NK cells represent a special population of lymphoid cells which are amongst the first line of defenders against microbial or virally infected cells as well as tumor formation. NK cells are able to survey the body and discern the health of cells and tissue through surface proteins interacting with activating and inhibitory receptors on the NK cell surface. Of these receptors, the NCRs are vital in activating NK cells and evoking powerful effector functions in the way of cellular cytotoxicity and cytokine secretion. Due to the extreme deadliness of NK cytotoxicity, effector functions are highly regulated by the balance of signaling from inhibitory and activating receptors binding respective ligands on the surface of target cells. It is in the later stages of this fine balance where signals transduced through NKp44 influence NK cell functions by conveying either an activating signal through recognition of an unknown ligand, or an inhibitory signal through recognition of PCNA 
In this report we have identified interaction between NKp44 on NK cells and HLA I in complex with PCNA on target cells. We demonstrate that DB cells uniformly express PCNA on the cell surface and Class I HLA antibody, W6/32, blocks binding of NKp44-Ig fusion protein to the surface of DB cells (). We further show HLA I and PCNA colocalize on the surface of DB cells as visualized through confocal microscopy (). This association was confirmed in DB cells as well as DU145 prostate cancer cells through coimmunoprecipitation (). Finally, we show HLA I in association with PCNA participates in induction of inhibitory signaling through NKp44 on NK92-MI cells () and primary NK cells ().
Rosental et al.
recently reported PCNA expression on the surface of cancer cells, but only in the presence of NKp44 expressing NK cells or over expression of PCNA through transfection 
. In this report, we find PCNA is constitutively expressed on the cell surface of DB cells without gene transfection or the presence of NK cells. We propose several key aspects that may account for this difference. First, Rosental et al.
utilized cell lines of epithelial origin to demonstrate their results 
. We have utilized a B cell lymphoma cell line of hematopoietic/immunologic origin. DB cells used in this report differ greatly from the epithelial cells used in Rosental et al.
DB cells arise from a completely separate line of progenitor cells, thus executing different genetic expression programs as compared to cells of epithelial origin. DB cells also grow in large globular clumps of cells in suspension, thus exhibiting completely different morphology and growth kinetics than epithelial cells, which grow as a flattened single cell layer in tissue culture. Furthermore, we have confirmed in that MCF-7 cells grown on a cover slip do not express PCNA on the cell surface. We have also seen this result in DU145 cells grown on a cover slip (data not shown). As seen in , DB cells do express PCNA on the cell surface. Surface PCNA expression in DB cells was confirmed by flow cytometry, which demonstrated one uniform population, indicating all DB cells in culture express PCNA on their cell surface (). As DB cells replicate very quickly, endogenous PCNA levels are likely higher than other slower replicating cell lines, which may account for the steady expression of PCNA on the surface of DB cells. Additionally, PCNA expression is also dependent on p53 status of the cell. Wild type p53 specifically trans-activates the PCNA promoter at p53 response elements when p53 levels are low and represses PCNA expression at high levels of p53 during cellular stress, preventing replication 
. However, mutant p53 nonspecifically trans-activates the PCNA promoter, resulting in rampant expression of PCNA in malignant cells 
. In this light, cell surface PCNA may be an indicator of overall cell health, p53 status, and a marker for early onset of cancer transformation. Previous reports have also indicated Bat3, which controls acetylation and thus activation of p53, also colocalizes with HLA I on the surface of tumor and dendritic cells, where it serves as a ligand for NKp30 
. This suggests NCR ligands are a complex involving HLA I and DAMP molecules, which bind or colocalize to HLA I forming a complex of molecules enabling recognition by the NCRs. Since NKp30 recognizes Bat3 and NKp44 recognizes PCNA, the NCRs may recognize a multitude of different DAMP molecules, leading to diverse activation or inhibition of NK effector functions.
Modulation of NK cell activity may not only depend on the DAMP molecule associated with HLA I, but also the NCR that is recognizing the motif. Due to the dual nature of NKp44 signaling, it will be of interest to determine if recognition of PCNA, HLA I, or the motif as a whole is responsible for inhibition of NK cytotoxicity. Neither NKp30 nor NKp46 has been reported to contain an ITIM sequence; however, an immunosuppressive isoform of NKp30 resulting from a single-nucleotide polymorphism in the 3′ untranslatable region has been reported 
. Thus, NCRs may exhibit both inhibitory and activating functions. Whether the divergence of NKp44 functions depends on the individual DAMP molecule associated with HLA I or solely the presence of HLA I, implicating a unique activating ligand, remains to be elucidated. Interestingly, DAMP molecules High-Mobility Group Protein B1 and S100A8/9 have the ability to bind heparin sulfate and heparin sulfate proteoglycans, which are known to be coligands involved in NCR dependent recognition of tumor cells, resulting in secretion of IFN-γ but not cytotoxicity 
. We postulate a DAMP molecule may be the missing link in heparin sulfate dependent recognition of tumor cells, which would then elicit full NK cell effector function.
These results also bring further attention and new function to DAMP molecules, or proteins which are located and function intracellularly, but somehow localize to the extracellular membrane, despite lacking a traditional secretory leader sequence 
. These proteins are released by cells which have become injured in the absence of infection due to ischemia, hypoxia, transformation, chemotherapy, or other trauma 
. Aspects of PCNA transport are still a mystery as the protein does not contain a secretory sequence or a nuclear localization signal, even though the majority is located inside the nucleus 
. In regards to PCNA transport in DB cells, our attempts to inhibit cell surface PCNA were unsuccessful using Brefeldin A, indicating PCNA is not transported via the Golgi apparatus. Unconventional protein transport to the cell surface can occur however, by either direct transport from the cytoplasm across the plasma membrane, lysosomal secretion, exosome derived bodies, or vesicle shedding 
Analogous to Toll Like Receptors recognizing pathogen-associated molecular patterns, the NCRs may represent a class of receptors that participate in pattern recognition of DAMP molecules, whose identities may reflect the intracellular health of a cell in addition to the traditional method of HLA I presenting self peptide. In this manner, HLA I may also present DAMP molecules for identification by the NCRs, which would then regulate NK cell function potentially dependent on the DAMP molecule present and the NCR engaged. Knowledge of the identities and nature of DAMP molecules that bind to HLA I or other cell surface molecules to form ligands for the NCRs will shed light on NK cell recognition of target cells under healthy and disease conditions. This study further highlights the elegant complexity of NK cell recognition and activation by target cells.
In this study, we have identified interactions between PCNA and HLA I on the extracellular surface of tumor cells, which forms a complex ligand for NKp44. We also demonstrated that PCNA localization to the extracellular surface does not absolutely require the presence of an NKp44 expressing NK cell or over expression due to gene transfection. Interaction between NKp44 and PCNA in complex with HLA I resulted in inhibition of NK cell cytotoxicity.