Pharmacological agents with widely differing known modes of action can induce HbF production in vivo and in vitro. Recent data from this laboratory and others suggests that common pathways of action may be responsible for HbF induction by disparate drugs. A second therapeutic conundrum is that some of these agents seem to induce cell-type specific responses which may be useful in a variety of unrelated diseases. For example butyrate compounds (BU) show promise in treating X-linked adrenoleukodystropy (XALD), cystic fibrosis, sickle cell disease (SCD), inflammatory bowel disease, cancer, and inborn errors of metabolism to name a few. These observations suggest that certain drugs may initiate a common set of cellular responses that are read out differently depending upon the cellular context. In this study we tested various pharmacologic agents for their ability to ameliorate disease phenotypes and identified a common set of cellular responses. Although all agents were cytotoxic at high dosage, we studied effects of non-toxic low dosages. We examined two divergent inherited disorders, XALD and SCD. Cultured skin fibroblasts from XALD patients and knockout mice, and XALD mouse tissues were utilized in XALD studies. All agents studied [BU, trichostatin A (TSA), and hydroxyurea (HU)] normalized the elevated levels of very long-chain fatty acids (VLCFA) characteristic of XALD. These agents also induced peroxisome proliferation by a mechanism that requires the up-regulation of PEX 11α and is distinct from the well known PPARα peroxisomal proliferation pathway. Using a fluorescent probe for mitochondrial mass (Mitotraker), we also demonstrated a 2–3 fold increase in mitochondrial mass following treatment with these drugs. Using mitochondrial inhibitors (antimycin A), mitochondrial mutations, and auxotrophic growth we further demonstrated that the ability of these drugs to normalize VLCFA levels in XALD was dependent on the increase in mitochondrial mass and/or activity.

Using cellular model systems relevant to SCD including CD34 selected cell cultures derived from SCD patients and K562 cells, a HbF induction model, we demonstrate that HU, BU, TSA, 5-azacytadine and hemin stimulate HbF induction as indicated by an increase in F-containing cells. Interestingly, as in the ALD fibroblasts, these agents also stimulate mitochondrial proliferation 2–3 fold in these cell systems as assayed by Mitotraker staining and induce peroxisomal biogenesis. Antimycin A treatment of the K562 cells using concentrations that had no significant effect on cell division, viability or baseline HbF production blunted the ability of these agents to cause mitochondrial expansion and completely inhibited the ability of these agents to induce HbF. Taken together these studies demonstrate that common cellular responses (mitochondrial biogenesis and peroxisomal proliferation) occur in diverse cell types in response to treatment with disparate therapeutic agents. Furthermore, this common response appears to be necessary for the ultimate effect of disease amelioration by these drugs, specifically correction of VLCFA levels in ALD fibroblasts and HbF induction in erythroid cells.

Author notes

Corresponding author