IFT88 deficiency in proximal tubular cells exaggerates cisplatin-induced injury by suppressing autophagy

Abstract

Primary cilia are widely regarded as specialized sensors in differentiated cells that have been implicated in the regulation of cell proliferation, differentiation, and viability. We have previously shown that shortening of primary cilia sensitizes cultured kidney tubular cells to cisplatin-induced apoptosis. Intraflagellar transport 88 (IFT88) is an essential component for ciliogenesis and maintenance. Here, we have further examined the effect of proximal tubule-specific IFT88 ablation on cisplatin-induced acute kidney injury (AKI). In this study, more severe AKI occurred in IFT88 knockout mice than age- and sex-matched wild-type mice. Mechanistically, cisplatin stimulated autophagy in kidney tubular cells as an intrinsic protective mechanism. However, renal autophagy was severely impaired in IFT88 knockout mice. In cultured HK-2 cells, cisplatin induced more apoptosis when IFT88 was knocked down. Tat-beclin 1 peptide, a specific autophagy activator, could partially prevent IFT88-associated cell death during cisplatin treatment, although cilium length was not improved significantly. Reexpression of IFT88 partially restored autophagy in IFT88 knockdown cells and suppressed apoptosis during cisplatin treatment. Taken together, these results indicate that defective autophagy in IFT88-deficient kidney cells and tissues contributes to the exaggerated AKI following cisplatin exposure.

NEW & NOTEWORTHY Almost every cell has one hair-like, nonmotile antenna projecting from the cell surface, named the primary cilium. In kidney tubular cells, the primary cilium has a protective role, but the underlying mechanism is unclear. This study shows that a short cilium leads to the suppression of autophagy, which is responsible for the heightened injury sensitivity. These findings provide the clues of how to manipulate primary cilium and autophagy to save kidneys.

INTRODUCTION

The primary cilium is one type of specific organelle protruding from the cell surface in almost every cell in the human body. Structurally, it is composed of α-tubulin-based microtubules, motor proteins, and intraflagellar transport (IFT) particles inside encapsulated by specialized cell membrane outside. Originally, nonstructural polycystin-1 and polycystin-2 proteins, responsible for adult polycystic kidney disease, were localized on the primary cilia and form a protein complex to regulate flow-induced Ca²⁺ signaling (1, 2). Further intensive studies have disclosed that dysfunction of many cilia-associated proteins leads to a large spectrum of human genetic diseases, named ciliopathies (3). Functionally, primary cilia have been widely regarded as not only cellular sensors but also a type of organelle important for cell proliferation, differentiation, apoptosis, and autophagy.

Autophagy is a finely regulated, degradative process for maintaining cellular homeostasis and function largely by a family of autophagy-related proteins. Dysfunction of autophagy is associated with a variety of pathological conditions, such as aging, neurodegenerative diseases, infections, and cancers. In kidneys, autophagy is important to renal physiology and autophagy dysfunction contributes to a series of renal diseases such as acute kidney injury (AKI), polycystic kidney disease, diabetic nephropathy, and renal fibrosis (4–8).

AKI is a common kidney disease mainly caused by sepsis, renal ischemia, or exposure to nephrotoxins. Cisplatin is a chemical drug that is used clinically to treat various types of cancers. However, cisplatin induces AKI or nephrotoxicity, which restricts its clinical use and efficacy. Mechanistic studies have disclosed a variety of factors, processes, and signaling pathways in cisplatin-induced AKI (9–29). We and others (30–33) have recently found that shortening of primary cilia in the kidney sensitizes cells to apoptosis and that there is a reciprocal regulation of primary cilia and autophagy, and autophagy protects kidneys from different types of injury. In this study, we tested the hypothesis that shortening of cilia in kidney proximal tubule cells by ablation of IFT88 may sensitize cisplatin-induced injury. We further examined the alteration of autophagy as an underpinning mechanism.

METHODS

Antibodies and Reagents

Primary antibodies were obtained as follows: cleaved caspase-3 (no. 9661) and GAPDH (no. 2118) from Cell Signaling, acetylated tubulin (ac-tubulin; no. 7451) and β-actin (no. A2228) from Sigma, IFT88 (no. 13967-1-AB) from Proteintech Group, light chain (LC)3B (no. NB100-2220) from Novus Biologicals, and fluorescein Lotus tetragonolobus agglutinin (LTA; no. FL-1321) from Vector Laboratories. Secondary antibodies for Western blot analysis were from Jackson ImmunoResearch (West Grove, PA), for immunohistochemistry from Dako North America (Santa Clara, CA), and for immunofluorescence staining from Chemicon (Temecula, CA), respectively. Other reagents were from Sigma unless specifically indicated.

Generation of Conditional IFT88 Knockout Mice

Floxed IFT88 mice were purchased from the Jackson Laboratory. To knock out IFT88 from kidney proximal tubular epithelial cells, IFT88 mice were crossed with phosphoenolpyruvate carboxykinase (PEPCK)-Cre mice (34) to generate IFT88 wild-type (WT) and knockout (KO) mice. The IFT88-floxed allele product was amplified as a DNA band of ∼410 bp by PCR using one pair of primers (5′- GACCACCTTTTTAGCCTCCTG-3′ and 5′- AGGGAAGGGACTTAGGAATGA-3′). The Cre gene product was detected at ∼420 bp by PCR using one pair of primers (5′- ACCTGAAGATGTTCGCGATTATCT-3′ and 5′- ACCGTCAGTACGTGAGATATCTT-3′). Animal experiments were conducted according to a protocol approved by the Institutional Animal Care and Use Committee of Charlie Norwood Veterans Affairs Medical Center. All mice were housed in the animal facility of the Charlie Norwood VA Medical Center.

Cisplatin-Induced AKI

IFT88 mice had free access to water and regular chow. Among the four groups of mice (4–5 mice/group), two groups of WT and KO mice were injected intraperitoneally with 30 mg/kg cisplatin, whereas the other two groups were injected with saline as controls. Mice were euthanized at 72 h after cisplatin administration to collect blood for measurement of blood urea nitrogen (BUN) and serum creatinine (SCr) using a commercial kit (Stanbio Laboratories, Boerne, TX) and the picric acid method established in this laboratory. Snap-frozen kidney samples in liquid nitrogen were used for biochemistry analysis, whereas samples fixed in 4% paraformaldehyde were for tissue section examination.

Apoptosis Assay

For TUNEL on kidney tissue sections, the In Situ Cell Death Detection Kit from Roche was used. Briefly, fixed kidney tissues were further processed in the Electron Microscopy and Histology Core of Augusta University. Kidney samples were embedded in paraffin, sectioned at 7 μm (Leica RM2025), and dried overnight before deparaffinization and rehydration through graded concentrations of ethanol. For morphological analysis of apoptosis in cultured cells, cells were stained with Hoechst 33342 and observed under an EVOS microscope (Thermo Fisher Scientific, Waltham, MA). Typical apoptosis was indicated by cell shrinkage and blebbing, nuclear condensation, and fragmentation.

Cell Culture and Treatment

Stable IFT88 knockdown cells derived from HK-2 cells (CRL-2190, American Type Culture Collection), including one clone targeting the IFT88 3′-untranslated region (3′-UTR); sequence (5′- GCCTTATGAGATCATCCTCAT-3′), were made as previously described (33) and maintained in DMEM-F-12 media supplemented with 10% FBS and puromycin (0.5 µg/mL). Cells were treated with 50 µM cisplatin for 8 h in combination with 25 µM scrambled or tat-beclin 1 peptide.

Reexpression of IFT88 in 3′-UTR Knockdown Cells

IFT88 plasmid (pEF5-FRT-TagRFP-IFT88), purchased from Addgene (no. 61684), was transfected into 3′-UTR knockdown cells and enriched with hygromycin (100 µg/mL), followed by cisplatin treatment for 8 h. Treated cells were used for biochemical and morphological analyses.

Immunoblot Analysis

Proteins from kidney tissues and cells were extracted with the SDS lysis buffer as previously described (32). Frozen kidney tissues in SDS buffer were homogenized with Bullet Blender Homogenizer (Next Advance, Troy, NY) for 2 min. Extracted samples were centrifuged at 15,000 relative centrifugal force (rcf) for 15 min at 4°C, and supernatants for protein analysis were transferred to a new tube. Protein samples (100–150 µg per lane for tissues, 15 µg for cultured cells) were electrophoresed in a 10% bis-Tris gel and transferred to PVDF nylon membranes (Bio-Rad, Hercules, CA). After being blocking with 5% nonfat dry milk (Lab Scientific, Livingston, NJ) in PBS for 1 h, the filter was incubated with primary antibody and washed with PBS-0.1% Tween 20 (MP Biomedicals, Santa Ana, CA). The filter was finally incubated with secondary antibody conjugated with horseradish peroxidase. The bound antibodies were detected with Detection Reagent (Pierce, Rockford, IL). Quantification of the band intensity was accomplished with National Institutes of Health ImageJ software.

Hematoxylin and Eosin Staining, Immunofluorescence Staining, and Confocal Microscopy

The histology of kidney tissues was examined by hematoxylin and eosin staining. Histological changes were evaluated by the percentage of renal tubules showing tubular lysis, loss of the brush border, cast formation, and other signs of injury. For immunofluorescence staining, tissue antigen retrieval was done by boiling slides in retrieval buffer [10 mM sodium citrate (pH 6.0)] and 0.05% Tween 20, and cultured cells on coverslips were fixed with 4% paraformaldehyde for 15 min and permeabilized with 0.1% Triton X-100 for 5 min. Samples were incubated with primary antibodies for 1 h after 1 h of BSA blocking, followed by a complete wash with PBS and then incubated with secondary antibodies with or without LTA (1:200) for another 1 h. Before being mounted with the Prolong Gold antifade reagent, cell nuclei were stained with Hoechst 33342 (H-1399, Invitrogen). All these procedures were performed at room temperature. For image analysis, Zeiss Axio fluorescence and confocal microscopes (Carl Zeiss, Thornwood, NJ) were used. The confocal microscope was equipped with the laser scanning microscopy (LSM) Image analysis system and used for the measurement of cilium length.

Immunohistochemistry

Kidney sections were treated as described above for antigen retrieval, followed by incubation with blocking buffer containing 2% BSA, 0.2% milk, and 2% normal goat serum in PBS with 0.8% Triton X-100. Tissue sections were incubated with IFT88 antibody overnight at 4°C, followed by incubation with horseradish peroxidase-conjugated secondary antibody. After incubation for signal amplification using the TSA Biotin System (Marlborough, MA), slides were treated with a VECTASTAIN Elite ABC Kit and developed with ImmPACT DAB Peroxidase Substrate (Vector Laboratories). The reaction was stopped by putting slides in water, followed by LTA staining.

Statistics

Data are expressed as means ± SD. A grouped or paired t test was calculated. P < 0.05 was considered statistically significant.

RESULTS

KO of IFT88 Sensitizes Renal Proximal Tubular Cells to Cisplatin Nephrotoxicity

To study the function of IFT88 in cisplatin-induced kidney injury, we established conditional KO mice with specific ablation of IFT88 in renal proximal tubule cells. After mouse breeding, IFT88 homozygous mice with or without PEPCK-Cre expression in proximal tubular cells were obtained (Fig. 1, A and B). By immunoblot analysis, we detected IFT88 in kidney proximal tubular cells in WT but not KO mice (Fig. 1C). To further confirm IFT88 KO in proximal tubular cells, we performed immunohistochemistry with IFT88 antibody. It was obvious that IFT88 was detected in LTA-positive tubules in WT mice but not in KO mice (Fig. 1D). To study the impact of IFT88 KO on tubular cells, we injected cisplatin or vehicle intraperitoneally into mice. After 3 days of cisplatin administration, renal function in both WT and KO mice deteriorated, but particularly in KO mice (Fig. 2). Both BUN and SCr levels in KO mice were higher than in WT mice, whereas before cisplatin injection, these two genotypes of mice showed normal and low BUN and SCr. Consistently, hematoxylin and eosin staining showed more severe kidney tubular damage in IFT88 mice after cisplatin treatment (Fig. 3A). By TUNEL staining, more apoptotic cells were observed in kidney tissues in IFT88 KO mice than in WT mice (Fig. 3B).

Figure 1.

Generation of conditional intraflagellar transport 88 (IFT88) knockout (KO) mice. A: to establish conditional KO mice with IFT88 deletion in proximal tubular cells, floxed IFT88 homozygous mice were crossed with phosphoenolpyruvate carboxykinase (PEPCK)-Cre transgenic mice to generate IFT88 wild-type (WT) and IFT88 KO mice. B: by PCR, a floxed IFT88 band was detected in both WT and KO mice and a PEPCK-Cre band only in the latter mice. C: by immunoblot (IB) analysis, IFT88 was confirmed to be removed in the IFT88 KO kidney cortex. β-Actin was used as a loading control. D: by immunohistochemistry, IFT88 expression was detected in WT mice at proximal tubular cells labeled by Lotus tetragonolobus agglutinin (LTA) but not in KO mice. Scale bar = 20 µm.

Figure 2.

Kidney function in intraflagellar transport 88 (IFT88) wild-type (WT) and IFT88 knockout (KO) mice after cisplatin treatment. To study the impact of IFT88 KO on kidney function, cisplatin was intraperitoneally injected into IFT88 WT and KO mice. Blood urea nitrogen (BUN) and serum creatinine (SCr) were measured under normal conditions and after 3 days of cisplatin. A: no BUN and SCr differences were detected between IFT88 WT and KO mice under physiological conditions. n = 4 animals for each group. B: BUN and SCr demonstrated higher levels in IFT88 KO mice than in IFT88 WT mice. n = 5 animals for each group. *P < 0.05.

Figure 3.

Cisplatin-induced kidney tubule injury and apoptosis in intraflagellar transport 88 (IFT88) wild-type (WT) and knockout (KO) mice. A: by hematoxylin and eosin staining, both WT and KO mice showed tubular injury, with the latter particularly more severe than the former by injury index. No obvious morphological differences between WT and KO mouse kidneys were observed under physiological conditions. n = 4 animals for each group. B: more apoptotic cells were found in KO mice than in WT mice by TUNEL staining. Cell nuclei were stained with Hoechst 33342. n = 12 fields of view for the IFT88 WT group; n = 15 fields of view for the IFT88 KO group. *P < 0.05; **P < 0.01. Scale bars = 100 µm in A and 20 µm in B. LTA, Lotus tetragonolobus agglutinin.

Shorter Primary Cilium Length in Renal Proximal Tubular Cells After Cisplatin Treatment

To evaluate the effect of cisplatin on cilium length in IFT88 WT and KO mice, we measured cilium length after cisplatin administration. Obviously, after cisplatin treatment, both IFT88 WT and KO mice showed reduced cilium length compared with mice under normal conditions. Of note, cilium length in IFT88 KO mice was lower than in WT mice after cisplatin administration (Fig. 4).

Figure 4.

Renal cilia in intraflagellar transport 88 (IFT88) wild-type (WT) and IFT88 knockout (KO) mice after cisplatin treatment. To study the effect of cisplatin on cilium length, cilium length was measured in different groups of mice. No differences in cilium length were found between WT and KO mice under physiological conditions. However, cilium length was reduced significantly after cisplatin administration, especially in the IFT88 KO group of mice. n, counted cilia number. **P < 0.01, KO vs. WT mice; ###P < 0.001, cisplatin vs. vehicle. Scale bar = 10 µm. ac-tubulin, acetylated tubulin; LTA, Lotus tetragonolobus agglutinin.

Autophagy Suppression in IFT88 KO Mice After Cisplatin Injection

Autophagy is an essential process for cellular homeostasis and function, and dysfunction of autophagy results in a series of pathological conditions. Autophagy is one key mechanism in preserving renal function from injury. To test autophagy status after cisplatin administration in IFT88 KO mice, we performed immunoblot analysis and immunofluorescence staining on kidney sections (Fig. 5, A and B). As shown in Fig. 5A, after cisplatin treatment, autophagy was upregulated compared with mice under physiological conditions. Furthermore, compared with WT mice after cisplatin treatment, IFT88 KO mice demonstrated suppressed LC3B II expression. By immunofluorescence staining, similar findings were found (Fig. 5B). No obvious LC3B signal was detected in both IFT88 WT and KO mice under physiological conditions, whereas after cisplatin treatment, more LC3B signals were found in IFT88 WT mice than in KO mice.

Figure 5.

Autophagy in intraflagellar transport 88 (IFT88) wild-type (WT) and IFT88 knockout (KO) mice after cisplatin treatment. To examine autophagy status, immunoblot (IB) analysis and immunofluorescence staining with light chain (LC)3B antibody were performed. A: by IB analysis, no big differences of LC3B II were found between IFT88 WT and KO mice. n = 4 for each group. B: LC3B signal was suppressed in IFT88 KO mice compared with WT mice. Cell nuclei were stained with Hoechst 33342. n = 11 fields of view for the IFT88 WT group; n = 12 fields of view for the IFT88 KO group. *P < 0.05 and **P < 0.01, KO vs. WT mice. Scale bar = 10 µm. LTA, Lotus tetragonolobus agglutinin.

Rescue of Apoptotic Cells by Autophagy Enhancer Tat-Beclin 1 Peptide in HK-2 Cells

Because primary cilia and autophagy regulate reciprocally and autophagy plays a protective role in AKI (6, 30–32), we determined whether the higher sensitivity of kidney tubular cells in IFT88 KO mice to cisplatin is due to the suppression of autophagy in kidney tubular cells. To this end, we examined the effect of tat-beclin 1 peptide, a specific autophagy activator, in IFT88 knockdown HK-2 cells (Fig. 6D). We first tested the effects of cisplatin on autophagy and cilium length and autophagy inhibitor (bafilomycin) on cilium length in HK-2 cells. It was obvious that cisplatin treatment activated autophagy and suppressed cilium length and that autophagy inhibitor shortened cilium length (Fig. 6, A–C). As shown in Fig. 6E, tat-beclin 1 peptide improved the survival of HK-2 cells during cisplatin treatment, including WT and knockdown IFT88 cells. Notably, more IFT88 knockdown cells survived cisplatin treatment, in contrast to cisplatin-treated WT cells. This was further confirmed by immunoblot analysis of cleaved caspase 3, indicative of caspase activation (Fig. 6G). As previously shown (32), tat-beclin 1 peptide increased cilium length in WT cells upon 8-h cisplatin treatment. In IFT88 knockdown cells, tat-beclin 1 failed to promote the growth of cilia (Fig. 6F).

Figure 6.

Enhancement of autophagy alleviates cell death in intraflagellar transport 88 (IFT88) knockdown human kidney (HK)-2 cells. To determine whether depression of autophagy is partially responsible for increased apoptosis in IFT88 knockout cells, we enhanced autophagy by tat-beclin 1 peptide and then examined apoptosis. A: immunoblot (IB) analysis showed induced autophagy in cisplatin-treated HK-2 cells (50 µM, 8 h). B: cisplatin suppressed cilium length in HK-2 cells (50 µM, 8 h). C: bafilomycin (Baf; 25 nM, 24 h) inhibited cilium length in HK-2 cells. D: IB analysis confirmed the reduction of IFT88 in shIFT88 knockdown cells. E: IFT88 knockdown cells were treated with cisplatin alone or in combination with tat-beclin 1 peptides. Compared with control cells, more dead cells were found in cisplatin-treated IFT88 knockdown cells (red arrow). After tat-beclin 1 peptide treatment, both control and IFT88 knockdown cells demonstrated resistance to cisplatin treatment, especially knockdown cells. F: after tat-beclin 1 peptide treatment, cilium length was elongated, whereas tat-beclin 1 peptide had a minimal effect on cilium length in IFT88 knockdown cells. G: IB analysis confirmed autophagy augment and apoptosis after tat-beclin 1 peptide treatment. Cell nuclei were stained with Hoechst 33342. **P < 0.01 and ***P < 0.001, knockdown vs. scramble; #P < 0.05, tat-beclin 1 vs. scramble. Scale bar = 5 µm in A–C, 100 µm in E, and 10 µm in F. ac-tubulin, acetylated tubulin; cleaved cas-3, cleaved caspase-3; LC3B, light chain 3B.

Rescue Effects of Reexpression of IFT88 in IFT88 Knockdown Cells

Finally, we performed a rescue experiment by reexpressing IFT88 in IFT88 knockdown cells (Fig. 7). To this end, shRNA targeting the 3′-UTR of IFT88 (shIFT883UTR) was transfected into HK-2 cells. Cells were then transfected with IFT88 for reexpression (Fig. 7A). After cisplatin treatment, IFT88-reexpressing cells had less apoptosis than IFT88 knockdown cells, as shown by immunoblot analysis of active/cleaved caspase 3 (Fig. 7B) and cell morphology (Fig. 7C). Moreover, IFT88 reexpression partially restored autophagy, as shown by LC3B blots (Fig. 7B). This rescue experiment provides additional evidence to indicate that an autophagy abnormality in IFT88 knockdown cells and KO mice is at least partially responsible for increased cell death and worsening of renal function.

Figure 7.

Effects of reexpression of intraflagellar transport 88 (IFT88) in IFT88 knockdown cells. HK-2 cells were transfected with shRNAs targeting the 3′-untranslated region (3′-UTR) of the IFT88 gene (shIFT883UTR) or control shRNAs (shControl). shIFT883UTR cells were then transfected with IFT88 for reexpression. A: immunoblot (IB) analysis verified the expression of IFT88 protein tagged with red fluorescent protein (RFP) after transfection into shIFT883UTR cells. *Endogenous IFT88. Arrowhead, RFP-tagged exogenous IFT88. B: IFT88-reexpressing cells had more light chain (LC)3B and active/cleaved caspase-3 (cleaved cas-3). C, left: IFT88-reexpressing cells showed less apoptosis in morphological analysis. Apoptotic cells are indicated by red arrows. Scale bar = 100 µm. Quantification is shown on the right. **P < 0.01. D: working model. IFT88 knockout (KO) suppresses ciliogenesis in kidney tubule cells, causing the suppression of autophagy through the reciprocal regulation between cilia and autophagy that involves the Erk1/2-mammalian target of rapamycin (mTOR) signaling pathway. Autophagy is an intrinsic protective mechanism in kidney tubule cells. Consequently, IFT88 KO cells with suppressed autophagy become more sensitive to cisplatin injury. IFT88 OE, IFT88 overexpression.

DISCUSSION

We have previously found that cultured cells with short cilia are more vulnerable to cisplatin exposure and identified that Erk1/2 plays an important role in modulating cilium length (33). However, more detailed understanding about the association of cilia and apoptosis in cisplatin-induced kidney injury was elusive. In the present study, we show that kidney proximal tubule-specific ablation of IFT88 exaggerated AKI following cisplatin exposure. Mechanistically, our data suggest that defective autophagy in IFT88-deficient kidney cells and tissues contributes to their injury sensitivity.

Autophagy is a conserved cellular process of cytoplasmic degradation that is widely regarded as an intrinsic mechanism for cells to survive various stresses. For instance, after acute ischemic kidney injury or cisplatin insult, autophagy responds quickly so that renal cells can survive the harsh conditions (6, 23) as found in our study. The intriguing finding is the defective response of autophagy in IFT88 KO renal tubular cells, accompanied by deterioration of renal function and increased apoptosis, compared with the WT control. These observations suggest that cilia-short kidney tubular cells may sensitize themselves to cisplatin through autophagy inhibition. This hypothesis is further supported by the tat-beclin 1 peptide experiment, because autophagy enhancement with tat-beclin 1 peptide, a specific autophagy activator (35, 36), partially rescued IFT88 knockdown HK-2 cells. More specifically, reexpression of IFT88 increased autophagy and reduced cell death in IFT88 knockdown cells. How does IFT88 KO suppress autophagy? IFT88 KO may inhibit autophagy via a series of events (Fig. 7). IFT88 KO first shortens cilia and then may activate Erk1/2, as previously reported (33, 37). Another possibility is that shortened cilia suppress autophagy through a different mechanism, because there is a reciprocal regulation of cilia and autophagy (31, 32). Please note that Erk1/2 may regulate autophagy through mammalian target of rapamycin signaling as well (38–40). But is the pathway that IFT88 KO inhibits autophagy via primary cilia the only mechanism? Currently, we do not have the answer, because almost every cell has a cilium and we cannot modulate IFT88 without affecting cilia. Certainly, it cannot be excluded that IFT88 KO may inhibit autophagy directly or indirectly via other pathways, such as AMP-activated protein kinase and AKT (38–40). There is evidence that IFT88 deficiency exaggerates cisplatin-induced kidney injury via mitochondrial dysfunction (44). In this regard, it is interesting to note that there is an intimate connection between defective autophagy and mitochondrial dysfunction and an organelle cross-talk between cilia and mitochondria. Especially, mitophagy is a specific form of selective autophagy that clears up damaged or dysfunctional mitochondria. Defective autophagy or mitophagy leads to the accumulation of dysfunctional mitochondria, resulting in excessive oxidative stress, aberrant metabolism, and loss of cellular homeostasis. The importance of autophagy and mitophagy in mitochondrial quality control in kidney injury has been suggested by recent studies (8, 41–43).

In our study, we found no differences in cilium length between IFT88 WT and KO mice under normal conditions. This may suggest that IFT88 removal in proximal tubular cells does not affect cilia maintenance for the observed period. One possibility for this observation is that the role of IFT88 is compensated redundantly by other IFT proteins and regulators. Another possibility is that the effect of IFT88 on cilium length is too minimal to detect during the period from IFT88 KO (PEPCK-Cre expression at 3 wk after birth, Ref. 34) to experiments (2.5–3 mo). It is not surprising that after cisplatin administration, cilium length is shortened. A more interesting point is that cilium length is more significantly reduced in IFT88 KO cells after cisplatin treatment, suggesting that cisplatin has more impact on cilia defective in cilia components. This could be the consequence of more preapoptotic cells in IFT88 KO cells. Another interesting point is that the autophagy activator tat-beclin 1 promotes cilium length in normal cells but not in IFT88 knockdown cells, suggesting that IFT88 protein is more important for cilia assembly in cultured cells. Thus, without enough components for cilia assembly, it is not surprising that tat-beclin 1 failed to promote cilium length in IFT88 knockdown cells. Our experiments provide further insights about the role of cilia, particularly in cisplatin-induced AKI, and the significance of autophagy (Fig. 7D). The findings support the strategy of kidney protection during cisplatin chemotherapy by enhancing autophagy and ciliogenesis.

GRANTS

This work was supported in part by the Department of Veterans Affairs (Grants BX000319 and BX005236) and by the National Institutes of Health (Grants DK058831 and DK087843). Z.D. is a recipient of a Senior Research Career Scientist award from the Department of Veterans Affairs.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

S.W. performed experiments; S.W., S.Z., and Z.D. analyzed data; S.W. and Z.D. drafted the manuscript; all authors edited, revised, and approved the final manuscript.