Search

Lung cancer is the leading cause of cancer death worldwide, with lung adenocarcinoma being the most common subtype. Many oncogenes and tumor suppressor genes are altered in this cancer type, and the discovery of oncogene mutations has led to the development of targeted therapies that have improved clinical outcomes. However, a large fraction of lung adenocarcinomas lacks mutations in known oncogenes, and the genesis and treatment of these oncogene-negative tumors remain enigmatic. Here, we perform iterative in vivo functional screens using quantitative autochthonous mouse model systems to uncover the genetic and biochemical changes that enable efficient lung tumor initiation in the absence of oncogene alterations. Generation of hundreds of diverse combinations of tumor suppressor alterations demonstrates that inactivation of suppressors of the RAS and PI3K pathways drives the development of oncogene-negative lung adenocarcinoma. Human genomic data and histology identified RAS/MAPK and PI3K pathway activation as a common feature of an event in oncogene-negative human lung adenocarcinomas. These Onc-negativeRAS/PI3K tumors and related cell lines are vulnerable to pharmacologic inhibition of these signaling axes. These results transform our understanding of this prevalent yet understudied subtype of lung adenocarcinoma. Significance: To address the large fraction of lung adenocarcinomas lacking mutations in proto-oncogenes for which targeted therapies are unavailable, this work uncovers driver pathways of oncogene-negative lung adenocarcinomas and demonstrates their therapeutic vulnerabilities.

Table S2. The information of sgRNA vectors in the high content Lenti-sgTS102/Cre pool.

(Updated, with minor text changes and rearrangements.) Figure S1-S4 (related to Figure 1) provide details on the selected genes and method. Figure S5 (related to Figure 2) shows the percentiles of all sgRNAs. Figure S6-S8 (related to Figure 3) present Stag2 validation and mechanistic investigations. Figure S9 (related to Figure 4) presents the percentiles of all genotypes at 26-week stage. Figure S10-S11 (related to Figure 5) suggest that genes that regulate tumor number are consistent across datasets. Figure S12 (related to Figure 6) provides additional evidence for the genes that regulate rare large tumors. Figure S13-S14 measure the influences of crowding and sex on Tuba-seq. Figure S15-S16 and Table S3 evaluate the specificity and sensitivity of Tuba-seq. Figure S17 (related to Figure 7) shows the complementarity of Tuba-seq to human datasets.

Supplementary tables 1-6 summarise the kinase panel screen data (1) mouse PK parameters (2) and further details of the anti-tumour activity (5) conferred by ASTX029, plus details of cell lines used in this study (3 and 4) and further details of the effects of ASTX029 and other MAPK inhibitors in models of MAPK resistance (6).

A) Structure of JNJ-42541707, a structurally related compound to JNJ-42756493. B) 72h growth inhibition (IC50) of JNJ-42541707 against 236 cancer cell lines from multiple origins color coded based on FGFR1,2,4 mRNA overexpression and FGFR WT.

Summary of gene sets generated from previously published analyses of the LKB1-deficient state in human lung adenocarcinoma cell lines and tumors as well as genetically engineered mouse models.

Cancer genotyping has identified a large number of putative tumor suppressor genes. Carcinogenesis is a multistep process, but the importance and specific roles of many of these genes during tumor initiation, growth, and progression remain unknown. Here we use a multiplexed mouse model of oncogenic KRAS–driven lung cancer to quantify the impact of 48 known and putative tumor suppressor genes on diverse aspects of carcinogenesis at an unprecedented scale and resolution. We uncover many previously understudied functional tumor suppressors that constrain cancer in vivo. Inactivation of some genes substantially increased growth, whereas the inactivation of others increases tumor initiation and/or the emergence of exceptionally large tumors. These functional in vivo analyses revealed an unexpectedly complex landscape of tumor suppression that has implications for understanding cancer evolution, interpreting clinical cancer genome sequencing data, and directing approaches to limit tumor initiation and progression.Significance:Our high-throughput and high-resolution analysis of tumor suppression uncovered novel genetic determinants of oncogenic KRAS–driven lung cancer initiation, overall growth, and exceptional growth. This taxonomy is consistent with changing constraints during the life history of cancer and highlights the value of quantitative in vivo genetic analyses in autochthonous cancer models.This article is highlighted in the In This Issue feature, p. 1601

Supplementary figures 1-7 show the effects of AST029 on ERK signalling in RAS-mutant cell lines (1) and tumour xenograft tissue (3), ASTX029 plasma PK linearity (2), ASTX029 activity in a cell line panel showing MAPK mutation status (4) the effects of ASTX029 on mouse body weight (5), PKPD effects of ASTX029 following b.i.d dosing to Colo205 tumour-bearing mice (6) and PD effects of ASTX029 following qd dosing to A375R tumour-bearing mice (7).

In vitro kinase inhibition by DiscoverX KinomeScan assay. For details see (19)

Inhibitory activity of Brivanib and JNJ-42756943 in kinase (top) and BaF3 kinase dependent proliferation (bottom) assays and ratio of FGFRs/VEGFR2 activities.

JNJ-42756493 anti-proliferative activity against cancer cells lines from multiple origins.. Detailed data supporting Figure 2.

Supplementary Table from Combinatorial Inactivation of Tumor Suppressors Efficiently Initiates Lung Adenocarcinoma with Therapeutic Vulnerabilities

Lung cancer is the leading cause of cancer death worldwide, with lung adenocarcinoma being the most common subtype. Many oncogenes and tumor suppressor genes are altered in this cancer type, and the discovery of oncogene mutations has led to the development of targeted therapies that have improved clinical outcomes. However, a large fraction of lung adenocarcinomas lacks mutations in known oncogenes, and the genesis and treatment of these oncogene-negative tumors remain enigmatic. Here, we perform iterative in vivo functional screens using quantitative autochthonous mouse model systems to uncover the genetic and biochemical changes that enable efficient lung tumor initiation in the absence of oncogene alterations. Generation of hundreds of diverse combinations of tumor suppressor alterations demonstrates that inactivation of suppressors of the RAS and PI3K pathways drives the development of oncogene-negative lung adenocarcinoma. Human genomic data and histology identified RAS/MAPK and PI3K pathway activation as a common feature of an event in oncogene-negative human lung adenocarcinomas. These Onc-negativeRAS/PI3K tumors and related cell lines are vulnerable to pharmacologic inhibition of these signaling axes. These results transform our understanding of this prevalent yet understudied subtype of lung adenocarcinoma.Significance:To address the large fraction of lung adenocarcinomas lacking mutations in proto-oncogenes for which targeted therapies are unavailable, this work uncovers driver pathways of oncogene-negative lung adenocarcinomas and demonstrates their therapeutic vulnerabilities.

Supplementary Data from Combinatorial Inactivation of Tumor Suppressors Efficiently Initiates Lung Adenocarcinoma with Therapeutic Vulnerabilities

Supplementary Figure from Combinatorial Inactivation of Tumor Suppressors Efficiently Initiates Lung Adenocarcinoma with Therapeutic Vulnerabilities

Phagocytosis is a key macrophage function, but how phagocytosis shapes tumor-associated macrophage (TAM) phenotypes and heterogeneity in solid tumors remains unclear. Here, we utilized both syngeneic and novel autochthonous lung tumor models in which neoplastic cells express the fluorophore tdTomato (tdTom) to identify TAMs that have phagocytosed neoplastic cells in vivo. Phagocytic tdTompos TAMs upregulated antigen presentation and anti-inflammatory proteins, but downregulated classic proinflammatory effectors compared to tdTomneg TAMs. Single-cell transcriptomic profiling identified TAM subset-specific and common gene expression changes associated with phagocytosis. We uncover a phagocytic signature that is predominated by oxidative phosphorylation (OXPHOS), ribosomal, and metabolic genes, and this signature correlates with worse clinical outcome in human lung cancer. Expression of OXPHOS proteins, mitochondrial content, and functional utilization of OXPHOS were increased in tdTompos TAMs. tdTompos tumor dendritic cells also display similar metabolic changes. Our identification of phagocytic TAMs as a distinct myeloid cell state links phagocytosis of neoplastic cells in vivo with OXPHOS and tumor-promoting phenotypes.

Lung cancer is a lethal and genomically-complex disease. Structural genomics has largely advanced our knowledge of genomic alterations, yet the function of a majority of altered genes remains less clear. Previous in silico and in vitro functional genomics data often lead to contradictory conclusions on gene functions. Genetically-engineered mouse models are reliable approaches for in vivo functional analyses, but development of these models are lagging behind due to the throughput limit. To overcome this throughput limit, we developed tumor barcoding and ultradeep barcode sequencing (Tuba-seq) that precisely quantifies the growth metrics of hundreds of tumor genotypes, which is a huge leap forward. Through this approach, we have begun a journey to create a quantitative functional taxonomy of tumor suppression in oncogenic KRAS-driven lung cancer. For example, STAG2 and CDKN2C emerged as novel functional tumor suppressor genes in the lung, when they were often overlooked by computational analyses due to relatively low mutation prevalence. Interestingly, STK11 and PTEN, both playing an important role in tumor growth, exhibit distinct roles in tumor initiation. These findings suggest that structural genomics is not sufficient to predict cancer driver genes, and calls for closer investigation of tumor suppressor functions in specific tumorigenesis stages. Furthermore, the quantitative nature of our data has enabled systematic characterization of interactions between tumor suppressor genes. For instance, RNF43 exhibits different tumor suppression modes in the presence or absence of STK11 or TRP53, while TRP53 can play opposite roles in PTEN- and RB1-deficient tumors. In addition, Foggetti et al. (2021) reported that tumor suppressors can play opposite roles in the contexts of different oncogenes. Collectively, these findings suggest that cooccurring mutations shift the functional landscape of tumor suppressors even in the same pathological subtype of cancer. Given the genomic diversity of lung cancer patients, driver genes may change case by case. We are now investigating the molecular mechanisms underlying these tumor suppressors and their genetic interactions. Our findings underscore the necessity of determining the consequences of enormous combinations of genomic alterations in their natural environment, which is challenging but critical for understanding cancer evolution, interpreting clinical cancer genome sequencing data, and directing approaches to limit tumor initiation and progression. Citation Format: Hongchen Cai, Su Kit Chew, Chuan Li, Christopher W. Murray, Laura Andrejka, Jess D. Hebert, Min K. Tsai, Rui Tang, Nicholas W. Hughes, Emily G. Shuldiner, Emily L. Ashkin, Shi Ya C. Lee, Maryam Yousefi, Dmitri A. Petrov, Charles Swanton, Monte W. Winslow. A journey to deconvolute the multifaceted functions and context-dependency of cancer driver genes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 827.

Lung cancer is the leading cause of cancer death worldwide, with lung adenocarcinoma being the most common subtype. Many oncogenes and tumor suppressor genes are altered in this cancer type and the discovery of oncogene mutations has led to the development of targeted therapies and better clinical outcomes. However, a large fraction of lung adenocarcinomas lacks mutations in known oncogenes, and the genesis and treatment of these oncogene-negative tumors remain enigmatic. Here, we perform iterative in vivo functional screens with quantitative autochthonous mouse model systems to uncover the genetic and biochemical changes that enable efficient lung tumor initiation in the absence of oncogene alterations. Through the generation of hundreds of diverse combinations of tumor suppressor alterations, we demonstrate that inactivation of suppressors of the RAS and PI3K pathways drive the development of oncogene- negative lung adenocarcinoma. Pathway-level human genomic data analysis and histology identified RAS/MAPK and PI3K pathway activation as a common event in oncogene-negative human lung adenocarcinomas. We demonstrate that oncogene-negative tumors and cell lines with activated RAS/MAPK and PI3K pathways are vulnerable to pharmacological inhibition of these signaling axes. Collectively, these results transform our understanding of this prevalent yet understudied subtype of lung adenocarcinoma. STATEMENT OF SIGNIFICANCEMany lung adenocarcinomas do not have mutations in known proto-oncogenes, and as a result, targeted therapies are unavailable for treating these patients. Here, we uncover driver pathways in a subset of these oncogene-negative lung adenocarcinomas and demonstrate the therapeutic value of inhibiting these pathways.

LKB1 is among the most frequently altered tumor suppressors in lung adenocarcinoma. Inactivation of Lkb1 accelerates the growth and progression of oncogenic KRAS-driven lung tumors in mouse models. However, the molecular mechanisms by which LKB1 constrains lung tumorigenesis and whether the cancer state that stems from Lkb1 deficiency can be reverted remains unknown. To identify the processes governed by LKB1 in vivo, we generated an allele which enables Lkb1 inactivation at tumor initiation and subsequent Lkb1 restoration in established tumors. Restoration of Lkb1 in oncogenic KRAS-driven lung tumors suppressed proliferation and led to tumor stasis. Lkb1 restoration activated targets of C/EBP transcription factors and drove neoplastic cells from a progenitor-like state to a less proliferative alveolar type II cell-like state. We show that C/EBP transcription factors govern a subset of genes that are induced by LKB1 and depend upon NKX2-1. We also demonstrate that a defining factor of the alveolar type II lineage, C/EBPα, constrains oncogenic KRAS-driven lung tumor growth in vivo. Thus, this key tumor suppressor regulates lineage-specific transcription factors, thereby constraining lung tumor development through enforced differentiation.

Using data from real-world practice, this analysis compared outcomes in patients with rheumatoid arthritis (RA) initiating treatment with an oral Janus kinase inhibitor, tofacitinib, in combination with persistent, discontinued, or interrupted treatment with oral methotrexate (MTX).This retrospective claims analysis (MarketScan® databases) included data from US patients with RA and at least one prescription claim for tofacitinib, dated between January 1, 2013, and April 30, 2017. Eligible patients were continuously enrolled for ≥12 months before and after treatment initiation, and initiated tofacitinib in combination with oral MTX, with at least two prescription claims for each. Patients were grouped according to treatment pattern (MTX-Persistent, MTX-Discontinued, or MTX-Interrupted). Tofacitinib treatment persistence, adherence, and effectiveness, as well as all-cause and RA-related health care costs, were assessed.A total of 671 patients were eligible for inclusion; 504 (75.1%) were MTX-Persistent; 131 (19.5%), MTX-Discontinued; and 36 (5.4%), MTX-Interrupted. Rates of tofacitinib treatment persistence, adherence, and effectiveness at 12 months were similar between the MTX-Persistent and MTX-Discontinued cohorts. The percentage of patients switched from tofacitinib to another advanced disease-modifying antirheumatic drug within 12 months of tofacitinib initiation was greater in the MTX-Persistent cohort compared with that in the MTX-Discontinued cohort. RA-related health care costs at 12 months post-initiation were significantly greater in the MTX-Persistent cohort compared with those in the MTX-Discontinued cohort.The findings from this analysis of real-world data indicate that patients who initiate tofacitinib in combination with oral MTX may discontinue MTX and still experience outcomes similar to those in patients who persist with MTX, with lesser RA-related health care costs. These results support those from a previous clinical study on methotrexate withdrawal in patients with RA (NCT02831855).

Cell-cell interactions influence all aspects of development, homeostasis, and disease. In cancer, interactions between cancer cells and stromal cells play a major role in nearly every step of carcinogenesis. Thus, the ability to record cell-cell interactions would facilitate mechanistic delineation of the role of the cancer microenvironment. Here, we describe GFP-based Touching Nexus (G-baToN) which relies upon nanobody-directed fluorescent protein transfer to enable sensitive and specific labeling of cells after cell-cell interactions. G-baToN is a generalizable system that enables physical contact-based labeling between various human and mouse cell types, including endothelial cell-pericyte, neuron-astrocyte, and diverse cancer-stromal cell pairs. A suite of orthogonal baToN tools enables reciprocal cell-cell labeling, interaction-dependent cargo transfer, and the identification of higher order cell-cell interactions across a wide range of cell types. The ability to track physically interacting cells with these simple and sensitive systems will greatly accelerate our understanding of the outputs of cell-cell interactions in cancer as well as across many biological processes., eLife digest It takes the coordinated effort of more than 40 trillion cells to build and maintain a human body. This intricate process relies on cells being able to communicate across long distances, but also with their immediate neighbors. Interactions between cells in close contact are key in both health and disease, yet tracing these connections efficiently and accurately remains challenging. The surface of a cell is studded with proteins that interact with the environment, including with the proteins on neighboring cells. Using genetic engineering, it is possible to construct surface proteins that carry a fluorescent tag called green fluorescent protein (or GFP), which could help to track physical interactions between cells. Here, Tang et al. test this idea by developing a new technology named GFP-based Touching Nexus, or G-baToN for short. Sender cells carry a GFP protein tethered to their surface, while receiver cells present a synthetic element that recognizes that GFP. When the cells touch, the sender passes its GFP to the receiver, and these labelled receiver cells become ‘green’. Using this system, Tang et al. recorded physical contacts between a variety of human and mouse cells. Interactions involving more than two cells could also be detected by using different colors of fluorescent tags. Furthermore, Tang et al. showed that, alongside GFP, G-baToN could pass molecular cargo such as proteins, DNA, and other chemicals to receiver cells. This new system could help to study interactions among many different cell types. Changes in cell-to-cell contacts are a feature of diverse human diseases, including cancer. Tracking these interactions therefore could unravel new information about how cancer cells interact with their environment.

Cancer genotyping has identified a large number of putative tumor suppressor genes. Carcinogenesis is a multistep process, but the importance and specific roles of many of these genes during tumor initiation, growth, and progression remain unknown. Here we use a multiplexed mouse model of oncogenic KRAS–driven lung cancer to quantify the impact of 48 known and putative tumor suppressor genes on diverse aspects of carcinogenesis at an unprecedented scale and resolution. We uncover many previously understudied functional tumor suppressors that constrain cancer in vivo. Inactivation of some genes substantially increased growth, whereas the inactivation of others increases tumor initiation and/or the emergence of exceptionally large tumors. These functional in vivo analyses revealed an unexpectedly complex landscape of tumor suppression that has implications for understanding cancer evolution, interpreting clinical cancer genome sequencing data, and directing approaches to limit tumor initiation and progression. Significance: Our high-throughput and high-resolution analysis of tumor suppression uncovered novel genetic determinants of oncogenic KRAS–driven lung cancer initiation, overall growth, and exceptional growth. This taxonomy is consistent with changing constraints during the life history of cancer and highlights the value of quantitative in vivo genetic analyses in autochthonous cancer models. This article is highlighted in the In This Issue feature, p. 1601

The initiation of cell division integrates a large number of intra- and extracellular inputs. D-type cyclins (hereafter, cyclin D) couple these inputs to the initiation of DNA replication1. Increased levels of cyclin D promote cell division by activating cyclin-dependent kinases 4 and 6 (hereafter, CDK4/6), which in turn phosphorylate and inactivate the retinoblastoma tumour suppressor. Accordingly, increased levels and activity of cyclin D–CDK4/6 complexes are strongly linked to unchecked cell proliferation and cancer2,3. However, the mechanisms that regulate levels of cyclin D are incompletely understood4,5. Here we show that autophagy and beclin 1 regulator 1 (AMBRA1) is the main regulator of the degradation of cyclin D. We identified AMBRA1 in a genome-wide screen to investigate the genetic basis of the response to CDK4/6 inhibition. Loss of AMBRA1 results in high levels of cyclin D in cells and in mice, which promotes proliferation and decreases sensitivity to CDK4/6 inhibition. Mechanistically, AMBRA1 mediates ubiquitylation and proteasomal degradation of cyclin D as a substrate receptor for the cullin 4 E3 ligase complex. Loss of AMBRA1 enhances the growth of lung adenocarcinoma in a mouse model, and low levels of AMBRA1 correlate with worse survival in patients with lung adenocarcinoma. Thus, AMBRA1 regulates cellular levels of cyclin D, and contributes to cancer development and the response of cancer cells to CDK4/6 inhibitors. AMBRA1 is the main regulator of the degradation of D-type cyclins, and loss of AMBRA1 promotes cell proliferation and tumour growth, and reduces the sensitivity of cancer cells to inhibition of CDK4 and CDK6.

Cell-cell interactions influence all aspects of development, homeostasis, and disease. In cancer, interactions between cancer cells and stromal cells play a major role in nearly every step of carcinogenesis. Thus, the ability to record cell-cell interactions would facilitate mechanistic delineation of the role of cancer microenvironment. Here, we describe GFP-based Touching Nexus (G-baToN) which relies upon nanobody-directed fluorescent protein transfer to enable sensitive and specific labeling of cells after cell-cell interactions. G-baToN is a generalizable system that enables physical contact-based labeling between various cell types, including diverse cancer-stromal cell pairs. A suite of orthogonal baToN tools enables reciprocal cell-cell labeling, interaction-dependent cargo transfer, and the identification of higher-order cell-cell interactions across a wide range of cell types. The ability to track physically interacting cells with these simple and sensitive systems will greatly accelerate our understanding of the outputs of cell-cell interactions in cancer as well as across many biological processes.

We present a novel crystallographic screening methodology (MiniFrags) that employs high-concentration aqueous soaks with a chemically diverse and ultra-low-molecular-weight library (heavy atom count 5-7) to identify ligand-binding hot and warm spots on proteins. We propose that MiniFrag screening represents a highly effective method for guiding optimisation of fragment-derived lead compounds or chemical tools and that the high screening hit rates reflect enhanced sampling of chemical space.

The kinase LKB1 is a critical tumor suppressor in sporadic and familial human cancers, yet the mechanisms by which it suppresses tumor growth remain poorly understood. To investigate the tumor-suppressive capacity of four canonical families of LKB1 substrates in vivo, we used CRISPR/Cas9-mediated combinatorial genome editing in a mouse model of oncogenic KRAS-driven lung adenocarcinoma. We demonstrate that members of the SIK family are critical for constraining tumor development. Histologic and gene-expression similarities between LKB1- and SIK-deficient tumors suggest that SIKs and LKB1 operate within the same axis. Furthermore, a gene-expression signature reflecting SIK deficiency is enriched in LKB1-mutant human lung adenocarcinomas and is regulated by LKB1 in human cancer cell lines. Together, these findings reveal a key LKB1–SIK tumor-suppressive axis and underscore the need to redirect efforts to elucidate the mechanisms through which LKB1 mediates tumor suppression. Significance: Uncovering the effectors of frequently altered tumor suppressor genes is critical for understanding the fundamental driving forces of cancer growth. Our identification of the SIK family of kinases as effectors of LKB1-mediated tumor suppression will refocus future mechanistic studies and may lead to new avenues for genotype-specific therapeutic interventions. This article is highlighted in the In This Issue feature, p. 1469

The ability of cancer cells to establish lethal metastatic lesions requires the survival and expansion of single cancer cells at distant sites. The factors controlling the clonal growth ability of individual cancer cells remain poorly understood. Here, we show that high expression of the transcription factor ARNTL2 predicts poor lung adenocarcinoma patient outcome. Arntl2 is required for metastatic ability invivo and clonal growth in cell culture. Arntl2 drives metastatic self-sufficiency by orchestrating the expression of a complex pro-metastatic secretome. We identify Clock as an Arntl2 partner and functionally validate the matricellular protein Smoc2 as a pro-metastatic secreted factor. These findings shed light on the molecular mechanisms that enable single cancer cells to form allochthonous tumors in foreign tissue environments.

We describe here the identification and characterization of 2 novel inhibitors of the fibroblast growth factor receptor (FGFR) family of receptor tyrosine kinases. The compounds exhibit selective inhibition of FGFR over the closely related VEGFR2 receptor in cell lines and in vivo. The pharmacologic profile of these inhibitors was defined using a panel of human tumor cell lines characterized for specific mutations, amplifications, or translocations known to activate one of the four FGFR receptor isoforms. This pharmacology defines a profile for inhibitors that are likely to be of use in clinical settings in disease types where FGFR is shown to play an important role. Mol Cancer Ther; 10(9); 1542–52. ©2011 AACR.

The originally published article failed to acknowledge the equal first authorship contribution of I. P. Winters and C. W. Murray. The article has now been corrected online. The editors apologize for this error.

The RAS-RAF-MEK-ERK signalling cascade is activated through mutations in RAS or RAF in over 30% of cancers. The successful development of inhibitors of BRAF and MEK kinases has led to effective treatment particularly of melanomas whose tumor growth is driven by activating mutations in BRAF such as V600E. Despite these successes, resistance emerges after several months, leading to increased signaling through ERK1/2. This has prompted the development of direct inhibitors of ERK1/2, several of which are in early clinical trials. The majority of clinical ERK1/2 inhibitors are ATP competitive, blocking ERK1/2 catalytic phosphorylation of downstream substrates such as RSK, but do not modulate phosphorylation of ERK1/2 by MEK. Crystal structural studies performed by us and others on the pERK1/2 modulating inhibitor SCH772984 suggested that it induces a conformational change in the glycine-rich loop of ERK2, which leads to Tyr36 becoming tucked under the loop and creating a new binding pocket. We hypothesized that this binding mode might underlie the ability of SCH772984 to block the phosphorylation of ERK1/2, and initiated a fragment-based approach to develop novel, orally bioavailable inhibitors that elicit a similar conformational change and also modulate the phosphorylation of ERK1/2. Using screening methods including high-throughput X-ray crystallography and biophysical assays, we identified fragments binding to both the hinge and the inducible pocket of ERK2. Progressive rounds of structure-guided fragment optimization and growing led to an understanding of inhibitor structure determinants required to induce the conformational change in ERK2. These efforts, together with iterative optimization in a screening cascade including measurement of pRSK and pERK levels and antiproliferative activity in RAS and BRAF mutant cells, led to the discovery of a novel series of pERK modulating ERK1/2 inhibitors. The lead compound shows low nanomolar potency in biochemical ERK1/2 assays and an excellent kinome selectivity profile. In BRAF and RAS mutant cell lines, the lead shows low nanomolar cell proliferation IC50 values, while sparing cell lines not driven by the MAPK pathway. The lead exhibits robust antitumor activity upon oral dosing in a range of subcutaneous xenograft models including the mutant BRAF colorectal line Colo205, providing a promising basis for further optimization towards clinical pERK1/2 modulating ERK1/2 inhibitors. Citation Format: Tom D. Heightman, Valerio Berdini, Hannah Braithwaite, Ildiko Buck, Megan Cassidy, Juan Castro, Aurélie Courtin, James Day, Charlotte East, Lynsey Fazal, Brent Graham, Charlotte Griffiths-Jones, John Lyons, Vanessa Martins, Sandra Muench, Joanne Munck, David Norton, Marc O'Reilly, Nick Palmer, Puja Pathuri, Mike Reader, David Rees, Sharna Rich, Caroline Richardson, Harpreet Saini, Neil Thompson, Nicola Wallis, Hugh Walton, Nicola Wilsher, Alison Woolford, Chris Murray. Fragment-based discovery of a highly potent, orally bioavailable ERK1/2 inhibitor that modulates the phosphorylation and catalytic activity of ERK1/2 [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr B161.

The MAPK pathway is commonly hyper-activated in human cancers due to the occurrence of oncogenic mutations in RAF, RAS and the upregulation of RTKs. The therapeutic potential of MAPK pathway inhibition has been demonstrated by the clinical efficacy of RAF and MEK1/2 (MEK) inhibitors in the treatment of BRAF-mutant melanoma. However, response to such agents is short-lived due to the onset of resistance mechanisms, which in the majority of cases result in the reactivation of ERK1/2 (ERK) signalling. Therefore, the direct targeting of ERK is an attractive therapeutic approach to overcoming the limitations of RAF or MEK inhibitors. Here, we describe a novel, potent, and selective ERK inhibitor, which inhibits both ERK catalytic activity and also the phosphorylation of ERK by MEK. Using fragment-based drug discovery we have developed a selective ERK inhibitor, which inhibits in vitro ERK catalytic activity with a low nM IC50 value. This lead compound has strong antiproliferative effects in a wide range of MAPK-activated cell lines, including the BRAF-mutant cell lines A375 (melanoma) and Colo205 (colorectal), the KRAS-mutant cell lines HCT116 (colorectal), Calu6 (lung) and Panc05.04 (pancreatic), and the NRAS-mutant cell line Ma-mel-27 (melanoma). The lead compound potently inhibits ERK cell signalling. The potent (nM) inhibition of RSK phosphorylation (a direct ERK substrate) was confirmed in A375 (BRAF-mutant melanoma) cells, using MSD analysis. In addition to inhibiting downstream ERK signalling, we demonstrated by ELISA and Western blotting that the lead compound confers a decrease in phospho-ERK levels in both BRAF-mutant and KRAS-mutant cell lines. We investigated the biochemical mechanism of the modulation of ERK phosphorylation in vitro and demonstrated that the compound prevents the phosphorylation of ERK by MEK (at key ERK activation loop residues, T202/Y204), without directly inhibiting MEK activity. The compound was profiled in a range of subcutaneous xenograft models including A375 (BRAF-mutant melanoma) and Calu-6 (KRAS-mutant lung). Once-daily oral dosing of the lead compound conferred significant antitumor activity in a range of in vivo efficacy studies. The compound potently inhibited the phosphorylation of downstream ERK substrates (including RSK) in tumor xenograft tissue. There was a clear relationship between in vivo compound concentrations and the modulation of ERK substrate phosphorylation. Furthermore, as was demonstrated in vitro, we confirmed that in addition to inhibiting ERK catalytic activity the compound potently inhibited the phosphorylation of ERK itself, in both KRAS and BRAF-mutant tumor xenografts. Here, we characterize a novel, highly potent, selective ERK inhibitor, which inhibits both ERK catalytic activity and also the upstream phosphorylation of ERK by MEK. These data support the further optimization of this series of compounds for clinical development. Citation Format: Joanne M. Munck, Valerio Berdini, Luke D. Bevan, Hannah Braithwaite, Ildiko M. Buck, Megan Cassidy, Juan Castro, Aurelie Courtin, James E. Day, Charlotte East, Lynsey Fazal, Brent Graham, Charlotte M. Griffiths-Jones, Tom D. Heightman, Chris J. Hindley, Birikiti Kidane, Justyna Kucia-Tran, John F. Lyons, Vanessa Martins, Sandra Muench, Chris W. Murray, David Norton, Marc O'Reilly, Nick Palmer, Puja Pathuri, Mike Reader, David C. Rees, Sharna J. Rich, Caroline J. Richardson, Harpreet K. Saini, Alpesh Shah, Lukas Stanczuk, Neil T. Thompson, Hugh Walton, Nicola E. Wilsher, Alison J. Woolford, Nicola G. Wallis. Characterization of a novel ERK1/2 inhibitor, which modulates the phosphorylation and catalytic activity of ERK1/2 [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr B154.

Many proteins undergo small side chain or even backbone movements on binding of different ligands into the same protein structure. This is known as induced fit and is potentially problematic for virtual screening of databases against protein targets. In this report we investigate the limits of the rigid protein approximation used by the docking program, GOLD, through cross-docking using protein structures of influenza neuraminidase. Neuraminidase is known to exhibit small but significant induced fit effects on ligand binding. Some neuraminidase crystal structures caused concern due to the bound ligand conformation and GOLD performed poorly on these complexes. A `clean' set, which contained unique, unambiguous complexes, was defined. For this set, the lowest energy structure was correctly docked (i.e. RMSD < 1.5 A away from the crystal reference structure) in 84% of proteins, and the most promiscuous protein (1mwe) was able to dock all 15 ligands accurately including those that normally required an induced fit movement. This is considerably better than the 70% success rate seen with GOLD against general validation sets. Inclusion of specific water molecules involved in water-mediated hydrogen bonds did not significantly improve the docking performance for ligands that formed water-mediated contacts but it did prevent docking of ligands that displaced these waters. Our data supports the use of a single protein structure for virtual screening with GOLD in some applications involving induced fit effects, although care must be taken to identify the protein structure that performs best against a wide variety of ligands. The performance of GOLD was significantly better than the GOLD implementation of ChemScore and the reasons for this are discussed. Overall, GOLD has shown itself to be an extremely good, robust docking program for this system.

When a small molecule binds to a protein, it loses a significant amount of rigid body translational and rotational entropy. Estimates of the associated energy barrier vary widely in the literature yet accurate estimates are important in the interpretation of results from fragment-based drug discovery techniques. This paper describes an analysis that allows the estimation of the rigid body entropy barrier from the increase in binding affinities that results when two fragments of known affinity and known binding mode are joined together. The paper reviews the relatively rare number of examples where good quality data is available. From the analysis of this data, we estimate that the barrier to binding, due to the loss of rigid-body entropy, is 15-20 kJ/mol, i.e. around 3 orders of magnitude in affinity at 298 K. This large barrier explains why it is comparatively rare to observe multiple fragments binding to non-overlapping adjacent sites in enzymes. The barrier is also consistent with medicinal chemistry experience where small changes in the critical binding regions of ligands are often poorly tolerated by enzymes.

This paper describes the application of PRO_LEADS to the flexible docking of ligands into crystallographically derived enzyme structures that are assumed to be rigid. PRO_LEADS uses a Tabu search methodology to perform the flexible search and an empirically derived estimate of the binding affinity to drive the docking process. The paper tests the extent to which the assumption of a rigid enzyme compromises the accuracy of the results. All-pairs docking experiments are performed for three enzymes (thrombin, thermolysin and influenza virus neuraminidase) based on six or more ligand-enzyme crystal structures for each enzyme. In 76% of the cases, PRO_LEADS can successfully identify the correct ligand conformation as the lowest energy configuration when the enzyme structure is derived from that ligand's crystal structure, but the methodology only docks 49% of the cases successfully when the ligand is docked against enzyme crystal structures derived from other ligands. Small movements in the enzyme structure lead to an under-prediction in the energy of the correct binding mode by up to 14 kJ/mol and in some cases this under-prediction can lead to the native mode not being recognised as the lowest energy solution. The type of movements responsible for mis-docking are: the movement of sidechains as a result of changes in C alpha position; the movement of sidechains without changes in C alpha position; the movement of flexible portions of main chains to facilitate the formation of hydrogen bonds; and the movement of metal atoms bound to the enzyme active site. The work illustrates that the assumption of a rigid active site can lead to errors in identification of the correct binding mode and the assessment of binding affinity, even for enzymes which show relatively small shift in atomic positions from one ligand to the next. A good docking code, such as PRO_LEADS, can usually dock successfully if there is induced fit in relatively rigid enzymes but there remains the need to develop improved strategies for dealing with enzyme flexibility. The work implies that treatments of enzyme flexibility which focus only on sidechain rotations will not deal with the critical shifts responsible for mis-docking of ligands in thrombin, thermolysin and neuraminidase. The paper demonstrates the utility of all pairs docking experiments as a method of assessing the effectiveness of docking methodologies in dealing with enzyme flexibility.

This paper tests the performance of a simple empirical scoring function on a set of candidate designs produced by a de novo design package. The scoring function calculates approximate ligand-receptor binding affinities given a putative binding geometry. To our knowledge this is the first substantial test of an empirical scoring function of this type on a set of molecular designs which were then subsequently synthesised and assayed. The performance illustrates that the methods used to construct the scoring function and the reliance on plausible, yet potentially false, binding modes can lead to significant over-prediction of binding affinity in bad cases. This is anticipated on theoretical grounds and provides caveats on the reliance which can be placed when using the scoring function as a screen in the choice of molecular designs. To improve the predictability of the scoring function and to understand experimental results, it is important to perform subsequent Quantitative Structure-Activity Relationship (QSAR) studies. In this paper, Bayesian regression is performed to improve the predictability of the scoring function in the light of the assay results. Bayesian regression provides a rigorous mathematical framework for the incorporation of prior information, in this case information from the original training set, into a regression on the assay results of the candidate molecular designs. The results indicate that Bayesian regression is a useful and practical technique when relevant prior knowledge is available and that the constraints embodied in the prior information can be used to improve the robustness and accuracy of regression models. We believe this to be the first application of Bayesian regression to QSAR analysis in chemistry.

This paper describes a novel methodology, PRO_SELECT, which combines elements of structure-based drug design and combinatorial chemistry to create a new paradigm for accelerated lead discovery. Starting with a synthetically accessible template positioned in the active site of the target of interest, PRO_SELECT employs database searching to generate lists of potential substituents for each substituent position on the template. These substituents are selected on the basis of their being able to couple to the template using known synthetic routes and their possession of the correct functionality to interact with specified residues in the active site. The lists of potential substituents are then screened computationally against the active site using rapid algorithms. An empirical scoring function, correlated to binding free energy, is used to rank the substituents at each position. The highest scoring substituents at each position can then be examined using a variety of techniques and a final selection is made. Combinatorial enumeration of the final lists generates a library of synthetically accessible molecules, which may then be prioritised for synthesis and assay. The results obtained using PRO_SELECT to design thrombin inhibitors are briefly discussed.

This paper describes the implementation and comparison of four heuristic search algorithms (genetic algorithm, evolutionary programming, simulated annealing and tabu search) and a random search procedure for flexible molecular docking. To our knowledge, this is the first application of the tabu search algorithm in this area. The algorithms are compared using a recently described fast molecular recognition potential function and a diverse set of five protein-ligand systems. Statistical analysis of the results indicates that overall the genetic algorithm performs best in terms of the median energy of the solutions located. However, tabu search shows a better performance in terms of locating solutions close to the crystallographic ligand conformation. These results suggest that a hybrid search algorithm may give superior results to any of the algorithms alone.

This paper describes the development of a simple empirical scoring function designed to estimate the free energy of binding for a protein-ligand complex when the 3D structure of the complex is known or can be approximated. The function uses simple contact terms to estimate lipophilic and metal-ligand binding contributions, a simple explicit form for hydrogen bonds and a term which penalises flexibility. The coefficients of each term are obtained using a regression based on 82 ligand-receptor complexes for which the binding affinity is known. The function reproduces the binding affinity of the complexes with a cross-validated error of 8.68 kJ/mol. Tests on internal consistency indicate that the coefficients obtained are stable to changes in the composition of the training set. The function is also tested on two test sets containing a further 20 and 10 complexes, respectively. The deficiencies of this type of function are discussed and it is compared to approaches by other workers.

Two new computational tools, PRO_PHARMEX and PRO_SCOPE, for use in active-site-directed searching of 3D databases are described. PRO_PHARMEX is a flexible, graphics-based program facilitating the extraction of pharmacophores from the active site of a target macromolecule. These pharmacophores can then be used to search a variety of databases for novel lead compounds. Such searches can often generate many 'hits' of varying quality. To aid the user in setting priorities for purchase, synthesis or testing, PRO_SCOPE can be used to dock molecules rapidly back into the active site and to assign them a score using an empirical scoring function correlated to the free energy of binding. To illustrate how these tools can add value to existing 3D database software, their use in the design of novel thrombin inhibitors is described.

In some instances, peptides can play an important role in the discovery of lead compounds. This paper describes the peptide design facility of the de novo drug design package, PRO_LIGAND. The package provides a unified framework for the design of peptides that are similar or complementary to a specified target. The approach uses single amino acid residues, selected from preconstructed libraries of different residues and conformations, and places them on top of predefined target interaction sites. This approach is a well-tested methodology for the design of organics but has not been used for peptides before. Peptides represent a difficulty because of their great conformational flexibility and a study of the advantages and disadvantages of this simple approach is an important step in the development of design tools. After a description of our general approach, a more detailed discussion of its adaptation to peptides is given. The method is then applied to the design of peptide-based inhibitors to HIV-1 protease and the design of structural mimics of the surface region of lysozyme. The results are encouraging and point the way towards further development of interaction site-based approaches for peptide design.

Recently, the development of computer programs which permit the de novo design of molecular structures satisfying a set of steric and chemical constraints has become a burgeoning area of research and many operational systems have been reported in the literature. Experience with PRO-LIGAND-the de novo design methodology embodied in our in-house molecular design and simulation system PRO-METHEUS-has suggested that the addition of a genetic algorithm (GA) structure refinement procedure can 'add value' to an already useful tool. Starting with the set of designed molecules as an initial population, the GA can combine features from both high- and low-scoring structures and, over a number of generations, produce individuals of better score than any of the starting structures. This paper describes how we have implemented such a procedure and demonstrates its efficacy in improving two sets of molecules generated by different de novo design projects.

Many proteins undergo small side chain or even backbone movements on binding of different ligands into the same protein structure. This is known as induced fit and is potentially problematic for virtual screening of databases against protein targets. In this report we investigate the limits of the rigid protein approximation used by the docking program, GOLD, through cross-docking using protein structures of influenza neuraminidase. Neuraminidase is known to exhibit small but significant induced fit effects on ligand binding. Some neuraminidase crystal structures caused concern due to the bound ligand conformation and GOLD performed poorly on these complexes. A 'clean' set, which contained unique, unambiguous complexes, was defined. For this set, the lowest energy structure was correctly docked (i.e. RMSD1.5 A away from the crystal reference structure) in 84% of proteins, and the most promiscuous protein (1mwe) was able to dock all 15 ligands accurately including those that normally required an induced fit movement. This is considerably better than the 70% success rate seen with GOLD against general validation sets. Inclusion of specific water molecules involved in water-mediated hydrogen bonds did not significantly improve the docking performance for ligands that formed water-mediated contacts but it did prevent docking of ligands that displaced these waters. Our data supports the use of a single protein structure for virtual screening with GOLD in some applications involving induced fit effects, although care must be taken to identify the protein structure that performs best against a wide variety of ligands. The performance of GOLD was significantly better than the GOLD implementation of ChemScore and the reasons for this are discussed. Overall, GOLD has shown itself to be an extremely good, robust docking program for this system.

When a small molecule binds to a protein, it loses a significant amount of rigid body translational and rotational entropy. Estimates of the associated energy barrier vary widely in the literature yet accurate estimates are important in the interpretation of results from fragment-based drug discovery techniques. This paper describes an analysis that allows the estimation of the rigid body entropy barrier from the increase in binding affinities that results when two fragments of known affinity and known binding mode are joined together. The paper reviews the relatively rare number of examples where good quality data is available. From the analysis of this data, we estimate that the barrier to binding, due to the loss of rigid-body entropy, is 15-20 kJ/mol, i.e. around 3 orders of magnitude in affinity at 298 K. This large barrier explains why it is comparatively rare to observe multiple fragments binding to non-overlapping adjacent sites in enzymes. The barrier is also consistent with medicinal chemistry experience where small changes in the critical binding regions of ligands are often poorly tolerated by enzymes.

This paper describes the further development of the functionality of our in-house de novo design program, PRO_LIGAND. In particular, attention is focussed on the implementation and validation of the ‘directed tweak’ method for the construction of conformationally flexible molecules, such as peptides, from molecular fragments. This flexible fitting method is compared to the original method based on libraries of prestored conformations for each fragment. It is shown that the directed tweak method produces results of comparable quality, with significant time savings. By removing the need to generate a set of representative conformers for any new library fragment, the flexible fitting method increases the speed and simplicity with which new fragments can be included in a fragment library and also reduces the disk space required for library storage. A further improvement to the molecular construction process within PRO_LIGAND is the inclusion of a constrained minimisation procedure which relaxes fragments onto the design model and can be used to reject highly strained structures during the structure generation phase. This relaxation is shown to be very useful in simple test cases, but restricts diversity for more realistic examples. The advantages and disadvantages of these additions to the PRO_LIGAND methodology are illustrated by three examples: similar design to an alpha helix region of dihydrofolate reductase, complementary design to the active site of HIV-1 protease and similar design to an epitope region of lysozyme.

An approach to de novo molecular design, PRO-LIGAND, has been developed that, in the environment of a large, integrated molecular design and simulation system, provides a unified framework for the generation of novel molecules which are either similar or complementary to a specified target. The approach is based on a methodology that has proved to be effective in other studies--placing molecular fragments upon target interaction sites-but incorporates many novel features such as the use of a rapid graph-theoretical algorithm for fragment placing, a generalised driver for structure generation which offers a large variety of fragment assembly strategies to the user and the pre-screening of library fragments. After a detailed description of the relevant modules of the package, PRO-LIGAND's efficacy in aiding rational drug design is demonstrated by its ability to design mimics of methotrexate and potential inhibitors for dihydrofolate reductase and HIV-1 protease.