Introduction

Efforts to monitor cellular immune responses to disease or vaccination often rely on the measurement of antigen-specific T cells. A variety of methods is available to identify antigen-specific CD4⁺ T-cell responses¹; however, these methods have limitations, such as an inability to assess the phenotype of responding cells (ELISA, ELISPOT) or the need for lethal fixation and permeabilization (intracellular cytokine staining)¹. Peptide–major histocompatibility complex class II multimers may also be used to identify antigen-specific CD4⁺ T-cell responses, but few multimers are available, knowledge of the subject's major histocompatibility complex haplotype is required and the method does not necessarily identify functionally responsive cells¹. For these reasons, new methods that overcome these limitations, and that can replace the measurement of multiple functional responses, are needed.

Recently activated CD4⁺ T cells express CD154 (CD40L), which provides costimulatory signals to activate B cells^{2, 3} and antigen-presenting cells (APCs)⁴; resting CD4⁺ T cells do not express CD154. Therefore, de novo-synthesized CD154 can identify antigen-specific CD4⁺ T cells in a stimulation assay. Previous assays for CD154 were limited by the highly transient nature of CD154 (refs. 5,6), which is rapidly internalized (and degraded) after surface expression⁶ or is secreted⁷. We overcame this limitation by including fluorescently conjugated antibody to CD154 (anti-CD154) and monensin in the culture during the stimulation¹. Using this method, cells that synthesize CD154 at any time during stimulation are subject to labeling and, once labeled, these cells retain their fluorescence.

In principle, as CD154 molecules are expressed on the cell surface, they complex with fluorescently labeled antibodies in the culture. In rhesus macaque samples, these complexes remain stable on most antigen-responsive cells for approximately 8 h poststimulation⁸; however, for human and mouse cells, CD154-antibody complexes become unstable unless monensin is present, such that less than half of the antigen-responsive cells are identified (see ANTICIPATED RESULTS). The precise mechanisms by which monensin acts are not known. Monensin may preserve internalized CD154-antibody complexes by inhibiting the acidification of endosomes⁹; however, even protease- or acid-resistant conjugates of CD154 (such as Alexa 594) require monensin. Therefore, it is possible that monensin prevents degradation of the CD154 antibody itself or that it blocks the secretion of CD154 (ref. 6). In any case, the monensin requirement precludes simultaneous use of this technique with methods that capture antigen-specific cells on the basis of cytokine secretion (i.e., cytokine capture assays). In some settings, however, nearly all cytokine-producing cells express CD154 (see ANTICIPATED RESULTS), thereby obviating the need for cytokine capture assays. In addition, once samples assayed by this method are removed from medium containing monensin, they secrete cytokines and proliferate normally. Thus, assays that cannot be used simultaneously may be used sequentially.

It is important to note that the method described here is not lethal to cells, because fixation and permeabilization are not required. Thus, antigen-specific T cells can be purified for cloning, further culture or RNA expression evaluation. In addition, this method may be used for stimulation times as long as 24 h, making it useful when highly synchronized responses are not present (such as assays requiring viral delivery of antigen to stimulate CD4⁺ T cells). The assay is compatible with fresh, cryopreserved, restimulated or cloned cells and has been tested on human^{1, 10}, mouse¹⁰ and rhesus macaque⁸ samples.

The assay is also compatible with intracellular cytokine staining¹, although it cannot be used with brefeldin A, which completely blocks surface expression of CD154. Using the assay to characterize CD4 responses, we found that, for the total response elicited by superantigen, cells that expressed tumor necrosis factor (TNF)-, interleukin (IL)-2 or interferon (IFN)- were predominantly CD154⁺ (ref. 1). Furthermore, CD154 was also found on a proportion of cells that expressed none of the above cytokines, suggesting that CD154 identifies antigen-responsive T cells that have other effector functions¹. For vaccine- or pathogen-specific CD4⁺ T cells, we found substantial heterogeneity in the expression profiles of CD154 and cytokines¹. This assay may provide additional information about antigen-elicited CD4 function by quantifying cells that are capable of stimulating APCs through CD40. Most importantly, the assay provides a simple means to isolate viable antigen-specific CD4⁺ T cells.

Materials

Reagents

• Fresh or cryopreserved cells (or cell lines) derived from humans, mice or rhesus macaques
• PBS, without calcium or magnesium (Bio-Whittaker, cat. no. 17-512F)
• DMSO, anhydrous >99% (Sigma, cat. no. 276855)
• Anti-CD154 phycoerythrin (TRAP1 clone; BD-Pharmingen, cat. no. 555700); see REAGENT SETUP
• Monensin (Golgi-stop; BD-Pharmingen, cat. no. 554724); see REAGENT SETUP
  Caution Monensin is a known toxin and highly flammable. Avoid contact with skin, eyes or mucous membranes; keep away from sources of ignition.
  Critical This assay is only compatible with monensin; it cannot be used with brefeldin A.
• Staphylococcus enterotoxin B (SEB) solution: dilute SEB (Sigma, cat. no. S4881) to 0.1 mg ml^{- 1} in PBS or DMSO, store at - 30 °C in 100-l aliquots (each aliquot contains 10 g)
  Caution Toxic; avoid inhalation, ingestion or contact with skin, eyes or mucous membranes.
• Peptide or whole antigen (as determined by investigator): usually diluted to a concentration of 200 g ml^{- 1} in DMSO or dH₂O
  Critical If DMSO is used as a diluent, prepare peptide stock so that the final concentration of DMSO in each culture well does not exceed 1%.
• Costimulatory antibodies: anti-CD28 (BD-Pharmingen, cat. no. 555725) and anti-CD49d (BD-Pharmingen, cat. no. 555501); see REAGENT SETUP
• Culture medium: RPMI 1640 medium, supplemented with 10% heat-inactivated newborn calf serum (HINCS), 100 U ml^{- 1} penicillin/streptomycin and 2 mM L-glutamine (GIBCO/BRL)
• 1 Experimental medium (for experiments with <10 wells only): culture medium supplemented with 0.66 l ml^{- 1} (2 M) monensin
  Critical Prepare on the day of the experiment.
• 2 Experimental medium (for experiments with 10 wells only): culture medium supplemented with 1.33 l ml^{- 1} (4 M) monensin
  Critical Prepare on the day of the experiment.
• 2 SEB medium (for experiments with 10 wells only): for 1 ml, add 20 l SEB solution to 980 l 2 experimental medium; scale volumes as necessary
  Critical Prepare on the day of the experiment.
• 2 Peptide-costimulation medium (for experiments with 10 wells only): add an appropriate concentration of each peptide (for 1 ml of medium, usually 2 g (1.2 M), but investigators should determine the ideal concentration of peptide (or whole antigen) for their experimental systems) to 2 experimental medium, along with 2 g each of anti-CD28 and anti-CD49d per ml; scale volumes as necessary
  Critical Prepare on the day of the experiment.
• 2 Costimulation-only medium (for experiments with 10 wells only): for 1 ml, add 2 g anti-CD28 and 2 g anti-CD49d to 2 experimental medium, along with the volume of peptide diluent used in 2 peptide-costimulation medium; scale volumes as necessary
  Critical Prepare on the day of the experiment.
• Staining medium: see REAGENT SETUP
• Fluorescently labeled antibodies appropriate for flow cytometry (such as antibodies to CD4 and CD3 as well as a marker for cell viability)

Equipment

• 96-well plates
• Flow cytometer

Reagent Setup

• Anti-CD154 phycoerythrin All investigators should titrate CD154 antibodies under the assay conditions described here before experimental use; we recommend starting the titration at 20 l reagent per 200 l culture, and using twofold dilutions down to 0.3 l reagent per 200 l culture. The optimal concentration is that which gives the brightest specific (CD154⁺) signal, with the dimmest background (CD154^- ) signal.
  Critical For experiments on human cells, we compared three commercially available conjugates and found that the phycoerythrin-labeled TRAP1 clone from Pharmingen gave the best signal. Investigators working with animal samples should determine which commercially available conjugates provide the strongest signal in their assay systems. (Mouse samples have been assayed successfully with the TRAP1 clone.)
• Costimulatory antibodies: anti-CD28 and anti-CD49d Antibodies to costimulatory molecules are commonly used to enhance cytokine production by T cells during short (<12 h) stimulations with peptide antigens. A number of factors influence the degree to which costimulation helps, including the number and type of APCs in the sample, duration of stimulation, strength of the response, properties of the antigen and which costimulatory receptors are targeted. Investigators may wish to determine whether the method of costimulation presented here is best for their particular assay systems. The use of costimulatory antibodies is not likely to affect results adversely, so we recommend them for any peptide stimulations performed with this protocol.
• Monensin We used the amount of monensin (2 M final concentration) recommended by the manufacturer; however, we have found that the assay can work with slightly lower concentrations of monensin. Investigators may wish to determine the optimal level of monensin needed in their systems.
• Staining medium RPMI 1640 medium, supplemented with 4% HINCS, HEPES buffer (25 mM) and sodium chloride (113 mM) and lacking biotin, L-glutamine, phenol red, riboflavin and sodium bicarbonate (GIBCO/BRL, custom order). Note that the authors use this staining medium for most of their immunofluorescence staining protocols. Individual laboratories may substitute any staining medium that they commonly use.

Procedure

1. Points from here (point 1) up to and including point 3 are related to Preparation of reagents and platesTiming: 1 hLabel a separate 96-well plate for each stimulation condition.
2. Determine the total number of wells that will be plated by taking into consideration the number of samples, the stimulation conditions and controls necessary (Table 1), and the maximum number of cells that can be cultured in one well (2 10⁶ cells). When more than 2 10⁶ cells are required for a particular stimulation condition, split cells across multiple wells.

   Table 1: Summary of the experimental conditions necessary for the assay.

   

   
   Full table
   
   
   Critical step Include a matched costimulation-only control for each donor's stimulated sample. SEB controls are not required for each peptide or donor, but run these controls for at least one sample in every experiment.
3. The cells should be prepared by one of two ways, for a small (<10 wells) experiment (A) or for a large (10 wells) experiment (B).
   1. Small (<10 wells) experimentTiming: 30 min
      1. To determine the number of cells needed, multiply the number of wells to be plated by 2 10⁶.Troubleshooting
      2. Resuspend this number of cells to a concentration of 10 10⁶ cells per ml in 1 experimental medium.
      3. Aliquot 200 l of cell suspension to each well.
      4. For peptide-costimulation wells, 0.2 g peptide is usually added per well. Investigators should determine the optimal amount of peptide for their experimental systems. Also, add 0.2 g each of costimulatory antibodies anti-CD28 and anti-CD49d to each of these wells.
      5. For costimulation-only wells (i.e., negative controls), add 0.2 g each of anti-CD28 and anti-CD49d. Also add the same volume of peptide diluent (DMSO or dH₂O) as used for the peptide-costimulation wells.Troubleshooting
      6. Add appropriate volume of anti-CD154–phycoerythrin (as determined by preliminary titration experiments; see REAGENT SETUP) to every well.Troubleshooting
         Critical step Anti-CD154 must be titrated under the conditions described in this protocol, because the amount necessary for this assay may differ considerably from what is used in standard surface staining.
      7. For SEB wells, add 2 l SEB solution.
         Critical step SEB should be added last, using a pipettor dedicated to SEB use only. Perform this step only after other plates have been put into incubator.Troubleshooting
   2. Large (10 wells) experimentTiming: 1 h
      1. The various 2 media (see MATERIALS) are used for large experiments. Calculate the total volume needed for each medium tube by multiplying the number of wells for each condition by 100 l. Set aside the appropriate volume of each 2 medium.
      2. To determine the number of cells needed, multiply the number of wells to be plated by 2 10⁶.
      3. Resuspend this number of cells to a concentration of 20 10⁶ cells per ml culture medium.
      4. Aliquot 100 l of cell suspension to each well.
      5. Add appropriate volume of anti-CD154 phycoerythrin (as determined by preliminary titration experiments; see REAGENT SETUP) to every well.Troubleshooting
         Critical step Anti-CD154 must be titrated under the conditions described in this protocol, because the amount necessary for this assay may differ considerably from what is used in standard surface staining.
      6. To each well, add 100 l of the 2 medium appropriate for that particular stimulation condition. See MATERIALS and Table 1.Troubleshooting
         Critical step 2 SEB medium should be added last, using a pipettor dedicated to SEB use only. Perform this step only after other plates have been put into incubator.
4. Points from here (point 4) up to and including point 4 are related to StimulationTiming: 6–24 hMove plates to 37 °C, 5% CO₂ incubator (Fig. 1).

   Figure 1: Optimization of CD154 detection.

   

   (a) Time course of SEB-induced CD154 expression (red) and background fluorescence from unstimulated cultures treated with monensin (blue), as measured by our method. Background fluorescence (unstimulated, monensin-exposed cultures) increased during the first 2 h of coculture but subsequently stabilized. (b) Kinetics of CD154 expression within CD4⁺ T cells, as measured by our method. Peak expression of CD154 was detected after 6 h of stimulation; CD154 expression was detected in 15% of all CD4⁺ T cells. Maximal amounts of CD154 were still observed after 24 h of stimulation. (c) CD154 expression as measured by surface and intracellular (IC) methods and by the method described here (- or + monensin (Mon)). Surface and IC methods identified CD154 expression in only 3% and 6%, respectively, of CD4⁺ T cells; in the same sample, our assay method identified 30% of the CD4⁺ T cells when monensin was included. When monensin is not included in the culture, fewer antigen-responsive CD4⁺ T cells are detected. Data are representative of two to four experiments. Modified, with permission, from ref. 1.

   Full size image (76 KB)
   
   Pause Point Plates must incubate at least 6 h to observe maximal CD154 staining. Plates may incubate for as long as 24 h, with no appreciable loss in CD154 signal. When longer stimulations are necessary, investigators should determine time course of CD154 expression in their experimental systems.
5. Points from here (point 5) up to and including point 10 are related to StainingTiming: 1–1.5 hAfter stimulation, remove plates from incubator and transfer contents of each well to an appropriately labeled test tube.
6. Add 1 ml staining medium to each tube.
7. Centrifuge tubes at 800g for 5 min, at approximately 25 °C.
8. Aspirate fluid and resuspend each pellet in 50 l staining medium.
9. Investigators may now prepare a cocktail of fluorescently labeled antibodies appropriate for their experimental systems and flow cytometry equipment. At a minimum, CD154 signals should be measured in cells stained with antibodies to CD4; however, the best results are obtained when cells are also stained with anti-CD3 and a marker of cell viability.
10. Stain cells and analyze by flow cytometry as per the individual laboratory's protocol.

Timing

It takes 30 min to prepare the plates. Plating requires 30 min (<10 wells) or 1 h (10 wells). Stimulation and staining require 6–24 h and 1–1.5 h, respectively.

Troubleshooting

Troubleshooting advice can be found in Table 2

Table 2: Troubleshooting table.

Full table

Anticipated results

A critical component of this assay is the measurement of background CD154 fluorescence in matched unstimulated controls treated with monensin. Through the first 2 h of monensin exposure, background levels of CD154 staining increase dramatically in CD4⁺ T cells, but background fluorescence stabilizes after this time¹ (Fig. 1a, blue dots). Between 5 and 8 h, peak amounts of CD154 (well above the amount observed in monensin-only samples) are observed in the SEB-stimulated samples (Fig. 1a, red dots), and these levels persist for more than 24 h (ref. 1) (Fig. 1b). To determine the proportion of CD154⁺ cells in a stimulated sample, investigators should first superimpose the histograms from the stimulated and unstimulated samples (cultured for the same incubation period), and then apply a gate encompassing only those events from the stimulated sample that do not overlap with the unstimulated sample (Fig. 1a, black box). In general, we observed similar kinetics for human cells stimulated with peptide or whole antigen¹, but investigators may still wish to perform preliminary studies to determine the kinetics of CD154 expression in their own experimental systems.

When compared with other methods of measuring CD154 expression, such as surface or intracellular staining, this assay detects a higher proportion of CD154-expressing cells¹, especially when monensin is included in the culture (Fig. 1c). In addition, the CD154 signal that is observed using this method is consistently brighter and lasts longer than that observed with standard surface or intracellular staining¹ (Fig. 1c). Additionally, the assay is compatible with intracellular cytokine staining techniques and may be a surrogate for the measurement of multiple diverse cytokines. Specifically, in SEB-stimulated cells, the vast majority of cells that produce any combination of IFN-, TNF- and IL-2 also expressed CD154 (ref. 1) (Fig. 2). Thus, CD154 identifies most of the CD4⁺ T cells that respond to antigen, regardless of which cytokines they elucidate. The assay also identifies cells that express CD154 but lack expression of the cytokines measured. These cells are likely to have other effector functions¹.

Figure 2: Relationship between CD154 and cytokine expression.

(a) Gating strategy used to identify lymphocytes, based on forward-scatter (FS) and side-scatter (SS) properties, and CD4⁺ T cells, based on coexpression of CD3 and CD4. Note that those cells that have downregulated CD3 expression (as a result of SEB stimulation) are included in the analysis. (b) Measurement of CD154 by our method is compatible with intracellular cytokine staining for TNF-, IL-2 and IFN-. (c) CD3 and CD154 expression within the total CD4⁺ T cell population (left), cells producing at least one of the cytokines measured (middle) and cells producing none of the cytokines measured (right). In this sample, approximately 11% of CD3⁺ CD4⁺ T cells expressed CD154 after SEB stimulation. Cells expressing at least one cytokine were mostly (70.6%) CD154⁺, whereas cells expressing none of the measured cytokines were mostly (but not exclusively) CD154^- . Data are representative of ten experiments. Data first published in ref. 1.

Full size image (78 KB)

There are conditions where expression of CD154 and cytokines is not tightly coupled. In our study of HIV-negative individuals, who were assayed soon after vaccination with a particular HIV subunit vaccine, a large proportion of CD4⁺ T cells that were specific for the vaccine epitope expressed IL-2 without expressing CD154 (ref. 1). Similarly, in HIV-positive individuals, reduced expression of CD154 parallels immunodeficiency^{11, 12, 13}, and a preponderance of CD154^- cytokine-producing cells is observed¹⁴. In these settings, CD154 expression may not be a surrogate marker for cytokine production, but its measurement is still likely to provide biologically important information about the ability of certain CD4⁺ T cells to stimulate APCs.

In human and mouse cells, optimal CD154 detection requires monensin. In settings where monensin is not compatible with an investigator's experimental system, however, alternative methods for CD154 detection are available. For example, CD154 expression may be assayed by standard surface staining, although this significantly underestimates the proportion of responding cells³. Intracellular staining of CD154 captures more antigen-responsive cells; however, intracellular assays are lethal to cells. Addition of anti-CD40 to the culture prevents the internalization of CD154 and seems to identify most antigen-responsive CD4⁺ T cells¹⁰ without killing them, but this technique could potentially dysregulate antigen presentation.

We found that the effects of monensin exposure are transient, suggesting that assays that are incompatible with monensin could be used after cells are isolated using the method described here. For example, we found that cells assayed by this method could be sorted, and their ability to secrete multiple cytokines was maintained for over 30 h¹. Additionally, the assay conditions did not affect the proliferative capacity of these cells over long-term culture¹.