GastroenterologyCopy of e-mail Notification zg04366

RUSH: Gastroenterology article <# 54080> for proofing by Protzer=====Dear Author,

The proof of your article, to be published by Elsevier in Gastroenterology, is available as a "PDF" file at the following URL:http://rapidproof.cadmus.com/RapidProof/retrieval/index.jsp

Login: your e-mail passwordPassword: ----

The site contains 1 file. You will need to have Adobe Acrobat Reader software to read these files. This is free software and is available for user download at: http://www.adobe.com/products/acrobat/readstep.html

Within 48 hours, please return the following to the address given below:

1) Corrected PDF set of page proofs2) If you are making extensive corrections to figures, please send as electronic files (JPG preferred). If you are making minor corrections to figures, please indicate corrrections in your reply. If your article contains color illustrations and you would like to receive proofs of these illustrations, please contact us within 48 hours.

After accessing the PDF file, please:

1) Carefully proofread the entire article, including any tables, equations, figure legends and references.2) Ensure that your affiliations and address are correct and complete.3) Check that any Greek letter, especially "mu", has translated correctly;4) Verify all scientific notations, drug dosages, and names and locations of manufacturers;5) Be sure permission has been procured for any reprinted material.6) Answer all author queries completely. They are listed on the last page of the proof;

You may choose to list the corrections (including the replies to any queries) in an e-mail and return to me using the "reply" button. Using this option, please refer to the line numbers on the proof. If, for any reason, this is not possible, mark the corrections and any other comments (including replies to questions) on a printout of the PDF file and fax this to Lisa Traynor (215-239-3388) or mail to the address given below.

Do not attempt to edit the PDF file (including adding <post-it> type notes).

If you have any problems or questions, please contact me. PLEASE ALWAYS INCLUDE YOUR ARTICLE NUMBER (located in the subject line of this e-mail) WITH ALL CORRESPONDENCE.

Sincerely,

Lisa TraynorJournal ManagerElsevier Inc.1600 John F Kennedy BlvdSuite 1800Philadelphia, PA 19103-2899Phone: 215-239-3387, Fax: 215-239-3388gastroproduction@elsevier.com

TI

FA

*IV

BtrconsitdtrcdHhcftcsdphaWmcctacTpdu

Thcl

GASTROENTEROLOGY 2008;xx:xxx

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657

AQ: 1

AQ: 2

123456789

1011

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

cells Redirected Against Hepatitis B Virus Surface Proteins Eliminatenfected Hepatocytes

ELIX BOHNE,*,‡ MARKUS CHMIELEWSKI,‡,§ GREGOR EBERT,*,‡ KATJA WIEGMANN,� TIMO KÜRSCHNER,¶

NDREAS SCHULZE,# STEPHAN URBAN,# MARTIN KRÖNKE,‡,� HINRICH ABKEN,‡,§ and ULRIKE PROTZER,*,**

Molecular Infectiology, ‡Center for Molecular Medicine Cologne (CMMC), §Tumorgenetics—Department of Internal Medicine I, �Institute for Medical Microbiology,mmunology, and Hygiene, University Hospital Cologne, Koeln; ¶Center of Molecular Biology Heidelberg (ZMBH), University of Heidelberg, Heidelberg; #Molecular

Firology, University Hospital Heidelberg, Heidelberg; and **Institute for Virology, Technical University of Munich/GSF Munich, München, Germany

rccpaiccfvpsDsbhfmf

twihtpp(tHp

achtMPrc

BA

SIC–L

IVER

,PA

NCREA

S,A

ND

BIL

IARY

TRA

CT

12131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657

AQ: 4

UNCORREC

ackground & Aims: The final goal in hepatitis Bherapy is eradication of the hepatitis B virus (HBV)eplication template, the so-called covalently closedircular DNA (cccDNA). Current antiviral treatmentf chronic hepatitis B depends on interferon � orucleoside analogues inhibiting the viral reverse tran-criptase. Despite treatment, cccDNA mostly persistsn the host cell nucleus, continues to produce hepa-itis B surface antigen (HBsAg), and causes relapsingisease. We therefore aimed at eliminating persis-

ently infected hepatocytes carrying HBV cccDNA byedirecting cytolytic T cells toward HBsAg-producingells. Methods: We designed chimeric T-cell receptorsirected against HBV surface proteins present onBV-infected cells and used them to graft primaryuman T cells with antibody-like specificity. The re-eptors were composed of a single chain antibodyragment directed against HBV S or L protein fusedo intracellular signalling domains of CD3� and theostimulatory CD28 molecule. Results: Our resultshow that these chimeric receptors, when retrovirallyelivered and expressed on the cell surface, enablerimary human T cells to recognize HBsAg-positiveepatocytes, release interferon � and interleukin 2,nd, most importantly, lyse HBV replicating cells.

hen coincubated with HBV-infected primary hu-an hepatocytes, these engineered, antigen-specific T

ells selectively eliminated HBV-infected and thusccDNA-positive target cells. Conclusions: Elimina-ion of HBV cccDNA-positive hepatocytes followingntiviral therapy is a major therapeutic goal inhronic hepatitis B, and adoptive transfer of grafted

cells provides a promising novel therapeutic ap-roach. However, T-cell therapy may also cause liveramage and therefore needs further preclinical eval-ation.

he human hepatitis B virus (HBV) is a small, envel-oped, and noncytopathic virus, with a very narrow

ost range and strong liver tropism causing acute andhronic liver disease. Worldwide, approximately 350 mil-

ion patients are infected with HBV. A cytotoxic T-cell

TED P

ROOesponse is thought to be responsible for both viral

learance and liver injury during HBV infection.1 A poly-lonal and multispecific T-cell response was observed inatients who had cleared acute infection,2 whereas a weaknd oligoclonal response was described in chronicallynfected individuals.3 The ongoing inflammation inhronic infections results in cirrhosis and hepatocellulararcinoma. Current therapeutic options include inter-eron (IFN) � and nucleos(t)ide analogues inhibiting theiral reverse transcriptase. Nucleos(t)ide analogues im-air the viral life cycle but do not target the viral tran-cription template, the HBV covalently closed circularNA (cccDNA), in the host cell nucleus. From the epi-

omal cccDNA template, a new HBV replication cycle cane initiated after the end of therapy, often resulting in aepatitis flare. Therefore, elimination of persistently in-

ected hepatocytes is necessary to cure the disease,4 whichay either be achieved by induction or by adoptive trans-

er of T cells, which eliminate infected hepatocytes.Here, we aimed at developing genetically modified cy-

olytic T cells carrying a chimeric T-cell receptor (cTCR),hich targets HBV surface proteins present on HBV-

nfected cells. HBV-infected cells continuously produceepatitis B surface antigen (HBsAg) from the cccDNAemplate even when HBV replication subsides.5 HBsAg isredominantly composed of the HBV small surface (S)rotein with trace amounts of middle and large surface

L) proteins. If patients did not seroconvert from HBsAgo antibodies against hepatitis B surface antigen (anti-Bs), a high number of hepatocytes (5%–30%) remainedositive for HBV S protein even after long-term, highly

Abbreviations used in this paper: anti-HBs, antibodies against hep-titis B surface antigen; cccDNA, covalently closed circular DNA; cTCR,himeric T-cell receptor; GALV, Gibbon ape leukemia virus; HBeAg,epatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepa-itis B virus; IFN, interferon; IL, interleukin; L, HBV large surface protein;HC, major histocompatibility complex; NF-�B, nuclear factor �B;BL, peripheral blood lymphocytes; PHH, primary human hepatocytes;cDNA, relaxed circular DNA; S, HBV small surface protein; scFv, singlehain antibody fragment.

© 2008 by the AGA Institute0016-5085/08/$32.00

doi:10.1053/j.gastro.2007.11.002

pihpa

cnctvf

dhrtaet

fiostltmCtt

fppadsl

Cst(s�pf

scp

ftGtmUD(bc

fsceDcDn(Dtat(FSwL

fchuqpsm(vp

rsbarewbdm

BA

SIC–LIV

ER,

PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

2 BOHNE ET AL. GASTROENTEROLOGY Vol. xx, No. x

585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113

AQ: 5

585960616263646566676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113

F1

AQ: 6

AQ: 7

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

UNCORREC

otent antiviral therapy.6 We therefore think that target-ng S antigen-positive cells and thus residual infectedepatocytes harboring HBV cccDNA is a promising ap-roach to ultimately eliminate HBV infection followingntiviral therapy.

Adoptive T-cell therapy proved safe to reconstituteellular immunity against cytomegalovirus after alloge-eic bone marrow transplantation and effective in theontrol of Epstein-Barr virus-associated posttransplanta-ion lymphoproliferative disorders.7,8 These findings pro-ide the rationale to develop adoptive T-cell approachesor malignancies but also for chronic infectious diseases.

Repopulation with receptor modified T cells, trans-uced to express MART-1-specific TCR, resulted inoming of T cells into the liver and induced profoundegression of advanced melanoma.9 The next genera-ion of adoptive T-cell therapies will likely rely on thebility to endow “fit” cells with elevated cell-surfacexpression of high-affinity, specific TCRs by gene-ransfer technology.10

Artificial receptor constructs with antibody-like speci-cities and a compound signalling domain11 allow rec-gnition of native, nonprocessed antigen on the cellurface. Thereby, they work independent of classical an-igen processing and allow targeting cells with estab-ished immune escape mechanisms such as down-regula-ion of major histocompatibility complex (MHC)

olecules or reduced endolysosomal antigen processing.himeric receptors were used to specifically target T cells

o cancer or human immunodeficiency virus antigens onhe cell surface.12,13

Here, we use cTCRs composed of single chain antibodyragment (scFv) antibodies against S and L HBV surfacerotein, respectively, fused to transmembrane and cyto-lasmic domains of the costimulatory CD28 moleculend to the CD3� signalling domain. We supposed theesigned cTCRs to recognize HBV S or L protein on theurface of HBV-infected hepatocytes and to licence cyto-ytic T cells to eliminate these cells.

Materials and MethodsCell Culture Conditions and HBV InfectionHuman embryonic kidney cells HEK 293T (ATCC

RL-1573), HepG2 hepatoma cells (ATCC HB-8065), andtably HBV transfected HepG2.2.15 cells14 were main-ained in Dulbecco’s modified Eagle mediumDMEM)/5% glucose (wt/vol), 10% (vol/vol) fetal bovineerum (FBS), penicillin (100 U/mL), streptomycin (100g/mL), nonessential amino acids (0.1 mmol/L), sodiumyruvate (1 mmol/L), and L-glutamine (2 mmol/L) (allrom Biochrom AG, Berlin, Germany).

Primary human hepatocytes (PHH) were isolated by atandard 2-step collagenase perfusion and differentialentrifugation from liver tissue of patients undergoing

artial hepatectomy for metastasis resection after in- d

TED P

ROO

F

ormed consent as approved by the local Ethics Commit-ee. PHH medium was Williams E Medium (InvitrogenmbH, Karlsruhe, Germany) supplemented with L-glu-

amine (5 mmol/L), glucose (0.06% [wt/vol]), HEPES (23mol/L, pH7.4), gentamycin (50 �g/mL), penicillin (50I/mL), streptomycin (50 �g/mL), inosine (37 �mol/L),MSO (1.75%), hydrocortisone (4.8 �g/mL), and insulin

1 �g/mL). Cells were infected with HBV as describedefore.15 For cocultures with redirected T cells, we usedortison-free medium.

In cell culture medium of PHH, alanine aminotrans-erase (ALT) levels were measured using the Reflotronystem (Roche Diagnostics, Mannheim, Germany). Liveells were stained with calceine and dead cells withthidium homodimer-1 (Molecular Probes, Eugene, OR).NA from cell lysates was prepared by standard phenol-

hloroform extraction. HBV cccDNA and total HBVNA was quantified by Light Cycler PCR (Roche Diag-ostic) as described.16 Hepatitis B e (HBeAg) and surface

HBsAg) antigens were determined by ELISA (Abbott,artford, UK). HBV core protein and albumin were de-

ected by Western blot using monoclonal antibodygainst human albumin (DAKO, Carpinteria, CA) or an-iserum raised against truncated HBV core proteinkindly provided by Michael Nassal, University ofreiburg, Germany). Chemiluminescence signals (Super-ignal West Dura, Pierce Biotechnology, Rockford, NJ)ere quantified using the Gel Doc 2000 System (Bio-Radaboratories, München, Germany).

Construction and Transduction of ChimericT-Cell ReceptorsscFv C8 was selected from a scFv library generated

rom peripheral blood lymphocytes (PBL) of HBsAg-vac-inated individuals, and scFv 5a19 was derived from aybridoma expressing the monoclonal antibody 5a1917

sing phage display screening. DNA encoding scFv se-uences were cloned 3= of cytomegalovirus (CMV) I/Eromotor and a �-leader sequence into a cTCR con-truct18 and inserted into the retroviral expression plas-

id pBullet19 (Figure 1A). Gibbon ape leukaemia virusGALV) envelope pseudotyped, amphotropic retroviralectors were produced by cotransfecting the packaginglasmids pHIT60 and pCOLT.Primary human PBL were freshly prepared from pe-

ipheral blood using Ficoll-Hypaque (Biochrom AG) den-tity gradient centrifugation. Non-T cells were depletedy plastic adherence, and T cells were stimulated withnti-CD3 antibody OKT3 and interleukin (IL)-2 beforeetroviral transduction as described previously.18,19 Tovaluate transduction efficiencies, T cells were stainedith PE-labelled murine anti-CD3, -CD4, or CD8 anti-odies (DakoCytomation, Glostrup, Denmark). Trans-uced cells were identified with FITC-labelled anti-hu-an IgG antibodies recognizing the extracellular Fc-

erived spacer of the cTCR, counterstained with

pFnlH

acwoecoagXaw(ci

iBv(t(i

ttCwcsntsHbts

cvCdF

FSip((atmflauwSauIdcp

BA

SIC–L

IVER

,PA

NCREA

S,A

ND

BIL

IARY

TRA

CT

Month 2008 REDIRECTED T CELLS AND HEPATOCYTES 3

114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169

AQ: 8

AQ: 9

AQ: 18

114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

TED P

ROO

F

ropidium iodine, and analysed using a FACS Cantolow Cytometer (BD Bioscience, Franklin Lakes). Alter-atively, cells were counterstained with DAPI and ana-

yzed using a fluorescent microscope (IX 81; Olympus,amburg, Germany).

T-Cell Activation and Cytotoxicity AssayTarget cells were plated with 5 � 104 cells/well on

collagen type I coated 96-well plate and cultured untilonfluency was reached. Redirected T cells (effector cells)ere added in dilutions to the target cells after removalf IL-2. Grafted T cells were only used if transductionfficacy was �25%. Dependent on the transduction effi-iency, T cells were diluted with unmodified PBL tobtain effector to target ratios (E:T ratio) between 0.02nd 2:1 as indicated. Specific cytotoxicity of receptorrafted T cells against target cells was monitored by aTT based colorimetric assay (Roche Diagnostics GmbH)s described in detail.18 To assay proliferation, T cellsere stained with CarboxyFluorescein Succinimidyl Ester

CFSE; Molecular Probes) before coculture with targetells and analyzed after 48 hours by flow cytometry gat-ng on receptor grafted T cells.

T-cell subpopulations were obtained by magnetic sort-ng using MACS CD4 and CD8 Microbeads (Miltenyi,ergisch Gladbach, Germany). Degranulation of acti-ated T cells was stained with LAMP-2 specific antibodieseBioscience, San Diego, CA). IFN-�, tumor necrosis fac-or (TNF) �, and IL-2 were detected by Cytoset-ELISABiosource, Camarillo). IFN-� secretion was analyzed us-ng MACS IFN-� secretion assay (Miltenyi).

ResultsGrafting of Primary Human T Cells WithChimeric TCRsFigure 1A schematically depicts the retroviral vec-

or construct containing the cTCR constructs used inhis study. The established cTCR BW431/26scFv-Fc-D28-�,18 recognizing carcinoembryonic antigen (CEA)as designated �CEA cTCR and used as a specificity

ontrol because CEA is not expressed on hepatocytes.cFv BW431/26 was replaced by HBV-specific scFv recog-izing HBV S (scFv C8) or L protein (scFv 5a19), respec-ively. scFv C8 binds to a conformational epitope pre-umably in the “a” determinant of S protein; it recognizesBV genotypes A and B better than D, and subtype adw

etter than ayw (data not shown). Antibody 5a19 bindso amino acid (aa) 37 to 43 in the preS1 region of HBVubtype ayw better than adw.

Primary human T cells were engineered to express theTCR by retroviral transduction with amphotropic retro-iruses pseudotyped with a GALV envelope. Up to 60% ofD3� cells expressed cTCRs following retroviral trans-uction as determined by flow cytometry (Figure 1B).

UNCORREC

igure 1. Grafting of primary human T cells with chimeric TCR. (A)chematic representation of the chimeric TCR (cTCR) constructs

nserted into the retroviral vector plasmid pBullet; long terminal re-eat (LTR), splice donor (SD), splice acceptor (SA), packaging signal

�) (20). The cTCR is composed of an N-terminal leader sequenceL�), heavy (VH) and light (VL) chain variable regions of the single chainntibody fragment (scFv), the Fc spacer domain (Fc) of human IgG1,ransmembrane and intracellular regions of the CD28 signalling do-ain (CD28), and the CD3� signalling domain (CD3�). (B) Immuno-

uorescence microscopy of cells stained with the FITC-conjugatednti-human IgG antibody and DAPI (left: �S-C8 transduced; right:nmodified T cell). (C) Lymphocytes obtained from a healthy donorere retrovirally grafted with cTCR containing scFv recognizing HBVprotein (�S-C8), HBV L protein (�L-5a19), or carcinoembryonic

ntigen (�CEA) or used as control (PBL). Flow cytometric analysissing a PE-conjugated anti-CD3 and a FITC-conjugated anti-human

g-Fc antibody, which detects the extracellular IgG1 CH2CH3 spaceromain of the receptors, were performed to identify cTCR grafted Tells. Data are presented as dot blots, and percentages of doubleositive cells are given.

luorescence staining using anti-human Fc antibodies

vc

tawcp�HsIt5AciItws

cHpcwbou

tocicsIg(

ttstbl5aUol

ca(dTtc�

FH�

BA

SIC–LIV

ER,

PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

4 BOHNE ET AL. GASTROENTEROLOGY Vol. xx, No. x

170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225

AQ: 10

AQ: 11

F2

170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225

F3

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

UNCORREC

isualized surface distribution of the cTCRs on graftedells (Figure 1C).

To monitor the capability of the cTCRs to recognizeheir target antigens and to induce T-cell activation uponntigen recognition, redirected T cells were challengedith medium containing subviral and viral HBV parti-

les, which carry S and to a low extend middle and Lroteins on their surface. Both HBV-specific cTCRs—S-C8 and �L-5a19 —recognized HBV surface proteins onBV particles leading to T-cell activation (as demon-

trated by nuclear factor �B [NF-�B]) and release ofFN-� (see Supplementary Figure 1 online at www.gas-rojournal.org) and TNF-�. cTCR �S-C8, but not �L-a19, grafted T cells secreted IL-2 (data not shown).lthough we observed a low-level NF-�B activation in Tells grafted with the CEA-specific cTCR and in unmod-fied PBL, this activation did not lead to secretion ofFN-�, TNF-�, or IL-2. From these data, we concludedhat primary human T cells were successfully graftedith our cTCR constructs that recognize native HBV

urface proteins resulting in T-cell activation.

Redirected T Cells Are Activated by and LyseHBV-Producing Cell LinesTo test their cytotoxic capabilities, redirected T

ells were cocultured with HBV-replicatingepG2.2.15 hepatoma cells or as a control with HBV�arental HepG2 cells. Microscopic evaluation revealedytotoxic lesions in monolayers of HBV� target cellshen T cells were grafted with �S-C8 or �L-5a19 cTCRut not when T cells were grafted with the �CEA cTCRr when unmodified T cells or HBV� target cells were

igure 2. Cytotoxic activity of engineered T cells redirected against HBepG2.2.15 and parental HepG2 cells (HBV�, lower panel), coculturL-5a19, or �CEA, respectively, or unmodified control cells (PBL).

sed (Figure 2). T

TED P

ROO

F

Next, we performed a time course experiment cocul-uring grafted and control T cells with HepG2.2.15 cellsr with HepG2 cells at E:T ratio 4:1. �S-C8 cTCR graftedells started to release IFN-� after 24 hours, accumulat-ng to 22.4 ng/mL at 72 hours. �L-5a19 cTCR grafted Tells showed a delayed response, with IFN-� secretiontarting after 36 hours and reaching levels of 9.9 ng/mL.FN-� secretion was neither observed from �CEA cTCRrafted nor unmodified T cells nor on HBV� target cellsdata not shown).

To quantify cytotoxicity, redirected T cells were cocul-ured at different E:T ratios using constant numbers ofarget cells for 72 hours (Figure 3A–C). To reflect aituation, which is likely to be achieved by adoptiveransfer of grafted T cells in vivo, we used E:T ratiosetween 0.2:10 and 1.5:10. �S-C8 cTCR grafted T cells

ysed up to 70% and �L-5a19 cTCR grafted T cells up to8% of HBV� target cells, whereas only minimal killingctivity was observed on HBV� target cells (Figure 3A).sing either �CEA cTCR grafted or unmodified T cellsn HBV� cells, background levels of up to 20% target cell

ysis were observed.IFN-� was released in a dose-dependent fashion upon

ontact with HBV� target cells by �S-C8 cTCR graftednd to a lower extend by �L-5a19 cTCR grafted T cellsFigure 3B). �S-C8 cTCR grafted T cells secreted IL-2 in aose-dependent fashion, whereas �L-5a19 cTCR graftedcells secreted only low levels of IL-2 (Figure 3C). Nei-

her �CEA grafted nor unmodified T cells nor T cells inontact with HBV� target cells released IFN-� or IL-2.S-C8 cTCR and to a lower extent �L-5a19 cTCR grafted

ace proteins. Light microscopy of HBV producing (HBV�, upper panel)r 72 hours with redirected T cells (E:T � 2:1) carrying cTCR �S-C8,

V surfed fo

cells proliferated upon antigen recognition on HBV�

cd

tctmUgcccLaccalT

eHIr

rpinwcrtw

FL((mg 0) ou

BA

SIC–L

IVER

,PA

NCREA

S,A

ND

BIL

IARY

TRA

CT

Month 2008 REDIRECTED T CELLS AND HEPATOCYTES 5

226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281

F4

AQ: 19

226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

UNCORREC

ells as determined by flow cytometric analysis of CFSEilution (Figure 3D).Next, we determined which T-cell subset contributed

o IFN-� secretion and cytotoxicity (Figure 4). By flowytometry, we identified IFN-� secreting cells and cyto-oxic activity by cell surface staining of degranulation

arker lysosomal membrane protein-2 (LAMP-2).20

pon contact with HBsAg, almost all �S-C8 cTCRrafted T cells (35.5% CD4�, 58.8% CD8�) actively se-reted IFN-� but none of the �CEA cTCR grafted controlells (Figure 4A). In addition, �S-C8 cTCR but not �CEATCR grafted CD4� as well as CD8� T cells displayedAMP-2 on their surface (Figure 4B), indicating cytotoxicctivity. To verify their cytotoxicity, CD4� and CD8� Tells were isolated by magnetic cell sorting followingTCR grafting. When cocultured with target cells, CD4�s well as CD8� �S-C8 cTCR grafted T cells specificallyysed HBV replicating HepG2.2.15 cells (Figure 4C).

igure 3. Cytotoxic T-cell response and cytokine secretion. T cells graprotein (�L-5a19, open circle) or against CEA (�CEA, solid square) o

upper panel) HepG2.2.15 cells or HBV� (lower panel) HepG2 cells. T cE:T) ratios and cocultured for 72 hours. (A) Specific lysis of target cells in

edia was measured by ELISA. Mean � SD is given. (D) Antigen-specrafted T cells after 48 hours. One representative staining (E:T � 0.8:1

aken together, cTCR binding to HBV surface proteins 5

TED P

ROO

F

nable grafted CD4� and CD8� T cells to recognizeBV-infected cells, to proliferate, to secrete the cytokines

FN-� and IL-2, and, last but not least, to lyse HBV-eplicating cells.

HBV-Infected Primary Human HepatocytesAre Specifically Eliminated by Redirected TCellsOur next aim was to assess the cytotoxicity of

edirected T cells toward the natural host cells of HBV,rimary human hepatocytes, and to study whether HBV-

nfected hepatocytes are selectively and efficiently elimi-ated. We therefore infected primary human hepatocytesith wild-type HBV leading to 5% to 15% of the hepato-

ytes expressing HBV antigen.15 After 3 days, we addededirected T cells at effector to target ratio 2:1 and cocul-ured them for 96 hours. To determine hepatocyte injury,e measured ALT levels in the cell culture media (Figure

ith the cTCR directed either against HBV S (�S-C8, open diamond) orodified cells (PBL, solid rectangle) were cultured together with HBV�ere added in different dilutions to obtain indicated effector to target cellallel assays is shown. Secretion of IFN-� (B) and IL-2 (C) into cell cultureroliferation was determined by flow cytometry of CFSE stained, cTCRt of 4 stainings is shown.

fted wr unmells w3 parific p

A) of a series of independent experiments using hepa- F5

tpbb�iruu�0Itgciorbccmpgb

F�cctCLfsHp

Fairc�tihieaHElcfc

BA

SIC–LIV

ER,

PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

6 BOHNE ET AL. GASTROENTEROLOGY Vol. xx, No. x

282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337

AQ: 12

AQ: 20

282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337

AQ: 21

AQ: 22

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

TED P

ROO

F

ocytes from 3 different donors. ALT levels were com-ared with that obtained when all hepatocytes were killedy induction of apoptosis using an anti-CD95 anti-ody.15 ALT levels significantly increased when �S-C8 orL-5a19 cTCR grafted T cells were subjected to HBV

nfected cells (P � .01, Student t test), whereas ALT levelsemained unchanged when noninfected hepatocytes weresed as target cells or when �CEA cTCR grafted ornmodified T cells were used as effectors. In addition,S-C8 or �L-5a19 cTCR grafted T cells secreted 1.3 �.008 ng/mL and 0.8 � 0.007 ng/mL (mean � SD)FN-�, respectively, if incubated with HBV-infected hepa-ocytes, indicating antigen-specific activation of cTCRrafted T cells (Figure 5B). Again, �S-C8 but not �L-5a19TCR grafted T cells secreted IL-2 (Figure 5C). HBV-nfected hepatocytes released HBsAg and HBeAg at levelsf approximately 0.35 �g/mL and 26 ng/mL medium,espectively, indicating that even high amounts of solu-le HBsAg do not block activity of �S-C8 TCR grafted Tells. Despite cytotoxic activity of �S-C8 or �L-5a19TCR grafted T cells, HBsAg and HBeAg in cell cultureedia were only slightly reduced (Figure 5D and E). A

ossible explanation is the stability of the secreted anti-en and that hepatoctyes still released antigen shortlyefore dying.

igure 5. HBV-infected primary human hepatocytes are eliminated byntigen-specific, engineered T cells. Primary human hepatocytes were

nfected with HBV and cultured for 3 days prior to the addition ofedirected T cells (HBV� target cells: black columns; HBV� controlells: grey columns). T cells grafted with cTCR �S-C8, �L-5a19, orCEA, respectively, or unmodified cells (PBL) were added as an effector

o target cell ratio of 2:1 and cocultured for 96 hours. (A) Liver transam-nase (ALT) levels were determined in cell culture media as a marker forepatocyte lysis. Mean values and standard deviations obtained from 3

ndependent infection experiments are given. For comparison, ALT lev-ls of �CD95-treated cells undergoing apoptosis are shown. (B) IFN-�nd (C) IL-2 were measured in the culture supernatant by ELISA. (D)BeAg and (E) HBsAg were determined in cell culture supernatants byLISA. (F) Western blot analysis of HBV-infected PHH cells for intracel-

ular albumin and HBV core proteins. (G) HBV rcDNA and (H) HBVccDNA were quantified by real-time PCR in cellular DNA preparationsrom infected hepatocyte cultures upon coculture with redirected T

UNCORREC

igure 4. Phenotypic analysis of T-cell subsets. HBV-specific,S-C8 cTCR-grafted T cells and control T cells grafted with �CEATCR were cocultured with HBV� HepG2.2.15 or HBV� HepG2ells for 48 hours. (A) IFN-�-secreting cells were PE labelled usinghe MACS IFN-� secretion assay and detected by flow cytometry. (B)ytotoxic T cells were identified by staining for CD4, CD8, andAMP-2 by flow cytometry. (C) CD4� and CD8� cells were isolatedrom cTCR-grafted T cells or unmodified cells (PBL) by magnetic cellorting and incubated for 72 hours with HBV� HepG2.2.15 orBV� HepG2 cells. Specific lysis (mean � SD) of target cells in 2arallel assays is shown.

ells.

t1Icc1ccWdiHcn

Hgciar

clc�es

pisH

aaaHttrmteasraiarm

5sspihmtwfe

5ohtroreimdvpTaTs

lTm�lccvcvc

rhucHthomgPt

BA

SIC–L

IVER

,PA

NCREA

S,A

ND

BIL

IARY

TRA

CT

Month 2008 REDIRECTED T CELLS AND HEPATOCYTES 7

338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393

AQ: 13

AQ: 14

338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393

AQ: 15

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

UNCORREC

To control functional integrity of hepatocytes, we de-ermined intracellular albumin. This also disregarded the0% to 20% of nonhepatocytes present in PHH cultures.15

n hepatocytes cocultured with �S-C8 cTCR grafted Tells, we observed a 16%–17% reduction in hepatocytesocultured with �L-5a19 cTCR grafted T cells and a0%–14% reduction of albumin compared with hepato-ytes cocultured with �CEA cTCR grafted or control Tells as quantified by chemiluminescence imaging of

estern blots (Figure 5F). Together with the ALT activityetermined (Figure 5A), this indicated that only a minor-

ty of hepatocytes was lysed. Regarding that 0.35 �g/mLBsAg relate to 4.4 � 104 subviral particles per hepato-

yte, we concluded that binding of subviral particles didot induce unspecific killing of noninfected hepatocytes.In contrast to albumin, HBV core protein as well asBV DNA levels was markedly reduced. �S-C8 cTCR

rafted T cells reduced HBV core by 73% and �L-5a19TCR grafted T cells by 57% (Figure 5F). They reducedntracellular HBV relaxed circular DNA (rcDNA) by 82%nd 72%, respectively, as quantified by real-time PCRelative to an external standard (Figure 5G).

Most remarkably, the replication template, HBVccDNA, was eliminated (�99.99% reduction, detectionimit 0.0025 copies per cell) by �S-C8 cTCR grafted Tells and reduced to 0.4 copies/cell (�80% reduction) byL-5a19 cTCR grafted T cells (Figure 5H). In controlxperiments, average levels of 2.5 copies/cell were ob-erved.

These data indicate that T cells grafted with HBV Srotein-specific cTCR �S-C8 are able to eliminate HBV-

nfected primary human hepatocytes. A lesser, but alsopecific, effect was observed for T cells equipped with theBV L protein specific cTCR �L-5a19.

DiscussionIn this study, we compared cTCRs directed

gainst 2 different surface antigens of HBV, the large Lnd the small S protein, with respect to their ability torm primary T cells against HBV-infected hepatocytes.BV surface antigens S and L are constantly present on

he surface of HBV replicating cells21 because of theargeting of HBV surface proteins into the endoplasmiceticulum (ER) membrane and steady exchange of ER

embranes with the plasma membrane.22 We aimed atargeting HBV S and L protein positive hepatocytes toliminate residual HBV-replicating cells by inducing anrtificial cytotoxic T-cell response mediated by antigen-pecific cTCRs. Using S- and L-specific scFv as antigenecognition domains of the cTCRs, we demonstratedntigen-specific activation of engineered T cells resultingn cytokine secretion, expansion of this T-cell populationnd, most importantly, cytotoxic elimination of HBV-eplicating hepatoma cells as well as HBV-infected pri-

ary human hepatocytes. m

TED P

ROO

F

For our study, we selected 2 scFv designated C8 anda19 for grafting specificity to the recombinant TCR.cFv C8 was selected because it is directed against auperficial epitope in the “a” determinant of HBV Srotein, which is expressed at very high amounts by

nfected cells. Among all scFv tested, scFv C8 showedighest binding efficiency and specificity and provedost effective and specific in targeting and T-cell activa-

ion. To obtain a suitable scFv against HBV L protein,hich is expressed in markedly lower amounts by in-

ected cells, we used hybridoma cells producing well-stablished monoclonal antibody 5a19.17

Antigen recognition by T cells bearing �S-C8 and �L-a19 cTCR, respectively, resulted in high-level secretionf the proinflammatory cytokine IFN-�. Because IFN-�as a direct antiviral effect on HBV replication in hepa-ocytes,23 we assume that IFN-� secretion by activated,edirected T cells contributes to virus control. Activationf the CD28 signalling domain in the cTCR construct24

esulted in secretion of IL-2. Cytokine levels were mark-dly higher if T cells grafted with �S-C8 cTCR were usedn comparison with �L-5a19 cTCR grafted T cells. The

inor activation potential of the �L-5a19 TCR might beue to the lower abundance of L protein on the surface ofiral and subviral particles, which contain 83%–94% Srotein, but only 1%–2% L, and 5%–15% middle protein.25

ime course experiments accordingly showed a delayedctivation kinetic of T cells with the L protein-specificCR �L-5a19 in comparison with the HBV S protein-

pecific cTCR.cTCR grafted T cells bound subviral HBV particles

eading to activation of NF-�B and secretion of IFN-�.his is in contrast to observations that soluble, unlikeembrane-bound, CEA does not lead to activation of

CEA cTCR26 and that soluble monomeric HIV-1 enve-ope protein gp120 bound less efficiently to a specificTCR than the respective transmembrane protein.13 Be-ause TCR-complex clustering is required for T-cell acti-ation via genuine TCRs27 as well as via recombinantTCRs,28 our data imply that the surface area of HBViral and subviral particles is sufficient to induce TCRlustering.

To address the question of whether activation of redi-ected T cells results in nonspecific lysis of uninfectedepatocytes in the neighborhood of infected cells, wesed primary human hepatocytes as target cells, whichan be infected with HBV and support all steps of theBV replication cycle. In cell culture, only a minority (5%

o 15%) of hepatocytes is productively infected, althoughepatocytes are incubated with HBV at high multiplicityf infection.15 When cocultured with HBV-infected pri-ary human hepatocytes, �S-C8 as well as �L-5a19 cTCR

rafted, but not �CEA cTCR grafted, T cells or controlBL, secreted IFN-�. However, they lysed only a propor-ion of the hepatocytes. Mild ALT activity in cell culture

edia and reduction of albumin production by hepato-

ct�s9csh

wteecbsaswt

bmnHwblspHrTpcic

icuditlrHoti

cstcf

utriAol

ct

1

1

1

1

1

BA

SIC–LIV

ER,

PA

NCREA

S,A

ND

BILIA

RY

TRA

CT

8 BOHNE ET AL. GASTROENTEROLOGY Vol. xx, No. x

394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449

AQ: 16

394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449

AQ: 17

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

UNCORREC

ytes, as well as microscopic evaluation, indicated thathe majority of hepatocytes remained intact. In contrast,S-C8 cTCR grafted T cells reduced HBV cccDNA repre-enting the nuclear persistence form of HBV by at least9.99% to undetectable levels. �L-5a19 cTCR grafted Tells reduced HBV cccDNA by more than 80%. Thistrongly argues for a specific elimination of HBV-infectedepatocytes by redirected T cells.Although we propose that HBV-infected hepatocytes

ere eliminated by �S-C8 grafted T cells, HBV core pro-ein and HBV rcDNA remained detectable. A possiblexplanation may be the mode by which redirected T cellsliminate their target cell. We found activation of effectoraspases 3 and 7 in target hepatocytes and were able tolock cytotoxicity by a pancaspase inhibitor (data nothown). We therefore expect an induction of caspasectivated DNAses,29 which will preferentially degrade hi-tone-bound nuclear cccDNA but not HBV rcDNA,hich is contained in viral capsids and therefore pro-

ected from DNAse activity.Because �L-5a19 and �S-C8 cTCR grafted T cells

ound subviral particles, we asked whether this bindingight either block T-cell effector functions or induce

onspecific killing of noninfected cells. High levels ofBsAg were present in the supernatants of coculturesith primary hepatocytes. These amounts are compara-le with levels achieved after antiviral therapy with pegy-

ated interferon � and nucleoside analogues6 and corre-pond to approximately 40,000 subviral HBV particleser hepatocyte or 20,000 per T cell. This vast excess ofBV particles did not impair the cytotoxic activity of

edirected T cells, indicating that they did not block-cell effector functions. Furthermore, HBV cccDNA-ositive cells were selectively lysed, although all hepato-ytes were subjected to high amounts of HBsAg. Thisndicates that soluble HBsAg itself does not mediate theytotoxic effect.

Although the in vitro performance of redirected T cellss very promising, several safety precautions have to beonsidered before translation into the clinics. In partic-lar, one needs to explore how strictly redirected T cellsifferentiate between infected hepatocytes and neighbor-

ng noninfected cells, which may bind HBV particles onheir surface. In addition, hepatocytes in the inflamediver may be more sensitive to cytolysis. We think it iseasonable to consider T-cell therapy after suppression ofBV replication by antiviral therapy. However, it remains

pen as to how efficiently redirected T cells will invadehe liver parenchyma and whether they will escape silenc-ng in an organ, which modulates immune responses.

Taken together, we have generated and qualitativelyompared cTCRs with specificities for 2 different HBVurface proteins. These TCRs were retrovirally deliveredo primary human T cells, activated them to secreteytokines, and enabled them to kill stably HBV trans-

ected as well as HBV infected primary hepatocytes. Sol-

TED P

ROO

F

ble HBV particles bound to cTCR but did not blockhem. The cytotoxic response was specific toward HBV-eplicating cells and was sufficient to eliminate HBV-nfected cells from a primary human hepatocyte culture.dditionally, antigen recognition resulted in expansionf redirected T cells, a key requisite for effective repopu-

ation when used for an adoptive transfer.

Appendix

Supplementary data

Supplementary data associated with this articlean be found, in the online version, at doi:10.1053/j.gas-ro.2007.11.002.

References

1. Chisari FV, Ferrari C. Hepatitis B virus immunopathology. SpringerSemin Immunopathol 1995;17:261–281.

2. Rehermann B, Fowler P, Sidney J, et al. The cytotoxic T lympho-cyte response to multiple hepatitis B virus polymerase epitopesduring and after acute viral hepatitis. J Exp Med 1995;181:1047–1058.

3. Penna A, Chisari FV, Bertoletti A, et al. Cytotoxic T lymphocytesrecognize an HLA-A2-restricted epitope within the hepatitis B virusnucleocapsid antigen. J Exp Med 1991;174:1565–1570.

4. Wieland SF, Spangenberg HC, Thimme R, et al. Expansion andcontraction of the hepatitis B virus transcriptional template ininfected chimpanzees. Proc Natl Acad Sci U S A 2004;101:2129–2134.

5. Ganem D, Prince AM. Hepatitis B virus infection—natural historyand clinical consequences. N Engl J Med 2004;350:1118–1129.

6. Wursthorn K, Lutgehetmann M, Dandri M, et al. Peginterferon�-2b plus adefovir induce strong cccDNA decline and HBsAgreduction in patients with chronic hepatitis B. Hepatology 2006;44:675–684.

7. Rooney CM, Smith CA, Ng CY, et al. Use of gene-modified virus-specific T lymphocytes to control Epstein-Barr-virus-related lym-phoproliferation. Lancet 1995;345:9–13.

8. Walter EA, Greenberg PD, Gilbert MJ, et al. Reconstitution ofcellular immunity against cytomegalovirus in recipients of alloge-neic bone marrow by transfer of T-cell clones from the donor.N Engl J Med 1995;333:1038–1044.

9. Morgan RA, Dudley ME, Wunderlich JR, et al. Cancer regression inpatients after transfer of genetically engineered lymphocytes.Science 2006;314:126–129.

0. Gattinoni L, Powell DJ Jr, Rosenberg SA, et al. Adoptive immuno-therapy for cancer: building on success. Nat Rev Immunol 2006;6:383–393.

1. Eshhar Z, Waks T, Gross G, et al. Specific activation and targetingof cytotoxic lymphocytes through chimeric single chains consist-ing of antibody-binding domains and the or subunits of theimmunoglobulin and T-cell receptors. Proc Natl Acad Sci U S A1993;90:720–724.

2. Abken H, Hombach A, Heuser C, et al. Tuning tumor-specific T-cellactivation: a matter of costimulation? Trends Immunol 2002;23:240–245.

3. Masiero S, Del Vecchio C, Gavioli R, et al. T-cell engineering by achimeric T-cell receptor with antibody-type specificity for the HIV-1gp120. Gene Ther 2005;12:299–310.

4. Sells MA, Chen ML, Acs G. Production of hepatitis B virus parti-cles in Hep G2 cells transfected with cloned hepatitis B virus

DNA. Proc Natl Acad Sci U S A 1987;84:1005–1009.

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

fgD(

t

sCfha

ER,

AN

DA

CT

Month 2008 REDIRECTED T CELLS AND HEPATOCYTES 9

450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505

450451452453454455456457458459460461462463464465466467468469470471472473474475476477

AQ: 3

tapraid2/zg0-gast/zg0-gast/zg099908/zg04366d07z goodj S�1 11/15/07 Art: 54080

5. Schulze-Bergkamen H, Untergasser A, Dax A, et al. Primary hu-man hepatocytes—a valuable tool for investigation of apoptosisand hepatitis B virus infection. J Hepatol 2003;38:736–744.

6. Untergasser A, Zedler U, Langenkamp A, et al. Dendritic cellstake up viral antigens but do not support the early steps ofhepatitis B virus infection. Hepatology 2006;43:539–547.

7. Pizarro JC, Vulliez-le Normand B, Riottot MM, et al. Structural andfunctional characterization of a monoclonal antibody specific forthe preS1 region of hepatitis B virus. FEBS Lett 2001;509:463–468.

8. Hombach A, Sent D, Schneider C, et al. T-cell activation byrecombinant receptors: CD28 costimulation is required for inter-leukin 2 secretion and receptor-mediated T-cell proliferation butdoes not affect receptor-mediated target cell lysis. Cancer Res2001;61:1976–1982.

9. Weijtens ME, Willemsen RA, Hart EH, et al. A retroviral vectorsystem “STITCH” in combination with an optimized single chainantibody chimeric receptor gene structure allows efficient genetransduction and expression in human T lymphocytes. Gene Ther1998;5:1195–1203.

0. Betts MR, Brenchley JM, Price DA, et al. Sensitive and viableidentification of antigen-specific CD8� T cells by a flow cytomet-ric assay for degranulation. J Immunol Methods 2003;281:65–78.

1. Chu CM, Liaw YF. Membrane staining for hepatitis B surfaceantigen on hepatocytes: a sensitive and specific marker of activeviral replication in hepatitis B. J Clin Pathol 1995;48:470–473.

2. Gorelick FS, Shugrue C. Exiting the endoplasmic reticulum. MolCell Endocrinol 2001;177:13–18.

3. Dumortier J, Schonig K, Oberwinkler H, et al. Liver-specific ex-pression of interferon following adenoviral gene transfer con-trols hepatitis B virus replication in mice. Gene Ther 2005;12:668–677.

UNCORREC

TED P

ROO

F

4. Hombach A, Muche JM, Gerken M, et al. T cells engrafted with arecombinant anti-CD30 receptor target autologous CD30(�) cu-taneous lymphoma cells. Gene Ther 2001;8:891–895.

5. Ganem D, Schneider R. Hepadnaviridae. The viruses and theirreplication. Field’s virology. 4th ed. Philadelphia: Lipincott-Raven,2001.

6. Hombach A, Koch D, Sircar R, et al. A chimeric receptor that selec-tively targets membrane-bound carcinoembryonic antigen (mCEA) inthe presence of soluble CEA. Gene Ther 1999;6:300–304.

7. Kolanus W, Romeo C, Seed B. T-cell activation by clusteredtyrosine kinases. Cell 1993;74:171–183.

8. Hombach A, Kohler H, Rappl G, et al. Human CD4� T cells lysetarget cells via granzyme/perforin upon circumvention of MHCclass II restriction by an antibody-like immunoreceptor. J Immunol2006;177:5668–5675.

9. Enari M, Sakahira H, Yokoyama H, et al. A caspase-activatedDNase that degrades DNA during apoptosis, and its inhibitorICAD. Nature 1998;391:43–50.

Received November 17, 2006. Accepted October 11, 2007.Address requests for reprints to: Ulrike Protzer, MD, Molecular In-

ectiology, Institute for Medical Microbiology, Immunology, and Hy-iene, University Hospital Cologne, Haus 37, Kerpener Str. 62;-50931 Koeln, Germany. e-mail: ulrike.protzer@uni-koeln.de; fax:

49) 221-4787288.Supported by the Medical Faculty of the University of Cologne and by

he Deutsche Forschungsgemeinschaft.The authors thank A. Budkowska for kindly providing the L Protein

pecific hybridoma cell line 5a19, Andreas Hombach for providing theEA-specific cTCR construct and for technical advice, Diana Wagneror excellent technical assistance, Eva Gückel and Gregor Hohn forelp with the experiments, Rolf Kaiser for performing serological HBVssays, and Percy Knolle for helpful discussions.Conflicts of interest: No conflicts of interest to disclose.

BA

SIC–L

IVPA

NCREA

S,BIL

IARY

TR

478479480481482483484485486487488489490491492493494495496497498499500501502503504505

JOBNAME: AUTHOR QUERIES PAGE: 1 SESS: 1 OUTPUT: Thu Nov 15 08:53:53 2007/tapraid2/zg0�gast/zg0�gast/zg099908/zg04366d07z

AQ18— Please clarify what is meant by „„(20)“ in the following: „„.. . long terminal repeat (LTR),splice donor (SD), splice acceptor (SA), packaging signal (() (20).”

AQ19— Figure 3 has been redrawn to match journal style, please check that legend correspondswith figure.

AQ20— Figure 4 has been redrawn to match journal style, please check that legend correspondswith figure.

AQ21— Figure 5 has been redrawn to match journal style, please check that legend correspondswith figure.

AQ22— Please verify that the following is meant: „„.. . or unmodified cells (PBL) were added asan effector. . .”

AQ1— Please approve running title.

AQ2— Please expand ZMBH and GSF in affiliation,if possible

AQ3— Please verify e-mail address and fax number and give permission to print.

AQ4— Please verify that the following is meant: „„.. . carrying a chimeric T-cell receptor (cTCR). ..”

AQ5— Please verify that „„supposed“ is meant in the following: „„We supposed the designedcTCRs to recognize HBV S or L protein. . .”

AQ6— Please verify that the following is meant: „„Gibbon ape leukaemia virus (GALV) envelopepseudotyped, amphotropic retroviral vectors. .. “

AQ7— Please verify that „„anti“ is not meant for „„CD8“ in the following: „„.. . stained with PE-labelled murine anti-CD3, -CD4, or CD8 antibodies. . .”

AQ8— Please provide abbreviation for state in the following and all other locations ofmanufacturers of brand-name products that are located in United States: (BD Bioscience,Franklin Lakes).

AQ9— „„{Pizarro, 2001 #73}„„ has been deleted from the following. If meant as other thanreference, please replace with additional information provided for clarity: “.. . to amino acid

AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES 1

JOBNAME: AUTHOR QUERIES PAGE: 2 SESS: 1 OUTPUT: Thu Nov 15 08:53:53 2007/tapraid2/zg0�gast/zg0�gast/zg099908/zg04366d07z

(aa) 37 to 43 in the preS1 region of HBV subtype ayw better than adw {Pizarro, 2001 #73.

AQ10— Please verify that the following is correct as meant: „„.. ., which carry S and to a lowextend middle and L proteins on their surface.”

AQ11— Please verify that HBV negative (-), and not hyphen, is meant in the following andthroughout article when used: „„.. . or as a control with HBV( parental HepG2 cells.”

AQ12— Please verify that the following is meant: „„A possible explanation is the stability of thesecreted antigen and that hepatoctyes still released antigen shortly before dying.”

AQ13— Please verify that the following is meant: „„This also disregarded the 10% to 20% ofnonhepatocytes present in PHH cultures.15”

AQ14— Please verify that the following is meant: „„.. . ., we observed a 16%–17% reduction inhepatocytes cocultured with (L-5a19 cTCR grafted T cells and a 10%–14% reduction ofalbumin compared with hepatocytes cocultured with (CEA cTCR. . .”

AQ15— Please verify that the following is meant: „„.. ., although hepatocytes are incubated withHBV at high multiplicity of infection.15”

AQ16— Please verify that „„cytolysis“ is meant in the following: „„.. . .hepatocytes in the inflamedliver may be more sensitive to cytolysis.”

AQ17— Please check updated reference 9— from Pubmed, OK?

AUTHOR QUERIES

AUTHOR PLEASE ANSWER ALL QUERIES 2