Fluorescent target array T helper assay: A multiplex flow cytometry assay to measure antigen-specific CD4+ T cell-mediated B cell help in vivo

Journal of Immunological Methods 387 (2013) 181–190

Contents lists available at SciVerse ScienceDirect

Journal of Immunological Methods journal homepage: www.elsevier.com/locate/jim

Research paper

Fluorescent target array T helper assay: A multiplex flow cytometry assay to measure antigen-specific CD4 + T cell-mediated B cell help in vivo☆ Benjamin J.C. Quah ⁎, Danushka K. Wijesundara, Charani Ranasinghe, Christopher R. Parish Department of Immunology, John Curtin School of Medical Research, Australian National University, Canberra, ACT, 2601, Australia

a r t i c l e

i n f o

Article history: Received 17 September 2012 Received in revised form 18 October 2012 Accepted 22 October 2012 Available online 30 October 2012 Keywords: T helper cells B cell activation Fluorescent target array Multiplex CFSE

a b s t r a c t CD4 + T cells play a central role in regulating the immune response. Their effector function is commonly assessed by their capacity to secrete cytokines detected by ELISPOT and intracellular cytokine staining. However, one aspect of their effector function that is often overlooked is their ability to help activation of cognate B cells directly, a process that is initiated through the engagement of their T cell-receptor (TCR) with cognate peptide presented on major histocompatibility complex class II (MHC-II) molecules by B cells. Here we report a method to monitor CD4 + T cell-mediated B cell help in vivo using a multiplex high throughput assay. This assay utilizes a fluorescent target array (FTA), which is composed of lymphocytes labeled with numerous (>200) unique fluorescence signatures that can be delineated in a single recipient animal based on combination labeling with the three vital dyes carboxyfluorescein diacetate succinimidyl ester (CFSE), CellTrace Violet (CTV) and Cell Proliferation Dye eFluor 670 (CPD). By pulsing different B cell populations in a FTA with titrated amounts of cognate MHC-II binding peptides, CD4 + T cell help could be assessed by measuring induction of the B cell activation markers CD69 and CD44 by antibody labeling and flow cytometry. We call this the “FTA T helper assay”, and have found it to be a robust and sensitive assay to measure CD4 + T cell helper activity across a multitude of peptide-pulsed B “target” cells in real time in vivo. Furthermore, the technique can be used simultaneously with the FTA killing assay that measures cytotoxic T cell function, to provide a comprehensive tool for measuring both CD4 + and CD8 + T cell activity during an immune response in vivo. © 2012 Elsevier B.V. All rights reserved.

1. Introduction CD4+ T cells play a central role in immune regulation, which is underscored by their capacity to differentiate into multiple T cell subsets with unique function (Zhu et al., 2010). TH1, TH2 and TH17 CD4+ effector T cell subsets are characterized by their production of the cytokines IFN-γ, IL-4 and IL-17/22, respectively (Sakaguchi et al., 2008; Fazilleau et al., 2009; Zhu et al., ☆ Research support: This work was supported by a Project Grant to BQ and CP and a Program Grant to CP from the National Health and Medical Research Council (NHMRC) of Australia. ⁎ Corresponding author at: Department of Immunology, John Curtin School of Medical Research, Building 131, Garran Road, The Australian National University, Canberra ACT 2601 Australia. Tel.: +61 2 6125 4716; fax: +61 2 6125 2595. E-mail address: [email protected] (B.J.C. Quah). 0022-1759/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jim.2012.10.013

2010). iTREGS (Sakaguchi et al., 2008) and TFH (Fazilleau et al., 2009) CD4+ effector T cell subsets are notable for their capacity to suppress immune responses (iTREGS) and regulate B cell differentiation (TFH). CD4+ effector T cells are an important indicator in immunotherapy design, but are often overshadowed due to the emphasis placed on generating robust cytotoxic T lymphocyte (CTL) and antibody responses (Rappuoli, 2007). Commonly, CD4+ T cell effector activity is measured by MHC-II/peptide tetramer reactivity (Nepom, 2012) and their capacity to produce cytokines as measured by ELISPOT (Czerkinsky et al., 1983) and intracellular cytokine staining (Prussin and Metcalfe, 1995). However, one aspect of their effector function that is often overlooked is their ability to help activation of cognate B cells directly. CD4+ T helper (TH) cells modulate B cell differentiation into plasma cells, dictating their fate as extrafollicular short-lived plasma cells or long-lived

182

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

plasma cells and their capacity to switch antibody classes and become high affinity B cells (Fazilleau et al., 2009; Zhu et al., 2010). The fundamental interaction that initiates these events is via the TCR of the effector TH cell engaging cognate peptide presented on MHC-II by the B cell. As a result of this antigenspecific interaction the B cell is “helped” to undergo activation and differentiation via receptor ligand interactions and paracrine cytokine signaling (Fazilleau et al., 2009; Zhu et al., 2010). Given the importance of this interaction, we sought to investigate whether a method could be developed to monitor TH cell-mediated B cell help in an in vivo setting in a multiplex high throughput assay that could be combined with other T cell effector read outs by flow cytometry. Recently we utilized the three vital dyes CFSE, CTV and CPD to label lymphocytes with numerous (> 200) unique fluorescence signatures based on the different fluorescence emission spectra of the dyes and using multiple fluorescence intensities of the dyes (Quah et al., 2012; Quah et al., 2012). These lymphocyte fluorescent target arrays (FTAs) were used as targets, after pulsing them with MHC-I binding peptides, in an in vivo CTL killing assay (Quah et al., 2012). Since more than 200 targets could be detected by this technique the assay allowed the simultaneous measurement of the in vivo killing of many target cell clusters pulsed with numerous peptides at different concentrations and the inclusion of many replicates. To expand on this technique, here we report the use of the FTA to measure the capacity of TH cells to activate B cells within an FTA pulsed with cognate MHC-II binding peptides through measuring upregulation of the B cell activation markers CD69 and CD44. We found the assay to be highly sensitive, detecting antigen-specific, peptide dose-dependent upregulation of activation markers on B cells induced by scarce TH cells in vivo. The FTA T helper assay generates reproducible results that correlate well with the number of antigen-specific CD4 + T effector cells generated in vivo. This assay can be coupled with the FTA killing assay to allow the simultaneous measurement of TH cell and CTL effector activity against a multitude of target cells pulsed with a broad concentration range of several different MHC binding peptides in real time in vivo, making it a valuable screening tool for assessment of immune responses. 2. Methods

separation (Miltenyi Biotec) as previously described (Quah et al., 2004).

2.3. Peptides and virus Peptides were synthesized at the Australian Cancer Research Foundation Biomolecular Resource Facility, JCSMR, ANU. Peptides included the MHC-II-binding peptides ISQAVH AAHAEINEAGR (ISQA), which is recognized by the OT-II Tg TCR in B6 (IA b) mice, and PVGEIYKRWIILGLN (Gag Th, a HIV gag epitope recognized by CD4 + T cells in BALB/c (H-2 d) mice (Mata and Paterson, 1999)) and the MHC-I-binding peptides SIINFEKL (which is recognized by the OT-I Tg TCR in B6 (K b) mice), ASNENMDAM (NP68, an in Flu NP epitope recognized by CD8 + T cells in B6 (D b) mice), SPYAAGYDL (F2L, a Vaccinia virus epitope recognized by CD8 + T cells in BALB/c (L d) mice), SPGAAGYDL (F2L mod, a Modified Vaccinia Ankara Virus epitope recognized by CD8 + T cells in BALB/c (L d) mice), AMQMLKETI (HIV gag, a HIV gag epitope recognized by CD8 + T cells in BALB/c (K d) mice), AMQMLKDTI (HIV gag mut, a HIV gag subtype C variant of HIV gag (Earl et al., 2009)), VGPTPVNII (HIV pol, a HIV pol epitope recognized by CD8 + T cells in BALB/c (D d) mice), and RGPGRAFVTI (HIV envelop (env), a HIV env epitope recognized by CD8 + T cells in BALB/c (D d) mice (Takeshita et al., 1995)). Fowl pox virus (FPV)-HIV and Vaccinia virus (VV)-HIV (VV-HIV (Gag/Pol) and VV-HIV (Env)) stocks were prepared as described previously (Ranasinghe et al., 2006). The VV-OVA construct (Restifo et al., 1995) was kindly provided by Dr D. C. Tscharke. 2.4. Fluorescent dye labeling CFSE, CTV and CPD dye labeling was as described previously with slight modification (Quah and Parish, 2012). Briefly, splenocytes in 20 °C RPMI 1640 (Invitrogen) supplemented with 10% fetal calf serum (FCS), were labeled with 0–66,000 nM of each dye in 1–2 mL aliquots for 5 min and then washed ≥3 times (Quah and Parish, 2010, 2012). The membrane intercalating dye DiI (Invitrogen) was used at 14 μM to label cells as described by the manufacturer.

2.1. Animals

2.5. FTA preparation

Mice were obtained from the Australian National University (ANU) Bioscience Services, ANU. Mice were housed and handled according to the guidelines of the ANU Animal Experimentation Ethics Committee. Mouse strains used were C57BL/6 (B6), B6.CD45.1 (B6 congenic for CD45.1) and BALB/c. Transgenic (Tg) mouse strains were OT-II (TCR-Tg specific for ISQAVHAAHAEINEAGR (ISQA)/IAb on a B6 (CD45.1+/2+) background (Barnden et al., 1998)), OT-I (TCR-Tg specific for SIINFEKL/Kb on a B6 background (Hogquist et al., 1994)). Male mice were used at 6–10 weeks of age.

FTA preparation was as described previously (Quah et al., 2012). Briefly, 2 mL aliquots of splenocytes were initially labeled with several concentrations (dependent on the type of FTA) of CTV. Cells were then split equally and labeled with several concentrations of CFSE and then washed once. Cell aliquots were then incubated with MHC-I and MHC-II binding peptides for 1 h at 37 °C, and washed twice at 4 °C, with the first wash being through a FCS cushion. Samples to be labeled with the same concentration of CPD were then pooled and washed once more, before being labeled with various concentrations of CPD. After washing the cell samples twice, all aliquots were pooled and washed again. Where necessary, pooled cells were also labeled with DiI. The FTA was then counted and resuspended at up to 25 × 10 7cells/mL ready for injection.

2.2. Cell preparation Lymphocytes were obtained from spleen and CD4 + T cells and CD8 + T cells enriched from lymphocytes by MACS

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

2.6. In vivo assays Pooled OT-II and OT-I T cells (5 × 10 7 or less) were adoptively transferred into the lateral tail vein of host mice in 200 μL of PBS using a 29 gauge needle and syringe. Approximately 0.5 h post adoptive transfer of T cells, host mice were challenged with protein antigen in the form 25 μg of ovalbumin (OVA, Sigma) co-injected with 1 μg of LPS (Sigma, LPS was included to act as a danger signal and enhance lymphocyte responses) in a 200 μL bolus or immunized with 5 × 10 6 PFU FPV-HIV or a combination of 2.5 × 10 6 PFU of VV-HIV (Gag/Pol) and 2.5 × 10 6 PFU of VV-HIV (Env) injected intramuscularly (i.m.). FTAs were injected i. v. into host mice at a total cell number of 5 × 10 7 or less and left in vivo for 18 h before FTA cells were assayed for target cell help and killing. 2.7. Antibody and MHC-I tetramer labeling of cells Splenocytes were initially labeled with SIINFEKL-peptideK b-APC (for OT-I T cell detection) MHC-I tetramers for 25 min at 37 °C and then washed twice, ready for antibody staining. Antibodies used to delineate cell populations included antiCD4-PE-Cy7, anti-CD4-Brilliant Violet 605, anti-CD4-Alexa Fluor 700, anti-CD8α-APC-eFluor 780, anti-CD44-biotin, antiB220-PerCP-Cy5.5, anti-B220-Alexa Fluor 700, anti-CD45.1PE-Cy7, anti-CD45.2-Horizon V500, anti-CD69-PerCp-Cy5.5, anti-CD69- Brilliant Violet 605 and anti-Vα2-PE, purchased from either BD Bioscience, eBioscience or Biolegend. Cell viability was assessed with the dye Hoechst 33258 (1 μg/ml, Calbiochem-Behring Corp.). Cells were labeled with antibodies and Hoechst 33258 for 20–30 min on ice then washed twice, as previously described (Quah et al., 2004). The second step reagent streptavidin-PE-Alexa Fluor 610 (Invitrogen), was incubated with cells for 20–30 min on ice and then the cells washed twice before analysis. 2.8. Flow cytometry and data analysis Analytical flow cytometry was performed using a Fortessa flow cytometer (Becton Dickinson). Spectral overlap of each fluorochrome was monitored using single stained controls and compensation performed to help reduce spectral “spillover” by mean fluorescence intensity (MFI) value comparisons. Post acquisition gating was used to analyze cell subsets using FlowJo software (Tree Star, OR) and detailed examples of the gating strategies employed in this study are depicted in Supplementary Fig. 1 and Supplementary Fig. 2. The flow cytometer underwent periodic quality control assessment using 8 channel fluorescent beads throughout the period of experimentation. 2.9. T cell help and % specific killing calculations Levels of T cell help were assessed on the basis of FTA B cell expression of activation markers, including CD69 and CD44, detected by antibody labeling and flow cytometry. Expression of the activation markers in the figures was calculated by subtracting the activation marker geometric mean fluorescence intensity (GMFI) on FTA B cells from naïve mice from that on FTA B cells from primed mice. The %

183

specific killing was assessed as previously described (Quah et al., 2012) using the following formula: 2

0

Targetsþpeptide primed

.

13

6 B B % specif ic killing ¼ 6 41−@ Targetsþpeptide .

Targetsþnil primed

naive

C7 C7  100⋅ A5

Targetsþnil naive

2.10. Statistical analysis Since the sample size of the data sets precluded the use of a normality test, nonparametric tests were performed for statistical analysis. The Mann–Whitney test was performed to test for statistically significant differences between two samples. The Spearman's rank-order correlation coefficient was calculated to assess for the monotonic relationship between two variables. Statistical values as well as area under curve (AUC) values were calculated using GraphPad Prism Software. 3. Results 3.1. Establishing a FTA T helper assay using TCR transgenic T cells To determine if FTAs could be used to assess CD4 + T cell-mediated B cell help, a FTA was constructed by labeling spleen cells from B6.CD45.1 mice using combinations of 5 fluorescence intensities of CFSE, 6 intensities of CTV and 6 intensities of CPD to generate a three dimensional array of 180 distinguishable cell clusters (Fig. 1a). For the purpose of peptide pulsing, this FTA was divided into 6 repeats (based on CPD fluorescence intensity) of 30 target clusters (based on CFSE and CTV fluorescence intensity) and these 30 clusters pulsed with MHC-binding peptides. This allowed for a titration of 7 concentrations of two MHC-II binding peptides (ISQA and Gag Th) to assess T cell “help” by TH cells (Fig. 1b). In addition, 7 concentrations of two MHC-I binding peptides (SIINFEKL and NP68) to measure T cell “killing” by CTLs were included to determine if the FTA could be used to measure CTL activity in unison with TH cell activity (Fig. 1b). These titrations were repeated 6 times within the FTA (based on CPD fluorescence). All 180 FTA cell clusters were pooled and injected into B6 host mice containing CD4 + T cells from OVA (ISQA)-specific OT-II TCR Tg mice and CD8+ T cells derived from OVA (SIINFEKL)-specific OT-I TCR-Tg mice. To generate effector cells, host mice were challenged 3 days before FTA injection with either OVA (with LPS) or VV-OVA (a Vaccinia virus expressing recombinant OVA). As a control an additional host mouse received no antigen (referred to as the naïve mouse). 18 h after FTA transfer, spleen cells from host mice were harvested and MHC-II-peptide pulsed FTA B cells assessed for activation by measurement of the early activation marker CD69 and MHC-I-peptide pulsed FTA cells (B and T cells) assessed for death. Since the FTA was constructed using B6.CD45.1 mice and these injected into B6 mice, FTA target clusters could easily be delineated from host mouse lymphocytes by CD45.1/2 markers using flow cytometry (Supplementary Fig. 1). As an example of how the FTA was used to assess B cell activation, histogram

184

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

Fig. 1. Setup of a 180 FTA T helper assay. Splenocytes from B6.CD45.1 mice were labeled with combinations of CTV (0 nM, 350 nM, 1295 nM, 4792 nM, 17,729 nM and 65,595 nM), CFSE (0 nM, 200 nM, 1000 nM, 5000 nM and 25,000 nM) and CPD (0 nM, 100 nM, 669 nM, 2541 nM, 9657 nM and 36,698 nM), and pulsed with peptides to generate a total of 180 target cell clusters. All target cells were pooled and injected i.v. into three B6 host mice that had three days earlier been injected i.v. with 2 × 107 CFSE-labeled pooled T cells derived from OT-II and OT-I mice. At the time of T cell injection, one host mouse was also injected i.v. with OVA (25 μg) and LPS (1 μg) and another host mouse was injected with 5 × 106 PFU of VV-OVA i.m. to prime injected T cells (Primed). The third animal received no antigen and acted as a control (Naïve). 18 h after FTA injection, splenocytes were collected and target cells delineated from host splenocytes by CD45 allotype antibody labeling and flow cytometry (SFig. 1). (a) 3D plots, 2D plots and histogram plots of the fluorescence intensities of FTA cells from the naïve animal. (b) 2D plots of the fluorescence intensities of target cells from the naïve and VV-OVA-primed host animals and the associated layout of peptide loading titrations mapped out on the basis of CTV and CFSE fluorescence. This layout was repeated 6 times, with each repeat being tagged with one of the 6 fluorescence intensities of CPD shown in a). Each target cell cluster was pulsed with either 0.14 μM to 100 μM, at 3-fold dilution steps, of each of the MHC-II peptides ISQA or Gag Th, or 0.00001 μM to 10 μM, at 10-fold dilution steps, of each of the MHC-I peptides SIINFEKL or NP68. Two target cell clusters were also left “unpulsed” (Nil) and served as controls. (c) An example of histogram analysis of the FTA T helper assay, with B220+ FTA cells being assessed for CD69 upregulation from the OVA (+LPS) primed animal compared to those from the naïve animal. Geometric mean fluorescence intensity (GMFI) of each histogram is shown for FTA B220+ cells from primed and naïve animals. (d) An example of histogram analysis of the FTA killing assay, with B220+ FTA cells being assessed for death from the VV-OVA primed animal compared to those from the naïve animal.

analysis of FTA B cell expression of CD69 is shown in Fig. 1c. FTA B cells pulsed with the OT-II-specific ISQA peptide clearly showed a peptide dose-dependent upregulation of CD69 in OVA-primed animals compared to naive animals. B cell activation was not observed when B cells were pulsed with a HIV gag-derived CD4+ T cell epitope not recognized by OT-II CD4 + T cells (Fig. 1c). In addition, histogram analysis showed that cells within the same FTA, pulsed with the OT-I-specific SIINFEKL MHC-I binding peptide showed a peptide dosedependent reduction in number in primed animals compared to naïve animals (Fig. 1d). In contrast, cells within the same FTA pulsed with the control flu-specific NP68 MHC-I binding peptide were not killed (Fig. 1d). From these data the ability of OT-II CD4 + T cells to help FTA B cells upregulate CD69 and OT-I CD8+ T cells to kill FTA cells

was calculated for all FTA cell clusters in the OVA-primed and VV-OVA-primed animals relative to the naïve animal as described in Methods. In addition, the OT-II and OT-I lymphocytes transferred into the recipient animals were CFSE labeled to allow their proliferative responses to be monitored at the time of FTA harvest. Assessing the CD4 + T cell response by CFSE-dilution revealed that OT-II T cells had proliferated extensively in response to both OVA and VV-OVA immunization at 4 days post priming (Fig. 2a). However, at this time point the recovery of OT-II CD4+ T cells from VV-OVA-primed animal (0.016% of CD4 + T cells) was ~6 folds lower than those recovered from OVA-primed mice (0.11% of CD4 + T cells) (Fig. 2b). Despite the low number of potential CD4+ T effector cells generated from these immunizations, the FTA T helper assay was capable of detecting “help” mediated by CD4 + T cell

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

185

Fig. 2. The FTA T helper assay allows the detection of B cell activation by low frequency TH cells. Flow cytometry analysis was used to assess the capacity of OT-II TH cell to help FTA B cells and assess the capacity of OT-I CTL to kill FTA cells from the experiment described in Fig. 1. (a) CD4+ OT-II and CD8+ OT-I T cells were discriminated from host splenocytes by CD45 allotype antibody labeling or MHC-I tetramer binding, respectively and assessed for cell division based on CFSE dilution 4 days after OVA (+LPS) or VV-OVA immunizations. (b) The number of recovered CD4+ OT-II and CD8+ OT-I T cells was assessed as the % of total CD4+ or CD8+ T cells (respectively) within the spleen. (c) B cell activation in the FTA T helper assay, as measured by the GMFI of CD69 expression on FTA B220+ cells from OVA (+LPS) or VV-OVA immunized mice above the GMFI of CD69 expression on FTA B220+ cells from naïve mice. Left panel displays B cell activation from all 6 intra-animal replicates, while the right panel shows the mean and standard error of mean of the same data. (d) % Specific killing of FTA cells pulsed with different concentrations of SIINFEKL and NP68. Left panel displays % specific killing of all 6 intra-animal replicates, while the right panel shows the mean and standard error of mean of the same data. (e) Area under curve values for T help and T killing assays depicted in c) and d). Mean and standard error of mean are shown for AUC values of intra-animal replicates.

effectors generated by OVA and even VV-OVA priming (Fig. 2c). Indeed, B cells pulsed with the OT-II-specific ISQA peptide showed a peptide dose-dependent upregulation of CD69 across all 6 intra-animal replicates. In contrast, B cells pulsed with the control Gag Th MHC-II peptides did not upregulate CD69. An advantage of using an FTA to monitor CD4 + T cells responses is that it allows multiple concentrations of cognate peptides to be used in many replicates thereby allowing robust summary statistics to be generated. One helpful statistic in this respect is the measurement of the area under each response curve to give a measure of the cumulative T cell “help” for a particular peptide-bearing B cell population. This calculation showed that the OVA priming regime generated an ~3 fold higher cumulative TH cell response against the ISQA epitope than the VV-OVA immunization strategy (Fig. 2e). It should be noted that without transferred OT-II cells, neither the OVA nor the

VV-OVA immunization regimes gave any detectable TH cell responses (data not shown). Another advantage of the large number of targets in a FTA is that they can be used to potentially monitor CD4+ T help responses and CTL killing responses simultaneously. In this example both the OVA and VV-OVA immunization strategies generated extensive OT-I CD8+ T cell proliferative responses 4 days post priming (Fig. 2a). However, at this time point the recovered OT-I CD8+ T cells from VV-OVA-primed animals (4.0% of total CD8+ T cells) was ~11 folds higher than those recovered from OVA-primed mice (0.34% of total CD8+ T cells) (Fig. 2b). Both immunization regimes showed a peptide dose-dependent killing of SIINFEKL-pulsed targets, but not NP68 peptide-pulsed control targets, across all 6 intra-animal replicates within the FTA. However, the VV-OVA immunization regime generated effectors that resulted in a >2 fold higher

186

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

cumulative killing response than the OVA immunization (Fig. 2e). The above examples show that the FTA assay can generate reproducible intra-animal replicate measurements of TH cell activity in vivo. To test the reproducibility of the FTA T helper assay between animals, 6 replicate mice were immunized with OVA (with LPS) after adoptive transfer of OT-II TCR Tg T cells (as above) and effector T cells from each replicate animal assessed for their capacity to help MHC-II peptide-pulsed B cells in a FTA relative to an unimmunized control mouse (Fig. 3). In this experiment it was found that the FTA T helper

assay gave reproducible detection of B cell activation in all six mice (Fig. 3a). However, it was noted that 3 mice (mouse 2, 6 and particularly mouse 4), had a statistically significant difference (P≤ 0.05) in the AUC measurement of cumulative B cell activation compared to mouse 3 (Fig. 3d, right axis). This was despite all of the OT-II CD4+ T cell populations in each of these replicate mice having very similar proliferation profiles in terms of the proportion of cells undergoing division, as assessed by CFSE dilution (Fig. 3b). However, further analysis of the actual number of T cells present in each generation of the dividing OT-II CD4+ T cell population, revealed variability in

Fig. 3. The FTA T helper assay generates reproducible inter-animal results that correlate with the frequency of TH cells in vivo. Flow cytometry analysis was used to assess the capacity of OT-II TH cells to help FTA B cells in 6 replicate mice 4 days after immunization with OVA (+LPS) analogous to the experiment described in Fig. 1 and including targets pulsed with 1.65 μM to 400 μM, at 3-fold dilution steps, of each of the MHC-II peptides ISQA or Gag Th. An unimmunized mouse served as a control. (a) B cell activation in the FTA T helper assay, as measured by the GMFI of CD69 expression on FTA B220+ cells from the 6 replicate OVA (+LPS)-immunized mice above the GMFI of CD69 expression on FTA B220+ cells from naïve mice. (b) CD4+ OT-II T cells were discriminated from host splenocytes by CD45 allotype antibody labeling and assessed for cell division based on CFSE dilution. (c) The total number of CD4+ OT-II T cells in each generation of division was assessed in each spleen of the 6 animal replicates. (d) The total number of CD4+ OT-II T cells was assessed in each spleen of the 6 animal replicates as well as the AUC values for T help responses as depicted in a). Means and standard errors of means are shown for AUC values of intra-animal replicates and P values are also shown from statistical analyses comparing the intra-animal replicates of mouse 3 with the intra-animal replicates of all other animals. The two-tailed Spearman's rank-order correlation coefficient of the corresponding AUC values and total spleen OT-II CD4+ T cell numbers of the 6 animal (rs) was +0.9429 with a corresponding P value of 0.0167, suggesting a strong and significant positive monotonic relationship between the two variables.

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

the number of effector cells present in the spleen of each animal (Fig. 3c and d, left axis). This variability in the number of potential effector cells correlated with the magnitude of B cell activation measured in each animal replicate (Fig. 3d). This demonstrates, as expected, that the overall magnitude of T cell help measured by the FTA T helper assay is dependent on the number of effector CD4+ T cells. Based on these data it appears that the FTA T helper assay has the capacity to measure TH cell activity in vivo across a broad range of epitope concentrations in many replicates, and can be performed simultaneously with the FTA killing assay to measure CTL activity. The FTA T help assay is capable of detecting CD4 + T cell effector activity even when effector T cells are less than 1/5000 of the total CD4 + T cell pool, suggesting it is a highly sensitive assay. This assay also gives reproducible results, both within an animal and between animals, that correlate well with the number of antigenspecific CD4+ T effector cells generated in vivo. 3.2. Detecting TH responses in polyclonal T cell populations The immunization regimes used in the above examples utilized TCR-Tg T cells as effector cells. It was uncertain from these studies whether the FTA T helper assay could be used to detect CD4 + T cell responses generated from a polyclonal repertoire. This is particularly important given the potential use of this assay as a high throughput multiparameter screening tool for different vaccination regimes. To this end, the FTA T helper assay was assessed for its capacity to detect TH cell responses generated by potential vaccine regimes against HIV. Two vectors currently undergoing investigation for their efficacy in generating HIV-specific immunity are recombinant fowl pox virus (FPV) and recombinant Vaccinia virus (VV) constructs that express HIV epitopes (Ranasinghe et al., 2006). Here we generated immune responses in BALB/c mice using two such vaccine constructs and assessed T cell responses using an expanded FTA assay. The FTA was constructed using combinations of 7 fluorescence intensities of CFSE, 6 intensities of CTV and 6 intensities of CPD to label BALB/c splenocytes to generate a three dimensional array of 252 distinguishable cell clusters (Fig. 4a). For the purpose of peptide pulsing, this FTA was divided into 6 repeats (based on CPD fluorescence intensity) of 42 target clusters (based on CFSE and CTV fluorescence intensity) and these 42 clusters pulsed with titrations of 7 different MHC-binding peptides (detailed in the Methods section) (Fig. 4b). The Gag Th MHC-II BALB/c-specific peptide was included to measure TH cell response against a known CD4 + T cell HIV epitope (Mata and Paterson, 1999). In addition the following 6 MHC-I BALB/ c-specific peptides were included: HIV gag (the HIV gag immunodominant epitope), HIV gag mut (a variant of HIV gag that has E at position 7 altered to D that results in a lower avidity epitope), F2L (the immunodominant VV epitope), F2L mut (a lower avidity version of F2L), HIV pol (a HIV pol epitope) and HIV env (a HIV envelop epitope), to measure CTL activity. The pooled FTA cell clusters were transferred into syngeneic BALB/c host mice 6 days post vaccination with either FPV-HIV or VV-HIV injected intramuscularly (i.m.). To allow the FTA to be detected in syngeneic hosts, a 4th dye, (DiI), was used to label all target cells (see Supplementary Fig. 2 for gating strategy). After 18 h in vivo, FTA cells from

187

host spleens were harvested and assessed for B cell activation and cell death (example of histogram analysis depicted in Fig. 4c and d). Analysis of the capacity of the two viral constructs to initiate CD4 + T cell effector function against the Gag Th epitope showed VV-HIV vaccination induced measurable CD69 upregulation on B cells in a peptide-dose dependent manner across all 6 intra-animal replicates (Figs. 4c and 5a, left panel). Similarly B cells also upregulated the activation marker CD44 in a peptide-dose dependent manner across all 6 intra-animal replicates (Figs. 4c and 5a, right panel). We have also used CD62L down regulation with similar success in measuring B cell activation in analogous assays (data not shown). In contrast, the capacity of the FPV-HIV vaccination to induce CD4 + T cell effector function against the Gag Th epitope was very poor (Fig. 5a). Similarly, the capacity of the FPV-HIV vaccination to induce CTLmediated killing of all HIV and VV epitope pulsed targets within the FTA was also very poor (Fig. 5b, left panel). VV-HIV vaccination, however, induced strong CTL killing activity against the wild type VV epitope F2L, the F2L mutant epitope and the HIV gag wild type epitope (Figs. 4d and 5b, right panel). VV-HIV vaccination also appeared to elicit a low but detectable response to the lower avidity HIV gag mutant epitope, but generated a poor CTL response against the Pol and Env HIV epitopes (Fig. 5b, right panel). These results are well summarized by the cumulative AUC statistics of each data set (Fig. 5c). From these data it is clearly evident that the FTA T helper assay can be used to measure TH cell activity generated from a polyclonal repertoire of CD4 + T cells in vivo. This assay can also be performed simultaneously with an extensive FTA killing assay to obtain an overall assessment of T cell effector activity generated by different vaccination regimes. The FTA could thus be of considerable value for assessment of immunotherapies. 4. Discussion Here we report on a novel assay to measure the capacity of CD4 + T effector cells to help B cells in vivo using a multiplex flow cytometry FTA assay. This assay is an extension of the FTA assay we have used previously to monitor multiparameter CTL killing activity in vivo (Quah et al., 2012). The construction of these live cell FTAs has been made possible with the availability of the vital dyes CFSE, CTV and CPD, which have different fluorescence emission spectra with minimal spectral overlap and label cells with exceptionally low fluorescence variability and low cell toxicity (Quah et al., 2012). All three dyes can therefore be used to label cells with up to 6–7 different fluorescence intensities of each dye that, when used in combination, can generate >250 unique fluorescence signatures that can be discriminated in a single recipient animal by flow cytometry. Also we have shown here that cells within an FTA can be pulsed with a wide range of concentrations of MHC-I and MHC-II binding peptides and thereby act as targets for effector CD8+ and CD4+ T cells. Unlike other cell-based multiplex fluorescence assays (Krutzik and Nolan, 2006) the FTA is comprised of viable cells, allowing them to be administered in animals and measure T cell responses in situ in real time in vivo. We have found the FTA T helper assay to be highly sensitive, detecting activation of B cells via CD4 + TH cells

188

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

Fig. 4. Setup of a 252 FTA T helper and T killing assay. Splenocytes from BALB/c mice were labeled with combinations of CTV (0 nM, 350 nM, 1295 nM, 4792 nM, 17,729 nM and 65,595 nM), CFSE (0 nM, 79 nM, 315 nM, 1106 nM, 3859 nm, 13,505 nm and 47,269 nM) and CPD (0 nM, 106 nM, 690 nM, 2738 nM, 10,262 nM and 38,506 nM), and pulsed with peptides to generate a total of 252 target cell clusters. Target cells were also labeled with DiI for identification in host splenocytes. FTA cells were injected i.v. into three BALB/c host mice that had 6 days earlier been injected i.m. with either VV-HIV (comprised of a mixture of 2.5 × 106 PFU of VV-HIV (Gag/Pol) and 2.5 × 106 PFU of VV-HIV (Env)) or 5 × 106 PFU FPV-HIV to prime the animals. The third animal received no antigen and acted as a control (Naïve). 18 h after FTA injection, splenocytes were collected and target cells delineated from host splenocytes by DiI label using flow cytometry (SFig. 2). (a) 3D plots, 2D plots and histogram plots of the fluorescence intensities of FTA target cells from the naïve animal. (b) 2D plots of the fluorescence intensities of target cells from the naïve and VV-HIV-primed animals and the associated layout of peptide loading titrations mapped out on the basis of CTV and CFSE fluorescence. This layout was repeated 6 times, with each repeat being tagged with the 6 fluorescence intensities of CPD shown in a). Each target cell cluster was pulsed with either 1.65 μM to 400 μM, at 3-fold dilution steps, of the MHC-II peptides Gag Th, or 0.01 nM to 1000 nM, at 10-fold dilution steps, of each of the MHC-I peptides listed. One target cell cluster was also left “unpulsed” (Nil) and served as controls. (c) An example of histogram analysis of the FTA T helper assay, with B220+ FTA cells being assessed for CD69 and CD44 upregulation from the VV-HIV-primed animal compared to those from the naïve animal. (d) An example of histogram analysis of the FTA killing assay, with B220+ FTA cells being assessed for death from the VV-HIV primed animal compared to those from the naïve animal.

that represent b0.02% of the tissue CD4 + T cell pool. Furthermore, the T helper assay can be used in combination with a CTL killing assay in the same FTA to measure responses from a normal repertoire of polyclonal T cells responding to different vaccination regimes. As with the FTA killing assay, the robustness of the FTA T helper assay is dependent on the fidelity of the fluorescent signature of each cell cluster within the FTA (Quah et al., 2012). However, unlike the FTA killing assay, which essentially monitors the disappearance of cell clusters from the FTA as a readout, the FTA T helper assay relies on measuring, by flow cytometry, the upregulation of B cell activation markers, such as CD69 and CD44. Thus, an important consideration for the T helper assay is the high amount of spectral spillover that can occur from brightly labeled CFSE,

CTV and CPD cell clusters into spectrally adjacent fluorescence detectors used to measure B cell activation markers. It is important, therefore, that compensation of spectral spillover is performed correctly. The complexities of these spectral spillover issues can be minimized by choosing antibody-fluorochrome conjugates that emit at wavelengths that have minimal spectral overlap with the labeling dyes. This issue is also somewhat minimized by calculating the expression of each activation marker in FTAs from primed animals relative to those present in FTAs from naïve animals. Another variable that can have a profound influence on the robustness of the FTA T helper assay are fluctuations in the fluorescence intensity that is detected between samples and even during each sample acquisition by the flow cytometer. We have observed that occasionally fluorescence

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

189

Fig. 5. The FTA T helper assay allows the detection of B cell activation by TH cells generated from a polyclonal repertoire of CD4+ T cells. Flow cytometry analysis was used to assess the capacity of VV-HIV and FPV-HIV vaccine regimes to generate TH cells and CTL activity in the FTA assay experiment described in Fig. 4. (a) B cell activation in the FTA T helper assay, as measured by the GMFI of CD69 and CD44 expression on FTA B220+ cells from immunized mice above the GMFI of CD69 and CD44 expression on FTA B220+ cells from naïve mice. (b) % Specific killing of FTA cells pulsed with MHC-I binding peptides. (e) Area under curve values for T help and T killing assays depicted in a) and b). Panels show the mean and standard error of mean of 6 intra-animal replicates.

intensity detection can fluctuate even when the cytometer is maintained at a high standard. Although this is an infrequent occurrence and may be dependent on a variety of factors including sample preparation, previous use of the machine and previous maintenance, it is nonetheless an important consideration and should be monitored as a quality control measure. This can be achieved by monitoring the mean fluorescence intensity of constant parameters (for example, CFSE, CTV and CPD or inclusion of 8 channel rainbow beads

can be used for this purpose) during each sample acquisition to assess for fluorescence fluctuations over time. With these precautions noted, we have found the FTA assay to be quite robust, giving consistent results over time and between animal replicates. This coupled with its high throughput multiplex nature and versatility in simultaneously measuring CD8 + and CD4 + T cell effector activity, make it a promising screening tool for in vivo effector T cell activity especially when evaluating immune responses to infection or

190

B.J.C. Quah et al. / Journal of Immunological Methods 387 (2013) 181–190

vaccination. Indeed, currently there are very few methods that can monitor T cell effector function in vivo, particularly in the case of CD4 + T cell effector function. The techniques of ELISPOT (Czerkinsky et al., 1983; Klinman, 2008) and intracellular cytokine staining (ICS) (Prussin and Metcalfe, 1995; Foster et al., 2007) by flow cytometry are two commonly used techniques to measure CD4 + T cell effector function. They both have the advantage of being capable of measuring cytokine production at a single cell level and ICS in particular can be used as a multiparametric assay for assessing production of several cytokines at once. However, these techniques normally require ex vivo stimulation of cells to induce cytokine production and, in the case of ICS, the use of secretion inhibitors to trap cytokines intracellularly is essential (Foster et al., 2007). It is, therefore, unclear how well these assays relate to actual cytokine secretion during immune responses in vivo. Therefore, the FTA T helper assay offers an advantage in that it is a relatively easy assay to assess CD4 + T cell effector function across a broad range of antigen concentrations in real time in vivo. Furthermore, it assesses an aspect of CD4 + T cell effector function that is often overlooked, namely the capacity of TH cells to induce B cell activation. It should also be noted that the FTA T helper assay is not mutually exclusive to the ICS assay and these techniques could be used in combination. Indeed, we have shown here how the phenotype of TH cells can be monitored (in this case measurement of their proliferation by CFSE dilution) at the same time as monitoring their capacity to activate B cells via the FTA T helper assay. Ultimately, it will be important to ascertain how these assays relate to actual immune response outcome and to assess their utility as correlates of successful immunity. This is, of course, an ongoing question for many immunological assays. Finally, while the FTA T helper assay has been used here to monitor the magnitude of TH cell responses, it could also be used to detect T–B cell interaction defects, such as those responsible for aberrant germinal center formation in SLAM-associated protein (SAP) mutant mice (Kageyama et al., 2012). It is also possible that the FTA technology outlined here could be modified to benefit many other areas of biology (Krutzik and Nolan, 2006; Gilbert et al., 2009).

Acknowledgments This work was supported by a Project grant to BQ and CP from the National Health and Medical Research Council (NHMRC) of Australia. We wish to thank Harpreet Vohra and Michael Devoy for their excellent maintenance of the JCSMR FACS laboratory, and Kerong Zhang and Cameron McCrae of the Australian Cancer Research Foundation Biomolecular Resource Facility, JCSMR, ANU, for the construction of MHC-I tetramers.

Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jim.2012.10.013.

References Barnden, M.J., Allison, J., Heath, W.R., Carbone, F.R., 1998. Defective TCR expression in transgenic mice constructed using cDNA-based alpha- and beta-chain genes under the control of heterologous regulatory elements. Immunol. Cell Biol. 76 (1), 34. Czerkinsky, C.C., Nilsson, L.A., Nygren, H., Ouchterlony, O., Tarkowski, A., 1983. A solid-phase enzyme-linked immunospot (elispot) assay for enumeration of specific antibody-secreting cells. J. Immunol. Methods 65 (1–2), 109. Earl, P.L., Cotter, C., Moss, B., VanCott, T., Currier, J., Eller, L.A., McCutchan, F., Birx, D.L., Michael, N.L., Marovich, M.A., Robb, M., Cox, J.H., 2009. Design and evaluation of multi-gene, multi-clade HIV-1 MVA vaccines. Vaccine 27 (42), 5885. Fazilleau, N., Mark, L., McHeyzer-Williams, L.J., McHeyzer-Williams, M.G., 2009. Follicular helper T cells: lineage and location. Immunity 30 (3), 324. Foster, B., Prussin, C., Liu, F., Whitmire, J.K., Whitton, J.L., 2007. Detection of intracellular cytokines by flow cytometry. (Chapter 6) In: Coligan, John E., et al. (Ed.), Current Protocols in Immunology (Unit 6 24). Gilbert, D.F., Wilson, J.C., Nink, V., Lynch, J.W., Osborne, G.W., 2009. Multiplexed labeling of viable cells for high-throughput analysis of glycine receptor function using flow cytometry. Cytometry. Part A 75 (5), 440. Hogquist, K.A., Jameson, S.C., Heath, W.R., Howard, J.L., Bevan, M.J., Carbone, F.R., 1994. T cell receptor antagonist peptides induce positive selection. Cell 76 (1), 17. Kageyama, R., Cannons, J.L., Zhao, F., Yusuf, I., Lao, C., Locci, M., Schwartzberg, P.L., Crotty, S., 2012. The receptor ly108 functions as a sap adaptordependent on-off switch for T cell help to B cells and NKT cell development. Immunity 36 (6), 986. Klinman, D., 2008. Elispot assay to detect cytokine-secreting murine and human cells. (Chapter 6) In: Coligan, John E. (Ed.), Current protocols in immunology (Unit 6 19). Krutzik, P.O., Nolan, G.P., 2006. Fluorescent cell bar coding in flow cytometry allows high-throughput drug screening and signaling profiling. Nat. Methods 3 (5), 361. Mata, M., Paterson, Y., 1999. Th1 T cell responses to HIV-1 gag protein delivered by a Listeria monocytogenes vaccine are similar to those induced by endogenous listerial antigens. J. Immunol. 163 (3), 1449. Nepom, G.T., 2012. MHC class II tetramers. J. Immunol. 188 (6), 2477. Prussin, C., Metcalfe, D.D., 1995. Detection of intracytoplasmic cytokine using flow cytometry and directly conjugated anti-cytokine antibodies. J. Immunol. Methods 188 (1), 117. Quah, B.J., Parish, C.R., 2010. The use of carboxyfluorescein diacetate succinimidyl ester (CFSE) to monitor lymphocyte proliferation. J. Vis. Exp.: JoVE 44. Quah, B.J., Parish, C.R., 2012. New and improved methods for measuring lymphocyte proliferation in vitro and in vivo using CFSE-like fluorescent dyes. J. Immunol. Methods 379 (1–2), 1. Quah, B., Ni, K., O'Neill, H.C., 2004. In vitro hematopoiesis produces a distinct class of immature dendritic cells from spleen progenitors with limited T cell stimulation capacity. Int. Immunol. 16 (4), 567. Quah, B.J., Wijesundara, D.K., Ranasinghe, C., Parish, C.R., 2012. Fluorescent target array killing assay: a multiplex cytotoxic T-cell assay to measure detailed T-cell antigen specificity and avidity in vivo. Cytometry. Part A 81 (8), 679. Ranasinghe, C., Medveczky, J.C., Woltring, D., Gao, K., Thomson, S., Coupar, B.E., Boyle, D.B., Ramsay, A.J., Ramshaw, I.A., 2006. Evaluation of fowlpox-vaccinia virus prime-boost vaccine strategies for high-level mucosal and systemic immunity against HIV-1. Vaccine 24 (31–32), 5881. Rappuoli, R., 2007. Bridging the knowledge gaps in vaccine design. Nat. Biotechnol. 25 (12), 1361. Restifo, N.P., Bacik, I., Irvine, K.R., Yewdell, J.W., McCabe, B.J., Anderson, R.W., Eisenlohr, L.C., Rosenberg, S.A., Bennink, J.R., 1995. Antigen processing in vivo and the elicitation of primary CTL responses. J. Immunol. 154 (9), 4414. Sakaguchi, S., Yamaguchi, T., Nomura, T., Ono, M., 2008. Regulatory T cells and immune tolerance. Cell 133 (5), 775. Takeshita, T., Takahashi, H., Kozlowski, S., Ahlers, J.D., Pendleton, C.D., Moore, R.L., Nakagawa, Y., Yokomuro, K., Fox, B.S., Margulies, D.H., et al., 1995. Molecular analysis of the same HIV peptide functionally binding to both a class I and a class II MHC molecule. J. Immunol. 154 (4), 1973. Zhu, J., Yamane, H., Paul, W.E., 2010. Differentiation of effector CD4 T cell populations (*). Annu. Rev. Immunol. 28, 445.