Optimization of storage and shipment of cryopreserved peripheral blood mononuclear cells from HIV-infected and uninfected individuals for ELISPOT assays

Abstract

Functional immunologic assays using cryopreserved peripheral blood mononuclear cells (PBMC) are influenced by blood processing, storage and shipment. The objective of this study was to compare the viability, recovery and ELISPOT results of PBMC stored and shipped in liquid nitrogen (LN/LN) or stored in LN and shipped on dry ice (LN/DI) or stored at -70°C for 3 to 12 weeks and shipped on DI (70/DI 3 to 12); and to assess the effect of donor HIV infection status on the interaction between storage/shipment and the outcome measures. PBMC from 12 HIV-infected and 12 uninfected donors showed that LN/LN conferred higher viability and recovery than LN/DI or 70/DI 3, 6, 9 or 12. LN/DI PBMC had higher viability than any 70/DI PBMC. The PBMC viability and recovery linearly decreased with the duration of storage at -70°C from 3 to 12 weeks. This effect was more pronounced in samples from HIV-infected than uninfected donors. Results of ELISPOT assays using CMV pp65, CEF and Candida albicans antigens were qualitatively and quantitatively similar across LN/LN, LN/DI and 70/DI 3. However, ELISPOT values significantly decreased with the duration of storage at -70°C both in HIV-infected and uninfected donors. ELISPOT results also decreased with PBMC viability <70%.

Figure 1

Viability and viable recovery of cryopreserved PBMC from HIV-infected and uninfected subjects according to shipment and storage conditions. Data derived from 12 HIV+ (HIV-infected) and 12 HIV− (HIV-uninfected) subjects are presented as mean and 95% CI of the viability and viable recovery on Day 1= day of thawing and on Day 2= after overnight incubation, when functional assays were set up. Cells were cryopreserved at a central laboratory and shipped to the testing laboratories in the conditions indicated on each graph: LN/DI=stored in liquid nitrogen and shipped on dry ice; LN/LN=stored and shipped in liquid nitrogen; 70/DIstored at −70°C and shipped on dry ice. 3, 6, 9, 12 indicate the numbers of weeks of storage at −70°C before shipment. The testing laboratories thawed and assayed samples from each subject in the same run. Figures 1a and 1c show that the viability of samples from HIV-uninfected individuals did not differ among LN/LN, LN/DI and 70/DI 3 samples on Day 1. On Day 2, there were no significant differences in LN/LN and LN/DI sample viability, but 70/DI 3 samples had significantly lower viability than LN/LN or LN/DI samples (p of 0.03 and 0.04, respectively). The Day 2 viability of 70/DI samples significantly decreased with the time of storage at −70°C (p=0.01), but the Day 1 did not (p=0.56). Figures 1b and 1d show that both Day 1 and Day 2 viability of cells from HIV-infected individuals was significantly higher in LN/LN compared with 70/DI 3 samples (p of 0.05 and 0.04, respectively). Viability of LN/DI samples was similar with that of LN/LN samples on Day 1, but lower on Day 2 (p of 0.1 and 0.03, respectively). No differences were observed in LN/DI and 70/DI 3 samples. There were sharp linear decreases of viability over time of storage at −70°C on Day 1 and Day 2 (p of 0.01 and 0.004, respectively). Figures 1e and 1g show that the viable recovery of samples from HIV-uninfected individuals did not differ among LN/LN, LN/DI and 70/DI 3 samples on Day 1. On Day 2, there were no significant differences in LN/LN and LN/DI sample viable recovery, but 70/DI 3 samples had significantly lower viable recovery than LN/LN or LN/DI samples (p of 0.02 and 0.05, respectively). The Day 2 viable recovery of 70/DI samples significantly decreased with the time of storage at −70°C (p =0.007), but the Day 1 did not (p=0.60). Figures 1f and 1h show that both Day 1 and Day 2 viable recovery of cells from HIV-infected individuals was significantly higher in LN/LN compared with 70/DI 3 samples (p of <0.01 and 0.02, respectively). Viable recovery of LN/DI samples was lower than that of LN/LN samples on Day 1, but similar on Day 2 (p of 0.01 and 0.09, respectively). No differences were observed in LN/DI and 70/DI 3 samples. There were sharp linear decreases of viable recovery over time of storage at −70°C on Day 1 and Day 2 (p of 0.0098 and 0.0078, respectively).

Figure 1

Viability and viable recovery of cryopreserved PBMC from HIV-infected and uninfected subjects according to shipment and storage conditions. Data derived from 12 HIV+ (HIV-infected) and 12 HIV− (HIV-uninfected) subjects are presented as mean and 95% CI of the viability and viable recovery on Day 1= day of thawing and on Day 2= after overnight incubation, when functional assays were set up. Cells were cryopreserved at a central laboratory and shipped to the testing laboratories in the conditions indicated on each graph: LN/DI=stored in liquid nitrogen and shipped on dry ice; LN/LN=stored and shipped in liquid nitrogen; 70/DIstored at −70°C and shipped on dry ice. 3, 6, 9, 12 indicate the numbers of weeks of storage at −70°C before shipment. The testing laboratories thawed and assayed samples from each subject in the same run. Figures 1a and 1c show that the viability of samples from HIV-uninfected individuals did not differ among LN/LN, LN/DI and 70/DI 3 samples on Day 1. On Day 2, there were no significant differences in LN/LN and LN/DI sample viability, but 70/DI 3 samples had significantly lower viability than LN/LN or LN/DI samples (p of 0.03 and 0.04, respectively). The Day 2 viability of 70/DI samples significantly decreased with the time of storage at −70°C (p=0.01), but the Day 1 did not (p=0.56). Figures 1b and 1d show that both Day 1 and Day 2 viability of cells from HIV-infected individuals was significantly higher in LN/LN compared with 70/DI 3 samples (p of 0.05 and 0.04, respectively). Viability of LN/DI samples was similar with that of LN/LN samples on Day 1, but lower on Day 2 (p of 0.1 and 0.03, respectively). No differences were observed in LN/DI and 70/DI 3 samples. There were sharp linear decreases of viability over time of storage at −70°C on Day 1 and Day 2 (p of 0.01 and 0.004, respectively). Figures 1e and 1g show that the viable recovery of samples from HIV-uninfected individuals did not differ among LN/LN, LN/DI and 70/DI 3 samples on Day 1. On Day 2, there were no significant differences in LN/LN and LN/DI sample viable recovery, but 70/DI 3 samples had significantly lower viable recovery than LN/LN or LN/DI samples (p of 0.02 and 0.05, respectively). The Day 2 viable recovery of 70/DI samples significantly decreased with the time of storage at −70°C (p =0.007), but the Day 1 did not (p=0.60). Figures 1f and 1h show that both Day 1 and Day 2 viable recovery of cells from HIV-infected individuals was significantly higher in LN/LN compared with 70/DI 3 samples (p of <0.01 and 0.02, respectively). Viable recovery of LN/DI samples was lower than that of LN/LN samples on Day 1, but similar on Day 2 (p of 0.01 and 0.09, respectively). No differences were observed in LN/DI and 70/DI 3 samples. There were sharp linear decreases of viable recovery over time of storage at −70°C on Day 1 and Day 2 (p of 0.0098 and 0.0078, respectively).

Figure 2

Effect of storage and shipment conditions on ELISPOT results. Data were derived from cryopreserved PBMC of 12 HIV-infected and 12 uninfected subjects. Cells were cryopreserved at a central laboratory and shipped to the testing laboratories in the conditions indicated on each graph: LN/DI=stored in liquid nitrogen and shipped on dry ice; LN/LN=stored and shipped in liquid nitrogen; 70/DIstored at −70°C and shipped on dry ice. 3, 6, 9, 12 indicate the numbers of weeks of storage at −70°C before shipment. The testing laboratories thawed and assayed samples from each subject in the same run. Results were pooled and presented as means and 95% CI of the entire population, because there were no differences by HIV status (p>0.5). Fig 2a shows that CMV pp65 peptide- stimulated ELISPOT values were not significantly different in LN/LN, LN/DI and 70/DI 3 samples. There was a linear decrease of ELISPOT values with the time of storage at −70°C (p=0.03). Fig 2b shows that CEF peptide-stimulated ELISPOT values were similar in LN/LN, LN/DI and 70/DI 3 samples. There was a trend linear decrease of ELISPOT values with the time of storage at −70°C that did not reach statistical significance (p=0.08). Fig 2c shows that candida- stimulated ELISPOT values were similar in LN/LN, LN/DI and 70/DI 3 samples. There was a linear decrease of ELISPOT values with the time of storage at −70°C (p=0.007).

Figure 3

Effect of shipment and storage of cryopreserved PBMC on the reproducibility of ELISPOT results relative to the LN/LN gold standard. Data were derived from PBMC obtained from 12 HIV-infected and 12 uninfected individuals. Cells were cryopreserved at a central laboratory and shipped to the testing laboratories in the conditions indicated on each graph: LN/DI=stored in liquid nitrogen and shipped on dry ice; LN/LN=stored and shipped in liquid nitrogen; 70/DIstored at −70°C and shipped on dry ice. 3, 6, 9, 12 indicate the numbers of weeks of storage at −70°C before shipment. The testing laboratories thawed and assayed samples from each subject in the same run. Data are presented as the proportion of samples in each storage/shipment condition that had a <2-fold difference in SFC compared with the gold standard results obtained in the LN/LN samples. 95, 94 and 81% of LN/DI samples had SFC within 2 fold of the LN/LN sample SFC for CMV pp65, CEF and candida, respectively. For 70/DI 3, corresponding numbers were 76, 81 and 81%, respectively. The proportion of samples with ELISPOT results within 2 fold of the LN/LN gold standard decreased with longer time of storage at −70°C.