A NKp80-based identification strategy reveals that CD56neg NK cells are not completely dysfunctional in health and disease.

iScience Article A NKp80-Based Identification Strategy Reveals that CD56 neg NK Cells Are Not Completely Dysfunctional in Health and Disease Ane Orrantia, In ˜igo Terre ´n, Alicia IzquierdoLafuente,...,Juan C. Garcı ´a-Ruiz, Olatz Zenarruzabeitia, Francisco Borrego francisco.borregorabasco@ osakidetza.eus HIGHLIGHTS NKp80 is a more precise marker than CD16 in order to identify CD56 neg NK cells CD16, but no NKp80, is downmodulated after cryopreservation and cell activation CD56 neg NK cells effector functions are not as diminished as previously described Orrantia et al., iScience 23, 101298 July 24, 2020 ª2020 The Authors. https://doi.org/10.1016/ j.isci.2020.101298 ll OPEN ACCESS iScience Article A NKp80-Based Identification Strategy Reveals that CD56 neg NK Cells Are Not Completely Dysfunctional in Health and Disease Ane Orrantia, 1 In ˜igo Terre ´n, 1 Alicia Izquierdo-Lafuente, 1 Juncal A. Alonso-Cabrera, 1 Victor Sanda ´, 1 Joana Vitalle ´, 1 Santiago Moreno, 2 Marı ´a Tasias, 3 Alasne Uranga, 4 Carmen Gonza ´lez, 4 Juan J. Mateos, 5 Juan C. Garcı ´a-Ruiz, 5 Olatz Zenarruzabeitia, 1 and Francisco Borrego 1,6,7, * SUMMARY Natural killer (NK) cells are usually identified by the absence of other lineage markers, due to the lack of cell-surface-specific receptors. CD56 neg NK cells, classically identified as CD56 neg CD16+, are very scarce in the peripheral blood of healthy people but they expand in some pathological conditions. However, studies on CD56 neg NK cells had revealed different results regarding the phenotype and functionality. This could be due to, among others, the unstable expression of CD16, which hinders CD56 neg NK cells’ proper identification. Hence, we aim to determine an alternative surface marker to CD16 to better identify CD56 neg NK cells. We have found that NKp80 is superior to CD16. Furthermore, we found differences between the functionality of CD56 neg NKp80+ and CD56 neg CD16+, suggesting that the effector functions of CD56 neg NK cells are not as diminished as previously thought. We proposed NKp80 as a noteworthy marker to identify and accurately re-characterize human CD56 neg NK cells. INTRODUCTION Natural killer (NK) cells are large granular lymphocytes that have the ability to recognize and kill transformed and virus-infected cells without prior sensitization (Caligiuri, 2008;Freud et al., 2017). In addition, they also produce and secrete a variety of cytokines and chemokines that modulate the immune response (Bancroft, 1993;Biron et al., 1999;Cooper et al., 2001;Robertson and Ritz, 1990). In human healthy adults, they comprise 5%–15% of circulating lymphocytes, being together with T cells and B cells, one of the three major lymphoid linages. However, in contrast to T and B cells, NK cells are a member of the innate lymphoid cells (ILCs) family, of which the main characteristic is the absence of rearranged antigen receptors encoded by the recombination activating genes (RAG) (Artis and Spits, 2015). Currently, ILCs are classified into five different subsets, depending on their effector functions, the cytokine pattern they secrete, and the transcription factors they need to develop and differentiate. The five subsets are NK cells, Group 1 ILC (ILC1), ILC2, ILC3, and lymphoid tissue-inducer (LTi) cells (Colonna, 2018;Vivier et al., 2018). NK cells and ILC1 are cells that produce interferon-g(IFN-g) as their signature cytokine and need the T-bet transcription factor to develop. On the other hand, ILC1 exhibit very little or no cytotoxic activity due to the low or zero levels of perforin and granzymes they express (Fuchs, 2016;Spits et al., 2016). Yet, the distinction between ILC1 and NK cells could be problematic because they express similar cell surface markers (Mjo ¨sberg and Spits, 2016;Spits et al., 2013;Trabanelli et al., 2018;Vivier et al., 2018). Nevertheless, in humans the NKp80 cell surface receptor is expressed on NK cells and seems to be an NK-cell-specific marker among human ILCs (Freud et al., 2016,2017). Furthermore, NK cells express both T-bet and Eomesodermin (Eomes) transcription factors, whereas ILC1 only express T-bet (Artis and Spits, 2015;Bal et al., 2020;Colonna, 2018;Mjo ¨sberg and Spits, 2016;Spits et al., 2013;Vivier et al., 2018). Commonly, as there is no known specific surface receptor that leads to NK cell identification within peripheral blood mononuclear cells (PBMCs), they are phenotypically defined as CD56+ cells that do not express other lineage markers, such as those that are specific for T cells (CD3), B cells (CD19), and myeloid cells (CD14). Furthermore, CD56 in combination with CD16, the low-affinity Fc gamma receptor IIIa, is generally used to distinguish different NK cell subsets that are present in healthy human peripheral blood (Freud 1 Biocruces Bizkaia Health Research Institute, Immunopathology Group, Barakaldo 48903, Spain 2 Ramo ´n y Cajal Health Research Institute (IRYCIS), Ramo ´n y Cajal University Hospital, Madrid 28034, Spain 3 Hospital Universitari i Politecnic La Fe, Valencia 46026, Spain 4 Biodonostia Health Research Institute, Donostia University Hospital, Donostia-San Sebastia ´n 20014, Spain 5 Biocruces Bizkaia Health Research Institute, Hematological Cancer Group, Cruces University Hospital, Barakaldo 48903, Spain 6 Ikerbasque, Basque Foundation for Science, Bilbao 48013, Spain 7 Lead Contact *Correspondence: francisco.borregorabasco@ osakidetza.eus https://doi.org/10.1016/j.isci. 2020.101298 iScience 23, 101298, July 24, 2020 ª2020 The Authors. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). 1 ll OPEN ACCESS et al., 2017;Montaldo et al., 2013). In this way, and based on the expression of these markers, two major subsets are identified: CD56 bright CD16+/and CD56 dim CD16+. In addition, two more subsets have been described: CD56 neg (CD56 neg CD16+) and unconventional CD56 dim (CD56+CD16) NK cells (Hu et al., 1995;Lugli et al., 2014;Roberto et al., 2018;Di Vito et al., 2019). However, none of these markers are specific to NK cells. For instance, although all ILCs are also negative for the abovementioned linage markers (Spits et al., 2013;Trabanelli et al., 2018;Vivier et al., 2018), CD56 is expressed by a subset of ILC3, known as NCR+ ILC3 and which, as NK cells, expresses natural cytotoxicity receptors (NCR) NKp30, NKp44, and NKp46 (Spits et al., 2013;Trabanelli et al., 2018;Vallentin et al., 2015;Vivier et al., 2018). On the other hand, although CD16 is widely accepted to be an NK-cell-specific marker among ILCs (Spits et al., 2016;Trabanelli et al., 2018;Vivier et al., 2018), it is known that CD16 could be downregulated following target cell activation (Borrego et al., 1994;Grzywacz et al., 2007;Peruzzi et al., 2013;Romee et al., 2013;Zhou et al., 2013) and cryopreservation (Lugthart et al., 2015). Therefore, the lack of marker specificity makes the process of identifying NK cells quite challenging and highlights the need to start using other set of cell surface markers. The CD56 neg subset expresses NK-cell-associated surface markers, such as CD16, CD94, and NKp46, in addition to the transcription factors Eomes and T-bet (Tarazona et al., 2002;Voigtetal.,2018). In healthy individuals, the presence of CD56 neg NK cells in blood is very rare (Bjo ¨rkstro ¨metal.,2010; Campos et al., 2014;Mu ¨ller-Durovic et al., 2019). However, years ago, in patients with chronic human immunodeficiency virus (HIV)-1 infection, a significant expansion of CD56 neg NK cells was reported (Hu et al., 1995;Lugli et al., 2014), which was associated with high HIV-1 viral load (Alter et al., 2005;Barker et al., 2007;Mavilio et al., 2005). Indeed, long-term non-progressors and patients who successfully suppress viral load after highly active antiretroviral therapy have CD56 neg NK cells levels comparable to the ones found in non-infected subjects (Alter et al., 2005;Brunetta et al., 2009). However, patients who fail to suppress viral load upon treatment have similar CD56 neg numbers to the ones with persistent viremia (Alteretal.,2005). In addition, in chronically HIV-1-infected individuals who developed broadly neutralizing antibodies (bnAbs), a high proportion of NK cells have a CD56 neg phenotype, whereas in patients who do not have bnAbs the proportion of CD56 neg NK cells was lower, although still high when compared with HIV-1 seronegative subjects (Bradley et al., 2018). Elevated frequencies of the CD56 neg subset has also been described in hepatitis C virus (HCV)-monoinfected and HCV/HIV1-coinfected people (Gonzalez et al., 2008,2009). Furthermore, the abnormal expansion of these cells correlated with monoinfected patient’s ability to respond to pegylated-IFNaand ribavirin treatment (Gonzalez et al., 2009). Moreover, treatments that suppress HCV replication decreases the number of CD56 neg NK cells in HCV-/HIV-1-coinfected patients (Gonzalez et al., 2008). On the other hand, it has been described that aging and human cytomegalovirus status has an effect on the frequency and distribution of NK cell subsets, increasing the percentage of CD56 neg NK cells (Campos et al., 2014;Mu ¨ller-Durovic et al., 2019). Studies with similar cohorts of patients differ in the frequency, functionality, and phenotype of the CD56 neg NK cell subset, which could be due to different gating strategies used for the identification of CD56 neg NK cells (Bjo ¨rkstro ¨m et al., 2010;Eller et al., 2009;Mavilio et al., 2005;Milush et al., 2013). It is very important to note that the abovementioned studies, and many others, have identified this NK cell subset as CD56 neg CD16+, and depending on the studies, they have or have not included in the gating strategy an exclusion channel for the exclusion of T cells, B cells, and/or monocytes within the cells of interest, i.e. CD56 neg NK cells. In addition, it is very well known that CD16 is downregulated by cryopreservation (Lugthart et al., 2015), after cytokine activation and target cell stimulation (Borrego et al., 1994;Grzywacz et al., 2007;Peruzzi et al., 2013;Romee et al., 2013;Zhou et al., 2013), and therefore the usage of this marker could lead to an inaccurate identification of the CD56 neg NK cells and inconsistent results. In this work, we have explored the possibility of using NKp80 as a marker to better identify CD56 neg NK cells. NKp80, the product of the KLRF1 gene, is an activating receptor expressed by virtually all mature human NK cells (Vitale et al., 2001).NKp80marksacriticalstepinNKcelldevelopment,asit defines functionally mature NK cells (Freud et al., 2016), and is an NK-cell-specific marker among human innate lymphoid cells (ILCs) (Vivier et al., 2018). We show that NKp80 is a more precise marker than CD16 in order to identify CD56 neg NK cells and that it is not downregulated after sample cryopreservation or cell activation. Importantly, using the NKp80 marker for the identification, we have demonstrated that the effector functions of CD56 neg NK cells are not as diminished as previously thought, both in health and in disease. ll OPEN ACCESS 2iScience 23, 101298, July 24, 2020 iScienc e Article RESULTS The NKp80 Receptor Is Superior to CD16 for the Identification of Circulating CD56 neg NK Cell Subset in Healthy People The CD16 receptor has traditionally been used, in combination with CD56, to identify the circulating NK cell subsets, with CD56 neg NK cells defined as CD56 neg CD16+ (Bjo ¨rkstro ¨m et al., 2010). However, CD16 is well known to be downregulated in some situations, such as, cryopreservation, after target cell stimulation and compounds, cell surface receptor, and cytokine activation (Borrego et al., 1994;Grzywacz et al., 2007; Lugthart et al., 2015;Peruzzi et al., 2013;Romee et al., 2013;Zhou et al., 2013). CD16 is shed from the cell surface as a consequence of matrix metalloproteinases activation, such as MT6 (also known as MMP25) and ADAM17 (Grzywacz et al., 2007;Peruzzi et al., 2013;Romee et al., 2013). With the aim to identify a more accurate marker with a more stable expression, we first compared CD16 with NKp80 receptor to identify CD56 neg NK cells in healthy donors. Our gating strategy included an exclusion channel (viability, CD3, CD14, and CD19) that allowed us to specifically study non-T, non-B, non-monocytes viable cells (Figure S1A). As previously described (Lugthart et al., 2015), CD16 expression was downregulated in cryopreserved samples; however, the expression of NKp80 was not significantly altered after cell freezing (Figure S2), suggesting that this receptor is more suitable for the detection of CD56 neg NK cells when it concerns to frozen cells. Very importantly, although no differences were seen regarding the percentage of CD56 neg NK cells selected using both markers (Figure S1B), there was a significantly higher frequency of Eomes+ cells in the CD56 neg NKp80+ subpopulation than in CD56 neg CD16+ cells (Figure 1A). Eomes is a specific intracellular marker for the detection of NK cells within the ILCs, given that it is a T-box transcription factor needed for the development and function of NK cells, whereas for example, ILC1 do not Figure 1. NKp80 Better Identifies CD56 neg NKCellsthanCD16inHealthyIndividuals (A) Bar graph showing the percentage of Eomes+ cells within CD56 neg CD16+ and CD56 neg NKp80+ populations. (B) Left part, representative contour plot showing the Eomes expression versus the size (FSC-A) of CD56 neg CD16+ cells. Data from a representative healthy donor is shown. Right part, bar graph showing the median of FSC-A parameter within CD56 neg CD16+Eomesand CD56 neg CD16+Eomes+ populations. (C) Bar graph showing the percentage of CD123 + cells within CD56 neg CD16+Eomes+, CD56 neg CD16+Eomes, CD56 neg NKp80+Eomes+, and CD56 neg NKp80+Eomespopulations. (D) Bar graph showing the percentage of Eomes+ cells within CD56 neg CD16+ and CD56 neg NKp80+ populations with or without the addition of anti-CD123 mAb to the exclusion channel (Exclusion C.). The mean with the standard error of the mean (SEM) is represented, except for (B) in which the median is represented. Each dot represents a donor. *p < 0.05, **p < 0.01, ****p < 0.0001, ns: not significant. ll OPEN ACCESS iScience 23, 101298, July 24, 2020 3 iScienc e Article express Eomes (Artis and Spits, 2015;Bal et al., 2020;Colonna, 2018;Mjo ¨sberg and Spits, 2016;Spits et al., 2013;Vivier et al., 2018). As the percentage of Eomes+ cells within the CD56 neg CD16+ subset was low, we considered the possibility that other CD16+ non-NK cells could have been selected using this gating strategy. This hypothesis was strengthened by the fact that within the CD56 neg CD16+ population, the Eomes  cells had larger size than Eomes+ cells (Figure 1B). Thus, we studied the expression of CD123 receptor (a-chain of the interleukin 3 receptor) expressed, among others, in plasmacytoid dendritic cells (pDCs) and basophils, which are characterized by a larger size and granularity (Collin et al., 2013;Han et al., 2008;McKenna et al., 2005;Vitalle ´et al., 2019a;Zenarruzabeitia et al., 2019). Results showed that CD56 neg CD16+Eomescells expressed CD123, in contrast to CD56 neg NKp80+Eomescells that barely did (Figure 1C). Furthermore, the addition of an anti-CD123 mAb to the exclusion channel revealed that the frequency of CD56 neg CD16+Eomes+ cells significantly increased but still tended to be lower compared with CD56 neg NKp80+Eomes+ cells (Figure 1D). These results suggested that the inaccuracy in the identification of the CD56 neg NK cell subset using the CD16 marker in the gating strategy is due to the selection of Eomescells that, at least partially, could be pDCs and/or basophils, which are characterized by the expression of CD123. Given that there are no significant differences in the frequency of CD56 neg CD16+ and CD56 neg NKp80+ cells (Figure S1B), but the latter expressed significantly higher levels of Eomes (Figure 1A), we analyzed if NKp80 is inclusive of the CD56 neg CD16+ subset. Results showed no significant differences in the frequency of CD16+NKp80,CD16NKp80+, and CD16+NKp80+ subsets within the CD56 neg cells and that half of the CD56 neg NKp80+ NK cells also co-express CD16 (Figure 2A).Importantly,when the expression of Eomes was analyzed within these three subsets, we found that the frequency of Eomes+ cells was very low in CD16+NKp80cells and significantly higher in both CD16NKp80+ and CD16+NKp80+ cells (Figure 2B). Moreover, although CD16+NKp80cells included a significant frequency of CD123+ cells (50%), both CD16NKp80+ and CD16+NKp80+ subsets comprised negligible levels of CD123+ cells (Figure 2C). Altogether, these results suggest that although CD16 and NKp80 do not completely identify the same CD56 neg cells, NKp80 is more precise for the identification of CD56 neg NK cells. More recently, it was shown that including CD7 as an additional marker to the CD56 neg CD16+ cell subset was an effective method to accurately identify CD56 neg NK cells (Milush et al., 2009,2013). Therefore, we compared the frequency of Eomes+ cells using CD7 or NKp80 markers to identify the CD56 neg NK cell subpopulation. There were no significant differences between CD7+CD56 neg CD16+ and CD56 neg NKp80+ cells in terms of Eomes expression, indicating that both strategies were equally effective for the identification of CD56 neg cells in these specific experimental settings. However, when we only used the CD7 marker instead of NKp80 the frequency of Eomes+ cells in the CD7+CD56 neg population was much lower than in both CD7+CD56 neg CD16+ and CD56 neg NKp80+ cells (Figure 3A). On the other hand, combining the CD7 and CD16 markers to identify the CD56 neg NK cell subset could be an obstacle in certain situations, especially due to the unstable expression of the CD16 receptor after cryopreservation and cell stimulation, and also the need for an additional monoclonal antibody to the panel and an extra flow cytometer detector, which could be overcome by only using NKp80 as a marker for the identification of CD56 neg NK cells. Next, we studied the CD300a (Dimitrova et al., 2016;Vitalle ´et al., 2019b;Zenarruzabeitia et al., 2016)and 2B4 (CD244) (Endt et al., 2007) receptors that are also expressed in NK cells, although not exclusively, as markers for the identification of the CD56 neg NK cells. Results showed a lower frequency of Eomes+ cells both in CD56 neg CD300a+ and in CD56 neg 2B4+ cells compared with CD56 neg NKp80+ cells (Figures 3Band 3C). Moreover, the addition of an anti-CD123 mAb to the exclusion channel minimally increased the frequency of Eomes+ cells (Figure 3D). Altogether, our results demonstrate that NKp80 is the best cell surface marker to identify the CD56 neg NK cell subset with a very high certainty and accuracy. CD56 neg NKp80+ Cells Are Expanded in HIV-Infected People and Patients with Multiple Myeloma As CD56 neg NK cells are infrequent in the peripheral blood of healthy donors, we next evaluated the accuracy of the NKp80 receptor to identify the expanded CD56 neg NK cells in pathological conditions, such as HIV infection and multiple myeloma (Figure 4A; Table S1). No significant differences were noticed in the frequency of Eomes+ cells between CD56 neg CD16+ and CD56 neg NKp80+ subpopulations in untreated ll OPEN ACCESS 4iScience 23, 101298, July 24, 2020 iScienc e Article HIV-1 infected subjects. However, the frequency of Eomes+ cells was significantly higher in CD56 neg NKp80+ than in CD56 neg CD16+ cells in HIV-1 infected subjects under combined antiretroviral therapy (cART) (Figure 4B). The differences between patient groups could be explained because the relative frequency of CD56 neg CD16+ non-NK cells (Eomes) is lower in untreated patients due to a higher expansion of the CD56 neg CD16+ NK cells (Eomes+) (Alter et al., 2005;Barker et al., 2007;Mavilio et al., 2005). Thus, CD16 could only serve to identify CD56 neg NK cells in certain pathological conditions in which this subset is very highly expanded.In addition, a higher frequency of Eomes+ cells within CD56 neg NKp80+ cellsincomparisonwithCD56 neg CD16+ cells was also noticeable in multiple myeloma patients (Figure 4C), in which CD56 neg NK cell expansion is more similar to the one of HIV-1 infected subjects under cART (Figure 4A). These findings suggest that NKp80, as demonstrated in healthy donors, is a noteworthy alternative to CD16 as a marker to identify CD56 neg NK cells also in disease. CD56 neg NKp80+ Cells Are More Functional than CD56 neg CD16+ Cells Although surface receptor profiling and proteomic analyses indicate that CD56 neg have a phenotypic relationship to CD56 dim (Voigt et al., 2018), CD56 neg NK cells have been described as functionally impaired compared with CD56 dim NK cells (Alter et al., 2005;Bjo ¨rkstro ¨m et al., 2010;Mavilio et al., 2005;Milush et al., 2013). However, others have proposed that these cells are skewed rather than dysfunctional (Eller Figure 2. CD56 neg CD16+NKp80Subset Mostly Include Non-NK Cells (A) Bar graph showing the percentage of CD56 neg CD16+NKp80,CD56 neg CD16NKp80+ and CD56 neg CD16+NKp80+ subsets within the CD56 neg cells. (B) Bar graph showing the percentage of Eomes+ cells within the CD56 neg CD16+NKp80,CD56 neg CD16NKp80+, and CD56 neg CD16+NKp80+ populations. (C) Bar graphs showing the percentage of CD123+ cells within the CD56 neg CD16+NKp80,CD56 neg CD16NKp80+, and CD56 neg CD16+NKp80+ populations. The mean with the standard error of the mean (SEM) is represented. Each dot represents a donor. ***p < 0.001, ****p < 0.0001, ns: not significant. ll OPEN ACCESS iScience 23, 101298, July 24, 2020 5 iScienc e Article et al., 2009). These differences may be due to inaccurate identification of CD56 neg NK cells using the CD16 marker. Therefore, we studied the effector functions of CD56 neg CD16+ and CD56 neg NKp80+ cells from healthy donors and HIV-1 infected subjects, by measuring degranulation (CD107a) (Alter et al., 2004) and production of TNF and IFNgafter cytokines and K562 target cell stimulations (Terre ´n et al., 2018). First, we wanted to compare CD16 and NKp80 expression downregulation after NK cell stimulation. Results showed that NKp80 was not significantly downregulated after K562 cell line and cytokine stimulation, whereas CD16 expression significantly decreased after stimulation with both stimuli (Figures 5Aand5B). In terms of functionality, we observed that CD56 neg NKp80+ cells exhibited higher production of TNF and IFNgthan CD56 neg CD16+ cells in treated HIV-1 infected subjects and that they showed a tendency to a higher production of both cytokines in untreated subjects (Figure 6A). Furthermore, in healthy donors, both cytokine production and the degranulationcapabilitytendedtobehigherintheCD56 neg NKp80+ cellsthaninCD56 neg CD16+ cells (Figure 6B). Finally, we compared the functionality of the CD56 dim and CD56 neg NK cells. Very importantly, our results showed that, although CD56 neg NK cells have lower effector functions than CD56 dim NK cells (Figure S3) in healthy donors and HIV-infected people, their functionality is much lower when we used the CD16-based gating strategy to identify CD56 neg NK cells than when we used Figure 3. NKp80 Better Identifies CD56 neg NK Cells than CD7, CD300a, and 2B4 (CD244) in Healthy Individuals (A) Bar graph showing the percentage of Eomes+ cells within CD7+CD56 neg CD16+, CD56 neg CD7+, CD56 neg CD16+, and CD56 neg NKp80+ populations. (B) Bar graph showing the percentage of Eomes+ cells within CD56 neg CD300a+, CD56 neg CD16+, and CD56 neg NKp80+ populations. (C) Bar graph showing the percentage of Eomes+ cells within CD56 neg 2B4+, CD56 neg CD16+, and CD56 neg NKp80+ populations. (D) Bar graphs showing the percentage of Eomes+ cells within CD56 neg CD300a+ and CD56 neg 2B4+ populations with or without the addition of anti-CD123 mAb to the exclusion channel (exclusion C.). The mean with the standard error of the mean (SEM) is represented. Each dot represents a donor. *p < 0.05, **p < 0.01, ***p < 0.001, ns: not significant. ll OPEN ACCESS 6iScience 23, 101298, July 24, 2020 iScienc e Article the NKp80-based gating strategy. Altogether, our results indicate that CD56 neg NK cells, defined as viable CD3-CD19-CD14-CD56 neg NKp80+ cells, are significantly less dysfunctional than previously thought. DISCUSSION It is of utmost importance to correctly phenotype the different subpopulations of immune cells not only in healthy people but also in disease situations. In the latter, variations in the frequency of cells subsets and in their effector functions, in comparison to healthy state, are frequently observed. These variations can help to understand the pathogenesis of diseases, and moreover, these changes can serve as biomarkers for the diagnosis, prognosis, and/or to determine the efficacy of the treatment. Some immune cell types are characterized by the expression of specific lineage cell surface markers. For example, T cells are CD3+, whereas other cell types do not express CD3. However, a specific NK cell surface marker has not been described yet. In general terms, the minimum requirement to define circulating human NK cells is based on the expression of CD56 and the absence of the CD3 marker, because an important subpopulation of T cells expresses CD56 (Ortaldo et al., 1991). However, there are other cells, such as ILC3, which can also express CD56 (Artis and Spits, 2015;Spits et al., 2013;Vallentin et al., 2015;Vivier et al., 2018). On the other hand, according to the expression of CD56 and CD16, NK cells have been classified into four subpopulations: CD56 bright (CD56 bright CD16+/), CD56 dim (CD56 dim CD16+), unconventional CD56 dim Figure 4. NKp80 Better Identifies CD56 neg NKCellsthanCD16inPathologicalConditions (A) Bar graphs showing the percentage of CD56 neg CD16+ (left part) and CD56 neg NKp80+ (right part) subsets within total NK cells from healthy donors (HD), untreated HIV-1 infected people (HIV), HIV-1-infected patients under cART (cART) and multiple myeloma patients. (B) Bar graphs showing the percentage of Eomes+ cells within CD7+CD56 neg CD16+, CD56 neg CD7+, CD56 neg CD16+, and CD56 neg NKp80+ populations in untreated HIV-1-infected subjects (HIV) and HIV-1-infected patients under cART (cART). (C) Bar graph showing the percentage of Eomes+ cells within CD56 neg CD16+ and CD56 neg NKp80+ populations in multiple myeloma patients. The mean with the standard error of the mean (SEM) is represented. Each dot represents a donor. *p < 0.05, **p < 0.01, ns: not significant. ll OPEN ACCESS iScience 23, 101298, July 24, 2020 7 iScienc e Article (CD56 dim CD16), and CD56 neg (CD56 neg CD16+) (Freud et al., 2017;Hu et al., 1995;Montaldo et al., 2013; Roberto et al., 2018;Di Vito et al., 2019). Finally, although the Eomes transcription factor is also expressed in CD4+ and CD8+ T cells (Knox et al., 2014;Narayanan et al., 2010), it is used as a specific intracellular marker of NK cells within the ILCs. CD56 neg cells represent a very low percentage of NK cells in peripheral blood from healthy people (Bjo ¨rkstro ¨m et al., 2010;Campos et al., 2014;Mu ¨ller-Durovic et al., 2019). However, in certain diseases there is a very significant expansion of this cell subpopulation (Alter et al., 2005;Barker et al., 2007;Forconi et al., 2018;Hu et al., 1995;Lugli et al., 2014;Mavilio et al., 2005). To our knowledge, in previous publications, the expression of the CD16 marker have so far been used in all the gating strategies to identify the CD56 neg NK cell subset. Some authors have made use of a strategy based on only three markers (CD3, CD56, and CD16) (Alter et al., 2005,2006;Barker et al., 2007;Campos et al., 2014;Frias Figure 5. CD16, But no NKp80, Is Downregulated after K562 Cell Line and IL-12+IL-18 Cytokine Stimulation (A) Representative pseudocolor plot graphs comparing the expression of CD16 and NKp80 in non-stimulated condition, and after K562 cell line and IL-12+IL-18 cytokine stimulation. Data from a representative healthy donor is shown. (B) Histograms showing the median fluorescence intensity (MFI) of CD16 and NKp80 on NK cells in non-stimulus condition and after K562 cell line and IL-12+IL-18 cytokine stimulation. Data from a representative healthy donor is shown. ll OPEN ACCESS 8iScience 23, 101298, July 24, 2020 iScienc e Article iScience, Volume 23 Supplemental Information A NKp80-Based Identification Strategy Reveals that CD56 neg NK Cells Are Not Completely Dysfunctional in Health and Disease Ane Orrantia, Iñigo Terrén, Alicia Izquierdo-Lafuente, Juncal A. Alonso-Cabrera, Victor Sandá, Joana Vitallé, Santiago Moreno, María Tasias, Alasne Uranga, Carmen González, Juan J. Mateos, Juan C. García-Ruiz, Olatz Zenarruzabeitia, and Francisco Borrego Viable/CD3/14/193.96 NK cells CD56neg NK cells A B Figure S1. Identification of CD56neg NK cells. Related to Figure 1. (A) Pseudocolor and contour plot graphs representing the gating strategy utilized for the identification of CD56neg NK cells. Data from a representative cryopreserved sample from a healthy donor is shown. Lymphocytes were electronically gated based on their forward and side scatter parameters and then single cells were selected. To identify NK cells, the population negative for the exclusion channel (viability, CD3, CD14 and CD19) was selected. Then CD56neg NK cells were identified using different gating strategies. (B) Percentage of CD56negCD16+ and CD56negNKp80+ cells in healthy donors. Bar graph showing the percentage of CD56negCD16+ and CD56negNKp80+ cells in healthy donors. The mean with the standard error of the mean (SEM) is represented. Each dot represents a donor. ns: not significant. A CD56 CD16 NKp80 B Fig. S2. CD16 but not NKp80 is downregulated after cell cryopreservation. Related to Figure 1. (A) Representative pseudocolor plot graphs comparing the expression of CD16 and NKp80 in fresh and cryopreserved samples. Data from a representative healthy donor is shown. (B) Histograms showing the median fluorescence intensity (MFI) of CD16 and NKp80 on NK cells in fresh and cryopreserved samples. Data from a representative healthy donor is shown. CD56dim CD56neg K562 cell line CD16 based gating IL-12 + IL-18 NKp80 based gating HIV cART HD HIV cART HD HIV cART HD HIV cART HD HIV cART HD HIV cART HD Fig. S3. Degranulation (CD107a) and cytokine production by NK cells in response to the K562 cell line and IL-12+IL-18 cytokine stimulation. Related to Figure 6. Bar graphs showing the percentage of CD56dim and CD56neg NK cells positive for CD107a and TNF after K562 cell line stimulation and IFNγ after IL-12+IL-18 cytokine stimulation from HIV-1 infected subjects (HIV), HIV-1 infected patients under cART (cART) and healthy donors (HD). The mean with the standard error of the mean (SEM) is represented. Each dot represents a donor. Table S1. Clinical data of untreated HIV-1 infected subjects, under cART HIV-1 infected patients and multiple myeloma patients. Related to Figure 4 and Figure 6. Untreated HIV-1 subjects HIV-1 patients on cART Multiple myeloma patients Median Range (minmax) Median Range (minmax) Median Range (minmax) Sex Female: n=0 Male: n=9 - Female: n=1 Male: n=7 - Female: n=6 Male: n=3 - Age (years) 39 (32-59) 49 (34-56) 63 (53-74) cART (months) - - 13 (10-27) - - TRANSPARENT METHODS Contact for reagents and resource sharing Further information and request for resources and reagents should be directed to and will be fulfilled by the Lead Contact, Francisco Borrego ([email protected]). This study did not generate new unique reagents. Experimental Model and Subject Details For this study, buffy coats from 24 healthy adult donors and cryopreserved peripheral blood mononuclear cells (PBMCs) from 9 multiple myeloma patients were collected through the Basque Biobank for Research (http://www.biobancovasco.org), which complies with the quality management, traceability and biosecurity, set out in the Spanish Law 14/2007 of Biomedical Research and in the Royal Decree 1716/2011. The study was approved by the Basque Ethics Committee for Clinical Research (PI2014017 and PI+CES+INC-BIOEF 2017-03). All subjects provided written and signed informed consent in accordance with the Declaration of Helsinki. In addition, cryopreserved PBMCs from healthy donors (n=5), untreated HIV-1 infected subjects (n=9) and patients under cART (n=8) were provided by the HIV BioBank integrated in the Spanish AIDS Research Network (RIS) (Supplementary Information, Appendix I). Samples were processed following current procedures and frozen immediately after their reception. All patients participating in the study gave their informed consent and protocols were approved by institutional ethical committees. All HIV-1 infected patients were asymptomatic when the sample was collected, were not coinfected with hepatitis C virus (HCV), had more than 200 CD4+ T cells/mm3 and they had never been diagnosed with AIDS. Untreated HIV-1 infected subjects had detectable viremia (>10,000 HIV-RNA copies/ml) and they had never been treated with cART, while patients under cART had undetectable viremia and had been treated with cART at least for 6 months. Clinical data of HIV-1 infected patients were obtained from the RIS database. Clinical data are shown in Table S1. Antibodies and reagents For flow cytometry-based procedures, the following fluorochrome-conjugated anti-human monoclonal antibodies (mAbs) were used: Brilliant Violet (BV)421 anti-CD56 (NCAM 16.2), BV510 anti-CD3 (UCHT1), BV510 anti-CD14 (MФP9), BV510 anti-CD19 (SJ25C1), BV510 antiCD123 (9F5), PE anti-CD123 (9F5), PE anti-CD7 (M-T701) and PerCP-Cy5.5 anti-IFNγ (B27) from BD Biosciences; FITC anti-CD16 (B73.1) and APC anti-TNF (MAb11) from BioLegend; PE anti-CD300a (E59.126) and PE anti-2B4 (clone C1.7) from Beckman Coulter; PE anti-CD107a (REA792) and PE-Vio770 anti-NKp80 (4A4.D10) from Miltenyi Biotec; eFluor660 anti-Eomes (WD1928) from eBioscience. Dead cells were detected with the LIVE/DEAD™ Fixable Aqua Dead Cell Stain Kit for 405nm excitation from Invitrogen, following manufacturer’s protocol. The following reagents were also used: Foxp3/Transcription Factor Staining Buffer Set from eBioscience; Brilliant Stain Buffer, BD GolgiStop™ Protein Transport Inhibitor (monensin), BD GolgiPlug™ Protein Transport Inhibitor (brefeldin A) and BD Perm/Wash™ Buffer from BD Bioscience; and paraformaldehyde (PFA) from Sigma-Aldrich/Merck. Methods Details Peripheral blood mononuclear cell isolation. Fresh PBMCs from healthy donors were obtained from buffy coats by Ficoll (GE Healthcare) density gradient centrifugation and cryopreserved in Fetal Bovine Serum (FBS) (GE Healthcare Hyclone) with 10% Dimethylsulfoxide (DMSO) (Thermo Scientific Scientific). Flow cytometry: Phenotypical studies. For phenotypical studies, cryopreserved PBMCs from healthy donors, HIV-1-infected subjects and multiple myeloma patients were thawed at 37ºC and washed twice with RPMI 1640 medium with L-Glutamine (Lonza). Then, cells were incubated for 1 hour at 37ºC with 10U DNase (Roche) in R10 medium (RPMI 1640 medium containing GlutaMAX from Thermo Fisher Scientific, 10% FBS and 1% Penicillin-Streptomycin from Thermo Fisher Scientific). Afterwards, cells were counted and washed with Phosphate Buffered Saline (PBS) (Gibco, Thermo Fisher Scientific). Then, dead cells were excluded by using the LIVE/DEAD reagent (Invitrogen, Thermo Fisher Scientific). For the staining of NK cell surface markers, cells were first washed with PBS containing 2.5% of Bovine Serum Albumin (BSA) (Millipore) and then incubated for 30 minutes at 4ºC with fluorochrome-conjugated mAbs. To identify NK cells, first viable cells that were negative for CD3, CD14 and CD19 were electronically gated, and then, by using the antiCD56 mAb in combination with mAbs against CD16, NKp80, CD300a, 2B4 and/or CD7, NK cells were classified in three subsets: CD56bright, CD56dim and CD56neg. After this, cells were washed again with 2.5% BSA in PBS and fixed and permeabilized with Foxp3/Transcription Factor Staining Buffer Set (eBioscience, Thermo Fisher Scientific) following manufacturer’s recommendations. Finally, cells were stained using anti-Eomes mAb for 30 minutes at room temperature (RT) and washed with Permeabilization Buffer 1x (eBioscience). Sample acquisition was carried out in a MACSQuant Analyzer 10 flow cytometer (Miltenyi Biotec). Flow cytometry: Functional assays. For functional assays, after DNase treatment, PBMCs from HIV-1-infected subjects and healthy donors were counted and plated at 0.5 x 106 cells/well in 48 well plates in NK cell culture medium (RPMI 1640 medium with GlutaMAX, 10% FBS, 1% penicillin streptomycin, 1% nonessential amino acids and 1% Sodium-Pyruvate). PBMCs were then primed with interleukin (IL)- 15 (10ng/mL) and cultured for 20 hours. For cytokine stimulation, IL-12 (10ng/mL) and IL-18 (50ng/mL) were also added to plated PBMCs. For target cell stimulation, K562 cells were added after the 20 hours of culture in IL-15 at Effector:Target (E:T) 1:1 ratio (0.5x106 PBMCs and 0.5x106 K562 cells). Then, IL-12+IL-18 and K562 stimulated PBMCs were cultured for 6 hours. CD107a was added at the start of the co-culture period and protein transport inhibitors were added after 1 hour for the rest of the incubation time following manufacturer’s protocol. Afterwards, viability and surface marker staining was performed as explained above. For intracellular staining, cells were fixed with 4% PFA for 15 minutes on ice and then washed twice with 2.5% BSA in PBS. After this, cells were permeabilized with BD Perm/Wash Buffer 1X for 15 minutes at RT. Finally, the corresponding mAbs were added for 30 minute and cells were washed with BD Perm/Wash Buffer 1X before acquisition in the MACSQuant Analyzer 10 flow cytometer (Miltenyi Biotec). The percentage of positive cells for CD107a, IFNγ and TNF was calculated after subtracting the non-stimulus condition. Quantification and Statistical Analysis. Data were analysed using FlowJo™ v10.4.1. GraphPad Prism v8.01 software was used for graphical representation and statistical analysis. As specified in all figure legends, each dot in the graphs represents a donor. Data were represented showing means ± standard error of the mean (SEM) or median as indicated in the figure legend. Prior to statistical analyses, data were tested for normal distribution with Kolmogórov-Smirnov normality test. In the case of multiple myeloma patients, an outlier was identified and removed using Grubb test (alpha=0.05). If data were normally distributed, t test for paired values was used to determine significant differences. Non-normal distributed data were compared with Wilcoxon matched-pairs signed rank test. Kruskal-Wallis test was used for multiple comparisons of non-normal data (Figure 4A). *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001. Appendix I: CoRIS Members Executive committee Santiago Moreno, Inma Jarrín, David Dalmau, Maria Luisa Navarro, Maria Isabel González, Federico Garcia, Eva Poveda, Jose Antonio Iribarren, Félix Gutiérrez, Rafael Rubio, Francesc Vidal, Juan Berenguer, Juan González, M Ángeles Muñoz-Fernández. Fieldwork data management and analysis Inmaculada Jarrin, Belén Alejos, Cristina Moreno, Carlos Iniesta, Luis Miguel Garcia Sousa, Nieves Sanz Perez, Marta Rava BioBanK HIV Hospital General Universitario Gregorio Marañón M Ángeles Muñoz-Fernández, Irene Consuegra Fernández Hospital General Universitario de Alicante (Alicante) Esperanza Merino, Gema García, Irene Portilla, Iván Agea, Joaquín Portilla, José SánchezPayá., Juan Carlos Rodríguez, Lina Gimeno, Livia Giner, Marcos Díez, Melissa Carreres, Sergio Reus, Vicente Boix, Diego Torrús Hospital Universitario Central de Asturias (Oviedo) Víctor Asensi, Eulalia Valle, María Eugenia Rivas Carmenado, Tomas Suarez-Zarracina Secades, Laura Pérez Is Hospital Universitario 12 de Octubre (Madrid) Rafael Rubio, Federico Pulido, Otilia Bisbal, Asunción Hernando, Lourdes Domínguez, David Rial Crestelo, Laura Bermejo, Mireia Santacreu Hospital Universitario de Donostia (Donostia-San Sebastián) José Antonio Iribarren, Julio Arrizabalaga, María José Aramburu, Xabier Camino, Francisco Rodríguez-Arrondo, Miguel Ángel von Wichmann, Lidia Pascual Tomé, Miguel Ángel Goenaga, Mª Jesús Bustinduy, Harkaitz Azkune, Maialen Ibarguren, Aitziber Lizardi, Xabier Kortajarena., Mª Pilar Carmona Oyaga, Maitane Umerez Igartua Hospital General Universitario De Elche (Elche) Félix Gutiérrez, Mar Masiá, Sergio Padilla, Catalina Robledano, Joan Gregori Colomé, Araceli Adsuar, Rafael Pascual, Marta Fernández, José Alberto García, Xavier Barber, Vanessa Agullo Re, Javier Garcia Abellan, Reyes Pascual Pérez, María Roca Hospital General Universitario Gregorio Marañón (Madrid) Juan Berenguer, Juan Carlos López Bernaldo de Quirós, Isabel Gutiérrez, Margarita Ramírez, Belén Padilla, Paloma Gijón, Teresa Aldamiz-Echevarría, Francisco Tejerina, Francisco José Parras, Pascual Balsalobre, Cristina Diez, Leire Pérez Latorre., Chiara Fanciulli Hospital Universitari de Tarragona Joan XXIII (Tarragona) Francesc Vidal, Joaquín Peraire, Consuelo Viladés, Sergio Veloso, Montserrat Vargas, Montserrat Olona, Anna Rull, Esther Rodríguez-Gallego, Verónica Alba., Alfonso Javier Castellanos, Miguel López-Dupla Hospital Universitario y Politécnico de La Fe (Valencia) Marta Montero Alonso, José López Aldeguer, Marino Blanes Juliá, María Tasias Pitarch, Iván Castro Hernández, Eva Calabuig Muñoz, Sandra Cuéllar Tovar, Miguel Salavert Lletí, Juan Fernández Navarro. Hospital Universitario La Paz/IdiPAZ Juan González-Garcia, Francisco Arnalich, José Ramón Arribas, Jose Ignacio Bernardino de la Serna, Juan Miguel Castro, Ana Delgado Hierro, Luis Escosa, Pedro Herranz, Víctor Hontañón, Silvia García-Bujalance, Milagros García López-Hortelano, Alicia González-Baeza, Maria Luz Martín-Carbonero, Mario Mayoral, Maria Jose Mellado, Rafael Esteban Micán, Rocio Montejano, María Luisa Montes, Victoria Moreno, Ignacio Pérez-Valero, Guadalupe Rúa Cebrián, Berta Rodés, Talia Sainz, Elena Sendagorta, Natalia Stella Alcáriz, Eulalia Valencia. Hospital Universitari MutuaTerrassa (Terrasa) David Dalmau, Angels Jaén, Montse Sanmartí, Mireia Cairó, Javier Martinez-Lacasa, Pablo Velli, Roser Font, Marina Martinez, Francesco Aiello Hospital Universitario de La Princesa (Madrid) Ignacio de los Santos, Jesus Sanz Sanz, Ana Salas Aparicio, Cristina Sarria Cepeda, Lucio Garcia-Fraile Fraile, Enrique Martín Gayo. Hospital Universitario Ramón y Cajal (Madrid) Santiago Moreno, José Luis Casado Osorio, Fernando Dronda Nuñez, Ana Moreno Zamora, Maria Jesús Pérez Elías, Carolina Gutiérrez, Nadia Madrid, Santos del Campo Terrón, Sergio Serrano Villar, Maria Jesús Vivancos Gallego, Javier Martínez Sanz, Usua Anxa Urroz, Tamara Velasco Hospital General Universitario Reina Sofía (Murcia) Enrique Bernal, Alfredo Cano Sanchez, Antonia Alcaraz García, Joaquín Bravo Urbieta, Angeles Muñoz Perez, Maria Jose Alcaraz, Maria del Carmen Villalba. Hospital Nuevo San Cecilio (Granada) Federico García, José Hernández Quero, Leopoldo Muñoz Medina , Marta Alvarez, Natalia Chueca, David Vinuesa García , Clara Martinez-Montes., Carlos Guerrero Beltran, Adolfo de Salazar Gonzalerz, Ana Fuentes Lopez Centro Sanitario Sandoval (Madrid) Montserrat Raposo Utrilla, Jorge Del Romero, Carmen Rodríguez, Teresa Puerta, Juan Carlos Carrió, Mar Vera, Juan Ballesteros, Oskar Ayerdi. Hospital Universitario Son Espases (Palma de Mallorca) Melchor Riera, María Peñaranda, Mª Angels Ribas, Antoni A Campins, Carmen Vidal, Francisco Fanjul, Javier Murillas, Francisco Homar., Helem H Vilchez, Maria Luisa Martin, Antoni Payeras. Hospital Universitario Virgen de la Victoria (Málaga) Jesús Santos, Crisitina Gómez Ayerbe, Isabel Viciana, Rosario Palacios, Carmen Pérez López, Carmen Maria Gonzalez-Domenec Hospital Universitario Virgen del Rocío (Sevilla) Pompeyo Viciana, Nuria Espinosa, Luis Fernando López-Cortés. Hospital Universitario de Bellvitge (Hospitalet de Llobregat) Daniel Podzamczer, Arkaitz Imaz, Juan Tiraboschi, Ana Silva, María Saumoy, Paula Prieto Hospital Costa del Sol (Marbella) Julián Olalla Sierra, Javier Pérez Stachowski., Alfonso del Arco, Javier de la Torre, José Luis Prada, José María García de Lomas Guerrero Hospital General Universitario Santa Lucía (Cartagena) Onofre Juan Martínez, Francisco Jesús Vera, Lorena Martínez, Josefina García, Begoña Alcaraz, Amaya Jimeno. Complejo Hospitalario Universitario a Coruña (Chuac) (A Coruña) Angeles Castro Iglesias, Berta Pernas Souto, Alvaro Mena de Cea. Hospital Universitario Virgen de la Arrixaca (El Palmar) Carlos Galera, Helena Albendin, Aurora Pérez, Asunción Iborra, Antonio Moreno, Maria Angustias Merlos, Asunción Vidal, Marisa Meca Hospital Universitario Infanta Sofia (San Sebastian de los Reyes) Inés Suárez-García, Eduardo Malmierca, Patricia González-Ruano, Dolores Martín Rodrigo, Mª Pilar Ruiz Seco. Hospital Universitario Príncipe de Asturias (Alcalá de Henares) José Sanz Moreno, Alberto Arranz Caso, Cristina Hernández Gutiérrez, María Novella Mena. Hospital Clínico Universitario de Valencia (València) María José Galindo Puerto, Ramón Fernando Vilalta, Ana Ferrer Ribera. Hospital Reina Sofía (Córdoba) Antonio Rivero Román, Antonio Rivero Juárez, Pedro López López, Isabel Machuca Sánchez, Mario Frias Casas, Angela Camacho Espejo Hospital Universitario Severo Ochoa (Leganés) Miguel Cervero Jiménez, Rafael Torres Perea Nuestra Señora de Valme (Sevilla) Juan A Pineda, Pilar Rincón Mayo, Juan Macias Sanchez, Nicolas Merchante Gutierrez, Luis Miguel Real, Anais Corma Gomez, Marta Fernandez Fuertes, Alejandro Gonzalez-Serna