1-s2.0-S2212877825002030-main-html.html

Glycolytic activation of β-cell Na

-ATPases containing β1-subunits accelerates Na

extrusion, prolonging the duration of Ca

oscillations but decreasing insulin secretion

Matthew T. Dickerson

, Prasanna K. Dadi

, Reagan P. McDevitt

1,2

, Jordyn R. Dobson

, Soma

Behera

, Spencer J. Peachee

, Shannon E. Gibson

, Tenzin Wangmo

, David A. Jacobson

Molecular Physiology and Biophysics, Vanderbilt University, Nashville, TN, USA.

University of Mississippi Medical Center School of Medicine, Jackson, MS, USA.

Abstract

Electrogenic Na

ATPases (NKAs) control β-cell Ca

influx and insulin secretion by integrating

the signal strength of stimulatory G protein (G

)-coupled ligands (e.g., GLP-1, glucagon) and

inhibitory G protein (G

)-coupled ligands (e.g., somatostatin, epinephrine). However, there is a

significant gap in our understanding of how specific NKA subunits contribute to β-cell function.

Here, we demonstrate that the NKA β1-subunit (NKAβ1) is highly expressed and functional at

the plasma membrane of mouse and human β-cells. β-cell-specific NKAβ1 knockout improves

glucose tolerance and hepatic insulin sensitivity, coinciding with enhanced first- and second-

phase glucose-stimulated insulin secretion (GSIS). Electrophysiological studies reveal that β-cell

NKAβ1 enhances somatostatin-induced NKA currents, increases action potential

afterhyperpolarization amplitude, and accelerates action potential frequency. Loss of NKAβ1

delays glucose-stimulated Ca

entry by impairing glycolysis-dependent NKA activation and

reduces Na

clearance efficiency during Ca

oscillations, resulting in prolonged silent phases.

Journal Pre-proof

1. Introduction

Pancreatic β-cell insulin secretion is coordinated by tightly regulated ion fluxes that translate

metabolic signals into electrical excitability [1–3]. Electrogenic Na⁺/K⁺-ATPases (NKAs) are

essential for maintaining β-cell ionic homeostasis and establishing the resting membrane

potential (

) [4–6]. For each ATP hydrolyzed by NKA three Na

ions are pumped out and two K⁺

ions are pumped in, leading to a net outflow of positive charge that hyperpolarizes β-cell

The activity of other ion channels and transporters that control β-cell excitability and insulin

secretion are also indirectly affected by NKA control of Na

and K

levels including ATP-sensitive

ATP

) channels, voltage-gated Ca²⁺ channels (Ca

), and ion transporters [7–15]. The core

functional NKA protein complex contains a catalytic α-subunit, an auxiliary β-subunit, and in

some cases a tissue-specific regulatory γ-subunit (FXYD protein). Core NKA complexes are

capable of forming further homo- and hetero-dimeric assemblies, which display enhanced

functionality [16,17]. Mouse and human β-cells primarily express NKAα1 subunits (encoded by

ATP1A1

) and NKAβ1 subunits (encoded by

ATP1B1

), although NKAβ2 and NKAβ3 are also

expressed at lower levels [18–20]. While NKAα1 consumes ATP to drive Na⁺ efflux and K⁺ influx,

NKA holoenzyme plasma membrane trafficking and enzymatic activity are tightly regulated by

NKAβ1 [4–6]. However, despite the importance of NKAβ1 to NKA function, its contribution to β-

cell physiology remains unexplored.

NKAβ1 is a type II membrane protein with a single transmembrane-spanning segment, a

small N-terminal cytosolic region, and a large C-terminal extracellular domain [4–6]. The

extracellular domain of NKAβ1 displays extensive N-linked glycosylation in both mice and

humans [21–23]. Glycosylation of NKAβ1 directly affects NKA structure and function and also

Journal Pre-proof

promotes intercellular adherens junction formation through interactions with E-cadherins and

other NKAβ1 proteins [21,22,24]. E-cadherin-based intercellular bridges play an important role

in facilitating glucose-stimulated insulin secretion (GSIS) in rodent and human β-cells [25–27].

Furthermore, potentiators of insulin secretion (e.g., glucose, cAMP) enhance E-cadherin-based

junction formation in rat β-cells [26]. Interestingly, adherens junctions are located in lateral

zones between β-cells along with glucose transporters, suggesting tight coupling between

glucose sensing and NKA activity [27,28].

All NKA β-subunit isoforms contain three conserved disulfide bonds in the extracellular

domain that are essential for interactions with the NKA α-subunit [5]. NKAβ1 has a seventh

cysteine (C46) located in the transmembrane domain, which undergoes oxidative

glutathionylation; C46 is also a single amino acid away from the NKAβ1 residue that forms a

hydrogen bond with NKAα1 [29]. NKAβ1 C46 is glutathionylated by NADPH oxidase in a PKC-

dependent manner during the E2 (outward facing; low Na

affinity) to E1 (inward facing; high

affinity) conformational change of the NKA holoenzyme [29]. This post-translational

modification decreases NKA activity by weakening the association between NKAβ1 and NKAα1

and by augmenting intracellular K

-mediated inhibition of the E2 to E1 conformational change.

Removal of basal C46 glutathionylation by enzymes such as glutaredoxin (Grx) and glutathione

reductase (GSR) may conversely increase the NKAα1 binding affinity of NKAβ1 and promote

plasma membrane NKA activity [29].

NKA control of cation handling is essential for β-cell function and survival. The net

outflow of positive charge generated by NKA activity mediates β-cell

hyperpolarization.

Active NKAs also return elevated Na

levels resulting from Na

influx through voltage-dependent

Journal Pre-proof

(Na

) channels and Ca

-dependent Na

-permeable transient receptor potential melastatin

(TRPM) 4/5 channels during Ca

oscillations to basal levels. This electrogenic process is

governed by the kinetic parameters of β-cell NKA (e.g., V

max

- maximum rate of Na

transfer;

1/2

- substrate concentration where NKA operates at half of V

max

; K

a,Na

and K

a,K

- affinity

constants reflecting how strongly NKA binds Na

and K

), which are tuned by NKA subunit

composition; particularly by the high ion affinity NKAβ1 subunit [30,31]. Moreover, NKAβ1

promotes NKA trafficking to and retention in the plasma membrane [32,33]. Thus, the

predominance of the NKAβ1 subunit in β-cells ensures robust Na

clearance as well as rapid and

efficient

hyperpolarization following Ca

oscillation termination.

Interestingly, β-cell

Atp1b1

mRNA expression decreases rapidly (within six hours)

following exposure to inflammatory cytokines (i.e., IFN-γ and IL-1β), suggesting a possible link

between NKA activity and β-cell dysfunction [34]. Partial NKA inhibition with cardiac glycosides

100

(e.g. ouabain), and presumably NKAβ1 deficiency, decreases pump activity at the plasma

101

membrane, leading to elongated silent phases [35,36]. This is a consequence of NKA being

102

required to remain active longer to restore intracellular ion homeostasis before initiation of the

103

subsequent Ca

oscillation. Elevated basal Na⁺ levels under these conditions may also impede

104

extrusion through the Na

/Ca

exchanger (NCX) and/or deplete mitochondria Ca

105

stimulating mitochondrial Na

/Ca

exchanger (NCLX) activity, which could impair metabolic

106

function [7,37]. Reduced ATP availability would lead to increased ATP-sensitive K

ATP

) channel

107

activity as well as decreased sarcoplasmic/endoplasmic reticulum Ca

-ATPase (SERCA) and

108

plasma membrane Ca

-ATPase (PMCA) activity, all of which could further disrupt β-cell Ca

109

handling [38–40].

110

Journal Pre-proof

β-cell NKA activity is regulated by a range of receptor-mediated processes. For example,

111

inhibitory (G

-) G protein-coupled receptors (GPCRs) such as somatostatin receptors (SSTRs) and

112

α2-adrenergic receptors (α2-ADRs) activate β-cell NKA through Src kinase (Src)-dependent

113

phosphorylation of NKAα1 tyrosine 10 (Y10) [41–43]. NKAα1 Y10 can also be directly

114

phosphorylated via receptor tyrosine kinases including insulin receptors (IRs) and epidermal

115

growth factor receptors (EGFRs) [41,44,45]. Whereas stimulatory (G

-) GPCRs including

116

glucagon-like peptide-1 receptors (GLP-1Rs) inhibit β-cell NKA via cAMP-dependent protein

117

kinase A (PKA)-mediated phosphorylation of serine 943 (S943) [41,46]. NKAβ1 interacts with

118

GLP-1Rs in mouse insulinoma cells [47], which suggests that GLP-1R-mediated elevations in β-

119

cell Ca

are regulated by local cAMP production near NKA. Indeed, β-cell cAMP oscillations are

120

paralleled by fluctuations in NKA activity, demonstrating the importance of PKA control of NKA

121

function [41]. Interestingly, in the presence of GLP-1, GLP-1Rs form a complex with β-arrestin1

122

and Src [48]. Formation of this complex with NKA may facilitate Src phosphorylation and/or

123

regulate NKA recycling [48,49]. However, it remains to be determined whether NKAβ1-GLP-1R

124

interactions influence β-cell function.

125

Here, we demonstrate that NKAβ1 modulation of NKA function is key to maintenance of

126

β-cell ionic homeostasis and regulation of insulin secretion. Genetic ablation of NKAβ1

127

decreases the magnitude of somatostatin-activated, ouabain-sensitive β-cell NKA currents.

128

Diminished NKA activity in NKAβ1KO β-cells leads to cytosolic accumulation of Na

and ATP and

129

results in extended periods of NKA activity in response to glucose stimulation and G

-GPCR

130

signaling. Furthermore, G

-GPCR-mediated β-cell NKA inhibition is attenuated, which may be

131

due to disruption of interactions between NKAβ1 and GLP-1Rs. Importantly, β-cell NKAβ1

132

Journal Pre-proof

knockout enhances islet GSIS, resulting in improved glucose tolerance and insulin sensitivity

133

vivo

. Together, these findings illuminate the importance of NKAβ1 in controlling β-cell electrical

134

excitability, Ca

handling, and insulin secretion.

135

136

2. Materials and methods

137

2.1. Chemicals and reagents

138

Unless otherwise noted all chemicals and reagents were purchased from Sigma-Aldrich (St.

139

Louis, MO) or Thermo Fisher (Waltham, MA). Clonidine hydrochloride (Clon), forskolin (Fsk), and

140

ouabain (Oua) were purchased from R&D Systems (Minneapolis, MN). Liraglutide (Lira) was

141

purchased from Novo Nordisk (Plainsboro, NJ). α-ketoisocaproic acid was purchased from

142

Cayman Chemical (Ann Arbor, MI). Diazoxide was purchased from TCI America (Portland, OR).

143

miR30-based adenoviral constructs expressing an mCherry fluorescent tag along with shRNAs

144

targeting human ATP1B1 mRNA (hATP1B1 shRNA 1: TGCTCACCATCAGTGAATTTAA, hATP1B1

145

shRNA 2: TACGTATGGGACCTACACTTAA, and hATP1B1 shRNA 3: AAGTTGGAAATGTGGAGTATTT;

146

Vector ID: VB220630-1580pgt; VectorBuilder, Chicago, IL) or scramble shRNAs with no human

147

mRNA target (scramble shRNA (3x): ACCTAAGGTTAAGTCGCCCTCG; Vector ID: VB220516-

148

1177dyr; VectorBuilder) were generated.

149

150

2.2. Animals

151

All mice were 8- to 16-week-old, age-matched males on a mixed C57Bl6/J (Stock #: 000664, The

152

Jackson Laboratory (JAX), Bar Harbor, ME), 129S1/SvImJ (Stock #: 002448, JAX) background.

153

Transgenic mice with β-cell-specific knockout of

Atp1b1

(gene encoding NKAβ1; NKAβ1

) were

154

Journal Pre-proof

generated by crossing animals with a loxP-flanked

Atp1b1

exon 4 (129S-Atp1b1

tm1.1Zboro

/J, Stock

155

#: 029525, JAX)[50] with mice expressing an Ins1cre (B6(Cg)-Ins1

tm1.1(cre)Thor

/J; Stock #: 026801;

156

JAX)[51]. All animals were housed in a Vanderbilt University IACUC (protocol # M2200007-00)

157

approved facility on a 12-hour light/dark cycle with

ad libitum

access to standard chow (Lab

158

Diets, 5L0D). Mice were humanely euthanized by cervical dislocation followed by

159

exsanguination. To preserve islet ion channel function mice were not treated with anesthesia.

160

161

2.3. Human Donors

162

All studies detailed here were approved by the Vanderbilt University Health Sciences Committee

163

Institutional Review Board (IRB# 110164). Healthy human islets were provided from multiple

164

isolation centers by the Integrated Islet Distribution Program (IIDP). Deidentified human donor

165

information is provided in Table 1. The IIDP obtained informed consent for deceased donors in

166

accordance with NIH guidelines prior to reception of human islets for our studies. The

167

deidentified healthy human pancreas samples stained in Fig. 1 were obtained from the NCI

168

funded Cooperative Human Tissue Network (CHTN) (https:// www. chtn.org/). Written consent

169

was obtained for deceased donors by the CHTN prior to reception of human pancreatic tissue.

170

171

2.4. Islet isolation and culture

172

Mouse pancreata were digested with collagenase P (Roche; Basel, Switzerland) and islets were

173

isolated using density gradient centrifugation [52–54]; mouse islets were cultured in RPMI-1640

174

(Corning) media with 5.6 mM glucose supplemented with 15% FBS, 100 IU/mL penicillin, and

175

100 mg/mL streptomycin (RPMI) at 37 °C, 5% CO

. Islets were cultured in polymer-coated 35

176

Journal Pre-proof

mm glass-bottomed dishes (CellVis, Mountain View, CA) that had been treated with 10 µL/mL

177

murine extracellular matrix (ECM) extract and 20 µL/mL bovine plasma fibronectin for 1 hour at

178

37 °C. All experiments were completed within 48 hours of islet isolation.

179

180

2.5. Human pseudoislet production

181

Upon arrival, human islets were allowed to recover for at least 2 hours in CMRL-1066 (Corning,

182

Cleveland, TN) media containing 5.6 mM glucose and supplemented with 20% fetal bovine

183

serum (FBS), 100 IU/mL penicillin, 100 mg/mL streptomycin, 2 mM Gluta-MAX, 2 mM HEPES,

184

and 1 mM sodium pyruvate (CMRL) at 37 °C, 5% CO

. Human pseudoislets were prepared as

185

previously described [55]. Briefly, human islets were washed with Versene, dispersed with

186

TrypLE Express, and washed with magnetic-activated cell sorting (MACS) buffer (PBS

187

supplemented with 0.5% FBS, 100 IU/mL penicillin, and 100 mg/mL streptomycin). To selectively

188

label human β-cells, the dispersed islet cells were combined with 5 µg/mL mouse anti-human

189

NTPDase3 antibody (CHU de Quebec-Universite Laval, Quebec, Canada) in MACS buffer then

190

rotated at 4 °C for 30 minutes. Antibody-labeled islet cells were washed twice with MACS buffer

191

and incubated with anti-mouse IgG2a+b conjugated magnetic microbeads (catalog #: 130-047-

192

202; Miltenyi Biotec, Gaithersburg, MD) at 4 °C for 15 minutes. Microbead-antibody-islet cell

193

complexes were washed twice with MACS buffer to remove unbound microbeads then

194

microbead-labeled β-cells were collected utilizing LS columns (catalog #: 130-042-401; Miltenyi

195

Biotec) and a QuadroMACS™ Seperator (catalog #: 130-091-051; Miltenyi Biotec); the unbound

196

non-β-cell fraction was also collected. The β-cell fraction was transduced with hATP1B1 or

197

scramble shRNA adenoviruses for 2 hours at 37 °C, washed twice with pre-warmed CMRL,

198

Journal Pre-proof

combined at a ratio of 60:40 with untransduced non-β-cells in 96 well EZSPHERE microplates

199

(catalog #: AG4860-900SP; AGC Techno Glass, Japan), and cultured 5 days at 37 °C, 5% CO

200

Samples were supplemented with fresh pre-warmed CMRL every 48 hours. Some shRNA

201

transduced human β-cells were cultured in coated 35 mm glass-bottomed dishes for

202

intracellular Ca

imaging as described in section 2.4.

203

204

2.6. qRT-PCR

205

Human β-cells were transduced for 2 hours with adenoviral shRNA constructs in non-coated 24

206

well plates and cultured in for 72 hours in CMRL supplemented with 5.6 mM glucose at 37 °C,

207

5% CO

. Total RNA was isolated from the shRNA transduced human β-cells utilizing RNeasy

208

Micro kits (catalog #: 74004; Qiagen, Ann Arbor, MI) as per manufacturer’s instructions. cDNA

209

was prepared using iScript cDNA synthesis kits (catalog #: 1708891; Bio-Rad) according to

210

manufacturer’s instructions. Quantitative real-time PCR reactions were performed on a CFX

211

Opus 96 Real-Time PCR system (Bio-Rad) utilizing iTaq™ Universal SYBR® Green supermix

212

(catalog #: 172-5120; Bio-Rad) and primers specific to human

ATP1B1

(forward:

213

CCCAAATGTCCTTCCCGTTCAG; reverse: GCAGGAGTTTGCCATAGTACGG),

ATP1A1

(forward:

214

GGCAGTGTTTCAGGCTAACCAG; reverse: TCTCCTTCACGGAACCACAGCA), and

GAPDH

(forward:

215

GTCTCCTCTGACTTCAACAGCG; reverse: ACCACCCTGTTGCTGTAGCCAA).

ATP1B1

and

ATP1A1

216

mRNA expression relative to

GAPDH

was calculated using the 2

-ΔΔCT

method and normalized to

217

expression in human β-cells transduced with scramble shRNAs [56].

218

219

2.7. Immunofluorescence imaging

220

Journal Pre-proof

Paraffin-embedded human pancreas sections were processed and probed as previously

221

described [57]. Following rehydration, tissue sections were subjected to heat-induced epitope

222

retrieval at 100 °C for 12 min. Pancreas sections were stained with primary antibodies (1:100

223

rabbit anti-ATP1B1 (catalog #: 15192-1-AP; Proteintech, Rosemont, IL), 1:1000 guinea pig anti-

224

insulin (catalog #: 20-IP35; Fitzgerald, North Acton, MA), and 1:200 rabbit anti-glucagon (catalog

225

#: 2760S; Cell Signaling Technology, Davers, MA), followed by secondary antibodies (1:500

226

donkey anti-mouse Alexa Fluor 647 (catalog #: 715-606-150; Jackson ImmunoResearch, West

227

Grove, PA), 1:500 donkey anti-guinea pig Alexa Fluor 488 (catalog #: 706-546-148; Jackson

228

ImmunoResearch), and 1:500 donkey anti-rabbit Alexa Fluor 550 (catalog #: 711-166-152;

229

Jackson ImmunoResearch)). Immunofluorescence images were collected with a Zeiss LSM 710

230

META inverted confocal microscope (x40 magnification).

231

232

2.8. NKAβ1 immunoblotting

233

Mouse islets were isolated, lysed on ice in RIPA buffer supplemented with 20 μL/mL Halt

234

protease/phosphatase inhibitor cocktail (catalog #: PI78442; Thermo Fisher) and islet cell lysates

235

were resolved on a nitrocellulose membrane. Immunoblots were blocked for 1 h in phosphate-

236

buffered saline with 0.1% Tween 20 (PBST) supplemented with 3% powdered milk (blocking

237

solution). All primary and secondary antibodies were diluted in blocking solution. Immunoblots

238

were probed with 1:500 mouse anti-ATP1B1 primary (catalog #: MA3-928; Thermo Fisher)

239

followed by 1:2500 goat anti-mouse HRP-conjugated secondary (catalog #: W4021; Promega,

240

Madison, WI). ATP1B1 protein bands were visualized with SuperSignal™ West Pico Plus utilizing

241

a Bio-Rad Digital ChemiDoc MP (ChemiDoc). Immunoblots were stripped with Restore™

242

Journal Pre-proof

Western Blot Stripping Buffer and reprobed with 1:1000 anti-GAPDH HRP-conjugated primary

243

(catalog #: MA5-15738-HRP; Thermo Fisher). Uncropped and unprocessed immunoblot scans

244

are displayed in Supplementary Fig. 1.

245

246

2.9 NKAβ1-GLP-1R coimmunoprecipitation

247

Mouse islets were isolated and lysed on ice in RIPA buffer. Lysates were centrifuged at 12,000

248

rpm for 10 min at 4

C to remove cell debris. 5 µg rabbit anti-ATP1B1 primary (Proteintech) was

249

added to the supernatant and incubated for 1 h at 4

C, followed by addition of 40 µL protein

250

A/G magnetic beads (catalog #: PI88802; Pierce). The mixture was gently agitated overnight at

251

C. The beads were washed three times with cold RIPA buffer, and bound proteins were eluted

252

for 45 min at room temperature under reducing conditions. Immunoprecipitated proteins were

253

resolved on a nitrocellulose membrane and probed as detailed in 2.8. Immunoblots were

254

probed with 1 µg/mL goat anti-GLP-1R primary (catalog #: GTX17606; GeneTex, Irvine, CA)

255

followed by 1:2500 donkey anti-goat HRP-conjugated secondary (catalog #: GTX232040-01;

256

GeneTex). Immunoblots were stripped and reprobed with 1:500 rabbit anti-ATP1B1 primary

257

(Proteintech) followed by 1:2500 goat anti-rabbit HRP-conjugated secondary. Uncropped and

258

unprocessed immunoblot scans are displayed in Supplementary Fig. 2.

259

260

2.10. Intraperitoneal glucose/pyruvate tolerance test

261

Mice were fasted for 5 h before experiments. Dextrose (Hospira, Lake Forest, IL) was injected

262

intraperitoneally at a dose of 2.0 g/kg body weight; sodium pyruvate was dissolved in sterile

263

Journal Pre-proof

saline and injected intraperitoneally at a dose of 2.0 g/kg body weight. Blood samples were

264

collected from the tail vein and blood glucose was measured using an Aimstrip Plus glucometer.

265

266

2.11. Dynamic insulin secretion assay

267

Mouse islets were isolated and incubated overnight in RPMI supplemented with 5.6 mM

268

glucose and 0.5 mg/mL BSA at 37 °C and 5% CO

. Human pseudoislets were incubated overnight

269

in CMRL supplemented with 5.6 mM glucose and 0.5 mg/mL human albumin at 37 °C and 5%

270

. Mouse islets and human pseudoislets were immobilized in a P-4 gel matrix (catalog #: 150-

271

4124; Bio-Rad, Hercules, CA) inside BioRep perifusion chambers within the 37 °C temperature-

272

controlled enclosure of a BioRep perifusion system. All chambers were perifused at a flow rate

273

of 120 µL/min with DMEM containing (mM) 0.5 CaCl

and 10.0 HEPES supplemented with 10%

274

FBS and 0.5 mg/mL BSA (or human albumin for human pseudoislets). Perifusion solutions were

275

supplemented with glucose concentrations and treatments as described in figure legends.

276

Perfusates were collected in 96 well plates at 4 °C using a robotic fraction collector. Mouse islets

277

and human pseudoislets were recovered at the end of each experiment and total insulin was

278

extracted with 1.5% acid ethanol. Perfusate insulin secretion and total islet insulin were

279

measured with mouse insulin ELISAs (catalog #: 12-12470-10; Mercodia, Winston-Salem, NC).

280

281

2.12. Patch-clamp electrophysiology

282

Patch electrodes (8-20 MΩ; 10 mV test pulse) were backfilled with intracellular solution

283

containing (mM) 90.0 KCl, 50.0 NaCl, 1.0 MgCl

, 10.0 EGTA, 10.0 HEPES, and 0.005 amphotericin

284

B (adjusted to pH 7.2 with KOH). Mouse islets were patched in Krebs-Ringer HEPES buffer (2 mL;

285

Journal Pre-proof

KRHB) containing (mM) 119.0 NaCl, 4.7 KCl, 2.0 CaCl

, 1.2 MgSO

, 1.2 KH

, and 10.0 HEPES

286

(pH 7.35 adjusted by NaOH) supplemented with indicated glucose concentrations; a perforated

287

whole-cell patch-clamp technique was utilized to record β-cell membrane potential (

) in

288

current-clamp mode using an Axopatch 200B amplifier with pCLAMP10 software (Molecular

289

Devices). Islet cells that did not display electrical activity at 2 mM glucose (G) were identified as

290

β-cells. After a perforated patch configuration was established (seal resistance >1.0 GΩ; leak<

291

20.0pA)

depolarization and action potential (AP) firing were induced by exchanging the bath

292

solution with KRHB supplemented with 20 mM glucose and 1 mM tolbutamide (Tolb). After AP

293

firing was observed, the amplifier was switched to voltage-clamp mode;

was held at −60 mV

294

and the membrane voltage was ramped from −100 mV to −50 mV every 15 s for at least 3 min

295

and the resulting β-cell currents recorded. The amplifier was then returned to current-clamp

296

mode and

was recorded. Mouse islets were perifused with indicated treatments and

297

changes in β-cell

and currents measured as specified in figure legends.

298

299

2.13. Intracellular Ca

and Na

imaging

300

For intracellular Ca

imaging, mouse islets and human β-cells were loaded with a ratiometric

301

Fura-2 AM Ca

indicator (2 μM) for 20 min prior to the start of an experiment at 37 °C, 5% CO

302

For intracellular Na

imaging, mouse islets were loaded with a ratiometric SBFI AM Na

indicator

303

(50 µM) for 1 hour before beginning an experiment at 37 °C, 5% CO

. Before each experiment,

304

culture media was replaced with KRHB supplemented with indicated glucose concentrations

305

and treatments; after 10 min, the mouse islets and human β-cells were treated as detailed in

306

figure legends. Fura-2 AM and SBFI AM fluorescence (Ex: 340 nm and 380 nm; Em: 510 nm)

307

Journal Pre-proof

were measured every 5 sec with a Nikon Ti2 epifluorescence microscope equipped with a Prime

308

95B camera with a 25 mm CMOS sensor and Nikon Elements software (x10 magnification; Nikon

309

Ti2); the ratios of Fura-2 AM and SBFI AM fluorescence excited at 340 nm and 380 nm were

310

utilized as indicators of intracellular Ca

and Na

respectively.

311

312

2.14. Intracellular ATP:ADP ratio and cAMP imaging

313

Immediately following isolation, mouse islets were transduced for 2 hours with an adenoviral

314

vector encoding a ratiometric Perceval HR ATP:ADP ratio indicator [58] or overnight with a

315

baculoviral vector encoding a cytosolic cADDis cAMP indicator (catalog #: U0200G; Montana

316

Molecular, Bozeman, MT) [59]. Islet culture media was replaced with KRHB supplemented with

317

the indicated glucose concentrations 10 min before each experiment; the islets were then

318

treated as indicated. Perceval HR fluorescence (Ex: 405 nm and 480 nm; Em: 510 nm) or cADDis

319

fluorescence (Ex: 480 nm; Em: 510 nm) were measured every 10 sec with a Nikon Ti2-E

320

microscope equipped with a Crest X-Light V3 spinning disk confocal system, an Andor Sona

321

sCMOS camera, and Nikon Elements AR software (x20 magnification; Nikon spinning disk). The

322

ratio of Perceval HR fluorescence excited at 480 nm and 405 nm was used as an indicator of

323

intracellular ATP:ADP ratio while cADDis fluorescence excited at 480 nm was employed as an

324

indicator of cytosolic cAMP.

325

326

2.15. Statistical analysis and modeling

327

Islet intracellular Ca

, Na

, ATP:ADP ratio, cAMP, and NKA immunofluorescence were analyzed

328

using Nikon Elements software and the ImageJ Fiji image processing pack. Axon Clampfit

329

Journal Pre-proof

software was utilized to quantify β-cell NKA currents and

. Islet Ca

plateau fraction was

330

defined as the fraction of time during which islet Ca

was ≥50% of islet Ca

oscillation

331

amplitude. Islet Ca

influx as well as islet ATP:ADP ratio increases were modeled with a

332

logarithmic variable slope (4 parameter) equation (1).

333

𝐼𝑠𝑙𝑒𝑡 [𝐶𝑎

] 𝑜𝑟 [𝐴𝑇𝑃: 𝐴𝐷𝑃 𝑟𝑎𝑡𝑖𝑜] 𝑖𝑛𝑐𝑟𝑒𝑎𝑠𝑒 = 𝑀𝑖𝑛 +

(𝑀𝑎𝑥 − 𝑀𝑖𝑛) × 𝑇𝑖𝑚𝑒

𝐾

𝐸𝐶

𝐾

+ 𝑇𝑖𝑚𝑒

𝐾

(1)

334

Islet ATP:ADP ratio decreases were modeled with a logarithmic variable slope (4 parameter)

335

equation (2).

336

𝐼𝑠𝑙𝑒𝑡 [𝐴𝑇𝑃: 𝐴𝐷𝑃 𝑟𝑎𝑡𝑖𝑜] 𝑑𝑒𝑐𝑟𝑒𝑎𝑠𝑒 = 𝑀𝑖𝑛 +

𝑀𝑎𝑥 − 𝑀𝑖𝑛

(

𝐼𝐶

𝑇𝑖𝑚𝑒

)

𝐾

+ 1

(2)

337

Islet Na

decreases were modeled with a one-phase exponential decay equation (3).

338

𝐼𝑠𝑙𝑒𝑡 [𝑁𝑎

] 𝑑𝑒𝑐𝑟𝑒𝑎𝑠𝑒 = 𝑀𝑎𝑥 + ([𝑁𝑎

]

− 𝑚𝑎𝑥) × 𝑒𝑥𝑝

(−𝐾×𝑇𝑖𝑚𝑒)

(3)

339

Immunoblots were analyzed using Bio-Rad Image Lab 5.0. Figures were prepared utilizing Adobe

340

Illustrator. Statistical analyses were carried out utilizing Microsoft Excel and GraphPad Prism

341

9.2.0 as indicated in figure legends; data were compared utilizing paired or unpaired two-

342

sample t-tests, one-sample t-tests, or two-way analysis of variance (ANOVA) with Šidák’s post-

343

hoc multiple comparisons tests. Data were normalized when appropriate as indicated in figure

344

legends. Unless stated otherwise, data are presented as mean values ± standard error (SEM) for

345

the specified number of samples (

). Differences were considered significant for

≤ 0.05.

346

347

3. Results

348

3.1. NKAβ1 is expressed in mouse and human β-cells

349

Journal Pre-proof

To assess islet NKAβ1 protein expression, mouse and human pancreas sections were stained for

350

insulin, glucagon, and NKAβ1. In both α- and β-cells, as well as in other islet cell types, NKAβ1

351

was expressed predominantly at the plasma membrane (Fig. 1A, 1B). NKAβ1 was not detected

352

in surrounding acinar cells; however, staining was observed in a subpopulation of exocrine cells.

353

The staining pattern of these exocrine cells closely resembled that of cytokeratin 19 in

354

pancreatic ductal cells [60], although further characterization is necessary to definitively

355

establish their identity. NKAβ1

pancreas tissue was then stained to confirm β-cell-specific

356

ablation of NKAβ1. NKAβ1 staining was undetectable in most β-cells, whereas strong plasma

357

membrane expression persisted in α-cells and other non-β-cell islet cells (Fig. 1C).

358

Immunoblotting of Ins1cre control islet lysates revealed a prominent band near the expected

359

molecular weight of NKAβ1 (~35 kDa; Fig. 1D). Band intensity was markedly decreased in lysates

360

from NKAβ1

islets, confirming efficient β-cell

Atp1b1

deletion (Fig. 1D). The residual signal

361

from NKAβ1

islets is attributable to NKAβ1 expression in α-cells and other non-β-cells within

362

the islet.

363

364

3.2. NKAβ1 limits mouse β-cell glucose-stimulated insulin secretion

365

To examine the

in vivo

effects β-cell NKAβ1 knockout, blood glucose levels in NKAβ1

and

366

Atp1b1

(fl/fl) littermate control mice were measured during intraperitoneal glucose tolerance

367

tests (IPGTTs). Glucose excursions were attenuated in NKAβ1

mice compared to controls (Fig.

368

1E; -47.9 ± 11.2% AUC from t = 0-60 min;

= 0.008), suggesting increased glucose-stimulated

369

insulin secretion (GSIS) and/or enhanced peripheral insulin sensitivity. To evaluate insulin

370

sensitivity, hepatic gluconeogenesis was measured utilizing intraperitoneal pyruvate tolerance

371

Journal Pre-proof

tests (IPPTTs) [61]. Because hepatic gluconeogenesis is inhibited by insulin signaling [62],

372

reduced blood glucose excursions in NKAβ1

mice following IPPTTs indicate improved hepatic

373

insulin sensitivity (Fig. 1F; -49.1 ± 8.5% AUC from t = 0-60 min;

= 0.006). Next, dynamic GSIS

374

from isolated NKAβ1

and age-matched Ins1cre control islets was measured to determine the

375

impact of NKAβ1 ablation on β-cell secretory function. First-phase GSIS (5-10 min) from

376

NKAβ1

islets increased modestly (Fig. 1G, 1H; 10.1 ± 2.3 µg/L at t = 8 min,

= 0.011; 5.7 ± 1.7

377

at t = 9 min,

= 0.029; 1.6 ± 0.5 at t = 10 min,

= 0.041), whereas second-phase GSIS (13-20

378

min) more than doubled (Fig. 1G, 1H; 240.0 ± 60.5% AUC from t = 13-20 min,

= 0.014). Insulin

379

secretion in response to KCl-mediated depolarization was indistinguishable between NKAβ1

380

and control islets, suggesting that altered insulin secretion is due to effects on β-cell

(Fg. 1G,

381

1H). Total insulin content was equivalent in NKAβ1

and control islets, indicating that insulin

382

biosynthesis and processing are likely unaffected (Fig. 1I).

383

384

3.3.

NKAβ1 expression enhances somatostatin-induced β-cell NKA currents

385

Knockout of β-cell NKAβ1 enhanced islet GSIS but did not alter depolarization-induced insulin

386

secretion, suggesting that NKAβ1 may limit secretion by promoting NKA-mediated β-cell

387

hyperpolarization. To test this, we explored the impact of NKAβ1 on β-cell electrical excitability

388

and NKA currents. NKAβ1

and control β-cell

were similar under basal (2 mM glucose) and

389

stimulatory conditions (20 mM glucose with 1 mM tolbutamide (Tolb); Fig. 2A-2C). Furthermore,

390

NKA activation with 200 nM somatostatin (SST) induced hyperpolarization, and NKA inhibition

391

with 200 µM ouabain (Oua) caused depolarization, to the same extent in both genotypes (Fig.

392

2A-2C). Under stimulatory conditions, NKAβ1

β-cell Oua-sensitive currents were reduced (-4.4

393

Journal Pre-proof

± 2.0 pA at -75 mV,

= 0.043; Fig. 2D). SST-induced NKA activation increased the difference in

394

Oua-sensitive currents between NKAβ1

and control β-cells (8.5 ± 3.5 pA between -100 and -50

395

mV,

= 0.027; Fig. 2E). Key parameters defining β-cell action potential (AP) shape were also

396

quantified (Fig. 2F-2K). Under stimulatory conditions, NKAβ1

β-cells displayed reduced

397

afterhyperpolarization (AHP) amplitude (-0.8 ± 0.4 pA,

= 0.027; Fig. 2G) and slower

398

instantaneous AP frequency (-0.8 ± 0.3 Hz,

= 0.013; Fig. 2J). These NKAβ1-dependent effects

399

were abolished when NKA was inhibited with Oua. In control β-cells, Oua-mediated NKA

400

inhibition diminished AP amplitude (-2.7 ± 1.1 mV;

= 0.027; Fig. 2F) and AHP amplitude (-0.7 ±

401

0.1 mV;

< 0.0001; Fig. 2G), decelerated AP upstroke (-0.4 ± 0.1 mV/ms;

= 0.001; Fig. 2H) and

402

downstroke (0.3 ± 0.1 mV/ms;

= 0.046; Fig. 2I), decreased AP firing frequency (-1.1 ± 0.2 Hz;

403

< 0.0001; Fig. 2J), and prolonged intervals between APs (183.0 ± 70.0 ms;

= 0.018; Fig. 2K), all

404

independently of NKAβ1. These findings indicate that NKAβ1

) regulates SSTR-mediated

405

activation of β-cell NKA currents and

) modulates β-cell AP kinetics during Ca

oscillations.

406

407

3.4. NKAβ1 facilitates islet glucose-stimulated Ca

influx in a glycolysis-dependent manner

408

As we have demonstrated that NKA activity regulates β-cell Ca

oscillations [41], we next

409

examined the impact of NKAβ1 on β-cell glucose-stimulated Ca

entry. Transient decreases in β-

410

cell Ca

preceding glucose-stimulated Ca

influx are well-documented and are attributed to

411

uptake of cytosolic Ca

into the endoplasmic reticulum (ER) through sarco/endoplasmic

412

reticulum Ca

ATPases (SERCAs) [63]. In NKAβ1

islets, the duration of these glucose-

413

dependent Ca

decreases was prolonged (133.2 ± 48.6 seconds,

< 0.0001; Fig. 3A, 3B),

414

suggesting that NKA activity also regulates β-cell Ca

handling while

is hyperpolarized. The

415

Journal Pre-proof

NKA protein complex has been shown to interact with several glycolytic enzymes [64] and be

416

preferentially powered by glycolytic ATP [65]; therefore, to determine whether the delay in

417

glucose-stimulated Ca

entry into NKAβ1

islets reflects altered glycolysis, β-cells were

418

stimulated with the non-glycolytic metabolic fuel, alpha-ketoisocaproate (α-KIC). Both NKAβ1

419

and control islets exhibited near-instantaneous Ca

entry in response to α-KIC (Fig. 3A, 3B).

420

Substitution of α-KIC in place of glucose did not affect total Ca

influx (Fig. 3A, 3C) but

421

accelerated the rate of islet Ca

entry independently of NKAβ1 (386.3 ± 303.8% for NKAβ1

422

islets;

= 0.0013; 296.0 ± 154.4% for control islets;

= 0.0069; Fig. 3A, 3D). These results

423

indicate that transient NKA activation fine-tunes the initiation of β-cell glucose-stimulated Ca

424

entry and suggest a role for NKAβ1 in coupling the pump to glycolytic ATP.

425

426

3.5. β-cell NKA complexes containing β1-subunits increase ATP consumption

427

During glycolysis, ADP and phosphoenolpyruvate (PEP) are consumed by pyruvate kinases (PKs)

428

to generate pyruvate and ATP [66]. At low glucose, β-cell workload is modest, and ADP levels

429

can become a limiting substrate for glycolysis and mitochondrial metabolism. Because NKA

430

activity maintains β-cell

in a hyperpolarized state at low glucose, ATP consumption by the

431

pump could impact ADP availability and, consequently, β-cell metabolism. To test whether

432

NKAβ1 knockout delays glucose-stimulated Ca

entry by depleting β-cell ADP levels, islet

433

ATP:ADP ratio was measured. The ATP:ADP ratio in NKAβ1

islets was elevated compared to

434

control islets at both low (2 mM; 45.1 ± 21.2%,

= 0.0017; Fig. 3E, 3F) and high glucose (9 mM;

435

23.6 ± 22.5%,

= 0.0012; Fig. 3E, 3F). Following glucose stimulation, the ATP:ADP ratio rose

436

rapidly in both genotypes, indicating that overall ATP generation and glycolysis remain intact.

437

Journal Pre-proof

Moreover, rates of glucose-stimulated ATP generation (Fig. 3E, 3G), ATP consumption during

438

periods of high Ca

(Fig. 3E, 3H), and ATP depletion after mitochondrial uncoupling with FCCP

439

(Fig. 3E, 3I) were unchanged in NKAβ1

islets. Together, these results establish that reduced

440

NKA activity in NKAβ1

β-cells leads to ATP accumulation without impairing glucose-stimulated

441

ATP production. The data further support that glycolytic machinery remains functional in

442

NKAβ1

β-cells and reinforce that NKAβ1 modulates the preferential use of glycolytically

443

derived ATP by the pump.

444

445

3.6.

NKAβ1 promotes efficient β-cell Na

removal

446

Glucose stimulation of islets in the presence of the K

ATP

channel blocker Tolb transiently

447

prevents β-cell Ca

entry in an NKA-dependent manner [41,67]. Thus, we next set out to

448

determine how NKAβ1 affects NKA-mediated inhibition of Ca

entry. NKAβ1

islets displayed

449

prolonged suppression of glucose-stimulated Ca

influx (216.6 ± 91.9 seconds,

= 0.011; Fig.

450

4A, 4B). Substituting α-KIC in place of glucose greatly diminished the magnitude of NKA-

451

mediated β-cell Ca

decreases in both NKAβ1

(330.5 ± 99.5% AUC,

= 0.0005; Fig. 4C, 4D)

452

and control islets (189.6 ± 59.6% AUC,

< 0.0001; Fig. 4C, 4D), further indicating that β-cell

453

NKAs are preferentially fueled by glycolytic metabolism. Sodium entry through Na

and

454

TRPM4/5 channels contributes to β-cell

depolarization during glucose-stimulated Ca

455

oscillations [68,69], and subsequent NKA activation serves as the primary mechanism for

456

clearing accumulated Na

, facilitating

hyperpolarization and termination of Ca

influx. To

457

investigate the influence of NKAβ1 on β-cell Na

handling, islets were stimulated with glucose in

458

the presence of the K

ATP

channel activator diazoxide to clamp

in a hyperpolarized state and

459

Journal Pre-proof

prevent activation of non-NKA Na

conductances. Intracellular Na

was elevated in NKAβ1

460

islets compared to controls at both 2 mM (83.0 ± 40.2%,

= 0.0117; Fig. 4E, 4F) and 11 mM

461

glucose (623.8 ± 398.8%,

= 0.0083; Fig. 4E, 4F). Kinetic modelling of Na

efflux using a one-

462

phase exponential decay equation revealed decelerated Na

clearance in NKAβ1

islets (rate

463

constant (K) = -47.8 ± 9.4%,

= 0.0305; half-life = 89.8 ± 36.5%,

= 0.0185; Fig. 4E, 4G, 4H).

464

These findings indicate that in NKAβ1 helps maintain low basal intracellular Na

levels in β-cells

465

by promoting efficient Na

extrusion.

466

467

3.7.

β-cell NKAβ1 tunes inhibitory and stimulatory GPCR control of NKA activity

468

In β-cells, the NKA complex interacts with and is activated by Src tyrosine kinase in a G

-GPCR-

469

dependent manner (e.g., SSTRs, α2A-ADRs) [41]. Because NKAβ1 stabilizes the NKA complex at

470

the plasma membrane, its role in G

-GPCR regulation of β-cell NKA activity was examined. β-cell

471

plateau fraction, defined as the proportion of time that Ca²⁺ remains elevated during

472

glucose-stimulated oscillations, was lower in NKAβ1

islets than in controls (-30.0 ± 7.0%,

473

0.0005; Fig. 5A, 5B). In the presence of 20 nM SST, the difference in Ca²⁺ plateau fraction

474

between NKAβ1

and control islets increased further (-51.6 ± 13.0%,

= 0.0003; Fig. 5A, 5B),

475

and SST mediated a larger reduction Ca

plateau fraction in NKAβ1

than in control islets (31.0

476

± 17.3%,

= 0.032; Fig. 5A, 5C). Calcium plateau fraction was also diminished in NKAβ1

477

compared to control islets when G

-coupled α2-ADRs were activated with clonidine (-56.1 ±

478

8.5%,

= 0.0069; Fig. 5D, 5E), suggesting that NKAβ1 broadly regulates islet G

signaling across

479

β-cell G

-GPCRs. G

-coupled glucagon-like peptide-1 receptor (GLP-1R) signaling inhibits β-cell

480

NKA function in a protein kinase A (PKA)-dependent fashion, thereby enhancing glucose-

481

Journal Pre-proof

stimulated Ca

entry [41]. As NKAβ1 interacts with GLP-1Rs in MIN6 cells [47], we next tested

482

whether this interaction occurs in primary mouse islets. A protein band consistent with the

483

molecular weight of GLP-1R (~60 kDa) co-immunoprecipitated with NKAβ1 (Fig. 5F). In co-

484

immunoprecipitated samples, NKAβ1 migrated at ~100 kDa (compared to ~35 kDa in freshly

485

prepared islet lysates). The apparent mass shift likely reflects differences in glycosylation or

486

multimerization arising from distinct sample processing conditions. Fresh lysates were boiled at

487

95 °C to ensure complete protein denaturation, whereas co-IP eluates were collected at room

488

temperature to minimize release of protein A/G that could obscure target protein bands. Next,

489

the influence of NKAβ1 on GLP-1R-mediated control of β-cell Ca

handling was examined. In

490

the presence of Tolb and 20 mM glucose, SST induced larger decreases in Ca

plateau fraction

491

in NKAβ1

islets (18.3 ± 0.63%,

= 0.042; Fig. 5G, 5H). Subsequent stimulation of GLP-1R

492

signaling with liraglutide increased Ca

plateau fraction more in control than in NKAβ1

islets

493

(222.3 ± 88.7%,

= 0.025; Fig. 5G, 5I). Measurement of β-cell cAMP levels using a fluorescent

494

cADDis indicator showed that NKAβ1 enhances liraglutide-mediated GLP-1R signaling (36.9 ±

495

7.5%, P = 0.0192; Fig. 5J, 5L), whereas pharmacological activation of adenylyl cyclases with

496

forskolin was unaffected (Fig. 5K, 5L). These results indicate that NKAβ1 promotes β-cell cAMP

497

production, potentially through a specific interaction with GLP-1R signaling.

498

499

3.8.

NKAβ1 regulates human β-cell glucose-stimulated Ca

entry and insulin secretion

500

NKAβ1 is highly expressed in human islets [18,20]; therefore, its role in human β-cell Ca

501

handling and insulin secretion was evaluated. qRT-PCR analysis of

ATP1B1

and

ATP1A1

mRNA

502

expression in purified human β-cells confirmed efficient shRNA-mediated NKAβ1 knockdown (-

503

Journal Pre-proof

97.3 ± 1.2%,

< 0.0001; NKAβ1

βKD

; Fig. 6A). Loss of NKAβ1 increased intracellular Ca

504

dispersed human β-cells at both low (2 mM; 3.5 ± 1.6%,

= 0.0356; Fig. 6B, 6C) and elevated

505

(9mM; 6.0 ± 3.3%,

= 0.006; Fig. 6B, 6C) glucose. NKAβ1

βKD

β-cell Ca

was also elevated

506

following addition of 200 nM SST (3.5 ± 2.3%,

= 0.0377; Fig. 6B, 6C). To assess effects on

507

insulin secretion, NKAβ1

βKD

human pseudoislets were generated by reaggregating shRNA-

508

transduced β-cells with untransduced islet cells (e.g., α- and δ-cells). NKAβ1

βKD

pseudoislets

509

contained more insulin than scramble shRNA controls (79.0 ± 18.8%,

< 0.0084; Fig. 6D),

510

suggesting altered secretion and/or synthesis. When secretion was normalized to insulin

511

content, NKAβ1

βKD

pseudoislets exhibited reduced second-phase GSIS (11 mM glucose; t = 15-

512

30 min; -53.5 ± 9.2%,

= 0.011; Fig. 6E, 6F). Basal secretion (1 mM glucose), first-phase GSIS (11

513

mM glucose; t = 9-15 min), and depolarization-stimulated secretion (30 mM KCl; t = 72-81 min)

514

were unaffected. When normalized to pseudoislet number, NKAβ1

βKD

increased first-phase GSIS

515

in four of the six biological replicates (Fig. 6G, 6H). Despite heterogeneity in GSIS and KCl

516

responses that obscured changes in insulin AUC, NKAβ1

βKD

pseudoislets secreted more insulin at

517

t = 12 min during first-phase GSIS (75.5 ± 42.7%,

= 0.001; Fig. 6G, 6H) and at t = 75 min during

518

KCl stimulation (140.2 ± 90.1%,

= 0.044; Fig. 6G, 6H). These data demonstrate that NKAβ1

519

regulates insulin secretion in human β-cells, with effects distinct from those observe in mice,

520

likely due to differences in species biology, experimental conditions, and/or the timing of gene

521

suppression (acute shRNA-based knockdown in adult human β-cells vs. developmental knockout

522

in mice).

523

524

4. Discussion

525

Journal Pre-proof

Oscillations in NKA activity play an important role in glucose-stimulated β-cell Ca

oscillations

526

and thus pulsatile insulin secretion. Here we identify a central role for the NKAβ1 subunit in

527

coordinating glycemic and GPCR control of β-cell NKA activity. We show that glucose-dependent

528

glycolysis activates β-cell NKAs, transiently hyperpolarizing

independently of K

ATP

channels.

529

Knockout of NKAβ1, the predominant β-cell NKA β-subunit, prolongs glucose-mediated NKA

530

activation and extends silent phases between glucose-stimulated Ca

oscillations. The resulting

531

increase in both total insulin secretion and its pulsatility leads to improved hepatic insulin

532

sensitivity and enhanced glucose tolerance in NKAβ1

mice. Extended intervals of NKA activity

533

likely reflect reduced Na

pumping capacity stemming from decreased plasma membrane

534

NKA expression and/or compensatory replacement by the lower-capacity NKAβ3 subunit.

535

Moreover, ablation of β-cell NKAβ1 amplifies G

-GPCR-mediated inhibition of Ca

entry while

536

attenuating G

-GPCR-induced cAMP production and stimulation of Ca

influx. Together, these

537

findings establish NKAβ1 as a key determinant of β-cell excitability and responsiveness to

538

secretagogues and inhibitors, with direct consequences for pulsatile insulin release and

539

downstream metabolic regulation.

540

β-subunits are essential for proper assembly, folding, and trafficking of NKA complexes

541

to the plasma membrane. Without β-subunits, NKAα1 is retained and degraded in the ER,

542

reducing NKA activity at the plasma membrane [70]. Our data show that NKAβ1 enhanced β-cell

543

NKA currents during SSTR activation, though SST still hyperpolarized

via NKA complexes

544

containing other β-subunits (e.g., β2/β3) [71]. NKAβ1 enhanced β-cell NKA current amplitude

545

but did not affect resting

, which is likely due to the large conductance of K

ATP

channels

546

compared to low basal NKA workload. Under stimulatory conditions, elevated Na⁺ influx raises

547

Journal Pre-proof

NKA workload, presumably increasing its contribution to β-cell

hyperpolarization. Consistent

548

with this, NKA inhibition has been shown to increase β-cell Ca

entry and insulin secretion in a

549

glucose-dependent manner [36,72]. Our data suggest that NKA regulation of AP shape, in

550

addition to

hyperpolarization, tunes β-cell Ca

handling and insulin secretion. We found that

551

NKA activity augments β-cell AP and AHP amplitude, as observed in neurons and pancreatic α-

552

cells [73,74]. Furthermore, NKAβ1 enhanced β-cell AHP amplitude and increased the frequency

553

of AP firing. This likely reflects reduced voltage-dependent inactivation of Ca

channels and

554

elevated intracellular K

, which increases the driving force for K

efflux. However, further studies

555

are needed to clarify the precise mechanisms by which NKAβ1 regulates β-cell

and

556

excitability.

557

The interval between glucose stimulation and β-cell Ca

influx is often marked by

558

modest, transient

hyperpolarization [3,75,76], which Düfer et al. attributed to NKA activity

559

[77]. Glucose-dependent β-cell NKA activation also transiently decreases Ca

in islets

560

depolarized with the K

ATP

inhibitor tolbutamide [41,67]. However, this effect was not replicated

561

when oxidative phosphorylation was stimulated downstream of glycolysis, strongly suggesting

562

that β-cell NKAs are driven by a pool of cytosolic ATP near the plasma membrane. Indeed, there

563

is evidence that NKA interacts with glycolytic machinery, preferentially utilizes glycolytic ATP,

564

and enhances glycolytic flux when active [64,65,78]. Our data also indicate that NKAβ1 shortens

565

the duration of glucose-stimulated NKA activation. Similarly, NKAβ1 reduces the duration of G

566

GPCR-mediated NKA activation. During Ca²⁺ oscillations, NKA workload rises with Na⁺ influx, and

567

our findings indicate that NKAβ1 shortens the silent phases caused by NKA-mediated Na⁺

568

clearance. The duration of NKA-mediated hyperpolarization depends primarily on the time

569

Journal Pre-proof

required to restore baseline intracellular Na

levels [74]. Therefore, because NKAβ1

β-cells

570

display tonically elevated intracellular Na

levels and decelerated glucose-stimulated Na

efflux,

571

it would presumably take longer to eliminate excess Na

from the preceding electrically active

572

period, which would prolong

hyperpolarization between Ca

oscillations. This is consistent

573

with decreased β-cell Ca

oscillation frequency following treatment with non-saturating

574

ouabain concentrations [36]. Future experiments will fully elucidate how NKAβ1 regulates

575

metabolic control of β-cell NKA activity.

576

Both metabolic and GPCR control of NKA impact insulin secretion. Rapid first-phase

577

insulin secretion relies on a readily releasable pool (RRP) of pre-docked granules, whereas

578

prolonged second-phase secretion depends on ATP-driven recruitment from reserve pools

579

[3,79,80]. In NKAβ1

β-cells, reduced ATP consumption by plasma membrane NKAs elevates

580

the ATP:ADP ratio, which may increase cytosolic ATP near the exocytotic machinery. This surplus

581

ATP could enhance GSIS by

) facilitating ATP-dependent kinesin and myosin Va–mediated

582

granule trafficking [81],

) promoting cortical F-actin remodeling for granule docking [82], and

)

583

supporting SNARE complex turnover by NSF/α-SNAP [83]. NKAβ1

βKD

enhanced GSIS in only four

584

of six human pseudoislet preparations, likely reflecting heterogeneity in islet isolation, shipping,

585

and donor factors such as age, health, and genetics. In mouse β-cells, we find that glycemic and

586

-GPCR regulation of NKA activity progressively declines during prolonged culture (>72 hours),

587

likely due to loss of the native microenvironment (innervation, vascularization, extracellular

588

matrix). Similarly, such culture-related dysfunction may blunt the effects of NKAβ1

βKD

in human

589

pseudoislets, which are maintained in culture for five to seven days prior to perifusion assays.

590

Beyond total insulin output, pulsatile release patterns are also critical for glucose homeostasis,

591

Journal Pre-proof

and their disruption is linked to insulin resistance in T2D [84,85]. Interestingly, elevated levels of

592

an endogenous ouabain-like NKA inhibitor have been detected in the plasma of T2D patients

593

and may contribute to development of insulin resistance by impairing glucose-stimulated Ca

594

oscillations and reducing insulin pulsatility [82]. In contrast, NKAβ1 knockout did not disrupt

595

oscillations, but instead prolonged silent phases between them, and presumably increased

596

insulin secretion per oscillation. Systemic insulin sensitivity was also enhanced, as evidenced by

597

blunted hepatic gluconeogenesis in NKAβ1

mice. These findings raise the possibility that NKA

598

β-subunits influence systemic insulin action by altering β-cell secretion dynamics, warranting

599

further investigation beyond the islet.

600

-GPCRs such as SSTRs and α2-ADRs stimulate β-cell NKA activity by increasing Src-

601

dependent phosphorylation of NKAα1 at Y10 while reducing PKA-mediated phosphorylation at

602

S943 [41]. In contrast, G

-coupled GLP-1Rs promote PKA phosphorylation of S943. Because

603

NKAβ1 targets NKA complexes to caveolin-enriched lipid rafts that contain both Src and A-kinase

604

anchoring protein (AKAP)-tethered PKA [86–88], knockout of this subunit could alter the

605

compartmentalized signaling environment, thereby amplifying G

ligand effects by enhancing

606

Src-dependent NKA activation and/or upstream G

-GPCR signaling [86–88]. NKAβ1 also directly

607

interacts with GLP-1Rs, so its loss may impair local PKA signaling near NKA complexes [47]. Such

608

disruption could explain the blunted effects of liraglutide on Ca

entry and global cAMP

609

production observed in NKAβ1

islets. Notably, when forskolin was used to directly activate

610

adenylyl cyclases, cAMP production in NKAβ1

islets matched controls, suggesting that NKAβ1

611

regulation of G

signaling-mediated NKA inhibition could be specific to GLP-1Rs. However, β-cells

612

also express other G

-GPCRs, including glucagon receptors and glucose-dependent

613

Journal Pre-proof

insulinotropic polypeptide receptors (GIPRs) [18–20], that may localize to lipid rafts near AKAP-

614

tethered PKA and could likewise modulate NKA activity. Therefore, it will be important to

615

determine whether this mechanism extends to other G

-GPCRs. Furthermore, glycolytic

616

enzymes such as enolase 1 and ADP-dependent glucokinase have been shown to interact with

617

GLP-1Rs [47], suggesting that these complexes

fuel NKA activity with glycolytic ATP, and

618

stimulate glycolysis via NKA-derived ADP. These interactions likely complement other glycolytic

619

enzymes associated with NKA, positioning NKA-glycolytic enzyme complexes near AKAPs to fine-

620

tune ATP availability for G

-coupled cAMP production.

621

In summary, we identify NKAβ1 as an important regulator of β-cell NKA activity, excitability,

622

and insulin secretion, integrating metabolic and GPCR-derived signals (Fig. 7A, 7B). We

623

demonstrate that NKAβ1 promotes efficient Na

clearance during Ca

oscillations, thereby

624

shortening silent phases between insulin pulses. Moreover, NKAβ1 tunes β-cell responsiveness

625

to both inhibitory and stimulatory inputs, attenuating the effect of G

ligands while supporting

626

-mediated stimulation of Ca²⁺ influx and cAMP production. Finally, we show that NKAβ1

627

influences systemic glucose metabolism, with its loss improving hepatic insulin sensitivity and

628

glucose tolerance. Taken together, these findings establish NKAβ1-containing NKA complexes as

629

key molecular determinants of β-cell electrical and secretory dynamics.

630

631

5. Acknowledgements

632

This research was performed with the support of the Integrated Islet Distribution Program

633

(https://iidp.coh.org/). We especially thank the organ donors and their families. These studies

634

have been supported by a Vanderbilt Integrated Training in Engineering and Diabetes grant

635

Journal Pre-proof

(T32DK101003), an Initiative for Maximizing Student Development at Vanderbilt grant

636

(T32GM139800), a Multidisciplinary Training in Molecular Endocrinology grant (T32DK007563),

637

National Institutes of Health grants (DK-097392, DK-115620, DK-129340, and DK-136768), an

638

American Diabetes Association Grant (1-17-IBS-024), a Juvenile Diabetes Research Foundation

639

grant (2-SRA-2019-701-S-B), a grant from The Leona M & Harry B. Helmsley Charitable Trust

640

(2306-06066), and a Pilot and Feasibility grant through the Vanderbilt Diabetes Research and

641

Training Center grant (P60-DK-20593).

642

643

6. Author contributions

644

M.T.D. and D.A.J. formulated and designed experiments. M.T.D., P.K.D., R.P.M, J.R.D., S.B., S.J.P.,

645

S.E.G, and T.W. performed experiments. M.T.D. and D.A.J. analyzed data. M.T.D. and D.A.J.

646

interpreted experimental results. M.T.D. and D.A.J. prepared figures. M.T.D. and D.A.J. drafted

647

the manuscript. M.T.D., P.K.D., R.P.M, J.R.D., S.B., S.J.P., S.E.G, T.W., and D.A.J. approved the final

648

manuscript submitted for publication.

649

650

7. References

651

[1]

Jacobson, D.A., Philipson, L.H., 2007. Action potentials and insulin secretion: new

652

insights into the role of Kv channels. Diabetes, Obesity & Metabolism 9 Suppl 2(Suppl 2): 89–98,

653

Doi: 10.1111/j.1463-1326.2007.00784.x.

654

[2]

Campbell, J.E., Newgard, C.B., 2021. Mechanisms controlling pancreatic islet cell

655

function in insulin secretion. Nature Reviews. Molecular Cell Biology 22(2): 142–58, Doi:

656

10.1038/s41580-020-00317-7.

657

[3]

Rorsman, P., Ashcroft, F.M., 2018. Pancreatic β-Cell Electrical Activity and Insulin

658

Secretion: Of Mice and Men. Physiological Reviews 98(1): 117–214, Doi:

659

10.1152/physrev.00008.2017.

660

Journal Pre-proof

[4]

Contreras, R.G., Torres-Carrillo, A., Flores-Maldonado, C., Shoshani, L., Ponce, A., 2024.

661

Na+/K+-ATPase: More than an Electrogenic Pump. International Journal of Molecular Sciences

662

25(11): 6122, Doi: 10.3390/ijms25116122.

663

[5]

Fedosova, N.U., Habeck, M., Nissen, P., 2021. Structure and Function of Na,K-ATPase-The

664

Sodium-Potassium Pump. Comprehensive Physiology 12(1): 2659–79, Doi:

665

10.1002/cphy.c200018.

666

[6]

Obradovic, M., Sudar-Milovanovic, E., Gluvic, Z., Banjac, K., Rizzo, M., Isenovic, E.R.,

667

2023. The Na+/K+-ATPase: A potential therapeutic target in cardiometabolic diseases. Frontiers

668

in Endocrinology 14: 1150171, Doi: 10.3389/fendo.2023.1150171.

669

[7]

Scranton, K., John, S., Angelini, M., Steccanella, F., Umar, S., Zhang, R., et al., 2024.

670

Cardiac function is regulated by the sodium-dependent inhibition of the sodium-calcium

671

exchanger NCX1. Nature Communications 15(1): 3831, Doi: 10.1038/s41467-024-47850-z.

672

[8]

Chung, Y.J., Hoare, Z., Baark, F., Yu, C.S., Guo, J., Fuller, W., et al., 2024. Elevated Na is a

673

dynamic and reversible modulator of mitochondrial metabolism in the heart. Nature

674

Communications 15(1): 4277, Doi: 10.1038/s41467-024-48474-z.

675

[9]

Gamba, G., 2005. Molecular Physiology and Pathophysiology of Electroneutral Cation-

676

Chloride Cotransporters. Physiological Reviews 85(2): 423–93, Doi:

677

10.1152/physrev.00011.2004.

678

[10]

Parker, M.D., Boron, W.F., 2013. The Divergence, Actions, Roles, and Relatives of Sodium-

679

Coupled Bicarbonate Transporters. Physiological Reviews 93(2): 803–959, Doi:

680

10.1152/physrev.00023.2012.

681

[11]

Mollajew, R., Toloe, J., Mironov, S.L., 2013. Single KATP channel opening in response to

682

stimulation of AMPA/kainate receptors is mediated by Na+ accumulation and submembrane

683

ATP and ADP changes. The Journal of Physiology 591(10): 2593–609, Doi:

684

10.1113/jphysiol.2012.248369.

685

[12]

Catterall, W.A., Lenaeus, M.J., Gamal El-Din, T.M., 2020. Structure and Pharmacology of

686

Voltage-Gated Sodium and Calcium Channels. Annual Review of Pharmacology and Toxicology

687

60: 133–54, Doi: 10.1146/annurev-pharmtox-010818-021757.

688

[13]

Bobulescu, I.A., Di Sole, F., Moe, O.W., 2005. Na+/H+ exchangers: physiology and link to

689

hypertension and organ ischemia. Current Opinion in Nephrology and Hypertension 14(5): 485–

690

94, Doi: 10.1097/01.mnh.0000174146.52915.5d.

691

[14]

Tian, J., Xie, Z., 2008. The Na-K-ATPase and calcium-signaling microdomains. Physiology

692

(Bethesda, Md.) 23: 205–11, Doi: 10.1152/physiol.00008.2008.

693

Journal Pre-proof

[15]

Ruminot, I., Gutiérrez, R., Peña-Münzenmayer, G., Añazco, C., Sotelo-Hitschfeld, T.,

694

Lerchundi, R., et al., 2011. NBCe1 mediates the acute stimulation of astrocytic glycolysis by

695

extracellular K+. The Journal of Neuroscience: The Official Journal of the Society for

696

Neuroscience 31(40): 14264–71, Doi: 10.1523/JNEUROSCI.2310-11.2011.

697

[16]

Seflova, J., Habibi, N.R., Yap, J.Q., Cleary, S.R., Fang, X., Kekenes-Huskey, P.M., et al.,

698

2022. Fluorescence lifetime imaging microscopy reveals sodium pump dimers in live cells. The

699

Journal of Biological Chemistry 298(5): 101865, Doi: 10.1016/j.jbc.2022.101865.

700

[17]

Linnertz, H., Urbanova, P., Obsil, T., Herman, P., Amler, E., Schoner, W., 1998. Molecular

701

distance measurements reveal an (alpha beta)2 dimeric structure of Na+/K+-ATPase. High

702

affinity ATP binding site and K+-activated phosphatase reside on different alpha-subunits. The

703

Journal of Biological Chemistry 273(44): 28813–21, Doi: 10.1074/jbc.273.44.28813.

704

[18]

Segerstolpe, Å., Palasantza, A., Eliasson, P., Andersson, E.-M., Andréasson, A.-C., Sun, X.,

705

et al., 2016. Single-Cell Transcriptome Profiling of Human Pancreatic Islets in Health and Type 2

706

Diabetes. Cell Metabolism 24(4): 593–607, Doi: 10.1016/j.cmet.2016.08.020.

707

[19]

DiGruccio, M.R., Mawla, A.M., Donaldson, C.J., Noguchi, G.M., Vaughan, J., Cowing-

708

Zitron, C., et al., 2016. Comprehensive alpha, beta and delta cell transcriptomes reveal that

709

ghrelin selectively activates delta cells and promotes somatostatin release from pancreatic

710

islets. Molecular Metabolism 5(7): 449–58, Doi: 10.1016/j.molmet.2016.04.007.

711

[20]

Blodgett, D.M., Nowosielska, A., Afik, S., Pechhold, S., Cura, A.J., Kennedy, N.J., et al.,

712

2015. Novel Observations From Next-Generation RNA Sequencing of Highly Purified Human

713

Adult and Fetal Islet Cell Subsets. Diabetes 64(9): 3172–81, Doi: 10.2337/db15-0039.

714

[21]

Bab-Dinitz, E., Albeck, S., Peleg, Y., Brumfeld, V., Gottschalk, K.E., Karlish, S.J.D., 2009. A

715

C-terminal lobe of the beta subunit of Na,K-ATPase and H,K-ATPase resembles cell adhesion

716

molecules. Biochemistry 48(36): 8684–91, Doi: 10.1021/bi900868e.

717

[22]

Vagin, O., Tokhtaeva, E., Sachs, G., 2006. The role of the beta1 subunit of the Na,K-

718

ATPase and its glycosylation in cell-cell adhesion. The Journal of Biological Chemistry 281(51):

719

39573–87, Doi: 10.1074/jbc.M606507200.

720

[23]

Ramírez-Salinas, G., Shoshani, L., Rosas-Trigueros, J.L., Huerta, C.S., Martínez-Archundia,

721

M., 2025. In silico studies provide new structural insights into trans-dimerization of β1 and β2

722

subunits of the Na+, K+-ATPase. PloS One 20(4): e0321064, Doi: 10.1371/journal.pone.0321064.

723

[24]

Tokhtaeva, E., Sachs, G., Sun, H., Dada, L.A., Sznajder, J.I., Vagin, O., 2012. Identification

724

of the amino acid region involved in the intercellular interaction between the β1 subunits of

725

Na+/K+ -ATPase. Journal of Cell Science 125(Pt 6): 1605–16, Doi: 10.1242/jcs.100149.

726

Journal Pre-proof

[25]

Parnaud, G., Lavallard, V., Bedat, B., Matthey-Doret, D., Morel, P., Berney, T., et al., 2015.

727

Cadherin engagement improves insulin secretion of single human β-cells. Diabetes 64(3): 887–

728

96, Doi: 10.2337/db14-0257.

729

[26]

Bosco, D., Rouiller, D.G., Halban, P.A., 2007. Differential expression of E-cadherin at the

730

surface of rat beta-cells as a marker of functional heterogeneity. The Journal of Endocrinology

731

194(1): 21–9, Doi: 10.1677/JOE-06-0169.

732

[27]

Dissanayake, W.C., Sorrenson, B., Shepherd, P.R., 2018. The role of adherens junction

733

proteins in the regulation of insulin secretion. Bioscience Reports 38(2): BSR20170989, Doi:

734

10.1042/BSR20170989.

735

[28]

Geron, E., Boura-Halfon, S., Schejter, E.D., Shilo, B.-Z., 2015. The Edges of Pancreatic Islet

736

β Cells Constitute Adhesive and Signaling Microdomains. Cell Reports 10(3): 317–25, Doi:

737

10.1016/j.celrep.2014.12.031.

738

[29]

Figtree, G.A., Liu, C.-C., Bibert, S., Hamilton, E.J., Garcia, A., White, C.N., et al., 2009.

739

Reversible oxidative modification: a key mechanism of Na+-K+ pump regulation. Circulation

740

Research 105(2): 185–93, Doi: 10.1161/CIRCRESAHA.109.199547.

741

[30]

Jaisser, F., Jaunin, P., Geering, K., Rossier, B.C., Horisberger, J.D., 1994. Modulation of the

742

Na,K-pump function by beta subunit isoforms. The Journal of General Physiology 103(4): 605–

743

23, Doi: 10.1085/jgp.103.4.605.

744

[31]

Yu, C., Xie, Z., Askari, A., Modyanov, N.N., 1997. Enzymatic properties of human Na,K-

745

ATPase alpha1beta3 isozyme. Archives of Biochemistry and Biophysics 345(1): 143–9, Doi:

746

10.1006/abbi.1997.0255.

747

[32]

Zacherl, S., La Venuta, G., Müller, H.-M., Wegehingel, S., Dimou, E., Sehr, P., et al., 2015. A

748

direct role for ATP1A1 in unconventional secretion of fibroblast growth factor 2. The Journal of

749

Biological Chemistry 290(6): 3654–65, Doi: 10.1074/jbc.M114.590067.

750

[33]

Rajasekaran, S.A., Gopal, J., Willis, D., Espineda, C., Twiss, J.L., Rajasekaran, A.K., 2004.

751

Na,K-ATPase beta1-subunit increases the translation efficiency of the alpha1-subunit in MSV-

752

MDCK cells. Molecular Biology of the Cell 15(7): 3224–32, Doi: 10.1091/mbc.e04-03-0222.

753

[34]

Stancill, J.S., Kasmani, M.Y., Khatun, A., Cui, W., Corbett, J.A., 2021. Single-cell RNA

754

sequencing of mouse islets exposed to proinflammatory cytokines. Life Science Alliance 4(6):

755

e202000949, Doi: 10.26508/lsa.202000949.

756

[35]

Grapengiesser, E., 1998. Unmasking of a periodic Na+ entry into glucose-stimulated

757

pancreatic beta-cells after partial inhibition of the Na/K pump. Endocrinology 139(7): 3227–31,

758

Doi: 10.1210/endo.139.7.6106.

759

Journal Pre-proof

[36]

Grapengiesser, E., Berts, A., Saha, S., Lund, P.E., Gylfe, E., Hellman, B., 1993. Dual effects

760

of Na/K pump inhibition on cytoplasmic Ca2+ oscillations in pancreatic beta-cells. Archives of

761

Biochemistry and Biophysics 300(1): 372–7, Doi: 10.1006/abbi.1993.1050.

762

[37]

Nita, I.I., Hershfinkel, M., Lewis, E.C., Sekler, I., 2015. A crosstalk between Na

channels,

763

pump and mitochondrial Na

transporters controls glucose-dependent cytosolic and

764

mitochondrial Na

signals. Cell Calcium 57(2): 69–75, Doi: 10.1016/j.ceca.2014.12.007.

765

[38]

Baukrowitz, T., Tucker, S.J., Schulte, U., Benndorf, K., Ruppersberg, J.P., Fakler, B., 1999.

766

Inward rectification in KATP channels: a pH switch in the pore. The EMBO Journal 18(4): 847–53,

767

Doi: 10.1093/emboj/18.4.847.

768

[39]

Sorensen, T., Vilsen, B., Andersen, J.P., 1997. Mutation Lys758 --> Ile of the sarcoplasmic

769

reticulum Ca2+-ATPase enhances dephosphorylation of E2P and inhibits the E2 to E1Ca2

770

transition. The Journal of Biological Chemistry 272(48): 30244–53, Doi:

771

10.1074/jbc.272.48.30244.

772

[40]

Thomas, R.C., 2011. The Ca(2+): H(+) coupling ratio of the plasma membrane calcium

773

ATPase in neurones is little sensitive to changes in external or internal pH. Cell Calcium 49(6):

774

357–64, Doi: 10.1016/j.ceca.2011.03.004.

775

[41]

Dickerson, M.T., Dadi, P.K., Zaborska, K.E., Nakhe, A.Y., Schaub, C.M., Dobson, J.R., et al.,

776

2022. Gi/o protein-coupled receptor inhibition of beta-cell electrical excitability and insulin

777

secretion depends on Na+/K+ ATPase activation. Nature Communications 13(1): 6461, Doi:

778

10.1038/s41467-022-34166-z.

779

[42]

El Moussawi, L., Chakkour, M., Kreydiyyeh, S.I., 2018. Epinephrine modulates Na+/K+

780

ATPase activity in Caco-2 cells via Src, p38MAPK, ERK and PGE2. PloS One 13(2): e0193139, Doi:

781

10.1371/journal.pone.0193139.

782

[43]

Ferjoux, G., Lopez, F., Esteve, J.-P., Ferrand, A., Vivier, E., Vely, F., et al., 2003. Critical role

783

of Src and SHP-2 in sst2 somatostatin receptor-mediated activation of SHP-1 and inhibition of

784

cell proliferation. Molecular Biology of the Cell 14(9): 3911–28, Doi: 10.1091/mbc.e03-02-0069.

785

[44]

Féraille, E., Carranza, M.L., Gonin, S., Béguin, P., Pedemonte, C., Rousselot, M., et al.,

786

1999. Insulin-induced stimulation of Na+,K(+)-ATPase activity in kidney proximal tubule cells

787

depends on phosphorylation of the alpha-subunit at Tyr-10. Molecular Biology of the Cell 10(9):

788

2847–59, Doi: 10.1091/mbc.10.9.2847.

789

[45]

Petrič, M., Vidović, A., Dolinar, K., Miš, K., Chibalin, A.V., Pirkmajer, S., 2021.

790

Phosphorylation of Na+,K+-ATPase at Tyr10 of the α1-Subunit is Suppressed by AMPK and

791

Enhanced by Ouabain in Cultured Kidney Cells. The Journal of Membrane Biology 254(5–6):

792

531–48, Doi: 10.1007/s00232-021-00209-7.

793

Journal Pre-proof

[46]

Beguin, P., Beggah, A.T., Chibalin, A.V., Burgener-Kairuz, P., Jaisser, F., Mathews, P.M., et

794

al., 1994. Phosphorylation of the Na,K-ATPase alpha-subunit by protein kinase A and C in vitro

795

and in intact cells. Identification of a novel motif for PKC-mediated phosphorylation. The Journal

796

of Biological Chemistry 269(39): 24437–45.

797

[47]

Zhang, M., Robitaille, M., Showalter, A.D., Huang, X., Liu, Y., Bhattacharjee, A., et al.,

798

2014. Progesterone receptor membrane component 1 is a functional part of the glucagon-like

799

peptide-1 (GLP-1) receptor complex in pancreatic β cells. Molecular & Cellular Proteomics: MCP

800

13(11): 3049–62, Doi: 10.1074/mcp.M114.040196.

801

[48]

Talbot, J., Joly, E., Prentki, M., Buteau, J., 2012. β-Arrestin1-mediated recruitment of c-

802

Src underlies the proliferative action of glucagon-like peptide-1 in pancreatic β INS832/13 cells.

803

Molecular and Cellular Endocrinology 364(1–2): 65–70, Doi: 10.1016/j.mce.2012.08.010.

804

[49]

Kimura, T., Allen, P.B., Nairn, A.C., Caplan, M.J., 2007. Arrestins and spinophilin

805

competitively regulate Na+,K+-ATPase trafficking through association with a large cytoplasmic

806

loop of the Na+,K+-ATPase. Molecular Biology of the Cell 18(11): 4508–18, Doi:

807

10.1091/mbc.e06-08-0711.

808

[50]

Flodby, P., Kim, Y.H., Beard, L.L., Gao, D., Ji, Y., Kage, H., et al., 2016. Knockout Mice

809

Reveal a Major Role for Alveolar Epithelial Type I Cells in Alveolar Fluid Clearance. American

810

Journal of Respiratory Cell and Molecular Biology 55(3): 395–406, Doi: 10.1165/rcmb.2016-

811

0005OC.

812

[51]

Thorens, B., Tarussio, D., Maestro, M.A., Rovira, M., Heikkilä, E., Ferrer, J., 2015. Ins1(Cre)

813

knock-in mice for beta cell-specific gene recombination. Diabetologia 58(3): 558–65, Doi:

814

10.1007/s00125-014-3468-5.

815

[52]

Dickerson, M.T., Bogart, A.M., Altman, M.K., Milian, S.C., Jordan, K.L., Dadi, P.K., et al.,

816

2018. Cytokine-mediated changes in K+ channel activity promotes an adaptive Ca2+ response

817

that sustains β-cell insulin secretion during inflammation. Scientific Reports 8(1): 1158, Doi:

818

10.1038/s41598-018-19600-x.

819

[53]

Corbin, K.L., West, H.L., Brodsky, S., Whitticar, N.B., Koch, W.J., Nunemaker, C.S., 2021. A

820

Practical Guide to Rodent Islet Isolation and Assessment Revisited. Biological Procedures Online

821

23(1): 7, Doi: 10.1186/s12575-021-00143-x.

822

[54]

Vierra, N.C., Dadi, P.K., Jeong, I., Dickerson, M., Powell, D.R., Jacobson, D.A., 2015. Type 2

823

Diabetes-Associated K+ Channel TALK-1 Modulates β-Cell Electrical Excitability, Second-Phase

824

Insulin Secretion, and Glucose Homeostasis. Diabetes 64(11): 3818–28, Doi: 10.2337/db15-

825

0280.

826

Journal Pre-proof

[55]

Zaborska, K.E., Dadi, P.K., Dickerson, M.T., Nakhe, A.Y., Thorson, A.S., Schaub, C.M., et al.,

827

2020. Lactate activation of α-cell KATP channels inhibits glucagon secretion by hyperpolarizing

828

the membrane potential and reducing Ca2+ entry. Molecular Metabolism 42: 101056, Doi:

829

10.1016/j.molmet.2020.101056.

830

[56]

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using real-

831

time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods (San Diego, Calif.) 25(4):

832

402–8, Doi: 10.1006/meth.2001.1262.

833

[57]

Shi, S.R., Key, M.E., Kalra, K.L., 1991. Antigen retrieval in formalin-fixed, paraffin-

834

embedded tissues: an enhancement method for immunohistochemical staining based on

835

microwave oven heating of tissue sections. The Journal of Histochemistry and Cytochemistry:

836

Official Journal of the Histochemistry Society 39(6): 741–8, Doi: 10.1177/39.6.1709656.

837

[58]

Tantama, M., Martínez-François, J.R., Mongeon, R., Yellen, G., 2013. Imaging energy

838

status in live cells with a fluorescent biosensor of the intracellular ATP-to-ADP ratio. Nature

839

Communications 4: 2550, Doi: 10.1038/ncomms3550.

840

[59]

Tewson, P.H., Martinka, S., Shaner, N.C., Hughes, T.E., Quinn, A.M., 2016. New DAG and

841

cAMP Sensors Optimized for Live-Cell Assays in Automated Laboratories. Journal of

842

Biomolecular Screening 21(3): 298–305, Doi: 10.1177/1087057115618608.

843

[60]

Uhlén, M., Fagerberg, L., Hallström, B.M., Lindskog, C., Oksvold, P., Mardinoglu, A., et al.,

844

2015. Proteomics. Tissue-based map of the human proteome. Science (New York, N.Y.)

845

347(6220): 1260419, Doi: 10.1126/science.1260419.

846

[61]

Pereira, R.M., Rodrigues, K.C. da C., Sant’Ana, M.R., Peruca, G.F., Morelli, A.P., Simabuco,

847

F.M., et al., 2020. Strength exercise reduces hepatic pyruvate carboxylase and gluconeogenesis

848

in DIO mice. The Journal of Endocrinology 247(2): 127–38, Doi: 10.1530/JOE-20-0193.

849

[62]

Puigserver, P., Rhee, J., Donovan, J., Walkey, C.J., Yoon, J.C., Oriente, F., et al., 2003.

850

Insulin-regulated hepatic gluconeogenesis through FOXO1-PGC-1alpha interaction. Nature

851

423(6939): 550–5, Doi: 10.1038/nature01667.

852

[63]

Ravier, M.A., Daro, D., Roma, L.P., Jonas, J.-C., Cheng-Xue, R., Schuit, F.C., et al., 2011.

853

Mechanisms of control of the free Ca2+ concentration in the endoplasmic reticulum of mouse

854

pancreatic β-cells: interplay with cell metabolism and [Ca2+]c and role of SERCA2b and SERCA3.

855

Diabetes 60(10): 2533–45, Doi: 10.2337/db10-1543.

856

[64]

Wen, X.-P., Long, G., Zhang, Y.-Z., Huang, H., Liu, T.-H., Wan, Q.-Q., 2022. Identification of

857

different proteins binding to Na, K-ATPase α1 in LPS-induced ARDS cell model by proteomic

858

analysis. Proteome Science 20(1): 10, Doi: 10.1186/s12953-022-00193-3.

859

Journal Pre-proof

[65]

Sepp, M., Sokolova, N., Jugai, S., Mandel, M., Peterson, P., Vendelin, M., 2014. Tight

860

coupling of Na+/K+-ATPase with glycolysis demonstrated in permeabilized rat cardiomyocytes.

861

PloS One 9(6): e99413, Doi: 10.1371/journal.pone.0099413.

862

[66]

Lewandowski, S.L., Cardone, R.L., Foster, H.R., Ho, T., Potapenko, E., Poudel, C., et al.,

863

2020. Pyruvate Kinase Controls Signal Strength in the Insulin Secretory Pathway. Cell

864

Metabolism 32(5): 736-750.e5, Doi: 10.1016/j.cmet.2020.10.007.

865

[67]

Hellman, B., Dansk, H., Grapengiesser, E., 2018. Somatostatin promotes glucose

866

generation of Ca2+oscillations in pancreatic islets both in the absence and presence of

867

tolbutamide. Cell Calcium 74: 35–42, Doi: 10.1016/j.ceca.2018.05.007.

868

[68]

Shigeto, M., Ramracheya, R., Tarasov, A.I., Cha, C.Y., Chibalina, M.V., Hastoy, B., et al.,

869

2015. GLP-1 stimulates insulin secretion by PKC-dependent TRPM4 and TRPM5 activation. The

870

Journal of Clinical Investigation 125(12): 4714–28, Doi: 10.1172/JCI81975.

871

[69]

Zhang, Q., Chibalina, M.V., Bengtsson, M., Groschner, L.N., Ramracheya, R., Rorsman,

872

N.J.G., et al., 2014. Na+ current properties in islet α- and β-cells reflect cell-specific Scn3a and

873

Scn9a expression. The Journal of Physiology 592(21): 4677–96, Doi:

874

10.1113/jphysiol.2014.274209.

875

[70]

Morton, M.J., Farr, G.A., Hull, M., Capendeguy, O., Horisberger, J.-D., Caplan, M.J., 2010.

876

Association with {beta}-COP regulates the trafficking of the newly synthesized Na,K-ATPase. The

877

Journal of Biological Chemistry 285(44): 33737–46, Doi: 10.1074/jbc.M110.141119.

878

[71]

Yoshimura, S.H., Iwasaka, S., Schwarz, W., Takeyasu, K., 2008. Fast degradation of the

879

auxiliary subunit of Na+/K+-ATPase in the plasma membrane of HeLa cells. Journal of Cell

880

Science 121(Pt 13): 2159–68, Doi: 10.1242/jcs.022905.

881

[72]

Kajikawa, M., Fujimoto, S., Tsuura, Y., Mukai, E., Takeda, T., Hamamoto, Y., et al., 2002.

882

Ouabain suppresses glucose-induced mitochondrial ATP production and insulin release by

883

generating reactive oxygen species in pancreatic islets. Diabetes 51(8): 2522–9, Doi:

884

10.2337/diabetes.51.8.2522.

885

[73]

Briant, L.J.B., Dodd, M.S., Chibalina, M.V., Rorsman, N.J.G., Johnson, P.R.V., Carmeliet, P.,

886

et al., 2018. CPT1a-Dependent Long-Chain Fatty Acid Oxidation Contributes to Maintaining

887

Glucagon Secretion from Pancreatic Islets. Cell Reports 23(11): 3300–11, Doi:

888

10.1016/j.celrep.2018.05.035.

889

[74]

Gulledge, A.T., Dasari, S., Onoue, K., Stephens, E.K., Hasse, J.M., Avesar, D., 2013. A

890

sodium-pump-mediated afterhyperpolarization in pyramidal neurons. The Journal of

891

Neuroscience: The Official Journal of the Society for Neuroscience 33(32): 13025–41, Doi:

892

10.1523/JNEUROSCI.0220-13.2013.

893

Journal Pre-proof

[75]

Morales-Reyes, I., Atwater, I., Esparza-Aguilar, M., Pérez-Armendariz, E.M., 2022. Impact

894

of biotin supplemented diet on mouse pancreatic islet β-cell mass expansion and glucose

895

induced electrical activity. Islets 14(1): 149–63, Doi: 10.1080/19382014.2022.2091886.

896

[76]

Jacobson, D.A., Kuznetsov, A., Lopez, J.P., Kash, S., Ammälä, C.E., Philipson, L.H., 2007.

897

Kv2.1 ablation alters glucose-induced islet electrical activity, enhancing insulin secretion. Cell

898

Metabolism 6(3): 229–35, Doi: 10.1016/j.cmet.2007.07.010.

899

[77]

Düfer, M., Haspel, D., Krippeit-Drews, P., Aguilar-Bryan, L., Bryan, J., Drews, G., 2009.

900

Activation of the Na+/K+-ATPase by insulin and glucose as a putative negative feedback

901

mechanism in pancreatic beta-cells. Pflugers Archiv: European Journal of Physiology 457(6):

902

1351–60, Doi: 10.1007/s00424-008-0592-4.

903

[78]

Meyer, D.J., Díaz-García, C.M., Nathwani, N., Rahman, M., Yellen, G., 2022. The Na+/K+

904

pump dominates control of glycolysis in hippocampal dentate granule cells. eLife 11: e81645,

905

Doi: 10.7554/eLife.81645.

906

[79]

Rorsman, P., Renström, E., 2003. Insulin granule dynamics in pancreatic beta cells.

907

Diabetologia 46(8): 1029–45, Doi: 10.1007/s00125-003-1153-1.

908

[80]

Henquin, J.C., 2000. Triggering and amplifying pathways of regulation of insulin secretion

909

by glucose. Diabetes 49(11): 1751–60, Doi: 10.2337/diabetes.49.11.1751.

910

[81]

Varadi, A., Tsuboi, T., Rutter, G.A., 2005. Myosin Va transports dense core secretory

911

vesicles in pancreatic MIN6 beta-cells. Molecular Biology of the Cell 16(6): 2670–80, Doi:

912

10.1091/mbc.e04-11-1001.

913

[82]

Kalwat, M.A., Thurmond, D.C., 2013. Signaling mechanisms of glucose-induced F-actin

914

remodeling in pancreatic islet β cells. Experimental & Molecular Medicine 45(8): e37, Doi:

915

10.1038/emm.2013.73.

916

[83]

Banerjee, A., Barry, V.A., DasGupta, B.R., Martin, T.F., 1996. N-Ethylmaleimide-sensitive

917

factor acts at a prefusion ATP-dependent step in Ca2+-activated exocytosis. The Journal of

918

Biological Chemistry 271(34): 20223–6, Doi: 10.1074/jbc.271.34.20223.

919

[84]

Matveyenko, A.V., Liuwantara, D., Gurlo, T., Kirakossian, D., Dalla Man, C., Cobelli, C., et

920

al., 2012. Pulsatile portal vein insulin delivery enhances hepatic insulin action and signaling.

921

Diabetes.. 61(9): 2269–79, Doi: 10.2337/db11-1462.

922

[85]

Satin, L.S., Butler, P.C., Ha, J., Sherman, A.S., 2015. Pulsatile insulin secretion, impaired

923

glucose tolerance and type 2 diabetes. Molecular Aspects of Medicine 42: 61–77, Doi:

924

10.1016/j.mam.2015.01.003.

925

Journal Pre-proof

[86]

Liu, L., Askari, A., 2006. Beta-subunit of cardiac Na+-K+-ATPase dictates the

926

concentration of the functional enzyme in caveolae. American Journal of Physiology. Cell

927

Physiology 291(4): C569-578, Doi: 10.1152/ajpcell.00002.2006.

928

[87]

Nichols, C.B., Rossow, C.F., Navedo, M.F., Westenbroek, R.E., Catterall, W.A., Santana,

929

L.F., et al., 2010. Sympathetic stimulation of adult cardiomyocytes requires association of AKAP5

930

with a subpopulation of L-type calcium channels. Circulation Research 107(6): 747–56, Doi:

931

10.1161/CIRCRESAHA.109.216127.

932

[88]

Nie, Y., Bai, F., Chaudhry, M.A., Pratt, R., Shapiro, J.I., Liu, J., 2020. The Na/K-ATPase α1

933

and c-Src form signaling complex under native condition: A crosslinking approach. Scientific

934

Reports 10(1): 6006, Doi: 10.1038/s41598-020-61920-4.

935

936

8. Figure Captions

937

938

Figure 1. β-Cell NKAβ1 localizes to the plasma membrane and influences insulin secretion.

939

B, C

Representative immunofluorescent staining of control mouse pancreas (

), healthy

940

human pancreas (

), and NKAβ1

pancreas (

) sections for insulin (green), glucagon (magenta),

941

and NKAβ1 (red).

Immunoblot of Ins1cre and NKAβ1

islet lysates probed for NKAβ1 (~35

942

kDa) and GAPDH protein.

(Left) Blood glucose levels during intraperitoneal glucose tolerance

943

tests (IPGTTs) in

Atp1b1

flox/flox

(white,

= 10 mice) and NKAβ1

(red,

= 9 mice) mice; (Right)

944

weight of mice measured prior to IPGTT.

Blood glucose levels during intraperitoneal pyruvate

945

tolerance tests (IPPTTs) in

Atp1b1

flox/flox

(white,

= 6 mice) and NKAβ1

(red,

= 7 mice) mice.

946

Dynamic glucose-stimulated insulin secretion (GSIS) and KCl-stimulated insulin secretion from

947

Ins1cre (blue,

= 3 islet preps) and NKAβ1

(red,

= 3 islet preps) islets.

Insulin area under

948

the curve (AUC) from time course shown in

Total insulin content in Ins1cre (blue,

= 4 islet

949

preps) and NKAβ1

(red,

= 4 islet preps) islets. Statistical analyses were performed using

950

unpaired two-sided t-tests (

) or two-way ANOVA with Šidák’s post-hoc multiple

951

comparisons tests (

); *

<0.05, ***

<0.001.

952

953

Figure 2. NKAβ1 regulates β-cell NKA currents and action potential kinetics.

954

A, B

Representative Ins1cre (blue, (

)) and NKAβ1

(red, (

)) β-cell

recordings within intact

955

islets showing typical responses to 200 nM somatostatin (SST) and 100 µM ouabain (Oua) in the

956

presence of 20 mM glucose (20G) + 1 mM tolbutamide (Tolb). Whole-cell β-cell currents were

957

measured (indicated by arrows) in response to a voltage ramp protocol (inset).

Ins1cre (blue,

958

= 10 β-cells) and NKAβ1

β-cell (red,

= 10 β-cells)

measured with 2 mM glucose (2G), 20G

959

+ 1 mM Tolb, 20G + 1 mM Tolb + 200 nM SST, and 20G + 1 mM Tolb + 200 nM SST+100 µM Oua.

960

Oua-sensitive NKA currents measured in Ins1cre (blue,

= 10 β-cells) and NKAβ1

β-cells

961

(red,

= 10 β-cells) with 20G + 1 mM Tolb.

Oua-sensitive NKA currents measured in Ins1cre

962

(blue,

= 10 β-cells) and NKAβ1

β-cells (red,

= 10 β-cells) with 20G + 1 mM Tolb + 200 nM

963

SST.

F–K

Quantification of Ins1cre (blue,

= 10 β-cells) and NKAβ1

β-cell (red,

= 10 β-cells)

964

action potential (AP) parameters with 20G + 1 mM Tolb and 20G + 1 mM Tolb + 200 nM SST +

965

Journal Pre-proof

100 µM Oua: AP amplitude (

), afterhyperpolarization (AHP) amplitude (

), max rise slope (

966

max decay slope (

), instantaneous frequency (

), and interevent interval (

). Statistic al analyses

967

were performed using unpaired two-sided t-tests (

D, E

) or two-way ANOVA with Šidák’s post-

968

hoc multiple comparisons tests (

C, F-K

); *

<0.05, **

<0.01, ***

<0.001, ****

<0.0001.

969

970

Figure 3. NKAβ1 facilitates glycolysis-dependent Ca²⁺ influx and tunes β-cell ATP consumption.

971

Representative Fura-2 AM Ca

traces from Ins1cre (blue) and NKAβ1

(red) islets stimulated

972

with 9 mM glucose (9G) or 9 mM α-ketoisocaproate (α-KIC).

Duration of glucose- (

= 14 islet

973

preps) and α-KIC-induced (

= 3 islet preps) Ca²⁺ decreases prior to Ca

influx in Ins1cre (blue)

974

and NKAβ1

(red) islets.

Total glucose- (

= 10 islet preps) and α-KIC-induced (

= 3 islet

975

preps) Ca²⁺ influx (AUC) into Ins1cre (blue) and NKAβ1

(red) islets.

Rate of glucose- (

= 12

976

islet preps) and α-KIC-induced (

=3 islet preps) Ca²⁺ influx into Ins1cre (blue) and NKAβ1

(red)

977

islets.

Representative Perceval HR ATP:ADP ratio response from Ins1cre (blue) and NKAβ1

978

(red) islets with 2G, 9G, and 5 µM FCCP. Arrows indicate where the rates of ATP:ADP ratio

979

change presented in panels

, and

were measured.

ATP:ADP ratio in Ins1cre (blue,

= 4

980

islet preps) and NKAβ1

(red,

= 4 islet preps) islets with 2G and 9G relative to Ins1cre controls

981

at 2G.

Rate of glucose-stimulated ATP production in Ins1cre (blue,

= 4 islet preps) and

982

NKAβ1

(red,

= 4 islet preps) islets.

Rate of ATP consumption during glucose stimulation in

983

Ins1cre (blue,

= 4 islet preps) and NKAβ1

(red,

= 4 islet preps) islets.

Rate of ATP depletion

984

following FCCP treatment in Ins1cre (blue,

= 4 islet preps) and NKAβ1

(red,

= 4 islet preps)

985

islets. Statistical analyses were performed using unpaired two-sided t-tests (

G-I

) or two-way

986

ANOVA with Šidák’s post-hoc multiple comparisons tests (

B-D, F

); *

<0.05, **

<0.01,

987

****

<0.0001.

988

989

Figure 4. NKAβ1 promotes efficient β-cell Na⁺ extrusion.

990

Representative Fura-2 AM Ca²⁺ traces from Ins1cre (blue) and NKAβ1

(red) islets stimulated

991

with 20 mM glucose (20G) in the presence of 1 mM Tolb.

Duration of glucose-induced

992

suppression of Ca²⁺ influx into Ins1cre (blue,

= 5 islet preps) and NKAβ1

(red,

= 5 islet

993

preps) islets.

Representative Fura-2 AM Ca²⁺ traces from Ins1cre (blue) and NKAβ1

(red)

994

islets stimulated with 9 mM α-KIC in the presence of 1 mM Tolb.

Glucose- (

= 5 islet preps)

995

and α-KIC-induced (

= 3 islet preps) Ca²⁺ changes (AUC) in Ins1cre (blue) and NKAβ1

(red)

996

islets.

Average SBFI AM Na⁺ traces in Ins1cre (blue) and NKAβ1

(red) islets stimulated with

997

9G in the presence of 125 µM diazoxide (DZ).

Intracellular Na⁺ in Ins1cre (blue,

= 3 islet

998

preps) and NKAβ1

(red,

= 3 islet preps) islets with 2G and 11 mM glucose (11G) relative to

999

Ins1cre controls at 2G.

Na⁺ efflux rate constants for Ins1cre (blue,

= 3 islet preps) and

1000

NKAβ1

(red,

= 3 islet preps) islets.

Na⁺ clearance half-lives for Ins1cre (blue,

= 3 islet

1001

preps) and NKAβ1

(red,

= 3 islet preps) islets. Statistical analyses were performed using

1002

unpaired two-sided t-tests (

) or two-way ANOVA with Šidák’s post-hoc multiple

1003

comparisons tests (

); *

<0.05, **

<0.01, ***

<0.001, ****

<0.0001.

1004

1005

Figure 5. NKAβ1 tunes inhibitory and stimulatory GPCR regulation of β-cell Ca²⁺ handling.

1006

Representative Fura-2 AM Ca²⁺ traces from Ins1cre (blue) and NKAβ1

(red) islets with 9G

1007

with or without 20 nM SST.

Ca²⁺ plateau fraction for Ins1cre (blue,

= 4 islet preps) and

1008

NKAβ1

(red,

= 4 islet preps) islets with 9G and 9G + 20 nM SST.

SST-induced change in Ca²⁺

1009

Journal Pre-proof

plateau fraction for Ins1cre (blue,

= 4 islet preps) and NKAβ1

(red,

= 4 islet preps) islets.

1010

Representative Fura-2 AM Ca²⁺ traces from Ins1cre (blue) and NKAβ1

(red) islets during α2-

1011

ADR activation with 100 nM clonidine (Clon) in the presence of 9G and 1 mM Tolb.

Clon-

1012

induced change in Ca²⁺ plateau fraction in Ins1cre (blue,

= 3 islet preps) and NKAβ1

(red,

1013

3 islet preps) islets in the presence of 9G and 1 mM Tolb.

Immunoblot of NKAβ1 (and

1014

associated proteins) immunoprecipitated from Ins1cre islet lysates then probed for NKAβ1

1015

(~100 kDa) and GLP-1R (~60 kDa) protein.

Representative Fura-2 AM Ca²⁺ traces from Ins1cre

1016

(blue) and NKAβ1

(red) islets during GLP-1R activation with 200 nM liraglutide (Lira) in the

1017

presence of 20G, 1 mM Tolb, and 200 mM SST.

Ca²⁺ plateau fraction in Ins1cre (blue,

= 4

1018

islet preps) and NKAβ1

(red,

= 4 islet preps) islets in the presence of 20G, 1 mM Tolb, and

1019

200 nM SST.

Lira-induced change in Ca²⁺ plateau fraction in Ins1cre (blue,

= 4 islet preps) and

1020

NKAβ1

(red,

= 4 islet preps) islets in the presence of 20G, 1 mM Tolb, and 200 nM SST.

1021

Representative cADDis cAMP traces during stimulation with 200 nM Lira in the presence of 11G.

1022

Representative cADDis cAMP traces during stimulation with 5 µM forskolin (Fsk) in the

1023

presence of 11G.

cAMP responses (AUC) to Lira (

= 4 islet preps) and Fsk (

= 3 islet preps) in

1024

Ins1cre (blue) and NKAβ1

(red) islets. Statistical analyses were performed using unpaired two-

1025

sided t-tests (

) or two-way ANOVA with Šidák’s post-hoc multiple comparisons tests (

1026

); *

<0.05, **

<0.01, ***

<0.001.

1027

1028

Figure 6. NKAβ1 regulates human β-cell Ca²⁺ handling and insulin secretion.

1029

ATP1B1

and

ATP1A1

mRNA expression in purified human β-cells transduced with scramble

1030

shRNA (blue, n = 4 islet donors) or

ATP1B1

shRNA (red, n = 4 islet donors).

Representative

1031

Fura-2 AM Ca²⁺ traces from dispersed human β-cells transduced with scramble (blue) or

ATP1A1

1032

(red) shRNA with 2G, 9G, and 9G + 200 nM SST.

Intracellular Ca²⁺ (AUC) in human β-cells

1033

transduced with scramble (blue,

= 3 islet donors) and

ATP1B1

shRNA (red,

= 3 islet donors)

1034

with 2G, 9G, and 9G + 200 nM SST.

Insulin content of human pseudoislets containing β-cells

1035

transduced with scramble (blue,

= 6 islet donors) or

ATP1B1

shRNA (red,

= 6 islet donors).

1036

Dynamic GSIS and KCl-stimulated insulin secretion from human pseudoislets containing β-cells

1037

transduced with scramble (blue,

= 6 islet donors) or

ATP1B1

shRNA (red,

= 6 islet donors).

1038

Data were normalized to insulin content.

Insulin secretion (AUC) from time course shown in

1039

Dynamic GSIS and KCl-stimulated insulin secretion from human pseudoislets containing β-cells

1040

transduced with scramble (blue,

= 6 islet donors) or

ATP1B1

shRNA (red,

= 6 islet donors).

1041

Data were normalized to pseudoislet number.

Insulin secretion (AUC) from time course shown

1042

. Statistical analyses were performed using paired two-sided t-tests (

) or two-way

1043

ANOVA with Šidák’s post-hoc multiple comparisons tests (

); *

<0.05, **

<0.01,

1044

****

<0.0001.

1045

1046

Figure 7. Model of NKA β-subunit control of β-cell ion homeostasis and signaling.

1047

In control β-cells, NKAβ1-containing NKA complexes preferentially utilize glycolytically-derived

1048

ATP to maintain high intracellular K

and low intracellular Na

. G

signaling promotes cyclical NKA

1049

activity via Src-mediated NKA phosphorylation, while GLP-1R signaling reduces NKA activity

1050

through cAMP production and PKA-mediated phosphorylation.

In NKAβ1 knockout β-cells,

1051

remaining NKAβ3-containing NKA complexes exhibit diminished pumping capacity, leading to

1052

accumulation. Decreased coupling of NKA complexes to glycolytic ATP pools results in ATP

1053

Journal Pre-proof

Ins1cre

NKA

∆β

kDa

NKA

GAPDH

Atp1b1

flox/flox

NKA

∆β

Blood glucose (mg/dL)

400

300

200

100

Time post-pyruvate (Minutes)

80 100 120

25
20
15
10

5
0

100

Insulin secretion (µg/L)

Time (Minutes)

Ins1cre

NKA

∆β

16.7G

30 mM KCl

(0-4 Min)

16.7G

(5-10 Min)

16.7G

(13-20 Min)

11G+KCl

(32-38 Min)

60
40
20

500

400

300

200

Insulin secretion (AUC)

Ins1cre

NKA

∆β

Atp1b1

flox/flox

NKA

∆β

Blood glucose (mg/dL)

Time post-insulin (Minutes)

500
400
300
200
100

80 100 120

✱✱✱

✱

Insulin content

relative to controls

1.5

1.0

0.5

0.0

✱✱✱

✱

Ins1cre

NKA

∆β

NKA

Insulin

Glucagon

Merged

50 µm

Human pancreas

50 µm

NKA

Insulin

Glucagon

Merged

Mouse pancreas

50 µm

NKA

Insulin

Glucagon

Merged

Mouse pancreas

Mouse weight (g)

Journal Pre-proof

Ins1cre

NKA

1KO

Membrane potential (

)

-100

-80

-60

-40

-20

60 Seconds

Membrane potential (

)

-100

-80

-60

-40

-20

60 Seconds

Ins1cre

NKA

1KO

20G+1mM Tolb

20G+1 mM Tolb

200 nM SST

200 µM Oua

Ins1cre

NKA

1KO

Basal NKA

currents (pA)

SST

-activated NKA

currents (pA)

Membrane potential (mV)

-100 -90 -80 -70 -60 -50

Membrane potential (mV)

-100 -90 -80 -70 -60 -50

Ins1cre

NKA

1KO

Membrane potential (mV)

-100

-80

-60

-40

-20

20G+T

olb

+SST +Oua

-60

-100 mV

-50 mV

-60

Voltage ramp

✱

✱✱✱✱

✱✱✱

✱✱

✱

✱✱

✱

✱✱✱✱

✱✱

✱

✱✱

✱

peak amplitude (mV)

20G+T

olb

20G+T

olb

+Oua

20G+T

olb

20G+T

olb

+Oua

AHP

amplitude (mV)

20G+T

olb

20G+T

olb

+Oua

Max rise slope (mV/ms)

20G+T

olb

20G+T

olb

+Oua

-3

-2

-1

Max decay slope (mV/ms)

20G+T

olb

20G+T

olb

+Oua

Instantaneous frequency (Hz)

20G+T

olb

20G+T

olb

+Oua

0.0

0.5

1.0

Interevent interval (s)

1.5

2.0

Ins1cre

NKA

1KO

Journal Pre-proof

0.2 F/F

min

300 Seconds

0.2 F/F

min

300 Seconds

Ins1cre

NKA

1KO

9 mM

-KIC

Ins1cre

NKA

1KO

✱

✱✱

0.0

0.5

1.0

1.5

2.0

ATP:ADP

ratio relative

to control at 2G

2.5

Ins1cre

NKA

1KO

ATP:ADP Ratio (F

480

405

)

300 Sec

0.5

1.0

1.5

2.0

2.5

Ins1cre

NKA

1KO

5 µM FCCP

Glucose-stimulated

∆

TP:ADP

ratio) (Slope)

0.000

0.005

0.010

0.015

0.050

0.025

-induced

∆

TP:ADP) ratio (Slope)

-6

-4

-2

FCCP-mediated

∆

TP:ADP) ratio (Slope)

-2.5

-2.0

-1.5

-1.0

-0.5

0.0

200

400

600

response time (Sec)

✱✱✱✱

-KIC

800

Islet glucose-stimulated

influx (Slope)

-KIC

✱✱

3.0

3.5

100

150

Islet glucose-stimulated

influx (AUC)

200

-KIC

250

Journal Pre-proof

0.000

0.001

0.002

0.003

0.004

0.005

100

200

300

400

500

Rate constant (1/Sec)

Half-life (Sec)

Ins1cre

NKA

1KO

✱

Ins1cre

NKA

1KO

125 µM DZ

11G

✱

300 Sec

1.12
1.10
1.08
1.06
1.04
1.02
1.00

Islet Na

relative to control

0.2 F/F

min

600 Seconds

20G

1 mM Tolb

Ins1cre

NKA

1KO

Glucose-stimulated

decrease (Seconds)

200

400

600

800

Ins1cre

NKA

1KO

✱

1000

100

150

200

Islet Ca

(AUC)

✱

✱✱✱✱

✱✱✱

✱✱

20G

α-KIC

0.2 F/F

min

600 Seconds

1 mM Tolb

9 mM

-KIC

Ins1cre

NKA

1KO

Ins1cre

NKA

1KO

250

Islet Na

relative to control

11G

✱

✱✱

Journal Pre-proof

0.1 F/F

min

600 Seconds

11G

200 nM Lira

Ins1cre

NKA

∆β

0.1 F/F

min

600 Seconds

11G

5 µM Fsk

0.0

0.1

0.2

0.3

0.4

0.5

plateau fraction

Ins1cre

NKA

∆β

0.2 F/F

min

600 Seconds

✱✱✱

9G 9G+SST

20 nM SST

✱✱

✱✱✱

0.0

0.2

0.4

0.6

0.8

SST

-induced plateau

fraction decrease

✱

Ins1cre

0.0

0.1

0.2

0.3

0.4

plateau fraction

✱✱

Ins1cre
NKA

∆β

NKA

∆β

0.1 F/F

min

600 Seconds

9G + 1 mM Tolb

100 nM Clon

✱✱

1.0

0.5

0.2 F/F

min

600 Seconds

Ins1cre

NKA

∆β

✱

Ins1cre

NKA

∆β

0.0

0.1

0.2

0.3

0.4

0.5

0.6

plateau fraction

20G+Tolb+SST

1 mM Tolb+200 nM SST

200 nM Lira

20G

0.0

0.4

1.2

1.4

ΔCa

plateau fraction

1.8

2.0

✱

+Lira

1.6

1.0

Ins1cre

NKA

∆β

GLP-1R

kDa

100

NKA

Mouse

Islets

-GLP-1R (Co-IP)

-NKA

1 (IP)

Islet cAMP

(AUC)

Lira

✱

Ins1cre

NKA

∆β

100

200

300

Fsk

150

250

350

Journal Pre-proof

ATP1B1

ATP1A1

mRNA

expression relative

to scramble shRNA

0.0

0.5

1.0

1.5

Scramble

shRNA

ATP1B1

shRNA

Scramble shRNA

ATP1B1 shRNA

Human

-cell Ca

340

380

)

0.65

0.70

0.75

0.80

0.85

Time (Minutes)

1000

2000

3000

200 nM SST

9G+SST

0.0

0.5

1.0

1.1

1.2

1.3

relative to

scramble shRNA

(AUC)

Scramble shRNA

ATP1B1 shRNA

✱

✱✱

✱

Time (Minutes)

Insulin secretion (% conte

)

Scramble shRNA

ATP1B1 shRNA

11G

30mM KCl

Insulin secretion (AUC)

0.2

0.3

0.4

0.5

9-15

Min

15-30

Min

72-81

Min

✱

0.0

0.1

0.0

1.0

2.0

3.0

Insulin content relative

to scramble shRNA

✱✱

Scramble shRNA

ATP1B1 shRNA

Scramble shRNA

ATP1B1 shRNA

✱✱✱✱

2.0

Insulin secretion (AUC)

0.04

0.06

0.08

0.00

0.02

9-15

Min

15-30

Min

72-81

Min

✱

✱✱

0.000

0.002

0.004

Insulin secretion (pmol/islet

)

Time (Minutes)

0.006

Scramble shRNA

ATP1B1 shRNA

11G

30mM KCl

Scramble shRNA

ATP1B1 shRNA

✱

0.10

Journal Pre-proof

Wild type

-cell

NKA

1 knockout

-cell

Elevated

ATP:ADP ratio

Augmented

-GPCR-induced

NKA activation

Diminished

GLP-1R-mediated

NKA inhibition

Reduced NKA

pumping function

Journal Pre-proof