Because an immediate sign of mitochondrial uncoupling is the drop in cell phosphorylation potential, we evaluated the Rottlerin uncoupling effect by detecting the levels of all components of the adenylate pool using CZE (Fig. 4a) and then calculating the energy charge, represented by the ATP + ADP + AMP/0.5(ATP + AMP). Rottlerin uncoupling effect is clearly evident at the dose of 20 μM, appearing as a fall in the cellular energy charge (Fig. 4b). The capacity to form high-energy phosphates was maximally compromised after 20 min (24% decrease) and was recovered slowly thereafter. At 24 h, cells recovered about 95% of the basal energy status, likely by compensatory metabolism. Rottlerin, at the dose of 5 μM, exerted a less evident effect (not statistically significant) on the cellular energy status, which was weakly compromised after 20–30 min treatment (4–5% decrease) and was completely recovered after 40 min.

Fig. 4 Effect of Rottlerin on the energy charge (EC). Cells were treated with 5 and 20 μM Rottlerin (R5 and R20), and high-energy phosphates (ATP, ADP, and AMP) were determined by CZE, as described in “Materials and methods”. a Representative electropherogram of perchloric acid extracts from MCF-7 cells. A Control sample, B 30-min incubation with 20 μM Rottlerin. AU absorbance units. b The EC, represented by the ATP + ADP + AMP/0.5(ATP + AMP) ratio, was calculated at the indicated time points and expressed as % of the control (100%). Values are the average of three separate experiments and are expressed as mean ± SD. *p < 0.05.

Effect of mitochondrial uncoupling on MTT reduction

FCCP is a commonly used protonophore that collapses the mitochondria inner membrane potential disrupting mitochondria function (13). As shown in Fig. 5 , a 30-min treatment of MCF-7 cells with 5 and 20 μM FCCP caused a significant increase in MTT reduction, an effect similar to that achieved with Rottlerin at the same times and doses. This result suggests that the Rottlerin interference in the MTT assay is related to its ability to act as a mitochondrial uncoupler.

Fig. 5 Effect of Rottlerin and FCCP on MTT assay. Cells were treated for 30 min with 5 and 20 μM Rottlerin and FCCP (R5, R20 and F5, F20, respectively) or vehicle alone (DMSO and ethanol, respectively) and MTT reduction was evaluated as described in “Materials and methods”. Values are the average of three separate experiments in quadruplicate and are expressed as mean ± SD. *p < 0.05.

Discussion

The MTT assay is a widely used test to measure cell proliferation, cell viability/survival, or drug toxicity. In a recent paper, the use of MTT has been described as one of the major techniques for testing tumor cell resistance to anticancer agents (14). On the other hand, several papers have reported agents that increase MTT reduction without increasing cell viability (15). Moreover, some studies reported that certain plant extracts and redox-active polyphenols can interfere with the MTT assay because they directly reduce the tetrazolium salt even in the absence of cells (16–19). Our results demonstrate that Rottlerin (5 and 20 μM) also strongly enhance the formation of formazan crystals inside cells. However, Rottlerin failed in reducing tetrazolium salts in vitro, indicating that the effect observed in cultured cells is not due to a direct reducing action but that the presence of an intermediate molecule/organelle is needed.

Because the MTT assay is largely based on mithocondrial LDH activity, we hypothesized that Rottlerin could enhance LDH activity indirectly, by maintaining high levels of NADH, due to its H+ donor properties. Rottlerin indeed contains five phenolic hydroxyl groups (Fig. 1 ) that act as hydrogen donors in the scavenging of free radicals, such as 1,1-diphenyl-2-picryl-hydrazyl (9). The results presented in the current study exclude any Rottlerin modulation of LDH activity in vitro, further confirming the need of whole cells to show the increase in MTT reduction.

Therefore, knowing that Rottlerin can interfere in the respiratory chain by acting as an uncoupler of oxidation and phosphorylation (4), we hypothesized that the Rottlerin artifact in the MTT test could be the consequence of a direct action on mitochondrial respiration. Rottlerin, indeed, by dissipating the inner mitochondrial membrane potential, accelerates electron transfer and increases dehydrogenases activity, oxygen consumption, and NADH oxidation. The observed drop in the cell energy charge indicated that also in MCF-7 cells Rottlerin exerted an uncoupling effect, which accidentally enhanced MTT reduction by over-stimulated mitochondrial dehydrogenases. To verify this hypothesis, a comparative study between Rottlerin and the chemical uncoupler FCCP was performed. The results indicated that the overestimation of the MTT test was linked to Rottlerin uncoupling properties, since both uncouplers, at the same doses, enhanced MTT reduction after only 1-h incubation and roughly to the same extent.

The mitochondrial uncoupling was clearly evident with 20 μM Rottlerin but only weakly apparent with 5 μM Rottlerin. However, since this dose also caused overestimation of the MTT test, it can be hypothesized that even with 5 μM Rottlerin a mild uncoupling occurred in our cells, although could be promptly compensated, in terms of energy metabolism, by an increase in glycolytic ATP production.

In the light of these new findings, we revised, in retrospect, our previous results about the lack of effect of Rottlerin on MCF-7 cell viability (5), since the overestimation of the MTT assay likely masked possible toxic/apoptotic effects in this cell line (false or underestimated result).

Considering that different cells may be not equally sensitive to uncouplers, we also tested the interference of Rottlerin in the MTT assay in another cell type. The results demonstrate that Rottlerin overestimates the MTT assay also in HMVEC, suggesting that such an artifact occurs independently from the cell type.