. Transcriptional control of pancreatic β-cell identity and plasticity during the pathogenesis of type 2 diabetes. J Genet Genomics 2022;49:316-28.ArticlePubMed
• 35. Liu C, Xu S, Liu Y, Lu Z, Yang J. Mito-fission gene prognostic model for colorectal cancer. PeerJ 2025;13:e19522.ArticlePubMedPMCPDF
• 36. Eto K, Suga S, Wakui M, Tsubamoto Y, Terauchi Y, Taka J, et al. NADH shuttle system regulates K(ATP) channel-dependent pathway and steps distal to cytosolic Ca(2+) concentration elevation in glucose-induced insulin secretion. J Biol Chem 1999;274:25386-92.ArticlePubMed
• 37. Capito K, Hedeskov CJ. Inosine-stimulated insulin release and metabolism of inosine in isolated mouse pancreatic islets. Biochem J 1976;158:335-40.ArticlePubMedPMCPDF
• 38. Campbell IL, Taylor KW. A species difference in nucleoside phosphorylase activity and inosine-stimulated insulin secretion in isolated islets of Langerhans. Biochem J 1977;168:591-3.ArticlePubMedPMCPDF
• 39. Bozem M, Garrino MG, Henquin JC. Inosine partially mimics the effects of glucose on ionic fluxes, electrical activity, and insulin release in mouse pancreatic B-cells. Pflugers Arch 1987;410:457-63.ArticlePubMedPDF
• 40. Metz SA, Kowluru A. Inosine monophosphate dehydrogenase: a molecular switch integrating pleiotropic GTP-dependent beta-cell functions. Proc Assoc Am Physicians 1999;111:335-46.PubMed
• 41. Metz S, Holland S, Johnson L, Espling E, Rabaglia M, Segu V, et al. Inosine-5’-monophosphate dehydrogenase is required for mitogenic competence of transformed pancreatic beta cells. Endocrinology 2001;142:193-204.ArticlePubMed
• 42. Hove-Jensen B, Andersen KR, Kilstrup M, Martinussen J, Switzer RL, Willemoes M. Phosphoribosyl diphosphate (PRPP): biosynthesis, enzymology, utilization, and metabolic significance. Microbiol Mol Biol Rev 2017;81:e00040-16.ArticlePubMedPMCPDF
• 43. Gooding JR, Jensen MV, Dai X, Wenner BR, Lu D, Arumugam R, et al. Adenylosuccinate is an insulin secretagogue derived from glucose-induced purine metabolism. Cell Rep 2015;13:157-67.ArticlePubMedPMC
• 44. Saito T, Tanaka Y, Morishita Y, Ishibashi K. Proteomic analysis of AQP11-null kidney: proximal tubular type polycystic kidney disease. Biochem Biophys Rep 2018;13:17-21.ArticlePubMedPMC

Figure & Data

References

Citations

Citations to this article as recorded by  

About this article

Zhu J, Li C, Zhu S, Wu J, Zhu X, Zhuo S, Zhang Y, Meng Z, Chen Y. FAM151A Regulates Insulin Secretion in Pancreatic β-Cells and Is Implicated in Pathogenesis of Type 2 Diabetes Mellitus. Diabetes Metab J. 2026 Aug 6. doi: 10.4093/dmj.2025.1061. Epub ahead of print.

Received: Oct 22, 2025; Accepted: Mar 05, 2026

DOI: https://doi.org/10.4093/dmj.2025.1061.

Download citation ↓