Glucocorticoid metabolizing enzymes predict initial glucocorticoid sensitivity in childhood ALL Free

Introduction Synthetic glucocorticoids (GC) are crucial first-line treatment for childhood acute lymphoblastic leukemia (ALL). The response to initial GC treatment is a strong prognostic indicator of treatment outcome. Clinically, GC resistance remains a therapeutic problem, but the precise mechanism of GC resistance remains unknown. The GC metabolizing enzymes, 11beta-hydroxysteroid dehydrogenase types 1 and 2 (11b-HSD1 and 11b-HSD2) control intracellular GC levels. 11b-HSD1 regenerates active GC from inert 11keto-metabolites, whereas 11b-HSD2 inactivates GC. Therefore, we hypothesized that expression of 11b-HSDmRNAs is associated with GC sensitivity in childhood ALL.

Methods Diagnostic bone marrow samples from 228 patients with ALL were obtained from the TCCSG biobank. The study was approved by the Institutional Review Boards of all participating institutions. Leukemia cells were isolated by density gradient centrifugation. RNA was extracted and reverse transcribed. Microarray was performed to measure mRNAs. Patients were stratified by clinical responsiveness of initial 7 days prednisolone monotherapy. The cut-off point of peripheral blood blast count of 1,000/microL was used to assess GC-sensitivity.

Further investigation was performed for the patients who were admitted in Hokkaido University. GC sensitive case 1 was a 4-year-old girl with relapsed B-cell precursor ALL. Case 2 was an 11-year-old boy with T-cell ALL, who showed GC resistance following 7 days prednisone treatment. Unfortunately, he died during induction therapy due to brain hemorrhage. 11b-HSD1 and 11b-HSD2 mRNA were measured during initial GC treatment. In vitro GC sensitivity was determined by MTT assay. GC sensitive lymphoblastic leukemia CCRF-CEM and resistant MOLT4F cells were used for cell line experiments. mRNA levels were measured by qPCR.

Results Overall, 202 patients were GC-sensitive and the other 26 patients were GC-resistant. 11b-HSD2 mRNA levels were significantly higher in GC-sensitive than GC-resistant ALL cells at diagnosis, whereas there was no difference in expression of 11b-HSD1 between groups. The expression of GC receptor was slightly high in the GC-sensitive group compared to resistant group, but it did not achieve significance.

To investigate whether 11b-HSD1 and 11b-HSD2 are differentially regulated during initial GC treatment, we measured 11b-HSD1 and 11b-HSD2 mRNA levels in cells from representative patients. In GC-sensitive case 1, an increase in 11b-HSD1 mRNA levels was associated with a decrease in leukemia cell number during initial GC treatment; 11b-HSD2 mRNA levels decreased and became undetectable following treatment. In GC-resistant case 2, 11b-HSD2 mRNA levels increased concomitantly with the increase in leukemia cell number during GC treatment, whereas 11b-HSD1 mRNA levels were undetectable. Similar to patient samples, in GC sensitive CCRF-CEM cells, 11b-HSD1 mRNA levels increased and 11b-HSD2 mRNA levels decreased by dexamethasone (24 hours, 0.1-10μM). Increased 11b-HSD1/decreased 11b-HSD2 paralleled the degree of cell death. In contrast, 11b-HSD2 expression increased by dexamethasone significantly in GC-resistant MOLT4F cells and 11b-HSD2 knockdown by shRNA did not protect GC induced cell death.

Discussion GC-sensitive ALL had an unexpectedly higher levels of basal 11b-HSD2 expression, however, which can be a biomarker of initial GC sensitivity/resistance. The basal levels of 11b-HSD2 cannot be a cause of GC resistance in ALL, because it is supposed to inactivate GC action within cells. The changes of 11b-HSD1/2 expression during GC treatment could account for this paradox. Previously, we have shown that reciprocal regulation of 11b-HSD1/2 predicts GC sensitivity/resistance in childhood ALL. In vitro dexamethasone (24 hours, 1μM) increased 11b-HSD1/decreased 11b-HSD2 mRNA in GC sensitive ALL, whereas GC decreased 11b-HSD1/increased 11b-HSD2 in GC resistant ALL (Sai et al 2020 J Pediatr). Similar reciprocal regulation occurred in representative patients and cell lines in this study. Therefore, after initial GC treatment, levels of 11b-HSD2 in GC-resistant ALL patients increase, which can be a cause of GC resistance as seen in GC-resistant MOLT4F cells.

Conclusion 11b-HSD2 can be a biomarker of GC sensitivity at diagnosis and the reciprocal regulation of 11b-HSD1/2 is associated with initial GC sensitivity in childhood ALL. Our data may shed new light on the understanding of GC sensitivity/resistance in ALL.