Thomas, Philip J. Barter and Kerry-Anne RyeBlake J. Cochran, Radjesh J. Bisoendial, Liming Hou, Elias N. Glaros, Jérémie Rossy, Shane R.

Dependent Mechanism−via a G-Protein-cAMP-PKA-FoxO1 -CellsβApolipoprotein A-I Increases Insulin Secretion and Production From Pancreatic

Print ISSN: 1079-5642. Online ISSN: 1524-4636 Copyright © 2014 American Heart Association, Inc. All rights reserved.

Greenville Avenue, Dallas, TX 75231is published by the American Heart Association, 7272Arteriosclerosis, Thrombosis, and Vascular Biology

doi: 10.1161/ATVBAHA.114.3041312014;

2014;34:2261-2267; originally published online August 21,Arterioscler Thromb Vasc Biol. 

http://atvb.ahajournals.org/content/34/10/2261World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://atvb.ahajournals.org/content/suppl/2014/08/21/ATVBAHA.114.304131.DC1.htmlData Supplement (unedited) at:

  http://atvb.ahajournals.org//subscriptions/

at: is onlineArteriosclerosis, Thrombosis, and Vascular Biology Information about subscribing to Subscriptions:

  http://www.lww.com/reprints

Information about reprints can be found online at: Reprints: 

document. Question and AnswerPermissions and Rightspage under Services. Further information about this process is available in the

which permission is being requested is located, click Request Permissions in the middle column of the WebCopyright Clearance Center, not the Editorial Office. Once the online version of the published article for

can be obtained via RightsLink, a service of theArteriosclerosis, Thrombosis, and Vascular Biologyin Requests for permissions to reproduce figures, tables, or portions of articles originally publishedPermissions:

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

2261

Recent evidence suggests that high-density lipoproteins (HDL) play an important role in glucose metabolism and

that they may protect against the development of type 2 dia-betes mellitus.1 People with type 2 diabetes mellitus typically have lower HDL cholesterol levels than healthy subjects.2,3 Low plasma HDL cholesterol levels are also predictive of progression to type 2 diabetes mellitus in people with pre–diabetes mellitus.4

Increasing HDL levels with reconstituted HDL infusions consisting of apolipoprotein A-I (apoA-I), the main HDL apolipoprotein, complexed with phosphatidylcholine in peo-ple with type 2 diabetes mellitus reduces blood glucose lev-els, increases plasma insulin levels, and improves pancreatic β-cell function.5 Treatment with the cholesteryl ester transfer protein inhibitor, torcetrapib, which increases plasma HDL cholesterol levels by ≈70%, also improves glycemic control in people with type 2 diabetes mellitus.6

ApoA-I seems to be responsible for conferring antidiabetic properties to HDL. Thus, apoA-I–deficient mice have impaired glucose tolerance, decreased skeletal muscle and hepatic AMP-activated protein kinase (AMPK) phosphorylation, as well as increased hepatic glucose production.7 Mice that over-express human apoA-I, by contrast, have improved insulin sen-sitivity and do not become obese when fed a high-fat diet.8 A single infusion of apoA-I also improves glucose tolerance and increases insulin secretion in insulin-resistant mice.9

We have previously reported that treatment of isolated islets and the MIN6 clonal β-cell line with either apoA-I or reconstituted HDL increases insulin secretion and insulin gene transcription under both basal and glucose-stimulated conditions.10 These effects were dependent, at least in part, on the ATP-binding cassette transporters, ATP-binding cassette transporter A1 (ABCA1) and ATP-binding cassette trans-porter G1 (ABCG1), and scavenger receptor class B type 1

© 2014 American Heart Association, Inc.

Arterioscler Thromb Vasc Biol is available at http://atvb.ahajournals.org DOI: 10.1161/ATVBAHA.114.304131

Objective—Therapeutic interventions that increase plasma levels of high-density lipoproteins and apolipoprotein A-I (apoA-I) A-I, the major high-density lipoprotein apolipoprotein, improve glycemic control in people with type 2 diabetes mellitus. High-density lipoproteins and apoA-I also enhance insulin synthesis and secretion in isolated pancreatic islets and clonal β-cell lines. This study identifies the signaling pathways that mediate these effects.

Approach and Results—Incubation with apoA-I increased cAMP accumulation in Ins-1E cells in a concentration-dependent manner. The increase in cAMP levels was inhibited by preincubating the cells with the cell-permeable, transmembrane adenylate cyclase inhibitor, 2′5′ dideoxyadenosine, but not with KH7, which inhibits soluble adenylyl cyclases. Incubation of Ins-1E cells with apoA-I resulted in colocalization of ATP-binding cassette transporter A1 with the Gαs subunit of a heterotrimeric G-protein and a Gαs subunit-dependent increase in insulin secretion. Incubation of Ins-1E cells with apoA-I also increased protein kinase A phosphorylation and reduced the nuclear localization of forkhead box protein O1 (FoxO1). Preincubation of Ins-1E cells with the protein kinase A–specific inhibitors, H89 and PKI amide, prevented apoA-I from increasing insulin secretion and mediating the nuclear exclusion of FoxO1. Transfection of Ins-1E cells with a mutated FoxO1 that is restricted to the nucleus confirmed the requirement for FoxO1 nuclear exclusion by blocking insulin secretion in apoA-I–treated Ins-1E cells. ApoA-I also increased Irs1, Irs2, Ins1, Ins2, and Pdx1 mRNA levels.

Conclusions—ApoA-I increases insulin synthesis and secretion via a heterotrimeric G-protein-cAMP-protein kinase A-FoxO1–dependent mechanism that involves transmembrane adenylyl cyclases and increased transcription of key insulin response and β-cell survival genes. (Arterioscler Thromb Vasc Biol. 2014;34:2261-2267.)

Key Words: apolipoprotein A-I ◼ diabetes mellitus, type 2

Received on: October 6, 2013; final version accepted on: August 12, 2014.From the Lipid Research Group, Centre for Vascular Research (B.J.C., R.J.B., L.H., P.J.B., K.-A.R.) and Redox Cell Signaling Group, School of Medical

Sciences (E.N.G., S.R.T.), University of New South Wales, Sydney, Australia; Lipid Research Group, Heart Research Institute, Sydney, Australia (B.J.C., R.J.B., L.H., P.J.B., K.-A.R.); Australian Centre for Nanomedicine, Sydney, Australia (J.R.); Immune Imaging, Centenary Institute, Sydney, Australia (R.J.B.); and Faculty of Medicine, University of Sydney, Sydney, Australia (P.J.B., K.-A.R.).

The online-only Data Supplement is available with this article at http://atvb.ahajournals.org/lookup/suppl/doi:10.1161/ATVBAHA.114.304131/-/DC1.Correspondence to Kerry-Anne Rye, PhD, Lipid Research Group, Centre for Vascular Research, University of New South Wales, Sydney, New South

Wales, Australia 2052. E-mail k.rye@unsw.edu.au or karye@ozemail.com.au

Apolipoprotein A-I Increases Insulin Secretion and Production From Pancreatic β-Cells via a G-Protein-cAMP-

PKA-FoxO1–Dependent MechanismBlake J. Cochran, Radjesh J. Bisoendial, Liming Hou, Elias N. Glaros, Jérémie Rossy,

Shane R. Thomas, Philip J. Barter, Kerry-Anne Rye

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

2262 Arterioscler Thromb Vasc Biol October 2014

(SR-B1).10 However, the mechanisms underlying these effects remain uncharacterized. Given the current interest in apoA-I- and HDL-raising therapeutics, and their potential to improve glycemic control in people with type 2 diabetes mellitus,6 determining the mechanism by which apoA-I increases insu-lin synthesis and secretion without having deleterious effects on β-cell function is clearly of importance.

Previous reports in different cell types have provided clues about signaling pathways that may explain the antidiabetic properties of apoA-I. For example, the interaction of apoA-I with ABCA1 activates cAMP in cultured human fibroblasts and Chinese hamster ovary cells.11 ApoA-I also activates protein kinase A (PKA) and increases ABCA1 phosphoryla-tion and expression in RAW macrophages and fibroblasts.11,12 Other studies have established that apoA-I activates janus kinase 2 in baby hamster kidney cells,13 AMPK in C2C12 myocytes,7 and protein kinase C14 and cell division control protein 42 signaling in human fibroblasts.15 We, and others, have further reported that apoA-I regulates intracellular Ca2+ levels in MIN6 cells, baby hamster kidney cells, and RAW macrophages.10,16

The present study establishes that apoA-I increases insulin secretion in the Ins-1E clonal β-cell line and isolated primary islets via a heterotrimeric G-protein–cAMP–PKA signaling pathway. The results also show that the apoA-I–mediated increase in insulin synthesis in Ins-1E cells involves nuclear exclusion of forkhead box protein O1 (FoxO1) and increased transcription of insulin response genes and genes involved in β-cell function and survival.

Materials and MethodsMaterials and Methods are available in the online-only Supplement.

ResultsApoA-I Increases Insulin Secretion From Ins-1E Cells in an ABCA1- and SR-BI–Dependent MannerWe have previously reported that apoA-I increases insulin secretion from MIN6 cells.10 To confirm that this is also the case for Ins-1E cells, incubations were performed with increas-ing concentrations of apoA-I (Figure I in the online-only Data Supplement). After 1 hour of incubation with apoA-I (final concentration, 0.125 mg/mL), the insulin concentration in the medium increased from 46.5±5.6 to 89.1±10.1 ng insulin/mg protein (P<0.001). Incubation for 1 hour in the presence of apoA-I at a final concentration of 1 mg/mL further increased the insulin concentration in the medium to 164.4±9.0 ng insu-lin/mg protein (P<0.001).

We have also reported that apoA-I increases insulin secretion from MIN6 cells in an ABCA1- and SR-BI–dependent man-ner.10 To confirm this in Ins-1E cells, transfections were per-formed with scrambled small interfering RNA (siRNA), ABCA1 siRNA, SR-BI siRNA, and ABCG1 siRNA, which decreased ABCA1, SR-B1, and ABCG1 protein expression by 56.8±4.6%, 74.0±7.94%, and 70.1±5.2%, respectively (Figure IIA and IIB in the online-only Data Supplement). Transfection of Ins-1E cells with scrambled siRNA or ABCG1 siRNA had no effect on apoA-I–mediated insulin secretion. When, by contrast, ABCA1 and SR-BI were knocked down, insulin secretion decreased from 165.0±25.4 to 59.5±13.7 and 52.0±15.4 ng insulin/mg protein, respectively (P<0.005 for both versus scrambled siRNA-trans-fected cells; Figure IIC in the online-only Data Supplement).

As reported previously,10 incubation of Ins-1E cells with apoA-I under basal and high glucose conditions did not alter cellular cholesterol levels (Table II in the online-only Data Supplement). Although this suggests that apoA-I increases insulin secretion from Ins-1E cells by a mechanism that is inde-pendent of cellular cholesterol depletion, the possibility that the insulin secretion was increased because of alterations in the subcellular localization of cholesterol cannot be excluded.

ApoA-I Increases Insulin Secretion From Ins-1E Cells Independent of the Akt/Protein Kinase B, Janus Kinase 2, Protein Kinase C, and AMPK Signaling PathwaysBecause the Akt/protein kinase B (PKB) signaling pathway plays a key role in β-cell function and insulin secretion,17 and this pathway is activated by HDL in endothelial cells,18 we next asked whether this could explain the increased insulin secre-tion in apoA-I–treated Ins-1E cells. Figure III in the online-only Data Supplement shows that apoA-I does not increase Akt phosphorylation in Ins-1E cells under basal (2.8 mmol/L) glucose conditions. Similarly, apoA-I did not activate the protein kinase C, janus kinase 2, or AMPK pathways7,13,14 in Ins-1E cells (Figure III in the online-only Data Supplement).

ApoA-I Increases cAMP Accumulation in Ins-1E Cells in a Transmembrane Adenylyl Cyclase– Dependent MannerBecause apoA-I increases cAMP levels in fibroblasts and Chinese hamster ovary cells,11 cAMP levels were also quantified in Ins-1E cells. These experiments were conducted under low-glucose conditions to minimize glucose-mediated cAMP induction.19 Incubation with apoA-I increased Ins-1E cAMP levels in a con-centration-dependent manner (Figure 1A). At final apoA-I con-centrations of 0.5 and 1.0 mg/mL, Ins-1E cAMP levels increased from 20.1±4.8 to 57.6±5.9 pmol/mg cell protein (P<0.01) and 95.5±12.1 pmol/mg cell protein (P<0.001), respectively.

To determine whether the increase in cAMP levels involved transmembrane or soluble adenylyl cyclases, Ins-1E cells were preincubated with 2′5′ dideoxyadenosine or KH7, which respectively inhibit transmembrane and soluble adenylyl cyclases, before addition of apoA-I (Figure 1B). Incubation with apoA-I increased Ins-1E cAMP levels from 45.3±4.7 to 119.2±42.6 pmol/mg cell protein in the control cells, (P<0.05). When the cells were treated with KH7 before incubation with

Nonstandard Abbreviations and Acronyms

ABCA1 ATP-binding cassette transporter A1

AMPK AMP-activated protein kinase

apoA-I apolipoprotein A-I

FoxO1 forkhead box protein O1

HDL high-density lipoprotein(s)

PKA protein kinase A

SR-B1 scavenger receptor class B type 1

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

Cochran et al ApoA-I and Pancreatic β-Cell Signaling 2263

apoA-I, cAMP levels increased from 32.8±0.9 to 125.0±26.7 pmol/mg cell protein (P<0.01). Pretreatment with the trans-membrane adenylyl cyclase inhibitor 2′5′ dideoxyadenosine, by contrast, abolished the apoA-I–mediated increase in cAMP.

ApoA-I Increases Colocalization of ABCA1 With the Heterotrimeric G-Protein Gαs SubunitBecause the Gα

s subunit of heterotrimeric G-proteins can acti-

vate adenylyl cyclases, we next asked whether this is the link

between the interaction of apoA-I with ABCA1 and transmem-brane adenylyl cyclase activation. Transfection of Ins-1E cells with heterotrimeric G-protein Gα

s subunit siRNA reduced Gα

s

protein levels by 62.2±9.7% relative to control (Figure 2A and 2B). Incubation with apoA-I increased insulin secretion in the scrambled siRNA-transfected and -nontransfected cells to the same extent (Figure 2C). When cells in which the Gα

s sub-

unit was knocked down were incubated with apoA-I, insulin secretion decreased from 189.6±25.2 ng/mg protein (control) to 98.35±25.96 ng/mg protein (P<0.001 versus scrambled siRNA-transfected cells; Figure 2C). Incubation with apoA-I also increased the colocalization of ABCA1 and the Gα

s sub-

unit (Figure 2D), with the Pearson coefficient increasing from 0.04±0.03 to 0.17±0.05 in Ins-1E cells incubated with PBS (P<0.001 versus negative control) and to 0.3±0.04 in cells incubated with apoA-I (P<0.001 versus PBS; Figure 2E). The specificity of the anti-ABCA1 antibody was confirmed by establishing that colocalization of ABCA1 and the Gα

s subunit

was minimal when ABCA1 was knocked down in Ins-1E cells (Figure IV in the online-only Data Supplement).

ApoA-I Increases Insulin Secretion From Ins-1E Cells and Primary Islets in a PKA-Dependent MannerTo determine if PKA was activated by the increase in cAMP levels, Ins-1E cells were incubated with apoA-I under basal conditions. This increased PKA phosphorylation by 103.9±19.6% at 15 minutes (Figure 3A and 3B; P<0.001 versus control) and by 54.2±19.6% (P<0.001 versus control) at 30 minutes. PKA phosphorylation levels had returned to baseline by 60 minutes. Incubation of primary mouse islets with apoA-I for 15 minutes increased PKA phosphorylation by 76.4±15.1% (P<0.01; Figure 3C and 3D).

Incubations were also performed with the PKA-specific inhibitors, H89 and PKI amide. Incubation of Ins-1E cells with

Figure 1. Apolipoprotein A-I (ApoA-I) increases cAMP levels in Ins-1E cells in a transmembrane adenylyl cyclase–dependent manner. A, Ins-1E cells were incubated for 15 minutes with apoA-I (final concentration, 0.125–1 mg/mL) in the presence of 0.5 mmol/L IBMX. B, Ins-1E cells were preincubated for 30 minutes with 30 μmol/L KH7 and 50 μmol/L 2′5′ dideoxyadenosine (ddAdo) or PBS, then incubated for 15 minutes in the presence of 0.5 mmol/L IBMX with (black bars) or without (white bars) apoA-I (final concen-tration, 1 mg/mL). cAMP was extracted from cells with 0.1 mol/L HCl and quantified by ELISA. Values represent the mean±SD (n=4) of 3 separate experiments. *P<0.05, **P<0.01, ****P<0.001.

Figure 2. Apolipoprotein A-I (ApoA-I) induces cocalization of ATP-binding cassette transporter A1 (ABCA1) and a heterotrimeric Gαs subunit. A, Ins-1E cells were transfected with Gαs siRNA or scrambled siRNA, then lysed and subjected to Western blot-ting for Gαs and β-actin. B, Quantification of Gαs expression in siRNA-transfected cells. C, Trans-fected Ins-1E cells were incubated for 1 hour with-out (white bars) or with (black bars) apoA-I (final concentration, 1 mg/mL) under basal (2.8 mmol/L) glucose conditions. Insulin levels in the medium were quantified by radioimmunoassay. D, Ins-1E cells were seeded on coverslips, transfected with a vector encoding for hemagglutinin (HA)-Gαs, serum starved for 6 hours, then incubated for 5 minutes with or without apoA-I (final concentration, 1 mg/mL). The cells were fixed, immunostained with a DyLight-488–labeled ABCA1 antibody and an Alexa-657–labeled HA-tag antibody, and imaged using a total internal reflection fluorescence micro-scope. Representative images from 3 separate experiments are shown. Arrows indicate points of colocalization. Scale bar, 20 μm. E, Quantification of the Pearson coefficient. All values represent the mean±SD (n=4 for B and C, n=20 fields of view for E). ****P<0.001 compared with controls.

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

2264 Arterioscler Thromb Vasc Biol October 2014

apoA-I for 1 hour under basal conditions increased the concen-tration of insulin in the medium from 61.1±8.5 to 171.6±12.7 ng insulin/mg cell protein (Figure 3E, white bars, P<0.001). Preincubation of the cells with H89 and PKI amide (Figure 3E, black and gray bars, respectively) abolished the apoA-I–medi-ated increase in insulin secretion. Preincubation of Ins-1E cells (Figure 3F) and primary islets (Figure 3G) with H89 and PKI amide under high glucose conditions also abolished the apoA-I–mediated increase in insulin secretion (Figure 3F).

ApoA-I Mediates Nuclear Translocation of FoxO1 in a PKA-Dependent MannerAs reported previously, 94.9±8.4% of the total cellular FoxO1 was localized in the nucleus of Ins-1E cells after 6 hours of serum starvation (Figure 4A, top).20 The intracellular distri-bution of FoxO1 was not affected by incubation with H89 for 30 minutes. When, by contrast, the cells were incubated with apoA-I, only 14.3±18.8% of the FoxO1 remained in the nucleus (P<0.001 versus control). Preincubation of the cells with H89 before incubation with apoA-I increased the nuclear localization of FoxO1 to 86.8±12.7% (P<0.001 versus incu-bation with apoA-I alone).

Ability of ApoA-I to Increase Long-Term Insulin Secretion From Ins-1E Cells Is Dependent on Nuclear Exclusion of FoxO1To determine whether the ability of apoA-I to increase long-term insulin production depends on continued nuclear exclu-sion of FoxO1, Ins-1E cells were transiently transfected with pCMV5 vectors encoding for wild-type FoxO1, FoxO1-ADA (a transcriptionally active FoxO1 mutant that resides exclusively in the nucleus), or an empty pCMV5 vector. FoxO1 protein expression was comparable in the FoxO1 and

FoxO1-ADA transfected cells (Figure 5A and 5B). Incubation of the transfected cells with PBS had no effect on insulin secre-tion (Figure 5C, white bars). Incubation with apoA-I increased the insulin concentration in the medium from 287.8±62.4 and 259.6±59.0 to 836.2±34.5 and 789.0±98.5 ng insulin/mg cell protein in the cells transfected with empty vector and the vec-tor encoding for FoxO1, respectively (P<0.001 versus PBS for both; Figure 5C, black bars). Insulin secretion was reduced to 401.02±55.69 ng insulin/mg cell protein when the FoxO1-ADA–transfected cells were incubated with apoA-I (P<0.001).

ApoA-I Increases Expression of FoxO1 Target Genes and β-Cell Survival GenesAs reported previously,10 Ins1 and Ins2 mRNA levels increased 2.1-fold (P<0.001), and 2.8-fold (P<0.005), respectively, when Ins-1E cells were incubated for 16 hours with apoA-I. Insulin receptor substrate 1 (IRS1) and IRS2 mRNA levels increased 4.7-fold (P<0.01) and 2.9-fold, respectively (P<0.05) (Figure 6). Nuclear exclusion of FoxO1 by apoA-I also increased tran-scription of the β-cell survival factor Pdx1 5.3-fold (P<0.001).21

DiscussionEmerging evidence suggests that increasing plasma HDL cholesterol and apoA-I levels improves glycaemic control in people with type 2 diabetes mellitus.5,6 The present study indicates that this may relate to the ability of HDL and apoA-I to increase insulin secretion and production in pancreatic β-cells9,10,22 and the activation of a G-protein–cAMP–PKA–FoxO1 signaling pathway. The results also establish that apoA-I increases transcription of insulin production genes and expression of the β-cell survival gene, Pdx1.

Insulin secretion is impaired in pancreatic β-cells with elevated cholesterol levels,23 and this detrimental effect can be rectified by

Figure 3. Apolipoprotein A-I (ApoA-I) increases insulin secretion via a protein kinase A (PKA)–dependent mechanism. A, Ins-1E cells were cultured in serum-free medium for 12 hours, then incubated with apoA-I (final concentration, 1 mg/mL) for 5 to 60 minutes. Cell lysates were Western blotted for phosphorylated PKA (pPKA), total PKA, and β-actin. B, Quantification of PKA phosphoryla-tion in Ins-1E cells incubated for 5 to 60 minutes in the absence or presence of apoA-I. C, Isolated primary mouse islets (n=50) were cultured in serum-free medium for 24 hours, then incubated with apoA-I (final concentration, 1 mg/mL) for 15 min-utes. Cell lysates were Western blotted for pPKA, total PKA, and β-actin. D, Quantification of PKA phosphorylation in isolated primary islets. E and F, Ins-1E cells were preincubated for 30 minutes with H89 (final concentration, 20 μmol/L, black bars), PKI amide (final concentration, 5 μmol/L, gray bars), or an equivalent volume of dimethyl sulfoxide (DMSO; white bars), then incubated for 1 hour with apoA-I (final concentration, 1 mg/mL) or PBS under low (2.8 mmol/L, E) or high (25 mmol/L, F) glu-cose conditions. Insulin levels in the medium were quantified by radioimmunoassay (RIA). G, Primary mouse islets were preincubated for 30 minutes with H89 (final concentration, 20 μmol/L, black bars), PKI amide (final concentration, 5 μmol/L, gray bars),

or DMSO (white bars) in Krebs-Ringer bicarbonate HEPES (KRBH) buffer containing 1.67 mmol/L glucose, then incubated for 1 hour with apoA-I (final concentration, 1 mg/mL) or PBS in KRBH buffer containing 16.7 mmol/L glucose. Insulin levels in the medium were quanti-fied by RIA. Values represent the mean±SD (n=4) of 3 separate experiments. **P<0.01, ***P<0.005, ****P<0.001 compared with control.

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

Cochran et al ApoA-I and Pancreatic β-Cell Signaling 2265

removing cholesterol from the cells with agents such as cyclo-dextrin that act as cholesterol scavengers. Because apoA-I is the main acceptor of cholesterol that is exported from cells via ABCA1 and SR-B1, depletion of excess cholesterol is a potential mechanism by which apoA-I may have enhanced insulin secre-tion in Ins-1E cells. However, we have reported previously, and confirm that this is not the case in the present study, as incuba-tion with apoA-I increased insulin secretion in an ABCA1- and SR-B1–dependent manner without altering cellular cholesterol levels.10 This led us to ask whether apoA-I increases β-cell insu-lin secretion by activating cell signaling pathways.

Because the phosphoinositide 3-kinase (PI3K)/Akt signal transduction pathway plays a key role in the response of pan-creatic β-cells to mediators of insulin release,17 and this pathway is also activated when HDL bind to SR-B1 in endothelial cells24 and mononuclear cells,25 this was explored in the first instance. However, we found no evidence of activation of the PI3K/Akt by apoA-I in Ins-1E cells. Similarly, incubation of Ins-1E cells with apoA-I did not activate the protein kinase C, janus kinase 2, or AMPK pathways. This highlights the complex nature of these pathways in insulin secretion and their potential inhibition by phosphatases such as protein phosphatase 2A26,27 and negative insulin signaling feedback loops. It is also possible that apoA-I does not bind to Ins-1E cells with an affinity that is sufficient to activate these pathways.

Because activation of adenylyl cyclases, the secondary messenger cAMP, and PKA can also release insulin from β-cells28,29 and improve β-cell survival by increasing expres-sion of IRS-230 and reducing apoptosis,31 this pathway was also explored. Support for involvement of this pathway in the apoA-I–mediated increase in insulin secretion in Ins-1E cells comes from studies showing that HDL increase intracellular cAMP levels in cholesterol-loaded fibroblasts.32 The results showing that apoA-I activates ≥1 transmembrane adenylyl cyclases, increases intracellular cAMP levels, and activates PKA in Ins-1E cells are consistent with insulin secretion being increased by activation of this pathway. It is noteworthy that this is the same mechanism by which the incretin, glucagon-like peptide-1, increases β-cell insulin synthesis and secretion.33

One of the key findings to emerge from the present study is that the interaction of apoA-I with ABCA1 increases insu-lin secretion by activating a transmembrane adenylyl cyclase via the Gαs subunit of a heterotrimeric G-protein. Sequence analysis shows that ABCA1 shares a striking homology

Figure 4. Apolipoprotein A-I (ApoA-I) mediates nuclear exclusion of forkhead box protein O1 (FoxO1) in a protein kinase A–dependent manner. A, Ins-1E cells were seeded on coverslips, serum starved for 6 hours, preincubated with H89 (final concentration, 20 μmol/L) or an equivalent volume of dimethyl sulfoxide (DMSO) for 30 min-utes, then incubated for 30 minutes with or without apoA-I (final concentration, 1 mg/mL). The cells were fixed and immunostained with an anti-FoxO1 antibody, an Alexa-488–labeled secondary antibody, and DAPI (4’,6-diamidino-2-phenylindole). Representa-tive images of FoxO1 distribution from 3 separate experiments are shown. Scale bar, 20 μm. B, Quantification of nuclear FoxO1 local-ization. Values represent the mean±SD (n=30 cells). ****P<0.001 compared with controls.

Figure 5. Nuclear exclusion of forkhead box protein O1 (FoxO1) is required for increased long-term insulin secretion by apolipoprotein A-I (apoA-I). A, Ins-1E cells were transfected with vectors encoding wild-type FoxO1, constitutively expressed nuclear FoxO1-ADA or an empty vector. B, Western blotting and quantification of whole cell lysates for FoxO1 24 and 36 hours after transfection. C, Ins-1E cells were transfected with empty vector or vectors encoding wild-type FoxO1 or FoxO1-ADA, then incubated for 24 hours. The transfected cells were incubated for an additional 16 hours at 37°C in low-glucose (2.8 mmol/L) RPMI (Roswell Park Memorial Insti-tute)-1640 medium in the absence (white bars) or presence (black bars) of apoA-I (1 mg/mL). Insulin levels in the medium were deter-mined by RIA. Values represent the mean±SD (n=3) for 3 separate experiments. ****P<0.001 vs control.

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

2266 Arterioscler Thromb Vasc Biol October 2014

with regions of the glucagon-like peptide-1 receptor and the glucose-dependent insulinotropic polypeptide receptor, which also bind to the Gα

s heterotrimeric G-protein subunit34,35

(Figure V in the online-only Data Supplement). This binding site consists of a KLK/R motif in the first intracellular domain of ABCA1 that is N-terminally flanked by several hydropho-bic residues. When this is considered in light of the obser-vation that apoA-I also activates the cAMP-PKA signaling pathway, it follows that the immediate (1 hour) glucose-medi-ated increase in insulin secretion in Ins-1E cells is unlikely to involve K+

ATP channel closure,10,28 FoxO1 localization, or

increased insulin synthesis.Because PKA can also directly phosphorylate FoxO1 in

endothelial cells,36 we asked whether this might explain the previously reported apoA-I–mediated increase in insulin syn-thesis in MIN6 cells10 and in Ins-1E cells in the present study. The result in Figure 4 showing that apoA-I excludes FoxO1 from the nucleus of Ins-1E cells is consistent with this notion. It also agrees with the observed activation of transcription fac-tors that upregulate the Ins1 and Ins2 genes37 and the increase in Pdx1 mRNA levels. The combined effects of these events increase insulin gene transcription.38 As described previ-ously,39 this suggests that nuclear FoxO1 functions as an insu-lin repressor in Ins-1E cells. The ability of apoA-I to increase Pdx1 gene transcription also has the capacity to improve β-cell survival.40

In conclusion, this study sheds new light on the molecular basis by which apoA-I increases insulin synthesis and secre-tion in pancreatic β-cells and provides an insight as to why therapies that increase apoA-I and HDL levels improve gly-cemic control in people with type 2 diabetes mellitus and the metabolic syndrome. The results also identify apoA-I, and possibly apoA-I mimetic peptides, as incretin-like molecules that are potentially attractive agents for treating people with type 2 diabetes mellitus and poor glycaemic control.

AcknowledgmentsB.J. Cochran and K.-A. Rye designed the studies, analyzed and interpreted the results, and prepared the article. B.J. Cochran, R.J. Bisoendial, L. Hou, E.N. Glaros, and J. Rossy acquired the results. K.-A. Rye is the guarantor of this work and as such had full access to all the data in the study and takes responsibility for the integrity

of the data and the accuracy of the data analysis. P.J. Barter and S.R. Thomas interpreted the results and assisted with article preparation.

Sources of FundingThis work was supported by the National Health and Medical Research Council of Australia (grant 482800).

DisclosuresNone.

References 1. Drew BG, Rye KA, Duffy SJ, Barter P, Kingwell BA. The emerging role

of HDL in glucose metabolism. Nat Rev Endocrinol. 2012;8:237–245. 2. Drexel H, Aczel S, Marte T, Benzer W, Langer P, Moll W, Saely CH. Is

atherosclerosis in diabetes and impaired fasting glucose driven by ele-vated LDL cholesterol or by decreased HDL cholesterol? Diabetes Care. 2005;28:101–107.

3. Gatti A, Maranghi M, Bacci S, Carallo C, Gnasso A, Mandosi E, Fallarino M, Morano S, Trischitta V, Filetti S. Poor glycemic control is an indepen-dent risk factor for low HDL cholesterol in patients with type 2 diabetes. Diabetes Care. 2009;32:1550–1552.

4. Haffner SM, Stern MP, Hazuda HP, Mitchell BD, Patterson JK. Cardiovascular risk factors in confirmed prediabetic individuals. Does the clock for coronary heart disease start ticking before the onset of clinical diabetes? JAMA. 1990;263:2893–2898.

5. Drew BG, Duffy SJ, Formosa MF, et al. High-density lipoprotein mod-ulates glucose metabolism in patients with type 2 diabetes mellitus. Circulation. 2009;119:2103–2111.

6. Barter PJ, Rye KA, Tardif JC, Waters DD, Boekholdt SM, Breazna A, Kastelein JJ. Effect of torcetrapib on glucose, insulin, and hemoglo-bin A1c in subjects in the Investigation of Lipid Level Management to Understand its Impact in Atherosclerotic Events (ILLUMINATE) trial. Circulation. 2011;124:555–562.

7. Han R, Lai R, Ding Q, Wang Z, Luo X, Zhang Y, Cui G, He J, Liu W, Chen Y. Apolipoprotein A-I stimulates AMP-activated protein kinase and improves glucose metabolism. Diabetologia. 2007;50:1960–1968.

8. Ruan X, Li Z, Zhang Y, Yang L, Pan Y, Wang Z, Feng GS, Chen Y. Apolipoprotein A-I possesses an anti-obesity effect associated with increase of energy expenditure and up-regulation of UCP1 in brown fat. J Cell Mol Med. 2011;15:763–772.

9. Stenkula KG, Lindahl M, Petrlova J, Dalla-Riva J, Göransson O, Cushman SW, Krupinska E, Jones HA, Lagerstedt JO. Single injections of apoA-I acutely improve in vivo glucose tolerance in insulin-resistant mice. Diabetologia. 2014;57:797–800.

10. Fryirs MA, Barter PJ, Appavoo M, Tuch BE, Tabet F, Heather AK, Rye KA. Effects of high-density lipoproteins on pancreatic beta-cell insulin secretion. Arterioscler Thromb Vasc Biol. 2010;30:1642–1648.

11. Haidar B, Denis M, Marcil M, Krimbou L, Genest J Jr. Apolipoprotein A-I activates cellular cAMP signaling through the ABCA1 transporter. J Biol Chem. 2004;279:9963–9969.

12. See RH, Caday-Malcolm RA, Singaraja RR, et al. Protein kinase A site-specific phosphorylation regulates ATP-binding cassette A1 (ABCA1)-mediated phospholipid efflux. J Biol Chem. 2002;277:41835–41842.

13. Vaughan AM, Tang C, Oram JF. ABCA1 mutants reveal an interdepen-dency between lipid export function, apoA-I binding activity, and Janus kinase 2 activation. J Lipid Res. 2009;50:285–292.

14. Yamauchi Y, Hayashi M, Abe-Dohmae S, Yokoyama S. Apolipoprotein A-I activates protein kinase C alpha signaling to phosphorylate and stabi-lize ATP binding cassette transporter A1 for the high density lipoprotein assembly. J Biol Chem. 2003;278:47890–47897.

15. Nofer JR, Remaley AT, Feuerborn R, Wolinnska I, Engel T, von Eckardstein A, Assmann G. Apolipoprotein A-I activates Cdc42 signaling through the ABCA1 transporter. J Lipid Res. 2006;47:794–803.

16. Karwatsky J, Ma L, Dong F, Zha X. Cholesterol efflux to apoA-I in ABCA1-expressing cells is regulated by Ca2+-dependent calcineurin sig-naling. J Lipid Res. 2010;51:1144–1156.

17. Bernal-Mizrachi E, Fatrai S, Johnson JD, Ohsugi M, Otani K, Han Z, Polonsky KS, Permutt MA. Defective insulin secretion and increased sus-ceptibility to experimental diabetes are induced by reduced Akt activity in pancreatic islet beta cells. J Clin Invest. 2004;114:928–936.

Figure 6. Effects of apolipoprotein A-I (apoA-I) on gene expres-sion in Ins-1E cells. Ins-1E cells were incubated for 16 hours at 37°C in RPMI-1640 medium in the absence (white bars) or presence of apoA-I (final concentration, 1 mg/mL, black bars). Insulin receptor substrate 1 (IRS1), IRS2, Ins1, Ins2, and Pdx1 mRNA levels were determined by quantitative reverse transcrip-tase polymerase chain reaction. Values represent the mean±SD (n=3) for 3 separate experiments. *P<0.05, **P<0.01, ***P<0.005, ****P<0.0001 vs control.

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

Cochran et al ApoA-I and Pancreatic β-Cell Signaling 2267

18. Mineo C, Yuhanna IS, Quon MJ, Shaul PW. High density lipoprotein-induced endothelial nitric-oxide synthase activation is mediated by Akt and MAP kinases. J Biol Chem. 2003;278:9142–9149.

19. Ramos LS, Zippin JH, Kamenetsky M, Buck J, Levin LR. Glucose and GLP-1 stimulate cAMP production via distinct adenylyl cyclases in INS-1E insulinoma cells. J Gen Physiol. 2008;132:329–338.

20. Martinez SC, Cras-Méneur C, Bernal-Mizrachi E, Permutt MA. Glucose regulates Foxo1 through insulin receptor signaling in the pancreatic islet beta-cell. Diabetes. 2006;55:1581–1591.

21. Kitamura T, Nakae J, Kitamura Y, Kido Y, Biggs WH 3rd, Wright CV, White MF, Arden KC, Accili D. The forkhead transcription factor Foxo1 links insulin signaling to Pdx1 regulation of pancreatic beta cell growth. J Clin Invest. 2002;110:1839–1847.

22. Siebel AL, Natoli AK, Yap FY, Carey AL, Reddy-Luthmoodoo M, Sviridov D, Weber CI, Meneses-Lorente G, Maugeais C, Forbes JM, Kingwell BA. Effects of high-density lipoprotein elevation with cholesteryl ester transfer protein inhibition on insulin secretion. Circ Res. 2013;113:167–175.

23. Hao M, Head WS, Gunawardana SC, Hasty AH, Piston DW. Direct effect of cholesterol on insulin secretion: a novel mechanism for pancreatic beta-cell dysfunction. Diabetes. 2007;56:2328–2338.

24. Seetharam D, Mineo C, Gormley AK, Gibson LL, Vongpatanasin W, Chambliss KL, Hahner LD, Cummings ML, Kitchens RL, Marcel YL, Rader DJ, Shaul PW. High-density lipoprotein promotes endothelial cell migration and reendothelialization via scavenger receptor-B type I. Circ Res. 2006;98:63–72.

25. Sumi M, Sata M, Miura S, Rye KA, Toya N, Kanaoka Y, Yanaga K, Ohki T, Saku K, Nagai R. Reconstituted high-density lipoprotein stimulates dif-ferentiation of endothelial progenitor cells and enhances ischemia-induced angiogenesis. Arterioscler Thromb Vasc Biol. 2007;27:813–818.

26. Ammälä C, Eliasson L, Bokvist K, Berggren PO, Honkanen RE, Sjöholm A, Rorsman P. Activation of protein kinases and inhibition of protein phos-phatases play a central role in the regulation of exocytosis in mouse pan-creatic beta cells. Proc Natl Acad Sci U S A. 1994;91:4343–4347.

27. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB. Exploiting the PI3K/AKT pathway for cancer drug discovery. Nat Rev Drug Discov. 2005;4:988–1004.

28. Yajima H, Komatsu M, Schermerhorn T, Aizawa T, Kaneko T, Nagai M, Sharp GW, Hashizume K. cAMP enhances insulin secretion by an action on the ATP-sensitive K+ channel-independent pathway of glucose signal-ing in rat pancreatic islets. Diabetes. 1999;48:1006–1012.

29. Yang S, Fransson U, Fagerhus L, Holst LS, Hohmeier HE, Renström E, Mulder H. Enhanced cAMP protein kinase A signaling determines

improved insulin secretion in a clonal insulin-producing beta-cell line (INS-1 832/13). Mol Endocrinol. 2004;18:2312–2320.

30. Jhala US, Canettieri G, Screaton RA, Kulkarni RN, Krajewski S, Reed J, Walker J, Lin X, White M, Montminy M. cAMP promotes pancreatic beta-cell survival via CREB-mediated induction of IRS2. Genes Dev. 2003;17:1575–1580.

31. Mohanty S, Spinas GA, Maedler K, Zuellig RA, Lehmann R, Donath MY, Trüb T, Niessen M. Overexpression of IRS2 in isolated pancreatic islets causes proliferation and protects human beta-cells from hyperglycemia-induced apoptosis. Exp Cell Res. 2005;303:68–78.

32. Haidar B, Denis M, Krimbou L, Marcil M, Genest J Jr. cAMP induces ABCA1 phosphorylation activity and promotes cholesterol efflux from fibroblasts. J Lipid Res. 2002;43:2087–2094.

33. Wang X, Zhou J, Doyle ME, Egan JM. Glucagon-like peptide-1 causes pancreatic duodenal homeobox-1 protein translocation from the cyto-plasm to the nucleus of pancreatic beta-cells by a cyclic adenosine monophosphate/protein kinase A-dependent mechanism. Endocrinology. 2001;142:1820–1827.

34. Mathi SK, Chan Y, Li X, Wheeler MB. Scanning of the glucagon-like pep-tide-1 receptor localizes G protein-activating determinants primarily to the N terminus of the third intracellular loop. Mol Endocrinol. 1997;11:424–432.

35. Takhar S, Gyomorey S, Su RC, Mathi SK, Li X, Wheeler MB. The third cytoplasmic domain of the GLP-1[7-36 amide] receptor is required for coupling to the adenylyl cyclase system. Endocrinology. 1996;137:2175–2178.

36. Lee JW, Chen H, Pullikotil P, Quon MJ. Protein kinase A-alpha directly phosphorylates FoxO1 in vascular endothelial cells to regulate expres-sion of vascular cellular adhesion molecule-1 mRNA. J Biol Chem. 2011;286:6423–6432.

37. Kitamura YI, Kitamura T, Kruse JP, Raum JC, Stein R, Gu W, Accili D. FoxO1 protects against pancreatic beta cell failure through NeuroD and MafA induction. Cell Metab. 2005;2:153–163.

38. Iype T, Francis J, Garmey JC, Schisler JC, Nesher R, Weir GC, Becker TC, Newgard CB, Griffen SC, Mirmira RG. Mechanism of insulin gene regula-tion by the pancreatic transcription factor Pdx-1: application of pre-mRNA analysis and chromatin immunoprecipitation to assess formation of func-tional transcriptional complexes. J Biol Chem. 2005;280:16798–16807.

39. Buteau J, Spatz ML, Accili D. Transcription factor FoxO1 mediates glucagon-like peptide-1 effects on pancreatic beta-cell mass. Diabetes. 2006;55:1190–1196.

40. Fujimoto K, Polonsky KS. Pdx1 and other factors that regulate pancreatic beta-cell survival. Diabetes Obes Metab. 2009;11(suppl 4):30–37.

Therapeutic interventions that increase plasma high-density lipoprotein and apolipoprotein A-I levels are associated with improved glycemic control in people with type 2 diabetes mellitus. We, and others, have reported previously that apolipoprotein A-I, the main high-density lipoprotein apolipoprotein, increases insulin synthesis and secretion in pancreatic β-cells. This study identifies a molecular mechanism by which apolipopro-tein A-I mediates these effects that involves activation of a G-protein-cAMP-PKA-FoxO1–dependent signal transduction pathway and increased transcription of genes that are important for insulin production and β-cell survival. These findings provide an insight into the mechanism by which therapies that increase plasma apolipoprotein A-I levels may translate into improved glycemic control in people with type 2 diabetes mellitus.

Significance

by guest on October 1, 2014http://atvb.ahajournals.org/Downloaded from

1  

Supplementary Figure I: ApoA-I increases insulin release from Ins-1E cells in a concentration-dependent manner. Ins-1E cells were incubated for 1 h with apoA-I (final concentration 0.063-1.0 mg/mL). The insulin content in the medium was quantified by RIA. Values represent the mean±SD (n=4) of two separate experiments. **** P < 0.001 vs control.

0 0.063 0.125 0.25 0.5 10

50

100

150

200

Insu

lin (n

g/m

g pr

otei

n)

ApoA-I (mg/mL)

********

****

****

  2

Supplementary Figure II: ApoA-I increases insulin release from Ins-1E cells in an ABCA1- and SR-BI-dependent manner. (A and B) Ins-1E cells were transfected with ABCA1 siRNA, SR-BI siRNA and ABCG1 siRNA. The transfected cells were lysed and protein expression was quantified by western blotting. (C) Transfected cells were incubated in the absence or presence of apoA-I (final concentration 1 mg/mL) under basal (2.8 mM) glucose conditions and the insulin content in the media was quantified by RIA. Values represent the mean±SD (n=3) of three separate experiments. ***P<0.005, **** P < 0.001 vs control.

Control

Scram

bled si

RNA

ABCA1 siR

NA

SB-BI s

iRNA

ABCG1 siR

NA0

50

100

% e

xpre

ssio

n

******** ****

Control

Scram

bled si

RNA

ABCA1 siR

NA

SR-BI s

iRNA

ABCG1 siR

NA0

50

100

150

200

Insu

lin (n

g/m

g pr

otei

n)

*** ***

ABCA1

SR-BI

ABCG1

β-actin

ControlScrambled

siRNA siRNAA

B

C

210

62

98

38

MW(kDa)

  3

Supplementary Figure III: Incubation of Ins-1E cells with apoA-I does not activate the PI3K/Akt, PKC. JAK2 or AMPK signaling pathways. Ins-1E cells were cultured in serum-free medium for 12 h, then incubated with apoA-I (final concentration 1 mg/mL) for 5-60 min. Cell lysates were western blotted for phosphorylated Akt (pAkt), total Akt, phosphorylated PKC (pPKC), total PKC, phosphorylated JAK2 (pJAK2), total JAK2, phosphorylated AMPK (pAMPK), total AMPK and β-actin. The ratio of phosphorylated to total protein was determined and expressed relative to control. Values represent mean±SD (n=3).

0 5 10 15 30 600

50

100

150

200

Time (min)

pAkt

/Akt

(% v

s. c

ontr

ol)

0 5 10 15 30 600

50

100

150

Time (min)

pPK

C/P

KC

(%

vs.

con

trol

)0 5 10 15 30 60

0

50

100

150

Time (min)

pJA

K2/

JAK

2 (%

vs.

con

trol

)

0 5 10 15 30 600

50

100

150

Time (min)

pAM

PK

/AM

PK

(%

vs.

con

trol

)

pAkt

Akt

pPKC

PKC

pJAK2

JAK2

pAMPK

AMPK

β-actin

0 5 10 15 30 60 Time (min)MW(kDa)

62

62

98

98

98

98

62

62

49

38

  4

Supplementary Figure IV: Confirmation of ABCA1 antibody specificity. (A and B) Ins-1E cells were transfected with scrambled siRNA or ABCA1 siRNA. Transfected cells were lysed and protein expression was quantified by western blotting. (C) Ins-1E cells seeded on glass coverslips were transfected with siRNA. The cells were fixed, immunostained with a DyLight-488 labeled ABCA1 antibody and imaged using a total internal reflection fluorescence microscope. Representative images from three separate experiments are shown. Scale bar=20 µm. All values represent the mean±SD (n=3). ****P<0.001 compared to controls.

Control

Scram

bled si

RNA

ABCA1 siR

NA0

50

100

150

% e

xpre

ssio

n

Control

Scrambled siRNA ABCA1 siRNA

A B

C

ScrambledsiRNA

ABCA1siRNA

ABCA1

β-actin

210

38

****

MW(kDa)

Control

Control

Scram

bled si

RNA

ABCA1 siR

NA0

50

100

150

% e

xpre

ssio

n

Control

Scrablmed siRNA ABCA1 siRNA

A B

C

ScrabledsiRNA

ABCA1siRNA

ABCA1

β-actin

210

38

****

MW(kDa)

Control

  5

Supplementary Figure V: ABCA1 displays sequence homology to known Gαs subunit binding regions of the glucagon-like peptide-1 receptor (GLP-1R), gastric inhibitor polypeptide receptor (GIPR) and glucagon receptor (GCGR). Conserved residues are highlighted. Hydrophobic resides are underlined.

ABCA1 DVGLPPSKLK

GLP-1R VICIVVSKLK

GIPR ILGILLSKLR

GCGR IVQLLVAKLR

  6

Supplementary Table I: Sequence of oligonucleotide primers used for RT-PCR.

IRS1 5’-ACTGGACGTCACAGGCAGAATGA-3’ 5’-GGAGGAAGACGTGAGGTCCTGGT-3’

IRS2 5’-TGCGAACAGCCGTCGGTGAC-3’ 5’-GACCGGTGACGGCTGAACGG-3’

Ins1 5’-ACCTGGTGGAGGCTCTGTACCT-3’ 5’-TGCGGGTCCTCCACTTCACGA-3’

Ins2 5’-CGGAAACCATCAGCAAGCAGGTCA-3’ 5’-TCCACGCTCCCCACACACCA-3’

Pdx1 5’-GAAATCCACCAAAGCTCACG-3’ 5’-TGTAGGCTGTACGGGTCCTC-3’

β-actin 5’-CACCCGCGAGTACAACCTTC-3’ 5’-CCCATACCCACCATCACACC-3’

HSPcb 5’-GATTGACATCATCCCCAACC-3’ 5’-CTGCTCATCATCGTTGTGCT-3’

  7

Supplementary Table II: Incubation of Ins-1E cells with apoA-I does not alter cellular cholesterol levels.

Glucose Treatment Unesterified Cholesterol Esterified Cholesterol mM nM/mg protein nM/mg protein 2.8 PBS 230.4±56.9 nda 2.8 apoA-I 247.2±41.6 nda 25 PBS 184.1±22.3 nda 25 apoA-I 190.5±15.4 nda

Ins-1E cells were incubated at 37 °C for 1 h without or with apoA-I (final concentration 1 mg/mL) under basal (2.8 mM) or high (25 mM) glucose conditions. The cells were lysed and the cholesterol was extracted and quantified by HPLC. Values represent the mean±SD (n=3) of two separate experiments. nda: Not detected  

Materials and Methods Cell culture. Ins-1E rat insulinoma cells (provided by Prof. Claes Wollheim and Prof. Pierre Maechler, University of Geneva) were cultured at 37 °C in a humidified, 5% CO2 atmosphere with RPMI-1640 medium supplemented with 10% (v/v) FCS, 10 mM HEPES, 2 mM L-glutamine, 1 mM sodium pyruvate and 50 µM β-mercaptoethanol. Cells were used between passages 70-901. All incubations were performed in Krebs-Ringer bicarbonate HEPES (KRBH) medium containing 1 mM CaCl2, 5 mM NaHCO3, 25 mM HEPES (pH 7.4) with 0.1% (w/v) BSA under basal (2.8 mM) or high (25 mM) glucose conditions. ApoA-I. The apoA-I used in the studies was donated by CSL, Ltd, Parkville, Australia. The samples were chromatographed on a Q-Sepharose Fast-Flow column attached to an AKTA FPLC system prior to use2. The preparations were confirmed to be >95% pure by electrophoresis on 20% SDS-polyacrylamide PhastGels (GE Healthcare, Little Chalfont, United Kingdom) and Coomassie Blue staining. Islet Isolation. Pancreatic islets were isolated from 12 week old C57Bl6 male mice, essentially as previously described 3. Liberase TL (05401020001, Roche Applied Science, Indianapolis, IN) was injected into the pancreas, prior to excision and digestion at 37 °C for 12 min. Islets were purified using a custom made Histopaque 1100 solution (Sigma-Aldrich, St Louis, MO) and hand picked under a dissection microscope. This procedure was approved by the Sydney Local Health District Animal Ethics Committee. cAMP Assay. Ins-1E cells (0.5 × 106 cells/well) were incubated at 37 °C in the presence or absence of apoA-I. Cells were pre-incubated for 15 min with 2',5'-dideoxyadenosine (2’5’ ddAdo, final concentration 50 µM) (D7408, Sigma-Aldrich) or KH7 (13243, final concentration 30 µM) (Cayman Chemical, Ann Arbor, MI) to block transmembrane and soluble adenylyl cyclases, respectively. Control cells were pre-incubated with an equivalent volume of DMSO. All experiments were conducted in the presence of IBMX (final concentration 0.5 mM) (I5879, Sigma-Aldrich) to inhibit phosphodiesterase activity. At the completion of the incubations cAMP was extracted from the cells with 0.1 M HCl and quantified by ELISA (581001, Cayman Chemical).

TIRF microscopy. Ins-1E cells (1 × 106 cells/well) were grown on glass coverslips and transfected with pcDNA3-HA-Gαs4 (a gift from Dr. Denis Dupré, Dalhousie University, Canada) using Lipofectamine 2000 (11668019, Life Technologies, Carlsbad, CA). After 24 h, the cells were serum starved for 6 h and incubated with or without apoA-I for 5 min. The cells were fixed in 10% (v/v) neutral buffered formalin (HT501128, Sigma-Aldrich) for 10 min, permeabilized with PBS containing 0.3% (v/v) Triton X-100 and incubated overnight at 4 °C with a DyLight 488-conjugated rabbit anti-mouse ABCA1 antibody (1:250 dilution, NB400-105G, Novus Biologicals, Littleton, CO) and Alexa647-conjugated mouse anti-HA Tag antibody (1:100 dilution, 3444, Cell Signaling, Technology, Beverly, MA). The cells were washed and imaged on a total internal reflection fluorescence microscope (ELYRA; Zeiss) with a 100× oil-immersion objective with a numerical aperture of 1.46. Pearson’s

coefficient was calculated using the JACP plugin5 for Image J (NIH, MD). Images with one channel rotated 90 degrees were used as the negative control.

Cell signalling assays. Ins-1E cells were grown in 12-well plates (1 × 106 cells/well) to ∼70% confluency, serum-starved for 12 h, then incubated with apoA-I (final concentration 1 mg/mL) for the time periods indicated. When the incubations were complete the cells were lysed with ice-cold radioimmune precipitation assay buffer, electrophoresed on 4-12% gradient SDS-PAGE gels, transferred to nitrocellulose membranes, and probed with a rabbit anti-mouse PKA phosphorylated at Thr197 (pPKA) (1:1000, 5661, Cell Signaling Technology), rabbit anti-mouse PKA (1:1000, 5842, Cell Signaling Technology), rabbit anti-mouse Akt phosphorylated at Ser473 (pAkt) (1:1000, 4058, Cell Signaling Technology), rabbit anti-mouse Akt (1:1000, 4685, Cell Signaling Technology), rabbit anti-mouse PKC phosphorylated at Thr638 (pPKC) (1:1000, 9375, Cell Signaling Technology), rabbit anti-mouse PKC (1:1000, 2056, Cell Signaling Technology), rabbit anti-mouse JAK2 phosphorylated at Tyr1008 (pJAK2) (1:1000, 8082, Cell Signaling Technology), rabbit anti-mouse JAK2 (1:1000, 3230, Cell Signaling Technology), rabbit anti-mouse AMPK phosphorylated at Thr172 (pAMPK) (1:1000, 2535, Cell Signaling Technology), rabbit anti-mouse AMPK (1:1000, 5831, Cell Signaling Technology), and rabbit anti-mouse β-actin (1:5000, ab1801, Abcam, Cambridge, UK) antibodies. A polyclonal sheep anti-rabbit-HRP was used as the secondary antibody (1:5000, ab6795, Abcam). The blots were developed with ECL Prime (RPN2236, GE Healthcare Life Sciences, Little Chalfont, UK) and imaged using a ChemiDoc MP system (BioRad, Hercules, CA). Band intensities were quantified using ImageJ (NIH, MD). PKA activity was inhibited by pre-incubating Ins-1E cells or islets for 30 min with H89 (final concentration 20 µM) (9844, Cell Signaling Technology) or the PKA inhibitor, 14-22 amide (final concentration 5 µM) (476485 Merck Millipore, Billerica, MA). Control cells were incubated with an equivalent volume of DMSO. Insulin concentrations were measured by radioimmunoassay (RIA) (RI-13K, Merck). Cells were lysed as described above and cellular protein levels determined by BCA assay (23225, Thermo Scientific, Rockford, IL).

Confocal microscopy. Ins-1E cells (1 × 106 cells/well) were grown on glass coverslips, serum starved for 6 h and incubated with or without apoA-I and H89, or an equivalent volume of DMSO, for 30 min. The cells were fixed in 10% (v/v) neutral buffered formalin (HT501128, Sigma-Aldrich) for 10 min, permeabilized with PBS containing 0.3% (v/v) Triton X-100 and incubated overnight at 4 °C with a rabbit anti-mouse FoxO1 monoclonal primary antibody (1:500 dilution, 2880, Cell Signaling Technology). The cells were washed, incubated with an Alexa488-conjugated mouse anti-rabbit secondary antibody (1:500 diltuon, A-11034, Life Technologies) for 1 h at room temperature and mounted in DAPI-containing Vectashield mounting medium

(H-1200, Vector Laboratories, Burlingame, CA). Confocal images were acquired using a confocal laser scanning microscope (Leica, SPEII, objective 40x/1.15 NA Wetzlar, Germany) and processed using Volocity 3-D Image Analysis software (Perkin Elmer, Waltham, MA). Transfections. Ins-1E cells were grown to 70% confluency in 12-well plates (~1×106 cells/well) and transfected with pCMV5-FoxO1, pCMV5-FoxO1-ADA or an empty vector (Addgene, Cambridge, MA) using Lipofectamine 2000 (11668019, Life Technologies). Expression of FoxO1 in the transfected cells was quantified after 24 and 36 h. Cells were lysed as above and western blotted using a rabbit anti-mouse FoxO1 monoclonal antibody (1:1000). To determine whether apoA-I increases long-term insulin secretion in a FoxO1-dependent manner, Ins-1E cells that had been transfected with pCMV5-FoxO1, pCMV5-FoxO1-ADA or empty vector were incubated for a further 24 h. The cells were then incubated with apoA-I (final concentration 1 mg/mL) for an additional 16 h. The insulin concentration in the medium was determined by RIA and total cell protein as previously described. Ins-1E cells at 70% confluency were also transfected with ABCA1 siRNA (L-092341, Thermo Scientific), ABCG1 siRNA (L-093864, Thermo Scientific), and SR-BI siRNA (L-098018, Thermo Scientific) using DharmaFECT (T-200, Thermo Scientific) or the Gαs subunit siRNA (E-092648, Thermo Scientific) using Accell Delivery Media (B-005000, Thermo Scientific). Protein expression relative to β-actin was quantified 72 h after transfection by western blotting whole cell lysates using rabbit anti-mouse ABCA1 (NB400-105, Novus Biologicals), rabbit anti-human ABCG1 (sc-20795, Santa Cruz Biotechnology, Dallas, TX), rabbit anti-mouse SR-BI (ab52629, Abcam), rabbit anti-human Gαs subunit (bs3939R, Bioss Antibodies, Woburn, MA) and rabbit anti-mouse β-actin. Quantitative Real-Time PCR. Total RNA was extracted from Ins-1E cells using TRI reagent (T9424, Sigma-Aldrich). cDNA was synthesised with iScript (170-8891, BioRad) and gene expression levels were quantified with iQ SYBR Green Super mix (170-8887, BioRad) as previously described 6. Primer sequences are shown in Supplementary Table I. Quantification of Ins-1E cholesterol levels. Ins-1E cells (5×106 cells) were incubated in the presence of absence of apoA-I under basal (2.8 mM) and high (25 mM) glucose conditions for 1 h at 37 °C. Cells were washed with PBS and lysed with water. Cholesterol content of lysates was quantified by HPLC as described previously7. Statistics. Results are presented as mean±SD of at least three independent experiments. Multiple comparisons were analysed as appropriate by one-way ANOVA or two-way ANOVA with Tukey’s post-test. A value of P < 0.05 was considered significant.  

References

1. Merglen A, Theander S, Rubi B, Chaffard G, Wollheim CB, Maechler P. Glucose sensitivity and metabolism-secretion coupling studied during two-year continuous culture in ins-1e insulinoma cells. Endocrinology. 2004;145:667-678

2. Weisweiler P. Isolation and quantitation of apolipoproteins a-i and a-ii from human high-density lipoproteins by fast-protein liquid chromatography. Clin Chim Acta. 1987;169:249-254

3. Carter JD, Dula SB, Corbin KL, Wu R, Nunemaker CS. A practical guide to rodent islet isolation and assessment. Biol Proced Online. 2009;11:3-31

4. Deyts C, Vetrivel KS, Das S, Shepherd YM, Dupre DJ, Thinakaran G, Parent AT. Novel galphas-protein signaling associated with membrane-tethered amyloid precursor protein intracellular domain. J Neurosci. 2012;32:1714-1729

5. Bolte S, Cordelieres FP. A guided tour into subcellular colocalization analysis in light microscopy. Journal of microscopy. 2006;224:213-232

6. Fryirs MA, Barter PJ, Appavoo M, Tuch BE, Tabet F, Heather AK, Rye KA. Effects of high-density lipoproteins on pancreatic beta-cell insulin secretion. Arterioscler Thromb Vasc Biol. 2010;30:1642-1648

7. Kritharides L, Jessup W, Gifford J, Dean RT. A method for defining the stages of low-density lipoprotein oxidation by the separation of cholesterol- and cholesteryl ester-oxidation products using hplc. Anal Biochem. 1993;213:79-89