ROS1
Proto-oncogene tyrosine-protein kinase ROS is an enzyme that in humans is encoded by the ROS1 gene. It is a receptor tyrosine kinase (RTK) of the insulin receptor family, consisting of 2,347 amino acids and featuring an extracellular ligand-binding domain, a single transmembrane segment, and a cytoplasmic tyrosine kinase domain.[1] ROS1 plays a role in epithelial cell differentiation and activation of downstream signaling pathways that regulate cell growth, survival, and organ development, particularly upon binding ligands such as NELL2. In normal physiology, it is expressed in various tissues including the kidney, but its precise physiological functions remain under investigation.[2] Pathologically, ROS1 is significant in oncology due to gene fusions or rearrangements that occur in approximately 1–2% of non-small cell lung cancers (NSCLC), often in younger, never-smoking patients with adenocarcinoma histology. These fusions, such as CD74-ROS1 or EZR-ROS1, result in constitutively active kinases driving tumorigenesis. ROS1 rearrangements are also reported in other cancers including colorectal, gastric, and ovarian carcinomas, though less frequently. Targeted therapies, including tyrosine kinase inhibitors like crizotinib (FDA-approved in 2016 for ROS1-positive metastatic NSCLC), entrectinib, and repotrectinib, have shown efficacy in treating these fusion-positive tumors. As of 2025, ongoing clinical trials and initiatives like The ROS1ders are advancing research and patient advocacy for ROS1-driven cancers.[3]Molecular Biology
Gene Characteristics
The ROS1 gene, with official symbol ROS1 and Gene ID 6098 (updated September 9, 2025), is located on the long arm of human chromosome 6 at cytogenetic band 6q22.1, spanning approximately 139 kb from genomic coordinates 117,287,353 to 117,425,942 on reference assembly GRCh38.p14.[1] It encodes the ROS proto-oncogene 1, a receptor tyrosine kinase belonging to the insulin receptor family.[4] The gene consists of 43 exons in its primary transcript (ENST00000368508.7), though alternative splicing can produce variants with up to 46 exons.[5][1] Expression of ROS1 is primarily restricted during development to epithelial tissues, where it plays a role in morphogenic processes such as differentiation in organs like the epididymis, coinciding with key developmental events. In 2020, NELL2 was identified as the endogenous ligand for ROS1, which activates intracellular signaling to promote epididymal epithelial differentiation.[6][7][8] In adult human tissues, basal RNA expression is low and tissue-specific, with the highest levels observed in the lung (RPKM 10.3) and also notable in testis and epididymis, but minimal elsewhere.[1] However, ROS1 is upregulated in various tumor cell lines, reflecting its proto-oncogenic potential.[1] ROS1, with NELL2 as its endogenous ligand, exhibits strong evolutionary conservation across species, from Caenorhabditis elegans to Drosophila melanogaster and mammals, underscoring its fundamental role in cellular signaling. Its Drosophila homolog, sevenless (Sev), is a well-characterized receptor tyrosine kinase essential for photoreceptor development in the compound eye.[9] The encoded protein is a type I integral membrane receptor.[4]Protein Structure
The ROS1 protein is a 2,347-amino-acid polypeptide encoded by a gene located on chromosome 6q22.1.[10] It functions as a type I transmembrane receptor tyrosine kinase, comprising an N-terminal extracellular region spanning residues 1–1,859, a single transmembrane helix at residues 1,860–1,882, and an intracellular C-terminal region from residues 1,883–2,347.[10][11] The extracellular domain includes multiple fibronectin type III (FN3) domains, such as those at residues 99–177 and 195–272, which contribute to structural integrity and potential ligand interactions.[12] This region also contains predicted disulfide bonds and N-linked glycosylation sites (e.g., at asparagine residues in FN3 motifs), which stabilize the protein fold and facilitate proper membrane localization.[10] The intracellular tyrosine kinase domain spans residues 1,945–2,222 and encompasses key catalytic elements, including the ATP-binding site with the conserved GXGXXG motif (P-loop) at residues 1,952–1,957.[13] The activation loop within this domain, approximately residues 2,110–2,145, features tyrosine residues Y2,110, Y2,114, and Y2,115 that serve as sites for autophosphorylation, enabling kinase activation.[4] Beyond the kinase domain, the C-terminal tail (residues 2,223–2,347) includes additional regulatory tyrosines, such as Y2,274 and Y2,334, which undergo autophosphorylation to modulate signaling.[4]Physiological Function
Normal Roles
ROS1, functioning as a receptor tyrosine kinase (RTK), displays dynamic expression patterns that underscore its involvement in developmental processes. During fetal development, ROS1 exhibits high levels of expression in epithelial tissues of organs undergoing morphogenesis, such as the kidney, lung, and intestine, where it aligns temporally with critical differentiation and branching events.[14] In contrast, adult tissues show markedly lower expression, confined primarily to niches like the testis (particularly the epididymis) and select neural structures, including the cerebellum and peripheral neural tissue.[15][10] In kidney development, ROS1 is expressed in the ureteric bud, suggesting a potential involvement in inductive interactions between the ureteric bud and metanephric mesenchyme.[16] Its transient expression in the developing lung coincides with epithelial differentiation and morphogenetic events during embryogenesis.[14] These expression patterns highlight ROS1's potential RTK-mediated regulation of cellular processes essential for organogenesis, though its precise mechanisms in these contexts rely on unidentified ligands.[10] Studies in animal models further elucidate ROS1's physiological functions, particularly in reproductive and neural systems. Despite expression in developing kidney and lung, ROS1 knockout mice show no histological abnormalities in these organs, suggesting redundant mechanisms or non-essential roles. ROS1 knockout mice exhibit male sterility due to disrupted epithelial differentiation in the epididymis initial segment, leading to production of immotile spermatozoa with flagellar defects, indicating ROS1's necessity for spermatogenesis and post-testicular sperm maturation.[6] Additionally, expression in neural tissues suggests potential involvement in neuronal differentiation, as inferred from developmental patterns and the absence of overt neurological phenotypes in knockouts, pointing to compensatory mechanisms.[15] Overall, ROS1 supports non-pathological cell growth, survival, and migration in these contexts, maintaining tissue homeostasis, though its ligand-dependent activation in adult humans—recently linked to the endogenous ligand neural epidermal growth factor-like ligand 2 (NELL2)—remains under investigation in non-reproductive tissues.[10][17]Signaling Mechanisms
ROS1 activation is primarily ligand-dependent, with neural epidermal growth factor-like ligand 2 (NELL2) identified as an endogenous ligand that binds the extracellular domain, promoting receptor dimerization and subsequent autophosphorylation of intracellular tyrosine residues, such as Y2274 and Y2334.[4][17] This autophosphorylation event stabilizes the active conformation of the kinase domain, enabling transphosphorylation of additional sites including Y2114 and Y2115, which serve as docking platforms for downstream signaling adaptors.[4] Upon activation, ROS1 phosphorylates substrates that initiate multiple downstream pathways critical for cellular responses. The PI3K/AKT pathway is engaged through recruitment of p85, promoting cell survival and growth; the MAPK/ERK cascade is activated via GRB2/SOS/RAS coupling, driving proliferation; and STAT3 phosphorylation occurs directly or indirectly, influencing differentiation and gene expression.[4][18] Kinase activity of ROS1 is characterized by in vitro assays demonstrating sensitivity to ATP-competitive inhibitors, such as crizotinib with an IC50 of approximately 42 nM, which bind the conserved ATP-binding pocket in the kinase domain.[19] Structurally, activation involves a conformational shift in the activation loop from an inactive, disordered state to an ordered, extended configuration that aligns catalytic residues and facilitates substrate access.[20][4] Signaling is tightly regulated by negative feedback mechanisms, including dephosphorylation by protein tyrosine phosphatases (PTPs) such as SHP-1 and SHP-2, which bind autophosphorylation sites to attenuate kinase activity.[21] Additionally, ubiquitination targets activated ROS1 for proteasomal degradation, mediated by E3 ligases like the Cbl family, preventing sustained signaling.[22]Pathological Roles
Involvement in Cancer
ROS1 functions as a proto-oncogene, encoding a receptor tyrosine kinase whose aberrant activation drives oncogenesis by promoting uncontrolled cell proliferation, invasion, and metastasis through hyperactive downstream signaling pathways such as PI3K/AKT, MAPK/ERK, and STAT3.[23][4] In normal physiology, ROS1 signaling is tightly regulated, but in cancer, dysregulation hijacks these pathways to confer malignant properties, including enhanced survival and migratory capabilities independent of physiological ligands.[24][23] Alterations in ROS1 are detected in approximately 1-2% of non-small cell lung cancer (NSCLC) cases, predominantly adenocarcinomas, and occur at lower frequencies in other malignancies including glioblastoma, ovarian carcinoma, sarcoma (such as angiosarcoma and epithelioid hemangioendothelioma), colorectal cancer, and inflammatory myofibroblastic tumors.[24][23][4] These changes often involve chromosomal rearrangements, such as the FIG-ROS1 fusion originally identified in glioblastoma, which exemplify how ROS1 dysregulation contributes to tumorigenesis across diverse tissue types.[23][4] Beyond fusions, non-fusion mechanisms of ROS1 activation include gene amplification, activating point mutations, and overexpression, all of which can lead to ligand-independent dimerization and constitutive kinase activity.[25][23] For instance, amplification has been implicated in glioblastoma progression, while overexpression correlates with invasive phenotypes in cancers like metastatic oral squamous cell carcinoma, potentially through enhanced autophosphorylation and signaling without external stimuli.[26][4] ROS1 alterations are associated with aggressive disease characteristics, such as poorer disease-free survival in certain cohorts, yet they confer a favorable prognostic outlook in terms of response to targeted interventions compared to wild-type tumors, with studies showing prolonged overall survival in ROS1-positive NSCLC patients (e.g., median 69.8 months versus 13.7 months in negatives in a cohort treated with crizotinib).[27][28][29] This duality underscores ROS1's role as a driver of malignancy that is therapeutically exploitable, highlighting the need for molecular profiling to identify actionable alterations.[24][29]Fusion Variants and Prevalence
ROS1 fusion genes arise primarily through genomic rearrangements involving the ROS1 tyrosine kinase domain on chromosome 6q22, resulting in chimeric proteins with constitutive kinase activity that drives oncogenesis. These fusions are most prevalent in non-small cell lung cancer (NSCLC), occurring in approximately 1-2% of cases overall, with frequencies up to 2.6% reported in some cohorts.[30] In NSCLC, ROS1 fusions are detected almost exclusively in the adenocarcinoma subtype, with a prevalence of about 2.5-2.8% in adenocarcinomas compared to less than 1% in squamous cell carcinomas.[31] The most common ROS1 fusion variants in NSCLC involve a limited set of 5' fusion partners, with over 50 recurrent and rare types identified across studies as of 2025.[32] These fusions typically result from intrachromosomal deletions or inversions within chromosome 6q22, preserving the entire ROS1 kinase domain while replacing the transmembrane and extracellular regions with dimerization motifs from the partner gene, leading to ligand-independent activation. The predominant variants include CD74-ROS1 (38-54% of ROS1-positive cases), EZR-ROS1 (13-24%), SDC4-ROS1 (9-13%), and SLC34A2-ROS1 (5-10%), with less frequent partners such as GOPC (FIG)-ROS1 (around 3-5%).[30][33][34]| Fusion Variant | Approximate Frequency in ROS1+ NSCLC | Partner Gene Location |
|---|---|---|
| CD74-ROS1 | 38-54% | Chromosome 5p14.1 |
| EZR-ROS1 | 13-24% | Chromosome 6q25.1 |
| SDC4-ROS1 | 9-13% | Chromosome 20q11.21 |
| SLC34A2-ROS1 | 5-10% | Chromosome 4q21.1 |
| GOPC-ROS1 | 3-5% | Chromosome 6q21 |