Abstract 2592

Histone deacetylase inhibitors (HDACi), like suberoylanilide hydroxamic acid (SAHA), are new drugs in oncologic treatment protocols. Histone deacetylases together with histone acetyltransferases (HAT) regulate the acetylation status of distinct lysine residues in histones. Not only do they affect the plasticity of the chromosomes and thereby have a direct impact on gene expression, but also regulate several non-histone-proteins via hyperacetylation. Among these are proteins involved in gene expression like transcription factors. SAHA is a pan HDACi and displayed antiproliferative effects on transformed cells in vitro and in vivo. SAHA was the first HDACi that was approved for treatment of refractory cutaneous T cell lymphoma. Interestingly, SAHA was reported to reduce the production of the proinflammatory cytokines TNFα, IFNγ, IL-1β and IL-12 at nanomolar concentrations, at which cell proliferation is not yet affected. Proinflammatory cytokines play a central role in the pathogenesis of acute GvHD. Indeed, it has been reported that SAHA reduced GvHD in mice while the important graft-versus-leukemia-effect (GvL) was not impaired.

A cellular tool for treatment of acute GvHD after allogenic stem cell transplantation are multipotent mesenchymal stromal cells (MSC). Their immunosuppressive effects are conducted by an altered cytokine profile of peripheral blood mononuclear cells (PBMC). We analysed the effect of SAHA on T cells directly as well as on MSC. Both cell types tolerated low micromolar doses of SAHA well. Morphology and immunophenotype of MSC was not considerably affected. For the treatment of GvHD a synergistic or additive effect of SAHA and MSC on immune effector cells would be desirable. Instead, we found that SAHA reduced the suppressive effects of MSC on PBMC.

Therefore, we focused to a closely related cell type under investigation in our lab, tumor stromal cells (TStrC). In previous studies, we found that tumor stromal cells (TStrC) also impaired T cell and NK cell function. As TStrC share several properties with MSC, we hypothesized that in this particular setting SAHA may facilitate tumorimmunological responses by interfering with the immune suppression exerted by TStrC. We isolated TStrC from neuroblastoma patients and analysed the effect of SAHA on these TStrC including cytokine patterns. In supernatants from TStrC cultures, we were not able to detect any of the cytokines IL-1β, IL-2, IL-4, IL-10, IL-12p70, IL-15, IL-17, TNFα, RANTES and IFNγ. Comparable to MSC, TStrC reduced the production of the proinflammatory cytokines IL-12p70, TNFα and IFNγ and the anti-inflammatory IL-10 in PBMC. In contrast, the level of IL-1β was markedly increased. Low micromolar concentrations of SAHA did neither affect the morphology and immune phenotype of TStrC nor impaired their proliferation. Pre-incubation of TStrC with SAHA resulted in an attenuated immune suppression by TStrC. Untreated and solvent-treated TStrC inhibited the proliferation of PBMC stimulated by IL-2/OKT3. PBMC co-cultivated with SAHA-treated TStrC showed a significantly enhanced, but not fully restored proliferation in a concentration-dependend manner.

Functional assays showed that NK cell cytotoxicity against the leukemia cell line K562 was impaired by TStrC. The activating NK cell receptors NKp44 and NKp46 were downregulated while the inhibitory receptor NKG2A remained unaffected. Pre-incubation of TStrC with SAHA partially restored NK cell cytotoxicity. In multiplex analyses, we investigated the cytokine profiles of TStrC and SAHA-treated TStrC monocultures as well as co-cultures of PBMC with TStrC and previously SAHA-treated TStrC. Control and SAHA-treated TStrC did not produce any of the cytokines IL-1β, IL-2, IL-4, IL-10, IL-12p70, IL-15, IL-17, TNFα, RANTES or IFNγ at a detectable level. Co-incubation of PBMC with SAHA-treated TStrC changed their cytokine profile as compared to co-incubation with untreated TStrC. PBMC produced higher levels of the cytokines IL-10, TNFα and IFNγ when co-cultivated with SAHA-treated TStrC. The secretion of IL-1β was also increased. Taken together, these findings suggest that SAHA may modify the tumor microenvironment at low micromolar concentrations to allow for more efficient anti-tumor immune responses.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.