Quantitative proteomics reveals novel interaction partners of Rac1 in pancreatic β-cells: Evidence for increased interaction with Rac1 under hyperglycemic conditions

Abstract

Rac1, a small G protein, regulates physiological insulin secretion from the pancreatic β-cell. Interestingly, Rac1 has also been implicated in the onset of metabolic dysfunction of the β-cell under the duress of hyperglycemia (HG). This study is aimed at the identification of interaction partners of Rac1 in β-cells under basal and HG conditions. Using co-immunoprecipitation and UPLC-ESI-MS/MS, we identified 324 Rac1 interaction partners in INS-1832/13cells, which represent the largest Rac1 interactome to date. Furthermore, we identified 27 interaction partners that exhibited increased association with Rac1 in β-cells exposed to HG. Western blotting (INS-1832/13 cells, rat islets and human islets) and co-immunoprecipitation (INS-1832/13 cells) further validated the identity of these Rac1 interaction partners including regulators of GPCR-G protein-effector coupling in the islet. These data form the basis for future investigations on contributory roles of these Rac1-specific signaling pathways in islet β-cell function in health and diabetes.

1. Introduction

Insulin secretion from the pancreatic β-cell is regulated principally by the extracellular concentration of glucose. However, the molecular and cellular mechanisms underlying the stimulus-secretion coupling of glucose-stimulated insulin secretion (GSIS) remain only partially understood. It is widely accepted that GSIS is mediated largely via the generation of soluble second messengers, such as cyclic nucleotides and hydrolytic products synthesized by phospholipases A₂, C and D (Jitrapakdee et al., 2010; Prentki et al., 2013; Berggren and Leibiger, 2006; Regazzi et al., 2016; Wang and Thurmond, 2009). The principal signaling cascade involves the glucose-transporter protein (i.e., Glut-2)-mediated entry of glucose into the β-cell resulting in an increase in the intracellular ATP/ADP ratio that is consequential to glucose metabolism via the glycolytic and the tricarboxylic acid cycle pathways. Such an increase in ATP levels culminates in the closure of membrane-associated ATP-sensitive potassium channels resulting in membrane depolarization followed by influx of the extracellular calcium through the voltage-gated calcium channels on the plasma membrane. A net increase in the intracellular calcium that occurs via the influx of extracellular calcium into the cytosolic fraction of the stimulated β-cell, in addition to the mobilization of calcium from the intracellular storage compartments, has been shown to play critical roles in GSIS (Jitrapakdee et al., 2010; Prentki et al., 2013; Berggren and Leibiger, 2006; Regazzi et al., 2016; Wang and Thurmond, 2009).

Multiple studies have provided convincing evidence to suggest that small G-proteins (e.g., Cdc42 and Rac1) play a significant regulatory role in cytoskeletal remodeling thereby favoring mobilization of secretory granules to the plasma membrane for fusion and release of their cargo into circulation. Published evidence also suggests novel regulatory roles for ADP-ribosylation factor 6 (Arf6) in insulin secretion from the islet β-cell (Kalwat and Thurmond, 2013; Kowluru, 2010, 2017). In this context, specific regulatory proteins/factors for G-proteins, namely guanine nucleotide exchange factors (GEFs; Tiam1, Vav2, β-PIX, Epac and ARNO) and guanine nucleotide dissociation inhibitors (GDIs; Rho GDIα, caveolin-1) have been identified and studied extensively in the islet β-cell (Wang and Thurmond, 2009; Kalwat and Thurmond, 2013; Kowluru, 2010, 2017; Jayaram et al., 2011). In further support of key regulatory roles for Rac1 in physiological insulin secretion in rodent and human islets (Kalwat and Thurmond, 2013; Kowluru, 2010, 2017) are the studies by Asahara et al. (2013) demonstrating impaired glucose tolerance and hypoinsulinemia in Rac1-null (βRac1^−/−) mice. Consistent with findings described above, only glucose-induced, but not KCl-induced, insulin secretion was inhibited significantly in islets from βRac1^−/− mice. The β-cell mass or islet density remained unchanged in these mice. siRNA-mediated knockdown of Rac1 in INS-1 cells also resulted in a significant defect in glucose-induced, but not KCl-induced, insulin secretion. Based on these findings, it was concluded that Rac1 plays a key regulatory role in insulin secretion primarily by regulating cytoskeletal organization (Asahara et al., 2013). In this context, Greiner et al. (2009) provided evidence to suggest that Rac1-null mice exhibited marked alterations in islet morphogenesis. Taken together, the above-described findings from multiple laboratories involving pharmacological and molecular biological tools as well as knockout animal models provide compelling evidence for novel regulatory roles for Rac1 in islet function, including GSIS (Wang and Thurmond, 2009; Kalwat and Thurmond, 2013; Kowluru, 2010, 2017; Jayaram et al., 2011; Asahara et al., 2013; Greiner et al., 2009).

It is noteworthy that, in addition to its positive modulatory role in insulin secretion, Rac1 has also been implicated in the metabolic dysregulation of the β-cell, specifically at the level of phagocyte-like NADPH oxidase (Nox2)-mediated generation of reactive oxygen species (ROS) thereby creating oxidative stress, mitochondrial dysfunction culminating in the functional abnormalities and eventual demise of the islet β-cell (Kowluru and Kowluru, 2014; Newsholme et al., 2009; Xiang et al., 2010). Data accrued from several recent investigations have implicated sustained activation of Rac1, which is seen under metabolic stress conditions (e.g., chronic hyperglycemia, lipotoxicity and exposure to biologically active sphingolipids like ceramide and proinflammatory cytokines), promotes activation of stress kinases (e.g., p38, JNK1/2 and p53) leading to β-cell dysfunction (Syed et al., 2010, 2011; Sidarala et al., 2015; Sidarala and Kowluru, 2017a, 2017b; Subasinghe et al., 2011; Kowluru and Kowluru, 2018). Together, these findings have led us to propose both “friendly” and “unfriendly” roles of Rac1 in islet β-cell function (Kowluru, 2011). Despite the aforestated evidence for critical regulatory roles of Rac1, potential identity and involvement of interaction between Rac1 and other proteins/factors in islet β-cells under hyperglycemic conditions remain an important, but understudied area of islet biology.

Co-immunoprecipitation (Co-IP) followed by mass spectrometry based proteomics has emerged as a powerful approach to map proteinprotein interactions (Marcilla and Albar, 2013; Wepf et al., 2009). However, most of these studies have utilized protein overexpression and/or epitope tagged bait proteins (Geetha et al., 2011). Recently, we developed a straightforward, label-free approach combining Co-IP with HPLC-ESI-MS/MS, without use of protein overexpression or protein tags, to investigate changes in the abundance of endogenous protein associate with a bait protein (Caruso et al., 2014). Earlier studies from our laboratory have optimized this label-free approach, and discovered novel IRS-1 interaction partners in skeletal muscle from healthy, obese non-diabetic and type 2 diabetic patients (Caruso et al., 2014). We have also used this approach to identify novel interaction partners of protein phosphatase 2A in insulin-secreting pancreatic β-cells (Zhang et al., 2016). In the current study, we used this proteomics approach to identify protein interaction partners of Rac1 in insulin-secreting INS-1832/13 cells exposed to basal or HG conditions. We also validated findings from proteomics approach by immunological approaches (Co-IP and Western blotting) in INS-1832/13 cells, normal rat islets and human islets. Our findings from these studies identified novel interaction partners for Rac1 under normal conditions. We also identified increased association of specific signaling proteins with Rac1 in INS-1832/13 cells exposed to HG conditions that may promote functional dysregulation and apoptosis in these cells.

2. Materials and methods

Antibodies directed against G protein-coupled receptor kinase-interacting protein 1 (GIT1; SC-365084), GIT1/2 (SC-135925), β-Pak–interacting exchange factor (β-PIX; SC-393184), Gαq11/14 (SC-365906) and Karyopherin-α2 (SC-55538) were from Santa Cruz Biotechnology, Inc. Anti-Rac1 GAP (A5298) was from Abclonal. Normal mouse IgG (No. 12–371) and Rac1 antibody (No. 05–389) were from Millipore. Co-Immunoprecipitation kit was obtained from Thermofisher Scientific.

2.1. Rat islets and human islets

Islets from normal male Sprague Dawley rats (~ 6 weeks old; Harlan Laboratories, Oxford, MI, USA) were isolated by the collagenase digestion method (Jayaram et al., 2011; Syed et al., 2011; Sidarala et al., 2015; Sidarala and Kowluru, 2017a). Human islets were obtained from PRODO Laboratories (Irvine, CA, USA). Islets from two human donors were used in these studies. The first donor was a 49-year old Hispanic male (64”, 179 lbs with a BMI of 30.3 and HbA1c of 5.8; Prodo Labs# HP-18196–01). The second donor was a 54-year old Caucasian male (71”, 160 lbs with a BMI of 22.3 and HbA1c of 5.8; Prodo Labs# HP-18164–01). Studies involving human islets were conducted according to the guidelines established by the US Department of Health and Human Services/NIH and were approved by the Research Service Biosafety Committee at the John D. Dingell VA Medical Center. All protocols employed in these studies were reviewed and approved by the Institutional Animal Care and Use Committee at Wayne State University and the Research Service Biosafety Committee at the John D. Dingell VA Medical Center.

2.2. INS-1832/13 cell culture and lysis

INS-1832/13 cells were kindly provided by Prof. Chris Newgard. INS-1832/13 cells were cultured in RPMI-1640 medium containing 10% FBS supplemented with 100 IU/ml penicillin and 100 IU/ml streptomycin, 1 mM sodium pyruvate, 50 μM 2-mercaptoethanol and 10 mM HEPES (pH 7.4). Prior to the treatment with basal glucose (LG; 2.5 mM) or high glucose (HG; 25 mM), cells were starved overnight in a low glucose/low serum growth medium (2.5 mM glucose and 2.5% FBS). The cells were then treated with growth medium containing 10% FBS with either 2.5 mM or 25 mM glucose for 24–48 h, and were homogenized in RIPA lysis buffer or IP lysis buffer containing protease inhibitors. Protein quantification was done as previously described (Zhang et al., 2016).

2.3. Identification of Rac1 interaction partners by Co-IP and HPLC-ESI-MS/MS

This is carried out according to the methods we published earlier (Caruso et al., 2014; Zhang et al., 2016) with modifications due to a newer mass spectrometry instrument, a Thermo Finnigan LTQ-Orbitrap Lumos, was used to generate UPLC-ESI-MS/MS data (see ESM Methods). In brief, the proteomics’ studies were conducted as follows: immunoprecipitation of the “bait” protein (Rac1), at the endogenous level; followed by 1D-SDS-PAGE to separate co-interaction proteins; ingel trypsin digestion to generate peptide fragments; and UPLC-ESI-MS/ MS analysis to identify co-immunoprecipitating proteins. Multiple biological comparisons and immunoprecipitation of NIgG (as non-specific control) were used to minimize false positives. Extensive bioinformatics and literature search were used to integrate biological and proteomics data and to identify pathways/functional categories, in which identified Rac1 interaction partners were involved, that were impacted by high glucose treatment. Additional experimental details are provided in Supplemental Materials.

2.4. Bioinformatics

Pathway analysis on high glucose-responsive Rac1 interaction partners were performed using Ingenuity Pathway Analysis (Ingenuity Systems, Inc., Redwood City, CA; www.ingenuity.com/), a bioinformatics analysis software package (Zhang et al., 2016) that contains biological and chemical interactions and functional annotations created by manual curation of the scientific literature (Zhang et al., 2016).

Reported Rac1 partners in rat, mouse and/or human cells/tissues were retrieved from 3 large interaction databases, STRING (https://string-db.org/), INTACT (http://www.ebi.ac.uk/intact) and BioGRID (https://thebiogrid.org/) and mapped to the Rac1 partners identified in this study. Default settings were used for the 3 databases except for the String database, for which only the interaction partners with minimum interaction score of 0.90 were considered. In addition, the approach used to identify the interaction and PMIDs for corresponding references in INTACT and BioGRID databases were also retrieved, if such information is available.

Subcellular localization enrichment analysis: was performed using the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8, https://david.ncifcrf.gov/.

2.5. Confirmation of Rac1 interaction partners by Co-IP and Western blotting

INS-1832/12 cells treated with either LG or HG were lysed in IP lysis buffer. Equal amounts of lysates prepared from these cells were precleared by rotation for 30 min at 4 °C with 10 μl of a 50% (w/v) slurry of protein G or A/G-agarose beads. After centrifugation, the precleared lysates were incubated overnight with antibodies directed against various signaling proteins or with normal mouse IgG along with protein G or A/G agarose beads at 4 °C with constant rotation. Precleared cell lysate without antibody was used as a negative control. Following incubation, beads were washed in IP lysis buffer, boiled in SDS sample buffer for 5 min and subjected to SDS-PAGE. Proteins were transferred to nitrocellulose membrane. Relative abundance of proteins in the immunoprecipitates was determined by Western blotting.

2.6. Western blotting

Following incubations, the cells were harvested and lysed in RIPA buffer containing 1 mg/ml protease inhibitor cocktail, 1 mM NaF, 1 mM PMSF and 1 mM Na₃VO₄. Cellular lysate proteins (30–50 μg) were separated by SDS-PAGE and electro-transferred onto the nitrocellulose membrane. The membranes were then blocked with 1% casein in 0.2X PBS or 3% BSA in PBS-T for 1 h at room temperature. Blots were then incubated overnight at 4 °C with appropriate primary antibody (1:3000 dilutions for GIT1, GIT1/2, β-PIX, Gα11/14, and karyopherin-α2 and 1:000 dilution for RacGAP1 in 1.5% BSA in PBS-T). The membranes were washed three times for 5 min each with PBS-T and probed with appropriate HRP-conjugated secondary antibody in 1.5% BSA in PBS-T at room temperature for 1 h. After washing, the target proteins were detected by chemiluminescence.

2.7. Statistical analysis of experimental data

All the data were presented as mean ± SEM from number of experiments as indicated in the text. The statistical significance of the differences between the experimental conditions was determined by t-test or ANOVA where appropriate. The p values < 0.05 was considered as statistically significant.

3. Results

3.1. Identification of Rac1 interaction partners in INS-1832/13 cells

As indicated in Fig. 1, INS-1832/13 cells were exposed to either LG or HG conditions for 48 h. Lysates derived from these cells were precleared with normal IgG, followed by co-immunoprecipitation with Rac1 antiserum. The proteins eluted from both anti-Rac1 and NIgG protein-A beads were resolved on 1D-SDS-PAGE and subjected to in-gel trypsin digestion. The resulting peptides were desalted and analyzed by UPLC-ESI-MS/MS.

Fig. 1. Experimental design of the study.

INS-1832/13 cells were treated with basal (2.5 mM) or HG (25 mM) for 48 h. Rac1 is immunoprecipitated and the proteins associated with Rac1 were identified by HPLC-ESI-MS/MS (see Methods for additional details).

As depicted in Fig. 2, a total of 1588 proteins were identified in anti-Rac1 immuno-precipitates with minimum 2 unique peptides with FDR at 0.01 in at least one Rac1 immunopreciptates after excluding common contaminants. Among those, 435 proteins had an enrichment ratio greater than 10. Out of these, 324 proteins were identified with a peak area (PA) in more than half (e.g., > 4 out of 8) Rac1 immunoprecipitates, which were classified as Rac1 interaction partners (Table 1 and Supplemental Table 1). To the best of our knowledge, this represents the largest Rac1 interactome to date in the pancreatic β-cell. Among the 324 Rac1 interaction partners, 39 were listed as Rac1 interaction partners in rat, mouse and/or human cells/tissues in the 3 large interaction partner databases, while 285 were not, and thus appeared to be novel.

Fig. 2.

Determination of interaction partners of Rac1 in INS-1832/13 cells exposed to basal or HG conditions.

Table 1.

The 324 proteins that met the criteria for classification as Racl interaction partners.

Gene Name	Protein Name	Mol. Weight [kD]	Enrichment ratio	Reported as an interaction partner of Racl in 3 large interaction databasesa
A1I3	Alpha-1-inhibitor 3	165.76	Infiniteb
AIM	Alpha-1-macroglobulin	167.12	Infinite
ACIN1	Apoptotic chromatin condensation inducer 1	151.04	Infinite
ACSL1	Long-chain-fatty-acid-CoA ligase 1	78.18	Infinite
ACTA1	Actin, alpha	42.05	Infinite	Yes
ACTG1	Actin, cytoplasmic 2	41.79	13000	Yes
ACTN1	Alpha-actinin-1	102.96	40	Yes
ACTN3	Actinin alpha 3	103.04	Infinite	Yes
ACTN4	Alpha-actinin-4	104.91	13	Yes
ACTR3	Actin-related protein 3	47.58	22	Yes
AFDN	Afadin	207.68	51
AHNAK	AHNAK 1	581.12	Infinite
ALDOA	Fructose-bisphosphate aldolase A	39.35	Infinite
AP2M1	AP-2 complex subunit mu	49.65	Infinite
AQR	Aquarius intron-binding spliceosomal factor	170.76	Infinite
ARHGAP11A	Rho GTPase-activating protein 11A	109.72	Infinite	Yes
ARHGEF6	Rho guanine nucleotide exchange factor 6	89.78	Infinite	Yes
ARHGEF7	Rho guanine nucleotide exchange factor 7	79.81	Infinite	Yes
ARPC1A	Actin-related protein 2/3 complex subunit 1A	41.60	Infinite	Yes
ARPC5	Actin-related protein 2/3 complex subunit 5	16.32	Infinite	Yes
ASCC3L1	U5 small nuclear ribonucleoprotein 200kDa helicase	244.87	190
ATP2A1	Sarcoplasmic/endoplasmic reticulum calcium ATPase 1	109.41	Infinite
ATP5A1	ATP synthase subunit alpha, mitochondrial	59.75	Infinite
ATXN2	Ataxin 2	117.30	36
BAT1A	Class I histocompatibility antigen, Non-RT1.A alpha-1 chain	38.95	Infinite
BCAS2	BCAS2, pre-mRNA-processing factor	26.10	46
BCLAF1	BCL2-associated transcription factor 1	184.71	Infinite
C1QB	Complement C1q subcomponent subunit B	26.65	Infinite
C3	Complement C3	186.32	Infinite
CAD	DNA fragmentation factor subunit beta	243.37	Infinite
CAMK2B	Calcium/calmodulin-dependent protein kinase type II subunit beta	72.83	Infinite	Yes
CAMSAP3	Calmodulin-regulated spectrin-associated protein family, member 3	132.03	Infinite
CCNA2	Cyclin A2 variant	43.86	Infinite
CDK5RAP3	CDK5 regulatory subunit-associated protein 3	57.13	Infinite
CELF2	CUGBP Elav-like family member 2	56.96	Infinite
CHUK	Conserved helix-loop-helix ubiquitous kinase (Predicted)	84.79	Infinite
CIRBP	Cold-inducible RNA-binding protein	18.61	Infinite
CKM	Creatine kinase M-type	43.02	36
CLINT1	Enthoprotin	52.14	230
CLP1	Polyribonucleotide 5-hydroxyl-kinase Clp1	47.74	Infinite
CLTB	Clathrin light chain B	25.12	Infinite
CLTC	Clathrin heavy chain 1	191.56	16	Yes
CLU	Clusterin	51.42	Infinite
CMAS	N-acylneuraminate cytidylyltransferase	48.13	Infinite
CMSS1	Uncharacterized protein C3orf26 homolog	31.50	Infinite
CNOT1	CCR4-NOT transcription complex, subunit 1	266.86	Infinite
CORO2A	Coronin	61.99	Infinite
CORO2B	Coronin	54.96	Infinite
CP	Ceruloplasmin	123.67	Infinite
CPSF1	Cleavage and polyadenylation specific factor 1, 160kDa (Predicted), isoform CRA_a	160.75	Infinite
CPSF4	Cleavage and polyadenylation specificity factor subunit 4	27.42	Infinite
CREBBP	CREB-binding protein	265.63	Infinite
CSDE1	Cold shock domain-containing protein E1	88.89	14
CSNK1D	Casein kinase I isoform delta	49.12	Infinite
CTTN	Cortactin isoform B	61.08	190	Yes
DBN1	Drebrin	72.60	100
DCAF7	DDB1 and CUL4-associated factor 7	38.93	Infinite
DDX1	ATP-dependent RNA helicase DDX1	82.50	31
DDX17	DEAD (Asp-Glu-Ala-Asp) box polypeptide 17, isoform CRA_a	72.64	22
DDX23	DEAD (Asp-Glu-Ala-Asp) box polypeptide 23 (Predicted), isoform CRA_b	95.50	Infinite
DDX3Y	DEAD-Box Helicase 3	73.15	Infinite
DDX41	DEAD (Asp-Glu-Ala-Asp) box polypeptide 41	69.80	Infinite
DDX47	DEAD (Asp-Glu-Ala-Asp) box polypeptide 47, isoform CRA_a	50.70	12
DDX6	DEAD-box helicase 6	54.24	13
DHX9	DEAH (Asp-Glu-Ala-His) box polypeptide 9 (Predicted)	131.73	140
DIDO1	Death inducer-obliterator 1	243.29	Infinite
DKC1	H/ACA ribonucleoprotein complex subunit 4	56.58	Infinite
DNAJC9	DnaJ (Hsp40) homolog, subfamily C, member 9 (Predicted)	29.98	Infinite
DNMT1	DNA (cytosine-5)-methyltransferase 1	182.77	Infinite
DOT1L	Histone-lysine N-methyltransferase, H3 lysine-79 specific	166.82	Infinite
EDC4	Enhancer of mRNA-decapping protein 4	163.25	12
EFHD2	EF-hand domain-containing protein D2	26.76	Infinite
EHMT1	Euchromatic histone lysine methyltransferase 1	138.91	Infinite
EHMT2	Euchromatic histone lysine methyltransferase 2	131.83	Infinite
EIF2C2	Protein argonaute-2	99.54	Infinite
EIF3D	Eukaryotic translation initiation factor 3 subunit D	63.99	Infinite
EIF4G1	Eukaryotic translation initiation factor 4 gamma, 1	175.70	18
ELAVL2	ELAV-like protein 2	42.53	Infinite
ELAVL4	ELAV-like protein 4	42.37	Infinite
ELP2	Elongator complex protein 2	91.75	Infinite
ELP3	Elongator complex protein 3	62.36	Infinite
ENO3	Beta-enolase	47.01	Infinite
EP300	E1A-binding protein	77.34	Infinite
EPB4.1L3	Erythrocyte protein band 4.1-like	107.07	14
ERI1	3–5 exoribonuclease 1	39.58	Infinite
ETV6	Transcription Factor ETV6	53.03	Infinite
EXOSC2	Exosome component 2	32.69	Infinite
EXOSC5	Exosome component 5	25.22	Infinite
EXOSC8	Exosome component 8	26.14	Infinite
FAM120A	Family with sequence similarity 120A	122.14	30
FAM160A2	FTS and Hook-interacting protein	86.27	Infinite
FAM98A	Protein FAM98A	55.07	Infinite
FHOD3	Formin homology 2 domain-containing 3	168.46	Infinite
FIP1L1	Pre-mRNA 3-end-processing factor FIP1	64.96	Infinite
FLII	Flightless I homolog	144.86	Infinite
FLNB	Filamin, beta (Predicted)	277.84	Infinite	Yes
FXR2	FMR1 autosomal homolog 2	74.37	58
FYN	Tyrosine-protein kinase Yes	60.73	Infinite	Yes
G3BP2	GTPase activating protein (SH3 domain) binding protein 2	50.77	Infinite
GAK	Cyclin-G-associated kinase	143.71	Infinite
GAPDH	Glyceraldehyde-3-phosphate dehydrogenase	35.83	Infinite
GAPVD1	GTPase-activating protein and VPS9 domains 1	165.16	23
GEMIN5	Gem (nuclear organelle)-associated protein 5	166.06	51
GIT1	ARF GTPase-activating protein GIT1	85.23	Infinite	Yes
GIT2	G protein-coupled receptor kinase interacting ArfGAP 2	84.52	Infinite	Yes
GKAP1	G kinase-anchoring protein 1	41.93	Infinite
GNAQ	Guanine nucleotide-binding protein G(q) subunit alpha	42.16	Infinite
GPR158	Probable G-protein-coupled receptor 158	134.83	Infinite
GRAMD1A	GRAM domain-containing protein 1A	89.52	Infinite
GRAMD3	GRAM domain-containing protein 3	48.05	Infinite
GRSF1	G-rich RNA sequence-binding factor 1	52.91	16
GTSE1	G-2 and S-phase-expressed 1	79.98	Infinite
H1FX	H1 histone family, member X	20.49	Infinite
H2AFJ	Histone H2A	14.15	Infinite
HIRA	Histone cell cycle regulator	111.68	Infinite
HNRNPC	Heterogeneous nuclear ribonucleoprotein C	33.06	24
HNRNPH3	Heterogeneous nuclear ribonucleoprotein H3 (2H9) (Predicted), isoform CRA_c	36.87	Infinite
HNRNPL	Heterogeneous nuclear ribonucleoprotein L	67.90	62
HNRNPM	Heterogeneous nuclear ribonucleoprotein M	77.63	13
HNRNPR	Heterogeneous nuclear ribonucleoprotein R	70.87	Infinite
HNRNPUL2	Heterogeneous nuclear ribonucleoprotein U-like 2	84.86	15
HSPA5	78 kDa glucose-regulated protein	72.35	Infinite
IKBKAP	Elongator complex protein 1	149.20	Infinite
IKBKB	Inhibitor of nuclear factor kappa-B kinase subunit beta	86.89	Infinite
IKBKG	NF-kappa-B essential modulator	48.07	Infinite
ILF2	Interleukin enhancer-binding factor 2	43.06	13
ILF3	Interleukin enhancer-binding factor 3	98.08	120
INA	Alpha-internexin	56.12	Infinite
INPPL1	Phosphatidylinositol 3,4,5-trisphosphate 5-phosphatase 2	139.14	Infinite
ITGB6	Integrin beta-6	85.96	Infinite
KAB	KARP-1-binding protein 1	125.96	Infinite
KEAP1	Kelch-like ECH-associated protein 1	69.40	Infinite
KHDRBS1	KH domain-containing, RNA-binding, signal transduction-associated protein 1	48.32	Infinite
KIDINS220	Kinase D-interacting substrate of 220 kDa	196.41	Infinite
KIF23	Kinesin-like protein	107.99	Infinite
KIF2A	Kinesin-like protein KIF2A	83.93	Infinite
KIFC1	Kinesin-like protein KIFC1	76.14	Infinite
KLHL22	Kelch-like protein 22	78.70	Infinite
KLHL7	Kelch-like protein 7	65.97	Infinite
KPNA1	Importin subunit alpha-1	60.14	Infinite
KPNA4	Importin subunit alpha	57.92	Infinite	Yes
KPNA7	Importin subunit alpha-7	59.89	Infinite	Yes
LAD1	Ladinin-1	57.40	Infinite
LANCL2	LanC lantibiotic synthetase component C-like 2	50.97	Infinite
LARP1	La ribonucleoprotein domain family, member 1	106.94	41
LARP5	La ribonucleoprotein domain family, member 4B	81.08	15
LARS	Leucyl-tRNA synthetase	134.28	11
LDHA	L-lactate dehydrogenase	36.45	Infinite
LIMA1	Epithelial protein lost in neoplasm	83.80	95	Yes
LIMCH1	LIM and calponin homology domains 1	121.50	Infinite
LMO7	LIM domain 7	156.76	220
LOC100359980	Similar to Myosin light chain 1 slow a (Predicted)	22.81	Infinite
LOC100360682	Small nuclear ribonucleoprotein polypeptide E	10.76	Infinite
LRRFIP2	Leucine-rich repeat flightless-interacting protein 2	49.77	Infinite
LYAR	Cell growth-regulating nucleolar protein	43.68	Infinite
LZTS2	Leucine zipper putative tumor suppressor 2	72.58	Infinite	Yes
MAGMAS	Mitochondrial import inner membrane translocase subunit TIM16	13.77	Infinite
MATR3	Matrin-3	94.50	70
MISP3	MISP Family Member 3	23.49	Infinite
MLLT10	Histone lysine Mmethyltransferase DOT1L cofactor	103.18	Infinite
MLLT6	MLLT6, PHD Finger Containing	99.17	Infinite
M0V10	Mov10 RISC complex RNA helicase	113.76	Infinite
MPRIP	Myosin phosphatase Rho-interacting protein	117.01	Infinite
MRPL11	39S ribosomal protein L11, mitochondrial	20.75	Infinite
MRPL16	39S ribosomal protein L16, mitochondrial	28.90	Infinite
MRPL21	Mitochondrial ribosomal protein L21 (Predicted), isoform CRA_a	23.39	Infinite
MRPS22	Mitochondrial ribosomal protein S22	41.24	Infinite
MTA2	Metastasis-associated gene family, member 2	74.96	Infinite
MUC13	Mucin-13	57.54	Infinite
MYEF2	Myelin expression factor 2	63.38	35
MYH10	Myosin-10	229.00	1800	Yes
MYH14	Myosin heavy chain 14	228.91	220	Yes
MYH4	Myosin-4	222.69	38
MYH6	Myosin-6	223.38	Infinite
MYH7	Myosin-7	222.90	Infinite
MYH9	Myosin-9	226.34	39	Yes
MYL1	Myosin light chain 1/3, skeletal muscle isoform	20.68	250
MYL12B	Myosin regulatory light chain 12B	19.78	Infinite	Yes
MYL6	Myosin light polypeptide 6	17.01	13	Yes
MYLK	Myosin light chain kinase	214.91	Infinite	Yes
MYLPF	Myosin regulatory light chain 2, skeletal muscle isoform	18.97	Infinite
MYO18A	Myosin XVIIIa	233.38	Infinite
MYO1B	Unconventional myosin-Ib	131.92	67
MYO1C	Unconventional myosin-Ic	119.81	44	Yes
MYO1D	Unconventional myosin-Id	116.09	250
MY05A	Unconventional myosin-Va	212.15	Infinite
MZT2B	Mitotic spindle organizing protein 2B	16.41	Infinite
NAB1	NGFI-A-binding protein 1	54.03	Infinite
NAP1L1	Nucleosome assembly protein 1-like 1	45.37	Infinite
NCBP1	Nuclear cap-binding protein subunit 1	91.91	14
NCK2	Non-catalytic region of tyrosine kinase adaptor protein 2 (Predicted)	42.91	Infinite	Yes
NDUFA8	NADH dehydrogenase [ubiquinone] 1 alpha subcomplex subunit 8	22.12	Infinite
NHP2	H/ACA ribonucleoprotein complex subunit 2	17.29	Infinite
NO66	Lysine-specific demethylase NO66	67.45	Infinite
N0L5A	Nucleolar protein 5A	65.40	27
NOL6	Neuroprotective protein 1	128.07	Infinite
NOP16	Nucleolar protein 16	21.12	Infinite
N0P58	Nucleolar protein 58	60.07	32
NPAP60	Nuclear pore complex protein Nup50	49.75	Infinite
NSUN4	NOL1/NOP2/Sun domain family, member 4 (Predicted)	42.61	Infinite
NUFIP2	NUFIP2, FMR1-interacting protein 2	75.66	Infinite
NUP153	Nuclear pore complex protein Nup153	152.83	Infinite
OTUD4	OTU deubiquitinase 4	122.72	Infinite
PABPC1	Polyadenylate-binding protein 1	70.70	10
PABPC4	Polyadenylate-binding protein 4	70.83	23
PABPN1	Poly(A) binding protein, nuclear 1, isoform CRA_a	32.31	Infinite
PAIP1	Polyadenylate binding protein-interacting protein 1 (Predicted)	46.09	Infinite
PAK1	Serine/threonine-protein kinase PAK 1	60.58	Infinite	Yes
PAK3	Serine/threonine-protein kinase PAK 3	60.71	Infinite	Yes
PCBP3	Poly (RC) binding protein 3	33.78	Infinite
PCF11	PCF11 cleavage and polyadenylation factor subunit	172.66	Infinite
PDS5B	Sister chromatid cohesion protein PDS5 homolog B	164.52	15
PLRG1	Pleiotropic regulator 1	57.19	Infinite
PLS3	Plastin-3	70.75	Infinite
POLR2A	DNA-directed RNA polymerase	217.20	Infinite
PPIE	Peptidyl-prolyl cis-trans isomerase	33.37	Infinite
PPP1R12A	Protein phosphatase 1 regulatory subunit 12A	115.27	Infinite	Yes
PPP1R12B	Protein phosphatase 1 regulatory subunit	111.39	Infinite	Yes
PPP1R9A	Neurabin-1	122.73	Infinite
PPP1R9B	Neurabin-2	89.66	Infinite
PRMT5	Protein arginine N-methyltransferase 5	72.69	Infinite
PRPF19	Pre-mRNA-processing factor 19	55.24	13
PRPF8	Pre-mRNA processing factor 8, isoform CRA_a	273.61	Infinite
PRPH	Peripherin	53.55	25
PRPS1	Ribose-phosphate pyrophosphokinase 1	34.81	Infinite
PRRC2A	Protein PRRC2A	229.17	21
PRRC2B	Protein Prrc2b	243.10	Infinite
PTBP3	Polypyrimidine tract-binding protein 3	59.81	Infinite
PTGFRN	Prostaglandin F2 receptor negative regulator	98.67	27
PTMA	Prothymosin alpha	12.38	Infinite
PTOV1	Prostate tumor-overexpressed gene 1 protein homolog	46.85	Infinite
PVALB	Parvalbumin alpha	11.93	24
PXN	Paxillin	64.06	Infinite	Yes
PYGM	Glycogen phosphorylase, muscle form	97.32	Infinite
QN1	Protein QN1 homolog	161.00	Infinite
RAB11A	Ras-related protein Rab-11B	24.49	Infinite
RAB8A	Ras-related protein Rab-8A	23.67	Infinite
RAI14	Ankycorbin	109.13	Infinite
RALY	RALY heterogeneous nuclear ribonucleoprotein	33.05	13
RBFOX2	RNA binding protein fox-1 homolog 2	45.55	Infinite
RBM15	RNA binding motif protein 15	105.76	Infinite
RBM47	RNA-binding motif protein 47	64.09	Infinite
RBMX	RNA-binding motif protein, X chromosome	42.26	Infinite	Yes
RECQL	ATP-dependent DNA helicase Q1	69.64	25
RGD1304704	LRRGT00192	28.17	11
RHOA	Transforming protein RhoA	21.78	Infinite	Yes
RPL12	60S ribosomal protein L12	17.85	Infinite
RPL14	60S ribosomal protein L14	23.78	Infinite
RPL19	60S ribosomal protein L19	23.47	Infinite
RPL21	60S ribosomal protein L21	18.68	Infinite
RPL26	60S ribosomal protein L26	17.26	Infinite
RPL27A	60S ribosomal protein L27a	16.59	Infinite
RPL31	60S ribosomal protein L31	16.32	Infinite
RPL36	60S ribosomal protein L36	12.41	Infinite
RPS27L	40S ribosomal protein S27-like	9.48	Infinite
RQCD1	Cell differentiation protein RCD1 homolog	33.60	Infinite
RRP7A	Ribosomal RNA-processing 7 homolog A	32.34	Infinite
SAFB	Scaffold attachment factor B1	105.52	190
SCRIB	Scribbled planar cell polarity protein	179.85	Infinite	Yes
SEC16A	Protein transport protein sec16	253.29	Infinite
SEC61A1	Protein transport protein Sec61 subunit alpha isoform 1	52.26	Infinite
SF3B1	Splicing factor 3 b, subunit 1	145.83	12
SGPL1	Sphingosine-1-phosphate lyase 1	63.76	65
SHMT1	Serine hydroxymethyltransferase	75.37	Infinite
SHPRH	SNF2 histone linker PHD RING helicase	194.70	Infinite
SIPA1L3	Signal Induced Proliferation Associated 1 Like 3	194.93	Infinite
SMARCA4	Transcription activator BRG1	181.40	Infinite
SMARCB1	SWI/SNF related, matrix associated, actin dependent regulator of chromatin, subfamily b, member 1	43.16	Infinite
SMC1A	Structural maintenance of chromosomes protein 1A	143.20	16
SMU1	WD40 repeat-containing protein SMU1	57.54	Infinite
SND1	Staphylococcal nuclease domain-containing protein 1	101.95	Infinite
SORBS1	Sorbin and SH3 domain-containing protein 1	143.37	Infinite
SPECC1L	Cytospin-A	124.34	Infinite
SPNA2	Spectrin alpha chain	286.71	13
SPOUT1	SPOUT Domain Containing Methyltransferase 1	41.97	Infinite
SRP68	Signal recognition particle subunit SRP68	70.49	11
SRPK1	SFRS protein kinase 1	73.93	Infinite
SRSF1	Serine/arginine-rich splicing factor 1	27.74	13
SRSF5	Serine/arginine-rich splicing factor 5	30.89	Infinite
STAU1	Protein Phosphatase 1, Regulatory Subunit 150	61.33	23	Yes
STAU2	Double-stranded RNA-binding protein Staufen homolog 2	62.68	52
STK38L	Serine/Threonine Kinase 38 Like	53.76	Infinite
SURF4	Surfeit 4, isoform CRA_a	83.26	Infinite
SVIL	Supervillin; Membrane-Associated F-Actin Binding Protein P205	242.05	Infinite
SYCP2	Synaptonemal complex protein 2	172.59	Infinite
SYNPO2	Synaptopodin-2	136.02	160
T	T brachyury transcription factor	47.43	Infinite
TAF9	Transcription initiation factor TFIID subunit 9	28.99	Infinite
TARDBP	TAR DNA Binding Protein	32.15	43
TBC1D5	TBC1 domain family, member 5	93.06	Infinite
TFRC	Transferrin receptor protein 1	85.88	Infinite
THBS3	Thrombospondin 3	104.13	Infinite
THOC2	THO complex 2	185.06	Infinite
THOC5	THO complex subunit 5 homolog	78.66	Infinite
THOC6	THO complex subunit 6 homolog	37.42	12
TMEM189	Ubiquitin-conjugating enzyme E2 variant 2	16.36	Infinite
TMOD2	Tropomodulin-2	39.49	20
TMOD3	Tropomodulin-3	39.47	Infinite
TNKS1BP1	Tankyrase 1-binding protein 1	180.01	34
TNNC2	Troponin C2	18.10	19
TNNT3	Troponin T	30.75	Infinite
TPM2	Tropomyosin beta chain	32.84	Infinite
TPM4	Tropomyosin alpha-4 chain	28.51	Infinite
TRA2A	Transformer 2 Alpha Homolog	32.58	Infinite
TRIM25	Tripartite motif-containing 25	71.92	Infinite
TRIM27	Tripartite motif-containing 27	58.55	Infinite
UBTD1	Ubiquitin domain-containing protein 1	25.96	Infinite
UFL1	E3 UFMl-protein ligase 1	89.58	Infinite
UPF1	UPF1, RNA Helicase And ATPase	122.64	35
USP10	Ubiquitin carboxyl-terminal hydrolase 10	87.31	Infinite
VIL1	Villin 1	92.66	11
VPS13A	Vacuolar Protein Sorting 13 Homolog A	166.62	370
WDR77	Methylosome protein 50	37.73	Infinite
WIZ	Widely-interspaced zinc finger motifs	103.64	Infinite
YLPM1	YLP motif-containing protein 1	238.98	Infinite
YTHDC1	YTH domain-containing protein 1	85.90	Infinite
YTHDF1	YTH N (6)-methyladenosine RNA-binding protein 1	60.91	Infinite
ZFP275	Zinc finger protein 275	55.99	Infinite
ZFP462	Zinc finger protein 462	283.91	Infinite
ZFR	Zinc finger RNA-binding protein	116.91	Infinite
A0A0G2JTG4	Uncharacterized protein	14.89	Infinite
A0A0G2K3S3	Uncharacterized protein	11.19	Infinite
A0A0G2K7N7	Uncharacterized protein	31.06	Infinite
A0A0G2K8A9	Uncharacterized protein	23.79	Infinite

^aSTRING, IntACT and BioGRID databases.

^bProteins only identified in the Rac1 immunoprecipitates, but not in the NIgG immunoprecipitates.

Data in Fig. 3A represent a graphical depiction of eleven significantly enriched pathways for the 324 Rac1 interaction partners identified using Ingenuity Pathway Analysis (IPA). The number of proteins identified as Rac1 interaction partners in a particular pathway is indicated above the bar. Based on the known biological functions of Rac1, we identified 30 interaction partners of Rac1 involved in the actin cytoskeletal signaling pathways. It is noteworthy that our studies identified Rac1 interaction partners involved in calcium (17 partners), RhoGDI (20 partners), PAK (13 partners) and Rho GTPase (19 partners) signaling pathways. Indeed, published evidence implicates these signaling pathways in islet β-cell function, including GSIS (Kalwat and Thurmond, 2013; Kowluru, 2010, 2017). Emerging evidence in multiple cell types including the islet β-cell appear to indicate localization of Rac1 in various subcellular compartments including cytosol, plasma membrane, mitochondria and nucleus (Kowluru et al., 2003; Abdrabou and Wang, 2018; Payapilly and Malliri, 2018; Baidwan et al., 2017). Fig. 3B is representation of significantly enriched subcellular localizations for the Rac1 interaction partners in INS-1832/13 cells. Pathways of actin cytoskeleton and GPCR signaling generated using IPA are provided as Supplemental Figs. 1 and 2, respectively.

Fig. 3.

Graphical depiction of significantly enriched pathways and subcellular localizations for the Rac1 interaction partners.

Panel A: Significantly enriched pathways for the Rac1 interaction partners identified using Ingenuity Pathway Analysis. The number of proteins identified as Rac1 interaction partners in a particular pathway is indicated above the bar. The greater the -Log (p value) value (e.g., the smaller p value), the less likely a pathway is significantly enriched just by chance.

Panel B: Significantly enriched subcellular localizations for the Rac1 interaction partners identified using the Database for Annotation, Visualization and Integrated Discovery (DAVID) v6.8, https://davidncifcrf.gov/ . The number of proteins identified as Rac1 interaction partners in a particular subcellular localization is indicated above the bar. The greater the -Log (p value) value (e.g., the smaller p value), the less likely a pathway is significantly enriched just by chance.

3.2. Identification of HG-responsive Rac1 interaction partners in INS-1832/13 cells

As described in Fig. 2 and 142 out of 243 interaction partners, had a fold change greater than 1.5 (i.e., 1.5 fold increase) or less than 0.67 (i.e., 1.5 fold decrease) when comparing lysates from cells treated with LG to HG conditions. Among these, 27 proteins showed a significant difference in Rac1 interaction in response to HG conditions (p < 0.05), and therefore, they were considered as “HG-responsive interaction partners” (Table 2 and Supplemental Table 2). Among these 27 Rac1 interaction partners with a significant difference between LG and HG treatments, 9 were listed as Rac1 interaction partners in rat, mouse and/or human cells/tissues in the 3 large interaction partner databases, while 18 were not, and thus appeared to be novel. Fig. 4 A represents a volcano plot displaying the 324 Rac1 interaction partners in INS-1832/ 13 cells exposed to either basal or HG conditions. Based on their known regulatory roles, these 27 Rac1 interaction partners were grouped into four major categories (Fig. 4 B), namely cytoskeletal proteins (e.g., cortactin, paxillin), nuclear proteins (e.g., importin; also referred to as karyopherin), G-proteins and their regulatory factors (e.g., the α-sub-unit of the trimeric G-protein Gq and the small G-protein Rho) and others (e.g., protein phosphatase 1 regulatory subunit 12A). Our findings also suggested that there were 13 interaction partners with a fold change greater than 5 (also with p < 0.05) in response to HG conditions. Interestingly, 7 out of these 13 proteins exhibited greater than 10 fold increase in the association with Rac1 under HG conditions (Table 2). They are: actin-related protein, LAD1 ladinin-1, leucine-rich repeat flightless-interacting protein 2, protein phosphatase 1 regulatory subunit 12A, peripherin, spectrin alpha chain and supervillin.

Table 2.

The 27 Racl interaction partners with a significant difference between high glucose (HG) and low glucose (LG) treated INS1 cells. All these interaction partners have a enrichment ratio > 10 compared to the NIgG control, detected in > 4 out of the 8 Racl IP samples, and also have a fold change for HG vs. LG greater than 1.5 (i.e., 1.5 fold increase) or less than 0.67 (i.e., 1.5 fold decrease). Data are given as fold changes (means ± SEM). Peak area for each protein identified in a specific IP sample was normalized against the peak area for Rac1 identified in the same sample (See Methods). The normalized peak area for each Rac1 interaction partner was compared between the HG and LG groups to assess effects of HG on protein-protein interactions involving Rac1. Mean of the normalized peak area for each Rac1 interaction partner in the LG sample was set to 1.00, and all the fold changes were relative to LG, n = 4, p < 0.05.

Gene Name	Protein Name	LG	HG	p-value	Reported as an interaction partner of Rac1 in 3 large interaction databasesa
ACTR3	Actin-related protein 3	1.0010.35	13.74±5.28	0.03255	Yes
ARHGEF6	Rho guanine nucleotide exchange factor 6	1.00±0.19	1.66±0.1	0.03836	Yes
ARPC5	Actin-related protein 2/3 complex subunit 5	1.00±0.23	9.114.15	0.01267	Yes
CTTN	Cortactin isoform B	1.00±0.27	8.62±2.12	0.00729	Yes
DDX3Y	DEAD-Box Helicase 3	1.00±0.2	1.79±0.14	0.03469
FLII	Flightless 1 homolog	1.00±0.26	3.98±1.02	0.01481
GNAQ	Guanine nucleotide-binding protein G(q) subunit alpha	1.00+0.20	9.82+3.36	0.00143
KPNA1	Importin subunit alpha-1	1.00±0.15	2.97±0.78	0.02214
KPNA7	Importin subunit alpha-7	1.00±0.23	2.79±0.75	0.02914	Yes
LAD1	Ladinin-1	1.00±0.15	28.05±11.7	0.01261
NUFIP2	NUFIP2, FMRl-interacting protein 2	1.00±0.16	1.94±0.31	0.03059
LRRFIP2	Leucine-rich repeat flightless-interacting protein 2	1.00±0.17	12.24±2.77	0.00022
MYOIB	Unconventional myosin-lb	1.00±0.39	8.12±2.97	0.03365
PABPC4	Polyadenylate-binding protein 4	1.00±0.17	1.92±0.25	0.02448
PLS3	Plastin-3	1.00±0.01	2.34±0.27	0.01301
PPP1R12A	Protein phosphatase 1 regulatory subunit 12A	1.00±0.13	14.12±2.18	0.00005	Yes
PRPH	Peripherin	1.00±0.18	17.72±8.66	0.01884
PXN	Paxillin	1.00±0.24	2.95±0.27	0.00564	Yes
RAI 14	Ankycorbin	1.00±0.01	8.73±1.75	0.00378
RALY	RALY heterogeneous nuclear ribonucleoprotein	1.00±0,17	1.66±0.17	0.02867
RPL27A	60S ribosomal protein L27a	1.00±0.33	2.86±0.51	0.02238
RHOA	Transforming protein RhoA	1.00±0.34	2.38±0.43	0.04885	Yes
SCRIB	Scribbled planar cell polarity protein	1.00±0.18	1.83±0.22	0.03841	Yes
SPNA2	Spectrin alpha chain	1.00±0.38	11.34±3.24	0.01484
SRSF5	Serine/arginine-rich splicing factor 5	1.00±0.16	2.11±0.45	0.04826
SVIL	Supervillin; Membrane-Associated F-Actin Binding Protein P205	1.00±0.27	14.69±4.95	0.02317
TRIM27	Tripartite motif-containing 27	1.00±0.25	5.07±1.04	0.00538

^*:Green highlighted proteins are the 7 RAC1 interaction partners with p<0.01 HG vs. LG.

^a:STRING, IntACT and BioGRID databases

Fig. 4.

HG-responsive interacting proteins of Rac1 in insulin-secreting INS-1832/13 cells

Panel A: Volcano plot for the 324 Rac1 interaction partners with/out high glucose treatments. The green points indicate Rac1 interaction partners with both large magnitude fold-changes (log 2 of fold change, x axis), and high statistical significance (-log10 of p value, y axis). The horizontal red line shows where p = 0.05 with points above the line having p < 0.05 and points below the line having p > 0.05: log (0.05, 10) = 1.30. The vertical red lines show where a fold change for HG vs. LG greater than 1.5 (i.e., 1.5 fold increase) or less than 0.67 (i.e., 1.5 fold decrease): log (1.5, 2) = 0.585. Those points having a fold-change < 1.5 and/or p > 0.05 are shown in gray and those with a fold-change > 1.5 and p < 0.05 are shown in green.

Panel B: Based on the identity and functions of the HG-responsive proteins identified in the current study, we broadly classified those proteins into four major categories, including G-proteins (subunits and regulatory factors), cytoskeletal proteins, nuclear proteins and others (see Table 2 for additional information on these proteins). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

3.3. Experimental determination of the protein abundance of Rac1 interaction partners in INS-1832/13 cells, normal rat islets and human islets

In the next series of experiments, we validated our proteomics data (above) by determining the expression/protein abundance of select signaling proteins via Western blotting of lysates derived from INS-1832/13 cells, normal rat islets and human islets. Data in Fig. 5 A suggest that GIT1/2, known GTPase activating proteins for Arf6, a small G-protein that we have implicated in physiological insulin secretion (Jayaram et al., 2011) are expressed in all three cell types studied. We also detected expression of β-PIX, a guanine nucleotide exchange factor for Cdc42 and Rac1 in all the three insulin-secreting cells studied. The expression of α-subunit of the trimeric G-protein Gαq/11/14, which we have shown to be one of the high glucose-responsive proteins in INS-1832/13 cells by proteomics (Fig. 4) was also verified in INS 1832/ 13 cells, normal rat islets and human islets. In addition, we noted the expression of Rac1 GTPase-activating protein (RacGAP1) in all the 3 insulin-secreting cells including human islets. Lastly, we provide the first evidence for the expression of karyopherin, a scaffolding protein, which has been implicated in the nuclear import of Rac1 under conditions of cell dysfunction and apoptosis (see below). Levels of expression of Rac1 in INS-1832/13 cells, normal rat islets and human islets are shown in Fig. 5B. Together, our Western blot data (Fig. 5 A and B) further strengthen our proteomics data. It is noteworthy that our current studies are the first to report expression of GIT1/2, RacGAP1 and karyopherin in clonal β-cells, rodent islets and human islets.

Fig. 5.

Immunological detection of select Rac1 interacting partners in INS-1832/13cells, normal rodent islets and human islets.

Panel A: Lysates derived from INS-1832/13 cells, normal rat islets and human islets were subjected to SDS-PAGE and identified by Western blotting method using specific antibodies directed against each of the proteins indicated (see Methods for additional details). Representative blots from multiple studies are provided in the figure.

Panel B: Rac1 expression in INS-1832/13 cells, normal rat islets and human islets was determined by Western blotting. The membranes were reprobed with anti-actin antibody to serve as loading control. Representative blots from multiple studies are provided in the figure.

3.4. Experimental validation of Gαq/11/14 as a high glucose-responsive Rac1 interaction partner in INS-1832/13 cells

Our proteomics data in Fig. 4 identified Gαq11/14 as one of the high glucose-responsive interaction partners of Rac1 in INS-1832/ 13 cells. Furthermore, our Western blot data (Fig. 5) suggested high degree of expression of this trimeric Ga subunit in INS-1832/13 cells, normal rat islets and human islets. Therefore, we then asked whether hyperglycemic exposure of INS-1832/13 cells leads to increased expression of this protein and/or whether it interacts with Rac1 under those conditions. Data from these studies are included in Fig. 6. First, we noticed no significant effects of hyperglycemic conditions on the expression of Gαq/11/14 (Panel A). Pooled data from 4 independent repeats are represented in Fig. 6 A. Data depicted in Fig. 6 B indicate no significant changes in the expression levels of two HG-responsive proteins (β-PIX and GIT1/2) that we identified in the current study. Interestingly, however, data from Co-IP experiments revealed that HG conditions (24 h) led to increased association of Gαq/11/14 with Rac1 (Panel C). Pooled data from multiple experiments are included in Panel D. Note that data depicted in Panels C and D represent relative abundance of Rac1 in INS 1832/13 lysates immunoprecipitated with an antiserum directed against Gαq/11/14. We next undertook a reverse Co-IP approach in which relative abundance of Gαq/11/14, GIT1/2 and β-PIX in immunoprecipitates derived from LG/HG exposed INS 1832/13 cells using the Rac1 antiserum. Data in Fig. 7 further demonstrate increased association of Rac1 with these three proteins in INS-1832/13 cells exposed to HG conditions. Together, data depicted in Figs. 6 and 7 validate our findings from proteomics experiments.

Fig. 6. Immunological validation of proteomics data Panel.

Panel A: HG conditions do not increase the expression of Gαq/11/14 in INS-1832/13 cells. INS-1832/13 cells were cultured in the presence of basal glucose (2.5 mM; LG) or high glucose (25 mM; HG) for 24 h (see Methods). Cells were lysed and subjected to SDS-PAGE and probed with an antiserum directed against Gaq11/14. GAPDH was used as a loading control. A representative blot from four experiments was shown here. The graph represents the pooled data (mean ± SEM) from four experiments in which relative intensity of Gαq11/14 was normalized to GAPDH.

Panel B: Expression of β-PIX and GIT1/1 under HG conditions in INS-1832/ 13 cells. INS-1832/13 cells were cultured in the presence of basal glucose (2.5 mM; LG) or high glucose (25 mM; HG) for 24 h (see Methods). Cells were lysed in RIPA buffer and subjected to SDS-PAGE and probed with an antiserum directed against β-PIX and GIT1/2. Actin was used as a loading control. A representative blot from four experiments is shown here.

Panel C: High glucose promotes an interaction between Rac1 and Gαq11/14. INS-1 cells were cultured in the presence of either LG (2.5 mM) or HG (25 mM) for 24 h and lysed in IP lysis buffer. Equal volume of precleared cell lysates was immunoprecipitated with anti-Gαq11/14 antibody or normal mouse IgG monoclonal antibody. Precleared cell lysate without antibody was used as negative control. The resulting immunoprecipitates were subjected to SDS-PAGE and analyzed by western blotting with antibody against Rac1. Total cell lysates were probed for Rac1 and Gαq11/14. A representative blot from four experiments is shown.

Panel D: Densitometric quantification of Rac1 abundance in the immunoprecipitates of Gαq11/14 from LG and HG exposed cells. Data are shown mean ± SEM for four independent experiments. * represents p < 0.05.

Fig. 7.

Co-IP experiments to demonstrate increased association between Rac1 and Gαq/11/14, GIT1/2 and β-PIX in INS-1832/13 cells under HG-treatment conditions. INS-1832/13 cells were cultured in the presence of either LG (2.5 mM) or HG (25 mM) for 24 h, and lysed in IP lysis buffer. Equal volume of precleared cell lysates were immunoprecipitated with anti-Rac1 serum or normal mouse IgG monoclonal antibody. Precleared cell lysates without antibody were used as negative controls. The resulting immunoprecipitates were subjected to SDS-PAGE and analyzed by Western blotting with antibody against Gαq/11/14, GIT1/2 and β-PIX. A representative blot from four experiments is shown.

4. Discussion

Using pharmacological, biochemical, and molecular biological approaches, several earlier studies have implicated small G proteins (e.g., Cdc42, Rac1, Arf6) in actin cytoskeletal rearrangements, vesicular transport and fusion of insulin-containing secretory granules to the plasma membrane for fusion and secretion of insulin (Kalwat and Thurmond, 2013; Kowluru, 2010, 2017). In the light of emerging evidence, which indicates regulatory roles of Rac1 in islet β-cell function in health and under metabolic stress (e.g., exposure to hyperglycemia; 8,22), we undertook the current studies to identify interaction partners of Rac1 in insulin-secreting INS-1832/13 cells under normal and HG conditions. It should be noted that, earlier studies from multiple laboratories including our own have documented significant impairment in GSIS in clonal β-cells, rodent islets and human islets following exposure to HG conditions (Sidarala et al., 2015; Sidarala and Kowluru, 2017a; Marshak et al., 1999; Khadija et al., 2014; Kong et al., 2015; Elumalai et al., 2017), Furthermore, we have demonstrated defects in mitochondrial (caspase-3 activation) and nuclear (lamin-B degradation) integrity and function under these conditions culminating in cell demise (Sidarala et al., 2015; Sidarala and Kowluru, 2017a; Khadija et al., 2014).

Using a quantitative proteomic approach (Co-IP and HPLC-ESI-MS/ MS), we identified a total of 324 Rac1 interaction partners in pancreatic β-cells. These Rac1 partners play contributory roles in the regulation of at least 11 signaling pathways, including those involved in actin cytoskeletal rearrangements, calcium- and G protein-dependent signaling. More importantly, we have identified 27 proteins that appear to exhibit increased association with Rac1 under hyperglycemic conditions. Identification of these interacting factors/partners provide fresh insights into delineation of specific signaling pathways that underlie islet β-cell dysregulation and demise under metabolic stress. For brevity, the following discussion is focused on potential roles of five Rac1 partners (i.e., GIT1/2, β-PIX, Gαq/11/14, Rac1GAP and karyopherin) that we have validated via immunological approaches (Figs. 5–7).

4.1. Potential regulatory roles of GIT1/2 in islet function

G protein-coupled receptor kinase-interacting protein-1 (GIT1) is a multidomain protein, which has been shown to interact with a variety of signaling molecules including β-PIX, PAK1, focal adhesion kinase, phospholipase-γ and mitogen-activated protein kinase-1. Functional activation of GITs has been shown to be precisely regulated via tyrosine phosphorylation (Hoefen and Berk, 2006; Zhou et al., 2016). Earlier studies have demonstrated that GITs also subserve the functions as GTPase-activating proteins (GAPs), specifically for Arf6 (Hoefen and Berk, 2006; Zhou et al., 2016). Published evidence indicates that GIT1 exerts both positive as well as negative modulatory roles in actin-driven membrane protrusions by increasing and decreasing activation of Rac1 and/or Cdc42, respectively (Hoefen and Berk, 2006; Zhou et al., 2016). Together, above evidence implicates GITs in the functional activation of Arf6-Cdc42-Rac1 signaling pathways. In this context, earlier studies have investigated roles of Cdc42, Rac1 and Arf6 in GSIS in clonal β-cells, rodent islets and human islets (Kalwat and Thurmond, 2013; Kowluru, 2010, 2017; Jayaram et al., 2011; Jayaram and Kowluru, 2012; Lawrence and Birnbaum, 2003; Ma et al., 2010; Suijun et al., 2014; Guo et al., 2018). Data from these investigations have yielded valuable insights into various regulatory proteins/factors that appear to mediate the activation-deactivation steps of these G proteins. These include GDIs, GEFs, and GAPs. For example, original investigations from our laboratory have reported novel regulatory roles for Arf nucleotide binding site opener (ARNO), a GEF for the small G-protein Arf6, in the functional activation of Arf6 in clonal β-cells and primary rodent and human islets (Jayaram et al., 2011). We reported that ARNO-Arf6 signaling axis controls GSIS by promoting sequential activation of Arf6, Cdc42 and Rac1 with 1, 3 and 15min of exposure to stimulatory glucose concentrations, respectively. Based on these findings and data from complementary studies, we concluded that ARNO-mediates the sequential activation of Arf6, Cdc42 and Rac1 culminating in GSIS (Jayaram et al., 2011). Furthermore, using a selective pharmacological inhibitor of ARNO/Arf6 signaling axis (e.g., secinH3), we reported marked inhibition by secinH3 of glucose-induced phospholipase D (PLD) activation, ERK1/2 phosphorylation and dephosphorylation of cofilin, suggesting that Arf6/ARNO signaling mediates PLD, ERK1/2 and cofilin activation in islet β-cells. In addition, secinH3 blocked glucose-induced Nox2 activation and associated ROS generation, thus placing Nox2 downstream to Arf6/ARNO signaling step (Jayaram and Kowluru, 2012). Together, our previous studies identified signaling steps downstream to ARNO/Arf6 axis leading to insulin secretion. Data from our current (proteomics as well as immunological) investigations provide evidence that GIT1/2 is Racl interacting partner and that its association with Rac1 appears to increase significantly under high glucose exposure conditions. In endothelial cells and non-small-cell lung cancer cells, GIT1-PIX axis has been shown to activate Rac1/Cdc42/cortactin to promote membrane protrusions and cell migration (Majumder et al., 2014). Our current investigations also identified cortactin as one of the interaction partners for Rac1 (Table 1). In endothelial cells, GIT2 has been shown to Vav2-mediated activation of Rac1 induced by C-X-C Motif Chemokine Receptor 2 (Li et al., 2018). Indeed, our recent studies have identified Vav2 as one of the GEFs for Rac1 in pancreatic β-cell (Veluthakal et al., 2015). Therefore, it is likely that GIT1/2 play critical roles in regulation of Arf6, Rac1 and Cdc42 in the islet. Methodical investigations are underway to precisely assess the roles of GIT1/2 in islet β-cell function under normal and metabolic stress conditions.

4.2. Potential modulatory roles of β-PIX in islet function

The p21-activated kinase-interacting exchange factor (β-PIX), a known GEF for Racl and Cdc42, has been implicated in the regulation of cell function in many cell types (Rathor et al., 2017; Černohorská et al., 2016). Recent investigations by Kepner et al. (2011) have provided novel insights into regulatory roles of β-PIX in physiological insulin secretion. These studies did not investigate putative roles for β-PIX as a GEF for Rac1 in the islet β-cell. However, published evidence clearly identified Cdc42 activation as upstream to Rac1 activation in the cascade of events leading to GSIS (Wang and Thurmond, 2009; Kowluru, 2010, 2017; Jayaram et al., 2011). Therefore, β-PIX might regulate Rac1 functions via the intermediacy of Cdc42. It is noteworthy, however, that recent investigations by Rathor and associates (Rathor et al., 2017) highlighted novel interactions between β-PIX, GIT1 and Rac1 in intestinal epithelial restitution after wound healing. Based on data from complementary studies, these researchers have demonstrated that increased association between β-PIX and GIT1 results in the stimulation of intestinal epithelial restitution by activating Rac1. Along these lines, studies by Cernohorska and coworkers have demonstrated critical roles for GIT1, β-PIX, and p21 protein (Cdc42/Rac1)-activated kinase 1 in microtubule nucleation. Further, they reported that such a signaling step correlated with recruitment of γ-tubulin to the centrosome. The authors proposed that GIT1/β-PIX signaling proteins with PAK1 kinase represent a novel regulatory mechanism of microtubule nucleation in interphase cells (Černohorská et al., 2016).

4.3. Potential regulatory roles of Gαq/11/14 in islet function and dysfunction

We also identified Gαq as one of the interacting partners of Rac1 in INS-1832/13cells. Our findings also suggested that it is expressed in normal rodent and human islets. Using immunocytochemical approaches Skoglund and associates (Skoglund et al., 1999) have reported expression of Gαq in both α-and β-cells of the islet. Using INS-1E β-cells, Shapiro and coworkers have demonstrated regulatory roles of Gαq in fatty acid (palmitate)-induced insulin secretion (Shapiro et al., 2005). Specifically, they implicated novel roles for Gαq-phospholipase C pathway in GPR40-mediated effects of palmitate in promoting intracellular calcium levels and insulin secretion. These studies provided compelling evidence on the identification of Gαq as the putative trimeric G protein that coupled palmitate to its intracellular regulatory signaling pathways. Along these lines, McNelis et al. (2015) have reported regulatory roles for Gαq in (S)-2-(4-chlorophenyl)-3,3-dimethyl-N-(5-phenylthiazol-2-yl) butanamide, a specific agonist for GPR43, induced inositol triphosphate and calcium levels and associated insulin secretion in Min6 cells, murine islets and human islets. Taken together, these studies implicate roles for Gαq in GPR-coupled signaling events leading to insulin secretion. Published evidence in other cell types highlights potential cross-talk between Gαq and Rac1 signaling pathways in transmission of a variety of signaling events. For example. Harmon and Ratner (2008) have implicated Gαq activation in HIV-1 envelope glycoprotein-mediated Rac1 activation and cytoskeletal rearrangement. Sabbatini and associates (Sabbatini et al., 2010) have demonstrated regulatory roles for Gα13 and Gαq in cholecystokinin-mediated activation of Rac1 in mouse pancreatic acini. Despite this evidence in other cells, potential cross-talk between Gαq and Rac1 has not been established in the islet β-cell function and insulin secretion. Furthermore, it is likely that increased association between these two signaling proteins that we observed in pancreatic β-cells under HG conditions could contribute to the genesis of islet dysfunction. Future studies will validate such a postulation.

4.4. Potential roles of karyopherin and RacGAP1 in islet dysfunction under the duress of HG

Two recent studies from our laboratory have provided evidence to indicate that HG conditions promote alterations in subcellular localization of Rac1, a signaling step that we proposed to contribute to cell dysfunction and demise. In the first, we demonstrated that HG conditions promote translocation, phosphorylation and accumulation of p53, a pro-apoptotic gene, in the nuclear fraction in INS-1832/13 cells. Interestingly, EHT1864, a known inhibitor of Rac1 activation (by specifically inhibiting the GDP/GTP exchange), markedly attenuated phosphorylation, but not translocation of p53. These studies implicate potential regulation of p53 phosphorylation by Rac1 in the nuclear fraction under HG conditions (Syed et al., 2011). Second, using pharmacological and molecular biological approaches, we reported mistargeting of biologically-active Rac1 to the nuclear compartment in INS-1832/13 cells, normal rat islets and human islets following exposure to HG conditions (Baidwan et al., 2017). In this context, a growing body of experimental evidence suggests involvement of specific transport proteins that mediate translocation of cytosolic Rac1 to the nuclear compartment. Sandrock et al. (2010) identified the nuclear import receptor karyopherin alpha2 (KPNA2) as a direct interaction partner of Rac1. Using a variety of experimental techniques they demonstrated that the C-terminal polybasic region of Rac1 contains a nuclear localization signal (NLS), whereas Rac2 and Rac3 lack a functional NLS and do not bind to KPNA2. They demonstrated that the presence of the NLS in Rac1 determines the specificity of the interaction and is a prerequisite for the nuclear import. Compatible with our observations (Li et al., 2018), Sandrock et al. (Veluthakal et al., 2015) demonstrated that although this interaction is independent of the Rac1 GDP/GTP loading, the induction of the translocation requires Rac1 activation. Using LC-MS/MS analysis, these investigators have also identified several proteins in the nuclear fraction that interacted with only biologically-active mutant of Rac1. Some of these interaction partners included IQGAP1, IQGAP2, IQGAP3, PIX, GIT1 and KPNA2. Lastly, our studies have also identified RacGAP1 as one of the interacting partners of Rac1. Earlier studies by Kawashima and Kitamura (2008) have demonstrated that Rac1 and RacGAP1 promote nuclear transport of transcriptional factors (e.g., phospho-STAT3). Interestingly, they demonstrated a critical role for importina (KPNA2) in the nuclear import of p-STAT-3 mediated by Rac1/RacGAP1 (Kawashima and Kitamura, 2008). Taken together, the data accrued in the current investigations (via proteomics and immunological approaches) are expected to provide a platform for future investigations to further validate the roles of these Rac1 interaction partners not only in physiological insulin secretion, but also in the pathogenesis of islet dysfunction under conditions of metabolic stress.

4.5. Identification of mitochondrial matrix proteins as potential interaction partners of Rac1

Even though not significantly enriched, our proteomics data captured a number of mitochondrial proteins as interacting partners of Racl in INS-1832/13 cells. Along these lines, original studies from our laboratory have provided the first evidence for localization of Rho G proteins, specifically Cdc42 and Racl in the mitochondrial fraction derived from insulin-secreting βTC-3 cells (Kowluru et al., 2003). Based on complementary physiological and microscopic studies, we proposed that activation of Racl and Cdc42 may be necessary for coupling mitochondrial events to insulin secretion. Velaithan and associates have demonstrated interaction between Racl and BCL2 in BCL-2 overexpressing B-cell lymphoma cells, which, in turn, has been shown to promote BCL-2 mediated generation of superoxide within the mitochondria (Velaithan et al., 2011). Data accrued from studies by Chong et al. (2018) have implicated Rac1 in the generation of mitochondrial superoxides mediated via activation of NADPH oxidase. More recent investigations from our laboratory have demonstrated inhibition of reactive oxygen species, mitochondrial DNA damage and cell demise in bovine retinal endothelial cells chronically exposed to hyperglycemic conditions (Kowluru et al., 2014). Future studies will assess putative regulatory roles of Rac1 in mitochondrial dysregulation of the islet β-cell under the duress of metabolic stress.

4.6. Potential knowledge gaps and future goals

Several lines of experimental evidence suggest that Rac1 undergoes post-translational modifications including prenylation, carboxylmethylation, palmitoylation, phosphorylation and SUMOylation (Olson, 2018; Kowluru and Kowluru, 2015; Castillo-Lluva et al., 2010). At present, we are unsure whether a particular type and/or the degree of these post-translational modification of Rac1 is altered under HG conditions, which could affect binding of various interacting partners with Rac1. This remains a potential caveat in these investigations. Methodical investigations are underway in our laboratory to systematically address potential effects of post-translational modifications of this GTPase on its interaction with islet endogenous proteins under metabolic stress.

In conclusion, data accrued from our current investigations provide the first global analysis of Rac1 interaction partners in β-cells under basal and hyperglycemic conditions. We identify a number of high glucose-responsive Rac1 interaction partners in INS-1832/13 cells. Our immunological findings in rat and human islets validate proteomics data. We envision that these observations form basis for future investigations of these Rac1 interaction partners in furthering our understanding of regulatory roles of specific signaling pathways in islet β-cell function in health and diabetes.

Abbreviations used

Arf6

ADP ribosylation factor 6

ARNO

Arf6 nucleotide binding site opener

CID

Collision-induced dissociation

Co-IP

Co-immunoprecipitation

FDR

False discovery rate

GAP

GTPase-activating protein

GDI

Guanine nucleotide dissociation inhibitor

GEF

Guanine nucleotide exchange factor

GIT

G protein-coupled receptor kinase-interacting protein-1

GPCR

G protein-coupled receptor

GSIS

Glucose-stimulated insulin secretion

HG

Hyperglycemic

UPLC-ESI-MS/MS

Ultra-performance liquid chromatography-electrospray-tandem mass spectrometry

LTQ

Linear ion trap

NIgG

Normal mouse IgG

NLS

Nuclear localization signal

NOX2

Phagocyte-like

NADPH

oxidase

PA

Peak area

PAK

p21-activated kinase

β-PIX

β-Pak-interacting exchange factor

ROS

Reactive oxygen species

Tiam1

T-cell lymphoma invasion and metastasis-inducing protein 1

Vav2

Vav guanine nucleotide exchange factor 2