MAP4K4 is a novel MAPK/ERK pathway regulator required for lung adenocarcinoma maintenance

About 76% of patients with lung adenocarcinoma harbor activating mutations in the receptor tyrosine kinase (RTK)/RAS/RAF pathways, leading to aberrant activation of the mitogen‐activated protein kinase (MAPK) pathways particularly the MAPK/ERK pathway. However, many lung adenocarcinomas lacking these genomic mutations also display significant MAPK pathway activation, suggesting that additional MAPK pathway alterations remain undetected. This study has identified serine/threonine kinase mitogen‐activated protein 4 kinase 4 (MAP4K4) as a novel positive regulator of MAPK/ERK signaling in lung adenocarcinoma. The results showed that MAP4K4 was drastically elevated in lung adenocarcinoma independently of KRAS or EGFR mutation status. Knockdown of MAP4K4 inhibited proliferation, anchorage‐independent growth and migration of lung adenocarcinoma cells, and also inhibited human lung adenocarcinoma xenograft growth and metastasis. Mechanistically, we found that MAP4K4 activated ERK through inhibiting protein phosphatase 2 activity. Our results further showed that downregulation of MAP4K4 prevented ERK reactivation in EGFR inhibitor erlotinib‐treated lung adenocarcinoma cells. Together, our findings identify MAP4K4 as a novel MAPK/ERK pathway regulator in lung adenocarcinoma that is required for lung adenocarcinoma maintenance.


Introduction
Lung cancer is the most common cause of cancerrelated mortality, and the most common form of lung cancer in the United States among both men and women is adenocarcinoma. The most common somatic aberrations in lung adenocarcinoma are recurrent mutations in the RTK [including epidermal growth factor receptor (EGFR)]/RAS/RAF pathways, which occurred in 76% of cases (Cancer Genome Atlas Research, 2014). RAF serine/threonine kinase phosphorylates MEK to activate MEK, and activated MEK then phosphorylates ERK, leading to its activation (Shaul and Seger, 2007;Wellbrock et al., 2004). Because of its pivotal role in cancer genesis and maintenance, the EGFR/RAS/RAF/MEK/ERK signaling cascade has been the subject of intensive research and pharmaceutical scrutiny to identify effective target-based therapy for cancer treatment (Dhillon et al., 2007;Roberts and Der, 2007;Samatar and Poulikakos, 2014). However, the full potential of this signaling pathway as effective targets for cancer therapy has not been realized and many challenges and hurdles remain.
For example, recent clinical trial result showed that the response rate of non-small-cell lung cancer patients to vemurafenib (BRAF inhibitor) was 42% (Hyman et al., 2015), suggesting that not every BRAF mutant non-small-cell lung cancer patients responded to BRAF-targeted therapy; RAS oncogene is still considered 'undruggable', with no effective RAS inhibitors in clinical use; in EGFR-targeted therapy, although initial response rates are high (71.2-83%) among patients with EGFR mutations, all patients eventually develop resistance (Chong and Janne, 2013). The most common mechanism of resistance to EGFR inhibitors is a secondary mutation in EGFR itself (T790M, threonine-to-methionine amino acid change at position 790), which renders the activated kinase insensitive to these agents (Kobayashi et al., 2005;Yu et al., 2013).
Together, there is an urgent and unmet need for novel therapeutic approaches to effectively block EGFR/RAS/RAF/MEK/ERK signaling. New insights into the regulation of mitogen-activated protein kinase (MAPK)/ERK signaling pathway could reveal new therapeutic point of intervention, therefore are greatly needed. Given that many lung adenocarcinomas lacking EGFR/RAS/RAF mutations also displayed significant MAPK activation (Cancer Genome Atlas Research, 2014), it is reasonable to predict that there are additional, still undetected MAPK pathway regulators.
The serine/threonine kinase MAP4K4 is a member of the Ste20p (sterile 20 protein) family. MAP4Ks were discovered in 1995 as a key kinase in the mating pathway in Saccharomyces cerevisiae (Wu et al., 1995). MAP4K4 is best known as an inflammation-related kinase: MAP4K4 promotes vascular inflammation (Roth Flach et al., 2015) and siRNA targeting macrophage MAP4K4 suppresses systemic inflammation (Aouadi et al., 2009). Although emerging evidence suggests that MAP4K4 is involved in many aspects of cell functions and processes, information regarding MAP4K4 in different types of cancer is very limited. Results from one previous study showed that MAP4K4 protein level was elevated in lung adenocarcinoma and its elevation was negatively associated with patients' prognosis (Qiu et al., 2012). But except for the correlative observations, whether and how MAP4K4 contributes to lung adenocarcinoma remain to be determined. The current study has identified MAP4K4 as a novel activator of MAPK/ERK signaling in lung adenocarcinoma. The results from functional in vitro and in vivo studies showed that MAP4K4 was required for the maintenance of the malignant phenotype of lung adenocarcinoma. Our findings suggest that pharmacological inhibition of MAP4K4 could be an effective approach to targeting lung adenocarcinoma, as a stand-alone therapy or in combination with other treatment.

Plasmid, transient transfection, lentivirus production, and infection
The expression plasmid of HA-MAP4K4 was constructed by inserting human MAP4K4 sequence into pcDNA3.1-HA. The expression plasmid of constitutively active ERK2 (plasmid #40819) was purchased from Addgene (Cambridge, MA, USA). Plasmid transfections were performed with Polyjet In Vitro DNA Transfection Reagent (SignaGen Laboratories, Rockville, MD, USA) according to the manufacturer's instruction. Cell lines stably overexpressing MAP4K4 were established by G418 selection. Lentiviral-based shRNAs targeting MAP4K4 were purchased from Sigma Aldrich (St. Louis, MO, USA). Lentiviruses encoding the shRNA against MAP4K4 or control-shRNA were produced in HEK293T cells transfected with the lentiviral vector expressing MAP4K4-shRNA or control-shRNA using the third-generation packaging systems (Addgene). The viral particle-containing media was filtered through syringe filters and subsequently used to infect target cells. Cell lines stably expressing shRNA were established by puromycin selection.

Tissue microarray and immunohistochemistry staining
Human lung adenocarcinoma tissue microarray (TMA) containing 44 cases of lung adenocarcinoma, 44 matched adjacent normal lung tissues, and three cases of normal lung tissues was purchased from US Biomax Inc. (Derwood, MD, USA) TMA slides containing 136 cases of human lung adenocarcinoma with KRAS mutation, EGFR mutation or KRAS, and EGFR wild-type were prepared as described previously (Villaruz et al., 2016). All TMA slides were formalinfixed and paraffin-embedded. Immunohistochemistry (IHC) was performed as described previously (Gao et al., 2015). Briefly, paraffin sections were deparaffinized and rehydrated. Antigen retrieval was performed in pressure cooker for 15 min in 20 mM Tris/ EDTA buffer (pH 9.0). The sections were incubated overnight with MAP4K4 antibody (Biorbyt, San Francisco, CA, USA) in a humidified chamber at room temperature. After washed with PBS, sections were incubated with HRP-labeled anti-rabbit secondary antibody (DAKO Envision+ system, Carpinteria, CA, USA) for 1 h at room temperature. Color visualization was performed with liquid DAB chromogen in imidazole/HCL buffer (pH 7.5) containing hydrogen peroxide until the brown color fully developed. The sections were counterstained with hematoxylin and coverslipped with permanent mounting media. The intensity of TMA staining was scored as 0 (negative), 1 (weak), 2 (moderate), and 3 (strong).

Cell proliferation assay
Cells were seeded in 96-well plates at a density of 3000 cells per well and incubated at 37°C for indicated time. Ten microlitre MTT (Sigma Aldrich) was then added into each well. Medium containing MTT was replaced with 100 lL per well DMSO after incubation at 37°C for 4 h. The absorbance was measured at 490 nm with iMark TM Microplate Absorbance Reader (Bio-Rad, Hercules, CA, USA). Three independent experiments were performed in triplicate. COMPUSYN software (http://www.combosyn.com), which is based on the Median-Effect Principle (Chou) and the Combination Index-Isobologram Theorem (Chou-Talalay) (Chou, 2010), was employed to calculate the combination index (CI) values, which were used to determine the drug combination effect: CI < 0.75 indicates synergism; CI = 0.75À1.25 indicates additive effects; and CI > 1.25 indicates antagonism.

In vitro cell invasion assay
Twelve-well falcon permeable supports with 8-lm pores and Matrigel were purchased from Corning Life Sciences (Lowell, MA, USA). The inserts were coated with 300 lgÁmL À1 Matrigel diluted with cold FBS-free RPMI-1640 medium, then placed into 12-well plates on ice. The Matrigel was allowed to solidify at 37°C for 3 h. After 1 9 10 5 indicated cells suspended in 500 lL FBS-free RPMI-1640 medium were seeded in the insert chambers, 1 mL RPMI-1640 with 10% FBS was added into the wells. Cells on the upper surface of the inserts were gently wiped away with PBS-saturated cotton swabs after incubation at 37°C for 16-20 h. Cells on the lower surface of the insert were fixed and stained with crystal violet solution (0.05% crystal violet, 1% formaldehyde, and 1% methanol). The stained cells were photographed with a Leica DFC420 C camera connected to a Leica DMI300 B microscope. Three independent experiments were performed in duplicate.

Wound healing assay
Twenty-four-well wound healing inserts were purchased from Cell Biolabs, Inc. (San Diego, CA, USA), and used according to the manufacturer's protocol. Briefly, inserts were placed in the 24-well plates before seeding 5 9 10 5 indicated cells in each well. The inserts were carefully removed after incubation at 37°C for 24 h. The cells were rinsed with culture medium and then incubated with 10 lM of mitomycin C (Cayman Chemical) at 37°C for 2 h. Cells were photographed 0, 24, and 48 h after mitomycin C treatment with a Leica DFC420 C camera connected to a Leica DMI300 B microscope. Percentage of wound area was measured with IMAGEJ (https://imagej.net/Downloads for free). Three independent experiments were performed in triplicate.

Anchorage-independent growth assay
The cell lines were trypsinized and 3 9 10 4 cells were suspended in DMEM containing 0.4% low melting point (LMP) agarose (Promega, Madison, WI, USA). Two milliliter of the above cell suspension was plated over 3 mL bottom agar containing 0.8% LMP agarose in a 60-mm dish. Two independent experiments were performed in quadruplicate. The dishes were incubated at 37°C with medium replenished every two days until colony formed. The colonies were stained with 0.01% crystal violet solution. Representative pictures were taken under stereomicroscope (Olympus SZX10, Center Valley, PA, USA).
2.9. PP2A phosphatase activity assay PP2A phosphatase activity was determined using the PP2A immunoprecipitation phosphatase assay kit (Millipore, Billerica, MA, USA) according to the manufacturer's protocol. Briefly, lysates containing 200 lg protein were incubated with 4 lg anti-PP2A and 40 lL protein A/agarose slurry in Ser/Thr assay buffer for 2 h at 4°C with constant rocking. After washed with Ser/Thr assay buffer for three times, immunoprecipitated PP2A was incubated with 750 lM phosphopeptide (K-R-pT-I-R-R) for 10 min at 30°C in a shaking incubator. Twenty-five microlitre of the reaction mixture was transferred into three wells of the 96well microtiter plate, and the plate was then incubated with 100 lL of malachite green phosphate detection solution for 15 min at room temperature. Free phosphate was quantified by measuring the absorbance at 650 nm with iMark TM Microplate Absorbance Reader (Bio-Rad).

RAS activity assay
RAS activity was determined using the RAS activation assay kit (Millipore) according to the manufacturer's protocol. Briefly, the cell lysates were gently rotated with RAF-1 RBD (RAS binding domain) agarose, which binds to active, GTP-bound RAS, at 4°C for 3 h. The precipitates were then subjected to SDS/ PAGE and immunoblotted with anti-RAS antibody.

Animal experiments
All animal procedures were carried out according to the protocols approved by the Institutional Animal Care and Use Committee at the University of Pittsburgh. Four-to five-week-old nude mice (Charles River) housed in a sterile environment were used for the experiments. For xenograft experiments, 19 10 6 H1975 shRNA control cells or MAP4K4-knockdown cells were subcutaneously injected into the left or right flanks of the mice, respectively. Eight mice were used. Tumor size was determined by caliper measurements twice a week. Tumor volume was calculated using the formula: V = ½ 9 a 9 b 2 , where a and b denote the largest and smallest axis of the tumor, respectively. Mice were euthanized 28 days after implantation. Tumors were excised, followed by cryogenic grinding with liquid nitrogen to prepare whole-cell lysates for immunoblotting with indicated antibodies.
For experimental metastasis, 1 9 10 6 H1975 shRNA control cells or MAP4K4-knockdown cells were injected into the tail vein of each mouse, six mice in each group. Mice were euthanized 12 weeks after injection. Mice lungs were fixed with intratracheal instillation of 10% formalin as described previously (Braber et al., 2010). Briefly, the caudal vena cava of the animals was cut to prevent the flow of blood into the bases of the lungs. A cannula was inserted into the trachea and fixed with a ligature; 10% formalin was gently injected through the cannula to inflate the lungs. Lungs and heart were removed en bloc and placed in a glass vial filled with 10% formalin. Twenty-four hours later, 10% formalin was replaced by 70% ethanol. Representative pictures of the lungs were taken. All lungs were embedded by paraffin, and hematoxylin and eosin (H&E) staining was performed. The number of surface lung tumors was counted lobe by lobe while excising the lungs. H&Estained slides were scanned by Aperio Digital Pathology Slide Scanner (Leica Biosystems Inc., Buffalo Grove, IL, USA), and tumor diameters were measured with ImageScope.

Statistics
Data were presented as mean AE SD. The difference between two groups was evaluated with Student's ttest (two-tailed). The difference in the distribution of positively stained TMA samples among three score groups was evaluated with chi-square test. P values less than 0.05 were considered statistically significant.

MAP4K4 elevation in lung adenocarcinoma is independent of KRAS or EGFR mutation status
We performed IHC analysis of MAP4K4 on human lung adenocarcinoma TMAs containing 44 cases of cancer tissues on a slide. Consistent with prior study (Qiu et al., 2012), we found that protein expressions of MAP4K4 were nearly undetectable in normal lung tissues and tumor-adjacent normal tissues but were drastically increased in tumor tissues. Strikingly, nearly 100% of the tumor cells were stained positive for MAP4K4. Representative IHC staining of MAP4K4 is shown in Fig. 1A, and score distributions are summarized in Fig. 1B. Of note, the specificity of anti-MAP4K4 for IHC was verified by western blot analysis. We then tested whether MAP4K4 elevation was associated with KRAS or EGFR mutation. The TMA slides, obtained from the University of Pittsburgh Lung Specialized Program of Research Excellence (SPORE) (Villaruz et al., 2016) and constructed using randomly selected archival lung adenocarcinomas with known EGFR and KRAS mutation status, were stained for MAP4K4. As shown in Fig. 1C, the rate of MAP4K4 overexpression did not appear to vary  To determine whether MAP4K4 is functionally involved in lung adenocarcinoma cells, we performed a variety of in vitro assays to evaluate the impact of shRNA knockdown of MAP4K4 on cell functions including proliferation, anchorage-independent growth, migration, and invasion. Three cell lines H23 (KRASmutated), H1975 (EGFR-mutated), and H1793 (wildtype KRAS and EGFR) were used as cellular models.
The results of MTT assays showed that downregulation of MAP4K4 significantly inhibited cell proliferation regardless of KRAS or EGFR mutation status ( Fig. 2A). Furthermore, MAP4K4 knockdown suppressed EGF-induced growth of H1793 cells (Fig. S1). Downregulation of MAP4K4 also substantially inhibited anchorage-independent growth of lung adenocarcinoma cells (Fig. 2B). Results from wound healing assay and cell invasion assay further showed that MAP4K4 deficiency jeopardized the abilities of lung adenocarcinoma cells to migrate and to invade (Fig. 2C,D). Next, we examined the effect of MAP4K4-specific small-molecule inhibitor, PF-06260933, on lung adenocarcinoma cell functions. As shown in Fig. S2, treatment of H1975 cells with PF-06260933 significantly suppressed cell proliferation, anchorage-independent growth, cell migration, and cell invasion. Together, the above results from in vitro cell function analyses suggest that MAP4K4 could be a previously unrealized key player in the regulation of cell functions of lung adenocarcinoma.

MAP4K4 activates the MAPK/ERK pathway in lung adenocarcinoma cells
Identification of downstream signaling mediators of MAP4K4 is crucial for understanding the mechanisms underlying MAP4K4 regulation of lung adenocarcinoma cells. It has been shown that MAP4K4 can induce ERK, JNK, and p38 activation in TNF-ainduced signaling (Bouzakri and Zierath, 2007;Bouzakri et al., 2009;Tesz et al., 2007). Whether these kinases function as MAP4K4 downstream effector in lung adenocarcinoma remains to be determined. To gain initial clues, we compared cellular MAP4K4 levels to the activation status of MAPK/ERK1/2, MEK1/2, MAKP/JNK1/2, MAPK/p38 as well as the serine/threonine protein kinase AKT in lung adenocarcinoma cell lines, evaluated by immunoblotting using specific antibodies against phosphorylated ERK, MEK, JNK, p38, or AKT. The results showed that high cellular levels of MAP4K4 appeared to coincide with high levels of phosphorylated ERK1/2, but not phosphorylation of MEK1/2, JNK, p38, or AKT (Fig. 3A), suggesting that ERK is likely downstream signaling mediator of MAP4K4. We next took genetic manipulation strategies-shRNA-knockdown and overexpression approachesto further validate ERK as MAP4K4 downstream signaling effectors. As shown in Fig. 3B, in cell lines expressing relatively high MAP4K4, shRNA downregulation of MAP4K4 greatly reduced the level of phosphorylated ERK1/2 without changing phosphorylation of MEK and RAS activity (Fig. 3B), indicating that ERK1/2 is the downstream signaling mediator, and regulation of ERK1/2 by MAP4K4 is independent of RAS/MEK/ERK axis. Consistent with this, overexpression of MAP4K4 in cell lines expressing relatively low MAP4K4 substantially increased the level of phosphorylation ERK1/2 without changing MEK phosphorylation and RAS activity (Fig. 3C). Functionally, overexpression of MAP4K4 in these cell lines could also promote cell proliferation (Fig. S3). To further confirm that ERK is a downstream effector of MAP4K4 signaling in lung adenocarcinoma cells, we transfected MAP4K4-knockdown cells with plasmid expressing constitutively active ERK2 L73P/S151D , and the results showed that overexpression of active ERK2 antagonized MAP4K4 downregulation-mediated inhibition of cell functions (Fig. 3D-F). Together, these results identified MAP4K4 as a novel activator of MAPK/ERK pathway in lung adenocarcinoma cells.
To further test that MAP4K4 bypasses MEK to activate ERK, we tested whether overexpression of MAP4K4 prevents MEK inhibitor-mediated inhibition of ERK. The results showed that ectopic expression of MAP4K4 in A549 cells partially prevented MEK inhibitor trametinib-mediated inhibition of ERK phosphorylation (Fig. S4A), supporting that MAP4K4 activation could bypass the effect of MEK inhibitor. Consistent with this, we found that concurrent inhibition of MAP4K4 and MEK with specific inhibitors exerted synergistic inhibitory effect on H23 cell proliferation (Fig. S4B).
The most common mechanism for acquired resistance to EGFR inhibitors such as erlotinib is the mutation of threonine 790 to a methionine (T790M) in the kinase domain of the EGFR gene, which results in increased kinase affinity for ATP, thus decreasing the sensitivity to ATP-competitive inhibitors (Yun et al., 2008). The current study showed that in T790Mmutated EGFR-TKI-resistant cell line H1975, downregulation of MAP4K4 inhibited ERK activation in the presence and absence of erlotinib treatment (Fig. 3G), implying that MAP4K4 bypasses EGFR to promote ERK activation; that is, MAP4K4 functioned in parallel with EGFR to regulate ERK activation, which in turn suggests that suppressing ERK reactivation through targeting MAP4K4 in lung adenocarcinoma could be a novel strategy to overcome EGFR-TKI resistance.

MAP4K4 activates ERK1/2 through inhibiting PP2A activity
We examined how MAP4K4 regulates ERK activation. As shown in Fig. 4A, shRNA knockdown of MAP4K4 did not inhibit MEK activation (represented as phosphorylated MEK), thus excluding the possibility that MAPK induces ERK activation through MEK. MKP3 (MAP kinase phosphatase 3), also known as dual specificity phosphatase 6 (DUSP6), is the ERK-specific MAP kinase phosphatase (Muda et al., 1996(Muda et al., , 1998. Our results show that downregulation of MAP4K4 did not change MKP3 level (Fig. 4A), suggesting that MAP4K4 unlikely regulates ERK activation by modulating MKP3. Phosphatase PP2A is a confirmed tumor suppressor (Sangodkar et al., 2016), and it was reported to dephosphorylate both MEK and ERK (Bae and Ceryak, 2009;Letourneux et al., 2006;Millward et al., 1999;Tay et al., 2012). Knockdown of MAP4K4 did not change the protein levels of the PP2A catalytic C subunit and regulatory B56b subunit (Fig. 4A). However, as shown in Fig. 4B, PP2A activity was upregulated in MAP4K4-knockdown cells and downregulated in MAP4K4-overexpressing cells. ERK activation can be rescued by PP2A inhibitor OA in MAP4K4-knockdown cells and suppressed by PP2A activator FTY720 in MAP4K4-overexpressing cells ( Fig. 4C). Together, the above results indicate that MAP4K4 regulates ERK activation through modulating PP2A phosphatase activity.

Downregulation of MAP4K4 inhibits in vivo xenograft growth and metastasis of lung adenocarcinoma cells accompanied by suppressed ERK activation
We further conducted in vivo experiments to validate the functional importance of MAP4K4 in lung adenocarcinoma cells. In xenograft experiments, H1975 shRNA control cells or MAP4K4-knockdown cells were subcutaneously injected into the left or right flanks of the mice, respectively. As shown in Fig. 5A, knockdown of MAP4K4 in H1975 cells caused significant reduction in tumor size. This result indicated that MAP4K4 is required for in vivo xenograft growth of H1975 cells. Immunoblotting analysis of H1975 xenograft tumors revealed that ERK activation was inhibited in MAP4K4-knockdown cells (Fig. 5B). In experimental metastasis study, H1975 shRNA control cells or MAP4K4-knockdown cells were injected into the tail vein of each mouse. Results of this assay indicated that knockdown of MAP4K4 greatly impaired the ability of H1975 cells to metastasize to lung (Fig. 5C-F). Together, these findings provided first evidence that MAP4K4 regulates lung adenocarcinoma cell functions in vivo, presumably through inducing ERK activation.

Discussion
To dramatically change the outlook for patients with lung adenocarcinoma, it is essential to identify target that may be exploitable in relapsed EGFR-mutant tumors as well as in lung adenocarcinomas lacking mutations in druggable targets. To this end, the current study has identified MAP4K4 as a previously unrealized ERK pathway activator and it is required for the maintenance of malignant phenotype of lung adenocarcinoma in vitro and in vivo. Our results indicate that MAP4K4 elevation in lung adenocarcinoma is independent of somatic mutations of KRAS or EGFR. Importantly, we find that MAP4K4 regulation of MAPK/ERK pathway is also independent of KRAS or EGFR mutation status, suggesting that MAP4K4targeted therapy could be an effective treatment for a broad range of patients with lung adenocarcinoma.
Therapies targeted toward the genetic features of a tumor can be highly effective; therefore, cancer-relevant mutations/cancer driver genes are considered the best candidates for the development of targeting drugs. However, recent findings suggest that somatically mutated genes are not necessarily the most critical genes for tumor maintenance (Bossi et al., 2016). Consistent with this, we have identified MAP4K4 as an import protein required for lung adenocarcinoma maintenance. This finding might have important clinical implications: It suggests MAP4K4-targeted therapy as a novel treatment option for the RAS-mutated lung adenocarcinoma patients, which currently has few treatment options (Ahearn et al., 2012).
MAP4K4 activates MAPK/JNK through TAK1 (a MAP3K) and MKK4/MKK7 (MAP2Ks) in HEK293T cells (Yao et al., 1999). In contrast, the current study indicates that MAP4K4 bypasses the three-tiered kinase canonical cascade (RAF/MEK/ERK) to activate MAPK/ERK by regulating PP2A activity. Regardless of the detailed underlying mechanism, it appears that downstream signaling selection by MAP4K4 and how MAP4K4 activates its target are cell type-, cancer type-, or context-specific.
Several MAP4K4-specific small-molecule inhibitors have been available now (Gao et al., 2016). However, the potent antitumor properties of these inhibitors remain to be determined. Our novel findings that PF-06260933 significantly suppressed lung cancer cell functions and had synergistic inhibitory effect with MEK inhibitor on lung cancer cell proliferation suggest that MAP4K4-specific inhibition could be a promising approach to treating lung adenocarcinoma.

Conclusions
The current study has identified MAP4K4 as a novel MAPK/ERK activator. Targeting MAP4K4 could be a promising alternative mean of ERK pathway inhibition and MAP4K4 coinhibition could prevent or circumvent resistance to vertical inhibition of EGFR/ RAS/RAF/MEK pathway in a broad range of patients with lung adenocarcinoma.

Supporting information
Additional Supporting Information may be found online in the supporting information tab for this article: Fig. S1. MAP4K4 knockdown suppresses EGFinduced lung adenocarcinoma cells growth. Fig. S2. MAP4K4 inhibitor suppresses lung adenocarcinoma cell functions. Fig. S3. MAP4K4 overexpression promotes proliferation of lung adenocarcinoma cells. Fig. S4. Concurrent inhibition of MAP4K4 and MEK exerts synergistic effect on lung adenocarcinoma cells.