USP10-mediated deubiquitination and activation of KRAS mutants promotes colorectal cancer via a novel USP10/KRAS positive feedback circuit

Abstract

Kirsten rat sarcoma viral oncogene homolog (KRAS) mutation is associated with the poor prognosis of colorectal cancer (CRC) patients, but the therapeutic strategies targeting KRAS are limited, and novel intervention strategies are urgently needed. The dysfunction of deubiquitinases (DUBs) is widely involved in the progression of malignancy, and DUBs are considered ideal anti-tumor targets due to their well-defined structures and catalytic sites. In our study, through DUB inhibitors screening and liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis, we identified that ubiquitin-specific protease 10 (USP10) functions as a potent DUB regulating KRAS mutants' activity. Mechanistically, USP10 directly binds to and promotes KRAS variants' activity across different mutants by removing the latter’s non-proteolytic ubiquitination chains mainly containing K6, K11, K27 and K29-linkage; while the activated KRAS mutants reciprocally upregulate USP10 levels by phosphorylating the latter at Thr42/Ser337, therefore forming a positive feedback circuit and synergistically promoting KRAS-mutant CRC growth. Moreover, we found that USP10 is elevated in KRAS-mutant CRC tissues and depletion of USP10 preferentially impeded KRAS-mutant CRC growth in vitro/in vivo. Our findings not only uncover the critical roles of the USP10/KRAS positive feedback circuit in promoting KRAS-mutant CRC growth, but also offer novel therapeutic strategies for CRC patients harboring KRAS variants across different mutants by targeting USP10.

Graphical abstract

USP10 promotes KRAS mutants’ activity by removing K6/K11/K27/K29-linked polyubiquitination, and the activated KRAS mutants reciprocally upregulate USP10 by phosphorylating Thr42/Ser337, therefore forming a positive feedback circuit and promoting KRAS-mutant colorectal cancer growth.

1. Introduction

Colorectal cancer (CRC) is the third most common cancer and the second leading cause of cancer death worldwide^1,2. Despite great progress in targeted therapies, effective therapeutic strategies targeting Kirsten rat sarcoma viral oncogene homolog (KRAS) mutant CRC remain challenging3, 4, 5. Approximately 50% of CRC patients harbor KRAS mutations, and these patients show poor response to current chemotherapy and molecular-targeted drugs, resulting in poor prognosis6, 7, 8, 9. Therefore, it is urgent to explore and develop novel and potent therapeutic strategies for CRC patients harboring KRAS-mutants.

Despite the success of KRAS^G12C covalent inhibitors, sotorasib and adagrasib, which benefit those patients harboring KRAS^G12C mutations^8,10, 11, 12, the incidence of KRAS^G12C only accounts for ∼3% in CRC patients. Moreover, ∼50% of KRAS^G12C patients rapidly develop acquired resistance after treatment with sotorasib13, 14, 15, indicating most of the CRC patients cannot benefit from KRAS^G12C covalent inhibitors. Son of sevenless 1 (SOS1) functions as the key effector promoting KRAS activity by modulating guanosine diphosphate/guanosine triphosphate (GDP/GTP) exchange and is considered an ideal target for KRAS intervention. However, BI-1701963, a previously considered promising SOS1 inhibitor, was ultimately terminated in phase I due to restraining normal KRAS activity, suggesting the impairment of normal KRAS activity leads to severe toxicity. Therefore, the intervention strategies of drugging KRAS mutants but sparing wild-type KRAS are urgently needed.

Ubiquitination modifications are the key processes in regulating KRAS activation, influencing the carcinogenic activity of KRAS by regulating its abundance or binding affinity to downstream effectors^7,16,17. Ubiquitination modifications are primarily controlled by E3 ubiquitin ligases and deubiquitinating enzymes (DUBs), which are considered ideal drug targets due to their well-defined structures and catalytic active sites^18,19. To date, several E3 ligases that regulate KRAS have been identified, including leucine zipper-like transcriptional regulator 1 (LZTR1)²⁰, beta-transducing repeat-containing protein (β-TRCP)²¹ and WD40-repeat protein 76 (WDR76)²², but developing the activators of these E3 enzymes remain a huge challenge. To date, the DUBs that directly regulate KRAS mutants remains less understood. A previous study reported that OTU domain-containing ubiquitin aldehyde-binding protein 1 (OTUB1) is involved in regulating KRAS²³, but OTUB1 modulates the membrane localization and activation of wild-type KRAS, not KRAS mutants, by inhibiting the enzymatic activity of the E2 enzymes ubiquitin-conjugating enzyme 13 (UBC13) and ubiquitin-conjugating enzyme E2 D3 (UBE2D3). Recently, ubiquitin-specific proteases 7 (USP7) was reported to stabilize KRAS and promote non-small cell lung cancer cell proliferation²⁴, but wild-type KRAS is also regulated by USP7, suggesting the potential risks of targeting USP7. So far, the specific DUBs that directly regulate KRAS mutants but spare wild-type KRAS remain unclear. Therefore, exploring DUBs that directly regulate KRAS mutants’ activity while sparing wild-type KRAS is expected to provide potential therapeutic targets for KRAS-mutant CRC.

Ubiquitin-specific proteases 10 (USP10) belong to the USP family, the largest subfamily of DUBs. In recent years, the roles of USP10 in various malignant tumors have been gradually reported, revealing its dual functions. USP10 acts as a tumor suppressor by regulating Krüppel-like factor 4 (KLF4) stability and suppressing lung tumorigenesis²⁵. In addition, USP10 antagonizes the transcriptional activity of c-Myc by preventing sirtuin 6 (SIRT6) ubiquitination and subsequent proteasomal degradation²⁶. Conversely, USP10 also functions as an oncogenic factor. Weisberg et al.²⁷ found that USP10 promoted acute myeloid leukemia progression by deubiquitinating and stabilizing oncogenic FMS-like tyrosine kinase-3 (FLT3); inhibition of USP10 significantly induced the FLT3 ubiquitination and degradation. Consistent with these findings, our previous study also revealed that USP10 promotes hepatocellular carcinoma proliferation by preventing the ubiquitination and degradation of Yes-associated protein (YAP) and its paralog transcriptional coactivator with PDZ-binding motif (TAZ)²⁸. Yuan et al.²⁹ found that USP10 is a critical DUB regulating Tumor protein 53 (TP53) stability by counteracting MDM2-mediated TP53 nuclear export and degradation. They demonstrated that USP10 suppresses cell proliferation in tumors with wild-type TP53, but promotes tumor growth in TP53-mutant tumor. These studies collectively indicate that the tumor-promoting or suppressing effect of USP10 is largely attributed to its substrates and the genetic context of the specific tumor. However, the roles of USP10 in KRAS-mutant CRC remain poorly understood.

In the current study, through a comprehensive screening of DUB inhibitors and liquid chromatography-tandem mass spectrometry (LC–MS/MS) analysis, we identified USP10 as a key promoter of KRAS-mutant CRC growth. Mechanistically, USP10 directly interacts with KRAS mutants and removes their non-proteolytic ubiquitination chains, predominantly containing Ub-K6, K11, K27, and K29 linkages. This deubiquitination enhances the activity of KRAS mutants. Conversely, activated KRAS mutants modulate USP10 protein levels and stability by promoting its phosphorylation at Tyr42 and Ser337, establishing a positive feedback circuit and synergistically driving KRAS-mutant CRC progression. Moreover, USP10 inhibition by RNA interference preferentially impedes the proliferation of KRAS-mutant CRC cells. Collectively, our findings identify USP10 as a novel regulator of oncogenic KRAS mutants and provide a potential therapeutic target for KRAS-mutant CRC.

2. Materials and methods

2.1. Cell culture

HEK-293T (RRID: CVCL_UE07), HCT-116 (RRID: CVCL_0291), HT-29 (RRID: CVCL_0320), SW480 (RRID: CVCL_0546), GP2d (RRID: CVCL_2450), SW837 (RRID: CVCL_1729), and SW620 (RRID: CVCL_0547) cells were originally obtained from the Cell Bank of Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). HEK-293T and SW480 cells were cultured in DMEM medium, while HT-29 and SW620 cells were maintained in RPMI-1640 medium and HCT-116 cells were maintained in McCoy’s 5A medium. All cells were supplemented with 10% (v/v) FBS at 37 °C, in a humidified 5% CO₂ incubator, and routinely monitored negative for mycoplasma contamination. Authentication of the cell lines was confirmed by short tandem repeating profiling every year according to the standard of the international Cell Line Authentication Committee.

2.2. Antibodies and reagents

The antibodies against USP10 (#8501, RRID: AB_10949976), MEK (#4694, RRID: AB_10695868), p-MEK (#9154, RRID: AB_2138017), ERK (#4695, RRID: AB_390779), p-ERK (#4370, RRID: AB_2315112), and p-S/T (#9631, RRID: AB_330308) were purchased from Cell Signaling Technology (Danvers, MA, USA). Antibodies against HA (#db2603, RRID: AB_3451976) and GAPDH (#db106, RRID: AB_3271576) were purchased from Diagbio Biosciences (Hangzhou, China). Antibodies against KRAS (#sc-30, RRID: AB_627865), ubiquitin (Ub) (#sc-8017, RRID: AB_628423), and RAF1 (#sc-7267, RRID: AB_628196) were obtained from Santa Cruz Biotechnology (Dallas, TX, USA). Antibodies against MYC (#A00704, RRID: AB_914461) and FLAG (#A00187, RRID: AB_1720813) were obtained from Gen-Script (Nanjing, China). Antibodies against KRAS (#a12704, RRID: AB_2861676) were purchased from Abclonal (Wuhan, China). Protein synthesis inhibitor cycloheximide (#SBR00013) and puromycin(#P8833) were obtained from Sigma–Aldrich (St. Louis, MO, USA). Proteasome inhibitor MG132 (#S2619) and MAPK inhibitors PD98059 (#S1177) and SB203580 (#S1076) were obtained from Selleck (Houston, TX, USA). USP10 inhibitor Spautin-1 (#HY-12990) and other DUB inhibitors were purchased from MedChemExpress (NJ, USA). RAF1 RBD agarose beads (#ab211176) were purchased from Abcam (Cambridge, UK).

2.3. Gene transfection and lentivirus infection

JET-PRIME (Polyplus, Strasbourg, France, #114-01) was utilized to transfect indicated plasmids according to the manufacturer’s instructions. The gene-specific shRNA sequences used in this study were as follows:

shUSP10 #1: GCCTCTCTTTAGTGGCTCTTT,

shUSP10 #2: GCTGTGGATAAACTACCTGAT.

2.4. Liquid chromatography-tandem-mass spectrometry (LC–MS/MS) analysis

To identify the potential proteins that interacted with KRAS, KRAS^G12V-Flag and vector were transfected into HEK-293T cells for 24 h. Then cells were lysed with RIPA lysis buffer (50 mmol/L Tris-Base, 150 mmol/L NaCl, 5 mmol/L EDTA, 1% NP40, 0.1% SDS, pH 7.4) and incubated with anti-Flag agarose beads to purify KRAS protein. The immunoprecipitated complexes were subjected to trypsin digestion for LC–MS/MS analysis. To identify the potential ubiquitin sites of KRAS^G12V, KRAS^G12V-Flag and HA-ub were transfected into HEK-293T cells for 24 h. Then cells were lysed with 4% SDS and incubated with anti-Flag agarose beads to purify ubiquitinated-KRAS protein. The immunoprecipitated complexes were subjected to trypsin digestion for LC–MS/MS analysis. LC–MS/MS analysis was performed by Jingjie PTM BioLab (Hangzhou, China). The operation protocol information, as well as procedures, was carried out according to the instructions provided by the company.

2.5. Immunoprecipitation and immunoblotting analysis

The immunoprecipitation (IP) assay and immunoblotting (IB) analysis were performed as previously described²⁸. For endogenous immunoprecipitation, cell lysates were extracted, and the concentration of protein was measured by the BCA Protein Quantification Kit (YEASEN, 20201ES86). An equal amount of cell lysate was incubated with the agarose beads coupling anti-KRAS antibodies at 4 °C for overnight. The immunocomplexes were washed 8 times with RIPA lysis buffer (50 mmol/L tris-base, 150 mmol/L NaCl, 5 mmol/L EDTA, 1% NP40, 0.1% SDS, pH 7.4) followed by immunoblotting analysis.

2.6. Immunohistochemistry

The immunohistochemistry analysis was performed as described previously³⁰. In brief, tissue slides were first de-paraffinized and immersed in PBS. The slides were heated in a microwave oven for 15 min in Citrate Antigen Retrieval buffer. After cooling to room temperature, the slides were incubated with 3% H₂O₂ to block endogenous peroxidase activity, and then incubated with 10% goat serum to block non-specific staining for 30 min, respectively. Subsequently, the slides were incubated with primary antibodies at 4 °C overnight and indicated secondary antibodies at room temperature for 1 h. After washing with PBS for 15 min, the slides were then exposed for 2–3 min to 3,3ʹ-diaminobenzidine tetrahydrochloride (DAB) and rinsed off in deionized water to terminate the DAB reaction. The evaluation of the IHC staining was performed by a pathologist who was blind from the gene information of these patients.

2.7. Cell proliferation and colony formation assays

CRC cells infected with the indicated lentivirus were seeded at 1000 cells/well in 96-well plates, and the optical density value was measured every day. The cell proliferation was assessed by sulforhodamine B (SRB) assay, in brief, the cell plates were fixed with 100 μL 10% trichloroacetic acid solution (pre-cooled at 4 °C) each well for 1 h at 4 °C, then the supernatant was discarded, and the cells were washed three times with double-distilled water and then dried. Subsequently, 100 μL 1% SRB staining solution was added to each well, and the plates were left at room temperature for 20 min. After that, the wells were washed clean with 1% glacial acetic acid and dried. Finally, 100 μL of tris-base solution was added to each well to dissolve the SRB staining solution bound to the cells. The cell plates were then placed on a room temperature shaker for 20 min, and the absorbance was measured at a wavelength of 515 nm. For colony formation assay, CRC cells infected with indicated shRNA were plated with 1000 cells/well in 6-well plates for 2 weeks, and were stained by SRB to count the number of colonies.

Inhibition rate (IR, %)=(Absorbance_control–Absorbance_treatment)/Absorbance_control × 100 (1)

Δ Inhibition rate (ΔIR) = IR_Mut–IR_WT (2)

2.8. Real-time PCR

Total RNA was freshly extracted by TRIzol (Invitrogen, Carlsbad, CA, USA). Using TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix Kit (Transgen) to carry out reverse transcription PCR and acquire cDNA. Subsequently, SYBR-Green was applied to real-time PCR analysis. The sequences of primers were used as follows:

USP10, forward primer: ATTGAGTTTGGTGTCGATGAAGT;

USP10, reverse primer: GGAGCCATAGCTTGCTTCTTTAG;

KRAS, forward primer: ACAGAGAGTGGAGGATGCTTT;

KRAS, reverse primer: TTTCACACAGCCAGGAGTCTT;

GAPDH, forward primer: GGAGCGAGATCCCTCCAAAAT;

GAPDH, reverse primer: GGCTGTTGTCATACTTCTCATGG.

2.9. In vitro deubiquitination assay

Flag-tagged KRAS and HA-tagged ubiquitin were transfected into HEK-293T cells. After 24 h, cells were harvested and lysed with 4% SDS buffer, and the extracts were incubated with anti-flag beads at 4 °C for overnight. The poly-ubiquitinated KRAS protein was incubated with bacterial-expressed rhUSP10 for 2 h at 37 °C in vitro, followed by immunoblotting analysis.

2.10. Antitumor activity in vivo

BALB/c-Nude mice (RRID: IMSR_RJ: BALB-C-NUDE, 4–5 weeks) were purchased from GemPharmatech Co., Ltd. (Jiangsu, China) and maintained in a pathogen-free animal facility. All animal experiments were performed according to the regulations of the Institutional Animal Care and Use Committee (IACUC). The protocols for the animal study and field sample permit were approved by the Innovation Institute for Artificial Intelligence in Medicine of Zhejiang University (approval number: DW22071101, China). BALB/c-Nude mice were subcutaneously injected with 1 × 10⁶ CRC cells in 200 μL medium. When tumor volume reached 50 mm³, these mice were randomly divided into two groups and underwent intratumor injection of the scramble or shUSP10 lentivirus every two days. At the endpoint of the experiment, the mice were humanely euthanized via cervical dislocation. Tumor volume (V) was calculated as Eq. (3):

V= (Length × Width²)/2 (3)

2.11. Patient-derived xenograft models

The patient-derived xenograft study had obtained informed consent from all participants and was approved by the Ethics Committee of the Hangzhou First People’s Hospital (ethics approval license: KY-20220105-0011-01, China). The gender, age, and body weight of participants are not variables in our study. The tumor specimens from KRAS-mutant CRC patient were cut into pieces of 1 mm³ under sterile conditions and subcutaneously transplanted into BALB/c-Nude mice. When tumor volume reached around 50 mm³, these mice were randomly divided into two groups and underwent intratumor injection of the scramble or shUSP10 lentivirus every two days. At the endpoint of the experiment, the mice were humanely euthanized via cervical dislocation. Tumor volume was calculated as described previously.

2.12. In vitro ubiquitin 7-amido-4-methylcoumarin (Ub-AMC) assay

293T cells transfected with vector or USP10-WT/C424A-Flag were lysed in RIPA lysis buffer (50 mmol/L tris-base, 150 mmol/L NaCl, 5 mmol/L EDTA, 1% NP40, 0.1% SDS, pH 7.4), and protein quantification was measured by BCA Protein Quantification Kit (YEASEN, 20201ES86). An equal amount of cell lysate was incubated with anti-Flag sepharose for 2 h at 4 °C. The immunocomplexes were washed 6 times with RIPA lysis buffer and then subjected to an in vitro Ub-AMC assay. For the in vitro Ub-AMC assay, the equal immunocomplexes were incubated in reaction buffer (50 mmol/L tris-HCl, 5 mmol/L MgCl₂, 2 mmol/L DTT, 2 mmol/L ATP, pH 7.5) at 37 °C for 1 h. Reactions were initiated by adding 2 μmol/L Ub-AMC (Boston Biochem, #U-550). Fluorescence levels were measured continuously at 25 °C using a Tecan Spark Microplate Reader at an excitation wavelength of 345 nm and an emission wavelength of 445 nm.

2.13. Statistical analysis

Numbers of biological replicates in different experiments are included in each figure legend and supplementary figure legend. No statistical methods were used to predetermine sample sizes, but our sample sizes are similar to those generally employed in this field. No codes have been used in our study. All experiments were randomized and blinded where possible, and no data were excluded. All experiments were repeated at least three times in our study, and the independent experiment results are presented as the mean ± standard deviation (SD) or mean ± standard error of mean (SEM). Two-tailed unpaired Student’s t-test was used for calculating statistical significance of two groups, while one-way ANOVA was used when multiple groups were compared. All statistical results were conducted using GraphPad Prism (RRID: SCR_002798). Differences in means were considered significant when P < 0.05 (∗P < 0.05, ∗∗P < 0.01, and ∗∗∗P < 0.001).

3. Results

3.1. Identification of USP10 as an essential regulator of KRAS mutants

Given KRAS mutation contributing to its robust tumor-promoting activities, we conducted paired HT-29 cell lines harboring KRAS^{WT/G12C/G12D/G12V} (KRAS^WT/Mut) and performed a whole-cell phenotypic screen to identify the underlying DUBs that are essential for KRAS mutant but not available to KRAS wild-type CRC (Fig. 1A and Supporting Information Fig. S1A). We evaluated DUB inhibitors for their selective activity in inhibiting the proliferation of HT-29-KRAS^Mut cells by screening DUB inhibitors library. From this screen, Spautin-1, previously reported as a USP10/USP13 inhibitor, emerged as a distinct hit, selectively impeding the proliferation of HT-29-KRAS^Mut cells (Fig. 1A). Additionally, LC–MS/MS analysis identified more than 300 proteins interacting with KRAS mutants, including OTUB1, a DUB reported to modulate the membrane localization of KRAS wild-type but not that of KRAS mutants by inhibiting E2 enzymes UBC13 and UBE2D3 activity, indicating the reliability of LC–MS/MS results (Supporting Information Table S1). Among these identified proteins, the protein score of USP10 ranks second only to that of KRAS (Fig. 1B). Together, the results of the DUB inhibitors screening and LC–MS/MS analysis suggest that USP10 functions as a potential DUB regulating KRAS mutants (Fig. 1C).

Figure 1.

Identification of USP10 as an essential regulator of KRAS mutants. (A) The schematic diagram of DUB inhibitors (DUBi) library screening and identification Spautin-1 selectively inhibiting KRAS-mutant CRC cell growth. HT-29 cells were infected with the lentivirus that stably over-expressed HT-29-KRAS^{WT/G12C/G12D/G12V} (HT-29-KRAS^WT/Mut) and screened by using puromycin, then these cells were seeded with 1000 cells/well in 96-well plates, and then were treated with 10 μmol/L DUB inhibitor for 72 h, followed by SRB assays. (B) The underlying proteins interacting with oncogenic KRAS were identified by LC–MS/MS. HEK-293T cells were transfected with KRAS^G12V-Flag and vector for 24 h, then cells were lysed with RIPA lysis buffer and incubated with anti-Flag agarose beads. The immunoprecipitated KRAS-Flag was captured using SDS-PAGE and the gel was subjected to trypsin digestion for LC–MS/MS analysis. (C) The Venn diagram illustrate the potential targets that are involved in oncogenic KRAS-induced cell proliferation and interact with KRAS mutants. (D) USP10 predominantly interacts with mutant KRAS in vivo. HEK-293T cells were co-transfected with indicated expression plasmids and the cell lysates were subjected to IP assays. (E) USP10 interacts with KRAS mutants at endogenous levels. Cell lysates of SW480 and SW620 were incubated with the affinity gel conjugated with KRAS antibody. Proteins retained on the affinity gel were subjected to IB analysis.

To clarify whether USP10 functions as a DUB directly regulating KRAS mutants, we first examined the interaction between USP10 and KRAS mutants/wild-type. As shown in Fig. 1D, the substantial interaction between USP10 and KRAS mutants has been observed while there is a very weak binding between USP10 and KRAS wild-type; moreover, the interaction between USP10 and KRAS was independent of USP10’s enzymatic activity (Fig. S1B), Meanwhile, the endogenous interaction between USP10 and KRAS mutants in KRAS-mutant CRC cells has also been confirmed by co-immunoprecipitation assays (Fig. 1E and Fig. S1C). Collectively, these data demonstrated that USP10 is a bona fide binding protein of KRAS.

3.2. USP10 positively regulates KRAS mutants’ activity

Given that USP10 interacts with KRAS mutants, we sought to determine whether USP10 regulates the turnover of KRAS mutants. Initially, we investigated the effect of USP10 on KRAS mutants’ levels and found that USP10 knockdown did not affect KRAS protein levels (Fig. 2A). Subsequently, we introduced wild-type USP10 and its enzymatic mutant USP10-CA and found that USP10-CA almost lost whole enzymatic activity, which are consistent with the observations in other studies (Supporting Information Fig. S2A)^28,31,32. More importantly, Overexpression of USP10 and USP10-CA failed to pose any effect on KRAS protein levels (Fig. 2B). Meanwhile, we also examined the effect of USP10 and its inhibitor Spautin-1 on KRAS mRNA levels. As shown in Fig. 2C and Fig. S2B and S2C, neither depletion of USP10 nor overexpression of USP10-WT/CA or its inhibitor affects KRAS mRNA levels. These results indicate that USP10 does not regulate the transcriptional levels and turnover of KRAS mutants.

Figure 2.

USP10 positively regulates KRAS mutants' activity. (A) Depletion of USP10 posed little effect on KRAS protein levels. SW480 and SW620 cells were infected with the lentivirus encoding the indicated shRNA, and the cell lysates were subjected to IB analysis. (B) Overexpression of USP10 posed little effect on KRAS protein levels. SW480 and SW620 cells were transfected with USP10-WT or USP10-CA, and the cell lysates were subjected to IB analysis. (C) Depletion of USP10 posed little effect on KRAS mRNA levels. SW480 and SW620 cells were infected with the lentivirus encoding the indicated shRNA, and cell lysates were subjected to real-time PCR assays (Mean ± SD, n = 3). (D) Depletion of USP10 remarkably sustained MAPK signaling. SW480 and SW620 cells were infected with indicated shRNAs, and the cell lysates were subjected to IB analysis. (E) USP10 inhibitor Spautin-1 significantly suppressed MAPK signaling. SW480 and SW620 cells were treated with 25 μmol/L Spautin-1 for the indicated time, and the cell lysates were subjected to IB analysis. (F) Overexpression of USP10 significantly promoted MAPK signaling in an enzymatic activity-dependent manner. SW480 and SW620 cells were transfected with the indicated plasmid, and the cell lysates were subjected to IB analysis. n.s., P > 0.05.

Subsequently, we further examined the effect of USP10 on KRAS activity by measuring its classical downstream MAPK pathway. As shown in Fig. 2D and Fig. S2D, depletion of USP10 remarkably decreased the phosphorylation of MAPK kinase (MEK) and extracellular signal-regulated kinase (ERK), suggesting the inhibition of the MAPK pathway. Additionally, we also introduced USP10 inhibitor Spautin-1 and found Spautin-1 also significantly suppressed the MAPK pathway (Fig. 2E). Correspondingly, overexpression of USP10 promoted MAPK pathway activation, whereas its enzymatic mutant USP10-CA did not have similar effect (Fig. 2F). Meanwhile, we further investigated the effect of USP10 knockdown on PI3K–AKT signaling, another classical downstream pathway of KRAS. As shown in Fig. S2E, depletion of USP10 remarkably downregulated p-AKT levels, indicating the significant inhibition of USP10 knockdown on PI3K–Akt signaling.

Taken together, these results suggest that USP10 positively regulated the activity of KRAS mutants, but imposed minimal effect on their protein levels.

3.3. USP10 directly deubiquitinates KRAS mutants in an enzymatic activity-dependent manner

The aforementioned data demonstrated that USP10 interacts with KRAS mutants and positively regulates their activity. Considering USP10 functions as a DUB, we hypothesized that USP10 might promote MAPK activation by deubiquitinating KRAS mutants; therefore, we examined the effect of USP10 on KRAS mutants' ubiquitination levels. As shown in Fig. 3A and B and Fig. S2F, overexpression of wild-type USP10, but not its enzymatic mutant CA, significantly decreased KRAS mutants' ubiquitination levels, indicating USP10 regulated KRAS ubiquitination levels in an enzymatic activity-dependent manner. Conversely, depletion of USP10 remarkably increased KRAS mutants' polyubiquitylation levels (Fig. 3C and Fig. S2G). Additionally, we tested the effect of Spautin-1 on KRAS ubiquitination levels. As shown in Fig. 3D, Spautin-1 reversed the decreased ubiquitination levels of KRAS mutants induced by USP10 overexpression. These results collectively suggest that USP10 negatively regulates KRAS mutants’ ubiquitination levels.

Figure 3.

USP10 directly deubiquitinates KRAS in an enzymatic activity-dependent manner. (A, B) Overexpression of USP10-WT, but not USP10-CA, significantly decreases KRAS mutants' ubiquitination levels. Cells transfected with the indicated plasmids were treated with MG132 (10 μmol/L, 8 h) before harvested. Total cell lysates were immunoprecipitated with anti-Flag affinity gels, then the proteins remaining on gels were subjected to IB analysis. (C) Depletion of USP10 dramatically increased KRAS mutants' ubiquitination. HEK-293T cells infected with control/USP10 shRNA were transfected with the indicated plasmids and treated with MG132 (10 μmol/L, 8 h) before harvested. Cell lysates were immunoprecipitated with anti-Flag affinity gels, and the proteins remaining on gels were subjected to IB analysis. (D) USP10 inhibitor Spautin-1 significantly increased KRAS mutants' ubiquitination. HEK-293T cells transfected with the indicated plasmids were treated with MG132 (10 μmol/L, 8 h) and Spautin-1 (50 μmol/L, 10 h) before harvested. Cell lysates were immunoprecipitated with anti-Flag affinity gels, then the protein remaining on gels were immunoblotted with the indicated antibodies. (E) Bacterial-expressed recombinant human USP10 (rhUSP10) effectively removed the polyubiquitination of KRAS mutants in vitro. HEK-293T cells were transfected with indicated expression plasmids and cell lysates were incubated with anti-Flag affinity gel for overnight. Then, proteins retained on anti-Flag affinity gel were incubated with rhUSP10 in 37 °C for 2 h, followed by IB assays. (F) Overexpression of USP10 efficiently removes the K6-, K11-, K27- and K29-linked polyubiquitin chains of KRAS mutants. Cells transfected with the indicated plasmids were treated with MG132 (10 μmol/L, 8 h) before harvested. Total cell lysates were immunoprecipitated with anti-Flag affinity gels, then the proteins remaining on gels were immunoblotted with the indicated antibodies.

To further demonstrate the deubiquitination activity of USP10 towards KRAS mutants, we performed in vitro deubiquitination assays using a cell-free system. We found that recombinant USP10 protein effectively removed the polyubiquitin chains from KRAS mutants in vitro (Fig. 3E). Subsequently, we investigated the types of ubiquitin chains that USP10 removes from KRAS mutants. As shown in Fig. 3F, USP10 primarily removed the non-proteolytic ubiquitination chains, including Ub-K6, K11, K27 and K29-linked chains. To further identify the ubiquitin sites of KRAS mutants, we performed LC–MS/MS analysis and found that K42, K128, K147 and K176 are the underlying ubiquitin sites of KRAS (Supporting Information Table S2). Subsequently, we constructed the ubiquitin site mutants of KRAS (KRAS-4KR/KRAS-K42R/K128R/K147R/K176R) and performed in vivo ubiquitination assays. As shown in Fig. S2H, USP10 overexpression significantly removed the polyubiquitination chains of KRAS, while posing minimal effect on KRAS-4KR, demonstrating K42, K128, K147 and K176 are the main ubiquitin sites of KRAS regulated by USP10.

These results demonstrated that USP10 deubiquitinates KRAS mutants by removing non-proteolytic ubiquitination chains, main including K6, K11, K27 and K29-linked ubiquitin in an enzymatic activity-dependent manner.

3.4. USP10 knockdown significantly suppresses KRAS-mutant CRC growth in vitro and in vivo

Given that USP10 positively regulates KRAS mutants’ activity, we next investigated the roles of USP10 in KRAS-mutant CRC growth. As shown in Fig. 4A and B and Supporting Information Fig. S3A-S3D, depletion of USP10 remarkably inhibited the proliferation and colony formation of KRAS mutant CRC, indicating that USP10 knockdown significantly suppresses KRAS-mutant CRC growth in vitro. Subsequently, we also performed an in vivo xenograft experiment to investigate the effect of USP10 on the growth of KRAS-mutant CRC xenografts. As shown in Fig. 4C–E and Supporting Information Table S3, depletion of USP10 significantly inhibited the tumor growth of SW480 and SW620 xenografts, as reflected by the decrease in relative tumor volume and tumor weight. Subsequently, we further analyzed the intratumor proteins and found that USP10 depletion significantly downregulated intratumor p-MEK and p-ERK levels (Fig. 4F).

Figure 4.

USP10 knockdown significantly suppresses KRAS-mutant CRC growth in vitro and in vivo. (A, B) Depletion of USP10 suppressed the proliferation of KRAS-mutant CRC cells in vitro. SW620, SW480 and HCT-116 cells were infected the lentivirus that encoding control/USP10 shRNA, followed by cell proliferation assays and colony formation assays (mean ± SD; n = 3). (C–E) Depletion of USP10 significantly arrested the growth of KRAS-mutant CRC xenografts in vivo. (C) The bearing-tumor mice and tumor images. (D) The relative tumor volume of the indicated groups. (E) The tumor weight of indicated groups (mean ± SEM; n = 10/group). (F) USP10 knockdown remarkably inhibited intratumor MAPK signaling. Immunoblot analysis of related protein levels in SW480 xenografts and SW620 xenografts. (G) The schematic diagram of PDX models. (H, I) Depletion of USP10 significantly inhibited the growth of KRAS-mutant CRC PDX in vivo (mean ± SEM; n = 6/group). (H) The relative tumor volume of indicated groups. (I) The tumor weight of indicated groups. (J) USP10 knockdown remarkably restrained intratumor MAPK signaling. Immunoblot analysis of related protein levels in PDXs. (K) USP10 levels in tumor tissues are significantly higher than those of in normal tissues in CRC patients. (L) USP10 is significantly overexpressed in tumor tissue compared to normal tissue in CRC patients. Representative section of H&E staining and USP10 staining in CRC patient tissues. T: tumor tissues; N: para-carcinoma tissues; scale bars: 250 μm. (M) USP10 level is inversely correlated with the overall survival of CRC patients. The correlation analysis between USP10 expression and overall survival of colorectal cancer patients in the Kaplan–Meier Plotter database (split by the best cutoff). ∗∗P < 0.01; ∗∗∗P < 0.001.

To further investigate USP10’s impact on KRAS-mutant CRC growth in a preclinical setting, we constructed a primary patient-derived xenograft (PDX) from a clinical KRAS-mutant colorectal cancer patient (Fig. 4G). As shown in Fig. 4H–J and Table S3, similar results were obtained, indicating that USP10 depletion inhibited KRAS-mutant CRC xenografts growth by restraining KRAS activity. Collectively, these results confirmed that USP10 plays a crucial role in promoting KRAS mutant CRC growth in vitro and in vivo.

To gain further insight into the relevance between USP10 and CRC pathogenesis, we analyzed USP10 levels in normal and tumor tissues of CRC patients by using the Cancer Genome Atlas Program (TCGA) database. As shown in Fig. 4K, CRC tumor tissues exhibited higher USP10 levels than normal tissues. Additionally, we collected clinical CRC patient samples and performed the IHC assays to comparatively investigate USP10 expression in tumor versus adjacent normal tissues. As shown in Fig. 4L, USP10 expression levels were significantly higher in tumor tissue compared to adjacent normal tissues, consistent with the results obtained from the TCGA database. Moreover, we found that high USP10 levels inversely correlated with the overall survival and relapse-free survival of CRC patients (Fig. 4M and Fig. S3E), highlighting that USP10 is involved in the malignant progression of CRC and is closely associated with the poor prognosis of CRC patients.

3.5. KRAS mutants upregulate USP10 levels and stability by promoting the phosphorylation of USP10 at Tyr42 and Ser337

The significant differences in binding amount between USP10 and KRAS wild-type/mutant led us to hypothesize that USP10 plays distinct roles in KRAS wild-type and mutant CRC. To test this hypothesis, we collected the clinical samples of KRAS wild-type/mutant CRC patients and compared the expression levels of USP10 by performing immunohistochemistry (IHC) assays. As shown in Fig. 5A, KRAS-mutant CRC samples exhibited higher USP10 levels compared to KRAS wild-type samples, suggesting that KRAS mutants may function as a driver promoting USP10 overexpression in KRAS-mutant CRC.

Figure 5.

KRAS mutants upregulates USP10 levels and stability by promoting the phosphorylation of USP10 at Tyr42 and Ser337. (A) KRAS-mutant CRC patients displayed higher USP10 levels compared to KRAS wild-type patients. Immunohistochemistry (IHC) staining of intratumor USP10 in KRAS mutants/wild-type CRC patients (mean ± SD, n = 6). (B) Overexpression of KRAS mutants significantly upregulated USP10 protein levels. HT-29 cells were transfected with the indicated plasmids and cell lysates were subjected to IB assays. (C) Overexpression of KRAS mutants, not KRAS wild-type, significantly promoted USP10 phosphorylation. HEK-293T cells transfected with indicated plasmids were immunoprecipitated with anti-Flag agarose beads and the protein remaining on anti-Flag agarose beads were subjected to IB assays. (D) MAPK inhibitor PD98059 reversed the upregulation of USP10 phosphorylation induced by KRAS mutants. HEK-293T cells transfected with indicated plasmids were treated with DMSO, PD98059 (10 μmol/L) for 24 h before harvested, then cell lysates were immunoprecipitated with anti-Flag agarose beads and immunoblotted with the indicated antibody. (E) T42 and S337 are the main phosphorylation sites of USP10 regulated by KRAS mutants. HEK-293T cells transfected with indicated plasmids were immunoprecipitated with anti-Flag agarose beads and the protein remaining on anti-Flag agarose beads was subjected to IB assays. TA&SA: T42A&S337A. (F) KRAS mutants significantly increased the protein levels and stability of USP10-WT, but not USP10-T42A&S337A. HT-29 cells expressing USP10-WT/TA&SA were infected with lentivirus encoding Vector or KRAS^G12V-Myc and treated with 20 μg/mL CHX for indicated time and the cell lysates were immunoblotted with the indicated antibody. TA&SA: T42A&S337A. ∗∗P < 0.01.

To decipher the underlying molecular mechanism of USP10 overexpression in KRAS-mutant CRC, we examined the effect of KRAS wild-type/mutant on USP10 levels. As shown in Fig. 5B, overexpression of KRAS mutants, but not wild-type KRAS, remarkably upregulated USP10 protein levels, indicating that KRAS mutants specifically enhance USP10 protein levels. To further investigate whether KRAS mutants mediated USP10 upregulation at the posttranslational levels or transcriptional levels, we next examined the effect of KRAS mutants/wild-type on USP10 mRNA levels. As shown in Supporting Information Fig. S4A, neither KRAS mutants nor KRAS wild-type influenced USP10 mRNA levels, implying that KRAS mutants upregulated USP10 levels at the posttranslational levels rather than the transcriptional levels.

Phosphorylation cascade signaling is the key process by which KRAS mutants regulate downstream proteins; we therefore examined the effect of KRAS mutants on USP10 phosphorylation levels. As shown in Fig. 5C, overexpression of KRAS mutants, but not KRAS wild-type, significantly increased the phosphorylation levels of USP10; moreover, this effect could be weakened by MEK inhibitor PD98059 (Fig. 5D), indicating that the upregulation of USP10 phosphorylation levels is mediated by the MAPK signaling pathway. Subsequently, we attempted to identify the underlying kinase-mediated KRAS upregulating USP10 phosphorylation. Given the MEK inhibitor can weaken the effect of KRAS mutants-induced USP10 phosphorylation upregulation, we focused on the kinases associated with the RAF–MEK–ERK pathway. We performed LC–MS/MS analysis and found Mitogen-activated protein kinase 1 (MAPK1/ERK2) interacts with USP10 (Supporting Information Table S4). Subsequent co-immunoprecipitation assay also demonstrated the interaction between USP10 and MAPK1 (Fig. S4B). More importantly, we found MAPK1 overexpression promoted USP10 phosphorylation (Fig. S4B), and MAPK1 depletion attenuated the upregulation of USP10 phosphorylation induced by KRAS mutants to a certain extent (Fig. S4C). These data collectively demonstrated KRAS mutants upregulate USP10 relying on MAPK1 at least in part.

Next, we sought to identify the specific phosphorylation sites mediated KRAS mutants upregulating USP10. Previous studies have shown that USP10 is primarily phosphorylated at Thr42 (T42), Ser76 (S76) and Ser337 (S337) ^29,33. Among these phosphorylation sites, Thr42 and Ser337 are known to affect USP10 stability²⁹, while Ser 76 affects USP10’s activity³³. We mutated these three phosphorylation sites to Alanine (Ala), thereby abolishing their phosphorylation. Interestingly, co-mutation at T42 and S337 (TA&SA) abolished the KRAS mutants-induced upregulation of USP10 phosphorylation, while the Ser 76 mutation did not have a similar effect (Fig. 5E and Fig. S4D), indicating Thr42 and Ser337 are the primary phosphorylation sites through which KRAS mutants upregulate USP10. We also investigated whether the phosphorylation at these sites affects USP10’s binding and deubiquitinating activity towards KRAS. As shown in Fig. S4E and S4F, the co-mutation of Thr42 and Ser337 did not influence USP10’s interaction with KRAS, nor did it impair its deubiquitinating activity towards KRAS, while the mutation of Ser76 exerted a weaker deubiquitinating activity towards KRAS due to its impaired enzymatic activity (Fig. S4G), which is consistent with the observation in other study³³, further demonstrating the effect of USP10 on KRAS mutants relying on USP10’s enzymatic activity. Considering that KRAS mutants upregulate USP10 at the posttranslational level rather than the transcriptional level, we tested whether the phosphorylation influences USP10 stability. We found that the co-mutation of Thr42 and Ser337 significantly decreased USP10 stabilization; moreover, overexpression of KRAS mutants remarkably increased USP10 stabilization, but this effect was greatly diminished for TA&SA (Fig. 5F).

Taken together, these data demonstrated KRAS mutants upregulating USP10 levels and stability by phosphorylating the latter at Thr42 and Ser337.

3.6. USP10 inhibition preferentially inhibits the growth of KRAS-mutant CRC in vitro and in vivo

Aforementioned data demonstrated that USP10 and KRAS mutants form a positive feedback circuit. Given the distinct regulatory mechanisms between USP10 and KRAS wild-type/mutants, we speculated that KRAS-mutant CRC may be more sensitive to USP10 inhibition. As shown in Fig. 6A and Supporting Information Fig. S5A, depletion of USP10 resulted in a striking reduction of cell growth in KRAS-mutant CRC cells, whereas it only exhibited slight effect in KRAS wild-type CRC cells. Similar results were observed in the colony formation assay (Fig. 6B), indicating depletion of USP10 preferentially inhibits the proliferation of KRAS-mutant CRC cells in vitro. To further validate this finding, we conducted in vivo xenografts experiments. As presented in Fig. 6C–E, Fig. S5B and Supporting Information Table S5, depletion of USP10 exhibited stronger anti-tumor activity in KRAS-mutant CRC xenografts compared to KRAS wild-type xenografts.

Figure 6.

USP10 inhibition preferentially inhibits the growth of KRAS-mutant CRC in vitro and in vivo. (A, B) Depletion of USP10 selectively inhibited KRAS-mutant CRC proliferation. HT-29 cells stably expressing KRAS^{WT/G12C/G12D/G12V} were infected with the lentivirus encoding control/USP10 shRNA and then these cells were subjected to cell proliferation assays (A) and colony formation assays (B). (C–E) Depletion of USP10 selectively suppressed KRAS-mutant CRC xenografts in vivo (mean ± SEM, n = 10/group). (C) The tumor images of indicated groups. (D) The relative tumor volume of indicated groups. (E) The tumor weight of indicated groups. ∗P < 0.05; ∗∗∗P < 0.001.

Collectively, these data underscore the crucial roles of USP10 in KRAS-mutant CRC and show that USP10 inhibition preferentially impairs KRAS-mutant CRC growth in vitro and in vivo.

3.7. USP10 is elevated in KRAS-mutant CRC tissues

To confirm our findings in a clinical context, we collected KRAS wild-type and mutant CRC tissue samples. IHC staining revealed that KRAS-mutant CRC tissues exhibit higher USP10 levels compared to KRAS wild-type CRC tissues (Fig. 7A–D). Approximately 50% of KRAS-mutant CRC patients displayed high (USP10 +++) or medium (USP10 ++) USP10 expression, whereas this population was less than 15% in KRAS wild-type patients. These findings further support our conclusion that KRAS mutants function as a driver promoting USP10 overexpression in KRAS-mutant CRC.

Figure 7.

USP10 is elevated in KRAS-mutant CRC tissues. (A, B) IHC assays were performed to detect the expression of USP10 and KRAS in KRAS-mutant (n = 36) and wild-type (n = 36) CRC tissues. (C) KRAS-mutant CRC tissues displayed higher USP10 levels than KRAS wild-type. IHC score was evaluated by multiplication of positive staining proportions (1 score, <25%; 2 score, 25%–50%; 3 score, 50%–75%; 4 score, 75%–100%) and protein expression intensity (1 score, weak staining; 2 score, moderate staining; 3 score, high staining). –, negative expression (1–3 score); +, low expression (4–6 score); ++, medium expression (7–9 score); +++, high expression (10–12 score). (D) Statistical analysis of USP10 expression in KRAS wild-type/mutant CRC specimens. (E) The schematic diagram of USP10/KRAS positive feedback circuit promoting the proliferation of KRAS-mutant CRC. ∗∗∗P < 0.001.

4. Discussion

KRAS is the most frequently mutant driver in human tumors, with approximately 50% of CRC patients harboring KRAS mutation. However, existing therapeutic therapies have shown limited efficacy in treating KRAS-mutant colorectal cancer. Although KRAS^G12C covalent inhibitors have successfully treated patients with this specific mutation, other KRAS-mutant tumors with a higher mutation frequency, such as KRAS^G12D and KRAS^G12V, have not benefited from these treatments. Consequently, identifying novel and effective therapeutic targets for these KRAS-mutant tumor is of critical importance. In our study, we demonstrated that USP10 depletion preferentially inhibited the growth of KRAS-mutant CRC, suggesting USP10 could be a potential therapeutic target for these tumors.

The dysregulation or abnormal activation of DUBs has been widely implicated in tumor progression and malignancy34, 35, 36. DUBs perform their biological functions by deubiquitinating substrates, thus regulating their abundance, subcellular localization, and activity³⁷. Given their well-defined structure and enzymatic sites, DUBs are considered the ideal drug targets. To date, approximately 60 DUB inhibitors have been reported, however, most of these inhibitors are non-selective, targeting multiple DUBs. In our study, through screening DUB inhibitors and LC–MS/MS analysis, we demonstrated that USP10 functions as a crucial DUB regulating the activity of KRAS mutants. USP10 belongs to the ubiquitin-specific proteases (USPs), the largest subfamily of DUBs. To date, no specific inhibitors targeting USP10 have been reported and spautin-1 has been reported to inhibit the deubiquitinating activity of both USP10 and USP13^38,39, highlighting the scarcity of specific inhibitors targeting USP10. Therefore, to minimize off-target toxicity, there is an urgent need to develop specific DUB inhibitors for USP10 and other potential DUB targets.

Emerging evidence indicates that ubiquitination modification plays a crucial role in regulating both the abundance and function of KRAS. Yin et al.¹⁶ found that the monoubiquitination of KRAS at lysine 104 alters its conformation and promotes its interaction with guanine nucleotide exchange factors. Another study also reported the monoubiquitination of KRAS at Lys147 and Lys117 promoting the binding affinity between KRAS and PI3K as well as RAF, suggesting that the monoubiquitination of KRAS at Lys104, Lys117 and Lys147 promoted its activity⁴⁰. So far, several E3 ligases, including LZTR1^20,41,42, β-TRCP²¹ and WDR76²², have been reported to regulate KRAS. Despite these findings, the DUBs directly regulating KRAS, particularly the mutant KRAS, remain elusive. A recent study reveals that USP7 deubiquitinates and stabilizes KRAS by removing the K48-linked polyubiquitin chains from residue K147, ultimately promoting non-small cell lung cancer growth²⁴. However, the specific DUBs directly regulating KRAS mutants' activity without affecting its abundance are still unknown. In our study, we highlight USP10 as a key DUB that modulates KRAS through a direct interaction and maintains the latter’s activation. Moreover, we found that the physical association between USP10 and KRAS mutants persists even when USP10’s catalytic site is mutated, demonstrating that their protein–protein interaction is independent of USP10’s enzymatic activity. In contrast, USP10’s ability to reduce KRAS ubiquitination and modulate its activity requires catalytic activity, as USP10-CA fails to deubiquitinate KRAS. This dichotomy is consistent with prior reports for other deubiquitinases^43,44, where substrate binding and enzymatic regulation of DUBs are generally separable events. Although the mutation of the catalytic site does not affect the binding between deubiquitinases and their substrates, the catalytically inactive mutants of DUBs lose their deubiquitinating enzyme activity and cannot effectively remove the ubiquitin chains of substrates, thereby failing to exert any substantial impact on substrates' levels and/or activity.

Polyubiquitination is one of the most important post-translational modifications regulating protein homeostasis and/or activity. K48- and K63-linked polyubiquitination have been widely reported to be involved in proteasomal degradation and DNA repair, and the functions of other non-canonical polyUb chains linked via K6-, K11-, K27-, K29-, K33- still need to be further clarified⁴⁵. K6-linked ubiquitination has been demonstrated to control the process of autophagy and DNA damage response; K11-linked ubiquitination is implicated in the regulation of cell cycle and proteasomal degradation; K27-linked ubiquitination functions as a major player of innate immunity by regulating NF-κB and IRF pathways; K29-linked ubiquitination of proteins have been implicated in signaling transduction and neurodegenerative disorders, while K33-linked ubiquitination remains the least studied of all Ub linkage types and may involve in protein trafficking and autophagy regulation⁴⁶. In our study, we found that K6-, K11-, K27-, and K29-linked polyubiquitination mediated KRAS activity, and USP10 promoted KRAS activity by removing these polyubiquitination chains, but the molecular mechanisms and action modes of these types of polyubiquitin chains regulating KRAS activity need to be further investigated in the future.

USP10 exhibits context-dependent roles in tumorigenesis, with its functional duality stemming from substrate specificity and cellular genetic background. Emerging evidence demonstrates that USP10 can act as either an oncogene or a tumor suppressor in different malignancies. For instance, Cao et al.³² found that USP10 promoted tumor progression by stabilizing mitosis-related protein Anillin (ANLN) in esophageal squamous cell carcinoma. Some lines of evidence reveal the roles of USP10 in modulating the tumor microenvironment, as it could facilitate NF-κB activation through NLR family pyrin domain-containing 7 (NLRP7) deubiquitination, thereby inducing tumor-associated macrophage polarization⁴⁷. Our previous study demonstrated that USP10 promoted hepatocellular carcinoma proliferation by preventing the ubiquitination and degradation of transcription coactivator YAP/TAZ²⁸. Conversely, USP10 also exerts tumor-suppressive effect by antagonizing c-Myc through SIRT6 and TP53 stabilization²⁶, while paradoxically promoting TP53-mutant tumor growth despite stabilizing wild-type TP53²⁹. These studies collectively suggested that USP10’s biological consequences are dictated by its substrate repertoire and the molecular landscape of a particular tumor type. In the present study, we identify KRAS mutants as key substrates of USP10 in KRAS-mutant CRC, where depletion of USP10 selectively inhibited KRAS-mutant CRC growth by restraining KRAS mutants' activity. Given the dual roles of USP10 in tumors, future development of USP10-targeted therapies should incorporate comprehensive genomics profiling to ensure tumor-specific efficacy.

The roles of USP10 in tumor progression have been extensively documented, yet the mechanisms underlying its high expression in tumors remain ambiguous. In our study, we demonstrated that KRAS mutants function as a driver promoting USP10 overexpression in KRAS-mutant CRC. Several previous studies have revealed that KRAS mutation primarily influences downstream protein regulation through a phosphorylation cascade. KRAS mutants-mediated ERK activation promotes the latter’s binding and subsequent phosphorylation of F-box and WD repeat domain-containing 7 (FBW7) at Thr205, leading to the degradation of FBW7⁴⁸. Additionally, KRAS mutations are implicated in regulating MYC levels through ERK1/2-mediated phosphorylation of MYC at Ser62⁴⁹. These findings collectively demonstrate that KRAS mutations predominantly modulate protein abundance via a phosphorylation cascade. In our study, we have shown that KRAS mutations upregulated USP10 protein stability by enhancing its phosphorylation at Thr42 and Ser337. However, the precise mechanisms by which phosphorylation at these specific sites influences USP10’s stability are yet to be elucidated and warrant further exploration. Such insights could unlock deeper understanding of KRAS signaling and open new therapeutic avenues for treatment strategies targeting KRAS-mutant malignancies, particularly CRC, through modulation of USP10 and the delineated feedback loop.

5. Conclusions

Our study has revealed USP10 as a novel and crucial target for combating KRAS-mutant CRC. Mechanistically, USP10 forms an essential part of a positive feedback circuit with KRAS mutants, operating through a distinct mechanism in which USP10 directly binds to KRAS mutants. It facilitates KRAS mutant activation by selectively removing its non-proteolytic ubiquitin chains linked via Ub-K6, K11, K27, and K29. This deubiquitination promotes the activation of KRAS mutants, which in turn enhances USP10’s stability through phosphorylation at Thr42 and Ser 337. The circuit of USP10/KRAS-mutant is vital, highlighting a synergistic interaction by these two proteins that promotes the growth of KRAS-mutant CRC both in vitro and in vivo (Fig. 7E). Importantly, disrupting this feedback loop using RNA interference against USP10 significantly hinders the growth of KRAS-mutant CRC cells, underscoring the pivotal role of the USP10/KRAS feedback mechanism and introducing a promising therapeutic approach for treating KRAS-mutant CRC by targeting USP10. This strategy presents a potent avenue for therapeutic intervention, providing a foundational step towards improved treatment outcomes for patients with the challenging mutations of KRAS.

Author contributions

Conceptualization: Tao Yuan, Weihua Wang, Qiaojun He, Bo Yang, Hong Zhu; Methodology: Tao Yuan, Weihua Wang, Qiaojun He, Bo Yang, Hong Zhu; Investigation: Tao Yuan, Weihua Wang, Ruilin Wu, Yue Liu, Junwei Fu, Jiamin Du, Meijia Qian, Jia’er Wang, Yubo Zhang, Qiaojun He, Bo Yang, Hong Zhu; Visualization: Tao Yuan, Weihua Wang, Qiaojun He, Bo Yang, Hong Zhu; Funding acquisition: Tao Yuan, Bo Yang, Hong Zhu; Project administration: Wencheng Kong, Ronggui Hu, Tianhua Zhou, Qiaojun He, Bo Yang, Hong Zhu; Supervision: Wencheng Kong, Ronggui Hu, Tianhua Zhou; Writing – original draft: Tao Yuan, Weihua Wang, Qiaojun He, Bo Yang, Hong Zhu; Writing – review & editing: Tao Yuan, Qiaojun He, Bo Yang, Hong Zhu.

Conflicts of interest

The authors declare no conflicts of interest.

Appendix A. Supporting information

The following is the Supporting information to this article: