Peptidome analysis reveals the involvement of endogenous peptides in mouse pancreatic dysfunction with aging

Abstract

Type 2 diabetes (T2D) is a glucose regulation disorder that has significantly enhanced mortality and the global disease burden. The prevalence of T2D has increased worldwide and is higher in the elderly. The function of pancreatic islets decreases with age, which is one important reason for the occurrence of diabetes in the elderly. Recently, peptidome analysis has attracted attention. However, the role of age‐related peptides in pancreatic dysfunction has not been investigated extensively. Here, we conducted a comparison of endogenous peptides between pancreas from adult and aging mice by liquid chromatography tandem mass spectrometry (LC‐MS/MS). A total of 2,089 peptides originating from 1,280 protein precursors were identified, of which 232 were upregulated and 183 were downregulated in the aging mice (fold change ≥ 2 and p < 0.05), suggesting that the expression of pancreatic peptides in mice varied with age. The molecular weight of most peptides was <3.0 kDa, and the isoelectric point distribution had a bimodal characteristic. Further analysis of cleavage site patterns indicated that proteases cleaved pancreatic proteins according to their rules. Moreover, Gene Ontology and pathway analyses showed that the differentially expressed peptides potentially had specific effects on pancreatic dysfunction. Some differential peptides were located within the domains of precursor proteins that were closely associated with the development of diabetes. We believe that our research may advance the current understanding of pancreas‐derived peptides and that certain peptides may be involved in the etiology of diabetes.

• 1.Differences in the pancreatic peptidomes between adult and aging mice were analyzed for the first time.
• 2.The differentially expressed peptides reveal potential contributions to pancreatic dysfunction.
• 3.Some peptides are predicted to play a bioactive role in the development of diabetes, although these results must be verified by functional investigations.

1. INTRODUCTION

Type 2 diabetes (T2D) is a common metabolic disease that is characterized by hyperglycemia. T2D affects multiple organ systems and has become a serious healthcare problem in our society. The prevalence of T2D has been increasing in recent decades. In 2013, the global number of diabetic patients was 382 million, and this figure is predicted to ascend to 592 million by 2035 (Guariguata et al., 2014). T2D is associated with significantly increased mortality from multiple diseases, including diabetic ketoacidosis or coma, chronic kidney disease, ischemic heart disease, stroke, infection, and certain cancers (Bragg et al., 2017). Therefore, effective strategies must be identified for the prevention and treatment of diabetes.

Previous studies showed that the prevalence of T2D in the elderly was much higher than that in the total adult population (20.2% vs. 10.9%; Wang et al., 2017) and that the absolute risk of cardiovascular disease was increased in the older population compared with that in younger persons (Orces & Lorenzo, 2017). Thus, aging is an important risk factor for T2D. Among these aging‐related contributors, deterioration of pancreatic function with increased age is a remarkable enabling factor for impaired glucose metabolism. Recently, several researchers have paid increasing attention to the effects of aging on the pancreas itself, especially on insulin secretion, the β‐cell mass, and the proliferative or regenerative capacity of β cells (Kalyani & Egan, 2013; Kushner, 2013). Therefore, exploring the mechanism of T2D from the perspective of pancreatic dysfunction will provide new targets for the prevention and treatment of T2D, especially in the elderly.

Endogenous peptides are a type of bioactive substances that can regulate many genes, produce a functional effect, and participate in various disease pathways. Recently, peptides have gained increasing interest as therapeutics for diabetes. For instance, glucagon‐like peptide 1 (GLP‐1) was demonstrated to be a potent insulinotropic peptide and became an effective therapeutic strategy for many subjects with T2D (Cho, Fujita, & Kieffer, 2014). Amylin, which is a peptide hormone produced in the pancreas, has well‐established physiological roles in glycemic regulation (Mietlicki‐Baase, 2016). Moreover, other endogenous peptides, such as leptin (Xu et al., 2018), ghrelin (Yada et al., 2014), cholecystokinin (Rehfeld, 2016), and the neuropeptide Y family of peptides (Khan, Vasu, Moffett, Irwin, & Flatt, 2017), are promising peptides in antidiabetes treatment. These findings inspired our interest in studying endogenous peptides associated with diabetes. However, little is known about the age‐related peptidome associated with pancreatic dysfunction and diabetes at present.

In this study, liquid chromatography tandem mass spectrometry (LC‐MS/MS) was used to compare endogenous peptides derived from the pancreas of adult and aging mice. The basic features and cleavage sites of the differentially expressed peptides were also analyzed. In addition, we characterized the differential peptides using Gene Ontology (GO) and pathway analyses and explored potential bioactive peptides involved in diabetes. The data may help clarify the underlying mechanism of age‐related pancreatic dysfunction from a peptidomic perspective and find novel peptides for diabetes treatment.

2. MATERIALS AND METHODS

2.1. Animal experiment and sample preparation

Male C57BL/6 mice aged 8 weeks (n = 5) and 44 weeks (n = 5) were purchased from the Cavens Lab Animal Co., Ltd. (Changzhou, China). The mice were maintained with a 12‐hr light:12‐hr darkness cycle and were fed standard laboratory diet freely. After 1 week of acclimatization, glucose tolerance test (GTT) and insulin tolerance test (ITT) were performed. For the GTT, the mice were fasted for 16 hr, and the fasting blood glucose was tested before an intraperitoneal injection of glucose (2.0 g/kg). After administration, the blood glucose concentration was measured at 30, 60, and 120 min. For the ITT, after a 4‐hr fast, the blood glucose levels were measured before and 30, 60, 90, and 120 min after an intraperitoneal injection of insulin (0.75 U/kg). Blood samples were taken from a small incision at the end of tail to measure blood glucose using a glucometer (Roche, Basel, Switzerland). Finally, mice that had fasted for 12 hr were killed by cervical dislocation. Blood was obtained from the left ventricle and centrifuged, and sera were stored at −80°C for the subsequent analyses. The serum glucose, insulin, and C‐peptide levels were determined using ELISA kits (all from Millipore, Billerica, MA). The pancreatic tissue was harvested, washed, and trimmed of fat. The tissue samples were washed with phosphate‐buffered saline supplemented with a protease inhibitor mixture (Roche) and rapidly frozen in liquid nitrogen, followed by storage at −80°C before analysis. All animal experiments were approved by the Institutional Animal Care and Use Committee (IACUC) of Nanjing Medical University (Approval Number: IACUC‐1605987).

2.2. Immunohistochemistry and analysis

Specimens from pancreatic tissue were fixed in 4% paraformaldehyde for 24 hr and embedded in paraffin. Sections of 6‐μm thickness were cut on a microtome (Finesse 325; Thermo Fisher Scientific, Waltham, MA). Subsequently, the sections were deparaffinized and submitted to antigen retrieval. Endogenous peroxidase activity was suppressed with a solution of 3.0% hydrogen peroxide in methanol. The sections were blocked with 3% bovine serum album for 30 min before incubated overnight at 4°C with rabbit anti‐insulin antibody (1:500; Santa Cruz Biotechnology, Santa Cruz, CA), followed by incubation with horseradish peroxidase‐conjugated goat anti‐rabbit immunoglobulin G (IgG; Sigma‐Aldrich, Poole, United Kingdom) at room temperature for 1 hr. Finally, insulin‐positive cells were stained by 3,3′‐diaminobenzidine to a brown color and then counterstained with hematoxylin. Digital images were captured of ×200 and ×400 using a digital slide scanner (3DHISTECH, Budapest, Hungary). The percentage of insulin‐positive cells to pancreatic tissue were analyzed via computer‐assisted software (3DHISTECH, QuantCenter software) by counting five sections per mouse in respective groups.

2.3. Peptides extraction and purification

Pancreatic tissue samples from adult and aging mice (n = 3) were homogenized in lysis buffer (Beyotime Biotechnology, Shanghai, China) mainly consisting of 1% Triton X‐100, 1% deoxycholate, and 0.1% sodium dodecyl sulfate, and then sonicated for 3 min in an ice water bath. The lysate was centrifuged at 12,000g for 3 min at 4°C, and the supernatant was collected. Subsequently, the samples were passed through centrifugal concentrators (Amicon Ultra‐15; Millipore) with a 10‐kDa molecular weight cutoff (MWCO). Next, the flow‐through was collected, and the peptides were purified using C18 columns (Acclaim PepMap 100 C18, 75 μm × 150 mm; Thermo Fisher Scientific). Finally, the supernatant was vacuum‐dried with the Speed‐Vac system (RVC 2‐18; Martin Christ, Osterode, Germany) and frozen to −80°C immediately.

2.4. Tandem mass tag (TMT) labeling and LC‐MS/MS analysis

The dried peptides were dissolved in 0.1% formic acid, filtered through a 0.45‐μm membrane, reduced with 10 mM dithiothreitol for 1 hr at 60°C, and alkylated with 55 mM iodoacetamide at room temperature for 45 min. Afterward, the peptides were desalted, lyophilized, and labeled with the TMT reagent (Thermo Fisher Scientific) according to the manufacturer’s indications. The TMT results were generated from analysis of isobaric tag combinations. The labeled samples were injected into the LTQ‐Orbitrap Velos mass spectrometer (Thermo Fisher Scientific). A full‐scan analysis was conducted in the m/z range from 600 to 5,000 at 2 spectra/s. For the MS analysis, 2,000 single excitation spectra were accumulated from 10 random positions of each sample, and 200 laser pulses were applied to each position. All acquired data were transformed and analyzed using the PEAKS software, V7.0 (Bioinformatics Solutions, Waterloo, Ontario, Canada).

2.5. Bioinformatics analysis

To calculate the molecular weight (MW) and isoelectric point (pI) of each differentially expressed peptide, the online pI/MW tool (http://web.expasy.org/compute_pi/) was used. Using the online tools SMART (http://smart.embl‐heidelberg.de/) and UniProt (http://www.uniprot.org/), we investigated whether the differential peptide sequences were located within the functional domains of their precursor proteins. Gene Ontology analysis (http://geneontology.org) was applied to explore potential physiological functions of the peptide precursors, including cellular components, molecular functions, and biological processes. In addition, the Kyoto Encyclopedia of Genes and Genomes (KEGG; http://www.kegg.jp/) was used to analyze the precursor proteins of the differential peptides and to gain an overview of the regulated pathways.

2.6. Statistical analysis

The experimental data were analyzed using Student’s paired t test and p < 0.05 was considered statistically significant. The threshold for screening differentially expressed peptides was a fold change ≥ 2.0. All data are expressed as the mean ± standard deviation in this study.

3. RESULTS

3.1. Pancreatic β‐cell function decreases with age

Compared with those of the adult mice, the aging mice exhibited increased fasting blood glucose and decreased insulin and C‐peptide levels (p < 0.05; Figure 1a–c). The GTT demonstrated that glucose tolerance was decreased in the aging mice because the blood glucose levels were higher throughout the glucose challenge and were significantly increased at 120 min (p < 0.05; Figure 1d). In the ITT, we found significantly enhanced blood glucose levels in the aging mice compared with those of the adult mice, especially at 30 and 120 min after insulin injection (p < 0.05; Figure 1e), which indicated that a setback of insulin sensitivity occurred in the aging mice. The areas under the curve (AUCs) for glucose in the GTT and ITT were significantly elevated in the aging group compared with those of the adult group (p < 0.05). To investigate whether there are any age‐related changes in the insulin‐positive cells in adult and aging mice, we stained pancreas sections with anti‐insulin antibody. Typical micrographs showed the distribution of insulin immunolabeling in islets from adult and aging mice. Quantitative analysis showed that the percentage of insulin‐positive cells was significantly reduced in aging mice compared with adult mice (p < 0.05; Figure 1f), suggesting that β cells gradually decreased with age.

Figure 1.

Pancreatic β‐cell function decreases with age. (a) The fasting blood glucose, (b) fasting insulin, and (c) fasting C‐peptide levels were compared between adult (8 weeks old, n = 5) and aging (44 weeks old, n = 5) male C57BL/6 mice. (d) The glucose tolerance test (GTT), and (e) insulin tolerance test (ITT) were performed after 16‐ and 4‐hr‐long fasts, respectively. The areas under the curve (AUCs) for glucose from the GTT and ITT were also calculated. (f) Micrographs showed the pattern and distribution of insulin‐immunoreactive cells in the pancreatic islets of adult and aging mice. Representative images are shown at ×200 and ×400 magnification, respectively. The percentage of insulin‐positive cells to the pancreas was calculated via computer‐assisted software. The data are presented as the mean ± SD, *p < 0.05 versus the adult group [Color figure can be viewed at wileyonlinelibrary.com]

3.2. Peptide identification

In total, 2,089 peptides originating from 1,280 precursor proteins were identified from the six samples (Supporting Information Table 1). Among them, 415 peptides originating from 346 precursor proteins showed significant differences (fold change ≥ 2 and p < 0.05), of which 232 peptides were upregulated and 183 were downregulated in the aging mice (Figure 2a). All of the differential peptides are listed in Supporting Information Table 2, and peptides with a fold change ≥ 5 are visualized by heat maps (Figure 2b,c).

Figure 2.

Differentially expressed peptides between adult and aging mice. (a) Of the 2,504 nonredundant peptides, 415 peptides were significantly differentially expressed in the aging mice (fold change ≥ 2 and p < 0.05), including 232 upregulated and 183 downregulated peptides. The degree of abundance of the (b) upregulated, and (c) downregulated peptides (fold change ≥ 5) is colored based on the heat map scale (upregulated: red, downregulated: blue). The data are visualized based on log₁₀ values [Color figure can be viewed at wileyonlinelibrary.com]

3.3. Features of the differentially expressed peptides

By analyzing the differentially expressed peptides in the mouse pancreas, we observed that both the MWs and pIs had broad distributions, although most of the peptides were distributed in the ranges of 0.5–1.3 kDa in MW and 5.0–6.0 and 8.0–9.0 in pI (Figure 3a,b). Because the amino acid composition and MW distribution contribute to the specific pI distribution, the MW versus pI distribution was also investigated (Figure 3c). A total of 17 overexpressed and 11 underexpressed peptides in the aging mouse pancreas that are located in the domains of their respective precursor proteins are listed in Table 1, including the peptide sequence, precursor protein, domain location, domain description, and fold change.

Figure 3.

Features of the differentially expressed peptides. (a) Molecular weights (MWs), (b) isoelectric points (pIs), and (c) scatter plots of MW versus pI of the differentially expressed peptides in the upregulated and downregulated groups. Distributions of the cleavage sites in the (d) upregulated, and (e) downregulated groups. The four cleavage sites are the C‐terminal amino acid of the preceding peptide, the N‐terminal and C‐terminal amino acids of the identified peptide and the N‐terminal amino acid of the subsequent peptide. (f) Peptide numbers as the same as those of the precursor protein [Color figure can be viewed at wileyonlinelibrary.com]

Table 1.

Differentially expressed peptides located in the functional domains of their precursor proteins

Peptide sequence	Precursor protein	Domain location	Domain description	Fold change
Upregulated peptides
KKTLDNDIMLI	TRY10	24–244	Peptidase S1	∞
KTSPF	SDK1	173–259	Ig‐like C2 type 2	6.35
TVYVI	NISCH	12–122	PX	5.00
QTVQNLTVPGG	TNC	1,167–1,259	Fibronectin type‐III 7	3.73
KPSIAAVVGSM	AGO3	518–819	Piwi	3.19
ILMPGG	CNTNAP3	207–345	Laminin G	3.18
NHAAVPREQVT	IL6ST	26–120	Ig‐like C2‐type	3.16
KLGDHLNSI	PTPN13	3–190	KIND	3.15
KRSPH	PDZD2	334–419	PDZ	2.95
SLGTLSDLS	KMT2B	1,733–1,789	FYR N‐terminal	2.59
KAGDIITVLE	CASKIN1	281–347	SH3	2.55
VEAMAVKKPG	ACAP1	266–362	PH	2.44
TQSDKVTLKGAK	HDLBP	800–863	KH 9	2.38
LAGHN	PHLPP1	895–914	LRR	2.36
FIGLN	IBTK	769–837	BTB 2	2.11
HALSVGPQTTT	COL12A1	336–426	Fibronectin type‐III 2	2.11
NRITAATHVH	STARD9	3–384	Kinesin motor	2.11
Downregulated peptides
RGGGRVI	PLXNB3	1,003–1,134	IPT/TIG 3	∞
HITKTT	ADCY1	257–455	CYCc	8.19
KSSLLS	ABCC1	644–868	ABC transporter 1	4.51
KNAAPI	AGK	58–199	DAGKc	3.84
HLFLPFSY	APOB	46–672	Vitellogenin	3.30
YGKVF	RPS6KB2	67–328	Protein kinase	2.88
HDAVVFLIEQL	ELN	3,663–3,744	FYR C‐terminal	2.85
QTNKVPH	MYO10	1,551–1,699	MyTH4	2.65
ISAAAAL	MAPK7	55–347	Protein kinase	2.13
FRQIQVV	NRK	1,138–1,425	CNH	2.05
HILAGNSPPLF	DCHS1	680–784	Cadherin 7	2.02

3.4. Cleavage site patterns of the differentially expressed peptides

Next, we searched for cleavage site distributions at the carboxyl terminal end (C‐terminal) and amino terminal end (N‐terminal) of each identified peptide in the upregulated and downregulated groups (Figure 3d,e). In both the upregulated and downregulated groups, lysine (K) and leucine (L) were the most abundant N‐ and C‐terminal amino acids of the identified peptides, respectively. The major C‐terminal amino acid of the preceding peptide and the N‐terminal amino acid of the subsequent peptide were valine (V) and proline (P) in the upregulated group but were glutamic acid (E) and asparagine (A) in the downregulated group, respectively. We detected several peptides from the same precursor protein. In Figure 3f, the top ten precursor proteins are listed; titin (TTN) yielded the highest number of related peptides.

3.5. GO and pathway analyses of the peptide precursors

In the biological process category, regulation of Rho protein signal transduction, regulation of membrane potential, cell adhesion, collagen fibril organization, transmembrane transport, regulation of microtubule depolymerization, regulation of cell migration, cellular response to DNA damage stimulus, endodermal cell differentiation, and calcium ion transmembrane transport were the dominant subcategories (Figure 4a). The cellular components included cytoplasm, collagen trimer, proteinaceous extracellular matrix, postsynaptic density, nucleoplasm, sarcolemma, basement membrane, cytoskeleton, microtubule, and microtubule‐associated complex (Figure 4b). For the molecular functions, ATP binding, ATPase activity, extracellular matrix (ECM) structural constituent, protein binding, nucleotide binding, Rho guanyl‐nucleotide exchange factor activity, poly(A) RNA binding, microtubule binding, actin binding, and ion channel activity were the most abundant subcategories (Figure 4c). The pathway analysis showed that the precursor proteins were involved in protein digestion and absorption, the ECM‐receptor interaction, ATP‐binding cassette (ABC) transporters, calcium signaling pathway, circadian entrainment, focal adhesion, lysine degradation, and PI3K‐Akt signaling pathway (Figure 4d).

Figure 4.

Gene Ontology (GO) and pathway analyses of the differentially expressed peptides based on analysis of their precursor proteins. The top ten GO analysis terms in the (a) biological process, (b) cellular component, and (c) molecular function categories between the adult and aging mice. (d) Differentially expressed precursor proteins were mapped to canonical pathways using the Kyoto Encyclopedia of Genes and Genomes tool

4. DISCUSSION

Over the past few decades, the number of patients with T2D has steadily risen. Although current agents can obtain a certain degree of hypoglycemic effect, they have various side effects and cannot prevent or delay pancreatic dysfunction. As mentioned above, some peptide drugs provide promising new regimes for diabetes treatment. Moreover, peptides can perform a wide range of functions, including stabilizing mitochondrial proteins, regulating signal transduction, maintaining intracellular calcium levels, and stimulating glucose uptake (Ferro, Rioli, Castro, & Fricker, 2014). Hence, we attempted to identify endogenous peptides to facilitate the elucidation of molecular mechanisms and the development of new therapeutics for diabetes.

The prevalence of T2D significantly increases with age. However, the underlying mechanisms are not completely understood. Insulin resistance has been considered to result from a reduced lean muscle mass, increased adiposity, and decreased physical activity, which contributes to glucose metabolic disorder (Scheen, 2005). However, insulin resistance alone has been demonstrated to not be sufficient to cause T2D. Pancreatic dysfunction is increasingly recognized as another key factor in abnormal glucose metabolism due to aging (De Tata, 2014). In accordance with previous studies, our study also detected elevated blood glucose and declining insulin levels as well as decreased insulin‐positive cells in aging mice, suggesting age‐related impairment of glucose metabolism and β‐cell function.

In the present study, we compared endogenous peptides between pancreas from adult and aging mice. We found that a 10‐kDa MWCO filter could remove redundant proteins from the mouse pancreas without compromising peptide recovery. Finally, we identified a total of 2,504 nonredundant peptides. More peptides were identified in our study than those in previous studies on endogenous pancreatic peptides (Budde et al., 2005; Minerva et al., 2011). The MWs of most of the identified peptides were less than 3.0 kDa, and the peptides were mainly distributed in the range of 0.5–1.0 kDa. This finding proved that the method used in our study was reliable for peptide extraction. The differential peptides had a bimodality feature of the pI distribution, with an almost blank space near the neutral pI. Some researchers have considered that this pattern is due to a lack of amino acids with neutral pH and is related to different subcellular locations (Wu et al., 2006).

A domain is a region with a specific sequence and structure in a protein that can exist, evolve and function independently of the rest of the protein chain. We identified 28 peptide sequences located in domains. Studies have shown that some precursor proteins play a role in diabetes. For example, adenylate cyclase 1 (ADCY1) plays a pivotal role in increasing β‐cell proliferation and mass as well as insulin secretion in both the MIN6 and INS‐1 pancreatic β‐cell lines (Kitaguchi, Oya, Wada, Tsuboi, & Miyawaki, 2013; Cromer et al., 2015). Interleukin‐6 receptor subunit β (IL6ST), which is also named glycoprotein 130, is required for stimulation of glucagon secretion, and α‐cell‐specific deletion of IL6ST results in protection from streptozotocin‐induced diabetes, suggesting that IL6ST may contribute to hyperglycemia by stimulating α cells to secrete glucagon (Chow et al., 2014; Keller et al., 2018). PDZ domain‐containing protein 2 (PDZD2) expression is plentiful in pancreatic β cells. PDZD2‐deficient mice showed increased glucose tolerance and basal insulin secretion after fasting (Tsang et al., 2010). However, confirmation experiments for these peptides are essential in future studies.

Proteins produce smaller fragments, such as peptides, through hydrolysis catalyzed by proteases. Proteases play a key regulatory role in the development and progression of disease, and different cleavage sites may reflect different protease activities (van den Berg & Tholey, 2012). We analyzed the cleavage sites of each differentially expressed peptide. The results showed that the cleavage site patterns of the mouse pancreatic peptidome differed from those of the serum and liver peptidomes (Hu et al., 2007; Villanueva et al., 2006). This discrepancy is not unexpected, because proteases vary in different body fluids and tissues. In our study, some peptides originated from the same precursor protein. TTN produced the most peptides, implying that this protein was most susceptible to cleavage by endogenous enzymes. We consider that this phenomenon is also mainly related to proteolytic processing.

Quantitative changes in peptide levels can reflect biological functions in vivo. Our results showed that 415 peptides, accounting for approximately 16.6% of the total peptides, exhibited significant changes in expression. Segmental functional peptides were reported to have biological functions related to their precursor proteins. The mitochondrial‐derived peptides MOTS‐c (Lee et al., 2015) and slit2 fragment (Svensson et al., 2016) play roles similar to those of their precursors, whereas peptide RT53 plays an opposite role to its precursor (Jagot‐Lacoussiere, Kotula, Villoutreix, Bruzzoni‐Giovanelli, & Poyet, 2016). Therefore, peptides can play the same or contrary functions as their protein precursors. As shown in Figure 2b, 18 peptides were expressed at a distinctly higher level and 35 peptides were expressed at significantly lower levels in the aging mouse pancreas (fold change ≥ 5). Many studies have discussed the relationship between these precursor proteins and diabetes. For example, ryanodine receptor 1 (RYR1) expression in pancreatic islets and β‐cell lines is essential for activating insulin secretion by inducing calcium release from secretory vesicles (Mitchell, Lai, & Rutter, 2003). E3 ubiquitin‐protein ligase (HUWE1) expression in the pancreas is crucial for the determination of the β‐cell mass; HUWE1‐deficient mice showed a decline in the β‐cell mass with age (Wang et al., 2014).

GO analysis showed that the biological processes of the differential peptides were tightly related to regulation of Rho protein signal transduction and membrane potential. Rho/Rho‐kinase activation is involved in pancreatic β‐cell dysfunction and inhibition of insulin biosynthesis (Nakamura et al., 2006). Pancreatic β cells are electrically excitable and are combined with alterations in the blood glucose levels via stimulation or suppression of insulin secretion by changes in the membrane potential (Rorsman & Braun, 2013). This result is in accordance with the molecular function analysis since the top terms “ATP binding” and “ATPase activity” are involved in the process of membrane potential changes induced by iron channel activation (Rorsman, Ramracheya, Rorsman, & Zhang, 2014). The calcium signaling pathway (Rutter et al., 2017) and phosphatidylinositol 3 kinase/protein kinase B (PI3K/Akt; Foukas & Withers, 2010) signaling pathway, which were identified in the pathway analysis, are essential for regulating the β‐cell mass and insulin secretion. Furthermore, one study showed that pancreatic regeneration in the aged pancreas was markedly attenuated, most likely due to decreased PI3K/Akt activation (Watanabe, Saito, Rychahou, Uchida, & Evers, 2005). These results suggest that these differential peptides participate in multiple diabetes‐related signaling pathways and may exert important regulatory effects.

The purpose of our study was to reveal the physiological age‐related endogenous peptide differences in the pancreas and explore the possible association with pancreatic dysfunction and diabetes. Besides, we hypothesized that pancreatic peptide in diabetic mice induced by high‐fat diet may also be different. In the future, we will do further research in this aspect.

In conclusion, using LC‐MS/MS technology, we investigated the peptide profiles in pancreas from adult and aging mice for the first time. A total of 415 peptides derived from 346 protein precursors were significantly differentially expressed (fold change > 2.0, p < 0.05). After analyzing the physical and chemical properties of these peptides, we applied a comprehensive method to analyze the functions of the differential peptides according to their precursor proteins. The results suggest that a few peptides may take part in pancreatic dysfunction and diabetes. Further studies are required to reveal the functions and mechanisms of these peptides and their correlations with diabetes.

CONFLICTS OF INTEREST

The authors declare that there are no conflicts of interest.

AUTHOR CONTRIBUTIONS

Y. H. and C. W. conceived of and designed the experiments; F. H. P. and X. H. performed the experiments; J. F., M. L. and L. G. analyzed the data; F. H. P., W. X. C. and H. Y. Y. performed the bioinformatics analysis; J. F. and Y. H. helped with the revision; and F. H. P. wrote the manuscript.

Supporting information

Supporting Information Table 1. The characteristics of all peptides identified by LC‐MS/MS

Supporting Information Table 2. Differentially expressed peptides between the adult and aging groups