IntroductionCollagen type VI (COL6), a critical component of the extracellular matrix, plays a crucial role in maintaining tissue structure and integrity1,2,3. Composed of three chains (α1, α2 and α3) encoded by Col6a1, Col6a2 and Col6a3, COL6 forms tetramers that aggregate into microfibrils within the extracellular space2,3. Endotrophin (ETP) is derived from the C5 domain of the COL6A3 chain and proteolytically cleaved from COL6 fibrils immediately after secretion4,5,6,7. This cleavage and the resulting separation suggest that ETP has a functional role distinct from COL6A3.Various Col6a3 mutant mouse models have been developed, differing substantially in their capacity to produce functional ETP. While these models have been used to study the biology of ETP, they clearly lack specificity due to manifest defects in COL6 function. Each model exhibits fundamentally different impairments in COL6 function and potentially distinct effects on ETP production8,9. The first model carries a hypomorphic Col6a3 allele (Col6a3hm), which leads to a near-complete absence of Col6a3 transcription, disrupted intracellular COL6 dimer and tetramer assembly and secretion and associated muscle and tendon defects8. The second model carries a disrupted Col6a3 allele (Col6a3d16), which causes an in-frame deletion of 18 amino acids at the N-terminus of the triple-helical domain of the COL6A3 parent chain9. While heterozygous Col6a3+/d16 and homozygous Col6a3d16/d16 mice exhibit normal intracellular COL6 dimer and tetramer assembly and secretion, they display disrupted extracellular microfibril formation along with muscle and tendon defects. Importantly, whereas the Col6a3hm allele is expected to decrease ETP production, the Col6a3d16 allele probably maintains ETP production. These findings suggest that the COL6A3 parent form is critical for muscle and tendon integrity, whereas ETP may not play a relevant role in this context. Nonetheless, neither model provides definitive insight into the contribution of ETP to the studied biological processes.To address these shortcomings, we now developed a novel Col6a3-Etp+mCherryCAAX mouse line using CRISPR–Cas9 genome editing to selectively modify the Col6a3 locus upstream of the endogenous ETP-coding sequence. In this model, ETP can be specifically deleted through Cre-mediated recombination, enabling functional studies of ETP without affecting the COL6A3 parent chain.ETP has been identified as a critical mediator of fibrosis, driving disease progression through fibrotic processes and inflammation and serving as a key factor in the pathophysiology of fibro-inflammatory disorders10,11,12. It stimulates fibroblasts to upregulate collagen I (COL1) synthesis, a key component of fibrosis, and elevated ETP levels have been associated with fibrosis and adverse outcomes in heart failure with preserved ejection fraction13. In addition, ETP is strongly associated with fibrosis-related conditions such as chronic kidney disease, where high ETP levels have been independently linked to increased mortality risk, underscoring its role in disease progression and prognosis14. Recently, our group demonstrated that targeting ETP with neutralizing antibodies reduces renal fibrosis and improves renal function in a mouse model of chronic kidney disease, further supporting the role of ETP in chronic fibro-inflammatory kidney diseases15. Collectively, these findings highlight the role of ETP as both a biomarker and a contributing factor in fibrotic conditions, providing a strong rationale for investigating its specific role in fibrosis independently of its parent gene, Col6a3.Here, we demonstrate that ETP is a critical driver of kidney fibrosis independent of COL6A3. Using our novel ETP-specific-knockout (ETPKO) mouse model, we show that ETP depletion reduces fibrotic protein expression and mitigates kidney fibrosis under ischemia–reperfusion injury (IRI) conditions. These findings establish ETP as a key mediator of fibrosis and validate ETPKO mice as a robust tool for studying fibrotic processes.Materials and methodsStudy approvalAll animal experimental protocols, including those for mouse use and euthanasia, were reviewed and approved by the Institutional Animal Care and Use Committee of the University of Texas Southwestern (UTSW) Medical Center under animal protocol numbers 2024-103554, 2024-103545-G and 2024-103559-G.Generation of the Col6a3-Etp+mCherryCAAX lineCol6a3-Etp+mCherryCAAX mice were generated using CRISPR–Cas9-based genome editing. A guide RNA sequence (CAGTTCAACCATCAACCTCA-[TGG]) was designed in silico using CRISPOR (www.crispor.tefor.net) to target the mouse Col6a3 open reading frame upstream of the start of the endogenous ETP-coding sequence, which naturally spans the last three exons of the gene. A donor plasmid containing a 350 bp left homology arm, a first in-frame lox2272 site, a copy of the full ETP-coding sequence with a stop codon, a second out-of-frame lox2272 site, a P2A-mCherryCAAX-coding sequence with a stop codon, a copy of the full Col6a3 3′ untranslated region (UTR), a rabbit β-globin poly(A) signal and a 350 bp right homology arm was synthesized by Genewiz (annotated sequence provided in the Supplement). This donor plasmid was used to prepare a linear, single-stranded repair construct using the Guide-it Long ssDNA Production System v2 with the following primers: 5′-TGGGTACTTAGGCTACACCCTG-3′ and 5′-GACCACAAGTCAACCCTAGCC-3′. An Alt-R CRISPR–Cas9 CRISPR RNA (crRNA) guide, trans-activating CRISPR RNA (tracrRNA and Cas9 protein were mixed with the repair construct and used for pronuclear injection into fertilized C57BL/6J eggs by the UTSW Transgenic Core. Injected eggs were transplanted into foster mothers, and the obtained offspring were screened for site-specific integration of the transgene by standard PCR. Candidate mice with apparent site-specific integration of the transgene were crossed to C57BL/6J mice. The offspring from these crosses were genotyped by PCR, and the full sequence of the integrated transgene, as well as the upstream and downstream regions, was verified by Sanger sequencing (Azenta/Genewiz). Sequence alignments were performed in SnapGene. Fully sequence-verified mice were crossed to C57BL/6J mice for at least two more generations. We used the following primers for routine genotyping of the transgene: G336 5′-TCTTCAGGCAGCACACCGAG-3′; G337 5′-TCACCATAGGACCGGGGTTTT-3′; and G338 5′-CTGAGGACCCCTTTGGAACTG-3′. These primers detect the modified (405 bp), nonmodified (241 bp) and knockout (161 bp) alleles. For further validation of recombination, the PCR products of the modified and knockout alleles were isolated using the Monarch PCR and DNA Cleanup Kit and analyzed by Sanger sequencing (Azenta/Genewiz). Before Cre recombination, the modified Col6a3-Etp+mCherryCAAX locus produces an messenger RNA that encodes the COL6A3 chain, a short 12-amino-acid ‘ITSYRILYTKLS’ linker derived from the in-frame lox2272 site and ETP. This mRNA possesses an untranslated region that includes the second out-of-frame lox2272 site, P2A-mCherryCAAX-coding sequence, Col6a3 3′UTR and a poly(A) tail. Importantly, the last two exons of the Col6a3 gene that encompass most of the endogenous ETP-coding sequence do not contribute to the Col6a3-Etp+mCherryCAAX mRNA because transcription of the modified locus terminates upstream at the introduced rabbit β-globin poly(A) signal. Cre-mediated recombination of the lox2272 sites removes the ETP-coding sequence and brings the P2A-mCherryCAAX-coding sequence in-frame. The resulting mRNA encodes the COL6A3 chain, a ‘ITSYRILYTKLS-GSG-ATNFSLLKQAGDVEENPG*P’ stretch derived from the lox2272 site and P2A peptide and mCherryCAAX, followed by an untranslated region that includes the Col6a3 3′UTR and a poly(A) tail. The P2A peptide causes ribosomal skipping during translation, resulting in the separation of the membrane-targeted red fluorescent mCherryCAAX protein from the remainder of the COL6A3 chain (note the indicated ‘G*P’ cleavage site).Mouse maintenanceAll mice used in this study, including littermate controls, were maintained on a pure C57BL/6 genetic background. Mice were housed under barrier conditions on a 12–12-h light–dark cycle in a temperature-controlled environment (22 °C) with ad libitum access to autoclaved water and chow diet. The cages were changed every other week, and constant veterinary supervision was provided. The diets used in this study include regular chow diet.Genotyping PCRGenotyping PCR assays were performed as described previously16. Briefly, a small piece of the mouse tail tip was incubated in 50 mM NaOH at 95 °C for 1.5 h and neutralized with 10% vol/vol 1 M Tris–HCl (pH 8.0). The supernatant was used as a PCR template, and the products were analyzed on a 1–2% agarose gel stained with ethidium bromide. Primer sequences are listed in Supplementary Table 2.Unilateral kidney ischemia–reperfusion modelThe unilateral kidney ischemia–reperfusion model followed the protocol from previous studies17,18,19. A mixture of ketamine (25 mg/ml) and xylazine (2.5 mg/ml) was administered via intraperitoneal injection to anesthetize the mouse. The anesthetized mouse was placed on an infrared warming pad (Kent Scientific Corporation) to maintain body temperature at 36–37 °C. To prevent dryness during anesthesia, Puralube Vet Ointment was applied to the eyes. The hair around 1 cm lateral to the spine below the 13th rib was shaved, and the skin was cleaned three times using 75% ethanol followed by povidone-iodine. A small incision was made using scissors and the skin was bluntly separated from the peritoneum. The intestines were gently displaced toward the right side of the abdominal cavity using sterile, saline-moistened cotton swabs. The renal pedicles were exposed by carefully separating the fascia and adipose tissue using forceps. Renal ischemia was induced by applying a micro clip to the renal artery and vein, with successful ischemia visually confirmed by the gradual and uniform darkening of the kidney. After the designated ischemia duration, the clip was removed, and successful reperfusion was confirmed by the rapid color change of the kidney from dark red to dark pink. Finally, the peritoneum and skin were closed using Vicryl 5-0 sutures and clips, respectively.Tissue preparation for histological analysisThe mice were euthanized via cervical dislocation after isoflurane anesthesia. The tissues were fixed in 10% formalin for 24 h at room temperature, stored in 50% ethanol, embedded in paraffin and cut into 5 μm sections for histological analysis.Histopathological analysisFor histopathological phenotyping, two age-matched (10-week-old) mice of each sex per genotype were submitted to the UTSW ARC Diagnostic Lab. The mice were killed, and hematoxylin and eosin (H&E)-stained slides were prepared from multiple tissues. Each slide was reviewed by the UTSW ARC Diagnostic Lab to evaluate the phenotype.ImmunostainingImmunostaining assays were performed as described previously16. Briefly, 5 μm paraffin sections were deparaffinized, subjected to antigen retrieval and blocked in 10% goat serum. The slides were incubated overnight at 4 °C with anti-ETP (1:300), anti-α-SMA (1:500), anti-COL6 (1:500), anti-mCherry (1:500), anti-RFP (1:500), anti-PDGFRB (1:500), anti-CD31 (1:500) or anti-F4/80 (1:500) primary antibodies in 5% BSA, followed by goat-derived Alexa Fluor-labeled secondary antibodies (1:1,000) for 1 h at room temperature. After washing, slides were mounted with 4′,6-diamidino-2-phenylindole-containing medium and imaged using a Zeiss LSM880 confocal microscope provided by the UTSW Quantitative Light Microscopy Core Facility. Image analysis and quantification were performed using FIJI/ImageJ.H&E and picrosirius red stainingFor H&E staining, 5 μm paraffin sections were deparaffinized, rehydrated and stained using an H&E Staining Kit. For picrosirius red staining, 5 μm paraffin sections were deparaffinized, rehydrated and sequentially stained with Weigert’s hematoxylin, phosphomolybdic acid and picrosirius red, with rinses performed after each staining step. The sections were then rinsed in 0.1 N hydrochloric acid and 0.5% acetic acid, dehydrated in ethanol and processed for imaging. The entire kidney image was acquired using a Hamamatsu NanoZoomer 2.0 HT provided by the UTSW Whole Brain Microscopy Facility. Image analysis was performed using NDP.view2.Aorta microdissection and vessel-wall-thickness calculationThe mice were killed, and the aortas were carefully dissected and fixed 10% formalin for 24 h at room temperature. Following fixation, surrounding adipose tissue was gently removed, tissue from two anatomical regions was embedded in paraffin, cut into 5 μm paraffin sections and stained with H&E. The images were acquired on a Keyence BZ-X800 series microscope and analyzed using FIJI/ImageJ. Vessel-wall thickness was determined by two methods. First, by calculation of the distance between the external elastic lamina and the internal elastic lamina. Second, by calculation of half of the difference between the mean outer diameter and the mean luminal diameter, providing an average estimate across the vessel20. For the latter procedure, for each anatomical region prepared, four to five measurements of both the outer vessel diameter and luminal diameter were performed, including two measurements along perpendicular axes.Preparation of whole-cell extracts from tissues and immunoblottingFrozen tissues were pulverized and resuspended in RIPA buffer using a glass douncer on ice. The mixture was incubated at 4 °C for 20 min with gentle mixing, followed by centrifugation to remove insoluble material. The supernatant was collected as the soluble extract, and the protein concentrations were determined using a BCA protein assay. For western blotting, 20 μg of protein was separated on a 4–12% gradient SDS–polyacrylamide gel electrophoresis gel and transferred to a nitrocellulose membrane. The membranes were blocked in 5% nonfat dry milk in TBST and incubated overnight at 4 °C with anti-ETP (1:500), anti-α-SMA (1:1,000), anti-COL1 (1:1,000), anti-COL6 (1:1,000), anti-mCherry (1:1,000), anti-RFP (1:1,000) or anti-GAPDH (1:2,000) primary antibodies in TBST supplemented with 5% wt/vol BSA. After washing in TBST, HRP-conjugated secondary antibodies were applied, and the protein signals were detected using chemiluminescence on a Thermo Fisher Scientific iBright 1500 system.RT–qPCRTotal RNA was extracted using RNeasy Mini Kit, Trizol and EZ-10 DNAaway RNA Miniprep Kit, followed by cDNA synthesis with PrimeScript RT Master Mix. A quantitative PCR was performed using PowerUp SYBR Green Master Mix on a Thermo Fisher Scientific QuantStudio 6 Flex system. Primer sequences are listed in Supplementary Table 3.StatisticsThe statistical analyses were performed using Prism, applying two-tailed Student’s t-tests for pairwise comparison and one-way analysis of variance (ANOVA) for group comparison. The statistical significance was set at P