To our knowledge, this is the first report demonstrating that Prxs are oxidized and that TrxR is inhibited in Cr(VI)-treated cells. Since the TrxR protein level remains constant in these cells (), the decline in TrxR activity represents enzyme inhibition. This inhibition was not reversed by NADPH, its endogenous electron donor, or by removal of free Cr by overnight dialysis. The effects on TrxR may therefore be prolonged in cells, extending beyond the time of Cr(VI) exposure. Even intermittent exposure to Cr(VI), which would be consistent with occupational exposures, could therefore have prolonged effects on TrxR in the bronchial epithelium.
Since this is the first report on TrxR inhibition following Cr(VI) exposure, the mechanism by which TrxR is inhibited is not known. Cr(VI) is redox-active and there are multiple redox-active sites within TrxR, including the flavin (FAD), the C-terminal active site Cys-SeCys, and the N-terminal domain dithiol (-CVNVGC-) [43
]. All of these sites are necessary for electron flow within TrxR and therefore its activity, and disruption of any one of these could theoretically inhibit its activity. Of these, the active site is likely the most reactive. It is exposed on the surface of the enzyme [62
] and the selenol of the SeCys residue (pKa
ca. 5.2) [63
] should be ionized to selenolate at physiological pH making it more reactive. As an example, one inhibitor of TrxR, 2,4-dinitrochlorobenzene, irreversibly binds the SeCys [56
], but it does not modify the N-terminal dithiol [64
Previous work demonstrated the potential for purified rat TrxR1 to reduce Cr(VI) to Cr(V) [42
]. Diversion of electrons to Cr(VI) therefore represents one potential mechanism of inhibition, but it cannot explain the results here because free Cr(VI) had been removed by dialysis and NADPH did not restore enzyme activity. The inhibition of TrxR observed here is therefore not likely due to competitive inhibition, or to simple oxidation of the redox-active thiols or selenol as would occur when TrxR reduces endogenous substrates such as Trx. Instead, the results are consistent with a more prolonged inactivation of the enzyme. The intracellular reduction of Cr(VI) is known to lead to a variety of reactive species which can act as potent oxidants, including Cr(V), Cr(IV), and reactive oxygen species including HO•
]. Reaction of TrxR with one or more of these reactive species represents one possible mechanism for TrxR inhibition. Other mechanisms must also be considered, however, including indirect effects resulting from reactive species generation. We are not aware of studies which have examined interactions between Cr species and SeCys. However, Cr(III) potassium sulfate in high concentration in vitro with 30% H2
has been reported to convert thiols to disulfides in a reaction mechanism that may involve HO•
]. However, such disulfides within TrxR, or the thioselenide of its active site, should be reversible by NADPH, the endogenous electron donor for TrxR. Since NADPH did not restore TrxR activity, a disulfide type of TrxR oxidation is not predicted in Cr(VI)-treated cells.
A prolonged inhibition of TrxR could, however, contribute to Trx oxidation or to a decreased capacity to keep Trx reduced. Since Prxs are dependent on Trxs, they are therefore indirectly dependent on TrxR. In these studies, the examination of TrxR, Trxs, and Prxs within the same cells allowed for direct examination of the relationship of the effects of Cr(VI) exposure on these proteins. Given the dependence of Prxs on Trxs, the oxidation of Trxs could influence the redox status and function of the Prxs. Similarly, TrxR inhibition could influence the redox status and function of the Trxs. The effects of Cr(VI) on the Trx redox status may therefore be due to its ability to both enhance oxidant generation and to inhibit TrxR. Even though the TrxR assay does not distinguish between TrxR1 and TrxR2, there was a significant correlation between the % TrxR inhibited and the % Trx1 oxidized () (Spearmann nonparametric correlation r
= 0.7245, P
< 0.0001). This might imply that an inability of TrxR to maintain Trx1 in a reduced state contributes to Trx1 oxidation. This type of Trx1 oxidation should be reversible by disulfide reductants, which proved to be the case (). However, it was previously noted that Cr(VI) can oxidize purified Trx1 in vitro [42
], as do some other oxidants. Trx1 oxidation may therefore be initiated by Cr(VI) and/or other oxidizing species that are generated following Cr(VI) exposure, and this oxidation may be sustained or enhanced by the decreased ability of TrxR to regenerate reduced Trx1. In the absence of oxidant stress, however, TrxR inhibition on its own may not be sufficient to cause Trx oxidation. This proved to be the case in HeLa cells in which inhibition of TrxR by aurothioglucose, or TrxR knockdown by siRNA, were not sufficient to cause Trx1 oxidation [67
]. However, monomethylarsonous acid, which both inhibits TrxR and increases ROS generation, did cause Trx1 oxidation [67
Fig. 9 Relationship between the inhibition/oxidation of various proteins in Cr(VI)-exposed cells: TrxR and Trx1 (A), TrxR and Trx2 (B), Trx1 and Prx1 (C), and Trx2 and Prx3 (D). All points represent individual datapoints, including different treatments and their (more ...)
There was also a correlation between TrxR inhibition and Trx2 oxidation (, Spearmann nonparametric correlation r = 0.7003, P < 0.0001). It should be noted that in there are seven superimposed points at 0% TrxR inhibited and 0% Trx2 oxidized. Therefore most of the points cluster at the lower left and upper right of this graph, which could contribute to the apparent correlation between TrxR and Trx2. Also note that there were two points for which TrxR inhibition was almost complete but where Trx2 oxidation was only partial, and three for which Trx2 oxidation was complete but TrxR was only partially inhibited. Since the TrxR assay measures total TrxR, it does not reflect just the mitochondrial isoform (TrxR2) on which Trx2 depends. However, the relationship in and the ability to reverse Trx2 oxidation by disulfide reductants () suggest that Trx2 oxidation may be enhanced or sustained by a loss of reducing equivalents from TrxR2.
Cytosolic Prxs such as Prx1 should be dependent on Trx1 to maintain their redox state. For Cr(VI) treatments that oxidized up to 75% of Trx1, the redox state of Prx1 largely resembled that of untreated cells (). Therefore, even a minority of active Trx1 is sufficient to keep Prx1 largely reduced. However, when >75% of Trx1 was oxidized, Prx1 abruptly shifted to being mostly oxidized (). Thus, these cells can maintain Prx1 redox status until nearly all of the Trx1 is oxidized. This suggests that Prx1 oxidation largely results from lack of reducing equivalents from Trx1. Prx1 oxidation in these cells therefore serves as a functional indicator that reflects the loss of the vast majority of Trx1 activity.
Mitochondrial Prxs such as Prx3 should be dependent on Trx2 to maintain their redox state. In untreated cells and for Cr(VI) treatments that did not cause Trx2 oxidation, Prx3 redox state was maintained at 56.7 ± 10.3% oxidized (mean ± SD) (). The Prx3 redox state shifted to 94.4 ± 7.2% oxidized (mean ± SD) for treatments that fully oxidized Trx2 (). The Prx3 oxidation for these two groups was significantly different (P
< 0.0001). There were only a few points for which Trx2 oxidation was intermediate (between 0 and 100% oxidized), which were the result of 2.5 μM Cr(VI) for 16 hr; for these, Prx3 oxidation (65.5 ± 6.8%) was not different from treatments in which Trx2 was fully reduced (P
= 0.153), but it was different from those with 100% Trx2 oxidation (P
< 0.0001). Overall, the data imply that Prx3 redox state can be maintained when there is partial Trx2 oxidation, but is not maintained once Trx2 becomes fully oxidized. There was a correlation between Trx2 oxidation and Prx3 oxidation (, Spearmann nonparametric correlation r
= 0.8860, P
< 0.0001). However, given that 20 of the 24 points clustered at either 0% or 100% Trx2 oxidized, the relationship between Trx2 and Prx3 redox states may not necessarily be linear. What is clear is that Prx3 oxidation was significantly less in cells without Trx2 oxidation (56.7% ±10.2%) than in cells in which Trx2 oxidation was complete (94.4% ± 7.2%) (P
<0.0001). The reversibility of Prx3 oxidation () implies that Prx3 oxidation represents conversion to the disulfide form. This form would be expected to result from lack of reducing equivalents from Trx2, but other reasons for Prx3 oxidation remain possible. While peroxide substrates gain ready access to the active site thiols of Prxs, these thiols may be somewhat protected from other species. For example, the Prx2 active site thiols are not readily alkylated with iodoacetamide or various amino acid chloroamines, despite the high reactivity of these agents with other cellular –SH groups [68
Fig. 7 The oxidation of Prx3 following 16 hr exposure to Cr(VI) is reversible by disulfide reduction. In cells treated with HBSS (untreated) or 5 μM Cr(VI) for 16 hr, the redox state of Prx3 was assessed using the standard protocol with non-reducing (more ...)
The cytosolic and mitochondrial thioredoxin systems are not in redox equilibrium with each other, so it is possible to discern differential effects of oxidants on these subcellular compartments [38
]. Previous work with shorter Cr(VI) exposure demonstrated that Trx2 was more susceptible in that it could all be converted to the oxidized form, whereas a small amount of reduced Trx1 remained even after higher Cr concentrations [42
]. The additional exposures tested here support the greater susceptibility of Trx2, e.g. 100% of Trx2 was oxidized after 25 μM (6 hr)() or 5 μM (16 hr)(), whereas 83% and 79% of Trx1 was oxidized by these same treatments, repsectively. The differential oxidation of Prx3 vs. Prx1 further supports the enhanced susceptibility of the mitochondrial system, e.g. 98% of Prx3 was oxidized after 5 μM Cr(VI) (16 hr), vs. 56% oxidation of Prx1 (, ). Similarly, mitochondrial Trx2 is more susceptible than cytosolic Trx1 to oxidation by t
-butyl hydroperoxide or diamide [38
]. The data are in contrast to acrolein, however, which preferentially oxidizes Trx1 [41
]. Overall, these redox blots provide a means to assess the relative extent of redox stress in the cytosolic vs. mitochondrial compartments of Cr(VI)-exposed cells. The greater effects on the mitochondrial compartment could indicate enhanced oxidant generation inside the mitochondria and/or a decreased ability of the mitochondrial system to protect itself.
As indicated above, the oxidation of Prxs following Cr(VI) exposure likely reflects the loss of reducing equivalents from the respective Trxs. This is in keeping with a recent report by Cox et al. [69
] in which the pronounced inhibition of TrxR by auranofin or dinitrochlorobenzene, which are well-known inhibitors of TrxR, resulted in the preferential oxidation mitochondrial Prx3 vs. cytosolic Prx1 in human Jurkat cells. This differential effect did not correspond with a greater inherent sensitivity of mitochondrial TrxR, as the TrxR in isolated cytosol was marginally more sensitive to auranofin than was TrxR in isolated mitochondria [69
]. It is plausible that Trx2 was oxidized in cells in which Prx3 oxidation occurred, but this was not examined [69
]. Auranofin and dinitrochlorobenzene treatments that caused pronounced TrxR inhibition and Prx3 oxidation in Jurkat cells resulted in cell death [69
]. In this study, Cr(VI) treatments that caused prominent changes in TrxR activity also caused loss of cell viability. However, the 16 hr data suggest that Prx oxidation is not required for cytotoxicity, e.g. clonogenic survival under these conditions was significantly less with 2.5 μM Cr(VI) (16 hr), but Prx1 and Prx3 redox status were not significantly altered (). With the 6 hr treatments (25 or 50 μM Cr), there was pronounced cytotoxocity and pronounced effects on TrxR, Trxs and Prxs ().
Given the complexity of Cr chemisty and its potential to affect multiple cellular components, Cr(VI) cytotoxicity is likely multifactorial. The Cr(VI)-induced changes to the activity of TrxR and to the redox state of Trxs and Prxs have not been previously explored, but they could contribute to the cytotoxic effects of Cr(VI). Inhibition of TrxR increases the susceptibility to oxidants and favors apoptosis [43
], and inhibition or genetic suppression of Trx enhances oxidant sensitivity, oxidant stress and apoptosis [38
]. The oxidation of Prxs can also enhance apoptosis [49
]. Changes to each of these components could affect cell survival following Cr(VI) exposure. It remains to be determined if effects on some of these proteins are more critical than others in these cells. Loss of protection from oxidant stress is just one of the potential consequences of the effects of Cr(VI) on the TrxR/Trx/Prx system. Since mitochondria lack catalase, and they contain 30 times more Prx3 than glutathione peroxidase [49
], Prx3 may be a critical regulator of mitochondrial H2
]. Treatments which markedly oxidize Prx3 could therefore have important effects on this signaling. Besides compromising Prx function, the oxidation of Trxs could have other effects as well. Trx is an important hydrogen donor for ribonucleotide reductase [71
], so Trx oxidation could lead to growth arrest. Also, reduced Trx binds and inhibits ASK1, and we noted the dissociation of ASK1 from Trx1 for treatments that oxidized Trx1 (). This dissociation could imply the activation of ASK1 which could promote apoptosis. The effects on Prxs and ASK1 imply that other Trx-dependent proteins may also be affected by these treatments.
While TrxR inhibition could facilitate the oxidation of Trxs and Prxs, TrxR inhibition could have other effects as well. While the mechanism of inhibition of TrxR in Cr(VI)-treated cells remains to be determined, treatments that target the SeCys of the TrxR active site can induce cell death [72
]. Given these multiple possible effects, further studies on the consequences of the effects of Cr(VI) on the TrxR/Trx/Prx system are therefore warranted.
In summary, this report describes the effects of Cr(VI) on the thioredoxins, peroxiredoxins, and TrxR in human bronchial epithelial cells. Low micromolar concentrations of Cr(VI) caused oxidation of cytosolic and mitochondrial Trxs and Prxs and inhibited TrxR. The oxidation of Trx1, Trx2, and Prx3 were reversible in vitro, but TrxR inhibition was not easily reversed. These effects of Cr(VI) could compromise the normal function and survival of the bronchial epithelium, and could alter redox-sensitive cell signaling and the ability of these cells to tolerate other oxidant insults.