Taurine and proline promote lung tumour growth by co-regulating Azgp1/mTOR signalling pathway

Abstract

Accurate metabolic biomarkers for lung cancer prognosis remain scarce but crucial. Taurine and proline, two metabolites, are consistently elevated across various cancer stages in previous studies, hinting at their potential role in disease progression. This study is the first to reveal how these metabolites contribute to poor prognosis. Transcriptomic analysis uncovered that taurine and proline downregulated Zinc-α2-glycoprotein (Azgp1), a gene linked to key metabolic pathways. Additionally, Azgp1 could also significantly affect downstream lipid metabolic pathways in lung cancer. Both taurine and proline influenced lipid metabolism via mammalian target of rapamycin (mTOR). When Azgp1 was overexpressed, lung cancer progression slowed significantly, alongside reduced mTOR activity. These findings underscore the pro-cancer role of taurine and proline, highlighting the Azgp1/mTOR axis as a vital, yet overlooked, pathway in lung cancer. This study not only advances our understanding but also identifies new therapeutic avenues.

Fig. 1. LLC-xenografted mouse model preliminarily reveals the tumour-promoting potential of taurine and proline.

LLC cells were injected by subcutaneously into wild type C57BL/6 mice (5 × 10⁵ cells per mouse). On day 7, when xenografts reached a size of approximately 30 mm³, mice were randomly assigned to the model (n = 5), proline (n = 5), L-kynurenine (n = 5), and taurine groups (n = 5). Then mice received repeated drug intervention with either vehicle (PBS), taurine (25 mg/kg, i.p., A), proline (25 mg/kg, i.p., B), or L-kynurenine (25 mg/kg, i.p., C) every day on the 7th day (D). Body weight (E) and tumour volumes (F) of mice were examined every 3 days. After dissection, tumour weight (G) and tumour size (H) were significantly increased in taurine- or proline-treated group except L-kynurenine-treated group compared to the model group according to representative images of tumours from mice. And the organ/body ratios (I) was measured and calculated. Statistical significance was calculated by means of one-way ANOVA analysis. All data are presented as the mean ± SEM, *p < 0.05, **p < 0.01^, and ***p < 0.001.

Fig. 2. Orthotopic lung cancer mouse model further closer to clinical features confirms the tumour-promoting potency of taurine and proline.

Mice were then randomly assigned to the control group (n = 10), model group (n = 11), taurine group (n = 10), and proline group (n = 11). LLC cells were injected by lung parenchyma (3 × 10⁵ per mouse, A) into wild type C57BL/6 mice. Then mice received repeated drug intervention with either vehicle (PBS), or taurine (25 mg/kg, i.p) or proline (25 mg/kg, i.p) every day on the third day. Mice were weighed every 3 days for both mouse models (B). Tumour development in mice was monitored by micro-CT every 5 days (C). After dissection, the taurine and proline group revealed more macroscopic nodules in the lung nodules than the model group (D). Weights of lungs (E) and other organ/body ratios (F) of all mice were measured and calculated. In the micro-CT scans, green arrows point to mouse hearts in the different sections and red arrows point to tumours. Statistical significance was calculated by means of one-way ANOVA analysis for body weight, lung weight, and lung/body ratio. All data are presented as the mean ± SEM, *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 3. Azgp1 gene plays a vital role in the tumour-promoting effect caused by taurine and proline.

The gene expression profiles were established by comparing the differentially expressed genes (DEGs) of lung tumours from the taurine (25 mg/kg, n = 3), proline (25 mg/kg, n = 3), model (n = 3) and control (n = 3) group. In heat-map analysis, the comparison of the top 66 and top 10 regulatory genes of tumours revealed in control vs. model vs. proline (A) and control vs. model vs. taurine groups (B), respectively. In the histogram, 4 DEGs including Cxcl9, Zbp1, Hmgcs2, and Azgp1 were ultimately identified in control vs. model vs. proline groups (C). And 4 DEGs including Azgp1, Cd3d, Cd8a and Cyp2d10 were ultimately identified in control vs. model vs. taurine groups (D). The criteria of |log₂FC| ≥ 1 and p < 0.05 was utilised to identify the DEGs with biological significance. Azgp1 was down-regulated in human primary LUSC (E) and LUAD (F). Azgp1 mRNA expression was compared from 515 human primary LUAD tumours (or 503 human primary LUSC) and 59 (or 52) normal human lung tissues available at the TCGA database. Lung tumours with low Azgp1 expression correlated with poor survival (G). Statistical significance was calculated by means of one-way ANOVA or t-test analysis, *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 4. Tumour-promoting effects of taurine and proline are closely associated with the alteration of lipid metabolism mediated by Azgp1.

The metabolite profiles were generated by extracting and quantifying metabolites from lung tumours using UPLC-MS/MS across four groups: taurine (25 mg/kg, n = 3), proline (25 mg/kg, n = 3), model (n = 3), and control (n = 3). Statistics for metabolites analysed by a first stage of mass spectrometry (MS1) revealed significant up-regulation or down-regulation between different groups (A). Specifically, heat-map and volcano plot preliminarily illustrated hierarchical clustering of metabolites analysed by MS1 with significant regulation in abundance between model vs. control (B, E), taurine vs. model (C, F), and proline vs. model (D, G), respectively. Of these, metabolites analysed by a second stage of mass spectrometry (MS2) were further identified by heat-map analysis with significant regulation in abundance between taurine vs. model (H) and proline vs. model (I), respectively. Finally, the histogram summarised regulatory metabolites of MS2 between taurine vs. model (J) and proline vs. model (K), respectively. The significance of differences in metabolite concentrations between two phenotypes was calculated using two-tailed unpaired Student’s t test. The p value was adjusted for multiple tests using an FDR (Benjamini–Hochberg), *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 5. Azgp1 is significantly correlated with downstream lipid metabolic pathways in lung cancer patients.

Azgp1 is significantly correlated with lipid metabolic pathways in 1017 lung cancer samples (including LUAD and LUSC) according to GSEA (A–H). Enrichment scores and p value was calculated in default parameters. |NES| > 1, NOM p val < 0.05, FDR q val < 0.25 were considered to be significant functional pathways. mTOR was up-regulated in human primary LUAD (I) and LUSC (J). mTOR mRNA expression was compared from 515 human primary LUAD tumours (or 503 human primary LUSC) and 59 (or 53) normal human lung tissues available at the TCGA database. mTOR, p-mTOR, Azgp1, and β-actin protein expressions were evaluated by western blot assay (K, n = 3). The significance of differences was calculated using two-tailed unpaired Student’s t test, *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 6. Overexpression of Azgp1 potentially inhibits lung cancer progression.

LLC cells were transduced by lentivirus, and the resistant cells were selected with corresponding antibiotics to obtain mouse Azgp1 overexpression stable cell lines and the control virus transfected cell lines (A). In vitro experiments, quantitative RT-PCR was conducted to determine the expressions of Azgp1 in LLC (wt) and LLC (Azgp1 overexpression) cells (B). Western blot assay was applied to confirm that gene Azgp1 had been overexpressed (C). LLC (wt) and LLC (Azgp1 overexpression) cells viability for 72 h were detected by MTT assays (D). The colony formation and statistical analysis of LLC (wt) and LLC (Azgp1 overexpression) cells for 72 h (E). Experiments were independently repeated four times. Experiments were independently repeated three times. In vivo experiments, LLC (wt) or LLC (Azgp1 overexpression) cells were injected by subcutaneously into wild type C57BL/6 mice (5 × 10⁵ cells per mouse, n = 6 mice per group, F). Then body weight (G) and tumour volumes (H) of mice were examined every 3 days. After dissection, the representative images of tumours (I) and tumour weights (J) were measured and calculated. mTOR, p-mTOR, Azgp1, and β-actin protein expressions in tumours were evaluated by western blot assay (K, n = 3 mice per group). Statistical significance was calculated by means of ANOVA analysis. All data are presented as the mean ± SEM, *p < 0.05, **p < 0.01, and ***p < 0.001.

Fig. 7. Azgp1/mTOR axis is identified as an under-reported pathway for taurine and proline in promoting lung cancer progression.

Taurine and proline may inhibit AZGP1 function, triggering mTOR pathway activation and lipid metabolism disorder that ultimately promote lung cancer progression.