Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 24:13:RP95652.
doi: 10.7554/eLife.95652.

Metabolite profiling of human renal cell carcinoma reveals tissue-origin dominance in nutrient availability

Keene L Abbott #  1   2   3 , Ahmed Ali #  2   3 , Bradley I Reinfeld #  4   5   6 , Amy Deik  3 , Sonu Subudhi  7 , Madelyn D Landis  5 , Rachel A Hongo  5 , Kirsten L Young  5 , Tenzin Kunchok  8 , Christopher S Nabel  2   9   10 , Kayla D Crowder  8 , Johnathan R Kent  11 , Maria Lucia L Madariaga  11 , Rakesh K Jain  7 , Kathryn E Beckermann  5 , Caroline A Lewis  8 , Clary B Clish  3 , Alexander Muir  12 , W Kimryn Rathmell  5   13 , Jeffrey Rathmell  13   14 , Matthew G Vander Heiden  1   2   3   15
Affiliations

Metabolite profiling of human renal cell carcinoma reveals tissue-origin dominance in nutrient availability

Keene L Abbott et al. Elife. .

Abstract

The tumor microenvironment is a determinant of cancer progression and therapeutic efficacy, with nutrient availability playing an important role. Although it is established that the local abundance of specific nutrients defines the metabolic parameters for tumor growth, the factors guiding nutrient availability in tumor compared to normal tissue and blood remain poorly understood. To define these factors in renal cell carcinoma (RCC), we performed quantitative metabolomic and comprehensive lipidomic analyses of tumor interstitial fluid (TIF), adjacent normal kidney interstitial fluid (KIF), and plasma samples collected from patients. TIF nutrient composition closely resembles KIF, suggesting that tissue-specific factors unrelated to the presence of cancer exert a stronger influence on nutrient levels than tumor-driven alterations. Notably, select metabolite changes consistent with known features of RCC metabolism are found in RCC TIF, while glucose levels in TIF are not depleted to levels that are lower than those found in KIF. These findings inform tissue nutrient dynamics in RCC, highlighting a dominant role of non-cancer-driven tissue factors in shaping nutrient availability in these tumors.

Keywords: cancer; cancer biology; human; metabolism; tumor microenvironment.

Plain language summary

Cancer cells convert nutrients into energy differently compared to healthy cells. This difference in metabolism allows them to grow and divide more quickly and sometimes to migrate to different areas of the body. The environment around cancer cells – known as the tumor microenvironment – contains a variety of different cells and blood vessels, which are bathed in interstitial fluid. This microenvironment provides nutrients for the cancer cells to metabolize, and therefore influences how well a tumor grows and how it might respond to treatment. Recent advances with techniques such as mass spectrometry, which can measure the chemical composition of a substance, have allowed scientists to measure nutrient levels in the tumor microenvironments of mice. However, it has been more difficult to conduct such studies in humans, as well as to compare the tumor microenvironment to the healthy tissue the tumors arose from. Abbott, Ali, Reinfeld et al. aimed to fill this gap in knowledge by using mass spectrometry to measure the nutrient levels in the tumor microenvironment of 55 patients undergoing surgery to remove kidney tumors. Comparing the type and levels of nutrients in the tumor interstitial fluid, the neighboring healthy kidney and the blood showed that nutrients in the tumor and healthy kidney were more similar to each other than those in the blood. For example, both the tumor and healthy kidney interstitial fluids contained less glucose than the blood. However, the difference between nutrient composition in the tumor and healthy kidney interstitial fluids was insignificant, suggesting that the healthy kidney and its tumor share a similar environment. Taken together, the findings indicate that kidney cancer cells must adapt to the nutrients available in the kidney, rather than changing what nutrients are available in the tissue. Future studies will be required to investigate whether this finding also applies to other types of cancer. A better understanding of how cancer cells adapt to their environments may aid the development of drugs that aim to disrupt the metabolism of tumors.

PubMed Disclaimer

Conflict of interest statement

KA, AA, BR, AD, SS, ML, RH, KY, TK, KC, JK, MM, KB, CL, CC, AM, WR, JR No competing interests declared, CN Royalty and income from Cambridge Epigenetix and ThermoFisher and stock in Opko Health, RJ Consultant/SAB fees from Cur, DynamiCure, Elpis, SPARC, SynDevRx; owns equity in Accurius, Enlight, SynDevRx; served on the Board of Trustees of Tekla Healthcare Investors, Tekla Life Sciences Investors, Tekla Healthcare Opportunities Fund, Tekla World Healthcare Fund, and received Research Grants from Boehringer Ingelheim and Sanofi; no funding or reagents from these organizations were used in the study, MV Scientific advisor for Agios Pharmaceuticals, iTeos Therapeutics, Sage Therapeutics, Pretzel Therapeutics, Lime Therapeutics, Faeth Therapeutics, Droia Ventures, and Auron Therapeutics

Figures

Figure 1.
Figure 1.. Levels of metabolites in renal cell carcinoma (RCC) tumor interstitial fluid (TIF) are similar to those found in normal kidney interstitial fluid (KIF).
(A) Schematic depicting study design whereby samples collected from patients with RCC undergoing nephrectomy were used to derive TIF, KIF, and plasma. Samples were then subjected to polar metabolomics and lipidomics analyses. See Supplementary file 1 for patient information, and Supplementary file 2 for metabolite concentrations. (B) Principal component analysis of polar metabolites measured from the indicated RCC patient samples (n = 55 patients). For each sample, absolute levels of 98 polar metabolites were quantified by liquid chromatography/mass spectrometry (LC/MS). Data represent 55 TIF, 46 KIF, and 27 plasma samples. The 95% confidence interval is displayed. (C) Principal component analysis of lipid species measured from the indicated RCC patient samples (n = 38 patients). For each sample, relative levels of 195 lipids were assessed by LC/MS. Data represent 34 TIF, 25 KIF, and 18 plasma samples. The 95% confidence interval is displayed. (D) T-test analysis of polar metabolites (n = 98) that do or do not significantly differ in concentration between each site from all RCC patient samples (n = 55 patients). Cutoffs of |log2 fold change| >1 and adjusted p-value (false discovery rate-corrected) <0.05 were used to determine significant metabolites. p-values in the plot are derived from chi-squared statistical analysis. (E) T-test analysis of lipids (n = 195) that do or do not significantly differ in concentration between each site from all RCC patient samples (n = 38 patients). Cutoffs of |log2 fold change| >1 and adjusted p-value (false discovery rate-corrected) <0.05 were used to determine significant metabolites. p-values in the plot are derived from chi-squared statistical analysis. Panel A created with BioRender.com, and published using a CC BY-NC-ND license with permission.
Figure 1—figure supplement 1.
Figure 1—figure supplement 1.. Levels of metabolites in clear cell renal cell carcinoma (RCC) interstitial fluid are similar to those found in normal kidney interstitial fluid (KIF).
(A) Principal component analysis of polar metabolites measured from the indicated clear cell RCC patient samples (n = 41 patients). For each sample, absolute levels of 98 polar metabolites were quantified by liquid chromatography/mass spectrometry (LC/MS). Data represent 36 tumor interstitial fluid (TIF), 28 KIF, and 20 plasma samples. The 95% confidence interval is displayed. (B) Principal component analysis of lipid species measured from the indicated clear cell RCC patient samples (n = 28 patients). For each sample, relative levels of 195 lipids were assessed by LC/MS. Data represent 25 TIF, 18 KIF, and 15 plasma samples. The 95% confidence interval is displayed. (C) T-test analysis of polar metabolites (n = 98) that do or do not significantly differ in concentration between each site from clear cell RCC patient samples (n = 41 patients). Cutoffs of |log2 fold change| >1 and adjusted p-value (false discovery rate-corrected) <0.05 were used to determine significant metabolites. p-values in the plot are derived from chi-squared statistical analysis. (D) T-test analysis of lipids (n = 195) that do or do not significantly differ in concentration between each site from all clear cell RCC patient samples (n = 28 patients). Cutoffs of |log2 fold change| >1 and adjusted p-value (false discovery rate-corrected) <0.05 were used to determine significant metabolites. p-values in the plot are derived from chi-squared statistical analysis.
Figure 2.
Figure 2.. Assessment of metabolites that differ between renal cell carcinoma (RCC) interstitial fluid and normal kidney interstitial fluid (KIF).
Volcano plots depicting the log2 fold change in polar metabolite concentration (A) or relative lipid levels (B) between tumor interstitial fluid (TIF) and KIF from RCC patients (n = 55 patients in [A], n = 38 patients in [B]). Cutoffs of |log2 fold change| >1 and adjusted p-value (false discovery rate-corrected) <0.05 were used to select significantly altered metabolites. Metabolites or lipids highlighted in red and blue are significantly higher and lower in TIF compared to KIF, respectively. Full names of selected lipids: PC(O-34:2), PC(P-34:1)/PC(O-34:2); PC(O-40:7), PC(P-40:6)/PC(O-40:7); LPC(O-18:1), LPC(P-18:0)/LPC(O-18:1); PC(O-34:1), PC(P-34:0)/PC(O-34:1); PC(O-34:4), PC(P-34:3)/PC(O-34:4); PC(O-36:1), PC(P-36:0)/PC(O-36:1). (C–E) Levels of selected metabolites measured by liquid chromatography/mass spectrometry (LC/MS) in plasma, TIF, and KIF from matched RCC patients (n = 10 patients). Each point represents a value measured from one patient, and the red line represents the mean across all patients considered. p-values were derived from either mixed-effects analysis (kynurenine, glutathione, glucose) or repeated measures one-way analysis of variance (ANOVA) (2-hydroxyglutarate, lactate, arginine, citrulline, ornithine), depending on whether missing values were present, and were Tukey multiple comparisons corrected (ns, not significant; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001). (F) Normalized peak area values of cholesterol measured by LC/MS in plasma, TIF, and KIF from matched RCC patients (n = 6 patients). Each point represents a sample, and the red line represents the mean across all patients considered. p-values were derived from repeated measures one-way ANOVA with Tukey multiple comparisons correction (ns, not significant; ****p < 0.0001). Relative abundance of cholesterol (G) or cholesteryl esters (H) in TIF compared to KIF from matched RCC patients (n = 20 patients). The mean is presented ± standard error of the mean (SEM), and the black dotted line indicates a ratio of 1, representing no difference in lipid levels between TIF and KIF. p-values were derived from a one sample t-test compared to 1 (*p < 0.05).
Figure 2—figure supplement 1.
Figure 2—figure supplement 1.. Heatmaps of metabolites that differ between renal cell carcinoma (RCC) interstitial fluid and normal kidney interstitial fluid (KIF).
Heatmaps depicting relative concentrations of the indicated polar metabolites (A) or lipids (B) that differ between KIF and tumor interstitial fluid (TIF). Data within each row were Z-score normalized.
Figure 2—figure supplement 2.
Figure 2—figure supplement 2.. Assessment of metabolites that differ between plasma in patients with renal cell carcinoma (RCC) with plasma from normal individuals and from patients with non-small cell lung cancer (NSCLC).
(A, B) Volcano plots depicting the log2 fold change in polar metabolite concentration measured in plasma from patients with RCC (n = 27) compared to that measured in plasma from healthy adults (n = 10) (A), or measured in plasma from patients with RCC compared to that measured in patients with NSCLC (n = 20). Cutoffs of |log2 fold change| >1 and adjusted p-value (false discovery rate-corrected) <0.05 were used to select significantly altered metabolites. Metabolites highlighted in red and blue are significantly higher and lower in RCC plasma, respectively. Heatmaps depicting relative concentrations of the indicated polar metabolites that differ between healthy and RCC plasma (C) or between NSCLC and RCC plasma (D). Data within each row were Z-score normalized.
Figure 2—figure supplement 3.
Figure 2—figure supplement 3.. Plasma cystine concentration is affected by fasting.
(A) Concentration of cystine as reported in the Human Metabolome Database (HMDB) (n = 5) or measured in plasma from healthy adults (normal, n = 10), from renal cell carcinoma (RCC) patients (n = 27), or non-small cell lung cancer (NSCLC) patients (n = 20). Each point represents a sample from patient, and the red line represents mean across all samples considered. p-values were derived from Brown–Forsythe and Welch analysis of variance (ANOVA) test with Dunnett’s T3 multiple comparisons correction (ns, not significant; ***p < 0.001; ****p < 0.0001). (B) Concentration of cystine measured by liquid chromatography/mass spectrometry (LC/MS) in plasma from an overnight (>9.5 hr) fasted healthy adult and plasma from the same adult ~4 hr after eating (fed).
See this image and copyright information in PMC

Update of

Similar articles

See all similar articles

Cited by

References

    1. Abbott KL, Ali A, Casalena D, Do BT, Ferreira R, Cheah JH, Soule CK, Deik A, Kunchok T, Schmidt DR, Renner S, Honeder SE, Wu M, Chan SH, Tseyang T, Stoltzfus AT, Michel SLJ, Greaves D, Hsu PP, Ng CW, Zhang CJ, Farsidjani A, Kent JR, Madariaga MLL, Gramatikov IMT, Matheson NJ, Lewis CA, Clish CB, Rees MG, Roth JA, Griner LM, Muir A, Auld DS, Vander Heiden MG. Screening in serum-derived medium reveals differential response to compounds targeting metabolism. Cell Chemical Biology. 2023;30:1156–1168. doi: 10.1016/j.chembiol.2023.08.007. - DOI - PMC - PubMed
    1. Apiz Saab JJ, Dzierozynski LN, Jonker PB, AminiTabrizi R, Shah H, Menjivar RE, Scott AJ, Nwosu ZC, Zhu Z, Chen RN, Oh M, Sheehan C, Wahl DR, Pasca di Magliano M, Lyssiotis CA, Macleod KF, Weber CR, Muir A. Pancreatic tumors exhibit myeloid-driven amino acid stress and upregulate arginine biosynthesis. eLife. 2023;12:e81289. doi: 10.7554/eLife.81289. - DOI - PMC - PubMed
    1. Apiz Saab JJ, Muir A. Tumor interstitial fluid analysis enables the study of microenvironment-cell interactions in cancers. Current Opinion in Biotechnology. 2023;83:102970. doi: 10.1016/j.copbio.2023.102970. - DOI - PMC - PubMed
    1. Bergers G, Fendt SM. The metabolism of cancer cells during metastasis. Nature Reviews. Cancer. 2021;21:162–180. doi: 10.1038/s41568-020-00320-2. - DOI - PMC - PubMed
    1. Birsoy K, Possemato R, Lorbeer FK, Bayraktar EC, Thiru P, Yucel B, Wang T, Chen WW, Clish CB, Sabatini DM. Metabolic determinants of cancer cell sensitivity to glucose limitation and biguanides. Nature. 2014;508:108–112. doi: 10.1038/nature13110. - DOI - PMC - PubMed