Adoptive transfer of donor-derived virus- specific T-cells can treat or prevent severe infections in allogeneic hematopoietic stem cell transplant recipients, but maybe ineffective in HLA-disparate recipients if the T-cells are restricted by unshared alleles. We previously described an AAPC consisting of mouse 3T3 fibroblasts transduced to express human B7.1, LFA-3, ICAM-1, β2-microglobulin and the alpha chain of HLA A*0201 (

Papanicolaou et al. Blood 2003;102:2498
). To test the potential of AAPC to present CMV pp65 epitopes and elicit virus-specific T-cells restricted by other, less common alleles, we generated a panel of AAPC, each expressing one of a series of HLA class-I alleles, including HLA A*0201, A*0301, A*2402, B*0702, B*0801 and C*0401. We have subsequently tested the AAPC expressing HLA A*0201, A*2402 and B*0702 for their capacity to sensitize human T-cells when loaded with overlapping 15-mer peptides spanning the CMV pp65 sequence or transduced to express the CMV pp65 protein. T-cells from groups of up to 4 CMV seropositive donors, each expressing one of these HLA alleles, when sensitized with either peptide loaded or transduced AAPC expanded 15–38 fold in vitro over 21 days. The T-cells generated were predominantly CD8+ T-cells containing central memory and effector memory subsets. As shown in table.1, T-cells elicited by each AAPC exhibited a high level of HLA-restricted, CMV-specific cytotoxic activity despite the variability in the percentage of HLA-tetramer positive and IFNγ producing T-cells. Responses against unmodified autologous or allogeneic targets was <5%. Similarly, over 2% of the T-cells in the cultures generated IFNγ in response to secondary stimulation with autologous PBMC loaded with the peptide pool, versus 0.1–0.3% in controls. In comparative assays, the yields of specific CTL generated in response to AAPC transduced to express CMV pp65 were higher than those generated in response to pool-loaded AAPC in each of the 9 donors tested. Strikingly, staining by HLA-tetramers demonstrated that sensitization by mouse derived, CMV pp65 transduced AAPC generates high proportions of T cells (7–51%) recognizing epitopes known to be immunogenic when presented by that human HLA allele on CMV infected cells i.e. NLVPMAVTV by HLA A*0201, QYDPVAALF by A*2402 and RPHERNGFTV by B*0702.

These data indicate that the human HLA expressing 3T3 cells process CMV pp65 protein and transport and present epitopes which are similar or identical to those presented on human, CMV-infected cells. Thus, a replenishable panel of 3T3 based AAPC, each expressing a single HLA allele, can be used for immediate sensitization and efficient expansion of human CMV-specific T-cells. Use of this panel permits rapid generation of virus-specific T-cells of desired HLA restriction for adoptive immunotherapy and maybe of particular value for treatment of patients receiving grafts from HLA-disparate donors.

Table 1.

Analysis of CTL Generated with AAPC (Transduced with CMV pp65 or Peptide Pulsed)

DonorsFold T-cell Expansion% CD8CD8 % Tetramer [+]CD8 % IFNγ51-Cr Cytotoxicity
All values represent average +/− SEM 
A*0201 Donors N=4 28 +/− 3 84% +/− 3 51% +/− 0.1 2.6% +/− 0.03 54% +/− 0.1 
A*2402 Donors N=4 15 +/− 2 62% +/− 5 15% +/− 0.2 2.43% +/− 0.003 34% +/− 0.03 
B*0702 Donors N=1 38 85% 7% 2% 53% 
DonorsFold T-cell Expansion% CD8CD8 % Tetramer [+]CD8 % IFNγ51-Cr Cytotoxicity
All values represent average +/− SEM 
A*0201 Donors N=4 28 +/− 3 84% +/− 3 51% +/− 0.1 2.6% +/− 0.03 54% +/− 0.1 
A*2402 Donors N=4 15 +/− 2 62% +/− 5 15% +/− 0.2 2.43% +/− 0.003 34% +/− 0.03 
B*0702 Donors N=1 38 85% 7% 2% 53% 

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution