Unveiling unique protein and phosphorylation signatures in lung adenocarcinomas with and without ALK, EGFR, and KRAS genetic alterations

Abstract

Genetic alterations in key oncogenes have been frequently identified in lung adenocarcinoma (LUAD), including genes encoding epidermal growth factor receptor (EGFR), Kirsten rat sarcoma viral oncogene homolog (KRAS), and anaplastic lymphoma kinase (ALK). In this pilot study, we aimed to characterize the differences in enriched biological pathways and phosphorylation events between LUAD tumors harboring EGFR, KRAS, or echinoderm microtubule-associated protein-like 4 (EML4)-ALK oncogenic alterations and triple wild-type LUAD tumors (WT, without EML4-ALK, KRAS, or EGFR alterations) by mass spectrometry (MS)-based quantitative proteomics and phosphoproteomics. We analyzed tumor regions of 82 formalin-fixed paraffin-embedded (FFPE) tissue sections with 6, 23, 31, and 22 samples from the EML4-ALK, EGFR, KRAS, and WT sample groups, respectively. A total of 1377 to 2189 proteins and 73 to 1781 phosphosites were quantified in these analyses. Based on the results, the samples clustered according to their genetic alteration type, and EGFR-mutated samples showed unique protein expression patterns. Membrane organization, vesicle organization, and vesicle-mediated transport Gene Ontology Biological Process (GOBP) terms were significantly downregulated in EGFR-mutated samples compared to the other sample groups. Changes in 36 proteins and 52 phosphosites were also identified as potentially specific to a given genetic alteration. Many of these proteins have previously been linked to EGFR or KRAS mutations [e.g., cathepsin L, stimulator of interferon genes protein (STING)], whereas several phosphoproteins are associated with RNA splicing [e.g., serine/arginine-rich splicing factor 1 (SRSF1), SRSF2, and SRSF7 proteins]. Kinase-substrate enrichment analysis indicated altered activities of 10 kinases, including mitogen-activated protein kinases (MAPKs) and cyclin-dependent kinases (CDKs). For example, CDK2 activity was elevated in EML4-ALK samples compared to the other sample groups. Our results could provide significant insights into further studies that could contribute to developing improved diagnostic and therapeutic strategies for LUAD.

Keywords: FFPE tissue; cancer research; genetic alterations; lung adenocarcinoma; mass spectrometry; phosphoproteomics.

Fig. 1

(Phospho)proteomic analysis of lung adenocarcinoma (LUAD) tumors. (A) Schematic workflow of applied methodologies: Small regions of formalin‐fixed paraffin‐embedded (FFPE) tissue sections from LUAD tumor samples were selected and subjected to on‐surface tryptic digestion. The resulting peptides were enriched for phosphopeptides using TiO₂ spin tips, separating phosphopeptides from non‐phosphorylated peptides, both of which were then analyzed separately by mass spectrometry. (B) Distributions of the samples analyzed in the cohort based on the genetic alterations and subtypes. In the case of WT samples, NA refers to not applicable, while for KRAS samples, NA refers to not available.

Fig. 2

Proteomic analysis of the EML4–ALK, EGFR, KRAS, and triple wild‐type (WT) sample groups. (A) Uniform manifold approximation and projection (UMAP) representation of the 82 samples based on the 2021 proteins. Different colors mark the four sample groups (blue for EML4–ALK, yellow for EGFR, green for KRAS, and magenta for wild‐type). (B) Heatmap with hierarchical clustering of the 166 proteins significantly altered (false discovery rate < 0.05, fold change > 2) between the groups (log2 protein LFQ intensity values are Z‐scored).

Fig. 3

Gene set enrichment analysis (GSEA) dot plot of significantly enriched Gene Ontology Biological Process (GOBP) terms in EGFR‐WT comparison, where the x‐axis represents the normalized enrichment score (NES). The size of the dots corresponds to the number of genes associated with the term, and the color represents the NES.

Fig. 4

Euler plot of the 36 potential alteration‐specific proteins. Proteins highlighted in black were significantly under‐expressed, while those in red were significantly overexpressed within the specific sample group. The proteins positioned at the bottom of the figure exhibit distinct expression patterns across multiple groups. A blue line indicates significant differences in expression between the connected groups, with overexpression occurring in the sample groups highlighted in red.

Fig. 5

Phosphoproteomic analysis of the EML4–ALK, EGFR, KRAS, and WT sample groups. (A) UMAP representation of the 82 samples based on the 384 phosphosites quantified using maxquant and considered for statistical analysis. Different colors mark the four sample groups (blue for EML4–ALK, yellow for EGFR, green for KRAS, and magenta for wild‐type). (B) Heatmap with hierarchical clustering of the 183 phosphosites significantly altered (false discovery rate < 0.05, fold change > 2) between the groups (log2 phosphosite intensity values are Z‐scored).

Fig. 6

Euler plot of the 52 potential alteration‐specific phosphosites. Phosphosites highlighted in black showed significantly decreased phosphorylation, while those in red showed significantly increased phosphorylation within the specific sample group. The phosphosites positioned at the bottom of the figure exhibit distinct phosphorylation patterns across multiple groups. A blue line indicates significant differences in phosphorylation between the connected groups, with increased phosphorylation occurring in the sample groups highlighted in red.

Fig. 7

Visualization of functional enrichment analysis in string for Gene Ontology Biological Process (GOBP) terms of 44 phosphoproteins with potential alteration‐specific phosphosites. The size of the dots corresponds to the number of genes associated with the term, and the color represents the false discovery rate (FDR).

Fig. 8

Result of the kinase–substrate enrichment analysis (KSEA). Kinase scores based on KSEA of 202 kinase–substrate pairs (number of substrates ≥ 3). Blue bars for suppressed and red bars for elevated kinase activities (P < 0.05 based on z‐score transformation). Black bars for kinases without significantly altered activities.

Fig. 9

Immunohistochemistry staining of phospho‐p44/42 MAPK (Erk1/2) (Thr202/Tyr204) antibody to assess MAPK3/MAPK1 activity in LUAD tumors. Scale bar: 50 μm. Brown color represents positive staining.

Fig. 10

Immunohistochemistry staining of phospho‐p38 MAPK (Thr180/Tyr182) antibody to assess MAPK11 activity in LUAD tumors. Scale bar: 50 μm. Brown color represents positive staining.