of February 13, 2018.This information is current as

TransplantationAllogeneic and Xenogeneic Stem CellHumanization of Mice: Implications for

Mediated Immune−Dendritic Cell

Ganser and Renata StripeckeLaura Macke, Bala Sai Sundarasetty, Torsten Witte, Arnold Constanca Figueiredo, Florian Länger, Andreas Schneider,Münz, Ana Raykova, Carlos A. Guzmán, Peggy Riese, Gustavo Salguero, Anusara Daenthanasanmak, Christian

ol.1302887http://www.jimmunol.org/content/early/2014/04/16/jimmun

published online 16 April 2014J Immunol 

MaterialSupplementary

7.DCSupplementalhttp://www.jimmunol.org/content/suppl/2014/04/16/jimmunol.130288

        average*  

4 weeks from acceptance to publicationSpeedy Publication! •    

Every submission reviewed by practicing scientistsNo Triage! •    

from submission to initial decisionRapid Reviews! 30 days* •    

?The JIWhy

Subscriptionhttp://jimmunol.org/subscription

is online at: The Journal of ImmunologyInformation about subscribing to

Permissionshttp://www.aai.org/About/Publications/JI/copyright.htmlSubmit copyright permission requests at:

Email Alertshttp://jimmunol.org/alertsReceive free email-alerts when new articles cite this article. Sign up at:

Print ISSN: 0022-1767 Online ISSN: 1550-6606. Immunologists, Inc. All rights reserved.Copyright © 2014 by The American Association of1451 Rockville Pike, Suite 650, Rockville, MD 20852The American Association of Immunologists, Inc.,

is published twice each month byThe Journal of Immunology

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

by guest on February 13, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

The Journal of Immunology

Dendritic Cell–Mediated Immune Humanization of Mice:Implications for Allogeneic and Xenogeneic Stem CellTransplantation

Gustavo Salguero,* Anusara Daenthanasanmak,* Christian M€unz,† Ana Raykova,†

Carlos A. Guzman,‡ Peggy Riese,‡ Constanca Figueiredo,x Florian Langer,{

Andreas Schneider,* Laura Macke,* Bala Sai Sundarasetty,* Torsten Witte,‖

Arnold Ganser,* and Renata Stripecke*

De novo regeneration of immunity is a major problem after allogeneic hematopoietic stem cell transplantation (HCT). HCT mod-

eling in severely compromised immune-deficient animals transplanted with human stem cells is currently limited because of in-

complete maturation of lymphocytes and scarce adaptive responses. Dendritic cells (DC) are pivotal for the organization of

lymph nodes and activation of naive Tand B cells. Human DC function after HCT could be augmented with adoptively transferred

donor-derived DC. In this study, we demonstrate that adoptive transfer of long-lived human DC coexpressing high levels of human

IFN-a, human GM-CSF, and a clinically relevant Ag (CMV pp65 protein) promoted human lymphatic remodeling in immune-

deficient NOD.Rag12/2.IL-2rg2/2 mice transplanted with human CD34+ cells. After immunization, draining lymph nodes became

replenished with terminally differentiated human follicular Th cells, plasma B cells, and memory helper and cytotoxic T cells.

Human Igs against pp65 were detectable in plasma, demonstrating IgG class-switch recombination. Human T cells recovered

from mice showed functional reactivity against pp65. Adoptive immunotherapy with engineered DC provides a novel strategy for

de novo immune reconstitution after human HCT and a practical and effective tool for studying human lymphatic regeneration

in vivo in immune deficient xenograft hosts. The Journal of Immunology, 2014, 192: 000–000.

Transplantation of hematopoietic stem cells (HSC) usingHLA-matched G-CSF–mobilized peripheral blood hasshown long-term successful outcomes for treatment of

high-risk hematologic malignancies (1) and for restoring inborngenetic hematopoiesis dysfunctions (2). It is broadly estimatedthat only 30% of patients being currently transplanted have anHLA-identical donor. The remaining of the patients receives HSCfrom haploidentical donors from the family or from HLA-matchedor mismatched unrelated donors (3). Despite the high risk of acuteand chronic graft-versus-host disease (GvHD), when hematopoi-

etic stem cell transplantation (HCT) is performed with humanadult HSC from unrelated donors, high rates of clinical successhave been reported for chronic and acute myeloid leukemia (3).Because of long-term reconstitution of the donor graft and asso-ciated benefits of graft-versus-leukemia effects, HCT pioneeredseveral concepts of clinical immune therapies (4, 5). Nevertheless,a profound cellular immunodeficiency is commonly observedduring the first 10–15 wk posttransplantation because of cytotoxicchemo/radiotherapy preconditioning regimens and several monthsrequired for the low numbers of seeding stem cells to reconstitute

*Department of Hematology, Hemostasis, Oncology and Stem Cell Transplantation,Hannover Medical School, 30625 Hannover, Germany; †Institute of ExperimentalImmunology University of Zurich, CH-8057 Zurich, Switzerland; ‡Department ofVaccinology and Applied Microbiology, Helmholtz Centre for Infection Research,38124 Braunschweig, Germany; xDepartment of Transfusion Medicine, HannoverMedical School, 30625 Hannover, Germany; {Intitute for Pathology, Hannover Med-ical School, 30625 Hannover, Germany; and ‖Clinic for Immunology and Rheuma-tology, Hannover Medical School, 30625 Hannover, Germany

Received for publication October 30, 2013. Accepted for publication March 9, 2014.

This work was supported by the Deutsche Forschungsgemeinschaft/REBIRTHExcellence Cluster (to R.S. and C.F.). A.D. was a Ph.D. fellow of the Centerfor Infection Biology (Zentrum fuer Infectionsforschung) within the framework ofthe Hannover Biomedical Research School graduate program at the MedizinischeHochschule Hannover. This study was further supported by grants of the DeutscheForschungsgemeinschaft/SFB738 and Else-Kroener-Fresenius Stiftung (to R.S.).Support for C.M.’s laboratory was obtained from the National Cancer Institute (GrantR01CA108609), the Sassella Foundation (Grants 10/02, 11/02, and 12/02), CancerResearch Switzerland (Grant KFS-02652-08-2010), the Association for InternationalCancer Research (Grant 11-0516), Klinische Forschungsschwerpunkt-Programm(Multiple Sclerosis Clinical Research Priority Program) and Klinische For-schungsschwerpunkt-Programm (Hemato-Lymphoid Diseases Clinical Research Pri-ority Program) of the University of Zurich, the Vontobel Foundation, the BaugartenFoundation, the European Molecular Biology Organization Foundation, the SobekFoundation, Fondation Acteria, Novartis, and the Swiss National Science Foundation(Grants 310030_143979 and CRSII3_136241).

G.S. designed the experiments, conducted experiments, analyzed data, and wrote thefirst manuscript draft; A.D. assisted in the planning and execution of pp65-specific

immune monitoring assays; C.M. and A.R. performed analysis of B cell populationsand follicular T cells, interpreted data, and revised the manuscript; C.A.G. and P.R.performed human Ab detection and revised the manuscript; C.F. performed cytokinearray analyses; F.L. performed the preparation and microscopic analyses of the tissuesamples; A.S. performed immune-histochemistry analyses. L.M. coordinated clinicalsamples acquisition. B.S.S. performed the real-time PCR analyses; T.W. providedplasma samples from systemic lupus erythematosus patients; A.G. coordinated theprocurement of clinical G-CSF–mobilized PBL and revised the manuscript; and R.S.planned the project, designed the experiments, obtained funding, enrolled collabo-rators, participated in study design, interpreted the data, and edited the final revisedmanuscript.

Address correspondence and reprint requests to Dr. Renata Stripecke, Department ofHematology, Hemostasis, Oncology and Stem Cell Transplantation, RegenerativeImmune Therapies Applied/Hannover Medical School, Carl Neuberg Strasse 1 -OE 6860/D-30625 Hannover, Germany. E-mail address: stripecke.renata@mh-hannover.de

The online version of this article contains supplemental material.

Abbreviations used in this article: DC, dendritic cell; fLUC, firefly Luciferase;GvHD, graft-versus-host disease; HCT, hematopoietic stem cell transplantation;HIS, humanized immune system; HSC, hematopoietic stem cell; hu, human; ID-LV,integrase-defective lentiviral vector; LN, lymph node; LYVE, lymphatic endothelialcell; mo-DC, monocyte-derived DC; NRG, NOD.Rag12/2.IL-2rg2/2; PB, Pacificblue; SmyleDC, self-differentiated myeloid-derived lentivirus-induced DC; Tfh, fol-licular T helper.

Copyright� 2014 by TheAmerican Association of Immunologists, Inc. 0022-1767/14/$16.00

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1302887

Published April 16, 2014, doi:10.4049/jimmunol.1302887 by guest on February 13, 2018

http://ww

w.jim

munol.org/

Dow

nloaded from

the recipient (2, 6). Recovery of the T cell compartment initiallyrelies on the expansion of memory T cells prior to possible denovo activation of naive T cells, which may take up to 6 mo post-HCT (2, 6). Reconstitution of the B cell compartment and itsassociated humoral immunity may take up to 2 y post-HCT andis characterized by initial repopulation of transitional B cells inperipheral blood, followed by a late increase in frequencies ofmature B cells (7). This period of adaptive T and B cell immunecell deficiency is associated with high morbidity and high risk ofnonrelapse-associated mortality because of opportunistic viraland fungal infections (6, 8, 9). Thus, novel immunization strat-egies and predictive in vivo models aiming to improve andvalidate the recovery of adaptive immune responses are required.Dendritic cells (DC) play a central role in the direct stimulation

of adaptive immune responses against Ags and trigger the re-generation and remodeling of secondary and tertiary lymphatictissues that are key for immune synapses between T and B cells tooccur (10, 11). Because DC are a rare population in peripheralblood, ex vivo generation has been established with monocytesexposed to combinations of inflammatory cytokines that lead todifferentiation of postmitotic and nonreplicating DC displayingnatural myeloid DC phenotype and Ag presentation functions.One major inconvenience of clinical “conventional” monocyte-derived (mo-)DC is that migration to lymph nodes (LN) afteradministration is rather poor (12).Self-differentiated myeloid-derived lentivirus-induced DC

(SmyleDC) can be rapidly generated after overnight infection ofmonocytes with integrase-defective lentiviral vectors (ID-LV) coex-pressing human GM-CSF, IFN-a, and the CMV pp65 antigenic protein(13). The pp65 tegument protein is known to be efficiently presentedby several HLA types to public and private TCRs (14). SmyleDC werelong-lived in immune-deficient NOD.Rag12/2.IL-2rg2/2 (NRG) mice(.4 wk), secreted several inflammatory cytokines, and effectivelyenhanced the expansion of adoptively transferred human PBL(huPBL) for anti-pp65 T cell activation in vivo (13). In this study, wecompared the effects of immunization with conventional (mo-DC)versus SmyleDC after autologous G-CSF–mobilized human CD34+

HCT into NRG mice. Post-HCT immunization with long-lived Smy-leDC (but not mo-DC) stimulated the regeneration of draining LN andpromoted de novo adaptive human immune responses against theclinically relevant pp65 Ag. This model validates preclinically thepotency of a genetically modified DC product and its associatedimmune effects on a robust and predictive humanized immunesystem (HIS) mouse model.

Materials and MethodsLentiviral vector construction and integrase-defectivelentivirus production

The self-inactivating (SIN) lentiviral backbone vector and the mono-cistronic vectors expressing the CMV-pp65 were described previously (13,15). Construction of the bicistronic lentiviral vector expressing the human(hu)GM-CSF and of the huIFN-a (LV-G2a) interspaced with a P2A ele-ment (RRL-cPPT-CMV-hGMCSF-P2A-huIFN-a) was constructed andextensively characterized as described previously (13). The structural in-tegrity of all constructs was reconfirmed by restriction digestion and se-quencing analysis of the promoters and transgenes. Large-scale lentivirusproduction was performed by transient cotransfection of human embryonickidney 293T cells as formerly described (16). To generate integrase-defective lentivirus, four packaging plasmids were used in the cotrans-fection: the plasmid containing the lentiviral vector expressing the cytokines,the plasmid expressing gag/pol containing a D64V point mutation in theintegrase gene (pcDNA3g/pD64V.4xCTE), the plasmid expressing rev(pRSV-REV), and the plasmid encoding the VSV-G envelope (pMD.G).Virus supernatants were collected and concentrated by ultracentrifugation,and the titers were evaluated by assessing p24 Ag concentration with ELISA(Cell Biolabs, San Diego, CA). One microgram of p24 equivalent per mil-liliter corresponds to ∼1 3 107 infective viral particles/ml.

Human CD34-positive peripheral blood stem cell isolation

PBMCs were obtained from leukapheresis of hematopoietic adult stem celltransplantation adult donors subjected to HSC mobilization regimen withG-CSF (Granocyte; Chugai Pharma, Berkeley Heights, NJ). All studieswere performed in accordance with protocols approved by the HannoverMedical School Ethics Review Board. HSC CD34+ cells were positivelyselected by MACS using a CD34 magnetic cell isolation kit (MiltenyiBiotec, Bergisch Gladbach, Germany). After two rounds of positivemagnetic selection, cell purity obtained was .99% with a contamination ofCD3+ T cells below 0.2%, as evaluated by flow cytometry.

Generation of human mo- and Smyle DCs,

The autologous CD34-negative PBMC fraction was used for further positiveselection of CD14+ monocytes using CD14 isolation beads (MiltenyiBiotec). For lentiviral gene transfer, monocytes were kept in culture in6-well plates conferring low cell adherence (flat-bottom for suspensioncells; Sarstedt, N€umbrecht-Rommelsdorf, Germany) with serum-freeCellgro medium in the presence of recombinant huGM-CSF and IL-4 (50ng/ml each; Cellgenix, Freiburg, Germany) for 8 h prior to transduction(Supplemental Fig. 1). For generation of SmyleDC, 5 3 106 CD14+

monocytes were transduced with 2.5 mg/ml p24 equivalent (multiplicity ofinfection of 5) of both ID-LV-G2a and ID-LV-pp65 in the presence of 5 mg/mlprotamine sulfate (Valeant, D€usseldorf, Germany). After 16 h of trans-duction, SmyleDC were harvested by gently resuspension of the looselyadherent cells, washed twice with PBS, and further maintained in culturewith serum-free Cellgro medium. For production of conventional mo-DC,monocytes were incubated with ID-LV-pp65 as described above. Following16-h transduction, LV was removed, and cells were further maintained inculture for 7 d in the presence of recombinant huGM-CSF (50 ng/ml) andIFN-a (1000 U/ml; PBL IFNSource, Piscataway, NJ). Cytokines werereplenished every 3 d. For mouse immunizations, SmyleDC directly aftertransduction or mo-DC at day 7 of culture were resuspended in PBS andused for mice injection. Viability, DC immunophenotype, and cytokine re-lease were assessed in Smyle or mo-DCs after 7, 14, and 21 d of culture. Thenumber of viable counts was determined by trypan blue exclusion.

Mouse transplantation with human HSC

NOD.Cg-Rag1tm1Mom Il2rgtm1Wjl (NOD;Rag12/2;IL-2rg2/2, NRG) micewere bred in-house and maintained under pathogen-free conditions in anIVC system (BioZone, Burton-on-Trent, U.K.). All procedures involvingmice were reviewed and approved by the Lower Saxony and followed theguidelines provided by the Animal Facility at Hannover Medical School.For HSC transplantation, 4-wk-old mice were sublethally irradiated(450 cGy) using a [137Cs] column irradiator (Gammacell 3000 Elan; BestTheratronics, Ottawa, ON, Canada). Mouse recipients were i.v. injectedwith 5 3 105 human CD34+ cells into the tail vein. Mice were bled atdifferent time points (on weeks 6, 10, and 13) after human HSC trans-plantation to monitor the status of human hematopoietic cell engraftmentand were sacrificed at week 20 for final analyses. DC injections wereperformed at 10 wk after HSC transplantation, followed by a boost on theweek 11. Briefly, Smyle or mo-DC were collected from culture plates,resuspended at a concentration of 5 3 105 cells in 100 ml PBS, and s.c.injected into the mouse right hind limb using a 27-gauge needle.

Flow cytometry analyses

Engraftment of human hematopoietic cells in human HSC-reconstitutedmice was evaluated in peripheral blood, spleens, and LN using the fol-lowing mouse anti-human Abs: PerCP anti-CD45, Alexa700 anti-CD19,Pacific blue (PB) anti-CD4, allophycocyanin anti-CD3, PE-Cy7 anti-CD8, FITC anti-CD45RA, and PE anti-CD62L (BioLegend); and PEanti-CD14, FITC anti-Lineage positive, allophycocyanin anti-CD11c, andPE anti-CD123 (BD Biosciences). For characterization of human B cellssubpopulations, the next fluorochrome-conjugated Abs were used: PB anti-CD45, Brilliant Violet 605 anti-CD19, PE anti-CD27, PE-Cy7 anti-CD38,FITC anti-IgD, Alexa700 anti-IgG, allophycocyanin anti-IgM, PerCP-Cy5.5 anti-CD24, and allophycocyanin-C7 anti-CD3. Follicular Th cellswere characterized by staining with PB anti-CD45, Alexa700 anti-CD14/CD19, FITC anti-CD3, allophycocyanin-C7 anti-CD4, PerCP-Cy5.5 anti-CXCR5, allophycocyanin anti-PD1, and PE-Cy7 anti-ICOS. For peripheralblood analyses, blood was lysed by two rounds of incubation with eryth-rocyte lysis buffer (0.83% ammonium chloride/20 mM HEPES [pH 7.2])for 5 min at room temperature, followed by stabilization with cold PBSand centrifugation for 5 min at 3003 g. Cells were incubated with Abs for30 min at 4˚C. Harvested spleen cells were treated with erythrocyte lysisbuffer (0.83% ammonium chloride/20 mM HEPES [pH 7.2]) for 5 min,washed with PBS, and incubated with Abs for 30 min on ice. After

2 HUMAN ADAPTIVE RESPONSES IN MICE

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

a washing step, cells were resuspended in PBS and acquired in LSR-II orLSR Fortessa flow cytometers (BD Biosciences). For DC phenotypic char-acterization, the following anti-human Abs were used: allophycocyanin anti-CD11c, PE anti-CD14, allophycocyanin anti-CD3, PE anti-CD80, PerCPanti-HLA-DR, allophycocyanin anti-CD86, allophycocyanin anti-CD83 (BDBiosciences), and FITC anti-CMVpp65 (Pierce Biotechnology, Rockford,IL). For DC staining, cells were collected, washed once with PBS, and in-cubated with mouse IgG (50 mg/ml) on ice for 15 min, followed by incu-bation with Abs. Cells were washed, resuspended in cell fix solution (BDBiosciences), and further analyzed using a FACSCalibur cytometer. Allanalyses were performed using FloJo (Tree Star, Ashland, OR) software.

LN drainage analyses

Evaluation of the hind-limb lymphatic drainage was adapted from amethodology described previously (17). Briefly, mice were s.c. injectedwith 20–30 ml 5% Evans blue into the right hind limb. After injection, dyewas allowed to be taken up by lymphatic vessels for 30 min. Mice wereeuthanized and dissected to locate the inguinal and axillary draining LNand the lymphatic vessels.

SmyleDC in vivo migration analyses with in vivobioluminescence imaging

HIS-NRG mice were immunized with SmyleDC on week 10 and 11 andfurther rechallenged with SmyleDC expressing the firefly Luciferase(fLUC) on week 17. Bioluminescence emission mediated by Luciferaseafter catalysis of the Luciferin substrate was monitored every week for up to3 wk for bioluminescence analysis. Briefly, mice were anesthetized withketamine (100 mg/kg i.p.) and xylazine (10 mg/kg i.p.), and an aqueoussolution of d-luciferin (150 mg/kg i.p.) was injected 5 min before imaging.Animals were placed into a dark chamber of the charge-coupled devicecamera (IVIS; Xenogen, Cranbury, NJ), and grayscale body surface ref-erence images (digital photograph) were taken under weak illumination.After the light source was switched off, photons emitted from luciferase-expressing cells within the animal body and transmitted through the tissuewere quantified over a defined time of up to 5 min using the software pro-gram Living Image (Xenogen) as an overlay on Igor (Wavemetrics, Seattle,WA). For anatomical localization, a pseudocolor image representing lightintensity (blue, least intense; red, most intense) was generated in LivingImage and superimposed over the grayscale reference image. Quantifiedluminescence consists in averaged photon radiance on the surface of theanimal and is expressed as photons/s/cm2/sr, where sr = steradian.

Real-time PCR for analyses of lentiviral vector copies intissues

Total genomic DNA was extracted from LN single-cell suspensions usingthe QiaAmp DNA blood mini kit (Qiagen, Hilden, Germany), according tothe manufacturer’s instructions. Vector-derived copy numbers were de-termined by real-time PCR as described previously (18, 19). A total of 100ng/2 ml genomic DNA prepared from the above step were added to the13-ml RQ-PCR mix containing 7.5 ml SYBRTaq mix with 1 ml primermix for woodchuck hepatitis virus posttranscriptional regulatory ele-ment, 59-GAGGAGTTGTGGCCCGTTGT-39 (forward), and 59-TGACA-GGTGGTGGCAATGCC-39 (reverse); or polypyrimidine tract bindingprotein 2, 59-TCTCCATTCCCTATGTTCATGC-39 (forward), and 59-GTTCCCGCAGAATGGTGAGGTG-39 (reverse), adjusting the volume to13 ml with PCR-grade nuclease-free water. A plasmid vector (pCR4-TOPO;provided by Dr. M. Rothe, Department of Experimental Hematology,Hannover Medical School, Hannover, Germany) containing these twoamplicons was used for standard curves. All samples were analyzed withStepOnePlus Real-Time PCR system (Applied Biosystems). The cyclingconditions were 10 min at 95˚C, 40 cycles of 15 s at 95˚C, 20 s at 56˚C,and 30s at 65˚C. Results were quantified by making use of primer pair–specific real-time PCR efficiencies and by comparing sample CT values toa standard curve generated with the plasmid vector (pCR4-TOPO). Datawere analyzed by StepOnePlus software (Applied Biosystems). The sen-sitivity threshold of this method permits reliable quantifications of lenti-viral copies when more than one copy per 100 ng DNA is detectable.

Functional analyses of pp65-CTLs recovered from mouse LNsand spleen

For evaluation of T cell immune responses against CMV-pp65, spleno-cytes were harvested, stained with allophycocyanin-conjugated anti-humanCD3, and sorted using a XDP cell sorter (Beckman Coulter). Human CD3+

cells were activated by human anti–CD2/CD3/CD28-conjugated magneticbeads (Miltenyi Biotec) in a bead-to-cell ratio of 1:2 and cultured inX-Vivo medium in the presence of 200 ng/ml human (IL)-2, 5 ng/ml huIL-7,

and 5 ng/ml IL-15. T cells were further expanded by coculture with SmyleDCin a DC-T cell ratio of 1:10 for additional 7 d. LN cells were also harvestedand directly incubated with SmyleDC as described above. For CMV-specificIFN-g production, expanded T cells isolated from spleen, LN, or PBMCfrom CMV-reactive healthy donors (20,000 cells) were seeded on an anti-human IFN-g–coated 96-well ELISPOT plate and incubated overnight in thepresence of 10 mg/ml pp65 overlapping peptide pool (Miltenyi Biotec) orno peptide. Next day, cells were washed, and plates were further incubatedwith biotin-conjugated anti-human IFN-g Abs, followed by alkalinephosphatase–conjugated streptavidin. Plates were developed using NBT/5-bromo-4-chloro-3-indolyl phosphate liquid substrate and analyzed in anAELVIS ELISPOT reader (AELVIS, Hannover, Germany).

Histology and immunohistochemistry analyses of humanhematopoietic cell engraftment

LN from human HSC-reconstituted NRG or C57BL/6 wild-type mice wereharvested and embedded in optimal cutting temperature compound (OCTSakuraFinetek, Torrance, CA) for cryopreservation. Frozen sections (5 mm) were fixedby acetone and stained with monoclonal anti-human CD3 (eBioscience, SanDiego, CA), anti-human Pe–Texas Red-conjugated CD8, anti human CD11c(eBioscience), allophycocyanin anti-human CD19 (eBioscience), anti-mouselymphatic endothelial cell (LYVE)-1 (DakoCytomation), and anti-mouseCD31 (BD Bioscience). Immunofluorescence analyses were performed in anAxiocam fluorescence microscope (Zeiss), and images were created usingAxiowert software (Zeiss). For histopathological analyses of xenograft-versus-host disease, representative samples from skin and colon were harvested androutinely formalin fixed and paraffin embedded. Two-micrometer sectionswere cut from the blocks and stained for H&E. An experienced hema-topathologist reviewed the slides blinded to the treatment group of the animals.A semiquantitative score (modified after Lerner et al.) (20) was used to scorethe histological changes. Grade 1 is defined by single or multiple apoptoticfigures without architectural changes. Grade 2 shows multiple apoptotic figuresand dropout of crypts or skin appendages. Grade 3 shows additionally surfacenecrosis and severe loss of crypts or skin appendages.

Ig production in HSC-NRG mice

Plasma was harvested from HSC-NRG mice 20 wk after reconstitution andscreened by ELISA for the presence of total human IgM an total human IgGas described elsewhere (21). Total IgM and IgG determination was per-formed by coating 96-well plates either with AffiniPure F(ab9)2 fragmentgoat anti-human IgM (Fc5m specific; Jackson ImmunoResearch Labora-tories) or AffiniPure goat anti-human IgG (Fcg fragment specific; JacksonImmunoResearch Laboratories) at a concentration of 1.2 mg/ml. For thedetection of anti–pp65-specific IgG and IgM, plates were coated with1 mg/ml pp65 recombinant protein. After coating, the plates were washedin ELISAwash buffer (PBS, 0.5% Tween 20), blocked with 1% casein, andfurther incubated with serial dilution of mouse plasma (starting at a dilution of1:1:100 for total IgG or 1:25 for total IgM detection). Serum samples for pp65-specific Abs were diluted 1:5 or 1:10 for IgG or IgM detection, respectively.Enzyme-conjugated detection Abs were added at a dilution of 1:2500 for HRP-conjugated anti-IgG and a dilution of 1:5000 for HRP-conjugated anti-IgM(both from Jackson ImmunoResearch Laboratories). TMB substrate/stop so-lution (BioSource International) was used for the development of the ELISA.

Analyses of human cytokines

Detection of human Th1/Th2 cytokines in DC culture supernatants andmouse plasma was performed by fluorescent bead–based 14-plex Luminexassay, according to the manufacturer’s protocol (Millipore). The 14-plexassay measured the following cytokines GM-CSF, IL-4, TNF-a, IL-6, IL-8, MCP-1, IL-10, IL-1b, IL-5, IL-13, IFN-g, IL-7, IL-2, and IL-12(p70).Detection of IFN-a in DC supernatants and mouse plasma was performedby commercially available ELISA kit (Mabtech, Minneapolis, MN).

Statistical analyses

Parametric (t test) and nonparametric (Kruskall–Wallis) statistical analyseswere performed to compare the differences among groups for engraftmentof human hematopoietic lineages in NRG mice. Analyses were performedin Graph prism 5th version software. All tests were two-sided, and p ,0.05 was considered significant.

ResultsSmyleDC generation and characterization in vitro

Gene transfer of huGM-CSF, huIFN-a, and the CMV-pp65 viralAg into human monocytes using ID-LV generated long-lasting

The Journal of Immunology 3

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

SmyleDC in vitro and in vivo (13). Under these conditions, usingthe GFP as a quantitative gene reporter, we observed between 10and 50% transduction efficiency (13). We adapted our protocol togenerate DC from the CD34– fraction recovered from PBMCobtained from G-CSF–mobilized HSC donors (Supplemental Fig.1). Control mo-DC were produced by transduction of monocyteswith an ID-LV vector expressing pp65 (Fig. 1A) and maintained inculture in the presence of recombinant huGM-CSF/huIFN-a. ForSmyleDC generation, monocytes were in addition cotransducedwith the bicistronic ID-LV expressing huGM-CSF/huIFN-a trans-genes (Fig. 1A) and maintained in the absence of cytokines.Levels of accumulated huIFN-a (2.0 ng/ml) and GM-CSF (0.3ng/ml) were detected in continuous culture of SmyleDC for up to21 d (Fig. 1B). Compared with mo-DC, SmyleDC cultures dis-played significantly higher cell viability (day 7, 45 versus 36%,p . 0.05; day 14, 35 versus 14% p = 0.021; day 21 17 versus 5%,p , 0.05; Fig. 1C) and intracellular pp65 expression (peak on14 d, 55 versus 21%, p = 0.014; Fig. 1D). Although we have not

validated pp65 as a quantitative gene reporter in DC, these datacorrelate with our experience using GFP as a marking gene. After21 d of culture, both DC types lost expression of the monocyticmarker CD14+ (Fig. 1E), whereas DC surface markers CD11c,HLA-DR, CD86, CD83, and CD80 were persistently expressed onSmyleDC (Fig. 1F). Analyses of cytokines secreted by mo-DCand SmyleDC revealed high stable production of the chemo-attractant proteins MCP-1 and IL-8 at nanogram per milliliterlevels (Fig. 1G). mo-DC maintained the presence of recombinantcytokines also secreted high levels of the DC2-type ILs IL-6 andIL-4 (data not shown). Several other cytokines were detectable forboth cultures at lower picogram per milliliter concentrations (IL-7,IL-10, IL-12, IL-13, IL-1b, IL-2, IL-5, and IFN-g), indicatinga mixed DC1- and DC2-type cytokine pattern. SmyleDC longevityin vivo after s.c. administration into NRG mice had been reportedin our previous publication (13). In this study, SmyleDC wereproduced from monocytes generated with G-CSF–mobilized bloodand lentivirally marked with the fLUC transgene for sequential

FIGURE 1. Generation and analyses of human con-

ventional mo-DC and SmyleDC from G-CSF–mobilized

donors. (A) The monocistronic integrase-defective LV

encoding for CMV-pp65 protein LV-CMV-pp65 alone was

used to generate mo-DC, whereas LV-CMV-pp65 plus the

bicistronic LV vector encoding huGM-CSF and huIFN-a

were used to cotransduce monocytes and generate

SmyleDC. (B) huGM-CSF and IFN-a production were

measured weekly in supernatants from SmyleDC kept in

culture up to 21 d. (C) Cell viability of mo-DC and

SmyleDC represented by the percentage of viable cells

recovered weekly for up to 21 d. Expression of CMV-pp65

(D), CD14 (E), and stability of DC differentiation (F) in

DC cultures (measured by expression of CD11c, HLA-DR,

CD86, CD80, and CD83) were assessed weekly by flow

cytometry. (G) Cell supernatants obtained weekly from

SmyleDC were analyzed for cytokine and chemokine se-

cretion by bead array. Data represent the average of at least

three independent experiments from different donors. *p,0.05. (H) SmyleDC/fLUC produced with additional co-

transduction with a lentiviral vector expressing the firefly

luciferase were injected s.c. 1 d after transduction into

NRG mice. Mice (n = 6) were administered i.p. with Lu-

ciferin, and optical imaging analyses were conducted for

localization of the bioluminescence signal emitted by vi-

able cells. Bioluminescence signals detected in the region

of interest were measured on days 14, 30, and 45 after

SmyleDC/fLUC administration and were plotted for each

mouse. Bars represent the average and error bars indicate

SEM.

4 HUMAN ADAPTIVE RESPONSES IN MICE

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

noninvasive optical imaging analyses. Mice (n = 6) injected s.c.with SmyleDC/fLUC received Luciferin administration for opticalimaging analyses, and the bioluminescence signals emitted byviable cells were quantified (Fig. 1H). Bioluminescence signalwas detectable for several weeks only at the injection site. In vivoviability was highest 14 d after injection, gradually decreasing inthe next 30 d and eventually vanishing afterwards.

Transplantation of human CD34+ cells into NRG, followed bySmyleDC immunization results in increased T cell expansion

We transferred adult human CD34+ HSC that were positivelyselected twice and highly pure (.99%) into sublethally irradiated4-wk-old NRG mice. Ten weeks after CD34+ cell transplantation,human hematopoietic reconstitution reached plateau and becamestable (2–5% of PBMC corresponding to human CD45+ cells).We did not observe statistically significant differences in the fre-quency human CD45+ in PBL in the different study arms prior toDC immunization (Supplemental Table I). Thus, the mice wereallocated to the different immunization groups at random afterHCT, receiving prime-boost immunizations of DC produced withmonocytes of the same HCT donor (5 3 105 cells) by s.c. injec-tions into the right hind flank (Supplemental Fig. 1). We comparedthe long-term (20 wk) hematopoietic and immune reconstitutionof nonimmunized versus mice immunized with mo-DC andSmyleDC produced with monocytes from the same CD34+ donor.Ten weeks after SmyleDC immunization, .10 pg/ml levels ofhuGM-CSF, IL-5, MCP-1, and IFN-g and lower levels of severalother human factors (IL-12, IL-1b, IL-6, IL-10, IL-8, IL-4, andIL-13) were detectable in plasma, indicating a persisting effect ofSmyleDC immunization (Fig. 2A). At this time point, levels ofhuman cytokines in plasma of mice immunized with mo-DC, andcontrols were dramatically lower. On the 20-wk endpoint, we didnot observe significant differences in the frequency of humanCD45+, B (CD19+), and T cells (CD3+) between the control andmo-DC groups (Fig. 2B). However, the frequency of humanCD45+ cells was significantly higher for SmyleDC-immunizedmice. Notably, analyzing the content of human CD45+ cells, therelative frequency of human B cells (defined as CD19+) wassignificantly decreased (to 30% in SmyleDC) upon SmyleDCvaccination, whereas the relative frequency of human T cells wassignificantly elevated (to 50%) (Fig. 2C, 2D, SupplementalTable I). In fact, even 1 wk after SmyleDC prime/boost immu-nization (week 13 after HCT), the rise in T cell expansion wasalready significant. At week 20 post-HCT, CD3+CD4+ Th cellsrepresented the most frequent T cell population (average 30%) ofmice immunized with SmyleDC and CD3+CD8+ CTLs wereclearly detectable (average 20% of human T cells in PBL) (Fig.2E, 2F, Supplemental Table I).

Regeneration of LN and lymphatic flow after SmyleDCimmunization

One of the most striking findings upon postmortem analysis camefrom the examination of the peripheral LN in mice immunizedwith SmyleDC: LN were clearly visible at the inguinal, axillary,and iliac regions (Fig. 3A). It has long been known that NRG micelacking the expression of the common cytokine receptor g-chaindisplay a defective lymphoid development and inactive LN fol-licles (22), and even after human HCT, the regeneration of pe-ripheral LN is not rescued and LN are mostly small or notidentifiable at necropsy. Mice immunized with SmyleDC showedconspicuous active axillary and inguinal LN in up to 90 and 70%of the cohorts, respectively. mo-DC immunization resulted in onlyunilateral and less frequent axillary (66%) and inguinal (33%) LN(Fig. 3B). Remarkably, there was a strong correlation between the

SmyleDC immunization at the right side and the formation of LNat the same side of the corresponding lymphatic draining axis. Toconfirm a functional lymphatic drainage from the lower trunk(inguinal LN) to the upper trunk (distal axillary draining LN) at20 wk after HCT, we injected 5% Evans blue s.c. in NRG micenear the SmyleDC immunization site. The ink stained the draininginguinal LN adjacent to the injection site, and the blue signalmigrated through the lymphatic vessels to the distal axillary LN(Fig. 3C). Immune-competent C57BL/6 mice with a normallymphatic system showed a similar ink drainage pattern, whereasnonimmunized mice or mice immunized with mo-DC showedimpaired drainage (Fig. 3C).In view of these results, we examined by optical imaging

analyses, whether SmyleDC were able to engraft, migrate, andmaintain viable in lymphoid tissues. Because transplanted NRG didnot display LN in the absence of SmyleDC immunization, HISmice (n = 6) were immunized on weeks 10 and 11 after HCT,as described above. On week 17, when the regeneration of LNwas under way, mice were subsequently administered s.c. withSmyleDC/fLUC on the right flank (colocalizing with the previousinjection sites) and on the contralateral left flanks (Fig. 3D).Sequential in vivo bioluminescence analyses performed weeklydemonstrated that most of the signal was emitted from the s.c.

FIGURE 2. SmyleDC immunization augments the detection of human

cytokines in plasma and expansion of human T cells. (A) Human cytokines

were detected in plasma from control, mo-DC, or SmyleDC-immunized HIS-

mice (week 20 after HCT). Bars represent average of cytokine concentration

(picograms per milliliter), and error bars represent SEM. (B) Analyses of

engraftment of human CD45+, hematopoietic cells, and subfractions relative

to expression of CD19+ (C), CD3+ (D), CD4+ (E), and CD8+ (F) cells in

peripheral blood of control, mo-DC, or SmyleDC-immunized HIS mice be-

fore (week 10), 2 wk after (week 13), and 8 wk after prime/boost immuni-

zation (week 20). Data represent the distribution of control (n = 10), mo-DC

(n = 7), and SmyleDC (n = 22) immunized mice. *p , 0.05.

The Journal of Immunology 5

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

injection sites (Fig. 3D). Remarkably, bioluminescence signalspreading from the injection site on the right flank toward the

anatomical site of the inguinal draining LN on the right side could

be detected, with increasing intensity during the following weeks.

This migratory signal was significantly lower for the left contra-

lateral inguinal LN site (Fig. 3D). This noninvasive direct imaging

approach following the biodistribution of the cells in vivo dem-

onstrated that SmyleDC/fLUC displayed an effective migration

toward more active LN structures. Notwithstanding, the biolumi-

nescence analyses also showed that SmyleDC/fLUC were able,

despite at lower rates, to migrate from the s.c. site to the less

developed left inguinal LN. In addition, optical imaging analy-

ses of several explanted LN demonstrated that SmyleDC/fLUC

engrafted and were highly viable in the right axillary LN (re-

ceiving lymph from the lower limbs drained by the thoracic duct).

No bioluminescence signal was detectable in mesenteric LN, less

likely to drain lymph from the lower limbs. Real-time PCR

analyses of the adjacent draining inguinal LN showed detectable

lentiviral copies in two out of four mice (Table I). Two mice

showed signals below the reliable sensitivity threshold of our PCR

assay (less than one copy per 100 ng DNA).

Regenerated LN contain human T and B cells at differentstages of differentiation

Immunohistological analyses of LN explanted from SmyleDC-immunized mice revealed a massive infiltration of lymphocytes,but only a few regions resembling the anatomy of germinal centersobserved in normally developed LN obtained from wild-typeC57BL/6 mice (Fig. 4A, Supplemental Fig. 2A). Immunofluo-rescence analyses showed a predominant repopulation with humanCD3+ T cells. Human CD11c+ DCs were detected in the cortex orcolocalizing with mouse LYVE-1. We also detected vessel struc-tures positive for mouse endothelial CD31 marker (most likelyhigh endothelial venules). Flow cytometry analyses of LN re-vealed human CD45+ cells (77%), CD3+ T lymphocytes (73%),and only 3.8% CD19+ B cells (Fig. 4B). Within human CD3+

cells, 56% were CD4+ and 42% were CD8+ (Fig. 4C). Remarkablyfor both T cell subsets, we observed ∼80% CD45RA–CD62L–

effector memory cells, 10–20% central memory cells, and ,5%naive T cells (Fig. 4D, 4E). To identify follicular T helper (Tfh)cells, LN explanted from SmyleDC-immunized mice were pooledand analyzed for CD4+CXCR5+hiPD-1+ICOS+ cells, which cor-responded to 4.2% of the CD3+ population (Fig. 4F). A human

FIGURE 3. Analyses of LN and

lymphatic flow in HIS-NRG pro-

moted after SmyleDC immunization

and SmyleDC/fLUC engraftment and

migration. (A) Macroscopic detection

of inguinal, axillary, and iliac LN 8

wk after SmyleDC immunization. (B)

Frequency of mice with regenerated

LN in control (n = 10), mo-DC (n = 7),

and SmyleDC-injected (n = 22)

mice on the right side (same side of

DC injections, IS, n) or left side

(contralateral to the DC injections

CL, N). (C) Representative macro-

scopic detection of Evans blue ink

staining the adjacent draining ingui-

nal and flowing through lymphatic

vessels toward the axillary LN com-

paring with C57BL/6 mice and HIS-

NRG mice (controls or immunized

with mo-DC or SmyleDC). (D) NRG

mice transplanted with human CD34+

cells and immunized with SmyleDC

(weeks 10 and 11) were subsequently

(on week 17) administered with

SmyleDC/fLUC coexpressing Lucif-

erase in the right (previous DC in-

jection site) and left (contralateral

side) flank. Bioluminescence signals

detected in the injection site and in

adjacent anatomical LN location

were evaluated every week for up to

3 wk. The graphs show biolumines-

cence quantification in the LN loca-

tion (top) and in the injection site

(bottom) in both right and left flanks.

Bars represent the average of radi-

ance for n = 6 mice and error bars

indicate SEM. *p , 0.05, comparing

injection site versus contralateral for

the given time point.

6 HUMAN ADAPTIVE RESPONSES IN MICE

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

tonsil showed a Tfh cell frequency of ∼8% (Fig. 4F, SupplementalFig. 2C). We also examined B cell subpopulations in pooled LNand compared it with a human tonsil (Supplemental Fig. 2D).CD24hiCD38hi transitional B cells corresponded to a minor pop-ulation of total CD19+ cells in humanized LN (0.4%) and tonsil(2.4%) (Fig. 4G). Naive B cells (IgD+CD24intCD38int) were lessfrequent in humanized LN than in tonsil (5.9 versus 43.9%).

Surprisingly, we found dramatically higher frequencies of CD27hi

CD38hi terminally differentiated plasmablasts in humanized LN ascompared with tonsil (49.1 versus 0.7%).

Increased absolute numbers of human mature T and B cells canbe detected in spleen of SmyleDC-immunized mice

Twenty weeks post-HCT, SmyleDC-immunized mice showed.100-fold increase in the absolute numbers of human CD3+

T cells (858,487) in the spleen in comparison with nonimmunizedmice (p = 0.0028) (Fig. 5A). Mice immunized with mo-DC (6,394cells/spleen, 132-fold less, p = 0.02 versus SmyleDC) and controlmice (4,459 cells/spleen, 192-fold less, p = 0.0007 versusSmyleDC) failed to support high levels of expansion/homing ofCD3+ cells to the spleen (Supplemental Table I). In contrast to thelower relative frequency previously assessed in peripheral blood,the absolute CD19+ B lymphocyte content in spleens was signif-icantly higher in SmyleDC-immunized mice (406,672 cells/spleen) than in mo-DC–immunized (82,065 cells/spleen, 5-foldlower, p = 0.37) and control mice (15,639 cells/spleen, 26-foldlower, p = 0.0034). Histologically, human T and B cells werefound interacting within clusters resembling follicles (Supplemental

Table I. Analyses of LN for detection of lentiviral copies in LNsamples by real time PCR

Mouse/SampleLV Copies

(in 100 ng of Genomic DNA) Readout

Mouse I 0.05 NegativeMouse II 2.36 PositiveMouse III 1.47 PositiveMouse IV 0.05 NegativePBMNC control 0.12SmyleDC (day 7) 7.9293T standard 2.93

PBMNC, peripheral blood mononuclear cell; SmyleDC, self-differentiated mye-loid-derived lentivirus-induced dendritic cell.

FIGURE 4. Characterization of

human cells present in HIS-NRG–

regenerated LN induced by SmyleDC

immunization. (A) H&E and fluores-

cence immunohistological analyses

(CD3, CD11c, mouse [m]LYVE-1,

and mCD31) of LN from C57BL/6-

positive control mice versus HIS-

NRG mice immunized with Smy-

leDC. Images were acquired with an

Axiocam fluorescence microscope

(Zeiss) at original magnification 310

and analyzed using Axiowert soft-

ware (Zeiss). Averaged frequency of

human CD45+, CD3+, and CD19+

cells (B) and CD4+ and CD8+ T cells

(C) in LN (n = 4) recovered from

SmyleDC-immunized mice. (D and

E) Characterization of CD45RA+/

CD62L+ naive, CD45RA–CD62L+

central memory, and CD45RA–CD62L–

effector memory subpopulations in

CD4+ and CD8+ LN T cells (n = 4).

(F) Frequency of Tfh cells (expressing

CD4+CXCR5+hiPD-1+ICOS+) in CD3

were analyzed in pooled LN (n = 8)

obtained from SmyleDC-injected HIS

mice. Human tonsil cells were used as

positive control. (G) Relative frequen-

cies of CD24hiCD38hi transitional,

IgD+CD24intCD38int mature, and

CD27hiCD38hi plasmablasts in CD19+

cells from pooled humanized LN and

human tonsil cells.

The Journal of Immunology 7

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

Fig. 2B). The composition of T lymphocyte subsets was further an-alyzed. Average numbers of Th and CTL cells in spleens of SmyleDCmice were 370,086 and 81,649 cells/spleen, respectively (Fig. 5B,5C). This was significantly higher than the numbers found after mo-DC immunization (13,976 and 7,937 cells/spleen, p = 0.0046 and 0.038 versus SmyleDC, respectively) or in nonimmunized controls(1,943 for CD4+, p = 0.0035 and 52 for CD8+, p = 0.021 versusSmyleDC). Both naive and effector Th and CTL were increased interms of absolute cell numbers (Fig. 5B, 5C). Follicular T cells, whichare rarely observed in spleens of HIS-NRG mice, were detectable at9425 Tfh cells/spleen on average after SmyleDC immunization(compared with only 12 cells/spleen in the control group [p =0.0023]), and no detectable Tfh population in mo-DC group (Fig. 5D).Analyses of the endogenously reconstituted DC compartment

in spleen of HIS mice was performed as absolute cell numbers.We observed significantly higher numbers of CD45+Lin–CD11c+

myeloid DC in SmyleDC-immunized mice compared with controlmice (797 versus 298, respectively, p = 0.04; Fig. 5E). Similarly,

we observed significantly higher endogenous reconstitution ofCD45+Lin–CD11c+ plasmacytoid DC in SmyleDC-immunizedmice as compared with control mice (1527 versus 76.7 respec-tively, p = 0.0001; Fig. 5F). Remarkably, mice immunized withmo-DC showed very low numbers or undetectable myeloid orplasmacytoid DCs in spleen.Finally, detailed analysis of B cell subpopulations in spleen revealed

no significant differences in numbers of transitional B cell (control6,732, mo-DC 29,028, and SmyleDC 77,454 cells/spleen) (Fig. 5G).However, SmyleDC immunization led to a significant marked ex-pansion of mature B cells (83,454 cells/spleen) and plasmablasts/plasma cells (91,522 cells/spleen) compared with other groups.

SmyleDC immunization induces anti-pp65 T cell immuneresponses

We showed previously that immunization of NRG mice withSmyleDC prior adoptive human PBL/T cell transfer enhanced Ag-specific T responses in vivo (13). In this study, we evaluated

FIGURE 5. Analyses of the absolute numbers of hu-

man cell populations in spleen of HIS-NRG mice. (A)

Scatter plots representing total cell numbers of human

CD45+, CD3+, and CD19+ cells per spleens from con-

trol, mo-DC, and SmyleDC-immunized mice on week

20 after HCT. Cell counts per spleen from total,

CD45RA+/CD62L+ naıve, and CD45RA–CD62L– ef-

fector memory subpopulations in CD4+ (B) and CD8+

(C) T cells. (D) Total cell numbers of CD3+CD4+

CXCR5+hiPD-1+ICOS+ Tfh cells. Scatter plots repre-

senting total cell numbers of human CD45+/Lin–/CD11+

myeloid DC (E) and human CD45+/Lin–/CD123+ plas-

mocytoid DC (F) in spleens from control, mo-DC, or

SmyleDC-injected HIS-NRG on week 20 after HCT. (G)

Total cell counts of CD19+CD24hiCD38hi transitional,

CD19+IgD+CD24intCD38int mature, and CD19+CD27hi

CD38hi plasmablasts per spleens from control, mo-DC,

and SmyleDC-immunized mice. Bars and error bars

represent means and SEM, respectively. *p , 0.05.

8 HUMAN ADAPTIVE RESPONSES IN MICE

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

whether SmyleDC could stimulate endogenously developed hu-man T cells post-HCT reactive against pp65. Human CD3+ cellswere FACS sorted from spleen or LN 20 wk after HCT. To obtainenough T cell numbers for conducting the immune assays, T cellsfrom spleens were expanded for 3 d in the presence of anti–CD2/CD3/CD28-conjugated beads and further cocultured for 7 d withSmyleDC plus cytokines (IL-2, IL-7, and IL-15) (SupplementalFig. 1C). T cells were seeded on IFN-g–coated plates overnight inthe absence of antigenic stimulation or with a pp65 overlappingpeptide pool (pp65pp) and analyzed by IFN-g ELISPOT. Spleno-cytes from SmyleDC-immunized mice showed higher frequency ofpp65-reactive T cells than mo-DC–immunized mice (33.6 spotsversus 15.5 spots on average for triplicates, p = 0.25) or controlmice (less than 1 spot, p , 0.05) (Fig. 6A). T cells isolated fromindividual LN from SmyleDC-immunized mice (n = 4) and cul-tured for 7 d with SmyleDC plus cytokines also reacted againstpp65 peptides (pp65: 53 spots versus no Ag: 18.7 spots, p = 0.021). PBMC from CMV-reactive donors were used as positivecontrol for the assay (Fig. 6B).

SmyleDC immunization induces Ig and pp65-specific humoralresponses

We observed significant increase in the frequency of IgG memoryB cells in spleens as well as generation of plasma IgM and IgG uponSmyleDC immunization in comparison with control groups, whereIg levels were close to limit of detection (Fig. 6C–E). Ig reactivityspecific against pp65 was assessed using plasma obtained fromCMV-seropositive systemic lupus erythematosus patients as pos-itive controls for an in-house developed ELISA system. Whereasthere was no detectable signal in plasma from mo-DC–immunizedor control mice, anti-pp65 IgG and IgM were found in plasma of 4of 22 and 11 of 22 mice, respectively (Fig. 6F, 6G). Remarkably,despite these functional Ag-specific human T and B cell responses,we did not detect any clinical signs of GvHD in these mice asevaluated by weight monitoring from weeks 6 to 20 after HCT(Fig. 7A). Histopathology analyses to examine occurrence ofGvHD was performed in transplanted control (n = 3) andSmyleDC-immunized (n = 4) mice 20 wk after HCT. For all ex-amined animals of the control group, no significant histologicalabnormalities were observed. Grade 1 GvHD could be shown inthree of four intestinal specimen of the SmyleDC immunizationgroup (n = 4) mainly showing single apoptotic figures at the baseof colonic crypts (Fig. 7B). Grade 1 GvHD could be also ob-served in skin specimen of two of four mice in the SmyleDCgroup revealing apoptotic figures preferentially at the infundib-ulum of hair follicles (Fig. 7B). Higher morphological grades ofGvHD were not found. A cohort of 10 mice immunized withSmyleDC maintained for 40 wk after HCT did not show clinicalsigns or any macroscopic evidence of late-onset GvHD (data notshown).

DiscussionWe have previously demonstrated several properties of murine (23,24) and human (15, 25) lentivirus–induced DC that can autono-mously differentiate in vivo. Among those, one particular char-acteristic that drastically differentiates them from conventionalDCs (cultured in the presence of recombinant cytokines) is theirextended longevity in vivo for several weeks after s.c. injection. Inaddition, murine lentivirus–induced DCs injected s.c. into syn-geneic immunocompetent C57BL/6 mice showed higher migra-tion to draining LN than conventional DCs (23) and persistedviable in LN for several weeks (24). In the current studies, theseobservations were confirmed and expanded to a xenograft hu-manized mouse system, with additional implications.

The combination of GM-CSF and IFN-a was previously shownby other groups as recombinant cytokines (26, 27) and also by usas transgenic cytokines (13) conferring superior activation of DCsthan the more traditional GM-CSF and IL-4 combination. GM-CSFand IFN-a are very complementary in the activation of downstreamsignaling cascade to allow cell viability, sequential differentiationand maturation of DCs, bypassing the need of additional maturationsteps. Lentivirus-induced SmyleDC generated from peripheral bloodobtained from G-CSF–mobilized donors showed, as expected, highviability in vitro and in vivo, concurrent with a typical activated DCimmunophenotype (high expression of HLA-DR, CD80, CD86, and

FIGURE 6. Human cellular and humoral responses against pp65

detected in HIS-NRG mice after SmyleDC immunization. (A) Human

CD3+ cells were sorted from splenocytes from control (n = 4), mo-DC and

(n = 4), SmyleDC-immunized (n = 6) HIS mice 8 wk after immunization.

Following ex vivo expansion for 10 d with SmyleDC, restimulation in the

presence (pp65pp) or absence (NoAg) of CMV-pp65 overlapping pooled

peptides was performed overnight on anti–IFN-g–coated plates. Bars

represent average of IFN-g–positive spots for 2 3 104 cells. (B) Similar

procedures were performed with effector cells recovered from LN obtained

from SmyleDC-immunized HIS-NRG (n = 4 mice), which were assayed in

parallel with peripheral blood mononuclear cells obtained from a CMV-

seropositive human donor. Data represent averaged IFN-g–positive spots

in 2 3 104 cells. (C) Frequency of switched B cells expressing CD27+

CD38+IgG+ within human CD19+ cells. Quantification of total Ig G (D)

and IgM (E) in plasma from control and DC-immunized mice 8 wk after

immunization. Quantification of pp65-specific IgG (F) and pp65-specific

IgM (G) in plasma from control, mo-DC, and SmyleDC-injected mice

8 wk after immunization. Plasma samples from systemic lupus eryth-

ematosus (SLE) patients were used as positive controls for both pp65-

specific IgG and IgM analyses. Bars and error bars represent means and

SEM respectively. *p , 0.05.

The Journal of Immunology 9

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

CD83) and cytokine production (persistent detectable levels of IL-12 and IFN-g). Noteworthy, long-lived human-derived SmyleDC/fLUC also were capable to engraft and traffic locally and systemi-cally through the xenogeneic mouse lymphatic system, producingfunctional effects. Therefore, this preclinical validation study aimedto demonstrate efficacy of SmyleDC in the human HCT settingusing the HIS-NRG model exceeded our expectations, because a fullrange of T and B cell terminal adaptive immune reconstitutioneffects, including the development of peripheral LN, were observed.Both mo-DC and SmyleDC could be produced with monocytesisolated from the G-CSF–mobilized HCT donors. SmyleDC showedhigher viability in vitro and persistent autocrine activation for ex-pression of relevant immunologic cell surface markers and cyto-kines than mo-DC. Subcutaneous administration of SmyleDC 10 wkpost-HCT and plasma analyses another 10 wk later revealed sub-stantially higher levels of several human cytokines (GM-CSF, IL-5,MCP-1, IFN-g, IL-13, TNF-a, IL-8, and IL-4) than in mo-DC–immunized or control mice. This was associated with significantlyhigher frequencies of human T cells, higher absolute numbers ofeffector memory T cells, and detection of terminally differentiatedplasma B cells in blood, spleen, and LN. Immune monitoring 20 wkafter HCT showed that these cellular effects were accompanied withpp65-specific T cell and Ab responses. Taken together, althoughimmunization with mo-DC expressing pp65 showed some degree ofimmune modulation in comparison with nonimmunized mice, theeffects of SmyleDC were clearly more profound. These differenceshighlight the requirement of cell therapies involving postmitotic andnonreplicating APC to persist long enough after administration toefficiently signal, produce cytokines and chemokines, migrate, at-tract, and interact with other cells of the immune system for robustAg presentation.

These new in vivo findings in NRG mice also can lead to noveladvances to improve the generation of humanized mice witha functional immune system. Although several transgenic andvaccination approaches to induce immune reconstitution in im-munodeficient mice have been evaluated over the past decade,suitable in vivo experimental models to address human hemato-poietic development to terminally and functionally differentiated Tand B cells were lacking. To experimentally recapitulate humanimmune reconstitution after HCT in vivo, Shultz, Ishikawa, andcolleagues (28, 29) pioneered the transplantation of CD34+ HSCinto different types of immunodeficient mouse strains lacking thecommon IL-2Rg-chain (IL-2Rg) (NOD/Rag1null/IL2Rgnull-NRG,NOD/LtSz-scid/IL2Rgnull-NSG, or NOD/SCID/IL2Rgnull-NOG)resulting in reconstitution of human hematopoietic lineages 8–10wk after CD34+ cell transfer. However, regardless of the source ofHSC and the method for cell transplantation, humanized micedisplayed suboptimal levels of lymphocyte reconstitution, andlacked or had low levels of Ag-specific cellular and humoralresponses and overall anergy (30, 31). Factors that may impactin the inefficient lymphatic development in HIS mice includethe absence of human histocompatibility molecules and a poorhumanized cytokine environment. To overcome this deficiency,approaches including delivery of recombinant cytokines (32, 33),transplantation of fetal lymphatic tissue along with HPCs (34, 35) orthe use of transgenic strains constitutively expressing the MHCclass I (36) and II (37) or critical hematopoietic cytokines (38)have been described recently. Despite some improvement, thesesingle strategies allowed a limited improvement in B and T cellresponses against human viral challenges. In contrast, the adoptivelytransferred donor-derived SmyleDC combines a long-lived andmigratory cell entity expressing persistently MHC I and II, severalrelevant costimulatory molecules, inflammatory and hematopoieiticcytokines.As consistent with previous reports regarding HCT transplan-

tation with G-CSF–mobilized adult HSC, we observed low relativefrequencies of T lymphocytes in nonimmunized control mice(i.e., ,20%) which reflected also the results obtained with mo-DCimmunization. Conversely, as also reported, the relative frequencyof circulating human B cells (in this study defined as CD19+ cells)was commonly .80% in control mice. SmyleDC immunizationresulted into a decrease of the relative CD19+ B cell frequency inPBL, but was associated with an increase of the absolute B cellcontent in spleens and LN. Notably, the higher repopulation rateof T and B cells in these tissues, was correlated with matured cellimmune phenotypes associated with immune activation.Importantly, few reports have directly addressed the presence of

reconstituted lymphatic structures in HSC-transplanted mice (39–41). Although human CD34+ HCT can be improved by usinghuman cord blood or fetal liver in HIS models, thereby reachinghigher rates of human cell engraftment (.60%) than in our studies(using G-CSF–mobilized adult blood), reported LN structureswere anecdotal or not fully characterized. In a recent extensivestudy of cord blood–based HCT transplantation into newbornBALB/c –Rag2null Il2rgnull mice reported by Lang et al. (42),a detailed temporal analyses demonstrated that the occurrence ofLN structures was a late event, macroscopically detectable ∼15wk after HCT, and then further developing for at least 25 wk afterHCT. Another important observation of these studies was thatT cell engraftment in LN seemed to precede B cell engraftmentand subsequent functional B cell maturation for production of Igs(42). Similar to our data, this study also showed that, althoughclusters of human T and B cells could be observed in LN, thetissue architecture was not as well organized as in normal humanLN. Therefore, although a correlation between the occurrence of

FIGURE 7. Analyses of GvHD after SmyleDC immunization. (A)

Monitoring of weight changes in a cohort of control (n = 3), mo-DC (n =

3), and SmyleDC (n = 10) immunized mice for 20 wk after HCT. Arrows

indicate time of SmyleDC immunization. (B) Representative microg-

raphies of H&E-stained skin and colon samples obtained from NRG mice

after SmyleDC immunization 20 wk after HCT. Arrows show the presence

of single apoptotic figures indicative of mild grade 1 GvHD.

10 HUMAN ADAPTIVE RESPONSES IN MICE

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

LN and functional B (and in our studies also T) responses wasobserved, further studies will be required to more preciselycharacterize the complex cellular and molecular interactions be-tween the engrafted human hematopoietic compartment with theremaining mouse hematopoietic and nonhematopoietic cells (suchas stroma and endothelial cells) leading to LN regeneration.Nevertheless, our current data show that adoptive therapy withactivated and long-lasting donor-derived DC after HCT acceler-ated the regeneration of peripheral LN and lymphatic flow toactivate, mobilize, and finally mature lymphocytes toward fullimmune function in humanized mice. Thus, the presence of LNstructures and lymphatic development in humanized mouse modelsmay be a “conditio sine qua non” for exploring these models forpredictive studies of adaptive immunity.Although it was not the focus of our current Ag-specific immune

potency studies, we predict that the higher absolute naive andmemory/effector T cell counts found in spleen reflect a bona fidehigher thymic or extrathymic T cell development in these hu-manized mice. Therefore, from the clinical perspective, adoptiveSmyleDC therapy extends our previous work exploring them for(re)activation of adoptively transferred T cells toward de novoT cell development and immune regeneration.Ultimately, this novel modality of “human endogenously re-

generated systemic LN” will allow in the future more detailedmechanistic in vivo studies of the development of the humanimmune system, antigenic presentation, T and B cell terminalactivation, and useful interpretation of preclinical testing of HCTprotocols, vaccines, and immunomodulatory molecules.Besides the effects in adaptive immunity, it would also be in-

teresting to elucidate the role of the adoptive SmyleDC transfer inthe development of the myeloid compartment, because severaleffects in native immunity also may underlie the observed T andB cell maturation effects. Although we were able to observe in-creased myeloid and plasmacytoid DC compartments in spleenafter SmyleDC immunization, a more detailed study analyzing thedevelopment of monocytes, macrophages, neutrophils, and baso-phils as previously described for newborn HIS mice (43) would beinformative regarding innate immunity development in differenttissues.Development of relevant in vivo syngeneic and humanized

mouse models to predict potency and risks of genetically modifiedcell therapies is an ongoing focus of our laboratory (15, 23–25, 44).For donor-derived SmyleDC to be used in the context of alloge-neic stem cell transplantation, one of the upmost importantimmunotoxicity risks is the development of acute or chronicGvHD. When reconstituting a fully functional human immunesystem in a mouse after xenogeneic transplantation of HSC,GvHD affecting mice tissues is commonly a concern. After all, theMHC class I and II molecules of the recipient mouse are notmatched to those of the human donor at all. For this reason, ani-mals were assessed for signs and symptoms of GvHD. Trans-planted mice immunized with SmyleDC and observed macro-scopically for 20 wk for clinical signs (n = 22) or weight gain (n =10) did not show GvHD. The histopathology analyses (n = 4)indicated that the reconstituted human immune system causedonly mild signs of GvHD in some of the mice, even though it wasin other respects functional. SmyleDC-immunized mice main-tained for 40 wk after HCT (n = 10) showed no clinical signs ofGvHD (data not shown). Although preclinical immunotoxicitystudies would require further detailed analyses with larger num-bers of mice, exploring the use of the SmyleDC for acceleratingthe reconstitution of a functional immune system in a recipientafter HCT seems to be safe. Given the fact that MHC moleculesbetween mice and humans differ more that the MHC molecules of

different humans, it would be tempting to speculate that SmyleDCmight have the potential to induce tolerance of the transplantedimmune system for the tissue of the recipient. Thus, the clinicallytranslatable use of the SmyleDC may hypothetically offer thepotential to facilitate allogeneic HSC transplantations using HLA-mismatched donors and recipients.Incidentally, production of SmyleDC with a single tricis-

tronic lentiviral vector and using methods compliant with goodmanufacturing practices is ongoing for translation into clinicaltrials to immunize transplanted patients receiving stem cell graftsfrom CMV-seronegative donors or cord blood (A. Daenthasanmak,G. Salguero, and R. Stripecke, manuscript in preparation). Inaddition, SmyleDC/pp65 could be also explored in the future as anautologous cellular immune therapeutic product against glioma andbreast cancer, as CMV has been recently implicated as an eti-ological agent for development of these types of cancer.

AcknowledgmentsWe thank all members of the Regenerative Immune Therapies Applied

group and in particular Alexandra Ingendoh for the technical contributions

to the completion of this work; Dr. Dirk Wedekind from the MHH Animal

Facility for regulatory and technical assistance for the use of NRGmice; and

Dr. Matthias Ballmaier (MHH Cell Sorting Core Facility) and Stephanie

Vahlsing (MHH Transfusion Medicine) for technical assistance. We spe-

cially thank Dr. Anke Breithaupt from the ambulatory service of bone

marrow transplantation and the staff of Cytonet, who coordinated the

procurement of HSC donors.

DisclosuresThe authors have no financial conflicts of interest.

References1. Gill, S., and D. L. Porter. 2013. Reduced-intensity hematopoietic stem cell

transplants for malignancies: harnessing the graft-versus-tumor effect. Annu.Rev. Med. 64: 101–117.

2. Filipovich, A. 2008. Hematopoietic cell transplantation for correction of primaryimmunodeficiencies. Bone Marrow Transplant. 42(Suppl. 1): S49–S52.

3. Oevermann, L., P. Lang, T. Feuchtinger, M. Schumm, H. M. Teltschik,P. Schlegel, and R. Handgretinger. 2012. Immune reconstitution and strategiesfor rebuilding the immune system after haploidentical stem cell transplantation.Ann. N.Y. Acad. Sci. 1266: 161–170.

4. Storb, R., B. Gyurkocza, B. E. Storer, M. L. Sorror, K. Blume, D. Niederwieser,T. R. Chauncey, M. A. Pulsipher, F. B. Petersen, F. Sahebi, et al. 2013. Graft-versus-host disease and graft-versus-tumor effects after allogeneic hematopoieticcell transplantation. J. Clin. Oncol. 31: 1530‑1538.

5. Mielcarek, M., B. Storer, P. J. Martin, S. J. Forman, R. S. Negrin, M. E. Flowers,Y. Inamoto, T. R. Chauncey, R. Storb, F. R. Appelbaum, and W. I. Bensinger.2012. Long-term outcomes after transplantation of HLA-identical related G-CSF‑mobilized peripheral blood mononuclear cells versus bone marrow. Blood119: 2675–2678.

6. Seggewiss, R., and H. Einsele. 2010. Immune reconstitution after allogeneictransplantation and expanding options for immunomodulation: an update. Blood115: 3861–3868.

7. Marie-Cardine, A., F. Divay, I. Dutot, A. Green, A. Perdrix, O. Boyer,N. Contentin, H. Tilly, F. Tron, J. P. Vannier, and S. Jacquot. 2008. TransitionalB cells in humans: characterization and insight from B lymphocyte reconstitu-tion after hematopoietic stem cell transplantation. Clin. Immunol. 127: 14–25.

8. Roncarolo, M. G., S. Gregori, B. Lucarelli, F. Ciceri, and R. Bacchetta. 2011.Clinical tolerance in allogeneic hematopoietic stem cell transplantation. Immu-nol. Rev. 241: 145–163.

9. Mori, T., and J. Kato. 2010. Cytomegalovirus infection/disease after hemato-poietic stem cell transplantation. Int. J. Hematol. 91: 588–595.

10. Steinman, R. M. 2012. Decisions about dendritic cells: past, present, and future.Annu. Rev. Immunol. 30: 1–22.

11. Wendland, M., S. Willenzon, J. Kocks, A. C. Davalos-Misslitz, S. I. Hammerschmidt,K. Schumann, E. Kremmer, M. Sixt, A. Hoffmeyer, O. Pabst, and R. Forster. 2011.Lymph node T cell homeostasis relies on steady state homing of dendritic cells.Immunity 35: 945–957.

12. Verdijk, P., E. H. Aarntzen, W. J. Lesterhuis, A. C. Boullart, E. Kok, M. M. vanRossum, S. Strijk, F. Eijckeler, J. J. Bonenkamp, J. F. Jacobs, et al. 2009. Limitedamounts of dendritic cells migrate into the T-cell area of lymph nodes but havehigh immune activating potential in melanoma patients. Clin. Cancer Res. 15:2531‑2540.

The Journal of Immunology 11

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from

13. Daenthanasanmak, A., G. Salguero, S. Borchers, C. Figueiredo, R. Jacobs,B. S. Sundarasetty, A. Schneider, A. Schambach, B. Eiz-Vesper, R. Blasczyk,et al. 2012. Integrase-defective lentiviral vectors encoding cytokines inducedifferentiation of human dendritic cells and stimulate multivalent immuneresponses in vitro and in vivo. Vaccine 30: 5118–5131.

14. Trautmann, L., M. Rimbert, K. Echasserieau, X. Saulquin, B. Neveu,J. Dechanet, V. Cerundolo, and M. Bonneville. 2005. Selection of T cell clonesexpressing high-affinity public TCRs within human cytomegalovirus-specificCD8 T cell responses. J. Immunol. 175: 6123–6132.

15. Salguero, G., B. S. Sundarasetty, S. Borchers, D. Wedekind, B. Eiz-Vesper,S. Velaga, A. C. Jirmo, G. Behrens, G. Warnecke, A. K. Knofel, et al. 2011.Preconditioning therapy with lentiviral vector-programmed dendritic cellsaccelerates the homeostatic expansion of antigen-reactive human T cells inNOD.Rag1-/-.IL-2rgc‑/‑ mice. Hum. Gene Ther. 22: 1209–1224.

16. Stripecke, R. 2009. Lentiviral vector-mediated genetic programming of mouseand human dendritic cells. Methods Mol. Biol. 506: 139–158.

17. Harrell, M. I., B. M. Iritani, and A. Ruddell. 2008. Lymph node mapping in themouse. J. Immunol. Methods 332: 170–174.

18. Maetzig, T., M. Galla, M. H. Brugman, R. Loew, C. Baum, and A. Schambach.2010. Mechanisms controlling titer and expression of bidirectional lentiviral andgammaretroviral vectors. Gene Ther. 17: 400–411.

19. Rothe, M., I. Rittelmeyer, M. Iken, U. R€udrich, A. Schambach, S. Glage,M. P. Manns, C. Baum, M. Bock, M. Ott, and U. Modlich. 2012. Epidermalgrowth factor improves lentivirus vector gene transfer into primary mousehepatocytes. Gene Ther. 19: 425–434.

20. Lerner, K. G., G. F. Kao, R. Storb, C. D. Buckner, R. A. Clift, and E. D. Thomas.1974. Histopathology of graft-vs.-host reaction (GvHR) in human recipients ofmarrow from HL-A-matched sibling donors. Transplant. Proc. 6: 367–371.

21. Becker, P. D., N. Legrand, C. M. van Geelen, M. Noerder, N. D. Huntington,A. Lim, E. Yasuda, S. A. Diehl, F. A. Scheeren, M. Ott, et al. 2010. Generation ofhuman antigen-specific monoclonal IgM antibodies using vaccinated “humanimmune system” mice. PLoS One 5: 5.

22. Cao, X., E. W. Shores, J. Hu-Li, M. R. Anver, B. L. Kelsall, S. M. Russell,J. Drago, M. Noguchi, A. Grinberg, E. T. Bloom, et al. 1995. Defective lymphoiddevelopment in mice lacking expression of the common cytokine receptorgamma chain. Immunity 2: 223–238.

23. Koya, R. C., T. Kimura, A. Ribas, N. Rozengurt, G. W. Lawson, E. Faure-Kumar,H. J. Wang, H. Herschman, N. Kasahara, and R. Stripecke. 2007. Lentiviralvector-mediated autonomous differentiation of mouse bone marrow cells intoimmunologically potent dendritic cell vaccines. Mol. Ther. 15: 971–980.

24. Pincha, M., G. Salguero, D. Wedekind, B. S. Sundarasetty, A. Lin, N. Kasahara,M. H. Brugman, A. C. Jirmo, U. Modlich, R. Gutzmer, et al. 2011. Lentiviralvectors for induction of self-differentiation and conditional ablation of dendriticcells. Gene Ther. 18: 750–764.

25. Sundarasetty, B. S., V. K. Singh, G. Salguero, R. Geffers, M. Rickmann,L. Macke, S. Borchers, C. Figueiredo, A. Schambach, U. Gullberg, et al. 2013.Lentivirus-induced dendritic cells for immunization against high-risk WT1+

acute myeloid leukemia. Hum. Gene Ther. 24: 220–237.26. Lapenta, C., S. M. Santini, M. Spada, S. Donati, F. Urbani, D. Accapezzato,

D. Franceschini, M. Andreotti, V. Barnaba, and F. Belardelli. 2006. IFN-a‑conditioned dendritic cells are highly efficient in inducing cross-primingCD8+ T cells against exogenous viral antigens. Eur. J. Immunol. 36: 2046–2060.

27. Cobb, A., L. K. Roberts, A. K. Palucka, H. Mead, M. Montes, R. Ranganathan,S. Burkeholder, J. P. Finholt, D. Blankenship, B. King, et al. 2011. Developmentof a HIV-1 lipopeptide antigen pulsed therapeutic dendritic cell vaccine. J.Immunol. Methods 365: 27–37.

28. Ishikawa, F., M. Yasukawa, B. Lyons, S. Yoshida, T. Miyamoto, G. Yoshimoto,T. Watanabe, K. Akashi, L. D. Shultz, and M. Harada. 2005. Development offunctional human blood and immune systems in NOD/SCID/IL2 receptor gchain(null) mice. Blood 106: 1565–1573.

29. Shultz, L. D., M. A. Brehm, J. V. Garcia-Martinez, and D. L. Greiner. 2012.Humanized mice for immune system investigation: progress, promise andchallenges. Nat. Rev. Immunol. 12: 786–798.

30. Lepus, C. M., T. F. Gibson, S. A. Gerber, I. Kawikova, M. Szczepanik,J. Hossain, V. Ablamunits, N. Kirkiles-Smith, K. C. Herold, R. O. Donis, et al.2009. Comparison of human fetal liver, umbilical cord blood, and adult bloodhematopoietic stem cell engraftment in NOD-scid/gc‑/‑, Balb/c-Rag1‑/‑gc‑/‑, andC.B-17-scid/bg immunodeficient mice. Hum. Immunol. 70: 790–802.

31. Andre, M. C., A. Erbacher, C. Gille, V. Schmauke, B. Goecke, A. Hohberger,P. Mang, A. Wilhelm, I. Mueller, W. Herr, et al. 2010. Long-term human CD34+

stem cell-engrafted nonobese diabetic/SCID/IL-2R gnull mice show impairedCD8+ T cell maintenance and a functional arrest of immature NK cells. J.Immunol. 185: 2710–2720.

32. O’Connell, R. M., A. B. Balazs, D. S. Rao, C. Kivork, L. Yang, andD. Baltimore. 2010. Lentiviral vector delivery of human interleukin-7 (hIL-7) tohuman immune system (HIS) mice expands T lymphocyte populations. PLoSOne 5: e12009.

33. Chen, Q., M. Khoury, and J. Chen. 2009. Expression of human cytokines dra-matically improves reconstitution of specific human-blood lineage cells in hu-manized mice. Proc. Natl. Acad. Sci. USA 106: 21783–21788.

34. Hu, Z., and Y. G. Yang. 2012. Human lymphohematopoietic reconstitution andimmune function in immunodeficient mice receiving cotransplantation of humanthymic tissue and CD34+ cells. Cell. Mol. Immunol. 9: 232–236.

35. Biswas, S., H. Chang, P. T. N. Sarkis, E. Fikrig, Q. Zhu, and W. A. Marasco.2011. Humoral immune responses in humanized BLT mice immunized withWest Nile virus and HIV-1 envelope proteins are largely mediated via humanCD5+ B cells. Immunology 134: 419–433.

36. Shultz, L. D., Y. Saito, Y. Najima, S. Tanaka, T. Ochi, M. Tomizawa, T. Doi,A. Sone, N. Suzuki, H. Fujiwara, et al. 2010. Generation of functional human T-cellsubsets with HLA-restricted immune responses in HLA class I expressing NOD/SCID/IL2r gnull humanized mice. Proc. Natl. Acad. Sci. USA 107: 13022–13027.

37. Danner, R., S. N. Chaudhari, J. Rosenberger, J. Surls, T. L. Richie,T. D. Brumeanu, and S. Casares. 2011. Expression of HLA class II molecules inhumanized NOD.Rag1KO.IL2RgcKO mice is critical for development andfunction of human T and B cells. PLoS One 6: e19826.

38. Willinger, T., A. Rongvaux, H. Takizawa, G. D. Yancopoulos, D. M. Valenzuela,A. J. Murphy, W. Auerbach, E. E. Eynon, S. Stevens, M. G. Manz, andR. A. Flavell. 2011. Human IL-3/GM-CSF knock-in mice support human al-veolar macrophage development and human immune responses in the lung. Proc.Natl. Acad. Sci. USA 108: 2390–2395.

39. Singh, M., P. Singh, G. Gaudray, L. Musumeci, C. Thielen, D. Vaira,C. Vandergeeten, L. Delacroix, E. Van Gulck, G. Vanham, et al. 2012. An im-proved protocol for efficient engraftment in NOD/LTSZ-SCIDIL-2RgNULLmice allows HIV replication and development of anti-HIV Immune Responses.PLoS One 7: e38491.

40. Marodon, G., D. Desjardins, L. Mercey, C. Baillou, P. Parent, M. Manuel,C. Caux, B. Bellier, N. Pasqual, and D. Klatzmann. 2009. High diversity of theimmune repertoire in humanized NOD.SCID.gc‑/‑ mice. Eur. J. Immunol. 39:2136–2145.

41. Sun, Z., P. W. Denton, J. D. Estes, F. A. Othieno, B. L. Wei, A. K. Wege,M. W. Melkus, A. Padgett-Thomas, M. Zupancic, A. T. Haase, and J. V. Garcia.2007. Intrarectal transmission, systemic infection, and CD4+ T cell depletion inhumanized mice infected with HIV-1. J. Exp. Med. 204: 705–714.

42. Lang, J., M. Kelly, B. M. Freed, M. D. McCarter, R. M. Kedl, R. M. Torres, andR. Pelanda. 2013. Studies of lymphocyte reconstitution in a humanized mousemodel reveal a requirement of T cells for human B cell maturation. J. Immunol.190: 2090–2101.

43. Tanaka, S., Y. Saito, J. Kunisawa, Y. Kurashima, T. Wake, N. Suzuki,L. D. Shultz, H. Kiyono, and F. Ishikawa. 2012. Development of mature andfunctional human myeloid subsets in hematopoietic stem cell-engrafted NOD/SCID/IL2rgKO mice. J. Immunol. 188: 6145–6155.

44. Pincha, M., B. S. Sundarasetty, G. Salguero, R. Gutzmer, H. Garritsen, L. Macke,A. Schneider, D. Lenz, C. Figueiredo, R. Blasczyk, et al. 2012. Identity, potency,in vivo viability, and scaling up production of lentiviral vector-induced dendriticcells for melanoma immunotherapy. Hum. Gene Ther. Methods 23: 38‑55.

12 HUMAN ADAPTIVE RESPONSES IN MICE

by guest on February 13, 2018http://w

ww

.jimm

unol.org/D

ownloaded from