Genetic and Physiological Effects of Insulin-Like Growth Factor-1 (IGF-1) on Human Urate Homeostasis

Abstract

Significance Statement

Hyperinsulinemia induces hyperuricemia by activating net renal urate reabsorption in the renal proximal tubule. The basolateral reabsorptive urate transporter GLUT9a appears to be the dominant target for insulin. By contrast, IGF-1 infusion reduces serum urate (SU), through mechanisms unknown. Genetic variants of IGF1R associated with reduced SU have increased IGF-1R expression and interact with genes encoding the GLUT9 and ABCG2 urate transporters, in a sex-specific fashion, which controls the SU level. Activation of IGF-1/IGF-1R signaling in Xenopus oocytes modestly activates GLUT9a and inhibits insulin's stimulatory effect on the transporter, which also activates multiple secretory urate transporters—ABCG2, ABCC4, OAT1, and OAT3. The results collectively suggest that IGF-1 reduces SU by activating secretory urate transporters and inhibiting insulin's action on GLUT9a.

Background

Metabolic syndrome and hyperinsulinemia are associated with hyperuricemia. Insulin infusion in healthy volunteers elevates serum urate (SU) by activating net urate reabsorption in the renal proximal tubule, whereas IGF-1 infusion reduces SU by mechanisms unknown. Variation within the IGF1R gene also affects SU levels.

Methods

Colocalization analyses of a SU genome-wide association studies signal at IGF1R and expression quantitative trait loci signals in cis using COLOC2, RT-PCR, Western blotting, and urate transport assays in transfected HEK 293T cells and in Xenopus laevis oocytes.

Results

Genetic association at IGF1R with SU is stronger in women and is mediated by control of IGF1R expression. Inheritance of the urate-lowering homozygous genotype at the SLC2A9 locus is associated with a differential effect of IGF1R genotype between men and women. IGF-1, through IGF-1R, stimulated urate uptake in human renal proximal tubule epithelial cells and transfected HEK 293T cells, through activation of IRS1, PI3/Akt, MEK/ERK, and p38 MAPK; urate uptake was inhibited in the presence of uricosuric drugs, specific inhibitors of protein tyrosine kinase, PI3 kinase (PI3K), ERK, and p38 MAPK. In X. laevis oocytes expressing ten individual urate transporters, IGF-1 through endogenous IGF-1R stimulated urate transport mediated by GLUT9, OAT1, OAT3, ABCG2, and ABCC4 and inhibited insulin's stimulatory action on GLUT9a and OAT3. IGF-1 significantly activated Akt and ERK. Specific inhibitors of PI3K, ERK, and PKC significantly affected IGF-1 stimulation of urate transport in oocytes.

Conclusions

The combined results of infusion, genetics, and transport experiments suggest that IGF-1 reduces SU by activating urate secretory transporters and inhibiting insulin's action.

Introduction

IGFs, secreted mainly from liver cells in response to growth hormone (GH) stimulation,^1–3 play important roles in growth, differentiation, development, kidney function,⁴ and renal pathophysiology.⁵ IGF-1 also inhibits insulin secretion.⁶ IGF receptors are heterotetrameric proteins of the type II receptor tyrosine kinase family.^3,7–11 IGF-1 binds to the ubiquitously expressed type I receptor (IGF-1R)¹² with greater affinity than IGF-2, whereas IGF-2 binds to the type II receptor (IGF-2R) with greater affinity than IGF-1.⁸ IGF-1 binding to IGF-1R triggers autophosphorylation of several tyrosine residues in the intracellular kinase domain of the beta-subunit of the receptor,¹³ activating signaling through insulin receptor substrates (IRS) 1–4 and Shc-dependent activation of phosphatidylinositol 3-kinase (PI3-K) and Akt (or protein kinase B).^2,3,14–17

Multiple studies have shown an association between IGF-1 signaling and circulating serum urate (SU) levels. IGF-1 infusion in healthy individuals decreases SU concentration and inhibits insulin and GH secretion.⁶ Genetic variation in both the IGF1 and IGF1R genes associates with SU levels.^18–20 IGF1R signaling also reportedly activates two secretory urate transporters, OAT1 and OAT3,^21,22 through PI3K and/or PKCζ.^23–26

The net urinary excretion of urate and serum SU levels is determined in large part by the balance between urate reabsorption and secretion within the renal proximal tubule, each transport process mediated by a separate subset of urate transporters. The urate transporters URAT1 and OAT10 reabsorb urate at the apical membrane in exchange for intracellular monocarboxylate anion^27–29; the sodium monocarboxylate transporters SMCT1 and SMCT2 mediate the intracellular accumulation of said anions, leading to “transactivation” of apical urate exchange.^28,30 Coding variants of human URAT1²⁷ and OAT10³¹ with reduced transport function are associated with decreased SU, emphasizing the importance of these apical transporters in human urate homeostasis. At the basolateral membrane, GLUT9, a high-capacity electrogenic urate transporter,^28,32–35 functions as the exclusive transporter for exit of reabsorbed urate into blood. Variation in SLC2A9, the gene encoding GLUT9, has the biggest single-gene effect on SU in genome-wide association studies (GWAS).^36,37 In the process of urate secretion, the basolateral organic anion exchangers OAT1 and OAT3 import urate into proximal tubular cells^30,38 with subsequent secretion at the apical membrane through NPT1, NPT4,^39,40 ABCG2,^{28,29,41–45} and ABCC4.⁴⁶ In addition, loss-of-function variants in human ABCG2 are associated with a decrease in intestinal excretion of urate; secretion by the intestine is responsible for approximately 30% of whole-body urate excretion.⁴⁷

In this study, we examined the effect of genetic variation at IGF1R on the expression of the IGF1 receptor, in addition to genetic interactions between urate-associated variants at SLC2A9, ABCG2, and IGF1R. To confirm the direct involvement of IGF-1 signaling on the regulation of urate transport, we examined the effect of IGF-1 on urate transport in human cells. To identify the urate transporters that are regulated by IGF-1 signaling, we examined the effect of IGF-1 on urate transport activity of ten individual urate transporters expressed in Xenopus oocytes.

Materials and Methods

Genetic Analysis

Colocalization analyses of the SU GWAS signal at IGF1R and expression quantitative trait loci (eQTL) signals in cis using COLOC2⁴⁸ was performed based on the method previously described⁴⁹ using the summary statistics from a SU GWAS⁵⁰ and eQTL data from GTEx (Genotype-Tissue Expression) version 8. Signals of genetic association that had a posterior probability for the H4 hypothesis >0.8 were considered to have a shared causal variant. All association analyses described below were performed using R v4.0.2 software. A linear model was used to test for sex-specific associations in the UK Biobank dataset (project 12611)⁴⁹ between rs12908437 and SU using the lm function adjusting by age. A forest plot depicting the effects of rs12908437 on SU in male and female was generated using R package meta (v.3.0-2). Linear regression for each model of inheritance (additive, recessive, and dominant) was run for stratified genotypes of the urate-associated genotypes at SLC2A9, ABCG2, and IGF1R. Akaike information criterion (a model selection tool) was used to select the most likely model. Heterogeneity between male and female was assessed using Cochran heterogeneity (Q) statistic. For the genetic interaction analyses, an interaction term was added to linear regression association analysis of genetic variants against the SU level. Analysis of epistasis between pairs of single nucleotide polymorphisms (SNPs) was performed in the United Kingdom Biobank dataset in which an interaction term was added to linear regression association analysis of genetic variants against the SU level. The dataset analyzed was 472,684 individuals of European ancestry, of whom 54.5% were women. The average age was 56.8 years.

Animals, Cell Lines, and Reagents

Mature female Xenopus laevis frogs were purchased from NASCO (Fort Atkinson, WI). The human kidney proximal tubule epithelial cell line (PTC-05) was obtained from Ulrich Hopfer (Case Western Reserve University, Cleveland, OH). Human embryonic kidney cell line (HEK 293T) was obtained from ATCC (Manassas, VA). DMEM and HAM’S F12 media, FBS, insulin, human EGF, penicillin, and streptomycin were purchased from Invitrogen (Carlsbad, CA). Human recombinant IGF-1 (Catalog # I3769-50UG), type IV collagen, transferrin, dexamethasone, interferon-gamma, ascorbic acid, sodium selenite (Na₂SeO₃), and triiodothyronine (T3) were purchased from SIGMA (St Louis, MO). Affinity purified rabbit anti-SLC2A9/GLUT9 (Code #BMP027), anti-SLC22A12/URAT1 (Code #BMP064), and anti-SLC22A13/ORCTL3/OAT10 (Code #BMP065) antibodies were purchased from MBL (Medical & Biological Laboratories Co. Ltd., Woburn, MA). Rabbit anti-IGF1R (Catalog #9750S), anti-ABCG2 (Catalog #4477S), anti-ABCC4 (Catalog #12705S), anti-phospho-Akt (Ser473) (Catalog #4060T), anti-Akt (Catalog #9272S), anti-phospho-IRS1 (Ser318) (Catalog #5610), anti-IRS1 (Catalog #2382S), anti-phospho-p44/42 MAPK(ERK1/2) (Thr202/Tyr204) (Catalog #4370T), anti-p44/42 MAPK(ERK1/2) (Catalog #4695S), anti-phospho-p38 MAPK (Thr180/Tyr182) (Catalog #4511T), anti-p38 MAPK (Catalog #9212S), anti–IGF-1R beta (D23H3) (Catalog #9750S), and anti-GAPDH (Catalog #5174S) antibodies were purchased from Cell Signaling Technology (Danvers, MA). Affinity purified rabbit anti-NPT1/SLC17A1 antibody (Catalog #PA550829) was purchased from Invitrogen (Carlsbad, CA). HRP-conjugated anti-rabbit IgG secondary antibody (Catalog #172-1019) was purchased from BIO-RAD (Hercules, CA) and ECL from Thermo Scientific (Rockford, IL). Genistein (soybean), LY 294002, bisindolylmaleimide 1-hydrochloride (BM1), PD 98059, SB203580, and U0126 were purchased from Calbiochem (Bilerica, MA). The [¹⁴C]-urate (specific activity: 50 mCi/mmol) was purchased from Moravec Inc (Brea, CA).

Cell Culture

All cells were routinely maintained in their respective appropriate growth medium in a humidified incubator at 37°C with 5% CO₂. Human PTC-05 cells were grown in 75-cm² flask or on tissue culture-treated sterile porous polycarbonate membrane (75-mm-diameter insert) in sterile polystyrene dish (COSTER, Corning, NY) as described previously.⁵¹ HEK 293T cells and other cells as indicated were grown in DMEM following supplier's instructions.

Expression Constructs

For expression of human urate transporters (GLUT9a, GLUT9b, URAT1, OAT10, OAT1, OAT3, OAT4, NPT1, ABCC4, and ABCG2) in oocytes, the full-length coding region of complementary DNAs (cDNAs) was cloned into the pGEMHE vector, which optimizes cRNA for expression in Xenopus laevis oocytes.⁵²

Functional Expression in Xenopus Oocytes

Studies using X. laevis oocytes have been performed in accordance with the Guide for the Care and Use of Laboratory Animals as adopted and promulgated by the US National Institutes of Health. Mature female X. laevis frogs were subjected to partial ovariectomy under tricaine-S anesthesia (0.17% for 15–20 minutes) followed by defolliculation of the oocytes as described previously.²⁸ The various X. laevis expression constructs in pGEMHE were linearized by Not1, Nhe1, or EcoR1 digestion. The cRNAs were in vitro synthesized by using T7 RNA polymerase (mMESSAGE mMACHINE; Ambion, Austin, TX) following the supplier's protocol, isopropanol-precipitated, washed twice with 70% ethanol, dried, dissolved in sterile nuclease-free water, and then stored at −80°C. The yield and integrity of the capped cRNA samples were assessed by spectroscopy (at 260 nm) and 1% agarose–formaldehyde gel electrophoresis, respectively. Approximately 18 hours after isolation, oocytes were microinjected with 50 nl of sterile water or 50 nl of a cRNA solution containing 25/12.5 ng of the indicated cRNA using fine-tipped micropipettes by a microinjector (World Precision Instrument Inc. Sarasota, FL) and then incubated in ND96 medium supplemented with pyruvate for 24 hours for protein expression.

RNA Extraction and RT-PCR

Total RNA from PTC-05 and other cells as indicated was extracted using spin columns with the RNeasy Mini Kit (QIAGEN, GmbH, Germany) following the manufacturer's instructions. Approximately 2 µg of total RNA, isolated from cells, were primed with poly-dT and random hexamers and then reverse-transcribed using AMV reverse transcriptase (New England Biolabs, Ipswich, MA). Equal amount of cDNA was used for PCR amplification keeping a negative control lacking template cDNA. Primers used for RT-PCR are presented in Supplemental Materials. All PCR products were confirmed by cloning and sequencing.

Immunohistochemistry

For immunohistochemistry, normal human kidney sections were obtained by the Renal Pathology Division at Brigham and Women's Hospital through an institutional review board (IRB)-approved protocol (Dr. Astrid Weins). Paraffin sections were dewaxed in Xylene, washed six times in absolute ethanol, incubated in a 1:1 diluted solution of 3% hydrogen peroxide and absolute ethanol, and then washed in tap water for 3–5 minutes. After antigen retrieval was achieved in a pressure cooker in Dako citrate buffer pH 6.0 at 120°C for 30 seconds, sections were rinsed in TBST and then incubated with rabbit anti-IGF1R (1:50 dilution), rabbit anti-ABCG2 (1:100 dilution), rabbit anti-URAT1 (1:100 dilution), anti-OAT10 (1:50 dilution), or rabbit anti-ABCC4 (1:200 dilution) antibody for 45 minutes and then rinsed with TBST for 10 minutes. Sections were then exposed to Labeled Polymer HRP anti-rabbit secondary antibody (DakoCytomation, K4011), washed in TBS, and then incubated with DAKO's DAB+ (3,3′-diaminobenzidine [DAB+] substrate-chromogen that results in a brown-colored precipitate at the antigen site) for 3–5 minutes, followed by washing in tap water and counterstaining with hematoxylin. Images of human kidney section were taken using a Nikon microscope (ECLIPSE 90i) at 20× (0.5-mm aperture) lateral magnifications.

Uptake and Efflux Assays

The [¹⁴C]-urate uptake and efflux experiments in Xenopus oocytes were performed as described previously.^28,53 To examine the effect of insulin on [¹⁴C]-urate uptake mediated by endogenous functional urate transporters in PTC-05 and HEK 293T cells, equal numbers (3 × 10⁶) of PTC-05 or HEK 293T cells were incubated with insulin (0.1–1.5 µM) in potassium-free^54,55 and serum-free defined isotonic medium (143 mM NaCl, 1.8 mM CaCl₂, 1 mM MgCl₂, 5 mM HEPES, pH 7.4) for 30 min at 25°C and then subjected to [¹⁴C]-urate uptake in the same medium for 1h at 25°C. For uptake experiments, each group had three separate wells of cells (typically 3X10⁶/well), urate uptakes were expressed as pmol/2.5–3×10⁶ cells/h, results are reported as means±SE, and statistical significance is defined as two-tailed P <0.05; each experiment was repeated three times. The [¹⁴C]-urate uptake activity was measured in isotonic K⁺-free uptake medium containing 20 μM [¹⁴C]-urate in 12-well plates after 1 hour of incubation at approximately 25°C. All uptake experiments using oocytes included at least 20 oocytes in each experimental group, as described,^28,53 using 40 μM [¹⁴C]-urate. For [¹⁴C]-urate efflux studies, control oocytes or oocytes expressing ABCG2 or ABCC4 were preinjected with 50 nl of 1500 µM [¹⁴C]-urate dissolved in efflux medium (K⁺-free medium, pH 7.4). Statistical significance was defined as two-tailed P <0.05, and the results were reported as means±SE. All the uptake experiments shown were performed more than three times for confirmation; data shown for each figure are from a single representative experiment.

Western Blotting

Western blotting was performed as described previously.²⁸ Approximately 48 hours postmicroinjection, oocytes (approximately 100) were lysed using a Teflon homogenizer in lysis buffer (50 mM Tris-HCl, pH 7.5, 50 mM NaCl, 1 mM EDTA, pH 8, 1% Triton X-100) supplemented with protease inhibitor (Roche, Indianapolis, IN) and phosphatase inhibitor (Thermo Fisher Scientific, Waltham, MA) cocktails. Western blotting was performed using appropriate primary and secondary antibodies. Approximately 30 µg of total protein of lysates was loaded per lane and fractionated using 8% SDS/PAGE gel electrophoresis and then transferred to polyvinylidene difluoride membrane. All the Western blotting experiments were performed more than three times for confirmation; data shown for each figure are from a single representative experiment. Quantitative analysis of the intensity of protein bands in Western blots was performed using KwikQuant Image Manager software (Kindle Biosciences, Greenwich, CT).

Statistics

Statistical analyses including linear regressions and significance were determined by the Student t test using SigmaPlot software. Transformation of data and curve fitting were made with SigmaPlot (Systat Software, Bangalore, Karnataka, India).

Results

Genetic Interaction between Urate-Associated Variants at SLC2A9 and ABCG2 and IGF1R

Here, we validate colocalization of the urate association at IGF1R and IGF1R expression using the summary statistics from Tin et al.⁵⁰ (Figure 1A) and Gene and Tissue Expression v8 (Figure 1B). The urate raising T-allele at the lead SNP (rs12908437) decreases IGF1R expression in heart left ventricle. Using the UKBB dataset, we observed a stronger effect at IGF1R (rs12908437) in women than men (for the T-allele β_Men=2.24 μmol/L, SE=0.24, β_Women=2.86 μmol/L, SE=0.20, heterogeneity P = 0.046, Figure 1C). We investigated the two-locus genotype-dependent pattern of association for each of SLC2A9 and ABCG2 with rs12908437 at IGF1R. This was performed by sex-stratified linear regression of the nine genotype groups comparing the effect on SU to the reference group of homozygous urate-lowering genotype at each genetic variant (Table 1). We observed that the only additive model with no significant effect was that of the GG genotype at SLC2A9 in men; in the presence of the C-allele, there was an effect in men (β=2.34 μmol/L, P = 1.38 × 10⁻²¹), whereas there was no urate-raising effect of the IGF1R T-allele in the absence of the SLC2A9 C-allele (β=1.06 μmol/L, P = 0.20) (Table 2). In women, however, there was an effect in both the presence (β=2.84 μmol/L, P = 3.33 × 10⁻⁴⁵) and absence of the C-allele (β=3.09 μmol/L, P = 6.71 x 10⁻⁷). Within the SLC2A9 GG genotype group we then evaluated what inheritance model for a IGF1R C-allele best fitted the data using the Akaike information criteria method—in men, the dominant model was preferred (AIC_dom=169,212.6, AIC_add=169,214.1, AIC_rec=169,215.8), and in women, the additive model was preferred (AIC_dom=193,212.3, AIC_add=193,204, AIC_rec=193,211.9). There was a heterogenous effect between sexes in the recessive model (P= 4×10⁻⁴), possibly the additive model (P = 0.05) but not the dominant model (P = 0.3). Finally, we tested whether the male-specific effects were due to an interaction effect between the IGF1R and SLC2A9 genotypes in men and found no evidence for epistatic interaction (P_int = 0.15).

Figure 1.

Colocalization of the association of SU and IGF1R expression at the IGF1R locus. Association of SU⁵⁰ (A) and IGF1R expression (B) at the IGF1R locus (±500 kb). The strength of LD, as measured by the r2 with rs12908437, is represented by the color of each point according to the legend. The plot was generated using locuszoom. (C) Forest plot of rs12908437 association with SU in the UKBB dataset (European) stratified by sex. Associations were adjusted by age.

Table 1.

The nature of the interactions at SLC2A9 and ABCG2 with rs12908437 at IGF1R were investigated by standard linear regression of the nine genotype groups comparing the effect on urate to the reference group of homozygous urate-lowering genotype at each genetic variant

			SLC2A9 (rs4447862) C = Urate-Raising
			Women (Δ Urate μmol/L, [SE])				Men (Δ Urate μmol/L, [SE])
			GG	GC	CC		GG	GC	CC
IGF1R (rs12908437) T = urate-raising	CC	Ref	38.69 0.77	62.26 0.80	62.26	Ref	23.32 0.98	36.73 0.98	36.73
	TC	2.63 0.91	41.20 0.77	65.34 0.80	62.71	2.29 1.22	26.30 0.98	39.31 0.97	37.02
	TT	6.51 1.34	43.82 0.89	68.36 0.87	61.84	1.23 1.77	27.72 1.12	40.86 1.08	39.63
		6.501	5.13	6.11		1.23	4.40	4.13

			ABCG2 (rs2231142) T = Urate-Raising
			Women (Δ Urate μmol/L, [SE])				Men (Δ Urate μmol/L, [SE])
			GG	TG	TT		GG	TG	TT
IGF1R (rs12908437) T = urate-raising	CC	Ref	11.45 0.52	25.60 1.82	25.60	Ref	17.08 0.63	34.04 2.11	34.04
	TC	2.73 0.3	14.46 0.49	26.68 1.69	23.95	2.85 0.39	19.72 0.59	32.18 2.07	29.33
	TT	5.41 0.47	18.86 0.84	28.84 3.11	23.43	4.50 0.56	19.73 1.01	34.56 3.77	30.06
		5.41	7.41	3.24		4.50	2.65	0.52

Table 2.

The relationship of SLC2A9 with rs12908437 at IGF1R investigated by standard linear regression in the presence and absence of the C-allele at SLC2A9

	SLC2A9 (rs4447862) C = Urate-Raising
	Women (Δ Urate μmol/L)		Men (Δ Urate μmol/L)
	GG	GC/CC	GG	GC/CC
IGF1R (rs12908437)	3.09*	2.84*	1.06	2.34*

*Statistically significant (P <0.05) associations.

IGF-1 Stimulates Urate Uptake in Human Proximal Tubule Epithelial Cells (PTC-05) and HEK 293T Cells

We first examined the effect of human recombinant IGF-1 on [¹⁴C]-urate uptake in PTC-05 and HEK 293T cells, which express IGF-1R (Figure 2A) and multiple urate transporters (Figure 2, B and C). We detected a statistically significant increase in [¹⁴C]-urate uptake in both cells (Figure 2D) in the presence of extracellular IGF-1 (30–60 nM)^56,57 in potassium-free^54,55 and serum-free isotonic medium (pH 7.4). In transiently transfected HEK 293T cells overexpressing recombinant human IGF-1R (2.5-fold increase) (Figure 2E; Supplemental Figure 1A) and exposure to extracellular IGF-1 (60 nM) caused a 57% increase in [¹⁴C]-urate uptake activity (Figure 2, F and G). In PTC-05 cells, exposure to IGF-1 for approximately 1.5 hours induced the expression of GLUT9 (approximately 1.35-fold) and ABCC4 (approximately two-fold increase) (Figure 2C; Supplemental Figure 1, B and E). However, in HEK 293T cells, exposure to IGF-1 induced the expression of GLUT9 (approximately four-fold increase), ABCG2 (approximately four-fold), and NPT1 (approximately 1.8-fold increase) but decreased the expression of ABCC4 (approximately 43% decrease) (Figure 2C; Supplemental Figure 2). In both cell types, exposure to IGF-1 for 1.5 hours did not significantly change the expression level of IGF-1R (Figure 2A).

Figure 2.

IGF-1, through its receptor (IGF-1R), stimulates [¹⁴C]-urate uptake in human renal proximal tubule epithelial cells (PTC-05) and HEK 293T cells. (A) Detection of the endogenous IGF-1Rβ protein expression in PTC-05 and HEK 293T cells by Western blotting using anti-IGF-1Rβ antibody. Approximately 30 µg of total protein/lane of each cell lysate was analyzed. (B) RT-PCR detection of expression of mRNAs for endogenous urate transporters in human kidney, PTC-05, and HEK 293T cells using appropriate introns-spanning primer sets (see methods). Each negative control contains everything except the respective template cDNA. (C) Detection of indicated endogenous urate transporters and GAPDH protein expression in PTC-05 and HEK 293T cells by Western blotting using anti-GLUT9, anti-OAT10, anti-ABCG2, anti-ABCC4, anti-NPT1, and anti-GAPDH antibodies. (D) The total net [¹⁴C]-urate uptake activities of the endogenous urate transporters in PTC-05 and HEK 293T cells were measured in the absence and presence of IGF-1 (30–60 nM) in K⁺-free isotonic medium (pH 7.4) containing 20 μM [¹⁴C]-urate after 1 hour of incubation at approximately 25°C. (E) Overexpression of IGF-1Rβ protein expression in transiently transfected HEK 293T cells was detected by Western blotting using anti-IGF-1Rβ antibody. Approximately 30 µg of total protein/lane of each cell lysate was analyzed. (F) The total net [¹⁴C]-urate uptake activities of the endogenous urate transporters in HEK 293T cells and in transiently transfected HEK 293T cells overexpressing IGF-1R was measured in the absence and presence of IGF-1 (60 nM). (G) The time-course plot showing the stimulatory effect of IGF-1 (60 nM) on the total net [¹⁴C]-urate uptake by endogenous urate transporters in HEK 293T and in transiently transfected HEK 293T cells overexpressing IGF-1R. (H) Inhibition profile of the total net [¹⁴C]-urate uptake activities of endogenous urate transporters, in transiently transfected HEK 293T cells overexpressing IGF-1R, in the presence of uricosuric drugs and antiuricosuric agents. The stimulatory effect of IGF-1 (60 nM), on the total net [¹⁴C]-urate uptake by endogenous urate transporters in transiently transfected HEK 293T cells overexpressing IGF-1R, was measured in a K⁺-free isotonic uptake medium (pH 7.4) in the absence or presence of 10 mM nicotinate (Nico), 10 mM pyrazinoate (PZA), 10 mM salicylate (Sal), 100 µM benzbromarone (Benz), 100 µM tranilast (Trani), or 1.0 mM probenecid (Proben) in the extracellular medium. The uricosuric drug, benzbromarone, tranilast, or probenecid was dissolved in DMSO. The results of urate uptake are the average of three independent experiments±SEM. Asterisk (*), P <0.001 compared with appropriate control.

In transiently transfected HEK 293T cells overexpressing IGF-1R (Figure 2E), the IGF-1 stimulation of urate uptake was inhibited by cis-inhibiting and uricosuric compounds (Figure 2G).

IGF1 Stimulation of Urate Uptake in PTC-05 and HEK 293T Cells Requires Activation of IRS1, Akt, p44/42 MAPK (ERK1/2), and p38 MAPK

To elucidate the intracellular mechanism(s) whereby IGF-1 activates urate transporters in PTC-05 or HEK 293T cells, we exposed these cells to IGF-1 (60 nM) in potassium-free^54,55 and serum-free isotonic medium and then analyzed the cell lysates by Western blotting. The results show robust phosphorylation-dependent activation of insulin receptor substrate 1 (IRS1) (approximately 3.2-fold), Akt (approximately 23-fold), and ERK (approximately six-fold) by IGF-1 in PTC-05 cell (Figure 3, A and B; Supplemental Figure 3, A-C). In HEK 293T cells, we found approximately 3.3-fold activation of IRS1, 13-fold activation of Akt, and 5.2-fold activation for ERK by IGF-1 (Figure 3, A and C; Supplemental Figure 3, D-F). In addition, in HEK 293T cells, we detected approximately 14-fold activation of p38 MAPK by IGF-1 (Figure 3C; Supplemental Figure 3G).

Figure 3.

IGF-1 stimulates urate uptake in PTC-05 and HEK 293T cells through the activation of IRS1, Akt, p44/42 MAPK (ERK), and/or p38 MAPK. (A). IGF-1 (60 nM) activates IRS1 in PTC-05 and HEK 293Tcells. (B) IGF-1 (60 nM) activates Akt and ERK in PTC-05 cells. (C) IGF-1 (60 nM) activates Akt, ERK, and p38 MAPK in HEK293T cells. Western blot analyses of the lysates (approximately 30 µg of total protein/sample/lane) of PTC-05/HEK 293T cells, treated with IGF-1 (60 nM) in K⁺-free isotonic medium at approximately 25°C for varying intervals of time (0–60 min), were performed using rabbit anti-phospho IRS1 (Ser318), anti-phospho Akt (Ser473), anti-phospho p44/42 MAPK(ERK1/2) (Thr202/Tyr204), or anti-phospho p38 MAPK (Thr180/Tyr182) antibody. The total protein for IRS1, Akt, p44/42 MAPK(ERK1/2), or p38 MAPK was detected with rabbit anti-IRS1, anti-Akt, anti-p44/42 MAPK(ERK1/2), or anti-p38 MAPK antibody, respectively. (D) The stimulatory effect of IGF-1 (60 nM) on total net [¹⁴C]-urate uptake by endogenous urate transporters in transfected HEK 293T cells overexpressing IGF-1R was measured in a K⁺-free isotonic uptake medium (pH 7.4) in the absence or presence of the natural protein tyrosine kinase inhibitor (genistein; 75 µM), phosphatidylinositol 3-kinase (PI3K) inhibitor (LY 294002; 50 µM), MEK/ERK inhibitor (20 µM PD 98059 or U0126), p38 MAPK inhibitor (10 µM SB 203580), or protein kinase C inhibitor (10 µM bisindolylmaleimide I (BMI). Before the urate uptake experiment, cells were preincubated without or with IGF-1 (60 nM) and drugs as indicated for 30 minutes at approximately 25°C. The total net [¹⁴C]-urate uptake activity was measured after incubating cells for 1 hour in K⁺-free isotonic medium, containing 40 μM [¹⁴C]-urate at approximately 25°C. Results of uptake experiments are the average of three independent experiments±SEM. Asterisk (*), P<0.001 compared with appropriate control.

In transiently transfected HEK 293T cells overexpressing IGF-1R (Figure 2E), the IGF-1 stimulation of [¹⁴C]-urate uptake strongly inhibited by a protein tyrosine kinase (PTK) inhibitor (Genistein, 75 µM) or a phosphoinositide 3-kinase (PI3K) inhibitor (LY 294002, 50 µM) (Figure 3D). Moreover, the IGF-1 stimulation of urate transport was approximately 54–60% inhibited by a MEK/ERK inhibitor (PD 98059 or U0126, 20 µM) and 53% inhibited by a p38 MAPK inhibitor (SB 203580, 10 µM) (Figure 3D), but unaffected by protein kinase-C (PKC) inhibitor (10 µM bisindolylmaleimide 1).

IGF-1 Stimulates Urate Transport Mediated by GLUT9, OAT1, OAT3, ABCG2, and ABCC4

To identify urate transporters that are functionally activated by IGF-1, we exploited the endogenous IGF-1R signaling pathways of X. laevis oocytes.^58–62 X. laevis oocytes have been successfully used for decades for functional characterization of various urate transporters.^{27,28,32,33,53,63,64} Exposure to IGF-1 did not significantly change the expression level of IGF-1R in oocytes (Figure 4A). We individually expressed several urate transporters in oocytes by microinjecting their respective in vitro synthesized cRNA and incubating for 48 hours (Figure 4B). These oocytes were then exposed to 60nM IGF-1 and then subjected to urate uptake for 1 hour in the presence of IGF-1 (60nM) in K⁺-free medium. The results show approximately 2-fold, 1.4-fold, 1.7-fold, and 2.7-fold increase in the urate uptake activity of GLUT9a, GLUT9b, OAT1, and OAT3, respectively (Figure 4C and D), by IGF-1; this [¹⁴C]-urate uptake was linear over time (Figure 4, E and F). In addition, we found approximately 14% and 27% increases in the urate efflux activity of ABCG2 and ABCC4 (Figure 4, G and H) in the presence of IGF-1. However, IGF-1 did not affect the urate uptake activities of URAT1, OAT10, OAT4, and NPT1 (Figure 4C).

Figure 4.

IGF-1, through its receptor (IGF-1R), stimulates urate transport mediated by GLUT9a, GLUT9b, OAT1, OAT3, ABCC4, and ABCG2 expressed in Xenopus laevis oocytes. (A) Detection of IGF-1Rβ expression in Xenopus laevis oocytes by Western blot analyses of the lysates (approximately 30 µg of total protein/sample/lane) of oocytes, untreated or treated with IGF-1 (60 nM) at approximately 25°C in K⁺-free isotonic medium for 0–90 minutes, using anti-IGF-1Rβ antibody. (B) The integrity and relative amount of in vitro synthesized cRNAs of various urate transporters as indicated were examined by 1.0% agarose-formaldehyde gel electrophoresis before microinjection into oocytes. The expression of GLUT9a, GLUT9b, URAT1, or OAT10 protein, in the lysates (approximately 20 µg of total protein/sample/lane) of oocytes microinjected with their respective cRNAs, was verified by Western blot analyses using affinity purified anti-GLUT9, anti-URAT1, or anti-OAT10 antibodies. (C, D) The [¹⁴C]-urate transport activities of human GLUT9a, GLUT9b, URAT1, OAT10, OAT4, NPT1, murine OAT3, and OAT1 expressed from their respective microinjected cRNAs in oocytes were measured in K⁺-free isotonic medium in the absence and presence of IGF-1(30–60 nM). Each oocyte was microinjected with 50 nl of cRNA solution (approximately 25 ng of cRNA) of each indicated urate transporter and then incubated in ND96 medium supplemented with 2.5 mM pyruvate for 48 hours at 16–18°C. Oocytes expressing URAT1 and OAT10 were preloaded with nicotinate by overnight incubation in ND96 medium containing 10 mM nicotinate, and each oocyte expressing OAT4 was preloaded with 50 nl of 10 mM Na-maleate by microinjection before [¹⁴C]-urate uptake. (E) The time-course plot showing the stimulatory effect of IGF-1 (60 nM) on [¹⁴C]-urate uptake mediated by GLUT9a or GLUT9b expressed in oocytes. (F) The time-course plot showing the stimulatory effect of IGF-1 (60 nM) on [¹⁴C]-urate uptake mediated by human OAT1 or murine OAT3 expressed in oocytes. Before [¹⁴C]-urate uptake, oocytes were incubated in K⁺-free isotonic medium for 30 min at room temperature (approximately 25°C) in the absence or presence of IGF-1 (60 nM). The uptake of [¹⁴C]-urate was measured in K⁺-free isotonic medium, containing 40 μM [¹⁴C]-urate after 1 hour of incubation at room temperature (approximately 25°C). (G) IGF-1 (60 nM) stimulates the [¹⁴C]-urate efflux activity of ABCG2 and ABCC4. (H) The time-course plot showing the stimulatory effect of IGF-1 (60 nM) on [¹⁴C]-urate efflux mediated by human ABCG2 expressed in oocytes. The effect of IGF-1 (60 nM) on [¹⁴C]-urate efflux activity of human ABCG2 in oocytes was measured in K⁺-free isotonic medium (pH 7.4; see methods). Asterisk (*), P <0.001 compared with urate uptake/efflux in absence of IGF-1. All data are mean±SEM, with n=12–15 oocytes per group.

IGF-1 Activates Multiple Urate Transporters through Akt and ERK Pathways

To elucidate mechanisms whereby IGF-1/IGF-1R signaling functionally activates urate transporters in Xenopus oocytes, we treated oocytes with IGF-1 (60 nM) for varying intervals of time and examined the lysates by Western blotting for the activation of Akt, ERK, and p38 MAPK. We found approximately 24-fold phosphorylation-dependent activation of Akt (Ser-473) and 3.2-fold activation of p44/42 MAPK (Erk) (Thr202/Tyr204) in response to IGF-1 for 1h (Figure 5A). However, we could not detect the activation of p38 MAPK (Thr180/Tyr182) in oocytes (Figure 5A).

Figure 5.

IGF-1, through its receptor (IGF-1R), stimulates of urate transport mediated by GLUT9a, GLUT9b, OAT1, murine OAT3 (mOAT3), ABCG2, and ABCC4 through activation of Akt and/or p44/42 MAPK (ERK) in Xenopus oocytes. (A) IGF-1 (60 nM) activates Akt and ERK in oocytes. Western blot analyses of the lysates (approximately 30 µg of total protein/sample/lane) of oocytes, treated with IGF-1 (60 nM) at approximately 25°C in K⁺-free isotonic medium (see methods) for varying intervals of time (0–60 min), were performed using rabbit anti-phospho Akt (Ser473), anti-phospho p44/42 MAPK(ERK1/2) (Thr202/Tyr204), or anti-phospho p38 MAPK (Thr180/Tyr182) antibody. The total protein for Akt, p44/42 MAPK(ERK1/2), or p38 MAPK was detected with rabbit anti-Akt, anti-p44/42 MAPK(ERK1/2), or anti-p38 MAPK antibody, respectively. (B-F) The stimulatory effect of IGF-1 (60 nM) on [¹⁴C]-urate uptake mediated by GLUT9a, GLUT9b, OAT1, or murine OAT3 and [¹⁴C]-urate efflux by ABCG2 or ABCC4 expressed in oocytes was measured in a K⁺-free isotonic uptake medium (pH 7.4) in the absence and presence of the natural PTK inhibitor (genistein; 75 µM) or PI3K inhibitor (LY 294002; 50 µM), MEK/ERK inhibitor (20 µM PD 98059), p38 MAPK inhibitor (10 µM SB 203580), or protein kinase C inhibitor (10 µM bisindolylmaleimide I (BMI). Before the urate uptake experiment, oocytes were preincubated in a K⁺-free isotonic uptake medium (pH 7.4; see methods) without or with IGF-1 (60 nM) and indicated drugs in a 12-well plate for 30 min at approximately 25°C. The [¹⁴C]-urate uptake activity was measured after incubating oocytes for 1 hour at approximately 25°C in K⁺-free isotonic medium. The [¹⁴C]-urate efflux activity was measured as described in methods. The drugs were dissolved in DMSO to make stock solution. Asterisk (*), P <0.001 compared with urate uptake/efflux in absence of IGF-1. All data are mean±SEM, with n=12–15 oocytes per group.

In addition, in the presence of a PI3K inhibitor (LY 294002; 50 μM), IGF-1-stimulation of urate transport by GLUT9 isoforms, OAT3, OAT1, ABCG2, and ABCC4 was completely inhibited (Figure 5, B–F). PI3K inhibition also significantly inhibited their basal urate transport activities (Supplemental Figure 4). However, although the PTK inhibitor genistein almost completely inhibited the IGF-1 stimulation of urate uptake by GLUT9a, it inhibited only approximately 51% of the IGF-1–stimulated activity of GLUT9b (Figure 5B). The basal urate transport activity of GLUT9a was significantly affected by genistein without any effect on GLUT9b (Supplemental Figure 4). Genistein, however, almost completely inhibited IGF-1 stimulation of urate uptake activities of OAT1, OAT3, ABCG2, and ABCC4 and significantly affected their basal urate transport activities (Figure 5B and C, E and F).

In oocytes, the ERK inhibitor PD 98059 (20 µM) almost completely abrogated the IGF-1 stimulation of GLUT9b, OAT1, OAT3, ABCG2, and ABCC4, with significantly less effect on GLUT9a (approximately 50% inhibition) (Figure 5, C–F). However, although the ERK inhibitor also significantly affected the basal urate uptake activity of GLUT9b, OAT1, ABCG2, and ABCC4 (Figure 5, C–F; Supplemental Figure 4), it barely affected the basal urate uptake activity of GLUT9a (Figure 5D). The p38 MAPK inhibitor (SB 203580, 10 µM) had very little effect on GLUT9a (approximately 14% inhibition), significant effect on GLUT9b (approximately 54% inhibition), murine OAT3 (approximately 67% inhibition), and ABCC4 (approximately 34% inhibition), but no effect on IGF-1 activation of urate transport by OAT1 and ABCG2 (Figure 5, C–F). The PKC inhibitor (10 µM bisindolylmaleimide 1) almost completely abrogated the IGF-1 activation of GLUT9a and GLUT9b, with very little effect on ABCG2 (approximately 11% inhibition) but no effect on OAT1, OAT3, and ABCC4 (Figure 5, C–F).

IGF-1 Affects Insulin Stimulation of Urate Transport

To explore the interactions between IGF-1 and insulin, we measured [¹⁴C]-urate content in PTC-05 cells grown on porous polycarbonate membranes in the absence and presence of IGF-1 (60 nM), insulin (1.0 µM), or both. We found that IGF-1 significantly inhibited insulin's stimulatory effect on [¹⁴C]-urate uptake in permeable membrane attached PTC-05 cells (Figure 6A). In oocytes, we did not detect any additive stimulatory effect of IGF-1 over insulin's stimulatory effect on urate transport by GLUT9, OAT1, OAT3, ABCG2, and ABCC4. We did however detect significant blunting of insulin's stimulatory effect on urate transport by GLUT9a and OAT3 in the presence of IGF-1 (Figure 6, B–E).

Figure 6.

IGF-1 affects insulin stimulation of urate transport by GLUT9a and OAT3. (A) The total net [¹⁴C]-urate uptake activities of the endogenous urate transporters in PTC-05 cells were measured in the absence and presence of IGF-1 (60 nM), insulin (1.0 µM), or both in K⁺-free isotonic medium (pH 7.4) containing 20 μM [¹⁴C]-urate after 1 hour of incubation at approximately 25°C. The [¹⁴C]-urate uptake activities of human GLUT9a (B), GLUT9b (C), OAT1, murine OAT3 (D), and [¹⁴C]-urate efflux activities of human ABCG2 and ABCC4 (E), expressed in oocytes, were measured in K⁺-free isotonic medium in the absence and presence of IGF-1 (60 nM), insulin (1.0 µM), or both. Each oocyte was microinjected with 50 nl of cRNA solution (approximately 25 ng of cRNA) of each indicated urate transporter and then incubated in ND96 medium supplemented with 2.5 mM pyruvate for 48 hours at 16–18°C. Asterisk (*), P <0.001 compared with urate uptake/efflux in absence of IGF-1. All oocyte data are mean±SEM, with n=12–15 oocytes per group.

IGF-1 Has No Influence on Nicotinate Transport by OAT10, URAT1, SMCT1, and SMCT2

URAT1 and OAT10 not only transport urate^27,28,65 but also transport nicotinate.^27,28,65 To assess whether IGF-1 has any potential regulatory role on nicotinate transport, we examined the effect of IGF-1 on [¹⁴C]-nicotinate uptake activity of OAT10 and URAT1 along with two other human nicotinate transporters, SMCT1 and SMCT2,²⁸ in Xenopus oocytes. The results show no significant effect of IGF-1 on nicotinate uptake mediated by OAT10, SMCT1, SMCT2, or URAT1 (Figure 7).

Figure 7.

IGF-1 does not have any significant effect on nicotinate transporters. The [¹⁴C]-nicotinate transport activities of human URAT1, OAT10, SMCT1, and SMCT2, expressed from their respective microinjected cRNAs in oocytes, were measured in K⁺-free isotonic medium (see methods) in the absence and presence of IGF-1 (60 nM). The [¹⁴C]-nicotinate uptake activity was measured after incubating oocytes for 1 hours at approximately 25°C in K⁺-free isotonic medium. All data are mean±SEM with n=12–15 oocytes per group.

Discussion

The GH/IGF-1 axis, together with insulin and IGFBPs, acts in a coordinated manner to regulate a myriad of physiological activities. Although the IGF-1/IGF-1R axis affects circulating SU levels,^6,18–20,66 the underlying mechanisms have not been characterized. Negative and nonlinear correlations between IGF-1 and UA levels have however been reported in nondiabetic and T2DM adults.^66,67 IGF-1 also inhibits secretion of insulin,⁶ and insulin reduces renal fractional excretion of urate^68–70 by stimulating urate transport through GLUT9a and other reabsorptive urate transporters.⁶⁵ We hypothesized that IGF-1 also exerts direct effects on urate transport mechanisms.

The major GWAS identified and/or genetically relevant human urate transporters include GLUT9, URAT1, ABCG2, ABCC4, OAT10, NPT1, and NPT4.²⁹ We detect coexpression of multiple urate transporters studied herein (GLUT9, ABCC4, ABCG2, URAT1, and OAT10) with the IGF1R protein in human proximal tubular cells (Supplemental Figure 5). Exploiting the functional endogenous IGF-1R in X. laevis oocytes,^58–62 we found significant increases in urate transport activity of GLUT9a, GLUT9b, OAT1, mOAT3, ABCG2, and ABCC4 after treatment with IGF-1 (Figure 4, C–H), with no effects of IGF-1 on the urate transport activity of URAT1, OAT10, OAT4, or NPT1 (Figure 4C). In addition, we found significant inhibitory effects of IGF-1 on insulin stimulation of urate uptake by GLUT9a and OAT3 (Figure 6, B and D). The signaling mechanism(s) whereby IGF-1 inhibits insulin activation of urate transport are as yet unclear and clearly worthy of further investigation, given the roles of these hormones in metabolic syndrome. To the extent that we detected inhibitory effects of IGF-1 in cell culture systems, we hypothesize that IGF-1 antagonizes specific signaling pathways that we have previously implicated in insulin activation of urate transport.⁶⁵ Notably, however, despite considerable homology, the insulin receptor and IGF-1R affect on overlapping yet distinct sets of signaling pathways, with considerable opportunity for antagonistic interactions.⁷¹

In HEK 293T and PTC-05 cells, we observed modest but statistically significant stimulation of the net [¹⁴C]-urate uptake by IGF-1. In addition, as in oocytes, IGF-1 inhibited the stimulatory effect of insulin on urate uptake in PTC-05 cells⁶⁵ (Figure 6A). In transiently transfected HEK 293T cells overexpressing IGF-1R, the augmented IGF-1 stimulation of urate uptake confirms regulation of IGF-1/IGF-1R signaling on urate transport. Almost complete inhibition of IGF-1 stimulation of [¹⁴C]-urate uptake in transfected HEK 293T cells overexpressing IGF-1R by specific inhibitors of protein tyrosine kinase (PTK) or PI3K (Figure 3D), with modest inhibition by specific inhibitors of MEK/ERK or p38 MAPK, suggests that IGF-1R transmits signals to urate transporters partially through PI3K-PDK-1-Akt, partly through Grb2-SOS-Ras-MAPK, and partially through p38 MAPK pathways (Figure 8).

Figure 8.

Regulation of urate transport by IGF-1 signaling: IGF-1 binding to its receptor (IGF-1R) activates IGF-1/insulin receptor substrate (IRS), Akt, ERK1/2, p38 MAPK, and PKC. The activation of urate transport (reabsorption or secretion) by IGF-1/IGF-1R signaling is indicated. Genistein, a phosphotyrosine kinase (PTK)-specific inhibitor; LY 295002, a PI3K-specific inhibitor; PD98059, a MEK/ERK-specific inhibitor; SB 203580, a p38 MAPK-specific inhibitor; BMI, bisindolylmaleimide 1-hydrochloride, a protein kinase C (PKC)-specific inhibitor.

Inhibitors of protein tyrosine kinase (PTK) or PI3K almost completely abolished the IGF-1 stimulation of [¹⁴C]-urate uptake mediated by GLUT9a, OAT1, OAT3, and urate efflux mediated by ABCG2 and ABCC4, suggesting that protein tyrosine phosphorylation of IGF-1R regulates the activities of multiple urate transporters. Inhibition of ERK almost completely abrogated the IGF-1 activation of GLUT9b, OAT1, OAT3, ABCG2, and ABCC4, with lesser effect on GLUT9a (Figure 5, C–F), indicating differential regulation of GLUT9 isoforms by IGF-1 signaling. Inhibition of p38 MAPK had no effect on IGF-1 activation of GLUT9a, GLUT9b, OAT1, and ABCG2; however, it inhibited the IGF-1–stimulated activity of mOAT3 and ABCC4 (Figure 5, C–F). Protein kinase-C (PKC) inhibition almost completely abolished IGF-1 activation of GLUT9a and GLUT9b without affecting OAT1, OAT3, ABCG2, and ABCC4 (Figure 5, C–F) suggesting divergent regulation of urate reabsorption transporters and urate secretion transporters.

Previously, we have shown that the SU-influencing association¹⁹ at IGF1R is colocalized with IGF1R expression.⁴⁹ Here, we have validated the colocalization of the urate association at IGF1R and IGF1R expression using the summary statistics from the larger Tin et al.⁵⁰ SU GWAS (Figure 1A) and the Gene and Tissue Expression v8 database (Figure 1B). The urate raising T-allele at the lead SNP (rs12908437) decreases IGF1R expression in heart left ventricle. A limitation of this analysis, however, is that insufficient kidney samples were available in GTEx to allow testing for colocalization. Similar findings have been reported for the lead rs35767 SNP at the IGF1 locus; the urate-lowering allele is associated with increased circulating levels of IGF-1 protein.^18,72,73 The genetic data thus implicate increased activity of the IGF-1–associated signaling with lower SU. The genetic interaction data are consistent with a hypothesis whereby altered expression of IGF1R increases activity of GLUT9 and ABCG2 to influence urate levels. The SLC2A9-IGF1R genetic data are consistent with a hypothesis whereby altered expression of IGF1R influences activity of GLUT9 to influence urate levels in men. In the presence of the GG urate-lowering SLC2A9 genotype, the effect of the urate-raising IGF1R T-allele is negated, a phenomenon not observed in women. The T-allele associates with reduced expression of IGF1R, which we speculate in men to result in reduced GLUT9-mediated reuptake of urate from filtered urine when coinherited with the SLC2A9 GG genotype. We speculate that the main effect of the association of IGF1R with urate levels reflects the cumulative effect on activity of GLUT9. Of note, a complicating issue is that the GLUT9b isoform is apically expressed in the distal nephron,⁷⁴ where it may potentially play a secretory role, versus the reabsorptive role of GLUT9a in the proximal tubule. Ultimately, the physiological effect of variation in GLUT9 and ABCG2 activity on the urate-lowering response to IGF-1 will require infusion experiments on humans genotyped for the major urate-influencing variants in these genes, as using for studies on the effect of these genes on urate handling after inosine loading.⁷⁵

Compared with insulin's robust effect,⁶⁵ we found very modest activation of urate uptake by IGF-1 in PTC-05 and transfected HEK-293T cells or on GLUT9 in oocytes, with greater relative activation of the urate secretory transporters OAT1, OAT3, ABCG2, and ABCC4. Notably, net urinary urate excretion reflects the balance between the separate, competing reabsorptive and secretory pathways; with a typical fractional excretion of urate of approximately 10%, reabsorption predominates. However, with no effect of IGF-1 on apical reabsorptive transporters, lesser activation of the basolateral reabsorptive transporter GLUT9a (versus insulin) and activation of both basolateral and apical secretory pathways, we propose that IGF-1 affects the relative balance between reabsorption and secretion, favoring an increase in urate excretion and a reduction in SU. The inhibitory effect of IGF-1 on insulin stimulation of urate uptake in PTC-05 cells, GLUT9a-expressing and OAT3-expressing oocytes, and the absence of an additive stimulatory effect of IGF-1 on insulin's stimulatory effect on urate transport on other transporters (Figure 6) indicate the yin-yang of insulin and IGF-1 signaling and their complex effect on urate transport. We conclude that the evident preferential activation of secretory urate transport mechanisms, combined with inhibition of insulin stimulatory effects (Figure 6) and insulin secretion,⁶ collectively begin to explain IGF-1's role in reducing SU. In addition, given the prominent role of ABCG2 and GLUT9 in the intestinal secretion of urate,^47,75–77 we hypothesize that the activation of these transporters in the intestine by IGF-1 promotes intestinal excretion of urate.

In summary, based on genetics, coexpression and colocalization of IGF1R and urate transporters in human renal proximal tubule (Supplemental Figure 5), cellular transport physiology, prior infusion data,⁶ overexpression of IGF1R and urate transporters, and pharmacologic manipulation of signaling pathways, we propose that the urate-reducing effects of IGF-1 are explained by a predominant activation of multiple secretory transporters and inhibition of insulin effects on reabsorptive urate transport. As in other aspects of the metabolic syndrome, the effects of IGF-1 on SU are evidently antagonistic to those of insulin⁶⁵ (Figure 6). This highlights that the hyperuricemia because of metabolic syndrome has a complex pathophysiology, with several potential mediators and thus several potential avenues for therapy in gout and related disorders.

Supplementary Material

jasn-34-451-s001.pdf ^{(1.3MB, pdf)}

Disclosures

H.K. Choi reports Consultancy: Allena, Ani, Horizon, LG, Ono, and Protalix; Research Funding: Horizon; Honoraria: Allena, Ani, Horizon, LG, Ono, and Protalix; and Advisory or Leadership Role: Allena, Ani, Horizon, LG, Ono, and Protalix. T. Merriman reports Research Funding: VariantBio. D. Mount reports Consultancy: Allena Pharmaceuticals, Alnylam Pharmaceuticals, ANI Pharmaceuticals, and Horizon Pharma; and Honoraria: McGraw Hill and UpToDate. All remaining authors have nothing to disclose.

Funding

This work is supported by National Institutes of Health grant NIAMS 5P50AR060772-09.

Author Contributions

D.B. Mount, T.R. Merriman, and H.K. Choi conceptualized the study and were responsible for project administration; A.K. Mandal and M.P. Leask were responsible for data curation; A.K. Mandal, M.P. Leask, D.B. Mount and N.A. Sumpter were responsible for formal analysis; A.K. Mandal, M.P. Leask, T.R. Merriman, and D.B. Mount were responsible for investigation; A.K. Mandal and T.R. Merriman were responsible for methodology; D.B. Mount and T.R. Merriman were responsible for funding acquisition; D.B. Mount provided supervision; A.K. Mandal, M.P. Leask, D.B. Mount, and T.R. Merriman wrote the original draft; and A.K. Mandal, D.B. Mount, and H.K. Choi reviewed and edited the manuscript.

Data Sharing Statement

The original contributions presented in the study are included in the article/Supplemental Material; further inquiries can be directed to the corresponding author.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/D604.

Primer sets for RT-PCR

Supplemental Figure 1. Quantification of the protein levels of IGF-1R, GLUT9, OAT10, ABCG2, ABCC4 and NPT1, expressed in equal number of control and transfected HEK 293T cells or IGF-1-treated PTC-05 cells, in K⁺-free isotonic medium, by densitometric measurements of specific protein bands in the Western blots (∼30μg of total protein/sample/lane).

Supplemental Figure 2. Quantification of the protein levels of GLUT9, OAT10, ABCG2, ABCC4 and NPT1, expressed in equal number of control and IGF-1-treated HEK 293T cells, in K⁺-free isotonic medium, by densitometric measurements of specific protein bands in the Western blots (∼30μg of total protein/sample/lane).

Supplemental Figure 3. Quantification of the protein levels of phospho-IRS (p-IRS), p-Akt, p-ERK and p-p38 MAPK, in equal number of control and IGF-1-treated PTC-05 or HEK 293T cells, in K⁺-free isotonic medium, by densitometric measurements of specific protein bands in the Western blots (∼30μg of total protein/sample/lane).

Supplemental Figure 4. Basal urate transport activity of urate transporters requires some level of activation of Akt and ERK signaling pathways.

Supplemental Figure 5. Immunohistochemistry of human kidney section with anti-IGF1R, anti-ABCG2, anti-ABCC4, anti-GLUT9, anti-URAT1, or anti-OAT10 antibody at 20X magnification.