Aug
17
Written by:
Emily Sherman
8/17/2010 7:53 AM
If you have an autoimmune, infectious, or other disease, researchers need your help. Donate plasma and receive $400/donation for your time. Visit www.idonateplasma.com for details.
|
Source PDF with Charts: http://www.seracare.com/Portals/0/Papers/PBMC%20Functionality%20Article.pdf
CD4+ and CD8+ cells in cryopreserved human PBMC maintain full functionality in cytokine ELISPOT assays$
Christian R. Krehera,b, Markus T. Dittricha, Robert Guerkova, Bernhard O. Boehmb, Magdalena Tary-Lehmanna,*
a Department of Pathology, School of Medicine, Case Western Reserve University, BRB 928, 10900 Euclid Avenue, Cleveland, OH 44106, USA
b Section of Endocrinology, University Hospital of Ulm, Ulm 89081, Germany
Received 25 February 2002; received in revised form 28 April 2003; accepted 28 April 2003
Abstract
The frequency and the cytokine signature of antigen-specific T cells in the blood reflect the magnitude and the quality of T cell immunity in vivo. Recently, cytokine enzyme-linked immunospot (ELISPOT) assays performed on freshly isolated peripheral blood mononuclear cells (PBMC) emerged as a promising tool for monitoring these key parameters, providing direct feedback information on the efficacy of vaccinations and immune therapies. However, performing ELISPOT assays with freshly isolated cells is not readily feasible in the context of clinical trials. The ability to obtain valid ELISPOT data on cryopreserved samples would greatly enhance ex vivo immune monitoring capabilities. We have therefore systematically studied antigen-specific T cell responses in freshly isolated PBMC and after cryopreservation. Four healthy donors were selected that displayed T cell responses to six recall antigens. The antigen reactive T cells were defined as CD4 or CD8 cells, and their cytokine effector class was established measuring interferon (IFN)-g, interleukin (IL)-2, IL-4 and IL-5. The donors were bled at three different time points, and their PBMC were tested fresh and after freeze-thawing. The results showed that the frequencies and type 1/type 2 cytokine signatures of recall antigen-specific CD4 and CD8 cells are unaffected after cryopreservation. In contrast to these data obtained on human PBMC, cryopreservation of murine spleen cells causes a decrease in cytokine secretion.
1. Introduction
Enzyme-linked immunospot (ELISPOT) assays have recently emerged as a primary tool for monitoring antigen-specific T cell immunity directly ex vivo (McMichael and O’Callaghan, 1998; Whiteside, 2000; Barouch and Letvin, 2001). Operating at single- cell resolution, these assays directly visualize the cytokine production of individual antigen-specific T cells even if they occur at frequencies < 1:10,000, below the detection limit of flow cytometric approaches (Helms et al., 2000; Hesse et al., 2001).
With the exception of acute viral infections and allergies, the antigen-specific T cells in freshly isolated peripheral blood mononuclear cells (PBMC) frequently occur in this low frequency range (Helms et al., 2000). When PBMC are tested in ex vivo ELISPOT assays, the recall antigen activates the specific T cells to produce cytokines, and the fingerprint of this cytokine response is recorded. The assays’ 24–48 h duration is too short, however, to permit the antigenspecific T cells to engage in clonal expansion in vitro or to redifferentiate, which requires cell divisions and the opening of the chromatin structure (Agarwal and Rao, 1998). Therefore, such assays determine the frequency and cytokine signature of the antigen-specific T cell pool in vivo, recording two cardinal features of T cell-mediated immunity: the clonal sizes and the cytokine effector class. Obtaining information on these parameters is crucial for the comprehensive assessment of any immunotherapy. In the case of vaccinations, it is of interest, whether Th1- or Th2- type immunity has been engaged, and what the magnitude of the respective effector cell class is. In the case of therapies aiming at inactivating immune function (e.g. treating allergies, autoimmunity or transplant rejection), the above parameters provide insights as to whether the antigen-specific T cells have been ablated or deviated towards a different effector class.
In clinical trials, PBMC samples frequently have to be studied longitudinally, with the blood drawn at several time points in order to assess the efficacy of the treatment. Ideally, to minimize inter-assay variations, these samples should be tested side by side, in a single assay. This can be accomplished only with frozen samples. Additionally, clinical studies frequently involve multi-center trials. Therefore, there is either the need to establish a testing facility at each site or to ship samples to a core laboratory for immediate testing. The associated costs of such studies would be reduced substantially if frozen samples could be used. Classic ex vivo T cell monitoring tools such as proliferation assays have largely failed using freeze-thawed samples (Weinberg et al., 1998). One possible reason is that in these assays thawed T cells need to establish themselves in tissue culture and engage in several cycles of proliferation (Tary-Lehmann and Saxon, 1992). ELISPOT assays in contrast, measure a more limited function of the T cell: the short-term induction of cytokine production. Finally, with relatively little cell material and labor involved, ELISPOT assays would ideally lend themselves for high throughput analysis (Heeger et al., 2000) in core laboratories if they could be performed with cryopreserved PBMC. The experiments reported here were designed to establish whether ELISPOT assays performed with frozen PBMC accurately reflect the clonal sizes and the cytokine signatures of the antigen- specific CD4+ and CD8+ T cells in freshly isolated PBMC.
2. Materials and methods
2.1. Antigens, human PBMC donors, and mice HLA-A2 restricted peptides of Epstein–Barr virus (EBV) (EBV early lytic protein, peptide 1280–1288, GLCTLVAML) and cytomegalovirus (CMV) (pp65, peptide 495–503, NLVPMVATV) (Currier et al., 2002); as well as the H-2Kb restricted ovalbumin (OVA)-peptide (OVA: 257–264, SIINFEKL) (Fremont et al., 1995) were purchased from Princeton Biomolecules (Langhorne, PA). OVA was purchased from Sigma (St. Louis, MO) mumps antigen was from BioWhittaker (Walkersville, MA), Standardized Grass Mix Allergen was from Center Laboratories (Port Washington, NY) and Standardized Dust Mite Mix Allergen from ALK Abello (Round Rock, TX). CMV antigen was a gift of Michael Boeckh (Fred Hutchinson Cancer Research Center, Seattle, WA). Healthy human donors 23–45 years of age, males and females were screened for HLA-A2 positivity at the University Hospitals of Cleveland, Tissue Typing Facility. Three A2 positive and one A2 negative donor were selected. Heparinized peripheral blood was obtained by venipuncture and PBMC were isolated by standard Ficoll-Hypaque density gradient centrifugation. Within 5 h after venipuncture, PBMC were either tested in ELISPOT assays or frozen.
C57BL/6 mice were purchased from The Jackson Laboratories (Bar Harbor, ME) and maintained at the animal facility of Case Western Reserve University under specific pathogen-free conditions. The immunization of mice was done by i.p. injection of 200 Al of antigen at a final concentration of 500 Ag/ml (OVA) or 1 mg/ml (SIINFEKL) emulsified in adjuvant (Chu et al., 1997). The adjuvant was prepared by mixing 150 Ag of the CpG oligodeoxynucleotide TCCATGACGTTCCTGACGT (Princeton) per ml IFA (Gibco BRL, Grand Island, NY). All treatments of mice complied with institutional guidelines and human studies were performed with the approval of the Institutional Review Board of Human Studies at the University Hospitals of Cleveland.
2.2. Cell separations and freeze-thawing CD4+ and CD8+ T cells were isolated from PBMC by negative selection using RosetteSep enrichment cocktails (Stemcell Technologies, Vancouver, BC, Canada). The T Cell Depletion Cocktail (RosetteSep) was used to obtain antigen presenting cells (APC). The cell fractions isolated were of >96% purity.
For freezing, PBMC were suspended at a concentration of 2_107/ml in 2 ml of freezing medium A (60% FCS, 40% RPMI) at room temperature. An equal volume of freezing medium B (20% DMSO, 80% FCS), also at room temperature, was added drop wise, while gently mixing by shaking the tube. The resulting 4 ml cell suspension was pipetted in 1.5 ml aliquots into 1.8 ml cryovials (Greiner Labortechnik, Frickenhausen, Germany). The tubes were placed into a pre-chilled (4 jC) Nalgene Cryogenic Controlled- Rate Freezing Container (Fisher Scientific, Hanover Park, IL) that was placed into a _80 jC freezer. After 24 h, the samples were transferred to a liquid nitrogen tank for indefinite storage until testing. For thawing, cryotubes were placed in a 37 jC water bath and as soon as the samples were completely thawed they were pipetted into a 15-ml tube containing a 2-fold amount of complete RPMI medium (93% RPMI- 1640, 5% heat-inactivated AB serum, 1% L-glutamine, 1% Penicillin–Streptomycin) at room temperature. The cells were washed two times at room temperature. The cell recovery and viability were determined by acridine orange and ethidium bromide (Becton Dickenson, San Jose, CA) staining.
2.3. ELISPOT assays Human cytokine ELISPOT assays were performed as described previously (Helms et al., 2000). Briefly, ImmunoSpot plates (Cellular Technology, Cleveland, OH) were coated overnight at 4 jC with the cytokine- specific capture antibody in phosphate buffered saline (PBS) at concentrations specified below. The plates were blocked with bovine serum albumin (BSA) (10 g/l in PBS: PBS–BSA) for 1 h and washed 3_ with PBS. PBMC were plated in complete RPMI medium at 3_105 cells per well unless specified differently in the Section 3. Antigens were added to final concentrations that we have previously established optimal for T cell stimulation in ELISPOT assays: EBV and CMV peptides at 20 Ag/ml; and Dust Mite Mix and Grass Mix at 100 AU/ml.
Mumps antigen was used at a final dilution of 1:16, CMV antigen at 1:1200. We applied phytohemaglutinin (PHA) at a final concentration of 10 Ag/ml. The cells were cultured in an incubator at 37 jC. Interferon (IFN)-g and interleukin (IL)-2 assays were of 24 h culture duration; IL-4 and IL-5 assays were harvested after 48 h. Subsequently, the plates were washed 3_ with PBS, then 3_ with PBS–Tween (0.5%), and the detection antibodies (in PBS–BSA– Tween) were added at concentrations specified below.
After an overnight incubation at 4 jC, plates were 4_ washed with PBS–Tween and the Streptavidin– HRP conjugate (Dako, Carpenteria, CA) was added at a 1:2000 dilution in PBS–BSA for 2 h at room temperature. Afterwards, plates were washed 3_ with PBS–Tween followed by 3_ washing with PBS. The spots were visualized using the horseradish peroxidase (HRP)-substrate AEC (Pierce Pharmaceuticals, Rockford, IL). The AEC stock solution was prepared by dissolving 10 mg AEC in 1 ml N,N-dimethyl formamide (Fisher Scientific, Fair Lawn, NJ). For the actual development, 1 ml of the AEC stock solution was freshly diluted into 30 ml of 0.1 M sodium-acetate buffer (pH 5.0), filtered (0.45 Am), and mixed with 15 Al H2O2. A total of 200 Al of the AEC solution were C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 81 plated per well. The reaction was stopped by rinsing with distilled water when spots became clearly visible macroscopically (10 to 45 min, dependent on the cytokine). The plates were air-dried overnight before subjecting them to image analysis using a Series 1 ImmunoSpot Analyzer (Cellular Technology). To assess the amount of cytokine produced by single cells, we also measured the spot-size distribution, which is one of the built-in functions of the Immunospot software.
For human ELISPOT assays, the following capture antibodies were used: IFN-g (2G1 from Endogen, Woburn, MA) at 2 Ag/ml, IL-2 (5334.21 from R&D Systems, Minneapolis, MN) at 3 Ag/ml, IL-4 (8D4-8 from BD PharMingen, San Diego, CA) at 4 Ag/ ml, and IL-5 (TRFK5 hybridoma was obtained from ATCC, Rockville, MD, and the antibody was grown and purified in our laboratory) at 4 Ag/ml. For detection, we used biotinylated antibodies: IFN-g (B133.5 from Endogen, biotinylated in our laboratory) at 2 Ag/ml, IL-2 (BG5 from Endogen) at 0.125 Ag/ml, IL-4 (MP4- 25D2 from PharMingen) at 3 Ag/ml, and IL-5 (JES1- 5A10 from PharMingen) at 4 Ag/ml.
Murine ELISPOT assays were performed identically, except that 106 spleen cells were plated in HL- 1 culture medium containing 1% L-glutamine and 1% Penicillin–Streptomycin. The following mAbs (all from PharMingen) were used: IFN-g (R46A2) at 6 Ag/ml, IL-2 (JES6-1A12) at 2 Ag/m l, IL-4 (11B11) at 6 Ag/ml, and IL-5 (TRFK5) at 4 Ag/ml. The detection biotinylated mAB were: IFN-g (XMG1.1) at 1 Ag/ml, IL-2 (JES6-5H4) at 3 Ag/ml, IL-4 (BVD6-24G2) at 2 Ag/ml, and IL-5 (TRFK4). The Splenocytes were stimulated with OVA (50 Ag/ ml), OVA-peptide (10 Ag/ml), or anti-CD3 (1 Ag/ml, 2C11) as previously established in our laboratory (Forsthuber et al., 1996; Yip et al., 1999).
2.4. Image analysis of ELISPOT data We used a Series 1 ImmunoSpot Image Analyzer (Cellular Technology) specifically designed for the ELISPOT assay. Digitized images were analyzed for the presence of areas in which color density exceeds background by an amount set on the basis of the comparison of experimental wells (containing T cells, APC and antigen) and control wells (containing T cells and APC only). After background and noise subtraction, custom software is used to analyze spot morphology for circularity and density distribution to identify and separate touching and overlapping spots (Hesse et al., 2001; Karulin et al., 2000). Objects that meet these criteria are recognized as spots and counted. The measurement of spot-size distribution is also a built-in function of the software; it is based on the array of spot sizes in a given well sorted according to distinct size categories.
2.5. Statistics
ELISPOT frequencies were compared between groups using the t-test as calculated by SigmaStat (version 7.0, SPSS, Chicago, IL). Statistical significance was set at pV0.05. The coefficient of variation (CV) to assess the relative variability of intraassay variations was calculated according to the formula CV=(Standard deviation of spot counts from triplicate wells/Mean spot counts in the same triplicate wells)_100. The regression curves and r2 values for spot counts obtained testing fresh vs. frozen samples were calculated by SigmaPlot (version 7.0, SPSS).
3. Results and discussion
3.1. Improved recovery and functionality of PBMC when DMSO containing freezing medium is added at room temperature A number of active and passive cellular processes contribute to maintaining cell viability during freezethawing.
Osmotic and solute imbalances need to be controlled and the toxicity of dimethyl sulfoxide (DMSO) is to be balanced with its protective function, which depends on the stability and dynamics of biomembranes (Yu and Quinn, 1998; Wolfe and Bryant, 1999). Traditionally, to minimize toxic side effects of DMSO, ice-cold freezing medium (containing DMSO) has been added to cells metabolically inactivated by chilling on ice (Areman et al., 1988; Yokoyama, 1997).
We compared freezing cells according to this classic protocol, with adding freezing medium at room temperature. The cell recovery was 84% and the viability of the thawed cell population was f94% when the latter protocol was applied. In contrast, when chilled freezing medium was added, only 54% of the cells were recovered (Fig. 1). This difference was statistically highly significant ( pV0.001). When the thawed cells were tested in ELISPOT assays, the cells frozen according to the classic protocol showed reduced functionality as compared to the cells to which DMSO containing freezing medium was added at 25 jC (Table 1). As specified in this table, the temperature-dependent differences were statistically significant. Therefore, cell recovery and functionality in ELISPOT assays were critically temperature-dependent. These results suggest that the metabolic toxicity of DMSO is outweighed by a temperature-sensitive mechanism that conceivably involves membrane lipids. At room temperature membrane lipids are in a fluid state and the lipophilic DMSO can evenly integrate itself into membranes to exert cryoprotective effects. In contrast, chilling causes a membrane transition to a gel phase involving lipid crystallization that interferes with the integration of DMSO (Yu and Quinn, 1998;Wolfe and Bryant, 1999). Among the several parameters that we altered to optimize the recovery of functional T cells,
Table 1
Temperature of freezing medium affects functionality of thawed PBMC Temperature of freezing medium
25 jC 4jC Cytokine Medium Mumps Ag EBV peptide Dust Ag Medium Mumps Ag EBV peptide Dust Ag
IFN-g 4F1a 231F21 39F2 23F5 2F1 140F29 p = 0.012 44F7 ( p = 0.246)b 3F1 ( p = 0.200) IL-2 2F1 77F5 2F2 17F1 1F1 25F5 pV0.001 1F1 4F2 p = 0.005 IL-4 1F0 17F3 1F0 16F2 0F0 0F0 p = 0.001 0F0 0F0 pV0.001 IL-5 3F1 1F1 1F1 24F6 1F1 0F0 0F0 4F1 p = 0.024
a Result represent meanFS.D. of triplicate wells.
b p values not reaching statistical significance are in parenthesis.
Fig. 1. Increased recovery of cell numbers and functionality when DMSO containing freezing medium is added at room temperature. Freshly isolated PBMC were split into two equal aliquots. One aliquot was cooled down to 4 jC in a refrigerator (for 20 min); the other one was kept at room temperature. Equal volumes of 20% DMSO containing freezing medium were added at the same rate adjusted for the temperature of the cells (at 25 or 4 jC). The cells were frozen and thawed as described in Materials and methods. Cell viability was jugged by acridine orange and ethidium bromide staining. The number of recovered viable (gray bar), non-viable (shaded bar) and lost cells (black bar) is shown as a percentage of the total number of PBMC originally frozen. The p values for the temperature comparison of viable recovered cells is specified. The results are representative for three independent experiments performed. C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 83
the temperature at which freezing medium was added was found to be by far the most critical. The rate of adding DMSO containing freezing medium (30 s–2 min) and the speed of thawing (1–5 min) had only marginal effects on PBMC recovery and functionality in cytokine ELISPOT assays (data not shown).
3.2. Cryopreservation does not affect cytokine production by mitogen-stimulated human PBMC By cross-linking sugars on membrane glycoproteins, mitogens (e.g. PHA) activate CD4+ and CD8+ cells and cells of the innate immune system. Such polyclonal activation of PBMC results in a level of cytokine production that is measurable by enzymelinked immunosorbent assays (ELISAs) (ELISPOT assays have >100-fold higher sensitivity than ELISAs (Tanguay and Killion, 1994)) but the immune diagnostic value of mitogen stimulations is limited. Previous studies have measured the effects of freezing on cytokine production by mitogen stimulated PBMC with contradictory results (Cillari et al., 1988; Venkataraman, 1995; Sobota et al., 1997; Wang et al., 1998).
We compared the frequencies of IFN-g, IL-2, IL-4, and IL-5 producing cells (Fig. 2A–D) and the per cell cytokine output (Hesse et al., 2001) of the individual cells (the inserted spot size distribution curves) in fresh vs. thawed PBMC after stimulation with PHA. At all cell concentrations tested, the fresh and freeze-thawed PBMC yielded comparable results.
Fig. 2. Fresh and thawed PHA-stimulated PBMC show similar frequencies of cytokine producing cells and per cell cytokine output. The same PBMC were tested either fresh (black circles) or thawed, 1 week after freezing (gray circles). ELISPOT assays were performed measuring IFN-g (A), IL-2 (B), IL-4 (C), or IL-5 (D). The number of cytokine spots per well ( Y-axis) is plotted against the cell number plated (X-axis). Standard deviations of spot numbers per well for triplicate wells were < 20%, and the r-values for regression lines were >0.95 (not shown). The inserts show the spot size distribution for the fresh and thawed cells in all wells. The data shown were obtained testing PBMC of a healthy volunteer and were reproduced on the same donor three times, and three times with PBMC of four different healthy donors. 84 C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93
However, when similar experiments were done with murine spleen cells or PBMC, a fundamentally different outcome was observed. As opposed to the f85% cell recovery after the cryopreservation of human PBMC, 66% of murine spleen cells and only 7% of murine PBMC were recovered. The recovery was better when DMSO was added at 25 jC (Fig. 3A).
When stimulated with anti-CD3 antibody, the frequency of IFN-g, IL-2, and IL-4 producing cells was reduced in spleen cells by up to 79% (Fig. 3B). Because of the >90% cell loss, we could not subject murine PBMC to functional assays. Similar results were obtained with C57BL/6 and BALB/c mice, irrespective of whether DMSO containing freezing medium was added cold or at room temperature. Therefore, murine spleen cells seem to be more sensitive to loss of function in ELISPOT assays during freeze-thawing than human PBMC.
Fig. 3. Cryopreservation impairs the cytokine response in murine splenocytes. (A) Freshly isolated mice spleen were split into two equal aliquots and processed as described in Fig. 1. The number of recovered viable (gray bar), non-viable (shaded bar) and lost cells (black bar) is shown as a percentage of the total number of cells originally frozen. The p values for the temperature comparison of viable recovered cells are specified. (B) Spleen cells of six naive C57BL/6 mice were pooled and tested directly ex vivo (closed bars) in the ELISPOT assays specified. The cell pool was frozen within 5 h after isolation, thawed a week later, and retested (gray bars). For fresh and thawed samples, the cell numbers were adjusted to 1 million viable cells per well. The mean number of spots per well and the standard deviation from triplicate wells is shown. The data are representative for three independent experiments. C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 85
3.3. Defining type 1/type 2 cytokine profile of recall responses in pre-freeze PBMC The primary scope of immunodiagnostics is to define the type 1/type 2 cytokine signature of antigen- specific CD4+ and CD8+ cells. IFN-g ELISPOT assays have recently been validated by us (Helms et al., 2000) as well as by others (McCutcheon et al., 1997; Smith et al., 2001; Currier et al., 2002) and have come into broad use in the scientific community. However, measuring IFN-g alone or even together with IL-4 may not sufficiently characterize T cell immunity. Thus, there is increasing evidence that the Th1/Th2 model does not account for the whole spectrum of cytokine expression patterns of T cells (Kelso, 1995). The coexpression of cytokines within the type 1 or the type 2 cassette can be stochastic (Kelso et al., 1995) or even mutually exclusive (Karulin et al., 2000); for example, neither do Th1 cells necessarily coexpress IL-2 and IFN-g nor Th2 cells obligatory coexpress IL-4 and IL-5 (Bucy et al., 1994; Jung et al., 1995). Therefore, we first closely characterized the cytokine expression patterns of recall antigen-specific T cells and subsequently tested whether these patterns are main-tained after freeze-thawing. Performing IFN-g, IL-2, IL-4, and IL-5 ELISPOT assays on freshly isolated PBMC of four healthy donors, we tested a series of antigens to which individuals frequently become environmentally sensitized.
The four protein antigens—when added extracellularly— are presented primarily on MHC class II molecules and the two MHC Class I restricted peptides that can directly bind to HLA-A2 molecules were tested as specified in Table 2. All four donors responded to mumps antigen with a classic type 1- polarized response, producing IFN-g and IL-2, but no IL-4 or IL-5. CMV antigen triggered an additional IL-4 response in the absence of IL-5 production in donor #1. These latter responses were therefore of a mixed Th0 type (however without involving IL-5 that is canonically linked to the Th0 response; Mosmann and Coffman, 1989). This anti-CMV response appeared to be specific: while donor #1 displayed a vigorous response, the other three individuals (#2, 3 and 4) did not respond at all. Similarly, while three donors (#1, 3 and 4) did not respond to grass allergen, donor #2 displayed a strong, highly type 2 polarized response, producing IL-4 and IL-5 in addition to IL-2, but no IFN-g. Donor #3 (unlike all the other donors) responded to dust mite antigen, with a Th0-type cytokine signature. Three of the four donors were HLA-A2 positive (donors #1–3). The A-2- restricted peptide of EBV induced IFN-g production in donors #1, 2, 3, but not in the A-2 negative donor #4. The A-2-restricted CMV peptide triggered a response in donor #1, but not in the A-2 positive donors #2 and #3; only IFN-g was detected. All antigen-induced responses shown in Table 2 are statistically significant compared to the medium background ( pV0.001). Overall, these cytokine sig-
Table 2
Frequenciesa and type 1/type 2 cytokine signatures of ELISPOT assays in freshly isolated PBMC Antigen Donor Frequencies per 300,000 PBMCb IFN-g IL-2 IL-4 IL-5 Mumps antigen 1 57F4c 43F4 –d – 2 85F11 46F2 – – 3 219F25 124F18 – – 4 115F21 55F9 – –
CMV antigen 1 197F35 57F6 18F6 – 2 – – – – 3 – – – – 4 – – – – EBV peptide 1 41F2 – – – 2 143F14 – – – 3 32F2 – – – 4 – – – –
CMV peptide 1 76F17 – – – 2 – – – – 3 – – – – 4 – – – – Grass allergen 1 – – – – 2 – 20F4 21F5 51F8 3 – – – – 4 – – – – Dust allergen 1 – – – – 2 – – – – 3 25F15 44F5 10F3 43F5 4 – – – – a Background subtracted. b Results represent meanFS.D. of triplicate wells. c Values < 10. d All recall Ag-induced spot counts are significantly increased ( pV0.001) over medium background counts.
86 C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93
natures were reproducible in six independent repeat experiments establishing the unique and characteristic cytokine profile for each donor–antigen combination on which the subsequent comparison of fresh vs. frozen samples could be based.
3.4. Defining CD4+ and CD8+ phenotypes of the recall antigen-specific T cells in pre-freeze PBMC CD4+ and CD8+ cells might show different sensitivity to freeze-thawing. We therefore performed cell
Fig. 4. Defining the CD4+/CD8+ phenotype of the recall antigen-specific cytokine producing cells. PBMC of donor #2 were tested as
unfractionated PBMC (black bars) and after cell separation into CD4+ (fine hatched bars) and CD8+ cells (hatched bars). T cell-depleted PBMC (98% purity) were plated as APC with the purified CD4+ cells (97% purity) or CD8+ cells (99% purity). The PBMC and cell fractions were tested for the production of the cytokines specified after stimulated with mumps antigen (A), EBV peptide (B), or grass allergen (C). The data shown are the means and standard deviations of triplicate wells after subtracting the spots in medium wells. The data are representative of three independent experiments.
Table 3
Frequenciesa of antigen-specific IFN-g producing cells in 300,000 fresh and thawed PBMCb
Experiment 1 Experiment 2 Experiment 3 Antigen Donor Fresh Frozen RIc Fresh Frozen RI Fresh Frozen RI
Mumps Ag 1 16F8 13F9 81 57F4 32F8 56 58F9 84F32 145 2 33F4d1 97F20d1 294 85F11d2 46F5d2 54 26F2 33F0 127
3 284F36 227F21 80 219F25 215F15 98 199F2 272F4 137 4 84F11 81F16 96 115F21 85F11 74 25F5 24F2 96 CMV Ag 1 294F46 357F84 121 197F35 190F14 96 445F21 416F33 93 EBV peptide 1 61F17 46F6 75 41F2d3 15F6d3 37 49F8 65F5 133
2 413F25 357F8 86 143F14 127F10 89 306F12 321F15 105 3 83F11d4 35F2d4 42 32F2 23F9 72 65F0 56F2 86
CMV peptide 1 36F3d5 42F3d5 117 76F17 48F8 63 83F18 101F20 122 Grass Ag 2 5F1 14F5 280 9F1 11F2 122 9F1 5F0 56
Dust Ag 3 12F3d6 19F5d6 158 25F15 19F8 76 10F5 5F1 50 There was no statistically significant difference between fresh and frozen recall responses except d1 p = 0.003, d2 p = 0.004, d3 p = 0.010, d4 p = 0.002, d5 p = 0.048, d6 p = 0.034.
a Background subtracted.
b Results represent mean of spot numbersFS.D. of triplicate wells.
c Recovery index (%).
C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 87
separation experiments to clearly define which of the above cytokine recall responses was CD4+ or CD8+ cell derived.
Fig. 4 shows the results obtained with cells of donor #2, who displayed responses to several recall antigens. The IFN-g response of the PBMC to mumps antigen was recovered entirely in the CD4+ cell fraction (Fig. 4A). Moreover, the frequency of mumps antigeninduced IFN-g producing cells was 3.2-fold higher in the CD4+ cell fraction than in the PBMC (f30% of the PBMC are CD4+ cells), consistent with the 3.3-fold higher frequency of CD4+ cells in the cell sample tested. The IL-2 secreting mumps protein-specific cells were also entirely recovered in the CD4+ cell fraction (Fig. 4A). In contrast, the reactivity to the EBV peptide in this donor resided entirely in the CD8+ cell fraction with a 6.6-fold higher frequency of EBV-specific spots than in the unseparated PBMC (Fig. 4B; f20% of PBMC are CD8+ cells, corresponding to a 5-fold enrichment of CD8+ cells in the test population).
PBMC of this donor also displayed IL-4 and IL-5 production in response to grass allergen. The frequency of IL-4 and IL-5 spots was 3.2- and 3.1-fold enriched in the CD4+ cell fraction vs. the PBMC, but 18% of the IL-5-producing cells were recovered in the CD8+ cell fraction (when numbers are corrected for the relative frequency of CD8+ cells present in PBMC (f20%) vs. the purified CD8+ cells). The type 2 cytokine response to grass allergen in this individual was therefore primarily mediated by CD4+ cells, but it also entailed a minor, but clear cut CD8+ cell component (Fig. 4C). Testing of the other donors gave similar
Table 4
Frequenciesa of antigen-specific IL-2 producing cells in 300,000 fresh and thawed PBMCb
Experiment 1 Experiment 2 Experiment 3
Antigen Donor Fresh Frozen RIc Fresh Frozen RI Fresh Frozen RI
Mumps Ag 1 16F3 14F3 88 43F4d1 30F1d1 70 31F2 10F1 32
2 20F8 20F4 100 46F2d2 33F4d2 72 24F4 19F1 79
3 118F5d3 75F5d3 64 124F18 122F5 98 141F1 91F3 65
4 28F6 29F2 104 55F9 49F3 89 13F2d5 6F1d5 46
CMV Ag 1 59F6d4 38F2d4 64 57F6 69F13 121 52F1 43F4 83
Grass Ag 2 10F6 6F3 60 20F4 25F4 125 8F2 2F0 25
Dust Ag 3 18F4 15F1 83 44F5 41F8 93 22F2 21F8 95
There was no statistically significant difference between fresh and frozen recall responses except d1 p = 0.011, d2 p = 0.014, d3 p = 0.001, d4
p = 0.002, d5 p = 0.002.
a Background subtracted.
b Results represent mean of spot numbersFS.D. of triplicate wells.
c Recovery index (%).
Table 5
Frequenciesa of antigen-specific IL-4 producing cells in 300,000 fresh and thawed PBMCb
Experiment 1 Experiment 2 Experiment 3
Antigen Donor Fresh Frozen RIc Fresh Frozen RI Fresh Frozen RI
CMV Ag 1 32F3 27F6 84 18F6 20F2 111 22F4 34F6 155
Grass Ag 2 7F3 7F3 100 21F5 31F4 148 9F1 9F1 100
Dust Ag 3 14F0 15F2 107 10F3d1 18F5d1 180 14F1 21F4 150
There was no statistically significant difference between fresh and frozen recall responses except d1 p = 0.032.
a Background subtracted.
b Results represent meanFS.D. of triplicate wells.
c Recovery index (%).
88 C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93
results (data not shown): the response of donor #1 to CMVantigen (IFN-g, IL-2, IL-4) and the responses of donors #1 and #3 to mumps antigen (IFN-g and IL-2) were found to be entirely CD4+, while the IFN-g response of both donors to EBV peptide was entirely CD8+ cell derived. The response of donor #3 to dust allergen was mixed, with 69% of IL-4 producing cells residing in the CD4+ cell fraction and 31% in the CD8+ cell fraction (data not shown). Overall, these experiments performed on freshly isolated PBMC permitted us to establish donor-antigen combinations for the subsequent studies addressing how CD4+ and CD8+ T cells, and the individual cytokines are affected by freeze-thawing.
3.5. Cryopreservation does not affect cytokine production by recall antigen-stimulated human PBMC
PBMC from the aforementioned four donors were tested repeatedly as freshly isolated and cryopreserved cells. Peripheral blood samples were obtained from donors at three different time points 3–4 weeks apart. One aliquot of PBMC was tested within 5 h of isolation; the other aliquot was frozen within 5 h, and thawed after 7 days for testing. The results obtained for IFN-g, IL-2, IL-4, and IL-5 are summarized in Tables 3–6, respectively. Data are shown only for those antigen–cytokine combinations to which the respective donors were found to respond (see Table 2). The spot counts obtained with the fresh and the frozen cells are shown. Additionally the percentage of spots recovered in the frozen sample relative to the fresh sample was calculated as a recovery index: RI=(Number of spots in thawed sample/Number of spots in fresh sample)_100%.
Overall, comparing 69 data points for fresh vs. frozen in Tables 3–6, there was no statistical significance for 57 data points. Twelve comparisons showed statistically significant results for the testing of fresh vs. frozen cells. However, these differences did not reproduce for different bleeds in repeat experiments done on an individual, or were not consistent for the different individuals in a group. Therefore, the occasional statistically significant differences seem to be attributable to experimental variations in a particular assay (see below).
Table 7 summarizes the results for CD4+ And CD8+ cells. Because the PBMC response to mumps antigen and to CMV antigen was entirely CD4+ cellmediated, the mean and standard deviations of the recovery indices for these antigens (Tables 3–5) were calculated for the individual cytokines from the data obtained with these antigens in three repeat experiments. Cumulative results for CD4+ cells are shown in Table 7. Similarly, because the PBMC response to the peptides of EBV and CMV was entirely CD8+ cell-mediated, the mean and standard deviations of the recovery indices for these peptides (Table 3) were
Table 6
Frequenciesa of antigen-specific IL-5 producing cells in 300,000 fresh and thawed PBMCb
Experiment 1 Experiment 2 Experiment 3
Antigen Donor Fresh Frozen RIc Fresh Frozen RI Fresh Frozen RI
Grass Ag 2 18F7 21F1 117 51F8 59F2 116 27F4 9F1 33
Dust Ag 3 15F1 21F6 140 43F5 30F6 70 35F7 21F4 60
In all instances there was no statistically significant difference between responses of fresh and frozen cells.
a Background subtracted.
b Results represent meanFS.D. of triplicate wells.
c Recovery index (%).
Table 7
Recovery indices for the frequencies of antigen-specific CD4+ and
CD8+ cells in PBMC after freeze-thawing
Recovery indices (%)
Cell population IFN-g IL-2 IL-4 IL-5
CD4+ T cells 110aF58 78aF23 117bF36 89cF41
CD8+ T cells 86dF30 80cF34 127cF36 89cF41
a CD4+ cell response to mumps Ag and CMV Ag.
b CD4+ cell response to CMV Ag.
c CD4+ and CD8+ cell response to grass Ag and dust Ag.
d CD8+ cell response to peptides of EBV and CMV.
C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 89
calculated for IFN-g from the data obtained testing the PBMC as the cumulative results for CD8+ cells (Table 7).
To compare the intra-assay variations of the spots induced by recall antigens in fresh vs. frozen cells, the coefficient of variation (CV) was calculated which measures the variability relative to the magnitude of the spot counts. For fresh and frozen cells, the regression curves followed a similar pattern showing an hyperbolic inverse relationship between variability and spot counts per well. When the number of antigen-specific cells was low within the 100 Al of cell suspension plated per well, the intra assay variability was higher than at the higher frequencies of antigen-specific cells (Fig. 5A and B). When fresh vs. frozen samples were compared for the mean number of spots induced in each recall antigen-induced response, a close to perfect linear correlation was seen (Fig. 5C and D). Therefore, the testing of fresh and frozen cells gave equivalent results for IFN-g and IL- 2. Also, the allergen-induced frequencies of IL-4 and IL-5 producing cells did not show statistically significant differences between fresh and the frozen samples (Tables 5 and 6). However, because of the lower
Fig. 5. Statistical analysis of recall-antigen-induced IFN-g and IL-2 spots in fresh and frozen samples. The data from each individual in Table 3 (IFN-g) and Table 4 (for IL-2) are plotted for each antigen, comparing the spots induced in fresh and frozen cells. In (A) and (B), the intra-assay variability is compared. The coefficient of variation was calculated as a measure of variability relative to the magnitude of the response (CV = Standard deviation of spot counts from triplicate wells/Mean spot counts in the same triplicate wells_100) and plotted against the corresponding mean spot numbers. The regression curves and r2 values are shown. In (C) and (D), the mean spot counts obtained for each fresh vs. frozen determination are plotted against each other. Regression curves and r2 values are as specified. 90 C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93
numbers of antigens and responsive individuals, these IL-4 and IL-5 responses were not subject to further detailed statistical analysis. (The range of CV values for IL-4 and IL-5 was comparable to IFN-g and IL-2, data not shown.) Overall, the variations were remarkably low in comparison with other bioassays available for testing T cell function in freshly isolated cell material.
3.6. Cryopreservation abrogates cytokine production by recall antigen-stimulated murine splenocytes While CD4+ and CD8+ cells in human PBMC showed unimpaired antigen-specific cytokine recall responses after freezing, murine T cells behaved differently. We studied the well-defined T cell response to OVA in C57BL/6 mice. Immunization with OVA induces a CD4+ cell-dominated response (Karulin et al., 2000; Cottrez et al., 2000). The cytokine signature of freshly isolated spleen cells obtained from OVA-immunized mice was characterized by the production of IFN-g, IL-2, IL-4, and IL- 5 (Fig. 6A). The injection of the Kb restricted OVA: 257–264 peptide (SIINFEKL) in adjuvant triggers CD8+ cells that also produce all four of these cytokines. In contrast to the freshly isolated spleen cells that yielded frequencies of cytokine-producing cells in the 13–155 per million frequency range, the thawed spleen cells showed marked loss of function. Except for IL-2 that was reduced by 51%, all other cytokines within the protein-induced CD4+ response became undetectable. Similarly, the production of IFN-g, IL-2, IL-4, and IL-5 by the OVA-peptidespecific CD8+ T cells was reduced by 71%, 65%, 96% and 100%, respectively (Fig. 6B). Similar results were obtained with BALB/c mice immunized with OVA and its MHC class II restricted peptide OVA: 323–339.
We tested whether the deficient function of the thawed cells resulted from an APC defect. Adding naive fresh spleen cells to the thawed cells did not recover their function (data not shown), excluding an APC deficiency and suggesting that the murine T cells themselves lose functionality in ELISPOT assays after cryopreservation.
3.7. Concluding remarks
The data presented here show that freshly isolated human CD4+ and CD8+ T cells maintain full functionality in cytokine ELISPOT assays after thawing, provided the PBMC are frozen by adding the DMSO containing freezing medium at room temperature. Therefore, frozen PBMC samples are well suited to establish the frequencies and cytokine signatures of antigen-specific Th1, Th2, Tc1, and Tc2 cells. These data were obtained with PBMC of healthy individuals. It remains to be established whether T cells in patients
Fig. 6. T cells in murine splenocytes lose function after freeze-thawing. C57BL/6 mice were immunized with OVA (A) or with the SIINFEKL OVA-peptide (B) as described in Materials and methods. One month later, one-half of the spleen cells was tested directly ex vivo (black bars); the other half of the cells was frozen and tested 1 week later (gray bars). The data represent means and standard deviations of triplicate wells from spleens pooled from six mice. The results are representative of three experiments performed.
C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 91
with various diseases or in individuals undergoing pharmacologic treatments also display this resistance to cryopreservation.
Our data have important implications for vaccine trials involving healthy individuals. Cryopreservation of PBMC permits testing of individuals before and after immunization for side by side comparison of samples in a single experiment, avoiding inter-assay variation. Furthermore, frozen aliquots of the same blood draw permit reproduction of data for each time point in successive experiments. Last, frozen samples should become indispensable for multi-center clinical studies with frozen PBMC shipped to a central laboratory for high throughput testing under standardized conditions. The maintained functional activity of CD4+ and CD8+ cells after freeze-thawing contributes to the value of cytokine ELISPOT assays for low frequency measurements as a monitoring tool of cellular immunity.
Acknowledgements
We thank Paul V. Lehmann and the members of our laboratory for valuable discussions and Earl Sigmund for editorial assistance.
References
Agarwal, S., Rao, A., 1998. Modulation of chromatin structure regulates cytokine gene expression during T cell differentiation. Immunity 9, 765– 775.
Areman, E.M., Simonis, T.B., Carter, C.S., Read, E.J., Klein, H.G., 1988. Bulk cryopreservation of lymphocytes in glycerol. Transfusion 28, 151– 156.
Barouch, D.H., Letvin, N.L., 2001. CD8+ cytotoxic T lymphocyte responses to lentiviruses and herpesviruses. Curr. Opin. Immunol. 13, 479–482.
Bucy, R.P., Panoskaltsis-Mortari, A., Huang, G.Q., Li, J., Karr, L., Ross, M., Russell, J.H., Murphy, K.M., Weaver, C.T., 1994. Heterogeneity of single cell cytokine gene expression in clonal T cell populations. J. Exp. Med. 180, 1251–1262.
Chu, R.S., Targoni, O.S., Krieg, A.M., Lehmann, P.V., Harding, C.V., 1997. CpG oligodeoxynucleotides act as adjuvants that switch on T helper 1 (Th1) immunity. J. Exp. Med. 186, 1623–1631.
Cillari, E., Liew, F.Y., Lo, C.P., Milano, S., Mansueto, S., Salerno, A., 1988. Suppression of IL-2 production by cryopreserved peripheral blood mononuclear cells from patients with active visceral leishmaniasis in Sicily. J. Immunol. 140, 2721– 2726.
Cottrez, F., Hurst, S.D., Coffman, R.L., Groux, H., 2000. T regulatory cells 1 inhibit a Th2-specific response in vivo. J. Immunol. 165, 4848– 4853.
Currier, J.R., Kuta, E.G., Turk, E., Earhart, L.B., Loomis-Price, L.,
Janetzki, S., Ferrari, G., Birx, D.L., Cox, J.H., 2002. A panel of MHC class I restricted viral peptides for use as a quality control for vaccine trial ELISPOT assays. J. Immunol. Methods 260, 157– 172.
Forsthuber, T., Yip, H.C., Lehmann, P.V., 1996. Induction of TH1 and TH2 immunity in neonatal mice. Science 271, 1728– 1730.
Fremont, D.H., Stura, E.A., Matsumura, M., Peterson, P.A., Wilson, I.A., 1995. Crystal structure of an H-2Kb-ovalbumin peptide complex reveals the interplay of primary and secondary anchor positions in the major histocompatibility complex binding groove. Proc. Natl. Acad. Sci. U. S. A. 92, 2479– 2483.
Heeger, P.S., Valujskikh, A., Lehmann, P.V., 2000. Comprehensive assessment of determinant specificity, frequency, and cytokine signature of the primed CD8 cell repertoire induced by a minor transplantation antigen. J. Immunol. 165, 1278–1284.
Helms, T., Boehm, B.O., Asaad, R.J., Trezza, R.P., Lehmann, P.V., Tary-Lehmann, M., 2000. Direct visualization of cytokine-producing recall antigen-specific CD4 memory T cells in healthy individuals and HIV patients. J. Immunol. 164, 3723–3732.
Hesse, M.D., Karulin, A.Y., Boehm, B.O., Lehmann, P.V., Tary- Lehmann, M., 2001. A T cell clone’s avidity is a function of its activation state. J. Immunol. 167, 1353– 1361.
Jung, T., Schauer, U., Rieger, C., Wagner, K., Einsle, K., Neumann, C., Heusser, C., 1995. Interleukin-4 and interleukin-5 are rarely co-expressed by human Tcells. Eur. J. Immunol. 25, 2413– 2416.
Karulin, A.Y., Hesse, M.D., Tary-Lehmann, M., Lehmann, P.V., 2000. Single-cytokine-producing CD4 memory cells predominate in type 1 and type 2 immunity. J. Immunol. 164, 1862– 1872.
Kelso, A., 1995. Th1 and Th2 subsets: paradigms lost? Immunol. Today 16, 374– 379.
Kelso, A., Groves, P., Troutt, A.B., Francis, K., 1995. Evidence for the stochastic acquisition of cytokine profile by CD4+ T cells activated in a T helper type 2-like response in vivo. Eur. J. Immunol. 25, 1168– 1175.
McCutcheon, M., Wehner, N., Wensky, A., Kushner, M., Doan, S., Hsiao, L., Calabresi, P., Ha, T., Tran, T.V., Tate, K.M., Winkelhake, J., Spack, E.G., 1997. A sensitive ELISPOT assay to detect low-frequency human T lymphocytes. J. Immunol. Methods 210, 149– 166.
McMichael, A.J., O’Callaghan, C.A., 1998. A new look at T cells. J. Exp. Med. 187, 1367– 1371.
Mosmann, T.R., Coffman, R.L., 1989. TH1 and TH2 cells: different patterns of lymphokine secretion lead to different functional properties. Annu. Rev. Immunol. 7, 145–173.
Smith, J.G., Liu, X., Kaufhold, R.M., Clair, J., Caulfield, M.J., 2001. Development and validation of a gamma interferon ELISPOT assay for quantitation of cellular immune responses to varicella-zoster virus. Clin. Diagn. Lab. Immunol. 8, 871– 879.
Sobota, V., Bubenik, J., Indrova, M., Vlk, V., Jakoubkova, J., 1997. Use of cryopreserved lymphocytes for assessment of the immunological effects of interferon therapy in renal cell carcinoma patients. J. Immunol. Methods 203, 1 –10. 92 C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93
Tanguay, S., Killion, J.J., 1994. Direct comparison of ELISPOT and ELISA-based assays for detection of individual cytokine-secreting cells. Lymphokine Cytokine Res. 13, 259– 263.
Tary-Lehmann, M., Saxon, A., 1992. Human mature T cells that are anergic in vivo prevail in SCID mice reconstituted with human peripheral blood. J. Exp. Med. 175, 503–516.
Venkataraman, M., 1995. Effects of cryopreservation on immune responses: VIII. Enhanced secretion of interferon-gamma by frozen human peripheral blood mononuclear cells. Cryobiology 32, 528–534.
Wang, S.Y., Hsu, M.L., Tzeng, C.H., Hsu, H.C., Ho, C.K., 1998. The influence of cryopreservation on cytokine production by human T lymphocytes. Cryobiology 37, 22– 29.
Weinberg, A., Betensky, R.A., Zhang, L., Ray, G., 1998. Effect of shipment, storage, anticoagulant, and cell separation on lymphocyte proliferation assays for human immunodeficiency virus-infected patients. Clin. Diagn. Lab. Immunol. 5, 804–807.
Whiteside, T.L., 2000. Immunologic monitoring of clinical trials in patients with cancer: technology versus common sense. Immunol. Invest. 29, 149– 162.
Wolfe, J., Bryant, G., 1999. Freezing, drying, and/or vitrification of membrane–solute-water systems. Cryobiology 39, 103– 129.
Yip, H.C., Karulin, A.Y., Tary-Lehmann, M., Hesse, M.D., Radeke, H., Heeger, P.S., Trezza, R.P., Heinzel, F.P., Forsthuber, T., Lehmann, P.V., 1999. Adjuvant-guided type-1 and type-2 immunity: infectious/noninfectious dichotomy defines the class of response. J. Immunol. 162, 3942– 3949.
Yokoyama,W.M., 1997. Cryopreservation of cells. In: Coligan, J.E., Kruisbeek, A.M., Margulies, D.H., Shevach, E.M., Strober, W. (Eds.), Current Protocols in Immunology. Wiley, New York, pp. A.3.G.1– A.3.G.3.
Yu, Z.W., Quinn, P.J., 1998. The modulation of membrane structure and stability by dimethyl sulphoxide. Mol. Membr. Biol. 15, 59– 68. C.R. Kreher et al. / Journal of Immunological Methods 278 (2003) 79–93 93
Source: http://www.seracare.com/Portals/0/Papers/PBMC%20Functionality%20Article.pdf