USP10 drives cancer stemness and enables super-competitor signalling in Colorectal Cancer

doi:10.21203/rs.3.rs-3742742/v1

Download PDF

Article

USP10 drives cancer stemness and enables super-competitor signalling in Colorectal Cancer

https://doi.org/10.21203/rs.3.rs-3742742/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 23 Oct, 2024

Read the published version in Oncogene →

You are reading this latest preprint version

The contribution of deubiquitylating enzymes (DUBs) to b-Catenin stabilization in intestinal stem cells and colorectal cancer (CRC) is poorly understood. Here, and by using an unbiassed screen, we discovered that the DUB USP10 stabilizes b-Catenin specifically in APC-truncated CRC in vitro and in vivo. Mechanistic studies, including in vitro binding together with computational modelling, revealed that USP10 binding to b-Catenin is mediated via the unstructured N-terminus of USP10 and is outcompeted by intact APC, favouring b-catenin degradation. However, in APC-truncated cancer cells USP10 binds to b-catenin, increasing its stability which is critical for maintaining an undifferentiated tumour identity. Elimination of USP10 reduces the expression of WNT and stem cell signatures and induces the expression of differentiation genes. Remarkably, silencing of USP10 in murine and patient-derived CRC organoids established that it is essential for NOTUM signalling and the APC super competitor-phenotype, reducing tumorigenic properties of APC-truncated CRC. These findings are clinically relevant as patient-derived organoids are highly dependent on USP10, and abundance of USP10 correlates with poorer prognosis of CRC patients. Our findings reveal, therefore, a role for USP10 in CRC cell identity, stemness, and tumorigenic growth by stabilising b-Catenin, leading to aberrant WNT signalling and degradation resistant tumours. Thus, USP10 emerges as a unique therapeutic target in APC truncated CRC.

USP10

CTNNB1

DUB

colorectal cancer

WNT

APC

Colorectal cancer (CRC) is the third most common cancer for both sexes, with 8% estimated new cases and 8 to 9% estimated deaths in 2021 ¹. It is widely accepted that environmental risk factors, such as diet, obesity, low level of exercise, alcohol and tobacco consumption increase the risk of developing CRC². Besides these environmental risk factors, inherited forms of CRC, characterized by distinct mutations, occur at a lower frequency ³. Although most patients have an overall good survival after diagnosis, only 10% patients with progressed disease survive longer than 5 years ⁴. This poor prognosis for late-stage patients punctuates the necessity of revealing novel and exploitable vulnerabilities in CRC.

In 80% of cases, CRC is characterized by hyperactivation of WNT signalling ⁵. This is predominantly caused by truncating mutations in the tumour suppressor gene Adenomatous Polyposis Coli (APC). These truncating mutations are causative to the impaired ability of the WNT destruction complex to degrade the WNT effector β-Catenin, causing the onset of degradation resistant tumours, or short, DRTs ^{5, 6}. Several ‘hotspot’ mutations have been described for APC, including the catenin inhibitory domain (CID), resulting in various variants of APC truncation identified in patients ⁷. This domain contains the β-Catenin binding 20 amino acid repeats (20 AAR). Consequently, APC-truncated tumour cells accumulate the co-activator β-Catenin, which translocates to the nucleus, displaces the TLE co-repressor and associates with the transcription factor TCF-4/LEF-1 to promote the expression of WNT pathway target genes ^{8, 9}. The accumulation of β-Catenin results in elevated cellular proliferation independent of WNT ligand availability and is believed to be a major event in disease onset of CRC ^{10, 11}.

Truncating mutations within APC impact β-Catenin abundance or activity and subsequently oncogenic transformation ^{9, 12–14 15}, as these cells only exhibit a limited ability to ubiquitylate β-Catenin ⁷. Despite a basal post transcriptional modification (PTM) activity retained by truncated APC ⁷, several E3 ligases were described in the past to modulate β-Catenin ubiquitylation independent of the WNT-destruction complex ^{16, 17}, including, but not limited to, HUWE1 ¹⁸, FBXW7 ¹⁹, JADE-1 ²⁰, UBR5 ²¹ and RNF4 ²².

Apart from ubiquitin ligases, several deubiquitylating enzymes (DUBs) were reported to deubiquitinate β-Catenin ²³, such as USP7 ¹³ or USP20 ²⁴, thereby extending the regulation of β-Catenin activity or abundance. However, whether the APC truncation status impacts the stabilization of β-Catenin by DUB enzymes is unknown. The possibility arises that truncated APC not only prevent the physiological degradation of β-Catenin, but also serves as a scaffold for β- Catenin regulating DUBs, enhancing β-Catenin accumulation in CRC, in an APC truncation specific manner. We investigated whether partial loss of discreet domains within APC allow de novo protein-protein interactions, towards which tumours could become addicted.

Here, we identified the DUB Ubiquitin Specific Protease 10 (USP10) as a direct binder of β-Catenin. The protein interaction between the DUB and the WNT effector only occurs in cells which have lost all AAR domains within APC, mutations present in around 30% of all CRC patients ^{25, 26}. USP10 is crucial in the tumorigenic signalling of β-Catenin and its abundance strongly enhanced WNT target gene expression and is required to maintain a stemness-like expression profile, in CRC and murine tumour organoids, respectively. Moreover, this mechanisms is evolutionarily conserved as elimination of Usp10 suppressed the gut progenitor hyperproliferation phenotype and reduced survival in a homozygous Apc (Apc^Q8/Q8) truncated D.melanogaster model ²⁷. Controlling USP10 abundance suppressed patient organoid growth and WNT and EMT signalling, while upregulating the unfolded stress response. Loss of USP10 reduced NOTUM signalling^{28, 29}, a prerequisite for competitive tumour cell growth, and prevented tumour engraftment and growth in vivo.

Overall, our study demonstrates that USP10 stabilises β-Catenin, promotes WNT-signalling and cancer stemness, essential to tumour onset and development in an APC^ΔAAR-truncation dependent manner. Taken together our results suggest that targeting USP10 could thus be a viable treatment option for the large cohort of CRC patients carrying APC^ΔAAR-truncating mutations.

USP10 is a novel regulator of β-Catenin signalling in CRC and correlates with poor patient survival

In colorectal cancer, increased protein stability of the WNT effector β-Catenin is critical for tumorigenesis and mediated either by loss of function mutations and truncations within the APC gene or via mutations of the degron motive within CTNNB1, the gene encoding β-Catenin. To investigate whether the nature of CTNNB1 APC mutations impact the stability and ubiquitylation of β-Catenin in CRC, we conducted an endogenous ubiquitin TUBE (tandem ubiquitin binding entity) assay in a panel of human CRC lines, comprising β-Catenin mutant lines (HCT116 and LS174T), or cell lines varying in the truncation length of APC, DLD-1, SW480, SW620, Colo320 and HT-29, respectively (Fig. 1a). Remarkably, and irrespective of genetic alteration we were able to detect poly-ubiquitylation of β-Catenin (Fig. 1a). This suggested that protein stability of the WNT effector could be altered in a Ubiquitin Proteasome System (UPS) specific fashion likely by additional factors. To discover such regulators, we performed a human DUB siRNA screen in the APC truncated CRC line HT-29, followed by assessing the residual protein abundance of β-Catenin (Fig. 1b, c). Analysis of the screen identified the deubiquitylase USP10 as a positive regulator of β-Catenin stability, along with previously identified DUBs (Fig. 1c and S1a, b). Indeed, we verified that loss of USP10 resulted in the depletion of the cytosolic and nuclear pool of β-Catenin, as seen by immunofluorescence (Fig. 1c).

Since USP10 was not implicated in intestinal homeostasis and β-Catenin signalling, next, we determined expression level of USP10 by interrogating publicly available patient data of colorectal cancer ²⁵. While USP10 was rarely mutated in CRC, it was predominantly upregulated, along with CTNNB1, when compared to adjacent wild-type tissue (Fig. 1d and S1c), and USP10 and CTNNB1 demonstrated a significant degree of correlation of expression in CRC samples (Fig. 1e, n^T=275, n^WT = 349, Spearman coefficient R = 0.79). Remarkably, elevated expression of USP10 in CRC is a strong indicator of overall poor patient survival in CRC (Fig. 1f), especially in the molecular subtypes CMS2-4 (S1e). Prompted by this observation, we next studied USP10 and β-Catenin levels by Immunohistochemistry (IHC) of tissue micro arrays (TMA) of CRC patients comprising non-transformed and tumour tissue. Not only was a difference in tissue architecture observed in CRC tissue samples, but USP10 and β-Catenin indeed showed a significant upregulation in CRC when compared to the adjacent tissue (Fig. 1g). Elevated protein abundance was furthermore analysed by using human samples from CRC resection surgeries, subjected to immunoblotting against endogenous USP10, β-Catenin and MYC. Non-transformed, adjacent tissue served as control. USP10, β-Catenin and the oncogene MYC were increased in tumour-samples compared to matched non-transformed tissue samples (Fig. 1h).

Publicly available data did highlight that expression of USP10 and CTNNB1 were elevated irrespective of CRC stage (Figure S1f). This was further validated using single cell sequencing data from two individual datasets, which demonstrated a tumour-specific increase in USP10 expression when compared to non-transformed, normal tissue (Figure S1g). We tested this observation in human and murine intestinal wild typic and tumour organoid models regarding the regulation of USP10 (Fig. 1i). Similar to the observation in patent-derived primary resected CRC tumours, the endogenous protein levels of USP10 was significantly increased in tumour derived organoids, compared to non-oncogenic (Fig. 1i). This was further supported by analysing publicly available expression data of patient derived non-oncogenic and CRC organoids ³⁰ (Fig. 1j).

These data propose that USP10 is a novel regulator of β-Catenin stability, and a putative involvement of USP10 in WNT signalling, intestinal homeostasis and carcinogenesis.

Genetically engineered murine models of intestinal cancer demonstrate the upregulation of USP10 as an early event in CRC formation

To further investigate the expression of USP10 in the intestine, we examined weather USP10 is expressed in the intestinal stem cell niche and analysed unperturbed crypts (Fig. 2a, b and S2a, b). Using fractionation of murine small intestine followed by immunoblotting we observed specific expression of Usp10 in crypts but not in villi (Fig. 2a). We confirmed that Usp10 was highly abundant in intestinal crypts when compared to the villus (Fig. 2b and S2a, b), and nuclear localised in intestinal stem cells using immunofluorescence (Usp10⁺/β-Catenin^nuclear; Usp10⁺/Cd44^high; Usp10⁺/Lysozyme⁻; Fig. 2b and S2b).

Intrigued by the expression of USP10 in unperturbed intestinal tissue, we tested whether the abundance of USP10 is enriched upon transformation of intestinal cells. First, we performed histopathologic analysis of individual tumours in animals, where carcinogenesis was induced either by acute CRISPR editing of Apc, causing a truncation at exon 10 (APC^ex10), or by loss of heterozygosity of Apc in a well-established mouse model of spontaneous intestinal cancer, Apc^min/+ (Fig. 2c-f, S2c-f). Irrespective of genetic alteration of Apc, causal to carcinogenesis we observed that the protein level of Usp10 was significantly upregulated in tumours within the GI tract, along with elevated protein levels of β-Catenin, when compared to non-transformed or non-oncogenic intestinal epithelium, respectively (Fig. 2e, f and S2e, f).

Next, we wondered if discreet genetic alterations could be a contributor to USP10 upregulation. To address this question, we used murine wild type organoids (Cas9) and employed CRISPR gene editing to generate Apc:Kras^G12D (AK) or Apc:Trp53:Kras^G12D (APK), rorganoids (Fig. 2g, h and S2g). Loss of Apc induce morphologic changes of wild type murine organoids and alleviated the requirement for WNT activating components in the growth medium. This is in accordance with previous reports, and likewise mutation of Kras to Kras^G12D upregulated endogenous Erk1/2 signalling and alleviated the requirement for EGF supplementation (Figure S2g) ³¹. In accordance the protein level Iof Usp10 and β-Catenin was significantly enriched in transformed organoids when compared to parental control (Fig. 2h).

Thus, the upregulation of USP10 in colorectal cancer is an early event, caused by oncogenic transformation irrespective of genetic driver complexity, and coincides with elevated abundance of the WNT effector β-Catenin.

Truncation of APC allows for de novo protein-protein interaction between USP10 and β-Catenin in CRC

We hypothesized that β-Catenin directly interacts with USP10 in CRC and tested whether mutations within either β-Catenin or APC are a prerequisite to enable a protein-protein interaction. To this end, we co-immunoprecipitation endogenous USP10 and β-Catenin in the human CRC lines HCT116 (CTNNB1^mutant/APC^wildtype) and HT-29 (CTNNB1^wildtype/APC^truncated, Fig. 3a, b). While USP10 and β-Catenin were singly immunoprecipitated in HCT116, no co-precipitation was observed. In contrast, USP10 co-immunoprecipitated with endogenous β-Catenin in HT-29 cells (Fig. 3a). This observation highlighted the possibility that the truncation status and length of APC is involved in the interaction between USP10 and β-Catenin. To examine this point, we used a panel of human CRC lines, comprising LS174T, DLD-1, SW480, SW620, Caco-2 and Colo320, which harbour varying truncation mutations within APC (Figure S3a). Endogenous co-immunoprecipitation of USP10 and β-Catenin only occurred in Colo320, confirming that a proximal truncation in APC is a prerequisite, as cell lines carrying distal deletions, such as DLD-1, Caco-2 and SW480/SW620, failed to co-immunoprecipitate USP10 with β-Catenin (Figure S3b). The truncation status within APC could have potential therapeutic implications, since analysing patient survival data and stratification of CRC patients regarding APC status (truncation within first 1000 amino acids or after, or carrying point mutations), indicated a trend towards shorter survival for short APC variant carriers (S3a).

Furthermore, based on the genetic alterations reported for HT-29 and Colo320, we concluded that the 15- and 20 β-Catenin binding amino acid Armadillo repeats (AAR) domains within APC are required to directly compete for binding of USP10 to β-Catenin. Hence, we postulate that the putative de novo interaction between USP10 and β-Catenin requires the loss of the AAR domains. Proximity ligation assays (PLA) between USP10 and β-Catenin in either control or CRISPR mediated APC-truncated HCT116 cells confirmed that APC competed with USP10 for binding to β-Catenin (Fig. 3c, d).

To further interrogate the interaction and map the USP10 binding site required for β-Catenin interaction, we conducted a µSPOT protein binding assay ³². Here, the intrinsically disordered regions (IDR) of USP10, along with the IDR sequences of a known β-Catenin binder, USP7 ¹³, or USP36 (an additional DUB comprised of large unstructured regions), were displayed as overlapping peptide libraries and probed with recombinant β-Catenin (Fig. 3e). We identified USP10 residues ⁷QYIFGDFSPDEFNQF²¹ (Fig. 3f-h) to mediate direct and robust binding to β-Catenin. Intriguingly, by assessing the binding affinity of β-Catenin towards its known interactors Axin1, APC or TCF4 we found that the presence of the USP10 peptide sequence did squelch binding of β-Catenin away from AXIN1 and APC, pointing towards a high affinity of β-Catenin towards USP10 and TCF4 (Figure S3d, e). This was further studied using AlphaFold2 MultimerV1.0 (AF2M) ³³, with the complete sequences of USP10 and β-Catenin as input, predicts the same residues within USP10 to engage with β-Catenin, as identified by µSPOT protein binding assay (Figure S3f, g). Remarkably, this binding site is overlapping with APC and AXIN1 binding to β-Catenin (Figure S3f, g).

Taken together, we discovered a direct USP10-β-Catenin interaction as well as that both USP10 with APC compete for the same β-Catenin binding site. Thus, lending a molecular explanation for the observed indirect β-Catenin stabilizing effect of APC truncations.

Acute deletion of Usp10 in intestinal stem cells of D. melanogaster rescues hyperplasia and lethality of the Apc ^Q8/Q8 model

As USP10 and the entire Wnt pathway is highly conserved between species we used D. melanogaster to investigate its involvement in intestinal homeostasis and hyperproliferation upon loss of function mutations within APC, ³⁴ (Fig. 4 and S4). Firstly, we assessed the impact of shRNA-mediated elimination of dUsp10 on intestinal progenitor homeostasis. Intestinal progenitor cells were marked by GFP expression, driven under the control of the escargot regulatory region (esg::GAL > GFP), and immunofluorescence against armadillo, the fly ortholog to β-Catenin, that is expressed on the surface of intestinal stem cells (ISCs, Figure S4a). Expression of shRNA against Usp10 resulted in a marked reduction of ISCs when compared to a LacZ control shRNA (Figure S4a and b).

Moreover, we tested for a genetic interaction between APC truncation and USP10 in a tumour-like setting using the Apc^Q8 hyperplasia model ³⁵. The allele Apc^Q8 harbours a premature stop codon leading to a significant truncation of Apc and loss of the β-catenin binding sites. Immunofluorescent analysis of D. melanogaster midguts revealed that heterozygous loss of Apc had a minor effect on overall tissue homeostasis highly similar to wildtype midguts. In contrast, midguts derived from adult animals carrying a homozygous LOF truncating mutation within Apc (Apc^Q8/Q8) presented an entirely disorganised intestine (Fig. 4a). This midgut was robustly populated by escargot-positive progenitors, many of them expressing the stem cell marker and Notch ligand Delta. (Fig. 4a, b). Knockdown of dUsp10, however, suppressed the stem cell and progenitor expansion observed in homozygous Apc^Q8, and animals presented a midgut resembling a normal appearance (Fig. 4a, b). Analysis of isolated midguts from either Apc^Q8/+, Apc^Q8/Q8 or Apc^Q8/Q8 flies expressing an shRNA against dUsp10 (Usp10i;Apc^Q8/Q8) in intestinal stem cells indicated a significant reduction in overall Usp10 transcript abundance, along with reduced expression of the stem cell marker escargot (Fig. 4c and S4e).

Lastly, we investigated the impact of dUsp10 deletion on overall survival in the background of Apc-truncation driven hyperplasia model. While the survival of heterozygous Apc^Q8/wt flies was similar to wildtype files, homozygous Apc^Q8/Q8 mutation were characterized with a temperature-sensitive lethality (Fig. 4d and S4f). Strikingly, expression of an shRNA against dUsp10 in intestinal progenitors (Usp10i;Apc^Q8/Q8) restored longevity, likely by negating the adverse effects on overall tissue homeostasis and growth imprinted by Apc^Q8/Q8 (Fig. 4d).

These data demonstrate an epistatic genetic linkage between USP10 and truncated APC that is required for ectopic stem cell proliferation.

USP10, via controlling β-Catenin protein stability, regulates WNT signalling and stemness signature genes

To further elucidate the function of USP10 in CRC in the context of APC truncation, we deleted endogenous USP10 by co-targeting of exon 2 and 10 in HT-29 and HCT116, respectively (Fig. 5a and S5a, b). Depletion of USP10 in HT-29 resulted in a marked reduction of β-Catenin protein, but not its mRNA transcript, along with reduction in the CRC protein marker and WNT target gene LGR5 (Fig. 5a, b). Loss of USP10 enhanced overall ubiquitylation of β-Catenin (Fig. 5c) and accelerated protein turnover in HT-29 cells (Fig. 5d and quantified in 5e). Depletion of USP10 in HCT116 a cell line that harbour non-truncated Apc, however, had no effect on overall β-Catenin abundance nor ubiquitylation (Figure S5b-d), confirming the dependency of the USP10-β-Catenin interaction on APC-truncation. Interestingly, while cells deleted for USP10 by CRISPR mediated targeting did show reduced abundance and increased ubiquitylation of β-Catenin, longitudinal propagation of HT-29^ΔUSP10 was not possible. Targeted cells within a heterogeneous cell pool were rapidly outcompeted by wildtype cells (Figure S5e). This is in line with previous reports of cell lethality upon USP10 loss ³⁶. To by-pass this long-term lethality, we used an inducible knock down system, comprising two independent shRNA against USP10, to acutely deplete the DUB in HT-29 (Figure S5f). USP10 depleted HT-29 showed a significantly reduced proliferation, when compared to control vector transduced cells (Figure S5g).

A stem cell niche-specific contribution of USP10 was further supported by analysing the whole proteome of HT-29 cells treated with either non-targeting (ctrl) or USP10 siRNA for 24 hours (Figure S5h, i). Among the downregulated proteins were proteins associated with the stem cell niche, including TCF4 (TCF7L2), TNFRSF21, NOTCH2, LGR4, CD44, along with reduced protein level of the proto-oncogene MYC, a direct target of WNT signalling (Figure S5h).

To investigate the extent of regulation of the WNT effector β-Catenin by USP10, and using a gain-of-function approach, we conditionally overexpressed either wild type (USP10^WT) or a catalytic inactive variant of USP10 (USP10^CA) in the CRC line HT-29. Conditional increase in USP10 led to an increase in β-Catenin abundance on protein as well as mRNA level (Fig. 5f, g). The catalytic activity of USP10 is required to facilitate these effects on β-Catenin, as USP10^CA failed to stabilise β-Catenin (Fig. 5f, g). Expression of USP10 significantly enhanced overall proliferation of HT-29 cells, when compared to vector or catalytic inactive mutant control cells (Fig. 5h). Given that β-Catenin directly controls intestinal homeostasis and stem- and cancer cell maintenance, next, we tested if USP10 affects the expression of essential stem- and CRC pathways. Immunofluorescence imaging of HT-29 expressing either USP10^WT or USP10^CA demonstrated that proteins associated with the CSC stem niche, such as β-Catenin, OLFM4, LGR5, ASCL2 or CD44 were significantly upregulated in a USP10^WT dependent fashion (Fig. 5i, k).

These observations establish that USP10 regulates the ubiquitylation and abundance of β-Catenin in an APC truncation dependent manner, promoting the expression of WNT pathway and (cancer) stem cell signatures and CRC growth.

USP10 is required to maintain CRC cell identity, stemness and tumour growth

To investigate the clinical relevance and dependency of human CRC tumours towards USP10 in a patient-relevant setting, we used patient-derived organoids (Fig. 6a). The patient organoid line P6T carries mutations comparable to HT-29; a truncating mutation resulting in a short APC variant (R^876*) and a longer variant (P^1420fs), making it a suitable candidate to test USP10 dependency. 3 weeks post infection and selection with either a non-targeting control shRNA (shNTC) or an shRNA targeting USP10, patient-derived organoids were analysed (Fig. 6b-h). Loss of USP10 significantly reduced overall organoid numbers and size (Fig. 6c, d). Transcriptomic analysis of P6T^shNTC and P6T^shUSP10 organoids revealed that USP10 is involved in the regulation of WNT signalling, differentiation and stem cell maintenance (Fig. 6f, g). Stem cell-related genes, such as LGR5, LEF1, AXIN2 or LRIG1 were reduced upon loss of USP!0, while the expression of differentiation associated genes, such as MUC2 or KRT20, were enriched (Fig. 6f). Furthermore, loss of USP10 led to enriched gene sets associated with stress signalling, such as unfolded protein response and reactive oxygen species signalling in P6T tumour organoids (Fig. 6h). These observations are in line with the results obtained from HT-29 cells and clearly demonstrate that USP10 is involved in the maintenance/propagation of a pro-tumorigenic signature, supporting stem cell-like features of cells expressing high levels of USP10 that is required for the tumorigenic state.

In addition, we tested whether these observations are conserved in murine models of CRC using murine intestinal organoid cultures (Fig. 2m). APK9 organoids were transduced with AAV encoding either shRNA against Usp10 or non-targeting control, respectively (Fig. 6i-m and S6a). Depletion of Usp10 was confirmed by immunoblotting (Fig. 6j). Transcriptomic analysis of APK9 organoids revealed that Usp10 is required to maintain WNT signalling, as loss of Usp10 significantly reduced this pathway (Fig. 6k-m). Knock down of Usp10 resulted in reduced abundance of β-Catenin as well WNT signalling target genes, such as Myc and Ccnd2 (Fig. 6mand S7b). Overall, reduction of Usp10 in APK9 organoids reduced signatures associated with stemness and induced the expression of differentiation gene signatures (Fig. 6k, l and S6b).

These results show that USP10, in CRC at least, contributes to the control of differentiation and can be linked to intestinal cancer cell identity. Hence, USP10, via β-Catenin, promotes intestinal cancer stemness and propagation.

Loss of Usp10 opposes competitor signalling and restores a wild-typic niche

Recently, it was shown that cancer cell eliminate the non-transformed intestinal stem cells by clonal competition derived by Apc-dependent Notum signalling, that induces the death of the naïve stem cells termed super competitor phenotype. This impact of the cancer cell on non-transformed neighbouring naive intestinal stem cells was shown to be crucial for tumour development ^{37, 38 28}. Given the strong impact on WNT signalling and the extended control of β-Catenin by USP10, we examined whether Usp10 is required for the super-competitor phenotype and, specifically, if silencing of USP10 could oppose this signalling axis. Towards this end, we cultured wild type organoids in the presence of pre-conditioned medium from either APK^shNTC or APK^shUsp10 organoids and assessed wild type organoid survival (Fig. 6f and S6d). While established wild type organoids grew in ENR medium, exposure to APK^shNTC derived medium rapidly resulted in wild type organoid loss (Fig. 7a, b and S7a, b)). Remarkably, when cultured in medium from APK^shUsp10 organoids, most wild type organoids survived an extended time under these conditions (Fig. 7a, b and S7a,). Analysis of the transcriptome of APK^shUsp10−2 organoids revealed that Notum, along with genes associated with a super competitor signature, such as Dkk2, Dkk3 or Wif1 ^{28, 29}, were downregulated upon loss of Usp10 (Figure S7b). This is in line with the observation that NOTUM was downregulated in patient derived CRC tumour organoids upon silencing of USP10 (Fig. 6f).

To further investigate in vivo if reduced WNT signalling, along with decreased NOTUM expression does affect tumour growth, we transplanted APK9 organoids stably expressing shNTC or shUsp10, respectively, in immune-competent C57Bl6/J mice (Fig. 7c). 24 weeks post-transplant mice were sacrificed, and tissue samples analysed. Mice transplanted with non-targeting control shRNA developed a large tumour, in line with previous reports for this genotype^39–41. Loss of Usp10, while resulting in tumour organoid engraftment, led to smaller lesions (Fig. 7d-f). Overall, tumour cells derived from shNTC organoids showed an enriched abundance of Usp10 and β-Catenin, which is in line with our previous observations (Fig. 7e, f). Loss of Usp10 led to a reduction in β-Catenin abundance (Fig. 7e, f). Remarkably, the NTC infected organoids showed morphologic resemblance to invasive tumours, while APK9^shUsp10 showed an enriched abundance of mucus-secreting cells (Fig. 7e, star).

USP10 is required to maintain active WNT signalling, driving a stemness-like signature and is required to stabilise via β-Catenin and drive oncogenesis. Its loss ameliorates tumour growth abrogate the super competitor phenotpye of APC, and initiates the differentiation of tumours. Overall, USP10 presents a vulnerability for CRC, at point of induction and propagation.

Colorectal cancer is mostly driven by deregulated WNT signalling, resulting in hyper-activation of the proto-oncoprotein β-Catenin and constant transcription of its target genes ¹¹. This is predominantly initiated via loss of function mutations within the tumour suppressor gene APC; around 80% of all CRC patients harbour these mutations ^{4, 26}. It is noteworthy that truncation mutations vary regarding localisation and the therefore resulting remaining domains within the APC protein ⁷. Recent studies have reported that tumour progression, loss-of-heterozygosity, and deregulation of WNT signalling are indeed affected by the length of the remaining tumour suppressor ⁴². Furthermore, aberrant WNT signalling, caused by APC truncation, is a prerequisite for the establishing of a super competitor cell phenotype, manifesting CRC onset ^{28, 29, 32}. Despite our growing understanding of the underlying genetic causalities of colorectal cancer, the identification of mechanisms at the basis of this phenomenon and identification of suitable therapeutic targets presents a challenge that yet needs to be overcome ⁴³.

One promising strategy is targeting enhanced onco-protein stability in cancer, in particular via the Ubiquitin Proteasome System (UPS). Despite the loss of either APC or mutations within the degron motive of β-Catenin, the WNT effector is still ubiquitylated, at least, in CRC-derived cell lines ^{7, 9}. Several E3 ligases have been reported in the past to ubiquitylate and thereby regulate the stability and activity of this proto-oncogene ^18–22. This opens an intriguing possibility to target the protein levels of β-Catenin by interfering with enzymes conferring stabilisation. As a potential therapeutically relevant druggable family of proteins DUBs are of interest, as this class of enzymes opposes substrate ubiquitylation and can contribute to protein stabilisation and activation ⁴⁴.

In this study, we investigated the possibility to control the abundance of β-Catenin protein via DUBs that are relevant to CRC that is driven by loss of the tumour suppressor APC. Via unbiassed screen we discovered that USP10 enhances of β-Catenin abundance. USP10 was reported to control the protein stability of TP53 ^{45, 46} and contribute to control of autophagy ^{47, 48}, DNA damage ^{49, 50} and metabolic signalling ^{50, 51}, all processes that tumour cells highly rely on ⁵². Analysing publicly available patient data and samples from local CRC patients, we found that USP10 is frequently upregulated in human CRC tumours and is often co-expressed with β-Catenin. Intriguingly, by utilising a limited panel of human CRC cell lines, we observed that USP10 only interacted with β-Catenin when the truncation mutation within APC resulted in loss of the AAR domains. This observation was further corroborated by using CRISPR targeting of APC in otherwise APC non-mutant CRC lines. Microarray-based binding assays identified residues 7–21 of the unstructured N-terminus of USP10 to directly interact with β-Catenin. Comparison of the resulting structural model with the APC-β-Catenin complex lends a molecular explanation for the observed direct competition with APC. An overlapping binding site of APC and USP10 on β-Catenin would also rationalize the need for APC truncation for successful USP10 co-immunoprecipitation. Additionally, this observation could explain how APC mutations indirectly control β-Catenin abundance without affecting the ubiquitylation level per se. Taken together, our data extends the role of WNT signalling and β-Catenin in CRC and proposes patient stratification towards USP10 dependency. This is a novel path to the possibility of β-Catenin and allows to transform APC from a diagnostic marker towards an actionable vulnerability within CRC. Furthermore, our observation reports a direct interaction between USP10 and β-Catenin, confirming that both proteins do regulate tumour-intrinsic processes. Recently, this axis has been linked to drive NSCLC progression and metastasis⁵³. Here, in contrast to our work, USP10 stabilized HDAC7, which in turn controlled β-Catenin abundance.

Recently, an alternative mechanism of regulation of the WNT pathway by USP10 was reported ⁵⁴. Here, during zebrafish development, USP10 contribute to the degradation of β-Catenin by recruiting and stabilising AXIN1 binding to β-Catenin. This is an interesting observation and highlights the possibility of substrate and functional specificity in a tissue specific context. This has been reported for additional enzymes belonging to the UPS family, such as USP28^55–57. In cancer, the function of USP10 as a tumour suppressor or proto-oncogene, is still not fully solved. Recent work demonstrated both functions in the same tumour entity, such as NSCLC^{53, 58} or CRC^{54, 59}. The reason for these differences is not well understood but its investigation and vigorous testing is required to elucidate the therapeutic potential of this deubiquitylase, USP10.

While USP10 regulates the abundance of β-Catenin by affecting its ubiquitylation, its loss had widespread consequences for APC truncated CRC. Altering USP10 abundance affected the expression of genes associated with proliferation, stemness and disease progression. USP10 imprints the cellular identity of cancer cells, as loss of USP10 altered the transcriptional profiles towards a non-transformed, differentiated state. Not only does USP10 control tumour intrinsic pathways and biological processes regulated by WNT, loss of USP10 suppressed cell death of non-transformed cells exposed to cultured medium from tumour organoids. Given that the super competitor signalling cascade ^{28, 29, 38} leads to the secretion of signalling molecules initiating cell death, loss of USP10 controls extrinsic signalling mechanisms and thereby opposes tumour growth. It is worth noting that the catalytic activity of USP10 was required for the effects observed; overexpression of the catalytic inactive form had no effect on global β-Catenin abundance nor on proliferation. Hence, the catalytic active site presents a suitable target site ⁶⁰.

The observed effects in gene expression in the USP10 shRNA cell lines and organoids were a direct consequence of reduced β-Catenin protein abundance. These processes are highly conserved, as by employing D.melanogaster intestinal hyperproliferation models and silencing of Usp10 in the intestinal stem cell niche demonstrated that targeting USP10 in vivo, indeed, did ameliorate the APC phenotype caused by homozygous loss of the tumour suppressor (APC^Q8/Q8) ³⁴. This observation is further supported by the organoid transplant model presented. The observed phenotype is in line with previous reports that inhibition NOTUM in established or transplanted CRC tumours does suppress tumour progression⁴⁰. Here, pharmacologic targeting of USP10 by small molecule inhibitors⁶¹ could extend the spectrum of possible intervention and open new avenues for CRC therapy.

Utilizing human primary patient material, colorectal cancer cell lines as well as genetically tailored murine organoids and Drosophila melanogaster as a model system for aberrant WNT signalling in the intestine, we unravelled a novel protein-protein interaction to which CRC are addicted. Taken together, our study provided robust in vitro and in vivo evidence that USP10 functions as a driver of CRC and could serve as a therapeutic relevant target in a distinct subset of APC-truncated patient cohort.

Conflict of Interest: The authors declare no potential conflicts of interest.

Ethics approval and Consent to participate

In vivo experiments concerning CRISPR mediated oncogenesis or the Apc^min/+ mouse model were approved by the Regierung Unterfranken and the ethics committee under the license numbers 2532-2-555, 2532-2-556, 2532-2-694 and 2532-2-1002. The mouse strains used for this publication are listed. All animals are housed in standard cages in pathogen‐free facilities on a 12‐h light/dark cycle with ad libitum access to food and water. FELASA2014 guidelines were followed for animal maintenance.

Mouse experiments concerning AOM/DSS induced oncogenesis were reviewed and approved by the Regierungspräsidium Darmstadt, Darmstadt, Germany.

Human colorectal cancer samples, irrespective of sex, were obtained from the Pathology Department at the University Hospital Würzburg (Germany). Informed consent was obtained from all patients. Experiments agreed with the principles set out in the WMA Declaration of Helsinki and the Department of Health and Human Services Belmont Report. Samples were approved under Ethics Approval 17/01/2006 (University Hospital Würzburg).

P6T human colorectal cancer organoids were established in a previous study³⁰ and obtained a following material transfer agreement with Hubrecht Organoid Technology. The collection of colorectal tissue for the generation of CRC organoids was performed according to the guidelines of the European Network of Research Ethics Committees (EUREC) following European, national and local law. In all cases, patients signed informed consent after ethical committees approved the study protocols.

Consent for publication

We have obtained consent to publish this paper from all the study participants.

Author contributions

Conceptualization: M.R., M.E.D.

Methodology: M.R. (in vitro, Co-IP, in vivo, TUBE, organoids, RNA Seq., Bioinformatic); O.H. (in vivo IHC and IF), S.T. (D. melanogaster), C.P.G. & A.C.J. (Mass Spec.), D.S. & S.L. (PLA); C.F. (in vitro), N.P. (in vitro), M.P. & F.G. (in vivo transplant), J.B. & M.M. (organoids & Bioinformatic), A.N. & E.W. (bioinformatic); C.S. & H.M. (SPOT Array), C.S.V. (Operetta system), C.A. (loading Ilumina), V.B. & M.Ro. (human FFPE samples), A.W. (resection material for human organoids).

Formal analysis: M.R. (in vitro, in vivo); M.R., J.B. (Bioinformatic), O.H., M.Ro. and M.E.D. (Pathology); S.T., A.O. & P.G. (D.melanogaster); C.S. & H.M. (SPOT); C.P.G.& I.D. (Mass Spec), M.P. & F.G. (in vivo)

Investigation: M.R., O.H., H.M., A.O. M.Ro., P.G., M.E.D.

Resources: A.W., M.P., V.L., M.M., M.Ro., H.M., F.G., A.O., P.G., M.E.D.

Writing-original draft: M.E.D.

Writing-review and editing: M.R., V.L., A.W., M.M., M.E., H.M., F.G., M.Ro., A.O., P.G. M.E.D.

Supervision: M.E.D.

Funding acquisition: M.E.D.

Acknowledgements

The Operetta High Content Microscope was funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) -440766788 (INST 93/1023-1 -FUGG).

Financial Support: MR is funded by the DKH MSNZ Wuerzburg, DFG-GRK 2243 and IZKF B335. OH is supported by the German Cancer Aid via grant 70112491 and 70114554. M.E. is supported by the TransOnc priority programme of the German Cancer Aid within grant 70112951 (ENABLE). AO and MED are funded by the German Israeli Foundation grant 1431 1431 and ICRF project grant to AO. Open Access funding enabled and organized by Projekt DEAL. H.M.M and C.S. are funded by the DFG (DFG MA6957/1-1).

Sung, H. et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 71, 209-249 (2021).
Half, E., Bercovich, D. & Rozen, P. Familial adenomatous polyposis. Orphanet J Rare Dis 4, 22 (2009).
Siegel, R.L., Miller, K.D., Fuchs, H.E. & Jemal, A. Cancer Statistics, 2021. CA: A Cancer Journal for Clinicians 71, 7-33 (2021).
Novellasdemunt, L., Antas, P. & Li, V.S. Targeting Wnt signaling in colorectal cancer. A Review in the Theme: Cell Signaling: Proteins, Pathways and Mechanisms. Am J Physiol Cell Physiol 309, C511-521 (2015).
Abu Ahmad, Y., Oknin-Vaisman, A., Bitman-Lotan, E. & Orian, A. From the Evasion of Degradation to Ubiquitin-Dependent Protein Stabilization. Cells 10 (2021).
Ranes, M., Zaleska, M., Sakalas, S., Knight, R. & Guettler, S. Reconstitution of the destruction complex defines roles of AXIN polymers and APC in beta-catenin capture, phosphorylation, and ubiquitylation. Mol Cell 81, 3246-3261 e3211 (2021).
Morin, P.J. et al. Activation of beta-catenin-Tcf signaling in colon cancer by mutations in beta-catenin or APC. Science 275, 1787-1790 (1997).
Yang, J. et al. Adenomatous polyposis coli (APC) differentially regulates beta-catenin phosphorylation and ubiquitination in colon cancer cells. J Biol Chem 281, 17751-17757 (2006).
Zhao, H. et al. Wnt signaling in colorectal cancer: pathogenic role and therapeutic target. Mol Cancer 21, 144 (2022).
Zhan, T., Rindtorff, N. & Boutros, M. Wnt signaling in cancer. Oncogene 36, 1461-1473 (2017).
Hankey, W., Frankel, W.L. & Groden, J. Functions of the APC tumor suppressor protein dependent and independent of canonical WNT signaling: implications for therapeutic targeting. Cancer Metastasis Rev 37, 159-172 (2018).
Novellasdemunt, L. et al. USP7 Is a Tumor-Specific WNT Activator for APC-Mutated Colorectal Cancer by Mediating beta-Catenin Deubiquitination. Cell Rep 21, 612-627 (2017).
Barua, D. & Hlavacek, W.S. Modeling the effect of APC truncation on destruction complex function in colorectal cancer cells. PLoS Comput Biol 9, e1003217 (2013).
Schneikert, J., Grohmann, A. & Behrens, J. Truncated APC regulates the transcriptional activity of beta-catenin in a cell cycle dependent manner. Hum Mol Genet 16, 199-209 (2007).
Gao, C., Xiao, G. & Hu, J. Regulation of Wnt/beta-catenin signaling by posttranslational modifications. Cell Biosci 4, 13 (2014).
Chen, C. et al. YAP-dependent ubiquitination and degradation of beta-catenin mediates inhibition of Wnt signalling induced by Physalin F in colorectal cancer. Cell Death Dis 9, 591 (2018).
Dominguez-Brauer, C. et al. E3 ubiquitin ligase Mule targets beta-catenin under conditions of hyperactive Wnt signaling. Proc Natl Acad Sci U S A 114, E1148-E1157 (2017).
Jiang, J.X. et al. Tumor suppressor Fbxw7 antagonizes WNT signaling by targeting beta-catenin for degradation in pancreatic cancer. Tumour Biol 37, 13893-13902 (2016).
Chitalia, V.C. et al. Jade-1 inhibits Wnt signalling by ubiquitylating beta-catenin and mediates Wnt pathway inhibition by pVHL. Nat Cell Biol 10, 1208-1216 (2008).
Shearer, R.F., Iconomou, M., Watts, C.K. & Saunders, D.N. Functional Roles of the E3 Ubiquitin Ligase UBR5 in Cancer. Mol Cancer Res 13, 1523-1532 (2015).
Thomas, J.J. et al. RNF4-Dependent Oncogene Activation by Protein Stabilization. Cell Rep 16, 3388-3400 (2016).
Park, H.B., Kim, J.W. & Baek, K.H. Regulation of Wnt Signaling through Ubiquitination and Deubiquitination in Cancers. Int J Mol Sci 21 (2020).
Wu, C. et al. USP20 positively regulates tumorigenesis and chemoresistance through beta-catenin stabilization. Cell Death Differ 25, 1855-1869 (2018).
Cerami, E. et al. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2, 401-404 (2012).
Gao, J. et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 6, pl1 (2013).
Ahmed, Y., Hayashi, S., Levine, A. & Wieschaus, E. Regulation of armadillo by a Drosophila APC inhibits neuronal apoptosis during retinal development. Cell 93, 1171-1182 (1998).
van Neerven, S.M. et al. Apc-mutant cells act as supercompetitors in intestinal tumour initiation. Nature 594, 436-441 (2021).
Flanagan, D.J. et al. NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature 2021 594:7863 594, 430-435 (2021).
van de Wetering, M. et al. Prospective derivation of a living organoid biobank of colorectal cancer patients. Cell 161, 933-945 (2015).
Matano, M. et al. Modeling colorectal cancer using CRISPR-Cas9-mediated engineering of human intestinal organoids. Nat Med 21, 256-262 (2015).
Schulte, C. et al. Multivalent binding kinetics resolved by fluorescence proximity sensing. Commun Biol 5, 1070 (2022).
Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583-589 (2021).
Martorell, O. et al. Conserved mechanisms of tumorigenesis in the Drosophila adult midgut. PLoS One 9, e88413 (2014).
McCartney, B.M. et al. Testing hypotheses for the functions of APC family proteins using null and truncation alleles in Drosophila. Development 133, 2407-2418 (2006).
Higuchi, M. et al. USP10 Is an Essential Deubiquitinase for Hematopoiesis and Inhibits Apoptosis of Long-Term Hematopoietic Stem Cells. Stem Cell Reports 7, 1116-1129 (2016).
Flanagan, D.J. et al. NOTUM from Apc-mutant cells biases clonal competition to initiate cancer. Nature 594, 430-435 (2021).
Schmitt, M. et al. Colon tumour cell death causes mTOR dependence by paracrine P2X4 stimulation. Nature (2022).
Watanabe, S. et al. Transplantation of intestinal organoids into a mouse model of colitis. Nat Protoc 17, 649-671 (2022).
Tian, Y. et al. APC and P53 mutations synergise to create a therapeutic vulnerability to NOTUM inhibition in advanced colorectal cancer. Gut (2023).
O'Rourke, K.P. et al. Transplantation of engineered organoids enables rapid generation of metastatic mouse models of colorectal cancer. Nat Biotechnol 35, 577-582 (2017).
Albuquerque, C. et al. The 'just-right' signaling model: APC somatic mutations are selected based on a specific level of activation of the beta-catenin signaling cascade. Hum Mol Genet 11, 1549-1560 (2002).
Guren, M.G. The global challenge of colorectal cancer. Lancet Gastroenterol Hepatol 4, 894-895 (2019).
Clague, M.J., Urbe, S. & Komander, D. Breaking the chains: deubiquitylating enzyme specificity begets function. Nat Rev Mol Cell Biol 20, 338-352 (2019).
Li, H. et al. Destabilization of TP53 by USP10 is essential for neonatal autophagy and survival. Cell Rep 41, 111435 (2022).
Yuan, J., Luo, K., Zhang, L., Cheville, J.C. & Lou, Z. USP10 regulates p53 localization and stability by deubiquitinating p53. Cell 140, 384-396 (2010).
Jia, R. & Bonifacino, J.S. The ubiquitin isopeptidase USP10 deubiquitinates LC3B to increase LC3B levels and autophagic activity. J Biol Chem 296, 100405 (2021).
Liu, J. et al. Beclin1 controls the levels of p53 by regulating the deubiquitination activity of USP10 and USP13. Cell 147, 223-234 (2011).
Zhang, M. et al. Ubiquitin-specific Peptidase 10 (USP10) Deubiquitinates and Stabilizes MutS Homolog 2 (MSH2) to Regulate Cellular Sensitivity to DNA Damage. J Biol Chem 291, 10783-10791 (2016).
Bhattacharya, U., Neizer-Ashun, F., Mukherjee, P. & Bhattacharya, R. When the chains do not break: the role of USP10 in physiology and pathology. Cell Death Dis 11, 1033 (2020).
Deng, M. et al. Deubiquitination and Activation of AMPK by USP10. Mol Cell 61, 614-624 (2016).
Hanahan, D. Hallmarks of Cancer: New Dimensions. Cancer Discov 12, 31-46 (2022).
Guo, K. et al. HDAC7 promotes NSCLC proliferation and metastasis via stabilization by deubiquitinase USP10 and activation of beta-catenin-FGF18 pathway. J Exp Clin Cancer Res 41, 91 (2022).
Wang, Y. et al. USP10 strikes down beta-catenin by dual-wielding deubiquitinase activity and phase separation potential. Cell Chem Biol (2023).
Saei, A. et al. Loss of USP28-mediated BRAF degradation drives resistance to RAF cancer therapies. J Exp Med 215, 1913-1928 (2018).
Diefenbacher, M.E. et al. The deubiquitinase USP28 controls intestinal homeostasis and promotes colorectal cancer. J Clin Invest 124, 3407-3418 (2014).
Prieto-Garcia, C. et al. Inhibition of USP28 overcomes Cisplatin-resistance of squamous tumors by suppression of the Fanconi anemia pathway. Cell Death Differ 29, 568-584 (2022).
Sun, J. et al. USP10 inhibits lung cancer cell growth and invasion by upregulating PTEN. Mol Cell Biochem 441, 1-7 (2018).
Reissland, M. et al. Stabilisation of β-Catenin-WNT signalling by USP10 in APC-<em>truncated</em> colorectal cancer drives cancer stemness and enables super-competitor signalling. bioRxiv, 2023.2002.2010.527983 (2023).
Lange, S.M., Armstrong, L.A. & Kulathu, Y. Deubiquitinases: From mechanisms to their inhibition by small molecules. Mol Cell 82, 15-29 (2022).
Weisberg, E.L. et al. Inhibition of USP10 induces degradation of oncogenic FLT3. Nat Chem Biol 13, 1207-1215 (2017).
Varga, J. et al. AKT-dependent NOTCH3 activation drives tumor progression in a model of mesenchymal colorectal cancer. J Exp Med 217 (2020).

There is NO conflict of interest to disclose.

Download PDF

Journal Publication

published 23 Oct, 2024

Read the published version in Oncogene →

Editorial decision: revise
04 Jan, 2024
Review #3 received at journal
02 Jan, 2024
Review #2 received at journal
29 Dec, 2023
Review #1 received at journal
16 Dec, 2023
Reviewer #3 agreed at journal
16 Dec, 2023
Reviewer #2 agreed at journal
15 Dec, 2023
Reviewer #1 agreed at journal
15 Dec, 2023
Reviewers invited by journal
15 Dec, 2023
Submission checks completed at journal
12 Dec, 2023
Editor assigned by journal
12 Dec, 2023
First submitted to journal
12 Dec, 2023

You are reading this latest preprint version

USP10 drives cancer stemness and enables super-competitor signalling in Colorectal Cancer

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Results

Truncation of APC allows for de novo protein-protein interaction between USP10 and β-Catenin in CRC

USP10, via controlling β-Catenin protein stability, regulates WNT signalling and stemness signature genes

USP10 is required to maintain CRC cell identity, stemness and tumour growth

Loss of Usp10 opposes competitor signalling and restores a wild-typic niche

Discussion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1