SIRT5 Regulates Energy Metabolism and Survival in Acute Lymphoblastic Leukemia Cells

Acute Lymphoblastic Leukemia (ALL) is the most common cancer in children, with a second incidence peak in adults over 60 years old. Long-term survival with currently available therapy regimens is >90% in children, but <50% in adults, reflecting differences in disease biology and tolerance of aggressive multi-agent chemotherapy regimens. While CAR-T cell therapy provides effective salvage to patients with relapsed B-cell ALL (B-ALL), outcome in relapsed T-ALL is exceptionally poor. In addition, therapy-related chronic functional impairment in pediatric ALL survivors are common. Less toxic, and more effective therapies are needed to improve ALL survival, particularly in adults, and reduce toxicity and functional impairment in all patients.

There is evidence that energy restriction creates a barrier to malignant transformation as B cell precursors undergo somatic hypermutation and during T cell receptor diversification. Mutational inactivation of B-lineage transcription factors such as IKZF1 and PAX5 in B-ALL and aberrant expression of T-lineage regulators such as TLX1/3 in T-ALL disables this metabolic brake, promoting malignant transformation of B and T lymphoid precursors, respectively. Metabolism is being recognized as an attractive target in cancer therapy, but metabolic plasticity often limits the efficacy of targeting single metabolic pathways. SIRT5 is a lysine deacylase that regulates multiple metabolic pathways, including energy metabolism. We have previously reported that a subset of acute myeloid leukemia (AML) cases depends on SIRT5 as a master regulator of central energy metabolism, implicating SIRT5 as a therapy target in AML. We hypothesized that this metabolic vulnerability may extend to ALL.

To study the function of SIRT5 in ALL, we stably expressed doxycycline (Dox)-inducible shRNAs targeting SIRT5 in ALL cell lines and primary samples. Knockdown (KD) of SIRT5 significantly inhibited growth and increased apoptosis in B-ALL and T-ALL cell lines. Similar results were seen with primary cells relapsed ALL, whereas SIRT5 KD had no effect on colony formation by normal CD34⁺ cord blood cells. We also injected mice with Jurkat T-ALL cells expressing Dox-inducible shSIRT5 and luciferase. Dox-induced SIRT5 KD abrogated disease initiation and eliminated established disease (Figure 1A).

To determine whether ALL cells require SIRT5 to maintain a metabolic state that is permissive to malignant transformation, we next evaluated the effect of SIRT5 disruption on energy metabolism, using Seahorse extracellular flux analysis. We found that SIRT5 deficient T-ALL cells had lower rates of both oxidative phosphorylation and glycolysis in comparison to SIRT5 wild type (WT) cells (Figure 1B).

Next, to evaluate dependency of SIRT5-KD on multiple genotypes and therapy-resistant ALL, we engineered 30 ALL cell lines to stably express doxycycline shSIRT5 representing B and T cell origin, diverse genotypes, and including drug resistant lines. We found that several B-ALL and T-ALL cell lines are highly dependent on SIRT5 for growth and survival, while others are not, showing a large range of sensitivity to SIRT5 KD. Therefore, to elucidate the potential mechanisms underlying SIRT5 dependence in ALL, we first analyzed whether SIRT5 KD differentially affects OXPHOS and glycolysis. Preliminary analysis of mitochondrial respiration upon SIRT5 KD suggested that OXPHOS and glycolysis are reduced in SIRT5-dependent cells, but not in SIRT5-independent cells upon SIRT5 KD.

Finally, to determine whether Sirt5 disruption affects energy metabolism in normal hematopoietic stem cells (HSCs), we generated a tamoxifen-inducible conditional hematopoietic cell-specific Sirt5 knockout (KO) model. In preliminary experiments, we found no difference in mitochondrial respiration between normal HSCs with or without genetic absence of Sirt5, suggesting that leukemic cells, but not normal hematopoietic stem cells are dependent on Sirt5 for regulation of energy metabolism.

Collectively, our results suggest that SIRT5 plays a critical oncogenic role in the development and maintenance of a subset of ALL, providing a rationale for the development of a clinical SIRT5 inhibitor. Targeting master regulators rather than single pathways may circumvent plasticity-based resistance in ALL.