ORIGINAL RESEARCH published: 25 February 2020 doi: 10.3389/fimmu.2020.00271 Characterization of a Cytomegalovirus-Specific T Lymphocyte Product Obtained Through a Rapid and Scalable Production Process for Use in Adoptive Immunotherapy Marta Grau-Vorster 1,2 , María López-Montañés 1,2 , Ester Cantó 1,2 , Joaquim Vives 1,3,4 , Irene Oliver-Vila 1 , Pere Barba 5 , Sergi Querol 1 and Francesc Rudilla 1,2* 1 Cell Therapy Service, Banc de Sang i Teixits, Barcelona, Spain, 2 Transfusion Medicine Group, Vall d’Hebron Research Edited by: Institute, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Musculoskeletal Tissue Engineering Group, Vall d’Hebron Nadia Caccamo, Research Institute (VHIR), Universitat Autònoma de Barcelona, Barcelona, Spain, 4 Medicine Department, Universitat University of Palermo, Italy Autònoma de Barcelona, Barcelona, Spain, 5 Hematology Department, Hospital Universitari Vall d’Hebron, Universitat Reviewed by: Autònoma de Barcelona, Barcelona, Spain Gabriella Pietra, University of Genoa, Italy Petra Reinke, Immunosuppressed patients are susceptible to virus reactivation or de novo infection. Charité – Universitätsmedizin Berlin, Adoptive immunotherapy, based on virus-specific T lymphocytes (VST), can prevent or Germany treat viral diseases. However, donor availability, HLA-compatibility restrictions, high costs, *Correspondence: and time required for the production of personalized medicines constitute considerable Francesc Rudilla frudilla@bst.cat limitations to this treatment. Ex vivo rapid and large-scale expansion of VST, compliant with current good manufacturing practice (cGMP) standards, with an associated cell Specialty section: donor registry would overcome these limitations. This study aimed to characterize a This article was submitted to T Cell Biology, VST product obtained through an expansion protocol transferable to cGMP standards. a section of the journal Antigenic stimulus consisted of cytomegalovirus (CMV) pp65 peptide pool-pulsed Frontiers in Immunology autologous dendritic cells (DCs) derived from monocytes. G-Rex technology, cytokines Received: 02 August 2019 IL-2, IL-7, and IL-15, and anti-CD3 and anti-CD28 antibodies were used for culture. Accepted: 03 February 2020 Published: 25 February 2020 At day 14 of cell culture, the final product was characterized regarding T cell subsets, Citation: specificity, and functionality. The final product, comprised mainly CD4+ and CD8+ T Grau-Vorster M, López-Montañés M, lymphocytes (49.2 ± 24.7 and 42.3 ± 25.2, respectively). The culture conditions made Cantó E, Vives J, Oliver-Vila I, Barba P, Querol S and Rudilla F (2020) it possible to achieve at least a 98.89-fold increase in pp65-specific CD3+ IFN-γ+ cells. Characterization of a These cells were specific, as pp65-specific cytotoxicity was demonstrated. Additionally, Cytomegalovirus-Specific T in complete HLA mismatch and without the presence of pp65, alloreactivity resulted in Lymphocyte Product Obtained Through a Rapid and Scalable <5% cell lysis. In conclusion, a cGMP scalable process for the generation of a large Production Process for Use in number of doses of CMV-specific cytotoxic T cells was successfully performed. Adoptive Immunotherapy. Front. Immunol. 11:271. Keywords: virus specific T lymphocytes (VST), antigen presenting cells (APC), adoptive immunotherapy, specificity, doi: 10.3389/fimmu.2020.00271 cytotoxicity, peripheral blood mononuclear cells (PBMC), alloreactivity Frontiers in Immunology | www.frontiersin.org 1 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST INTRODUCTION Protocol optimization focused on reducing production time and maximizing product functionality. We also aimed to generate Immunodeficient patients are susceptible to infection by large amounts of cells to ensure the availability of several doses cytomegalovirus (CMV) and other viruses. Herpesviruses, such for a single patient if needed. Importantly, this method is readily as CMV, are often asymptomatic or mild in healthy individuals. adaptable to clinical implementation in compliance with current Nevertheless, in the context of immunocompromised patients, good manufacturing practice (cGMP) legislation and the product viral infections can be a severe cause of morbidity and mortality may be used as a safe and effective antiviral agent for patients at (1, 2). CMV infection rates indicate that the virus is common in high risk of disease due to CMV infection. most of the population and that incidence increases with age (3). The impaired immune systems of immunocompromised patients are unable to eradicate or limit the virus. Antiviral pharmacologic MATERIALS AND METHODS agents are effective against only some of these viruses; their use is PBMC Handling costly, associated with significant toxicities and does not provide Peripheral blood mononuclear cells (PBMC) were collected long-term protection (4, 5). Adoptive immunotherapy based on from the peripheral blood of 9 CMV+ seropositive healthy virus-specific T lymphocytes (VST) is therefore an attractive donors with donor informed consent and the corresponding option, as T lymphocytes can confer long-term protection against approval of the ethics committee (complies with good clinical the development of viral disease (6–8). practice CPMP/ICH/135/95 and Real Decreto 1090/2015). Donor lymphocyte infusion (DLI) is a conventional PBMC isolation, cryopreservation, and thawing were performed therapy for leukemia relapse after Hematopoietic Stem Cell as described in previous studies (23). Positive serology for CMV Transplantation (HSCT) and it has also shown activity against IgG was confirmed using chemiluminescence (Abbot, Abbot virus infections in this setting (9, 10). But the use of non-specific Park, Illinois, USA). PBMC from each donor were used to obtain lymphocytes to treat viral infection has the risk to develop CMV-specific VST, moDCs and phytohemagglutinin (PHA) graft-versus-host disease (GvHD) (11). In addition, DLI may lymphoblasts. The determination of cell concentration was not be effective when stem cell donors are seronegative as it performed by flow cytometry (Perfect-Count MicrospheresTM ; is the case in umbilical cord blood transplantation. However, Cytognos, Salamanca, Spain) or Neubauer chamber counting. therapeutic alternative based on antigen-specific T cells is safe The percentage of cell viability was determined either by and effective, as it does not increase the risk of GvHD and 7-aminoactinomycin D (7AAD; BD Biosciences, San Jose, shows overall response rates of 90% (12). Direct selection with California, USA) staining or Trypan Blue (GE Healthcare, USA). peptide-multimers (13) or the cytokine capture system (14, 15) Data was analyzed using CellQuest Pro software version 5.2.1 are currently used to obtain VST. Both procedures present (BD Biosciences). certain limitations, such as the challenge of finding an available compatible and seropositive donor, and the amount of final Obtaining and Pulsing Dendritic Cells product obtained, which defines the number of doses. While Freshly isolated PBMC were initially seeded for 2 h to allow multimer selection presents the disadvantage of being limited to the adherence of monocytes. Non-adherent PBMC were then certain HLA types, the cytokine capture system will only select collected and cryopreserved. Monocytes were cultured for 8 days T cells producing a specific cytokine. Another hurdle of these to obtain mature dendritic cells (DC), as described elsewhere methodologies is the very low frequency of CMV-specific T cells (24). Mature DC were pulsed for 1–2 h at 37◦ C, 5% CO2 , with present in peripheral blood. According to Gamadia et al. (16), the 10 µg/mL of PepTivator R CMV pp65 (Miltenyi Biotec, Bergisch frequency of CD8+ interferon gamma (IFN-γ) producing cells Gladbach, Germany), referred to here as pp65. PepTivator CMV ranged between 0.18 and 0.80%, with similar observed CD4+ pp65 is based on peptide pools of mainly 15-mer peptides with IFN-γ+ frequencies. 11-amino acid (aa) overlap, covering the complete sequence of In order to create a VST bank, the cell expansion protocol the pp65 protein of human cytomegalovirus. Pulsed-DC were must be well-established and must minimize the presence of used for PBMC coculture. alloreactive T cells in the final product, and the expanded T cell populations must contain both CD8+ and CD4+ cells to ensure Expansion Protocol an effective response to the infection (17, 18). As shown in Figure 1, thawed PBMC were seeded at 0.25- We describe a process for generating a single preparation of 1.0E+06 cells/cm2 at a DC:PBMC ratio of 1:10 in a G-Rex antiviral T lymphocytes (CD4+ and CD8+ ) that is consistently culture system (Wilson Wolf Manufacturing, New Brighton, specific for immunodominant and subdominant antigens derived MN). Fresh medium was composed of Roswell Park Memorial from CMV, a frequent cause of post-transplant morbidity or Institute GlutaMAX media (RPMI, Gibco Laboratories, NY, death. This approach uses a standardized mix of peptides USA) supplemented with 10% (v/v) human AB serum (hSerAB) focusing on pp65 protein, the major immunodominant CMV (Banc de Sang i Teixits, Barcelona, Spain). After 5 days of antigen, presented by mature monocyte-derived dendritic culture, the medium was completely replaced with RPMI+10% cells (moDC), and a combination of cytokines to promote hSerAB supplemented with IL-2 120 U/mL (Miltenyi Biotec), the activation, survival, and expansion of T cells. After a IL-7 4,400 U/mL (R&D Systems, MN, USA), and IL-15 bibliographic search of the media supplements most commonly 80 U/mL (Miltenyi Biotec). Moreover, monoclonal antibodies used for lymphocyte expansion, cytokines IL-2, IL-7, and anti-CD3 (1 µg/mL) and anti-CD28 (2 µg/mL) were added IL-15 (19–22) were selected for VST activation and expansion. (Biolegend, San Diego, California, USA). On days 7, 9, Frontiers in Immunology | www.frontiersin.org 2 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST FIGURE 1 | Expansion protocol schematic for the rapid generation of virus-specific T cells (VST). Starting material based on HLA typed peripheral blood IgG+ for CMV. Isolation of PBMC and selection of monocytes by plastic adherence. Non-adherent PBMC cryopreserved, and attached monocytes differentiated to DC. Seeding performed in a G-Rex device in a ratio 1:10 pp65-pulsed DC:thawed PBMC. Expansion lasted 14-day, with addition at day 5 of anti-CD3+ anti-CD28 antibodies, and IL-2, IL-7, IL-15. Renewal of cytokines was performed at days 5, 7, 9, and 12. At day 14 the final product was characterized and cryopreserved. and 12, an aliquot of 100–300 µL was taken from each Data acquisition was performed using a Miltenyi MACS culture for cell counting and viability analysis, and fresh Quant flow cytometer. Phenotype was analyzed using FlowJo medium consisting of RPMI+10% hSerAB with the addition software v10. of IL-2 120 U/mL, IL-7 4,400 U/mL and IL-15 80 U/mL was added. Culture split was performed when cell density reached Kinetics of Cytokine Secretion 8E+06 cells/cm2 . Cryopreserved expanded cells were stimulated with CMV pp65 peptide pool (Miltenyi) 50 ng/mL at a cell concentration of 1E+06 cells/mL in a 96-well plate. Supernatant samples were Obtaining and Pulsing Lymphoblasts collected at 0, 6, 10, 24, and 48 h. All samples were centrifuged Fresh or cryopreserved PBMC were adjusted to 2E+06 cells/mL at 1,000 g for 15 min at 4◦ C. To remove aggregates or debris, and stimulated with PHA 5 µg/mL. After 24 h, cells were washed samples were centrifuged again at 10,000 g for 10 min at 4◦ C twice with Dulbecco’s phosphate-buffered saline solution (PBS, and stored at −80◦ C until analysis. Samples were thawed Gibco Laboratories) by centrifugation at 340 g, RT, for 10 min. just once and kept on ice before assay. The Bio-Plex ProTM The solution was then adjusted again to 2E+06 cells/mL with Human Cytokine Th1/Th2 (Bio-Rad Laboratories, Hercules, RPMI+10%hSerAB and supplemented with IL2 (100 U/mL). CA, USA) was used for the determination of levels of 9 At least 4 mL of this sample was seeded (2 mL/well) in cytokines: IL-2, IL-4, IL-5, IL-10, IL-12 (p70), IL-13, IFN-γ, a 24-well plate and the medium was replenished every 2– GM-CSF, and TNF-α in a multiplex assay using a Luminex 4 days. The culture was stored in the incubator at 37◦ C, 100IS analyzer (Luminex Corp. Austin. TX, USA) according to 5% CO2 as long as G-Rex culture lasted. On the last day the manufacturer’s instructions. Duplicates were tested for each of culture, half of the blasts were pulsed with 1 µg/mL of sample. Data analysis was performed using Bioplex Manager pp65 peptide pool (Miltenyi) for 1–2 h at 37◦ C 5% CO2 and Software v6.1 (Bio-Rad Laboratories Inc.). the other half were left non-pulsed. When blasts were used in an allogeneic context, samples were irradiated at 30 Gy Enzyme-Linked Immunospot (ELISPOT) for 15 min. ELISPOT (MABTECH, Nacka, Sweden) assay was performed in order to determine the specificity of T cells against CMV pp65 peptide by means of interferon gamma secretion. Cells PBMC Immunophenotype were stimulated with 50 ng/mL pp65 peptide pool (Miltenyi) and For the phenotype analysis, PBMC were collected and stimulated left overnight at 2E+05 cells/well. Plates were read in an AID with 1 µg/mL of pp65 for 6 h at 37◦ C, 5% CO2 , at 1E+07 ELISPOT reader (AID GMBH, Strassberg, Germany) following cells/mL. After 3 h of stimulation, 1 µg/mL of brefeldin A manufacturer’s instructions. was added. The cells were then washed and permeabilized using Cytofix/Cytoperm solution (BD). Staining consisted of CD107a Degranulation Assay Live/Dead and monoclonal antibodies: CD45, CD8, CD3, IFN- The final product was incubated for 4 h at 37◦ C with γ, CD19, CD20, CD56, CD4, FOXP3, CD25, C-C chemokine 10% CO2 in the presence of monensin (5 ng/mL, Sigma- receptor 7 type (CCR7), and CD45RA (Miltenyi Biotech). Aldrich, Saint Louis, Missouri, USA) and anti-CD107a FITC For the analysis of the frequency of HLA- (BD Biosciences Pharmingen) with PepTivator R CMV pp65 A∗ 0201/NLVPMVATV CD8+ T cells, APC labeled Pro5 (Miltenyi). Stimulation with PHA (Sigma-Aldrich) was used as a MHC Pentamer (Proimmune, Oxford, UK) was used. Staining positive control for degranulation. Unstimulated cells were used was performed according to manufacturers’ instructions. as a negative control for degranulation. Data acquisition was Frontiers in Immunology | www.frontiersin.org 3 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST performed using a MACSQuant instrument (Miltenyi Biotech). 42.2-fold total expansion within 14 days. The growth kinetics Data analysis was performed using FlowJo software v10. of these cells is shown in Figure 2A. Cells grow slowly up to day 7 and start proliferating exponentially at day 9, with Cytotoxicity Assay Based on Flow the highest proliferation at days 12–14 of culture (Figure 2A). Cytometry More specifically, the overall expansion factor of CD3+ cells for A cytotoxicity assay was performed to detect the manufactured PBMC cocultured with pulsed DC was an average of 49.3 ± 48.6 lymphocytes’ ability to attack either autologous or allogeneic (median, 13.6; min, 6.7; max, 127.2; n = 9) while for PBMCs lymphoblasts with or without presentation of CMV pp65 cocultured with non-pulsed DC, it was an average of 9.6 ± 6.3 peptides (Miltenyi). Lymphoblasts were labeled with CFSE (median, 8.6; min, 4.1; max, 17; n = 4). (CellTraceTM CFSE Cell Proliferation Kit; Invitrogen, Waltham, MA, USA), as already reported (23). Allogeneic and autologous Final Product Mainly Comprising T Cells non-pulsed lymphoblasts were stained with a high concentration The presence of a variety of cell populations in the final product of CFSE (2.5 µM), while pp65 peptide pool-pulsed lymphoblasts (CD4+ T cells, CD8+ T cells, NK cells, B cells and Treg were labeled with a low concentration of CFSE (0.25 µM) (25). cells) was evaluated. Expanded cells largely consisted of CD3+ Fluorescence loss in the target population was monitored by T cells (mean, 96.9 ± 1.9%) containing both CD4+ (mean, means of flow cytometry in order to quantify the percentage 49.2 ± 24.7%) and CD8+ (mean, 42.3 ± 25.2%) populations of specific cytotoxicity (26). Two different ratios of target (T) (Figure 2B). Compared to the phenotype of the initial product, and effector (E) cells were tested: T:E 1:5 and 1:10, maintaining significant differences were found in the CD3+ cell subset, which 1E+04 target cells in both conditions. Triplicates were performed had largely expanded (p = 0.0004), with no changes among the for each condition. Samples were seeded in a 96-well plate and CD4+ cell subset (p = 0.2891) but with a significant increase in were incubated for 4 and 24 h of coculture. Data acquisition was the CD8+ cell population (p = 0.0315). B cells, NK cells, and Treg performed using a FACS Calibur instrument (BD). Data analysis cells did not expand and, consequently, their presence decreased was performed using FlowJo software v10. to almost undetectable levels in the final product (B cells, 0.3 ± Percentage of specific cytotoxicity was calculated using the 0.6%; p = 0.0013; NK cells, 0.5 ± 0.6%; p = 0.0225; and Treg cells, ratio of pulsed and non-pulsed target cells without the presence 0.1 ± 0.1%; p = 0.0395). of effector cells as a baseline, according to the following equation (27): Expanded VST Cells Show Antiviral Specificity Using IFN-γ Sample(CFSElow/CFSEhigh)   Cell lysis (%) = 100 − 100∗ In order to test the lymphocytes’ CMV specificity, cells were re- Baseline(CFSElow/CFSEhigh) exposed to pp65-pepmix and activation was measured using IFN- (1) γ intracellular production by flow cytometry. IFN-γ secretion of CD3+ , CD3+ CD4+ , and CD3+ CD8+ populations is shown HLA Typing in a representative dot plot (Figure 2C). In the independent HLA typing for all samples was performed using the next expansions performed, at day 14 of coculture with pp65-pulsed generation sequencing method. Briefly, DNA samples were moDC, the total number of CD3+ IFN-γ+ , CD3+ CD4+ IFN- amplified by means of multiplex PCR using an in-house γ+ , and CD3+ CD8+ IFN-γ+ cells specific for CMV-pp65 was strategy. After DNA library preparation (GenDX, Utrecht, The significantly higher compared to day 0 (Figure 2D) (p = 0.0002). Netherlands), pooled samples were paired-end sequenced using Absolute values of CD3+ IFN-γ+ , CD4+ IFN-γ+ and CD8+ IFN- a Miseq (Illumina, San Diego, California, USA) following γ+ cells were calculated from day 14 with respect to day 0. The manufacturer’s instructions. Data analysis was performed fold expansion results for CD3+ IFN-γ+ cells were min, 98.89; using NGSengine software (GenDX) and IMGT HLA data max, 12,528; and median, 1480.18. The results for CD4+ IFN-γ+ versions 3.29.0–3.35.0. cells were min, 60.37; max, 6158.77; and median, 2082.53. The results for CD8+ IFN-γ+ cells were min, 41.51; max, 32146.13; Data Analysis median, 562.41; n = 9. Microsoft Office Excel R and GraphPad Prism 6 were used for A summary of the results for 9 donor-independent expansions the analysis of the results and for plot generation. Statistical is shown in Figure 2E. Our data confirmed that, with the significance was set at: ∗ p < 0.05, ∗∗ p < 0.01, and ∗∗∗ p < 0.001, expansion system used, the final product contains antigen- Mann-Whitney test. specific IFN-γ-producing T cells (CD3+ IFN-γ+ 25.3 ± 16.2%) with no significant differences between compartments RESULTS (CD3+ CD4+ IFN-γ+ , 15.7 ± 18.1%; CD3+ CD8+ IFN-γ+ , 29.3 ± 20.3%; p = 0.3536). Moreover, significant differences were Large Expansion of VST in 14-Day found regarding specificity gain in all subsets after 14-day of Coculture culture expansion (CD3+ p < 0.0001, CD3+ CD4+ p < 0.0001, Nine VST batches with pp65 specificity were manufactured and CD3+ CD8+ p < 0.0001). from nine healthy CMV+ donors. We obtained a median of Moreover, we analyzed the frequency of HLA- 43.3E+06 cells (min., 4.5E+06 cells; max., 100.1E+06 cells) A∗ 0201/NLVPMVATV pentamer+ CD8+ T cells. According from an initial seed of 1E+06 PBMC, representing an average to the results, cells expanded with pp65-pulsed DC show a Frontiers in Immunology | www.frontiersin.org 4 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST FIGURE 2 | Pp65-specific T cell expansion. (A) Number of total cells at different time points during the 14-day culture. PBMC cocultured with: pp65 peptide pool-pulsed DC (black, n = 9) and nonpulsed DC (gray, n = 4). (B) Percentage of general cell subpopulations pre- (day 0) and post-expansion (day 14). T cells: CD3+ , CD4+ : CD3+ CD4+ , CD8+ : CD3+ CD8+ , NK: CD3− CD56+ , B cell: CD3− CD19+ CD20+ , Treg: CD3+ CD4+ CD25+ FOXP3+ (n = 9). (C) Gating strategy for IFN-γ secreting cells from one representative donor. Dot plots show the response of T lymphocytes against pp65 antigen stimulation. (D) Number of VST at day 0 and after the 14-day culture (n = 8). (E) Percentage of IFN-γ secreting population cells at days 0 and 14 of culture (n = 9). *p < 0.05, **p < 0.01 and ***p < 0.001 (Mann-Whitney test). higher percentage of antigen specific CD8+ T lymphocytes than identified and classified into four different subsets based on pre-expansion cells (Supplementary Figure 4). the expression of CCR7 and CD45RA markers, following the gating strategy shown in Supplemental Figure 1. Here, CMV-Specific IFN-γ+ T Cells Are Effector naïve T cells (TN ): CD45RA+ CCR7+ , central memory T cells Memory T Cells (TCM ): CD45RA− CCR7+ , effector memory T cells (TEM ): To thoroughly characterize the expanded product, CD3+ , CD45RA− CCR7− , and terminally differentiated effector CD3+ CD4+ , and CD3+ CD8+ T cell populations were memory T cells (TEMRA ): CD45RA+ CCR7− . Frontiers in Immunology | www.frontiersin.org 5 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST significant changes (p = 0.1797). For CD3+ CD8+ cells, the TCM subset showed no significant changes (p = 0.9372) and the TEMRA subset showed a significant decrease (p = 0.0022). Interferon gamma, used here as a marker of CMV T cell specificity, was expressed by both CD3+ CD4+ and CD3+ CD8+ T cells. The T cell subsets expressing IFN-γ corresponded overwhelmingly to the TEM subset: CD3+ , 91.15 ± 5.29%; CD3+ CD4+ , 90.95 ± 5.25%; and CD3+ CD8+ , 89.03 ± 7.62% (Figure 3B). Moreover, when comparing the phenotype of pre- expanded cells, cells expanded with nonpulsed DC, and cells expanded with pp65-pulsed DC, we observed that the main T cell subset in specific expanded cells correspond to TEM together with a higher expression of IFN-γ expression. Cells expanded with non-pulsed DC and pre-expansion cells, have a lower expression of IFN-γ and a higher percentage of TN (Supplementary Figure 5). If we focus on the pp65-pulsed DC sample, and analyze the non-responsive cells (CD3+ IFN-γ− ), the T cell subsets are distributed similarly to the responsive cells (CD3+ IFN-γ+ ) but with a lesser percentage of TEM , and a higher percentage of other T cell subsets, including TN (Supplementary Figure 6). Furthermore, the population of expanded cells that expressed PD1 corresponded to the IFN-γ+ population (Supplementary Figure 7). Functional and Specific Final Product Against pp65 CMV Secretion kinetics of Th1 and Th2 cytokines after stimulation with pp65 of 2 cryopreserved batches of expanded cells at different time points (0, 6, 10, 24, and 48 h) are shown in Figure 4. Cytokine secretion had high concentrations of IFN-γ and TNF-α detected with a secretion peak after 6 h of stimulation. Furthermore, IL-2 was secreted in a similar manner but in much lower concentrations. T cell response to pp65 was also assessed as a potency assay using IFN-γ ELISPOT for cryopreserved samples from day 0 and 14-day expanded product. Based on the number of spot forming colonies (SFC), a significant increase (p = 0.0022) in reactivity was detected after expansion (Figure 5A). A similar difference in FIGURE 3 | Frequency of different T cell differentiation subset phenotypes in reactivity was observed compared to the control of expanded cells the final product. (A) Comparison between pre-expansion and post-expansion with nonpulsed DC (Supplementary Figure 8). for CD3+ , CD3+ CD4+ , and CD3+ CD8+ subpopulations (n = 6). (B) T cell A CD107a degranulation assay was performed to check differentiation phenotypes in post-expansion IFN-γ+ T cells (n = 4). TN indicates naïve T cell (CCR7+ CD45RA+ ); TCM , central memory T cell the functionality of three final products. The expression of (CCR7+ CD45RA− ); TEM , effector memory T cell (CCR7− CD45RA− ); and CD8+ CD107a+ was detected in the presence of CMV pp65 TEMRA , terminally differentiated effector memory T cell (CCR7− CD45RA+ ). peptide pool (Figure 5B), while only a slight expression **p < 0.01 (Mann-Whitney test). was shown when cells were left unstimulated. However, this assay showed high variability between effector cell batches (CD8+ CD107a+ 22.4 ± 25.71, n = 3, and CD4+ CD107a+ 8.67 After expansion, the TN subset decreased significantly in all ± 11.74, n = 3) (Supplementary Figure 2). T cell populations (CD3+ p =0.0022, CD3+ CD4+ p = 0.0043, We took advantage of the cytotoxicity assay to show that our and CD3+ CD8+ p = 0.0022) (Figure 3A). TEM increased product was able to induce cell death specifically in the target significantly in CD3+ , CD3+ CD4+ , and CD3+ CD8+ cells cell population (see gating strategy in Supplemental Figure 3). (p = 0.0050, 0.0022, and 0.0022, respectively). In general, CD3+ Both fresh and cryopreserved effector cells induced lysis subpopulations TCM and TEMRA decreased significantly after among peptide-pulsed blasts, in a similar manner (fresh expansion (p = 0.0087). For CD3+ CD4+ cells, the TCM subset vs. cryopreserved: 4-h p = 0.7, 24-h p = 0.8). This further decreased significantly (p = 0.0022) and TEMRA showed no supports the feasibility of a ready-to-use cryopreserved cell-bank Frontiers in Immunology | www.frontiersin.org 6 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST FIGURE 4 | Polyfunctionality of the final thawed product tested using cytokine secretion kinetics after antigenic stimulation. Test performed with Luminex technology. Cytokines tested: IL-2, IL-4, IL-5, IL-10, IL-12 (p70), IL-13, IFN-γ, GM-CSF, and TNF-α. Cells were stimulated with pp65 50 ng/mL at a cell concentration of 1E+06 cells/mL in a 96-well plate. Graphs show the kinetics of the secreted cytokines in the supernatant (pg/mL) at each time: 0, 6, 10, 24, and 48 h. (A) Representation of secretion kinetics for cytokines IL-13, IFN-γ, GM-CSF, and TNF-α from 2 batches of final product. (B) Representation of secretion kinetics for cytokines IL-2, IL-4, IL-5, and IL-10 from 2 batches of final product. with verified post-thawing functional VSTs (Figure 5C). development in different centers as an effective and feasible The expanded product in the presence of autologous blasts therapy for immunocompromised patients with herpesvirus pulsed with the peptide resulted in a high percentage of infections (1, 5, 28–31). We showed a methodology based on cytotoxicity at both 4 and 24 h (23.45 ± 18.35% and 68.61 ± ex vivo culture expansion of low-frequency VST to clinical 19.76%, respectively). numbers. The process is based on technology that allows scaling We leveraged the cytotoxicity assay to assess in vitro up and transfer to cGMP standards. Peripheral blood was safety by testing the alloreactivity and specificity of the final used as starting material instead of leukapheresis, in order to product. Supplementary Tables 1, 2 show, respectively, the high facilitate donation compared to the direct selection protocol resolution HLA typing for expanded cells [used here as effector (32). Physiological antigen presenting cells, DC, are used as [E] cells] and PHA blasts [used here as target [T] cells]. Fresh potent stimulators for specific T cell activation, together with and cryopreserved expanded cells were cocultured for 24 h with an unspecific stimulation on day 5 based on anti-CD3 and allogeneic blasts (with 2–3 HLA matches) at ratios of (T:E) 1:5 anti-CD28 antibodies. Pp65 peptide pool was used to pulse and 1:10 to test for alloreactivity (Figure 5D), which resulted in DC for antigen specific stimulation due to being the most cell lysis below 5% in all cases. However, a correlation between immunogenic protein of CMV (33). Nevertheless, immediate the number of HLA mismatches and percentage of cell lysis was early 1 protein (IE-1) specific T cells have been demonstrated not observed (Pearson correlation: ratio 1:5, p = 0.8211; ratio to have a protective effect (34). Therefore, and since the use 1:10, p = 0.9248; data not shown). Further characterization was of both pp65 and IE-1 peptides has been proven successful by performed by coculture of the final product with allogeneic blasts Tzannou and colleagues (35), our future approach would be to (2–5 HLA match), either pulsed with pp65 peptide pool or non- use both peptides for the generation of CMV specific T cells. pulsed. Expanded T cells specifically induced cell lysis of the The protocol described in this study allowed us to produce in allogeneic blasts pulsed with pp65 regardless of the number of vitro safe and effective VST after a short expansion of only 14- HLA matches (Figure 5E). Similarly, a correlation between the day showing an advantage in time over protocols based on longer number of HLA matches and the percentage of cell lysis was not expansions (6, 33). The use of G-Rex technology enabled us observed (Pearson test, p = 0.6460). to grow cells with a larger volume of media compared to the traditional plastic-based culture and therefore obtain a higher DISCUSSION density. The final product was thoroughly characterized for IFN- γ expression among CD8+ cells, which are the cells that most In recent years, expansion protocols for the creation of VST express this cytokine, although the CD4+ subset is also essential third-party banks for adoptive immunotherapy have been under for orchestrating the immune response (17, 18). According to Frontiers in Immunology | www.frontiersin.org 7 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST FIGURE 5 | Potency and functional assays performed with the expanded product. (A) IFN-γ ELISPOT for D0 and D14 thawed cells (n = 6). (B) Example of a degranulation assay (CD8+ CD107a+ ) dot plot for effector cells (VST), unstimulated (left plot) and stimulated with pp65 (right plot). (C) Cytotoxicity assay representation for pp65-pulsed autologous blasts specific lysis [4-h ratio, 1:5 (T:E); n = 3; 24-h ratio, 1:5 (T:E); n = 2]. (D) Cytotoxicity assay representation for irradiated allogeneic blasts unspecific cell lysis with 2–3 match (n = 6, fresh n = 2, thawed n = 4). (E) Cytotoxicity assay representation for effector cells (E): VST, cocultured with target cells (T), pp65-pulsed allogeneic blasts. Percentage of cell lysis calculated from a ratio of 1:10 (T:E). **p < 0.01 (Mann-Whitney test). the average of 42.2 fold-expansion obtained from the 9 samples, interesting that the final product contained low levels of TN cells, a single blood donation may yield up to 200 doses of 2E+07 thereby reducing the probability of GvHD (38). The final product VST/m2 , an average fold expansion higher than other expansion is mainly composed of TEM , which provide the effector function protocols; which involves substantial cost reduction compared needed. TEM are known to mediate protective memory and they to other methodologies like direct selection. Enhancement of migrate to inflamed peripheral tissues and display immediate final product reactivity against CMV pp65 peptides was also effector functions (39). The highest percentage of IFN-γ secretion confirmed by means of IFN-γ ELISPOT assays. The presence was attributed to this subset, which can therefore rapidly control of T helper 1 (Th1) CD4+ cells (CD4+ IFN-γ+ ), present in our the infection. Despite the low concentration of TCM in the final final product, is associated with control of persistent infections product, we hypothesize that these cells may engraft and confer (36). The proliferation of NK and B cells was prevented and, long-lasting protection with later differentiation to effector despite the use of IL-2, regulatory T cells were not present in the cells upon antigenic stimulation. In vivo T cell persistence of final product, thus facilitating the activity of the VSTs generated third-party partially HLA-matched VST has been confirmed by (37). Since naïve T cells are potentially alloreactive, it is also other authors for up to 12 weeks (40). Ex vivo expanded cells Frontiers in Immunology | www.frontiersin.org 8 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST could also be manufactured, in the case of hematopoietic stem Future perspectives include the translation of VST production cell transplantation, if available, from the same donor of the and validation to cGMP standards. Future prospects also include transplant, for which rates of functional VST engraftment have the generation of other virus-specific T cells, such as EBV, AdV, been shown to persist for up to 9 years (41). The option of ex vivo and BKV, or even multivirus-specific T cells as shown feasible in expansion on purpose for a single patient or in an autologous the literature (5). context is also conceivable. However, although personalized In summary, the use of pp65 peptide pool pulsed DCs expansions probably have some advantages, such as choosing the and the addition of anti-CD3 and anti-CD28 antibodies in the highest HLA matching, this would require manufacturing time presence of IL-2, IL-7, and IL-15 resulted in >98.89-fold CD3+ that is otherwise absent in a third-party bank. IFN-γ+ cell expansion at day 14. An average of 25.3% IFN- Final product polyfunctionality was demonstrated by the γ was achieved in 9 expansions, with very small amounts of detection of several secreted cytokines after cell stimulation. CD56+ , CD19+ /CD20+ , or CD4+ CD25+ FOXP3+ cells. TEM Among the inflammatory cytokines tested, we found not only cell subpopulations were increased after expansion, while TN IFN-γ after stimulation with pp65 but also high concentrations subpopulations were decreased in the final product. Moreover, of TNF-α. Products with production of multiple cytokines have the reactivity of the product against CMV peptide pp65 was been shown to provide a more effective immune response against enhanced after short culture expansion tested using the IFN-γ a pathogen (42). ELISPOT assay. In terms of effectivity and in vitro safety, strong We showed that the product has the potential to specifically cytotoxicity was shown when specific T cells were cocultured with kill pp65 peptide pool-pulsed autologous cells. To emulate pulsed pp65 autologous blasts. Final product degranulation was allogeneic therapy conditions, we tested pp65-pulsed target confirmed in line with cytotoxicity results. When expanded cells cells with 2–3 match with our effector cells. The response were cultured with non-pulsed allogeneic blasts, a maximum of varied from the HLA allele matched rather than with the 5% lysis was observed. Furthermore, cytotoxicity of expanded number of HLA matching, in line with findings in the literature cells with pp65-pulsed allogeneic blasts showed specific cell lysis. that certain pp65 epitopes are HLA restricted and some of In conclusion, we have shown a feasible and scalable process that them are immunodominant epitopes (33, 43–45). Moreover, can be easily transferred to cGMP standards and that generates a the results shown by degranulation assays further confirmed safe and functional product. that the VST obtained act as antigen-specific effector cells; cells degranulate, and therefore have the capacity to strike pp65 DATA AVAILABILITY STATEMENT pulsed cells. As for any other therapy it is essential that our product All datasets generated for this study are included in the shows not only efficacy but safety. Alloreactivity assays were article/Supplementary Material. performed to characterize the in vitro safety of the product. When testing alloreactivity, we observed that effector cells did not lyse target cells that were not pulsed with the peptide pool, AUTHOR CONTRIBUTIONS independently of the number of HLA matches. The similar Experiments were designed by MG-V, ML-M, IO-V, and FR. functionality observed for both the fresh and cryopreserved Experimental data were generated by MG-V, ML-M, and IO-V, product demonstrates the feasibility of generating a VST bank and analysis was performed by MG-V, ML-M, and EC. MG-V, that maintains the product properties. The lack of correlation ML-M, SQ, PB, and FR prepared the manuscript. All authors observed between HLA matching and specific cell lysis suggests have reviewed the manuscript and have made substantial and that HLA restriction of immunodominant epitopes is more intellectual contributions to the work. important than higher HLA compatibility between the donor and the patient in order to induce specific lysis. On the contrary, higher HLA matching would ensure engraftment of donor T FUNDING lymphocytes as well as long-term protection. Alloreactivity is an important parameter to consider when using cell therapies Work in our laboratory was supported by the Spanish Cell as a treatment, as GvHD is one of the major concerns when Therapy Network (TerCel, file No. RD16/0011/0028), and infusing allogeneic cells. Approaches to avoid GvHD include carried out as part of AdvanceCat with the support of ACCIÓ donor-patient HLA matching and reducing the number of TN (Catalonia Trade & Investment; Catalan government) under cells present in the final product. Moreover, alloreactivity in the Catalonian ERDF operational program (European Regional vitro is reduced by ex vivo culturing with virus-specific stimuli, Development Fund) 2014-2020. Our laboratory was certified as unspecific cell lysis induced by the expanded cells is <5%. by the Catalan government as a Consolidated Research Group Our results complement the cytotoxicity assays performed, which (ref. 2017SGR719). ensured the ability of VST to kill target cells without attacking those cells not loaded with the antigen. Some reports indicate that ACKNOWLEDGMENTS virus-specific memory T cells can exert allo-HLA reactivity (46). Nevertheless, published studies report no increased incidence of The authors would like to thank the current members of GvHD, even when the donor had HLA mismatch, or toxicity Servei de Teràpia Cel·lular (Banc de Sang i Teixits, Barcelona, related to VST infusion (1, 5, 8, 15, 29, 47). Spain) for their technical support and advice; members of the Frontiers in Immunology | www.frontiersin.org 9 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST Histocompatibility and Immunogenetics Laboratory for their SUPPLEMENTARY MATERIAL technical support; E. Valdivia and N. Martínez, for supplying blood samples; Dr. D. Benítez-Ribas, Dr. Georgina Flórez- The Supplementary Material for this article can be found Grau, and Dr. Alejandro Madrigal for their technical support online at: https://www.frontiersin.org/articles/10.3389/fimmu. and advice. 2020.00271/full#supplementary-material REFERENCES 15. Peggs KS, Thomson K, Samuel E, Dyer G, Armoogum J, Chakraverty R, et al. Directly selected cytomegalovirus-reactive donor T cells confer 1. Withers B, Clancy L, Burgess J, Simms R, Brown R, Micklethwaite K, et al. rapid and safe systemic reconstitution of virus-specific immunity following Establishment and operation of a third party virus-specific T-cell bank within stem cell transplantation. Clin Infect Dis. (2011) 52:49–57. doi: 10.1093/cid/ an allogeneic stem cell transplant program. Biol Blood Marrow Transplant. ciq042 (2018) 24:2433–42. doi: 10.1016/j.bbmt.2018.08.024 16. Gamadia LE, Rentenaar RJ, Baars PA, Remmerswaal EB, Surachno S, 2. Sukdolak C, Tischer S, Dieks D, Figueiredo C, Goudeva L, Heuft Weel JF, et al. Differentiation of cytomegalovirus-specific CD8+ T cells HG, et al. CMV-, EBV- and ADV-specific T cell immunity: screening in healthy and immunosuppressed virus carriers. Blood. (2001) 98:754– and monitoring of potential third-party donors to improve post- 61. doi: 10.1182/blood.V98.3.754 transplantation outcome. Biol Blood Marrow Transplant. (2013) 19:1480–92. 17. Pourgheysari B, Piper KP, McLarnon A, Arrazi J, Bruton R, Clark F, et al. doi: 10.1016/j.bbmt.2013.07.015 Early reconstitution of effector memory CD4+ CMV-specific T cells protects 3. Colugnati FA, Staras SA, Dollard SC, Cannon MJ. Incidence against CMV reactivation following allogeneic SCT. Bone Marrow Transplant. of cytomegalovirus infection among the general population and (2009) 43:853–61. doi: 10.1038/bmt.2008.403 pregnant women in the United States. BMC Infect Dis. (2007) 7:71. 18. Novy P, Quigley M, Huang X, Yang Y. CD4 T cells are required for CD8 T cell doi: 10.1186/1471-2334-7-71 survival during both primary and memory recall responses. J Immunol. (2007) 4. Fujita Y, Rooney CM, Heslop HE. Adoptive cellular immunotherapy 179:8243–51. doi: 10.4049/jimmunol.179.12.8243 for viral diseases. Bone Marrow Transplant. (2008) 41:193– 19. Hanley PJ, Cruz CR, Savoldo B, Leen AM, Stanojevic M, Khalil M, et al. 8. doi: 10.1038/sj.bmt.1705906 Functionally active virus-specific T cells that target CMV, adenovirus, 5. Papadopoulou A, Gerdemann U, Katari UL, Tzannou I, Liu H, and EBV can be expanded from naive T-cell populations in cord Martinez C, et al. Activity of broad-spectrum T cells as treatment blood and will target a range of viral epitopes. Blood. (2009) 114:1958– for AdV, EBV, CMV, BKV, and HHV6 infections after HSCT. 67. doi: 10.1182/blood-2009-03-213256 Sci Transl Med. (2014) 6:242ra83. doi: 10.1126/scitranslmed.30 20. Szmania S, Galloway A, Bruorton M, Musk P, Aubert G, Arthur A, 08825 et al. Isolation and expansion of cytomegalovirus-specific cytotoxic T 6. Mackinnon S, Thomson K, Verfuerth S, Peggs K, Lowdell M. Adoptive lymphocytes to clinical scale from a single blood draw using dendritic cellular therapy for cytomegalovirus infection following allogeneic stem cell cells and HLA-tetramers. Blood. (2001) 98:505–12. doi: 10.1182/blood. transplantation using virus-specific T cells. Blood Cells Mol Dis. (2008) 40:63– V98.3.505 7. doi: 10.1016/j.bcmd.2007.07.003 21. Krug LM, Dao T, Brown AB, Maslak P, Travis W, Bekele S, et al. WT1 peptide 7. Smith C, Økern G, Rehan S, Beagley L, Lee SK, Aarvak T, et al. Ex vivo vaccinations induce CD4 and CD8 T cell immune responses in patients with expansion of human T cells for adoptive immunotherapy using the novel mesothelioma and non-small cell lung cancer. Cancer Immunol Immunother. Xeno-free CTS immune cell serum replacement. Clin Transl Immunol. (2015) (2010) 59:1467–79. doi: 10.1007/s00262-010-0871-8 4:e31. doi: 10.1038/cti.2014.31 22. Gerdemann U, Keirnan JM, Katari UL, Yanagisawa R, Christin AS, Huye LE, 8. Gerdemann U, Katari UL, Papadopoulou A, Keirnan JM, Craddock JA, et al. Rapidly generated multivirus-specific cytotoxic T lymphocytes for the Liu H, et al. Safety and clinical efficacy of rapidly-generated trivirus- prophylaxis and treatment of viral infections. Mol Ther. (2012) 20:1622–32. directed T cells as treatment for adenovirus, EBV, and CMV infections after doi: 10.1038/MT.2012.130 allogeneic hematopoietic stem cell transplant. Mol Ther. (2013) 21:2113– 23. Oliver-Vila I, Ramírez-Moncayo C, Grau-Vorster M, Marín-Gallén 21. doi: 10.1038/mt.2013.151 S, Caminal M, Vives J. Optimisation of a potency assay for 9. Papadopoulos EB, Ladanyi M, Emanuel D, Mackinnon S, Boulad F, Carabasi the assessment of immunomodulative potential of clinical grade MH, et al. Infusions of donor leukocytes to treat Epstein-Barr virus-associated multipotent mesenchymal stromal cells. Cytotechnology. (2018) 70:31–44. lymphoproliferative disorders after allogeneic bone marrow transplantation. doi: 10.1007/s10616-017-0186-0 N Engl J Med. (1994) 330:1185–91. doi: 10.1056/NEJM199404283301703 24. Flórez-Grau G, Cabezón R, Borgman KJE, España C, Lozano JJ, Garcia-Parajo 10. Kolb HJ, Schattenberg A, Goldman JM, Hertenstein B, Jacobsen N, MF, et al. Up-regulation of EP 2 and EP 3 receptors in human tolerogenic Arcese W, et al. Graft-versus-leukemia effect of donor lymphocyte dendritic cells boosts the immunosuppressive activity of PGE 2. J Leukoc Biol. transfusions in marrow grafted patients. Blood. (1995) 86:2041–50. (2017) 102:881–95. doi: 10.1189/jlb.2A1216-526R doi: 10.1182/blood.V86.5.2041.bloodjournal8652041 25. Wonderlich J, Shearer G, Livingstone A, Brooks A, Soloski MJ, Presby MM. 11. Scarisbrick JJ, Dignan FL, Tulpule S, Gupta ED, Kolade S, Shaw B, et al. Induction and measurement of cytotoxic T lymphocyte activity. Curr Protoc A multicentre UK study of GVHD following DLI: rates of GVHD are high Immunol. (2018) 120:3.11.1–29. doi: 10.1002/cpim.38 but mortality from GVHD is infrequent. Bone Marrow Transplant. (2015) 26. Berraondo P, Nouzé C, Préville X, Ladant D, Leclerc C. Eradication of 50:62–7. doi: 10.1038/bmt.2014.227 large tumors in mice by a tritherapy targeting the innate, adaptive, and 12. Houghtelin A, Bollard CM. Virus-specific T cells for the regulatory components of the immune system. Cancer Res. (2007) 67:8847– immunocompromised patient. Front Immunol. (2017) 8:1272. 55. doi: 10.1158/0008-5472.CAN-07-0321 doi: 10.3389/fimmu.2017.01272 27. Chaoul, N., Fayolle, C., and Leclerc, C. (2016). In vivo OVA-specific cytotoxic 13. Cobbold M, Khan N, Pourgheysari B, Tauro S, McDonald D, Osman H, et al. CD8+ T cell killing assay. Bio-protocol. 6:e1838. doi: 10.21769/BioProtoc.1838 Adoptive transfer of cytomegalovirus-specific CTL to stem cell transplant 28. Leen AM, Bollard CM, Mendizabal AM, Shpall EJ, Szabolcs P, Antin JH, et al. patients after selection by HLA – peptide tetramers. J Exp Med. (2005) Multicenter study of banked third-party virus-specific T cells to treat severe 202:379–86. doi: 10.1084/jem.20040613 viral infections after hematopoietic stem cell transplantation Methods. Blood. 14. Feuchtinger T, Matthes-Martin S, Richard C, Lion T, Fuhrer M, Hamprecht K, (2014) 121:5113–24. doi: 10.1182/blood-2013-02-486324 et al. Safe adoptive transfer of virus-specific T-cell immunity for the treatment 29. Barker JN, Doubrovina E, Sauter C, Jaroscak JJ, Perales MA, Doubrovin M, of systemic adenovirus infection after allogeneic stem cell transplantation. Br et al. Successful treatment of EBV-associated posttransplantation lymphoma J Haematol. (2006) 134:64–76. doi: 10.1111/j.1365-2141.2006.06108.x after cord blood transplantation using third-party EBV-specific cytotoxic Frontiers in Immunology | www.frontiersin.org 10 February 2020 | Volume 11 | Article 271 Grau-Vorster et al. Production and Characterization of VST T lymphocytes. Blood. (2010) 116:5045–9. doi: 10.1182/blood-2010-04- 40. Tzannou I, Papadopoulou A, Naik S, Leung K, Martinez CA, Ramos 281873 CA, et al. Off-the-shelf virus-specific T cells to treat BK virus, human 30. Wilkie GM, Taylor C, Jones MM, Burns DM, Turner M, Kilpatrick D, et al. herpesvirus 6, cytomegalovirus, epstein-barr virus, and adenovirus infections Establishment and characterization of a bank of cytotoxic T lymphocytes after allogeneic hematopoietic stem-cell transplantation. J Clin Oncol. (2017) for immunotherapy of Epstein-Barr virus-associated diseases. J Immunother. 35:3547–57. doi: 10.1200/JCO.2017.73.0655 (2004) 27:309–16. doi: 10.1097/00002371-200407000-00007 41. Heslop HE, Slobod KS, Pule MA, Hale GA, Rousseau A, Smith CA, et al. 31. Haque T, Wilkie GM, Jones MM, Higgins CD, Urquhart G, Wingate P, et al. Long-term outcome of EBV-specific T-cell infusions to prevent or treat EBV- Allogeneic cytotoxic T-cell therapy for EBV-positive posttransplantation related lymphoproliferative disease in transplant recipients. Blood. (2010) lymphoproliferative disease: results of a phase 2 multicenter clinical 115:925–35. doi: 10.1182/blood-2009-08-239186 trial. Blood. (2007) 110:1123–32. doi: 10.1182/blood-2006-12- 42. Foley JF. Polyfunctional T cells. Sci Signal. (2012) 5:ec42. 063008 doi: 10.1126/scisignal.2002929 32. Tischer S, Priesner C, Heuft HG, Goudeva L, Mende W, Barthold M, et al. 43. Kondo E, Akatsuka Y, Kuzushima K, Tsujimura K, Asakura S, Tajima K, et al. Rapid generation of clinical-grade antiviral T cells: selection of suitable T-cell Identification of novel CTL epitopes of CMV-pp65 presented by a variety of donors and GMP-compliant manufacturing of antiviral T cells. J Transl Med. HLA alleles. Blood. (2004) 103:630–8. doi: 10.1182/blood-2003-03-0824 (2014) 12:336. doi: 10.1186/s12967-014-0336-5 44. Slezak SL, Bettinotti M, Selleri S, Adams S, Marincola FM, Stroncek DF. CMV 33. Koehne G, Hasan A, Doubrovina E, Prockop S, Tyler E, Wasilewski G, pp65 and IE-1 T cell epitopes recognized by healthy subjects. J Transl Med. et al. Immunotherapy with donor T cells sensitized with overlapping (2007) 5:17. doi: 10.1186/1479-5876-5-17 pentadecapeptides for treatment of persistent cytomegalovirus 45. Lehmann PV, Suwansaard M, Zhang T, Roen DR, Kirchenbaum GA, Karulin infection or viremia. Biol Blood Marrow Transplant. (2015) AY, et al. Comprehensive evaluation of the expressed CD8+ T cell epitope 21:1663–78. doi: 10.1016/j.bbmt.2015.05.015 space using high-throughput epitope mapping. Front Immunol. (2019) 34. Bunde T, Kirchner A, Hoffmeister B, Habedank D, Hetzer R, Cherepnev 10:655. doi: 10.3389/fimmu.2019.00655 G, et al. Protection from cytomegalovirus after transplantation is correlated 46. Amir AL, D’Orsogna LJ, Roelen DL, van Loenen MM, Hagedoorn RS, de Boer with immediate early 1-specific CD8 T cells. J Exp Med. (2005) 201:1031– R, et al. Allo-HLA reactivity of virus-specific memory T cells is common. 6. doi: 10.1084/jem.20042384 Blood. (2010) 115:3146–57. doi: 10.1182/blood-2009-07-234906 35. Tzannou I, Watanabe A, Naik S, Daum R, Kuvalekar M, Leung 47. Bollard CM, Heslop HE. T cells for viral infections after allogeneic KS, et al. “Mini” bank of only 8 donors supplies CMV-directed hematopoietic stem cell transplant. Blood. (2016) 127:3331–40. T cells to diverse recipients. Blood Adv. (2019) 3:2571–80. doi: 10.1182/blood-2016-01-628982 doi: 10.1182/bloodadvances.2019000371 36. Snell LM, Osokine I, Yamada DH, De la Fuente JR, Elsaesser HJ, Brooks Conflict of Interest: PB declares having received honoraria from Amgen, Celgene, DG. Overcoming CD4 Th1 cell fate restrictions to sustain antiviral CD8 T Gilead, Incyte, Jazz Pharmaceuticals, MSD, Novartis, Pfizer and Roche, not related cells and control persistent virus infection. Cell Rep. (2016) 16:3286–96. with the present article. PB has received funding from the Carlos III FIS16/01433 doi: 10.1016/j.celrep.2016.08.065 Health Institute, Asociación Española contra el Cáncer (Ideas Semilla 2019) and a 37. Dittmer U, He H, Messer RJ, Schimmer S, Olbrich AR, Ohlen PERIS 2018-2020 grant from the Generalitat de Catalunya (BDNS357800). C, et al. Functional impairment of CD8+T cells by regulatory T cells during persistent retroviral infection. Immunity. (2004) The remaining authors declare that the research was conducted in the absence of 20:293–303. doi: 10.1016/S1074-7613(04)00054-8 any commercial or financial relationships that could be construed as a potential 38. Distler E, Bloetz A, Albrecht J, Asdufan S, Hohberger A, Frey M, conflict of interest. et al. Alloreactive and leukemia-reactive T cells are preferentially derived from naïve precursors in healthy donors: implications Copyright © 2020 Grau-Vorster, López-Montañés, Cantó, Vives, Oliver-Vila, Barba, for immunotherapy with memory T cells. Haematologica. (2011) Querol and Rudilla. This is an open-access article distributed under the terms 96:1024–32. doi: 10.3324/haematol.2010.037481 of the Creative Commons Attribution License (CC BY). The use, distribution or 39. Sallusto F, Geginat J, Lanzavecchia A. Central memory and effector reproduction in other forums is permitted, provided the original author(s) and the memory T cell subsets: function, generation, and maintenance. Annu copyright owner(s) are credited and that the original publication in this journal Rev Immunol. (2004) 22:745–63. doi: 10.1146/annurev.immunol.22.012703. is cited, in accordance with accepted academic practice. No use, distribution or 104702 reproduction is permitted which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 11 February 2020 | Volume 11 | Article 271