The impact of experimental manipulation on neutrophil immunophenotyping

The interest of developing functional and other experimental procedures on polymorphonuclear neutrophils (PMN) for the study of inflammatory disorders and primary immunodeficiencies is constantly increased. However, these experimental manipulations may induce significant changes on neutrophils’ immunophenotyping and function. For this purpose, we investigated the effect of different experimental procedures (such as resting state, isolation and incubation with and without inflammatory stimuli) on the immunophenotype of PMN of healthy individuals, evaluating the expression pattern of specific granulocytic markers, important for chemotaxis, adhesion, opsonization, migration and phagocytosis. Our results indicate that the effect of experimental manipulation on PMN should always be taken into consideration in order to avoid misinterpreting the experimental findings. *Correspondence to: Matthaios Speletas, Department of Immunology and Histocompatibility, Faculty of Medicine, School of Health Sciences, University of Thessaly, Larissa, Greece, Tel: +302410685687; E-mail: maspel@med.uth.gr


Introduction
Polymorphonuclear neutrophils (PMN) are major effector cells in innate host defense against invading microorganisms. They migrate towards the site of infection and inflammation in order to confront the pathogens. PMN recognize pathogens through various receptors, such as scavenger, complement and immunoglobulin receptors, eliminating them either through mechanisms that involve degranulation, phagocytosis and oxidative killing into the phagosomes, or through the formation of neutrophil extracellular traps (NETs) [1,2].
During recent years there is an increasing interest in the development of procedures, where human PMN are being studied in the context of inflammatory disorders and primary immunodeficiencies [3,4]. In these procedures, the expression of cell surface molecules, such as adhesion molecules, immunoglobulin and complement receptors, is investigated along with functional responses to inflammatory stimuli [3][4][5]. In order to safely draw conclusions when investigating the role of PMN in health and disease, the in vitro data should reflect the in vivo behavior. A few previous studies have shown that specific experimental manipulation of PMN might induce changes in the expression pattern of several surface antigens [6][7][8].
Therefore, the aim of our study was to investigate more precisely the impact of different experimental procedures, such as isolation and incubation with and without inflammatory stimuli, on the immunophenotype of PMN of healthy individuals, evaluating the expression pattern of specific granulocytic markers, important for chemotaxis, adhesion, opsonization, migration and phagocytosis.

Subjects
Eight milliliters of peripheral blood were collected using heparinized syringe from 14 healthy individuals (male/female: 4/10, mean age: 41.5 years, range: 21-65) with no history of autoimmunity, chronic inflammatory disorder, or malignancy. All individuals were receiving no medication during blood collection, except from two subjects under common anti-hypertensive treatment. The absence of infection/inflammation was also verified after clinical examination and the measurement of C-reactive protein (CRP) serum levels, using commercially available immunonephelometric assays (Immulite-2000, Siemens Medical Solutions, Llanberis, Gwynedd, UK), according to the manufacturer's instructions. All subjects provided written informed consent. The study was conducted in accordance with the principles of the Helsinki Declaration and was approved by the Institutional Review Board of the University Hospital of Larissa, Greece.

Flow cytometry analysis
Immunophenotyping was performed by flow cytometry on Coulter FC-500 instrument (Epics XL-MCL, 4 color analysis, Beckman-Coulter/ BC, Hialeah, FL, USA) using a multistaining protocol and commercially available reagents. Mouse anti-human immunoglobulin G (IgG) monoclonal antibodies were used for the detection of CD10 (clone: ALB1), CD11b (clone: Bear1), CD14 (clone: RMO52), CD16 (clone: 3G8), CD18 (clone: 7E4), CD64 (clone: 22), and CD66b (clone: 80H3). All the above antibodies were purchased by BC and were conjugated with the appropriate fluorochrome (fluorescein isothiocyanate, FITC; phycoerythrin, PE; PE-cyanine5 PE-Cy5). The percentage of fluorescent cells and the mean fluorescence intensity (MFI) were determined in each case corrected for background fluorescence, using FITC, PE, and PE-Cy5-labelled control antibodies. All lysis and staining procedures were performed at room temperature. The data analyzed were the percentile (%) expression of each marker and the intensity of expression determined by the equation: (%) expression × MFI /100. In particular, we evaluated the expression pattern of the aforementioned markers on PMN in four different conditions: (1) resting (basal) state, where PMN had not undergone any treatment, (2) normal density PMN, where neutrophil phenotype reflects changes due to the isolation procedure, (3) culture-control PMN, where neutrophil phenotype reflects changes under culture conditions without any trigger, and (4) culture-lipopolysaccharide (LPS) PMN, where neutrophil phenotype reflects changes under LPS trigger. Experimental manipulation of specimens was always begun into two hours after blood collection.
Therefore, the heparinized whole-blood samples were separated in three parts. The first part (approximately 2 mL) was immediately used for immunophenotyping of PMN at resting (basal) state. The second part (4 mL) was used for PMN isolation and the third part (approx. 0.5-1 mL) was used in culture assays, as described below: In resting (basal) state, the erythrocytes were initially removed using NH 4 Cl/KHCO 3 lysis solution, and white blood cells (WBC) were washed twice with 1× PBS. After resuspension, WBC were stained with the specific monoclonal antibodies (CD10, CD11b, CD14, CD16, CD18, CD64, and CD66b) for 15 minutes. After being washed with PBS, WBCs were re-suspended in PBS and a flow cytometric evaluation in basal state was performed. Approximately 10,000 PMN were acquired per sample and cells were initially gated according to forward and side scatter characteristics.
PMN isolation was performed using the double ficoll-gradient density centrifugation. More specifically 3 mL of Histopaque 1119 (Sigma-Aldrich, St. Louis, Missouri, USA) were layered on top of 3 mL Histopaque 1077 g/ml (Sigma-Aldrich). Four mL of blood specimen were diluted with 2 mL of 1 × PBS (in a total volume of 6 mL) and were layered on the top of the ficoll layers. After a centrifugation step at room temperature for 30 minutes at 2,000rpm, the PMN were isolated from the interface between the two ficoll layers and washed with 1× PBS for 10 minutes at 1.600 rpm. Finally, cells were re-suspended in 1 × PBS for subsequent immunophenotyping with the specific monoclonal antibodies, as described above. In case that erythrocytes were present, a lysis step with NH 4 Cl/KHCO 3 , before staining procedure, was performed.
PMN culture assays: In a 24-well tissue culture plate, 2 × 10 5 WBC (containing at least 1-1.5 × 10 5 PMN) were seeded in a total volume of 500 1L Iscoves Basal Medium (Biochrom, Berlin, Germany), containing 6% normal human serum (NHS) with or without the presence of LPS (100 ng/mL, E.Coli serotype 026:B6, Sigma-Aldrich). After 10 minutes of pre-incubation in order to let the cells rest in a humidified atmosphere (37 0 C, 8% CO 2 ), WBC were treated with LPS for 1h. Afterwards, cells were collected, washed, lysed in order to remove erythrocytes, stained with the specific antibodies and the proper isotype controls, and evaluated by flow cytometry, as described above.

Statistical analysis
Wilcoxon non-parametric match-paired analysis was performed for the comparison of different experimental conditions. Presented data refer to Intensity of expression defined by the formula: Intensity of expression = % of expression × MFI. Statistical analyses and graphs were made on GraphPad-6 Prism software (CA, USA). For all analyses, alpha was set at 0.05 (2-sided).

Expression of immunoglobulin receptors (CD16, CD64)
The Fc receptor with low affinity for IgG (FCGR3 or CD16; OMIM *610665) is expressed constitutively on mature PMN, and its crosslinking leads to degranulation and activation of the respiratory burst [9]. As presented in Figure 1A, CD16 was significantly upregulated after isolation, as well as after incubation of whole blood without any trigger (p = 0.001 and p = 0.002, respectively). However, when PMN were treated with LPS, they exhibited a lower intensity of expression comparing to both untreated incubated PMN (p = 0.001) and PMN in resting (basal) state (p = 0.005). CD64 is a high affinity IgG receptor (OMIM *601502) that is expressed only in activated granulocytes and is a useful marker in infectious conditions [10,11], and sepsis [12]. We did not find any remarkable CD64 expression in all experimental conditions, apart from two (2) individuals exhibiting over 20% expression in basal state (20.3% and 42.4%, respectively), with its intensity of expression to being decreased during both incubation conditions (with and without LPS, Figure 1B).

Expression of maturation markers (CD10)
CD10 (OMIM *120520) is a specific marker that is indicative of maturation state of PMN [13]. We observed that the majority of PMN express CD10 in all cases, and the intensity of its expression is significantly increased after PMN isolation (p < 0.001) ( Figure 1C). On the other hand, no significant differences were observed considering the other experimental conditions ( Figure 1C).
We observed that the intensity of CD18 expression was higher after isolation (p < 0.001) and incubation of whole blood neutrophils, with (p < 0.001) or without LPS (p = 0.002), comparing to resting cells. The difference of expression was not significant when the comparison was made between neutrophils under culture conditions (p = 0.090) ( Figure  1D). Furthermore, a remarkable increase of the intensity of CD11b expression was also observed in all experimental conditions ( Figure  1E). CD66b (OMIM: 615747) is a specific granulocytic marker that is upregulated after LPS activation, reflecting a degranulation phenomenon [19]. It belongs to Carcino-Embryonic Antigen Cell Adhesion Molecule (CEACAM) family and its stimulation induces oxidative burst and increases neutrophil adhesion via a CD18-dependent pathway [20]. We observed that CD66b expression was also significantly increased in all experimental conditions, compared to resting cells ( Figure 1F). In addition, a remarkable increase of CD66b expression was found in LPStreated PMN compared to untreated ones in culture conditions (p < 0.001), although a notable variation of CD66b expression was observed after LPS exposure (as presented in detail, in Figure 1F).

Discussion and conclusion
PMN are cells of the innate immune system comprising the first line host defense against invading microorganisms. Their life begins in the bone marrow, where progenitor cells of myeloid origin differentiate into mature PMN, entering the circulation as terminally differentiated cells. They die by apoptotic cell death, having a short lifespan of 6-12 hours in the periphery, unless they enter the tissue phase following an inflammatory stimulus, where they display a prolonged lifespan. After exposure to inflammatory stimuli, PMN can produce cytokines and chemokines, regulating inflammation [21,22]. There is increased evidence that changes in the expression of several molecules can be used as a quantifiable index of PMN activation, after treatment with different stimuli, or can be interpreted as a "primed" status of the cell [21]. PMN activation and degranulation result in the exposure of their contents in both the plasma membrane and the extracellular milieu; however, the level of expression of different PMN markers usually reaches a plateau and it is not altered by subsequent stimuli [23].
In our study, we investigated simultaneously the effect of common experimental conditions on healthy individuals' PMN. Our findings indicate that the expression of PMN surface molecules are affected by their experimental manipulation. In particular, we observed that CD10 and CD16 are more prone to unspecific priming due to isolation procedure, while the adhesion molecules CD11b and CD66b are more susceptible to culture conditions, even in the absence of any trigger. On the other hand, CD66b expression exhibits a great variation of its intensity in response to LPS, while CD64 is expressed in low levels on PMN and it is not affected by experimental manipulation, even by LPS exposure. These findings should be taken into consideration when studying PMN defects, since abnormalities of PMN compartment have been implicated in the pathogenesis and/or phenotype of several diseases, including sepsis [24] and immunodeficiencies [25]. PMN defects can affect the response to pathogens through several mechanisms, like phagocytosis, chemotaxis and transmigration [26], or disturb the dialectic communication of neutrophils with the other counterparts of the immune system [27]. There are several reports in the literature analyzing the effect of several experimental conditions on PMN functions, including isolation [28,29], the usage of different anticoagulants [8], the method of fixation or the effect of temperature alterations [21] on PMN immunophenotyping, or functional responses to Ca 2+ homeostasis, ROS production, phagocytosis and/or bacterial ingestion [28]. However, the data from different studies cannot be compared and most importantly they lead to controversial and opposite findings. In our study, we focused on PMN immunophenotyping, as it is a fast procedure to detect possible defects, and for this purpose we analyzed common surface markers indicative of PMN functional capacity. According to the data presented herein, we recommend that a minimal experimental manipulation of PMN should be chosen when investigate their functional capacity, to both avoid the unspecific priming and to draw safer conclusions when interpreting in vivo phenomena.