Identification and characterisation of transient receptor potential melastatin 2 and CD38 channels on natural killer cells using the novel application of flow cytometry

Background Natural Killer (NK) cells are effector lymphocytes of the innate immune system and are subclassed into CD56BrightCD16Dim/− and CD56DimCD16+ NK cells. Intracellular calcium (Ca2+) is fundamental to regulate a number of intracellular signalling pathways and functions in NK cells, which are essential in mediating their natural cytotoxic function. Transient receptor potential melastatin 2 (TRPM2) is a Ca2+-permeable non-selective cation channel that possesses a critical role in calcium-dependent cell signalling to maintain cellular homeostasis. TRPM2 and CD38 protein surface expression has yet to be determined on NK cells using flow cytometry. Characterisation of TRPM2 has been previously identified by in vivo models, primarily using methods such as genetic remodification, immunohistochemistry and whole cell electrophysiology. The aim of this study was to develop an in vitro methodology to characterise TRPM2 and CD38 surface expression on NK cell subsets using an antibody that has not been previously applied using flow cytometry. Results At 2 h/1 h, TRPM2 (Fig. 2 A, B, p < 0.05) and TRPM2/CD38 (Fig. 3A, B, p < 0.05) surface expression significantly increased between 1:300 and 1:50 at 2 h/1 h. TRPM2/CD38 surface expression furthermore increased between 1:100 and 1:50 at 2 h/1 h (Fig. 3A, p < 0.05). Interestingly, TRPM2/CD38 surface expression significantly decreased from 1:50 to 1:5 on CD56BrightCD16Dim/− NK cells. These consistent findings highlight that 1:50 is the optimal antibody dilution and threshold to measure TRPM2 and TRPM2/CD38 surface expression on NK subsets. 2 h/1 h was determined as the optimal incubation period to ensure a sufficient timeframe for maximal antibody binding and surface expression. Conclusion For the first time, we describe an in vitro methodology to characterise TRPM2 and CD38 surface expression on NK cells in healthy participants. Finally, using an antibody that has not been previously applied in flow cytometry, we determined an antibody concentration and incubation time that is robust, rapid and sensitive for the application of flow cytometry. Electronic supplementary material The online version of this article (10.1186/s12865-019-0293-0) contains supplementary material, which is available to authorized users.


Background
Natural Killer (NK) cells are effector lymphocytes of the innate immune system found in peripheral blood, bone marrow, spleen, and lymph nodes. In peripheral blood, NK cells represent 15% of lymphocytes and are phenotypically distinguished by the surface expression of CD56 (neural cell-adhesion molecule) and CD16 (FcγIII receptor, the low affinity receptor of IgG) receptors. Thereby, NK cells are subclassed into CD56 Bright CD16 Dim/− and CD56 Dim CD16 + NK cells which represent respectively 10 and 90% of NK cells in peripheral blood [1]. NK cells have diverse biological functions, which include recognizing and killing virally infected or transformed cells. The former NK population is primarily involved in immunosurveillance and cytokine production, whereas the latter are cytotoxic and kill infected, tumour or 'missing self' cells [2]. Intracellular calcium (Ca 2+ ) mobilisation is required to regulate a number of intracellular signalling pathways in NK cells, such as the antibody dependent cellular cytotoxicity (ADCC) or mitogen-activated protein kinase pathway, which are essential for the development of immune synapse formation, cytokine production and cytotoxic activity [1]. Intracellular Ca 2+ is also required for the target cell adhesion, granule polarization and degranulation, all of which are necessary for NK cells to mediate their natural cytotoxicity [1,3].
Transient receptor potential melastatin 2 (TRPM2) is a Ca 2+ -permeable nonselective cation channel that is characterised with a unique C-terminal ADP-ribose (ADPR) pyrophosphate domain [4]. TRPM2 is synergistically activated by intracellular ADPR and Ca 2+ within the plasma membrane and/or lysosomal compartments. Binding of ADPR to TRPM2 opens the channel and allows the permeation of sodium (Na 2+ ), potassium (K + ) and Ca 2+ into the cell and hydrolysis of ADPR to ribose 5-phosphate and adenosine monophosphate (AMP) [5]. Previous investigations have shown that TRPM2 mediates a novel anti-tumour mechanism in NK cells in synergy with CD38, a multifunctional ectoenzyme using Nicotinamide adenine dinucleotide (NAD + ) as a substrate to catalyse the production of ADPR, cyclic ADPR (cADPR) and Nicotinic acid adenine dinucleotide phosphate (NAADP) [6]. Rah et al. (2016) demonstrated that CD38 facilitates the production of ADPR, which in turn mobilizes intracellular Ca 2+ and can activate TRPM2 resulting in cytolytic degranulation and antitumor activity of NK cells [6].
TRPM3 surface expression on CD56 Bright CD16 Dim/− and CD56 Dim CD16 + NK cells has been identified on healthy participants by flow cytometry [23,24]. Flow cytometry has been the preferred technology for determining and quantifying homogenous cell subsets [25] due to its single-cell levelled analysis for multiple characteristics, such as cellular features, organelles, and structural components [25]. This sensitive and specific feature enables prompt and accurate quantification, analytical precision, superior throughput, and reproducibility [26], all of which are advantageous for unique and rare cell populations, such as NK cells. Current flow cytometer technologies can detect up to eighteen colours in one flow assay. Thus, the scientific prospects not only lie in biomedical research, but also for clinical applications of diagnostic value [25].
Currently, there are no in vitro models that have characterised endogenous TRPM2 and CD38 surface expression on human NK cells. Thus, the aim of this present study was to develop a methodology to characterise TRPM2 and CD38 surface expression on human NK cells using flow cytometry. This investigation may facilitate a better understanding of the role of TRPM2 and CD38 in disease pathology involving immune cells such as NK cells.

Results
Immunophenotype of TRPM2 and CD38 receptors on NK cell subsets by flow cytometry CD3 − /CD56 + NK cells were sorted into CD56 Dim CD16 /+ and CD56 Bright CD16 Dim/− NK cell subsets using CD56 (Pe-Cy7) and CD16 (BV650). Five antibody controls were performed to determine an individualised positive TRPM2 and TRPM2/CD38 gate for each participant. Antibody controls included an unstained tube (unlabelled NK cells); secondary tube (conjugated secondary antibody FITC); and a FMO tube (CD3, CD56, CD16 and CD38). (b) Normal rabbit serum was used at comparable dilutions as the primary TRPM2 antibody to measure TRPM2 and TRPM2/CD38 surface expression on NK subsets. (c) Normalised TRPM2 and TRPM2/CD38 surface expression was calculated by compensating the percentage of fluorescence spill over into the B525_50 (TRPM2) and V525_50 (CD38) detectors from the TRPM2 antibody stained tube on both NK subsets.

Discussion
This investigation reports, for the first time, the identification of TRPM2 and CD38 surface expression on human NK cell subsets in healthy participants. This paper is also the first to develop a methodology that quantifies TRPM2 and CD38 surface expression with an antibody that has not been previously applied using flow cytometry. This novel method may have significant implications for analysing TRPM2 and CD38 surface expression in vitro and may facilitate a better understanding of the role of TRPM2 and CD38 in disease pathology involving immune cells such as NK cells.
In order to characterise TRPM2 surface expression, an extracellular TRPM2 antibody was preferred to prevent non-specific binding. The predominant clonality available on the market is polyclonal intracellular TRPM2 antibodies. Intracellular TRPM2 ion channels were not investigated as cell fixation and permeabilisation provides access to intracellular antigens. As TRPM2 is also localised on intracellular compartments, such as the endoplasmic reticulum and lysosome, cell permeabilisation can enable non-specific binding and activation of these intracellular TRPM2 channels, which potentially can mediate a number of downstream signalling pathways, such as Ca 2+ influx (15). Thus, a rabbit IgG polyclonal extracellular TRPM2 antibody (Thermo Fisher Scientific, USA, OST00112W) was chosen due to its ready availability and extracellular binding, specifically to the third extracellular loop of the human TRPM2 receptor.
One limitation of the primary TRPM2 antibody was the absence of a determined antibody concentration. According to Thermo Fisher Scientific, "antibody concentrations in ascites fluid, culture supernatant and serum are not determined due to various proteins in serum which makes it impossible to acquire an accurate concentration of a specific antibody". Due to the absence of a determined antibody concentration, a TRPM2 isotype control could not be performed. However, as the primary TRPM2 antibody contains rabbit serum, normal rabbit serum (Thermo Fisher Scientific, USA, 01-6101) was used at comparable dilutions as the primary TRPM2 antibody. This negative control was used to distinguish any non-specific binding, as well as determine an individual positive TRPM2 and TRPM2/CD38 gate for each participant (Fig. 4b, c). Additionally, an unstained tube; a secondary tube; and a FMO control (Fig. 4a) were performed for each participant to compensate any potential fluorescence spill over (Additional file 1: Figure S1, Additional file 2: Figure S2, Additional file 3: Figure S3, Additional file 14: Figure S14, Additional file 15: Figure S15, Additional file 16: Figure S16).
Interestingly, a normal distribution curve was observed on both NK subsets for TRPM2 and dual expression with CD38 at 2 h/1 h. Comparatively, receptor surface expression remained relatively constant at 1 h/30 min on both NK subsets. This observation supported the significant decrease in TRPM2/CD38 surface expression from 1:50 to 1:5 on CD56 Bright CD16 Dim/− NK cells (Fig. 3a, p < 0.05). Importantly, this result demonstrates an inverse relationship between antibody concentration and receptor expression and highlights 1:50 as the threshold antibody dilution for TRPM2 (Fig. 1).
In contrast there was a significant increase in TRPM2 surface expression with 1:300 at 1 h/30 min (Fig. 2b, p < 0.05), but not with dual expression with CD38 (Fig. 2b, p < 0.05). This sole result revealed a difference in receptor surface expression between incubation periods. As CD56 Bright CD16 Dim/− NK cells are less abundant than the CD56 Dim CD16 + subset, the percentage of receptor expression increases with limited cells detected. Moreover, the percentage of receptor expression increases for rarer channels. Given TRP ion channels are relatively scarce, particularly on lymphocytes, a longer incubation time is required to ensure optimal binding and subsequent surface expression. Moreover, the consistent pattern with the 1:50 TRPM2 dilution on both NK subsets justified 2 h/1 h as the optimal incubation period to ensure a sufficient timeframe for maximal antibody binding and surface expression.
Despite tested applications for western blot and immunohistochemistry assays, no additional studies have published the use of the OST00112W TRPM2 antibody. Future directions include the examination of TRPM2 and CD38 channels on additional lymphocytes, as well as investigate the manufacturer's tested applications to further assess antibody specificity.

Conclusion
This novel methodology is the first to identify and characterise TRPM2 and TRPM2/CD38 surface expression on human NK cells in healthy participants. This pilot investigation is also the first to use a TRPM2 antibody that has not been previously applied in flow cytometry, as well as determine the optimal primary TRPM2 antibody dilution and incubation time. This method provides an in vitro alternative using flow cytometry to characterise TRPM2 in a rapid, robust and sensitive fashion. This pilot investigation provides insight for possible improvement in antibody design to facilitate a more accurate assessment of TRPM2 and CD38 surface expression.

Study participants
From 150 screened Australian participants, ten healthy participants were selected for this pilot investigation. Two participants were excluded due to outlier values during data analysis. Participants were sourced from the National Centre of Neuroimmunology and Emerging Full blood count parameters were measured for each healthy participant. All participant results were within the specified reference ranges for each parameter. There were no significant differences between healthy participants for these reporting parameters (

Peripheral blood mononuclear cell isolation and natural killer cell isolation
Peripheral blood mononuclear cells (PBMCs) were isolated from whole blood by centrifugation over a density gradient medium (Ficoll-Paque Premium; GE Healthcare, Uppsala, Sweden) to separate granulocytes as previously described [27,28]. PBMCs were stained with trypan blue stain (Invitrogen, Carlsbad, CA) to determine total cell count and cell viability and adjusted to a final concentration of 5 × 10 7 cells/ml. NK cells were isolated from PBMCs using an EasySep Negative Human NK Cell Isolation Kit (Stemcell Technologies, Vancouver, BC, Canada) as previously described [27,28].

Statistical analysis
Pilot data from this investigation were analysed using SPSS version 24 (IBM Corp, Version 24, Armonk, NY, USA) and GraphPad Prism, version 7 (GraphPad Software Inc., Version 7, La Jolla, CA, USA). Shapiro-Wilk normality tests were conducted to determine the distribution of data, in addition to skewness and kurtosis tests to determine data normality. The independent Mann-Whitney U test was performed to determine the statistical significance between groups in TRPM2 parameters on NK cells. Conversely, the Kruskal Wallis H test was used to determine