The axolotl kidney: a novel model to study kidney regeneration

Chronic kidney disease remains a major public health burden worldwide with no ideal treatment besides dialysis or kidney transplantation.1Mehta R.L. Cerda J. Burdmann E.A. et al.International Society of Nephrology's 0by25 initiative for acute kidney injury (zero preventable deaths by 2025): a human rights case for nephrology.Lancet. 2015; 385: 2616-2643Abstract Full Text Full Text PDF PubMed Scopus (654) Google Scholar,2Naved B.A. Bonventre J.V. Hubbell J.A. et al.Kidney repair and regeneration: perspectives of the NIDDK (Re)Building a Kidney consortium.Kidney Int. 2022; 101: 845-853Abstract Full Text Full Text PDF PubMed Scopus (10) Google Scholar Promoting kidney regeneration is a promising strategy to treat patients with end-stage renal disease. Mammalian kidney can partially repair damaged nephrons in the condition of minor injury and progress to irreversible renal fibrosis with severe injury.3Humphreys B.D. Valerius M.T. Kobayashi A. et al.Intrinsic epithelial cells repair the kidney after injury.Cell Stem Cell. 2008; 2: 284-291Abstract Full Text Full Text PDF PubMed Scopus (696) Google Scholar,4Kumar S. Liu J. Pang P. et al.Sox9 activation highlights a cellular pathway of renal repair in the acutely injured mammalian kidney.Cell Rep. 2015; 12: 1325-1338Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar By contrast, axolotl (Ambystoma mexicanum) exhibits a remarkable ability to regrow multiple body parts, including its entire limbs, the tail, ocular tissues, liver, heart, and even brain.5Gerber T. Murawala P. Knapp D. et al.Single-cell analysis uncovers convergence of cell identities during axolotl limb regeneration.Science. 2018; 362eaaq068Crossref Scopus (187) Google Scholar, 6Nowoshilow S. Schloissnig S. Fei J.F. et al.The axolotl genome and the evolution of key tissue formation regulators.Nature. 2018; 554: 50-55Crossref PubMed Scopus (289) Google Scholar, 7Wei X. Fu S. Li H. et al.Single-cell stereo-seq reveals induced progenitor cells involved in axolotl brain regeneration.Science. 2022; 377eabp9444Crossref Scopus (14) Google Scholar, 8Otsuki L. Tanaka E.M. Positional memory in vertebrate regeneration: a century's insights from the salamander limb.Cold Spring Harb Perspect In Biol. 2022; 14: a040899Crossref PubMed Scopus (6) Google Scholar, 9Lust K. Maynard A. Gomes T. et al.Single-cell analyses of axolotl telencephalon organization, neurogenesis, and regeneration.Science. 2022; 377eabp9262Crossref PubMed Scopus (5) Google Scholar,S1 However, little is known about the ability of axolotl kidney regeneration after injury. In this study, we showed that the axolotl kidney shares structural, molecular, and functional similarities with the mammalian kidney. We further showed that axolotl nephrons can regenerate after severe renal injury. We also found that many genes involved in mammalian kidney development are upregulated at the early stage of axolotl renal tubule regeneration. The axolotl has one pair of "pear-shaped" kidneys that are located on the dorsal side of the body cavity, as shown in Figure 1a and b. Hematoxylin and eosin staining (see Supplementary Methods) showed that the oval-shaped glomeruli (black dotted circle) are located in the middle portion of the kidney, and the proximal tubules with brush borders and the distal tubules are positioned in the outside of the kidney (Figure 1c). Hematoxylin and eosin staining also showed that the axolotl nephron is composed of a corpuscle containing the glomerulus and Bowman's capsule, and a renal tubule composed of the glomerulotubular junction, the proximal tubule, and the distal tubule (Figure 1c). Ultrastructural examination of axolotl kidney using transmission electron microscopy showed that the proximal tubule has the brush borders composed of several neatly arranged microvilli, whereas the distal tubule has regular lumen and a few short microvilli (Figure 1d). Transmission electron microscopy examination (see Supplementary Methods) also showed the normal podocyte cell body, the intact primary foot processes of podocytes, the secondary foot processes of podocytes, and the slit diaphragm structures (Figure 1e). Immunofluorescence staining results showed that podocin, the cell marker of podocyte in mammal, was highly expressed in the axolotl glomerulus (Figure 1f). Megalin, a multifunctional endocytic clearance receptor for circulating proteins, was highly expressed in the proximal tubules costained with peanut agglutinin (PNA; Figure 1g). Na-K-2Cl cotransporter 2 (NKCC2), a marker for the thick ascending limb of the loop of Henle, was specifically expressed in the tubules connecting with proximal tubules (Figure 1h). Calbindin, a marker for the distal tubules in mammal, was specifically expressed in the tubules costained with Dolichos Biflorus Agglutinin (DBA; Figure 1i). We found that the axolotl kidney could filtrate and reabsorb small (10,000 and 40,000 molecular weight) but exclude large (70,000 and 155,000 molecular weight) fluorescent-labeled dextrans, suggesting that the axolotl slit diaphragm functions as a size-based filtration diaphragm (Figure 1j). These data suggested that the axolotl kidney shares remarkable structural, molecular, and functional similarities with the mammalian kidney. The axolotl renal tubular injury model was established by i.p. injection of gentamicinS2,S3 (see Supplementary Methods, Supplementary Figure S1). As shown in Figure 2a, massive disorganization of proximal tubules occurred 7 days after injection (dpi), and numerous basophilic cellular aggregates were formed between the injured tubules at 14 dpi. These basophilic cellular aggregates possess faint lumen morphology at 30 dpi and then enlongated into immature tubules at 60 days after injury. Many regenerated proximal tubules with brush border were observed at 90 dpi (Figure 2a). The Paller's score showed that severe damage in the kidney structure occurred at 7 dpi (Figure 2c). The quantification of hematoxylin and eosin staining also suggested that tubular regeneration was arising from these basophilic cellular aggregates, which will then form lumens and elongate into mature tubules (Figure 2d). Cell death and proliferation are 2 major events in the process of tissue regeneration/repair after injury. Terminal deoxynucleotidyl transferase–mediated dUTP nick end-labeling (TUNEL)–positive cells were present at 3 dpi, and the number of TUNEL-positive cells peaked at 5 dpi. The number of TUNEL-positive cells significantly decreased by 7 dpi and returned to the basal level by 14 dpi (Supplementary Figure S2A and D). The percentage of 5-ethynyl-2′-deoxyuridine (EdU+) cells increased gradually from 3 to 7 dpi, and EdU+ cells were scattered throughout the whole kidney. The percentage of cells with EdU incorporation peaked at 14 dpi, and EdU+ cells started emerging in some tubular structures. The EdU-labeling signal in EdU+ cells at 14 dpi was much weaker than that of EdU+ cells at 7 dpi, indicating that these cells were proliferating frequently at this time point (Supplementary Figure S2B, C, and E). These results showed that renal cells were undergoing drastic proliferation during the renal regeneration process after injury. To provide a comprehensive profile of the regulatory network at the transcriptional level, we performed total mRNA sequencing (bulk RNA sequencing) of the whole kidney from both 14 days after gentamicin injection group and the control group. The transcriptome expression profiles showed 2444 upregulated genes and 2668 downregulated genes comparing the injured samples with control samples (Supplementary Figure S3A). The transcriptomic data were deposited and are available at National Center for Biotechnology Information with the Gene Expression Omnibus accession number GSE228583. The downregulated genes, such as Lrp2 and Slc12a3, were exclusively related to tubular function (Supplementary Figure S3B). Gene Ontology enrichment analysis showed that upregulated genes were assembled in metabolic processes, including oxidation-reduction process, fatty acid metabolic process, adenosine triphosphate metabolic process, and extracellular matrix organization process (Supplementary Figure S3C and D). In addition, the enriched Gene Ontology terms were associated with cell cycle, DNA replication, and cell cycle processes (Supplementary Figure S3E). Several nephrogenesis-related genes (such as Ccnd1, Top2a, Sox9, Bmp7, and Wnt4) were significantly upregulated in the gentamicin injection group (Supplementary Figure S3B), which further was confirmed by quantitative real-time polymerase chain reaction, Western blot, immunostaining, and in situ hybridization (Supplementary Figure S4). These results suggest that nephrogenesis-related genes may contribute to the axolotl kidney regeneration after severe renal tubule injury. The axolotl renal glomerular injury model was established by tail i.v. injection of doxorubicin.S4 As shown in Figure 2e, compared with the control kidney, periodic acid–Schiff staining showed that the glomeruli were separated from the Bowmen's capsules and there was some protein cast accumulated in the lumen 1 week after injury. The interstitium matrix was thicker than that of the control group 2 weeks after injection, and the damage was much more severe. The regenerated glomeruli were much smaller than those of the control group 4 weeks after injury. The size of glomeruli was similar to that of the control group 8 weeks after injury (Figure 2f). The expression level of podocin protein was recovered with the progress of the glomerular regeneration (Figure 2g). The morphology of the glomerulus was also recovered 8 weeks after injury (Figure 2h and i). These results suggested that renal glomerulus can regenerate after doxorubicin-induced renal glomerular injury. Our renal protein filtration and reabsorption analysis also showed that the glomerular filtration function was recovered (Supplementary Figure S5). In this study, we systemically characterized the axolotl kidney and showed that it shares remarkable structural, molecular, and functional similarities with the mammalian kidney. Structurally, the axolotl kidney has the basic unit of the mammalian kidney, the nephron, which is composed of a glomerulus and a tubule system. The podocyte, one of the major cell types in the glomeruli, has a similar foot processes as that of mammalian kidney. Its glomerular filtration barrier is also similar to that of the mammalian kidney. The epithelium of its renal proximal tubules is converted to brush border that greatly increases the luminal surface area, increasing the efficiency of reabsorption. Molecularly, the axolotl kidney expresses tubular mature markers (megalin and Na-K-2Cl cotransporter 2 [NKCC2]) and podocyte proteins (podocin), indicative of the similarity with mammalian kidney. Functionally, similar to the mammalian kidney, the axolotl kidney has both glomerular ultrafiltration and protein reabsorption functions of renal proximal tubule. Given the remarkable similarity between the axolotl kidney and the mammalian kidney, our findings suggest that the axolotl kidney could be an ideal model to study renal physiology, pathology, and regeneration in vivo. In this study, we also demonstrated that the axolotl nephrons regenerate after severe renal injury. First, in the gentamicin-induced nephrotoxicity model, we showed that the renal tubules regenerate 2 months after severe renal injury. Our bulk RNA-sequencing results showed the upregulation of genes involved in cell proliferation, extracellular matrix remodeling, and nephrogenesis.4Kumar S. Liu J. Pang P. et al.Sox9 activation highlights a cellular pathway of renal repair in the acutely injured mammalian kidney.Cell Rep. 2015; 12: 1325-1338Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar,S3,S5–S11 The coordinated activation of this gene network may suppress renal fibrosis and promote kidney regeneration, and ultimately determine whether the finale is the perfect kidney regeneration or renal fibrosis. The activation of nephrogenesis-related genes, such as Sox9 and Wnt4, during axolotl kidney regeneration also suggests that the function of these genes is evolutionarily conserved.4Kumar S. Liu J. Pang P. et al.Sox9 activation highlights a cellular pathway of renal repair in the acutely injured mammalian kidney.Cell Rep. 2015; 12: 1325-1338Abstract Full Text Full Text PDF PubMed Scopus (126) Google Scholar,S7,S12–S14 Later, we showed the regeneration of glomeruli in doxorubicin-induced glomerular injury model. Our results showed that glomeruli were regenerated and much smaller than that of the control group 4 weeks after injury. The size of regenerated glomerulus is similar to that of the control group 8 weeks after injury. The expression level of podocin was also recovered with the progress of the glomerular regeneration. So far, we can only distinguish the regenerated glomeruli from the undamaged ones based on the size of the glomerulus because we do not have the transgenic axolotl lines that can be used to trace the regeneration of the glomeruli. In summary, in this study, we showed that the axolotl kidney shares remarkable similarities with mammalian kidney, and the axolotl kidney regenerates after severe injury. Our study is the first to show that the axolotl could serve as a novel model organism for studying kidney regeneration after injury. Studies of kidney regeneration in axolotl will provide invaluable information for the understanding of the molecular mechanism underlying kidney regeneration. All the authors declared no competing interests. This work was supported by the National Key Research Plan (2017YFA0104602 to FZ); the National Natural Science Foundation of China Key Program (82030022 to FFH); the Program of Introducing Talents of Discipline to Universities, 111 Plan (D18005 to FFH); the Guangdong Key Program of Precision Medicine (2022 to FFH); Guangdong Provincial Clinical Research Center for Kidney Disease (2020B1111170013 to FFH); the National Key Research and Development Program of China (2019YFE0106700 and 2021YFA0805000 to J-FF); the National Natural Science Foundation of China (31970782 and 92268114 to J-FF); and the High-Level Hospital Construction Project of Guangdong Provincial People's Hospital (DFJHBF202103 and KJ012021012 to J-FF). We thank Fang Yang and Miaomiao Zhou for their assistance in electron microscopy. We also thank Professor Youhua Liu and Professor Haining Zhu for discussions on the project. FZ and FFH conceptualized the paper. LC, JL, MK, YO, JD, XH, and SG performed the experiments and validated the study. FZ, LC, YO, XH, JL, JD, and SG performed the formal analysis. LC, YO, JL, JD, XH, and SG investigated. FZ and J-FF provided resources. FZ curated the data. FZ and LC wrote the original draft and visualized the study. FZ, LC, and FFH reviewed and edited. FZ, J-FF, and FFH supervised. FZ and FFH performed project administration and acquired funding. Download .docx (18.23 MB) Help with docx files Supplementary File (Word) Supplementary Methods. Supplementary Figure S1. Establishment of the axolotl severe acute kidney injury (AKI) model by injecting different doses of gentamicin. (A) Experimental scheme of the optimization of the gentamicin concentration. (B) Hematoxylin and eosin (HE) staining showed the injury of renal tubules with different doses of gentamicin. The slightly dilated tubules were found at 200 μg/g. The lumen of many renal tubules was dilated at 500 μg/g (one black asterisk). Cellular debris (2 black asterisks) was observed in the proximal tubule lumen at 1 and 2 mg/g. Bars = 100 μm. (C) Detection of the reabsorption function of the proximal tubules in kidneys injected with different doses of gentamicin. The 10-kDa dextran was not found in any renal tubules at 1 and 2 mg/g. Bars = 100 μm. Supplementary Figure S2. Cell proliferation and apoptosis in the axolotl kidney after gentamicin-induced kidney injury. (A) Terminal deoxynucleotidyl transferase–mediated dUTP nick end-labeling (TUNEL) analysis of cell apoptosis at different time points after injury. The proximal tubules were counterstained with PNA (green). The peak of cell apoptosis was detected at 5 days after injection (dpi). The level of apoptotic cells decreased by 7 dpi and returned to the basal level by 14 dpi. Bars = 50 μm. (B) Experimental scheme of the examination of cell proliferation at different time points after injury. (C) EdU incorporation analysis of cell proliferation at different time points after injury. The 10-kDa rhodamine B isothiocyanate (RITC)–dextran was injected before injury to label all the renal proximal tubules. The percentage of cells with EdU incorporation peaked at 14 dpi, and EdU+ cells started to form tubular-like structures (white dotted circles) at 14 and 21 dpi. Bars = 50 μm. (D) Quantification of the percentage of TUNEL+ cells in the proximal tubules. (E) Quantification of the percentage of EdUpositive cells in the whole kidney. Data are presented as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001. Data in (D) were acquired from 5 individual replicates, and data in (E) were acquired from 6 individual replicates. NS, no significance. Supplementary Figure S3. The transcriptional regulation network of the axolotl kidney after gentamicin-induced kidney injury. (A) Volcano plot from RNA sequencing showed the differentially expressed genes from the control kidney and gentamicin-induced injury kidney. (B) Heat map showing the expression of the differentially expressed genes in gentamicin-induced injury group versus control group. (C–E) Gene Ontology (GO) enrichment analysis showed upregulated genes assembled in several biological processes. (C) The enriched GO terms were associated with metabolic processes, including oxidation-reduction process, fatty acid metabolic process, and adenosine triphosphate (ATP) metabolic process. (D) The enriched GO terms were associated with the extracellular matrix organization process. (E) The enriched GO terms were associated with cell cycle, DNA replication, and cell cycle processes. Supplementary Figure S4. Validation of the expression of Wnt pathway components and SOX9 in regenerating kidneys. (A) Quantitative data of the relative CCND1, TOP2A, and BMP7 mRNA levels in gentamicin-induced injury group versus control group using quantitative real-time polymerase chain reaction (RT-qPCR). (B) Representative Western blot images showing the increased expression of WNT4 in the kidney at 14 days after injection (dpi). (C) Quantitative data of the relative WNT4 mRNA levels in gentamicin-induced injury group versus control group. (D) Quantitative data showing the expression of WNT4 protein by 14 dpi compared with the control kidney. (E) Representative micrographs showing the double immunostaining of PNA (red) with WNT4 and active β-catenin at 14 and 60 dpi. Higher-magnification images of rectangles were shown (bottom). Bars = 25 μm. (F) Quantitative data of the relative SOX9 mRNA levels in gentamicin-induced injury group versus control group. (G) Representative micrographs showing the double immunostaining of PNA (red) with SOX9 14 and 60 days after gentamicin-induced injury. Higher-magnification images of rectangles were shown (bottom). Bars = 25 μm. (H) In situ hybridization showing the expression of SOX9 in axolotl kidney at 14 dpi. Higher-magnification images of rectangles were shown (bottom). Bars = 50 μm. Data are presented as mean ± SD. ∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001. Data were acquired from 5 to 6 individual replicates. NS, no significance. Supplementary Figure S5. The regenerated renal glomeruli are functional after doxorubicin-induced kidney injury. (A) Schematic diagram of the analysis of the glomerular filtration function. (B) Functional analysis of regenerated renal glomeruli. The white dotted circles indicated the glomerulus, and the white arrowheads indicated the proximal tubules with 150-kDa fluorescein isothiocyanate (FITC)–dextran accumulation. Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI) (blue). Higher-magnification images of rectangles were shown. Bars = 100 μm. Supplementary References.