Circulating succinate-modifying metabolites accurately classify and reflect the status of fumarate hydratase-deficient renal cell carcinoma

Abstract

Germline or somatic loss-of-function mutations of fumarate hydratase (FH) predispose patients to an aggressive form of renal cell carcinoma (RCC). Since other than tumor resection there is no effective therapy for metastatic FH-deficient RCC, an accurate method for early diagnosis is needed. Although MRI or CT scans are offered, they cannot differentiate FH-deficient tumors from other RCCs. Therefore, finding noninvasive plasma biomarkers suitable for rapid diagnosis, screening, and surveillance would improve clinical outcomes. Taking advantage of the robust metabolic rewiring that occurs in FH-deficient cells, we performed plasma metabolomics analysis and identified 2 tumor-derived metabolites, succinyl-adenosine and succinic-cysteine, as excellent plasma biomarkers for early diagnosis. These 2 molecules reliably reflected the FH mutation status and tumor mass. We further identified the enzymatic cooperativity by which these biomarkers are produced within the tumor microenvironment. Longitudinal monitoring of patients demonstrated that these circulating biomarkers can be used for reporting on treatment efficacy and identifying recurrent or metastatic tumors.

Keywords: Cancer; Genetic diseases; Metabolism; Molecular diagnosis; Oncology.

Figure 1. Pathological characteristics of the RCC cohort.

(A) Kaplan-Meier analysis of OS following diagnosis for patients with stage I/II FH-deficient RCC versus those with stage III/IV FH-deficient RCC (log-rank test, P = 0.007). (B) Expression of mutant FH proteins in FH-null mouse cells. (C) Relative in vitro enzymatic activity of mutant FH proteins normalized to WT activity. The experiments were performed independently 3 times. All data are presented as the mean ± SEM. (D) H&E and IHC staining of 2SC and FH tissue from patient PID332 p.N154H. Scale bars: 200 μm and 50 μm (insets). (E) Negative correlation (cor.) between intracellular fumarate (Fum) levels and variant FH enzymatic activity. (F) Migratory capability of cells in Matrigel. Scale bar: 600 μm. (G) Negative correlation between cellular migration and variant FH enzymatic activity in cells. norm., normalized.

Figure 2. Identification of potential liquid biopsy biomarkers in RCC.

(A) PCA of plasma samples (1,637 features) from NC individuals and patients with RCC categorized according to FH mutation status, tumor mass, and disease stage (represented by the size of each dot). (B) Venn diagram of altered plasma metabolites of FH-MT versus NC samples (blue) and FH-MT versus FH-WT samples (yellow) and tumor size correlation analysis (red). (C) The top 20 ROCAUC-ranked plasma metabolites discriminating FH-MT from FH-WT and NC samples. (D) Heatmap classification of FH-MT, FH-WT, and NC samples based on the metabolites in C. Scale bar: log₂(normalized abundance). (E) Regularized partial correlation network of significantly altered metabolites in B. Each node represents a metabolite, and each edge represents the strength of the partial correlation coefficient between 2 compounds that were mapped into biochemical pathways. The size of each circle represents the strength of the correlation with tumor burden.

Figure 3. Association between FH deficiency and circulating biomarker levels in mice with CDX tumors.

(A and B) Schematic overview of the generation of orthotopic kidney tumor xenografts by implantation of transformed epithelial kidney cells with the genotypes HRAS^G12V Fh1^Δ/Δ (n = 5 mice implanted) and HRAS^G12V Fh1^fl/fl (n = 5 mice implanted). Individual tumor sizes at the endpoints were documented. Scale bar: 1 cm. (C) Longitudinal monitoring of the indicated plasma metabolites in mice bearing either WT FH tumors (HRAS^G12V Fh1^fl/fl, n = 5) or FH-deficient tumors (HRAS^G12V Fh1^Δ/Δ, n = 5). (D) Correlation between the maximal measured level of each metabolite during the course of the study and tumor volume (cm³) at the endpoint in each HRAS^G12V Fh1^Δ/Δ–engrafted mouse. Mal, malate; Rel., relative.

Figure 4. Relation between FH-MT and plasma biomarker levels in mice with PDX tumors.

(A) Schematic overview of the generation of subcutaneous xenografts by transplantation of human kidney tumors with the genotypes FH-WT ccRCC (n = 5 mice engrafted) or FH-MT RCC (n = 5 mice engrafted). Surgical resection of the subcutaneous FH-MT PDX was performed upon reaching the endpoint, and mice were sacrificed 1 week after the resection surgery. (B) Growth rate of each PDX tumor of the indicated genotype. (C) Representative images of individual tumors at the endpoints. (D and E) Longitudinal monitoring of the indicated plasma metabolites in mice bearing a PDX of either a FH-WT ccRCC tumor (n = 5 mice) or a FH-deficient tumor (n = 5 mice). (F) Comparisons of plasma metabolites over time in mice engrafted with FH-MT tumors (n = 5). All data are presented as the mean ± SEM. *P < 0.05 and **P < 0.01, by paired, 2-tailed Student’s t test, with P value–adjusted Bonferroni correction. Conc., concentration.

Figure 5. The biochemical conversion of plasma molecules in mice.

(A) Schematic representation of the in vivo metabolic conversion of labeled intravenously injected metabolites in mice. LC/MS, liquid chromatography/mass spectrometry. (B–D) Mice were injected with 1,4-¹³C₂-fumarate (B), ¹³C₃¹⁵N₁ suc-cys (C), or ¹³C₂¹⁵N₁ suc-ado (D), and the injected metabolic tracer as well as its metabolic products were analyzed. (E and F) In vitro metabolic conversion of fumarate into malate in mouse plasma (E) or human plasma (F) incubated with 1,4-¹³C₂-fumarate for the indicated duration. In each assay, the buffer was used as a nonenzyme control. All experiments were performed independently 3 times. All data are presented as the mean ± SEM.

Figure 6. Circulating suc-cys is a product of the enzymatic cascade of the GGT1-DPEP1 axis.

(A) Schematic representation of intravenous injection of suc-GSH into mice and the metabolic fate in vivo. (B) Analysis of the in vitro assay of suc-GSH in mouse plasma. (C) Analysis of the in vitro enzymatic conversion of suc-GSH into suc-cys-gly by the recombinant proteins. (D) Analysis of the in vitro enzymatic conversion of suc-cys-gly into suc-cys by the recombinant human proteins. (E) Assessment of the ability of the indicated mouse tissue homogenates to catabolize the conversion of suc-GSH into suc-cys-gly and suc-cys. (F) mRNA levels of different transpeptidases and dipeptidases in various mouse tissues (normalized to liver) were determined by quantitative PCR (qPCR). (G) Immunoblotting was used to measure GGT1 and DPEP1 expression. All experiments were performed independently 3 times. All data are presented as the mean ± SEM.

Figure 7. Plasma suc-cys is generated by GGT1-DPEP1 cooperativity in the kidney.

(A and B) Representative IHC staining for GGT1 (A) and DPEP1 (B) in healthy human kidney tissue showing strong plasma membrane and cytoplasmic positivity in renal tubular cells. (C and D) Representative IHC staining for GGT1 (C) and DPEP1 (D) in FH-deficient kidney tumor tissue and adjacent normal tissue from 1 patient. All experiments were performed independently 3 times. (E) A workflow of suc-cys generation from suc-GSH via peptidases. Scale bars: 800 μm (left) and 50 μm (right) (images in A–D).

Figure 8. Circulating suc-ado and suc-cys levels can be used to accurately identify FH-deficient RCC.

(A) Box plot analysis showing the distribution of plasma metabolites across NC, FH-MT, and FH-WT. Wilcoxon rank-sum test P values with Bonferroni correction were calculated. (B and C) Logistic regression ROCAUC analyses of metabolites in FH-MT versus NC samples (B) and FH-MT versus FH-WT samples (C). (D) Spearman’s correlation coefficient between tumor volume (log₂ mm³) and potential plasma biomarker levels (log₂ ng/mL). (E) Scatter plot analysis showing the distribution of the levels of plasma metabolites between FH-MT carriers (no tumor), NCs, and patients with FH-WT or FH-MT RCC. All data are presented as the mean ± SEM. **P < 0.01, ***P < 0.001, and ****P < 0.0001, by unpaired, 2-tailed Student’s t test, with P value–adjusted Bonferroni correction. (F) Scatter plot analysis showing the distribution of the levels of plasma metabolites across NCs and patients with FH-WT, FH-MT stage I/II, or FH-MT stage III/IV. All data are presented as the mean ± SEM. ****P < 0.0001, by unpaired, 2-tailed Student’s t test, with P value–adjusted Bonferroni correction.

Figure 9. Plasma suc-ado and suc-cys levels can be used to accurately monitor tumor progression.

(A–C) Longitudinal monitoring of the dynamics of plasma suc-ado and suc-cys during multiple clinical events in 3 patients: PID522 (A), PID555 (B), and PID563 (C). In each panel, the top row shows the distinct time of each clinical event, the middle row shows examples of medical imaging at the indicated clinical events, and the lower row shows the “wave plot,” the height of which represents the concentration of plasma metabolites at the indicated time point.