• View in gallery

    Cluster plots of coexpressed genes in banked tissue samples of the IKs (n = 6) and CNIKs (5) from 6 cats that had undergone 90 minutes of unilateral (right) RI to experimentally induce CKD 6 months before tissue collection (RI group) and in banked tissue samples of the right kidneys (control kidneys; 9) from 9 healthy cats (control group), with each plot representing 1 of the 20 gene clusters that explained most of the variances among samples. In each plot, the y-axis represents transformed, normalized gene expression, the x-axis represents the 20 individual renal tissue samples (CNIKs, sample Nos. 1 through 5; IKs, sample Nos. 6 through 11; and control kidneys, sample Nos. 12 through 20), the red lines represent genes that had a cluster membership Tmax value < 0.8, and the black lines represent genes that had a cluster membership Tmax value ≥ 0.8. Adequate RNA integrity was not confirmed for the remaining CNIK sample; therefore, it was not included in the analysis.

  • 1. Brown CA, Elliott J, Schmiedt CW, et al. Chronic kidney disease in aged cats clinical features, morphology, and proposed pathogeneses. Vet Pathol 2016;53:309326.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2. Lulich J, Osborne C, O'Brien T, et al. Feline renal failure: questions, answers, questions. Compend Contin Educ Pract Vet 1992;14:127153.

    • Search Google Scholar
    • Export Citation
  • 3. Marino CL, Lascelles BDX, Vaden SL, et al. Prevalence and classification of chronic kidney disease in cats randomly selected from four age groups and in cats recruited for degenerative joint disease studies. J Feline Med Surg 2014;16:465472.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4. Bartges JW. Chronic kidney disease in dogs and cats. Vet Clin North Am Small Anim Pract 2012;42:669692.

  • 5. Levey AS, Eckardt K-U, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005;67:20892100.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Eddy AA. Overview of the cellular and molecular basis of kidney fibrosis. Kidney Int Suppl (2011) 2014;4:28.

  • 7. Fogo AB. Mechanisms of progression of chronic kidney disease. Pediatr Nephrol 2007;22:20112022.

  • 8. Basile DP, Bonventre JV, Mehta R, et al. Progression after AKI: understanding maladaptive repair processes to predict and identify therapeutic treatments. J Am Soc Nephrol 2016;27:687697.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Gewin L, Zent R, Pozzi A. Progression of chronic kidney disease: too much cellular talk causes damage. Kidney Int 2017;91:552560.

  • 10. Lawson J, Elliott J, Wheeler-Jones C, et al. Renal fibrosis in feline chronic kidney disease: known mediators and mechanisms of injury. Vet J 2015;203:1826.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11. Sánchez-Lara AC, Elliott J, Syme HM, et al. Feline chronic kidney disease is associated with upregulation of transglutaminase 2: a collagen cross-linking enzyme. Vet Pathol 2015;52:513523.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12. Lawson JS, Syme HM, Wheeler-Jones CPD, et al. Urinary active transforming growth factor β in feline chronic kidney disease. Vet J 2016;214:16.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13. Mimura I, Nangaku M. The suffocating kidney: tubulointerstitial hypoxia in end-stage renal disease. Nat Rev Nephrol 2010;6:667678.

  • 14. Hewitson TD. Renal tubulointerstitial fibrosis: common but never simple. Am J Physiol Renal Physiol 2009;296:F1239F1244.

  • 15. McLeland SM, Cianciolo RE, Duncan CG, et al. A comparison of biochemical and histopathologic staging in cats with chronic kidney disease. Vet Pathol 2015;52:524534.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16. van Beusekom CD, Zimmering TM. Profibrotic effects of angiotensin II and transforming growth factor beta on feline kidney epithelial cells. J Feline Med Surg 2019;21:780787.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17. Lawson JS, Liu H-H, Syme HM, et al. The cat as a naturally occurring model of renal interstitial fibrosis: characterisation of primary feline proximal tubular epithelial cells and comparative pro-fibrotic effects of TGF-β1. PLoS One 2018;13:e0202577.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Arata S, Ohmi A, Mizukoshi F, et al. Urinary transforming growth factor-β1 in feline chronic renal failure. J Vet Med Sci 2005;67:12531255.

  • 19. Habenicht LM, Webb TL, Clauss LA, et al. Urinary cytokine levels in apparently healthy cats and cats with chronic kidney disease. J Feline Med Surg 2013;15:99104.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Lourenço BN, Coleman AE, Schmiedt CW, et al. Profibrotic gene transcription in renal tissues from cats with ischemia-induced chronic kidney disease. Am J Vet Res 2020;81:180189.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Lourenço BN, Coleman AE, Tarigo JL, et al. Evaluation of profibrotic gene transcription in renal tissues from cats with naturally occurring chronic kidney disease. J Vet Intern Med 2020;34:14761487.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Schmiedt CW, Brainard BM, Hinson W, et al. Unilateral renal ischemia as a model of acute kidney injury and renal fibrosis in cats. Vet Pathol 2016;53:87101.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23. Brown CA, Rissi DR, Dickerson VM, et al. Chronic renal changes after a single ischemic event in an experimental model of feline chronic kidney disease. Vet Pathol 2019;56:536543.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24. DiBartola SP, Rutgers HC, Zack PM, et al. Clinicopathologic findings associated with chronic renal disease in cats: 74 cases (1973–1984). J Am Vet Med Assoc 1987;190:11961202.

    • Search Google Scholar
    • Export Citation
  • 25. Kuncio GS, Neilson EG, Haverty T. Mechanisms of tubulointerstitial fibrosis. Kidney Int 1991;39:550556.

  • 26. Fine LG, Bandyopadhay D, Norman JT. Is there a common mechanism for the progression of different types of renal diseases other than proteinuria? Towards the unifying theme of chronic hypoxia. Kidney Int Suppl 2000;75:S22S26.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27. Fine LG, Norman JT. Chronic hypoxia as a mechanism of progression of chronic kidney diseases: from hypothesis to novel therapeutics. Kidney Int 2008;74:867872.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28. Palm F, Nordquist L. Renal tubulointerstitial hypoxia: cause and consequence of kidney dysfunction. Clin Exp Pharmacol Physiol 2011;38:474480.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29. Cowgill LD, Polzin DJ, Elliott J, et al. Is progressive chronic kidney disease a slow acute kidney injury? Vet Clin North Am Small Anim Pract 2016;46:9951013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30. Kaballo MA, Elsayed ME, Stack AG. Linking acute kidney injury to chronic kidney disease: the missing links. J Nephrol 2017;30:461475.

  • 31. Cisek K, Krochmal M, Klein J, et al. The application of multi-omics and systems biology to identify therapeutic targets in chronic kidney disease. Nephrol Dial Transplant 2016;31:20032011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32. Ghosh S, Chan CK. Analysis of RNA-seq data using TopHat and Cufflinks. Methods Mol Biol 2016;1374:339361.

  • 33. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 2010;26:139140.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34. Love M, Anders S, Huber H. Differential analysis of count data—the DESeq2 package. Available at: bioconductor.jp/packages/2.14/bioc/vignettes/DESeq2/inst/doc/DESeq2.pdf. Accessed May 6, 2021.

    • Search Google Scholar
    • Export Citation
  • 35. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 1995;57:289300.

    • Search Google Scholar
    • Export Citation
  • 36. Mi H, Lazareva-Ulitsky B, Loo R, et al. The PANTHER database of protein families, subfamilies, functions and pathways. Nucleic Acids Res 2005;33(suppl 1):D284D288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 37. Rau A, Maugis-Rabusseau C. Transformation and model choice for RNA-seq co-expression analysis. Brief Bioinform 2018;19:425436.

  • 38. Schurch NJ, Schofield P, Gierliński M, et al. Evaluation of tools for differential gene expression analysis by RNA-seq on a 48 biological replicate experiment. arXiv 2015;1505.02017. doi:10.1261/rna.053959.115.

    • Search Google Scholar
    • Export Citation
  • 39. Argraves WS, Greene LM, Cooley MA, et al. Fibulins: physiological and disease perspectives. EMBO Rep 2003;4:11271131.

  • 40. Neiman M, Hedberg JJ, Dönnes PR, et al. Plasma profiling reveals human fibulin-1 as candidate marker for renal impairment. J Proteome Res 2011;10:49254934.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 41. Scholze A, Bladbjerg E-M, Sidelmann JJ, et al. Plasma concentrations of extracellular matrix protein fibulin-1 are related to cardiovascular risk markers in chronic kidney disease and diabetes. Cardiovasc Diabetol 2013;12:6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 42. Zhang ZX, Shek K, Wang S, et al. Osteopontin expressed in tubular epithelial cells regulates NK cell-mediated kidney ischemia reperfusion injury. J Immunol 2010;185:967973.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 43. Pichler R, Giachelli CM, Lombardi D, et al. Tubulointerstitial disease in glomerulonephritis. Potential role of osteopontin (uropontin). Am J Pathol 1994;144:915926.

    • Search Google Scholar
    • Export Citation
  • 44. Kahles F, Findeisen HM, Bruemmer D. Osteopontin: a novel regulator at the cross roads of inflammation, obesity and diabetes. Mol Metab 2014;3:384393.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 45. Yan X, Sano M, Lu L, et al. Plasma concentrations of osteopontin, but not thrombin-cleaved osteopontin, are associated with the presence and severity of nephropathy and coronary artery disease in patients with type 2 diabetes mellitus. Cardiovasc Diabetol 2010;9:70.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 46. Wang IC, Hsu WL, Wu PH, et al. Neutrophil gelatinase associated lipocalin in cats with naturally occurring chronic kidney disease. J Vet Intern Med 2017;31:102108.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 47. Cranenburg EC, Koos R, Schurgers LJ, et al. Characterisation and potential diagnostic value of circulating matrix Gla protein (MGP) species. Thromb Haemost 2010;104:811822.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 48. Schurgers LJ, Barreto DV, Barreto FC, et al. The circulating inactive form of matrix gla protein is a surrogate marker for vascular calcification in chronic kidney disease: a preliminary report. Clin J Am Soc Nephrol 2010;5:568575.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 49. Miyata KN, Nast CC, Dai T, et al. Renal matrix Gla protein expression increases progressively with CKD and predicts renal outcome. Exp Mol Pathol 2018;105:120129.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 50. Actor JK, Hwang S-A, Kruzel ML. Lactoferrin as a natural immune modulator. Curr Pharm Des 2009;15:19561973.

Advertisement

Analysis of genes associated with proinflammatory and profibrotic pathways upregulated in ischemia-induced chronic kidney disease in cats

View More View Less
  • 1 From the Department of Small Animal Medicine and Surgery, College of Veterinary Medicine, University of Georgia, Athens, GA 30602
  • | 2 From the Department of Pathology, College of Veterinary Medicine, University of Georgia, Athens, GA 30602
  • | 3 From the Department of Physiology and Pharmacology, College of Veterinary Medicine, University of Georgia, Athens, GA 30602
  • | 4 From the Department of Plant Biology, Franklin College of Arts and Sciences, and Georgia Genomics and Bioinformatics Core, University of Georgia, Athens, GA 30602

Abstract

OBJECTIVE

To use RNA sequencing (RNAseq) to characterize renal transcriptional activities of genes associated with proinflammatory and profibrotic pathways in ischemia-induced chronic kidney disease (CKD) in cats.

SAMPLES

Banked renal tissues from 6 cats with experimentally induced CKD (renal ischemia [RI] group) and 9 healthy cats (control group).

PROCEDURES

Transcriptome analysis with RNAseq, followed by gene ontology and cluster analyses, were performed on banked tissue samples of the right kidneys (control kidneys) from cats in the control group and of both kidneys from cats in the RI group, in which unilateral (right) RI had been induced 6 months before the cats were euthanized and the ischemic kidneys (IKs) and contralateral nonischemic kidneys (CNIKs) were harvested. Results for the IKs, CNIKs, and control kidneys were compared to identify potential differentially expressed genes and overrepresented proinflammatory and profibrotic pathways.

RESULTS

Genes from the gene ontology pathways of collagen binding (eg, transforming growth factor-β1), metalloendopeptidase activity (eg, metalloproteinase [MMP]-7, MMP-9, MMP-11, MMP-13, MMP-16, MMP-23B, and MMP-28), chemokine activity, and T-cell migration were overrepresented as upregulated in tissue samples of the IKs versus control kidneys. Genes associated with the extracellular matrix (eg, TIMP-1, fibulin-1, secreted phosphoprotein-1, matrix Gla protein, and connective tissue growth factor) were upregulated in tissue samples from both the IKs and CNIKs, compared with tissues from the control kidneys.

CONCLUSIONS AND CLINICAL RELEVANCE

Unilateral ischemic injury differentially altered gene expression in both kidneys, compared with control kidneys. Fibulin-1, secreted phosphoprotein-1, and matrix Gla protein may be candidate biomarkers of active kidney injury in cats.

Abstract

OBJECTIVE

To use RNA sequencing (RNAseq) to characterize renal transcriptional activities of genes associated with proinflammatory and profibrotic pathways in ischemia-induced chronic kidney disease (CKD) in cats.

SAMPLES

Banked renal tissues from 6 cats with experimentally induced CKD (renal ischemia [RI] group) and 9 healthy cats (control group).

PROCEDURES

Transcriptome analysis with RNAseq, followed by gene ontology and cluster analyses, were performed on banked tissue samples of the right kidneys (control kidneys) from cats in the control group and of both kidneys from cats in the RI group, in which unilateral (right) RI had been induced 6 months before the cats were euthanized and the ischemic kidneys (IKs) and contralateral nonischemic kidneys (CNIKs) were harvested. Results for the IKs, CNIKs, and control kidneys were compared to identify potential differentially expressed genes and overrepresented proinflammatory and profibrotic pathways.

RESULTS

Genes from the gene ontology pathways of collagen binding (eg, transforming growth factor-β1), metalloendopeptidase activity (eg, metalloproteinase [MMP]-7, MMP-9, MMP-11, MMP-13, MMP-16, MMP-23B, and MMP-28), chemokine activity, and T-cell migration were overrepresented as upregulated in tissue samples of the IKs versus control kidneys. Genes associated with the extracellular matrix (eg, TIMP-1, fibulin-1, secreted phosphoprotein-1, matrix Gla protein, and connective tissue growth factor) were upregulated in tissue samples from both the IKs and CNIKs, compared with tissues from the control kidneys.

CONCLUSIONS AND CLINICAL RELEVANCE

Unilateral ischemic injury differentially altered gene expression in both kidneys, compared with control kidneys. Fibulin-1, secreted phosphoprotein-1, and matrix Gla protein may be candidate biomarkers of active kidney injury in cats.

Introduction

Chronic kidney disease occurs commonly in aged domestic cats.1–3 The term CKD defines an alteration of the renal function, structure, or both that has been present for several months, irrespective of the cause.4,5 Regardless of its primary underlying cause, CKD is irreversible and slowly progressive.4 Molecular mechanisms of the progression of CKD in models for people and rodents have been reported6–9; however, few studies10–12 have evaluated these mechanisms in cats.

Tubulointerstitial inflammation and fibrosis, typical histologic features of end-stage renal disease in people,13,14 are observed in most cats with CKD, including early stages of CKD.1,15 Prior studies16–21 assessed a limited number of specific profibrotic and proinflammatory mediators in cats with CKD. For example, compared with clinically normal cats, cats with CKD have higher urine concentrations of TGF-β1 and interleukin-8 and lower urine concentration of VEGF-A,18,19 and these findings suggest that these mediators may influence the progression of CKD.

Our group recently described a model of ischemia-induced CKD in cats in which a single 90-minute episode of transient, unilateral RI induces chronic tubulointerstitial inflammation and fibrosis,22,23 which are changes that mimic those of naturally occurring kidney disease in cats and people.1,15,24,25 Our findings indicate that there was an upregulation of transcription of specific profibrotic genes (including TGF-β1, MMP-2, MMP-7, MMP-9, and TIMP-1) and downregulation of VEGFA that persisted for 6 months following the ischemic episode.20 A subsequent study21 of cats with naturally occurring CKD shows similar patterns of differential regulation. Importantly, unilateral RI caused changes in gene transcription of TIMP-1 and VEGF-A that affected both kidneys.20 This information supports the role of renal hypoxia in the intrinsic progression of renal disease26–28 and the notion that acute kidney injury and CKD are interconnected syndromes.29,30

Comprehensive analysis of the renal transcriptome following RI may identify novel biomarkers, therapeutic targets, or both for cats with CKD.31 The aim of the present study was to use RNAseq to characterize renal transcriptional activities of genes associated with proinflammatory and profibrotic pathways in ischemia-induced CKD in cats. We hypothesized that renal tissues from cats with experimentally induced CKD would have greater genetic expression of molecular pathways associated with inflammation and fibrosis than would renal tissues from control cats and that the genetic expression of these pathways would differ for renal tissue samples obtained of both the IKs and CNIKs from cats with experimentally induced CKD versus those obtained from control cats.

Materials and Methods

Samples

We used banked tissue samples of left and right kidneys from cats that underwent 90 minutes of uni-lateral (right) RI 6 months before tissue collection (RI group; n = 6) and banked tissue samples of right kidneys (control kidneys) from healthy cats (control group; 9), of which all tissues samples except those from 1 control cat had been evaluated in previous studies.20,23 The kidney tissue samples from the cats in the RI group were further subgrouped on the basis of whether they were from the directly affected IK versus the CNIK.

Clinicopathologic analyses reflective of renal function (including serum and urine biochemical analyses) had been performed within 4 days before euthanasia; for analysis purposes, urine specific gravity measurements > 1.060 were assigned a value of 1.061. In those studies,20,23 the cats were euthanized by IV administration of pentobarbital, and the kidneys were harvested immediately thereafter. All kidneys were grossly and histologically evaluated by 2 board-certified veterinary pathologists (CAB and DRR), as described.20,23

Sample processing and RNAseq

Sample processing and total RNA extraction were performed as previously reported.20 Briefly, each kidney was sectioned, and then one-fourth to one-half was minced and immediately placed in RNA stabilization solution.a Following overnight incubation at 4°C, samples preserved in RNA stabilization solution were decanted, individually homogenized with the use of a mortar and pestle, divided into aliquots of 30 mg, and stored at −80°C until analysis. Total RNA was extracted from 1 aliquot of renal homogenate per sample with the use of a commercially available RNA extraction kit,b then quantified with a spectrophotometer.c Integrity of the isolated RNA was assessed first by visualization of 18S and 28S ribosomal bands on 1.2% agarose gels, then by analysis with a bioanalyzer systemd that assigns an RNA integrity number ranging from 1 to 10 to each sample. Samples with an RNA integrity number ≥ 5.5 were used in the RNAseq analysis.

The Georgia Genomics and Bioinformatics Core performed the RNAseq library preparation and the RNAseq. Briefly, total RNA samples were poly(A) enriched before cDNA synthesis to minimize rRNA in the final libraries. The cDNA samples were converted to barcoded, stranded RNAseq libraries with the use of a commercially available kite per the manufacturer's instructions. A quantitative PCR assay library kitf and fragment analysis kitg were used to assess the RNAseq libraries for concentration and size distribution, respectively. Equimolar amounts of all libraries were pooled in a final RNAseq pool. The concentration and size distribution of the final pool were assessed, and then samples were sequencedh with the use of a paired-end, 75-cycle sequencing protocol.

Statistical analyses

Statistical analyses of demographic and clinical data were performed with available software.i For each cat group (RI and control cats), distributions of data for continuous variables were examined for normality with the Shapiro-Wilk test and by visual assessment of histograms and normal Q-Q plots. Data were compared between groups with the Mann-Whitney U test.

Bioinformatic analyses— Sequencing data were demultiplexed with the use of available softwarej and on the basis of the sample-barcodes map to assign reads to their samples. The RNAseq reads from each sample were quality-trimmed with standard soft-warek to remove low-quality bases, homopolymers, and artifacts.

The trimmed RNAseq data were analyzed as described.32 Briefly, standard software was used to mapl quality reads to the reference feline genome,m then identify splice junctions and isoforms and develop the per-gene count matrix.n Next, 2 algorithmso,p were used as described33,34 to perform normalization and identify differentially expressed genes among the samples. In the differential expression analysis, the pairwise comparisons used were IK versus CNIK, IK versus control kidneys, and CNIK versus control kidneys. For each algorithm, the appropriate design matrix for each comparison was used to account for independence (control kidneys vs IKs or CNIKs) and nonindependence (IKs vs CNIKs) of the observations. To account for multiple testing errors, an FDR of ≤ 0.05 was used. Multiplicity correction was performed by applying the Benjamini-Hochberg method on values of P to control for the FDR.35,q

The resultant lists of significantly (FDR ≤ 0.05) upregulated genes from the pairwise comparisonso were used for gene ontology analysis with a classification systemr as described.36 A tool within the classification system was used to identify gene ontology classifications that were overrepresented in these lists of significantly upregulated genes, compared with the reference feline genome. This was performed with the use of the Fisher exact test to determine the probability that the number of genes observed in each category occurred by chance (randomly), as determined by the reference feline genome and by applying the Benjamini-Hochberg method to control for FDR < 0.05.

Coexpression clustering analysis— A single list without duplicative inclusion of significantly differentially expressed genes was compiled with the use of a union operation to combine the 2 gene sets yielded from the comparisons analyses algorithms.o,p Raw read counts were obtained from the RNAseq mapping data and normalized with the use of RNAseq library size and gene length count to obtain the transcripts-per-million–scaled matrix. Next, cluster analysis on the transcripts-per-million–scaled matrix was performed with an available software package,s as described,37 and because only a subset of genes was expected to have been assigned to biologically interpretable groups, the Tmax was used in this analysis to assess for cluster stability. Genes assigned to clusters with a Tmax value greater than the threshold (≥ 0.8) were considered and subsequently visualized, and the Tmax values (range, 0 to 1) implied the level of confidence (low to high, respectively) in the cluster membership. The gene coexpression clustering criteria parameters selected were a K-means algorithm for the cluster method, a cluster range to test of 2 to 25 clusters, and a transformation model of log centered log ratio; a minimum cutoff of 5 for a normalized count (removing genes with low expression from the analysis); the number of starts set at 1,000; and the iteration maximum of 10,000.

Results

Samples

Clinical information was collected from the medical records for cats in the RI group (6 spayed females) and the control group (9 sexually intact females). The median age and body weight were significantly (P = 0.002 and P = 0.006, respectively) greater for the RI group (520 days and 4.72 kg) versus the control group (270 days and 3.59 kg), and the median urine protein-to-creatinine ratio was significantly (P = 0.004) lower for the RI group (0.10; reference range, < 0.5) versus the control group (0.14). The median urine specific gravity and concentrations of serum creatinine and SUN did not substantially differ for cats in the RI group (1.056, 1.4 mg/dL, and 24 mg/dL, respectively) versus the control group (1.055, 1.0 mg/dL, and 28 mg/dL, respectively). Data for all 6 cats in the RI group and 8 of the 9 cats in the control group had been reported previously.20,23 The cats in the RI group had undergone ovariohysterectomy during the same anesthetic event as their surgically induced RI, and a progressive decline in glomerular filtration rate and an increase in serum creatinine concentration were documented for cats of the RI group during the original 6-month study.23 Additionally, histologic examination of the IKs from cats in the RI group revealed lesions that mimicked those of naturally occurring CKD (variable numbers of obsolescent glomeruli, tubulointerstitial inflammation, and fibrosis), whereas no abnormalities were identified on histologic reexaminations of the CNIKs from cats in the RI group or the kidneys from cats in the control group.

Differentially expressed genes

Adequate RNA integrity was confirmed for all but 1 CNIK sample; therefore, samples from 6 IKs, 5 CNIKs, and 9 control kidneys were evaluated. The differential expression analysis performed with the use of 2 different algorithms to conduct pairwise comparisons between IK, CNIK, and control kidneys revealed a set of 4,739o and a set of 5,529p differentially expressed genes.

In the data set obtained with the algorithmo preferred for experiments with fewer than 12 biological replicates,38 the IKs had 1,743 upregulated and 1,196 downregulated genes, compared with those of the control kidneys; the CNIK had 209 upregulated and 291 downregulated genes, compared with those of the control kidneys; and the IKs had 849 upregulated and 451 downregulated genes, compared with those of the CNIKs.

Findings from RNAseq indicated that the MMP-2 gene was upregulated in samples from IKs, compared with CNIKs; the MMP-7 and MMP-9 genes were up-regulated in samples from IKs, compared with both CNIKs and control kidneys; the TIMP-1 gene was up-regulated in samples from both the IKs and CNIKs, compared with control kidneys; and the TGF-β1 gene was upregulated in samples from IKs, compared with control kidneys. When these findings were compared with those of a previous study20 that used reverse transcription PCR assays to evaluate differential transcription of genes in tissue samples from these same kidneys, similar patterns of regulation were identified (Supplementary Table S1, available at: avmajournals.avma.org/doi/suppl/10.2460/ajvr.82.7.589).

Gene ontology analysis

Gene ontology analysis of the differentially expressed genes that were upregulated in the IKs versus control kidneys revealed an overrepresentation of several proinflammatory and profibrotic pathways (Table 1). When overrepresented and upregulated genes in the IKs, compared with control kidneys, were considered by gene ontology category (ie, biological processes, molecular function, or cellular component) and pathway, notable genes identified and related to biological processes included CXCL9, CXCL10, CXCL11, and CXCL16 (pathway, T-cell chemotaxis); T-cell–specific surface glycoproteins CD28 and CD5 (pathway, T-cell costimulation); TGF-β1 and TGF-β3, and vimentin (pathway, positive regulation of collagen biosynthetic process); CD74 (pathway, positive regulation of neutrophil chemotaxis); and platelet-derived growth factor receptor-α (pathway, platelet degranulation). Within the molecular function category, notable genes identified included TGF-β1, TGF-β2, TGF-β3, fibroblast growth factor-10 and -11, insulin-like growth factor-1, VEGF-C, hepatocyte growth factor, and macrophage colony-stimulating factor 1 (pathway, growth factor activity); MMP-7, MMP-9, MMP-11, MMP-13, MMP-16, MMP-23B, and MMP-28 (pathway, metalloendopeptidase activity); SPP1 (pathway, extracellular matrix binding); and connective tissue growth factor, fibronectin, vascular cell adhesion protein-1, and intercellular adhesion molecule-1 (pathway, integrin binding). As for the cellular component category, pertinent genes included TIMP-1, MGP, and fibulin-1 (pathway, extra-cellular matrix). Likewise, when overrepresented and upregulated genes in the IKs, compared with the CNIKs, were considered by gene ontology category and pathway, notable genes included MMP-2, MMP-7, MMP-9, MMP-13, and MMP-19 (pathway, metalloendopeptidase activity); SPP1 (pathway, extracellular matrix binding); and MGP (pathway, extracellular matrix; Table 2).

Table 1

Select overrepresented pathways, grouped by gene ontology category, that were upregulated in banked tissue samples of the IKs from 6 cats that had undergone 90 minutes of unilateral (right) RI to experimentally induce CKD 6 months before tissue collection (RI group), compared with results for tissue samples of the right kidneys (control kidneys) from 9 healthy cats (control group).

GO categoryOverrepresented pathway*Fold enrichmentRaw P valueFDR
Biological processDendritic cell chemotaxis (GO:0002407)11.551.70E-033.49E-02
T-cell chemotaxis (GO:0010818)10.831.39E-044.63E-03
Positive regulation of macrophage chemotaxis (GO:0010759)10.51.20E-055.59E-04
Platelet degranulation (GO:0002576)9.632.68E-034.91E-02
T-helper I-type immune response (GO:0042088)9.029.38E-042.23E-02
T-cell migration (GO:0072678)8.892.88E-051.19E-03
T-cell costimulation (GO:0031295)8.021.38E-032.94E-02
Positive regulation of collagen biosynthetic process (GO:0032967)7.221.96E-033.86E-02
Positive regulation of epithelial to mesenchymal transition (GO:0010718)6.082.22E-046.81E-03
Positive regulation of granulocyte chemotaxis (GO:0071624)5.956.13E-041.58E-02
Positive regulation of neutrophil chemotaxis (GO:0090023)5.956.13E-041.57E-02
Regulation of T-cell proliferation (GO:0042129)4.291.55E-071.13E-05
T-cell activation (GO:0042110)4.283.87E-II4.88E-09
Positive regulation of angiogenesis (GO:0045766)3.082.26E-046.87E-03
Molecular functionToll-like receptor binding (GO:0035325)9.632.20E-041.90E-02
Collagen binding (GO:0005518)6.85.28E-081.96E-05
Chemokine binding (GO:0019956)6.743.38E-042.41E-02
Extracellular matrix binding (GO:0050840)5.873.45E-066.09E-04
Integrin binding (GO:0005178)4.383.10E-066.38E-04
Chemokine activity (GO: 0008009)4.512.74E-042.12E-02
Heparin binding (GO:0008201)3.512.14E-052.65E-03
Growth factor activity (GO:0008083)2.942.05E-052.62E-03
Metalloendopeptidase activity (GO:0004222)2.495.88E-043.89E-02
Cellular componentT-cell receptor complex (GO:0042101)9.632.20E-046.49E-03
Integrin complex (GO:0008305)7.475.28E-082.50E-06
Major histocompatibility complex class II protein complex (GO:0042613)7.226.89E-041.84E-02
Extracellular matrix (GO:0031012)3.689.26E-213.83E-18

Reported as the pathway and the 7-digit unique identifier prefix with GO.

The raw P value was determined with the Fisher exact test.

Reported as the exponential notation for powers of 10 (eg, 1.70E-4 means 1.70 × 10−4).

E = Exponential notation. GO = Gene ontology.

Table 2

Select overrepresented pathways, grouped by gene ontology category, that were upregulated in tissue samples of the IKs (n = 6) versus the CNIKs (5) from the 6 cats in the IR group described in Table 1. Adequate RNA integrity was not confirmed for the remaining CNIK sample; therefore, it was not included in the analysis.

GO categoryOverrepresented pathway*Fold enrichmentRaw P value†‡FDR‡
Biological processEstablishment of T-cell polarity (GO:0001768)21.921.14E–034.04E-02
Positive regulation of interleukin-2 biosynthetic process (GO:0045086)16.73.26E-041.49E-02
T-cell costimulation (GO:0031295)16.246.10E-053.82E-03
T-cell chemotaxis (GO:0010818)14.614.76E-042.03E-02
Positive regulation of macrophage chemotaxis (GO:0010759)10.631.21E-034.16E-02
Positive regulation of neutrophil migration (GO:1902624)10.318.00E-054.78E-03
Positive regulation of granulocyte chemotaxis (GO:0071624)10.318.00E-054.76E-03
Positive regulation of interleukin-10 production (GO:0032733)10.318.00E-054.74E-03
Platelet activation (GO:0030168)5.844.14E-041.83E-02
Molecular functionCytokine receptor activity (GO:0004896)6.945.23E-106.47E-07
Cellular componentT-cell receptor complex (GO:0042101)16.246.10E-053.17E-03
Major histocompatibility complex class II protein complex (GO:0042613)14.611.69E-051.04E-03
Integrin complex (GO:0008305)6.059.18E-043.04E-02
Extracellular matrix (GO:0031012)3.181.09E-082.26E-06

See Table 1 for key.

For the CNIKs versus control kidneys, there were fewer differentially expressed genes. Results of gene ontology analysis indicated that overrepresented pathways pertained, once again, to the regulation of extracellular matrix deposition (Table 3). Notable genes included in the overrepresented pathways included TIMP-1, TIMP-2, connective tissue growth factor, thrombospondin-1, MGP, MMP-28, and fibulin-1 (cellular component category; pathway, extracellular matrix).

Table 3

Select overrepresented pathways upregulated in tissue samples of the CNIKs (n = 5) versus the control kidneys described in Tables 1 and 2.

GO categoryOverrepresented pathway*Fold enrichmentRaw P value†‡FDR‡
Biological processRegulation of cell migration (GO:0030334)4.191.94E-055.04E-02
Molecular functionProtein complex binding (GO:0032403)3.862.23E-054.14E-02
Cellular componentExtracellular matrix (GO:0031012)6.641.63E-092.70E-06

See Table 1 for key.

Gene coexpression clustering analysis

A union list of 4,801 differentially expressed genes was compiled. After analysis with various transformation and cluster models, the selected number of clusters was 20, which explained most of the variances among samples (Figure 1). Genes of interest, including SPP1 and MGP, were identified in cluster 2 (Table 4) and cluster 10 (Table 5), respectively, with greater expression detected in tissue samples from IKs versus CNIKs or control kidneys and in samples from CNIKs versus control kidneys.

Figure 1
Figure 1

Cluster plots of coexpressed genes in banked tissue samples of the IKs (n = 6) and CNIKs (5) from 6 cats that had undergone 90 minutes of unilateral (right) RI to experimentally induce CKD 6 months before tissue collection (RI group) and in banked tissue samples of the right kidneys (control kidneys; 9) from 9 healthy cats (control group), with each plot representing 1 of the 20 gene clusters that explained most of the variances among samples. In each plot, the y-axis represents transformed, normalized gene expression, the x-axis represents the 20 individual renal tissue samples (CNIKs, sample Nos. 1 through 5; IKs, sample Nos. 6 through 11; and control kidneys, sample Nos. 12 through 20), the red lines represent genes that had a cluster membership Tmax value < 0.8, and the black lines represent genes that had a cluster membership Tmax value ≥ 0.8. Adequate RNA integrity was not confirmed for the remaining CNIK sample; therefore, it was not included in the analysis.

Citation: American Journal of Veterinary Research 82, 7; 10.2460/ajvr.82.7.589

Table 4

List of genes identified in cluster 2 as coexpressed in tissue samples of the IKs (n = 6), CNIKs (5), and control kidneys (9) described in Tables 1 and 2.

Gene symbol or locusDescription*
SPP1Secreted phosphoprotein-1
PLET1Placenta expressed transcript 1
LOC109493266Undetermined
LOC105261057Undetermined
LOC105260308Undetermined
LOC105260274Undetermined
LOC105259930Undetermined
LOC101098498Predicted: Neutrophil gelatinase-associated lipocalin-like
LOC101090949Undetermined
LOC101086711Undetermined
JCHAINJoining chain of multimeric IgA and IgM

Genes described as undetermined were unknown in the reference annotation.

Table 5

List of genes identified in cluster 10 as coexpressed in tissue samples of the IKs (n = 6), CNIKs (5), and control kidneys (9) described in Tables 1 and 2.

Gene symbol or locationDescription*
CCL21C-C motif chemokine ligand 21
CD53Leukocyte surface antigen CD53
CD74Leukocyte surface antigen CD74
CFDComplement factor D
CH13L1Chitinase 3 like I
CLDN3Claudin 3
CLEC2DC-type lectin domain family 2 member D
CORO1ACoronin IA
DRAHLA class II histocompatibility antigen, DR a chain-like
IL7RInterleukin 7 receptor
IRF8Interferon regulatory factor 8
KRT7Keratin 7
LCP1Lymphocyte cytosolic protein I
LOC101098301DLA class II histocompatibility antigen, DR-I P chain
LOC102899262Undetermined
LOC109494204Undetermined
LTFLactotransferrin
MFAP4Microfibril-associated protein 4
MGPMatrix Gla protein
P116Peptidase inhibitor 16
P13Peptidase inhibitor 3
PTPRCProtein tyrosine phosphatase, receptor type C
SLC26A3Solute carrier family 26 member 3
SLC34A2Solute carrier family 34 member 2
TACSTD2Tumor-associated calcium signal transducer 2

Genes described as undetermined were unknown in the reference annotation.

DLA = Dog leukocyte antigen. HLA = Human leukocyte antigen.

Discussion

The results of the present study identified an extensive list of proinflammatory and profibrotic pathways upregulated in tissue samples of the kidneys from cats that 6 months previously underwent transient, unilateral RI to experimentally induce CKD. Although 6 months had elapsed between the ischemic event and harvest of renal tissues, a union set of 4,801 genes, many of which are involved in the regulation of inflammatory cell chemotaxis, cytokine and growth factor activities, and deposition of extracellular matrix, was identified as differentially expressed in pairwise comparisons of results for the IKs, CNIKs, and control kidneys. Although the direct ischemic injury was induced in only the right kidneys (the IKs) of the cats in the RI group, proinflammatory and profibrotic pathways were also upregulated in the CNIKs, compared with results for control kidneys. Findings of the present study together with those of a prior study,20 support acute kidney injury and CKD as closely related syndromes.29,30 Further, the present study identified additional cytokines that may be implicated in the progression of CKD in cats.

Fibulin-1, an extracellular matrix protein prominently expressed in blood vessels,39 was upregulated in renal tissue samples from cats in the RI group versus the control group. Relatedly, in people, high plasma fibulin-1 concentration has been detected in individuals with various types of renal disease (glomerulonephritis, diabetic nephropathy, obstructive uropathy, and analgesic abuse nephropathy); therefore, this protein has been suggested as a marker for renal impairment.40 Also in people, a study41 shows that plasma concentrations of fibulin-1 inversely correlate with glomerular filtration rate and positively correlate with systemic blood pressure and other hemodynamic markers of cardiovascular risk in CKD and diabetes. Investigation of fibulin-1 plasma concentration is warranted to assess this protein's potential as a biomarker for renal disease in cats.

In the present study, SPP1, which is an extracellular matrix protein and proinflammatory cytokine also known as osteopontin, was upregulated in both kidneys from the RI group, compared with control kidneys. In rodents and people, SPP1 is highly expressed in renal tubular epithelial cells and has been suggested as an important factor in the initial inflammatory responses involving natural killer cell activity in RI reperfusion injury.42 In experimental models of glomerulonephritis in rodents, expression of SPP1 correlates with the severity of the associated tubulointerstitial injury.43 Findings in people and rodents suggest that SPP1 also contributes to the pathogenesis of diabetic nephropathy.44 In people with diabetes, the plasma concentration of SPP1 is independently associated with the presence and severity of nephropathy and coronary heart disease.45 Although data regarding SPP1 in cats with CKD is lacking, gene coexpression clustering analysis in the present study revealed that the SPP1 gene is coexpressed with a predicted locus for neutrophil gelatinase–associated lipocalin. Because urinary neutrophil gelatinase–associated lipocalin has gained attention as a marker for the prediction of progression of CKD in cats,46 SPP1 may similarly offer promise as a biomarker for progressive CKD in cats.

Our results indicated that MGP, an important vitamin K–dependent local inhibitor of vascular calcification,47 was also upregulated in tissue samples from the IKs and CNIKs, compared with those from the control kidneys. Circulating concentrations of an inactive form of MGP were positively and independently associated with aortic calcification in people with various stages of CKD and proposed as a biomarker of vascular calcification in people with renal disease.48 Also in people, expression of MGP correlates with interstitial fibrosis, tubular atrophy, acute tubular injury, and interstitial inflammation and is positively associated with an increased risk of end-stage renal disease.49 In our study, MGP was coexpressed with several genes that encode for proteins involved in inflammatory processes, such as lactotransferrin (a mediator of immune response to injury),50 interleukin-7 receptors, and interferon regulatory factor 8; therefore, further investigation is warranted regarding the potential use of MGP as a marker for ongoing renal inflammation and fibrosis in cats.

There were limitations to the present study. First, because banked tissues were available from only a small number of cats with experimentally induced CKD,23 the number of biological replicates in each group was relatively small. Second, a sham-operated control group was not available. Therefore, it was possible that some changes identified for the CNIKs, compared with the control kidneys, may have been from global decreases in renal perfusion under anesthesia, rather than a result of the intentional ischemia induced in the IK. Additionally, cats in the RI group were older and all spayed, whereas the control group consisted of younger and sexually intact female cats. Such differences in age and reproductive status may have influenced gene transcription results; however, it appeared unlikely that the transcriptional regulation of renal inflammation and fibrosis would have been impacted by these factors, particularly given the fact that the upregulation of several of the identified proinflammatory and profibrotic pathways was for the IKs versus the CNIKs from cats in the RI group. Overall, the cats from which the renal tissues were obtained were young, and the question remains as to whether aged cats would have similar patterns of transcriptional activities following RI. Another limitation was that because tissues were collected at a single, relatively late, time point after the ischemic event, potential transient changes in the transcription of genes that promote inflammation and fibrosis earlier in the renal inflammatory processes may not have been detected. Further, the present study was at the transcriptional level and may not have reflected changes in protein expression in the kidneys or changes that could occur in the kidneys of cats with naturally occurring CKD.

Findings from the present study supported our hypotheses that renal tissues from cats with experimentally induced CKD would have greater genetic expression of molecular pathways associated with inflammation and fibrosis than would renal tissues from control cats and that the genetic expression of these pathways would differ for tissue samples obtained from both the IKs and CNIKs from cats with experimentally induced CKD versus those from control cats. Additionally, our results suggested that acute, transient ischemic renal injury leads to persistent and wide-ranging changes in transcription of those genes that promote inflammation and fibrosis. Fibulin-1, SPP1, and MGP may prove to be useful biomarkers of active kidney injury in cats, as they are in people. Future evaluation of fibulin-1, SPP1, and MGP in renal tissues, plasma, or both from cats with spontaneous CKD is warranted to explore their potential as renal biomarkers in cats. Additionally, future prospective studies may include a larger cohort, age- and gender-matched controls, and samples collected at multiple and earlier time points following RI.

Acknowledgments

This work was performed at the University of Georgia College of Veterinary Medicine and was internally funded. Dr. Bianca Lourenço was the recipient of a Boehringer Ingelheim Postdoctoral Scholarship.

The authors declare that there were no conflicts of interest. Presented in part as an oral abstract at the 28th European College of Veterinary Internal Medicine - Companion Animals Annual Congress, Rotterdam, Netherlands, September 2018.

Abbreviations

CKD

Chronic kidney disease

CNIK

Contralateral nonischemic kidney

CXCL

C-X-C motif chemokine ligand

FDR

False discovery rate

IK

Ischemic kidney

MGP

Matrix Gla protein

MMP

Matrix metalloproteinase

RI

Renal ischemia

RNAseq

RNA sequencing

SPP1

Secreted phosphoprotein-1

TGF-β

Transforming growth factor-β

TIMP

Tissue inhibitor of metalloproteinase

Tmax

Maximum conditional probability

VEGF

Vascular endothelial growth factor

Footnotes

a.

RNAlater stabilization solution, Quiagen, Valencia, Calif.

b.

RNeasy Plus Mini Kit, Quiagen, Valencia, Calif.

c.

NanoDrop spectrophotometer, Thermo Fisher Scientific, Waltham, Mass.

d.

2100 Bioanalyzer system, Agilent Technologies, Palo Alto, Calif.

e.

KAPA mRNA HyperPrep Kit, Kapa Biosystems, Woburn, Mass.

f.

KAPA qPCR library quantification kit, Kapa Biosystems, Wo-burn, Mass.

g.

High Sensitivity NGS Fragment Analysis kit, Advanced Analytical Technologies, Ankeny, Iowa.

h.

NextSeq 500 platform, Illumina Inc, San Diego, Calif.

i.

JMP Pro, version 14.1.0, SAS Institute Inc, Cary, NC.

j.

bcl2fastq, version 2, Illumina Inc, San Diego, Calif.

k.

Trimmomatic, version 0.36, Bolger AM, Lohse M, Usadel B. Available at: usadellab.org/cms/index.php?page=trimmomatic. Accessed Apr 15, 2017.

l.

TopHat, version 2.1.1, Center for Computational Biology, Johns Hopkins University, Baltimore, Md. Available at: ccb.jhu.edu/software/tophat/index.shtml.

m.

Felis_catus_8.0, GCF_000181335.2_Felis_catus_8.0_ge nomic.gff. Available at: www.ncbi.nlm.nih.gov/assembly/GCF_000181335.2/. Accessed Apr 15, 2017.

n.

Cufflinks, version 2.2.1. Available at: cole-trapnell-lab.github.io/cufflinks/. Accessed Apr 15, 2017.

o.

edgeR: Empirical analysis of digital gene expression data in R, R package, version 3.14, Chen Y, Lun A, McCarthy D, et al. Available at: bioconductor.org/packages/release/bioc/html/edgeR.html. Accessed Apr 15, 2017.

p.

DESeq2: Differential gene expression analysis based on the negative binomial distribution, R package, version 1.30.0, Love M, Anders S, Huber W. Available at: bioconductor.org/packages/release/bioc/html/DESeq2.html. Accessed Apr 15, 2017.

q.

P.adjust function, version 3.3.2, R Core Team, R Foundation for Statistical Computing, Vienna, Austria. Available at: www.R-project.org/. Accessed Apr 15, 2017.

r.

PANTHER Classification System. Available at: go.pantherdb.org/. Accessed Mar 14, 2018.

s.

CoSeq: Co-expression analysis of sequencing data, version 1.5.1, Rau A, Maugis-Rabusseau C, Godichon-Baggioni A. Available at: bioconductor.org/packages/release/bioc/html/coseq.html. Accessed Aug 8, 2018.

References

  • 1. Brown CA, Elliott J, Schmiedt CW, et al. Chronic kidney disease in aged cats clinical features, morphology, and proposed pathogeneses. Vet Pathol 2016;53:309326.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 2. Lulich J, Osborne C, O'Brien T, et al. Feline renal failure: questions, answers, questions. Compend Contin Educ Pract Vet 1992;14:127153.

    • Search Google Scholar
    • Export Citation
  • 3. Marino CL, Lascelles BDX, Vaden SL, et al. Prevalence and classification of chronic kidney disease in cats randomly selected from four age groups and in cats recruited for degenerative joint disease studies. J Feline Med Surg 2014;16:465472.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 4. Bartges JW. Chronic kidney disease in dogs and cats. Vet Clin North Am Small Anim Pract 2012;42:669692.

  • 5. Levey AS, Eckardt K-U, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005;67:20892100.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 6. Eddy AA. Overview of the cellular and molecular basis of kidney fibrosis. Kidney Int Suppl (2011) 2014;4:28.

  • 7. Fogo AB. Mechanisms of progression of chronic kidney disease. Pediatr Nephrol 2007;22:20112022.

  • 8. Basile DP, Bonventre JV, Mehta R, et al. Progression after AKI: understanding maladaptive repair processes to predict and identify therapeutic treatments. J Am Soc Nephrol 2016;27:687697.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 9. Gewin L, Zent R, Pozzi A. Progression of chronic kidney disease: too much cellular talk causes damage. Kidney Int 2017;91:552560.

  • 10. Lawson J, Elliott J, Wheeler-Jones C, et al. Renal fibrosis in feline chronic kidney disease: known mediators and mechanisms of injury. Vet J 2015;203:1826.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 11. Sánchez-Lara AC, Elliott J, Syme HM, et al. Feline chronic kidney disease is associated with upregulation of transglutaminase 2: a collagen cross-linking enzyme. Vet Pathol 2015;52:513523.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12. Lawson JS, Syme HM, Wheeler-Jones CPD, et al. Urinary active transforming growth factor β in feline chronic kidney disease. Vet J 2016;214:16.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 13. Mimura I, Nangaku M. The suffocating kidney: tubulointerstitial hypoxia in end-stage renal disease. Nat Rev Nephrol 2010;6:667678.

  • 14. Hewitson TD. Renal tubulointerstitial fibrosis: common but never simple. Am J Physiol Renal Physiol 2009;296:F1239F1244.

  • 15. McLeland SM, Cianciolo RE, Duncan CG, et al. A comparison of biochemical and histopathologic staging in cats with chronic kidney disease. Vet Pathol 2015;52:524534.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 16. van Beusekom CD, Zimmering TM. Profibrotic effects of angiotensin II and transforming growth factor beta on feline kidney epithelial cells. J Feline Med Surg 2019;21:780787.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 17. Lawson JS, Liu H-H, Syme HM, et al. The cat as a naturally occurring model of renal interstitial fibrosis: characterisation of primary feline proximal tubular epithelial cells and comparative pro-fibrotic effects of TGF-β1. PLoS One 2018;13:e0202577.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18. Arata S, Ohmi A, Mizukoshi F, et al. Urinary transforming growth factor-β1 in feline chronic renal failure. J Vet Med Sci 2005;67:12531255.

  • 19. Habenicht LM, Webb TL, Clauss LA, et al. Urinary cytokine levels in apparently healthy cats and cats with chronic kidney disease. J Feline Med Surg 2013;15:99104.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20. Lourenço BN, Coleman AE, Schmiedt CW, et al. Profibrotic gene transcription in renal tissues from cats with ischemia-induced chronic kidney disease. Am J Vet Res 2020;81:180189.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21. Lourenço BN, Coleman AE, Tarigo JL, et al. Evaluation of profibrotic gene transcription in renal tissues from cats with naturally occurring chronic kidney disease. J Vet Intern Med 2020;34:14761487.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22. Schmiedt CW, Brainard BM, Hinson W, et al. Unilateral renal ischemia as a model of acute kidney injury and renal fibrosis in cats. Vet Pathol 2016;53:87101.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23. Brown CA, Rissi DR, Dickerson VM, et al. Chronic renal changes after a single ischemic event in an experimental model of feline chronic kidney disease. Vet Pathol 2019;56:536543.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 24. DiBartola SP, Rutgers HC, Zack PM, et al. Clinicopathologic findings associated with chronic renal disease in cats: 74 cases (1973–1984). J Am Vet Med Assoc 1987;190:11961202.

    • Search Google Scholar
    • Export Citation
  • 25. Kuncio GS, Neilson EG, Haverty T. Mechanisms of tubulointerstitial fibrosis. Kidney Int 1991;39:550556.

  • 26. Fine LG, Bandyopadhay D, Norman JT. Is there a common mechanism for the progression of different types of renal diseases other than proteinuria? Towards the unifying theme of chronic hypoxia. Kidney Int Suppl 2000;75:S22S26.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27. Fine LG, Norman JT. Chronic hypoxia as a mechanism of progression of chronic kidney diseases: from hypothesis to novel therapeutics. Kidney Int 2008;74:867872.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28. Palm F, Nordquist L. Renal tubulointerstitial hypoxia: cause and consequence of kidney dysfunction. Clin Exp Pharmacol Physiol 2011;38:474480.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29. Cowgill LD, Polzin DJ, Elliott J, et al. Is progressive chronic kidney disease a slow acute kidney injury? Vet Clin North Am Small Anim Pract 2016;46:9951013.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30. Kaballo MA, Elsayed ME, Stack AG. Linking acute kidney injury to chronic kidney disease: the missing links. J Nephrol 2017;30:461475.

  • 31. Cisek K, Krochmal M, Klein J, et al. The application of multi-omics and systems biology to identify therapeutic targets in chronic kidney disease. Nephrol Dial Transplant 2016;31:20032011.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 32. Ghosh S, Chan CK. Analysis of RNA-seq data using TopHat and Cufflinks. Methods Mol Biol 2016;1374:339361.

  • 33. Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor package for differential expression analysis of digital gene expression data. Bioinformatics 2010;26:139140.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34. Love M, Anders S, Huber H. Differential analysis of count data—the DESeq2 package. Available at: bioconductor.jp/packages/2.14/bioc/vignettes/DESeq2/inst/doc/DESeq2.pdf. Accessed May 6, 2021.

    • Search Google Scholar
    • Export Citation
  • 35. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc Series B Stat Methodol 1995;57:289300.

    • Search Google Scholar
    • Export Citation
  • 36. Mi H, Lazareva-Ulitsky B, Loo R, et al. The PANTHER database of protein families, subfamilies, functions and pathways. Nucleic Acids Res 2005;33(suppl 1):D284D288.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 37. Rau A, Maugis-Rabusseau C. Transformation and model choice for RNA-seq co-expression analysis. Brief Bioinform 2018;19:425436.

  • 38. Schurch NJ, Schofield P, Gierliński M, et al. Evaluation of tools for differential gene expression analysis by RNA-seq on a 48 biological replicate experiment. arXiv 2015;1505.02017. doi:10.1261/rna.053959.115.

    • Search Google Scholar
    • Export Citation
  • 39. Argraves WS, Greene LM, Cooley MA, et al. Fibulins: physiological and disease perspectives. EMBO Rep 2003;4:11271131.

  • 40. Neiman M, Hedberg JJ, Dönnes PR, et al. Plasma profiling reveals human fibulin-1 as candidate marker for renal impairment. J Proteome Res 2011;10:49254934.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 41. Scholze A, Bladbjerg E-M, Sidelmann JJ, et al. Plasma concentrations of extracellular matrix protein fibulin-1 are related to cardiovascular risk markers in chronic kidney disease and diabetes. Cardiovasc Diabetol 2013;12:6.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 42. Zhang ZX, Shek K, Wang S, et al. Osteopontin expressed in tubular epithelial cells regulates NK cell-mediated kidney ischemia reperfusion injury. J Immunol 2010;185:967973.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 43. Pichler R, Giachelli CM, Lombardi D, et al. Tubulointerstitial disease in glomerulonephritis. Potential role of osteopontin (uropontin). Am J Pathol 1994;144:915926.

    • Search Google Scholar
    • Export Citation
  • 44. Kahles F, Findeisen HM, Bruemmer D. Osteopontin: a novel regulator at the cross roads of inflammation, obesity and diabetes. Mol Metab 2014;3:384393.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 45. Yan X, Sano M, Lu L, et al. Plasma concentrations of osteopontin, but not thrombin-cleaved osteopontin, are associated with the presence and severity of nephropathy and coronary artery disease in patients with type 2 diabetes mellitus. Cardiovasc Diabetol 2010;9:70.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 46. Wang IC, Hsu WL, Wu PH, et al. Neutrophil gelatinase associated lipocalin in cats with naturally occurring chronic kidney disease. J Vet Intern Med 2017;31:102108.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 47. Cranenburg EC, Koos R, Schurgers LJ, et al. Characterisation and potential diagnostic value of circulating matrix Gla protein (MGP) species. Thromb Haemost 2010;104:811822.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 48. Schurgers LJ, Barreto DV, Barreto FC, et al. The circulating inactive form of matrix gla protein is a surrogate marker for vascular calcification in chronic kidney disease: a preliminary report. Clin J Am Soc Nephrol 2010;5:568575.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 49. Miyata KN, Nast CC, Dai T, et al. Renal matrix Gla protein expression increases progressively with CKD and predicts renal outcome. Exp Mol Pathol 2018;105:120129.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 50. Actor JK, Hwang S-A, Kruzel ML. Lactoferrin as a natural immune modulator. Curr Pharm Des 2009;15:19561973.

Supplementary Materials

Contributor Notes

Address correspondence to Dr. Lourenço (lourenco@uga.edu).