Prevention of hepatocyte allograft rejection in rats by transferring adenoviral early region 3 genes into donor cells


  • Elena V. Mashalova,

    1. Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, NY
    Search for more papers by this author
  • Chandan Guha,

    1. Department of Radiation Oncology, Albert Einstein College of Medicine, Bronx, NY
    2. Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, NY
    Search for more papers by this author
  • Namita Roy-Chowdhury,

    1. Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, NY
    2. Department of Medicine, Albert Einstein College of Medicine, Bronx, NY
    3. Department of Molecular Genetics, Albert Einstein College of Medicine, Bronx, NY
    Search for more papers by this author
  • Laibin Liu,

    1. Department of Radiation Oncology, Albert Einstein College of Medicine, Bronx, NY
    Search for more papers by this author
  • Ira J. Fox,

    1. Department of Surgery, University of Nebraska Medical Center, Omaha, NE
    Search for more papers by this author
  • Jayanta Roy-Chowdhury,

    Corresponding author
    1. Marion Bessin Liver Research Center, Albert Einstein College of Medicine, Bronx, NY
    2. Department of Medicine, Albert Einstein College of Medicine, Bronx, NY
    3. Department of Molecular Genetics, Albert Einstein College of Medicine, Bronx, NY
    • Marion Bessin Liver Research Center, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461
    Search for more papers by this author
    • fax: 718-430-8975

  • Marshall S. Horwitz

    1. Department of Microbiology and Immunology, Albert Einstein College of Medicine, Bronx, NY
    2. Department of Pediatrics, Albert Einstein College of Medicine, Bronx, NY
    Search for more papers by this author
    • Deceased.

  • This research was conducted by E. V. Mashalova in partial fulfillment of the requirements for the degree of doctor of philosophy at the Sue Golding Graduate Division of Medical Sciences, Albert Einstein College of Medicine, Yeshiva University, Bronx, NY.

  • Potential conflict of interest: Nothing to report.


Hepatocyte transplantation is being evaluated as an alternative to liver transplantation for metabolic support during liver failure and for definitive treatment of inherited liver diseases. However, as with liver transplantation, transplantation of allogeneic hepatocytes requires prolonged immunosuppression with its associated untoward effects. Therefore, we explored strategies for the genetic modification of donor hepatocytes that could eliminate allograft rejection, obviating the need for immunosuppression. Products of early region 3 (AdE3) of the adenoviral genome are known to protect infected cells from immune recognition and destruction. In the present study we showed that immortalized rat hepatocytes that had been stably transduced with AdE3 before transplantation into fully MHC-mismatched rats are protected from allograft rejection. Quantitative real-time PCR analysis showed that a similar number of engrafted AdE3-transfected hepatocytes had survived in syngeneic and allogeneic recipients. AdE3 expression did not reduce expression of MHC class I on the surfaces of donor hepatocytes. Consistent with this, the in vivo cytotoxic cell–mediated alloresponse was attenuated but not abolished in recipients of AdE3-transfected allogeneic hepatocytes. In contrast, graft survival correlated with a marked reduction in cell-surface localization of Fas receptor in the transplanted cells and inhibition of Fas-mediated apoptosis, which are related to the antiapoptotic functions of the AdE3 proteins. CONCLUSION:AdE3 gene products prevent hepatocyte allograft rejection mainly by protecting the cells from the effector limb of the host immune response and could be used as a tool to facilitate allogeneic hepatocyte transplantation. (HEPATOLOGY 2007;45:755–766.)

Transplantation of isolated hepatocytes is being evaluated clinically as a minimally invasive procedure to provide immediate metabolic support until a donor liver becomes available for transplantation or to permanently ameliorate liver-based inherited metabolic diseases.1 However, like whole organs, hepatocytes are subject to allograft rejection, requiring long-term administration of immunosuppressive agents with their associated untoward effects.2 Engrafted hepatocytes elicit an alloresponse pattern that is different from that of other transplanted tissues, such as vascularized heart and nonvascularized pancreatic islets.3–5 For example, depletion of CD4+ T cells or blocking CD28/B7 costimulation, which ameliorates allograft rejection of other transplanted tissues, has been inadequate to prevent hepatocyte allorejection.2 Experimental and clinical findings indicate that apoptosis is the predominant mechanism of hepatocyte allograft rejection and suggest death receptors such as Fas6–9 and TNF receptor type 1 (TNF-R1)10 play a key role.

Genetic modification of donor cells to prevent their recognition and/or destruction by the host immune system could provide an attractive alternative to systemic immunosuppression. In the present study, we evaluated the possibility of preventing allograft rejection by expressing nonsecreted immunomodulatory gene products of adenovirus early region 3 (AdE3)11 in immortalized donor hepatocytes. Several properties of the AdE3-encoded proteins are pertinent. The AdE3 glycoprotein 19K (gp19K) binds to the heavy chain of histocompatibility complex (MHC) class I, thereby down-regulating its cell-surface expression,12, 13 and interferes with presentation of MHC class I antigens by inhibiting tapasin-mediated peptide processing.14 The AdE3–receptor internalization and degradation (RID)α/β complex (10.4K/14.5K) inhibits cell-surface localization of death receptors such as Fas,15–17 TNF-related apoptosis-inducing ligand receptor-1 and receptor-2 (TRAIL-R1/2),18, 19 and TNF-R1,20, 21 thereby inhibiting the apoptosis induced by their respective ligands. AdE3-RIDα/β also inhibits TNF-α-induced apoptosis through a mechanism that does not primarily involve receptor down-regulation.22 AdE3-14.7K also inhibits apoptosis mediated by TNF-α,23 Fas ligand (FasL),24 and TRAIL19 by interacting with intracellular components of the signaling pathways.11, 25 We showed that AdE3-expressing pancreatic islets derived from transgenic mice are not rejected after transplantation into allogeneic recipients.26 In addition, transgenic expression of AdE3 in β-cells prevented or delayed the onset of type 1 diabetes in 2 murine models.27, 28 On the basis of these findings, we hypothesized that ex vivo transfer of AdE3 genes into donor hepatocytes should prevent their rejection after transplantation into allogeneic recipients by suppressing their allorecognition and/or protecting them from killing by effector T lymphocytes. To test this, we stably transfected the AdE3 region of the adenoviral genome into 2 conditionally immortalized rat hepatocyte lines that were then transplanted into allogeneic hosts in order to evaluate their ability to engraft, survive, and proliferate and to evoke host cell-mediated alloresponse. In addition, we identified AdE3 functions that could explain allograft survival.


AdE3, adenovirus early region 3; FasL, Fas ligand; gp19K, glycoprotein 19K; RID, receptor internalization and degradation; Tag, SV40 T antigen.

Materials and Methods

Hepatocyte Lines.

The Immortalized Gunn Rat Hepatocyte (IGRH) line was derived from hepatocytes isolated from the liver of inbred Gunn rats (Wistar-RHA background) and conditionally immortalized as described previously.29 The Rat Hepatocyte 69 (RH69) line was derived from hepatocytes isolated from the liver of an inbred Lewis rat and conditionally immortalized using a recombinant Moloney's murine leukemia virus–based retroviral vector, SSR69, which contains the SV40 T antigen (Tag) gene flanked by loxP sites, as described previously.30, 31

The IGRH-E3 and RH69-E3 hepatocyte lines, which contain the entire AdE3 transcription unit expressed from CMV promoter, were generated by stable transfection of the IGRH and RH69 hepatocyte lines, respectively. The vector for delivery of the AdE3, pΔE1-CMV-E3, was generated previously.32 Cultured IGRH and RH69 hepatocytes were cotransfected with pΔE1-CMV-E3 and pcDNA3.1/Zeo (at a ratio of 5:1; Invitrogen) with FuGENE reagent (Roche Diagnostics Corp.) at a reagent/plasmid DNA ratio of 3:1, and colonies representing clonal cell lines were selected with Zeocin (Invitrogen) treatment at a final concentration of 400 μg/ml. In parallel, IGRH and RH69 cells were transfected with pcDNA3.1/Zeo alone, selected for Zeocin resistance, and used as AdE3-negative “parental cells” in the transplantation studies and various in vitro experiments.

Monolayers of IGRH and IGRH-E3 hepatocytes were maintained in Dulbecco's Modification of Eagle's Medium (Cellgro) supplemented with 10% Fetal Bovine Serum (FBS; Gemini Bio-Products), 50 U/ml penicillin (Cellgro), and 50 μg/ml streptomycin (Cellgro) in a humidified incubator at 33°°C and 5% CO2. Monolayers of RH69 and RH69-E3 hepatocytes were maintained in Waymouth's medium (Invitrogen) supplemented with 10% FBS, 50 U/ml penicillin, 50 μg/ml streptomycin, 0.1 mM insulin-transferrin-selenium (Invitrogen), and 0.1 mM dexamethasone (Sigma-Aldrich) in a humidified incubator at 37°C and 5% CO2.


Inbred Gunn rats and Wistar-RHA rats (congeneic to the Gunn rats) were obtained from our colony at the Special Animal Core of the Marion Bessin Liver Research Center of the Albert Einstein College of Medicine. Inbred Fisher 344 (F344) rats and Lewis rats were purchased from Charles River Laboratories. The animals were housed at the Institute for Animal Studies of the Albert Einstein College of Medicine. The experimental protocols were approved by the Institutional Animal Care and Use Committee.

Hepatocyte Transplantation.

Rats were anesthetized by ether inhalation and 20 × 106 viable hepatocytes suspended in 0.5 ml of PBS were transplanted as described previously.33 In some experiments, to promote preferential proliferation of transplanted hepatocytes in the livers, the rats were subjected to 68% hepatectomy and hepatic irradiation (50 Gy) 48 hours before intrasplenic injection of 5 × 106 hepatocytes.

Adenovirus Infection.

Ad/E3, Ad/RID, and Ad/null viruses, which were described previously,34 were kind gifts of William S. M. Wold (Saint Louis University, St. Louis, MO). Infection to control AdE3 protein expression was done with 2,000 particles/cell (MOI ∼ 20) for 36 hours.

Lentivirus Transduction.

A second-generation lentiviral vector expressing GFP under the control of an intronless elongation factor 1-alpha promoter (EF1-α short) was generated by standard methods in 293 cells.35 IGRH and IGRH-E3 cells were grown in culture and infected with the lentiviral vector at an MOI of 10 in growth medium containing 8 μg/ml polybrene (Sigma-Aldrich).

CFSE Labeling.

RH69 and RH69-E3 hepatocytes were grown in culture and labeled as adherent cells with CFSE (CFDA SE; Molecular Probes) at a final concentration of 20 μM according to the manufacturer's instructions. Rat splenocytes were suspended at 50 × 106 cells/ml in PBS with 5% FBS and incubated with CFSE at a final concentration of either 5 μM (for high-fluorescence intensity) or 0.5 μM (for low-fluorescence intensity) for 5 minutes at room temperature while protected from light.


Five hundred nanograms of total RNA was reverse-transcribed with the Superscript system (Invitrogen) using oligo (dT) priming. The cDNA products subsequently were used for PCR analysis with the following oligonucleotide primer sequences: primers 53 and 87, which were described previously36 for the detection of AdE3 sequences and mRNA; CAATGCCTGTTTCATGCC (forward) and CCTGGAATAGTCACCATG (reverse) for Tag; CGTTCTGGTTCGATACACC (forward) and GAAGTCACCCATCACCGTC (reverse) for albumin; AGTTAGTCGAGTCACAGCTGG (forward) and GCAGACGTCATCATTCCAG (reverse) for the asialoglycoprotein receptor.

Quantitative Real-Time PCR.

Oligonucleotide primer sequences for detecting Tag DNA were described previously.37 PCR was carried out in glass capillaries with the SYBR Green PCR Master Mix (Applied Biosystems) using 600 ng of DNA and 0.2 μM (final concentration) of each primer. Assays were performed in quadruplicate in a 7900HT thermal cycler (Applied Biosystems) under the following conditions: initial denaturation for 5 minutes at 94°C, followed by 50 cycles of 15 seconds at 94°C and 60 seconds at 60°C. Fluorescence emissions were collected every 7 seconds for the length of the run for each reaction well. Each DNA sample was also analyzed for rat β-actin DNA using the following oligonucleotide primer sequences: forward, 5′- AAGTCCCTCACCCTCCCAAAAG-3′; reverse, 5′- AAGCAATGCTGTCACCTTCCC-3′. Standard curve analysis for absolute quantification of Tag copy number per input DNA was performed with genomic DNA isolated from the Tag-positive cell line (IGRH; 1 copy Tag/genome), which was 10-fold serially diluted in SV40-negative DNA to produce equal amounts of 600 ng of the DNA template per reaction in each dilution sample. Amplification plots and predicted threshold cycles (Ct values) from the exponential phase of the PCR were analyzed using Sequence Detection Software (version 2.2; Applied Biosystems). Amplification specificity was determined by melting curve analysis using Dissociation Curve software (Applied Biosystems).

Western Blot Analysis.

Between 5 and 10 μg of protein was separated on 12% SDS polyacrylamide gels (BioRad) and transferred electrophoretically to Hybond C membranes (Amersham). The primary antibodies used were: Fas (M-20, 5 μg/ml final concentration; Santa Cruz Biotechnology), β-tubulin (H-235, 200 μg/ml at 1:1,000 dilution; Santa Cruz Biotechnology), rabbit antisera to RIDβ (1:500; Genemed Synthesis Inc.), rabbit antisera to gp19K (1:2,000; Covance), and rabbit antisera to 14.7K (1:1,000; Covance). The blots were incubated with a horseradish peroxidase–conjugated secondary antibody (Amersham) and developed with Supersignal West Pico Substrate (Pierce).

Flow-Cytometric Analysis.

MHC class I and Fas staining were performed with biotin-conjugated anti-rat RT1A (OX-18, 1μg/1 × 106cells; BD Biosciences) and anti-CD95 (IPO-4, 1μg/ml × 106cells; GeneTex, Inc.), respectively. For negative controls, the primary antibodies were substituted with appropriate isotype controls. To identify cells undergoing apoptotic death, Annexin V-PE/7-amino-actinomycin D (7-AAD) double staining was performed using an Annexin V-PE Apoptosis Detection Kit I (BD Pharmingen). Briefly, cells were grown in culture and treated with Fas agonistic antibody (IPO-4, 1 μg/mL; GeneTex Inc.) for 16 hours, stained with Annexin V-PE and 7-AAD according to the manufacturer's instructions, and subjected to flow cytometry analysis using FACScalibur (Becton-Dickinson) and CellQuest software (BD Biosciences).

Indirect Immunofluorescence Staining and Microscopy.

Staining was performed on either methanol-fixed 5-μm liver cryosections or cells, grown on coverslips using mouse mAb against Tag (Ab-2, 5 μg/ml final concentration; Oncogene Research Products) or rabbit polyclonal antibody against rat FasR (A-20, 200 μg/ml at 1:50 dilution; Santa Cruz Biotechnology). The secondary antibody was Cy5-conjugated antimouse IgG (2 μg/mL) or Cy3-conjugated antirabbit IgG (3 μg/ml; Jackson ImmunoResearch). For negative controls, appropriate isotypes were used. Fluorescently labeled tissue sections and cells were mounted using ProLong Antifade reagent (Molecular Probes). Microscopy was performed with an Olympus IX70 inverted microscope. Image recording was performed by a 12-bit Cooke Sensicam QE cooled CCD cameras and IPLab software. Laser capture dissection of desired cells from fluorescently stained frozen sections was performed using Laser Capture Microscopy (Leica) according to the manufacturer's standard protocol.

In Vivo Cytotoxicity Assay.

To prepare target cells for in vivo alloantigen stimulation and evaluation of cytotoxic activity, erythrocytes were removed from Wistar-RHA rat spleen cell suspensions by osmotic lysis. The cells were then labeled with 0.5 μM of CFSE for low-fluorescence intensity (the target), as described in the CFSE labeling section. Syngeneic control cells were prepared from Lewis rat/spleen cell suspensions and labeled with 5 μM CFSE for high-fluorescence intensity (the control). For adoptive transfer, 6-8 × 107 cells from each population were mixed at a 1:1 ratio in 0.5 ml of PBS and injected into the tail vein of Lewis rats that had been previously transplanted with either IGRH or IGRH-E3 hepatocyte allografts (Wistar RHA strain) by intrasplenic injection as described in the Hepatocyte Transplantation section. No preparative treatment was used in these experiments. Naive (nontransplanted) Lewis rat recipients were used as controls. Recipients were sacrificed 20 hours after adoptive transfer, the spleens were harvested, and splenic cell suspensions were prepared. Allospecific cytotoxicity was determined by detecting the differentially labeled fluorescent allogeneic target and syngeneic control cell populations in the spleen cell suspensions using flow-cytometric analysis. FACS analysis was performed with FACScalibur (Becton-Dickinson) and CellQuest (BD Biosciences). The percent specific lysis of fluorescent allogeneic target cells in each rat was calculated as follows:

equation image

where A is the number of targets per number of controls in nonimmune recipients.

Statistical Analysis.

Data are presented as means ± SDs and were analyzed by the 2-tailed Student's t test. P values less than 0.05 were considered statistically significant.

Results and Discussion

Stable Transfection of Conditionally Immortalized Rat Hepatocytes with AdE3 Genes.

A plasmid, pΔE1-CMV-E3, expressing AdE3 genes from adenovirus type 2 from a CMV immediate early promoter32 was transfected into 2 conditionally immortalized rat hepatocyte lines, IGRH (Wistar RHA haplotype) and RH69 (Lewis haplotype). The IGRH and RH69 cell lines had been generated previously by retrovector-mediated transduction with thermoliable SV40 T-antigen (Tag)29 and loxP -flanked wild-type Tag,30 respectively. Stably transfected cell clones (Supplemental Fig. 1B), which expressed all AdE3 mRNA, as indicated by the typical splicing pattern of the AdE3 transcript (Supplemental Fig.1A-C), were further assessed for AdE3 protein expression, IGRH-E3(3), expressed RIDβ, and 14.7K (Supplemental Fig.1D) and a high level of gp19K protein (Supplemental Fig.1E). Similarly, an AdE3-transfected RH69 clone (RH69-E3[19]) expressed both AdE3 mRNA (not shown) and AdE3 protein (Supplemental Fig.1F). The IGRH-E3(3) and RH69-E3(19) lines were used in the transplantation studies.

Figure 1.

In vivo trafficking and engraftment capacity of intrasplenically injected AdE3-transduced hepatocytes. (A) Appearance of IGRH-E3 hepatocytes in vitro after Lentivector-mediated transduction with GFP (fluorescent microscopy, ×60). (B) Detection of GFP-labeled IGRH-E3 hepatocytes in the liver of syngeneic hosts 72 hours after intrasplenic injection. (C) Detection of GFP-labeled IGRH-E3 hepatocytes in the liver of syngeneic hosts 3 weeks after intrasplenic injection. (D) Detection of CFSE-labeled RH69-E3 hepatocytes in syngeneic hosts' spleen 72 hours after intrasplenic injection. (E) Detection of CFSE-labeled RH69-E3 hepatocytes in syngeneic hosts' liver 72 hours after intrasplenic injection. (F) Detection of CFSE-labeled RH69-E3 hepatocytes in syngeneic hosts' spleen 2 weeks after intrasplenic injection. Green fluorescence (top) and phase-contrast (bottom) images of the same field are shown (fluorescent microscopy, ×10). (G) RT-PCR detection of albumin in RH69 and RH69-E3 cells in culture and after engraftment in the spleen. (H) RT-PCR detection of asialoglycoprotein receptors (asgpRs) in RH69 and RH69-E3 cells in culture and after engraftment in the spleen.

AdE3 Gene Expression Did Not Affect Migration, Engraftment, and Expression of Hepatocyte-Specific Genes in Immortalized Hepatocytes.

Following intrasplenic injection, a small percentage of primary hepatocytes engrafted in the spleen, whereas a larger fraction entered the portal blood flow and engrafted in the liver.38, 39 Of the 2 immortalized hepatocyte lines that we used in this study, the IGRH cells engrafted in the liver, whereas the RH69 cells engrafted mainly in the spleen, with a small minority of the cells engrafting in the liver (J. Roy-Chowdhury and I.J. Fox, unpublished data). Mechanisms underlying this difference are unknown and are being investigated in our laboratory. To determine whether the expression of AdE3 genes affected the engraftment of IGRH, IGRH-E3 hepatocytes were labeled by lentiviral transduction of the green fluorescent protein (GFP) and injected into the splenic pulp of syngeneic Gunn rats (Wistar-RHA background; n = 3; Fig. 1A). GFP-positive cells were detected in the liver after 72 hours (Fig. 1B) and were found to be stably engrafted 3 weeks later (Fig. 1C). Similarly, RH69-E3 cells labeled with a fluorescent dye, CFSE, before injection into the splenic pulp of syngeneic Lewis rat recipients were found in large numbers in the spleen 72 hours (Fig. 1D) and 2 weeks later (Fig. 1F), whereas only occasional CFSE-labeled cells were seen in the liver (Fig. 1E). Thus, AdE3 expression did not significantly alter the engraftment properties of either cell line.

Next, we compared the expression profile of albumin and the asialoglycoprotein receptor (ASGR) in RH69 and RH69-E3 cells in vitro and in vivo after engraftment in the spleen. RT-PCR on cultured RH69 and RH69-E3 cells and cells isolated from the spleen 1 week after transplantation indicated that albumin and ASGR were expressed to the same extent in the AdE3-transduced and parental cells in culture and in vivo (Fig. 1G-H). Thus, AdE3 gene expression did not adversely affect the functionality of these hepatocytes.

AdE3 Protected IGRH-E3 Hepatocytes Engrafted in Liver from Allograft Rejection and Permitted Their Proliferation.

Like primary hepatocytes, Tag-immortalized hepatocytes are MHC class I positive/MHC class II negative and evoke a similar degree of alloreactivity.40 To compare the engraftment potential and survival of the transplanted AdE3-transduced and parental hepatocytes without subjecting the host to any preparative treatment, we transplanted 20 × 106 IGRH-E3 or IGRH cells into MHC-matched (Wistar RHA) or MHC-disparate recipients (Lewis). Quantitative PCR analysis showed that 20%-25% of the injected IGRH and IGRH-E3 cells engrafted in livers (data not shown), indicating that hepatectomy is not necessary for engraftment. In the absence of any preparative treatment, the ratio of host hepatocytes to transplanted cells 1 week after transplantation was similar to that 4 weeks after transplantation in syngeneic hosts. In contrast, AdE3-transduced IGRH hepatocytes survived in allogeneic hosts, whereas the parental IGRH cells became undetectable 4 weeks after transplantation. Furthermore, survival of AdE3-transfected hepatocytes in allogeneic hosts was similar to that in syngeneic hosts in the absence of any preparative regimen.

However, although the IGRH cells engrafted without any preparative treatment of the host livers, they did not proliferate in quiescent livers. Stimulation of mitosis alone does not result in preferential proliferation of transplanted hepatocytes because the host cells proliferate to a similar extent, and the liver does not exceed its body-weight-appropriate mass.33 Preferential proliferation of the engrafted cells can be induced by inhibiting the proliferative capacity of the host hepatocytes by maneuvers, such as hepatic irradiation, followed by mitotic stimulation, for example, by partial hepatectomy.33 We applied these principles, which were established in our previous studies with primary hepatocytes,33 to promote proliferation of the immortalized hepatocytes. The recipients were subjected to 68% hepatectomy and preparative hepatic irradiation (HIR) 2 days before transplantation. We transplanted 5 × 106 IGRH-E3(3) cells (Wistar RHA haplotype) or parental IGRH cells (n = 3) into fully MHC-disparate Lewis (n = 7) or F344 (n = 3) rats. A control group included sham-operated rats without hepatocyte transplantation. RT-PCR of Tag RNA isolated from the spleens and livers of the allogeneic recipients showed that the parental IGRH cells had become undetectable 2 weeks after transplantation (Fig. 2A), indicating allorejection. In contrast, the IGRH-E3 hepatocytes were observed in the livers 9 weeks after transplantation (Fig. 2A). The caudate lobes, which were shielded during hepatic irradiation because of technical constraints, exhibited weaker signals than did the other lobes, indicating a lower level of hepatic repopulation. RT-PCR analysis of RNA from the recipient livers showed positive signals for AdE3 mRNA 9 weeks after transplantation (Fig. 2B).

Figure 2.

IGRH-E3 hepatocytes transplanted into the liver were protected from allograft rejection. (A) Tag mRNA detected by RT-PCR. (B) AdE3 mRNA detected by RT-PCR. A representative result from at least 3 animals in each group is shown. Indirect immunofluorescence staining using a Tag-specific primary mAb and Cy5-conjugated secondary Ab (red) 9 weeks after allogeneic transplantation in: (C) livers from allogeneic recipients of IGRH cells and (D) liver from a nontransplanted control. Livers from allogeneic recipients of IGRH-E3 cells: (E) ×10 and (F) ×60 magnification of the same image. Overlay of Cy5 and phase-contrast images of the same field are shown as indicated. Representative result from at least 3 animals per group shown. Real-time PCR using Tag gene as a marker to detect transplanted cells: (G) surviving engrafted IGRH and IGRH-E3 hepatocytes shown as a percent of all endogenous cells in livers of syngeneic (syn) and allogeneic (allo) recipients 9 weeks after transplantation (data show the mean of normalized Tag copies + SEMs of 3 independent experiments); (H) comparison of the distribution of IGRH-E3 hepatocytes engrafted in livers of syngeneic and allogeneic recipients 9 weeks after transplantation (each sample analyzed in quadruplicate; data show means of normalized Tag copies + SEMs of 3 independent experiments). Abbreviations: RPL, right posterior lobe; RAL, right anterior lobe; CL, caudate lobe.

Next, we used immunofluorescence staining of liver cryosections to identify the engrafted cells. Although the thermolabile Tag in the IGRH cells was mostly degraded at physiological temperatures, there was enough residual protein for detection. Recipient livers showed clusters of IGRH-E3 allografts 9 weeks after transplantation (Fig. 2E). Because transplanted hepatocytes initially integrated into liver cords as single cells, the presence of clusters indicated proliferation of the engrafted IGRH-E3 cells (Fig. 2F). In contrast, the IGRH allografts were undetectable (Fig. 2C,D).

To compare the survival of IGRH and IGRH-E3 cells in the livers of allogeneic versus syngeneic hosts, we performed quantitative real-time PCR with a sensitivity to detect as few as 2 copies of the Tag gene in genomic DNA, normalized to β-actin as an endogenous reference gene. The engrafted cells made up 0.1% of the total number of liver cells 72 hours after transplantation (data not shown). The proportion of engrafted cells increased to 1.2% for IGRH cells and 1.5% for IGRH-E3 cells in syngeneic recipients 4 weeks after transplantation, indicating preferential proliferation of the transplanted cells under preparative treatment of the host livers. Four weeks after transplantation, the IGRH cells were undetectable in allogeneic recipients. In contrast, the level of repopulation of the livers with allogeneic IGRH-E3 cells was not significantly different from that in syngeneic hosts (data not shown). Nine weeks after transplantation, the parental IGRH hepatocytes made up 2.6% of the total number of liver cells in syngeneic hosts but were undetectable in allogeneic recipients (Fig. 2G). In contrast, the engrafted IGRH-E3 cells made up 3.8% and 1.9% of the total number of liver cells in allogeneic F344 and Lewis rat hosts, respectively (Fig. 2G). These results were not significantly different from the level of repopulation found after transplantation of IGRH-E3 cells in syngeneic recipients (2.8%). Furthermore, the distribution of IGRH-E3 hepatocytes in the liver lobes of the syngeneic and allogeneic hosts was similar (Fig. 2H).

AdE3 Protected RH69-E3 Hepatocytes Engrafted in Spleens from Allograft Rejection.

To evaluate whether AdE3 could protect hepatocellular allografts in the spleen, we transplanted RH69-E3 Lewis (n = 5) or parental RH69 (n = 5) cells into fully MHC-disparate Wistar RHA rats. Immunofluorescence analysis of the recipient spleens using Tag as a marker of the transplanted cells revealed that a large fraction of the transplanted RH69-E3 hepatocytes remained viable in the allogeneic hosts 8 weeks after transplantation (Fig. 3A), whereas the parental RH69 hepatocytes were rejected (Fig. 3B,C).

Figure 3.

RH69-E3 cells transplanted into the spleen were protected from allograft rejection. Indirect immunofluorescence analysis of spleen cryosections with a Tag-specific primary mAb and a Cy3-conjugated secondary Ab (red): (A) recipients of RH69-E3 cells, (B) recipients of RH69 cells, and (C) control animal. Cy3 images (top) and corresponding phase-contrast images (bottom) displayed (×40). Representative result from at least 5 animals shown. Detection of (D) Tag mRNA, (E) AdE3 mRNA, and (F) albumin and asialoglycoprotein receptors (asgpR) by RT-PCR in Tag-positive transplanted hepatocytes dissected from the spleen of the allogeneic recipients using Lazer Capture Microscopy. Two representative results from at least 5 animals shown.

Eight weeks after transplantation, the Tag-positive cells were dissected from the spleens of allogeneic recipients using laser capture microscopy and analyzed for the presence of Tag mRNA (Fig. 3D), AdE3 mRNA (Fig. 3E), and 2 liver-specific genes, albumin and ASGR (Fig. 3F) by RT-PCR. The results confirmed that the Tag-positive cell clusters were derived from the transplanted RH69-E3 hepatocytes and that the engrafted cells continued to express AdE3 and liver-specific genes after repeated cell divisions.

These findings are consistent with our previous observations that AdE3 when expressed as a transgene abrogated autoimmunity and prevented allograft rejection of pancreatic islets in mice.26–28 Similarly, a human tumor cell line infected with Ad/E3 recombinant virus exhibited prolong survival in immunocompetent mice.34 We considered several known potential mechanisms by which AdE3 gene products could affect different steps of the host alloresponse. AdE3 encodes nonsecreted proteins that are likely to exert their protective effects by targeting processes in the cells where they are expressed. Therefore, we examined whether AdE3 expression down-regulated cell-surface MHC class I via gp19K, thereby abrogating host immune response and/or whether they down-regulated cell-surface death receptors, consequently protecting the allograft from apoptotic signaling by FasL and TNF-α.

AdE3 Expression in IGRH and RH69 Hepatocytes Did Not Alter Cell-Surface Expression of MHC Class I.

MHC class I is important in initiating the T-cell-mediated immune response against allografts, and AdE3 gp19K is known to inhibit cell-surface localization of MHC class I. However, flow-cytometric analysis did not reveal any alteration in cell-surface expression of MHC class I molecules in IGRH-E3 (Fig. 4A) or RH69-E3 cells (Fig. 4B). We found no reduction in cell-surface MHC class I even after overexpression of gp19K by infecting IGRH and RH69 cells with recombinant Ad/E3 or wild-type adenoviruses (data not shown).

Figure 4.

Expression of AdE3 gene products in donor hepatocytes did not alter cell-surface levels of MHC class I molecules (rat RT1A) in vitro but attenuated the alloreactive response in allogeneic hosts in vivo. Flow-cytometric analysis of expression of MHC class I expression on surfaces of (A) IGRH-E3 cells (filled) and IGRH cells (black line) and (B) RH69-E3 cells (filled) and RH69 cells (black line) (histogram overlay shown). Analysis of alloreactive response in vivo with representative histogram plots showing the ratio of CFSElow (Wistar RHA target) to CFSEhigh (Lewis control) spleen cells obtained from (C) naive Lewis rats, (D) Lewis recipients of IGRH allografts (Wistar RHA), and (E) Lewis recipients of IGRH-E3 allografts (Wistar RHA) 7 days after hepatocyte transplantation (A value and percent specific lysis indicated). (F) Analysis of alloreactive response in vivo showing kinetics of alloresponse in Lewis rat recipients of IGRH (open bars) and IGRH-E3 (filled bars) allografts. Values represent means ± SDs of specific lysis of target cells in 3 rats per group.

The apparent discrepancy between our findings and previous studies describing gp19K-dependent down-regulation of MHC class I12, 13 could have resulted from the difference in the binding affinity of gp19K to MHC class I heavy chains among different animal strains and human cells of different HLA types.41

AdE3 Gene Expression in Donor Cells Attenuated But Did Not Abolish Host Alloresponse.

To evaluate the effect of AdE3 gene expression in the donor hepatocytes on the induction of alloreactive cytotoxic responses in vivo, we performed a modified in vivo cytotoxicity assay. A mixture of CFSE-labeled allogeneic (Wistar RHA) and syngeneic (Lewis) splenocytes were adoptively transferred into Lewis rats that had received IGRH or IGRH-E3 allografts (Wistar RHA strain) at various times before adoptive transfer. The presence of donor-specific alloreactive cell-mediated cytotoxicity evoked by the transplanted hepatocellular allografts should result in lysis of subsequently transferred splenocytes from the allogeneic but not the syngeneic donor strain. Nearly all transferred splenocytes of the Wistar RHA strain were eliminated in Lewis rats that had received IGRH allografts 7 days before the adoptive transfer, indicating Wistar RHA-specific cytotoxic alloreactivity (Fig. 4D). As expected, the alloresponse was undetectable in control-naive Lewis rats (Fig. 4C), indicating that the cytotoxicity measured in recipients of IGRH hepatocytes was allospecific. Cell-mediated cytotoxic activity in the recipients of IGRH allografts was vigorous, with 81.5% and 51.8% lysis 7 and 14 days after transplantation, respectively (Fig. 4F). The alloreactive cytotoxic activity in the recipients of IGRH-E3 allografts was significantly lower, with 39.5% and 16.7% lysis 7 and 14 days after transplantation, respectively (Fig. 4E-F).

The lack of down-regulation of cell-surface MHC class I by AdE3 expression and the incomplete attenuation of in vivo cytotoxic activity indicate that although gp19K may have potentially exerted some protective effect in vivo, allograft survival did not seem to have resulted from blocking allo-MHC class I recognition in our immortalized hepatocytes. Consistent with this, we previously observed that pancreatic β cells from gp19K-deleted AdE3-transgenic mice were still protected from destruction by host autoimmune responses.42

AdE3-Mediated Protection from Allograft Rejection Correlated with Loss of Cell-Surface Fas Receptors in Transplanted Cells and Inhibition of Fas-Mediated Apoptosis.

Triggering of apoptosis through binding of FasL,10 which is present in alloreactive effector immune cells, with Fas, which is expressed constitutively on surfaces of hepatocytes, is critical in hepatocyte allograft rejection. In human liver transplant recipients, hepatocyte apoptosis correlates with Fas receptor up-regulation.7 In animal studies, blockade of Fas-FasL interaction by mAb or siRNA increases allograft survival of hepatocytes in the spleen.8, 9 Also, the protective effect of tacrolimus is correlated with its ability to inhibit Fas-induced apoptosis in human hepatocytes.6 Therefore, in view of the known inhibition of FasL-mediated killing by the AdE3-RIDα/β complex, we evaluated the effect of AdE3 expression on cell-surface localization of Fas in cultured cells and in vivo after engraftment, as well as on Fas-mediated apoptosis in vitro. Flow-cytometric analysis showed cell-surface Fas expression in RH69 and IGRH cells, whereas cell-surface Fas was markedly down-regulated in both RH69-E3 and IGRH-E3 hepatocytes (Fig. 5A). Immunofluorescence analysis revealed that in parental RH69 and IGRH hepatocytes, the cellular pool of Fas extended to the surfaces of the cells, whereas in RH69-E3 and IGRH-E3 cells, Fas was largely limited to intracellular locations (Fig. 5B). The AdE3 transfection did not affect the total Fas protein in the cells, as determined by Western blot (data not shown). These results suggested that AdE3 proteins did not inhibit Fas synthesis but prevented its localization to plasma membranes.

Figure 5.

Expression of AdE3 genes in donor hepatocytes inhibited cell-surface expression of Fas receptor. (A) Flow-cytometric analysis of cell-surface levels of Fas in AdE3-transduced RH69 and IGRH hepatocytes and the corresponding parental cell lines. The data are representative of 2 independent experiments. (B) Analysis of Fas cellular distribution in hepatocytes grown in culture by indirect immunofluorescence staining with Fas-specific Ab and Cy3-conjugated secondary Ab (red). Cy3 and phase-contrast images of the corresponding fields are shown as indicated (×60). (C) Modulation of cell-surface levels of Fas receptor in AdE3-transduced hepatocytes 5 days after engraftment in the spleen. CFSE-labeled RH69 or RH69-E3 cells, recovered from spleens of recipient rats 5 days after transplantation, were stained with anti-Fas-PE and analyzed by 2-color flow cytometry. Plots are gated on CFSE-positive (transplanted) cells. Overlay of the histograms is shown. The data are representative of 2 independent experiments. (D) Detection of Fas-induced apoptosis in culture AdE3-transduced and parental RH69 and IGRH hepatocytes treated with agonistic Fas antibody, stained with Annexin V-PE and 7-ADD, and analyzed by 2-color flow cytometry. Plots are gated on 7-ADD–negative cells. Data are representative of 3 experiments.

To evaluate the effect of AdE3 gene expression on cell-surface localization of Fas in vivo, we labeled RH69 and RH69-E3 cells with a fluorescent dye, CFSE, before injection into the splenic pulp of syngeneic recipients. Single-cell suspensions of the spleens of recipient rats 5 days after engraftment were subjected to 2-color flow cytometry to identify the CSFE-labeled transplanted cells (green) and measure cell-surface Fas expression (red). The results showed that the cell-surface Fas in the engrafted RH69-E3 cells was markedly down-regulated compared with that in the parental RH69 cells (Fig. 5C). The difference was more pronounced in vivo than in the cultured cells (Fig. 5A and C).

To determine whether the loss of cell-surface Fas could protect the AdE3-transduced hepatocytes from Fas-induced apoptosis, cultured IGRH-E3 and RH69-E3 cells and their corresponding parental cell lines were treated with a Fas agonistic antibody, IPO-4 (GeneTex), for 16 hours. Flow-cytometric analysis of Annexin V expression and internalization of a vital dye, 7-AAD, revealed that after IPO-4 treatment, large fractions of the IGRH and RH69 cells were undergoing apoptosis (Annexin V-positive and 7-ADD-negative; Fig. 5D). In contrast, the corresponding AdE3-transduced hepatocytes remained viable and remarkably refractory to Fas-induced apoptosis (Annexin V-negative and 7-ADD-negative).

These findings suggest that depletion of cell-surface Fas and its retention in hepatocytes by RIDα/β expression are important in AdE3-mediated protection of allografted hepatocytes from FasL-mediated killing.

AdE3 Gene Products Inhibited TNF-α/TNFR1 Apoptotic Signaling.

TNF-α/TNFR1 signaling is another known mediator of apoptosis during liver transplant rejection.43 AdE3-RIDα/β is known to down-regulate cell-surface TNF-R1.20, 21 In addition, 14.7K interferes with down-stream TNF-α signaling events.11 In the present study, we examined the effect of AdE3 gene expression on the proapoptotic branch (c-Jun NH2-terminal kinase [JNK]/AP-1) of the TNF-α/TNF-R1 signaling pathway,44 specifically by determining TNF-α-induced phophorylation of c-Jun in the hepatocyte lines in vitro. Western blot analysis revealed TNF-α-mediated phosphorylation of c-Jun in the IGRH and RH69 hepatocytes, but not in the IGRH-E3 and RH69-E3 hepatocytes (data not shown), indicating that AdE3 gene products inhibited the apoptotic JNK/AP-1 pathway.

Our results provide a “proof of principle” that expression of AdE3 genes in donor hepatocytes out of context with other adenoviral genes protects these cells from immune rejection after engraftment in the livers or spleens of allogeneic recipients. Together, the data indicate that the AdE3-mediated protection against the effector mechanisms of the host alloresponses is sufficient to prolong allograft survival despite immune recognition of the transplanted cells. In addition, AdE3 gene products are known to inhibit expression of inflammatory chemokines, such as IP10, MCP-1, and IL-8, which may attenuate the recruitment of inflammatory cells.11 Because the AdE3 gene products are not secreted, their expression in donor hepatocytes should prevent allograft rejection without interfering with the host immune system, providing a rationale for potential clinical applications of AdE3 in allogeneic hepatocyte transplantation. For this purpose, we envision extending these studies to the transduction of primary hepatocytes with AdE3 genes, using integrating gene transfer vectors.


Dedicated to the memory of Marshall S. Horwitz. We thank Dr. Teresa DiLorenzo and Dr. Steven Porcelli for their helpful discussions.