Comparing methods for ex vivo characterization of human monocyte phenotypes and in vitro responses

Immunobiology 220 (2015) 1305–1310

Contents lists available at ScienceDirect

Immunobiology journal homepage: www.elsevier.com/locate/imbio

Short communication

Comparing methods for ex vivo characterization of human monocyte phenotypes and in vitro responses Lisa Johnston a , Scott A. Harding a,b , Anne Camille La Flamme a,c,∗ a b c

School of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington, New Zealand Wellington Cardiovascular Research Group, Wellington Hospital, Private Bag 7902, Wellington, New Zealand Malaghan Institute of Medical Research, P.O. Box 7060, Wellington, New Zealand

a r t i c l e

i n f o

Article history: Received 27 May 2015 Received in revised form 20 July 2015 Accepted 23 July 2015 Available online 29 July 2015 Keywords: Monocyte Cell purification Ex vivo activation In vitro stimulation Monocyte activation phenotype

a b s t r a c t Monocytes are key innate effector cells and their phenotype and function may be a useful biomarker of disease state or therapeutic response. However, for such an assay to be clinically feasible it needs to be simple and reproducible, which this study aimed to address. Peripheral blood mononuclear cells (PBMC)2 isolated from whole blood using Histopaque-1077 or cell preparation tubes (CPT) showed no difference in the ex vivo monocyte activation marker expression or in vitro responses; however, a delayed isolation using CPT significantly altered ex vivo and in vitro phenotypes and responses. Furthermore, purification of monocytes using CD14+ microbeads resulted in a loss of CD14low CD16+ monocytes compared to PBMC samples. Thus, the use of CPT reduced complexity and time compared to Histopaque, and PBMC isolation allowed the analysis of all 3 major monocyte subsets. Finally, because the delayed isolation of PBMC from CPT significantly altered monocytes, time delays should be standardized. © 2015 Elsevier GmbH. All rights reserved.

1. Introduction Monocytes are key innate effector cells that are central to many immune responses and play an essential role in the effective control and clearance of pathogens from the body (Shi and Pamer, 2011). However, monocytes may also contribute to inflammatory conditions and have been implicated in the pathogenesis of diseases such as Crohn’s disease, MS and atherosclerosis (Woollard and Geissmann, 2010; Zhou et al., 2009; Bar-Or et al., 2003). Three subpopulations of monocytes have been described based upon CD14 and CD16 expression: classical (CD14++ CD16− ), intermediate (CD14++ CD16+ ) and non-classical or “patrolling” (CD14− CD16+ ) (Wong et al., 2012). Each of these subpopulation has specialized functions, phenotypic markers, and distinct disease associations (Wong et al., 2012). Previous studies have revealed the remarkable multipotency of monocytes in response to altering inflammatory environments (Avraham-Davidi et al., 2013; Weber et al., 2007;

Jakubzick et al., 2013), and because their phenotype and function may reflect the individual’s immune environment, this information may be useful as a marker of disease state (e.g. Crohn’s disease, hypercholesterolemia, sepsis, multiple sclerosis) (Sawada-Hase et al., 2000; Garlichs et al., 2001; Danikas et al., 2008; Chuluundorj et al., 2014). The ability to assess monocyte responses in patients is a potentially powerful biomarker to assess disease progression or treatment response in a range of diseases. However, many studies have relied on isolation of PBMC by Histopaque gradient followed by CD14+ microbead monocyte purification (Chuluundorj et al., 2014; Kim et al., 2004; Giulietti et al., 2007). This method to isolate monocytes is time consuming, technical and expensive. Thus, for such assays to be clinically feasible, they need to be simple, reproducible and practical for the clinical environment. To address this need, this study aimed to develop a simplified, reproducible method to assess peripheral monocyte responses in humans.

2. Methods Abbreviations: CPT, cell preparation tubes; HCA, hierarchical cluster analysis; LPS, lipopolysaccharide; MFI, mean fluorescent intensity; MS, multiple sclerosis; PBMC, peripheral blood mononuclear cells. ∗ Corresponding author at: School of Biological Sciences, Victoria University of Wellington, P.O. Box 600, Wellington 6140, New Zealand. E-mail address: anne.lafl[email protected] (A.C. La Flamme). http://dx.doi.org/10.1016/j.imbio.2015.07.014 0171-2985/© 2015 Elsevier GmbH. All rights reserved.

2.1. Subjects Healthy volunteers were aged between 22 and 51 years (n = 5; 60% female). For the ex vivo and cluster analyses, four of these

1306

L. Johnston et al. / Immunobiology 220 (2015) 1305–1310

Fig. 1. Gating strategy for monocyte activation markers (A–I) and monocyte populations (J–N) in PBMC and isolated monocyte samples. (A) Live, singlet cells were gated using forward scatter properties (FSC-H and FSC-A). In both PBMC (B) and isolated monocyte samples (C), CD14low−++ and CD11b+ cells were considered monocytes. Once gated, the monocyte activation markers were assessed. The light grey histograms represent stained samples, and the dark grey histograms are isotype controls. Shown are representative flow plots (D–I) from monocytes, and the change in MFI (MFI) between the specific antibody and isotype control is shown in the upper right hand corner of each plot. CD14+ CD16− (J), CD14+ CD16+ (K), and CD14int CD16+ (L) monocytes were gated as shown (M) from isolated monocytes or PBMC isolated on day 1 (J–N) or 2 (N). *p < 0.05 by two-way ANOVA (J & L) or one-way ANOVA with Neuman–Keuls Multiple Comparison Test (N).

healthy subjects are shown (age 22–51; 50% female). The study was approved by the Victoria University of Wellington Human Ethics Committee (RM20738).

2.2. Total monocyte and subset isolation and culture PBMC were isolated by either Histopaque-1077 (Sigma, St. Louis, MO, USA) gradient or by CPT (BD Biosciences, Franklin Lakes, NJ, USA). For Histopaque-1077 isolation, 15 ml heparinized whole blood was mixed with 25 ml phosphate buffered saline (PBS) without calcium or magnesium and layered onto 15 ml Histopaque1077. The gradient was centrifuged at 400 × g for 30 min at room temperature, and the PBMC collected from the interface. For CPT isolation, 15 ml of blood was collected into two CPT, inverted and spun at 1600 × g for 20 min at room temperature, and the PBMC collected from the interface. For d2 samples, CPT were stored at room temperature before PBMC isolation. Once isolated, the PBMC were washed with PBS and either resuspended in complete medium containing DMEM, 10% FCS, 100 U/ml penicillin plus 100 ␮g/ml streptomycin, 10 mM Hepes, 2 mM l-glutamine, and 50 ␮M 2-mercaptoethanol (all from Life Technologies, Carlsbad, CA, USA) for PBMC culture or washed with MACS isolation buffer for monocyte isolation. PBMC for monocyte isolation were incubated with human CD14 microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) at a concentration of 20 ␮l/107 cells for 15 min at 4 ◦ C. The magnetic separation was performed using LS columns (Miltenyi Biotec), and the bound cells were then washed and resuspended in complete medium. The isolated PBMC (1 × 106 cells/well) and monocytes (1 × 105 cells/well) were cultured in U-bottomed 96-well plates (BD Biosciences). All cells were primed with IFN-␥ (20 U/ml; Peprotech, Rocky Hill, NJ, USA) for 16 h at 37 ◦ C in 5% CO2 . After priming, the cells were stimulated with or without LPS from Escherichia coli (200 ng/ml; Sigma) for 24 h at 37 ◦ C in 5% CO2 .

2.3. Flow cytometry Cells were stained with the optimized primary antibodies or isotype control antibodies as described previously (Chuluundorj et al., 2014). Antibodies used were anti-CD11b (ICRF44), CD14 (M5E2), CD86 (2331(FUN-1)), CD64 (10.1), CCR2 (48607) plus isotypes (BD Biosciences) and anti-HLA-DR (TU36), ILT3 (ZM4.1), CD40 (5C3) and isotypes (Biolegend, San Diego, CA USA). Samples were run on a FACS Canto II flow cytometer using Diva software (BD Biosciences), and analyzed using FlowJo 5 software (Tree Star, Ashland, OR, USA). See Fig. 1 for PBMC and monocyte gating strategies. 2.4. Cytokine assays IL-12p40 ELISA reagents (BD Biosciences) were used according to the manufacturer’s protocol. 2.5. Statistical analyses Statistical analyses were performed using GraphPad Prism 6 (GraphPad, La Jolla, CA, USA). Student’s paired t-test was used when comparing 2 groups and one-way ANOVA for 3 or more. A two-way ANOVA was used to test group versus culture effects. Hierarchical cluster analysis (HCA) was performed using HCE3.5 (Math Works Inc., USA). An average linkage method and row and column clustering directions were used. In order to measure similarities and distances, Pearson correlation coefficient was selected, whereas the mosaic mapping method was chosen to display the data. 3. Results and discussion 3.1. Isolation of PBMC by Histopaque or CPT does not affect ex vivo monocyte markers or in vitro responses. PBMC and monocytes were isolated using standard Histopaque1077 gradient separation or CPT cell preparation tubes. Monocyte

L. Johnston et al. / Immunobiology 220 (2015) 1305–1310

1307

Fig. 2. No difference in ex vivo or in vitro expression of monocyte activation markers and cytokine production between CPT and Histopaque isolation. (A–F) Ex vivo expression of monocyte activation markers were measured by flow cytometry on PBMC isolated by Histopaque (H) or CPT (C) methods and expressed in MFI. Each data line represents an individual (n = 4). (G–I) In vitro monocyte expression of CD40 (G) and CD64 (H) as measured by flow cytometry and IL-12 production by ELISA (I) were measured following stimulation with LPS (L) or medium alone (M) of Histopaque and CPT-isolated PBMC. Shown are the individual paired values (n = 4 − 5); each symbol is a different individual. *p < 0.05 medium vs LPS by two-way repeated measure ANOVA. (J) HCA dendrogram and heat map of ex vivo monocyte activation marker expression on PBMC isolated by Histopaque (H) or CPT (C) methods. Green shows low expression, black – medium expression and red – high expression. The lowest expression is shown as bright green and the highest as bright red.

1308

L. Johnston et al. / Immunobiology 220 (2015) 1305–1310

Fig. 3. Ex vivo and in vitro expression of monocyte activation markers and cytokine production was similar between PBMC and purified monocytes. (A–F) Ex vivo expression of monocyte activation markers were measured by flow cytometry on PBMC and purified CD14+ monocytes (CD14) isolated by CPT and expressed in MFI. Each data line represents an individual (n = 4). (G–I) In vitro monocyte expression of CD40 (G) CD64 (H) and IL-12 (I) were measured following stimulation with LPS (L) or medium alone (M) of isolated PBMC or purified CD14+ monocytes from CPT. Shown are the individual paired values (n = 4 − 5); each symbol is a different individual. *p < 0.05 and **p < 0.01 medium compared to LPS by two-way ANOVA. For H, **p < 0.01 PBMC compared to CD14+ by two-way repeated measure ANOVA. (J) HCA dendrogram and heat map of ex vivo monocyte activation marker expression from PBMC and CD14 isolated by Histopaque (H) or CPT (C) on day 1 and PBMC isolated by CPT on Day 2. Green shows low expression, black – medium expression and red – high expression. The lowest expression is shown as bright green and the highest as bright red.

activation markers and cytokine production were compared between the two methods to ascertain whether CPT, the quicker and more convenient method, could be used instead of gradient separation. We first assessed wither isolation of PBMC by either

method altered the proportion of the three major monocyte subpopulations. While no difference in the proportion of CD14+ CD16− , CD14+ CD16+ , or CD14low CD16+ monocytes in PBMC isolated by either method was found, there was a significant reduction in

L. Johnston et al. / Immunobiology 220 (2015) 1305–1310

CD14low CD16+ and concurrent increase in CD14+ CD16− monocytes when monocytes were purified by CD14+ microbead isolation from either PBMC population (Fig. 1J–M). Ex vivo monocyte activation markers were measured immediately following PBMC isolation, and no difference in the level of CD40, HLA-DR, CD86, CD64, CCR2 and ILT3 was found between the gated monocyte populations in PBMC isolated by either techniques (Fig. 2A–F). Additionally, no difference was detected in the ex vivo expression of these markers on purified CD14+ monocytes from PBMC isolated by Histopaque or CPT (Supplementary Fig. 1A–F). To determine if the PBMC isolation method affected functional responses, in vitro activation of monocytes in the PBMC was assessed following LPS stimulation (Fig. 2G–I). As expected, monocyte expression of CD40 and production of IL-12 significantly increased while CD64 significantly decreased after LPS stimulation compared to medium alone (Chuluundorj et al., 2014) (Fig. 2G–I). HLA-DR and CCR2 expression were not altered by LPS stimulation (data not shown). Importantly, no significant differences in any of the in vitro responses to LPS were detected when comparing Histopaque and CPT-isolated PBMC (Fig. 2G–I) or purified CD14+ monocyte cultures (Supplementary Fig. 1G–I). Whilst the individual analysis of cytokines and surface markers is important in understanding the molecular changes occurring in monocytes, this analysis does not consider the pattern of expression in an individual subject. To compare the pattern of monocyte activation in PBMC isolated by different techniques, an HCA of the ex vivo monocyte phenotype was performed (Fig. 2J). This analysis indicated that CPT and Histopaque monocyte profiles of each individual were more similar than the patterns found by the method of isolation suggesting that either method of PBMC isolation resulted in a similar ex vivo phenotype. Taken together, these results demonstrate that the isolation technique has no significant impact on the measurement of monocyte activation either ex vivo or in vitro. Isolating PBMC by Histopaque gradient can be time consuming and sensitive to the user’s experience with gradient preparation while the extended handling of the sample to prepare and construct the gradient increases the risk of contamination. Alternatively, CPT allow for a direct venepuncture to centrifuge isolation method reducing the potential for contamination. Finally, whilst CPT are more expensive compared to standard vacutainers, with the reduced sample preparation and handling time and consumable use, CPT isolation becomes cost-effective compared to the Histopaque method.

3.2. Delayed isolation of PBMC from CPT affected monocyte activation marker expression and cytokine production With the nature of clinical samples, time delay from venepuncture to testing is common. This issue may be exacerbated if assays need to be performed in a central laboratory with samples from multiple locations. To ascertain whether the measurement of monocyte activation from CPT-isolated PBMC was altered by time delay; two blood samples were taken from each individual and centrifuged, and PBMC were then isolated that day (d1) or the following day (d2) post-venepuncture. This delayed isolation of PBMC from CPT led to an increased proportion of CD14+ CD16+ with a concurrent loss of CD14+ CD16− monocytes (Fig. 1N) indicating that the subpopulations were modified by sustained storage. Analysis of ex vivo monocyte activation markers demonstrated no significant difference in the expression of CD40, HLA-DR, CD86, CD64 or ILT3 between PBMC isolated on d1 and d2 when analysed individually (Supplementry Fig. 2A–F). In contrast, monocyte expression of CCL2 in PBMC on d2 was consistently and significantly reduced compared to the same individual’s PBMC on d1 (Supplementry Fig. 2E). While LPS stimulation resulted in increased

1309

expression of CD40 and decrease expression of CD64 compared to media alone (Supplementry Fig. 2G and H), PBMC on d2 did not produce IL-12 in response to LPS stimulation (Supplementry Fig. 2I). While the majority of ex vivo and in vitro monocyte activation markers showed no significant difference between d1 and d2 isolations, these analyses were done on the individual parameters in isolation. In contrast, an HCA analysis of the overall ex vivo monocyte activation profile showed that the d2-pattern (“Day 2”) of all subjects did not cluster closely with their d1 monocyte activation (“PBMC C”) profile (Fig. 3J). Therefore, while assessing parameters on an individual basis indicated that only CCR2 was significantly modified by the delayed isolation, once the overall pattern of expression was assessed by HCA, it was clear that delaying the isolation by even one day significantly altered the expression profile of monocytes as well as the subset distribution. This finding emphasizes the importance of standardizing the isolation time when comparing overall monocyte activation phenotypes. 3.3. CD14+ monocyte isolation from PBMC is not required to assess monocyte activation Previous studies have reported reproducible monocyte activation results in purified CD14+ monocyte samples ex vivo and following stimulation with LPS in vitro (Chuluundorj et al., 2014; Kim et al., 2004). However, further isolation of monocytes from PBMC adds time and expense to the assay and can only demonstrate direct effects of treatments/culture conditions on monocytes. Therefore, to ascertain if reproducible monocyte activation profiles could be obtained from PBMC, monocyte activation markers and cytokine production was measured in isolated PBMC and compared to purified CD14+ monocytes from the same blood sample. Freshly isolated cells were analyzed by flow cytometry, and the only significant difference found was a modest but consistent increase in ILT3 expression on purified CD14+ monocytes (Fig. 3A–F). After stimulation with LPS, there was a significant increase in CD40 and IL-12 and concurrent decrease in CD64 to media alone in PBMC and purified monocyte cultures (Fig. 3G & I). Interestingly although there was no significant difference between PBMC and purified monocytes for CD40 expression and IL-12 production and CD64 expression in vitro was significantly lower in the purified monocytes compared to PBMC (Fig. 3H). Finally, the HCA analysis demonstrated that the ex vivo activation patterns clustered together by the individual, irrespective of isolation technique (CPT vs. Histopaque) and whether the activation markers were measured in PBMC or purified monocyte samples (Fig. 3J). 4. Conclusion Peripheral monocyte responses could be measured in humans using a simple reproducible assay. The use of a CPT over a Histopaque-1077 gradient allowed for a simpler and quicker method of isolation whilst not altering the measurement of the monocyte response. Previous studies have also compared the use of CPT with standard density gradients; to analyze IFN-␥ production from T cells of HIV patients (Ruitenberg et al., 2006) and PBMC RNA levels from melanoma patients (de Vries et al., 2000), and these studies found equivocal results between the two methods. Although monocytes have a relatively short half-life compared to circulating T cells (van Furth and Cohn, 1968) and are not as robust to isolation methods, comparable results were obtained from monocytes isolated by both techniques. Moreover, the monocyte activation states and responses could be measured in PBMC without the need for further isolation although there are advantages and disadvantages to this approach. For example, the stimulation

1310

L. Johnston et al. / Immunobiology 220 (2015) 1305–1310

of monocytes within PBMC must take into account the direct and indirect effects on monocytes, which may better represent what occurs in vivo. Additionally, purification of monocytes from PBMC may not equally target all 3 monocyte populations as we found with the CD14+ microbead-isolated monocytes. Finally, the delayed isolation of PBMC from CPT was the key variable tested that had a distinct effect on the overall pattern of monocyte responses and thus, where possible any time delays should be standardized. Conflict of interest The authors have no conflict of interest to declare. Funding This study was funded by the University Research Fund from Victoria University of Wellington (to ACL and SH). Acknowledgement This study was funded by the University Research Fund from Victoria University of Wellington (to ACL and SH). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.imbio.2015.07. 014 References Avraham-Davidi, I., Yona, S., Grunewald, M., Landsman, L., Cochain, C., Silvestre, J.S., et al., 2013. On-site education of VEGF-recruited monocytes improves their performance as angiogenic and arteriogenic accessory cells. J. Exp. Med. 210 (November 18 (12)), 2611–2625. Bar-Or, A., Nuttall, R.K., Duddy, M., Alter, A., Kim, H.J., Ifergan, I., et al., 2003. Analyses of all matrix metalloproteinase members in leukocytes emphasize monocytes as major inflammatory mediators in multiple sclerosis. Brain 126 (December (Pt 12)), 2738–2749. Chuluundorj, D., Harding, S.A., Abernethy, D., La Flamme, A.C., 2014 Jul. Expansion and preferential activation of the CD14(+)CD16(+) monocyte subset during multiple sclerosis. Immunol. Cell Biol. 92 (July (6)), 509–517.

Danikas, D.D., Karakantza, M., Theodorou, G.L., Sakellaropoulos, G.C., Gogos, C.A., 2008. Prognostic value of phagocytic activity of neutrophils and monocytes in sepsis. Correlation to CD64 and CD14 antigen expression. Clin. Exp. Immunol. 154 (October (1)), 87–97. Garlichs, C.D., John, S., Schmeisser, A., Eskafi, S., Stumpf, C., Karl, M., et al., 2001. Upregulation of CD40 and CD40 ligand (CD154) in patients with moderate hypercholesterolemia. Circulation 104 (November 13 (20)), 2395–2400. Giulietti, A., van Etten, E., Overbergh, L., Stoffels, K., Bouillon, R., Mathieu, C., 2007. Monocytes from type 2 diabetic patients have a pro-inflammatory profile. 1,25-Dihydroxyvitamin D(3) works as anti-inflammatory. Diabetes Res. Clin. Pract. 77 (July (1)), 47–57. Jakubzick, C., Gautier, E.L., Gibbings, S.L., Sojka, D.K., Schlitzer, A., Johnson, T.E., et al., 2013. Minimal differentiation of classical monocytes as they survey steady-state tissues and transport antigen to lymph nodes. Immunity 39 (September 19 (3)), 599–610. Kim, H.J., Ifergan, I., Antel, J.P., Seguin, R., Duddy, M., Lapierre, Y., et al., 2004. Type 2 monocyte and microglia differentiation mediated by glatiramer acetate therapy in patients with multiple sclerosis. J. Immunol. 172 (June 1 (11)), 7144–7153. Ruitenberg, J.J., Mulder, C.B., Maino, V.C., Landay, A.L., Ghanekar, S.A., 2006. VACUTAINER CPT. and Ficoll density gradient separation perform equivalently in maintaining the quality and function of PBMC from HIV seropositive blood samples. BMC Immunol. 7, 11. Sawada-Hase, N., Kiyohara, T., Miyagawa, J., Ueyama, H., Nishibayashi, H., Murayama, Y., et al., 2000. An increased number of CD40-high monocytes in patients with Crohn’s disease. Am. J. Gastroenterol. 95 (June (6)), 1516–1523. Shi, C., Pamer, E.G., 2011. Monocyte recruitment during infection and inflammation. Nat. Rev. Immunol. 11 (November (11)), 762–774. Weber, M.S., Prod’homme, T., Youssef, S., Dunn, S.E., Rundle, C.D., Lee, L., et al., 2007. Type II. monocytes modulate T cell-mediated central nervous system autoimmune disease. Nat. Med. 13 (August (8)), 935–943. Wong, K.L., Yeap, W.H., Tai, J.J., Ong, S.M., Dang, T.M., Wong, S.C., 2012. The three human monocyte subsets: implications for health and disease. Immunol. Res. 53 (September (1–3)), 41–57. Woollard, K.J., Geissmann, F., 2010. Monocytes in atherosclerosis: subsets and functions. Nat. Rev. Cardiol. 7 (Feburary (2)), 77–86. Zhou, L., Braat, H., Faber, K.N., Dijkstra, G., Peppelenbosch, M.P., 2009. Monocytes and their pathophysiological role in Crohn’s disease. Cell Mol. Life Sci. 66 (January (2)), 192–202. de Vries, T.J., Fourkour, A., Punt, C.J., Ruiter, D.J., van Muijen, G.N., 2000. Analysis of melanoma cells in peripheral blood by reverse transcription-polymerase chain reaction for tyrosinase and MART-1 after mononuclear cell collection with cell preparation tubes: a comparison with the whole blood guanidinium isothiocyanate RNA isolation method. Melanoma Res. 10 (April (2)), 119–126. van Furth, R., Cohn, Z.A., 1968. The origin and kinetics of mononuclear phagocytes. J. Exp. Med. 128 (September 1 (3)), 415–435.