Abstract

   Tumor metastasis severely compromises the prognosis of patients with
   non-small cell lung cancer (NSCLC), yet the underlying mechanisms
   remain unclear. The biogenic amine 5-hydroxytryptamine (5-HT), derived
   from tryptophan (Trp), has been reported to promote tumor growth, but
   its role in NSCLC metastasis has not been fully elucitated. In this
   study, fecal metabolomic profiling revealed significant enrichment of
   Trp metabolism and serotonergic synapse pathways in patients with NSCLC
   metastasis. Notably, the levels of 5-HT in peripheral blood showed a
   positive correlation with fecal Trp concentrations in NSCLC patients.
   Functional experiments demonstrated that 5-HT promoted NSCLC cell
   migration, invasion, epithelial-mesenchymal transition (EMT), and
   metastasis by inhibiting the expression of small nuclear
   ribonucleoprotein polypeptide G (SNRPG). RNA sequencing (RNA-seq)
   further identified cyclin-dependent kinase 14 (CDK14) as a critical
   downstream target of SNRPG. SNRPG downregulated CDK14 expression,
   thereby restraining malignant phenotypes in NSCLC cells. Furthermore,
   SNRPG negatively regulated the protein levels of Wilms tumor 1 (WT1) by
   directly binding to it. Loss of SNRPG led to WT1 accumulation, which in
   turn bound to the CDK14 promoter and activated transcription,
   ultimately driving NSCLC metastasis. These findings collectively
   suggest that the 5-HT/SNRPG/WT1/CDK14 axis plays a pivotal role in
   promoting NSCLC metastasis and highlight 5-HT and SNRPG as potential
   therapeutic targets for NSCLC metastasis.

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s43556-025-00312-4.

   Keywords: 5-Hydroxytryptamine , Small nuclear ribonucleoprotein
   polypeptide G , Cyclin-dependent kinase 14 , Metastasis, Non-small cell
   lung cancer

Introduction

   Lung cancer is the leading cause of cancer-related deaths worldwide,
   with non-small cell lung cancer (NSCLC) comprising 80%–85% of cases
   [[40]1]. Approximately 50% of NSCLC patients present with metastatic
   disease at diagnosis, and the median survival is only 22.5 months
   [[41]2]. Despite advances in surgery, chemotherapy, radiotherapy,
   targeted therapies, and immunotherapies, the 5-year survival rate for
   NSCLC remains a mere 26.5% [[42]3]. Reliable biomarkers for diagnosing
   or prognosticating NSCLC metastasis remain absent, highlighting the
   urgent need to investigate the molecular mechanisms underlying
   metastasis to enable early detection and treatment.

   Recent studies have established a strong link between the human gut
   microbiota and the initiation, progression, metastasis, and treatment
   response of cancer [[43]4, [44]5]. The gut microbiota, consisting of
   trillions of microorganisms, produces metabolites that regulate host
   functions and maintain a dynamic balance within the human body [[45]6].
   Notably, gut microbiota and its metabolites have been implicated in
   promoting migration and invasion of colorectal cancer cells, thereby
   facilitating tumor metastasis [[46]7, [47]8]. Serotonin
   (5-hydroxytryptamine, 5-HT), primarily synthesized by enterochromaffin
   cells (ECs) in the intestines [[48]9], and its levels are modulated by
   the gut microbiota [[49]10]. Additionally, gut-derived 5-HT promotes
   tumor growth in cancers of the breast, lung, colorectal, and liver
   [[50]11]. Excessive 5-HT in the gastrointestinal tract has been shown
   to aggravate colorectal cancer progression by activating inflammatory
   pathways and enhancing cancer cell self-renewal via the Wnt/β-catenin
   signaling pathway [[51]12, [52]13]. While the role of 5-HT in small
   cell lung cancer proliferation has been observed in vitro, its
   involvement in NSCLC metastasis remains unclear [[53]14].

   Ribonucleoprotein (RNP) granules, crucial RNA–protein complexes,
   participate in various cellular processes [[54]15]. SNRPG, a member of
   the RNP family, contributes to the formation of U small nuclear RNPs
   (snRNPs), which serve as precursors for spliceosomes [[55]16]. Although
   SNRPG has been identified as a potential biomarker for Alzheimer’s
   disease, its involvement in tumorigenesis, particularly in NSCLC
   metastasis, remains unclear [[56]9, [57]17].

   This study revealed a significant association between peripheral blood
   5-HT levels and NSCLC metastasis, with elevated 5-HT expression
   correlating with shorter progression-free survival (PFS).
   Mechanistically, 5-HT inhibited SNRPG expression, thereby deregulating
   the negative regulation of SNRPG on WT1 protein. Subsequently,
   accumulated WT1 transcriptionally activated CDK14, mediating
   epithelial-mesenchymal transition (EMT). This cascade promoted NSCLC
   cell migration and invasion, thereby enhancing metastasis. These
   findings highlight the critical role of the 5-HT/SNRPG axis in NSCLC
   metastasis and suggest its potential as a prognostic biomarker for
   patients with NSCLC.

Results

Patients with NSCLC exhibiting high expression of 5-HT have a poor prognosis

   Fecal specimens from NSCLC patients were divided into two groups: the
   metastatic group (M group; n = 11) and the non-metastatic group (N
   group; n = 21). Untargeted metabolomic profiling using liquid
   chromatography quadrupole time-of-flight (LC-QTOF) identified 991
   metabolites across both positive and negative ionization modes. Partial
   least squares discriminant analysis (PLS-DA) effectively distinguished
   the M and N groups, revealing significant metabolic differences
   (R^2Y = 0.89 and Q^2Y = 0.78). Model validation confirmed its efficacy,
   with an R^2 value exceeding Q^2 and a negative Q^2 regression intercept
   (Fig. [58]1a and Fig. S1a). Among the 259 differentially expressed
   metabolites identified through volcano and violin plot analyses,
   tryptophan (Trp) was significantly upregulated in the M group
   (Fig. [59]1b, c and Table S1). Receiver operating characteristic (ROC)
   curve analysis demonstrated excellent accuracy for Trp, with an area
   under the ROC curve (AUC) of 0.991 (Fig. [60]1d). KEGG pathway
   enrichment analysis identified significant enrichment of the
   serotonergic synapse pathways in the M group (Fig. [61]1e).
   Furthermore, ELISA confirmed a positive correlation between peripheral
   blood 5-HT levels and fecal Trp content in NSCLC patients
   (Fig. [62]1f). Trp is primarily metabolized through the kynurenine,
   indole, and 5-HT pathways, and the gut microbiota influences its
   metabolic fate. Gut bacteria, including Lactobacillaceae,
   Bacteroidaceae, Lachnospiraceae, and Enterobacteriaceae, can metabolize
   Trp, promoting 5-HT generation and correlating with elevated plasma
   5-HT levels [[63]18, [64]19]. Based on these findings and literature on
   gut microbiota-mediated Trp metabolism [[65]20–[66]25], we speculate
   that the gut microbiota of patients with NSCLC metastasis may
   contribute to elevated peripheral blood 5-HT levels through Trp
   metabolism, although the specific microbial contributions warrant
   further investigation.

Fig. 1.

   [67]Fig. 1
   [68]Open in a new tab

   Patients with NSCLC exhibiting high expression of 5-HT have a poor
   prognosis. (a) PLS-DA separated the M and N groups into distinct
   clusters. (b) Volcano plot highlighted differential metabolites. (c)
   Trp concentrations were compared between these groups. (d) ROC analysis
   for tryptophan (left panel) and boxplot showing the distribution of
   tryptophan levels in non-metastatic (N) and metastatic (M) NSCLC
   patients (right panel). The red dotted line indicates the optimal
   cutoff value for distinguishing between the two groups. (e) Metabolite
   functions and pathways were studied using the KEGG database with
   enriched pathways shown in a bubble plot. (f) A scatter diagram
   illustrated the correlation between peripheral blood 5-HT levels and
   fecal Trp in NSCLC patients. (g) Peripheral blood 5-HT levels of
   healthy individuals and NSCLC patients. (h) The relationships between
   peripheral blood 5-HT levels and metastasis. (i) Survival curve
   analysis between peripheral blood 5-HT levels and PFS. (*P < 0.05, ns:
   not significant. N-HM: non-hepatic metastasis; HM: hepatic metastasis;
   N-BM: non-brain metastasis; BM: brain metastasis; N-BOM: non-bone
   metastasis; BOM: bone metastasis; N-PE: non-pleural effusion; PE:
   pleural effusion; N-AM: non-adrenal gland metastasis; AM: adrenal gland
   metastasis)

   To further explore the role of 5-HT in NSCLC, peripheral blood serum
   samples from 96 patients with NSCLC and 56 healthy individuals were
   assessed using ELISA. Peripheral blood 5-HT levels were significantly
   higher in patients with NSCLC than in healthy individuals (P < 0.01,
   Fig. [69]1g). Subsequent analyses examined the associations between
   5-HT levels and clinical variables, including grade, TNM staging, age,
   gender and distant organ metastasis. Notably, stage IV patients
   exhibited higher 5-HT levels than stage III patients, although the
   difference was not statistically significant (P > 0.05; Fig. S1b).
   Similarly, patients with M1 stage NSCLC had elevated 5-HT levels
   compared to M0 stage patients (P > 0.05; Fig. S1c). Additionally,
   patients with liver metastases had significantly higher circulating
   5-HT concentrations than those without liver metastases (P < 0.05;
   Fig. [70]1h). Peripheral blood 5-HT levels showed no significant
   association with gender and age among patients with NSCLC (P > 0.05;
   Fig. S1d). Survival curve analysis demonstrated high 5-HT expression
   correlated with significantly shorter PFS in NSCLC patients
   (Fig. [71]1i). These findings implicate peripheral blood 5-HT levels as
   a potential prognostic biomarker, with elevated levels indicating worse
   clinical outcomes in NSCLC.

5-HT promotes migration and invasion of NSCLC cells in vitro

   H1299 cells were treated with varying concentrations and time gradients
   of 5-HT. The proliferative effect was found to be both time- and
   dose-dependent, peaking at 72 h and 20 μM (Fig. S2). Although
   supra-physiological, this concentration was selected based on optimal
   in vitro responses and previous studies on 5-HT, allowing for the
   observation of cellular reactions that may occur under pathological
   conditions or in microenvironments with elevated local 5-HT
   concentrations. Migration and invasion assays were conducted after
   treating H1299 and A549 cells with 5-HT to explore its potential in
   promoting NSCLC metastasis. The results revealed a morphological
   transition in cells treated with 5-HT, from polygonal or
   cobblestone-like shapes to elongated spindle-shaped or fusiform forms
   compared to the PBS group (Fig. [72]2a). Wound healing and transwell
   assays were subsequently performed. 5-HT treatment markedly enhanced
   the migration and invasion capabilities of NSCLC cells in vitro
   (Fig. [73]2b, c). Additionally, in the 5-HT-treated group, E-cadherin
   was downregulated, while N-cadherin, vimentin, and Snail were
   upregulated (Fig. [74]2d, e). These results collectively suggest that
   5-HT promotes migration, invasion, and EMT in NSCLC cells.

Fig. 2.

   [75]Fig. 2.
   [76]Open in a new tab

   5-HT promotes migration and invasion of NSCLC cells in vitro. (a) The
   morphological transformation of H1299 and A549 cells was examined by a
   phase contrast microscope after being treated with 5-HT. Original
   magnification, 200 ×. Scale bar = 200 μm. (b, c) Transwell assays (b)
   of H1299 and A549 cells with 5-HT treatment. Original magnification,
   100 ×. Scale bar = 100 μm. Wound healing assays (c) for the evaluation
   of H1299 and A549 cells' migration ability. Original magnification,
   50 ×. Scale bar = 200 μm. (d, e) Protein levels were assessed by
   western blot (d) and immunofluorescence (e) after treatment with 5-HT.
   Original magnification, 200 ×. Scale bar = 25 μm. β-actin was used as
   an internal reference for the western blot. **P < 0.01

5-HT promotes migration and invasion of NSCLC cells by downregulating SNRPG

   RNA-seq analysis of 5-HT-treated H1299 cells identified differentially
   expressed genes (DEGs) potentially mediating migration and invasion in
   NSCLC cells (Table S2). KEGG pathway enrichment analysis highlighted
   spliceosome pathways as significant (Fig. [77]3a). Validation of
   candidate DEGs mRNA expression in 5-HT-treated H1299 cells using
   qRT-PCR revealed a notable downregulation of SNRPG mRNA (Fig. S3a).
   Western blot analysis further confirmed reduced SNRPG protein levels in
   both H1299 and A549 cells following 5-HT treatment (Fig. [78]3b),
   suggesting its involvement in regulating NSCLC metastasis and
   warranting further investigation. SNRPG was transiently silenced and
   overexpressed in H1299 and A549 cells using SNRPG siRNA and
   overexpression plasmids, respectively. Transfection efficiency was
   assessed at both mRNA and protein levels via qRT-PCR and Western blot
   analysis (Fig. S3b, c). Silencing SNRPG induced a morphological shift
   from epithelioid to spindled morphology in both A549 and H1299 cells
   (Fig. [79]3c) and significantly enhanced migration and invasion
   compared to controls (Fig. [80]3d, e). Conversely, SNRPG overexpression
   inhibited migration and invasion in A549 and H1299 cells (Fig. S3d, e).
   Immunoblotting and immunofluorescence analyses revealed changes in EMT
   markers: reduced E-cadherin and elevated N-cadherin, vimentin, and
   Snail expression upon SNRPG silencing, with the opposite pattern upon
   overexpression (Figs. [81]3f, g and S3f, g). These results highlight
   the critical role of SNRPG in modulating the migration, invasion and
   EMT of NSCLC cells. To determine whether 5-HT-induced promotion of
   NSCLC cell migration and invasion occurs through SNRPG downregulation,
   rescue experiments were performed. H1299 and A549 cells were
   co-transfected with SNRPG overexpression plasmids and treated with
   5-HT. Western blot confirmed the SNRPG protein expression
   (Fig. [82]3h). Combination treatment markedly attenuated 5-HT-induced
   migration, invasion and EMT (Figs. [83]3h, i and S3h). These results
   emphasize the role of 5-HT in promoting NSCLC cell migration, invasion,
   and EMT progression through SNRPG downregulation.

Fig. 3.

   [84]Fig. 3.
   [85]Open in a new tab

   5-HT promotes migration and invasion of NSCLC cells by downregulating
   SNRPG. (a) KEGG pathway enrichment analysis of RNA-seq. The numbers on
   the bars represent the differential genes involved. (b) Protein levels
   of SNRPG were assessed after being treated with 5-HT. (c) The
   morphological transformation of H1299 and A549 cells was examined after
   being transfected with siSNRPG. Original magnification, 200 ×. Scale
   bar = 200 μm. (d) Wound healing assays for the evaluation of SNRPG
   knockdown on H1299 and A549 cells' migration ability. Original
   magnification, 50 ×. Scale bar = 200 μm. (e) Transwell assays of H1299
   and A549 cells with transient SNRPG knockdown. Original magnification,
   100 ×. Scale bar = 100 μm. (f, g) Protein levels of EMT markers were
   assessed by western blot (f) and immunofluorescence (g). Original
   magnification, 200 ×. Scale bar = 25 μm. (h) Protein levels of SNRPG
   and EMT markers were assessed after treating with 5-HT and infecting
   with SNRPG overexpression plasmid by western blot. β-actin was used as
   an internal reference. (i) The migration and invasion of H1299 and A549
   cells were determined by the Transwell assay after treating with 5-HT
   and infecting with SNRPG overexpression plasmid. **P < 0.01,
   ***P < 0.001

   Additionally, Western blot analysis showed that 20 μM 5-HT treatment
   for 72 h suppressed SNRPG expression and promoted EMT in BEAS-2B cells,
   a normal lung epithelial cell line. Silencing SNRPG in BEAS-2B cells
   also promoted the EMT process (Fig. S3i). These results suggest that
   the 5-HT/SNRPG regulatory axis influences the epithelial-mesenchymal
   plasticity of both normal lung epithelial cells and NSCLC cells,
   providing insights into the early molecular events in NSCLC
   progression.

5-HT promotes NSCLC metastasis by downregulating SNRPG in vivo

   To evaluate the in vivo effects of altering SNRPG expression and
   modulating 5-HT levels, a stable SNRPG overexpression cell line was
   established in H1299-Luc cells (Fig. S4a). A mouse lung metastasis
   model was then created by intravenously injecting H1299 cells (NC) and
   H1299 cells overexpressing SNRPG (SNRPG_OE). After establishing
   xenograft mouse lung metastases, the mice were administered
   intraperitoneal injections of PBS or 5-HT every other day for a total
   of 10 injections and were monitored in real-time using in vivo imaging.
   After euthanasia, lung and liver tissues were harvested (Fig. [86]4a).
   An increase in the number of lung metastases was observed in the
   NC + 5-HT group, with some mice also exhibiting liver metastases
   compared to the NC + PBS group (Fig. [87]4b, c). This finding indicates
   that 5-HT promotes lung metastasis colonization in vivo and facilitates
   liver metastasis, consistent with previous observations of elevated
   peripheral blood 5-HT levels in patients with metastatic NSCLC. In
   contrast, the number of lung metastases and the weight of liver
   metastatic nodules decreased in the SNRPG_OE + PBS group relative to
   the NC + PBS group. In addition, compared with the NC + 5-HT group, the
   number of lung metastases in the SNRPG_OE + 5-HT group decreased.
   Representative macroscopic and live imaging further illustrated lung
   and liver metastases (Figs. [88]4c and S4b). Histological analysis
   confirmed the NSCLC metastasis-promoting effect of 5-HT (Fig. [89]4b).
   Ki-67 staining showed tumor formation in the lungs of all mice
   (Fig. [90]4d). Immunohistochemistry (IHC) revealed the highest SNRPG
   and E-cadherin expression in the SNRPG_OE + PBS group and the lowest in
   the NC + 5-HT group. However, expression levels were slightly higher in
   the SNRPG_OE + 5-HT group compared to the NC + 5-HT group. Conversely,
   vimentin showed the opposite pattern (Fig. [91]4d). In summary, these
   results suggest that SNRPG functions as a protective factor against
   NSCLC by mitigating the pro-metastatic effects of 5-HT in vivo.

Fig. 4.

   [92]Fig. 4.
   [93]Open in a new tab

   5-HT promotes NSCLC metastasis by downregulating SNRPG in vivo. (a)
   Illustration of the steps of the animal studies. (b) The representative
   images of HE staining of lung tissue of mice in each group (n = 5).
   Scale bar = 25 µM. (c) Macroscopic images of lungs and livers from a
   representative mouse from each group (n = 5). (d) Representative images
   of IHC staining in the lung specimens from mice in each group (n = 5).
   Scar bar = 25 µM. *P < 0.05, **P < 0.01, ns: not significant

SNRPG inhibits NSCLC metastasis by downregulating CDK14

   The mechanism by which SNRPG inhibits NSCLC metastasis was further
   explored. RNA-seq analysis of H1299 cells with SNRPG silencing
   identified key downstream effectors. Comparative analysis revealed
   significant transcriptional misregulation in cancer-related pathways
   (Fig. [94]5a), with DEGs listed in Table S3. Among upregulated genes,
   qRT-PCR confirmed increased CDK14 and PER2 mRNA in both A549 and H1299
   cells following SNRPG silencing, consistent with RNA-seq data (Fig.
   S5a). Given its stronger upregulation and reported association with
   lung cancer metastasis [[95]26], CDK14 was selected for further
   analysis. Western blot analysis confirmed a substantial increase in
   CDK14 protein levels after SNRPG silencing (Fig. [96]5b). Additionally,
   combined SNRPG knockdown and 5-HT treatment further elevated CDK14
   expression (Fig. S5b). IHC staining of mouse lung sections showed
   higher CDK14 staining in the NC + PBS group compared to the
   SNRPG_OE + PBS group (Fig. S5c). Kaplan–Meier survival analysis
   demonstrated that high CDK14 expression in patients with NSCLC was
   associated with poor prognosis (Fig. S5d). qRT-PCR and Western blot
   confirmed the transfection efficiencies of CDK14 silensing and
   overexpression in A549 and H1299 cells (Fig. S5e, f). After CDK14
   overexpression, H1299 and A549 cells exhibited a mesenchymal,
   spindle-shaped morphology, with the change being more pronounced in
   H1299 cells (Fig. [97]5c). Wound healing assays showed reduced closure
   in the siCDK14 group (Fig. S5g), and transwell assays revealed
   diminished migration and invasion upon CDK14 knockdown (Fig. [98]5d),
   while CDK14 overexpression enhanced these capabilities (Fig. S5h, i).
   Altered expression of EMT markers was also observed upon manipulation
   of CDK14 (Figs. [99]5e, f and S5j, k).

Fig. 5.

   [100]Fig. 5
   [101]Open in a new tab

   SNRPG inhibits NSCLC metastasis by downregulating CDK14. (a) KEGG
   pathway enrichment analysis of RNA-seq. The numbers on the bars
   represent the differential genes involved. (b) CDK14 expression was
   detected by western blot after transfecting with siSNRPG in H1299 and
   A549 cells. (c) The morphologic transformation was examined after
   infecting with CDK14 overexpression plasmid. Original magnification,
   200 ×. Scale bar = 200 μm. (d) Cell migration and invasion of H1299 and
   A549 cells were determined by the Transwell assay after transfecting
   with siCDK14. Original magnification, 100 ×. Scale bar = 100 μm. (e, f)
   Protein levels of CDK14 and EMT markers were assessed by western blot
   (e) and immunofluorescence (f) after transfecting with siCDK14.
   Original magnification, 200 ×. Scale bar = 25 μm. (g) Cell migration
   and invasion were determined after transfecting with siSNRPG and
   siCDK14. Original magnification, 100 ×. Scale bar = 100 μm. (h) Protein
   levels were assessed after transfecting with siSNRPG and siCDK14. (i)
   Cell migration and invasion were determined after treating with 5-HT
   and CDK14 siRNA. Original magnification, 100 ×. Scale bar = 200 μm. (j)
   Protein levels were assessed between the 5-HT + siCDK14 group and the
   5-HT group. β-actin was used as an internal reference for the western
   blot. *P < 0.05, **P < 0.01, ***P < 0.001

   Rescue experiments confirmed that SNRPG inhibits NSCLC cell migration
   and invasion through CDK14 downregulation. Co-silencing SNRPG and CDK14
   significantly reduced migration and invasion compared to the siSNRPG
   group (Figs. [102]5g and S5l). Western blot analysis revealed elevated
   E-cadherin expression and decreased levels of N-cadherin, vimentin, and
   Snail upon simultaneous silencing of SNRPG and CDK14, compared to the
   siSNRPG group (Fig. [103]5h), indicating a reversal of the EMT process
   and inhibition of NSCLC metastasis. In addition, GSEA analysis of the
   RNA-seq data showed a negative correlation between SNRPG expression and
   EMT (Fig. S5m). These results indicate that SNRPG inhibits the
   migration, invasion and EMT of NSCLC cells by down-regulating CDK14.
   Further functional experiments demonstrated that siCDK14 could reverse
   the effects of 5-HT in promoting migration, invasion, and EMT in NSCLC
   cells (Fig. [104]5i, j). These results suggest that 5-HT promotes the
   migration, invasion and EMT of NSCLC cells by up-regulating CDK14.

SNRPG negatively regulates WT1 to suppress CDK14 transcription and inhibits
NSCLC cell migration and invasion

   To elucidate the mechanism by which SNRPG regulates CDK14 expression,
   Co-IP was performed in A549 cells, but no direct interaction between
   SNRPG and CDK14 was detected (Fig. S6a), implying that SNRPG may affect
   the function of CDK14 through indirect regulation. The CDK14 promoter
   sequence was retrieved from the UCSC database, and potential
   transcription factors were predicted using the PROMO and JASPAR
   databases ([105]http://jaspar.genereg.net/). Venn analysis identified
   WT1 and SP1 as candidate transcription factors through which SNRPG may
   regulate CDK14 expression (Fig. [106]6a). SNRPG knockdown increased WT1
   but decreased SP1 protein levels (Fig. [107]6b). qRT-PCR showed
   unchanged WT1 mRNA levels but reduced SP1 mRNA levels following SNRPG
   knockdown in both H1299 and A549 cells (Fig. S6b). Regarding the
   observed discrepancy between WT1 mRNA and protein expression, it is
   speculated that post-translational modifications may be involved. Co-IP
   confirmed a direct interaction between SNRPG and WT1 in A549 cells
   (Fig. [108]6c). However, confocal imaging revealed that SNRPG was
   localized in the cytoplasm, while WT1 was localized in the nucleus of
   A549 cells (Fig. [109]6d). To further investigate the effect of SNRPG
   knockdown on WT1 subcellular localization, subcellular fractionation
   and immunofluorescence staining experiments were performed. WT1
   expression was upregulated in both the nucleus and cytoplasm following
   SNRPG silencing (Fig. [110]6e, f). This implies that SNRPG may regulate
   the protein abundance of WT1 by affecting WT1 stability, translational
   efficiency, or nucleoplasmic shuttling. This regulation may occur in
   the cytoplasm and affect the transcriptional activity of WT1 by
   influencing the total amount of WT1 entering the nucleus. Then,
   correlation analysis using GEPIA and TCGA databases revealed a negative
   correlation between SNRPG and WT1 expression, and a positive
   correlation between WT1 and CDK14 expression (Fig. S6c). In both H1299
   and A549 cells, WT1 knockdown not only reduced intracellular CDK14
   expression but also reversed the increase in CDK14 levels caused by
   SNRPG knockdown (Fig. [111]6g, h). Kaplan–Meier survival analysis
   showed that high WT1 expression was associated with poor prognosis in
   patients with NSCLC (Fig. S6d). qRT-PCR and Western blot were used to
   determine the transfection efficiencies of WT1 silencing and
   overexpression in A549 and H1299 cells (Fig. S6e, f). WT1 knockdown
   upregulated E-cadherin expression and decreased N-cadherin, vimentin,
   and Snail expression (Fig. S6g, h). Rescue experiments demonstrated
   that WT1 knockdown inhibited NSCLC cell migration and invasion, while
   co-silencing of SNRPG and WT1 reversed the enhanced migration and
   invasion observed with SNRPG knockdown alone (Fig. [112]6i). Through
   JASPAR database prediction, we identified potential WT1 transcription
   factor binding sequences in the CDK14 promoter region (Fig. [113]6j).
   To assess the regulatory role of the WT1 transcription factor on the
   CDK14 promoter, a dual-luciferase reporter system was employed. Results
   showed that in the group co-transfected with pGL4-CDK14 and
   pcDNA3.1-WT1, the Fluc/Rluc ratio was significantly higher than in the
   empty vector control group (pcDNA3.1), indicating that WT1 specifically
   activates the CDK14 promoter. In contrast, when pGL4-mCDK14 was used,
   the Fluc/Rluc ratio significantly decreased even with WT1
   overexpression, confirming that WT1 regulation of CDK14 is dependent on
   specific binding sites within its promoter (Fig. [114]6k). Similarly,
   ChIP analysis revealed that, compared to the IgG control, WT1 was
   significantly enriched at the CDK14 promoter (Fig. [115]6l).

Fig. 6.

   [116]Fig. 6
   [117]Open in a new tab

   SNRPG negatively regulates WT1 to suppress CDK14 transcription and
   inhibits NSCLC cell migration and invasion. (a) Venn analysis for
   identifying the transcription factor of CDK14. (b) WT1 and SP1
   expression was detected by western blot after transfection with
   siSNRPG. (c) Immunoprecipitation using a SNRPG antibody showed SNRPG
   and WT1 association. (d) Colocalization between SNRPG and WT1 was
   examined by confocal microscopy. Scale bar = 20 µm. Nucleus were
   visualized with DAPI (blue). (e) Protein expressions of WT1 in the
   nucleus and cytoplasm were detected after knockdown of SNRPG in H1299
   cells and A549 cells. (f) Protein levels of WT1 were assessed by
   immunofluorescence after transfection with siSNRPG. (g) Protein levels
   of CDK14 were detected by western blot after transfection with siWT1.
   (h) Protein levels were assessed after transfection with siSNRPG and
   siWT1. (i) Cell migration and invasion were determined after
   transfection with siSNRPG and siWT1. Original magnification, 100 ×.
   Scale bar = 100 μm. (j) The potential WT1 binding motifs in the CDK14
   promoter region. (k) A dual-luciferase reporter gene assay was used to
   evaluate changes in luciferase activity. (l) ChIP assay. β-actin was
   used as an internal reference for the western blot. **P < 0.01,
   ***P < 0.001

   These findings indicate that SNRPG negatively regulates the protein
   level of WT1 through directly binding to WT1, thereby weakening the
   transcriptional activation effect on the CDK14 promoter, and ultimately
   inhibiting the migration, invasion, EMT and metastasis of NSCLC cells.
   These results reveal the regulatory mechanism of the SNRPG/WT1/CDK14
   axis in NSCLC metastasis, highlighting the value as potential
   therapeutic targets in NSCLC.

Discussion

   Approximately 90% of deaths in patients with NSCLC result from tumor
   metastasis worldwide [[118]27]. Despite therapeutic advances,
   overcoming NSCLC metastasis remains a significant challenge.
   Identifying therapeutic targets is crucial to improving patient
   outcomes. In this study, patients with metastatic NSCLC exhibited a
   broader spectrum of gut microbiota metabolites, particularly L-Trp, and
   enrichment in the serotonergic synapse pathway. Furthermore, peripheral
   blood 5-HT levels positively correlated with fecal Trp content in NSCLC
   patients. The gut microbial influence Trp metabolism, predominantly via
   the kynurenine and 5-HT pathways [[119]28]. 5-HT can be produced
   directly by microorganisms using intracavity tryptophan or tryptophan
   expression synthase [[120]29]. Compared to specific pathogen-free mice,
   gut ECs in germ-free (GF) mice exhibit larger morphology, implying that
   gut microbes may influence the development and function of ECs
   [[121]30]. Tryptophan hydroxylase (TPH) 1, predominantly expressed by
   ECs, suggests that gut microbiota can modulate the 5-HT pathway through
   TPH. For example, Clostridium ramosum enhances 5-HT production in colon
   ECs [[122]31]. TPH1 knockout (TPH1^−/−) mice exhibit a significant
   reduction in 5-HT concentrations [[123]32], with notable differences in
   microbial composition between TPH1^−/− and TPH1^+/- littermates
   [[124]33]. Therefore, we speculate that the gut microbiota of patients
   with NSCLC metastasis may facilitate 5-HT production through Trp
   metabolism.

   5-HT is linked to tumor initiation, progression, and prognosis. It
   serves as a sensitive marker for liver metastasis in colorectal cancer
   [[125]34], promotes prostate cancer cell migration via MAPK and
   PI3K/AKT pathways [[126]35], and enhances lung cancer cell
   proliferation via HTR activation [[127]36]. This study first links 5-HT
   to liver metastasis in NSCLC, where it enhances migration and invasion,
   thereby facilitating EMT, underscoring its potential as both a
   prognostic marker and a therapeutic target [[128]37, [129]38].

   However, a key limitation of this study is the inability to establish
   causality between elevated 5-HT levels and NSCLC metastasis. While
   elevated 5-HT levels were observed in metastatic patients, the
   cross-sectional design cannot confirm whether these levels precede
   metastasis or result from it. Future prospective longitudinal studies
   tracking 5-HT levels from diagnosis through disease progression are
   needed to clarify this relationship. Interventional studies targeting
   the 5-HT pathway could provide further evidence of causality.
   Additionally, the small sample size limits the statistical power, and
   larger, more homogeneous cohorts are required to confirm these findings
   and explore the role of peripheral 5-HT in cancer progression.

   To elucidate the downstream regulatory mechanism through which 5-HT
   promotes NSCLC metastasis, RNA-seq was performed on NSCLC cells exposed
   to 5-HT, implicating SNRPG as a potential effector. Located on
   chromosome 2p13.3, SNRPG is a core component of uridine-rich snRNP
   complexes, serving as a precursor to spliceosomes [[130]16, [131]39].
   Despite limited research on the role of SNRPG in tumorigenesis,
   previous studies have hinted at its relevance in breast cancer, NSCLC,
   and colorectal cancer, with Lan et al. suggesting its oncogenic
   potential, particularly in glioblastoma [[132]16, [133]40, [134]41]. In
   the present study, a novel protective role for SNRPG was identified in
   NSCLC. While other pathways showed higher statistical significance in
   the KEGG analysis, SNRPG was prioritized based on its consistent and
   robust expression changes in our validation experiments. Additionally,
   the potential connection between SNRPG and 5-HT-mediated NSCLC
   metastasis represents a novel and unexplored area of research, which
   may provide valuable insights into the mechanisms driving cancer
   progression. Our research results indicated that overexpression of
   SNRPG significantly impeded NSCLC metastasis and partially countered
   5-HT-mediated metastasis, highlighting its pivotal role in 5-HT-induced
   NSCLC metastasis. These divergent roles of SNRPG underscore its
   functional diversity across various cancer types. Further experiments
   demonstrated that 5-HT treatment suppressed SNRPG expression and
   promoted EMT in normal lung epithelial cells. SNRPG silencing also
   promoted EMT in normal lung epithelial cells, suggesting that the
   5-HT/SNRPG axis also influences epithelial-mesenchymal plasticity in
   normal lung epithelial cells. These findings may provide insights into
   the early molecular mechanisms that contribute to NSCLC progression.

   To clarify the mechanism by which SNRPG inhibits the metastasis of
   NSCLC, we conducted RNA-seq analysis on SNRPG knockdown cells.
   Integrating KEGG pathway enrichment with experimental verification, we
   identified CDK14 as a key target. CDK14, also known as PFTK1, is a
   member of the cell division cycle 2-related protein kinase family and
   plays a pivotal role in regulating the cell cycle during proliferation
   [[135]42]. Its overexpression in NSCLC, esophageal cancer, ovarian
   cancer, and gliomas has been associated with enhanced tumor
   proliferation, metastasis, and poor clinical outcomes [[136]26,
   [137]43]. Silencing CDK14 reduces migration, invasion, and EMT
   progression in NSCLC cells [[138]26], consistent with our current
   study. Interestingly, rescue experiments revealed that CDK14 plays a
   critical role in SNRPG-mediated suppression of metastasis, despite no
   direct protein binding between SNRPG and CDK14, implying that SNRPG may
   affect the function of CDK14 through indirect regulation. In addition,
   5-HT promoted the malignant phenotype of NSCLC cells by upregulating
   CDK14. These new findings strengthen our understanding of the
   mechanisms underlying the role of the 5-HT/SNRPG/CDK14 axis in
   regulating NSCLC cell behavior and provide a theoretical basis for the
   development of therapeutic strategies targeting this pathway.

   Based on the screening of transcription factors and functional
   validation, our study revealed that WT1 was the core transcription
   factor by which SNRPG regulated the expression of CDK14. WT1 encodes a
   zinc finger transcription factor that is upregulated in various tumors,
   including lung cancer [[139]44–[140]47]. While Chen et al. found that
   WT1 inhibited E-cadherin expression in NSCLC cells, thereby enhancing
   their invasive and migratory capabilities, and Xu et al. reported that
   WT1 promoted NSCLC cell proliferation by upregulating Cyclin D1 and
   p-pRb expression [[141]48, [142]49]. Our study suggested that silencing
   SNRPG in NSCLC cells increased WT1 protein levels without affecting
   mRNA levels. Co-IP results confirmed that SNRPG directly interacted
   with WT1, but co-localization experiments revealed that SNRPG was
   localized in the cytoplasm, while WT1 was located in the nucleus.
   Further subcellular fractionation and immunofluorescence staining
   showed that in NSCLC cells, the protein levels of WT1 in both the
   nucleus and cytoplasm were up-regulated when SNRPG was knocked down.WT1
   is known to shuttle between the nucleus and cytoplasm, and its
   different isoforms are involved in various aspects of gene expression
   regulation, including transcription and RNA processing
   [[143]50–[144]52]. Therefore, we speculate that in the cytoplasm, WT1
   may interact with SNRPG. This interaction may affect the protein
   stability, translation, or nuclear-cytoplasmic shuttling of WT1 to
   regulate the protein abundance of WT1, thereby influencing its
   distribution and function within cells. Nuclear accumulation of WT1 may
   enhance its binding to the CDK14 promoter, activating the
   transcriptional activity of CDK14 and thereby promoting the malignant
   phenotype of NSCLC cells. This was further supported by dual-luciferase
   reporter assays demonstrating that WT1 specifically activated the CDK14
   promoter through defined binding sites, and ChIP experiments showing
   significant enrichment of WT1 at the CDK14 promoter. Importantly, WT1
   knockdown not only reduced CDK14 expression but also reversed the
   increase in CDK14 caused by SNRPG knockdown, establishing the complete
   regulatory pathway. Future work will explore how SNRPG regulates the
   protein level of WT1, as well as the specific function of WT1 in the
   cytoplasm.

   This study has several limitations. Our in vivo experiments only used
   the SNRPG overexpression model and lacked knockdown model validation.
   Future research should include stable knockdown or knockout of SNRPG in
   animal models to confirm the anti-cancer function and therapeutic
   potential. Additionally, the absence of comparisons with other cancers
   or non-cancer diseases limits our ability to determine whether elevated
   fecal Trp and peripheral blood 5-HT were specific biomarkers for NSCLC
   metastasis. Future studies should include diverse patient cohorts
   across different cancer types and stages, as well as individuals with
   inflammatory or metabolic disorders, to assess the specificity and
   sensitivity of these potential biomarkers. While correlations between
   fecal Trp, peripheral blood 5-HT, and NSCLC metastasis are observed,
   further experiments, including fecal microbiota transplantation or
   metagenomic analysis, are needed to directly identify that gut
   microbiota promotes 5-HT production via tryptophan metabolism.

   In conclusion, we demonstrate that gut microbiota may regulate 5-HT
   biosynthesis through Trp metabolism. Elevated 5-HT significantly
   inhibits the expression of SNRPG and deregulates its negative
   regulation of WT1 protein. Accumulated WT1 transcriptionally activates
   CDK14, driving the EMT process and accelerating NSCLC metastasis
   (Fig. [145]7). These important findings provide new perspectives and
   potential intervention targets for the prevention and treatment of
   NSCLC metastasis.

Fig. 7.

   [146]Fig. 7
   [147]Open in a new tab

   Illustration of the potential mechanisms by which 5-HT induces NSCLC
   metastasis. In NSCLC patients, gut microbiota may promote Trp
   metabolism to generate 5-HT. 5-HT suppresses SNRPG expression, thereby
   releasing the inhibition of WT1 protein. Subsequent WT1 activation
   stimulates CDK14 transcription, promoting EMT, and enhances metastatic
   capabilities of NSCLC cells—including migration, invasion, and distant
   dissemination

Materials and methods

Patients

   Ethical approval for this study was granted by the Ethics Committee of
   the Second Affiliated Hospital of Dalian Medical University (No.
   2023–132). A total of 152 participants were enrolled from December 2022
   to May 2023, comprising 56 healthy individuals and 96 patients with
   NSCLC. Metabolomics analysis was conducted on 32 patients with NSCLC,
   while ELISA testing was performed on all participants. Demographic data
   and criteria are provided in Tables S4, S5, and Supplementary materials
   and methods.

Fecal and blood sample collection from cancer individuals

   Fecal and blood samples were collected prior to the first treatment.
   Stool samples were preserved in 2-ml cryovials, transported in liquid
   nitrogen, and stored at − 80 °C. Blood samples were drawn into
   centrifuge tubes, allowed to clot and separate at 37 °C for 1 h, and
   then subjected to two rounds of centrifugation at 3000 rpm and
   12,000 rpm for 10 min at room temperature. The resulting supernatant
   was collected as serum and stored at − 80 °C.

Untargeted metabolomic profiling

   This work was supported by Biomarker Technologies (Beijing, China).
   High-resolution mass spectrometry was performed using the Waters Xevo
   G2-XS QTOF mass spectrometer, which collects both primary and secondary
   mass spectrometry data in MSe mode under the control of MassLynx V4.2
   acquisition software. In each data acquisition cycle, dual-channel data
   were collected at low and high collision energies simultaneously. The
   low collision energy was set to 2 V, and the high collision energy
   range was 10–40 V, with a scanning frequency of 0.2 s per mass
   spectrum. The ESI ion source parameters were as follows: Capillary
   voltage: 2000 V (positive ion mode) or −1500 V (negative ion mode);
   cone voltage: 30 V; ion source temperature: 150 °C; desolvent gas
   temperature: 500 °C; backflush gas flow rate: 50 L/h; desolventizing
   gas flow rate: 800 L/h. Detailed experimental procedures are provided
   in Supplementary materials and methods.

RNA sequencing analysis

   This work was supported by Biomarker Technologies (Beijing, China).
   Total RNA from cell samples was extracted using the TRIzol Reagent
   (Life Technologies, California, USA), following the manufacturer’s
   instructions. RNA concentration and purity were assessed using a
   NanoDrop 2000 (Thermo Fisher Scientific, Wilmington, DE). RNA integrity
   was evaluated with the RNA Nano 6000 Assay Kit on the Agilent
   Bioanalyzer 2100 system (Agilent Technologies, CA, USA). Sequencing
   libraries were constructed using the Hieff NGS Ultima Dual-mode mRNA
   Library Prep Kit for Illumina (Yeasen Biotechnology, Shanghai, China),
   in accordance with the manufacturer’s instructions. Index codes were
   added to attribute sequences to specific samples. PCR amplification was
   performed with Phusion High-Fidelity DNA Polymerase, universal PCR
   primers, and index (X) primers. Following PCR, products were purified
   using the AMPure XP system, and library quality was assessed on the
   Agilent Bioanalyzer 2100 system. Libraries were sequenced on the
   Illumina NovaSeq6000 platform (PE150 sequencing platform, Origin of San
   Diego), generating 150 bp paired-end reads as per the manufacturer’s
   instructions. Raw reads were processed using a bioinformatics pipeline
   tool on the BMKCloud online platform ([148]www.biocloud.net). Detailed
   experimental procedures are provided in Supplementary materials and
   methods.

Lentivirus expressing transduction

   NC and SNRPG overexpression lentiviral vectors (SNRPG_OE) were
   purchased from Genechem (Shanghai, China). The lentiviral generation
   system employed a second-generation approach, with a plasmid to
   packaging vector to envelope ratio of 30:16:12, as specified by the
   manufacturer. The 293T cell Line was used as the interim cell Line.
   Lentiviral particle collection details can be found in Supplementary
   materials and methods. The viral titer was 2× 10^9 TU. H1299 cells were
   transfected with either NC or SNRPG_OE lentivirus. Stable transfection
   was selected using puromycin (2 μg/mL), and the cells were cultured
   with 1 μg/mL puromycin following selection.

Animal experiments

   Female NYG mice (4–6 weeks old) were purchased from Liaoning Changsheng
   Biotechnology Co., Ltd. (Liaoning, China), with an average initial body
   weight of 20.3 ± 2.5 g. The mice were housed in a barrier system and
   fed irradiated purified feed. They were randomly divided into four
   groups: NC + PBS, NC + 5-HT, SNRPG_OE + PBS, and SNRPG_OE + 5-HT, with
   5 mice per group. A mouse lung metastasis model was established via
   tail vein injection. Cells were resuspended in PBS, and 600× 10^4
   SNRPG-overexpressing cells or control cells were injected into each
   group. Lung colonization was monitored weekly using bioluminescence
   imaging (BLI). Mice were anesthetized with Tribromoethanol (#MA0478,
   2.5%, Meilunbio) at a dose of 750 mg/kg. After anesthesia, D-luciferin
   potassium salt D (Yeasen Biotechnology, Shanghai, China) was
   administered via intraperitoneal injection, and images were captured
   using the IVIS Spectrum Xenogen system (PerkinElmer, MA, USA).
   Bioluminescence images were analyzed using Living Image software
   (version 2.50). 5-HT salt (0.5 mg/kg; Selleck, Shanghai, China) was
   administered intraperitoneally every other day, starting after
   confirmation of lung colonization, as previously described [[149]53].
   Lung metastases were monitored weekly using BLI for 4 weeks. Euthanasia
   was carried out by cervical dislocation. The study was approved by the
   Animal Ethics Review Committee of Dalian Medical University (No.
   [150]AEE23068).

Immunohistochemistry (IHC)

   Paraffin sections (3 μm thick) were deparaffinized, rehydrated, and
   incubated in a 3% H[2]O[2] methanol solution for endogenous peroxidase
   activity inhibition. Antigen retrieval was performed with sodium
   citrate antigen retrieval solution (50X, #C1032, Solarbio Science &
   Technology, Beijing, China). The sections were blocked with 5% BSA
   (Bioss) for 2 h and then incubated overnight at 4 °C with the following
   primary antibodies: anti-SNRPG (1:50 dilution; Proteintech), anti-CDK14
   (1:1000 dilution; Proteintech), anti-E-cadherin (1:400 dilution; Cell
   Signaling Technology), anti-vimentin (1:200 dilution; Cell Signaling
   Technology), and anti-Ki-67 (1:500 dilution; Abcam, UK). IHC was
   performed using the UltraSensitive™ SP IHC Kit (#KIT-9710,
   Biotechnologies, Fuzhou, China) and DAB solution (#DA1010, Solarbio
   Science & Technology), following incubation with biotinylated secondary
   antibodies. DAB solution was applied to the tissue, and positive
   brown-yellow staining was observed under a microscope. The reaction was
   terminated by rinsing with water, and the tissue was counterstained
   with hematoxylin (#KGA223, Key Gen Biotech, China) for 40 s, followed
   by differentiation with differentiation solution (#C0161S, Beyotime
   Biotechnology) for 10 s, with both steps followed by rinsing with
   running water. After dehydration and clearing, the tissue was mounted
   with neutral balsam (#G8590, Solarbio Science & Technology). IHC
   results were independently evaluated by two blinded pathologists and
   scored. The immunoreactive score (IRS), proposed by Remmele and Stegner
   [[151]54], was used to quantify the staining.

Luciferase reporter assay

   293T, A549, and H1299 cells were seeded in 6-well plates and
   transfected with pGL4-basic (empty vector), pGL4-CDK14 (WT1-binding
   wild-type CDK14 promoter), or pGL4-mCDK14 (mutant promoter), along with
   the phRL-TK Renilla luciferase plasmid as an internal control. After
   48 h, the cells were washed with ice-cold PBS, lysed, and centrifuged
   at 12,000 rpm for 10 min. The supernatants were analyzed for both
   Firefly and Renilla luciferase activities, and the data were normalized
   to Firefly/Renilla relative light unit (RLU) ratios. Three biological
   replicates were performed to assess the WT1-mediated transcriptional
   regulation of CDK14.

Chromatin Immunoprecipitation (ChIP)

   Cells were harvested and cross-linked with 1% formaldehyde (v/v) at
   room temperature for 10 min. Following cross-linking, the cells were
   resuspended in cell lysis buffer containing protease inhibitors and
   centrifuged to remove the supernatant and isolate the nuclei. The
   nuclear pellet was resuspended in lysis buffer and incubated on ice for
   10 min. Chromatin was sheared by sonication into fragments, and
   fragment size was verified by electrophoresis using a small aliquot of
   the sample. The sonicated lysate was centrifuged (4 °C, 12,000 g,
   10 min), and the supernatant was collected and diluted tenfold with
   ChIP dilution buffer. A portion of the sample was reserved as the
   "Input" control. Target protein-specific antibodies were added, and
   incubation was carried out with rotation at 4 °C overnight (isotype IgG
   served as an NC). Pre-equilibrated protein G-agarose beads were added,
   and incubation was continued for an additional hour at 4 °C. The
   antibody-protein-DNA complexes were then eluted from the beads.
   Proteinase K was added to both the bead-bound complexes and Input
   samples, and the mixtures were incubated at 65 °C for 2 h. Finally, DNA
   was purified using a DNA purification kit.

Data analysis

   Data were analyzed using GraphPad Prism version 9 software. All
   experiments were conducted with at least three biological replicates,
   and all quantitative data were presented as mean ± standard deviation.
   Statistical significance between two groups was assessed using a
   Student’s unpaired t-test. Correlation analysis was performed using
   Spearman's correlation analysis. A P-value of < 0.05 was considered
   statistically significant (*P < 0.05, **P < 0.01, ***P < 0.001).

Supplementary Information

   [152]Supplementary Material 1.^ (4.4MB, docx)

Acknowledgements