A flow cytometry‐based assay to determine the phagocytic activity of both clinical and nonclinical antibody samples against Chlamydia trachomatis

Abstract Globally, an estimated 131 million new cases of chlamydial infection occur annually. Chlamydia trachomatis infection can cause permanent damage to the fallopian tubes in woman, resulting in infertility and a risk of ectopic pregnancy. There is a great need for a vaccine against Chlamydia trachomatis and as a result there is a need for assays to evaluate functional immune responses for use in future clinical trials and epidemiological studies. Antibodies play a crucial role in the defense against infection and can be protective by several functions, including phagocytosis and neutralization. Vaccine development could greatly benefit from a method to measure functional C. trachomatis‐specific antibodies in a large number of samples. In the current in vitro antibody protection assays, which measure the capacity of antibodies to facilitate phagocytic uptake of C. trachomatis, the phagocytosed bacteria have to be counted manually. This is both labor demanding, time consuming, and it prevents high‐throughput usage of this method. In this study, we, therefore, developed a simple and rapid flow cytometry based assay to measure the capacity of antibodies to mediate Fc‐receptor dependent phagocytosis. This method is highly reproducible and suitable to analyze large numbers of clinical and nonclinical samples. © 2018 The Authors. Cytometry Part A Published by Wiley Periodicals, Inc. on behalf of ISAC.

neutralization during an infection with C. trachomatis (11). Such roles could be to increase T cell activation, induce killing of epithelial cells via antibody-dependent cell-mediated cytotoxicity (12,13) or phagocytosis of antibody-coated chlamydiae (14).
Ab induced phagocytosis is mediated by the constant part of the Ab, which binds to Fc-receptors on the surface of the phagocytic cell. A phagocytosis assay measures the ability of an Ab to increase uptake of the bacteria into a phagocytic cell, and counting of the engulfed bacteria often involves visualization via immunofluorescence staining, followed by manually counting the bacterial inclusions via microscopy. This method of bacterial counting is a time consuming, labor demanding, low-throughput method, and only a limited number of samples can be handled. Efficient vaccine antigen discovery and development require more effective highthroughput assays.
Here we present a simple, rapid flow cytometry (FCM) based assay to measure the capacity of Abs to mediate Fcreceptor dependent phagocytosis. This method is highly reproducible and suitable to analyze large numbers of clinical and nonclinical samples.
HL-60 cells, a human promyeloblast cell line, were obtained from ATCC. These cells were maintained and stimulated identical to the PLB-985 cells.

CFSE Staining
3.37 3 10 9 IFU of SvD bacteria were washed in PBS at 20,000 g, 48C for 20 min. The pellet was resuspended in 500 ml of a 2 mM carboxyfluorescein diacetate succinimidyl ester (CFDA SE) solution (Vybrant V R CFDA SE Cell Tracer Kit, Thermofischer Scientific; diluted in PBS) and incubated for 30 min at 378C. The CFDA SE solution was prewarmed to 378C. CFDA SE is cell permeable and as soon as it enters cells, its acetate group cleaved by intracellular esterases to form the amin-reactive fluorescent product carboxyfluorescein succinimidyl ester (CFSE). To quench unbound CFSE, 500 ml of icecold PBS containing 10% BSA was added, followed by a centrifugation at 20,000 g, 48C for 20 min. The pellet was washed once more in ice-cold PBS containing 10% BSA. The bacteria were fixed by resuspending them in 500 ml of 4.2% formaldehyde (Cytofix V R , BD, San Jose, CA). After 20 min incubation at 48C the bacteria were washed with PBS. The pellet was resuspended in 250 ml PBS and stored at 48C until usage. 50% of the originally amount of SvD bacteria is lost due to the staining procedure.

FCM Based Phagocytosis Assay
The assay was performed in a 96 U-well Nunclon TM delta surface plate (Thermofischer Scientific) with a total volume of 200 ml. CFSE-labeled SvD bacteria and serum samples were diluted in PLB-985 assay media, mixed 1:1 and incubated for 40 min at 378C on a rocker table. 100 ml of DMF-stimulated PLB-985 cells at a concentration of 2 3 10 6 cells/ml were then mixed with 40 ml of the bacteria-serum mix. Assay medium was added to each well up to a total volume of 200 ml. The 96 U-well assay plate was incubated for 4 h at 378C on a rocker table. Afterwards, cells were immediately washed with PBS and kept at 48C from there on.
For some controls, the stimulated PLB-985 cells were preincubated for 30 min with different dilutions of human Fc receptor binding inhibitor monoclonal Ab (Thermofischer Scientific, Cat. 14-9161-73) or with 20 ml of the actin inhibitor Cytochalasin D (Sigma-Aldrich, St. Louis, MO) before the addition of bacteria-serum mix.

FCM Analysis to Determine Phagocytosis
All samples were measured with a BD FACSCanto equipped with a high throughput sample reader (HTS). Acquiring Software was BD FACSDIVA version 8.0.1. Analysis of the FCS files were performed with FlowJo version 10.3 (FlowJo, LLC, Ashland, Oregon).
The entire staining procedure was performed at 48C. The PBS was removed and the cells were resuspended in 50 ml fixable viability dye eFluor V R 780 (Thermofischer Scientific; Cat. 65-0865-14). After 15 min cells were washed with FACSbuffer (PBS with 2% FBS, 0.1% sodium azide, 1 mM EDTA). The cells were then fixed for 20 min with BD Cytofix V R (containing 4.2% formaldehyde) and washed in PBS. Finally, the cells were resuspended in 130 ml PBS. 80 ml of the stained samples were acquired with the HTS. PLB-985 cells were gated on FSC-A versus SSC-A. Doublets and triplets were excluded by gating on FSC-A versus FSC-H. Dead cells were excluded by gating on the negative population on the APC-Cy7 channel. The CFSE signal was then measured in the FITC channel. The baseline for chlamydia-positive cells was set by controls that incubated the phagocytic cells only with the CFSE-labeled bacteria and without any serum.

Ethical Statement
The protocol and procedures employed were reviewed and approved by our institutional review committee. Regarding the two human Abs used in the study, they have been described previously (17) in a study that was approved by the Local Ethical Committee for Copenhagen (01-008/03). The procedures followed were in accordance with the ethical standards of Local Ethical Committee for Copenhagen on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

Labeling C. trachomatis with CFSE
To make the phagocytosis assay as simple as possible, we decided to use fluorescently labeled bacteria, thus avoiding the step of intracellular staining of phagocytosed bacteria.
The first objective was to label the bacteria, and measure the labeling by flow cytometry. To label the bacteria, we used CFSE (as explained in materials and methods). Labeling with CFSE occurs inside the bacteria, making it a good method to visualize antibody mediated phagocytosis of bacteria, as it does not interact with surface antigens. At first, C. trachomatis (SvD) bacteria were stained with CFSE and analyzed by flow cytometry. The gating strategy is shown in Figure 1A. For comparison, bacteria were stained with an anti-C. trachomatis LPS Ab and some were co-stained with the anti-C. trachomatis

Original Article
LPS Ab and CFSE-labeled. The data showed an efficient labeling with CFSE, and an almost complete co-staining of anti-C. trachomatis LPS on CFSE-labeled bacteria (Fig. 1A). Moreover, CFSE labeling and fixation did not affect phagocytosis of the bacteria, as we observed similar phagocytosis of CFSE labeled and nonlabeled bacteria (Supporting Information Fig. S1).

Antibody Mediated Uptake of Bacteria
We next coated the CFSE labeled SvD bacteria with different concentrations of rabbit polyclonal Abs directed against Hirep1. The Hirep1 vaccine construct is based on the C. trachomatis Major Outer Membrane Protein (MOMP), and is known to induce neutralizing antibodies [(6) and Olsen et al. unpublished observations)]. We measured the uptake of Abcoated bacteria by PLB-985 cells, an immature myeloid cell line, which was differentiated by DMF treatment for 5 days into terminally mature neutrophils, closely mimicking the functions of blood neutrophils (18). The measurement was performed by flow cytometry and the gating strategy for measuring uptake of CFSE labeled bacteria into PLB-985 cells is shown in Figure 1B.
Control stainings are shown in Figure 1C. It shows a comparison of PLB-985 cells with PLB-985 cells being subjected to CFSE-labeled C. trachomatis bacteria (noncoated or precoated with anti-Hirep1 serum or na€ ıve serum). Although some phagocytosis was observed by adding CFSE labeled bacteria (noncoated or coated with na€ ıve serum) to PLB-985 cells, the phagocytosis increased when adding C. trachomatis bacteria that had been coated with anti-Hirep1 serum (Fig. 1C). At MOIs of 10, 20, or 40, anti-Hirep1 Abs induced a strong phagocytosis. Approximately 58% of PLB-985 cells stained positive for C. trachomatis (at MOI 10) (Fig. 2A). Decreasing the concentration of the Abs led, as expected, to a decrease in phagocytosis to 37% at a serum dilution of 1:10,000. Significantly less positive cells were observed in the presence of serum from na€ ıve animals ( Fig. 2A). We choose a MOI of 10 as the bacteria-to-cell ratio we would use in this assay. Furthermore, we noted that the phagocytosing capability of the PLB-985 cells increased following stimulation with DMF (Fig.  2B). To determine the variability of the phagocytosis assay it was repeated in five independent experiments with anti-Hirep1 Ab coated CFSE-labeled bacteria at a serum dilution of 1:10 and bacteria concentration of MOI 10 with DMF stimulated PLB-985 cells. The intra-assay coefficient of variation (CV) was between 1.13 and 2.77% and the interassay CV was 3.76% (Table 1). We, therefore, conclude that the assay is reproducible using a serum dilution of 1:10 and bacteria concentration of MOI 10. Using serum dilutions of 1:100-1:10,000 gave, as expected, a reduced phagocytosis and a higher CV value, although even with a dilution of 1:10,000, the assay was reproducible (Supporting Information Fig. S2).
We next compared the anti-Hirep1 Ab with a rabbit antiserum directed against another chlamydial surface-exposed antigen, the CT043 antigen (19). Interestingly, in contrast to the anti-Hirep1 Ab, an Ab directed against the CT043 antigen did not induce phagocytosis, demonstrating that being surface exposed does not automatically lead to phagocytosis. We also tested a control serum from rabbits vaccinated with a tuberculosis antigen ["H56" (20)]. As expected, this Ab did also not induce phagocytosis (Figs. 2C and 2D).
Finally, we also tested the FCM assay with HL-60 cells, another neutrophilic cell line that is used in several laboratories. CFSE labeled bacteria coated with the anti-Hirep1 Ab were added to DMF stimulated HL-60 cells and phagocytosis was measured by FCM, as explained above. The results showed that at a MOI of 40 HL-60 and PLB-985 cells showed a similar phagocytosis (Fig. 3). However, at a MOI of 10, PLB-985 cells showed a significantly higher phagocytosis compared to HL-60 cells.
Taken together, using a rabbit Ab specific for MOMP, we could show phagocytosis of the CFSE labeled bacteria. The percentage of phagocytosing cells was dependent on the amount of Ab used. Compared to the na€ ıve rabbit control serum, no increase in phagocytosis was observed with antibodies directed against a tuberculosis antigen or with an anti-CT043 Ab.

Inhibition of Phagocytosis
We next wanted to explore the mechanism behind the observed phagocytosis. As expected, the Fcc receptors (CD16 and CD32) were expressed on PLB-985 cells and CD64 on $40% of the cells (Fig. 4A). To confirm that the phagocytosis was dependent on Fcc receptors, we examined the effect of inhibiting the binding of anti-Hirep1 Abs to the Fcc receptors, on the ability of anti-Hirep1 Abs to induce phagocytosis. Stimulated PLB-985 cells were preincubated with three different dilutions of human Fcc receptor binding inhibitor polyclonal Ab (1:4, 1:10, 1:100), prior to incubation with CFSElabeled bacteria coated with the anti-Hirep1 Ab. The result showed that in the absence of preincubation with Fcc receptor inhibitor, approximately 60% of the PLB-985 cells were SvD positive (Fig. 4B). Preincubation with Fcc receptor inhibitor reduced the phagocytosis in a concentration dependent manner (Figs. 4B and 4D) from 60% to 11%. Thus, efficient phagocytosis was dependent on an intact interaction between the Ab coated bacteria and the Fcc receptors.
Several reports have shown that Fc receptor mediated internalization involve polymerization of actin (21,22). Therefore, preventing actin polymerization should inhibit internalization. To test this, PLB-985 cells were preincubated with Cytochalasin D, an inhibitor of actin polymerization, before the addition of bacteria-serum mix. At all concentrations tested, Cytochalasin D blocked Ab mediated phagocytosis of the labeled bacteria (Figs. 4C and 4D).
In summary, PLB-985 cells stained positive for CD16 and CD32 and partially for CD64. Moreover, using the FCM phagocytosis assay, we show that anti-Hirep1 Ab induced  Original Article phagocytosis could be blocked by an Fcc receptor inhibitor, or by inhibiting the polymerization of actin.

Phagocytosis Mediated by Human and Murine Serum
An obvious use for a FCM based phagocytosis assay is the high-throughput testing of serum samples from human donors, or samples from mice vaccinated with different vaccine candidates. It was, therefore, important to show that the assay is also suitable for human and murine serum. We used serum from mice infected with C. trachomatis SvD and serum from mice immunized with a construct containing the neutralizing epitope in the VD4 region of MOMP, "VD4p4." In addition, we tested a murine monoclonal Ab specific for C. trachomatis LPS. The assay was performed in the same way as shown above with the rabbit serum. CFSE-labeled SvD C. trachomatis bacteria were coated with diluted serum samples (1:10-1:10,000), and subsequently incubated with the DMF-stimulated PLB-985 cells for 4 h, before being subjected to FCM analysis.
Both serum from infected and immunized mice induced phagocytosis (Fig. 5A). A murine anti-C. trachomatis LPS mAb also induced phagocytosis in a dose-dependent manner (Fig. 5A).
We also tested serum from a human donor with a confirmed genital SvD infection. We choose a donor that we have previously shown to possess Abs directed against MOMP SvD (16,17). In contrast to serum from a nonexposed "na€ ıve" human donor, the serum from the exposed human donor induced a strong phagocytosis leading to 52.6% positive PLB-985 cells (Fig. 5B).
Taken together, the assay showed to be applicable for both human and murine serum. Murine Abs directed against the VD4 region from MOMP induced phagocytosis of the C. trachomatis bacteria, whereas serum from nonvaccinated mice did not. Anti-LPS Abs also induced phagocytosis.

DISCUSSION
The flow cytometry assay was recently shown to be an alternative to microscopy in terms of counting and titrating bacteria (23). In this article, we show that flow cytometry is also suitable for high-throughput testing of the phagocytosing ability of an Ab directed against an intracellular bacterium, in our case C. trachomatis.
For the FACS based phagocytosis assay, we used CFSE labeled bacteria. While others have also used CFSE labeling to visualize the appearance of bacteria in target cells (24), there are other options, such as fluorescein isothiocyanate (FITC) or a pH sensitive dye pHrodo TM (25,26). FITC is an amine group reactive compound that binds to every protein, which is also true for CFSE. However, we used CFDA SE, which is a cell permeable dye. Once inside the bacteria, intracellular esterases cleave the acetate groups, which results in the reactive CFSE form. Thus, this method provides an intrabacterial labeling, while FITC stains the outside of the bacteria. The pH sensitive pHrodo dye is labeling the bacteria in the same way as CFDA SE. However, the dye will change its excitation maximum according to the pH, and this allows a distinction between bacteria in a neutral pH environment, such as on the surface of cells, or bacteria in an acidic intracellular environment, such as the phagolysosomes. In future experiments, we will compare the different staining methods. It should be noted that the pHrodo dye in general is more expensive than CFDA SE, which is an important factor to consider in the development of a high throughput assay, to be used with many samples.
We observed 50-70% of phagocytosing cells after coating bacteria with serum of vaccinated or infected animals/individuals in our in vitro assay. Phagocytosis was observed not only with human serum, but also with murine and rabbit serum. Figure 5. Phagocytosis assay using human and murine serum. CFSE-labeled SvD bacteria at a MOI of 10 were incubated with (A) serum from infected/vaccinated mice or a monoclonal mouse-a-C. trachomatis LPS Ab, or (B) serum from an infected human donor. Serum of na€ ıve mice/ humans were used as negative controls. Murine sera were titrated from 1:10 to 1:10000 (C. trachomatis LPS mAb was pre-diluted to 1 ng/ll) and human sera from 1:100 to 1:10000. Phagocytosis was measured by flow cytometry. (C) Representative pseudo-color dot plots of the phagocytosis assay with all sera at a dilution of 1:10 (for human sera 1:100). Mean and SD are shown at a sample size of n 5 3 (A and B).

Original Article
The ability of IgG from different species to bind to human Fc receptors is in agreement with previous studies (27)(28)(29)(30). Fabbrini et al. developed an Ab dependent phagocytosis assay for Group B Streptococcus (26). They used the human neutrophillike HL-60 cell line. They also tested different murine and rabbit sera from vaccinated animals and achieved a phagocytic uptake of approximately 60%. We also tested the HL-60 cell line, and found HL-60 cells also efficiently phagocytosed Ab coated bacteria, although requiring higher MOIs than with PLB-985 cells. The specificity and sensitivity of the FCM-based assay was not compared with the standard microscopy assay.
The phagocytosis in our assay was Fcc receptor and actin polymerization dependent as blocking FcR binding or actin polymerization prevented uptake of the bacteria (Fig. 4). The FcR blocking reagent is known to bind Fcg receptors, indicating a role for these receptors in mediating the phagocytosis. All the Fcc receptors are capable of endocytosis, and the precise role for each of the individual Fcc receptors in phagocytosing C. trachomatis requires further experiments (31,32). Interestingly, the FACS assay could distinguish between a binding nonphagocytosing Ab (directed against CT043) and a binding phagocytosing Ab (directed against Hirep1) (Fig.  2C). As phagocytosis require crosslinking of the Fc receptor (33), it could be speculated that the lack of phagocytosis with the CT043 Ab is due to the CT043 antigen being too far apart on the chlamydial surface for the binding Abs to crosslink the Fc receptor on the phagocyte.
We also tested an LPS mAb and found that despite a complete lack of neutralizing capability (data not shown), this mAb was, however, able to induce phagocytosis of C. trachomatis (Fig. 5). Thus, this mAb could be used to selectively study the phagocytosis of C. trachomatis.
In conclusion, we have developed a simple, rapid, reproducible flow cytometric based assay that use cultivable effector/ target cells and is able to handle a large number of samples.