FAM86A methylation of eEF2 links mRNA translation elongation to tumorigenesis

Summary

eEF2 post-translational modifications (PTMs) can profoundly affect mRNA translation dynamics. However, the physiologic function of eEF2K525 trimethylation (eEF2K525me3), a PTM catalyzed by the enzyme FAM86A, is unknown. Here, we find that FAM86A methylation of eEF2 regulates nascent elongation to promote protein synthesis and lung adenocarcinoma (LUAD) pathogenesis. The principal physiologic substrate of FAM86A is eEF2, with K525me3 modeled to facilitate productive eEF2-ribosome engagement during translocation. FAM86A depletion in LUAD cells causes 80S monosome accumulation and mRNA translation inhibition. FAM86A is overexpressed in LUAD and eEF2K525me3 levels increase through advancing LUAD disease stages. FAM86A knockdown attenuates LUAD cell proliferation and suppression of the FAM86A-eEF2K525me3 axis inhibits cancer cell and patient-derived LUAD xenograft growth in vivo. Finally, FAM86A ablation strongly attenuates tumor growth and extends survival in KRAS^G12C-driven LUAD mouse models. Thus, our work uncovers an eEF2 methylation-mediated mRNA translation elongation regulatory node and nominates FAM86A as an etiologic agent in LUAD.

Graphical Abstract

eTOC blurb:

FAM86A trimethylates the essential mRNA translation elongation factor eEF2 at K525 (eEF2Km525me3), a PTM of unknown function. Francis et. al. show that FAM86A generation of eEF2K525me3 regulates the mRNA translation elongation step to promote global protein synthesis and lung cancer pathogenesis in vivo, including in a KRAS.G12C-driven GEM tumor model.

Introduction

Protein lysine methylation is the addition of one, two, or three methyl moieties to the ε-nitrogen of a lysine side chain, forming mono-, di-, and tri-methylated derivatives^1–3. Lysine methylation is a common post-translational modification in humans, generated by dozens of lysine methyltransferases (KMTs) encoded by the human genome. Human KMTs, via their catalytic activities, regulate diverse biological processes ranging from chromatin biology to signal transduction^4,5. Accordingly, the dysregulation of many KMTs has been linked to human cancers and other diseases^1,3. One main biological process regulated by lysine methylation is mRNA translation, wherein many KMTs target protein components of the translation machinery^4,6,7. Here we focus on the human seven-beta-strand (7βS) KMT FAM86A (also named eEF2KMT) that methylates eukaryotic translation elongation factor 2 (eEF2) at lysine 525 (K525)^8–10.

The GTPase eEF2 is a ubiquitously expressed and essential protein that orchestrates tRNA translocation within the 80S ribosome during mRNA translation elongation^11–14. eEF2, in addition to its N-terminal GTP-binding domain (Domain I), harbors four additional small C-terminal structural domains, hereto referred to as Domains II-V (Figure S1A). eEF2-GTP binds at the A site in elongating ribosomes to facilitate tRNA translocation from the hybrid P/E and A/P states to the E and P sites, respectively, which vacates the A site to allow accommodation of another amino-acyl tRNA^11,14,15. eEF2’s GTPase activity provides mechanical energy that facilitates translocation, though GTP hydrolysis per se can be uncoupled from successful translocation events^12,13,16,17. In addition to promoting translocation efficiency, as codon-anticodon interactions are disrupted and then reformed during translocation, eEF2 also impacts mRNA translation fidelity by preventing frameshifting. While eEF2 is essential for cell and organismal viability, mutations in eEF2 are linked to neurodevelopmental disorders and eEF2 is frequently upregulated in cancer^18–20.

Several PTMs are implicated in regulating eEF2 function. For example, biosynthesis of the unique modified amino acid diphthamide (DPH) on eEF2 at residue 715 (eEF2H715DPH) (Figure S1B) is generated by a cascade involving seven distinct enzymes¹¹. This PTM is linked to maintenance of translocation accuracy²¹, and while DPH synthesis is not essential in yeast or in human cancer cell lines, eEF2H715DPH is required for mammalian development¹¹. Besides DPH, eEF2 is phosphorylated by EF2K at T56 (eEF2T56ph) (Figure S1B), which inhibits eEF2-mediated elongation and influences in a context-dependent manner cancer cell behaviors^22–26. In contrast to DPH and T56ph, virtually nothing is known about the biological function of K525me3, a PTM located in eEF2 Domain III that helps synchronize ribosome binding with GTP hydrolysis (Figure S1B)^15,16.

FAM86A is the human homolog of the S. cerevisiae protein Efm3, which in yeast generates high stoichiometry trimethylation of eEF2K509⁸. FAM86A was previously shown to methylate human eEF2 in vitro^8,9. Recently, we reported that extensive intermolecular interaction between human FAM86A and eEF2, including the FAM86 motif that is independent of the FAM86A catalytic domain, promotes efficient catalysis on eEF2K525 and is required for eEF2K525me3 generation in the HEK293T human cell line¹⁰. efm3 yeast knockout strains exhibited only mild functional phenotypes and no overall growth phenotype, suggesting that eEF2 methylation by Efm3 is generally dispensable in yeast⁸. However, to date no functional investigation of FAM86A has been reported in human cells or tissues.

Here we demonstrate that eEF2K525me3 is the principal physiologic substrate of FAM86A across diverse human cell lines (beyond HEK293T cells) and tissue samples. We show that loss of eEF2K525me3 in lung cancer cell lines inhibits global protein synthesis and provide evidence that eEF2 methylation regulates mRNA translation output by promoting translocation at a nascent elongation step. We find that FAM86A, eEF2, and eEF2K525me3 levels are upregulated in several human cancer cell lines including LUAD lines and FAM86A depletion with loss of eEF2K525me3 inhibits proliferation of LUAD cell lines. Further, we show that FAM86A knockdown inhibits human lung cancer cell and patient-derived xenograft (PDX) tumor growth, and Fam86a knockout dramatically suppresses tumorigenesis and significantly prolongs survival in a mouse model of KRAS^G12C-driven LUAD. Thus, our work uncovers a pivotal physiologic role for eEF2 methylation in the regulation of mRNA translation and identifies this activity as a potential targetable vulnerability to treat lung cancer.

Results

eEF2K525 methylation is a principal physiologic activity of FAM86A

FAM86A was previously shown to methylate eEF2K525 in vitro⁸ and we recently demonstrated that FAM86A knockdown in HEK293T cells depleted endogenous eEF2K525me3¹⁰. Here, we observe FAM86A dependency for eEF2K525 trimethylation in six additional human cell types comprising one diploid and five cancer cell lines (Figure 1A). As FAM86A is ubiquitously expressed in human cells and tissues (Figure S1C–D), our results are consistent with broad FAM86A regulation of eEF2 methylation. We note that the levels of total eEF2 protein are not impacted by FAM86A depletion (Figure 1A).

Figure 1. Evidence that the principal physiologic activity of FAM86A is methylation of eEF2K525.

A. FAM86A is required for eEF2K525me3 generation in multiple human cell lines. Western blots of whole cell extracts (WCE) from the indicated cell lines expressing CRISPR-Cas9 and two independent sgRNAs targeting FAM86A or a control sgRNA.

B. In vitro methylation assays with recombinant GST-FAM86A and FLAG-eEF2 as substrate. Top panel, ³H-SAM was used as methyl donor and the methylation was visualized by autoradiography. Bottom panel, Coomassie stain of proteins present in the reactions.

C. In vitro methylation assays as in (B) with recombinant wild-type GST-FAM86A or catalytically dead GST-FAM86A_WDA as indicated.

D. Reconstitution of eEF2K525me3 by FAM86A in HEK293T cells. Western blots of WCEs from control- or FAM86A-depleted HEK293T cells complemented with the indicated vector. E. Western blots of the indicated HEK293T WCEs and antibodies.

F. In vitro methylation assay as in (B) with 3μg of recombinant GST-FAM86A and 25μg of indicated WCEs as in (E).

G. In vitro methylation assay as in (F). eEF2 methylation detected by Western blot with eEF2K525me3 antibody and a single band the size of eEF2 by autoradiography.

See also Figure S1

Consistent with previous work^8,10, recombinant FAM86A methylates recombinant eEF2 in vitro in the presence of ³H radiolabeled methyl-donor S-adenosylmethionine (³H-SAM) (Figure 1B; S1E). Substitution of eEF2K525 to an arginine (eEF2_K522R) abolishes FAM86A methylation activity (Figure 1B). Based on FAM86A structure¹⁰ and sequence homology with other 7βS KMTs (Figure S1F–G), we generated a putative FAM86A catalytic mutant harboring W139A and D188A substitutions (hereafter FAM86A_WDA) and found that recombinant wild-type FAM86A, but not FAM86A_WDA, methylated recombinant eEF2 in vitro (Figure 1C). Complementation of FAM86A-depleted HEK293T cells with CRISPR-resistant wild-type, but not catalytically dead FAM86A_WDA, reconstituted eEF2K525me3 levels (Figure 1D). Thus, FAM86A, in a catalytic activity-dependent manner, is required for maintenance of physiologic levels of eEF2K525me3 in cells.

To test if FAM86A has additional substrates besides eEF2, a CRISPR-guided biochemical approach was taken. We used recombinant FAM86A to perform in vitro methylation assays with ³H-SAM on extracts from control or FAM86A-depleted HEK293T cells as substrate (Figure 1E–F). A single prominent band the size of eEF2 (~95kDa) was detected only in the reaction using FAM86A-depleted extract as substrate, with no methylated protein observed in reactions with the control extract (Figure 1F). Consistent with previous findings⁸, we postulated that endogenous eEF2K525 is methylated at nearly saturating stoichiometry in control cells that express FAM86A and thus cannot be further modified in control extracts. Accordingly, using either recombinant wild-type or catalytically dead mutant FAM86A to perform in vitro methylation assays on extracts from FAM86A-depleted HEK293T cells as substrate, the only band specifically detected in wild-type FAM86A versus FAM86A_WDA reactions was the 95kDa band (Figure 1G). Moreover, eEF2K525me3 was specifically detected by an eEF2K525me3 antibody only in reactions with wild-type FAM86A but not with FAM86A_WDA (Figure 1G). FAM86A also does not methylate the elongation factor eEF1A, histones and several other proteins in vitro (data not shown). Collectively, these data suggest that the principal physiologic catalytic activity of FAM86A is eEF2K525 methylation.

FAM86A methylation of eEF2 regulates mRNA translation

eEF2 is an indispensable translation elongation factor that fundamentally regulates protein synthesis¹¹. However, whether FAM86A-catalyzed methylation of eEF2 impacts mRNA translation and protein synthesis is unknown. Notably, FAM86A knockdown in two independent LUAD cell lines (NCIH2122 and A549) strongly reduced global protein synthesis as monitored by detection of puromycin-labeled newly synthesized polypeptides²⁷ (Figure 2A). This effect was observed in cells growing in serum-rich media, following overnight serum starvation, and was most pronounced following re-stimulation of starved cells with serum-rich media wherein protein synthesis is activated (Figure 2A). We observed no FAM86A-dependent change in the eEF2-deactivating modification T56ph (Figure S2A). To measure protein synthesis by a method independent of puromycylation, cells were pulsed with the methionine analog L-azidohomoalanine (AHA), with AHA incorporation into newly synthesized proteins detected with streptavidin by first clicking the AHA azide group to biotin-alkyne²⁸. In these assays, FAM86-depleted NCIH2122 and A549 cells showed decreased protein synthesis, mirroring the results obtained in puromycylation experiments (Figure 2B).

Figure 2. FAM86A methylation of eEF2 regulates protein synthesis.

A. FAM86A knockdown decreases protein synthesis in LUAD cells. Western blots of WCEs following puromycylation assays in H2122 and A549 cells grown in serum-rich media, serum starved, or serum re-stimulation following starvation as indicated. Asterisks: non-specific band; Arrow: indicates FAM86A.

B. AHA labeling under baseline conditions in serum-rich media shows decrease in protein synthesis upon depletion of FAM86A in H2122 and A549 cells as in (A). WCE from the indicated cell lines probed with the indicated antibodies. Asterisks: non-specific band; Arrow: indicates FAM86A.

C. Polysome profile of cytosolic extracts from either control- or FAM86A-depleted H2122 cells fractionated on 10%–50% sucrose gradients and absorbance measured at 254nm. Representative profile is shown for three biological replicates.

D. Quantification of 80S peak area relative to combined polysome peak area. n=3 biological replicates. n.s. not significant; two-tailed unpaired Student’s t test. Data represented as mean±SEM.

E. Timeline schematic of HHT assay.

F. Western blots of H2122 WCEs following puromycylation assay as in (A) with HHT as described in (D).

G. Overall position of eEF2 in the 80S human ribosome A site and eEF2-rpS3 interaction. Structure shown as a cartoon model; eEF2 and rpS23 shown as spheres. eEF2 domain I: dark blue; Domains II–V: light blue. eEF2K525: red. RpS23: yellow. Approximate lateral positions of E, P and A sites and approximate 40S and 60S subunit boundaries are labeled. PDB: 6Z6N.

H. Zoom-in stick view of eEF2K525 position near the acidic motif spanning rpS23 residues E95, E96, N97, and D98. PDB: 6Z6N.

I. Zoom-in electrostatic view of eEF2-K525 interaction with rpS23 similar to (G) comparing fit of unmodified (K525me0, left) versus methylated (K525me3, right) eEF2 in the rpS23 acidic pocket. Adapted from PDB: 6Z6N.

See also Figure S2

To gain insight into the mechanism behind the protein synthesis phenotype, we performed polysome profiling experiments on extracts from control and FAM86A-depleted H2122 LUAD cells. An accumulation of 80S ribosomes and a modest decrease in levels of individual 40S and 60S subunits was observed in FAM86A-depleted cells relative to control cells (Figure 2C–D; S2B). Surprisingly, given the inherent role of eEF2 in elongation, polysome levels were the same irrespective of FAM86A expression (Figure 2C; S2B). Similar polysome profiling results were observed in NCIH358 cells, a third LUAD cell line that also showed reduced protein synthesis upon FAM86A knockdown (Figure S2C–E); in our hands, A549 cells were not amenable to polysome profiling). Together, these data suggest that eEF2K525me3 promotes protein synthesis via a mechanism that may act during the initial phase of elongation.

Next, puromycylation assays were performed in cells treated with homoharringtonine (HHT), a small molecule protein synthesis inhibitor postulated to block early elongation steps and therefore used to specifically monitor mRNA translation occurring only on ribosomes already engaged in elongation^29,30 (Figure 2E). In these experiments, no difference was detected in puromycin labeling between control and FAM86A-depleted samples (Figure 2F), consistent with the polysome profiling experiments showing no FAM86A-dependent change in polysome population. These findings suggest a model in which eEF2K525me3 functions at a nascent elongation step, prior to the repetitive elongation cycle exhibited by active polysomes that is downstream of HHT blockade.

In a published cryo-EM structure of eEF2-bound 80S human ribosome³¹, eEF2 is positioned at the A-site with its N-terminal GTPase domain exposed to the solvent and the C-terminal domains II-V interacting with the ribosome (Figure 2G). In this structure, eEF2 forms several intermolecular interactions, including with the 28S rRNA, 18S rRNA, the small ribosomal proteins rpS30, rpS27a, rpS15, and rpS23, and the large ribosomal proteins rpL9 and rpL23. Notably, K525me3 is at the center of the interaction between eEF2 and a four-residue acidic motif in rpS23 (Figure 2H). This interaction is present in multiple independent published structures of the eEF2–80S complex, and the published electron density map of this structure is of sufficient resolution to detect that a trimethyl moiety on eEF2K525 favorably disperses the positive charge of eEF2K525 over the large, negatively charged rpS23 surface (Figure 2I, S2F–G)^32,33. We hypothesized that the K525me3-rpS23 interactions promote proper engagement of eEF2 in the 80S ribosome to thereby modulate eEF2 activity. To test this idea, eEF2 was purified from control or FAM86A-depleted cells (hence ±K525me3 (Figure S2H)) and eEF2-dependent in vitro GTPase hydrolysis was measured in the absence or presence of purified ribosomes, which stimulate eEF2’s GTPase activity³⁴. While the intrinsic GTPase activity of isolated eEF2 was unaffected by K525 methylation status, incubation with ribosomes stimulated eEF2 GTPase activity, with unmethylated eEF2 exhibiting two-fold greater GTPase activity than methylated eEF2 (Figure S2I). We considered that the enhanced GTPase activity of unmethylated eEF2 might come at a cost of diminished translocation fidelity, however, FAM86A depletion did not cause a difference in a translation elongation frameshifting assay (Figure S2J). Collectively, these observations suggest a model in which K525 methylation by FAM86A supports favorable eEF2-ribosome interactions that promote protein synthesis by helping couple GTP turnover with tRNA translocation during the first step of elongation (see Discussion).

FAM86A-eEF2K525me3 axis promotes Kras-driven LUAD tumorigenesis in vivo

eEF2 is frequently upregulated in cancer^18,19. Like eEF2, FAM86A is overexpressed in diverse cancers, including in ~70% of LUAD patient samples (Figure 3A, S3A). Further, eEF2K525me3 immunohistochemistry (IHC) signal is far stronger in human LUAD tissue compared to normal human lung tissue (Figure 3B) and eEF2K525me3 signal increases through advancing LUAD disease stages (Figure 3C). Moreover, the levels of FAM86A, eEF2, and eEF2-K525me3 are elevated in five distinct cancer cell lines as compared to non-transformed IMR-90 and WI38 normal diploid cell lines (Figure 3D). These observations suggest a potential role for the FAM86A-eEF2 methylation axis in influencing cancer cell phenotypes. Indeed, The Cancer Dependency Map (DepMap) project found that proliferation of >1000 human cancer cell lines, including all 53 LUAD cell lines in DepMAP, was inhibited by FAM86A depletion (Figure 3E). Specifically, the DepMap Chronos pipeline calculated a median FAM86A gene effect score of −0.86 in all cell lines and −0.90 in LUAD, values indicative of high confidence results (Figure 3E). For comparison, gene effect scores for an essential protein like eEF2 were naturally far more severe than for FAM86A (Figure S3B), whereas gene effect scores for enzymes that generate H715DPH and T56ph on eEF2 were far less severe compared to FAM86A (Figure S3C). We note that FAM86A depletion in IMR90 fibroblasts modestly decreased protein synthesis but did not impact cell proliferation (Figure S3D–E). Overall, these data support the notion that the K525me3 PTM represents an important mode of eEF2 regulation.

Figure 3. FAM86A-eEF2K525me3 axis is upregulated in LUAD and promotes LUAD cell and xenograft growth in vivo.

A. Categorical (top) and heatmap (bottom) representations of FAM86A mRNA expression in LUAD TCGA dataset. mRNA high: z score>2 over normal tissue.

B. Representative eEF2K525me3 IHC images of patient samples from normal lung tissue (n=6) and LUAD at advancing stages (n=70), scale bars: 50μm.

C. IHC chromogen stain intensity quantification shows increasing eEF2K525me3 levels in advancing stages of LUAD. P-values determined by two-way ANOVA with Tukey’s testing for multiple comparisons. Line: median.

D. Western blots with indicated antibodies of WCEs from human transformed (A549, H2122, T3M4, HT1080, and U2OS) and non-transformed (IMR90 and WI38) cell lines.

E. Density plot of Chronos gene effect scores for FAM86A in all Cancer Cell Line Encyclopedia lines (n=1095) and all LUAD cell lines (n=53) in DepMap. Black dashed line indicates value x=0.

F. Proliferation of control- versus FAM86A-depleted H2122 cells (Figure 1A for Western blots). n=3 biological replicates. p<0.01 for sgFAM86A-1/2 compared to sgControl; n.s. for sgFAM86A-1:sgFAM86A-2; two-tailed unpaired Student’s t test. Data are represented as mean±SEM.

G. Tumor volume quantification for A549 xenografts treated as indicated in NSG mice (n=5 mice per group). P-values were determined by two-way ANOVA. Data are represented as mean±SEM.

H. Western blots with indicated antibodies of indicated WCEs. Asterisks: non-specific band.

I. FAM86A catalytic activity is required for H2122 xenograft tumor growth. Tumor volume quantification as in (G) with cell lines as in (H). P-values determined by two-way ANOVA with Tukey’s testing for multiple comparisons. Data represented as mean±SEM.

J. Western blots with the indicated antibodies of WCEs from LUAD PDX (carrying mutations in Kras^G12C, p53^R273C, KMT2C^M3463*, MSH6 ^K1358Dfs*2) modified ex vivo to express sgRNA-FAM86A or sgRNA-control.

K. Tumor volume quantification for LUAD-PDX xenografts as in (J) in NSG mice (n=5 mice per treatment group). P-values determined by two-way ANOVA. Data represented as mean±SEM.

See also Figure S3

We next directly tested FAM86A-dependent phenotypes in cancer cells. FAM86A knockdown inhibited proliferation of H2122, A549 and H358 cells, three KRAS-mutant LUAD lines (Figure 3F; S3F–G), consistent with a potential role for FAM86A/eEF2K525me3 in LUAD neoplastic growth. Indeed, FAM86A depletion inhibited A549 and H2122 xenograft tumor growth (Figure 3G–I). Moreover, xenograft tumor growth of FAM86A-depleted H2122 cells was restored by complementation with CRISPR-resistant wild-type FAM86A, whereas catalytically inactive FAM86A_WDA failed to restore tumor growth (Figure 3H–I). Finally, FAM86A-knockdown in primary human LUAD PDX samples resulted in depletion of eEF2K525me3 levels and xenograft growth inhibition (Figures 3J–K). Together, these data suggest that FAM86A, via eEF2K525me3, promotes LUAD tumorigenesis.

To further investigate the in vivo role of eEF2K525me3 in cancer pathogenesis, we obtained a Fam86a knockout strain (Fam86a^−/−) from the international mouse phenotyping consortium³⁵. Notwithstanding the functions of FAM86A in the regulation of protein synthesis, proliferation, and xenograft growth in lung cancer cells, Fam86a^−/− mice are born at normal mendelian ratios, viable into adulthood, fertile, and show only a few minor phenotypes (Table S1). For example, sections from tissues normally associated with cancer drug toxicity concerns (intestine, liver, kidney, and heart) from FAM86A knockout mice showed normal histology, and proliferation and apoptosis rates that were indistinguishable from control mice (Figure S4A–L). We next crossed Fam86a^−/− mice with a KRAS^G12C-driven LUAD mouse model in which a conditional Kras^G12C mutant allele was designed to maintain normal Kras expression from the endogenous locus (Figure S4M). Briefly, a partial wild-type Kras cDNA encoding exons 2–4 (including the poly A signal), flanked by loxP sites, was inserted into intron 1 of the mutant Kras^G12C allele. The resulting animal model expresses wild-type Kras at physiologic levels driven by the endogenous promoter and does not display any discernable phenotype. Upon Cre-mediated recombination, the inserted cDNA sequence is removed, and the mutant Kras^G12C is expressed (Figure S4M). We interbred the Kras^G12C mutant mice with p53^loxP/loxP mutant strains to generate Kras;p53 mice to better recapitulate the natural history of aggressive disease seen in patients, including activation of MAPK (phospho-ERK1/2) and expression of a marker characteristic for LUAD (NKX2.2), but lacking a squamous cell lung carcinoma marker (Desmoglein 3) (Figure S4M–N). LUAD was induced by intratracheal lavage of adenovirus expressing the Cre recombinase (Ad-Cre) in adult Kras;p53 and Kras;p53;Fam86a (hereto named KP and KPF, respectively) mice (Figure 4A)³⁶. KP mutant mice developed widespread adenocarcinoma at 24 weeks after Ad-Cre infection, which was clearly visible at the whole organ level and by histopathology (Figure 4B). In KPF mice, FAM86A knockout resulted in loss of FAM86A and eEF2K525me3 signal in lung tissue lysates (Figure 4C). Moreover, tumor development was dramatically attenuated in KPF mutant mice compared to KP mice based on gross pathology, quantification of tumor number and burden, and analyses of cell proliferation and apoptosis (Figure 4B, D–G). In this aggressive malignancy model, the median survival of KP mice is 178 days with no mice surviving to 200 days (Figure 4H). In contrast, the majority of the KPF cohort (8 out of 10) survived the full duration of the study (past 300 days) with no signs of tumors (Figure 4H). For the two deceased KPF mice, they survived substantially longer than the longest surviving mice in the KP cohort (Figure 4H) and had far smaller tumors at the time of death relative to the massive high-grade tumors seen for animals in the control group (see Figure 4B), suggesting that lethality was a consequence of tumor growth in a physiologically vulnerable location (Figure 4H; data not shown). Together these in vivo data support a key enabling function for FAM86A in oncogenic KRAS-driven LUAD tumorigenesis.

Figure 4. Fam86a ablation inhibits LUAD tumorigenesis in vivo.

A. Experimental design to assess effects of Fam86a ablation on lung cancer pathogenesis in Kras;p53 model.

B. Representative macroscopic pathology, HE-stained sections and IHC with indicated antibodies of lungs from Kras;p53 (KP) and Kras;p53;Fam86a (KPF) mutant mice at 24 weeks post Ad-Cre induction (representative of n=6 mice per experimental group), scale bars: 2mm (macro) and 50μm (histology).

C. Western blots with indicated antibodies of representative from KP and KPF lung biopsy lysates at 24 weeks post Ad-Cre induction. Actin shown as loading control.

D - G. Quantification of tumor number, tumor burden, proliferation (pH3⁺ positive cells), and apoptosis (cleaved Caspase 3⁺ positive cells) in KP and KPF samples as in (B). Whiskers: 25th to 75th percentile, center line: median; n=6 mice for each group. P-values determined by two-tailed unpaired t-test.

H. Kaplan-Meier survival curves of KP (n=8) and KPF (n=10) mutant mice. P-values determined by log-rank test for significance.

See also Figure S4 and Table S1

Discussion

There is a growing appreciation that lysine methylation influences epigenetic regulation of gene expression at the level of mRNA translation and protein synthesis^1,6. In this study, we have provided evidence that cellular protein synthesis is strongly impacted by FAM86A-mediated methylation of eEF2K525. Based on knockdown and reconstitution experiments performed in independent cell lines, including primary lung cancer patient samples, as well as mouse tissue knockouts, we propose that like the yeast homologue Efm3, FAM86A is the sole enzyme responsible for generating endogenous eEF2K525me3. This notion is accordant with the exquisite selectivity of several other 7βS KMT family members such as METTl3^6,37,38. We further propose given i) FAM86A does not methylate histones, nucleosomes, eEF1A, ribosomes, or several other proteins we have tested directly (data not shown), ii) the only protein detected in the biochemical screen using whole cell extracts to be methylated by FAM86A is eEF2 (see Figure 1), and iii) the highly selective mode of binding between FAM86A and eEF2¹⁰, that the physiologic and pathologic functions of FAM86A reported here are specifically mediated by eEF2K525 methylation. However, we cannot rule out the possibility that FAM86A has an additional substrate/s that influence mRNA translation and/or cancer cell behaviors.

In contrast to H715DPH and T56ph, two well-characterized eEF2 PTMs with important regulatory functions, methylation at K525 promotes global mRNA translation output. Considering this observation and given the fundamental role of eEF2 in mediating tRNA translocation during elongation, it was unexpected for K525me3 loss to not cause an appreciable alteration in polysome populations. Moreover, polysome runoff experiments implicate K525me3 increasing mRNA translation output through a mechanism that occurs at a step prior to the actions of HHT. At the same time, the ribosome-dependent GTP hydrolysis activity of eEF2 is elevated, not decreased, upon loss of methylation. While the GTPase activity of eEF2 is generally synchronized with tRNA translocation, under certain conditions it can be decoupled from productive tRNA translocation^16,17. In this light, by leveraging available elegant structures of eEF2-engaged ribosomes, we postulate that stabilization of eEF2-rpS23 interactions by eEF2K525 methylation may help secure a favorable eEF2-ribosome engagement to synchronize GTP hydrolysis with productive tRNA translocation. We further speculate that this mechanism normally acts as a potential quality control checkpoint at the initial translocation step of elongation when the E site is not yet occupied by a tRNA. Future studies utilizing biophysical and structural approaches will be crucial for obtaining insights into the detailed mechanisms by which methylation of eEF2K525 regulates tRNA translocation fundamentals and elongating ribosomes^39,40.

Protein synthesis is among the most energy-intensive processes in cells, and its dysregulation by various mechanisms is a common factor in neoplastic diseases^41–43. Indeed, there is increasing evidence that fast-growing malignancies, like those driven by RAS-signaling, develop mechanisms to upregulate the mRNA translation machinery to meet their demand for increased protein synthesis. Thus, identifying clinically actionable molecular mechanisms such as methylation that upregulate protein synthesis activity in tumors may uncover opportunities to target specific cancer vulnerabilities with minimal on-target toxicity. Here, we have provided evidence for FAM86A-mediated methylation of eEF2 to serve as such a mechanism. Potentially acting at an early stage in elongation, eEF2K525me3 enhances mRNA translation to maintain robust protein synthesis in LUAD cell lines. Our data suggests that this activity promotes LUAD tumorigenesis and raises the possibility that cancers may become dependent on the increased activity of the FAM86A-eEF2-K525me3 pathway to drive malignant growth. As Fam86a knockout mice develop normally and are viable with only minor phenotypes, and key tissues such as heart, intestine, kidney, and liver show normal histology (Figures S4A–L and Table S1), the increase in protein synthesis driven by eEF2K525me3 appears to be dispensable for non-diseased tissue in vivo. These observations combined with the broad dependency of the 1095 cancer cell lines for FAM86A provide the rationale for efforts to develop FAM86A inhibitors. It may also be interesting to explore potential crosstalk between targetable mechanisms that increase mRNA translation capacity, for example upregulation of eIF4E and METT13-mediated eEF1A methylation^37,44. In this vein, loss of the tumor suppressor ARID1A has revealed a tumor vulnerability to eEF2 inhibition³⁰, raising the notion that the substantial population of ARID1A mutant tumors may be rendered particularly susceptible to FAM86A inhibition. Overall, we uncover a mechanism mediated by FAM86A-catalyzed methylation of eEF2 that potentially regulates elongation at an early stage to increase protein synthesis output.

Limitations of study

Several experiments aimed at developing mechanistic understanding into eEF2K525me3’s mode of action used puromycin and HHT as small molecule inhibitors of mRNA translation. However, certain aspects of the mechanism by which these drugs impact protein synthesis, particularly HHT, remain opaque. Thus, caution must be taken when interpreting results using these drugs. To address this issue for puromycin, we used AHA labeling as an orthogonal assay to assess protein synthesis. Future testing of hypotheses derived in part from negative results using HHT will need to include diverse strategies and consider alternative underlying mechanisms related to how eEF2K525 methylation may impact the elongation step of mRNA translation. Finally, while we previously modeled the FAM86A-eEF2 interaction in silico¹⁰, obtaining a physical crystal structure of the complex may further uncover the molecular basis for the selectivity of FAM86A for eEF2K525 versus other potential substrates.

STAR ★ METHODS

RESOURCE AVAILABILITY

Lead Contact

• Further information and requests for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Or Gozani (ogozani@stanford.edu).

Materials Availability

• Plasmids, antibody and cell lines generated in this study will be available from the lead contact upon request with a completed material transfer agreement.

• Mouse line generated in this study will be shared by the lead contact upon request with a completed material transfer agreement.

Data and Code Availability

• Original western blot images have been deposited at Mendeley and are publicly available as of the date of publication. The DOI is listed in the key resources table.

• This study did not generate any unpublished code, software, or algorithm. All utilized codes are publicly available as of the date of publication. DOIs are listed in the key resources table.

• Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.

KEY RESOURCES TABLE

REAGENT OR RESOURCE	SOURCE	IDENTIFIER
Antibodies
Biotin Micro Beads	Miltenyi Biotec	Cat# 130-042-401
CD45-Biotin	eBiosciences	Cat# 13-0451-81
CD31-Biotin	eBiosciences	Cat# 13-0319-80
Ter119-Biotin	eBiosciences	Cat# 13-5921-81
Cleaved Caspase 3	Cell Signaling Technologies	Cat# 9664
ERK1/2	Cell Signaling Technologies	Cat# 4695
phospho-ERK1/2	Cell Signaling Technologies	Cat# 4376
phospho-H3	Cell Signaling Technologies	Cat# 9701
Ki67	BD Bioscience	Cat# 550609
RPS6	Santa Cruz Biotechnology	Cat# sc-74459
RPL40	Abcam	Cat# ab109230
Peroxidase-conjugated streptavidin	Jackson ImmunoResearch	Cat# 016-030-084
Desmoglein 3	ThermoFisher	Cat# 32-6300
NKX2.2	Abcam	Cat# ab191077
FLAG	Sigma-Aldrich	Cat# F1804
GAPDH	Cell Signaling Technology	Cat# 2118
Tubulin	Millipore	Cat# 05-661
Actin	Cell Signaling Technologies	Cat# 4970
eEF2	Abcam	Cat# ab75748
eEF2K525me3	ABclonal Biotechnology	N/A
FAM86A	GeneMed Synthesis Inc	N/A
IRDye® 680RD Donkey anti-Rabbit IgG Secondary Antibody	LI-COR	Cat# 926-68073
IRDye® 800CW Donkey anti-Mouse IgG Secondary Antibody	LI-COR	Cat# 926-32212
Peroxidase AffiniPure Donkey Anti-Mouse IgG (H+L)	Jackson ImmunoResearch	Cat# 715-035-151
Peroxidase AffiniPure Donkey Anti-Rabbit IgG (H+L)	Jackson ImmunoResearch	Cat# 711-035-152
Bacterial and Virus Strains
BL21	Thermo Fisher Scientific	Cat# C6070-03
BL21-RIL	Agilent Technologies	Cat# 230240
DH5α	Thermo Fisher Scientific	Cat# K4520-1
Ad5-CMV-Cre	Baylor College of Medicine, Viral Vector Production Core	Cat# Ad5-CMV-Cre RRID:SCR_015037
Biological Samples
Human LUAD Tissue Array	MD Anderson Pathology	N/A
Chemicals, Peptides, and Recombinant Proteins
RPMI 1640 Medium	Corning	Cat# MT10040CV
DMEM Medium	Corning	Cat# MT10017CV
Fetal bovine serum	Thermo Fisher Scientific	Cat# 10500056
PBS	Corning	Cat# MT21031CV
HBSS	Thermo Fisher Scientific	Cat# 14025076
Trypsin-EDTA 0.25%	Corning	Cat# MT25053CI
Puromycin	Thermo Fisher Scientific	Cat# A1113802
Hygromycin B	Corning	Cat# 30240CR
Blasticidin	Thermo Fisher Scientific	Cat# A1113903
G418 Sulfate	Corning	Cat# MT30234CI
Complete Protease Inhibitor Cocktail	Sigma-Aldrich	Cat# 4693159001
Phosphatase Inhibitor Cocktail	Thermo Fisher Scientific	Cat# 78420
Bovine Serum Albumin (BSA)	Thermo Fisher Scientific	Cat# BP9703100
L-Reduced glutathione	Sigma-Aldrich	Cat# G4251-25G
S-adenosyl-methionine	New England Biolabs	Cat# B9003S
S-Adenosyl-l-[methyl-3H] methionine	American Radiolabeled Chemicals	Cat# ART0288
3X FLAG peptide	Sigma	Cat# F4799
IPTG	Fisher	Cat# I560000-25
l-azidohomoalanine (AHA)	Click Chemistry T ools	Cat# 1066-100
Biotin-PEG4-Alkyne	Click Chemistry T ools	Cat# TA105-25
Glutathione Sepharose 4B	Sigma-Aldrich	Cat# GE17-0756-01
Coomassie Plus Assay	Thermo Fisher Scientific	Cat# 23236
Coomassie GelCode Blue	Thermo Fisher Scientific	Cat# 24590
GTP solution	Thermo Fisher Scientific	Cat# R1461
Matrigel	Corning	Cat# 354248
TransIT-293	Mirus Bio	Cat# MIR-2706
Polybrene	Sigma-Aldrich	Cat# TR-1003-G
NP-40	Sigma-Aldrich	Cat# I8896
Phenylmethylsulfonyl fluoride (PMSF)	Sigma-Aldrich	Cat# P7626
DMSO	Sigma-Aldrich	Cat# D5879
16% Formaldehyde (w/v)	Sigma-Aldrich	Cat# F8775
PVDF membrane (0.2 μm)	BioRad	Cat# 1620177
PVDF membrane (0.45 μm)	Millipore	Cat# IPVH00010
Critical Commercial Assays
ATPase/GTPase Activity kit	Sigma	Cat# MAK1113
Dual-Glo Luciferase Assay System	Promega	Cat# E2920
QIAprep Spin Miniprep Columns	Qiagen	Cat# 27115
ZymoPURE Plasmid Miniprep Kit	Zymo	Cat# D4211
ZymoPURE II Plasmid Maxiprep Kit	Zymo	Cat# D4203
DNA PCR Purification Kit	Qiagen	Cat# 28106
PrestoBlue™ Cell Viability Reagent	ThermoFisher	Cat# A13261
Click Chemistry Reaction Buffer Kit	Click Chemistry T ools	Cat# 1001
DAB Substrate Kit	Abcam	Cat# ab64238
Vectastain ABC kit	Vector Laboratories	Cat# PK-6100
BCA Protein Assay Kit	Pierce	Cat# 23227
ECL Substrate	Amersham	Cat# RPN2106
PCR Mycoplasma Test Kit I/C	PromoKine	Cat# PK-CA91-1096
Deposited Data
Original Blot Images Data	This study	Mendeley Data DOI: 10.17632/c48sj885jr.1
Experimental Models: Cell Lines
Human: A549	ATCC	Cat# CCL-185
Human: NCI-H2122	ATCC	Cat# CRL-5985
Human: NCI-H358	ATCC	Cat# CRL-5807
Human: T3M4	Riken	Cat# RCB1021
Human: U2OS	ATCC	Cat# CRL- HTB-96
Human: HT1080	ATCC	Cat# CCL-121
Human: IMR90	ATCC	Cat# CCL-186
Human: WI-38	ATCC	Cat# CCL-75
Human: 293T	ATCC	Cat# CRL-3216
Experimental Models: Organisms/Strains
Mouse: KrascKI-G12C	This paper	N/A
Mouse: p53LoxP/LoxP	(Jonkers, et al., 2001)45	Strain# JAX 008462
Mouse: NOD.SCID-IL2Rg−/− (NSG)	The Jackson Laboratories	Strain# 005557
Oligonucleotides
sgRNA non-targeting (control) 5’-GGGCT ACTAGGATT CAATCT-3’	(Morgens, et al., 2017)46	N/A
sgRNA FAM86A-1 human 5’-AGCACGGCCAT CAT CTCCT A-3’	This paper	N/A
sgRNA FAM86A-2 human 5’-CACCGGGCATATTTGACGGA-3’	This paper	N/A
Recombinant DNA
Plasmid: pLentiCRISPRv2	Feng Zhang Lab	Cat# Addgene #52961
Plasmid: pLentiCRISPRv2 hygro	Gift from Brett Stringer	Cat# Addgene #98291
Plasmid: pCMV-dR8.2 dvpr	Bob Weinberg Lab	Cat# Addgene #8455
Plasmid: pCMV-VSV-G	Bob Weinberg Lab	Cat# Addgene #8454
Plasmid: pGEX-6P-1	GE Healthcare	Cat# 28-9546-48
Plasmid: pENTR3C	Thermo Fisher Scientific	Cat# A10465
Plasmid: pLenti6.2 V5-DEST	Thermo Fisher Scientific	Cat# V36820
Plasmid: pcDNA3.1(+)	Thermo Fisher Scientific	Cat# V7020
Plasmid: pSGDlucV3.0	Addgene	Cat# 119760
Software and Algorithms
Prism 7	GraphPad	https://www.graphpad.com/; RRID:SCR_002798
Excel for Mac 2016	Microsoft	https://www.microsoft.com/en-us/; RRID:SCR_016137
PreciPoint M8 ViewPoint	PreciPoint	http://www.precipoint.com/microsco py-software/viewpoint/
ImageJ – Fiji package	Freeware	http://fiji.sc; RRID:SCR_002285
Clustal Omega	European Bioinformatics Institute	http://www.ebi.ac.uk/Tools/msa/clustalo/; RRID:SCR_001591
Excel for Mac	Microsoft	https://www.microsoft.com/ RRID:SCR_016137
PyMOL	Schrodinger	http://www.pymol.org/ RRID:SCR_000305
RStudio	Posit, PBC	https://posit.co/ RRID:SCR_000432
Geneious Prime	Biomatters	http://www.geneious.com/ RRID:SCR_010519
Image Studio	LI-COR Biosciences	https://www.licor.com/bio/image-studio/ RRID:SCR_015795
Adobe Illustrator	Adobe, Inc	http://www.adobe.com/products/illustrator; RRID:SCR_010279

EXPERIMENTAL MODEL AND STUDY PARTICIPANT DETAILS

Animal models

Kras^cKI-G12C/+ conditional knock-in strain was generated using the targeting vector containing endogenous splice acceptor of intron 1, 5’UTR of exon 2, CDS and 3’UTR of mouse Kras, including a polyadenylation signal flanked by LoxP and Frt sequences followed by the Neo cassette flanked by SDA (self-deletion anchor) sites. DTA was used for negative selection. A glycine-to-cysteine substitution (GGT to TGT) at codon 12 was introduced into exon 2 in 3’ homology arm. C57BL/6 ES cells were used for gene targeting. Targeted ES cells were injected into C57BL/6 albino embryos, which were then re-implanted into CD-1 pseudo-pregnant females. Founder animals were identified by their coat color, their germline transmission was confirmed by breeding with C57BL/6 females and subsequent genotyping of the offspring. A combination of primers internal and external to the targeting vectors was used to amplify the allele and PCR products were sequenced to ensure the absence of other alternations.

Fam86a^tm1e.1 knockout strain was obtained from the international mouse phenotyping consortium (IMPC)³⁵. p53^LoxP/LoxP mice have been described before⁴⁵. All mice were maintained in a mixed C57BL/6 and 129/Sv background, and we systematically used littermates as controls in all the experiments. Immunocompromised NSG mice (NOD.SCID-IL2Rg−/−) mice were utilized for xenograft tumor growth studies. All experiments were performed on 6 to 10-week-old animals with balanced cohorts of male and female mice as our initial data did not indicate significant differences in disease progression or response to treatment between females or males. All animals were numbered and experiments were conducted in a blinded fashion. After data collection, genotypes were revealed and animals assigned to groups for analysis. For treatment experiments, mice were randomized. None of the mice with the appropriate genotype were excluded from this study or used in any other experiments. Mice had not undergone prior treatment or procedures. All mice were fed a standard chow diet ad libitum and housed in pathogen-free facility with standard controlled temperature, humidity, and light-dark cycle (12h) conditions with no more than 5 mice per cage under the supervision of veterinarians, in an AALAC-accredited animal facility at the University of Texas M.D. Anderson Cancer Center. All animal procedures were reviewed and approved by the MDACC Institutional Animal Care and Use Committee (IACUC 00001636, PI: Mazur).

Cell lines

293T (female, embryonic kidney), U2OS (female, 15 years old, osteosarcoma), T3M4 (male, age not reported, pancreatic cancer), and HT1080 (male, 35 years old, fibrosarcoma) cells were cultured in DMEM supplemented with 10% fetal bovine serum, 2mM glutamine, and 100 U/mL penicillin/streptomycin. NCI-H2122 (female, 46 years old, lung adenocarcinoma), A549 (male, 58 years old, lung adenocarcinoma) and NCI-H358 (male, bronchioalveolar carcinoma) cells were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum and 100 U/mL penicillin/streptomycin. IMR-90 (female, 16 weeks gestation, normal lung fibroblast) and WI-38 (female, 3-months gestation, normal lung fibroblast) were from ATCC and cultured in EMEM supplemented with 10% fetal bovine serum and 100 U/mL penicillin/streptomycin. All cells were cultured at 37 °C in a humidified incubator with 5% CO2. Cell lines were authenticated by short tandem repeat profiling and tested negative for mycoplasma.

Cancer cell-derived xenografts

LUAD cell lines A549 and H2122 xenograft tumors were generated by transplanting cells that were transduced with lentivirus expressing sgRNA/Cas9 and indicated reconstitution vectors expressing wildtype or mutant FAM86A to immunocompromised NSG mice. The transplantation was performed by subcutaneous injection of cells mixed with matrigel (1:1) to the flanks of mice. When tumors became palpable, they were calipered every 2 days to monitor growth kinetics. Tumor volume was calculated using the formula: Volume = (width)2 × length / 2 where length represents the largest tumor diameter and width represents the perpendicular tumor diameter.

Patient-derived cancer xenografts

PDXs were obtained from the NCI Patient-Derived Models Repository (PDMR), NCI-Frederick, Frederick National Laboratory for Cancer Research. The following know mutations were detected in PDX sample: KRASG12C, p53R273C, MSH6K1358Dfs*2, KMT2CM3463_Q3464delinsI*, TET1E1640*, MUTYH^X298_splice treatment naïve, Asian male, age 57. Briefly, surgically resected tumor specimens were obtained from deidentified patients with histologically confirmed LUADs. All tumor specimens were collected after written patient consent and in accordance with the institutional review board-approved protocols of the University of Texas M.D. Anderson Cancer Center (PA19–0435, PI: Mazur). Patient-derived xenograft tumors were generated and propagated by transplanting small tumor fragments isolated directly from surgical specimens subcutaneously into NSG mice as we established previously³⁷. For analysis of FAM86A knockdown growth, collected PDX tumors were minced using a razor blade and digested in collagenase digestion buffer at 37°C for 1 hour. Cells were passed through 100 μm and 40 μm cell strainers and centrifuged for 1200 rpm for 5 min. Cells were incubated in RBC lysis buffer for 2 min and then resuspended in 6 mL of media and spun through 0.5 mL of serum layered on the bottom of the tube to remove cellular debris. Contaminating human or mouse hematopoietic and endothelial cells (CD45, Ter119, CD31) are depleted using biotin-conjugated anti-mouse CD45, CD31 and Ter119 antibodies and separated on a MACS LS column using anti-biotin microbeads. Next, the cells were transduced with lentivirus expressing sgRNA/Cas9 and selected with puromycin for 72 hours. Next, the cells were collected, mixed with Matrigel (1:1) at a density of 2 ×10⁷ cells per ml and 100 μL transplanted subcutaneously to the hind flanks of NSG mice. When tumors became palpable, they were calipered to monitor growth kinetics. Tumor volume was calculated using the formula: Volume = (width)² × length / 2 where length represents the largest tumor diameter and width represents the perpendicular tumor diameter.

METHOD DETAILS

Transfection and viral transduction

Transient expression was performed using TransIT-293 following the manufacturer’s protocol. For CRISPR-Cas9 knockouts, virus particles were produced by co-transfection of 293T cells with the lentiCRISPR v2/puro or lentiCRISPR v2/hygro construct containing indicated sgRNAs, pCMV-VSV-G and pCMV-dR8.2 dvpr in a ratio of 5:1:4 by mass. For complementation, virus particles were produced in the same manner with pLenti6.2 V5-DEST. 48 hours after transfection, target cells were transduced with 0.45 μm filtered viral supernatant and 8 μg/mL polybrene and changed to fresh media 24 hours later. Cells were selected beginning 24 hours after media replacement with 2 μg/ml puromycin, 200 μg/ml Hygromycin B, or 8 μg/ml blasticidin.

Plasmids

Default protein sequences were FAM86A (UniProt ID: Q96G04) and eEF2 (Uniprot ID: P13639) unless mutations otherwise specified. Bacterial expression plasmids were created using pGEX-6P-1. Transient mammalian expression was conducted using pcDNA3.1(+) including the N-terminal FLAG sequence DYKDDDDK. The plasmid lentiCRISPRv2/puro was used by default for CRISPR-Cas9 knockdown, while lentiCRISPRv2/hygro was used for any cells subject to a puromycylation assay. The plasmid pLenti6.2 V5-DEST/blasticidin was used for stable expression in mammalian cells including reconstitution of FAM86A. Gene-coding inserts were amplified by PCR using cDNA from U2OS cells as templates. Single point mutations of FAM86A and eEF2 were generated by site-directed mutagenesis.

Immunoblot analysis

For western blot analysis, cells were lysed in RIPA buffer with 1 mM PMSF and protease inhibitor cocktail. Protein concentration was determined using the Pierce Coomassie Plus Assay. Protein samples were resolved by SDS-PAGE and transferred to a PVDF membrane (0.45 μm). The following antibodies were used (at the indicated dilutions): FAM86A (1:500), eEF2-K525me3 (1:1,000), eEF2 (1:10,000), eEF2-T56ph (1:1000), beta-tubulin (1:5,000), RPS6 (1:1000), RPL40 (1:5000), puromycin (1:10,000), Peroxidase-conjugated HRP (1:10,000), GAPDH (1:5,000), and actin (1:5,000). Mouse and Rabbit secondary antibodies were used at 1:10,000 dilution in combination with either Amersham ECL or visualized by fluorescence on a LiCOR Odyssey Fc.

Expression and purification of recombinant proteins

GST fusion proteins were expressed in BL21 Escherichia coli by overnight culture at 20 °C in LB medium (10 g/L tryptone, 5 g/L yeast extract, and 10 g/L NaCl) supplemented with 0.1 mM IPTG (isopropyl 1-thio-b-D-galactopyranoside), purified using Glutathione Sepharose 4B and eluted in 10 mM reduced glutathione. Protein concentrations were measured using Pierce Coomassie Plus Assay. For purification of FLAG-eEF2, eEF2 bearing an N-terminal FLAG tag was expressed by transient transfection in either wild-type or FAM86A-depleted 293T cells. 48 hours after transfection, eEF2 was isolated from WCEs using anti-Flag M2 magnetic beads according to the instructions of the manufacturer and eluted with 200 μM 3x Flag peptides. The resulting purified eEF2 was immediately visualized by SDS-PAGE for quality control and used for enzymatic reactions.

Sucrose cushion purification of human ribosomes

Approximately 5 million 293T cells were trypsinized from a 10 cm plate and pelleted by centrifugation 500g × 5 minutes. Cell pellets were washed in cold PBS and lysed by resuspension in Triton Lysis Buffer (20 mM Tris pH 7.5, 150 mM NaCl, 5 mM MgCl₂, and 1% Triton X) supplemented with 1 mM DTT, 1mM PMSF, and complete protease inhibitor. Lysates were clarified by centrifugation at 14,000 rpm × 10 minutes and protein concentration determined by Coomassie Plus Assay. 300 μg of lysate was loaded per 400 μL sucrose cushion (1M sucrose in Triton Lysis Buffer) in Beckman thickwall 8×34mm tubes. Samples were centrifuged at 100,000 rpm for 1 hour at 4 °C. Pelleted ribosomes were washed twice with water and resuspended in buffer containing 15 mM Tris pH 7.5, 150 mM NaCl, and 15 mM MgCl₂.

Protein sequence alignments

Proteins queries were FAM86A (Uniprot: Q96G04), efm3 (Uniprot: P47163), METTL19 (Uniprot: Q8IYL2), METTL20 (Uniprot: Q8IXQ9), CAMKMT (Uniprot: Q7Z624), METTL18 (Uniprot: O95568), and METTL22 (Uniprot: Q9BUU2). Sequences were aligned by Clustal Omega.

Protein modeling

Publicly available protein models for FAM86A (PDB: 8FZB) and the eEF2–80S ribosome (PDB: 6Z6N, 6D9J, 4V6X) were accessed and visualized using PyMOL. The electron density map of 6Z6N was downloaded directly from https://www.rcsb.org/structure/6z6n and methylation modeled onto eEF2-K525 using PyMOL.

Polysome profiling

Control- or FAM86A-depleted H2122 and H358 cells were seeded into six 15-cm Petri dishes (~10 million cells per dish) 24 hours prior to treatment with 100 μg/mL cycloheximide and incubated for 2 minutes at 37°C. Cells were washed and scraped in cold PBS containing 100 μg/mL cycloheximide, pelleted and lysed in lysis buffer (5 mM Tris pH7.4, 2.5 mM MgCl₂, 1.5 mM KCl, 100 mg/mL cycloheximide, 2 mM DTT, 0.5% Triton, 0.5% Na-DOC, 100 U/mL SUPERase In RNase inhibitor and protease inhibitors). Lysates were clarified for 10 minutes at 14,000 rpm at 4°C. RNA contents were determined by Nanodrop and 500 μg of RNA were loaded on 10%−50% sucrose gradients made in 15 mM Tris pH 7.4, 15 mM MgCl₂, 150 mM NaCl and prepared using a BioComp Gradient Station. Gradients were spun for 2 hours in a TH-641 rotor (Sorvall) at 40,000 rpm and 4°C. Gradients were analyzed (254 nm) and fractions collected with a BioComp Piston Gradient Fractionator.

In vitro methylation assays

In vitro methylation assays were performed by combining 3 μg of purified, recombinant proteins in methyltransferase assay buffer (50 mM Tris pH 8.0, 20 mM KCl, 5 mM MgCl₂, 1 mM DTT, and 10% glycerol) supplemented with 1 mM S-adenosyl-methionine (SAM) or 2 μCi of tritiated AdoMet. For assays with cell extracts, extracts were prepared in cell lysis buffer (50 mM Tris pH 8.0, 150 mM NaCl, 0.2 mM EDTA, 0.1% NP-40, and 10% glycerol), clarified by centrifugation at 14,000 rpm × 10 min, and 25 μg loaded into the reaction mixture. Reaction mixtures were incubated overnight at 30°C. Reactions were resolved by SDS-PAGE, followed by autoradiography, Coomassie stain or western blot analysis.

In vitro GTPase assays and Frameshifting assays

GTPase assays were performed using ATPase/GTPase Activity Assay Kit following the manufacturer’s protocol. Briefly, 0.5 μg of Flag-tagged eEF2-K525me0/3 was incubated with 50 μM GTP ± 5 μg purified HEK293T ribosomes at 37°C for 3 hours in 30 μL of manufacturer’s assay buffer. The reactions were terminated by adding 200 μL of the kit reagent and incubating for an additional 30 min at room temperature. Formation of hydrolyzed free phosphate was measured at a wavelength of 620 nm and readings of negative control reactions (lacking eEF2) were subtracted from the sample readings. Frameshifting assays were performed by transient transfection of H2122 cells (as above) with pSGDlucV3.0 and analyzed with the Dual-Glo Luciferase Assay System following the manufacturer’s protocol as previously described⁴⁷.

Proliferation Assays

Cells were seeded at 500,000 cells per 6 cm plates on day 0. Cell counts were acquired by Countess II FL Automated Cell Counter (Thermo Fisher Scientific) at indicated days for 10 days and graphed as fold-change relative to day 0. After each counting, the cells were maintained at a density between 400,000 and 600,000 cells/plate. Trypan blue was used to stain non-viable cells.

Cell Viability assays

Cells in triplicate were seeded at 5,000 cells per well in 96 well plate on day 0. Viability was measured at indicated times using PrestoBlue^® Cell Viability Assay (ThermoFisher Scientific) following the manufacturer’s protocol.

Puromycylation Assays

Adapted from²⁷, control- or FAM86A-depleted cells were seeded at 300,000 cells per 6 cm plate 24 hours prior to 24 hours of serum starvation. For serum re-stimulation, cells were changed to serum-rich media containing 10% fetal bovine serum for an optimized period of 1 hour prior to puromycin pulse. Puromycin pulses were performed by incubating the cells with 10 μg/mL puromycin for 15 min at 37°C. Cells were then washed with cold PBS and lysed in RIPA buffer supplemented with 1 mM PMSF and protease inhibitor mixture. Whole cell lysates (~10 μg each) were then assayed by western blot using the anti-puromycin antibody. For assays with HHT, cells were treated with HHT [10mM] and puromycin as above to label proteins.

AHA LABELING Assays

Control- or FAM86A-depeleted cells were grown in methionine free media for 1h and then cultured in 50 μM of AHA (L-azidohomoalanine) for 2 hours. RIPA buffer was used for harvesting whole cell lysates and click reactions were performed using Click Chemistry Reaction Buffer Kit (Click Chemistry Tools) following the manufacturer’s recommendation. Briefly, 100–200 mg of lysates were incubated with 40 mM Biotin-PEG4-Alkyne for 30 min. Total protein was extracted using methanol and chloroform. 1–5 mg of lysates were further assayed by western blot analysis using streptavidin-conjugated horseradish peroxidase.

Lung Adenocarcinoma Mouse Models

To generate tumors in the lungs of Kras^cKI-G12C/+, p53^LoxP/LoxP and Fam86a^tm1e.1 (Kras;p53 and Kras;p53;Fam86a) mutant mice, we used replication-deficient adenoviruses expressing Cre-recombinase (Ad-Cre) as previously described³⁶. Briefly, 8-week old mice were anesthetized by continuous gaseous infusion of 2% isoflurane for at least 10 min using a veterinary anesthesia system. The virus was delivered to the lungs by intratracheal instillation. Prior to administration, the virus was precipitated with calcium phosphate to improve the delivery of Cre by increasing the efficiency of viral infection of the lung epithelium. Mice were treated with one dose of 5 × 10⁶ PFU of Ad-Cre. Mice were analyzed for tumor formation and progression at indicated time points after viral infection.

Histology and Immunohistochemistry

Tissue specimens were fixed in 4% buffered formalin for 24 hours and stored in 70% ethanol until paraffin embedding. 3 μm sections were stained with hematoxylin and eosin (HE) or used for immunohistochemical studies. Human tissue sections were collected in accordance with the institutional review board-approved protocols of the University of Texas M.D. Anderson Cancer Center (PA19–0435, PI: Mazur). Immunohistochemistry (IHC) was performed on formalin-fixed, paraffin embedded mouse and human tissue sections using a biotin-avidin HRP conjugate (Vectastain ABC kit) method as described before⁴⁸. The following antibodies were used (at the indicated dilutions): cleaved Caspase 3 (1:500), pH3 (1:1,000), Ki67 (1:2,000), pERK1/2 (1:1000), eEF2K525me3 (1:1000). Sections were developed with DAB substrate and counterstained with hematoxylin. Pictures were taken using a PreciPoint M8 microscope equipped with the PointView software. Analysis of the tumor area and IHC analysis was done using ImageJ software. Quantification eEF2K525me3 IHC chromogen intensity was performed by measuring the reciprocal intensity of the chromogen stain as previously described⁴⁹.

Quantification and Statistical Analysis

Please refer to the Figure Legends for description of sample size (n) and statistical details. All values for n are for individual mice or individual samples. Sample sizes were chosen based on previous experience with given experiments. Cell culture assays have been performed in triplicates and in two independent experiments, unless stated otherwise. Differences were analyzed by log-rank, two-tailed unpaired Student’s t test, two-way ANOVA with Tukey’s testing for multiple comparisons using Prism 10 (GraphPad) or Rstudio (ggpubr package).

Highlights:

• FAM86A physiologically generates eEF2K525me3 in diverse cellular and tissue samples

• FAM86A-eEF2K525me3 pathway regulates elongation to promote protein synthesis

• Generation of a facile KRAS.G12C GEMM to model lung cancer in vivo

• Inhibition of FAM86A-eEF2K525me3 axis strongly suppresses lung cancer tumorigenesis