Inhibition of P21-activated kinases 1 and 4 synergistically suppress the growth of pancreatic cancer by stimulating anti-tumour immunity

Inhibition of P21-activated kinases 1 and 4 synergistically suppress the growth of pancreatic cancer by stimulating anti-tumour immunity | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Inhibition of P21-activated kinases 1 and 4 synergistically suppress the growth of pancreatic cancer by stimulating anti-tumour immunity Yi Ma, Chelsea Dumesny, Li Dong, Ching-Seng Ang, Khashayar Asadi, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-3974396/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 7 You are reading this latest preprint version Abstract Background: Pancreatic ductal adenocarcinoma (PDA) is one of the most lethal types of cancer, and KRAS oncogene occurs in over 90% of cases. P21-activated kinases (PAK), containing six members (PAK1 to 6), function downstream of KRAS. PAK1 and PAK4 play important roles in carcinogenesis, but their combinational effect remains unknown. In this study, we have determined the effect of dual inhibition of PAK1 and PAK4 in PDA progression using knockout (KO) cancer cell lines. Methods: Murine wild-type (WT) and PAK1KO pancreatic cancer cell lines were isolated from PAK1 +/+ and PAK1 -/- KPC ( LSL-Kras G12D/+ ; LSL-Trp53 R172H/+ ; Pdx-1-Cre ) mice. KPC PAK4KO and KPC PAK1&4 KO cell lines were generated from KPC WT and KPC PAK1KO cell lines respectively using the CRISPR-CAS9 gene knockout technique. PAK WT and KO cell lines were used in mouse models of pancreatic tumours. Cells and tumour tissue were also used in flow cytometry and proteomic studies. A human PDA tissue microarray was stained by immunohistochemistry. Results: Double knock out of PAK1 and PAK4 caused complete regression of tumour in a syngeneic mouse. PAK4KO inhibited tumour growth by stimulating a rapid increase of cytotoxic CD8+ T cell infiltration. PAK1KO synergistically with PAK4KO increased cytotoxic CD8+ T cell infiltration and stimulated a sustained infiltration of CD8+ T cells at a later phase to overcome the immune evasion in the PAK4KO tumour. The human PDA tissue microarray study showed the important role of PAK1 and PAK4 in intra-tumoral T-cell function. Conclusion: Our results demonstrated that dual inhibition of PAK1 and PAK4 synergistically suppressed PDA progression by stimulating cytotoxic CD8+ T cell response. p21-activated kinases1&4 intra-tumoral T cells pancreatic ductal adenocarcinoma Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Pancreatic ductal adenocarcinoma (PDA) is one of the most aggressive solid cancers, with a dismal five-year survival rate of 9.8%[ 1 ]. This is mainly due to late diagnosis and lack of effective treatment for advanced disease. While surgery can be curative, only 10–20% of patients have resectable tumours on diagnosis[ 2 ]. Despite advances in systemic therapy of solid malignancies, the development of targeted therapy in PDA remains stagnant. Gemcitabine-based chemotherapy and FOLFIRINOX are still the mainstay treatment of advanced PDA, but only extend overall survival on average, by two to three months[ 3 , 4 ]. P21-activated kinase (PAK) has emerged as a potential therapeutic target. PAKs are a group of serine/threonine kinases that act downstream of KRAS, while KRAS mutations occur in over 90% of PDA [ 5 – 7 ]. There are six members of the PAK family, which are categorized into two groups: group 1 (PAK1 to 3) and group 2 (PAK4 to 6)[ 8 ]. Among the PAK proteins, PAK1 and PAK4 are mostly investigated for their role in tumorigenesis. In PDA, PAK1 was shown to inhibit cancer cell apoptosis, activate pancreatic stellate cells and down-regulate intra-tumoral CD4 + and CD8 + T cell infiltration[ 9 – 11 ]. PAK4 was reported to play a role in PDA cell apoptosis, cell cycle arrest and cancer cell stemness[ 12 – 14 ]. Recently, PAK4 was shown to suppress T cell response in melanoma, prostate cancer, and glioblastoma, which is consistent with an observed upregulation of intra-tumoral CD8 + T cells in PDA mouse model by PAK4 inhibitor PF-3758309[ 10 , 15 – 18 ]. Despite the pro-tumorigenic role of PAK1 and PAK4 in PDA, the development of clinically effective PAK1 or PAK4 inhibitors has not been successful. Given that both PAK1 and PAK4 play important roles in PDA biology, we hypothesize that the two may promote PDA tumour growth synergistically. Using PAK1 and PAK4 double knockout (KO) cell lines, we investigated their combinational effect in PDA which will guide future development of PAK-targeted therapy. Methods and Methods Cell lines and cell culture Murine WT and PAK1KO pancreatic cancer cell lines were isolated from PAK1 +/+ and PAK1 −/− KPC mice as previously described[ 10 ]. KPC PAK4KO and KPC PAK1&4 KO cell lines were generated from KPC WT and KPC PAK1KO cell lines respectively using CRISPR-CAS technique. Cancer cells were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 5% fetal bovine serum (FBS) (Hyclone Laboratories, Melbourne, Australia) in a 37°C incubator under humidified 5% CO 2 . Animal studies All mouse studies were approved by the Austin Health Animal Ethics Committee (A2022/5797). Experimental C57BL6 and SCID mice were housed in the Austin Health Bioresource Facility and monitored for health criteria. For C57BL6 syngeneic mouse model, KPC WT, PAK1KO, PAK4KO and PAK1& 4 KO cells (0.5-1x10 6 cells/100ul/mouse) were injected subcutaneously to the flank of 7 weeks old male C57BL6 mice. The mice were observed for one week or 4–6 weeks. Tumour growths were monitored using a calliper. Tumour volume (mm 3 ) was calculated using the simplified ellipsoid formula: $$Volume \left(V\right)=Length \left(L\right)\times Width \left(W\right)^2\times 0.5$$ Tumour weight (g) was measured upon mice culling. For the SCID mouse model, KPC WT, PAK4KO and PAK1&4 KO cancer cells (5x10 5 cells/mouse) were injected subcutaneously into the flank of 6-week-old male SCID mice. Mice were monitored for the time indicated in the Result section. Tumour growth and weight were obtained as described above. Patient information collection and tissue microarray generation The collection of patient’s clinical information and generation of human PDA tissue microarray (TMA) were approved by the Austin Health Human Research Ethics Committee (HREC/73948/Austin- 2021). Patients who had surgical resection of PDA between 2008 and 2019 under Austin Health were identified. Patients who had premature mortality from either delayed surgical complications, background comorbidities, incorrect disease staging or missing survival data due to transfer of care were excluded. Once recruited, participants’ baseline demographics, disease staging and grading, treatment and survival information were retrospectively collected. The formalin-fixed PDA tumour blocks were assessed by a qualified anatomical pathologist, and three 1mm diameter cores were taken from each tumour sample to assemble TMA blocks. CRISPR-CAS gene knock-out To generate PAK4KO PDA cell lines, an inducible lentiviral CRISPR/Cas9 system was used as described previously[ 19 ]. Single-guide RNA (sgRNA) oligos targeting mouse PAK4 (guide 1: CCCGCGATAAGCGCCCACT; guide 2: CGAACGATGGTCTGGGGTC) were cloned into the BsmBI site of the pFgH1tUTG GFP lentiviral vector. KPC WT and KPC PAK1KO cells were infected with lentiviral constructs encoding Cas9 and mCherry, and a doxycycline-inducible sgRNA targeting PAK4 and GFP. mCherry and GFP double-positive cells were sorted using BD FACS Aria III (BD Biosciences, New Jersey, USA). Cell clones with PAK4KO were confirmed by immunoblot. Flow cytometry analysis Mice tumour specimens were minced and digested in a digestion buffer containing 1.25mg/ml collagenase IV (Worthington Biomedical, Lakewood, USA). Digested tumours were filtered and resuspended in FACS buffer to obtain single-cell suspension (Table S1 ). 1X10 6 cells per sample were blocked with 1µl of mouse FcR blocking reagent (Miltenyi, Biotec, Bergisch Gladbach, Germany). Cell viability was assessed using 1:500 dilution of Zombie UV™ Fixable Viability Dye (BioLegend, San Diego, USA). Fluorophore-labelled antibodies against CD45, B220, CD3, CD4, CD8, and PD1 (Table S3 ) were added and incubated on ice for 20 minutes. For FoxP3 staining, cells were permeabilized and fixed with eBioscience™ FoxP3/Transcription Factor Staining Buffer Set (Invitrogen, Waltham, USA) according to the manufacturer’s instructions. Cells were incubated in FoxP3 fixation/permeabilization solution for 60 minutes in the dark at room temperature (RT). Fluorophore-labelled FoxP3 antibody was added to permeabilized cells and incubated for 30 minutes in the dark at RT (Table S3 ). For staining of cytoplasmic markers, cells were permeabilized and fixed with eBioscience™ Intracellular Fixation and Permeabilization Buffer Set (Invitrogen, Waltham, USA) according to the manufacturer’s instruction. Cells were incubated in a fixation buffer for 60 minutes in the dark at RT. Fluorophore-labelled Granzyme B and Perforin antibodies were added to permeabilized cells and incubated for 30 minutes in the dark at RT (Table S3 ). Cells were resuspended in FACS buffer and analyzed by Cytek® Aurora flow cytometer (Cytek Biosciences, California, USA). FCS Express version 7.12.0007 (De Novo Software, Pasadena, USA) was used for manual gating and statistical analysis. Global and phospho-proteomic studies The proteomic studies were conducted consequently by these steps as Sample Preparation, Liquid Chromatograph Data Independent Acquisition Mass Spectrometry , Data Search and Bioinformatic Analysis , which were detailed in the Supplementary Methods. Analysis of TMA samples The TMA blocks of 5µM sections were stained using immunohistochemistry. Samples were boiled in Tris-EDTA buffer (Table S1 ) at 99˚C for 30 minutes and then blocked with Dako REAL™ peroxidase blocking solution (Agilent Technologies, Glostrup, Denmark) at RT for 15 minutes followed by 5% goat serum at RT for an hour for endogenous peroxidase quenching and protein blocking respectively. Primary antibodies against PAK1, PAK4, CD4 and CD8 (Table S2 ) were added to samples and incubated at RT for an hour before one hour incubation with Dako EnVision + System HRP Labelled Anti-Rabbit Polymer (Agilent Technologies, Glostrup, Denmark) at RT. DAB staining was achieved with EnVision FLEX DAB + substrate Chromogen System (Agilent Technologies, Glostrup, Denmark) and samples were counterstained with haematoxylin (Sigma-Aldrich, St Louis, USA). Whole slide images were captured with Aperio AT2 digital pathology slide scanner (Leica Biosystems, Wetzlar, Germany), and analysed by QuPath version 0.4.3[ 20 ]. . PAK1 and PAK4 expression were determined using mean DAB intensity of the TMA core, while CD4 + and CD8 + T cell levels were assessed by the percentage of positive stained cells against the total number of cells in the TMA core. Mean of all three replicates was calculated for each patient for PAK1, PAK4, CD4 and CD8 levels. Univariate and multivariate linear regression was used to assess correlation between individual variables and percentage of CD4 + or CD8 + cells. Correlation between individual variables and overall survival of patients was visualised with Kaplan Meier curve and tested by univariate and multivariate Cox proportional hazards regression. Interaction term was included for multivariate regression models to assess interaction between PAK1 and PAK4. Regression analysis was conducted using R version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria). Statistical analysis All in vitro experiments were repeated in three replicates. In vivo experiments included at least three mice per group. For continuous variables, mean ± standard deviation (SD) was reported for parametric data while median ± inter-quantile range (IQR) were reported for non-parametric data. Hypothesis testing was conducted by either two-sided t test or one way ANOVA for parametric data, Mann-whitney’s U test for non-parametric data and Chi-square test for categorical results. Linear regression model was fitted for cell and tumour growth curve, and correlation coefficients (i.e. slopes) of linear fit was compared between groups. Statistical analysis was conducted with GraphPad Prism version 10.0.2 (GraphPad Software, Boston, USA), Stata BE version 17.0 (Texas, USA) and R version 4.3.0. p-value < 0.05 was considered statistically significant. Results Inhibition of PAK1 and PAK4 synergistically suppressed PDA growth. KPC WT, PAK1KO, PAK4KO and PAK1&4 KO cell lines were validated by immunoblot (Fig. S1 a). PAK1KO reduced KPC cell growth by 72 hours, but PAK4KO increased cell growth by 72 hours (Fig. S1 b). Global proteomic analysis of PAK WT and KO cells showed significantly differential protein expression profiles of PAK KO cell lines from WT cell line (Fig S2 a-c). Functional enrichment for KEGG and reactome pathways indicated changes in programmed cell death (e.g., apoptosis) and cell cycle regulation (Fig. S2 d, e), which was validated by FACS analysis. PAK1KO and PAK1&4 KO increased the apoptosis and cell death while PAK4KO reduced cancer cell death and had no effect on the apoptosis (Fig S1 c, d). In a syngeneic mouse model (Fig. 1 a), PAK4KO suppressed the in vivo tumour growth (Fig. 1 a-d) significantly. Only 30% of the mice injected with PAK4KO cells developed tumours, which also grew significantly slower than the KPC WT- and PAK1KO-inoculated tumours, even with an extra two-weeks’ growth (Fig. 1 c-e). No tumour developed in mice injected with PAK1&4 KO cells. The growth curve (Fig. 1 e) showed a peak around one week after cell injection in PAK4KO and PAK1&4 KO injected mice followed by tumour regression. This suggested that PAK4KO inhibits PDA tumour growth through modulating the adaptive immune response, which is confirmed in a SCID mouse model (Fig. 1 f-j). In SCID mice, PAK4KO did not inhibit the tumour growth at all while PAK double KO decreased the tumour growth significantly compared to PAK4KO but not to WT (Fig. 1 g-j). These results indicated that PAK4KO inhibited PDA by stimulation of anti-tumour immunity and that inhibition of PAK1 and 4 synergistically suppressed PDA progression. Inhibition of PAK1 and PAK4 synergistically increased intra-tumoral CD8 + T cell infiltration. Given the findings that PAK4 KO tumour regressed one week after cell injection, we further examined PAK WT vs KO tumour growth within the first week (Fig. 2 a, c). Tumour growth was inhibited by PAK1KO, PAK4KO and PAK1&4 KO (Fig. 2 b, d). By one week, PAK double KO further suppressed the tumour growth compared to single PAK KO, indicating a synergistic effect of PAK1 and PAK4. One week tumour specimens were then digested into single cell suspension for FACS analysis of intra-tumoral lymphocyte infiltration (gating strategy shown in Fig. S4). By one week, PAK1KO increased B cell and CD4 + T cell infiltration while PAK4KO increased the infiltration of T cells, particularly CD8 + T cells (Fig. 2 e, f). PAK1&4 KO further increased total T cell and CD8 + T cell infiltration, compared to single PAK4KO, suggesting a synergistic effect of PAK1KO and PAK4KO on intra-tumoral CD8 + T cell infiltration. PAK4KO also increased the infiltration of active CD8 + T cells at one week, shown by increased levels of granzyme B+, and granzyme B and perforin double positive cytotoxic CD8 + T cells (Fig.S5a, b). However, Granzyme B + CD8 + T cells were reduced by PAK1KO, contributing to no change in granzyme B + CD8 + T cells and less increase in granzyme B and perforin double positive cytotoxic CD8 + T cells in double KO tumour (Fig. S5a, b). The level of granzyme B + CD4 + T cells were increased in both single PAK KO tumour, with further significant increase in PAK1&4 double KO (Fig. S5c, d). The granzyme B and perforin double positive CD4 + T cells was increased only in PAK1&4 double KO tumour. These data suggested that PAK1 and PAK4 inhibition have synergistic effect on cytotoxic CD8 + and CD4 + T cells infiltration[ 21 ]. PAK4KO also decreased the infiltration of regulatory CD4 + T cells (Treg) (Fig. S5e, f). PAK1KO decreased PD-1 + CD8 + T cells while PAK4KO increased PD1 + CD8 + T cells, and PD1 + T cells were known to have higher anti-tumour activity (Fig. S5g, h)[ 22 , 23 ]. Pancreatic cancer developed immune evasion after PAK4 inhibition. After initial regression, approximately 30% of mice developed a PAK4KO tumour (Fig. 1 e), indicating immune evasion. We conducted a proteomic study to compare PAK4KO versus WT tumours at one week and four weeks after cancer cell injection to investigate the differences in immune cell infiltration. Protein expressions of PAK4KO and WT tumours were much more different at one week than four weeks (Fig. 3 a). Four protein clusters were classified based on expression patterns and were visualized as heatmap and UMAP (Fig. S6a, b). All four clusters of proteins had different protein expression levels between PAK4 KO and WT tumours at one week, and these differences were reduced in clusters 1, 2 and 4 at four weeks (Fig. S6a). KEGG and reactome pathways were functionally enriched based on the significant proteins and identified immune-related pathways (Fig. S6c, d). Given that the data from FACS analysis showed a significant role of PAK4KO in CD8 + T cell mediated immune response (Fig. 2 ), T cell receptor signaling pathway was examined in greater detail by heatmap. Multiple protein targets that are known to play an important role in downstream signaling of T cell receptor (TCR) were identified (Fig. 3 b), such as the VAV family proteins (VAV1, VAV3) and proteins that are directly involved in MHC- TCR signaling complex (CD3g, ZAP70, LCK)[ 24 , 25 ]. The expression profiles of CD3g, ZAP70 and LCK were compared between WT and PAK4KO at one week and four weeks. CD3g and ZAP70 levels were significantly higher in PAK4KO tumour at one week but dropped to similar levels as WT tumour by four weeks (Fig. 3 c, d). There was no difference in LCK levels between WT and PAK4 KO tumours at both time points (Fig. 3 e). These findings from global proteomic analysis were further supported by phosphor-proteomic results. Differentially expressed phospho-sites between WT and PAK4 KO tumours were demonstrated at one and four weeks (Fig. S6e). From the significant phospho-sites identified across all four groups, 38 kinases and their activity were predicted (Fig. S6f). Among these predicted kinases, IKKε (IKBKE) and PKCθ (PRKCQ) were predicted to have significantly higher activity in PAK4 KO tumour at one week in comparison to WT tumour (Fig. S6g) indicating a higher activity of T cell receptor pathway in PAK4KO as both IKKε and PKCθ are activated upon activation of T cell receptor[ 26 – 28 ]. However, by four weeks, the kinase activities of IKKε and PKCθ were no longer different between PAK4 KO and WT tumours (Fig. S6h) suggesting an immune evasion of PAK4KO tumour after the initial response. This is confirmed by the finding that PAK4KO tumour grew at a similar pace to WT after escaping the initial regression (Fig. 3 f-j). The PAK WT cancer cells were injected two weeks after PAK4KO cells as PAK4KO tumour grew out of the initial regression after two weeks. When comparing the PAK4KO tumour with WT tumour, there was no difference in tumour weight (Fig. 3 i) though PAK4KO tumour still demonstrated a slightly slower growth rate in comparison to WT in tumour size (Fig. 3 h). These results suggested that PAK4KO tumour developed an immune evasion after an initial phase due to a reduced immune response induced by PAK4KO from one week to four weeks. PAK1 inhibition induced CD8 + T cell infiltration at a later phase. Global proteomic analysis of PAK WT and KO tumours also identified an up-regulation of PAK1 expression from 1week to 4 weeks in both WT and PAK4KO tumours, but slightly greater in PAK4KO tumour (Fig. 4 a, b). Functional enrichment of PAK4KO identified differential expression of proteins involved in T-cell receptor signalling pathways between one week and four weeks. PPI network was constructed and annotated by log2(FC) of individual proteins. VAV family proteins, CD3g, ZAP70 and LCK were down-regulated at four weeks in comparison to one week (Fig. 4 c), suggesting a downregulation of T-cell response in PAK4KO tumour over time. Furthermore, the upregulation of PAK1 over four weeks was found to be involved in T-cell receptor signaling pathways (Fig. 4 c), indicating that the increase of PAK1 level in PAK4KO tumour over time may contribute to their immune resistance. PAK4KO tumour still had slightly higher CD8 + T cell infiltration than WT tumour by four weeks, but the difference was a lot less compared to the increase at one week (Fig. 2 e, f, Fig. 4 d, e). Furthermore, the percentage of total T cells (CD45 + CD3+) in PAK4KO tumour was no longer different to WT tumour after four weeks. On the other hand, PAK1KO tumour demonstrated a significant increase of CD8 + T cell infiltration by three weeks (Fig. 4 d, e) compared to at one week (Fig. 2 e, f), suggesting a delayed CD8 + T cell response in PAK1 KO tumour. In addition, PAK4KO reduced the percentage of Treg and increased the level of PD1 + CD8 + T cells at one week (Fig. S5e-h), which were reversed by four weeks (Fig. S7a-d), suggesting a more immunosuppressive environment. On the other hand, although PAK1KO had a high level of Treg (Fig. S7a, b), it also had an increase in PD1 + CD8 + T cell level (Fig. S7c, d) from one week to three weeks, suggesting that CD8 + T cells had increased anti-tumour activity in PAK1KO tumour over time. Together these results indicated that PAK1KO caused a delayed infiltration of CD8 + T cells which may compensate for the reduced immune response to PAK4KO over time. Low PAK1 and PAK4 expressions improved T cell function in human pancreatic cancer. To study the roles of PAK1 and PAK4 in human PDA, 178 PDA cases were assessed from the TCGA database. High PAK1 and PAK4 gene expressions were associated with reduced survival in PDA patients (Fig. S8a, b). Functional enrichment of genes correlated with PAK1 (Fig. S8c) and PAK4 (Fig. S8d) revealed immune-related pathways for PAK4 but not PAK1. Tumour purity adjusted estimation of intra-tumoral B and T cell infiltration showed a negative correlation between PAK4 and CD8 + T cell infiltration, but not with B cells or CD4 + T cells (Fig. S8f). No correlation between PAK1 and lymphocyte infiltration was demonstrated (Fig. S8e). In a TMA analysis of PDA patients (Fig. 5 a), patients were grouped into low or high expression groups based on median of PAK1 or PAK4 intensity as well as median percentage of CD4 + or CD8 + positive cells (Fig. 5 b, Fig. 6 a). Individual study variables were compared between low versus high expression groups for PAK1 and PAK4 (Table S4 and S5). To assess the relationship between PAK expression and the level of T cell infiltration, both univariate and multivariate linear regression models were applied (Table S6 and S7). By univariate analysis, PAK4 expression was positively correlated with CD4 + and CD8 + T cell infiltration and PAK1 expression was positively correlated with CD4 + T cell infiltration. However, these correlations disappeared in multivariate analysis (Table S6 and S7). Overall survival of patients was demonstrated by Kaplan Meier curve (Fig. 5 c), with a number of patients at risk and events reported. While high levels of CD4 + and CD8 + T cell infiltration significantly improved overall survival (Fig. 5 f, g), PAK1 and PAK4 expression levels did not significantly affect patient survival by univariate analysis (Fig. 5 d, e). However, high PAK4 levels increased risk of death with multivariate analysis (Table 1 ). Furthermore, PAK1 and PAK4 demonstrated negative interaction (HR below 1), indicating that a reduction in the level of either PAK1 or PAK4 would lead to worse survival outcome for the other (Table 1 ). Table 1 Univariate and multivariate Cox regression for survival analysis Univariate Multivariate Variable Hazard ratio (HR) p-value Hazard ratio (HR) p-value Age 1 0.7 1.05 0.005** Sex - Male - Female - 0.89 - 0.6 - 0.80 - 0.5 PAK1 - Low - High - 1.06 - 0.8 - 1.39 - 0.5 PAK4 - Low - High - 1.05 - 0.8 - 2.58 - 0.04* PAK1: PAK4 interaction - - 0.2 0.021* CD4 - Low - High - 0.59 - 0.028* - 0.35 - 0.003** CD8 - Low - High - 0.56 - 0.016* - 1.15 - 0.7 Cancer Site - Head/neck - Body/tail - Multifocal - 0.46 2.13 - 0.092 0.5 - 2.31 2.67 - 0.2 0.4 Resection Margin - R0 - R1 - 2.02 - 0.003** - 2.95 - 0.003** T stage - 1 - 2 - 3 - 4 - 2.9 4 3.79 - 0.009** 0.003** 0.2 - 6.22 7.57 42.6 - 0.002** 0.002** 0.006** N stage - 0 - 1 - 2 - 1.24 1.67 - 0.5 0.078 - 1.47 2.15 - 0.3 0.052 M stage - 0 - 1 - 24.2 - 0.004** - 98.8 - < 0.001*** Grade - 1 - 2 - 3 - 2.08 3.56 - 0.3 0.081 - 0.59 1.63 - 0.7 0.7 Lymphovascular Invasion - Yes - No - 0.72 - 0.2 - 0.88 - 0.7 Perineural Invasion - Yes - No - 0.68 - 0.11 - 2.11 - 0.035* Adjuvant chemotherapy - Yes - No - 1.77 - 0.045* - 2.01 - 0.11 Abbreviations: CI: confidence interval. Statistical significance: p < 0.05 * , p < 0.01 ** , p < 0.001 *** . Finally, subgroup analysis of overall survival was conducted for levels of CD4 + and CD8 + T cells based on PAK1 or PAK4 expression. While high levels of CD4 + or CD8 + T cells failed to improve patient survival in the high PAK1 expression group (Fig. 6 b, c), high level of CD4 + T cells significantly improved overall survival in the low PAK1 subgroup (Fig. 6 b). There was also a trend suggesting high level CD8 + T cells was associated with better survival (p = 0.058) in low PAK1 patients (Fig. 6 c). On the other hand, PAK4 expression level did not affect the correlation between CD4 + T cells and patient survival (Fig. 6 d). However, despite not being statistically significant, the low PAK4 expression level resulted in a trend (p = 0.069) of improved overall survival by CD8 + T cells (Fig. 6 e). Discussion The individual roles of PAK1 and PAK4 in PDA tumorigenesis has been recognized in the literature[ 29 ]. However, development of selective PAK1 or PAK4 inhibitors for the treatment of solid tumours has not been successful[ 29 , 30 ]. The fact that both PAK1 and PAK4 have been implicated in intra-tumoral T cell response, suggested that PAK1 and PAK4 inhibition may function synergistically in suppressing the growth of PDA[ 10 , 16 , 31 ]. In this study, we demonstrated a synergistic effect of PAK1 and PAK4 inhibition in suppressing PDA growth in mice. PAK4KO stimulated the infiltration and activation of CD8 + T cells in tumour to a greater degree at an initial phase while PAK1KO caused an increased infiltration of active CD8 + T cells at a late phase. Together PAK1 and PAK4 double KO stimulated a sustained increase of infiltration of active CD8 + T cells, leading to a complete tumour regression. The results from a TMA of human PDA also confirmed the importance of PAK1 and PAK4 in intra-tumoral CD4 + and CD8 + T cell function, and the impact on overall survival of PDA patients. The fact that PAK4KO suppressed tumour growth in a syngeneic mouse model but not in SCID mice, indicates that PAK4KO regresses tumour growth through stimulating the anti-tumour immunity. An increased infiltration of cytotoxic CD8 + T cell was demonstrated in PAK4KO PDA tumour, which is consistent with the results from melanoma and prostate cancer[ 15 , 18 ]. However, an immune evasion developed in PAK4KO tumour (Fig. 3 f-j) due to a reduced infiltration and activation of CD8 + T cell at a later phase (four weeks, Fig. 4 d, e). While PAK1KO did not enhance CD8 + T cell infiltration significantly at the initial phase (one week, Fig. 2 e,f), it caused delayed increase in the infiltration of CD8 + T cell (Fig. 4 d,e), which is consistent with our previous findings in a syngeneic mice model[ 10 ]. This delayed effect of PAK1KO is likely to compensate for the reduced anti-tumour immune response by PAK4KO over time, leading to a sustained immune response to kill cancer cells. Therefore, the tumour was completed regressed by PAK1 and PAK4 double knockout (Fig. 1 e). While high PAK1 expression was previously shown to be negatively correlated with PDA patient survival, this was not found in our results from the human PDA TMA dataset[ 10 ]. After correcting for tumour stage, grade, and resection status, high PAK4 expression was noted to associate with worse outcomes which is the opposite to a previous report[ 32 ]. We also found a positive correlation between CD4 + or CD8 + T cell infiltration and PDA patient survival. Furthermore, low PAK1 or PAK4 levels enhanced the pro-survival effects of CD4 + and CD8 + T cells. This again highlights the important roles of PAK1 and PAK4 in T cell response in human PDA. Conclusion Our results identified a synergistic effect of PAK1 and PAK4 inhibition on PDA growth and T-cell immune response. It also indicated a rapid development of immune evasion with selective PAK4 inhibition in PDA, which may explain the failure of selective PAK4 inhibitors in clinical trials. However, pan-PAK inhibitors have a high level of toxicity as PAK2 inhibition can lead to endothelial cell dysfunction and vascular malformation[ 33 ]. Recent development of a selective PAK1 degrader by proteolysis-targeting chimera (PROTAC) technique has increased the hope of developing more selective PAK inhibitors[ 34 ]. Whether the combination of selective PAK1 and PAK4 degraders can offer greater clinical efficacy while minimizing side effects remains a question to be addressed and requires further research. Declarations Ethics approval and consent to participate The collection of patient’s clinical information and generation of human PDA tissue microarray (TMA) were approved by the Austin Health Human Research Ethics Committee (HREC/73948/Austin- 2021). Consent for publication NA Competing interests YM, CD, LD, CA, KA, YZ, MN and HH have no conflict of interest to declare. Funding This work was supported by grants from Pancare Foundation, Austin Medical Research Foundation (HH-2021, MN-2020) and MDHS (Medicine Dental Health Science, University of Melbourne) Seeding Ideas Grants (MN2020, HH and CA 2021). Author Contribution Y.M. and H.H. conceptualised and designed the study. Y.M., C.D., L.D., C.A., H.H. and K.A. were involved in sample preparation, conduction of experiments and data curation. Y.M., H.H. and Y.Z. analysed the data. Y.M. drafted the manuscript. H.H. and Y..Z modified the manuscript. All authors reviewed the manuscript. Acknowledgement YM is supported by Australian Government Research Training Program (RTP) Scholarship, Royal Australasian College of Surgeons (RACS) Foundation of Surgery Scholarship and Pancare Foundation Tim McGahan PhD Scholarship. HH is supported by the Henry Baldwin Cancer Research Trust Fund. Ms Shuji Jiang for design and creation of figures. Dr David Baloyan for advice and assistance on flow cytometry. Associate Professor Sarah Ellis for advice on immunohistochemistry and whole slide scanning. Ms Kat Hall for assistance in human ethics application. Graphical abstract was created by Biorender online platform ( https://www.biorender.com ). Availability of data and materials Data and study material will be available upon request from the corresponding author. References AIHW. Cancer in Australia 2019. Cancer series. Canberra: Australian Institute of Health and Welfare; 2019. Adamska A, Domenichini A, Falasca M. Pancreatic Ductal Adenocarcinoma: Current and Evolving Therapies. Int J Mol Sci. 2017;18(7). Von Hoff DD, Ervin T, Arena FP, Chiorean EG, Infante J, Moore M, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18):1691–703. Conroy T, Desseigne F, Ychou M, Bouche O, Guimbaud R, Becouarn Y, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817–25. Manser E, Leung T, Salihuddin H, Zhao ZS, Lim L. A brain serine/threonine protein kinase activated by Cdc42 and Rac1. Nature. 1994;367(6458):40–6. Menard RE, Mattingly RR. Cell surface receptors activate p21-activated kinase 1 via multiple Ras and PI3-kinase-dependent pathways. Cell Signal. 2003;15(12):1099–109. Maitra A, Hruban RH. Pancreatic cancer. Annu Rev Pathol. 2008;3:157–88. Yeo D, He H, Baldwin GS, Nikfarjam M. The role of p21-activated kinases in pancreatic cancer. Pancreas. 2015;44(3):363–9. Yeo D, Phillips P, Baldwin GS, He H, Nikfarjam M. Inhibition of group 1 p21-activated kinases suppresses pancreatic stellate cell activation and increases survival of mice with pancreatic cancer. Int J Cancer. 2017;140(9):2101–11. Wang K, Zhan Y, Huynh N, Dumesny C, Wang X, Asadi K, et al. Inhibition of PAK1 suppresses pancreatic cancer by stimulation of anti-tumour immunity through down-regulation of PD-L1. Cancer Lett. 2020;472:8–18. Jagadeeshan S, Subramanian A, Tentu S, Beesetti S, Singhal M, Raghavan S, et al. P21-activated kinase 1 (Pak1) signaling influences therapeutic outcome in pancreatic cancer. Ann Oncol. 2016;27(8):1546–56. Tyagi N, Bhardwaj A, Singh AP, McClellan S, Carter JE, Singh S. p-21 activated kinase 4 promotes proliferation and survival of pancreatic cancer cells through AKT- and ERK-dependent activation of NF-kappaB pathway. Oncotarget. 2014;5(18):8778–89. Mohammad RM, Li Y, Muqbil I, Aboukameel A, Senapedis W, Baloglu E, et al. Targeting Rho GTPase effector p21 activated kinase 4 (PAK4) suppresses p-Bad-microRNA drug resistance axis leading to inhibition of pancreatic ductal adenocarcinoma proliferation. Small GTPases. 2019;10(5):367–77. Tyagi N, Marimuthu S, Bhardwaj A, Deshmukh SK, Srivastava SK, Singh AP, et al. p-21 activated kinase 4 (PAK4) maintains stem cell-like phenotypes in pancreatic cancer cells through activation of STAT3 signaling. Cancer Lett. 2016;370(2):260–7. Abril-Rodriguez G, Torrejon DY, Liu W, Zaretsky JM, Nowicki TS, Tsoi J, et al. PAK4 inhibition improves PD-1 blockade immunotherapy. Nat Cancer. 2020;1(1):46–58. Abril-Rodriguez G, Torrejon DY, Karin D, Campbell KM, Medina E, Saco JD, et al. Remodeling of the tumor microenvironment through PAK4 inhibition sensitizes tumors to immune checkpoint blockade. Cancer Res Commun. 2022;2(10):1214–28. Ma W, Wang Y, Zhang R, Yang F, Zhang D, Huang M, et al. Targeting PAK4 to reprogram the vascular microenvironment and improve CAR-T immunotherapy for glioblastoma. Nat Cancer. 2021;2(1):83–97. Su S, You S, Wang Y, Tamukong P, Quist MJ, Grasso CS, et al. PAK4 inhibition improves PD1 blockade immunotherapy in prostate cancer by increasing immune infiltration. Cancer Lett. 2022;555:216034. Dong L, Vaux DL. Glucocorticoids can induce BIM to trigger apoptosis in the absence of BAX and BAK1. Cell Death Dis. 2020;11(6):442. Bankhead P, Loughrey MB, Fernandez JA, Dombrowski Y, McArt DG, Dunne PD, et al. QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017;7(1):16878. Takeuchi A, Saito T. CD4 CTL, a Cytotoxic Subset of CD4(+) T Cells, Their Differentiation and Function. Front Immunol. 2017;8:194. Simon S, Labarriere N. PD-1 expression on tumor-specific T cells: Friend or foe for immunotherapy? Oncoimmunology. 2017;7(1):e1364828. Simon S, Vignard V, Florenceau L, Dreno B, Khammari A, Lang F, et al. PD-1 expression conditions T cell avidity within an antigen-specific repertoire. Oncoimmunology. 2016;5(1):e1104448. Courtney AH, Lo WL, Weiss A. TCR Signaling: Mechanisms of Initiation and Propagation. Trends Biochem Sci. 2018;43(2):108–23. Tybulewicz VL. Vav-family proteins in T-cell signalling. Curr Opin Immunol. 2005;17(3):267–74. Coudronniere N, Villalba M, Englund N, Altman A. NF-kappa B activation induced by T cell receptor/CD28 costimulation is mediated by protein kinase C-theta. Proc Natl Acad Sci U S A. 2000;97(7):3394–9. Sgarbanti M, Marsili G, Remoli AL, Stellacci E, Mai A, Rotili D, et al. IkappaB kinase epsilon targets interferon regulatory factor 1 in activated T lymphocytes. Mol Cell Biol. 2014;34(6):1054–65. Zhang J, Feng H, Zhao J, Feldman ER, Chen SY, Yuan W, et al. IkappaB Kinase epsilon Is an NFATc1 Kinase that Inhibits T Cell Immune Response. Cell Rep. 2016;16(2):405–18. Ma Y, Nikfarjam M, He H. The trilogy of P21 activated kinase, autophagy and immune evasion in pancreatic ductal adenocarcinoma. Cancer Lett. 2022;548:215868. He H, Dumesny C, Ang CS, Dong L, Ma Y, Zeng J, et al. A novel PAK4 inhibitor suppresses pancreatic cancer growth and enhances the inhibitory effect of gemcitabine. Transl Oncol. 2021;16:101329. Su S, You S, Wang Y, Tamukong P, Quist MJ, Grasso CS, et al. PAK4 inhibition improves PD1 blockade immunotherapy in prostate cancer by increasing immune infiltration. Cancer Lett. 2023;555:216034. Park S, Kim JW, Kim H, Kim JW, Kim YJ, Lee KW, et al. Prognostic value of p21-activated kinase 4 in resected pancreatic cancer. APMIS. 2017;125(8):699–707. Radu M, Lyle K, Hoeflich KP, Villamar-Cruz O, Koeppen H, Chernoff J. p21-Activated Kinase 2 Regulates Endothelial Development and Function through the Bmk1/Erk5 Pathway. Mol Cell Biol. 2015;35(23):3990–4005. Chow HY, Karchugina S, Groendyke BJ, Toenjes S, Hatcher J, Donovan KA, et al. Development and Utility of a PAK1-Selective Degrader. J Med Chem. 2022;65(23):15627–41. Additional Declarations No competing interests reported. Supplementary Files SupplementaryDocuments.docx SupplementaryFigures.pdf Supplementaryuncroppedblots.pptx Cite Share Download PDF Status: Under Review Version 1 posted Editorial decision: Revision requested 21 Apr, 2024 Reviews received at journal 16 Mar, 2024 Reviewers agreed at journal 09 Mar, 2024 Reviewers invited by journal 09 Mar, 2024 Editor assigned by journal 06 Mar, 2024 Submission checks completed at journal 06 Mar, 2024 First submitted to journal 20 Feb, 2024 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-3974396","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":277773518,"identity":"5939dc11-b039-4cda-9576-30cf10920712","order_by":0,"name":"Yi Ma","email":"","orcid":"","institution":"University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Yi","middleName":"","lastName":"Ma","suffix":""},{"id":277773519,"identity":"7c530ebb-1c4c-4b69-9a09-d59dab51e9e2","order_by":1,"name":"Chelsea Dumesny","email":"","orcid":"","institution":"University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Chelsea","middleName":"","lastName":"Dumesny","suffix":""},{"id":277773520,"identity":"d7d3285e-ef87-46fc-bc1f-4e63cef484c3","order_by":2,"name":"Li Dong","email":"","orcid":"","institution":"University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Li","middleName":"","lastName":"Dong","suffix":""},{"id":277773521,"identity":"ad1081f7-51ba-485c-9d95-f05d2bc7bb65","order_by":3,"name":"Ching-Seng Ang","email":"","orcid":"","institution":"University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Ching-Seng","middleName":"","lastName":"Ang","suffix":""},{"id":277773522,"identity":"2090cf51-3e6a-45b7-bc8d-f9f98af860c4","order_by":4,"name":"Khashayar Asadi","email":"","orcid":"","institution":"Austin Health","correspondingAuthor":false,"prefix":"","firstName":"Khashayar","middleName":"","lastName":"Asadi","suffix":""},{"id":277773523,"identity":"d65ccbd0-63b7-49e9-957d-bd0b671d37e8","order_by":5,"name":"Yifan Zhan","email":"","orcid":"","institution":"Shanghai Huaota Biopharm","correspondingAuthor":false,"prefix":"","firstName":"Yifan","middleName":"","lastName":"Zhan","suffix":""},{"id":277773524,"identity":"ed93b0d3-6f45-423c-aa38-8a4b6add507f","order_by":6,"name":"Mehrdad Nikfarjam","email":"","orcid":"","institution":"University of Melbourne","correspondingAuthor":false,"prefix":"","firstName":"Mehrdad","middleName":"","lastName":"Nikfarjam","suffix":""},{"id":277773525,"identity":"3033be3c-63ab-4cd8-9655-5019423dbc8f","order_by":7,"name":"Hong He","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8ElEQVRIiWNgGAWjYBACAxDB2MAgByYZCg4AuQnEaTGGaDEgQUtiA4RLhBZz9t7DLxh32KRvuN3cuuGDwR0GfvYcA4afbbi1WPacS7NgPJOWu+HOwbabMwyeMUj2vDFg7MWjxeBGjpkBY9vh3A03Ettu8xgcBokYMPAS1vI/3QCk5Q9Qiz1QC+Nf/FqMHzC2HUgAa2EA2SKRY8CM15YzZ8wYEtuSDWcCtdzsMXjGI3HmWcFhmXN4tBzvMf7wsc1Onu9G+rMbPyruyPG3J298+KYMtxYgYJNIQOLxgIgDeDUwMDB/IKBgFIyCUTAKRjoAAKxHWz88KYxVAAAAAElFTkSuQmCC","orcid":"","institution":"University of Melbourne","correspondingAuthor":true,"prefix":"","firstName":"Hong","middleName":"","lastName":"He","suffix":""}],"badges":[],"createdAt":"2024-02-21 03:35:50","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-3974396/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-3974396/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":52453248,"identity":"871b8dd9-2e63-4f60-a24c-e655a4e9ce39","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":184322,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKnockout of PAK1 and PAK4 synergistically inhibited pancreatic cancer tumour growth. \u0026nbsp;a) \u003c/strong\u003eC57BL6 mice were injected at right flank with KPC WT (n=6), PAK1 KO (n=8), PAK4 KO (n=8) and PAK1 and 4 KO (n=8) cell lines. Mice injected with WT and PAK1 KO cells were culled at day 20, while mice bearing PAK4 KO tumours were culled at day 26. Tumour\u003cstrong\u003e \u003c/strong\u003ephotos \u003cstrong\u003e(b)\u003c/strong\u003e, size \u003cstrong\u003e(c)\u003c/strong\u003eand tumour weight \u003cstrong\u003e(d)\u003c/strong\u003e of KPC WT, PAK1 KO and PAK4 KO tumours were presented (PAK1 and 4 KO tumour regressed completely after two weeks).\u003cstrong\u003e E)\u003c/strong\u003eIndividual growth curve of KPC WT, PAK1 KO, PAK4 KO and PAK1 and 4 KO tumours in C57BL6 mice. \u003cstrong\u003eF)\u003c/strong\u003e SCID mice were injected at left sided abdomen with KPC WT (n=6), PAK4 KO (n=6), PAK1 and 4 KO (n=6) cell lines. Mice were culled at day 18. \u003cstrong\u003eG)\u003c/strong\u003e Individual growth curve of KPC WT, PAK4 KO and PAK1 and 4 KO tumours in SCID mice. \u003cstrong\u003eH-j) \u003c/strong\u003eTumour\u003cstrong\u003e \u003c/strong\u003ephotos, growth curve and tumour weight of KPC WT, PAK1 KO and PAK1 and 4 KO tumours in SCID mice. Statistical significance: *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001. All comparisons were made against WT unless otherwise indicated.\u003c/p\u003e","description":"","filename":"Fig.1.png","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/f8012ce7eea748ecb40c6bab.png"},{"id":52453245,"identity":"af9b4ab2-d626-4d45-823e-a7c1f34a9965","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":268582,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eKnockout of PAK1 and PAK4 KO synergistically stimulated CD8+ T cell infiltration in pancreatic cancer.\u003c/strong\u003e C57BL6 mice (4 in each group) were subcutaneously injected with KPC WT, PAK1 KO, PAK4 KO and PAK1 and 4 KO cell lines \u003cstrong\u003e(a)\u003c/strong\u003e at lower back and abdomen \u003cstrong\u003e(c)\u003c/strong\u003e. Mice were culled at day 7. Tumour size \u003cstrong\u003e(b)\u003c/strong\u003e and weight \u003cstrong\u003e(d) \u003c/strong\u003ewere combined for lower back and abdomen tumours from each mouse. Tumour tissues collected on mice culling were subjected to FACS analysis for tumour infiltrating lymphocytes. \u003cstrong\u003ee)\u003c/strong\u003e The infiltration of B cell (B220+), CD4+ and CD8+ T cells in WT, PAK1 KO, PAK4 KO and PAK1 and 4 KO tumours were demonstrated by UMAP. \u003cstrong\u003ef) \u003c/strong\u003eThe percentage of cell infiltrations were compared. Statistical significance: *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001. All comparisons were made against WT unless otherwise indicated.\u003c/p\u003e","description":"","filename":"Fig.2.png","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/b0176472255c67ce5fc10412.png"},{"id":52453461,"identity":"9b30dccd-5227-403e-b0d3-3b53eda71c1f","added_by":"auto","created_at":"2024-03-11 19:28:02","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":240781,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImmune evasion developed in PAK4 KO tumour.\u003c/strong\u003e A global proteomic analysis of the tumour tissues of WT and PAK4KO showed that more differentially expressed proteins between PAK4 KO and WT tumours at 1 week than 4 weeks \u003cstrong\u003e(a)\u003c/strong\u003e. \u003cstrong\u003eb)\u003c/strong\u003e Heatmap demonstrated individual protein expression identified in T cell receptor signaling KEGG pathway.\u003cstrong\u003e \u003c/strong\u003eThe expression of CD3g \u003cstrong\u003e(c)\u003c/strong\u003e, Zap 70 \u003cstrong\u003e(d)\u003c/strong\u003e were significantly increased in PAK4KO tumour at 1 week but not at 4 weeks. There was a trend of increment of LCK \u003cstrong\u003e(e)\u003c/strong\u003e in PAK4 KO tumour at 1 week rather than at 4 weeks.\u003cstrong\u003e f)\u003c/strong\u003e C57BL6 mice were injected at right flank with KPC WT (n=4), PAK4KO (n=16) cell lines. PAK4 KO cell line was injected 14 days before WT cell line injection (D0) and only five mice developed PAK4KO tumour. Mice were culled at day 21. \u003cstrong\u003eg-i) \u003c/strong\u003ePAK4 KO tumours that escaped immune surveillance showed similar growth as WT tumours demonstrated by tumour photos \u003cstrong\u003e(g)\u003c/strong\u003e, growth curve \u003cstrong\u003e(h)\u003c/strong\u003e and tumour weight \u003cstrong\u003e(i)\u003c/strong\u003e. \u003cstrong\u003ej)\u003c/strong\u003e Individual growth curve of KPC WT and PAK4 KO tumours. Statistical significance: *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001. All comparisons were made against WT unless otherwise indicated.\u003c/p\u003e","description":"","filename":"Fig.3.png","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/9da20c8044e87dcf8011e77f.png"},{"id":52453246,"identity":"3e4eb8c6-ed88-4229-800e-a7341ab958fb","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":254592,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003ePAK1 KO induced delayed CD8+ T cell infiltration.\u003c/strong\u003e \u003cstrong\u003ea-b)\u003c/strong\u003e PAK1 expression increased from 1 week to 4 weeks in both WT and PAK4 KO tumour from global proteomic analysis, but to a higher degree in PAK4 KO tumours. \u003cstrong\u003ec)\u003c/strong\u003e Protein-protein interaction network analysis of differentially expressed proteins of PAK4 KO tumours between 4 weeks and 1 week showed a role of PAK1 in T cell receptor signaling. Tumour specimens collected from the experiment described in Fig. 1a were subjected to FACS. The infiltration of B cell (B220+) and CD4+ T cell were significantly increased in PAK1 KO tumour rather than PAK4 KO tumour \u003cstrong\u003e(d, e)\u003c/strong\u003e. Although the infiltration of CD8+ T cell was still significantly increased in PAK4KO tumour \u003cstrong\u003e(d, e)\u003c/strong\u003e, its level has reduced in comparison to one week results (Fig 2e,f). Statistical significance: *p\u0026lt;0.05, **p\u0026lt;0.01, ***p\u0026lt;0.001. All comparisons were made against WT unless otherwise indicated.\u003c/p\u003e","description":"","filename":"Fig.4.png","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/407da0b583fcddda3a45ecba.png"},{"id":52453251,"identity":"d3d1d3c1-bf4e-4d97-b274-cc9c16b83f2e","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":136432,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eCorrelation of PAKs and T cells to the survival of pancreatic cancer patients. \u003c/strong\u003eThe data were obtained from a tissue microarray (TMA) study from human pancreatic cancer patients (n=100). \u003cstrong\u003ea)\u003c/strong\u003e Flow diagram demonstrating identification, exclusion, and inclusion of study participants. \u003cstrong\u003eb)\u003c/strong\u003e Table of median PAK1, PAK4, CD4 and CD8 levels as well as inter-quantile range (IQR). \u003cstrong\u003ec) \u003c/strong\u003eKaplan-meier curve demonstrating overall survival of all study participants, as well as number of participants at risk at each time point. \u003cstrong\u003ed-g)\u003c/strong\u003e Kaplan-meier curves demonstrating correlation between PAK1, PAK4, CD4 and CD8 levels and overall survival respectively. Hazard ratio (HR), confidence interval (CI) and p-values were reported for all survival analysis, with two-sided p-value below 0.05 considered statistically significant.\u003c/p\u003e","description":"","filename":"Fig.5.png","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/61f53f88075243cb9b3eaf32.png"},{"id":52453249,"identity":"7bdbc689-5ad6-4ce0-bfbb-03b51bf75c58","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":290074,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eImpact of PAK levels on the effects of T cells on patients’ survival.\u003c/strong\u003e The data were obtained from a tissue microarray (TMA) study from human pancreatic cancer patients (n=100). \u003cstrong\u003ea)\u003c/strong\u003eImmunohistochemistry (IHC) staining of low versus high PAK1, PAK4, CD4 and CD8 TMA cores. Kaplan-meier curve demonstrated the correlation between CD4 \u003cstrong\u003e(b)\u003c/strong\u003e or CD8 \u003cstrong\u003e(c) \u003c/strong\u003elevels to the overall survival in low versus high PAK1 expression subgroups. Similarly, Kaplan-meier curve demonstrated the correlation between CD4 \u003cstrong\u003e(d)\u003c/strong\u003e or CD8 \u003cstrong\u003e(e)\u003c/strong\u003e levels to the overall survival in low versus high PAK4 expression subgroups. Hazard ratio (HR), confidence interval (CI) and p-values were reported for all survival analysis, with two-sided p-value below 0.05 considered statistically significant.\u003c/p\u003e","description":"","filename":"Fig.6.png","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/c107a442a3ba8780e6f267d2.png"},{"id":52454122,"identity":"756767a7-fb34-4a5f-8274-18ef39674ac1","added_by":"auto","created_at":"2024-03-11 19:36:03","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1919811,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/8068455b-4b60-4fa5-806b-ad198f28c4f6.pdf"},{"id":52453252,"identity":"e8e8f2ae-9a9b-4b15-ac56-b2abd6e57a11","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"docx","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":68941,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryDocuments.docx","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/4776767b90dd9c9dbf03b4dd.docx"},{"id":52453250,"identity":"889d6e05-c999-484f-b772-2b7e1f2d6983","added_by":"auto","created_at":"2024-03-11 19:20:02","extension":"pdf","order_by":9,"title":"","display":"","copyAsset":false,"role":"supplement","size":1477449,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigures.pdf","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/057fd753d19098cd5707a7d9.pdf"},{"id":52453254,"identity":"ed368937-2d3b-48a9-9d10-a11ad8113e73","added_by":"auto","created_at":"2024-03-11 19:20:05","extension":"pptx","order_by":10,"title":"","display":"","copyAsset":false,"role":"supplement","size":26441418,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementaryuncroppedblots.pptx","url":"https://assets-eu.researchsquare.com/files/rs-3974396/v1/c7eeb2283223d5f04ed064bd.pptx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Inhibition of P21-activated kinases 1 and 4 synergistically suppress the growth of pancreatic cancer by stimulating anti-tumour immunity","fulltext":[{"header":"Background","content":"\u003cp\u003ePancreatic ductal adenocarcinoma (PDA) is one of the most aggressive solid cancers, with a dismal five-year survival rate of 9.8%[\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e]. This is mainly due to late diagnosis and lack of effective treatment for advanced disease. While surgery can be curative, only 10\u0026ndash;20% of patients have resectable tumours on diagnosis[\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e]. Despite advances in systemic therapy of solid malignancies, the development of targeted therapy in PDA remains stagnant. Gemcitabine-based chemotherapy and FOLFIRINOX are still the mainstay treatment of advanced PDA, but only extend overall survival on average, by two to three months[\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e, \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eP21-activated kinase (PAK) has emerged as a potential therapeutic target. PAKs are a group of serine/threonine kinases that act downstream of KRAS, while KRAS mutations occur in over 90% of PDA [\u003cspan additionalcitationids=\"CR6\" citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e]. There are six members of the PAK family, which are categorized into two groups: group 1 (PAK1 to 3) and group 2 (PAK4 to 6)[\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e]. Among the PAK proteins, PAK1 and PAK4 are mostly investigated for their role in tumorigenesis. In PDA, PAK1 was shown to inhibit cancer cell apoptosis, activate pancreatic stellate cells and down-regulate intra-tumoral CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cell infiltration[\u003cspan additionalcitationids=\"CR10\" citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. PAK4 was reported to play a role in PDA cell apoptosis, cell cycle arrest and cancer cell stemness[\u003cspan additionalcitationids=\"CR13\" citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e]. Recently, PAK4 was shown to suppress T cell response in melanoma, prostate cancer, and glioblastoma, which is consistent with an observed upregulation of intra-tumoral CD8\u0026thinsp;+\u0026thinsp;T cells in PDA mouse model by PAK4 inhibitor PF-3758309[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eDespite the pro-tumorigenic role of PAK1 and PAK4 in PDA, the development of clinically effective PAK1 or PAK4 inhibitors has not been successful. Given that both PAK1 and PAK4 play important roles in PDA biology, we hypothesize that the two may promote PDA tumour growth synergistically. Using PAK1 and PAK4 double knockout (KO) cell lines, we investigated their combinational effect in PDA which will guide future development of PAK-targeted therapy.\u003c/p\u003e"},{"header":"Methods and Methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eCell lines and cell culture\u003c/h2\u003e \u003cp\u003eMurine WT and PAK1KO pancreatic cancer cell lines were isolated from PAK1\u003csup\u003e+/+\u003c/sup\u003e and PAK1\u003csup\u003e\u0026minus;/\u0026minus;\u003c/sup\u003e KPC mice as previously described[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. KPC PAK4KO and KPC PAK1\u0026amp;4 KO cell lines were generated from KPC WT and KPC PAK1KO cell lines respectively using CRISPR-CAS technique. Cancer cells were cultured in Dulbecco\u0026rsquo;s Modified Eagle\u0026rsquo;s Medium (DMEM) supplemented with 5% fetal bovine serum (FBS) (Hyclone Laboratories, Melbourne, Australia) in a 37\u0026deg;C incubator under humidified 5% CO\u003csub\u003e2\u003c/sub\u003e.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003eAnimal studies\u003c/h2\u003e \u003cp\u003e All mouse studies were approved by the Austin Health Animal Ethics Committee (A2022/5797). Experimental C57BL6 and SCID mice were housed in the Austin Health Bioresource Facility and monitored for health criteria. For C57BL6 syngeneic mouse model, KPC WT, PAK1KO, PAK4KO and PAK1\u0026amp; 4 KO cells (0.5-1x10\u003csup\u003e6\u003c/sup\u003e cells/100ul/mouse) were injected subcutaneously to the flank of 7 weeks old male C57BL6 mice. The mice were observed for one week or 4\u0026ndash;6 weeks. Tumour growths were monitored using a calliper. Tumour volume (mm\u003csup\u003e3\u003c/sup\u003e) was calculated using the simplified ellipsoid formula:\u003cdiv id=\"Equa\" class=\"Equation\"\u003e\u003cdiv format=\"TEX\" class=\"mathdisplay\" id=\"FileID_Equa\" name=\"EquationSource\"\u003e\n$$Volume \\left(V\\right)=Length \\left(L\\right)\\times Width \\left(W\\right)^2\\times 0.5$$\u003c/div\u003e\u003c/div\u003e\u003c/p\u003e \u003cp\u003eTumour weight (g) was measured upon mice culling.\u003c/p\u003e \u003cp\u003eFor the SCID mouse model, KPC WT, PAK4KO and PAK1\u0026amp;4 KO cancer cells (5x10\u003csup\u003e5\u003c/sup\u003e cells/mouse) were injected subcutaneously into the flank of 6-week-old male SCID mice. Mice were monitored for the time indicated in the Result section. Tumour growth and weight were obtained as described above.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003ePatient information collection and tissue microarray generation\u003c/h2\u003e \u003cp\u003e The collection of patient\u0026rsquo;s clinical information and generation of human PDA tissue microarray (TMA) were approved by the Austin Health Human Research Ethics Committee (HREC/73948/Austin- 2021).\u003c/p\u003e \u003cp\u003ePatients who had surgical resection of PDA between 2008 and 2019 under Austin Health were identified. Patients who had premature mortality from either delayed surgical complications, background comorbidities, incorrect disease staging or missing survival data due to transfer of care were excluded. Once recruited, participants\u0026rsquo; baseline demographics, disease staging and grading, treatment and survival information were retrospectively collected. The formalin-fixed PDA tumour blocks were assessed by a qualified anatomical pathologist, and three 1mm diameter cores were taken from each tumour sample to assemble TMA blocks.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eCRISPR-CAS gene knock-out\u003c/h2\u003e \u003cp\u003eTo generate PAK4KO PDA cell lines, an inducible lentiviral CRISPR/Cas9 system was used as described previously[\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Single-guide RNA (sgRNA) oligos targeting mouse PAK4 (guide 1: CCCGCGATAAGCGCCCACT; guide 2: CGAACGATGGTCTGGGGTC) were cloned into the BsmBI site of the pFgH1tUTG GFP lentiviral vector. KPC WT and KPC PAK1KO cells were infected with lentiviral constructs encoding Cas9 and mCherry, and a doxycycline-inducible sgRNA targeting PAK4 and GFP. mCherry and GFP double-positive cells were sorted using BD FACS Aria III (BD Biosciences, New Jersey, USA). Cell clones with PAK4KO were confirmed by immunoblot.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003eFlow cytometry analysis\u003c/h2\u003e \u003cp\u003eMice tumour specimens were minced and digested in a digestion buffer containing 1.25mg/ml collagenase IV (Worthington Biomedical, Lakewood, USA). Digested tumours were filtered and resuspended in FACS buffer to obtain single-cell suspension (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e). 1X10\u003csup\u003e6\u003c/sup\u003e cells per sample were blocked with 1\u0026micro;l of mouse FcR blocking reagent (Miltenyi, Biotec, Bergisch Gladbach, Germany). Cell viability was assessed using 1:500 dilution of Zombie UV\u0026trade; Fixable Viability Dye (BioLegend, San Diego, USA). Fluorophore-labelled antibodies against CD45, B220, CD3, CD4, CD8, and PD1 (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e) were added and incubated on ice for 20 minutes.\u003c/p\u003e \u003cp\u003e For FoxP3 staining, cells were permeabilized and fixed with eBioscience\u0026trade; FoxP3/Transcription Factor Staining Buffer Set (Invitrogen, Waltham, USA) according to the manufacturer\u0026rsquo;s instructions. Cells were incubated in FoxP3 fixation/permeabilization solution for 60 minutes in the dark at room temperature (RT). Fluorophore-labelled FoxP3 antibody was added to permeabilized cells and incubated for 30 minutes in the dark at RT (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eFor staining of cytoplasmic markers, cells were permeabilized and fixed with eBioscience\u0026trade; Intracellular Fixation and Permeabilization Buffer Set (Invitrogen, Waltham, USA) according to the manufacturer\u0026rsquo;s instruction. Cells were incubated in a fixation buffer for 60 minutes in the dark at RT. Fluorophore-labelled Granzyme B and Perforin antibodies were added to permeabilized cells and incubated for 30 minutes in the dark at RT (Table \u003cspan refid=\"MOESM3\" class=\"InternalRef\"\u003eS3\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eCells were resuspended in FACS buffer and analyzed by Cytek\u0026reg; Aurora flow cytometer (Cytek Biosciences, California, USA). FCS Express version 7.12.0007 (De Novo Software, Pasadena, USA) was used for manual gating and statistical analysis.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec8\" class=\"Section2\"\u003e \u003ch2\u003eGlobal and phospho-proteomic studies\u003c/h2\u003e \u003cp\u003eThe proteomic studies were conducted consequently by these steps as \u003cem\u003eSample Preparation, Liquid Chromatograph Data Independent Acquisition Mass Spectrometry\u003c/em\u003e, \u003cem\u003eData Search\u003c/em\u003e and \u003cem\u003eBioinformatic Analysis\u003c/em\u003e, which were detailed in the Supplementary Methods.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec9\" class=\"Section2\"\u003e \u003ch2\u003eAnalysis of TMA samples\u003c/h2\u003e \u003cp\u003eThe TMA blocks of 5\u0026micro;M sections were stained using immunohistochemistry. Samples were boiled in Tris-EDTA buffer (Table \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003e) at 99˚C for 30 minutes and then blocked with Dako REAL\u0026trade; peroxidase blocking solution (Agilent Technologies, Glostrup, Denmark) at RT for 15 minutes followed by 5% goat serum at RT for an hour for endogenous peroxidase quenching and protein blocking respectively. Primary antibodies against PAK1, PAK4, CD4 and CD8 (Table \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003e) were added to samples and incubated at RT for an hour before one hour incubation with Dako EnVision\u0026thinsp;+\u0026thinsp;System HRP Labelled Anti-Rabbit Polymer (Agilent Technologies, Glostrup, Denmark) at RT. DAB staining was achieved with EnVision FLEX DAB\u0026thinsp;+\u0026thinsp;substrate Chromogen System (Agilent Technologies, Glostrup, Denmark) and samples were counterstained with haematoxylin (Sigma-Aldrich, St Louis, USA). Whole slide images were captured with Aperio AT2 digital pathology slide scanner (Leica Biosystems, Wetzlar, Germany), and analysed by QuPath version 0.4.3[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e]. .\u003c/p\u003e \u003cp\u003ePAK1 and PAK4 expression were determined using mean DAB intensity of the TMA core, while CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cell levels were assessed by the percentage of positive stained cells against the total number of cells in the TMA core. Mean of all three replicates was calculated for each patient for PAK1, PAK4, CD4 and CD8 levels. Univariate and multivariate linear regression was used to assess correlation between individual variables and percentage of CD4\u0026thinsp;+\u0026thinsp;or CD8\u0026thinsp;+\u0026thinsp;cells. Correlation between individual variables and overall survival of patients was visualised with Kaplan Meier curve and tested by univariate and multivariate Cox proportional hazards regression. Interaction term was included for multivariate regression models to assess interaction between PAK1 and PAK4. Regression analysis was conducted using R version 4.3.0 (R Foundation for Statistical Computing, Vienna, Austria).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec10\" class=\"Section2\"\u003e \u003ch2\u003eStatistical analysis\u003c/h2\u003e \u003cp\u003eAll \u003cem\u003ein vitro\u003c/em\u003e experiments were repeated in three replicates. \u003cem\u003eIn vivo\u003c/em\u003e experiments included at least three mice per group. For continuous variables, mean\u0026thinsp;\u0026plusmn;\u0026thinsp;standard deviation (SD) was reported for parametric data while median\u0026thinsp;\u0026plusmn;\u0026thinsp;inter-quantile range (IQR) were reported for non-parametric data. Hypothesis testing was conducted by either two-sided t test or one way ANOVA for parametric data, Mann-whitney\u0026rsquo;s U test for non-parametric data and Chi-square test for categorical results. Linear regression model was fitted for cell and tumour growth curve, and correlation coefficients (i.e. slopes) of linear fit was compared between groups. Statistical analysis was conducted with GraphPad Prism version 10.0.2 (GraphPad Software, Boston, USA), Stata BE version 17.0 (Texas, USA) and R version 4.3.0. p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05 was considered statistically significant.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003e \u003cb\u003eInhibition of PAK1 and PAK4 synergistically suppressed PDA growth.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eKPC WT, PAK1KO, PAK4KO and PAK1\u0026amp;4 KO cell lines were validated by immunoblot (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ea). PAK1KO reduced KPC cell growth by 72 hours, but PAK4KO increased cell growth by 72 hours (Fig. \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003eb). Global proteomic analysis of PAK WT and KO cells showed significantly differential protein expression profiles of PAK KO cell lines from WT cell line (Fig \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003ea-c). Functional enrichment for KEGG and reactome pathways indicated changes in programmed cell death (e.g., apoptosis) and cell cycle regulation (Fig. \u003cspan refid=\"MOESM2\" class=\"InternalRef\"\u003eS2\u003c/span\u003ed, e), which was validated by FACS analysis. PAK1KO and PAK1\u0026amp;4 KO increased the apoptosis and cell death while PAK4KO reduced cancer cell death and had no effect on the apoptosis (Fig \u003cspan refid=\"MOESM1\" class=\"InternalRef\"\u003eS1\u003c/span\u003ec, d).\u003c/p\u003e \u003cp\u003eIn a syngeneic mouse model (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea), PAK4KO suppressed the \u003cem\u003ein vivo\u003c/em\u003e tumour growth (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ea-d) significantly. Only 30% of the mice injected with PAK4KO cells developed tumours, which also grew significantly slower than the KPC WT- and PAK1KO-inoculated tumours, even with an extra two-weeks\u0026rsquo; growth (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ec-e). No tumour developed in mice injected with PAK1\u0026amp;4 KO cells. The growth curve (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee) showed a peak around one week after cell injection in PAK4KO and PAK1\u0026amp;4 KO injected mice followed by tumour regression. This suggested that PAK4KO inhibits PDA tumour growth through modulating the adaptive immune response, which is confirmed in a SCID mouse model (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ef-j). In SCID mice, PAK4KO did not inhibit the tumour growth at all while PAK double KO decreased the tumour growth significantly compared to PAK4KO but not to WT (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eg-j). These results indicated that PAK4KO inhibited PDA by stimulation of anti-tumour immunity and that inhibition of PAK1 and 4 synergistically suppressed PDA progression.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cb\u003eInhibition of PAK1 and PAK4 synergistically increased intra-tumoral CD8\u0026thinsp;+\u0026thinsp;T cell infiltration.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eGiven the findings that PAK4 KO tumour regressed one week after cell injection, we further examined PAK WT vs KO tumour growth within the first week (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ea, c). Tumour growth was inhibited by PAK1KO, PAK4KO and PAK1\u0026amp;4 KO (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eb, d). By one week, PAK double KO further suppressed the tumour growth compared to single PAK KO, indicating a synergistic effect of PAK1 and PAK4. One week tumour specimens were then digested into single cell suspension for FACS analysis of intra-tumoral lymphocyte infiltration (gating strategy shown in Fig. S4). By one week, PAK1KO increased B cell and CD4\u0026thinsp;+\u0026thinsp;T cell infiltration while PAK4KO increased the infiltration of T cells, particularly CD8\u0026thinsp;+\u0026thinsp;T cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee, f). PAK1\u0026amp;4 KO further increased total T cell and CD8\u0026thinsp;+\u0026thinsp;T cell infiltration, compared to single PAK4KO, suggesting a synergistic effect of PAK1KO and PAK4KO on intra-tumoral CD8\u0026thinsp;+\u0026thinsp;T cell infiltration.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePAK4KO also increased the infiltration of active CD8\u0026thinsp;+\u0026thinsp;T cells at one week, shown by increased levels of granzyme B+, and granzyme B and perforin double positive cytotoxic CD8\u0026thinsp;+\u0026thinsp;T cells (Fig.S5a, b). However, Granzyme B\u0026thinsp;+\u0026thinsp;CD8\u0026thinsp;+\u0026thinsp;T cells were reduced by PAK1KO, contributing to no change in granzyme B\u0026thinsp;+\u0026thinsp;CD8\u0026thinsp;+\u0026thinsp;T cells and less increase in granzyme B and perforin double positive cytotoxic CD8\u0026thinsp;+\u0026thinsp;T cells in double KO tumour (Fig. S5a, b). The level of granzyme B\u0026thinsp;+\u0026thinsp;CD4\u0026thinsp;+\u0026thinsp;T cells were increased in both single PAK KO tumour, with further significant increase in PAK1\u0026amp;4 double KO (Fig. S5c, d). The granzyme B and perforin double positive CD4\u0026thinsp;+\u0026thinsp;T cells was increased only in PAK1\u0026amp;4 double KO tumour. These data suggested that PAK1 and PAK4 inhibition have synergistic effect on cytotoxic CD8\u0026thinsp;+\u0026thinsp;and CD4\u0026thinsp;+\u0026thinsp;T cells infiltration[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e]. PAK4KO also decreased the infiltration of regulatory CD4\u0026thinsp;+\u0026thinsp;T cells (Treg) (Fig. S5e, f). PAK1KO decreased PD-1\u0026thinsp;+\u0026thinsp;CD8\u0026thinsp;+\u0026thinsp;T cells while PAK4KO increased PD1\u0026thinsp;+\u0026thinsp;CD8\u0026thinsp;+\u0026thinsp;T cells, and PD1\u0026thinsp;+\u0026thinsp;T cells were known to have higher anti-tumour activity (Fig. S5g, h)[\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e].\u003c/p\u003e \u003cp\u003e \u003cb\u003ePancreatic cancer developed immune evasion after PAK4 inhibition.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eAfter initial regression, approximately 30% of mice developed a PAK4KO tumour (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee), indicating immune evasion. We conducted a proteomic study to compare PAK4KO versus WT tumours at one week and four weeks after cancer cell injection to investigate the differences in immune cell infiltration. Protein expressions of PAK4KO and WT tumours were much more different at one week than four weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ea). Four protein clusters were classified based on expression patterns and were visualized as heatmap and UMAP (Fig. S6a, b). All four clusters of proteins had different protein expression levels between PAK4 KO and WT tumours at one week, and these differences were reduced in clusters 1, 2 and 4 at four weeks (Fig. S6a). KEGG and reactome pathways were functionally enriched based on the significant proteins and identified immune-related pathways (Fig. S6c, d).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eGiven that the data from FACS analysis showed a significant role of PAK4KO in CD8\u0026thinsp;+\u0026thinsp;T cell mediated immune response (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e), T cell receptor signaling pathway was examined in greater detail by heatmap. Multiple protein targets that are known to play an important role in downstream signaling of T cell receptor (TCR) were identified (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eb), such as the VAV family proteins (VAV1, VAV3) and proteins that are directly involved in MHC- TCR signaling complex (CD3g, ZAP70, LCK)[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. The expression profiles of CD3g, ZAP70 and LCK were compared between WT and PAK4KO at one week and four weeks. CD3g and ZAP70 levels were significantly higher in PAK4KO tumour at one week but dropped to similar levels as WT tumour by four weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ec, d). There was no difference in LCK levels between WT and PAK4 KO tumours at both time points (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ee).\u003c/p\u003e \u003cp\u003eThese findings from global proteomic analysis were further supported by phosphor-proteomic results. Differentially expressed phospho-sites between WT and PAK4 KO tumours were demonstrated at one and four weeks (Fig. S6e). From the significant phospho-sites identified across all four groups, 38 kinases and their activity were predicted (Fig. S6f). Among these predicted kinases, IKKε (IKBKE) and PKCθ (PRKCQ) were predicted to have significantly higher activity in PAK4 KO tumour at one week in comparison to WT tumour (Fig. S6g) indicating a higher activity of T cell receptor pathway in PAK4KO as both IKKε and PKCθ are activated upon activation of T cell receptor[\u003cspan additionalcitationids=\"CR27\" citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e]. However, by four weeks, the kinase activities of IKKε and PKCθ were no longer different between PAK4 KO and WT tumours (Fig. S6h) suggesting an immune evasion of PAK4KO tumour after the initial response. This is confirmed by the finding that PAK4KO tumour grew at a similar pace to WT after escaping the initial regression (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef-j). The PAK WT cancer cells were injected two weeks after PAK4KO cells as PAK4KO tumour grew out of the initial regression after two weeks. When comparing the PAK4KO tumour with WT tumour, there was no difference in tumour weight (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ei) though PAK4KO tumour still demonstrated a slightly slower growth rate in comparison to WT in tumour size (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eh). These results suggested that PAK4KO tumour developed an immune evasion after an initial phase due to a reduced immune response induced by PAK4KO from one week to four weeks.\u003c/p\u003e \u003cp\u003e \u003cb\u003ePAK1 inhibition induced CD8\u0026thinsp;+\u0026thinsp;T cell infiltration at a later phase.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eGlobal proteomic analysis of PAK WT and KO tumours also identified an up-regulation of PAK1 expression from 1week to 4 weeks in both WT and PAK4KO tumours, but slightly greater in PAK4KO tumour (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ea, b). Functional enrichment of PAK4KO identified differential expression of proteins involved in T-cell receptor signalling pathways between one week and four weeks. PPI network was constructed and annotated by log2(FC) of individual proteins. VAV family proteins, CD3g, ZAP70 and LCK were down-regulated at four weeks in comparison to one week (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec), suggesting a downregulation of T-cell response in PAK4KO tumour over time. Furthermore, the upregulation of PAK1 over four weeks was found to be involved in T-cell receptor signaling pathways (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ec), indicating that the increase of PAK1 level in PAK4KO tumour over time may contribute to their immune resistance.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003ePAK4KO tumour still had slightly higher CD8\u0026thinsp;+\u0026thinsp;T cell infiltration than WT tumour by four weeks, but the difference was a lot less compared to the increase at one week (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee, f, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed, e). Furthermore, the percentage of total T cells (CD45\u0026thinsp;+\u0026thinsp;CD3+) in PAK4KO tumour was no longer different to WT tumour after four weeks. On the other hand, PAK1KO tumour demonstrated a significant increase of CD8\u0026thinsp;+\u0026thinsp;T cell infiltration by three weeks (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed, e) compared to at one week (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee, f), suggesting a delayed CD8\u0026thinsp;+\u0026thinsp;T cell response in PAK1 KO tumour. In addition, PAK4KO reduced the percentage of Treg and increased the level of PD1\u0026thinsp;+\u0026thinsp;CD8\u0026thinsp;+\u0026thinsp;T cells at one week (Fig. S5e-h), which were reversed by four weeks (Fig. S7a-d), suggesting a more immunosuppressive environment. On the other hand, although PAK1KO had a high level of Treg (Fig. S7a, b), it also had an increase in PD1\u0026thinsp;+\u0026thinsp;CD8\u0026thinsp;+\u0026thinsp;T cell level (Fig. S7c, d) from one week to three weeks, suggesting that CD8\u0026thinsp;+\u0026thinsp;T cells had increased anti-tumour activity in PAK1KO tumour over time. Together these results indicated that PAK1KO caused a delayed infiltration of CD8\u0026thinsp;+\u0026thinsp;T cells which may compensate for the reduced immune response to PAK4KO over time.\u003c/p\u003e \u003cp\u003e \u003cb\u003eLow PAK1 and PAK4 expressions improved T cell function in human pancreatic cancer.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eTo study the roles of PAK1 and PAK4 in human PDA, 178 PDA cases were assessed from the TCGA database. High PAK1 and PAK4 gene expressions were associated with reduced survival in PDA patients (Fig. S8a, b). Functional enrichment of genes correlated with PAK1 (Fig. S8c) and PAK4 (Fig. S8d) revealed immune-related pathways for PAK4 but not PAK1. Tumour purity adjusted estimation of intra-tumoral B and T cell infiltration showed a negative correlation between PAK4 and CD8\u0026thinsp;+\u0026thinsp;T cell infiltration, but not with B cells or CD4\u0026thinsp;+\u0026thinsp;T cells (Fig. S8f). No correlation between PAK1 and lymphocyte infiltration was demonstrated (Fig. S8e).\u003c/p\u003e \u003cp\u003eIn a TMA analysis of PDA patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ea), patients were grouped into low or high expression groups based on median of PAK1 or PAK4 intensity as well as median percentage of CD4\u0026thinsp;+\u0026thinsp;or CD8\u0026thinsp;+\u0026thinsp;positive cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eb, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ea). Individual study variables were compared between low versus high expression groups for PAK1 and PAK4 (Table S4 and S5). To assess the relationship between PAK expression and the level of T cell infiltration, both univariate and multivariate linear regression models were applied (Table S6 and S7). By univariate analysis, PAK4 expression was positively correlated with CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cell infiltration and PAK1 expression was positively correlated with CD4\u0026thinsp;+\u0026thinsp;T cell infiltration. However, these correlations disappeared in multivariate analysis (Table S6 and S7).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eOverall survival of patients was demonstrated by Kaplan Meier curve (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ec), with a number of patients at risk and events reported. While high levels of CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cell infiltration significantly improved overall survival (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ef, g), PAK1 and PAK4 expression levels did not significantly affect patient survival by univariate analysis (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003ed, e). However, high PAK4 levels increased risk of death with multivariate analysis (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e). Furthermore, PAK1 and PAK4 demonstrated negative interaction (HR below 1), indicating that a reduction in the level of either PAK1 or PAK4 would lead to worse survival outcome for the other (Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eUnivariate and multivariate Cox regression for survival analysis\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eUnivariate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMultivariate\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eVariable\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eHazard ratio (HR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eHazard ratio (HR)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep-value\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAge\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.005**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eSex\u003c/p\u003e \u003cp\u003e- Male\u003c/p\u003e \u003cp\u003e- Female\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.89\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.80\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePAK1\u003c/p\u003e \u003cp\u003e- Low\u003c/p\u003e \u003cp\u003e- High\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.06\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.39\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePAK4\u003c/p\u003e \u003cp\u003e- Low\u003c/p\u003e \u003cp\u003e- High\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.05\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.58\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.04*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePAK1: PAK4 interaction\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.021*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCD4\u003c/p\u003e \u003cp\u003e- Low\u003c/p\u003e \u003cp\u003e- High\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.59\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.028*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.35\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.003**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCD8\u003c/p\u003e \u003cp\u003e- Low\u003c/p\u003e \u003cp\u003e- High\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.016*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCancer Site\u003c/p\u003e \u003cp\u003e- Head/neck\u003c/p\u003e \u003cp\u003e- Body/tail\u003c/p\u003e \u003cp\u003e- Multifocal\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.46\u003c/p\u003e \u003cp\u003e2.13\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.092\u003c/p\u003e \u003cp\u003e0.5\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.31\u003c/p\u003e \u003cp\u003e2.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.2\u003c/p\u003e \u003cp\u003e0.4\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eResection Margin\u003c/p\u003e \u003cp\u003e- R0\u003c/p\u003e \u003cp\u003e- R1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.02\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.003**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.95\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.003**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eT stage\u003c/p\u003e \u003cp\u003e- 1\u003c/p\u003e \u003cp\u003e- 2\u003c/p\u003e \u003cp\u003e- 3\u003c/p\u003e \u003cp\u003e- 4\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.9\u003c/p\u003e \u003cp\u003e4\u003c/p\u003e \u003cp\u003e3.79\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.009**\u003c/p\u003e \u003cp\u003e0.003**\u003c/p\u003e \u003cp\u003e0.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e6.22\u003c/p\u003e \u003cp\u003e7.57\u003c/p\u003e \u003cp\u003e42.6\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.002**\u003c/p\u003e \u003cp\u003e0.002**\u003c/p\u003e \u003cp\u003e0.006**\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eN stage\u003c/p\u003e \u003cp\u003e- 0\u003c/p\u003e \u003cp\u003e- 1\u003c/p\u003e \u003cp\u003e- 2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.24\u003c/p\u003e \u003cp\u003e1.67\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.5\u003c/p\u003e \u003cp\u003e0.078\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.47\u003c/p\u003e \u003cp\u003e2.15\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.3\u003c/p\u003e \u003cp\u003e0.052\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eM stage\u003c/p\u003e \u003cp\u003e- 0\u003c/p\u003e \u003cp\u003e- 1\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e24.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.004**\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e98.8\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001***\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eGrade\u003c/p\u003e \u003cp\u003e- 1\u003c/p\u003e \u003cp\u003e- 2\u003c/p\u003e \u003cp\u003e- 3\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.08\u003c/p\u003e \u003cp\u003e3.56\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.3\u003c/p\u003e \u003cp\u003e0.081\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.59\u003c/p\u003e \u003cp\u003e1.63\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.7\u003c/p\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eLymphovascular Invasion\u003c/p\u003e \u003cp\u003e- Yes\u003c/p\u003e \u003cp\u003e- No\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.72\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.2\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.88\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.7\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePerineural Invasion\u003c/p\u003e \u003cp\u003e- Yes\u003c/p\u003e \u003cp\u003e- No\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.68\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.11\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.035*\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAdjuvant chemotherapy\u003c/p\u003e \u003cp\u003e- Yes\u003c/p\u003e \u003cp\u003e- No\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e1.77\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.045*\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e2.01\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e-\u003c/p\u003e \u003cp\u003e0.11\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colspan=\"5\" nameend=\"c5\" namest=\"c1\"\u003e \u003cp\u003eAbbreviations: CI: confidence interval. Statistical significance: p\u0026thinsp;\u0026lt;\u0026thinsp;0.05\u003csup\u003e*\u003c/sup\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.01\u003csup\u003e**\u003c/sup\u003e, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001\u003csup\u003e***\u003c/sup\u003e.\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003eFinally, subgroup analysis of overall survival was conducted for levels of CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cells based on PAK1 or PAK4 expression. While high levels of CD4\u0026thinsp;+\u0026thinsp;or CD8\u0026thinsp;+\u0026thinsp;T cells failed to improve patient survival in the high PAK1 expression group (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb, c), high level of CD4\u0026thinsp;+\u0026thinsp;T cells significantly improved overall survival in the low PAK1 subgroup (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eb). There was also a trend suggesting high level CD8\u0026thinsp;+\u0026thinsp;T cells was associated with better survival (p\u0026thinsp;=\u0026thinsp;0.058) in low PAK1 patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ec). On the other hand, PAK4 expression level did not affect the correlation between CD4\u0026thinsp;+\u0026thinsp;T cells and patient survival (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ed). However, despite not being statistically significant, the low PAK4 expression level resulted in a trend (p\u0026thinsp;=\u0026thinsp;0.069) of improved overall survival by CD8\u0026thinsp;+\u0026thinsp;T cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003ee).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe individual roles of PAK1 and PAK4 in PDA tumorigenesis has been recognized in the literature[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. However, development of selective PAK1 or PAK4 inhibitors for the treatment of solid tumours has not been successful[\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e, \u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e]. The fact that both PAK1 and PAK4 have been implicated in intra-tumoral T cell response, suggested that PAK1 and PAK4 inhibition may function synergistically in suppressing the growth of PDA[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e]. In this study, we demonstrated a synergistic effect of PAK1 and PAK4 inhibition in suppressing PDA growth in mice. PAK4KO stimulated the infiltration and activation of CD8\u0026thinsp;+\u0026thinsp;T cells in tumour to a greater degree at an initial phase while PAK1KO caused an increased infiltration of active CD8\u0026thinsp;+\u0026thinsp;T cells at a late phase. Together PAK1 and PAK4 double KO stimulated a sustained increase of infiltration of active CD8\u0026thinsp;+\u0026thinsp;T cells, leading to a complete tumour regression. The results from a TMA of human PDA also confirmed the importance of PAK1 and PAK4 in intra-tumoral CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cell function, and the impact on overall survival of PDA patients.\u003c/p\u003e \u003cp\u003eThe fact that PAK4KO suppressed tumour growth in a syngeneic mouse model but not in SCID mice, indicates that PAK4KO regresses tumour growth through stimulating the anti-tumour immunity. An increased infiltration of cytotoxic CD8\u0026thinsp;+\u0026thinsp;T cell was demonstrated in PAK4KO PDA tumour, which is consistent with the results from melanoma and prostate cancer[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e]. However, an immune evasion developed in PAK4KO tumour (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003ef-j) due to a reduced infiltration and activation of CD8\u0026thinsp;+\u0026thinsp;T cell at a later phase (four weeks, Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed, e). While PAK1KO did not enhance CD8\u0026thinsp;+\u0026thinsp;T cell infiltration significantly at the initial phase (one week, Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003ee,f), it caused delayed increase in the infiltration of CD8\u0026thinsp;+\u0026thinsp;T cell (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003ed,e), which is consistent with our previous findings in a syngeneic mice model[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. This delayed effect of PAK1KO is likely to compensate for the reduced anti-tumour immune response by PAK4KO over time, leading to a sustained immune response to kill cancer cells. Therefore, the tumour was completed regressed by PAK1 and PAK4 double knockout (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003ee).\u003c/p\u003e \u003cp\u003eWhile high PAK1 expression was previously shown to be negatively correlated with PDA patient survival, this was not found in our results from the human PDA TMA dataset[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. After correcting for tumour stage, grade, and resection status, high PAK4 expression was noted to associate with worse outcomes which is the opposite to a previous report[\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e]. We also found a positive correlation between CD4\u0026thinsp;+\u0026thinsp;or CD8\u0026thinsp;+\u0026thinsp;T cell infiltration and PDA patient survival. Furthermore, low PAK1 or PAK4 levels enhanced the pro-survival effects of CD4\u0026thinsp;+\u0026thinsp;and CD8\u0026thinsp;+\u0026thinsp;T cells. This again highlights the important roles of PAK1 and PAK4 in T cell response in human PDA.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eOur results identified a synergistic effect of PAK1 and PAK4 inhibition on PDA growth and T-cell immune response. It also indicated a rapid development of immune evasion with selective PAK4 inhibition in PDA, which may explain the failure of selective PAK4 inhibitors in clinical trials. However, pan-PAK inhibitors have a high level of toxicity as PAK2 inhibition can lead to endothelial cell dysfunction and vascular malformation[\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e]. Recent development of a selective PAK1 degrader by proteolysis-targeting chimera (PROTAC) technique has increased the hope of developing more selective PAK inhibitors[\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e]. Whether the combination of selective PAK1 and PAK4 degraders can offer greater clinical efficacy while minimizing side effects remains a question to be addressed and requires further research.\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/h2\u003e\n\u003cp\u003eThe collection of patient\u0026rsquo;s clinical information and generation of human PDA tissue microarray (TMA) were approved by the Austin Health Human Research Ethics Committee (HREC/73948/Austin- 2021).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNA\u003c/strong\u003e\u003c/p\u003e\n\u003ch2\u003eCompeting interests\u003c/h2\u003e\n\u003cp\u003eYM, CD, LD, CA, KA, YZ, MN and HH have no conflict of interest to declare.\u003c/p\u003e\n\u003ch2\u003eFunding\u003c/h2\u003e\n\u003cp\u003eThis work was supported by grants from Pancare Foundation, Austin Medical Research Foundation (HH-2021, MN-2020) and MDHS (Medicine Dental Health Science, University of Melbourne) Seeding Ideas Grants (MN2020, HH and CA 2021).\u003c/p\u003e\n\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\n\u003cp\u003eY.M. and H.H. conceptualised and designed the study. Y.M., C.D., L.D., C.A., H.H. and K.A. were involved in sample preparation, conduction of experiments and data curation. Y.M., H.H. and Y.Z. analysed the data. Y.M. drafted the manuscript. H.H. and Y..Z modified the manuscript. All authors reviewed the manuscript.\u003c/p\u003e\n\u003ch2\u003eAcknowledgement\u003c/h2\u003e\n\u003cp\u003eYM is supported by Australian Government Research Training Program (RTP) Scholarship, Royal Australasian College of Surgeons (RACS) Foundation of Surgery Scholarship and Pancare Foundation Tim McGahan PhD Scholarship. HH is supported by the Henry Baldwin Cancer Research Trust Fund. Ms Shuji Jiang for design and creation of figures. Dr David Baloyan for advice and assistance on flow cytometry. Associate Professor Sarah Ellis for advice on immunohistochemistry and whole slide scanning. Ms Kat Hall for assistance in human ethics application. Graphical abstract was created by Biorender online platform (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://www.biorender.com\u003c/span\u003e\u003c/span\u003e).\u003c/p\u003e\n\u003ch2\u003eAvailability of data and materials\u003c/h2\u003e\n\u003cp\u003eData and study material will be available upon request from the corresponding author.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAIHW. Cancer in Australia 2019. Cancer series. Canberra: Australian Institute of Health and Welfare; 2019.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAdamska A, Domenichini A, Falasca M. Pancreatic Ductal Adenocarcinoma: Current and Evolving Therapies. Int J Mol Sci. 2017;18(7).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVon Hoff DD, Ervin T, Arena FP, Chiorean EG, Infante J, Moore M, et al. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med. 2013;369(18):1691\u0026ndash;703.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eConroy T, Desseigne F, Ychou M, Bouche O, Guimbaud R, Becouarn Y, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817\u0026ndash;25.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eManser E, Leung T, Salihuddin H, Zhao ZS, Lim L. A brain serine/threonine protein kinase activated by Cdc42 and Rac1. Nature. 1994;367(6458):40\u0026ndash;6.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMenard RE, Mattingly RR. Cell surface receptors activate p21-activated kinase 1 via multiple Ras and PI3-kinase-dependent pathways. Cell Signal. 2003;15(12):1099\u0026ndash;109.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaitra A, Hruban RH. Pancreatic cancer. Annu Rev Pathol. 2008;3:157\u0026ndash;88.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeo D, He H, Baldwin GS, Nikfarjam M. The role of p21-activated kinases in pancreatic cancer. Pancreas. 2015;44(3):363\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYeo D, Phillips P, Baldwin GS, He H, Nikfarjam M. Inhibition of group 1 p21-activated kinases suppresses pancreatic stellate cell activation and increases survival of mice with pancreatic cancer. Int J Cancer. 2017;140(9):2101\u0026ndash;11.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang K, Zhan Y, Huynh N, Dumesny C, Wang X, Asadi K, et al. Inhibition of PAK1 suppresses pancreatic cancer by stimulation of anti-tumour immunity through down-regulation of PD-L1. Cancer Lett. 2020;472:8\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJagadeeshan S, Subramanian A, Tentu S, Beesetti S, Singhal M, Raghavan S, et al. P21-activated kinase 1 (Pak1) signaling influences therapeutic outcome in pancreatic cancer. Ann Oncol. 2016;27(8):1546\u0026ndash;56.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTyagi N, Bhardwaj A, Singh AP, McClellan S, Carter JE, Singh S. p-21 activated kinase 4 promotes proliferation and survival of pancreatic cancer cells through AKT- and ERK-dependent activation of NF-kappaB pathway. Oncotarget. 2014;5(18):8778\u0026ndash;89.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMohammad RM, Li Y, Muqbil I, Aboukameel A, Senapedis W, Baloglu E, et al. Targeting Rho GTPase effector p21 activated kinase 4 (PAK4) suppresses p-Bad-microRNA drug resistance axis leading to inhibition of pancreatic ductal adenocarcinoma proliferation. Small GTPases. 2019;10(5):367\u0026ndash;77.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTyagi N, Marimuthu S, Bhardwaj A, Deshmukh SK, Srivastava SK, Singh AP, et al. p-21 activated kinase 4 (PAK4) maintains stem cell-like phenotypes in pancreatic cancer cells through activation of STAT3 signaling. Cancer Lett. 2016;370(2):260\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbril-Rodriguez G, Torrejon DY, Liu W, Zaretsky JM, Nowicki TS, Tsoi J, et al. PAK4 inhibition improves PD-1 blockade immunotherapy. Nat Cancer. 2020;1(1):46\u0026ndash;58.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAbril-Rodriguez G, Torrejon DY, Karin D, Campbell KM, Medina E, Saco JD, et al. Remodeling of the tumor microenvironment through PAK4 inhibition sensitizes tumors to immune checkpoint blockade. Cancer Res Commun. 2022;2(10):1214\u0026ndash;28.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa W, Wang Y, Zhang R, Yang F, Zhang D, Huang M, et al. Targeting PAK4 to reprogram the vascular microenvironment and improve CAR-T immunotherapy for glioblastoma. Nat Cancer. 2021;2(1):83\u0026ndash;97.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu S, You S, Wang Y, Tamukong P, Quist MJ, Grasso CS, et al. PAK4 inhibition improves PD1 blockade immunotherapy in prostate cancer by increasing immune infiltration. Cancer Lett. 2022;555:216034.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong L, Vaux DL. Glucocorticoids can induce BIM to trigger apoptosis in the absence of BAX and BAK1. Cell Death Dis. 2020;11(6):442.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBankhead P, Loughrey MB, Fernandez JA, Dombrowski Y, McArt DG, Dunne PD, et al. QuPath: Open source software for digital pathology image analysis. Sci Rep. 2017;7(1):16878.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTakeuchi A, Saito T. CD4 CTL, a Cytotoxic Subset of CD4(+) T Cells, Their Differentiation and Function. Front Immunol. 2017;8:194.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSimon S, Labarriere N. PD-1 expression on tumor-specific T cells: Friend or foe for immunotherapy? Oncoimmunology. 2017;7(1):e1364828.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSimon S, Vignard V, Florenceau L, Dreno B, Khammari A, Lang F, et al. PD-1 expression conditions T cell avidity within an antigen-specific repertoire. Oncoimmunology. 2016;5(1):e1104448.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCourtney AH, Lo WL, Weiss A. TCR Signaling: Mechanisms of Initiation and Propagation. Trends Biochem Sci. 2018;43(2):108\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTybulewicz VL. Vav-family proteins in T-cell signalling. Curr Opin Immunol. 2005;17(3):267\u0026ndash;74.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoudronniere N, Villalba M, Englund N, Altman A. NF-kappa B activation induced by T cell receptor/CD28 costimulation is mediated by protein kinase C-theta. Proc Natl Acad Sci U S A. 2000;97(7):3394\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSgarbanti M, Marsili G, Remoli AL, Stellacci E, Mai A, Rotili D, et al. IkappaB kinase epsilon targets interferon regulatory factor 1 in activated T lymphocytes. Mol Cell Biol. 2014;34(6):1054\u0026ndash;65.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang J, Feng H, Zhao J, Feldman ER, Chen SY, Yuan W, et al. IkappaB Kinase epsilon Is an NFATc1 Kinase that Inhibits T Cell Immune Response. Cell Rep. 2016;16(2):405\u0026ndash;18.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMa Y, Nikfarjam M, He H. The trilogy of P21 activated kinase, autophagy and immune evasion in pancreatic ductal adenocarcinoma. Cancer Lett. 2022;548:215868.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHe H, Dumesny C, Ang CS, Dong L, Ma Y, Zeng J, et al. A novel PAK4 inhibitor suppresses pancreatic cancer growth and enhances the inhibitory effect of gemcitabine. Transl Oncol. 2021;16:101329.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSu S, You S, Wang Y, Tamukong P, Quist MJ, Grasso CS, et al. PAK4 inhibition improves PD1 blockade immunotherapy in prostate cancer by increasing immune infiltration. Cancer Lett. 2023;555:216034.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark S, Kim JW, Kim H, Kim JW, Kim YJ, Lee KW, et al. Prognostic value of p21-activated kinase 4 in resected pancreatic cancer. APMIS. 2017;125(8):699\u0026ndash;707.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRadu M, Lyle K, Hoeflich KP, Villamar-Cruz O, Koeppen H, Chernoff J. p21-Activated Kinase 2 Regulates Endothelial Development and Function through the Bmk1/Erk5 Pathway. Mol Cell Biol. 2015;35(23):3990\u0026ndash;4005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChow HY, Karchugina S, Groendyke BJ, Toenjes S, Hatcher J, Donovan KA, et al. Development and Utility of a PAK1-Selective Degrader. J Med Chem. 2022;65(23):15627\u0026ndash;41.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"cell-communication-and-signaling","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ccas","sideBox":"Learn more about [Cell Communication and Signaling](http://biosignaling.biomedcentral.com/)","snPcode":"12964","submissionUrl":"https://submission.nature.com/new-submission/12964/3","title":"Cell Communication and Signaling","twitterHandle":"@bmc","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"p21-activated kinases1\u00264, intra-tumoral T cells, pancreatic ductal adenocarcinoma","lastPublishedDoi":"10.21203/rs.3.rs-3974396/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-3974396/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground:\u003c/strong\u003e Pancreatic ductal adenocarcinoma (PDA) is one of the most lethal types of cancer, and \u003cem\u003eKRAS\u003c/em\u003e oncogene occurs in over 90% of cases. P21-activated kinases (PAK), containing six members (PAK1 to 6), function downstream of KRAS. PAK1 and PAK4 play important roles in carcinogenesis, but their combinational effect remains unknown. In this study, we have determined the effect of dual inhibition of PAK1 and PAK4 in PDA progression using knockout (KO) cancer cell lines.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods:\u003c/strong\u003e Murine wild-type (WT) and PAK1KO pancreatic cancer cell lines were isolated from PAK1\u003csup\u003e+/+\u003c/sup\u003e and PAK1\u003csup\u003e-/-\u003c/sup\u003e KPC (\u003cem\u003eLSL-Kras\u003c/em\u003e\u003csup\u003e\u003cem\u003eG12D/+\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e; LSL-Trp53 \u003c/em\u003e\u003csup\u003e\u003cem\u003eR172H/+\u003c/em\u003e\u003c/sup\u003e\u003cem\u003e; Pdx-1-Cre\u003c/em\u003e) mice. KPC PAK4KO and KPC PAK1\u0026amp;4 KO cell lines were generated from KPC WT and KPC PAK1KO cell lines respectively using the CRISPR-CAS9 gene knockout technique. PAK WT and KO cell lines were used in mouse models of pancreatic tumours. Cells and tumour tissue were also used in flow cytometry and proteomic studies. A human PDA tissue microarray was stained by immunohistochemistry.\u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults:\u003c/strong\u003e Double knock out of PAK1 and PAK4 caused complete regression of tumour in a syngeneic mouse. PAK4KO inhibited tumour growth by stimulating a rapid increase of cytotoxic CD8+ T cell infiltration. \u0026nbsp;PAK1KO synergistically with PAK4KO increased cytotoxic CD8+ T cell infiltration and stimulated a sustained infiltration of CD8+ T cells at a later phase to overcome the immune evasion in the PAK4KO tumour. The human PDA tissue microarray study showed the important role of PAK1 and PAK4 in intra-tumoral T-cell function.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion:\u003c/strong\u003e Our results demonstrated that dual inhibition of PAK1 and PAK4 synergistically suppressed PDA progression by stimulating cytotoxic CD8+ T cell response.\u003c/p\u003e","manuscriptTitle":"Inhibition of P21-activated kinases 1 and 4 synergistically suppress the growth of pancreatic cancer by stimulating anti-tumour immunity","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-03-11 19:19:57","doi":"10.21203/rs.3.rs-3974396/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"decision","content":"Revision requested","date":"2024-04-21T17:52:00+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2024-03-16T21:36:12+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"d98d8c21-4a0b-45c3-ba07-5e65c2572b8a","date":"2024-03-09T21:42:43+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2024-03-09T21:38:01+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2024-03-07T04:28:01+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2024-03-07T04:28:01+00:00","index":"","fulltext":""},{"type":"submitted","content":"Cell Communication and Signaling","date":"2024-02-21T02:42:00+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"cell-communication-and-signaling","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"ccas","sideBox":"Learn more about [Cell Communication and Signaling](http://biosignaling.biomedcentral.com/)","snPcode":"12964","submissionUrl":"https://submission.nature.com/new-submission/12964/3","title":"Cell Communication and Signaling","twitterHandle":"@bmc","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"c8f97f19-e5ec-4310-b429-4c81e4853e1e","owner":[],"postedDate":"March 11th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2024-05-19T20:23:12+00:00","versionOfRecord":[],"versionCreatedAt":"2024-03-11 19:19:57","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-3974396","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-3974396","identity":"rs-3974396","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}