PCK1 is a key regulator of metabolic and mitochondrial functions in renal tubular cells

Abstract

Phosphoenolpyruvate carboxykinase 1 (PCK1 or PEPCK-C) is a cytosolic enzyme converting oxaloacetate to phosphoenolpyruvate, with a potential role in gluconeogenesis, ammoniagenesis, and cataplerosis in the liver. Kidney proximal tubule cells display high expression of this enzyme, whose importance is currently not well defined. We generated PCK1 kidney-specific knockout and knockin mice under the tubular cell-specific PAX8 promoter. We studied the effect of PCK1 deletion and overexpression at the renal level on tubular physiology under normal conditions and during metabolic acidosis and proteinuric renal disease. PCK1 deletion led to hyperchloremic metabolic acidosis characterized by reduced but not abolished ammoniagenesis. PCK1 deletion also resulted in glycosuria, lactaturia, and altered systemic glucose and lactate metabolism at baseline and during metabolic acidosis. Metabolic acidosis resulted in kidney injury in PCK1-deficient animals with decreased creatinine clearance and albuminuria. PCK1 further regulated energy production by the proximal tubule, and PCK1 deletion decreased ATP generation. In proteinuric chronic kidney disease, mitigation of PCK1 downregulation led to better renal function preservation. PCK1 is essential for kidney tubular cell acid-base control, mitochondrial function, and glucose/lactate homeostasis. Loss of PCK1 increases tubular injury during acidosis. Mitigating kidney tubular PCK1 downregulation during proteinuric renal disease improves renal function.

NEW & NOTEWORTHY Phosphoenolpyruvate carboxykinase 1 (PCK1) is highly expressed in the proximal tubule. We show here that this enzyme is crucial for the maintenance of normal tubular physiology, lactate, and glucose homeostasis. PCK1 is a regulator of acid-base balance and ammoniagenesis. Preventing PCK1 downregulation during renal injury improves renal function, rendering it an important target during renal disease.

INTRODUCTION

Chronic kidney disease (CKD) affects 10% of the world population and is responsible for large morbidity, mortality, and health expenses (1, 2). Metabolic shifts occurring in proximal tubular (PT) cells during CKD are associated with a worse renal prognosis (3). We recently described major alterations of the gluconeogenic pathway during CKD in a stage-dependent manner, which were associated with renal prognosis and systemic complications. Phosphoenolpyruvate carboxykinase 1 (PCK1 or PEPCK-C) downregulation was associated with a worse outcome (4).

PCK1 is a cytosolic enzyme that converts oxaloacetate (OAA) to phosphoenolpyruvate (5). Its role in gluconeogenesis is well established, with hypoglycemia in systemic deletion (6). Isolated hepatic deletion of PCK1 results in normoglycemia, implying other sources of glucose than the liver (7). In addition, PCK1 has a central role in energy production. This enzyme, by converting oxaloacetate, allows the continuous function of the tricarboxylic acid cycle through cataplerosis (5, 8). Finally, PCK1 may be crucial in fatty acid recycling (9). Although the role of PCK1 in the liver has been well described (10), its role in the kidney is less understood.

PCK1 is highly expressed in the kidney PT, which participates in glucose homeostasis in several ways, including gluconeogenesis using lactate, glutamine, and glycerol as substrates. PT cells also reabsorb glucose from the urine through Na⁺-glucose transporters 1 and 2 (SGLT1 and SGLT2, respectively) and secrete it into the blood through glucose transporters (11). Gluconeogenesis in the kidney is regulated by classical and nonclassical factors, among which metabolic acidosis is important (12, 13). Lactate is the preferred substrate under normal conditions and is taken up via specific transporters (14–17). Gluconeogenesis in the kidney is tightly coupled to ammoniagenesis, and glutamine is the preferred substrate for glucose production during metabolic acidosis (18). By virtue of its involvement in both pathways, PCK1 provides a potential mechanistic link between gluconeogenesis and ammoniagenesis, although this has been debated (19, 20). This may also be important in kidney disease where metabolic acidosis enhances renal function loss (21–23).

Given the importance of PCK1 in multiple PT physiological processes, we hypothesized that this enzyme may link metabolic alterations to CKD progression and that its replacement may be an important therapeutic target.

METHODS

Animal Models

All animal experiments were approved by the Institutional Ethical Committee of Animal Care in Geneva and Cantonal authorities (Authorization No. GE6A). All animals were housed at 20°C and had access to food and water ad libitum.

To generate PCK1 kidney tubule-specific knockout (KO) mice (PCK1^KO/KO mice), we used PCK1^fl/fl (generously donated by Mirko Trajkovski, University of Geneva, Geneva, Switzerland) in a C57BL/6 background with targeted insertion of LoxP sites between exons 3–4 and exon 5–6. Floxed mice were then crossed with Pax8-Cre mice to obtain a knockout strain with deletion of PCK1 in whole renal tubules (named PCK1 KO or KO in the text) (24). Similarly, to generate PCK1 kidney tubule-specific knockin (KI) mice (PCK1^KI/KI mice), we used PCK1^fl/fl (from Cyagen Biosciences, Santa Clara, CA) in a C57BL/6 background containing the CAG promoter-loxP-PGK-Neo-6*SV40 pA-loxP-Kozak-Mouse Pck1 CDS-rBG pA cassette into the ROSA26 gene (named PCK1 KI or KI in the text). Floxed mice were then crossed with Pax8-Cre mice to obtain a KI strain with the insertion of PCK1 in whole renal tubules. Eight-week-old male mice were used for experiments, and littermate wild-type (WT) PAX8^Cre- mice were used as controls (named PCK1 WT or WT in the text).

To induce acidosis, mice were fed a diet supplemented with 0.2 M HCl for 48 h. To do this, 150 mL of HCl (0.333 mM) were mixed with 100 g of food and then dried and given to mice. To generate a proteinuric progressive tubule-interstitial mouse model (POD-ATTAC), PCK1 WT and PCK KI mice were bred with POD-ATTAC mice (C57BL/6 background). At 8 wk old, transgenic male mice and their littermates received one intraperitoneal injection of B/B Dimerizer (Takara) at 0.2 mL/kg per day for 5 consecutive days, and 14 days after the first injection, urine was collected using metabolic cages. Mice were then euthanized.

PCR Primers and Genotyping

Genomic DNA from mouse phalange biopsies was isolated and used to perform PCR with the following primers: PCK1 KO (forward: 5′- TCTGTCAGTTCAATACCAATCT-3′ and reverse: 5′- AATGTTCTCTGCAAGTCCTGGTG-3′), PCK1 KI (forward: 5′- CTTTATTAGCCAGAAGTCAGATGC-3′ and reverse: 5′- GGTCAACACCGACCTCCCTTACG-3′), PAX8-Cre (forward: 5′- GAATCTCACACCTTCCAA-3′ and reverse: 5′- CGCACTTGCTTTCTCCCA-3′), and POD-ATTAC (forward: 5′- GAAAGTGCCCAAACTTCA-3′ and reverse: 5′- AACTGAGATGTCAGCTCATAGATGGGGG-3′). DNA amplicon was loaded onto an agarose gel (2%) containing Sybr Safe and run for 60 min at 100 V.

Real-Time Quantitative PCR

Total RNA was extracted from tissues (renal cortex area or liver) with TRIzol reagent (Invitrogen) according to the manufacturer’s instructions. After RNA dosage using Nanodrop (Thermo Fisher Scientific), 1 µg of total RNA was reverse transcripted using qScript cDNA supermix (Quanta-bio). cDNA was used to perform quantitative PCR in triplicate using PowerUp SYBR Green Master Mix (Applied Biosystems) and the QuantStudio 5 Real-Time PCR System (Thermo Fisher Scientific). The 2^−ΔΔCT method (where C_T is threshold cycle) was used to analyze relative changes in gene expression levels. Expression was normalized by the mean expression of the ribosomal protein lateral stalk subunit P0 (RLP0) housekeeping gene of the control group. The primers used in quantitative PCR were as follows: RLP0 (forward: 5′-AATCTCCAGAGGCACCATTG-3′ and reverse: 5′-GTTCAGCATGTTCAGCAGTG -3′); phosphoenolpyruvate carboxykinase 1 (PCK1; forward: 5′-CCATCCCAACTCGAGATTCTG-3′ and reverse: 5′-CTGAGGGCTTCATAGACAAGG-3′); kidney injury molecule-1 (KIM-1; forward: 5′-ACTAAGGGCTTCTATGTTGGC-3′ and reverse: 5′-AGCTTCAATCTTAGAGACACGG-3′); sodium-glucose cotransporter-2 (SLC5a2; forward: 5′-TGGAGAGAATGGAGCAACAC-3′ and reverse: 5′-CACAAGCCAACACCAATGAC-3′); pyruvate kinase M (PKM; forward: 5′-GGTGTTTGCATCTTTCATCCG-3′ and reverse: 5′-CTGGCCTCCAAGATCTCATC-3′); glucose-6-phosphatase, catalytic (G6PC; forward: 5′- AAAAAGCCAACGTATGGATTCCG-3′ and reverse: 5′-CAGCAAGGTAGATCCGGGA-3′); peroxisome proliferator-activated receptor-γ coactivator-1α (PGC1α; forward: 5′- AGTCCCATACACAACCGCAG-3′ and reverse: 5′- CCCTTGGGGTCATTTGGTGA-3′); acyl-CoA oxidase 1 (ACOX1; forward: 5′- CTTGGATGGTAGTCCGGAGA-3′ and reverse: 5′- TGGCTTCGAGTGAGGAAGTT-3′); and acyl CoA oxidase 2 (ACOX2; forward: 5′- GGCCAGGTTTCTGATGAAGA-3′ and reverse: 5′- CTTGGTGTGGCGAGATACAT-3′).

Western Blot Analysis

Kidney tissue samples were homogenized in cold lysis buffer on ice and then centrifuged at 13,000 g for 10 min. The protein supernatant was assayed using a BCA protein assay (Thermo Scientific, Pierce; 25 µg of proteins were then loaded into precast 4−10% bis-acrylamide gels (Bio-Rad), and migration was performed at 100 V for 1 h. The transfer was performed on nitrocellulose membrane (Bio-Rad) using blot transfer (Bio-Rad). After being washed with Tris-buffered saline (TBS)-0.1% Tween and blocked at room temperature in 10% milk and TBS-Tween for 1 h, the membrane was incubated overnight at 4°C with rabbit anti-PCK1 (ab70358, Abcam) or β-actin primary antibody (4970S, Cell Signaling) diluted in 5% milk and TBS-Tween (1/1,000 or 1/10,000). After washes, the membrane was then incubated for 1 h with goat anti-rabbit horseradish peroxidase secondary antibody (1/5,000, no. 554021, BD Pharm). Protein expression was detected with an ECL detection reagent (WesternBright Quantum, Advansta), whose chemiluminescence was revealed with the Fusion imaging system (Vilber). Protein quantification was performed using Evolution Capt-EDGE software (Viber). Protein expression was normalized to β-actin expression. Results are expressed as fold changes in protein expression compared with control samples.

Histology and Immunofluorescence

Kidneys were fixed in 4% paraformaldehyde, paraffin embedded, and then cut into 5-µm sections with a microtome.

To perform histological analysis, a median section of the kidney was selected and stained with hematoxylin and eosin solution. Slides were scanned with an Axioscan image scanner (Zeiss) at ×20 magnification. To quantify the surface of glomeruli, we delimited 20 glomeruli per slide for each mouse using Qpath software.

To perform SGLT2, megalin, NaPi-IIa, and Na⁺-K⁺-ATPase α₁-subunit immunofluorescence, 5-µm-thick sections from paraffin-embedded kidneys were deparaffinized and submitted to antigen retrieval in citrate buffer (pH 6) for 12 min at 110°C. Sections stained with rabbit anti-NaPi-IIa (25) were treated with PBS and 1% SDS and blocked in PBS and 1% BSA. The primary antibody was incubated overnight at 4°C, and sections were washed with hypertonic PBS [18 g NaCl-PBS (26)] and incubated with a secondary antibody at room temperature. The other antibody stainings were performed overnight at 4°C after cell permeabilization in 0.1% Triton X-100 and nonspecific site blockade in PBS with 1% BSA and 10% serum. Rabbit anti-SGLT2 (No. 24654-1-AP, Proteintech), mouse anti-megalin (sc-515772, Santa Cruz Biotechnology), and mouse anti-Na⁺-K⁺-ATPase α₁-subunit (No. 05-369, Merck Millipore) were used and detected with the following secondary antibodies: Alexa Fluor 647-conjugated donkey anti-rabbit (No. 711-606-152, Jackson ImmunoResearch), Alexa Fluor 647-conjugated donkey anti-mouse (A-31571, Thermo Fisher Scientific), and Alexa Fluor 568-conjugated goat anti-mouse (A-11004, Thermo Fisher Scientific). Slides were mounted with Dako glycergel (C056330-2, Agilent) and imaged using the AxioScan Z1 slide scanner.

For PCK1 immunofluorescence, 5-µm-thick sections from paraffin-embedded kidneys were deparaffinized and submitted to antigen retrieval in citrate buffer (pH 6). Sections were incubated with rabbit anti-PCK1 (Abcam) overnight at 4°C after cell permeabilization in 0.1% Triton X-100 and nonspecific site blockade in PBS with 1% BSA and 10% serum. After a wash, secondary antibody cyanine 3 was incubated for 1 h at room temperature. Slides were mounted and imaged using the AxioScan Z1 slide scanner.

For SGLT2 signal quantification, the percentage of SGLT2-positive pixels in three cortical regions per animal was measured using ImageJ software. The results were statistically analyzed using a parametric test on n = 4 animals. The megalin Intensity mean value was measured in the whole cortex from each animal using ZEN software to estimate PT length. The results were statistically analyzed using a parametric t test on n = 4 animals.

RNA Sequencing of Murine Kidneys

RNA quantification was performed from PCK1^WT/WT, PCK1^KO/KO, PCK1^WT/WT, and PCK1^KI/KI mouse kidneys with a Qubit fluorimeter (Thermo Fisher Scientific), and RNA integrity was assessed with a bioanalyzer (Agilent Technologies). The TruSeq mRNA stranded kit from Illumina was used for library preparation with 700 ng of total RNA as input. Library molarity and quality were assessed with Qubit and TapeStation using a DNA high-sensitivity chip (Agilent Technologies). Libraries were pooled at 2 nM and loaded for clustering on a single-read Illumina flow cell for an average of 25 million reads per sample. Reads of 100 bases were generated using TruSeq SBS chemistry on an Illumina HiSeq 4000 sequencer. After quality control, reads were mapped to the GRCm38 genome using TopHat software. Tables of counts were thus generated with HTSeq. Downstream analyses were performed using edgeR. Raw gene counts were normalized by the trimmed mean of M values method using the edgeR package, and counts are expressed in counts per million. Comparisons of gene expression were performed using a gene-wise negative binomial linear model with the quasi-likelihood test. Values are expressed as the trimmed mean of M value-normalized counts per million.

Blood and Urine Analysis

Mice were housed in metabolic cages for 48 h with food and water ad libitum. Urine was collected after a 24-h period, and parameter analyses were performed by the Zurich Integrative Rodent Physiology facility using a Beckman Coulter AU480 analyzer (https://www.zirp.uzh.ch/en/labanalysis/bloodchemistry/analyses.html). For blood analysis, mice were anesthetized with an intraperitoneal injection of a mix of ketamine and xylazine (100 µg/g and 5 μg/g, respectively), and blood was collected in Li-heparin microtubes by intracardiac punction. Blood analysis was performed using the EPOC system (Siemens) or by the Zurich Integrative Rodent Physiology Facility.

Cell Oxygen Consumption Rate (Seahorse) and ATP Measurement

The oxygen consumption rate (OCR) was detected using the Seahorse XF Cell Mito Stress Test (Agilent) according to the manufacturer’s recommendations. About 80,000 cortical cells from PCK1^WT/WT, PCK1^KO/KO, PCK1^WT/WT, and PCK1^KI/KI mice were seeded into XF cell culture microplates previously coated with CellTak (Corning), with Seahorse XF DMEM (pH 7.4) supplemented with glucose (5 mM), pyruvate (1 mM), and glutamine (2 mM). To do this, immediately after euthanasia, we dissected the renal cortex of each mouse. Mechanical dissociation was performed using the gentleMACS dissociator (Miltenyi Biotec). After filtration at 20 µM and being rinsed with PBS, cells were counted on a Malassez slide using trypan blue. Oligomycin (2 μM), FCCP (2 μM), and rotenone/antimycin A (0.5 μM) were then added to the plate separately following the manufacturer’s instructions. It was detected by the Seahorse Bioscience XF96 Extracellular Flux Analyzer. OCR was normalized to cell number.

To measure the amount of ATP in the renal cortex of PCK1^WT/WT, PCK1^KO/KO, PCK1^WT/WT, and PCK1^KI/KI mice, we used the ATP Assay Kit (ab83355, Abcam) following the manufacturer's recommendations. The amount of ATP was normalized by the amount of protein.

Transcutaneous Glomerular Filtration Rate Measurement

To measure the glomerular filtration rate (GFR), renal clearance of FITC-sinistrin was performed in live animals. Mice were anesthetized with isoflurane (3% vaporization) and then shaved on the right flank. A minicamera (NIC-Kidney device, Mannheim Pharma & Diagnostics, Mannheim, Germany) was attached on the flank with adhesive tape, and FITC-sinistin (Fresenius-Kebi) was injected in the tail at a concentration of 0.35 g/kg after 2 min of basal recording. After 90 min of recording, the camera was removed under anesthesia and the recording was analyzed with MP&D laboratory software (Mannheim Pharma & Diagnostics). GFR was calculated according to the appropriate formula and expressed as milliliters per minute per 100 grams of body weight.

Cortical Glucose and Lactate Measurements

To determine glucose and lactate in the renal cortex, we lysed a piece of the cortex using the corresponding buffer recommended by the manufacturer. We then performed a protein assay using the Bradford method. For glucose, the assay was performed with 30 µg protein using the glucose assay kit (ab65333, Abcam) and detected by fluorometry (excitation/emission: 535/587 nm). For lactate, the assay was performed with 2 µg protein using the lactate assay kit (ab65331, Abcam) and detected by fluorometry (excitation/emission: 535/587 nm).

Statistical Analyses

All data were analyzed using GraphPad Prism software and are expressed as means ± SD All experiments were performed at least three times. An ANOVA test with secondary Tukey analysis was performed in more than two groups. A Student’s t test was performed in a two-group analysis. P < 0.05 was considered significant.

RESULTS

Mice With Tubular Deletion of PCK1 Display Major Alterations of Renal Function, Acid-Base, Glucose, and Lactate Homeostasis

To study the role of PCK1 in the kidney, we generated PCK1 kidney tubule-specific KO and KI mice under the tubule-specific PAX8 promoter (Supplemental Figs. S1A and S2B) (24). As expected, PCK1 was not expressed in the kidneys of KO mice and overexpressed in KI mice (Fig. 1, A–D, and Fig. 2, A–D). The expression of other gluconeogenic and glycolytic enzymes was not altered by modifications of PCK1 expression (Supplemental Fig. S1 and Fig. 2M).

Figure 1.

Characteristics of phosphoenolpyruvate carboxykinase 1 (PCK1) knockout (KO) mice at baseline. A: Pck1 mRNA expression in the kidney cortex and liver of wild-type (WT) and KO mice (kidney cortex: n = 7 WT mice and n = 6 KO mice; liver: n = 9 WT mice and n = 7 KO mice). B: PCK1 protein expression in the kidney cortex of WT and KO mice (n = 7 WT mice and n = 6 KO mice). C: representative immunoblots of protein expression in the kidney cortex of WT and KO mice. D: immunofluorescence staining of PCK1 in the kidneys of WT and KO mice (nuclei in blue and PCK1 in gold). Representative images of kidney sections with hematoxylin and eosin (H&E) staining of WT and KO mice are shown. E: measured glomerular filtration rate of WT and KO mice (n = 9 WT mice and n = 6 KO mice). F: urinary albumin normalized to urinary creatinine in WT and KO mice (n = 8 WT mice and n = 5 KO mice). G: blood pH in WT and KO mice (n = 7 WT mice and n = 6 KO mice). H: blood bicarbonate concentration in WT and KO mice (n = 7 WT mice and n = 6 KO mice). I: blood chloride concentration in WT and KO mice (n = 7 WT mice and n = 6 KO mice). J: urinary anion gap in WT and KO mice (n = 8 WT mice and n = 5 KO mice). K: glucose concentration in the kidney cortex of WT and KO mice (n = 5 WT mice and n = 4 KO mice). L: lactate concentration in the kidney cortex of WT and KO mice (n = 4 WT mice and n = 4 KO mice). M: immunofluorescence staining of Na⁺-glucose transporter 2 (SGLT2) in the kidneys of WT and KO mice. N: glucose concentration in urine of WT and KO mice (n = 8 WT mice and n = 5 KO mice). O: lactate concentration in urine of WT and KO mice (n = 6 WT mice and n = 5 KO mice). P: glucose concentration in serum of WT and KO mice under fasting conditions (n = 5 WT mice and n = 5 KO mice). Q: lactate concentration in serum of WT and KO mice under fasting conditions (n = 5 WT mice and n = 6 KO mice). R: histogram representing SGLT2 pixels in the kidney cortex of WT and KO mice. S: heatmap displaying mRNA levels of genes of interest in the kidney cortex of WT and KO mice. Results are presented with error bars showing means ± SD. Statistical analysis was performed using a t test; ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Figure 2.

Characteristics of phosphoenolpyruvate carboxykinase 1 (PCK1) knockin (KI) mice at baseline. A: Pck1 mRNA expression in the kidney cortex of wild-type (WT) and KI mice (n = 9 WT mice and n = 7 KI mice). B: PCK1 protein expression in the kidney cortex of WT and KI mice (n = 3 WT mice and n = 3 KI mice). C: representative immunoblots of protein expression in the kidney cortex of WT and KI mice. D: immunofluorescence staining of PCK1 in the kidneys of WT and KI mice (nuclei in blue and PCK1 in gold). Representative images of kidney sections with hematoxylin and eosin (H&E) staining of WT and KI mice are shown. E: measured glomerular filtration rate of WT and KI mice (n = 7 WT mice and n = 8 KI mice). F: blood pH in WT and KI mice (n = 10 WT mice and n = 7 KI mice). G: blood bicarbonate concentration in WT and KI mice (n = 10 WT mice and n = 7 KI mice). H: blood chloride concentration in WT and KI mice (n = 10 WT mice and n = 7 KI mice). I: glucose concentration in the kidney cortex of WT and KI mice (n = 8 WT mice and n = 8 KI mice). J: lactate concentration in the kidney cortex of WT and KI mice (n = 8 WT mice and n = 8 KI mice). K: glucose concentration in serum of WT and KI mice under fasting conditions (n = 5 WT mice and n = 5 KI mice). L: lactate concentration in serum of WT and KI mice under fasting conditions (n = 5 WT mice and n = 7 KI mice). M: heatmap displaying mRNA levels of genes of interest in the kidney cortex of WT and KI mice. Results are presented with error bars showing means ± SD. Statistical analysis was performed using a t test; ns, not significant. *P < 0.05 and ***P < 0.001.

KO mice did not display any visible renal structural alterations (Fig. 1D). PT markers such as megalin, NaPi-IIA, and Na⁺-K⁺-ATPase were normally expressed (Supplemental Fig. S1E). Glomerular and PT surfaces appeared unchanged (Supplemental Fig. S2, A and B). PCK1 deletion resulted in a decreased measured GFR associated with albuminuria (Fig. 1, E and F). KO mice presented decreased blood pH and bicarbonate, hyperchloremia, and a positive urinary anion gap characteristic of renal tubular acidosis (Fig. 1, G–J, and Supplemental Table S1). KO mice also displayed a decrease in renal cortical tissue glucose levels and hyperlactatemia. Fasting glycemia was not different in KO mice, and liver PCK1 expression was maintained (Fig. 1, K, L, P, and Q). PCK1 deletion induced transcriptional kidney regulation (Supplemental Fig. S1C and Supplemental Table S2), including modifications in glucose and lactate tubular transporters (Fig. 1S). Loss of SGLT2 (Slc5a2), confirmed at the protein level, and loss of Na⁺-coupled monocarboxylate transporter 2 (SMCT2; Slc5a12), resulting in glycosuria and lactaturia (Fig. 1, M, N, O, R, and S), were prominent features.

In KI mice, kidney tubular overexpression of PCK1 induced no major changes in serum parameters and, notably, no fasting or fed hyperglycemia (Fig. 2, A, B, E, F, I, and L, and Supplemental Table S3). Furthermore, no significant kidney functional, transcriptional, or structural modifications were observed (Fig. 2, A–M, and Supplemental Fig. S1D and Supplemental Tables S3 and S4).

Loss of PCK1 Results in Profound Acidosis in Mice After HCl Ingestion

In RNA-sequencing analysis of the kidney cortex, PCK1 tubular deletion resulted in increased expression of Na⁺-coupled neutral amino acid transporter 3 (Snat3) and Na⁺-bicarbonate cotransporters (Nbce1 and Nbcn1, respectively), whereas no major change was observed in KI mice (Fig. 3, A and C). To further evaluate the impact of PCK1 on the regulation of acid-base balance and ammoniagenesis, we challenged our transgenic mice with an acidic load for 2 days (27, 28). As expected, HCI ingestion resulted in hyperchloremic metabolic acidosis in WT mice (Fig. 3, E–H, and Supplemental Table S5). PCK1 gene expression was strongly increased under acidotic conditions (Fig. 3, B and D). KO mice displayed more pronounced acidosis, decreased ammoniagenesis, and renal dysfunction with decreased creatinine clearance, elevated KIM-1 expression, and albuminuria (Fig. 3, I, J, M, N, Q, and R, and Supplemental Table S5). Histological experiments confirmed enhanced tubular vacuolization compatible with tubular injury and mitochondrial dysfunction that were more pronounced in KO mice (Supplemental Fig. S2A).

Figure 3.

Two-day HCl load in phosphoenolpyruvate carboxykinase 1 (PCK1) knockout (KO) and knockin (KI) mice. A: heatmap displaying mRNA levels of genes related to acid-base transporters in the kidney cortex of wild-type (WT) and KO mice under basal conditions. B: Pck1 mRNA expression in the kidney cortex after 2 days of normal diet (N diet) and HCl diet (H diet) in WT and KO mice (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 5 KO mice on H diet). C: heatmap displaying mRNA levels of genes of acid-base transporters in the kidney cortex of WT and KI mice under basal conditions. D: Pck1 mRNA expression in the kidney cortex after 2 days of N diet and H diet in WT and KI mice (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). E: blood pH in WT and KO mice after 2 days of N diet and H diet (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 4 KO mice on H diet). F: blood bicarbonate concentration in WT and KO mice after 2 days of N diet and H diet (n = 4 WT mice on N diet, n = 6 WT on H diet, and n = 4 KO on H diet). G: blood pH in WT and KI mice after 2 days on N diet and H diet (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). H: blood bicarbonate concentration in WT and KI mice after 2 days of N diet and H diet (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). I: serum anion gap in WT and KO mice after 2 days of N diet and H diet (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 4 KO mice on H diet). J: urine anion gap in WT and KO mice after 2 days of N diet and H diet (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 4 KO mice on H diet). K: serum and urine anion gap in WT and KI mice after 2 days of N diet and H diet (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). L: serum and urine anion gap in WT and KI mice after 2 days of N diet and H diet (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). M: creatinine clearance in WT and KO mice after 2 days of N diet and H diet (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 4 KO mice on H diet). N: albuminuria in WT and KO mice after 2 days of N diet and H diet (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 5 KO mice on H diet). O: creatinine clearance in WT and KI mice after 2 days of N diet and H diet (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). P: albuminuria in WT and KI mice after 2 days of N diet and H diet (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). Q: glucose concentration in serum with N diet and H diet in PCK1 KO mice (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 4 KO mice on H diet). R: kidney injury molecule-1 (Kim1) mRNA expression in the kidney cortex after 2 days of N diet and H diet in PCK1 KO mice (n = 4 WT mice on N diet, n = 6 WT mice on H diet, and n = 5 KO mice on H diet). S: glucose concentration in serum with N diet and HCl diet in PCK1 KI mice (n = 6 PCK1^WT/WT mice on N diet, n = 6 PCK1^WT/WT mice on H diet, and n = 6 PCK1^KI/KI mice on H diet). T: Kim1 mRNA expression in the kidney cortex after 2 days with N diet and H diet in PCK1 KI mice (n = 6 WT mice on N diet, n = 6 WT mice on H diet, and n = 6 KI mice on H diet). Results are presented with error bars showing means ± SD. Statistical analysis was performed using an ANOVA test with secondary Tukey analysis; ns, not significant. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Under acidotic conditions, PCK1 overexpression in the kidney did not affect the regulation of acid-base balance and ammoniagenesis. This may be explained by the resulting increase in PCK1 gene expression in WT mice to the levels of KI mice when subjected to acidosis. Nevertheless, hypoglycemia was prevented and KIM-1, a marker of renal injury, was lower in KI mice (Fig. 3, C, D, G, H, K, L, O, P, S, and T, and Supplemental Table S5), whereas histological aspect appeared unmodified (Supplemental Fig. S2B).

PCK1 Regulates Metabolism and Mitochondrial Function in Cortical Cells and Its Restoration Rescues Renal Disease

Metabolic acidosis induces tubular injury through alterations of mitochondrial function and metabolism (29). To explain the major effect of PCK1 on susceptibility to tubular injury during acidosis, we assessed mitochondrial function in KO and KI mice. To this end, we measured basal mitochondrial respiration and maximal respiration capacity in primary kidney cortical cells from these mice (Fig. 4, A and E). We observed that basal respiration was lower in KO cells compared with WT cells (Fig. 4, B and C). Similarly, the amount of ATP in renal cortex cells was significantly lower, demonstrating impaired mitochondrial function in KO mice (Fig. 4D). Conversely, KI mice had normal respiration and ATP levels (Fig. 4, F–H).

Figure 4.

Restoration of phosphoenolpyruvate carboxykinase 1 (PCK1) in the kidney in the renal disease model of glomerular lesions and mitochondrial function. A: cell oxygen consumption and extracellular acidification rate in kidney cortical cells of wild-type (WT) and knockout (KO) mice. B: quantification of basal respiration in kidney cortical cells of WT and KO mice. C: quantification of maximal respiration of kidney cortical cells of WT and KO mice. D: quantification of ATP in kidney cortical cells of WT and KO mice (n = 8 WT mice and n = 6 KO mice). E: cell oxygen consumption and extracellular acidification rate in kidney cortical cells of WT and knockin (KI) mice. F: quantification of basal respiration in kidney cortical cells of WT and KI mice. G: quantification of maximal respiration of kidney cortical cells of WT and KI mice. H: quantification of ATP in kidney cortical cells of WT and KI mice (n = 7 WT mice and n = 6 KI mice). I: measurement of loss of kidney function between day 0 (d0) and day 14 (d14) in healthy (WT CTL) and injured (WT POD and KI POD) mice (n = 6 WT CTL mice, n = 7 WT POD mice, and n = 9 KI POD mice). J: blood creatinine of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 6 WT CTL mice, n = 9 WT POD mice, and n = 10 KI POD mice). K: blood bicarbonate of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 6 WT CTL mice, n = 9 WT POD mice, and n = 10 KI POD mice). L: Pck1 mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). M: kidney injury molecule-1 (Kim1) mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). N: pyruvate kinase M (Pkm) mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). O: glucose-6-phosphatase, catalytic (G6pc) mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). P: peroxisome proliferator-activated receptor-γ coactivator-1α (Pgc1α) mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). Q: acyl-CoA oxidase 1 (Acox1) mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). R: acyl-CoA oxidase 2 (Acox2) mRNA expression in the kidney cortex of healthy (WT CTL) and injured (WT POD and KI POD) mice after 14 days (n = 7 WT CTL mice, n = 9 WT POD mice, and n = 7 KI POD mice). S: summary schema of the role of PCK1 in renal tubules. Results are presented with error bars showing means ± SD. Statistical analysis was performed as follows: for two groups, t test was used; and for more than two groups, an ANOVA test with secondary Tukey analysis was used. *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. AU, arbitrary units; ns, not significant; GFR, glomerular filtration rate; OCR, oxygen consumption rate; SGLT2, Na⁺-glucose transporter 2. SMCT2, Na⁺-coupled monocarboxylate transporter 2.

We have previously demonstrated that PCK1 expression decreases during CKD (4). To demonstrate that PCK1 may be an attractive therapeutic target, we aimed to restore PCK1 expression in the kidney during proteinuria-related renal disease. We thus exposed KI mice to proteinuric CKD using the POD-ATTAC mouse model (30). KI mice maintained normal PCK1 expression levels during CKD comparable to healthy mice, whereas CKD mice had a significant decrease in PCK1 expression (Fig. 4L). The loss of renal function was measured by the difference in GFR between day 0 and day 14. As expected, injured WT mice displayed a decrease in renal function. In KI mice, GFR and creatinine levels were better preserved (Fig. 4, I–K). KIM-1 expression was strongly induced in the kidneys of proteinuric mice, whereas in KI mice, this increase was milder (Fig. 4M). KI mice displayed more preserved kidney histology (Supplemental Fig. S2C). As observed in our previous studies, the different markers of glycolysis and fatty acid oxidation were modulated under proteinuric kidney disease, but overexpression of PCK1 did not modify their expression (Fig. 4, N–R).

DISCUSSION

In this article, we describe the role of PCK1 in kidney tubular cells using two transgenic mouse models. We first showed that loss of tubular PCK1 modifies renal function and that PCK1 is a key regulator of gluconeogenesis and ammoniagenesis and several major glucose and lactate transporters, notably SGLT2. We further demonstrated that kidney tubule-specific deletion of PCK1 is sufficient to increase sensitivity to tubular injury in a condition of stress such as metabolic acidosis. Moreover, we showed a major role for PCK1 regarding energy production in tubular cells. Finally, we identified that maintaining PCK1 expression during CKD attenuates the severity of renal disease, independently of other metabolic pathways.

We have previously demonstrated that PCK1 is expressed mostly in the PT in the mouse kidney (31). In several human and mouse single-cell databases available in the KIT project (https://humphreyslab.com/SingleCell/), this pattern was confirmed. During injury, we showed that PCK1 expression largely decreases during acute kidney disease and CKD in humans and mice (4, 31). Diabetic kidney disease may be an exception since renal PCK1 may be overexpressed in this case (32). Historically, PCK1 has been mainly associated with glucose production in the liver. However, liver-deficient mice present normoglycemia, perhaps due to extrahepatic compensation (7). Similar to liver-deficient mice, renal PCK1 KO mice were not hypoglycemic when fasted, in line with hepatic compensation and potential modification of glucose consumption (7, 10). Tubular PCK1 deletion resulted in systemic lactate accumulation, whereas overexpression of PCK1 rescued hypoglycemia during metabolic acidosis.

A crucial novel finding from our study was that PCK1 depletion causes striking downregulation of glucose and lactate transporters, resulting in glycosuria and lactaturia. SGLT2 has emerged as a critical target to modulate disease progression in the kidney, but very little is understood about its regulation. Our findings suggest a link between metabolic disturbances and SGLT2 and SMCT2 expression, which might be important for predicting responses to SGLT2 inhibitors in certain disease scenarios. Although the pathways of these regulations are not fully understood and deserve more study, a potential role of locally elevated lactate levels or modulations of key transcription factors by PCK1 may be involved. Moreover, the alteration of GFR in KO mice may be explained by SGLT2 loss and a consequent modification of tubuloglomerular feedback (33). Importantly, albuminuria was also elevated, suggesting a detrimental role of PCK1 deletion on renal health at baseline, in addition to tubuloglomerular feedback modulation.

A main phenotype of renal PCK1 loss was metabolic acidosis. Thus, our results show that PCK1 is important for both glucose metabolism and acid-base balance, implying that it provides an important mechanistic link between gluconeogenesis and ammoniagenesis. The previously described PCK1-independent ammoniagenesis is thus insufficient to compensate for metabolic acidosis in vivo (19).

Finally, PCK1 deletion increased sensitivity to renal injury during metabolic acidosis and proteinuria. Acidosis induces tubular injury through modifications of energy production and alterations of metabolism (29). We showed here that PCK1 also governs mitochondrial function in cortical cells, most likely through its cataplerotic function (8), and show that KO mice develop severe renal dysfunction during acidosis. We further demonstrated that mitigating the downregulation of PCK1 in a proteinuric model of CKD could partially rescue the disease in terms of measured renal function. This effect seemed independent of other metabolic pathways and may be related to the effect of PCK1 on energy production, together with the protection against metabolic acidosis.

Altogether, we demonstrate that PCK1 is a key regulator of acid-base balance and glucose/lactate homeostasis as well as mitochondrial function in tubular cells. PCK1 appears to play an important role in the pathology of CKD progression, linking the downregulation of the gluconeogenetic pathway observed in several types of renal disease to tubular injury and progression of renal disease.

SUPPLEMENTAL MATERIAL

Supplemental Figs. S1–S3: https://doi.org/10.6084/m9.figshare.22344232.v1.

Supplemental Tables S1–S6: https://doi.org/10.6084/m9.figshare.22344286.v1.

Supplemental legends: https://doi.org/10.6084/m9.figshare.22344343.v1.

DATA AVAILABILITY

Data will be made available upon reasonable request.

GRANTS

S.d.S. is supported by a grant from the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (SNSF 320030_204187). D.D. is the recipient of a grant from the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (32350_214547). A.F. is the recipient of a grant from the Schweizerischer Nationalfonds zur Förderung der Wissenschaftlichen Forschung (323530_191224) and from the Fondation Center de Recherches Médicales Carlos et Elsie de Reuter. C.A.W. received honoraria from Kyowa Kirin, Advicenne, and Medice. J.M.R. is supported in part by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01DK119497.

DISCLOSURES

A.F. and S.d.S. received consulting fees from Astellas, Switzerland. S.d.S. received consulting fees from Bayer, AstraZeneca, Boehringer-Ingelheim, and Otsuka. The other authors have no conflicts of interest to declare. The authors confirm that the results presented in this paper have not been published previously in whole or part, except in abstract format.

AUTHOR CONTRIBUTIONS

S.d.S. conceived and designed research; T.V., D.D., G.A., I.S., A.F., H.A., S.B., D.P., Q.G., and J.M.R. performed experiments; T.V., D.D., G.A., I.S., A.F., H.A., S.B. J.M.R., and S.d.S. analyzed data; T.V., D.D., D.L., C.A.W., A.M.H., and S.d.S. interpreted results of experiments; T.V. and D.D. prepared figures; T.V., D.D., and S.d.S drafted manuscript; T.V., D.D., D.L., C.A.W., A.M.H., and S.d.S edited and revised manuscript.