We read with great interest the article by Zangrossi et al.  on OCT4 transcription factor expression in peripheral blood mononuclear cells (PBMNCs) and were intrigued by their conclusion challenging the role of OCT4 as a marker of stem cell pluripotency. The six different methods used and the meticulous testing for the specificity of polymerase chain reaction (PCR) and immunodetection assays that was undertaken in this study leave no doubt that PBMNCs express OCT4; we fully agree with the suggested additional measures to ensure that the detected PCR products are true OCT4 sequences and do not derive from OCT4 pseudogenes. An extensive review of misleading results that, in fact, correspond to pseudogene sequences but were originally attributed to OCT4 expression in normal and neoplastic cells has been published recently .
However, the existence of pseudogenes is not the only banana peel when investigating OCT4 expression, either at the mRNA or at the protein level. Just after OCT4 discovery, it became clear that different mRNAs are generated by alternative splicing from the OCT4 gene , resulting into two OCT4 protein isoforms with different N termini but identical POU DNA-binding and C-transactivation domains: a long one, OCT4A (isoform 1), and a short one, OCT4B (isoform 2). The existence of these two mRNA and protein isoforms has been validated, as appears in the Gene database from the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov). Importantly, only the long isoform, OCT4A, is responsible for stemness properties  and can sustain stem cell renewal , whereas OCT4B cannot. OCT4B does not seem to function as a transcriptional activator and is localized in the cytoplasm .
However, throughout the study by Zangrossi et al. , there is no reference to these OCT4 isoforms, and it is clear that OCT4A, the pluripotency regulator, was not specifically considered. By contrast, this study, which was so well performed, provides strong evidence that, in fact, OCT4B is detected in PBMNCs, because (a) the homemade PCR protocol does not distinguish OCT4A from OCT4B, since the target sequence amplified is common in both isoforms (exons 3–5 in NM_002701.4 [OCT4A] and in NM_203289.3 [OCT4B]); (b) the premade TaqMan assay from Applied Biosystems (Hs00742896_s1; Foster City, CA, http://www.appliedbiosystems.com) detects OCT4B exclusively, since the target sequence corresponds only to NM_203289.3; and (c) the mouse monoclonal antibody from Chemicon that was used for fluorescent immunocytochemistry recognizes amino acids 143–359 on the human OCT4 (Chemicon, Temecula, CA, http://www.chemicon.com). As originally described  and later confirmed , this sequence is identical in OCT4A and OCT4B proteins, which means that the Chemicon antibody cannot distinguish between the two isoforms. The expected results with this antibody, according to the subcellular localization described for these isoforms [4, 5], would be OCT4A staining in the nucleus and OCT4B staining in the cytoplasm, which is exactly the finding in the beautifully double-stained immunocytochemistry pictures shown in the report of Zangrossi et al. . Finally, another strong point suggesting that, actually, OCT4B is expressed in PBMNCs is the lack of OCT4 nuclear binding capacity in this cell fraction; the antibody used for mobility shifting is the only one of the three used in this study that specifically recognizes OCT4A at its unique N terminus (amino acids 1–134 for the mouse monoclonal antibody from Santa Cruz Biotechnology Inc. [Santa Cruz, CA, http://www.scbt.com]). This antibody does not recognize OCT4B. Thus, if PBMNCs expressed OCT4A, a band would appear on the gel, as was the case with the nuclear fraction of murine embryocarcinoma cells that were expected to contain this molecule; if PBMNCs expressed OCT4B but not OCT4A, no band would be detected, as was the case in this study.
Because OCT4A and OCT4B share common structural regions but not functional properties [2, , –5], a distinct and specific reference to the identity of these OCT4 derivatives appears to be of primary importance. Thus, when challenging OCT4 as a marker of stem cell pluripotency, it is important to specify which OCT4 is meant. As we show here, the data on OCT4 expression in PBMNCs that are presented by Zangrossi et al.  correspond to OCT4B, which is not a marker of stem cell pluripotency [4, 5]. The demonstration of OCT4B expression in PBMNCs is important per se, showing that the OCT4 gene may not be silenced in a subset of these differentiated cells and indicating that this isoform of as yet unknown function may play a role in the activation of lymphocytic populations, especially CD3+ T cells, in the context of immune response. On the other hand, OCT4A still cannot be challenged as a marker of stem cell pluripotency, since it was apparently not found to be expressed in PBMNCs.
Vassiliki Kotoula*, Spyros I. Papamichos, Alexandros F. Lambropoulos, * Department of Pathology, General Regional Hospital Papageorgiou, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece, Laboratory of Molecular Biology, First Department of Obstetrics and Gynecology, General Regional Hospital Papageorgiou, School of Medicine, Aristotle University of Thessaloniki, Thessaloniki, Greece