The redox-active, anti-cancer drug Dp44mT inhibits T-cell activation and CD25 through a copper-dependent mechanism

We were particularly interested to read a recently published article in the FASEB J by Gundelach et al.¹ This study investigated the effect of the thiosemicarbazone iron chelator, di-2-pyridylketone 4,4-dimethyl-3-thiosemicarbazone (Dp44mT), which has been extensively investigated as an anti-cancer agent,^2–6 on the activation of isolated mouse splenic T cells in vitro. The authors observed that Dp44mT inhibited the up-regulation of surface CD25 expression after mitogenic stimulation, while CD69 expression was unaltered relative to the stimulated control.¹

Although the exact mechanism responsible in the specific inhibition of CD25 up-regulation by Dp44mT was not identified, the authors suggest that the inhibition of T-cell activation after Dp44mT treatment occurred via a mechanism that was iron-independent.¹ In fact, it was proposed that the ability of Dp44mT to bind copper and/or the capacity of the Dp44mT-copper complex to redox cycle played a role in the inhibition of T-cell activation. Gundelach et al.¹ also observed that the non-toxic copper chelators, tetrathiomolybdate and bathocuprione disulfonate, were able to mitigate the inhibitory effects of Dp44mT.¹ In agreement with our previous studies demonstrating the effect of Dp44mT on cellular thiol systems,^5,7 this same mechanism was shown to be involved in inhibiting CD25 up-regulation.¹ Collectively, these authors suggest from their in vitro studies that Dp44mT may act as an immunosuppressive agent that could complicate its use as a novel anti-tumor drug.¹

We would like to make several comments regarding the preliminary nature of these studies and the speculative nature of the conclusions. First, all of the studies reported by Gundelach et al. are limited to isolated T cells in vitro which provide a highly simplistic model of the intricate immune effector response against tumors. Indeed, a complex array of cellular interactions are associated with the immune response,^8–10 which cannot be predicted by utilizing isolated T cells alone. Hence, carefully performed in vivo studies should have been performed before speculating that Dp44mT interferes with the immune response and that this could affect the drug's potential as an anti-tumor agent. Such studies would need to include a comprehensive assessment of T cells (their various subsets), natural killer cells, macrophages, etc. and their tumor effector functions after administration of Dp44mT in vivo. With regard to the proposed effect of Dp44mT affecting T cells and then complicating its anti-tumor activity,¹ it is notable that we have demonstrated the marked efficacy of Dp44mT against tumors in nude mice, despite the fact that these animals have a significantly depressed immune system.³ Moreover, we have also demonstrated the anti-tumor effects of Dp44mT in vivo using mice with a normal immune system.⁴

The deficiencies in experimental design described above, and particularly the lack of implementing an in vivo model, are underlined by the fact that Dp44mT acts differently in vitro and in vivo. We have previously demonstrated that Dp44mT can induce marked iron chelation and tumor cell iron depletion in vitro.⁴ However, in clear contrast, a slight increase in tumor iron levels was observed upon administration of Dp44mT in mice in vivo.³ Thus, it may be expected that the immunosuppressant role of Dp44mT observed in vitro in isolated T cells¹ may be different to that found in vivo. Of interest, our studies using Dp44mT in vivo in mice did not demonstrate any significant differences in white blood cell counts relative to the control.^3,4

In addition to the preliminary nature of this article, some of the interpretation appears unclear and the exact mechanism involved remains unknown, as the authors do not take into account the previous studies examining the chemical and biochemical properties of Dp44mT.^2,4,5,7,11 For example, Gundelach et al.¹ use iron supplementation (ferric ammonium citrate (FAC)) of the culture system to determine whether the block in T-cell activation by Dp44mT was dependent on the ability of this ligand to chelate iron. The authors report that FAC did not have any effect on the inhibition of T cells by Dp44mT and this observation led them to conclude that iron-binding was not implicated.¹ Instead, they indicate that copper chelation alone was involved in the mechanism on inhibiting T-cell activation,¹ but do not clearly dissect whether it was due to copper depletion and/or due to the ability of Dp44mT to form a redox-active copper complex.^2,11 In fact, they show that the anti-oxidant, N-acetylcysteine, rescues the inhibition of T-cell activation, suggesting that redox activity may be involved in the mechanism.¹ It is important to realize that unlike the ligand, desferrioxamine, Dp44mT forms redox active complexes with both iron and copper that mediate many of its biological effects.^2,11 Hence, the inability of iron supplementation to rescue CD25 inhibition by Dp44mT cannot be simply interpreted to indicate that iron complexation was not involved. In fact, both redox-active iron and copper complexes form in cells^2,11 and reactive oxygen species (ROS) generated by both of these are important to assess in terms of the effect observed.

Considering the deficiencies above, additional experiments are essential in order to clarify the interpretation of the roles of these complexes and their redox activity in the effects observed. These include (1) assessment of the ability of the anti-oxidant, N-acetylcysteine, to rescue T-cell activation in the presence of either the pre-synthesized iron or copper complex of Dp44mT; (2) examination of T-cell activation upon treatment with redox-inactive desferrioxamine in comparison to its redox-inactive iron complex to definitively determine if iron chelation alone is sufficient;¹² (3) determination of whether copper supplementation (as CuCl₂) acts similarly to iron supplementation and is unable to rescue T-cell activation in the presence of Dp44mT; (4) the use of cell-permeable scavengers of ROS to block their effects in the presence of the iron or copper complexes, and conversely, to implement agents that generate ROS to assess if it is responsible for the effects observed; and (5) implementation of Dp2mT, an analogue of Dp44mT which cannot bind iron, as an important control.^3,4 Collectively, these further studies would examine the potential role of ROS generation in the inhibition of T-cell activation by Dp44mT via the formation of iron and/or copper complexes and the role of iron and/or copper chelation.

Finally, Gundelach et al.¹ propose it is important to examine the immunosuppressive effects of compounds being considered as novel chemotherapeutics, and therefore, have chosen to study Dp44mT. However, while the anti-tumor activity of Dp44mT has been extensively documented by our group and others,^3,4,6 we have also demonstrated that this agent induces cardiotoxicity at high, non-optimal intravenous doses³ and exhibits marked toxicity when given by the oral route.¹³ Due to these side effects, better tolerated thiosemicarbazones (e.g. di-2-pyridylketone 4-cyclohexyl-4-methyl-3-thiosemicarbazone (DpC)) have been developed that do not demonstrate cardiotoxicity and are effective by both the oral and intravenous routes.^14,15 Therefore, it would have been preferable to assess the effect of this latter compound on T-cell effector function rather than Dp44mT that has unfavorable pharmacological properties.

In summary, the preliminary in vitro experiments conducted by Gundelach et al. are inconclusive. The studies with isolated T cells should have been complemented by in vivo studies in animal models with and without tumors to determine if appropriate thiosemicarbazones (e.g. DpC) demonstrate immunosuppressive activity.