Comparative and Functional Evaluation of In Vitro Generated to Ex Vivo CD8 T Cells Dzana D. Dervovic, Maria Ciofani, Korosh Kianizad and Juan Carlos Zúñiga-Pflücker This information is current as of September 24, 2021. J Immunol published online 27 August 2012 http://www.jimmunol.org/content/early/2012/08/26/jimmun ol.1200979 Downloaded from

Supplementary http://www.jimmunol.org/content/suppl/2012/08/28/jimmunol.120097 Material 9.DC1

Why The JI? Submit online. http://www.jimmunol.org/

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 24, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published August 27, 2012, doi:10.4049/jimmunol.1200979 The Journal of Immunology

Comparative and Functional Evaluation of In Vitro Generated to Ex Vivo CD8 T Cells

Dzˇana D. Dervovic´,*,† Maria Ciofani,*,†,1 Korosh Kianizad,*,† and Juan Carlos Zu´n˜iga-Pflu¨cker*,†

The generation of the cytotoxic CD8 T cell response is dependent on the functional outcomes imposed by the intrathymic constraints of differentiation and self-tolerance. Although thymic function can be partly replicated in vitro using OP9-DL1 cell cultures to yield CD8 ab TCR-bearing cells from hematopoietic progenitor cells, a comprehensive and functional assessment of entirely in vitro generated CD8 T cells derived from bone marrow hematopoietic stem cells has not been established and remains controversial. In this study, we demonstrate that a phenotypic, molecular, and functional signature of in vitro derived CD8 T cells is akin to that of ex vivo CD8 T cells, although several significant differences were also observed. Transfer of in vitro derived CD8 T cells into

syngeneic and immunodeficient host mice showed no graft-versus-host response, whereas a robust homeostatic proliferation was Downloaded from observed, respectively. These findings, along with a diverse and broad TCR repertoire expressed by the in vitro derived CD8 T cells, allowed for the successful generation of Ag-specific T cells to be obtained from an entirely in vitro generated CD8 T cell pool. These findings support the use of Ag-specific in vitro derived effector CD8 T cells for immune reconstitution approaches, which would be amenable to further tailoring for their use against viral infections or malignancies. The Journal of Immunology, 2012, 189: 000–000. http://www.jimmunol.org/ cells are generated in the thymus after colonization from CD8 T cells is thought to be achieved through a neglect, select, the blood by bone marrow (BM)-derived progenitors (1– eliminate process that is partly dependent on the strength of signal T 6). In the thymus, T cell development can be character- induced by the self-peptide MHC (pMHC) complex. Taken to- ized as a series of distinct subsets based on the expression of cell gether, these processes determine not only survival, but also the surface proteins. The earliest T cell progenitors, double-negative appropriate lineage specification of ab T cells in the periphery, (DN) cells, lack expression of the TCR, CD4, or CD8. Subsequent which in general correlates with their phenotypic and functional signaling through the rearranged pre-TCR and growth factor specialization into MHCII-restricted CD4 Th cells or into MHCI- receptors induces cell proliferation and differentiation into double- restricted CD8 CTLs (11–19). positive (DP) cells that express both CD4 and CD8 coreceptors, as Moreover, the development of mature, functional, and self- by guest on September 24, 2021 well as the mature ab-TCR (7–9). Further differentiation is ini- restricted T cells, expressing a diverse Ag-specific repertoire, tiated by interaction of the TCR with the self-peptides presented which is essential for an efficient immune response, is achieved by MHC proteins, MHCII or MHCI, to generate mature CD4 and through signals induced via reciprocal interactions between T cell CD8 T cells, respectively (10). The generation of mature CD4 and precursors and multiple soluble as well as membrane-bound factors provided by cortical and medullary thymic stromal cells. Among these signals, Notch–ligand interactions are compulsory for T *Department of Immunology, University of Toronto, Toronto, Ontario M4N 3M5, lymphopoiesis (20–25). Canada; and †Sunnybrook Research Institute, Toronto, Ontario M4N 3M5, Canada With regard to T lymphopoiesis, Notch signaling has been 1Current address: Molecular Pathogenesis Program, The Kimmel Center for Biology implicated in B versus T, and ab versus gd T cell lineage com- and Medicine of the Skirball Institute, New York University School of Medicine, mitment, differentiation through early DN stages of thymocyte New York, NY. development, progression from DN to DP stages, CD4 Th1 versus Received for publication April 4, 2012. Accepted for publication July 25, 2012. Th2 cell specification, as well as regulation of cytolytic effector This work was supported by funds from the Canadian Institutes of Health Research function in CD8 T cells (22, 26–37). Similarly, ectopic expression (MOP-12927 and MOP-119538). D.D.D. was supported by a scholarship from the Lady Tata Memorial Fund; K.K. was supported by a Natural Sciences and Engineer- of the Notch ligand Delta-like 1 by mouse OP9 BM cells supports ing Research Council studentship; and J.C.Z.-P. is a recipient of a Canada Research efficient T lymphopoiesis from mouse fetal liver (FL) progenitors, Chair in Developmental Immunology. adult BM-derived hematopoietic stem cells (HSCs), embryonic The microarray data presented in this article have been submitted to the Gene Ex- stem cells, and cord blood-derived CD34+ into mature TCR ab+ pression Omnibus (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE37669) + under accession number GSE37669. CD8 T cells (22, 29, 38). However, the functional assessment of + Address correspondence and reprint requests to Dr. Juan Carlos Zu´n˜iga-Pflu¨cker, entirely in vitro generated CD8 T cells remains ambiguous and Sunnybrook Research Institute, 2075 Bayview Avenue, Room A3-31, Toronto, ON not fully characterized. M4N 3M5, Canada. E-mail address: [email protected] In this study, we demonstrate that ab+ CD8 T cell development The online version of this article contains supplemental material. proceeds normally in HSC/OP9-DL1 cocultures, leading to their Abbreviations used in this article: BM, bone marrow; DC, dendritic cell; DN, double- differentiation into functionally mature cells. The maturation negative; DP, double-positive; FL, fetal liver; Gzm-B, granzyme-B; HSC, hemato- poietic stem cell; hTRP-2, human TRP-2; LCMV, lymphocytic choriomeningitis status of in vitro generated CD8 T cells showed similarities to that viral; LN, lymph node; MFI, mean fluorescence intensity; pMHC, peptide MHC; of ex vivo CD8 T cells as determined by the surface expression QRT-PCR, quantitative real-time PCR; SP, single-positive; TRP-2, tyrosinase-related of differentiation markers, with some significant differences also protein-2. noted. At the molecular level, in vitro derived CD8 T cells por- Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 trayed predominantly a normal CD8-biased gene expression pro-

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1200979 2 IN VITRO GENERATED CD8 T CELLS

file, albeit Notch-regulated genes showed elevated expression. In cytometric cell sorting of CD117+ Sca-1high expressing cells, referred to as addition, in vitro stimulation assays confirmed that these CD8 the HSC-enriched fraction. Thymus and spleen were isolated from 4- to 8- T cells are functionally mature cells that can proliferate and express wk-old mice. Single-cell suspensions were disaggregated through a 40 mM nylon cell strainer with a syringe plunger and were washed further with CD8-specific activation markers. Correspondingly, we found that OP9 medium. Spleens were depleted of RBCs with ammonium chloride these cells mature into effector cells as defined by their capacity to lysing reagent (BD Pharmingen), as recommended by the supplier. Prior to produce antiviral cytokines and cytolytic effector molecules. cell sorting, OP9-DL1 cocultures were passaged through 40 mM nylon cell We also demonstrate that CD8 T cells generated in HSC/OP9- strainer. For the generation of dendritic cells (DCs), BM cells were iso- lated, as described above, and 5 3 106 cells were placed in 10 ml DMEM, DL1 cocultures are broadly tolerant to self-pMHC Ags, as assayed containing 10% FBS, with the addition of 5 ng/ml each, mouse IL-4, and by in vitro and in vivo responses to allogeneic or syngeneic APCs. mouse GM-CSF (PeproTech). Fresh media and cytokines were added on In addition, we observed augmented lymphopenia-driven prolif- days 3 and 6. In the last 12–16 h, DCs were activated by addition of 0.1 + eration and expansion of in vitro derived CD8+ T cells when mg/ml LPS (Sigma-Aldrich). At day 7, CD11c cells were FACS sorted or transferred into RAG22/2 host mice. In addition, we revealed the directly purified with CD11c microbeads, according to the manufacturer’s protocol (Miltenyi Biotec). presence of a diverse and polyclonal TCR repertoire in coculture- derived CD8 T cells. Finally, we show that in vitro derived CD8+ BM HSC/OP9-DL1 cell cocultures

T cells generated an enhanced primary Ag-specific response 5 BM-derived HSCs (10 cells) were cocultured on a monolayer of OP9-DL1 against tumor-associated Ag, tyrosinase-related protein-2 (TRP- cells in 10-cm plates, as previously described (39). 2), and lymphocytic choriomeningitis viral (LCMV) Ag g33 in vitro. Taken together, our results demonstrate that the use of Microarray analysis a Notch ligand-expressing stromal cell line allows for the efficient MACS molecular (Miltenyi Biotec) performed complete gene expression Downloaded from generation of Ag-specific, diverse, functionally competent, and analysis from mRNA samples obtained from CD8 T cells, using Agilent self-tolerant CD8+ T cells from adult BM-derived HSCs. whole mouse genome oligo microarrays. CD8 cells were obtained from the thymus (ex vivo), as indicated above, or from BM HSC/OP9-DL1 cocul- tures by first staining for CD4-FITC, CD8-allophycocyanin, and TCRb-PE; Materials and Methods following staining, the cells were immunomagnetically depleted with anti- Mice FITC magnetic beads and processed with an AutoMACS cell sorter to deplete CD4+ cells. The CD8-enriched fraction was then subjected to flow http://www.jimmunol.org/ CD1, BALB/c, C57BL/6 (B6), and B6-Ly5.1 congenic mice were purchased cytometric cell sorting with a FACSAria to isolate CD42CD8+ TCRb+ from The Jackson Laboratory. RAG2-deficient mice were bred and cells. Sort purity ranged from 98 to 99%. The RNA samples were quality maintained in our animal facility under specific pathogen-free conditions. checked via the Agilent 2100 Bioanalyzer platform (Agilent Technolo- All animal procedures were reviewed and approved by the Sunnybrook gies). The hybridization procedure was performed according to the Agilent Health Science Centre Animal Care Committee. 60-mer oligo microarray processing protocol using the Agilent Gene Ex- Cell lines pression Hybridization Kit (Agilent Technologies). The DNA microarray data reported in this paper have been deposited in Gene Expression Om- OP9-DL1 cells were generated from the OP9 BM stromal cell line and nibus Express (accession number GSE37669, http://www.ncbi.nlm.nih. maintained, as previously described (22). gov/geo/query/acc.cgi?acc=GSE37669). Flow cytometry Real-time quantitative PCR by guest on September 24, 2021

Biotin, FITC, PE, PE-Cy5, PE-Cy7, allophycocyanin, and allophycocyanin- high Cy7 mAbs were purchased from BD Biosciences or eBiosciences, except RNA was isolated from coculture-derived and sorted CD8 TCRb T cells for the anti-perforin mAb that was purchased from Kamyia Biomedical. using TRIzol reagent (Invitrogen). Total RNA was treated with Quanti Tect The following conjugated Abs were used: anti-B220 (RA3-6B2), anti-CD3e Reverse Transcription cDNA synthesis kit (Qiagen). Real-time quantitative PCR (QRT-PCR) reaction was carried out with SYBR Green (Invitrogen), (145-2C11), anti-CD4 (GK1.5, L3T4), anti-CD5 (Ly-1), anti-CD8a (LY-2, R Lyt-2, 53-6.7), anti-CD11a (M17/4, 2D7), anti-CD11b (M1/70), anti-CD16/ and analysis was performed using ABI Prism 7500 Sequence Detection CD32 (2.4G2), anti-CD19 (1D3), anti-CD25 (7D4), anti-CD27 (LG.7F9), System. Calculations were completed using relative quantification method, anti-CD28 (37.51), anti-CD44 (IM7), anti-CD45 (30-F11), anti-CD45.1 in which the samples were normalized to b-actin expression. Primer (A20), anti-CD45.2 (104), anti-CD54 (3E.2), anti-CD62L (MEL-14), anti- sequences for QRT-PCR were purchased from Invitrogen, and the Vb- CD69 ([1H].2F3), CD90.2 (53-2.1), anti-CD117 (2B8), anti-CD122 (TM- specific sequences used were as previously described (40). b1), anti-CD244 (eBio244F4), anti-TCRb (H57-597), anti–GR-1 (RB6- 8C5), NK1.1 (PK136), Sca-1 (E13-161.7), TER119, and anti-H2KbDb (28- T cell stimulation assay 8-6). Cells were stained by standard staining techniques and analyzed on Sorted CD8+ TCRbhigh T cells (5 3 104) were cultured in triplicates with a FACSCalibur or LSRII flow cytometers (BD Biosciences). Data files were plate-bound anti-CD3ε (10 mg/ml) and anti-CD28 (5 mg/ml) mAb in 200 analyzed with FlowJo (Tree Star). Dead cells were excluded from all data by ml OP9 media supplemented with 1 ng/ml IL-7 and 10 U/ml IL-2 forward and side scatter and propidium iodide or DAPI (Molecular Probes). (PeproTech). Cells were stained with appropriate Abs and analyzed by Cell sorting was performed with FACSDiVa or a FACSAria cell sorters (BD flow cytometry at the indicated time points. The average mean fluores- . Biosciences). Purity was typically 98% for all populations, as determined cence intensity (MFI) of each sample was plotted against the time point(s) by postsort analysis. from the analysis. For CFSE labeling, sorted CD8 TCRbhigh T cells (5 3 4 Intracellular staining 10 ) were labeled with 0.5 mM CFSE, as recommended by the supplier (Molecular Probes), and the number of cell divisions after stimulation was Coculture-derived and sorted CD8 TCRbhigh T cells (5 3 104) were stim- measured by flow cytometry and calculated by FlowJo software using the ulated with plate-bound anti-CD3 and anti-CD28 for 72 h, as described same constrains for all samples. Proliferation indexes were calculated below. The cells were surface stained with the appropriate mAbs, followed using the same parameters. by intracellular staining using fixation solution and permeabilization buffer, as indicated by the supplier (eBioscience). The secretion of cytokines in the Enzyme-linked immunosorbent assay last 6–18 h was blocked with addition of 10 mg/ml brefeldin A (Sigma- high 4 Sorted CD8 TCRb T cells (5 3 10 ) were stimulated with plate-bound Aldrich). Cells were then analyzed by flow cytometry. anti-CD3/CD28 for 72 h, as described above. The cytokine levels were Cell isolation and preparation determined by using a commercially available DuoSet ELISA Develop- ment System kit (R&D Systems). In all assays, sample concentrations were BM cells were harvested from the femur and tibia of 6- to 8-wk-old mice. calculated from a standard curve trendline equation using OD450nm The dissected long bones were crushed using a Pyrex glass stopper in readings for a given standard concentration. OD450nm raw values from a-MEM (Life Technologies) supplemented with 20% FBS (HyClone) and triplicate wells were subtracted from OD450nm readings obtained from 2 2.2 g/ml NaHCO3, referred to as OP9 media. Lineage-negative (Lin )BM triplicate wells of negative controls (media). To obtain final concentration, samples were enriched by immunomagnetic depletion of cells expressing these values were multiplied by the dilution factor. Results were plotted CD45R, CD19, CD3, NK1.1, TER119, CD11b, and Gr-1, followed by flow using one grouping variable scatter plot (Prism). The Journal of Immunology 3

Mixed lymphocyte reaction maturational intermediates appearing approximately the third + Sorted CD8 TCRbhigh cells (5 3 104) were cultured with g-irradiated (25 week, to mature CD8 single-positive (SP) cells that express high Gy) splenocytes (5 3 104) as stimulators in each well, in triplicates, of levels of TCR, which were clearly detectable at ∼1 mo from the a 96-well round-bottom plate. The cultures were incubated in OP9 media start of the cocultures (Supplemental Fig. 1). During the first 3 wk, for 48–72 h in a humidified 37˚C incubator in an atmosphere of 5% CO2 in there was a well-defined and ordered appearance of different DN 3 air. A total of 1 mCi in 10 ml[H]TdR was added to each well in a volume subsets, characterized by CD44 and CD25 expression, with DN2 of 200 ml for the final 18 h of culture. Cells were harvested using a mul- + + 2 tiple well harvester, and [3H]TdR incorporation was determined in liquid (CD44 CD25 ) cells appearing first (day 5), then DN3 (CD44 + 2 2 scintillation counter. The mean incorporation values of [3H]TdR in DNA CD25 ) cells (day 8), and finally DN4 (CD44 CD25 ), or pre- were plotted using a spreadsheet program (Prism). DP, cells within the first 2 wk of coculture. Cellular expansion Adoptive cell transfer obtained after 35–38 d of culture from BM-HSCs differentiating high high into T-lineage cells, and CD8 TCRb T cells ranged from 2 to Coculture-derived and sorted CD8 TCRb T cells were labeled with 10 3 4 3 2 3 6 3 10 -fold and 1 to 2 10 -fold, respectively. mM CFSE, as described above. The labeled cells (1–2 10 ) were washed + extensively in PBS and then injected in 300 ml PBS per mouse (B6, B6- In addition, by day 35, CD8 SP cells appeared to have attained Ly5.1, CD1, or RAG22/2) via the lateral tail vein. Three and/or 7 d after a mature phenotype, as 10–30% of these cells showed high levels of injection, LN and spleen cells were isolated, and single-cell suspensions TCRb expression (Fig. 1A, 1B). Further analysis of these in vitro were stained with specific mAbs for analysis by flow cytometry. generated CD8+ TCRb+ SP T cells showed that they expressed TCR spectratyping analysis a similar pattern of differentiation and maturation cell surface markers to that of ex vivo thymus CD8+ SP T cells, including similar Spectratyping was performed using the PCR primers and methodology levels of CD27, CD28, Qa2, and MHCI expression (Fig. 1C). Downloaded from published by Pannetier et al. (41), with some modifications. Briefly, 25 ng + cDNA from coculture-derived and sorted CD8 TCRbhigh T cells was In vitro derived CD8 SP T cells upregulated CD5 expression to subjected to 35 cycles of elongation for each of the 24 Vb-chains as fol- lower levels than that of CD8 cells from the thymus or spleen (Fig. lows: 30 s at 94˚C, 30 s at 55˚C, and 45 s at 72˚C. After the final cycle, the 1C, lower panel). Although the expression of CD24 (heat-stable Ag) mixture was heated for 5 min at 72˚C. A total of 5 ml of the primary PCR remained high on in vitro generated CD8 T cells as compared with product was subjected to additional 25 cycles of PCR amplification. The 6- FAM–labeled Cb primer was used in the second round of amplification. ex vivo CD8 SP cells, the clear upregulation of a broad set of dif-

Electrophoresis was performed on a 3730 XL DNA sequencer (Applied ferentiation cell surface markers (CD5, CD27, CD28, Qa2, MHCI) http://www.jimmunol.org/ Biosystems), and CDR3 length and fluorescence intensity of PCR products coincided with a mature CD8 T cell profile (Fig. 1C). Of note, were determined with GeneMapper software (Applied Biosystems). in vitro derived CD8 T cells lacked surface expression of CD44, Generation of Ag-specific CD8 T cells and pMHC-tetramer CD122, and CD244 (Fig. 1D), suggesting that in vitro derived CD8 staining T cells are likely to be akin to conventional, rather than innate-like CD8 T cells (42–45). Taken together, these findings indicate that CD11c+ DCs (5 3 105) were infected with 100 PFU/cell with an adeno- virus expressing the TRP-2 or LCMV-associated Ag gp33 (provided by OP9-DL1 cells can support the efficient differentiation of mature J. Bramson, McMaster University, Hamilton, ON, Canada), and cultured conventional CD8 T cells from BM-HSCs in vitro. with 0.5–1 3 106 coculture-derived and sorted CD8+ TCRbhigh cells for + high 7d.CD8 TCRb cells were washed and replated with adenovirus In vitro generated CD8 T cells display a similar gene by guest on September 24, 2021 expressing the TRP-2– or gp33–infected DCs every 7 d, as indicated. To expression profile to ex vivo CD8 cells detect the TRP-2–specific T cells, cells were stained with allophycocyanin- coupled H-2Kb/SVYDFFVWL (human TRP-2 [hTRP-2] 180–188) or We further characterized in vitro generated CD8 T cells by com- gp33–41 H-2Kb/KAVYNFATC as well as H2Kb/SIINFEKL (ova 257–264) paring their gene expression profile with that of ex vivo thymic CD8 b or (gp276–286) SGVENPGGYCL/H2D as a negative control, according T cells, using the Agilent’s whole mouse genome oligo microarray to the manufacturer’s protocol (Proimmune). comprised of 41,534 (60-mer) oligonucleotide probes, representing Statistics ∼22,000 mouse genes (Miltenyi Biotec). The scatter plot shown in The data are presented as mean 6 SEM. To determine statistical signifi- Fig. 2A provides an analysis of the differential expression patterns cance, a two-tailed Student t test was used for comparison between two between in vitro derived and ex vivo CD8 T cells for ∼41,000 experimental groups, using Prism software. Statistical significance was individual probes, in which the data correlation of log ratio in- , , , determined as p 0.05 (*p 0.05, **p 0.01). tensity 1 on the x-axis represents ex vivo derived and log ratio intensity 2 on the y-axis represents in vitro derived CD8 TCRbhigh Results cells (accession GSE37669). Of note, given the overall similarity in Generation of mature conventional CD8 T cells from BM-HSC/ phenotype between in vitro and ex vivo CD8 T cells (Fig. 1C), we OP9-DL1 cocultures found that most gene probes (.97%) were expressed at equivalent We have previously demonstrated that OP9-DL1 cells support T (#3-fold change) levels (blue). However, ∼1400 probes, or ∼3.5% lymphopoiesis from fetal liver (FL)-derived HSCs, resulting in the of the total number analyzed, whose levels differed by 3-fold or generation of a small number of mature CD8 T cells (22). Although greater, were either upregulated (red) or downregulated (green) by we later showed that adult BM-derived HSC/OP9-DL1 cocultures the in vitro generated CD8 T cells (Fig. 2A, Supplemental Table could also give rise to CD8 T cells, a detailed characterization and 1A). Given the distinctive feature of the inclusion of OP9 cells comprehensive analysis of the generation of these T cells were within these cocultures, and sort purities that may include 1–2% not performed and remained undefined whether functional T cells OP9 cells, a subtractive gene expression analysis between genes could be generated from adult hematopoietic progenitors. In this highly expressed by OP9 stromal cells and in vitro derived CD8 study, we sought to characterize the generation of mature func- T cells was performed (Supplemental Table 1B). This analysis 2 tional CD8 T cells from Lin Sca1+ CD117+ BM-derived HSCs resulted in the reduction of the number of differentially expressed cultured with OP9-DL1 cells. genes between in vitro and ex vivo CD8 cells to only ∼0.6%, which Flow cytometric analysis of the BM-HSC/OP9-DL1 cocultures will be described further in depth separately. performed at different time intervals revealed an apparently syn- In this study, we focused on the expression of several key chronous phenotypic progression through the various stages of transcription factors recognized to play critical roles in CD8 2 2 T cell development, commencing with the earliest CD4 CD8 (Runx1 and Runx3) or CD4 (GATA3 and ThPok) lineage speci- double-negative (DN) from days 5 to 23, through CD4+CD8+ DP fication and commitment (Fig. 2A, bar graph) (46–50). A 4-fold 4 IN VITRO GENERATED CD8 T CELLS

FIGURE 1. Development of mature conven- tional CD8+ TCRbhigh cells from BM-HSC/ OP9-DL1 cocultures. Lin2 CD117+ Sca1+ HSCs were isolated from the BM of 6- to 8-wk- old B6 mice and cocultured with OP9-DL1 cells for 35 d. (A) Flow cytometric analysis of CD4, CD8, and TCRb expression from in vitro de- rived (OP9-DL1) or ex vivo (thymus and spleen) cells. (B) Histograms showing TCRb expression are from CD42CD8+ gated cells, and bar graph Downloaded from with the corresponding MFI of TCRb expres- sion for the indicated samples (n . 3). (C and D) Histograms showing the expression of indi- cated cell surface markers from in vitro derived (OP9-DL1)orexvivo(thymusandspleen) CD42CD8+ TCRb+ gated cells. Below each http://www.jimmunol.org/ histogram, the corresponding MFIs are shown for the indicated samples (n . 3). Data are rep- resentative of at least three independent experi- ments. **p , 0.01. RCN, Relative cell numbers. by guest on September 24, 2021

increase in gata3 transcript levels by in vitro derived CD8 sion, cell division, and effector function of CD8 TCRbhigh cells in TCRbhigh cells was observed. Although GATA3 is a transcription response to TCR/CD3 stimulation, by anti-CD3 and anti-CD28 factor identified to play a role in CD4 T cell specification, ex- Ab-mediated engagement (Fig. 3A–D). As expected, in vitro de- pression of ThPok (zbtb7b), a transcription factor involved in CD4 rived CD8 T cells underwent proliferation and cell division, as T cell commitment, was not detected. In addition, expression measured by [3H]thymidine uptake and CFSE dilution, in re- levels of runx1 and runx3, transcription factors known to play sponse to TCR stimulation (Fig. 3A, 3B). Additionally, flow a role in CD8 T cell development, were detected from both CD8 cytometric analysis of stimulated in vitro generated CD8 T cells T cells, which further indicates a predominantly CD8-biased gene revealed a bona fide activation profile (Fig. 3C). In particular, the expression profile by in vitro derived CD8 T cells (Fig. 2B). acquisition of an activated effector profile was marked by the in- Transcript levels of CD8- or CD4-specific genes obtained from the creased expression of CD25 (IL-2Ra), CD44 (Pgp1), and CD11a gene-array analysis were verified by QRT-PCR (Fig. 2B). Our (LFA-1), and decreased expression of CD62L (L-selectin) upon combined findings suggest that transcription factors involved in TCR stimulation (Fig. 3C). Importantly, activated in vitro gener- CD8-lineage commitment are expressed by in vitro derived CD8 ated CD8 SP T cells produced IFN-g (Fig. 3D), and also expressed T cells in a similar fashion as found in CD8 SP thymocytes. high levels of granzyme-B (Gzm-B) and perforin, key effector However, some of the differences in gene expression that were molecules that define the cytolytic capacity of a CD8 T cell (Fig. detected from this analysis, such as the increased levels of gata3 3E). Of note, we consistently detected higher levels of Gzm-B and expression, are likely to be the result of ongoing Delta-like 1-in- perforin production by in vitro generated CD8 T cells. Because duced Notch signals provided by the OP9-DL1 cells (37) or un- in vitro CD8 T cells are exposed to continuous Notch signaling accounted contamination from OP9 cells. provided by the OP9-DL1 cells, these results are in line with re- cent reports suggesting a Notch-dependent induction of cytolytic In vitro derived CD8 T cells are functionally equivalent to molecules in CD8 T cells (34, 35). Thus, the effector profile ex vivo thymus-derived cells signature observed upon anti-CD3/CD28–induced activation of To evaluate the functional properties of CD8 T cells generated from in vitro generated CD8 T cells was similar to that of functionally BM-HSC/OP9-DL1 cocultures, we measured the cellular expan- mature ex vivo CD8 cells. The Journal of Immunology 5

host mice showed a substantial loss of CFSE (Supplemental Fig. 2B), confirming their ability to mount an effective alloresponse (Fig. 4A) and ability to become activated after in vivo transfer. Based on these findings, and consistent with the MLR assay analysis, the absence of CD8 T cell proliferation in vivo suggests that naive resting in vitro derived CD8 T cells can undergo tolerance induction. To address whether in vitro generated CD8 T cells are able to undergo homeostatic T cell proliferation (51, 52), 1–2 3 106 CFSE-labeled in vitro generated CD8 T cells were adoptively transferred into RAG2-deficient mice. Seven days after transfer, LNs and spleen were isolated and analyzed by flow cytometry. Notably, a population of donor T cells with little to no CFSE content was detected by flow cytometry (Fig. 4C, Supplemental Fig. 2C). These data suggest that in vitro generated CFSE-labeled CD8 T cells underwent several rounds of cell division after transfer into immunodeficient hosts. Of note, the transferred CD8 FIGURE 2. Gene expression profile of in vitro derived CD8 T cells T cells showed slightly upregulated CD44 levels, but failed to compared with ex vivo CD8 naive T cells. (A) Scatter plot of signal in- upregulate CD25 expression (Supplemental Fig. 2C), a phenotypic Downloaded from tensities obtained from hybridization of one of four gene expression feature of cells undergoing IL-7–driven homeostatic proliferation + high microarray experiments is shown, in which sorted CD8 TCRb cells under lymphopenic conditions. from day 35 HSC/OP9-DL1 cocultures and thymuses from B6 mice (4–5 wk old) were labeled with CY3 (green) and CY5 (red), respectively, and In vitro generated CD8 T cells display a broad TCR-Vb CDR3 used to hybridize onto an Agilent GPL4134 microarray. Selected fold diversity change (gene ratios; color coded as in the scatter plot) represents the mean value of four independent hybridizations for the indicated genes to Agilent In an attempt to characterize the TCR repertoire that is present http://www.jimmunol.org/ whole mouse genome oligo microarrays. (B) RNA transcript levels for the in CD8 T cells generated in vitro, we performed a CDR3 length indicated genes, measured by QRT-PCR from sorted CD8+ TCRbhigh cells spectratyping analysis of in vitro and ex vivo CD8 T cells. The derived from day 35 HSC/OP9-DL1 cocultures or ex vivo thymocytes, are analysis of 22 of 24 Vb-chains (Vb1-Vb20) showed that in vitro shown. Results of triplicate samples were normalized to b-actin expres- generated CD8 T cells displayed a broad and highly diverse TCR- sion. These data are representative of at least five independent experiments. Vb usage, with CDR3 lengths showing a normal Gaussian dis- tribution comparable to that of ex vivo CD8 T cells (Fig. 5). Notably, we found that nearly all of the Vb-chains showed poly- Generation of self-tolerant CD8 T cells in vitro clonal usage with similar CDR3 lengths to that of ex vivo CD8 To determine whether BM-HSC/OP9-DL1 coculture-derived CD8 T cells. These findings were supported by surface expression of by guest on September 24, 2021 T cells can discriminate between self and nonself Ags, we performed TCRa and TCRb tested at protein level (data not shown). Taken MLR cultures using in vitro generated CD8 T cells as the responders together, these results provide an important insight by demon- (Fig. 4A). Of note, C57BL/6 (B6; H2b)-derived HSCs were used to strating that a broad and diverse TCR repertoire is produced generate CD8 T cells in OP9-DL1 cocultures, and these cells dis- during in vitro T cell differentiation. played increased proliferation in response to allogeneic CD1 (H2q)- irradiated splenocytes used as stimulators. In contrast, no increased Generation of an Ag-specific response by CD8 T cells proliferation in response to syngeneic B6 stimulator cells was generated in vitro detected (Fig. 4A). The generation of alloreactive CD8 T cells that Based on the above findings showing that a broad TCR repertoire also lack self-responsiveness suggests that in vitro generated CD8 is generated in vitro, and that these CD8 T cells are functionally T cells underwent a series of successful positive and negative mature with a powerful effector machinery capable of recognizing selection events, two processes that are critical in the selection of and destroying cells expressing viral or tumor-associated Ags, we a functional and self-tolerant T cell repertoire. sought to assess the premise that an Ag-specific response can be The absence of overt self-reactivity in the MLR assays does not developed entirely in vitro. To this end, we examined whether a fully address whether all potentially self-reactive T cell clones are specific response can be generated against a tumor-associated Ag indeed eliminated during CD8 T cell maturation in vitro. For in- and viral Ag, the hTRP-2 (also known as dopachrome tautomerase) stance, it is not clear whether a potentially limited abundance of and LCMV-associated Ag gp33, respectively. self-peptides presented by MHCI on OP9-DL1 cells results in the For this approach, a bulk population of in vitro generated CD8 elimination of all self-reactive T cells. To test this premise, we T cells was stimulated (or primed) with BM-derived DCs that were transferred 1–2 3 106 CFSE-labeled in vitro generated or ex vivo induced to express full hTRP-2 protein or LCMV-specific gp33 CD8 T cells i.v. into syngeneic B6 mice. Three days later, donor epitope (KAVYNFATC) following infection with a recombinant CFSE-labeled T cell proliferation was evaluated by measuring adenovirus-based Ag delivery (Fig. 6). The Ag-specific response CFSE dilution of in vitro derived CD8 T cells in comparison with was measured by MHCI-specific tetramers (tet) containing im- noninjected and CFSE-labeled cells. As shown in Fig. 4B (and munodominant CD8 epitopes for both mouse and hTRP-2, Supplemental Fig. 2A), both in vitro and ex vivo CD8 T cells SVYDFFVWL (residues 180–188) and LCMV-specific gp33–41. showed a very small number of cells that had diminished CFSE At the peak of the response (day 14), the frequency of gp33-tet+ levels, as assessed by flow cytometry from lymph nodes (LNs), in vitro derived CD8 T cells reached an average of ∼10%, which suggesting that CD8 T cells, generated in vitro from B6 HSC/ was higher, but not statistically different, than that observed from OP9-DL1 cultures, failed to mount an overt antiself response ex vivo CD8 T cells (Fig. 6A). Similarly, frequency of TRP-2-tet+ in vivo. However, flow cytometry analysis of CFSE-labeled cells was higher, but not statistically different, for in vitro derived in vitro generated CD8 T cells transferred into H2-mismatched CD8 T cells than that observed from ex vivo CD8 T cells (Fig. 6B). 6 IN VITRO GENERATED CD8 T CELLS Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021

FIGURE 3. In vitro derived CD8+ TCRbhigh cells display similar functional properties to ex vivo CD8+ TCRbhigh cells. (A) Bar graphs shows [3H]TdR incorporation measured from anti-CD3/CD28 mAb-stimulated (ST) and unstimulated (UST) CD8+ TCRbhigh T cells, obtained in vitro (day 35 BM-HSC/ OP9-DL1 cocultures) or ex vivo (6-wk B6 thymocytes). Data represent the average values, and error bars represent SEM from triplicate cultures. (B) Flow cytometry analysis indicating the number of cell cycle divisions detected by labeling sorted CD8+ TCRbhigh T cells with 0.5 mM CFSE. Number of cell divisions (at top) was calculated by FlowJo software using the same constraints for all samples. Proliferation indexes (numbers at bottom left or right of each histograms) were calculated using the same parameters. Representative plots of stimulated and unstimulated cells are shown. (C) Bar graphs showing the changes in MFI values (from triplicate samples) of the indicated cell surface markers from stimulated (ST: white and black bars) or unstimulated (UST: light and dark gray bars) CD8+TCRbhigh cells, obtained as in (A), measured at 24, 48, and 72 h. Data are representative of at least three independent experiments. (D) Bar graph showing IFN-g production detected from culture supernatants of stimulated (ST) or unstimulated (UST) CD8+ TCRbhigh T cells, in vitro (white bars) and ex vivo (black bars, thymocytes). (E) Flow cytometric analysis of Gzm-B and perforin expression of stimulated (black lines) or unstimulated (shaded) CD8+ TCRbhigh T cells for 72 h, as indicated. Data are representative of at least five separate experiments.

Flow cytometric analysis using control tetramer peptides (gp276– ings clearly demonstrate that the use of a Notch ligand-expressing 286, or ova257–264) was used to show the specificity of the stromal cell line supports the generation of CD8 T cells that are responding T cells, which was low and not statistically different functionally competent, self-tolerant, bearing a diverse TCR rep- between each Ag-specific population. Taken together, these find- ertoire, and capable of mounting an Ag-specific response. The Journal of Immunology 7 Downloaded from http://www.jimmunol.org/

+ FIGURE 5. CDR3 spectratyping reveals a diverse TCR repertoire from FIGURE 4. Absence of overt self-reactivity by in vitro derived CD8 + high TCRbhigh T cells. (A) Bar graph shows [3H]TdR incorporation measured in vitro derived CD8 TCRb cells. Fluorescent electrophoresis profiles from a 3-d MLR assay using sorted CD8+ TCRbhigh T cells isolated from are displayed for the 22 Vb-specific RT-PCR products analyzed. cDNA was amplified from in vitro derived (HSC/OP9-DL1 d38 coculture; upper HSC/OP9-DL1 day 35 cocultures (in vitro) or from B6 thymocytes (ex + high vivo) mixed with either syngeneic (Syn) or allogeneic (Allo) irradiated panels) and ex vivo (B6 thymocytes; bottom panels) sorted CD8 TCRb splenocytes as stimulators. Data represent the average values, and error cells. The x-axis of each profile corresponds to the CDR3 length in nu- bars represent SEM from triplicate cultures. (B and C) Flow cytometric cleotides, and each peak is separated by three bases, corresponding to in- analysis of CFSE levels from labeled CD8+ TCRbhigh T cells, obtained frame transcripts.

in vitro (B6 BM-HSC/OP9-DL1 coculture, day 35) or ex vivo (B6 mice by guest on September 24, 2021 splenocytes), i.v. injected into nonirradiated B6 mice (B)orRAG22/2 mice of T cells. CD5 levels have been shown to correlate with the in- (C). Analysis of CFSE expression by CD8+ cells from LNs of injected (middle and bottom panels) or noninjected (top panel) mice after 3 d (B)or tensity of TCR signals mediating positive selection (55), which 7d(C). The data are representative of two to three mice analyzed indi- suggests that cells differentiating in vitro may experience TCR vidually. Experiment was repeated at least three times. signals of lower intensity, but sufficiently strong to allow for the generation of CD8 T cells. In terms of CD24 expression, one study has shown that recent Discussion thymic emigrants retain high expression of CD24, and that Whereas several previous studies have concluded that functionally downregulation takes place in the periphery (56). Despite the mature CD8 SP T cells can be generated from mouse FLs and evidence that functional properties might be diminished in CD8+ embryonic stem cells, as well as human cord blood-derived CD34+ CD24+ cells (e.g., reduction in proliferation and cytokine pro- HPCs, in this study, we demonstrate that OP9-DL1 cells support duction in response to stimulation) (56), our data support the efficient T cell differentiation from adult BM-derived HSCs, notion of functional autonomy despite the expression of CD24. giving rise to functionally mature CD8 SP cells. Additionally, the functional significance of this trait resides in the Recent studies have demonstrated phenotypic classification finding that CD24 expression is required for optimal T cell pro- of CD8 T cells based on the expression of activation markers liferation in lymphopenic host (57). Accordingly, we find that into innate CD8+CD44+CD122+ or CD8+NK1.1+CD44+CD122+ in vitro CD8+CD24+ when compared with ex vivo CD8+CD242- CD244+NKT-like CD8 T cells versus conventional CD8+CD442 derived T cells undergo enhanced proliferative responses in 2 2 2 CD122 NK1.1 cells; CD8+CD28 suppressor versus CD8+ RAG22/2 host mice. Although the role of CD24 on T cells is CD28+ effector cells (42, 44, 53, 54). An inclusive phenotypic largely unknown, our current understanding highlights the find- comparison in this study demonstrates that in vitro derived CD8 ing that functional CD8 T cell responses are not dependent on T cells exhibit all aspects of ex vivo conventional CD8 TCRbhigh downregulating of CD24 maturational marker per se. Of note, this cells recognized by high expression of CD5, CD27, CD28, Qa2, evidence is also supported by the ability of CD8 T cells differ- and MHCI, and lack of expression of innate phenotypic markers entiated from FL progenitors in OP9-DL1 cocultures to maintain such as CD44, CD122, and CD244. This analysis indicates that functional competence (e.g., undergo proliferation, upregulation the in vitro derived CD8 cells appear to conform to a conventional of activation markers, and generation of Gzm-B upon activation; CD8 phenotypic profile. data not shown). Conversely, several distinctive phenotypic features of in vitro Additionally, when compared directly, we find a similar pattern derived in comparison with ex vivo CD8 T cells are lower lev- of expression of the memory differentiation markers by in vitro els of CD5 expression, and the lack of downregulation of CD24 derived CD8 T cells to what is observed by ex vivo CD8 T cells. (heat-stable Ag), which has been associated with the mature status However, one of the differences was the altered expression of 8 IN VITRO GENERATED CD8 T CELLS

FIGURE 6. Ag-specific T cell responses by CD8+ TCRbhigh T cells generated from HSC/OP9-DL1 cocultures. (A and B) Flow cytometric analysis of CD8 expression and pMHC-tetramer reactivity, as indicated, from in vitro primed CD8+ TCRb+ cells is shown. CD8+ TCRbhigh cells from B6 splenocytes (ex vivo) and BM-HSC/OP9-DL1 day 35 cocultures (in vitro) were cultured with DCs infected with adenoviruses encoding specific peptide Ags (A) LCMV (gp33) and (B) hTRP-2 for 14 d. Lower panels of (A) and (B) show control staining with nonspecific pMHC tetramers (LCMV-gp276, and ova-257), as indicated. Data are representative of at least three separate experiments. Bar graphs show the percentage of pMHC-tetramer+ cells from the Downloaded from indicated samples obtained from separate experi- ments (n . 3; p . 0.05, no statistical difference between ex vivo and in vitro samples). http://www.jimmunol.org/

CD279 (PD1) by in vitro derived CD8 T cells (Supplemental Fig. progenitors using OP9-DL1 cells, the process of positive and 3A, 3B). Although expression of other inhibitory receptors such as negative selection of in vitro derived CD8 T cells is mainly dis- CD244 (2B4), CD160, CD152 (CTLA4), and Lag3 was not tested, cussed rather than experimentally demonstrated. Moreover, in- induction of inhibitory receptor PD1 correlates with T cell ex- sufficient numbers of CD8 T cells were used for in vitro analyses haustion (58). Whereas the mechanism that controls PD1 ex- only (22, 29, 60). In this study, we reveal that in vitro derived CD8 pression is largely unknown, it remains to be investigated whether T cells when transferred into syngeneic host mice remain resting, by guest on September 24, 2021 Notch signaling plays a role in the induction of this inhibitory which suggests that OP9-DL1 cells express an array of self-Ags at receptor. With this in mind, we noted that, following TCR stim- a sufficient level to eliminate potentially autoreactive clones. ulation, the addition of Notch signaling promoted expression of Moreover, previous studies have demonstrated that, in the pe- PD1 (Supplemental Fig. 3B). Despite this finding, our results also riphery, mature naive T cells, for their survival and homeostatic support previous reports that demonstrate Notch ligand-induced proliferation, require low-affinity interactions with self-pMHC activation, whereby higher expression of activating receptors (e.g., molecules involved in positive selection. This observation was CD28) and cytolytic molecules (e.g., granzyme A, Gzm-B, and confirmed by one study, which elegantly identified that the ho- perforin) may balance or counterbalance expression of the in- meostatic proliferation of adaptively transferred CD8 T cells is hibitory receptors resulting in unaltered or even enhanced CD8 abrogated in MHCI-deficient irradiated host (52). This would T cell-mediated overall response (59). Thus, the ability to fully suggest that in an immunodeficient host, proliferation of in vitro evaluate the differentiation status of in vitro derived CD8 T cells derived CD8 T cells is attained through the TCR interaction(s) might have an important implication in the prospective use of with self-pMHCI and the trophic effects of available cytokines in vitro derived human CD8 T cells for immune reconstitution. (e.g., IL-7, IL-15). In light of this finding, we hypothesize that In this study, we also analyzed a small fraction of genes obtained homeostatic proliferation of in vitro derived CD8 T cells was from the gene expression microarrray performed between in vitro mediated through contact(s) of TCR expressed by in vitro derived derived and ex vivo CD8 T cells. We noted that 52 genes were CD8 T cells with pMHC expressed either by APCs such as DCs in upregulated and 87 genes were downregulated by in vitro derived the periphery. Surprisingly, homeostatic proliferation of in vitro CD8 T cells. In-depth analysis and conformation of differentially derived CD8 T cells seems to be augmented in immunodeficient expressed genes by in vitro derived CD8 T cells will be charac- host. This may be due to either increased concentration of avail- terized in detail elsewhere. Nevertheless, this initial analysis points able IL-7 in vivo, high expression of CD24 (previously discussed), to several significant differences between in vitro and ex vivo CD8 or overt activation of TCR with pMHCI expressed in the periph- T cells. Among these, the highly elevated levels of granzyme A ery, although the latter possibility is not supported by the lack of detected by resting in vitro derived T cells point to a potential role CD25. Nevertheless, this finding highlights the importance of the of Notch signaling in regulating the expression of this effector- TCR/self-pMHCI recognition, which allows for proliferation of function gene. Other differences obtained from the gene micro- in vitro derived CD8 T cells to occur in the periphery. array analysis may be due to the presence of contaminating OP9 In addition to the ability of in vitro derived CD8 T cells to display cells, or due to ongoing Notch or cytokine signaling, from the self-tolerance and homeostatic proliferation in vivo, we demon- exogenously added IL-7. strate that in vitro derived CD8 T cells also express a diverse TCR Despite a number of studies that have demonstrated induction repertoire, an attribute ascribed to conventional CD8 T cells, which of CD8 T cell differentiation from different mouse or human can recognize peptides in the context of classical MHCI molecules The Journal of Immunology 9

(61). Of note, although innate CD8 T cells play an important role 21. Radtke, F., A. Wilson, G. Stark, M. Bauer, J. van Meerwijk, H. R. MacDonald, and M. Aguet. 1999. Deficient T cell fate specification in mice with an induced in the immune responses against several bacteria, viruses, and inactivation of Notch1. Immunity 10: 547–558. parasites, their TCR repertoire diversity is limited (62). In con- 22. Schmitt, T. M., and J. C. Zu´n˜iga-Pflu¨cker. 2002. Induction of T cell development trast, we demonstrate a broad and functional heterogeneity among from hematopoietic progenitor cells by delta-like-1 in vitro. Immunity 17: 749–756. 23. Artavanis-Tsakonas, S., M. D. Rand, and R. J. Lake. 1999. Notch signaling: cell CD8 T cells. Importantly, we also show that these in vitro derived fate control and signal integration in development. Science 284: 770–776. CD8 T cells have the ability to differentiate into Ag-specific, 24. Maillard, I., T. Fang, and W. S. Pear. 2005. Regulation of lymphoid development, gp33, and TRP-2, CD8 T cells. differentiation, and function by the Notch pathway. Annu. Rev. Immunol. 23: 945–974. In conclusion, important initial steps were made in identifying 25. Lehar, S. M., J. Dooley, A. G. Farr, and M. J. Bevan. 2005. Notch ligands Delta 1 phenotypic, molecular, and functional nature of in vitro derived and Jagged1 transmit distinct signals to T-cell precursors. Blood 105: 1440–1447. 26. Garbe, A. I., A. Krueger, F. Gounari, J. C. Zu´n˜iga-Pflu¨cker, and H. von Boehmer. CD8 T cells generated from BM-HSC/OP9-DL1 cocultures. Fur- 2006. Differential synergy of Notch and T cell receptor signaling determines ther investigation regarding primary and/or secondary Ag-specific alphabeta versus gammadelta lineage fate. J. Exp. Med. 203: 1579–1590. responses by in vitro derived CD8 T cells in vivo may enhance our 27. Radtke, F., A. Wilson, S. J. Mancini, and H. R. MacDonald. 2004. Notch reg- ulation of lymphocyte development and function. Nat. Immunol. 5: 247–253. understanding of additional inputs necessary for developing novel 28. Ciofani, M., T. M. Schmitt, A. Ciofani, A. M. Michie, N. Cuburu, A. Aublin, approaches for manipulating immune responses against foreign J. L. Maryanski, and J. C. Zu´n˜iga-Pflu¨cker. 2004. Obligatory role for cooperative Ags. signaling by pre-TCR and Notch during thymocyte differentiation. J. Immunol. 172: 5230–5239. 29. Schmitt, T. M., R. F. de Pooter, M. A. Gronski, S. K. Cho, P. S. Ohashi, and Acknowledgments J. C. Zu´n˜iga-Pflu¨cker. 2004. Induction of T cell development and establishment of T cell competence from embryonic stem cells differentiated in vitro. Nat. We thank Dr. J. Bramson for obtaining adenoviruses containing hTRP-2 Immunol. 5: 410–417. and LCMV-gp33. We thank G. Knowles, A. Khandani, and C. Macintosh 30. Robey, E., D. Chang, A. Itano, D. Cado, H. Alexander, D. Lans, G. Weinmaster, Downloaded from for expert assistance in cell sorting. Additionally, we are thankful to and P. Salmon. 1996. An activated form of Notch influences the choice between M. Motashami for critical review of the manuscript. CD4 and CD8 T cell lineages. Cell 87: 483–492. 31. Deftos, M. L., Y. W. He, E. W. Ojala, and M. J. Bevan. 1998. Correlating notch signaling with thymocyte maturation. Immunity 9: 777–786. Disclosures 32. Amsen, D., J. M. Blander, G. R. Lee, K. Tanigaki, T. Honjo, and R. A. Flavell. The authors have no financial conflicts of interest. 2004. Instruction of distinct CD4 T helper cell fates by different notch ligands on antigen-presenting cells. Cell 117: 515–526.

33. Amsen, D., A. Antov, D. Jankovic, A. Sher, F. Radtke, A. Souabni, http://www.jimmunol.org/ M. Busslinger, B. McCright, T. Gridley, and R. A. Flavell. 2007. Direct regu- References lation of Gata3 expression determines the T helper differentiation potential of 1. Abramson, S., R. G. Miller, and R. A. Phillips. 1977. The identification in adult Notch. Immunity 27: 89–99. bone marrow of pluripotent and restricted stem cells of the myeloid and lym- 34. Maekawa, Y., Y. Minato, C. Ishifune, T. Kurihara, A. Kitamura, H. Kojima, phoid systems. J. Exp. Med. 145: 1567–1579. H. Yagita, M. Sakata-Yanagimoto, T. Saito, I. Taniuchi, et al. 2008. Notch2 2. Fowlkes, B. J., L. Edison, B. J. Mathieson, and T. M. Chused. 1985. Early integrates signaling by the transcription factors RBP-J and CREB1 to promote T lymphocytes: differentiation in vivo of adult intrathymic precursor cells. J. T cell cytotoxicity. Nat. Immunol. 9: 1140–1147. Exp. Med. 162: 802–822. 35. Cho, O. H., H. M. Shin, L. Miele, T. E. Golde, A. Fauq, L. M. Minter, and 3. Watanabe, Y., Y. Aiba, and Y. Katsura. 1997. T cell progenitors in the murine fetal B. A. Osborne. 2009. Notch regulates cytolytic effector function in CD8+ liver: differences from those in the adult bone marrow. Cell. Immunol. 177: 18–25. T cells. J. Immunol. 182: 3380–3389. 4. Schwarz, B. A., and A. Bhandoola. 2006. Trafficking from the bone marrow to 36. Krueger, A., A. I. Garbe, and H. von Boehmer. 2006. Phenotypic plasticity of

the thymus: a prerequisite for thymopoiesis. Immunol. Rev. 209: 47–57. T cell progenitors upon exposure to Notch ligands. J. Exp. Med. 203: 1977– by guest on September 24, 2021 5. Zlotoff, D. A., and A. Bhandoola. 2011. Hematopoietic progenitor migration to 1984. the adult thymus. Ann. N. Y. Acad. Sci. 1217: 122–138. 37. Fang, T. C., Y. Yashiro-Ohtani, C. Del Bianco, D. M. Knoblock, S. C. Blacklow, 6. Zlotoff, D. A., A. Sambandam, T. D. Logan, J. J. Bell, B. A. Schwarz, and and W. S. Pear. 2007. Notch directly regulates Gata3 expression during T helper A. Bhandoola. 2010. CCR7 and CCR9 together recruit hematopoietic progeni- 2 cell differentiation. Immunity 27: 100–110. tors to the adult thymus. Blood 115: 1897–1905. 38. Van Coppernolle, S., G. Verstichel, F. Timmermans, I. Velghe, D. Vermijlen, 7. Petrie, H. T., F. Livak, D. Burtrum, and S. Mazel. 1995. T cell receptor gene M. De Smedt, G. Leclercq, J. Plum, T. Taghon, B. Vandekerckhove, and recombination patterns and mechanisms: cell death, rescue, and T cell produc- T. Kerre. 2009. Functionally mature CD4 and CD8 TCRalphabeta cells are tion. J. Exp. Med. 182: 121–127. generated in OP9-DL1 cultures from human CD34+ hematopoietic cells. J. 8. Wang, L., Y. Xiong, and R. Bosselut. 2010. Tenuous paths in unexplored terri- Immunol. 183: 4859–4870. tory: from T cell receptor signaling to effector gene expression during thymocyte 39. Holmes, R., and J. C. Zuniga-Pflucker. 2009. The OP9-DL1 system: generation selection. Semin. Immunol. 22: 294–302. of T-lymphocytes from embryonic or hematopoietic stem cells in vitro. Cold 9. Ellmeier, W., S. Sawada, and D. R. Littman. 1999. The regulation of CD4 and Spring Harb. Protoc. 2009: pdb prot5156. CD8 coreceptor gene expression during T cell development. Annu. Rev. Immu- 40. Johnston, S. L., N. D. Borson, and P. J. Wettstein. 1997. Spectratyping of TCRs nol. 17: 523–554. expressed by CTL-infiltrating minor histocompatibility antigen-disparate allog- 10. Singer, A., S. Adoro, and J. H. Park. 2008. Lineage fate and intense debate: rafts. J. Immunol. 159: 5233–5245. myths, models and mechanisms of CD4- versus CD8-lineage choice. Nat. Rev. 41. Pannetier, C., M. Cochet, S. Darche, A. Casrouge, M. Zo¨ller, and P. Kourilsky. Immunol. 8: 788–801. 1993. The sizes of the CDR3 hypervariable regions of the murine T-cell receptor 11. Starr, T. K., S. C. Jameson, and K. A. Hogquist. 2003. Positive and negative beta chains vary as a function of the recombined germ-line segments. Proc. Natl. selection of T cells. Annu. Rev. Immunol. 21: 139–176. Acad. Sci. USA 90: 4319–4323. 12. Sprent, J., and H. Kishimoto. 2002. The thymus and negative selection. Immunol. 42. Schwartzberg, P. L., K. L. Mueller, H. Qi, and J. L. Cannons. 2009. SLAM Rev. 185: 126–135. receptors and SAP influence lymphocyte interactions, development and function. 13. Surh, C. D., and J. Sprent. 1994. T-cell apoptosis detected in situ during positive Nat. Rev. Immunol. 9: 39–46. and negative selection in the thymus. Nature 372: 100–103. 43. Berg, L. J. 2007. Signalling through TEC kinases regulates conventional versus 14. Surh, C. D., B. Ernst, D. S. Lee, W. Dummer, and E. LeRoy. 2000. Role of self- innate CD8(+) T-cell development. Nat. Rev. Immunol. 7: 479–485. major histocompatibility complex/peptide ligands in selection and maintenance 44. Horai, R., K. L. Mueller, R. A. Handon, J. L. Cannons, S. M. Anderson, of a diverse T cell repertoire. Immunol. Res. 21: 331–339. M. R. Kirby, and P. L. Schwartzberg. 2007. Requirements for selection of 15. Ziegler, A., C. A. Mu¨ller, R. A. Bo¨ckmann, and B. Uchanska-Ziegler. 2009. conventional and innate T lymphocyte lineages. Immunity 27: 775–785. Low-affinity peptides and T-cell selection. Trends Immunol. 30: 53–60. 45. Prince, A. L., C. C. Yin, M. E. Enos, M. Felices, and L. J. Berg. 2009. The Tec 16. Robey, E., and B. J. Fowlkes. 1994. Selective events in T cell development. kinases Itk and Rlk regulate conventional versus innate T-cell development. Annu. Rev. Immunol. 12: 675–705. Immunol. Rev. 228: 115–131. 17. Peterson, P., T. Org, and A. Rebane. 2008. Transcriptional regulation by AIRE: 46. Bosselut, R. 2004. CD4/CD8-lineage differentiation in the thymus: from nuclear molecular mechanisms of central tolerance. Nat. Rev. Immunol. 8: 948–957. effectors to membrane signals. Nat. Rev. Immunol. 4: 529–540. 18. Palmer, E. 2003. Negative selection: clearing out the bad apples from the T-cell 47. Herna´ndez-Hoyos, G., M. K. Anderson, C. Wang, E. V. Rothenberg, and repertoire. Nat. Rev. Immunol. 3: 383–391. J. Alberola-Ila. 2003. GATA-3 expression is controlled by TCR signals and 19. Yasutomo, K., C. Doyle, L. Miele, C. Fuchs, and R. N. Germain. 2000. The regulates CD4/CD8 differentiation. Immunity 19: 83–94. duration of antigen receptor signalling determines CD4+ versus CD8+ T-cell 48. Kappes, D. J., X. He, and X. He. 2006. Role of the transcription factor Th-POK lineage fate. Nature 404: 506–510. in CD4:CD8 lineage commitment. Immunol. Rev. 209: 237–252. 20. Pui, J. C., D. Allman, L. Xu, S. DeRocco, F. G. Karnell, S. Bakkour, J. Y. Lee, 49. Kohu, K., T. Sato, S. Ohno, K. Hayashi, R. Uchino, N. Abe, M. Nakazato, T. Kadesch, R. R. Hardy, J. C. Aster, and W. S. Pear. 1999. Notch1 expression in N. Yoshida, T. Kikuchi, Y. Iwakura, et al. 2005. Overexpression of the Runx3 early lymphopoiesis influences B versus T lineage determination. Immunity 11: transcription factor increases the proportion of mature thymocytes of the CD8 299–308. single-positive lineage. J. Immunol. 174: 2627–2636. 10 IN VITRO GENERATED CD8 T CELLS

50. Sun, G., X. Liu, P. Mercado, S. R. Jenkinson, M. Kypriotou, L. Feigenbaum, 56. Houston, E. G., Jr., and P. J. Fink. 2009. MHC drives TCR repertoire shaping, but P. Gale´ra, and R. Bosselut. 2005. The zinc finger protein cKrox directs CD4 not maturation, in recent thymic emigrants. J. Immunol. 183: 7244–7249. lineage differentiation during intrathymic T cell positive selection. Nat. Immu- 57. Li, O., P. Zheng, and Y. Liu. 2004. CD24 expression on T cells is required for nol. 6: 373–381. optimal T cell proliferation in lymphopenic host. J. Exp. Med. 200: 1083–1089. 51. Surh, C. D., and J. Sprent. 2000. Homeostatic T cell proliferation: how far can 58. Blackburn, S. D., H. Shin, W. N. Haining, T. Zou, C. J. Workman, A. Polley, T cells be activated to self-ligands? J. Exp. Med. 192: F9–F14. M. R. Betts, G. J. Freeman, D. A. Vignali, and E. J. Wherry. 2009. Coregulation 52. Tan, J. T., E. Dudl, E. LeRoy, R. Murray, J. Sprent, K. I. Weinberg, and of CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral C. D. Surh. 2001. IL-7 is critical for homeostatic proliferation and survival of infection. Nat. Immunol. 10: 29–37. naive T cells. Proc. Natl. Acad. Sci. USA 98: 8732–8737. 59. Osborne, B. A., and L. M. Minter. 2007. Notch signalling during peripheral 53. Hu, D., K. Ikizawa, L. Lu, M. E. Sanchirico, M. L. Shinohara, and H. Cantor. T-cell activation and differentiation. Nat. Rev. Immunol. 7: 64–75. 2004. Analysis of regulatory CD8 T cells in Qa-1-deficient mice. Nat. Immunol. 60. Taghon, T., I. Van de Walle, G. De Smet, M. De Smedt, G. Leclercq, 5: 516–523. B. Vandekerckhove, and J. Plum. 2009. Notch signaling is required for prolif- 54. Rifa’i, M., Y. Kawamoto, I. Nakashima, and H. Suzuki. 2004. Essential roles of eration but not for differentiation at a well-defined beta-selection checkpoint CD8+CD122+ regulatory T cells in the maintenance of T cell homeostasis. J. during human T-cell development. Blood 113: 3254–3263. Exp. Med. 200: 1123–1134. 61. Baumgartner, C. K., and L. P. Malherbe. 2011. Antigen-driven T-cell repertoire 55. Azzam, H. S., A. Grinberg, K. Lui, H. Shen, E. W. Shores, and P. E. Love. 1998. selection during adaptive immune responses. Immunol. Cell Biol. 89: 54–59. CD5 expression is developmentally regulated by T cell receptor (TCR) signals 62. Kronenberg, M., and Y. Kinjo. 2009. Innate-like recognition of microbes by and TCR avidity. J. Exp. Med. 188: 2301–2311. invariant natural killer T cells. Curr. Opin. Immunol. 21: 391–396. Downloaded from http://www.jimmunol.org/ by guest on September 24, 2021 SUPPLEMENTARY FIGURES

Supplementary Figure 1. Differentiation of BM-derived LSK HSCs into T cells in vitro. LSK HSCs were isolated from BM of 6-8 weeks old B6 mice. Developmental progression of T- lineage cells from early DN stages identified by the expression of CD44 and CD25 (left panels), to later stages identified by the expression of CD4 and CD8 (middle panels), and TCRβ on CD4- CD8+ gated cells (right panels), was examined on the indicated time by flow cytometry. Numbers inside the plots show the percentage of cells within each quadrant. The data are representative of at least 6 independent experiments.

Supplementary Figure 2. A) Flow cytometric analysis representative of CFSE-labeled in vitro- derived CD8 cells before and after injection into syngeneic (B6) mice and cocultured with OP9- DL1 cells. Flow cytometric analysis for CD4 and CD8 expression was performed from LN- isolated cells three days after adoptive transfer (left panels). In addition, CD4- CD8+ gated cells were analyzed for CFSE (middle panels) and CD24 and CFSE (right panels) in CTLR-un- injected and injected B6 mice. B) Flow cytometric analysis of CD8 and MHCI (H2b donor cells) was performed from LN-isolated three days after adoptive transfer into H2-mismatched (H2d) host mice. In addition, donor MHCI+ CD8+ gated cells were analyzed for CFSE (right panel). C) Flow cytometric analysis for CD4 and CD8 expression of CFSE-labeled in vitro-derived CD8 cells before and after injection into Rag2-/- mice was performed from LN-isolated cells seven days after adoptive transfer (left panels). In addition, CD4- CD8+ gated cells were analyzed for CFSE (middle panels and CD25 and CD44 expression (right panels) in CTLR-un-injected and injected mice.

Supplementary Figure 3. Analysis of expression of activation and memory CD8 T cell markers. A) Surface expression of memory T cell markers on in vitro- versus ex vivo-CD8 cells was examined after anti-CD3/CD28 stimulation for three days, with a subsequent addition of IL- 15 for an additional seven days. Histograms are shown for the indicated markers (black filled lines for specific mAbs, and white not filled lines for control isotype mAbs), respectively. Below each histogram, the corresponding MFIs are shown for the indicated samples (n>3); white bars = OP9-DL1; black bars = thymus. p>0.05 for each comparison . B) Gene expression analysis of

1 CD279 and CD28 from splenic CD8 T cells with (DL1) or without (GFP) exposure (OP9 cell coculture) to Notch ligand in absence (UST) or presence (ST) of anti-CD3/CD28-mediated activation for 72 hours.

2 Supplemental Figure 1

3 Supplemental Figure 2

4

Supplemental Figure 3

5 Suppl. Table 1A. A list by go function (A - L) of ≥2 fold differences in gene expression between in-vitro and ex-vivo CD8 T cells, as indicated in Figure 2a.

Gene Gene description Fold change

A. Cell Cycle Phase

INHBA inhibin, beta A (activin A, activin AB alpha polypeptide) 9.85276 CDKN2A cyclin-dependent kinase inhibitor 2A (melanoma, p16, inhibits CDK4) 9.10808 CDKN2B cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4) 9.0640075 BCAT1 branched chain aminotransferase 1, cytosolic 7.6966475 CDKN2C cyclin-dependent kinase inhibitor 2C (p18, inhibits CDK4) 6.0233975 CDC25C cell division cycle 25C 5.9218275 CCNA2 cyclin A2 5.7743525 SPO11 SPO11 meiotic protein covalently bound to DSB homolog (S. cerevisiae) 5.3182075 UBE2C ubiquitin-conjugating enzyme E2C 5.28832 BUB1 BUB1 budding uninhibited by benzimidazoles 1 homolog (yeast) 5.05039 NEK2 NIMA (never in mitosis gene a)-related kinase 2 4.8729775 NUSAP1 nucleolar and spindle associated protein 1 4.7599775 BIRC5 baculoviral IAP repeat-containing 5 (survivin) 4.5812225 CDKN1A cyclin-dependent kinase inhibitor 1A (p21, Cip1) 4.44498 KIF1 kinesin family member 11 4.39156 CENPF centromere protein F, 350/400ka (mitosin) 4.877315 TPX2 TPX2, microtubule-associated, homolog (Xenopus laevis) 4.1644 AURKA aurora kinase A 4.128655 KIF15 kinesin family member 15 4.0956975 HSPA2 heat shock 70kDa protein 2 4.401835 KIF2C kinesin family member 2C 4.0071375 NCAPH non-SMC condensin I complex, subunit H 3.8782675 ANLN anillin, actin binding protein 3.7753575 BUB1B BUB1 budding uninhibited by benzimidazoles 1 homolog beta (yeast) 3.6641925 NPM2 nucleophosmin/nucleoplasmin, 2 3.4499625 PLK1 polo-like kinase 1 (Drosophila) 3.4448025 CDCA5 cell division cycle associated 5 5.2061625 CENPE centromere protein E, 312kDa 3.4123875 CDC25B cell division cycle 25B 4.1292225 TTK TTK protein kinase 3.3491925 MAD2L1 MAD2 mitotic arrest deficient-like 1 (yeast) 3.3164725 CDC6 CDC6 cell division cycle 6 homolog (S. cerevisiae) 3.23863 ESPL1 extra spindle poles like 1 (S. cerevisiae) 3.220265 KIF22 kinesin family member 22 3.147505 CDKN3 cyclin-dependent kinase inhibitor 3 (CDK2-associated dual specificity phosphatase 3.10436 DBF4 DBF4 homolog (S. cerevisiae) 3.0523325 DLG7 discs, large homolog 7 (Drosophila) 3.0280475 SMC4 structural maintenance of chromosomes 4 4.221345 CD28 CD28 molecule 2.83533 RAD54L RAD54-like (S. cerevisiae) 2.7618125 KNTC1 kinetochore associated 1 2.70689 MSH5 mutS homolog 5 (E. coli) 2.695945 SKP2 S-phase kinase-associated protein 2 (p45) 2.6078775 CDK2AP1 CDK2-associated protein 1 2.44583 GFI1 growth factor independent 1 2.405355 CDK6 cyclin-dependent kinase 6 2.264075 RAD21 RAD21 homolog (S. pombe) 2.1623725 ACVR1B activin A receptor, type IB -2.9400575 APBB2 amyloid beta (A4) precursor protein-binding, family B, member 2 (Fe65-like) -3.3662025 APBB1 amyloid beta (A4) precursor protein-binding, family B, member 1 (Fe65) -3.9051975 B. Cell Cycle Process

INHBA inhibin, beta A (activin A, activin AB alpha polypeptide) 9.85276 CDKN2A cyclin-dependent kinase inhibitor 2A (melanoma, p16, inhibits CDK4) 9.10808 CDKN2B cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4) 9.06400 BCAT1 branched chain aminotransferase 1, cytosolic 7.69664 MYH10 myosin, heavy chain 10, non-muscle 6.74921 CDKN2C cyclin-dependent kinase inhibitor 2C (p18, inhibits CDK4) 6.02339 CDC25C cell division cycle 25C 5.92182 CCNA2 cyclin A2 5.77435 SPO11 SPO11 meiotic protein covalently bound to DSB homolog (S. cerevisiae) 5.31820 UBE2C ubiquitin-conjugating enzyme E2C 5.28832 BUB1 BUB1 budding uninhibited by benzimidazoles 1 homolog (yeast) 5.05039 TUBE1 tubulin, epsilon 1 6.44061 NEK2 NIMA (never in mitosis gene a)-related kinase 2 4.87297 NUSAP1 nucleolar and spindle associated protein 1 4.75997 BIRC5 baculoviral IAP repeat-containing 5 (survivin) 4.58122 CDKN1A cyclin-dependent kinase inhibitor 1A (p21, Cip1) 4.44498 KIF11 kinesin family member 11 4.39156 FBXO5 F-box protein 5 4.24261 CENPF centromere protein F, 350/400ka (mitosin) 4.87731 TPX2 TPX2, microtubule-associated, homolog (Xenopus laevis) 4.1644 AURKA aurora kinase A 4.12865 KIF15 kinesin family member 15 4.09569 HSPA2 heat shock 70kDa protein 2 4.40183 KIF2C kinesin family member 2C 4.00713 NCAPH non-SMC condensin I complex, subunit H 3.87826 ANLN anillin, actin binding protein 3.77535 PRC1 protein regulator of cytokinesis 1 3.69609 BUB1B BUB1 budding uninhibited by benzi midazoles 1 homolog beta (yeast) 3.66419 NPM2 nucleophosmin/nucleoplasmin, 2 3.44996 PLK1 polo-like kinase 1 (Drosophila) 3.44480 CDCA5 cell division cycle associated 5 5.20616 CENPE centromere protein E, 312kDa 3.41238 CDC25B cell division cycle 25B 4.12922 TTK TTK protein kinase 3.34919 MAD2L1 MAD2 mitotic arrest deficient-like 1 (yeast) 3.31647 CDC6 CDC6 cell division cycle 6 homolog (S. cerevisiae) 3.23863 ESPL1 extra spindle poles like 1 (S. cerevisiae) 3.22026 KIF22 kinesin family member 22 3.14750 CDKN3 cyclin-dependent kinase inhibitor 3 (CDK2-associated dual specificity phosphatase3.1043 DBF4 DBF4 homolog (S. cerevisiae) 3.05233 DLG7 discs, large homolog 7 (Drosophila) 3.02804 SMC4 structural maintenance of chromosomes 4 4.22134 CD28 CD28 molecule 2.83533 RAD54L RAD54-like (S. cerevisiae) 2.76181 KNTC1 kinetochore associated 1 2.70689 MSH5 mutS homolog 5 (E. coli) 2.69594 SKP2 S-phase kinase-associated protein 2 (p45) 2.60787 STMN1 stathmin 1/oncoprotein 18 2.46244 CDK2AP1CDK2-associated protein 12.4458 GFI1 growth factor independent 1 2.40535 CDK6 cyclin-dependent kinase 6 2.26407 RAD21 RAD21 homolog (S. pombe) 2.16237 ACVR1B activin A receptor, type IB -2.94005 APBB2 amyloid beta (A4) precursor protein-binding, family B, member 2 (Fe65-like) -3.36620 APBB1 amyloid beta (A4) precursor protein-binding, family B, member 1 (Fe65) -3.90519 C. Chromosome

MYCN v-myc myelocytomatosis viral related oncogene, neuroblastoma derived (avian) 32.5775 FOXC1 forkhead box C1 11.4086 LIG4 ligase IV, DNA, ATP-dependent 10.7160 RGS12 regulator of G-protein signalling 12 5.90364 BUB1 BUB1 budding uninhibited by benzimidazoles 1 homolog (yeast) 5.05039 SYCE2 synaptonemal complex central element protein 2 4.63772 BIRC5 baculoviral IAP repeat-containing 5 (survivin) 4.58122 TMPO thymopoietin 4.39185 CENPF centromere protein F, 350/400ka (mitosin) 4.87731 TOP2A topoisomerase (DNA) II alpha 170kDa 5.71189 BUB1B BUB1 budding uninhibited by benzimidazoles 1 homolog beta (yeast) 3.66419 OIP5 Opa interacting protein 5 3.56387 NPM2 nucleophosmin/nucleoplasmin, 2 3.44996 INCENP inner centromere protein antigens 135/155kDa 3.43892 CDCA5 cell division cycle associated 5 5.20616 CENPE centromere protein E, 312kDa 3.41238 MAD2L1 MAD2 mitotic arrest deficient-like 1 (yeast) 3.31647 HMGB2 high-mobility group box 2 4.50891 KIF22 kinesin family member 22 3.14750 NCAPD2 non-SMC condensin I complex, subunit D2 3.24440 SMC4 structural maintenance of chromosomes 4 4.22134 PCNA proliferating cell nuclear antigen 2.77722 HMGB1 high-mobility group box 1 2.69610 CENPA centromere protein A 2.61545 HIST4H4 histone cluster 4, H4 2.61130 REPIN1 replication initiator 1 2.51438 CBX1 chromobox homolog 1 (HP1 beta homolog Drosophila ) 2.35252 RFC4 replication factor C (activator 1) 4, 37kDa 2.29726 D. Cytokine production

INHBA inhibin, beta A (activin A, activin AB alpha polypeptide) 9.85276 GHRL ghrelin/obestatin preprohormone 3.243745 CD28 CD28 molecule 2.83533 SOD1 superoxide dismutase 1, soluble (amyotrophic lateral sclerosis 1 (adult)) 2.48924 MAST2 microtubule associated serine/threonine kinase 2 2.1613825 SPN sialophorin (leukosialin, CD43) -2.6967525 MALT1 mucosa associated lymphoid tissue lymphoma translocation gene 1 -2.6581775 LTB lymphotoxin beta (TNF superfamily, member 3) -3.222655 CARD11 caspase recruitment domain family, member 11 -2.99291 TLR1 toll-like receptor 1 -3.9954475 IFNG interferon, gamma -4.2016375 BCL3 B-cell CLL/lymphoma 3 -7.38563 TLR7 toll-like receptor 7 -15.56835 CRTAM cytotoxic and regulatory T cell molecule -16.603945 ABCA1 ATP-binding cassette, sub-family A (ABC1), member 1 -38.1056625

E. Endopeptidase activity

GZMA granzyme A (granzyme 1, cytotoxic T-lymphocyte-associated serine esterase 3) 70.1783 CAPN5 calpain 5 14.8611 F2 coagulation factor II (thrombin) 13.8919 MEP1A meprin A, alpha (PABA peptide hydrolase) 7.96006 CASP1 caspase 1, apoptosis-related cysteine peptidase (interleukin 1, beta, convertase) 7.13850 HTRA1 HtrA serine peptidase 1 6.81966 MMP2 matrix metallopeptidase 2 (72kDa gelatinase, 72kDa type IV collagenase) 6.47186 CASP3 caspase 3, apoptosis-related cysteine peptidase 3.40276 MMP17 matrix metallopeptidase 17 (membrane-inserted) 3.38327 USP2 ubiquitin specific peptidase 2 2.67901 PRSS3 protease, serine, 3 (mesotrypsin) -2.48350 PRSS2 protease, serine, 2 (trypsin 2) -2.60609 KLK8 kallikrein 8 (neuropsin/ovasin) -3.94256 F12 coagulation factor XII (Hageman factor) -5.16673 MMP15 matrix metallopeptidase 15 (membrane-inserted) -5.62370 CTSS cathepsin S -9.98214 UCHL1 ubiquitin carboxyl-terminal esterase L1 (ubiquitin thiolesterase) -4.69385 F. Defense Response

IL17RB interleukin 17 receptor B 32.663 CXCL1 chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha) 20.592 HP haptoglobin 17.346 PTX3 pentraxin-related gene, rapidly induced by IL-1 beta 13.864 INHBA inhibin, beta A (activin A, activin AB alpha polypeptide) 9.8527 HDAC9 histone deacetylase 9 9.0966 ANXA1 annexin A1 7.8148 CLEC1B C-type lectin domain family 1, member B 7.5857 LY96 lymphocyte antigen 96 5.3219 GATA3 GATA binding protein 3 4.6799 CEBPB CCAAT/enhancer binding protein (C/EBP), beta 6.5041 ADORA3 adenosine A3 receptor 3.6276 GHRL ghrelin/obestatin preprohormone 3.2437 BLNK B-cell linker 3.1903 TYROBP TYRO protein tyrosine kinase binding protein 2.8048 ALOX5AP arachidonate 5-lipoxygenase-activating protein 2.7451 ADORA2A adenosine A2a receptor 2.7343 CCR4 chemokine (C-C motif) receptor 4 2.2958 CD97 CD97 molecule 2.2893 TRAT1 T cell receptor associated transmembrane adaptor 1 2.3338 CXCR4 chemokine (C-X-C motif) receptor 4 2.4903 RIPK2 receptor-interacting serine-threonine kinase 2 -2.1253 IRAK2 interleukin-1 receptor-associated kinase 2 -2.2162 DMBT1 deleted in malignant brain tumors 1 -2.3913 IL10RB interleukin 10 receptor, beta -2.2698 SPN sialophorin (leukosialin, CD43) -2.6967 NFE2L1 nuclear factor (erythroid-derived 2)-like 1 -2.7366 LGALS3BP lectin, galactoside-binding, soluble, 3 binding protein -2.8160 CD48 CD48 molecule -2.5697 RSAD2 radical S-adenosyl methionine domain containing 2 -3.1350 PGLYRP2 peptidoglycan recognition protein 2 -2.4933 NFKB1 nuclear factor of kappa light polypeptide gene enhancer in B-cells 1 (p105) -3.3413 LBP lipopolysaccharide binding protein -3.6546 NCF2 neutrophil cytosolic factor 2 (65kDa, chronic granulomatous disease, autosomal 2) -4.317 CXCL10 chemokine (C-X-C motif) ligand 10 -4.3779 CCR5 chemokine (C-C motif) receptor 5 -4.4087 PGLYRP1 peptidoglycan recognition protein 1 -4.4810 GPR68 G protein-coupled receptor 68 -5.0449 LSP1 lymphocyte-specific protein 1 -5.2905 CX3CR1 chemokine (C-X3-C motif) receptor 1 -5.3109 CCL4 chemokine (C-C motif) ligand 4 -5.3865 PRF1 perforin 1 (pore forming protein) -7.1928 CCR7 chemokine (C-C motif) receptor 7 -7.2684 CCL5 chemokine (C-C motif) ligand 5 -5.8798 MX2 myxovirus (influenza virus) resistance 2 (mouse) -8.0269 TLR7 toll-like receptor 7 -15.568 NCF1 neutrophil cytosolic factor 1, (chronic granulomatous disease, autosomal 1) -16.225 S100A9 S100 calcium binding protein A9 -16.374 CRTAM cytotoxic and regulatory T cell molecule -16.603 RNASE6 ribonuclease, RNase A family, k6 -21.472 MGLL monoglyceride lipase -5.0966 MX1 myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) -10.567 CD83 CD83 molecule -32.819 G. Extracellular Matrix

COL3A1 collagen, type III, alpha 1 (Ehlers-Danlos syndrome type IV, autosomal dominant) 21.7811 FBN1 fibrillin 1 18.4811 FBLN5 fibulin 5 16.6341 COL5A2 collagen, type V, alpha 2 16.1729 COL16A1 collagen, type XVI, alpha 1 11.4138 COL6A3 collagen, type VI, alpha 3 10.0125 COL1A2 collagen, type I, alpha 2 9.89489 NID2 nidogen 2 (osteonidogen) 9.73156 PRELP proline/arginine-rich end leucine-rich repeat protein 9.30025 POSTN periostin, osteoblast specific factor 8.5544 DST dystonin 13.7796 COL5A3 collagen, type V, alpha 3 7.3648 COL8A1 collagen, type VIII, alpha 1 7.3461 CTGF connective tissue growth factor 6.5866 COL5A1 collagen, type V, alpha 1 4.8760 DGCR6 DiGeorge syndrome critical region gene 6 4.3852 COL15A1 collagen, type XV, alpha 1 3.9739 ADAMTS5 ADAM metallopeptidase with thrombospondin type 1 motif, 5 (aggrecanase-2) 3.3933 COMP cartilage oligomeric matrix protein 2.7778 SOD1 superoxide dismutase 1, soluble (amyotrophic lateral sclerosis 1 (adult)) 2.48924 FBLN2 fibulin 2 2.4249 PRSS2 protease, serine, 2 (trypsin 2) -2.6060 FBLN1 fibulin 1 -3.6985 H. Extracellular Region

FSTL1 follistatin-like 1 56.8173 COL3A1 collagen, type III, alpha 1 (Ehlers-Danlos syndrome type IV, autosomal dominant) 21.7811 CXCL1 chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha) 20.5929 FBN1 fibrillin 1 18.4811 FBLN5 fibulin 5 16.6341 COL5A2 collagen, type V, alpha 2 16.1729 F2 coagulation factor II (thrombin) 13.8919 COL16A1collagen, type XVI, alpha 1 11.4138 COL6A3 collagen, type VI, alpha 3 10.0125 COL1A2 collagen, type I, alpha 2 9.89489 INHBA inhibin, beta A (activin A, activin AB alpha polypeptide) 9.8527 NID2 nidogen 2 (osteonidogen) 9.7315 PRELP proline/arginine-rich end leucine-rich repeat protein 9.3002 POSTN periostin, osteoblast specific factor 8.5544 IL9R interleukin 9 receptor 8.4709 DST dystonin 13.7796 MEP1A meprin A, alpha (PABA peptide hydrolase) 7.96006 COL5A3 collagen, type V, alpha 3 7.36481 COL8A1 collagen, type VIII, alpha 1 7.34612 HTRA1 HtrA serine peptidase 1 6.81966 CTGF connective tissue growth factor 6.58667 MMP2 matrix metallopeptidase 2 (gelatinase A, 72kDa gelatinase, 72kDa type IV collagenase) 6.4718 ANGPTL2 angiopoietin-like 2 5.9484 DLK1 delta-like 1 homolog (Drosophila) 5.6933 COL5A1 collagen, type V, alpha 1 4.8760 DGCR6 DiGeorge syndrome critical region gene 6 4.3852 COL15A1 collagen, type XV, alpha 1 3.9739 CCL2 chemokine (C-C motif) ligand 2 3.9280 CCL7 chemokine (C-C motif) ligand 7 6.8617 SCUBE1 signal peptide, CUB domain, EGF-like 1 3.4824 ADAMTS5 ADAM metallopeptidase with thrombospondin type 1 motif, 5 (aggrecanase-2) 3.3933 GHRL ghrelin/obestatin preprohormone 3.2437 COMP cartilage oligomeric matrix protein 2.7778 PTN pleiotrophin (heparin binding growth factor 8, neurite growth-promoting factor 1) 2.5699 SOD1 superoxide dismutase 1, soluble (amyotrophic lateral sclerosis 1 (adult)) 2.48924 FBLN2 fibulin 2 2.4249 RNH1 ribonuclease/angiogenin inhibitor 1 -2.1124 MFNG manic fringe homolog (Drosophila) -2.1688 PRSS3 protease, serine, 3 (mesotrypsin) -2.4835 PRSS2 protease, serine, 2 (trypsin 2) -2.6060 SPN sialophorin (leukosialin, CD43) -2.6967 LGALS3BP lectin, galactoside-binding, soluble, 3 binding protein -2.8160 LBP lipopolysaccharide binding protein -3.6546 FBLN1 fibulin 1 -3.6985 KLK8 kallikrein 8 (neuropsin/ovasin) -3.9425 CCL4 chemokine (C-C motif) ligand 4 -5.3865

I. Extracellular Space

FSTL1 follistatin-like 1 56.8173 CXCL1 chemokine (C-X-C motif) ligand 1 (melanoma growth stimulating activity, alpha) 20.5929 FBN1 fibrillin 1 18.4811 F2 coagulation factor II (thrombin) 13.8919 INHBA inhibin, beta A (activin A, activin AB alpha polypeptide) 9.85276 IL9R interleukin 9 receptor 8.4709 MEP1A meprin A, alpha (PABA peptide hydrolase) 7.9600 HTRA1 HtrA serine peptidase 1 6.8196 MMP2 matrix metallopeptidase 2 (gelatinase A, 72kDa gelatinase, 72kDa type IV collagenase) 6.47186 ANGPTL2 angiopoietin-like 2 5.9484 DLK1 delta-like 1 homolog (Drosophila) 5.6933 CCL2 chemokine (C-C motif) ligand 2 3.92805 CCL7 chemokine (C-C motif) ligand 7 6.8617 SCUBE1 signal peptide, CUB domain, EGF-like 1 3.4824 GHRL ghrelin/obestatin preprohormone 3.2437 PTN pleiotrophin (heparin binding growth factor 8, neurite growth-promoting factor 1) 2.5699 SOD1 superoxide dismutase 1, soluble (amyotrophic lateral sclerosis 1 (adult)) 2.4892 MFNG manic fringe homolog (Drosophila) -2.1688 PRSS3 protease, serine, 3 (mesotrypsin) -2.4835 PRSS2 protease, serine, 2 (trypsin 2) -2.6060 SPN sialophorin (leukosialin, CD43) -2.6967 LGALS3BP lectin, galactoside-binding, soluble, 3 binding protein -2.8160 LBP lipopolysaccharide binding protein -3.6546 FBLN1 fibulin 1 -3.6985 KLK8 kallikrein 8 (neuropsin/ovasin) -3.9425 CCL4 chemokine (C-C motif) ligand 4 -5.3865 J. Intrinsic to Membrane

SLC22A3 solute carrier family 22 (extraneuronal monoamine transporter), member 3 22.621 TSPAN2 tetraspanin 2 21.173 IL17RB interleukin 17 receptor B 32.663 EMP1 epithelial membrane protein 1 19.605 LIFR leukemia inhibitory factor receptor alpha 17.749 CTLA4 cytotoxic T-lymphocyte-associated protein 4 17.577 LDLR low density lipoprotein receptor (familial hypercholesterolemia) 16.794 ADRA2A adrenergic, alpha-2A-, receptor 16.742 SLC2A4 solute carrier family 2 (facilitated glucose transporter), member 4 13.678 SDC1 syndecan 1 11.981 CXADR coxsackie virus and adenovirus receptor 9.7010 IL9R interleukin 9 receptor 8.4709 IFNGR1 interferon gamma receptor 1 8.4393 GPR56 G protein-coupled receptor 56 8.1476 KCNK5 potassium channel, subfamily K, member 5 8.1001 MEP1A meprin A, alpha (PABA peptide hydrolase) 7.9600 P2RX1 purinergic receptor P2X, ligand-gated ion channel, 1 7.8888 CLEC1B C-type lectin domain family 1, member B 7.5857 LRP12 low density lipoprotein-related protein 12 7.3770 SLC43A1 solute carrier family 43, member 1 7.3544 GPR34 G protein-coupled receptor 34 7.0110 EPHB2 EPH receptor B2 6.6705 SLC16A4 solute carrier family 16, member 4 (monocarboxylic acid transporter 5) 6.6577 CLDN4 claudin 4 6.4198 PALM paralemmin 6.2482 ITGA9 integrin, alpha 9 6.0206 TSPAN4 tetraspanin 4 5.7314 DLK1 delta-like 1 homolog (Drosophila) 5.6933 STOM stomatin 5.6908 ITGAM integrin, alpha M (complement component 3 receptor 3 subunit) 5.1881 SLC1A4 solute carrier family 1 (glutamate/neutral amino acid transporter), member 4 5.1567 FADS2 fatty acid desaturase 2 5.1458 PTPRJ protein tyrosine phosphatase, receptor type, J 6.5723 NOTCH3 Notch homolog 3 (Drosophila) 4.7181 PTPRF protein tyrosine phosphatase, receptor type, F 15.063 GALNT3 UDP-N-acetyl-alpha-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 3 (GalNAc-T3) 4.6895 OSMR oncostatin M receptor 4.5286 ACVR2B activin A receptor, type IIB 4.5272 FCER1G Fc fragment of IgE, high affinity I, receptor for; gamma polypeptide 4.4416 KCNJ8 potassium inwardly-rectifying channel, subfamily J, member 8 4.2469 APH1B anterior pharynx defective 1 homolog B (C. elegans) 4.1589 RYK RYK receptor-like tyrosine kinase 4.0668 JAG2 jagged 23.881 GPR65 G protein-coupled receptor 65 -2.6630 SLC14A1solute carrier family 14 (urea transporter), member 1 (Kidd blood group) -2.6758 SPN sialophorin (leukosialin, CD43) -2.6967 STIM1 stromal interaction molecule 1 -2.6984 ABCA3 ATP-binding cassette, sub-family A (ABC1), member 3 -2.7201 HOMER1 homer homolog 1 (Drosophila) -2.7596 LAX1 lymphocyte transmembrane adaptor 1 -2.7719 CNGA1 cyclic nucleotide gated channel alpha 1 -2.8669 CD48 CD48 molecule -2.5697 BACE1 beta-site APP-cleaving enzyme 1 -2.8875 ACVR1B activin A receptor, type IB -2.9400 TGFBR2 transforming growth factor, beta receptor II (70/80kDa) -2.9679 CNR2 cannabinoid receptor 2 (macrophage) -2.9919 NINJ2 ninjurin 2 -3.0112 IL2RG interleukin 2 receptor, gamma (severe combined immunodeficiency) -3.0147 ADCY7 adenylate cyclase 7 -3.0325 ABCC1 ATP-binding cassette, sub-family C (CFTR/MRP), member 1 -2.9164 FXYD5 FXYD domain containing ion transport regulator 5 -3.1795 MADD MAP-kinase activating death domain -3.2400 TRPM1 transient receptor potential cation channel, subfamily M, member 1 -3.3001 GRM6 glutamate receptor, metabotropic 6 -3.3082 SLC16A6 solute carrier family 16, member 6 (monocarboxylic acid transporter 7) -3.1499 TNFRSF9 tumor necrosis factor receptor superfamily, member 9 -3.3520 ITGB7 integrin, beta 7 -3.4514 GPR125 G protein-coupled receptor 125 -3.4635 MAN1C1 mannosidase, alpha, class 1C, member 1 -3.6230 AQP9 aquaporin 9 -3.6750 EFNB1 ephrin-B1 -3.7980 MYH9 myosin, heavy chain 9, non-muscle -2.3688 KCNMB4potassium large conductance calcium-activated channel, subfamily M, beta member 4 -3.9151 TLR1 toll-like receptor 1 -3.9954 ENTPD1 ectonucleoside triphosphate diphosphohydrolase 1 -4.1620 GABRR2 gamma-aminobutyric acid (GABA) receptor, rho 2 -4.2771 CCR5 chemokine (C-C motif) receptor 5 -4.4087 ARL6IP5 ADP-ribosylation-like factor 6 interacting protein 5 -4.4815 CD6 CD6 molecule -2.1515 HVCN1 hydrogen voltage-gated channel 1 -4.5353 HS3ST3B1 heparan sulfate (glucosamine) 3-O-sulfotransferase 3B1 -4.6211 IL27RA interleukin 27 receptor, alpha -4.6793 FFAR2 free fatty acid receptor 2 -4.8795 GPR114 G protein-coupled receptor 114 -4.9633 GPR68 G protein-coupled receptor 68 -5.0449 GCNT1 glucosaminyl (N-acetyl) transferase 1, core 2 (beta-1,6-N-acetylglucosaminyltransferase) -5.0891 EBI2 Epstein-Barr virus induced gene 2 (lymphocyte-specific G protein-coupled receptor) -5.0969 CX3CR1 chemokine (C-X3-C motif) receptor 1 -5.3109 ABCB11 ATP-binding cassette, sub-family B (MDR/TAP), member 11 -5.3712 IL2RB interleukin 2 receptor, beta -5.3967 MMP15 matrix metallopeptidase 15 (membrane-inserted) -5.6237 SLC7A4 solute carrier family 7 (cationic amino acid transporter, y+ system), member 4 -5.8517 ART1 ADP-ribosyltransferase 1 -6.1205 CLEC2D C-type lectin domain family 2, member D -6.6682 CD55 CD55 molecule, decay accelerating factor for complement (Cromer blood group) -4.7856 CXCR3 chemokine (C-X-C motif) receptor 3 -7.2596 CCR7 chemokine (C-C motif) receptor 7 -7.2684 ITGAE integrin, alpha E (antigen CD103, human mucosal lymphocyte antigen 1; alpha polypeptide) -7.4872 NKG7 natural killer cell group 7 sequence -4.0919 CD7 CD7 molecule -7.8601 GPR55 G protein-coupled receptor 55 -8.0356 NFAM1 NFAT activating protein with ITAM motif 1 -4.6989 ACVRL1 activin A receptor type II-like 1 -8.1566 CD200 CD200 molecule -8.2861 UST uronyl-2-sulfotransferase -9.1877 F2RL1 coagulation factor II (thrombin) receptor-like 1 -9.1942 SLC22A2 solute carrier family 22 (organic cation transporter), member 2 -9.3357 CD52 CD52 molecule -9.4332 CCR8 chemokine (C-C motif) receptor 8 -10.647 F2R coagulation factor II (thrombin) receptor -10.831 ENG endoglin (Osler-Rendu-Weber syndrome 1) -11.325 KCNC1 potassium voltage-gated channel, Shaw-related subfamily, member 1 -5.5851 DSCAM Down syndrome cell adhesion molecule -13.402 TSPAN9 tetraspanin 9 -14.363 GGTLA1 gamma-glutamyltransferase-like activity 1 -8.0929 TNFRSF25 tumor necrosis factor receptor superfamily, member 25 -3.3936 SORL1 sortilin-related receptor, L(DLR class) A repeats-containing -21.931 CHL1 cell adhesion molecule with homology to L1CAM (close homolog of L1) -12.936 CXCR6 chemokine (C-X-C motif) receptor 6 -27.158 CD226 CD226 molecule -9.3528 CD83 CD83 molecule -32.819 ABCA1 ATP-binding cassette, sub-family A (ABC1), member 1 -38.105 ST3GAL6 ST3 beta-galactoside alpha-2,3-sialyltransferase 6 -62.100 K. Phase of Mitotic Cycle

CDKN2B cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4) 9.0640 CDC25C cell division cycle 25C 5.9218 CCNA2 cyclin A2 5.7743 UBE2C ubiquitin-conjugating enzyme E2C 5.2883 BUB1 BUB1 budding uninhibited by benzimidazoles 1 homolog (yeast) 5.0503 NEK2 NIMA (never in mitosis gene a)-related kinase 2 4.8729 NUSAP1 nucleolar and spindle associated protein 1 4.7599 BIRC5 baculoviral IAP repeat-containing 5 (survivin) 4.5812 KIF11 kinesin family member 11 4.3915 TPX2 TPX2, microtubule-associated, homolog (Xenopus laevis) 4.1644 AURKA aurora kinase A 4.1286 KIF15 kinesin family member 15 4.0956 KIF2C kinesin family member 2C 4.0071 NCAPH non-SMC condensin I complex, subunit H 3.8782 ANLN anillin, actin binding protein 3.7753 BUB1B BUB1 budding uninhibited by benzimidazoles 1 homolog beta (yeast) 3.6641 NPM2 nucleophosmin/nucleoplasmin, 2 3.4499 PLK1 polo-like kinase 1 (Drosophila) 3.4448 CDCA5 cell division cycle associated 5 5.2061 CENPE centromere protein E, 312kDa 3.4123 CDC25B cell division cycle 25B 4.1292 TTK TTK protein kinase 3.3491 MAD2L1 MAD2 mitotic arrest deficient-like 1 (yeast) 3.3164 ESPL1 extra spindle poles like 1 (S. cerevisiae) 3.2202 KIF22 kinesin family member 22 3.1475 DLG7 discs, large homolog 7 (Drosophila) 3.0280 SMC4 structural maintenance of chromosomes 4 4.2213 CD28 CD28 molecule 2.8353 KNTC1 kinetochore associated 1 2.7068 L. Phase

CDKN2B cyclin-dependent kinase inhibitor 2B (p15, inhibits CDK4) 9.0640 CDC25C cell division cycle 25C 5.9218 CCNA2 cyclin A2 5.7743 SPO11 SPO11 meiotic protein covalently bound to DSB homolog (S. cerevisiae) 5.3182 UBE2C ubiquitin-conjugating enzyme E2C 5.0503 NEK2 NIMA (never in mitosis gene a)-related kinase 2 4.8729 NUSAP1 nucleolar and spindle associated protein 1 4.7599 BIRC5 baculoviral IAP repeat-containing 5 (survivin) 4.5812 KIF11 kinesin family member 11 4.3915 TPX2 TPX2, microtubule-associated, homolog (Xenopus laevis) 4.1644 AURKA aurora kinase A 4.1286 KIF15 kinesin family member 15 4.0956 HSPA2 heat shock 70kDa protein 2 4.4018 KIF2C kinesin family member 2C 4.0071 NCAPH non-SMC condensin I complex, subunit H 3.8782 ANLN anillin, actin binding protein 3.7753 BUB1B BUB1 budding uninhibited by benzimidazoles 1 homolog beta (yeast) 3.6641 NPM2 nucleophosmin/nucleoplasmin, 2 3.4499 PLK1 polo-like kinase 1 (Drosophila) 3.4448 CDCA5 cell division cycle associated 5 5.2061 CENPE centromere protein E, 312kDa 3.4123 CDC25B cell division cycle 25B 4.1292 TTK TTK protein kinase 3.3491 MAD2L1 MAD2 mitotic arrest deficient-like 1 (yeast) 3.3164 ESPL1 extra spindle poles like 1 (S. cerevisiae) 3.2202 KIF22 kinesin family member 22 3.1475 DLG7 discs, large homolog 7 (Drosophila) 3.0280 SMC4 structural maintenance of chromosomes 4 4.2213 CD28 CD28 molecule 2.8353 RAD54L RAD54-like (S. cerevisiae) 2.7618 KNTC1 kinetochore associated 1 2.7068 MSH5 mutS homolog 5 (E. coli) 2.6959 RAD21 RAD21 homolog (S. pombe) 2.1623

Table 1B. Microarray gene expression analysis showing fold change in gene expression, ≥2 fold, of in vitro (Red) vs ex vivo (Green) derived CD8 T cells, as indicated in Figure 2A. The analysis shows fold change in gene expression after excluding genes that are highly expressed in OP9 genes only from the list shown in Table 1A.

Gene Name GENE Information Fold Change

GZMA granzyme A (granzyme 1, cytotoxic T-lymphocyte-associated serine esterase 3) 70.2 IL17RB interleukin 17 receptor B 32.7 MYCN v-myc myelocytomatosis viral related oncogene, neuroblastoma derived (avian) 32.6 SLC22A3 solute carrier family 22 (extraneuronal monoamine transporter), member 3 22.6 TSPAN2 tetraspanin 2 21.2 RAG1 recombination activating gene 1 21.0 CTLA4 cytotoxic T-lymphocyte-associated protein 4 17.6 CPA3 carboxypeptidase A3 (mast cell) 14.8 F2 coagulation factor II (thrombin) 13.9 GYS2 glycogen synthase 2 (liver) 12.2 CXADR coxsackie virus and adenovirus receptor 9.7 CDKN2A cyclin-dependent kinase inhibitor 2A (melanoma, p16, inhibits CDK4) 9.1 HDAC9 histone deacetylase 9 9.1 ODZ1 odz, odd Oz/ten-m homolog 1(Drosophila) 8.7 POSTN periostin, osteoblast specific factor 8.6 IL9R interleukin 9 receptor 8.5 KCNK5 potassium channel, subfamily K, member 5 8.1 P2RX1 purinergic receptor P2X, ligand-gated ion channel, 1 7.9 CLEC1B C-type lectin domain family 1, member B 7.6 DIO2 deiodinase, iodothyronine, type II 7.6 GPR34 G protein-coupled receptor 34 7.0 MAFB v-maf musculoaponeurotic fibrosarcoma oncogene homolog B (avian) 7.0 CLDN4 claudin 4 6.4 ITGA9 integrin, alpha 9 6.0 CD4 CD4 molecule 5.9 LSR lipolysis stimulated lipoprotein receptor 5.6 ITGAM integrin, alpha M (complement component 3 receptor 3 subunit) 5.2 GALNT3 UDP-N-acetyl-alpha-D-galactosamine:polypeptide N-acetylgalactosaminyltransferase 3 (GalNAc-T 4.7 KCNJ8 potassium inwardly-rectifying channel, subfamily J, member 8 4.2 JAG2 jagged 2 3.9 NCAPH non-SMC condensin I complex, subunit H 3.9 CAMK2A calcium/calmodulin-dependent protein kinase (CaM kinase) II alpha 3.9 ADORA3 adenosine A3 receptor 3.6 SCUBE1 signal peptide, CUB domain, EGF-like 1 3.5 IL12RB2 interleukin 12 receptor, beta 2 3.5 NCAPD2 non-SMC condensin I complex, subunit D2 3.2 BLNK B-cell linker 3.2 ACCN3 amiloride-sensitive cation channel 3 2.9 CD28 CD28 molecule 2.8 TYROBP TYRO protein tyrosine kinase binding protein 2.8 SLC23A1 solute carrier family 23 (nucleobase transporters), member 1 2.8 ALOX5AP arachidonate 5-lipoxygenase-activating protein 2.7 BARD1 BRCA1 associated RING domain 1 2.7 MSH5 mutS homolog 5 (E. coli) 2.7 TEC tec protein tyrosine kinase 2.6 PBX4 pre-B-cell leukemia transcription factor 4 2.5 CXCR4 chemokine (C-X-C motif) receptor 4 2.5 GFI1 growth factor independent 1 2.4 TRAT1 T cell receptor associated transmembrane adaptor 1 2.3 GCH1 GTP cyclohydrolase 1 (dopa-responsive dystonia) 2.3 CCR4 chemokine (C-C motif) receptor 4 2.3

ACP5 acid phosphatase 5, tartrate resistant -2.1 MFNG manic fringe homolog (Drosophila) -2.2 SELL selectin L (lymphocyte adhesion molecule 1) -2.2 STX2 syntaxin 2 -2.2 DMBT1 deleted in malignant brain tumors 1 -2.4 SYNGR1 synaptogyrin 1 -2.4 PRSS3 protease, serine, 3 (mesotrypsin) -2.5 PGLYRP2 peptidoglycan recognition protein 2 -2.5 CD48 CD48 molecule -2.6 KCNJ6 potassium inwardly-rectifying channel, subfamily J, member 6 -2.6 CD72 CD72 molecule -2.6 EDG6 endothelial differentiation, lysophosphatidic acid G-protein-coupled receptor, 6 -2.6 KRT18 keratin 18 -2.6 PRSS2 protease, serine, 2 (trypsin 2) -2.6 SEMA4F sema domain, immunoglobulin domain (Ig), transmembrane domain (TM) and short cytoplasmic d -2.6 GPR65 G protein-coupled receptor 65 -2.7 SLC14A1 solute carrier family 14 (urea transporter), member 1 (Kidd blood group) -2.7 SPN sialophorin (leukosialin, CD43) -2.7 DUSP2 dual specificity phosphatase 2 -2.7 LAX1 lymphocyte transmembrane adaptor 1 -2.8 CNGA1 cyclic nucleotide gated channel alpha 1 -2.9 CNR2 cannabinoid receptor 2 (macrophage) -3.0 CARD11 caspase recruitment domain family, member 11 -3.0 NINJ2 ninjurin 2 -3.0 IL2RG interleukin 2 receptor, gamma (severe combined immunodeficiency) -3.0 RSAD2 radical S-adenosyl methionine domain containing 2 -3.1 LTB lymphotoxin beta (TNF superfamily, member 3) -3.2 TRPM1 transient receptor potential cation channel, subfamily M, member 1 -3.3 GRM6 glutamate receptor, metabotropic 6 -3.3 TNFRSF9 tumor necrosis factor receptor superfamily, member 9 -3.4 DAPK2 death-associated protein kinase 2 -3.5 TMEM176B transmembrane protein 176B -3.5 AQP9 aquaporin 9 -3.7 TNF tumor necrosis factor (TNF superfamily, member 2) -3.7 KCNMB4 potassium large conductance calcium-activated channel, subfamily M, beta member 4 -3.9 KLK8 kallikrein 8 (neuropsin/ovasin) -3.9 NKG7 natural killer cell group 7 sequence -4.1 ENTPD1 ectonucleoside triphosphate diphosphohydrolase 1 -4.2 IFNG interferon, gamma -4.2 GABRR2 gamma-aminobutyric acid (GABA) receptor, rho 2 -4.3 F13A1 coagulation factor XIII, A1 polypeptide -4.3 NCF2 neutrophil cytosolic factor 2 (65kDa, chronic granulomatous disease, autosomal 2) -4.3 CXCL10 chemokine (C-X-C motif) ligand 10 -4.4 CCR5 chemokine (C-C motif) receptor 5 -4.4 ABCG1 ATP-binding cassette, sub-family G (WHITE), member 1 -4.5 PGLYRP1 peptidoglycan recognition protein 1 -4.5 HS3ST3B1 heparan sulfate (glucosamine) 3-O-sulfotransferase 3B1 -4.6 IL27RA interleukin 27 receptor, alpha -4.7 NFAM1 NFAT activating protein with ITAM motif 1 -4.7 CD55 CD55 molecule, decay accelerating factor for complement (Cromer blood group) -4.8 GPR114 G protein-coupled receptor 114 -5.0 GPR68 G protein-coupled receptor 68 -5.0 EBI2 Epstein-Barr virus induced gene 2 (lymphocyte-specific G protein-coupled receptor) -5.1 F12 coagulation factor XII (Hageman factor) -5.2 CX3CR1 chemokine (C-X3-C motif) receptor 1 -5.3 CCL4 chemokine (C-C motif) ligand 4 -5.4 KCNC1 potassium voltage-gated channel, Shaw-related subfamily, member 1 -5.6 SLC7A4 solute carrier family 7 (cationic amino acid transporter, y+ system), member 4 -5.9 ART1 ADP-ribosyltransferase 1 -6.1 PRF1 perforin 1 (pore forming protein) -7.2 CXCR3 chemokine (C-X-C motif) receptor 3 -7.3 CCR7 chemokine (C-C motif) receptor 7 -7.3 ITGAE integrin, alpha E (antigen CD103, human mucosal lymphocyte antigen 1; alpha polypeptide) -7.5 CD7 CD7 molecule -7.9 GPR55 G protein-coupled receptor 55 -8.0 CD200 CD200 molecule -8.3 F2RL1 coagulation factor II (thrombin) receptor-like 1 -9.2 SLC22A2 solute carrier family 22 (organic cation transporter), member 2 -9.3 CD226 CD226 molecule -9.4 CD52 CD52 molecule -9.4 CTSS cathepsin S -10.0 GGT1 gamma-glutamyltransferase 1 -10.0 MX1 myxovirus (influenza virus) resistance 1, interferon-inducible protein p78 (mouse) -10.6 CCR8 chemokine (C-C motif) receptor 8 -10.6 SCG5 secretogranin V (7B2 protein) -12.1 CHL1 cell adhesion molecule with homology to L1CAM (close homolog of L1) -12.9 DSCAM Down syndrome cell adhesion molecule -13.4 TLR7 toll-like receptor 7 -15.6 KLK7 kallikrein 7 (chymotryptic, stratum corneum) -15.9 NCF1 neutrophil cytosolic factor 1, (chronic granulomatous disease, autosomal 1) -16.2 S100A9 S100 calcium binding protein A9 -16.4 CRTAM cytotoxic and regulatory T cell molecule -16.6 RNASE6 ribonuclease, RNase A family, k6 -21.5 SORL1 sortilin-related receptor, L(DLR class) A repeats-containing -21.9 CXCR6 chemokine (C-X-C motif) receptor 6 -27.2 CD83 CD83 molecule -32.8 ST3GAL6 ST3 beta-galactoside alpha-2,3-sialyltransferase 6 -62.1