Introduction

Ubiquitination has emerged as a widely acknowledged signaling mechanism that controls the destiny of proteins by finely tuning both their “quantity” and “quality.” This intricate regulatory process encompasses a diverse range of cellular functions, including proteasomal protein degradation, receptor endocytosis, DNA repair, gene transcription, kinase activation, protein-protein interactions, and the assembly of signaling complexes [1,2,3]. Ubiquitination is a dynamic and reversible form of post-translational modification, which is meticulously governed by an array of enzymes, comprising E1s, E2s, and E3s. In contrast, deubiquitination refers to the removal of ubiquitin molecules from ubiquitin-labeled proteins. This process is catalyzed by deubiquitinating enzymes (DUBs) and serves as a counterbalancing mechanism for E3 ligase activities. Owing to their remarkable substrate specificity, E3 ligases and DUBs are promising targets for cancer treatment [4, 5]. Currently, anticancer drugs targeting the proteasome, E3 ligases, and DUBs have being actively developed, with their therapeutic potential being supported by animal experiments and clinical trials. Ubiquitin-specific peptidase 26 (USP26), a member of the deubiquitinase family of ubiquitin-specific proteases, is a ubiquitin hydrolase that plays a critical role in regulating cellular protein stability and function. A growing body of evidence suggests that USP26 is significantly involved in tumorigenesis and tumor progression. Recent study have demonstrated that USP26 deubiquitinates and stabilizes TAZ, thereby promoting the progression of anaplastic thyroid carcinoma (ATC) [6]. Additionally, another study has indicated that USP26 enhances the stability of Snail through deubiquitination, leading to epithelial-mesenchymal transition (EMT) and the promotion of metastasis in esophageal squamous cell carcinoma [7]. However, the precise role of USP26 in colorectal cancer (CRC) remains unclear.

Mitophagy, a specialized form of autophagy, is critical for maintaining mitochondrial quality. It serves the purpose of eliminating damaged or aging mitochondria, enabling cells to maintain their optimal energy metabolism and functionality [8, 9]. Previous research has established a significant correlation among mitochondrial dysfunction, mitochondrial DNA mutations, and cancer initiation [10,11,12]. Impaired mitochondria can disrupt energy metabolism, trigger oxidative stress, and hinder apoptosis, thereby facilitating the survival and proliferation of tumor cells [10, 11, 13]. Damaged mitochondria can be selectively removed by promoting mitophagy, preventing the accumulation of dysfunctional mitochondria, and ensuring the presence of healthy, fully functional mitochondria within cells. This, in turn, contributes to sustaining the cellular energy equilibrium, mitigating oxidative stress, and preserving cellular viability, ultimately acting as a tumorigenesis suppressor.

Numerous signaling pathways regulate mitophagy in organisms. Among these pathways, the Pink1/Parkin pathway is the most widely acknowledged and embraced route in mitophagy research [9, 14,15,16]. PTEN-induced kinase 1 (PINK1) and the E3 ubiquitin ligase PRKN (Parkin), both localized to the mitochondria, are well-established co-actors involved in the orchestration of dysfunctional mitochondrial elimination via mitophagy. However, the connection between PRKN-mediated mitophagy and tumorigenesis remains a subject of debate, with no definitive evidence currently available regarding the role of PRKN in tumor development. Some studies have reported diminished expression of PRKN in specific types of tumors [15, 17,18,19,20]. Conversely, other studies have linked PRKN overexpression with the progression of certain tumors. These findings suggest that PRKN’s role may be complex and context-dependent across different tumor types. Consequently, further research is required to elucidate the precise roles and underlying mechanisms of PRKN in tumorigenesis. These research endeavors have the potential to unveil the intricate role of PRKN and its associated mitophagy pathway within the realm of tumor biology, providing invaluable theoretical foundations for the development of novel strategies for tackling cancer.

In the present study, we provide compelling evidence revealing the overexpression of USP26 in CRC, with its elevated levels being closely associated with an unfavorable prognosis for CRC patients. We substantiate that USP26 actively facilitates CRC tumorigenesis, operating in a manner reliant upon the presence of PRKN protein. Furthermore, we investigated the intricate mechanics of USP26 by demonstrating its direct interaction with PRKN, which ultimately promotes its deubiquitination. This mechanistic insight has profound implications in PRKN-mediated mitophagy. Our findings suggested that USP26 plays a critical role in colorectal tumorigenesis, highlighting its potential as a promising therapeutic target for cancer treatment.

Results

USP26 is upregulated in CRC

In our exploration of the role of deubiquitinases (DUBs) in CRC tumorigenesis, we conducted a DUBs expression assay utilizing the GEO database. Based on a comparative analysis with normal tissue, the mRNA levels of DUBs in tumors were categorized into 10 expression modules according to their distribution characteristics. Notably, Module 1 exhibited the most significant increase in DUB expression (Fig. S1A). Subsequently, we analyzed the expression heterogeneity fold changes (FC) of 9 DUBs in Module 1, highlighting the pronounced position of USP26 (Fig. S1B). In addition, a DUBs PCR array analysis was performed on three paired CRC samples. The results revealed elevated expression of USP26, OTUB2, and USP10 in the tumor group, with USP26 exhibiting the most substantial increase (Fig. S1C).

To further elucidate the role of USP26 in tumorigenesis, we analyzed USP26 expression levels in both adjacent and tumor tissues using data from the TCGA database. Our findings revealed a significant upregulation of USP26 expression in tumor tissues compared to normal tissues, with particularly pronounced differences in CRC samples (Fig. 1A, B). Kaplan–Meier analysis further demonstrated that individuals with CRC displaying heightened USP26 expression experienced a less favorable disease-free survival outcome than those with lower USP26 expression (Fig. 1C). We procured 12 pairs of CRC and adjacent tissues for further analysis, which consistently revealed elevated levels of USP26 expression within the tumor tissues compared to the adjacent tissues (Fig. 1D, E). Immunohistochemical (IHC) analysis was also conducted to assess USP26 expression, and the results consistently indicated increased USP26 expression in the tumor tissues (Fig. 1F, G). Collectively, these data suggest that USP26 may be involved in the initiation and progression of CRC.

Fig. 1: USP26 is overexpressed in colon carcinoma.
figure 1

A The expression of USP26 was analyzed in various cancer tissues and adjacent tissues in the TCGA database (*p < 0.05, ***p < 0.001, t-test). B Evaluation of USP26 expression levels from a COAD cohort in the TCGA database (**p < 0.01, t-test). C Kaplan–Meier plot illustrating disease-free survival rates for COAD patients. The data were sourced from the GEO database, and p-values were calculated using log-rank tests. D, E Assessment of USP26 protein expression in tumor and adjacent specimens from clinical patients. Results are presented as mean ± SD, **p < 0.01. F, G Immunohistochemical (IHC) staining of USP26 in 23 paired of clinical specimens, along with statistical quantification results. Results are presented as mean ± SD, ***p < 0.001.

Full size image

Inhibition of USP26 attenuates CRC tumorigenesis in vivo and in vitro

Given the aberrant expression of USP26 in CRC, we investigated whether USP26 expression influenced the development and progression of CRC. To validate these hypothesis, we established stable USP26 knockdown cell lines (shUSP26#1 and shUSP26#2) and stable USP26 overexpression cell lines in SW48 and LoVo cells using lentiviral infection and confirmed their efficiency through western blot analysis (Fig. 2A and Fig. S2A). To assess the impact of USP26 on CRC cell proliferation, we conducted CCK8 assays, which revealed a decrease in the viability of USP26 knockdown cells and an increase in the viability of USP26 overexpressing cells, suggesting a direct influence of USP26 on CRC cell proliferation (Fig. 2B and Fig. S2B). These findings were consistent with the results of EDU assays in SW48 and LoVo cells, which demonstrated that silencing USP26 led to a reduction in cell proliferation, and overexpression of USP26 contributed to a significant increase in cell proliferation (Fig. 2C, D, and Fig. S2C, D). Furthermore, our investigations showed that USP26 knockdown impaired the clonogenic potential of tumor cells, whereas overexpression of USP26 increased their colony-forming ability (Fig. 2E, F, and Fig. S2E, F). To better replicate the tumor growth dynamics, we performed an in vivo analysis. Xenograft mouse models were established utilizing stable USP26 knockdown cell lines. In agreement with the in vitro results, USP26 suppression was found to diminish the growth rate, tumor size, and tumor weight of the xenografts (Fig. 2G–I).

Fig. 2: Inhibition of USP26 moderates CRC tumorigenesis in vivo and in vitro.
figure 2

A Western blot analysis of USP26 expression with USP26 knockdown in SW48 and LoVo cells. B Cell viability of SW48 and LoVo cells stably expressing shNC or shUSP26 was assessed using CCK8. C, D Representative images from the EDU assay performed on SW48 and LoVo cells in response to USP26 knockdown. Scale bars, 100 μm. E, F Evaluation of colony formation capability in SW48 and LoVo cells stably expressing shNC or shUSP26. Results represent the mean ± SD of three experiments (**P < 0.01, n = 3). G Xenograft tumors derived from LoVo cells, with or without USP26 depletion. H Quantification of tumor weights generated in (G) (n = 5). I Quantification of tumor volume generated in (G) (n = 5).

Full size image

USP26 suppresses the activation of mitophagy in CRC cells

Mitophagy is an evolutionarily conserved mechanism aimed at clearing damaged mitochondria, thereby curtailing the build-up of dysfunctional mitochondria, cellular oxidative stress, and genomic instability, all of which collectively suppress tumorigenesis [15]. Ubiquitination and deubiquitination play pivotal roles in mitophagy, although their underlying mechanisms have not been fully elucidated [1, 21]. In this study, we conducted a gene set enrichment analysis employing publicly available datasets and discerned a negative correlation between USP26 upregulation and mitophagy (Fig. 3A). To substantiate this discovery, we conducted western blot analysis in SW48 and LoVo cells, which revealed that USP26 inhibition resulted in decreased expression of the autophagy receptor P62, concomitant with an increase in LC3-II levels (Fig. 3B). Conversely, USP26 overexpression led to an elevated expression of the autophagy receptor P62 and a reduction in LC3-II levels (Fig. S3A). To confirm the influence of USP26 on mitophagy, we conducted a seahorse experiment. These experiments demonstrated an overall decline in oxygen consumption rate (OCR) within USP26-silenced cells, with significant reductions observed in maximal respiration, while an enormous increase was observed when USP26 was overexpressed (Fig. 3C, D, and Fig. S2B, C). Furthermore, USP26-silenced cells exhibited reduced mitochondrial ROS levels and diminished mitochondrial membrane potential, both of which are key indicators of mitophagy activation (Fig. 3E–H). USP26-overexpressed cells exhibited increased mitochondrial ROS levels and enhanced mitochondrial membrane potential (Fig. S2D–G). Electron microscopy revealed more autophagosomes in USP26-silenced LoVo cells than those in control cells (Fig. 3I). Additionally, the introduction of autophagy inhibitor 3-methyladenine (3-MA) reversed USP26-induced proliferation of SW48 and LoVo cells (Fig. 3J–M). In summary, these findings collectively affirm that USP26 silencing enhances mitophagy in CRC cells, maintains ROS homeostasis via mitophagy, and ultimately suppresses tumor cell proliferation.

Fig. 3: USP26 depletion promotes the activation of mitophagy.
figure 3

A GSEA analysis in two distinct GEO datasets (GSE29623 and GSE17526) reveals a significant association between USP26 levels and the activation of mitophagy signaling. B Western blot analysis of indicated proteins following USP26 intervention in SW48 and LoVo cell lines. C, D Measurement of oxygen consumption rate (OCR) in SW48 and LoVo cells stably expressing shNC or shUSP26 under basal conditions and in response to oligomycin, FCCP, and rotenone. E, F Quantification of mitochondrial superoxide (Mito-sox) staining in SW48 and LoVo cells transduced with control or USP26 shRNA, analyzed by flow cytometry (n = 3). G, H Quantification of JC-1 staining in SW48 and LoVo cells transduced with control or USP26 shRNA, analyzed by flow cytometry (n = 3). I Representative images from transmission electron microscopy showing autophagosomes (arrows) in LoVo cells with or without USP26 depletion. Scale bars: left, 2 μm; right, 500 nm. J, K SW48 and LoVo cells stably expressing shUSP26 were treated with 3-MA (an inhibitor of mitophagy), and cell viability was assessed using CCK8. L, M SW48 and LoVo cells stably expressing shUSP26 were treated with 3-MA, and colony formation capability was measured. Results are presented as mean ± SD, *p < 0.05, **p < 0.01, ***p < 0.001.

Full size image

USP26 directly interacts with and deubiquitinates PRKN

To further investigate the regulatory mechanism of USP26 in mitophagy, we conducted mass spectrometry analysis and applied mass spectrometry scoring to identify the proteins associated with USP26. Given the role of USP26 in the inactivation of mitophagy, we specifically selected proteins related to mitophagy for our study. This endeavor led to the identification of the protein PRKN (Table S1). Subsequently, we constructed overexpression plasmids, HA-USP26 and Flag-PRKN, and confirmed their interaction by IP (Fig. 4A). Moreover, we executed immunoprecipitation (IP) experiments to probe the physical interactions between endogenous USP26 and PRKN, yielding consistent results (Fig. 4B). To elucidate the specific USP26 domains involved in the interaction with PRKN, we generated a USP26 deletion mutant. Notably, full-length PRKN exhibited robust binding to the USP domains of USP26 (Fig. S4A, B). Conversely, among the different PRKN domains, including PRKN UBL, RING0, RING1, and RING2, only the UBL domain interacted with USP26 (Fig. S4C, D). Furthermore, in vitro GST-pull-down assays further delineated that USP26 directly interacted with PRKN (Fig. 4C). To further corroborate these findings, we used immunofluorescence to observe the colocalization of USP26 and PRKN in SW48 and LoVo cells (Fig. 4D). Collectively, these results provided compelling evidence that USP26 directly interacts with PRKN in CRC cells.

Fig. 4: USP26 directly interacts with and deubiquitinates PRKN.
figure 4

A Exogenous interaction between USP26 and PRKN was determined in HEK293T cells after co-transfection with Ha-USP26 and Flag-PRKN. B The endogenous interaction between USP26 and PRKN was demonstrated in the LoVo cell line using antibodies against USP26 and PRKN, respectively. C An in vitro binding assay was performed to verify the direct binding of USP26 and PRKN. D Confocal images show the relative localization of USP26 and PRKN in LoVo and SW48 cells. Scale bar, 20 μm. E The ubiquitination levels of PRKN decreased in HEK293T cells co-transfected with GFP-USP26 and Ha-PRKN. F USP26 removes PRKN-Ub conjugates in an in vitro ubiquitination assay. G Flag-PRKN and HA-Ub were co-expressed with USP26 or USP26 Mut in HEK293T cells. Cell lysates were subjected to ubiquitination assay and the ubiquitination level of PRKN was detected by Western blot.

Full size image

Because USP26 is a DUB, we hypothesized that it might deubiquitinate the PRKN protein. To substantiate this hypothesis, we conducted in vivo and in vitro ubiquitination assays, both of which revealed that USP26 had the capability to reduce the level of PRKN ubiquitination (Fig. 4E, F). Furthermore, we determined whether USP26 regulated deubiquitination through its deubiquitinase activity. We generated a USP26 mutant by replacing the catalytic cysteine residue with a serine residue (C304S). We co-transfected USP26 or USP26 mutants with Flag-PRKN and HA-ubiquitin into HEK293T cells. Ectopic expression of USP26 notably reduced the ubiquitination level of PRKN protein, but no such effect was observed in the USP26 mutant (Fig. 4G).

As all know, PRKN is an E3 ubiquitin ligase. We found that USP26 could deubiquitinate PRKN, leading to a decrease in its ubiquitination level. Therefore, we examined the ubiquitination levels and expression of PRKN-recognized substrates (RHOT1 and VDAC1) [22, 23]. We found that in USP26-silenced SW48 and LoVo cells, the ubiquitination levels of RHOT1 and VDAC1 were significantly increased and their protein expression levels decreased (Fig. S5A–C). Similarly, in USP26 overexpressing cells, the ubiquitination levels of RHOT1 and VDAC1 reduced significantly, whereas their protein expression levels increased. This suggests that USP26 decreases the ubiquitination level of PRKN, weakening its E3 ubiquitin ligase activity (Fig. S5D–F).

The ubiquitin-proteasome system (UPS) serves as the principal pathway for protein degradation within cells and is responsible for the degradation of over 80% of cellular proteins. Therefore, we used the protein synthesis inhibitor cycloheximide (CHX) to explore the impact of USP26 on the stability of the PRKN protein. Intriguingly, our findings indicated that the silencing of USP26 led to increased levels and enhanced stability of the PRKN protein (Fig. S5G). To explore the underlying mechanisms, we investigated two major pathways of intracellular protein degradation: the UPS and autophagy. Surprisingly, proteasome inhibitors, such as MG132 and lactacystin, failed to impede the degradation of PRKN by USP26 (Fig. S5H). Conversely, the autophagy inhibitor 3-MA demonstrated the capacity to inhibit the clearance of PRKN by USP26 (Fig. S5H). In conclusion, these findings reveal a novel mechanism by which USP26 targets PRKN for degradation via autophagy. However, this requires further investigation.

USP26 removes K27-linked ubiquitin conjugates from PRKN at K129

Previous studies have shown that all seven internal lysine residues (Lys 6, 11, 27, 29, 33, 48, and 63) of ubiquitin can potentially serve as sites for chain extension, mediating various outcomes, including proteasomal degradation pathways and non-degradative functions [21]. To identify the preferred ub sites targeted by USP26 for interaction with PRKN, we generated a series of hemagglutinin (HA)-tagged ub mutants. In these mutants, only one lysine (K) residue was preserved, whereas other lysine residues were mutated to arginine. These mutants were co-transfected with GFP-USP26 and ub into HEK293 cells for IP experiments, revealing that USP26 preferentially removed the K27 ubiquitin chain conjugated to PRKN (Fig. 5A, B). Subsequently, we validated this ub chain in USP26 knockdown cells, both with and without CCCP treatment (Fig. 5C).

Fig. 5: USP26 removes K27-linked ubiquitin conjugates from PRKN at K129.
figure 5

A, B The levels of different K-linked PRKN ubiquitination upon USP26 overexpression were assessed by Western blotting using an anti-HA antibody. C K27-linked PRKN ubiquitination in LoVo cells after USP26 depletion, with or without CCCP treatment for 1 h. D Identification of potential ubiquitination sites on PRKN through mass spectrometry and secondary mass spectrometry analysis of one potential ubiquitination residue. E HEK293T cells were transfected with Flag-PRKN(WT), Flag-PRKN(K27R), Flag-PRKN(K76R), Flag-PRKN(K129R), and Flag-PRKN(K408R), with or without GFP-USP26, to examine changes in ubiquitination. F HEK293T cells were transfected with Flag-PRKN(WT) and Flag-PRKN(K129R) in combination with shNC or shUSP26 to assess PRKN ubiquitination levels.

Full size image

To identify the specific sites of PRKN deubiquitination by USP26, Flag-PRKN protein was immunoprecipitated from HEK293T cells and subjected to mass spectrometry (Fig. S6A). Four lysine sites (K27, K76, K129, and K408) were found to be ubiquitinated, and the ubiquitination status at the sites of PRKN was confirmed using secondary mass spectrometry (Fig. 5D). Next, we explored potential target sites for the USP26-dependent deubiquitination of PRKN. Plasmids carrying K-to-R mutations at these four sites were constructed and co-transfected with USP26 to assess the ubiquitination status of PRKN by IP. The overexpression of USP26 downregulated the ubiquitination levels of the K27R, K76R, and K408R mutants, but had no effect on the ubiquitination level of the K129R mutant (Fig. 5E). Consistent with these experimental results, USP26 knockdown did not increase ubiquitination of the K129R mutant (Fig. S6B). Our study demonstrated that USP26 influences the ubiquitination of PRKN at K129.

Because USP26 primarily removes the K27 ubiquitin chain conjugated to PRKN, we aimed to confirm whether USP26 removes the K27 ubiquitin chain at lysine 129 of PRKN. Western blotting demonstrated that the Lys129 mutation led to a significant decrease in the levels of K27-linked ubiquitin conjugates in PRKN (Fig. 5F). These findings indicate that the ubiquitination of PRKN at Lys129 is regulated by USP26 through K27-linked ubiquitin. Next, we determined the functional significance of ubiquitination at K129 in PRKN. To achieve this, we established a cell line with endogenous PRKN knockdown and rescued it with either wild-type PRKN or K129R-mutant PRKN. K-to-R mutations are commonly considered ubiquitination-defective mutations. Immunoblotting results revealed that comparing to wild-type PRKN, K129R partially attenuated PRKN-mediated mitophagy in SW48 and LoVo cells (Fig. S6C). K129R displayed partial restoration of mitochondrial ROS levels in SW48 and LoVo cells (Figs. S6D, 6E). Furthermore, immunofluorescence colocalization experiments demonstrated that the ubiquitination-defective state of K129 mutant led to reduced translocation of PRKN to the mitochondria in LoVo cells (Fig. S6F).

Fig. 6: USP26-triggered tumorigenesis is mediated by PRKN.
figure 6

A Western blot analysis of indicated proteins with or without shUSP26 and shPRKN in SW48 and LoVo cell lines. B, C Measurement of oxygen consumption rate (OCR) in indicated cells. D, E Quantification of Mito-sox staining in indicated cells by flow cytometry (n = 3). F, G Detection of cell viability in indicated cells using the CCK8 assay. H, I Measurement of colony formation capability in indicated cells. J The images of xenograft tumors of nude mice. K Quantification of tumor weights generated in (J) (n = 5). L Quantification of tumor volume generated in (J) (n = 5). The results are presented as mean ± SD, **p < 0.01, ***p < 0.001, t-test.

Full size image

USP26-triggered tumorigenesis is mediated by PRKN

To further confirm the dependence of tumor formation on PRKN mediated by USP26, we performed concurrent knockdown of PRKN in USP26-silenced SW48 and LoVo cells. Western blot analysis revealed that the silencing of USP26 increased the expression of LC3-II and decreased the levels of the autophagy receptor p62, and this effect was rescued by silencing PRKN (Fig. 6A). To investigate the effect of PRKN on USP26-mediated mitophagy, we conducted a Seahorse experiment. The results showed that the silencing of PRKN almost completely reversed the reduction in OCR caused by USP26 silencing (Fig. 6B, C). Consistent with this finding, PRKN silencing attenuated the decrease in mitochondrial ROS levels induced by USP26 silencing (Fig. 6D, E). We performed cell viability assays and colony formation experiments using these cells. The results demonstrated that USP26 knockdown reduced the activity of CRC cells and weakened their clonogenic capacity, whereas the simultaneous silencing of PRKN blocked this attenuating effect (Fig. 6F–I). Supported by these data, we further implanted these cells into nude mice to establish a subcutaneous tumor model, and the results demonstrated that USP26 silencing significantly suppressed tumor progression, which was inhibited by PRKN intervention (Fig. 6J–L). Collectively, our data indicated that PRKN is necessary for USP26-mediated tumor formation.

Expression of PRKN is negatively correlated with USP26 in CRC

In our previous study, we observed significant upregulation of USP26 in various cancers, including CRC and lung adenocarcinoma. Moreover, Kaplan–Meier analysis demonstrated a positive correlation between high USP26 expression and poor survival prognosis in CRC. To further explore the relationship among USP26, PRKN, and colon tumors, we performed IHC staining using anti-USP26 and anti-PRKN antibodies in a CRC tissue microarray. Our study revealed higher expression levels of USP26 in CRC tissues than in adjacent tissues (Fig. 7A). Additionally, patients with low USP26 expression had significantly longer overall survival compared to those with high USP26 expression (Fig. 7B). Furthermore, IHC staining using anti-PRKN antibodies on the CRC tissue microarray showed a negative correlation between PRKN and USP26 protein levels in different specimens (Fig. 7C–E). This correlation further validated the regulatory relationship between USP26 and PRKN.

Fig. 7: Expression of PRKN is negatively correlated with USP26 in CRC.
figure 7

A USP26 IHC staining scores were detected in the tumor and adjacent tissues (n = 66). Student’s two-tailed t-test, p < 0.001. B Kaplan–Meier analysis of overall survival in a set of 66 colorectal cancer patients according to USP26 expression. Log-rank test, p = 0.035. C Representative images of IHC staining for USP26 and PRKN. D, E Quantitative IHC staining scores show the correlation between USP26 and PRKN using a microarray of colorectal cancer specimens. F The working model illustrating USP26-mediated PRKN K129 deubiquitination.

Full size image

In conclusion, our findings suggest that USP26-mediated deubiquitination of PRKN is a crucial factor in the development of CRC and that USP26 levels may serve as a predictive marker for cancer occurrence.

Discussion

Post-translational protein modifications play a pivotal role in the regulation of protein activity and tumorigenesis [24, 25]. Ubiquitination plays a central role in governing various vital biological processes, including protein degradation, stability, participation in signaling pathways, and DNA repair mechanisms. These processes are essential for maintaining normal cellular functions and have a significant effect on tumorigenesis. Notably, protein ubiquitination is a dynamic and reversible process that continuously involves cycles of ubiquitination and deubiquitination. E3 ubiquitin ligases are responsible for selectively mediating the ubiquitination of specific substrates while DUBs function to counteract this process by removing ubiquitin moieties. Due to their ability to precisely target substrates, both E3 ligases and DUBs hold great promise as potential therapeutic targets in cancer treatment [4, 5].

Currently, anticancer drugs have being actively developed that target various components of the ubiquitination and deubiquitination machinery, along with proteasome [26,27,28,29]. Preclinical studies and clinical trials have demonstrated its potential as an effective cancer treatment. In this study, we provide compelling evidence highlighting the correlation between increased USP26 expression and unfavorable prognosis for patients with CRC. Our investigations revealed that the loss of USP26 function hindered tumor formation both in vitro and in vivo. We observed a significant reduction in cell proliferation when USP26 is deficient, underscoring its critical role in promoting CRC tumorigenesis. These findings suggest that USP26 is a promising therapeutic target for CRC treatment.

Mitophagy and ubiquitination are interconnected cellular processes. Ubiquitination plays a crucial role in regulating the initiation and execution of mitophagy. During this process, proteins located on the mitochondrial surface undergo ubiquitination, enabling them to associate with autophagosomal membrane structures and form mitophagosomes. Subsequently, these mitophagosomes merge with lysosomes, resulting in mitochondrial degradation and recycling [30, 31]. A growing body of evidence supports the involvement of DUBs in the regulation [32,33,34,35]. For instance, previous studies have reported that USP8 interacts with PINK1 and Parkin, two critical regulatory factors, and contributes to the control of proteins related to mitochondrial structure, such as Drp1 and Mfn2 [36]. Moreover, USP30 and USP33 have been identified as DUBs that removes ubiquitin from PRKN, thereby inhibiting the selective clearance and degradation of mitochondria [37, 38]. In this study, we revealed a direct interaction between USP26 and PRKN. We demonstrated that USP26 acts as a deubiquitinase of PRKN, resulting in reduced PRKN activity and expression. Notably, PRKN is long-lived and is relatively resistant to degradation by cellular proteasomes. Additionally, PRKN undergoes atypical self-ubiquitination, which may partly account for the absence of further stabilization of PRKN upon USP26 overexpression [39,40,41]. Ubiquitin linkages, specifically the K6, K27, K29, and K63 ubiquitin chains, are associated with processes such as transport, signal transduction, DNA repair, and autophagy [42,43,44,45,46,47]. Our research revealed that USP26 could specifically remove the K27 ubiquitin chain from the PRKN-Ub conjugate, leading to the inhibition of mitophagy. Furthermore, we elucidated the functional significance of distinct ubiquitinated amino acid residues in PRKN. These findings establish a theoretical foundation for the development of potential USP26 inhibitors and inhibitory peptides aimed at disrupting the interaction between USP26 and PRKN as well as deubiquitinating specific residues. Moreover, our investigations revealed that USP26 primarily exerted its effects on CRC by modulating the protein PRKN. Inhibition of protein synthesis has been shown to prolong PRKN half-life in USP26-deficient cells. Additionally, decreased PRKN expression in CRC cells negated the inhibitory effects of the loss of USP26. These consistent results were corroborated by in vivo tumor formation experiments, providing further confirmation of USP26’s role as a tumor growth promoter.

In summary, our study sheds light on the role of USP26 in promoting tumor formation by modulating the PRKN-mediated mitophagy pathway. These findings offer fresh perspectives on the involvement of USP26 in the mitophagy signaling pathway within the context of CRC. These studies underscore the importance of mitophagy in the development of colorectal tumors. Consequently, the modulation of USP26 activity or gene expression, along with disruption of the interaction between USP26 and PRKN using interfering peptides, would emerge as a promising therapeutic strategy for CRC treatment.

Materials and methods

Cell lines and cell culture

Human cell lines SW48, LoVo, and HEK293T were procured from the American Type Culture Collection. Cells were cultured in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 50 U/mL penicillin-streptomycin. All cell cultures were maintained at 37 °C in a 5% CO2 incubator.

Reagents

CHX and proteasome inhibitor MG132 were purchased from Selleck Chemicals (Houston, TX, USA). The autophagy inhibitor 3-MA was purchased from Sigma-Aldrich (St. Louis, MO, USA). The antibody against USP26 (ab188239) was obtained from Abcam (Cambridge, UK). Antibodies against PRKN (#4211), Ha (#3724), Flag (#14793), LC3I/II, and GAPDH (#5174) were purchased from Cell Signaling Technology (Danvers, MA, USA). anti-P62 antibody (#51064-2-AP) was purchased from Proteintech (Wuhan, Hubei, China).

Plasmid and stable cell lines construction

The expression plasmids pcDNA3.1-USP26-HA, pcDNA3.1-USP26-GFP, and pcDNA3.1-PRKN-Flag were constructed by inserting the PCR products derived from HEK293T cell cDNA into pcDNA3.1 vector. Specifically, the pcDNA3.1-PRKN K129R-Flag vector was generated using the Mut Express II Fast Mutagenesis Kit V2 (Vazyme, China) based on the following sequences: 5′-ACAGCAGGAGGGACTCACCACCAGCTGGAAGT-3′ and 5′-TGAGTCCCTCCTGCTGTCAGTGTGCAGAATGA-3′. USP26- and PRKN-knockout cell lines were established using a lentiviral system (pLVX, psPAX2, and pMD2.G). Lentivirus particles were produced in HEK293T cells. SW48 and LoVo cells were infected and subsequently selected using 2 µg/ml of puromycin over 2 weeks. To generate lentiviral RNA (shRNA) targeting USP26, two distinct shRNA sequences were used: shUSP26#1: 5′-GATTGTTCGAGGTGTGTAAGCTTCA AGAGAGCTTACACACCTCGAACAATC-3′, shUSP26#2:5′-GCAAGACTGGGATATC TAAGTTTCAAGAGAACTTAGATATCCCAGTCTTGC-3′. Similarly, two different shRNA sequences were used for lentiviral shRNA targeting PRKN: shPRKN#1: 5′-GGAGGTGGTTGCTAAGCGACATTCAAGAGATGTCGCTTAGCAACCACCTCC-3′, and shPRKN#2: 5′-GGTCAAGAAATGAATGAAACTTTCAAGAGAATGGGCATTCA TTTCTTGACC-3′.

Immunohistochemistry assay

Paired carcinoma and adjacent tissue specimens were collected from patients diagnosed with CRC, who underwent surgical resection at the Department of Gastrointestinal Surgery, Wuhan Tongji Hospital. Additionally, a human CRC tissue microarray comprising 79 pairs primary tumors and adjacent normal tissues was procured from Wuhan Baiqiandu Technology Co., Ltd. (Cat. #80A1) and was used to assess USP26 and PRKN expression. The evaluation of IHC staining was performed independently by two gastrointestinal pathologists. An immunoreactive score was used to quantify IHC staining results. Staining intensity was categorized as follows: 0, no staining; 1, weak staining; 2, moderate staining; and 3, strong staining. The percentage of positively stained cells was rated on a scale of 1–4 (1, <10%; 2,11–50%; 3,51–75%; and 4, >75%). The final IHC score was calculated by multiplying the percentage of positively stained cells and the staining intensity score.

Cell counting kit-8 cell viability and colony formation assay

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8) assay (MCE). CRC cells (3 × 103 cells per 100 μL) were seeded into 96-well plates and allowed to incubate overnight. At the designated time points, 10 μL of CCK-8 solution, mixed with 90 μL of medium, was added to each well. Following a 2 h incubation at 37 °C, the optical density of the cells was measured at 450 nm.

For the colony formation assay, CRC cells were transformed into single-cell suspensions and plated in six-well plates at a density of 500 cells per well. The cells were then incubated at 37 °C in a 5% CO2 environment for a period of 14 days. Following incubation, cells were fixed with 4% paraformaldehyde for 30 min and stained with 0.1% crystal violet for 30 min at room temperature. Finally, colonies were counted and photographed.

Immunofluorescence assay

Cells were plated on coverslips and fixed in 4% paraformaldehyde for 30 min at room temperature. Subsequently, the cells were permeabilized using permeabilization buffer containing Triton X-100 for 5 min and blocked with 1% bovine serum albumin for 1 h at room temperature. Following this, the cells were subjected to overnight staining at 4 °C with the following primary antibodies: anti-USP26 (1:200, PM036, MBL), anti-PRKN (1:200, ab184699, Abcam), and anti-TOMM20 (1:200, 4377s, Cell Signaling Technology). After staining, the cells were rinsed twice with TBST and subsequently incubated for 1 h at room temperature with the following secondary antibodies: anti-rabbit Alexa Fluor Plus 488 (1:200, SA00006-2, Proteintech), anti-rabbit Alexa Fluor Plus 594 (1:200, SA00013-4, Proteintech), anti-mouse Alexa Fluor Plus 488 (1:200, SA00006-1, Proteintech), anti-mouse Alexa Flour 594 (1:200, SA00013-3, Proteintech), and DAPI (C1002, Beyotime). Images were acquired using a confocal laser scanning microscope (Olympus FLUOVIEW FV1000) equipped with a 60× oil objective, and the FV31S-SW Viewer software was used for subsequent analysis.

Western blot and immunoprecipitation

Cells were lysed in NP40 buffer containing 1% protease inhibitor cocktail and 1% phosphatase inhibitor. For Western blot analysis, proteins were separated by SDS-PAGE and transferred onto a PVDF membrane (Millipore, Billerica, MA, USA). After transfer, the membrane was blocked with 5% BSA at room temperature for 1 h. Subsequently, the membranes were incubated with specific primary antibodies, followed by incubation with horseradish peroxidase-conjugated secondary antibodies. Protein bands were visualized using chemiluminescence. For IP, the cell lysates were incubated with the corresponding antibodies and Protein A/G Magnetic Beads (Santa Cruz Biotechnology). This was followed by immunoblotting using the indicated antibodies. A comprehensive list of all antibodies used for western blot analysis is provided in Supplementary Table S1.

Seahorse analysis

The oxygen consumption rate (OCR)(pmol/min) was quantified using a Seahorse XF-96 metabolic extracellular flux analyzer through a mitochondrial stress test. To initiate the assay, cells (2 × 105 cells/well) were seeded in Seahorse 96-well plates coated with poly-L-lysine (Sigma–Aldrich). The cells were cultured in serum-free XF Base Media (Agilent) supplemented with 25 mM glucose, 1 mM pyruvate, and 2 mM glutamine for 1 h, thereafter the following compounds were added sequentially: 2.5 μM oligomycin, 2 μM fluoro-carbonyl cyanide phenylhydrazone (FCCP), and 0.5 μM rotenone plus 0.5 μM antimycin A.

Mouse tumor xenograft model

Female BALB/c nude mice aged 6–8 weeks were randomly used to establish xenograft models. On day 0, the mice were subcutaneously inoculated with 1 × 106 CRC cells. Tumor dimensions were recorded every 2 or 3 days using a digital caliper, and the results were expressed as tumor volume (calculated as width2 × length/2). For ethical considerations, mice were euthanized via CO2 inhalation after 24 or 28 days.

Mass spectrometry detection

Mass spectrometry was conducted at the National Protein Science Facility of the School of Life Sciences, Tsinghua University. Following IP, the resulting products were separated by SDS-PAGE and visualized using Coomassie blue staining. Specific protein bands of interest were excised, digested, reconstituted in 0.1% trifluoroacetic acid, and subsequently loaded onto a mass spectrometer (Orbitrap Fusion; Thermo Fisher Scientific) for mass spectrum analysis. The resulting MS data were then scrutinized against the target protein database sourced from UniProt using the in-house Proteome Discoverer software (Version PD1.4, Thermo Fisher Scientific, USA).

Statistical analysis

All statistical analyses were performed using the GraphPad Prism 8.0.2 software (GraphPad Software, Bethesda, MD, USA). Continuous data were reported as mean ± standard deviation and were subjected to statistical evaluation using either Student’s t-test (two-tailed) or analysis of variance, as appropriate. Survival analysis was conducted using Kaplan–Meier curves, and significance was determined using the log-rank test. P was set at <0.05 significant.