Figure 1.
In vivo evidence of thiaminase anticancer activity.
A. A Kaplan-Meyer plot of time to pre-defined tumor volume endpoint for subcutaneous MCF-7 breast cancer xenografts treated with thiaminase 2000 units SC QOD or buffer control. The median time to endpoint was 41 days for untreated control and 59 days (p = 0.03 Log rank test). B. A Kaplan-Meyer plot of time to pre-defined tumor volume endpoint for subcutaneous RS4 leukemia subcutaneous xenografts treated with thiaminase 850 units SC BIW or buffer control. The median time to endpoint was 16.5 days for the control group and not reached after 60 days of observation in the treated group (p<0.001 Log rank test). C. Primary ALL and AML specimens were thawed and plated in triplicate in two concentrations of thiaminase 0.4 units/ml and 4 units/ml, and assessed for viability at 48 hours relative to untreated cells. The ALL specimen with the asterisk was used for the in vivo study shown in Figure 1D. D. Primary ALL cells were injected IV on Day 1; three thiaminase treatments of 2000 units/kg SC were administered on days 17, 20 and 24. The animals were sacrificed on Day 33 and bone marrow was examined by flow cytometry for human ALL cells (percent human CD45); Untreated n = 4; vehicle treated n = 10, native thiaminase n = 8 (* p<0.01, Mann-Whitney test).
Figure 2.
Effect of thiamine antagonists on thiaminase activity.
A. Growth inhibition of Reh and RS4 leukemia cells in the presence or absence of thiaminase and either N3PT or oxythiamine. B. Growth inhibition of RS4 and Reh leukemia cells incubated in normal medium, thiamine-free medium or medium containing thiaminase prior to exposure to different concentrations of N3PT. C. RS4 subcutaneous xenografts showing untreated control, N3PT alone, 1k-PEGylated thiaminase and 1k-PEGylated thiaminase followed by N3PT. The time-to-endpoint was 16.5 days for control, 23 days for N3PT, 25 days for 1k-PEG thiaminase and 55 days for 1k-PEG thiaminase (p<0.01 Log rank test). D. Oxygen consumption rate of RS4 cells treated with thiaminase, N3PT or thiaminase followed by N3PT (clear: control; thin-stripe: thiaminase; thick- stripe: N3PT; solid: thiaminase + N3PT).
Figure 3.
OCR and ECAR of leukemia and breast cell lines incubated for 48(clear: control; thin-stripe: thiaminase; thick- stripe: rapamycin; solid: thiaminase + rapamycin).
A. Quantification of OCR parameters. The ATP-linked rate is the basal rate minus the rate measured after the addition of oligomycin. The maximal capacity is the rate measured after the addition of FCCP. The reserve capacity is the basal rate minus the FCCP rate. All data are the mean ± SEM, of triplicate samples and are representative of 3 independent experiments (†p<0.05, *p<0.01 two way ANOVA (Newman Kruskal Wallis test). B. Quantification of ECAR parameters. Glycolysis is the rate determine from subtracting the basal rate from the rate after the addition of glucose. Glycolytic reserve is the rate determined by subtracting the rate following the addition of oligomycin from the rate following the addition of glucose. All data are the mean ± SEM, of triplicate samples and are representative of 3 (†p<0.05, *p<0.01 two way ANOVA (Newman Kruskal Wallis test).
Figure 4.
Validation of thiaminase action in of RS4 leukemia cells and MCF-7 breast cancer cells by metabolomic analysis.
Both cell lines were analyzed under six conditions: control for 24 hours (C-24); incubation in thiaminase for 24 hours (T-24); control for 48 hours (C-48); thiaminase for 48 hours (T-48); rapamycin for 48 hours (R-48); and both rapamycin and thiaminase for 48 hours (R+T-48). The median is indicated by the bar in the center of the rectangle, the rectangle dimensions reflect the range of the two mid-quartile values, and the outer bars represent the ranges of all of the values. The data represent four independent experiments. For C-48 vs T-48 and C-48 vs T+R-48 comparisons, ** indicates p<0.05 and * indicates 0.05<p<0.1). A. Internal validation in RS4 cells, showing the expected decrease in thiamine and thiamine diphosphate in thiaminase treated cells, an increase in thiazole, the product of thiaminase cleavage of thiamine in the thiaminase-treated cells and the appearance of rapamycin only in the rapamycin treated cells. B. Validation of thiaminase-mediated inhibition of transketolase by demonstration of the accumulation of substrate surrogates ribose and ribulose in both RS4 and MCF-7 cell lines at 48 hours. C. Immunoblot analysis of thiamine pyrophosphate kinase (TPK1) and thiamine triphosphatase (THTPA) in RS4 and MCF-7 cells treated with thiaminase, rapamycin or both. D. Immunoblot analysis of pyruvate kinase isozymes M1 and M2 and carnitine palmitoyl transferase (CPT1) in RS4 and MCF-7 cells treated with thiaminase, rapamycin or both.
Figure 5.
Differential effects of thiaminase on branched chain amino acid catabolism.
A. Schematic diagram showing branched chain amino acid catabolism. B. Different metabolomic signatures indicating inhibition of BCKDH by thiaminase. The top three panels show accumulation of BCKDH substrates after 48-7 cells after thiaminase treatment- most notably isovalerylcarnitine, which is also reversed by rapamycin. C. An immunoblot of cytosolic and mitochondrial branched chain amino acid transferase (cBCAT and mBCAT), the enzymes that catalyzes the reactions that produce BCKDH substrates, and total and phosphorylated BCKDH subunit E1 (BCKD-E1 and pBCKD-E1, respectively) in RS4 leukemia cells treated with thiaminase, rapamycin or both. For C-48 vs T-48 and C-48 vs T+R-48 comparisons, ** indicates p<0.05 and * indicates 0.05<p<0.1). D. Cytotoxicity assay of RS4 and MCF-7 cells treated with increasing concentrations of thiaminase (in milliunits, mu) under control conditions or in medium that contains gabapentin.
Figure 6.
Effects of thiaminase on aromatic amino acid catabolism.
A. Accumulation of the products of phenylalanine (phenylpyruvate and phenyllactate) and tyrosine (hydroxyphenylpyruvate and hydroxyphenyllactate) catabolism in RS4 cells after treatment with thiaminase for 48 hours. B. Accumulation of tryptophan catabolites after thiaminase treatment, showing accumulation of indolelactate but not kynurenine. For C-48 vs T-48 and C-48 vs T+R-48 comparisons, ** indicates p<0.05 and * indicates 0.05<p<0.1).