Peroxiredoxin 4: a multifunctional biomarker worthy of further exploration

Prx4 should be of special interest to clinicians as it may represent a new marker candidate that could be highly valuable in predicting the prognosis of an adverse outcome to risk stratify patients and to monitor therapy. Indeed, serum and expression levels, as well as post-translational modifications, of the enzyme have been linked to a number of different diseases, ranging from primary care to advanced critical care topics (see Additional file 1). The biomarker performance of Prx4 in these patients is truly encouraging, given the high heterogeneity and potential confounders, such as co-morbidities and medication. Indeed, Prx4 expression in the mouse liver was responsive to different drugs, including the analgesic paracetamol and the chemotherapeutic etoposide phosphate [28]. Altogether, Prx4 is assumed to be important in many single settings, which is very promising, but requires further validation. The results found by Chang and colleagues [10], for instance, are methodologically questionable because they are based on an immunoassay that worked with diluted patient plasma directly coated to the solid phase, which strongly limits specific antibody binding. Moreover, findings from different cancer studies are in part contradictory, concluding either good [16, 18] or bad prognosis [14, 19] from increased Prx4 expression in tumor tissue. Specifically, sepsis [29], cancer [30] and cardiovascular disease [31] are candidate diseases where Prx4 measurement may be of particular use as they share redox and inflammatory dysregulation and frequently lack valid severity and risk assessment by non-invasive biomarkers.

An improved understanding of the redox regulation, structural assembly and activity of Prx4 would provide a better insight into analyzing Prx4 as a biomarker in tissue or serum. Conoidin A, a specific Prx activity inhibitor, is currently under evaluation [32]. Tavender and Bulleid found Prx4 oxidation to be indicative of the degree of oxidative stress in the ER [3]. Atherosclerotic plaque development (Full et al., unpublished data), viral infection of A549 lung adenocarcinoma epithelial cells [25] and stimulation of human umbilical vein endothelial cells (HUVEC) with H₂O₂ and hydroperoxides [33] resulted in oxidation and a shifted isoelectric point of Prx4, respectively. Moreover, selective changes for either a putative 31-kDa precursor or 27-kDa secretable Prx4 have been reported in lung cancer [34] and spermatogenesis disorders [35, 36]. Consequently, detection of Prx4 hyperoxidation and oligomerization [3, 25, 37, 38] could be a meaningful, albeit challenging, measure of thiol-redox imbalances and should be integrated into Prx4 analysis. Existing expression data, however, require careful interpretation in light of a newly discovered 29.5-kDa splice variant of Prx4 [39] that cannot be recognized by the antibodies used in the immunoassay that has been developed for the detection of Prx4 dimers and oligomers in serum [11].

The amount of secreted Prx4 is most likely proportional to the initial intracellular expression levels. With exceptions [40], it has been shown in different in vivo models of infection that Prx4 expression was further potentiated in organs with strong endogenous expression of the enzyme [26, 41]. Consequently, tissue with high basal or up-regulated cellular expression of Prx4 likely represents a source for circulating Prx4 (Figure 1). Okado-Matsumoto and colleagues hypothesized that the enzyme is bound to the endothelial cells of blood vessels and released in response to a changed redox environment [9], matched by the high expression of Prx4 in endothelial cells [2]. There is sufficient further evidence that augmented Prx4 serum levels may be a result of secretion in response to pro-oxidant and pro-inflammatory signaling cascades, possibly related to regulation of NF-κB [2, 4, 6] (Figure 1). Indeed, Prx4 serum levels are correlated with endogenous antioxidants (albumin, bilirubin) and markers of inflammation (C-reactive protein, interleukin-6) in critically ill patients [13]. Moreover, changes in Prx4 expression or oxidation resulted from viral infection [4, 6, 24‐27] and oxidative stimulation [33, 38, 42, 43] in vitro and in vivo. Preliminary data from Madin-Darby canine kidney strain I cells stably transfected with Prx4 revealed a significant dose-dependent increase of Prx4 secretion after stimulation with H₂O₂ (Schulte et al., unpublished data). In addition, intracellular Prx4 was associated with nitrotyrosine expression in bladder and ovarian cancer [14, 16, 44] and nitric oxide-dependent regulation in stimulated macrophages [43], suggesting nitrosative stress as another promoter of Prx4 secretion. Ultimately, circulating Prx4 may originate from non-specific membrane leakage due to tissue damage and apoptotic cell death, representing a consequence of the aforementioned stress stimuli (Figure 1).

Intracellular Prx4 is mainly localized in the ER, where it contributes to redox and disulfide regulation [3], whereas no definite function has yet been assigned for extracellular Prx4 [6, 9]. Preincubation of HUVEC with reduced Prx4 resulted in improved cell viability after H₂O₂ treatment [9]. Protective anti-inflammatory effects have been reported for Prx4, when injected or overexpressed during infection [4, 6, 26]. In addition, Prx4 expression has been shown to be relevant for the antioxidant effects of natural compounds [42, 45, 46]. Chaperone activity in the extracellular space is conceivable, but requires experimental confirmation [7, 47]. A recent publication by Palande and colleagues indirectly linked Prx4 to the innate immune system by showing that ER-localized Prx4 has an inhibitory effect on granulocyte colony-stimulating factor, which is a key regulator of neutrophil production [48]. However, Prx4 secretion could lead to a loss of intracellular or membrane-bound Prx4, which would result in a changed capacity to remove H₂O₂, to regulate redox mechanisms or to participate in signaling pathways in the cellular vicinity. The resulting cellular oxidative stress and disturbed signaling would initiate and exacerbate pathological processes, largely increasing the probability of adverse outcome and death. On this basis, a speculative model is suggested, concluding that increased serum levels of Prx4 as a result of unfavorable Prx4 translocation could indicate higher morbidity (Figure 2). Indeed, elevated Prx4 serum levels have been linked to adverse outcome and higher mortality in critical care and emergency settings [12, 13] as well as in the general population (Abbasi et al., unpublished data) and type 2 diabetes mellitus patients (Alkhalaf et al., unpublished data). In addition, elevated intracellular Prx4 levels in carcinoma tissue are associated with improved survival [16, 18], in contrast to the fact that tumor cells escape apoptosis by increased Prx4 expression [49‐52]. This is in better agreement with results from other settings reporting higher mortality in cancer patients with increased Prx4 expression levels in tumor tissue [14, 19]. Finally, there is evidence that Prx isoforms and other antioxidants can substitute for Prx4 [14, 18, 25, 34, 53], resulting in compensation for functional consequences of Prx4 release.