Studies in mouse models of kidney disease have identified homeo-domain interacting protein kinase 2 (HIPK2) as a key regulator of kidney fibrosis, and a potential drug target for future human therapeutics. By combining genetic analysis of a mouse model of HIV-associated nephropathy (HIVAN) with computational systems biology approaches, a multidisciplinary team in the U.S. and China identified HIPK2 upregulation in damaged mouse kidneys, and went on to confirm that increased levels of the protein are present in human patients with various kidney diseases.
Notably, the Mount Sinai School of Medicine (MSSM)-led team showed that knocking out HIPK2 in the mouse HIVAN model led to significantly improved renal function, and reduced proteinuria and kidney fibrosis. John Cijiang and colleagues describe their findings in Nature Medicine. Their paper is titled “A systems approach identifies HIPK2 as a key regulator of kidney fibrosis.”
The Tg26 mouse model develops kidney dysfunction that mimics the pathological renal changes in human kidney disease. Using this model, researchers at the MSSM’s department of nephrology carried out an integrated experimental and computational approach to identify and rank protein kinases that may regulate known changes in mRNA expression and transcription factor activation in this model.
They determined which genes were differentially expressed in Tg26 and wild-type mice, and then used the computational methods TRANSFAC matrices and chromatin immunoprecipitation (ChIP) Enrichment Analysis (ChEA) to predict which upstream transcription factors were most likely to be responsible. Protein-DNA interactions that were differentially regulated between Tg26 and wild-type kidneys were then profiled using a protein-DNA binding array.
Further analyses and data from protein-DNA interaction arrays for top-ranking transcription factors resulted in the identification of 18 transcription factors that could potentially be responsible for the pattern of kidney gene expression in Tg26 mice.
The next stage was to link the transcription factors to upstream regulatory pathways by constructing a protein-protein interaction subnetwork based on known interactions, and link in to this protein kinases believed to phosphorylate proteins within the subnetwork. Of these kinases, HIPK2 had 14 substrates within the network, representing about 50% of all the known substrates for HIPK2.
The researchers therefore focused more closely on HIPK2, and examined its expression in Tg26 kidneys. Immunostaining and western blot analyses indicated that the gene is expressed primarily in the kidney tubule region, rather than the glomeruli, and that HIPK2 protein expression is regulated at the post-translational level. Interestingly, infecting primary human renal tubular epithelial cells (hRTECs) with an engineered HIV virion led to increased HIPK2 protein levels, but not HIPK2 mRNA levels. Treating the cells with a proteosome inhibitor further increased HIPK2 expression, suggesting that HIV infection increases HIPK2 activity by reducing HIPK2 proteasomal degradation.
HIV infection promotes DNA damage in kidney cells through oxidative stress, and prior work has shown that DNA damage increases HIPK2 protein concentrations by inhibiting the ubiquitin ligase SIAH1. Given the fact that SIAH1 gene expression was reduced the kidney gene expression microarray data from Tg26 mice, the MSSM investigators looked at whether whether HIV-induced upregulation of HIPK2 in the kidneys of Tg26 also resulted from reduced SIAH1 concentrations.
Western blot analyses confirmed there was an inverse relationship between the levels of SIAH1 and HIPK2 protein levels. Moreover, infection of human RTECs by HIV suppressed SIAH1 protein and SIAH1 mRNA expression. Transfecting hRTECs with an SIAH1 expression construct led to increased HIPK2 protein levels, while transfecting cells with an SIAH1 gene-silencing siRNA resulted in reduced HIPK2 levels, “indicating that SIAH1 acts upstream of HIPK2,” the authors note.
Importantly, the team’s in vitro findings were confirmed by examining the renal expression of SIAH1 and HIPK2 in different human kidney diseases including HIVAN, FSGS, diabetic nephropathy, and severe IgA nephropathy (IgAN). Upregulation of HIPK2 in these tissues was associated with tubulointerstitial injury or fibrosis and glomerulosclerosis, whereas in patients with minimal change disease, which didn’t display upregulated HIPK2, there was no tubulointerstitial injury or fibrosis and glomerulosclerosis. Similarly, SIAH1 staining was lower in the tubules of diseased kidneys than in those of normal kidneys and was inversely related to the intensity of HIPK2 staining in all groups.
Further studies demonstrated that HIPK2 overexpression led to the increased RTEC apoptosis, which is a characteristic of tubulointerstitial injury in HIVAN. Conversely, cells overexpressing a kinase activity-defective HIPK2 demonstrated far lower levels of HIV-induced apoptosis. In addition overexpression of HIPK2 or TGF-β (a mediator of epithelial-to-mesenchymal transition [EMT] with which HIPK2 interacts), induced the expression of EMT markers. Conversely, overexpression of a kinase activity-deficient HIPK2 diminished the expression of EMT markers in cells infected with HIV or treated with TGF-β, when compared with cells that didn’t overexpress KD-HIPK2.
The researchers then generated HIPK2-knockout mice that still expressed the HIV transgene. Unlike Tg26 mice that typically develop proteinuria at four weeks of age, the HIPK2 knockout Tg26 animals demonstrated lower proteinuria, lower levels of the renal function marker serum urea nitrogen, reduced tubulointerstitial injury and fibrosis, and reduced podocyte hyperplasia and glomerulosclerosis. Importantly, the expression of EMT markers were elevated in Tg26 mice but know the KO-Tg26 animals, compared with wild-type mice.
Finally, the MSSM investigators tested whether HIPK2 was upregulated in other models of renal fibrosis, including a unilateral ureteral obstruction (UUO) model, and a folic acid-induced renal fibrosis model. As expected, HIPK2 was higher in the tubulointerstitium of wild-type UUO kidneys compared to the sham-operated wild-type kidneys, sham-operated KO-HIPK2 kidneys, or UUO KO-HIPK2 kidneys. Fibrosis was also increased in wild-type UUO kidneys, but not KO-HIPK2 kidneys, while SIAH1 expression was lower and HIPK2/EMT marker expression higher in wild-type-UUO kidneys than wild-type sham-operated kidneys.
Similarly, KO-HIPK2 mice were protected from folic-acid–induced kidney fibrosis, and in these animals serum urea nitrogen concentrations and the renal expression of EMT markers were also reduced following treatment with folic acid, when compared with wild-type mice treated with folic acid.
The authors believe that HIPK2 could represent a new therapeutic target for kidney disease, especially given that protein kinases are druggable targets. “We present here a systems approach to identifying upstream protein kinases based on genome-wide mRNA expression microarrays and DNA-protein arrays,” they state. “We identified and confirmed that HIPK2, a protein kinase previously unrecognized in kidney disease, has a crucial role in renal fibrosis. We elucidated the regulation of HIPK2 in kidney disease and in downstream signaling pathways that mediate HIPK2-induced apoptosis, as well as the expression of EMT markers in kidney cells.”