Role of the CDKL1-SOX11 signaling axis in acute kidney injury

Abstract

The biology of the cyclin-dependent kinase-like (CDKL) kinase family remains enigmatic. Contrary to their nomenclature, CDKLs do not rely on cyclins for activation and are not involved in cell cycle regulation. Instead, they share structural similarities with mitogen-activated protein kinases and glycogen synthase kinase-3, although their specific functions and associated signaling pathways are still unknown. Previous studies have shown that the activation of CDKL5 kinase contributes to the development of acute kidney injury (AKI) by suppressing the protective SOX9-dependent transcriptional program in tubular epithelial cells. In the current study, we measured the functional activity of all five CDKL kinases and discovered that, in addition to CDKL5, CDKL1 is also activated in tubular epithelial cells during AKI. To explore the role of CDKL1, we generated a germline knockout mouse that exhibited no abnormalities under normal conditions. Notably, when these mice were challenged with bilateral ischemia-reperfusion and rhabdomyolysis, they were found to be protected from AKI. Further mechanistic investigations revealed that CDKL1 phosphorylates and destabilizes SOX11, contributing to tubular dysfunction. In summary, this study has unveiled a previously unknown CDKL1-SOX11 axis that drives tubular dysfunction during AKI.

NEW & NOTEWORTHY Identifying and targeting pathogenic protein kinases holds potential for drug discovery in treating acute kidney injury. Our study, using novel germline knockout mice, revealed that Cdkl1 kinase deficiency does not affect mouse viability but provides protection against acute kidney injury. This underscores the importance of Cdkl1 kinase in kidney injury and supports the development of targeted small-molecule inhibitors as potential therapeutics.

INTRODUCTION

Acute kidney injury (AKI) is a common disease that is typically triggered by conditions associated with direct kidney injury, diminished blood flow, or urinary tract obstruction (1–4). The key pathological features of AKI include tubular cell death, endothelial dysfunction, and inflammation (5–11). AKI impacts 5–20% of hospitalized individuals and is commonly associated with conditions such as sepsis, cancer, cardiac surgery, and crush syndrome (1–3). With ∼2 million annual AKI-related deaths globally, the absence of effective treatments underscores the urgent need for therapeutics (4).

Causal pathogenic pathways have been previously identified in epithelial, endothelial, and immune cells (5–8). However, therapeutically targeting these pathways is challenging due to the limited druggability of many of the identified pathways. Among the ∼20,300 protein-coding genes, only ∼22% are considered viable drug targets (12). Within this druggable genome, protein kinases are one of the largest families, with over 80 kinase inhibitors approved for cancer and inflammatory diseases (13). Their pivotal roles in cellular functions and implication in various pathological conditions make kinase inhibitors appealing for focused therapeutic interventions in cancer, inflammatory, autoimmune, and kidney disorders (13–19).

Using an unbiased kinome-wide screen, we recently identified CDKL5 as a pathogenic driver of renal tubular epithelial cell (RTEC) dysfunction (19, 20). In mice, genetic and pharmacological cyclin-dependent kinase-like (Cdkl)5 inhibition alleviated AKI (19). In humans, a genome-wide association study linked higher CDKL5 expression in the tubular compartment to reduced kidney function (21). Although CDKL5 has emerged as a druggable target for AKI, very little is known about other CDKL family members. This study sought to illuminate the activation status and functional significance of CDKL family kinases. Our findings revealed that the understudied CDKL1 kinase is a promising therapeutic target for AKI and sheds light on its biological function in responding to cellular stress.

MATERIALS AND METHODS

Animal Experiments

Mouse experiments followed approved protocols by the Institutional Animal Care and Use Committees. Mice were kept in controlled conditions and fed a standard diet with unrestricted water access. CRISPR-Cas9-mediated germline Cdkl1 gene knockout (KO) was done at GemPharmatech (Stock No. T027309). Heterozygous Cdkl1 mice were bred in house for wild-type (WT) and KO littermates. Sox11-flox mice (Stock No. T010043, GemPharmatech) were crossed with Ggt1-Cre (No. 012841, Jackson Laboratories) to obtain control and tubular-specific Sox11 KO mice. Littermate controls were used in all experiments.

Rhabdomyolysis and Ischemia-Reperfusion Injury-Associated AKI

Ischemia-reperfusion injury (IRI)- and rhabdomyolysis-associated AKI were induced in male mice aged 8–12 wk, following methodologies delineated in recent publications (20, 22, 23). Investigators assessing experimental outcomes were blinded to the genotypes of the mice. In brief, for IRI, mice were anesthetized, and bilateral renal pedicles were clamped for 30 min, followed by clamp removal and wound closure. Body temperature was monitored and maintained at 37°C. Sham-operated groups underwent analogous procedures without renal pedicle clamping. Rhabdomyolysis was induced by intramuscular glycerol injection into hindlimbs (7.5 mL/kg of 50% glycerol). Blood and renal specimens were harvested at specified time points for subsequent analyses.

Assessment of Kidney Injury

Renal injury was evaluated through a combination of serum analysis, including blood urea nitrogen (BUN) and creatinine levels, and histological examination using hematoxylin and eosin staining. Mouse blood samples were obtained at specified intervals and analyzed for BUN and creatinine concentrations using the QuantiChrom Urea Assay Kit (DIUR-100) and enzymatic assay-based creatinine measurements (ab65340, Abcam). Histological assessment involved the collection of mouse kidneys at designated time points followed by paraffin embedding. Tissue sections (4 µm) were stained with hematoxylin and eosin using established protocols. Histopathological evaluation was performed in a blinded manner, examining 10 consecutive ×100 fields per section from a minimum of 3 mice per group. Tubular injury severity was determined by assessing the percentage of tubules exhibiting dilation, epithelial flattening, cast formation, loss of the brush border and nuclei, and denudation of the basement membrane. The extent of tissue damage was graded according to the percentage of affected tubules as follows: 0 for no damage, 1 for <25%, 2 for 25%–50%, 3 for 50%–75%, and 4 for >75%.

Kinase Assay

Renal tissues and cells underwent lysis using a buffer solution comprising 150 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1% (vol/vol) Triton X-100, 2.5 mM sodium pyrophosphate, 1 mM β-glycerol phosphate, 1 mM Na₃VO₄, 10 μg/mL leupeptin, 10 μg/mL aprotinin, 1 mM phenylmethylsulfonyl fluoride, 50 mM NaF, 0.2% (wt/vol) dodecyl β-d-maltoside, and 20 mM Tris (pH 7.5). Subsequently, the soluble extracts were subjected to immunoprecipitation targeting Cdkl1, Cdkl2, Cdkl3, Cdkl4, and Cdkl5. The antibodies used for immunoprecipitation were Cdkl1 (NBP2-32482, Novus), Cdkl2 (M11432, Boster Bio), Cdkl3 (M12184, Boster Bio), Cdkl4 (H00344387, Novus), and Cdkl5 (ab22453, Abcam). The Cdkl1 and Cdkl5 antibodies were validated using KO murine kidney lysates, whereas the Cdkl2-4 antibodies were validated with Western blot analysis using overexpression lysates from human embryonic kidney (HEK)-293 cells. Specifically, 500 µg of the protein lysate were incubated with 2 μg of either IgG or anti-Cdkl1–Cdkl5 antibodies at 4°C overnight, followed by the addition of 30 μL of agarose protein A/G beads. The resulting bead-bound immunoprecipitates were washed and collected via centrifugation. These immunoprecipitates were then introduced into a protein kinase reaction buffer supplemented with 20 µM ATP and myelin basic protein (Millipore) as the substrate and incubated at 30°C for 30 min. The ADP-Glo Kinase Assay (Promega) kit was used to quantify kinase activity through luminescent detection of ADP generated during the kinase reaction as recently described (19, 24). Upon termination of the reaction, Western blot analysis was conducted to assess the level of immunoprecipitated proteins. Relative kinase activity was determined by normalizing the luminescence representing kinase activity to the densitometry of the signal for Cdkl1–Cdkl5 in the immunoprecipitates.

Immunoblot Analysis

Whole cell lysates from renal cortical tissues were made in modified RIPA buffer [20 mM Tris•HCl (pH 7.5), 150 mM NaCl, 1 mM Na₂EDTA, 1 mM EGTA, 1% Nonidet P-40, 2.5 mM sodium pyrophosphate, 1 mM β-glycerophosphate, protease, and phosphatase inhibitors] supplemented with 1% SDS. Invitrogen bis-Tris gradient mini- or midi-gels were used for Western blot analysis, followed by detection by ECL reagent (Cell Signaling). The primary antibodies used for Western blot analysis were Sox11 (No. 518104), Sox4 (No. 130633), β-actin (No. 47778), and neutrophil gelatinase-associated lipocalin (No. 50351) from Santa Cruz Biotechnology; Sox9 (EPR14335-78) and Cdkl5 (ab22453) were from Abcam; and Cdkl1 (NBP2-32482) was from Novus. All primary antibodies were used at 1:1,000 dilution. Secondary antibodies were from Jackson ImmunoResearch and used at 1:2,000 dilutions. Uncropped images of Western blots are provided in Supplemental Fig. S5.

Site-Directed Mutagenesis and Cycloheximide Assay

The Sox11 construct containing the FLAG tag was procured from Origene (MR206174). The QuikChange II XL Site-Directed Mutagenesis Kit from Agilent Technologies was used to induce mutations in Cdkl1 phosphorylation sites using established protocols (22, 25). Mutagenesis primers were designed using the QuikChange primer design program and synthesized by Integrated DNA Technologies. Sequencing of mutant constructs was conducted to validate successful mutagenesis. For cycloheximide pulse-chase experiments, Boston University mouse proximal tubular cell (BUMPT) cells were transiently transfected with plasmids encoding either WT Sox11 with the FLAG tag or the S253D mutant using previously outlined procedures (19). Subsequently, cells were treated with 100 µg/mL cycloheximide, and lysates were collected at specified time intervals. Western blot analysis was performed using an antibody against the FLAG tag to assess protein expression levels.

Statistical Analysis

The data depicted in all graphs are expressed as means with SD. Statistical assessments were conducted using GraphPad Prism software. A significance threshold of P < 0.05 was adopted. To ascertain statistical significance between two groups, a two-tailed unpaired Student’s t test was used. For comparisons among two or more groups, one-way ANOVA followed by Tukey’s or Dunnett’s multiple-comparisons test was used. No outliers were omitted, and each experiment was replicated a minimum of three times.

RESULTS

Probing the Activation Status of the CDKL Family During AKI

The CDKLs are part of the larger CMGC (CDK, MAPK, GSK, and CLK) kinase family. The CMGC family encompasses various kinases, such as cyclin-dependent kinases, mitogen-activated protein kinases, glycogen synthase kinase, and CDKLs (26). Despite sharing similar structural domains, it is still unclear whether they perform similar functions. The members of the CDKL family, including CDKL5, exhibit structural similarities primarily in the kinase domain. This domain contains an ATP-binding site, a serine/threonine (S/T) motif, and a classic threonine/any amino acid/tyrosine (TXY) motif. In addition, most members have nuclear localization signals, indicating their potential activity within the cell nucleus (Fig. 1A) (27). The distinguishing factor among CDKL family members lies in the C-terminal region. CDKL5, for instance, possesses an exceptionally long C-terminal region, whereas the others vary in size and sequence similarity in this region.

Figure 1.

Cdkl1 is activated in the kidneys during acute kidney injury. A: schematic representation for the CDKL family of kinases. The red, blue, and green triangles represent the ATP-binding site, TXY motif, and putative nuclear localization signal, respectively. B–H: C57B/6 wild-type male mice (8–12 wk old) were either injected with 50% glycerol into the hindlimb muscle (Rhabdo) or challenged with bilateral kidney ischemia for 30 min followed by reperfusion (IRI) to induce kidney injury. Kidneys were collected 24 h after injury followed by analysis of BUN (B) and serum creatinine (C) to assess the extent of damage. Sham, Rhabdo, and IRI kidney tissue lysates were used for immunoprecipitation of Cdkl1 (D), Cdkl2 (E), Cdkl3 (F), Cdkl4 (G), and Cdkl5 (H), followed by in vitro kinase assays. In the bar graphs (n = 8 biological replicates from 3 repeated experiments), the experimental values are presented as means ± SD. The height of the error bar = 1 SD. P < 0.05 was indicated as statistically significant. Student’s t test was carried out, and statistical significance is indicated by ****P < 0.0001. BUN, blood urea nitrogen; CDKL, cyclin-dependent kinase-like; IRI, ischemia-reperfusion injury; NS, not significant; Rhabdo, rhabdomyolysis.

We have previously demonstrated that Cdkl5 activation in tubular cells is an early event during AKI development (19, 20). To assess the activation status of all Cdkl family members, we used well-established models involving bilateral IRI and rhabdomyolysis (20, 28). In these models, AKI developed at 24 h, as evidenced by elevated BUN and serum creatinine levels (Fig. 1, B and C). We used the kidney cortical tissues to conduct immunoprecipitation of Cdkl1–Cdkl5 proteins, followed by an in vitro kinase assay using myelin basic protein as a substrate. As reported in our previous studies, in these assays, we measured the amount of ATP to ADP conversion as the readout of kinase activity (19, 24, 29). Our results indicated that, in addition to Cdkl5, Cdkl1 was also robustly activated during AKI, whereas the kinase activity of other members remained unchanged (Fig. 1, D–H).

Cdkl1 Germline KO Mice Are Protected From Rhabdomyolysis and IRI-Associated AKI

The consequence of Cdkl1 gene deletion in mice is unknown. To address this, we obtained mice with CRISPR-Cas9-mediated Cdkl1 gene deletion (Fig. 2, A and B). Heterozygous mice were crossed, and homozygous KO (Cdkl1^−/−) mice were compared with WT littermates in all experiments. Mice were born at the expected Mendelian ratios, were viable, and appeared normal with no obvious phenotype. The kidney or body weight was not affected by Cdkl1 global gene deletion (Supplemental Fig. S1). WT and Cdkl1^−/− mice were monitored for up to 1 year and no obvious abnormalities were observed.

Figure 2.

Cdkl1 germline deletion provides protection from rhabdomyolysis- and IRI-associated acute kidney injury. Cdkl1 knockout mice were created by deleting exon (E)3 in the Cdkl1 gene to create a null transcript. A: illustration of the Cdkl1 knockout mouse gene structure with exon 3 deletion in red. B: representative Western blots of kidney tissues showing successful gene deletion. C–F: Cdkl1 knockout mice or littermate WT controls (8–12 wk old) were subjected to rhabdomyolysis-associated acute kidney injury by injecting 50% glycerol into the hind muscle or subjected to bilateral ischemia for 30 min followed by reperfusion. Blood and kidneys were collected at 24 h, and kidney function and damage were examined by BUN (C), serum creatinine (D), and histological analysis (H&E staining) (E and F). Asterisks indicate damaged tubules. Scale bar = 100 μm. In the bar graphs (n = 6 biological replicates from 3 repeated experiments), the experimental values are presented as means ± SD. The height of the error bar = 1 SD. P < 0.05 was indicated as statistically significant. Student’s t test was carried out, and statistical significance is indicated by *P < 0.05. BUN, blood urea nitrogen; CDKL, cyclin-dependent kinase-like; H&E, hematoxylin and eosin; IRI, ischemia-reperfusion injury; NS, not significant; Rhabdo, rhabdomyolysis; WT, wild-type.

Next, we treated WT and Cdkl1^−/− mice with intramuscular glycerol injection to induce rhabdomyolysis-associated AKI or 30 min of bilateral ischemia as reported in our previous studies (20, 22, 28). In both these models, AKI is induced within 24 h. The severity of injury and extent of kidney damage were assessed by measuring BUN, serum creatinine, and histological analysis. We found that Cdkl1^−/− mice had significantly less kidney damage in both rhabdomyolysis and IRI models compared with WT mice (Fig. 2, C–F). Cdkl1 gene deletion mitigated long-term survival after ischemic injury (Supplemental Fig. S2). These results provide strong evidence that Cdkl1 contributes to the pathogenesis of AKI.

Cdkl1 Phosphorylates Sox11 During AKI

The substrates of Cdkl1 remain unknown. In our previous work, we showed that Cdkl5 phosphorylates SRY-related HMG box gene 9 (Sox9) during AKI (19, 20, 22). To ascertain if Cdkl1 is also involved in Sox9 phosphorylation, we carried out immunoblot analysis of WT and Cdkl1 KO kidney tissues (Fig. 3A). We found that AKI-associated Sox9 phosphorylation was not influenced by Cdkl1 gene deletion. Interestingly, data analysis of single-cell RNA-sequencing studies from other groups and bulk RNA-sequencing studies from our group showed that along with Sox9, Sox4 and Sox11 are also upregulated in proximal tubular cells during AKI (23, 30). The gene expression data were confirmed by immunoblot analysis, which showed a significant induction of Sox9, Sox4, and Sox11 during AKI (Fig. 3A). The mRNA level of Sox11 was also significantly induced during AKI (Supplemental Fig. S3). Interestingly, Cdkl1 deficiency resulted in higher Sox11 protein induction in the kidneys (Fig. 3A). We next carried out immunoprecipitation experiments with Sox11 antibody followed by immunoblot analysis with a phosphoserine antibody, which showed a significant induction in Sox11 phosphorylation that was suppressed in Cdkl1 KO mice (Fig. 3A, bottom). These results suggest that Cdkl1 might regulate Sox11 stability through phosphorylation-mediated posttranslational regulation.

Figure 3.

Cdkl1 phosphorylates Sox11 at site Ser²⁵³. A: bilateral kidney ischemic injury, rhabdomyolysis-associated injury, or sham kidney tissues lysates from Cdkl1^−/− mice or control littermates were used for Western blot analysis of the indicated proteins. Kidney cortical lysates were also immunoprecipitated by Sox11 antibody to assess Sox11 and phospho-serine levels in WT and Cdkl1^−/− mice (bottom). B: purified Cdkl1, WT Sox11, and mutant Sox11 were coincubated for an in vitro kinase assay followed by Western blot and autoradiography of Sox11 phosphorylation. C: schematic of the mouse Sox11 protein sequence. HMG-Box indicates the high mobility group box domain, AR indicates the autoregulatory domains, and TAD indicates the transactivation domain. Protein sequence analysis showed the sequence surrounding Ser²⁵³ was highly conserved among organisms. D and E: analysis of Sox11 protein stability was performed by carrying out the cycloheximide (CHX) chase assay in BUMPT cells transfected with Flag-tagged WT and S253D phosphomimetic Sox11 mutant. The results indicated that the phosphomimetic mutant had reduced stability compared with WT Sox11. In the bar graph (n = 3 biological replicates from 3 repeated experiments), the experimental values are presented as means ± SD. The height of the error bar = 1 SD. P < 0.05 was indicated as statistically significant. Student’s t test was carried out, and statistical significance is indicated by *P < 0.05. CDKL, cyclin-dependent kinase-like; IP, immunoprecipitation; IRI, ischemia-reperfusion injury; Rhabdo, rhabdomyolysis; WT, wild-type.

We investigated whether recombinant Cdkl1 could catalyze the phosphorylation of Sox11. Our findings demonstrated that Cdkl1 is capable of phosphorylating Sox11 (Fig. 3, B and C), with the primary phosphorylation site identified as Ser²⁵³ (corresponding to Ser²⁸⁷ in human SOX11). Although previous studies have identified various sites of Sox11 phosphorylation in nonkidney tissues, Ser²⁵³ has not been previously studied (31–33). This site is present in the C-terminus in a domain with unknown function. In our experiments using cycloheximide assays (Fig. 3, D and E), we observed decreased stability in the Sox11-S253D phosphomimetic mutant. This suggests that the phosphorylation of Sox11 mediated by Cdkl1 might lead to reduced stability and functional suppression.

Sox11 Conditional KO Mice Exhibit Higher Rhabdomyolysis- and IRI-Associated AKI

To study the role of Sox11 in AKI, we next generated proximal tubule-specific conditional KO mice by crossing Sox11 floxed mice with Ggt1-Cre mice (Fig. 4, A and B). These mice had no renal abnormalities under normal conditions. We confirmed that kidney or body weight was not influenced by Sox11 gene deletion (Supplemental Fig. S4). However, when mice were challenged with rhabdomyolysis and bilateral IRI, they exhibited increased severity of AKI as measured by BUN, serum creatinine, and histological analysis (Fig. 4, C–F). Cumulatively, our findings imply that the phosphorylation of Sox11 by Cdkl1 may attenuate its protective capacity, thereby exacerbating AKI pathogenesis. Significantly, our investigation underscores the pivotal role of Sox11 as a protective gene in the context of AKI.

Figure 4.

Tubular Sox11 knockout mice exhibit elevated acute kidney injury. A: tubular-specific Sox11 conditional knockout (Sox11^cKO) mice were created by crossing Sox11-flox mice with Ggt1-Cre mice. B: representative Western blots of Rhabdo kidney tissues showing successful gene deletion. C–F: Sox11^cKO mice or littermate controls (8–12 wk old) were subjected to Rhabdo-associated acute kidney injury by injection of 50% glycerol into the hind muscle or subjected to bilateral ischemia for 30 min followed by reperfusion. Blood and kidneys were collected at 24 h, and kidney function and damage were examined by BUN (C), serum creatinine (D), and histological analysis (H&E staining) (E and F). Asterisks indicate damaged tubules. Scale bar = 100 μm. In the bar graphs (n = 6 biological replicates from 3 repeated experiments), the experimental values are presented as means ± SD. The height of the error bar = 1 SD. P < 0.05 was indicated as statistically significant. Student’s t test was carried out, and statistical significance is indicated by *P < 0.05. BUN, blood urea nitrogen; CDKL, cyclin-dependent kinase-like; Con, control; IRI, ischemia-reperfusion injury; NS, not significant; H&E, hematoxylin and eosin; Rhabdo, rhabdomyolysis; WT, wild-type.

DISCUSSION

The physiological role of Cdkl1 has remained unknown due to the lack of gene deletion mouse models. Although Sox11 has been investigated for its role in nephrogenesis, its contribution to the tubular stress response in AKI remains unknown (34–36). In the current study, we clarified the detrimental role of Cdkl1 kinase and the protective function of the Sox11 transcription factor in tubular dysfunction linked to AKI.

The CDKL family members were originally discovered in the quest for cdc2-related kinases, identified through the highly conserved proline-serine-threonine-alanine-isoleucine-arginine-glutamate (PSTAIRE) motif (37). This exploration resulted in the identification of CDKL1 (KKIALRE), CDKL2 (KKIAMRE), and CDKL3 (NKIAMRE) (37–39). Despite their sequence similarity to cdc2 kinases, CDKLs have not demonstrated evidence of cyclin binding or involvement in cell cycle regulation (40). The CDKL family expanded further with the identification of CDKL5 during the mapping of the Xp22 transcriptional region, which has been associated with various diseases (41). Tissue expression patterns within the CDKL family vary. CDKL1, CDKL2, and CDKL5 exhibit more ubiquitous expression across various tissues, including the brain and kidneys (38, 42, 43); in contrast, CDKL3 and CDKL4 are primarily restricted to germ cells, with lower expression in other tissues (27, 39, 43). Despite their prevalence, the physiological functions of CDKL members remain largely unknown, with only CDKL5 gene deletion mice having been previously developed (44–47).

Our work illuminates the physiological function of CDKL1 through the use of germline KO mice. Importantly, despite the absence of noticeable baseline phenotypic abnormalities in Cdkl1-deficient mice, they exhibit protection against kidney injury. In contrast, prior investigations in zebrafish have demonstrated that depleting CDKL1 using morpholino antisense oligonucleotides led to severe developmental defects, including malformations in the brain and eye regions (48). Remarkably, our Cdkl1^−/− mice show no apparent developmental or other abnormalities. This pattern mirrors observations related to CDKL5 gene deletion, wherein zebrafish exhibit skeletal and neuronal malformations, whereas Cdkl5^−/− mice remain predominantly normal without anatomic or developmental issues (44, 45, 49). One potential explanation for these differences might be compensation by other family members in mice. Consequently, it would be intriguing to generate double KO mice for both Cdkl1 and Cdkl5 to further investigate and characterize their phenotypes.

In contrast with mammals, Caenorhabditis elegans has only one CDKL gene, named CDKL-1. This worm CDKL-1 has combined features of mammalian CDKL1, CDKL2, CDKL3, and CDKL4. Intriguingly, worm CDKL-1 has been shown to be involved in cilium length regulation (40, 50, 51). Murine and human CDKL1 are distinct from worm CDKL-1, and based on the baseline phenotype, it is unlikely that it has any role in cilia development or regulation. However, further studies are needed to conclusively prove this notion. Interestingly, CDKL1 has putative nuclear localization sequences, suggesting a role in nuclear function. Previous studies have shown CDKL1 localization in both the cytosol and nucleus (38, 52). Other CDKL members have multiple nuclear localization signals as well and may translocate and have either nuclear or cytoplasmic substrates depending on the circumstance, as is the case with CDKL5 (53).

We propose that Cdkl1 may contribute to the development of AKI by phosphorylating and destabilizing the transcription factor Sox11. Sox11, a well-known regulator of kidney development and nephrogenesis, exhibits widespread expression in the early stages of nephrogenesis, decreasing as nephron tubule differentiation progresses (34–36). In adult tissues, Sox11 expression is minimal (34, 54). Notably, single-cell RNA-sequencing studies have shown an increase in Sox11 expression in tubular cells during AKI (30). Our bulk RNA-sequencing analysis of kidneys subjected to cisplatin, rhabdomyolysis, and IRI also revealed elevated levels of Sox11 (23). To investigate the function of Sox11, we carried out targeted gene deletion specifically within tubular cells, and these mice exhibited exacerbated AKI severity. Tubular Sox11 gene deletion did not lead to renal abnormalities under normal conditions, likely because Ggt1-Cre is activated postnatally. This suggests that, akin to Sox9, Sox11 acts as a protective factor that is reactivated in tubular cells during AKI. Future studies are required to identify downstream Sox11 target genes to understand its protective role during AKI.

Here, we present in vitro evidence suggesting that the phosphorylation of Sox11 at Ser²⁵³ by Cdkl1 may lead to reduced stability. This is supported by the in vivo observation that the total Sox11 protein upregulation in renal tissues after AKI was further elevated in the kidneys of Cdkl1 KO mice. Although the phosphorylation site is not present in the DNA-binding domain, future studies are required to examine the effect of Ser²⁵³ phosphorylation on the transcriptional activity of Sox11. Our study has two notable limitations. First, we were not able to generate a phospho-Ser²⁵³ Sox11 antibody, which prevented a direct assessment of Sox11 phosphorylation status during AKI. Second, we used germline Cdkl1 KO mice for our study, which does not directly prove the role of Cdkl1 activation in tubular cells. Nevertheless, both our past research and the current study suggest that CDKLs play a crucial role as stress-responsive nuclear kinases in tubular cells, influencing the essential Sox family of transcription factors.

Our study revealed a stress-responsive axis involving Cdkl1 and Sox11, pinpointing Cdkl1 as a potential target for drug intervention in AKI. In summary, this study provides the initial characterization of Cdkl1 germline KO mice, shedding light on its nonessential role in development and its consequential role in AKI pathogenesis.

DATA AVAILABILITY

Data will be made available upon reasonable request.

SUPPLEMENTAL MATERIAL

Supplemental Figs. S1–S5: https://doi.org/10.6084/m9.figshare.26081800.v2.

GRANTS

This work was supported by funds from the Ohio State University Cancer Center (to N.S.P.) and National Institutes of Health Grant R01DK132230 (to N.S.P.). J.A.S. was supported by a Predoctoral Fellowship 900765 from the American Heart Association and a CDKL5 Forum Junior Fellowship Award.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

J.A.S., N.S.P., and J.Y.K. conceived and designed research; J.A.S., G.V.M., T.V., A.J.D., D.Z.-O., N.S.P., and J.Y.K. performed experiments; J.A.S., G.V.M., T.V., A.J.D., D.Z.-O., N.S.P., and J.Y.K. analyzed data; J.A.S., A.J.D., D.Z.-O., N.S.P., and J.Y.K. interpreted results of experiments; J.A.S., N.S.P., and J.Y.K. prepared figures; J.A.S., N.S.P., and J.Y.K. drafted manuscript; J.A.S., A.J.D., D.Z.-O., N.S.P., and J.Y.K. edited and revised manuscript; J.A.S., N.S.P. and J.K. approved final version of manuscript.