Lower Vector Doses Research Articles

343. Characterization of Delivery Routes of AAV Vectors for Neonatal Gene Transfer

Gene transfer into a fetus or neonate is considered to be a fundamental approach to the treatment of genetic diseases, which cause irreversible manifestations in adulthood, and is potentially advantageous in transducing tissues with relatively low vector doses. Because of its safety and long-term transgene expression, adeno-associated virus (AAV) vector is one of the promising vehicles for this purpose. Although the potential utility has been suggested, systematic comparison of administration routes of AAV vectors has not been fully investigated, especially for neonates. In this study, we explored the optimal delivery route of AAV vectors for neonatal gene therapy in mice. For this purpose, vectors carrying human coagulation factor IX (hFIX) gene under cytomegalovirus (CMV)-derived promoter using either AAV1 or AAV5 capsids were administered into immune-competent neonatal mice (C57BL/6) within 24 hours after birth. As for delivery routes, injections into muscles of lower limbs, a cervical vein, and a peritoneal cavity were tested. Based on our previous observations, AAV1-CMV-hFIX vectors were used for intramuscular injection and AAV5-CMV-hFIX vectors for intravenous and intraperitoneal cavity injection. Each neonate mouse received a dose of 1 × 10(10) genome copy per body. Plasma concentrations of hFIX were periodically monitored after the vector injection. As a result, intramuscular injection showed the highest hFIX concentration (142.0 ± 67.5 ng/ml), followed by intraperitoneal cavity injection (21.8 ± 5.8 ng/ml) and intravenous injections (10.7 ± 1.5 ng/ml). Dose escalation study for intramuscular injections (5 × 10(10) genome copy of AAV1-CMV-hFIX) showed a higher hFIX concentration (243.2 ± 72.6 ng/ml), a level that is considered within therapeutic range, and the expression lasted for more than 24 weeks. Additionally, we examined the difference of transduction efficiency by sex, because liver-targeted transduction with AAV vectors is known to be inefficient for adult female mice. To validate the sex-related difference in gene expression in neonate and adult mice, each mouse was injected with 5 × 10(10) genome copy of AAV1-CMV-hFIX vectors intramuscularly or 3 × 10(11) genome copy of AAV5-CMV-hFIX vectors into peritoneal cavity. AAV5-CMV-LacZ vectors were also used for evaluating gene expression throughout the body. Our study revealed that the sex-related difference in transgene expression was observed for intraperitoneal cavity injection into adult mice but not into neonate mice 10 weeks following injection (male vs. female: 98.3 ± 24.8 vs. 10.8 ± 3.8 ng/ml in adult, 214.4 ± 42.6 vs. 178.2 ± 38.9 ng/ml in neonate, respectively). Interestingly, X-gal-staining after intraperitoneal cavity injection of AAV5-CMV-LacZ vectors showed that the sex-related difference in transgene expression was apparent in adult peritoneal epithelium, and that such a difference was not observed in groups with neonatal injection. These findings would be valuable for designing strategies of neonatal gene therapy in terms of delivery routes of AAV vectors.

868. Enhancement of Gene Therapy Approaches for the Correction of Farber Disease

Glycogen Storage Disease type II (GSD-II) results in the accumulation of glycogen within the lysosomes of primarily the cardiac and skeletal muscles. Functional inhibitions similar to those of muscular dystrophies are common in GSD-II patients. However the most detrimental form of GSD-II affects infants, who are completely lacking muscle expression of the glycogen reducing enzyme, acid-a-glucosidase (human GAA). Most affected infants die within the first year of life due to cardio-respiratory failure, while juvenile and adult forms will more likely progress to skeletal muscle inadequacy. Recent enzyme replacement therapy (ERT) with hGAA has prolonged the life span of affected infants, but long term care and large-scale production of the enzyme remain realistic caveats of ERT. Developments within our laboratory show the promise of adenovirus mediated gene transfer of the hGAA gene into liver tissues, and its ability for profuse and long-term expression of hGAA protein. Various improvements of the adenovirus including the removal of viral genes necessary for replication have yielded high potency and minimal immune responses when injected into GAA knockout (KO) murine models. Based upon these previous studies, we hypothesized that advanced generation, fully deleted adenovirus based vectors would have an even more decreased propensity to generate antivirus and/or transgene immune responses. We have therefore constructed a fully deleted adenovirus capable of expressing high levels of the hGAA gene, facilitated via use of a liver specific enhancer, promoter and other gene expression elements. Intravenous delivery of this vector into GAA knockout mice has produced long term hGAA protein secretion from the liver, (currently up to 56 days post injection). Longer term studies are currently in progress. Interestingly, at lower vector doses, antibody titers to hGAA are detectable as measured by ELISA, but at much higher vector doses anti-hGAA antibody no longer becomes detectable. This seeming paradox may be due to the fact that the large amounts of hGAA being secreted into the plasma from the liver of animals injected with higher doses of the vector, “sequesters” the low amounts of anti-hGAA antibody being produced within the same animals. We are currently investigating whether the low anti-hGAA antibody titers are correlated to specific properties of the fully deleted adenovirus vector system per se, and/or the various constituents utilized to induce hGAA expression from the vector. Furthermore, we will assess how the efficacy of the fully deleted Ad vectors might be impacted upon by the diminished immune responses afforded by their use. In summary, this unique gene transfer system may be a viable method for gene therapy of GSD-II, and perhaps may lead to an alternative or adjunctive therapy relative to current, ERT based strategies to treat GSD-II.

Tumor necrosis factor inhibitor gene transfer ameliorates lung graft ischemia-reperfusion injury

Objective Tumor necrosis factor is an important mediator of lung transplant ischemia-reperfusion injury, and soluble type I tumor necrosis factor receptor binds to tumor necrosis factor and works as a tumor necrosis factor inhibitor. The objectives of this study were to demonstrate that gene transfer of type I tumor necrosis factor receptor–IgG fusion protein reduces lung isograft ischemia-reperfusion injury and to compare donor endobronchial versus recipient intramuscular transfection strategies. Methods Three donor groups of Fischer rats (n = 6/group) underwent endobronchial transfection with either saline, 2 × 10 7 plaque-forming units of control adenovirus encoding β-galactosidase, or 2 × 10 7 plaque-forming units of adenovirus encoding type I tumor necrosis factor receptor–IgG fusion protein. Left lungs were harvested 24 hours later. Two recipient groups (n = 6/group) underwent intramuscular transfection with 2 × 10 7 plaque-forming units or 1 × 10 10 plaque-forming units of adenovirus encoding type I tumor necrosis factor receptor–IgG fusion protein 24 hours before transplantation. All donor lung grafts were stored for 18 hours before orthotopic lung transplantation. Graft function was assessed 24 hours after reperfusion. Transgene expression was evaluated by means of enzyme-linked immunosorbent assay and immunohistochemistry of type I tumor necrosis factor receptor. Results Endobronchial transfection of donor lung grafts with 2 × 10 7 plaque-forming units of adenovirus encoding type I tumor necrosis factor receptor–IgG fusion protein significantly improved arterial oxygenation compared with the saline and β-galactosidase donor groups (366.6 ± 137.9 vs 138.8 ± 159.9 and 140.6 ± 131.4 mm Hg, P = .009 and .010, respectively). Recipient intramuscular transfection with 1 × 10 10 plaque-forming units of adenovirus encoding type I tumor necrosis factor receptor–IgG fusion protein improved lung graft oxygenation compared with that seen in the low-dose intramuscular group (2 × 10 7; 320.3 ± 188.6 vs 143.6 ± 20.2 mm Hg, P = .038). Type I tumor necrosis factor receptor–IgG fusion protein was expressed in endobronchial transfected grafts. In addition, intramuscular type I tumor necrosis factor receptor–IgG fusion protein expression was dose dependent. Conclusions Donor endobronchial and recipient intramuscular adenovirus-mediated gene transfer of type I tumor necrosis factor receptor–IgG fusion protein improved experimental lung graft oxygenation after prolonged ischemia. However, donor endobronchial transfection required 500-fold less vector. Furthermore, at low vector doses, it does not create significant graft inflammation.

Widespread distribution of beta-hexosaminidase activity in the brain of a Sandhoff mouse model after coinjection of adenoviral vector and mannitol.

Sandhoff disease is a severe inherited neurodegenerative disorder resulting from deficiency of the beta-subunit of hexosaminidases A and B, lysosomal hydrolases involved in the degradation of G(M2) ganglioside and related metabolites. Currently, there is no viable treatment for the disease. Here, we show that adenovirus-mediated transfer of the beta-subunit of beta-hexosaminidase restored Hex A and Hex B activity after infection of Sandhoff fibroblasts. Gene transfer following intracerebral injection in a murine model of Sandhoff disease resulted in near-normal level of enzymatic activity in the entire brain at the different doses tested. The addition of hyperosmotic concentrations of mannitol to the adenoviral vector resulted in an enhancement of vector diffusion in the injected hemisphere. Adenoviral-induced lesions were found in brains injected with a high dose of the vector, but were not detected in brains injected with 100-fold lower doses, even in the presence of mannitol. Our data underline the advantage of the adjunction of mannitol to low doses of the adenoviral vector, allowing a high and diffuse transduction efficiency without viral cytotoxicity.

IFN-gamma sensitization of prostate cancer cells to Fas-mediated death: a gene therapy approach.

While human prostate cancers and cell lines express Fas, most of these cell lines are resistant to Fas-mediated death. In the present studies we addressed the ability of IFN-gamma to influence Fas-mediated cell death in prostate cancer cells. In vitro exposure of the human cell lines LNCaP and PC3 and the mouse cell line RM-1 to agonist anti-Fas antibody and/or soluble Fas ligand resulted in killing of only PC3 cells. However, preincubation with IFN-gamma resulted in synergistic killing in all three cell lines. In vitro treatment of RM-1 with a replication-incompetent adenovirus expressing mouse FasL (Ad.FasL) resulted in maximal cell kill near 40%, which correlated with baseline Fas expression. The addition of IFN-gamma enhanced cell kill to a degree consistent with the resulting higher levels of Fas and maintained synergistic killing at very low doses of vector. Co-inoculation of orthotopic RM-1 primary tumors with Ad.mFasL and an adenovirus expressing mouse IL-12 (Ad.mIL-12) to drive host production of IFN-gamma negated the survival advantage of Ad.mIL-12 alone. However, the staggered injection of Ad.mIL-12 and Ad.FasL achieved almost threefold higher levels of apoptosis in primary tumor tissue and doubled median survival. Therefore, IFN-gamma is capable of bestowing increased sensitivity to Fas-mediated cell death in prostate cancer cells and, in a gene therapy approach, may define a powerful tool to treat prostate cancers.

Selective ablation of human cancer cells by telomerase-specific adenoviral suicide gene therapy vectors expressing bacterial nitroreductase.

Reactivation of telomerase maintains telomere function and is considered critical to immortalization in most human cancer cells. Elevation of telomerase expression in cancer cells is highly specific: transcription of both RNA (hTR) and protein (hTERT) components is strongly upregulated in cancer cells relative to normal cells. Therefore, telomerase promoters may be useful in cancer gene therapy by selectively expressing suicide genes in cancer cells and not normal cells. One example of suicide gene therapy is the bacterial nitroreductase (NTR) gene, which bioactivates the prodrug CB1954 into an active cytotoxic alkylating agent. We describe construction of adenovirus vectors harbouring the bacterial NTR gene under control of the hTR or hTERT promoters. Western blot analysis of NTR expression in normal and cancer cells infected with adenoviral vectors showed cancer cell-specific nitroreductase expression. Infection with adenoviral telomerase-NTR constructs in a panel of seven cancer cell lines resulted in up to 18-fold sensitization to the prodrug CB1954, an effect that was retained in two drug-resistant ovarian lines. Importantly, no sensitization was observed with either promoter in any of the four normal cell strains. Finally, an efficacious effect was observed in cervical and ovarian xenograft models following single intratumoural injection with low doses of vector, followed by injection with CB1954.

The magnetofection method: using magnetic force to enhance gene delivery.

In order to enhance and target gene delivery we have previously established a novel method, termed magnetofection, which uses magnetic force acting on gene vectors that are associated with magnetic particles. Here we review the benefits, the mechanism and the potential of the method with regard to overcoming physical limitations to gene delivery. Magnetic particle chemistry and physics are discussed, followed by a detailed presentation of vector formulation and optimization work. While magnetofection does not necessarily improve the overall performance of any given standard gene transfer method in vitro, its major potential lies in the extraordinarily rapid and efficient transfection at low vector doses and the possibility of remotely controlled vector targeting in vivo.

Adenoviral-Induced Islet Cell Cytotoxicity Is Not Counteracted by Bcl-2 Overexpression

The ability to transfer immunoregulatory, cytoprotective, or anti-apoptotic genes into pancreatic islet cells may allow enhanced resistance against the autoimmune destruction of these cells in type 1 diabetes. We describe here an inducible transduction system for expression of the anti-apoptotic bcl-2 gene in insulin-producing cells as a potential tool for protecting against beta-cell death. Isolated pancreatic rat islet cells or rat insulinoma (RINm5F) cells were transduced using a progesterone antagonist (RU 486) inducible adenoviral vector system, expressing the bcl-2 gene. Bcl-2 overexpression was measured by Western blot assays and flow cytometry analysis. Following exposure to cytokines or to the mitochondrial uncoupler FCCP, cell survival was determined using fluorescence and electron microscopy, and a colorimetric assay (2,3-bis[2-methoxy-4-nitro-5-sulfophenyl]- 2H-tetrazolium-5-carboxanilide [XTT]-based) for cell viability. The mitochondrial membrane potential ((m)) was assessed using the lipophilic cationic membrane potential-sensitive dye JC-1. The adenoviral gene transfer system induced Bcl-2 expression in more than 70% of beta-cells and the protein expression levels were successfully regulated in response to varying concentrations of progesterone antagonist RU 486. Exposure of islet cells to proinflammatory cytokines IL-1beta, TNF-alpha, and IFN-gamma, or to the mitochondrial uncoupler FCCP resulted in disruption of the mitochondrial membrane potential ((m)) and beta-cell death. Bcl-2 overexpression stabilized (m) and prevented cell death in RINm5F cells but not in islet cells. In addition, prolonged in vitro culture revealed adenoviral-induced islet cell necrosis. The RU 486-regulated adenoviral system can achieve an efficient control of gene transfer at relatively low doses of the adenoviral vector. However, Bcl-2 overexpression in islet cells did not prevent adenoviral- or cytokine-induced toxicity, suggesting that the specific death pathway involved in adenoviral toxicity in beta-cells may bypass the mitochondrial permeability transition event.

Ectopic osteogenesis using adenoviral bone morphogenetic protein (BMP)-4 and BMP-6 gene transfer.

Bone morphogenetic proteins (BMPs) delivered on scaffolds can induce ectopic bone formation after subcutaneous injection. Adenoviral vectors (Ad) carrying BMP2, BMP7, and BMP9 cDNAs have been shown to produce bone through endochondral ossification. The present study was performed to elucidate the histological events leading to ectopic ossification for two novel first-generation adenoviral constructs encoding BMPs, AdBMP4 and AdBMP6. In vitro, the viral constructs produced and secreted the mature BMP4 and BMP6 proteins. In vivo, the calf muscles of athymic nude rats were injected with AdBMP4, AdBMP6, AdBMP2, or AdlacZ. Rats were sacrificed 3, 6, 9, 16, 21, 60, and 90 days postinjection. Whereas AdBMP4 produced ectopic bone through mechanisms similar to endochondral ossification, AdBMP6 seemed to induce bone by way of mechanisms similar to both intramembranous and endochondral ossification pathways. At the relatively low vector dose used in this study, AdBMP2 caused an initial recruitment of primitive mesenchymal cells, without further development to bone. From computed tomographic analysis, AdBMP6 produced the most rapid tissue calcification. The ultimate density of ectopic bone formed by AdBMP4 and AdBMP6 was comparable. The current study demonstrates that AdBMP4 and AdBMP6 are more potent than the prototypical osteogenic adenoviral vector AdBMP2 and seem to induce ectopic bone by different mechanisms.

Nonviral vector loaded collagen sponges for sustained gene delivery in vitro and in vivo.

Naked DNA and standard vectors have previously been used for gene delivery from implantable carrier matrices with great potential for gene therapeutic assistance of wound healing or tissue engineering. We have previously developed copolymer-protected gene vectors which are inert towards opsonization. Here we examine their potency in carrier-mediated gene delivery in comparison to standard vectors using a vector-loaded collagen sponge model. Equine collagen type I sponges were loaded by a lyophilization method with naked DNA, polyethylenimine (PEI)-DNA, DOTAP/cholesterol-DNA and copolymer-protected PEI-DNA. These preparations were characterized in terms of vector-release, cell growth on the matrices and reporter gene expression by cells colonizing the sponges in vitro and in vivo. Subcutaneous implantation of sponges in rats served as an in vivo model. At the chosen low vector dose, the loading efficiency was at least 86%. Naked DNA-loaded collagen matrices lost 77% of the DNA dose in an initial burst in aqueous buffer in vitro. The other preparations examined displayed a sustained vector release. There was no difference in cell growth and invasion of the sponges between vector-loaded and untreated collagen grafts. Reporter gene expression from cells colonizing the sponges in vitro was observed for not more than 7 days with naked DNA, whereas the lipoplex and polyplex preparations yielded long-term expression throughout the experimental period of up to 56 days. The highest expression levels were achieved with the PEI-DNA-PROCOP (protective copolymer) formulation. Upon subcutaneous implantation in rats, no luciferase expression was detected with naked DNA preparations. DOTAP/cholesterol-DNA and PEI-DNA-loaded implants lead to reporter gene expression for at least 3 days, but with poor reproducibility. PEI-DNA-PROCOP collagen matrices yielded consistently the highest reporter gene expression levels for at least 7 days with good reproducibility. With the preparation method chosen, lipoplex- and polyplex-loaded collagen sponges are superior in mediating sustained gene delivery in vitro and local transfection in vivo as compared to naked DNA-loaded sponges. Protective copolymers are particularly advantageous in promoting the tranfection capacity of polyplex-loaded sponges upon subcutaneous implantation, likely due to their stabilizing and opsonization-inhibiting properties.

A novel bystander effect involving tumor cell-derived Fas and FasL interactions following Ad.HSV-tk and Ad.mIL-12 gene therapies in experimental prostate cancer

To enhance the NK population induced by Herpes Simplex virus thymidine kinase (HSV-tk) gene transduction and ganciclovir (GCV) treatment, adenovirus-mediated (Ad) expression of IL-12 was added to Ad.HSV-tk + GCV as combination gene therapy. This approach resulted in improved local and systemic growth suppression in a metastatic model of mouse prostate cancer (RM-1). In vitro assay of tumor infiltrating lymphocytes noted superior lysis of both RM-1 and Yac-1 targets with combination therapy, but in vivo depletion of NK cells only negatively impacted on systemic growth inhibition. TUNEL assay of primary tumors noted induction of apoptosis between two and four times higher than controls lasting for 6-8 days post-vector injection. After demonstrating that Ad.HSV-tk/GCV and Ad.mIL-12-induced IFN-gamma independently up-regulated expression of FasL and Fas, respectively, studies examined tumor cell-mediated death through Fas/FasL-induced apoptosis as a mechanism of primary tumor growth suppression. In vitro, combination therapy at low vector doses resulted in synergistic growth suppression, which could be negated by the addition of anti-FasL antibody. In vivo co-inoculation of an adenovirus expressing soluble Fas resulted in combination therapy-treated tumors, which were three times larger than expected, and a reduction in apoptosis to baseline levels. In FasL knockout mice, combination therapy maintained the superior results experienced in wild-type mice, indicating that tumor cell, not host cell FasL, was responsible for Fas transactivation. Therefore, the combination of Ad.HSV-tk/GCV + Ad.mIL-12 results in enhanced local growth control via apoptosis due to tumor cell expression of Fas and FasL and improved anti-metastatic activity secondary to a strong NK response.

Lethal toxicity, severe endothelial injury, and a threshold effect with high doses of an adenoviral vector in baboons.

The effects of intravenous administration of a first-generation adenoviral vector expressing beta-galactosidase were compared in two baboons receiving a high dose or lower dose of vector, 1.2 x 10(13) or 1.2 x 10(12) particles/kg, respectively. The high-dose baboon developed acute symptoms, decreased platelet counts, and increased liver enzymes, and became moribund at 48 hr after injection, while the lower-dose baboon developed no symptoms. Expression of the beta-galactosidase transgene was prominent in liver, spleen, and endothelium of the arterial vasculature in the high-dose baboon, but was much more limited and spared the endothelium in the lower-dose baboon. Injury to the vascular endothelium was the most prominent abnormality in the high-dose baboon. Extensive histological studies provide a detailed picture of the pathology associated with a lethal dose of first-generation adenoviral vector in a primate.

Intratumoral administration of low doses of an adenovirus vector encoding tumor necrosis factor alpha together with naive dendritic cells elicits significant suppression of tumor growth without toxicity.

Although tumor necrosis factor alpha (TNF-alpha) is a potent cytokine with a myriad of innate immune antitumor properties, systemic administration of TNF-alpha is associated with significant toxicity, limiting the use of the TNF-alpha protein as an antitumor therapeutic. On the basis of the knowledge that dendritic cells (DCs) play a central role in initiating antitumor adaptive immune responses, we hypothesized that intratumoral administration of low doses of an adenovirus encoding TNF-alpha (AdTNF-alpha) together with syngeneic DCs would act synergistically to suppress preexisting tumors. As a model, four different tumor cell lines, all resistant in vitro to the TNF-alpha protein, were implanted in syngeneic mice, and established tumors received intratumor AdTNF-alpha alone or in combination with DCs. At high doses (10(9) PFU), AdTNF-alpha alone suppressed tumor growth, but was associated with systemic toxicity. A 100-fold lower AdTNF-alpha concentration (10(7) PFU) or high doses of the control vector AdNull had no systemic toxicity, but also minimal suppression of tumor growth. In contrast, local administration of the low dose (10(7) PFU) of AdTNF-alpha in combination with syngeneic DCs (AdTNF-alpha + DCs) elicited marked tumor suppression without toxicity. Administration of AdTNF-alpha + DCs into tumors elicited tumor-specific cytotoxic T cells and protected animals against subsequent challenge with the same tumor, suggesting that AdTNF-alpha + DC therapy induced tumor-specific adaptive immune host responses. Consistent with this concept, studies with syngeneic knockout mice showed that MHC class I molecules on DCs as well as CD8(+) T cells were necessary for the antitumor effect of intratumor AdTNF-alpha + DCs. These data demonstrate that the combination of intratumoral administration of the TNF-alpha cDNA together with naive DCs can evoke tumor suppression without systemic toxicity, providing a new paradigm for the use of TNF-alpha as antitumor therapy.

Improved muscle-derived expression of human coagulation factor IX from a skeletal actin/CMV hybrid enhancer/promoter

Hemophilia B is caused by the absence of functional coagulation factor IX (F.IX) and represents an important model for treatment of genetic diseases by gene therapy. Recent studies have shown that intramuscular injection of an adeno-associated viral (AAV) vector into mice and hemophilia B dogs results in vector dose-dependent, long-term expression of biologically active F.IX at therapeutic levels. In this study, we demonstrate that levels of expression of approximately 300 ng/mL (6% of normal human F.IX levels) can be reached by intramuscular injection of mice using a 2- to 4-fold lower vector dose (1 x 10(11) vector genomes/mouse, injected into 4 intramuscular sites) than previously described. This was accomplished through the use of an improved expression cassette that uses the cytomegalovirus (CMV) immediate early enhancer/promoter in combination with a 1.2-kilobase portion of human skeletal actin promoter. These results correlated with enhanced levels of F.IX transcript and secreted F.IX protein in transduced murine C2C12 myotubes. Systemic F.IX expression from constructs containing the CMV enhancer/promoter alone was 120 to 200 ng/mL in mice injected with 1 x 10(11) vector genomes. Muscle-specific promoters performed poorly for F.IX transgene expression in vitro and in vivo. However, the incorporation of a sequence from the alpha-skeletal actin promoter containing at least 1 muscle-specific enhancer and 1 enhancer-like element further improved muscle-derived expression of F.IX from a CMV enhancer/promoter-driven expression cassette over previously published results. These findings will allow the design of a clinical protocol for therapeutic levels of F.IX expression with lower vector doses, thus enhancing efficacy and safety of the protocol. (Blood. 2000;95:2536-2542)

Induction of human immunodeficiency virus type 1 (HIV-1)-specific cytolytic T lymphocyte responses in seronegative adults by a nonreplicating, host-range-restricted canarypox vector (ALVAC) carrying the HIV-1MN env gene.

CD8+ cytolytic T lymphocytes (CTL) are likely to be an important component of effective vaccines against human immunodeficiency virus type 1 (HIV-1). CTL can be induced most effectively with live virus vectors. However, because of concerns about the safety of such vectors, a nonreplicating canarypox vector (ALVAC) capable of expressing foreign genes in mammalian cells has been developed. This study evaluated the capacity of an ALVAC vector expressing the HIV-1MN envelope (env) glycoprotein to induce HIV-1-specific CTL in seronegative volunteers. Protocols were designed to determine whether immunization with ALVAC alone or in combination with subunit boosting could induce CTL in vaccinia-immune and -naive volunteers. A simple method for antigen-specific in vitro stimulation was used to detect CTL responses in HIV-1-seronegative vaccine recipients. The results indicate that low doses of a nonreplicating virus vector alone can elicit both CD4+ and CD8+ HIV-1-specific CTL in a subset of seronegative volunteers.