Ketohexokinase-Dependent Metabolism of Fructose Induces Proinflammatory Mediators in Proximal Tubular Cells

KHK-Mediated Monocyte Chemotactic Protein 1 Production Induced by Fructose

Our group reported previously that a high-fructose diet increases the renal expression of monocyte chemotactic protein 1 (MCP-1) in the remnant kidney model of rats and confirmed this effect in vitro in HK-2 cells.²⁸ MCP-1 is a potent chemoattractant of monocytes/macrophages and plays a key role in the proinflammatory response,²⁹ and its action is predominantly confined to the renal tubule, which is also a major site of production.³⁰ MCP-1 has been shown to play a crucial role in the inflammatory-mediated renal damage.³¹ To explore this effect further in renal epithelial cells and to determine whether it occurs with physiologic concentrations of fructose, we stimulated HK-2 cells with various concentrations of fructose for 72 h and found that 1 mM fructose (equal to the physiologic postprandial concentration in the peripheral blood²¹) induced more than a two-fold increase in MCP-1 production (Figure 1A). Further increasing the fructose concentration to 5 and 10 mM plateaued with an additional 30% increment, statistically not different from the MCP-1 level at 1 mM. Interestingly, glucose did not stimulate MCP-1 (Figure 1B) and even attenuated the effect of fructose at the concentration 25 mM (Figure 1C). Stimulation of mannitol also did not induce an elevation of MCP-1 level excluding an osmotic mechanism (Figure 1D). When we compared the effects of fructose and tagatose, which is also metabolized by KHK,³² we found that both sugars, at 5 mM, induced a significant increase in MCP-1 production (Figure 1E), suggesting that MCP-1 release depends largely on the first step of fructose metabolism because tagatose, after phosphorylation by KHK, is metabolized much slower than fructose-1-phosphate.³²

• Download figure
• Open in new tab
• Download powerpoint

Figure 1.

Fructose-induced MCP-1 production in HK-2 cells. (A) Dosage-dependent effect of fructose on MCP-1 production. Glucose content in the medium in this and other experiments unless indicated otherwise was 5 mM. (B) Difference in the effects of fructose and glucose on MCP-1 production. (C) Glucose attenuates effect of fructose on MCP-1 production. (D) d-Mannitol did not produce a significant effect on MCP-1 production, excluding the osmotic mechanism. (E) Similarity of the effects of fructose and tagatose on MCP-1 production. In all experiments, cells were treated with fructose, glucose, tagatose, and mannitol for 72 h. MCP-1 content in the medium was assayed by ELISA and is expressed as pg/mg of cellular protein content. Data are means ± SEM *P < 0.05 versus untreated cells or corresponding control, nonparametric U test and, in A, one-way ANOVA; ^#P = 0.06 versus 5 mM fructose + 5 mM glucose; n = 3, each experiment performed in triplicate.

Because KHK expression in the kidney and, specifically, in the proximal tubule is very high, we next sought whether metabolism by KHK is required for the proinflammatory effect of fructose to induce MCP-1 production. First, we detected expression of KHK in HK-2 cells and found that mRNA for KHK in these cells is expressed at a level comparable to the liver (Figure 2A). Next, we transfected HK-2 with four different small hairpin RNA (shRNA) constructs for human KHK containing a gene coding resistance to puromycin and used this latter property to select stably transfected clones in puromycin-containing medium. Among the resistant clones, two showed an almost absent expression of the mRNA for KHK (Figure 2B). Western blot analysis confirmed that >75% of the KHK expression was suppressed at the protein level (Figure 2C). Stable transfection of HK-2 cells with nonrelevant shRNA did not affect KHK expression showing the absence of nonspecific effects on transfection and expression of small interfering RNA duplexes in the cells (Figure 2, B and C). We also confirmed a reduction in the functional activity of KHK in the silenced clones as reflected by the absence of ATP depletion in response to fructose (Figure 2D) because ATP depletion by acute fructose loading is strictly KHK dependent.^22,23,33 That an increase in ATP was seen after fructose loading in these clones (Figure 2D) is because, in the absence of KHK, fructose enters the glycolytic cascade via hexokinase, resulting in an increase in net ATP production. Next, we incubated nontransfected HK-2 cells, KHK-silenced HK-2 clone, and cells stably expressing a nonrelevant shRNA with fructose. To obtain a robust response, we used a concentration of fructose 5 mM. In nontransfected cells and pRS-GFP shRNA clone, fructose induced MCP-1 production, whereas stable silencing of KHK abolished the effect of fructose even at 5 mM, indicating a specific requirement for KHK (Figure 2E).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2.

KHK is required for fructose-induced MCP-1 production in HK-2 cells. (A) Expression of mRNA for KHK and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) in HK-2 cells and human liver. Inverted images of the RT-PCR products on agarose gel are shown. (B) mRNA expression for KHK and GAPDH genes in nontransfected HK-2 cells and puromycin-resistant clones transfected with pRS-GFP and pRS-KHK. (C) Analysis of KHK silencing at the protein level. Representative blot and densitometric analysis (KHK/GAPDH ratio) are shown. (D) Functional test of KHK silencing: Effect of fructose on KHK-dependent ATP depletion. (E) Fructose-induced MCP-1 is abolished by KHK silencing. Cells were treated with fructose for 72h. MCP-1 content in the medium was assayed by ELISA and is expressed as pg/mg of cellular protein content or percentage of control (100%). *P < 0.05 versus untreated cells or corresponding control, nonparametric U test; n = 3, each experiment performed in triplicate.

Involvement of Reactive Oxygen Species in the KHK-Mediated MCP-1 Production Induced by Fructose

MCP-1 production can be blocked by antioxidants in proximal tubular cells and other cells.^34,35 We therefore examined fructose-induced MCP-1 production in the presence of antioxidants N-acetylcysteine, cell-permeable superoxide scavenger MnTMPyP, and inhibitors of NADPH oxidase (apocynin) and xanthine oxidoreductase (allopurinol). All tested antioxidants completely abrogated the effect of fructose, suggesting that fructose-induced MCP-1 stimulation in HK-2 cells is redox dependent and mediated by superoxide production (Figure 3A). Apocynin suppressed both basal and stimulated reactive oxygen species (ROS) production (Figure 3A).

• Download figure
• Open in new tab
• Download powerpoint

Figure 3.

Fructose-induced KHK-dependent MCP-1 production in HK-2 cells is mediated by ROS (A) Fructose-dependent MCP-1 increase is prevented by antioxidants [10 mM N-acetyl-cysteine (NAC), 25 μM Mn(II) tetrakis(1-methyl-4 to 6 pyridyl) porphyrin (MnTMPyP), 0.1 mM allopurinol, and 200 μM apocynin]. Cells were treated for 72 h. (B) ROS production by HK-2 cells in the presence of 5 mM fructose. ROS-induced fluorescence is shown on the left, and differential interference contrast (DIC) is shown on the right. Cells were treated with fructose for 2 h. (C) Dosage response of ROS production in HK-2 cells incubated with varying concentrations of fructose or d-mannitol for 2 h. (D) Time course of ROS production in HK-2 cells incubated in 5 mM fructose. The antioxidant tempol (5 mM) completely prevented an increase in the fluorescence in response to fructose, demonstrating specificity of ROS detection (data not shown). (E) ROS accumulation after 2 h of stimulation with 5 mM fructose in nontransfected HK-2 and HK-2 clones stably expressing KHK shRNA or GFP-shRNA as a control for irrelevant shRNA and transfection. (F) Effect of oxypurinol (0.1 mM) on fructose-induced ROS production. Data are expressed as arbitrary units of fluorescence and are means ± SEM (n = 3 performed in triplicate, each representing 20 to 30 cells). *P < 0.05 versus untreated cells or corresponding control; ^&P < 0.05 versus fructose-treated cells, nonparametric U test.

These results prompted us to examine the effect of fructose on ROS production in HK-2 using live cell imaging with the ROS-specific fluorescence probe 5 (and 6)-chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate-acetyl ester (CM-H₂DCFDA). As shown in Figure 3B, fructose induced an increase in the probe fluorescence, which was ROS specific because the antioxidant tempol prevented this increase (data not shown). At concentrations of 0.5 to 5.0 mM, fructose-induced ROS production was dose dependent. An increase in oxidants was observed with concentrations as low as 1 mM (physiologic) and reached a plateau between 5 and 10 mM (Figure 3C). In the presence of d-mannitol, ROS production was not affected (Figure 3C), excluding an osmotic mechanism in the effect of fructose. Time-course experiments showed that the level of ROS increased as early as 15 min and remained at the same level for approximately 4 h with return to basal levels after 16 h, when, most likely, all of the added fructose was metabolized (Figure 3D).

To test whether fructose-induced ROS production was KHK dependent, we incubated nontransfected cells, KHK-silenced HK-2 cells, and cells transfected with nonrelevant shRNA in the presence or absence of a high level of fructose (5 mM) for 2 h. Fructose produced a significant increase in ROS production in normal HK-2 and in pRS-GFP(−) clone cells, whereas no effect was observed in cells with silenced KHK expression (Figure 3E). Finally, to test whether fructose-induced ROS production is dependent on xanthine oxidoreductase (XOR) and urate production as suggested by results in Figure 3A, we tested the effect of fructose in the presence of oxypurinol (0.1 mM), a XOR inhibitor. Oxypurinol did not modify the level of ROS in untreated cells but partially reduced ROS accumulation in cells treated with 5 mM fructose, suggesting involvement of XOR in this pathway (Figure 3F).

XOR Expression and Uric Acid Production in Response to Fructose in Tubular Epithelial Cells

Fructose metabolism via KHK results in the stimulation of AMP-deaminase followed by a XOR-mediated increase in uric acid formation.^36,37 This process has been usually studied in the liver; however, KHK is highly expressed in HK-2 cells and fructose-induced MCP-1 and ROS production might be suppressed by XOR inhibition. We therefore examined the expression of XOR in HK-2 cells and measured uric acid, the XOR reaction product. Reverse transcriptase–PCR (RT-PCR) analysis confirmed modest expression of XOR (Figure 4A). To identify uric acid, we used mass spectrometry. Uric acid was identified in the extract of HK-2 cells as fragments at m/z 124 and 96 by mass spectrometry (MS)/MS analysis as described in the Concise Methods section (Figure 4B), confirming that XOR is active. Next we tested whether fructose increases uric acid production in these conditions and whether uric acid is responsible for the increase in MCP-1 expression. Uric acid was measured in fructose-treated cells by both enzymatic assay with uricase and by MS. Figure 4, C and D, shows that fructose treatment induced an increase in intracellular level of uric acid in HK-2 cells that was completely prevented by allopurinol (Figure 4D). To determine whether fructose can induce local accumulation of uric acid in the kidney, we measured its content in the crude kidney homogenates from rats pair fed with the high-fructose diet or standard rat diet for 16 wk; however, in this case, we observed only a tendency to an increase in the uric acid content from 0.122 ± 0.025 to 0.145 ± 0.031 nmol/mg protein (P < 0.1; n = 6).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4.

Fructose-induced intracellular uric acid production and XOR expression in HK-2 cells. (A) Expression of mRNA for XOR and β-actin in HK-2 cells. Inverted images of the RT-PCR products on agarose gel are shown. (B) Identification of uric acid in samples by MS/MS experiments. (C) Intracellular uric acid content in response to 1 and 5 mM fructose stimulation for 72 h. Data are means ± SEM and are expressed as nmol/mg of protein. *P < 0.05 versus untreated cells. (D) Analysis of the effect of fructose with and without allopurinol on the intracellular level of uric acid by tandem LC/MS/MS. Extracts contained the same amount of total protein. The x axis is the retention time, and the y axis is the intensity of the single-reaction monitoring peaks. The peak area is proportional to the quantity of uric acid.

Uric Acid Mimics the Effect of Fructose on MCP-1 Production

To test whether intracellular urate production induced by fructose affects MCP-1 production, we incubated cells with fructose with or without allopurinol (0.1 mM). As shown in Figure 5, allopurinol blocked fructose-induced MCP-1 production, suggesting that the fructose effect might be mediated by XOR-induced oxidants and/or increased uric acid production. Cell stimulation with 15 mg/dl uric acid mimicked the effect of fructose. Combining 5 mM fructose and 15 mg/dl uric acid, however, did not produce an additional effect on the MCP-1 levels, showing that effects of fructose and uric acid are not additive. The presence of 0.1 mM allopurinol in the latter combination partially reduced MCP-1 production, suggesting a partial involvement of XOR in this pathway (Figure 5).

• Download figure
• Open in new tab
• Download powerpoint

Figure 5.

Comparison of the effects of fructose and uric acid on MCP-1 production in HK-2 cells. Cells were treated with fructose and/or uric acid with or without allopurinol for 72 h. MCP-1 content in the medium was assayed by ELISA and is expressed as pg/mg of cellular protein. Data are means ± SEM; n = 3, each experiment is performed in triplicate. Statistical analysis was performed using nonparametric U test.

Cell Culture

HK-2, an immortalized cell line from normal adult human kidney that maintains phenotype of proximal tubular cells⁵⁰ and represents the best in vitro model of human proximal tubule cells, were obtained from American Type Culture Collection (Rockville, MD). Cells were grown to confluence in DMEM/F-12 (50/50) with l-glutamine and HEPES buffer (Mediatech, Herndon, VA) supplemented with 10% FBS; 100 U/ml penicillin; 100 g/ml streptomycin; and 2 mM l-glutamine and insulin (10 mg/L), transferrin (5.5 mg/L), and sodium selenite (6.7 μg/L). Cells were cultured at 37°C in 95% air and 5% carbon dioxide (CO₂) to 70 to 75% of confluence.

Animal Protocol

All animal protocols were approved by the University of Florida Institutional Animal Care and Use Committee. Twenty-four 150-g male Sprague-Dawley rats (Charles Rivers Laboratories, Wilmington, MA) were divided into four groups of six rats each and received a diet of either 60% fructose (F60) or standard rat food (control), which contains complex carbohydrates, instead of purified sugars (Harlan-Teklad, Madison, WI). The animals were pair fed to ensure equal food intake. Rats were maintained in temperature- and humidity-controlled specific pathogen–free conditions on a 12:12-h light-dark cycle. After 16 wk, the rats were anesthetized with isoflurane and killed. Samples of the kidney tissues were stored at −80°C until analysis.

Stable Transfection of HK-2 Cells and KHK Silencing with shRNA

For the silencing of KHK in HK-2 cells, we used HuSH shRNA KHK-specific constructs in the pRS backbone under control of U6 promoter (Origene, Rockville, MD). To achieve long-term silencing, we used stable transfection. Cells were transfected with plasmids for four different KHK shRNA and shRNA for nonrelevant gene (GFP; pRS-GFP) using Lipofectamine 2000 (Invitrogen, Carlsbad, CA) following the protocol of the manufacturer followed by the selection of the transfectants resistant to puromycin in the selective medium containing 1 μg/ml puromycin for 8 to 10 wk. The best KHK-silenced clones were selected and validated on the basis of the lowest KHK mRNA expression and the absence of the functional KHK activity (see Figure 2). KHK-silenced HK-2 clone selected for experiment (pRS-KHK) and the clone stably expressing nonrelevant shRNA were maintained in the medium with 0.5 μg/ml puromycin, which was removed before experiments.

RNA Isolation and RT-PCR

Total RNA extraction was performed with TRIZOL (Invitrogen). The RNA purity and concentration were evaluated by spectrophotometric measurements. KHK was detected in normal HK-2 and in clones using SuperScript One-Step RT-PCR with Platinum Taq System (Invitrogen) according to the protocol of the manufacturer with the following primers: KHK (Genebank NM_006488; 89 bp, fully compatible with KHK-A and KHK-C isoforms), forward primer 5′-TCATGGAAGAGAAGCAGATCC-3′ and reverse primer 5′-CTCCGAGTCCTCCTTAGGGTA-3′; glyceraldehyde-3-phosphate dehydrogenase (Genbank NM_002046; 355 bp), forward primer 5′-TCACCAGGGCTGCTTTTAAC-3′ and reverse primer 5′-TTCACACCCATGACGAACAT-3′; and XOR (Genbank NM_000379), forward primer 5′-CCTCCGCACAGATATTGTCAT-3′ and reverse primer 5′-GGGATCTTGTAGGTGCTAGGG-3′. The following cycling conditions were used for cDNA synthesis and PCR amplification: (1) Reverse transcription at 50°C for 30 min, (2) predenaturation at 94°C for 2 min, (3) denaturation at 94°C for 30 s, (4) annealing 57°C for 1 min, and (5) extension of primers at 72°C for 1 min. Steps 3 through 5 were repeated for 30 cycles. For DNA final extension, the products were held at 72°C for 5 min. The DNA fragments were separated by electrophoresis on 1.5% agarose gel with ethidium bromide (1 μg/ml) and then visualized by ultraviolet light. Total RNA form human liver used in this study was obtained from Ambion (Austin, TX).

Measurement of Intracellular ATP Content

Cellular ATP was measured by using an ATP bioluminescence assay (Roche assay kit CLS II; Roche, Mannheim, Germany), following the manufacturer's instructions. After the experimental treatments, cells were washed with ice-cold PBS and detached from the culture plates with 0.05% trypsin/0.02% EDTA, and then we added to the cell suspension boiling 200 mM Tris/8 mM EDTA (pH 7.75) solution and incubated for 2 min at 98°C. Cell debris was precipitated at 10,000 × g for 2 min. Then ATP was determined in the supernatant by adding luciferin/luciferase by automated injection in a luminometer (Optocomp II; MGM Instruments, Hamden, CT) with integration of the light signal for 2 s. The ATP concentration was calculated by interpolation from a standard curve. Cellular protein content was determined by using the BCA protein assay. The ATP content was normalized to protein concentration and was expressed as nanomoles per milligram of protein.

Uric Acid Assay

To assess intracellular uric acid content, we rinsed cells with ice-cold PBS and harvested them in buffer containing 25 mM HEPES (pH 7.1), 100 mM KCl, 1 mM dithiothreitol, and 0.1 mM EDTA. The cell extracts were homogenized by 100 strokes of the pestle and frozen/thawed in liquid nitrogen three to four times. The intracellular level of uric acid was measured using Amplex Red UA/Uricase Assay Kit from Invitrogen-Molecular Probes (Eugene, OR). The fluorescence was measured using excitation in the range of 530 to 560 nm and emission detection at approximately 590 nm. Uric acid concentration in the samples was determined from the standard curve. The uric acid levels were normalized for protein concentration and expressed as nanomoles per milligram of protein.

Detection of Uric Acid by MS

All samples were stored at −80°C before analysis. Before the analysis, they were thawed in a cold water bath and filtrated through a 0.2-μm Micro Centrifuge Nylon filter (Millipore, Billerica, MA) by centrifugation for 10 min at 14,000 rpm. The filtrates were sealed in amber glass vials using Teflon-lined caps for liquid chromatography (LC)-MS analysis. The calibrators were 0.050, 0.125, 0.250, 0.500, 1.000, and 5.000 μM uric acid in aqueous solution.

The LC-MS analyses were carried out with a ThermoFinnigan Surveyor liquid chromatography system (ThermoFinnigan, San Jose, CA) and a TSQ Quantum triple quadruple mass spectrometer (ThermoFinnigan) equipped with ESI interface operated in negative-ion mode. LC analyses were performed in a gradient elution mode using Phenomenex Luna 5μ C18 (2) 100A (150 mm × 4.6 mm) column (Phenomenex, Torrance, CA) coupled with a Phenomenex Luna C18,(2)5-μm particle size guard column. The mobile phase used included 5 mM ammonium acetate/0.1% acetic acid (A) and methanol (B). The mobile phase flow was 600 μl/min, and the sample injection volume was 20 μl. The gradient began at 95% A. The composition was linearly ramped to 25% B over the next 4.5 min, and it was maintained constant for 1.5 min. Then, it was reversed to the original composition of 95% A over 0.5 min, after which time it was kept constant for 0.5 min to reequilibrate the column. In the TSQ Quantum, nitrogen was used for both the sheath (60 psi) and auxiliary (20 units) gases. The heated capillary temperature was maintained at 350°C. The collision pressure was 1.5 U, and collision energy was 25 V. The single-reaction monitoring method was applied for the analysis. The ions selected for single-reaction monitoring were m/z 124.1 and 96.1. The operation of the LC-MS and data analysis were performed using ThermoFinnigan Xcalibur 1.4 software. The calibration was linear with the R² value of 0.998.

ROS Detection

We assessed intracellular ROS with the fluorescence probe CM-H₂DCFDA (Molecular Probes) as described previously in detail.⁴⁹ Briefly, at the end of treatments, cells were washed with HBSS followed by incubation in the presence of CM-H2DCFDA (5 μM) for 30 min in HBSS. Then cells were transferred to HBSS without probe, and green fluorescence was measured using an inverted microscope Axiovert 200 (Carl Zeiss Microimaging, Thornwood, NY) equipped with CCD camera and image acquisition/analysis software AxioVision 4.5 (Carl Zeiss Microimaging, Thornwood, NY).

ELISA for MCP-1

At the end of the treatments, media supernatants were collected, quantified, and centrifuged at 1000× g for 10 min to remove dead and nonadherent cells, and the remnant cell monolayer was lysed as described for protein determination. MCP-1 content in the supernatant was assayed by human MCP-1/CCL2 immunoassay Quantikine kit (R&D Systems, Minneapolis, MN) according to the manufacturer's protocol. Experiments were performed in triplicate and confirmed on three independent occasions. The MCP-1 content was normalized to protein content and was indicated as picograms per milligram of protein.

Statistical Analysis

At least three independent experiments were performed in triplicate each. Data were analyzed by one-way ANOVA followed by Fisher least significant test, unpaired t test, or Mann-Whitney U test, with P < 0.05 considered significant. Comparison between two values was performed by t test or U test. ANOVA was used to test differences among several means.