A Novel l-Asparaginase with low l-Glutaminase Coactivity Is Highly Efficacious against Both T- and B-cell Acute Lymphoblastic Leukemias In Vivo

Abstract

Acute lymphoblastic leukemia (ALL) is the most common type of pediatric cancer, although about 4 of every 10 cases occur in adults. The enzyme drug l-asparaginase serves as a cornerstone of ALL therapy and exploits the asparagine dependency of ALL cells. In addition to hydrolyzing the amino acid l-asparagine, all FDA-approved l-asparaginases also have significant l-glutaminase coactivity. Since several reports suggest that l-glutamine depletion correlates with many of the side effects of these drugs, enzyme variants with reduced l-glutaminase coactivity might be clinically beneficial if their antileukemic activity would be preserved. Here we show that novel low l-glutaminase variants developed on the backbone of the FDA-approved Erwinia chrysanthemi l-asparaginase were highly efficacious against both T- and B-cell ALL, while displaying reduced acute toxicity features. These results support the development of a new generation of safer l-asparaginases without l-glutaminase activity for the treatment of human ALL.Significance: A new l-asparaginase-based therapy is less toxic compared with FDA-approved high l-glutaminase enzymes Cancer Res; 78(6); 1549-60. ©2018 AACR.

Conflict of interest statement

Disclosure of Potential Conflicts of Interest

H.A. Nguyen, A.M. Schalk, Y. Su. and A. Lavie declare competing financial interest by being founders with equity stake in Enzyme by Design, Inc. a startup developing new L-asparaginases. Y. Saunthararajah is on the Scientific Advisory Board of Enzyme by Design, Inc. All other authors declare no competing financial interests.

©2018 American Association for Cancer Research.

Figures

Fig. 1. Engineered ErA variants have reduced L-glutaminase activity

(A) NMR spectroscopy was used to monitor the ability of various L-asparaginases to hydrolyze Gln. The starting Gln concentration (600 µM; labeled as 100%) was chosen to reflect physiological Gln levels. All enzymes were added to the same final concentration of 50 nM, which depending on molecular weight and L-asparaginase rate translates to 0.2 – 0.8 IU/ml. Under these conditions, ErA-WT fully hydrolyses the Gln in an hour (red trace). In contrast, ErA-DM exhibits negligible Gln hydrolysis (green trace), reduced even compared to EcA-WT (gray trace). (B) To further compare the L-glutaminase activity of the ErA variants relative to EcA-WT, an experiment was conducted at a higher enzyme concentration relative to that shown in panel A. Enzyme amounts were based on matched L-asparaginase activity level; with all enzymes at 15 IU/ml, EcA-WT completely depletes Gln within 45 minutes, whereas for both ErA-DM and ErA-TM, Gln levels are only reduced by 20% after 1 hour.

Fig. 2. The low L-glutaminase ErA-TM eliminates T-ALL LOUCY cells as effectively as the high L-glutaminase ErA-WT and with reduced toxicity

(A) Female mice tail vein-injected with luciferase-expressing LOUCY cells four weeks prior were treated daily with vehicle (n=3); ErA-WT (n=4); and ErA-TM (n=4) for 14 days (drug dose 50 IU/mouse/day; i.p.). For each group, the representative animal shown had the highest BLI signal at day 0 of treatment. BLIs from all animals are presented in Supplementary Fig. 4 and 6. The average BLI signal of each group at day 7 and 14 relative to the value at day 0 (day 0 = 1) was plotted with mean and standard deviation (SD). See Supplementary Biostatistics on imaging for detailed standard error analysis. (B) Relative BLI flux at day 14 between the vehicle, ErA-WT, and ErA-TM groups. The flux for the vehicle mice increased 10-fold relative to day 0. For both treated groups, the flux decreased dramatically relative to vehicle control (p-value <0.0001), returning to background levels by day 14, with no significant (ns) difference between the treated groups. Mean with SD were plotted. See Supplementary Biostatistics on imaging for detailed standard error analysis. (C) PB %huDC45+ levels were determined one week prior to treatment initiation (day -7), at treatment start (day 0), and at end of treatment (day 14). At day 0, all animals were highly engrafted, as indicated by %huCD45+ >8%. By day 14, for the vehicle-treated mice, the %huDC45+ increased to 40–60%, whereas for both treatment groups, the %huCD45+ was undetectable (p-value <0.0001 between vehicle- and enzyme-treated groups; ns between the two enzyme-treated groups). Mean with SD were plotted. All tests were set at controlling for probability of Type I error of 0.05. See Supplementary Biostatistics for more details. (D) At day 0, assessment of BM %huDC45+ in 3 mice with similar BLI flux as the ones used for treatment revealed high engraftment (gray boxes). At day 14, BM %huDC45+ remained high in the vehicle-treated mice, but was undetectable in both enzyme-treated groups (p-value <0.0001 between vehicle- and enzyme-treated groups; ns between the two enzyme-treated groups). Mean with SD were plotted. All tests were set at controlling for probability of Type I error of 0.05. See Supplementary Biostatistics for more details. (E) Spleens from the vehicle-treated mice were highly enlarged, whereas spleens from the ErA-WT and ErA-TM groups resembled normal mouse spleens in size. (F) H&E-stained paraffin sections of livers from vehicle-, ErA-WT- and ErA-TM-treated mice. Vehicle-treated animals had livers filled with deposits of lymphoblastic leukemic cells (arrows). In contrast, livers of mice treated with ErA-WT or ErA-TM had no detectable leukemic cells present; bar = 10 µm. (G) Female mice tail vein-injected with luciferase-expressing LOUCY cells four weeks prior were treated i.p. with ErA-WT (n=3); and ErA-TM (n=3) for 14 days (a total of 9 drug doses of 25 IU/mouse on days indicated by gray arrows). The average BLI (+SD) signal of each group at day 0, 4, 7, 11, 15, 18, 22 and 29 relative to the value at day 0 (day 0 = 1) is plotted. (H) Correlation between L-glutaminase activity and toxicity of the ErA variants. Weight loss (in grams, relative to day 0), an indicator of toxicity, was monitored in mice treated with vehicle (black trace), ErA-WT (red trace, L-glutaminase@Gln500µM=15.87 sec−1) and ErA-TM (blue trace, Gln500µM=0.01 sec−1). The pronounced daily weight loss in the ErA-WT-treated group is ameliorated in the ErA-TM-treated group by 0.29 g/day, p-value <0.0001.

Fig. 3. Similar in vivo stability for ErA-WT and ErA-TM and ability to deplete blood asparagine but dissimilar impact on glutamine homeostasis

(A) L-asparaginase activity measurement in blood plasma samples of mice obtained 24 h after injection of 50 IU of ErA-WT or ErA-TM from batch #1. Mean with SEM (standard error of the mean) is shown for each group and non-parametric Mann-Whitney test was used for statistics. (B) L-asparaginase activity measurement in blood plasma samples of mice obtained 1, 3, 7 and 14 days after injection of 50 IU of ErA-WT or ErA-TM from batch #2. (C) Determination of asparagine levels in blood plasma samples of mice prior to the treatment and at days 1, 3, 7 and 14 post administration of ErA-WT or ErA-TM. For each time point, the mean and standard deviation are shown. (D) Determination of aspartate levels in blood serum samples of mice prior to the treatment and at days 1, 3, 7 and 14 post administration of ErA-WT or ErA-TM. For each time point, the mean and standard deviation are shown. (E) Determination of glutamine levels in blood plasma samples of mice prior to the treatment and at days 1, 3, 7 and 14 post administration of ErA-WT or ErA-TM. For each time point, the mean and standard deviation are shown. (F) Determination of glutamate levels in blood plasma samples of mice prior to the treatment and at days 1, 3, 7 and 14 post administration of ErA-WT or ErA-TM. For each time point, the mean and standard deviation are shown.

Fig. 4. In a T-ALL PDX model, ErA-TM displays similar cell killing combined with reduced toxicity compared to ErA-WT

Female mice injected with primary T-ALL cells five weeks prior were treated daily with vehicle; ErA-WT or ErA-TM (n=5 for each group) for 13 days (drug dose 50 IU/mouse/day; i.p.). Cell debris and duplicates were gated out during flow cytometry data analysis. (A) PB %huCD45+ levels were determined once every week and at day 0, all animals were engrafted, as indicated by %huCD45+ ~4%. For the vehicle-treated mice, the %huCD45+ increased to ~40% by day 7 and ~80% by day 13, whereas for both treatment groups, the %huCD45+ stayed statistically indifferent to day 0 (p-value >0.2). No difference was detected between the two enzyme-treated groups at day 7 and day 13 (p-value >0.6). The %huCD45+ levels could not be analyzed for one mouse in the ErA-WT group and for one mouse treated with ErA-TM due to bad sample quality. (B) Similar to panel A, but for the BM at day 13. Whereas the BM of the vehicle group was full of cancer cells with %huCD45+ ~95%, the disease was largely controlled in the two enzyme-treated groups with %huCD45+ ~20 – 40%. (C) Similar to panel B, but for the spleen at day 13. The analysis of spleen sample of mouse 18 (ErA-WT-treated) was not included because of bad sample quality. Whereas the vehicle-treated group’s spleens were ~80% invaded with cancer cells, less than 20% of %huCD45+ was detected in the enzyme-treated groups. No significant difference was detected between the ErA-WT- and ErA-TM-treated groups (p-value ~0.2). (D) On sacrifice, spleens were harvested and weighed. Shown are the spleen weights for the 3 groups. Consistent with the cancer cell invasion data in panel C, the weight of spleens from the vehicle-treated group are significantly higher than the enzyme-treated groups, but no significant difference was detected between the two enzyme-treated groups (p-value >0.8). (E) Mice weight change, shown as % change relative to day 0, over the course of the 13-day treatment period is shown. On average, at day 13, ErA-WT- treated mice lost ~30% of body weight, whereas ErA-TM- treated mice lost ~15% (p-value <0.05). n.s. = not significant. In all panels, the color code depicting the vehicle-treated group, the ErA-WT- treated group and the ErA-TM- treated group is black, red and blue, respectively. Mean with SD were plotted.