HQ-thioethers are present in the bone marrow of rats following co-administration of PHE/HQ, the majority of which appear to be generated in situ and further metabolized via the mercapturic acid pathway [22
]. This pathway is important in modulating the reactivity of HQ-GSH conjugate. Based on the (re)activity of HQ-GSH conjugates, we speculated that some of the hematotoxic effects attributed to HQ (or 1,4-BQ) may, in fact, be mediated by their thiol conjugates. Indeed, as shown by Monks et al (this issue), HQ-GSH conjugates are far more efficient generators of superoxide anion than the HQ/1,4-BQ redox couple. Moreover, HQ-GSH conjugates are toxic to developing erythrocytes in vivo [22
]. Whether this is a direct effect of these conjugates in bone marrow, or whether additional tissues/organs contribute to the observed erythrotoxicity is not known. The hematopoietic microenvironment is regulated by stromal cells which secrete a variety of cytokines, including interleukin-1 (IL-1), tumor necrosis factor-α, granulocyte-macrophage colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and macrophage colony-stimulating factor (M-CSF). In addition, kidney peritubular cells produce and secrete the hormone, erythropoietin (EPO) [36
]. These growth factors interact with various hematopoietic stem/progenitor cells to control cellular proliferation and differentiation. Because HQ-GSH conjugates are capable of damaging cells within the proximal tubules of the kidney [21
], the major site of EPO production, we hypothesized that 2,3,5-GS-HQ and 2,6-GS-HQ might also be capable of indirectly inducing anemia by reducing serum EPO levels. Consistent with this reasoning, both 2,3,5-GS-HQ and 2,6-GS-HQ caused reductions in circulating EPO levels (). We have previously shown that quinone-thioethers are “erythrotoxic” based on a reduction in iron uptake [22
]. However, EPO drives RBC production and consequently the incorporation of iron. It is therefore possible that the reduced iron uptake involved both decreased production of EPO as well as direct toxicity to existing RBCs. In this report we have shown that quinone-thioethers induced a reduction in circulating lymphocytes, consistent with bone marrow toxicity, as previously reported [22
]. However a possible redistribution of lymphocytes from the peripheral blood into tissues cannot be ruled out.
Adduction of proteins by reactive electrophiles is not a random event, but rather specific proteins appear to be targeted. These protein targets may differ between different electrophiles, as illustrated with HQ-GSH and 4HNE modified proteins (), but may also exhibit overlap (eg the 50kD proteins). Chemical structure, reactivity, and ability to localize within the various intracellular compartments are several characteristics that likely govern which proteins are targeted by a given electrophile. We have provided evidence in support of the existence of “electrophile binding motifs” (EBMs) within proteins in the kidney that are selectively adducted by reactive electrophiles following GS-HQ administration [24
]. Whether similar EBMs exists within the bone marrow adductome deserves attention. With respect to the reactive metabolites derived from GS-HQ, they appear to selectively target (i) proteins with a high lysine (basic amino acid) content, and specifically (ii) proteins that contain lysine residues either flanking a potentially nucleophilic amino acid [KXK
], or containing two lysine residues preceded or followed by a nucleophilic amino acid [XKK
]. Although we have detected GS-HQ-adducted proteins in bone marrow of PHE/HQ treated rats () the identity of these proteins, and the site(s) of adduction, remain to be determined. One such target may be γ-GT. Indeed, HQ-GSH conjugates are substrates for, and inhibit, renal γ-GT [38
]. Consistent with this view, 2,3,5-GS-HQ inhibits γ-GT in bone marrow cell lysates and in HL-60 cells ().
Both “free” and protein-bound HQ-GSH conjugates (and metabolites thereof) retain the ability to redox cycle and generate ROS. One consequence of which is the initiation of lipid peroxidation. Cysteine thiols are common targets of reactive electrophilic metabolites, and of endogenous electrophiles, including lipid-derived α,β-unsaturated aldehydes, such as 4-hydroxynonenal (4HNE) and 4-oxononenal (4ONE), and ROS. In addition to the formation of HQ-GSH-derived protein adducts, PHE/HQ administration to rats also results in the formation of 4HNE-protein adducts (). Although in most cases, the toxicological significance of protein adducts remains uncertain, physiological concentrations of either 4HNE or 4ONE cause the cross-linking of bovine brain tubulin, and an inability of tubulin to polymerize [39
]. 4HNE also reduces ERK-1/2 phosphorylation causing a loss of activity and of nuclear localization. Interestingly, the loss of ERK activity is caused by a single 4HNE modification, on histidine 178. While the precise toxicological implication of proteins modified by 4HNE is not known, recent evidence suggests a role in the pathogenesis of several diseases [40