¹Immunotherapy and Gene Therapy Unit, Istituto Nazionale Tumori, Milano, Italy

²Vaccine Research Group, Division of Microbiology, GBF-German Research Centre for Biotechnology, Braunschweig, Germany

Correspondence to: P Paglia, Immunotherapy and Gene Therapy Unit, Department of Experimental Oncology, Istituto Nazionale Tumori, Via Venezian 1, 20133 Milano, Italy

^aPresent address: Department of Life Sciences, UEL, London, UK

Macrophages are normal targets for Salmonella during natural infections, and it has been demonstrated that attenuated bacteria can deliver nucleic acid vaccine constructs. Therefore, we assessed if attenuated Salmonella can be used for the in vivo delivery of transgenes to their natural cellular target, in an attempt to correct genetic defects associated with monocytes/macrophages. This system would offer the distinct advantage of achieving a specific targeting of defective cells in a non-invasive form. Using a reporter gene, we demonstrated that attenuated Salmonella could be used as an effective in vitro delivery system to transfer genetic material into nondividing cells like murine macrophages. In vivo, the oral administration of attenuated Salmonella allows targeted delivery of transgenes to macrophages and subsequently expression of transgenes at a systemic level. IFN-deficient mice (GKO) were thus selected as a model for the in vivo validation of the Salmonella-based delivery approach. Attenuated Salmonella, used as the carrier for a eukaryotic expression vector encoding the murine IFN gene, was able to restore the production of this cytokine in GKO macrophages. Their oral administration to IFN-deficient mice also re-established, in these immunocompromised animals, the natural resistance to bacterial infections. These results demonstrate, for the first time, that attenuated Salmonella can be successfully used in vivo as a DNA delivery system for the correction of a genetic defect associated with monocyte/macrophages. Gene Therapy (2000) 7, 1725-1730.

gene therapy; inherited disease; Salmonella; IFN; GKO mice

Introduction

The major alternative to traditional therapeutic interventions of phagocyte-associated inherited diseases is probably gene therapy.^1,2,3 It has been shown that bacteria, which have evolved the ability to enter mammalian cells, are able to transfer genetic material to host cells without major DNA rearrangements occurring during the trans-kingdom transfer process.^4,5,6 Attenuated Salmonella strains have been successfully used as a carrier system for the in vivo delivery of nucleic acid-based vaccines.⁷ This approach has proven to be efficacious in protecting against infectious diseases and for cancer immunotherapy.^8,9,10 Macrophages present at the bacterial portal of entry, the Peyer's patches, constitute not only the local cellular targets for Salmonella, but also the major in vivo bacterial 'reservoir' following their systemic dissemination.¹¹ For these reasons, macrophages can be considered an optimal target for a Salmonella-based gene therapy.

The basic challenge in gene therapy is to develop approaches for delivering genetic material to the appropriate cells of the patient in a way that is tissue/cell specific, efficient and safe. Vehicles, which encapsulate therapeutic genes for delivery to cells, are called vectors. Many of the vectors currently in use are based on attenuated or modified viruses, or synthetic vectors in which complexes of DNA, proteins and/or lipids are formed in particles. Tissue-specific vectors have only been partially obtained, by using carriers that in nature infect certain cell types, such as herpes virus does for cells of the nervous system or adenoviruses, which demonstrate tropism for lung mucosae.^12,13 The use of bacterial carriers fulfills the requirements of a natural, attenuated, infective agent working as a Trojan horse for a targeted gene delivery. This approach seems particularly promising for gene therapy of diseases in which affected tissues are the natural targets of these bacteria (eg monocytes, macrophages, hepatocytes and intestinal epithelia). Low-cost, non-invasive administration, and life-long suitability of the therapy, combined with easy preparation, storage and transport of the carrier, additionally support the attempt of using bacteria for gene therapy.

A number of inherited diseases including lysosomal storage disorders (LSD), hemochromatosis and immune functional disorders primarily affect monocyte/ macrophages.^14,15 The possibility of transducing these cells at high efficiency in vitro allows the study of the pathophysiology of such diseases, whereas the use of appropriate knock-out mice are useful to test functionally, at a phenotypic level, the in vivo effects of vector-mediated gene reconstitution. Therefore, we evaluated the usefulness of attenuated Salmonella strains as a vector system for gene replacement in IFN-deficient mice.^16,17 The results obtained prove the concept that Salmonella can be successfully used in vivo as a delivery system for the correction of a 'genetic disorder' associated with monocyte/macrophages.

Results

Salmonella mediates in vitro gene transfer into murine macrophages

Salmonella-based vectors have been used to transduce, at high efficiency, non-proliferating differentiated cells such as macrophages in vitro. To demonstrate the exchange of genetic material between prokaryotic and mammalian cells, peritoneal unelicited murine macrophages were infected with the attenuated Salmonella typhimurium aroA^- strain SL7207 or its isogenic derivative carrying the plasmid pCMV which contains the -galactosidase (gal) gene (lacZ) placed under the control of the eukaryotic expression promoter of cytomegalovirus (CMV). Infected macrophages were stained with X-gal to detect the expression of the gal transgene over a 48 h time-course. No gal-positive cells were detectable during the first 24-28 h after infection (data not shown). In contrast, after 48 h most of the macrophages (approximately 90%) express the transgene encoded by the eukaryotic expression vector, whereas no gal expression was detectable within control cells infected with SL7207 containing no plasmid (Figure 1, panels a and b). The diffuse staining of macrophage cytoplasm supports the fact that production of gal is due to eukaryotic transcription and translation of the transgene (Figure 1, panel c).

In comparison to the pioneer study published by Darji et al,⁸ the protocol employed here has been consistently refined to optimize gene transfer. In fact, several factors may affect the efficiency of this process in vitro and/or in vivo. We observed that a strict control of LPS content has a major impact on bacteria/macrophage interactions, mostly because free radicals, such as NO, may impair the viability of the vector. To this extent the use of low endotoxin level containing reagents was particularly useful. Another critical factor to be considered is the ratio between eukaryotic cells and bacteria during infection. Macrophages undergo apoptotic death when infected by a critical number of bacteria. Therefore, a very careful tuning of different aspects led to the establishment of an efficient infection protocol, which optimizes Salmonella-mediated gene transfer to either murine macrophages or human monocytes/macrophages¹⁸ and dendritic cells. It is also important to highlight that transfection efficiency using Salmonella as a delivery system was far beyond that obtained using traditional transfection approaches.

To extend this result and to verify the ability of Salmonella to deliver genes within differentiated murine cells, leading to a restored protein synthesis, the strain SL7207 was transformed with the pCG3 plasmid, which contains the gene coding for the murine IFN (mIFN) placed under the control of the CMV promoter. The growth kinetics of the two isogenic SL7202 strains carrying the pCMV or pCG3 plasmid were compared, to rule out any indirect effect resulting from an impaired viability of the bacterial carrier as a result of the mIFN gene itself. As shown in Figure 2 (left panel), no differences were detectable in the growth kinetics of SL7207 pCMV and pCG3 strains. Furthermore, when plated on brain heart infusion (BHI) agar, identical numbers of viable bacteria were recovered for the wild-type SL7207 strain and the isogenic derivatives carrying either the pCG3 or the pCMV plasmid (about 2 ´ 10⁸ CFU/ml when they reach an OD₆₀₀ of 0.4).

The carrier SL7207 (pCG3) was used to infect, in vitro, murine peritoneal macrophages obtained from mice knocked-out for the IFN gene (GKO). The transcription of the mIFN gene by infected macrophages was examined by RT-PCR. IFN mRNA was detected within macrophages infected with Salmonella carrying plasmid pC3G (Figure 3a, lane 3) but not pCMV (Figure 3a, lane 2). Furthermore, the restored production of IFN in transduced macrophages was confirmed by testing cell supernatant in an IFN capture ELISA (Figure 3b). In conclusion, in vitro Salmonella-mediated gene transfer allows GKO macrophages constitutively to produce and release IFN in culture supernatant starting 24 h after infection with the bacterial carrier.

In vivo targeted gene delivery mediated by attenuated Salmonella vector allows correction of a genetic defect and restores natural resistance to bacterial infection in GKO mice

We have previously shown that the oral administration of Salmonella carrying the GFP gene under the control of a eukaryotic expression promoter, leads to an in vivo expression of the transgene by about 20% of cells in the spleen and that macrophages and dendritic cells but not lymphocytes are major targets of Salmonella-mediated gene transfer in vivo.¹⁰ Therefore, the expression of the Salmonella-delivered transgene was not locally restricted at the first site of interaction between bacteria and the host cells in the gastrointestinal apparatus, regardless of the oral administration of the carrier. In addition, a consistent number of transfected cells was detectable in this model at the systemic level suggesting a larger application potential of gene delivery mediated by Salmonella. The experiments performed using GFP also showed that only with the help of an enhanced GFP were we able to detect in vivo, at systemic level, the expression of the transgene delivered by Salmonella.¹⁰ In fact, when the gal gene was used as a model, we were unable to localize in vivo transfected cells following oral administration. This may be due, at least in part, to a problem of sensitivity. However, when Salmonella was administered by the intraperitoneal route, the expressed transgene (ie gal) could also be detected in peritoneal macrophages. In the case of mIFN poor recovery of transfected cells and reduced sensitivity of the read-out system may also constitute limiting conditions. Thus, recombinant Salmonella carrying either the pCMV or pC3G vectors freshly grown (about 2 ´ 10⁸ CFU) were administered to GKO mice by intraperitoneal injection. mRNA signals and release of IFN were detected in macrophages recovered from the peritoneal cavity of GKO mice receiving the SL7207 (pCG3) strain but not the isogenic strain carrying pCMV plasmid (data not shown). These results can be considered as a proof-of-principle for usefulness of the Salmonella-mediated system to deliver the mIFN gene to GKO macrophages. To evaluate further the feasibility of using a Salmonella-based strategy to correct genetic defects at phenotypic level in vivo, bacteria carrying either the pCG3 or pCMV plasmid were administered orally to GKO mice. These animals lack IFN-mediated resistance to bacterial infections. GKO mice were fed four times at 7 day intervals, with approximately 10⁹ Salmonella CFU. This bacterial dose has been previously determined as LD₉₀ for attenuated aroA^- Salmonella in IFN-deficient mice (data not shown). As shown in Figure 4, more than 50% of mice receiving the control vector carrying the pCMV plasmid died because of the Salmonella infection before the end of the treatment, and no survivors were left within 50 days from initiation of treatment (100% lethality). Conversely, all GKO mice receiving Salmonella carrying the pC3G plasmid survived the treatment with a high bacteria dose (100% survival), confirming that macrophage-targeted delivery of IFN gene mediated by Salmonella was able to restore resistance to bacterial infection in IFN-deficient mice.

Discussion

Attenuated derivatives of invasive bacteria, such as Shigella, Salmonella and Listeria, have been previously used for the delivery of eukaryotic expression vectors either in vitro or in vivo to prevent infectious and neoplastic diseases.^{4,5,6,8,9,10,19,20} These strains were attenuated by deletion of genes involved in the production of metabolites essential for cell wall synthesis or replication within host cells. The resulting mutants maintain the ability to infect their hosts, but they cannot undergo subsequent replication and die within target cells due to the lack of essential metabolites. These strains can specifically target antigen-presenting cells, such as dendritic cells and macrophages, and epithelial cells such as hepatocytes and enterocytes. After being phagocytosed, bacteria survive within target cells to a limited extent by either modifying the phagosomal compartment or breaching the cytoplasm. Once they undergo lysis, the plasmid content is released, leading to expression of the encoded protein. Subsequent induction of antigen-specific humoral and cellular immune responses have been extensively documented.²¹ This suggested the potential usefulness of this type of vectors for targeted gene therapy.

Several features must be accomplished by an ideal gene therapy vector in terms of efficacy, safety, general preparation and handling. One of the most interesting aspects associated with the use of attenuated Salmonella as a vector is the possibility of administering these bacteria by the oral route, a strategy which has proved to be successful in terms of efficacy and acceptability.²² In fact, the anti-typhoid fever vaccine based in the attenuated strain Ty21a is one of the few live bacterial vaccines licensed for human use. This vaccine has been extensively used to immunize both adults and children.^23,24

In contrast to laboratory conditions, the clinical application of the available viral vectors has shown that the extent of gene transfer and expression levels are generally low.²⁵ These shortcomings mean that high doses of viral vectors or lipid/DNA complexes are usually required to achieve a measurable gene transfer in vivo. Use of high doses of any vector leads to concerns about its safety and highlights the need to define a therapeutic window, avoiding toxicity. We have demonstrated that Salmonella-mediated gene transfer is able to induce a consistent level of gene expression, which is detectable at systemic level at doses that are safe for the vaccine. The span of transgene expression and the potential adverse effects resulting from repeated dosing should also be carefully evaluated. The period in which a newly introduced gene is expressed is an important variable which needs to be taken into account and may differ within different targeted tissues. Using Salmonella, transgene expression was detectable in vivo up to 1 month following the last administration of the carrier.¹⁰ In addition, multiple administrations of the carrier did not appear to limit the extent of gene transfer. In fact, expression of the wild-type phenotype in knock-out mice is maintained in all animals receiving the therapy, from the start of the treatment for the entire duration of the experiment. Although the invasive nature of the carrier is advantageous for efficient DNA delivery, it also implies a potential risk for the host, due to the hypothetical reversion of the carrier to the virulent phenotype. However, the possibility of generating attenuated mutants carrying more than one independent deletion leading to an attenuated phenotype, eliminates the risk of reversion following horizontal gene transfer. In addition, different mutants have been identified, which are not harmful even for immunocompromised hosts.^26,27 Some safety considerations for the use of Gram-negative bacteria refer to the toxic effect of LPS. However, this concern has been ruled out by oral delivery and by the fact that Salmonella strains have been widely used as vaccines both in human and veterinary medicine. The potential for the integration of bacteria-delivered DNA into host cell genome and oncogenesis promotion are not well known. While injection of naked DNA has not resulted in genomic integration, in vitro delivery of eukaryotic expression vectors by attenuated Listeria resulted in chromosomal integration of the delivered gene, most probably because of the high number of plasmid copies delivered in a single cell.²¹ This aspect would need to be carefully addressed. In terms of clinical endpoints, even though the applications of gene therapy attempted to date have primarily been safety studies, clinical benefit has proved to be difficult to establish except for the gene therapy of adenosine deaminase deficiency^28,29 and cancer.³⁰ Here, we have shown the outcome of gene delivery in vivo in a knock-out model, which resembles to a good extent a genetic defect. The emerging results demonstrate, for the first time, the feasibility of using an attenuated bacterial carrier as a vector for gene therapy. Oral administration of S. typhimurium, used as a DNA delivery system, resulted in targeted gene expression within macrophages replenishing their immunological functions. In fact, the Salmonella-mediated gene transfer was completely efficacious in preventing animal death due to a bacterial infection.

Materials and methods

Bacterial carrier and plasmids

The auxotrophic S. typhimurium aroA strain SL7207 (S. typhimurium 2337-65 derivative hisG46, DEL407 [aroA::Tn10{Tc-s}]) was kindly provided by BAD Stocker (Stanford University, School of Medicine, CA, USA). Salmonella strains were routinely grown at 37°C in BHI broth or agar (Sigma-Aldrich, Milan, Italy), supplemented with 100 g per ml of ampicillin (Life Technologies, Milan, Italy) when required. The plasmid pCMV was employed to set up the gene transfer protocol in vitro. mIFN gene was cloned in pCDNA3 vector (Clontech Laboratories, Heidelberg, Germany) to provide the eukaryotic expression system for this cytokine (pCG3). This vector has been validated by transient transfection of Cos cells and detection of the mIFN released in the supernatant of transfected cells (not shown).

Bacterial growth

To examine the kinetic of growth of the isogenic SL7207 strains carrying different plasmids, a bacterial suspension was made with four to five colonies from BHI agar plates containing ampicillin (10 mg/l). Bacterial suspension was adjusted to OD₆₀₀ = 0.3 and therefore diluted 1:30 in a 30 ml final volume of BHI broth in a 125 ml Erlenmeyer flask. Ampicillin was added as appropriate at 10 mg/l. Flasks were incubated at 37°C with gentle shaking and the OD of the cultures were monitored at different time-points. The experiment was stopped when cultures reached an OD₆₀₀ of approximately 0.8. The number of viable bacteria in cultures reaching an OD₆₀₀ of approximately 0.4, was evaluated by plating culture aliquots (100 l per plate) serially diluted (1:10) in saline buffer on BHI agar plates. Colonies were enumerated after ON incubation of plates at 37°C.

Animals

Female BALB/c of 6-12 weeks of age were from Charles River (Calco, Italy); IFN knock-out (GKO) mice from Jackson Laboratory (Bar Harbor, ME, USA), were maintained by breeding at INT-Animal Facility. This study has been approved by the Institutional Ethic Committee for Animal Use in Experimental Research.

Assay for gene transfer in vitro

Murine cells have been exposed in vitro to recombinant Salmonella to test for the exchange of genetic material and expression of transgenes. Briefly, unelicited peritoneal macrophages, either from BALB/c or GKO mice, were obtained by washing the murine peritoneal cavity with cold PBS. Recovered cells were plated in Iscove's medium supplemented with heat inactivated FCS (5%), which had been previously selected as a low endotoxin-containing batch (Life Technologies). Cells were allowed to adhere to the plastic, then carefully washed to remove lymphocytes and other contaminating cells and infected, 24 h later, by replacing medium with antibiotic-free medium containing recombinant Salmonella at a 10:1 to 50:1 infection ratio. Infection was carried out for 30 min at 37°C. Wells were then carefully washed and refilled with gentamycin-containing fresh medium in order to kill any residual extracellular bacteria. Gene transfer was determined, starting from the time of gentamycin addition to cultures and up to 24-48 h following infection, by either RT-PCR using specific primers (Clontech Laboratories), gal Staining Set (Roche Molecular Biochemicals, Milan, Italy) an X-gal-based staining kit to detect gal-expressing cells or by ELISA, using specific monoclonal antibodies pair (clones R4-6A2 and XMG1.2; Pharmingen, San Diego, CA, USA), to detect IFN released in the supernatant of infected cells.

Assay for gene transfer in vivo

Exponentially growing recombinant Salmonella carrying either pCMV or pCG3 vectors (about 2 ´ 10⁸ CFU in 0.4 ml saline buffer) were administered in vivo to GKO mice by intraperitoneal injection following extensive washing and resuspension in PBS. Peritoneal macrophages were recovered from the peritoneal cavity 24-48 h after injection and plated at 0.5 ´ 10⁶ cells per well in 48-wells plated in Iscove's medium containing gentamycin. After 18 h incubation at 37°C, IFN presence was measured by ELISA in macrophage supernatants as previously described.

Monitoring of genetic defect correction in vivo

Recombinant Salmonella carrying either pCMV or pCG3 vectors were administered in vivo by feeding mice (20 animals per group) four times at 7 day intervals with a bacterial suspension containing approximately 10⁹ CFU in 30 l volume of bicarbonate buffer. This dose has been determined as LD₉₀ for Salmonella SL7207 in GKO mice (data not shown). Survival of animals in control and experimental groups was monitored.

Statistical analysis

Statistical differences between experimental groups were evaluated by non-parametric Mann and Whitney U test for the in vivo survival. A P value <0.05 was considered statistically significant.

Acknowledgements

This work was in part supported by grants from Telethon-Italy (Grant Nos A102 and A135), ISS-AIDS-1998, AIRC, CNR (PF-Biotechnology), HGF-Strategiefondsprojekt 98/03 'Infektionsabwehr und Krebsprävention' (P1.3) and the European Community (QLK2-CT-1999-00310).

1 Malech HL, Bauer TR Jr, Hickstein DD. Prospects for gene therapy of neutrophil defects. Semin Hematol 1997; 34: 355-361, MEDLINE

2 Meischl C, Roos D. The molecular basis of chronic granulomatous disease. Springer Semin Immunopathol 1998; 19: 417-434, MEDLINE

3 Platt F, Butters TD. New therapeutic prospects for the glycosphingolipid lysosomal storage diseases. Biochem Pharmacol 1998; 56: 421-430, MEDLINE

4 Grillot-Courvalin C et al. Functional gene transfer from intracellular bacteria to mammalian cells. Nat Biotechnol 1998; 16: 862-866, MEDLINE

5 Grillot-Courvalin C, Goussard S, Courvalin P. Bacteria as gene delivery vectors for mammalian cells. Curr Opin Biotechnol 1999; 10: 477-481, MEDLINE

6 Sizemore DR, Branstrom AA, Sadoff JC. Attenuated Shigella as a DNA delivery vehicle for DNA-mediated immunization. Science 1995; 270: 299-302, MEDLINE

7 Hackett J. Use of Salmonella for heterologous gene expression and vaccine delivery systems. Curr Opin Biotechnol 1993; 4: 611-615, MEDLINE

8 Darji A et al. Oral somatic transgene vaccination using attenuated S. typhimurium. Cell 1997; 91: 765-775, MEDLINE

9 Medina E et al. Salmonella vaccine carrier strains: effective delivery system to trigger anti-tumor immunity by oral route. Eur J Immunol 1999; 29: 693-699, MEDLINE

10 Paglia P et al. Gene transfer in dendritic cells, induced by oral DNA vaccination with Salmonella typhimurium, results in protective immunity against a murine fibrosarcoma. Blood 1998; 92: 3172-3176, MEDLINE

11 Jones BD, Falkow S. Salmonellosis: host immune responses and bacterial virulence determinants. Annu Rev Immunol 1996; 14: 533-561, MEDLINE

12 Jane SM, Cunningham JM, Vanin EF. Vector development: a major obstacle in human gene therapy (editorial). Ann Med 1998; 30: 413-415, MEDLINE

13 Robbins PD, Ghivizzani SC. Viral vectors for gene therapy. Pharmacol Ther 1998; 80: 35-47, MEDLINE

14 Michalski JC, Klein A. Glycoprotein lysosomal storage disorders: alpha- and beta-mannosidosis, fucosidosis and alpha-N-acetylgalactosaminidase deficiency. Biochim Biophys Acta 1999; 1455: 69-84, MEDLINE

15 Feder JN. The hereditary hemochromatosis gene (HFE): a MHC class I-like gene that functions in the regulation of iron homeostasis. Immunol Res 1999; 20: 175-185, MEDLINE

16 Nansen A et al. Role of interferon-gamma in the pathogenesis of LCMV-induced meningitis: unimpaired leucocyte recruitment, but deficient macrophage activation in interferon-gamma knock-out mice. J Neuroimmunol 1998; 86: 202-212, MEDLINE

17 Kjerrulf M et al. Interferon-gamma receptor-deficient mice exhibit impaired gut mucosal immune responses but intact oral tolerance. Immunology 1997; 92: 60-68, MEDLINE

18 Montosi G et al. Wild type Hfe protein normalizes transferrin iron accumulation in macrophages from subjects with hereditary hemochromatosis. Blood 2000; 96: 1119-1124, MEDLINE

19 Paglia P et al. The defined attenuated Listeria monocytogenes delta mp12 mutant is an effective oral vaccine carrier to trigger a long-lasting immune response against a mouse fibrosarcoma. Eur J Immunol 1997; 27: 1570-1575, MEDLINE

20 Guzman CA et al. Attenuated Listeria monocytogenes carrier strains can deliver an HIV-1 gp120 T helper epitope to MHC class II-restricted human CD4+ T cells. Eur J Immunol 1998; 28: 1807-1814, MEDLINE

21 Dietrich G et al. Delivery of DNA vaccines by attenuated intracellular bacteria. Immunol Today 1999; 20: 251-253, MEDLINE

22 Dilts DA et al. Phase I clinical trials of aroA aroD and aroA aroD htrA attenuated S. typhi vaccines; effect of formulation on safety and immunogenicity. Vaccine 2000; 18: 1473-1484, MEDLINE

23 Hone DM et al. A galE via (Vi antigen-negative) mutant of Salmonella typhi Ty2 retains virulence in humans. Infect Immun 1988; 56: 1326-1333, MEDLINE

24 Levine MM et al. Safety, infectivity, immunogenicity, and in vivo stability of two attenuated auxotrophic mutant strains of Salmonella typhi, 541Ty and 543Ty, as live oral vaccines in humans. J Clin Invest 1987; 79: 888-902, MEDLINE

25 Russell DW, Hirata RK. Human gene targeting by viral vectors. Nat Genet 1998; 18: 325-330, MEDLINE

26 Medina E et al. Pathogenicity island 2 mutants of Salmonella typhimurium are efficient carriers for heterologous antigens and enable modulation of immune responses. Infect Immun 1999; 67: 1093-1099, MEDLINE

27 VanCott JL et al. Regulation of host immune responses by modification of Salmonella virulence genes. Nature Med 1998; 4: 1247-1252, Article MEDLINE

28 Onodera M et al. Successful peripheral T-lymphocyte-directed gene transfer for a patient with severe combined immune deficiency caused by adenosine deaminase deficiency. Blood 1998; 91: 30-36, MEDLINE

29 Bordignon C et al. Gene therapy in peripheral blood lymphocytes and bone marrow for ADA-immunodeficient patients. Science 1995; 270: 470-475, MEDLINE

30 Bordignon C et al. Cell therapy: achievements and perspectives. Haematologica 1999; 84: 1110-1149, MEDLINE

Figure 1 Salmonella mediates delivery and expression of the reporter gene in vitro to macrophages. Detection of gal by X-gal staining in murine peritoneal macrophages infected in vitro with Salmonella carriers: cells infected with empty vector SL7207 (panel a) or SL7207 carrying the pCMV plasmid (panel b, original magnification ´ 40; panel c, original magnification ´ 100).

Figure 2 The growth kinetics of the Salmonella strain SL7207 is not impaired by the presence of mIFN gene coded by the pCG3 plasmid. The growth pattern of the Salmonella strains SL7207 carrying either pCMV or pCG3 were monitored by measuring OD₆₀₀ of broth cultures at different time intervals (panel a). To rule out leakiness in the control of the CMV promoter over the transcription of the mIFN gene, the presence of this cytokine was monitored by ELISA in the bacterial cultures of SL7207 [pCG3] reaching the OD600 of 0.4 and 0.6 (panel b). As positive control a supernatant from LPS-stimulated macrophages (5 ´ 10⁵ cells + 10 g/ml LPS, 48 h) was included in the ELISA assay.

Figure 3 Salmonella allows targeted IFN gene delivery and restores IFN production in GKO macrophages in vitro. (a) RT-PCR detecting IFN mRNA using cDNA obtained from control IFN template (lane 1), murine macrophages infected with Salmonella carrying either pCMV (lane 2) or pCG3 vectors (lane 3), and uninfected control GKO macrophages (lane 4). (b) ELISA determination of IFN released in supernatants of Salmonella infected GKO macrophages.

Figure 4 A genetic defect from macrophages can be corrected in vivo by Salmonella-mediated targeted gene delivery. Survival of GKO mice (20 animals per group) receiving S. typhimurium SL7207 carrying pCMV () or pCG3 () vectors.