Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Hygiene hypothesis wikipedia , lookup
Monoclonal antibody wikipedia , lookup
Molecular mimicry wikipedia , lookup
Immune system wikipedia , lookup
Immunosuppressive drug wikipedia , lookup
Polyclonal B cell response wikipedia , lookup
Adaptive immune system wikipedia , lookup
Innate immune system wikipedia , lookup
Cancer immunotherapy wikipedia , lookup
Psychoneuroimmunology wikipedia , lookup
Gene Therapy (1998) 5, 789–797 1998 Stockton Press All rights reserved 0969-7128/98 $12.00 http://www.stockton-press.co.uk/gt Superiority of the ear pinna over muscle tissue as site for DNA vaccination ¨ P Forg1,2, P von Hoegen1,3, W Dalemans3 and V Schirrmacher1 1 German Cancer Research Center (DKFZ), Division of Cellular Immunology, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany Three different vaccination sites were compared for efficiency of immunization with naked DNA. Using the bacterial lacZ gene as a model, all three sites of the mouse (skeletal muscle, dermis of abdominal skin or of the ear pinna) could express the gene product -gal but varied in expression time with muscle tissue showing the longest expression. Expression time, however, did not correlate with immune response intensity. The ear pinna was by far the most effective and muscle the least effective priming site for specific humoral and cytotoxic T cell-mediated immune responses. Following intra-pinna DNA inoculation, -gal expressing cells were detectable around the injection site and in the major draining lymph node. Efficiency of immunization was also dependent on the promoter and expression vector used. The cytomegalus virus promoter driven pCMV vector was superior to the Moloney murine leukemia virus LTR driven BAG vector. LacZ DNA immunization was also compared with cell-based vaccination with lacZ-transfected tumor cells, in which case again the pinna was the best site for inducing strong immune responses. Tumor-specific T cell responses could also be well induced in the pinna, leading to cytotoxic T lymphocyte induction and protective antitumor immunity. Thus, the pinna was found to be a privileged site for induction of antitumor responses and for genetic immunization, an important finding of immediate practical and potential future clinical implications. Keywords: lacZ gene; promoter; tumor; immunity; genetic immunization Introduction In 1990, Wolff and colleagues1 reported that nonreplicating DNA plasmids encoding reporter genes could be internalized and the encoded proteins expressed by muscle cells following injection of plasmids directly into muscle without the use of any transfection vehicle. The subsequent finding that naked plasmid DNA encoding a foreign protein could also be immunogenic upon intramuscular inoculation opened an entirely new area of research with great clinical impact. Somatic gene therapy is becoming increasingly important for potential treatment of genetic and acquired disorders. Without any specific delivery system the direct DNA transfer into tissues is followed by uptake of plasmid molecules into the cytoplasm and subsequently into the nucleus of cells. They are then translated into functional proteins, processed into peptides and presented by major histocompatibility (MHC) molecules to immune T cells.2 Direct DNA transfer has been successfully used for in vitro induction of both humoral and cellular immune responses to several antigens, including viruses such as influenza, hepatitis B, or HIV (human immunodeficiency virus), parasites, or tumor antigens.3–8 Additionally, systemic effects were observed after transfer of cytokine genes.9 Correspondence: V Schirrmacher Present addresses: 2Pharmacia and Upjohn GmbH, 91058 Erlangen, Germany; and 3SmithKline Beecham Biologicals, Rue de l’Institut 89, B-1330 Rixensart, Belgium Received 8 August 1997; accepted 18 December 1997 It was assumed at the beginning that striated muscle is the only tissue in vitro suitable to take up and express naked DNA.10 Soon, however, it turned out that other organs showed a similar ability for expression of a reporter gene.11 The skin as a possible target tissue for genetic immunization was successfully tested by Tang et al,12 by bombardment with gold microprojectiles coated with plasmid DNA. But gene expression in the skin was also observed when using needle injection of plasmid DNA,13,14 especially when applying high-frequency puncturing of the skin.15 For effectiveness of gene therapy the skin seems to be superior to muscle or other organs. Muscle is not considered to be a site specialized for antigen presentation because it contains few if any antigen presenting cells (APCs) or lymphocytes. Thus, bone marrow-derived nonmuscle cells were shown to be involved in the induction of cytotoxic T lymphocytes (CTL) following intramuscular DNA immunization.16–18 In contrast, the skin and mucous membranes are the physiological sites where most exogenous antigens are normally encountered. The skin-associated lymphoid tissues contain specialized cells such as keratinocytes, macrophages and Langerhans’ cells, which are involved in the initiation and further augmentation of immune responses. Langerhans’ cells carry the antigen from the skin to draining lymph nodes where they function as professional APCs for priming naive T lymphocytes. The transfection of skin cells by DNA is therefore expected to be an efficient route of immunization, mimicking a physiologic response as during infection. Taking advantage of the reporter gene lacZ which can Intra-pinna DNA vaccination ¨ P Forg et al 790 easily be monitored we compared the ability of skeletal muscle and different sites of skin to take up and express naked lacZ DNA and to induce antibody and CTL responses against the encoded model antigen -galactosidase (-gal). In previous studies with the ESb tumor system it was shown that the ear pinna is a privileged site for the induction of antitumor immunity,19 preventing the outgrowth of an otherwise lethal dose of tumor cells. Thus it was reasonable to test the ear pinna also as a putative site for DNA vaccination and to compare the strength of the immune response to that induced by -gal expressing tumor cells. Additionally, two different lacZ encoding vectors were compared for usage as vaccines. Here we demonstrate that the strongest immune responses to DNA vaccination were obtained after intra ear pinna (i.e.) and not after muscle immunization and that the immune response intensity did not correlate with the duration of gene expression in the injected tissues. Results Comparison of expression of lacZ DNA in skeletal muscle, abdominal skin and ear pinna after direct DNA transfer We first compared different tissues for their capability of taking up injected lacZ DNA and expressing the foreign gene over a certain period of time. Two different vectors were used: the retroviral vector BAG and the expression vector pCMV. Following inoculation, animals were killed at different time-points, and their frozen tissue sections stained for -gal-positive cells ex vivo (Figure 1a–f). With both lacZ encoding vectors a direct transfection was observed. Especially in the case of the intradermal injections at the abdominal skin or the ear pinna, the stained area appears to represent the direction of the needle during the prick (Figure 1b and c). An analysis of the ear pinna by X-gal-stained histological cross-sections (Figure 1e) revealed that many cell types within the dermis and epidermis took up and expressed the DNA without any preference. Expression was seen longer in muscle tissue (Figure 1a) (quadriceps at least for 4 weeks) than in the abdominal skin (Figure 1b) or the ear pinna (Figure 1c–e), but it was also dependent on the vector used. Expression vanished earlier with pCMV than with BAG. Three weeks after pinna injection, no blue cells could be detected. The kinetics of gene expression were studied in more detail for the ear pinna. Using pCMV, distinct expression as a blue spotted circular area had already occurred after 1 day. This characteristic pattern changed to a few lines (Figure 1d) within 10 days. After 14–21 days, in most cases, no more expression was visible. A similar gradual decline of gene expression over time was seen in nude mice. Figure 2 illustrates the relative strength and duration of -gal gene expression at the different injection sites. Detection of -galactosidase-positive cells in the draining lymph nodes after i.e. injection Analysis of the draining lymph nodes (dLn) after i.e. DNA injection revealed the presence of -gal-positive cells (Figure 1f). In control lymph nodes of nontreated ear pinnae such X-gal staining of cryosections was never a b c d e f Figure 1 Expression of -gal following injection of DNA into different murine tissues. (a) Whole organ staining of the quadriceps of the left hind leg ex vivo 30 days after injection of BAG. (b) Skin from the shaved flank 21 days after injection of BAG. (c and d) Ventral site of an ear pinna, injected with pCMV 1 day (c) and 10 days (d) before. (e) Tangential section of an ear pinna with -gal expressing cells (× 20). (f) A -galpositive cell in a cryosection of the draining lymph node detected by Xgal staining 3 days after injection of pCMV into the corresponding ear pinna (× 100). Figure 2 Kinetics of -gal expression after pCMV lacZ DNA inoculation into the ear pinna (i.e.), into the dermis of the flank (i.d.) or into muscle tissue (i.m.). Expression strength is expressed by an arbitrary scale of 1 (low), 2 (medium) and 3 (high). Early time-points (3–24 h) were only tested i.e. Expression duration was i.m. ⬎ i.d. ⬎ i.e. observed. The first time-point after i.e. pCMV inoculation when blue cells could be detected in the major dLn (superfiscia cervicalis) was at 1 day. The type of cell in the lymph node which expressed -gal could not be clearly identified. When the pinna was removed 30–60 min after Intra-pinna DNA vaccination ¨ P Forg et al injection, no blue cells were detectable at any time later in the dLn. This differed from a situation, in which lacZtransfected tumor cells were i.e. inoculated and in which stained cells could be seen in the dLn as early as 15 min after inoculation.20 Induction of -gal-specific antibodies The development of humoral immunity, in the form of specific IgG anti--gal antibodies, was tested using a sensitive -gal ELISA plate assay. Serum titers are given as relative values in comparison to a standard monoclonal anti--gal antibody at a concentration of 10−5 mg/ml. The efficacy of the different sites of injection with regard to induction of humoral immunity are compared in Figure 3. Independent of the injection site, a single inoculation of BAG induced distinct antibody levels. These were detectable after 2 weeks and increased constantly over the observed period of 12 weeks. While inoculation into the quadriceps (Figure 3a) or into the abdominal skin (Figure 3b) revealed a similar antibody response, i.e. immunization (Figure 3c) resulted in a much stronger response which reached a plateau after 4 weeks. Sera from mice immunized i.e. had to be diluted much further to reveal the whole titration curve. Inoculation of live ESbL-lacZ tumor cells into the ear pinna also resulted in the production of specific anti-- gal antibodies. In comparison with DNA-based vaccination, the antibody increase was more rapid (Figure 3d). Induction of a MHC class I restricted -gal peptidespecific CTL response To evaluate the development of a T cell-mediated immune response, splenocytes from mice immunized with BAG or pCMV intramuscular (i.m.), intradermal (i.d.) or i.e., were used to set up a CTL restimulation culture in the presence of a known Ld-restricted peptide of the dominant -gal-derived T cell epitope. As target cells for the subsequent 4h 51Cr release assay the lacZ expressing P13.1 cells were used and as negative control the respective nontransfected parental P815 mastocytoma cells. Spontaneous release was always below 10%. Results of representative experiments are shown in Figure 4. After i.m. immunization with BAG for 9–21 days no specific CTL responses could be generated (Figure 4a). In contrast, after i.d. immunization (Figure 4b) distinct responses were reproducibly observed with around 50% lysis at 50:1 effector to target cell ratio, when the immune spleen cells were used on day 21. The cytolytic activity generated in these bulk cultures was even higher when BAG immunization was performed i.e. (60–70% specific lysis, Figure 4c). Priming times with BAG shorter than 3 weeks was not sufficient for induction of strong CTL Figure 3 Kinetics of humoral anti--gal immune responses. Serum samples from three animals per group were tested in triplicate on -gal ELISA plates. The average values were calculated over the OD492 of monoclonal anti--gal at a concentration of 10−5 mg/ml. Single immunization with BAG i.m. (a), i.d. (b) and i.e. (c). (d) Cell-based immunization with ESbL-lacZ tumor cells. 791 Intra-pinna DNA vaccination ¨ P Forg et al 792 with ESbL-lacZ tumor cells, however, can be totally addressed to the effect of -gal since restimulation with the peptide is -gal specific. Figure 4 -gal peptide-specific CTL responses in lacZ-immunized mice. Splenocytes from three mice per group were isolated 3 weeks after a single injection of the lacZ vector BAG i.m. (a), i.d. flank (b), and i.e. (c) or 9 days after i.e. injection of ESbL-lacZ cells (d) or of the lacZ vector pCMV (e). The immune spleen cells were restimulated in bulk cultures for 5 days with 0.5 g/ml TPHPARIGL peptide. The thus generated Ld-peptiderestricted CTL effector cells were then tested in a 4 h 51Cr release assay against P13.1 cells expressing -gal (쮿), or against the parental P815 cells (쑗) as negative control. (f) Effect of resection of the DNA-injected ear pinna at different time-points after immunization. Mice were immunized as in (e). In separate groups of three mice each the pinna was resected either 1 h (왔), 24 h (쏔) or 72 h (쏆) after DNA inoculation. Effector cells were generated as in (e) and tested against P13.1 target cells. responses (data not shown). The strongest -gal-specific CTL response was observed after i.e. immunization with live lacZ-transfected ESbL lymphoma cells (Figure 4d). In addition to the strong immune response to the foreign lacZ gene product, strong ESb tumor-specific CTL immune responses are induced at this injection site.21 The observed CTL response (Figure 4d) after immunization Promoter-dependent differences in the induction of immune responses and relevance of antigen persistence When pCMV was used instead of BAG, a quicker and stronger -gal peptide-specific CTL response was observed. Thus, 30% specific lysis could be generated from immune spleen cells of mice primed for only 9 days by i.e. immunization (Figure 4e). Removal of the injected ear pinna at different time-points after pCMV injection revealed that the strength of the induced CTL responses increased with the time of pinna persistence. No CTL response was seen when the pinna was removed 1 h after DNA application, while weak responses were seen when the injected pinna was allowed to persist for 24 or 72 h (Figure 4e). Even a 72 h lasting persistence did not reach the same CTL capacity as a persistence for the whole priming time of 9 days. Thus, persistence of the DNAinjected pinna over a certain period of time (⬎72 h) appeared as a prerequisite for obtaining optimal CTL responses. When anti--gal immune responses were compared at the same time-point after i.e. injection of the two different lacZ expression vectors, stronger humoral (Figure 5a) and T cell-mediated responses (Figure 5b) were observed after immunization with pCMV than with BAG. A kinetic analysis of the pCMV-induced i.e. response revealed that good CTL responses were already obtainable after 6 days of DNA immunization. The lytic activity remained high (approximately 80% specific lysis) for at least 45 days (data not shown). Another point of interest was the influence of B and T cell-mediated immune responses of a second DNA injection. To test this, 14 days after the first i.e. injection of pCMV mice received the same dose of pCMV at the same site. As shown in Figure 5a, this boosting injection increased the antibody titer further. Similarly, a boosting injection of pCMV and a total priming time of 28 days led to a further increase in the peptide-generated CTL activity (Figure 5b). The lysis was antigen specific since nonimmunized mice as well as mice immunized with a nonrelevant plasmid which did not encode the lacZ gene displayed only background activity (data not shown). In summary, we demonstrate important parameters for optimal immune responses after genetic immunization with a model antigen: the site of injection, the promoter used and the number of injections performed. Mouse pinna as a privileged site for induction of antitumor responses The pinna was also a privileged site for induction of antitumor immune responses. Results obtained with the welldefined DBA/2 mouse lymphoma ESb which expresses a tumor characteristic Kd-restricted CTL epitope22 are shown in Table 1. While inoculation of 5 × 104 live tumor cells at s.c., i.m. or i.d. (flank) sites (groups I–III) led to progressive tumor growth and death from metastases within 2–3 weeks, i.e. inoculation (group IV) induced tumor resistance and specific protective antitumor immunity as revealed by tumor re-challenge experiments. Induction of protective immunity by i.e. tumor immunization was abrogated by sublethal irradiation Intra-pinna DNA vaccination ¨ P Forg et al (group VII) or pretreatment with depleting anti-CD4 (V) or anti-CD8 (VI) monoclonal antibodies (mAbs). ESbspecific CTL responses could be induced by i.e. priming and intraperitoneal (i.p.) restimulation (groups VIII–X). The peritoneal exudate cells mediated syngeneic antitumor cytotoxicity and lysed ESb, but not Eb lymphoma cells which express a different tumor-associated antigen (group VIII). Furthermore, the ESb-specific kill could be blocked by anti-MHC class I Kd but not anti-Dd mAbs, thus revealing the restricting molecule for the tumorspecific CTL as Kd. Discussion In order to optimize DNA vaccination protocols, we evaluated the efficacy of gene transfer in several tissues and followed subsequent immune responses to -gal as model antigen. Mice were inoculated into skeletal muscle, skin and ear pinna with two different vectors containing the bacterial lacZ gene under the transcriptional control of the Moloney murine leukemia virus LTR or the CMV promoter. The read out systems for the study were the examination of the lacZ gene expression, the production of anti--gal antibodies and the priming for gal peptide-specific CTL responses. In situ staining by X-gal revealed that all three tissues were able to take up and express the plasmid DNA. No differences were detected between the two vectors BAG and pCMV with regard to the quantity of lacZ mRNA in injected tissues (data not shown) and to the level of gene expression in situ. The skeletal muscle expressed gal in a broad area after injection of lacZ encoding vectors (Figure 1a), whereas the stained cells of skin and ear pinna were arranged around a well defined region (Figure 1b–d) which seems to reflect the direction of the needle during inoculation. Cross-sections of the ear pinna (Figure 1e) displayed distribution of -gal expressing cells throughout the whole organ, including the epidermis and dermis. The alteration of expression strength and pattern noticed in the ear pinna over time is in agreement with observations made by Hengge et al23 in pig skin, with a decline of expression after 3 weeks and a gradual reduction in size of the transfected stretch. One possibility could be that the decline of expression was due to an immunological process as was described in muscle where an inflammatory reaction occurred in response to the injection of naked DNA.24 This seems, however unlikely since we found the same decline of expression in the ear pinna of immune-deficient nude mice. Moreover, inflammatory reactions of the skin were not observed in this study. Another possibility is down-regulation of promoter activity. Transient expression of transgenes under the control of the CMV promoter was observed by other investigators.14,25 For clinical applications it may be considered an advantage when expression of a foreign gene is transient and is switched off after induction of immunity. It might be safer and easier to control. Immunizations could be repeated if necessary. Muscles as well as skin have already been described as targets for DNA transfer. The skin is an especially interesting tissue since it carries a variety of cells that could induce or be involved with immune responses, eg keratinocytes, macrophages or Langerhans’ cells. The thin ear pinna exhibits two layers of epidermis and dermis, separated by cartilage, thereby doubling the amount of professional APC. So far, the ear pinna tissue as a special site and form of the skin has not been investigated in gene therapy in detail although its anatomical properties and immunological features seem to predestine it for this purpose. It is easily accessible without the need for pre-treatments such as shaving, and has been demonstrated to be a privileged organ for the induction of protective antitumor immune responses.19,20,26 An interesting point is the detection of -gal expressing cells in the draining lymph node after i.e. injection of lacZ. We could not see such blue stained cells when the injected ear pinna was removed after 1 h. Our assumption is that these stained cells represent Langerhans’ cells or dermal dendritic cells which have been transfected directly and then migrated into the draining lymph node Figure 5 Comparison of i.e.-induced immune responses to different lacZ plasmid expression vectors. Humoral (a) and cell-mediated (b) anti--gal immune responses after a single i.e. immunization on day 0 with BAG (쮿) or pCMV (쑗) or after two i.e. immunizations on day 0 and 14 with pCMV (쎲). (a) Serum samples from three animals per group taken before (*) or 4 weeks after immunization were ELISA tested as in Figure 3. (b) -gal Ld peptide-restricted CTL responses were assayed as in Figure 4 with day 28 immune spleen cells against P13.1 target cells (*: lysis of P815 control cells). 793 Intra-pinna DNA vaccination ¨ P Forg et al 794 Table 1 Mouse pinna as a privileged site for induction of antitumor responses Group I II III IV V VI VII ESb tumor cell inoculuma Site Additional treatmentb % ESb tumor resistanced 5 × 104 5 × 104 5 × 104 5 × 104 5 × 104 5 × 104 5 × 104 s.c. i.m. i.d. i.e. i.d. i.e. i.e. — — — — d-1 ␣ CD4 d-1 ␣ CD8 d-1 5 Gy 0 0 0 80e 0 0 0 Restimulation % Specific cytotoxicity against ESb Eb Priming VIII IX X 5 × 104 5 × 104 5 × 104 i.e. i.e. i.e. ESb irr. d-12 i.p.c + medium ESb irr. d-12 i.p. + ␣Kd ESb irr. d-12 i.p. + ␣Dd 55.1 9.8 57.0 0.5 1.5 2.0 a DBA/2 mice were immunized with live syngeneic ESb 289 lymphoma cells on day 0 (d0) at the sites indicated. Separate groups of mice (10 per group) were pretreated i.p. 1 day before with 250 l ascites fluid of either anti-CD4 (GK 1.5) or antiCD8 (YTS) mAb or were pretreated by 5 Gy sublethal X-irradiation. c For generation of tumor-specific CTL activity, mice were restimulated with 1.5 × 107 X-irradiated (100 Gy) ESb tumor cells i.p. for 3 days and the harvested peritoneal effector cells (PEC) further cocultured for 3 days with ESb stimulator cells. 51Cr-labeled ESb target cells (which express a Kd-restricted tumor-specific CTL epitope) or Eb target cells (which express a different tumor-associated antigen) were preincubated for 45 min at 4°C with either medium or five-fold concentrated hybridoma supernatants before PEC effector cells were added in a 2.5-fold excess to the targets and incubated for 4 h. d Represents the per cent survivors after 60 days. Mortality occurred between days 8 and 18 and was due to metastases. e These mice developed specific protective antitumor immunity so that after re-challenge i.d. with either Eb or ESb lymphoma cells Eb but not ESb tumors developed. b to function as APCs. We cannot exclude the possibility that these cells might have taken up the protein from dying cells and have then migrated into the lymph node to present the antigen to lymphocytes but it is not likely that the -gal would then be in enzymatically active form that allows detection by X-gal staining. The same phenomenon, the presence of foreign DNA expressing dendritic cells in draining lymph nodes after cutaneous DNA-based immunization was recently reported by Condon et al.27 Other studies pointed towards an important role of host APC in the induction of T cell-mediated immune responses after i.m. DNA injection. Such APC did not need to be transfected themselves.16–18 The presence of the foreign antigen -gal in the lymph nodes was a first hint for induction and generation of an immune response. The generation of a specific humoral immune response against the encoded protein after genetic immunization was seen for every site of injection, but the titers were influenced by the injection site. Intramuscular and intradermal immunizations exhibited a similar development with a rather low antibody titer. Animals injected with DNA i.e. displayed an IgG antibody level after 2 weeks that was similar to that observed as maximal after i.m. or i.d. injections. This may reflect the strength of MHC class II-mediated helper T cell immunity induced by i.e. immunizations. A further substantial antibody response increase was demonstrated after injecting pCMV a second time 2 weeks after the initial DNA transfer. This is in contrast to Ertl and Xiang25 who saw no boosting effect and proposed the augmentation to be the result of a slow on-rate. Comparison of the titers of antibody induced after inoculation of BAG or pCMV, respectively, revealed the CMV promoter-driven expression to be more efficient. This may reflect a specific activity of the CMV promoter, confirming earlier results,28 and probably demonstrates the connection between level of expression and strength of the generated immune response. A similar conclusion concerning antigen dose dependency of the response can also be drawn from immunizations with -gal expressing tumor cells. Intra-pinna injection of a subtumorigenic dose of ESbL-lacZ as well as of ESb cells (Table 1) led to induction of antitumor immunity so that a subsequent tumor challenge at a tumorigenic site was rejected by syngeneic DBA/2 mice. This effect was dependent on CD4 and CD8 T cells since injection of depleting anti-CD4 or anti-CD8 mAbs would prevent induction of immunity against the ESbassociated tumor antigens (Table 1 and Ref. 29). The additional foreign protein -gal in ESbL-lacZ could induce additional anti--gal T cell responses (Figure 3d) thereby increasing the immunogenicity of the tumor cells.21 The production of anti--gal antibodies after tumor cell-based immunization was higher than after genetic immunization. After DNA injection, the number of cells expressing -gal is presumably lower than the 5 × 104 ESbL-lacZ tumor cells injected i.e. The induced anti--gal immune responses could also confer protective immunity against other -gal-transfected tumor cells (unpublished observations). Differences as seen with the humoral immune responses were also observed when CTL-mediated peptide-specific immune responses were tested. Muscle cells injected with BAG totally failed to induce MHC class I-(Ld) -gal peptide-restricted CTL activity, while after intradermal immunizations good CTL responses were elicted. Thus, while humoral responses were equivalent for i.m. and i.d. injections, induction of cytotoxic Intra-pinna DNA vaccination ¨ P Forg et al CD8+ T cell precursors differed distinctly between these two tissues. Reasons for this may be seen in the different APC content of the injection sites in combination with the rather low level of expression. A weak CTL response could be elicited i.m. when the stronger pCMV vector ¨ was applied (data not shown). Bohm et al,30 however, presented data in their system suggesting that lower expression stimulates only MHC class I-mediated T cell responses while a higher expression resulted in both humoral and cellular immunity. In contrast, in our system, the lower level of expression after i.m. injection of BAG seemed to induce preferably B cell-mediated responses. With regard to the peptide-specific CD8 T cellmediated CTL response, the intra-pinna inoculation site proved again to be superior to muscle and abdominal skin. The induced CTL response was nearly comparable to the response after injection of ESbL-lacZ cells so that in this case the amount of antigen was not as decisive as for the humoral responses. Again, a booster injection could further augment the CTL immune response. A main difference between the genetic and the cell-based immunization in the ear pinna relates to the effect of removal of the injected pinna at different time-points after injection. For direct DNA transfer we demonstrated the need for persistence of the pinna, possibly as an antigen reservoir, during the priming period for the development of a strong CTL response. Previously, we showed in a similar experimental set-up concerning tumor cellbased immunization in the ear pinna that the cytotoxic antitumor immune response is generated at comparable levels even if the pinna is removed after 1 h.20 In the case of i.e.-injected lacZ-tumor cells, lacZ-labeled cells could be detected in the draining lymph node and spleen as early as 15 min after injection,20 while in the case of i.e.injected lacZ DNA, at least 24 h were required before lacZ-labeled cells could be detected in the draining node. Thus, although the strength of immune responses induced either by DNA or cell-based immunization may be similar, the kinetics of induction and the mechanism of priming seem to be quite different. In summary, the ear pinna was demonstrated to be an optimal site for generation of humoral and cell-mediated immune responses after DNA vaccination. One reason for this may be the local and focal concentration of antigen in a restricted area connected with the one major draining lymph node (superfiscia cervicalis). This can result in fast stimulation of virgin T lymphocytes by antigen-loaded dendritic cells. After i.d. or i.m. injection, the response is likely distributed over several lymph nodes.31 The mechanical irritation by needle injection may induce local cytokine secretion and activate and recruit additional APC. Regarding the route of DNA administration, the experiences of different groups differ widely. Donnelly et al32 reported that i.m. treatment resulted in a better protection than i.d. application. There are hints33 that gene gun-mediated immunization gives similar effects as i.d. injections with predominantly Th2 immune responses, but with similar CTL activity as after i.m. injection. However, recently, i.d. injection of the lacZ gene was found to preferentially induce Th1 response.34 In accordance with this, we demonstrated recently a superiority of the i.e. over a s.c. tumor inoculation site for induction of a Th1 type cytokine response.26 Fynan et al35 observed that i.m. applications were weaker in elicting immune responses than gene gun treatment. Our results show that i.d. and especially i.e. injections, are the most efficient ways of inducing strong immune responses. In these situations, the entrance of the antigen via the skin may be efficient because it follows routes of natural infection. The type of antigen may also affect the type of immune response generated. However, the route of administration seems to have greater influence, pointing towards outstanding potential of the ear pinna as a DNA vaccination site. Materials and methods DNA The retrovirus-based vector BAG (approximately 11 kb)36 was obtained from Dr C Cepko (Dept. of Genetics, Harvard Medical School, Boston, MA, USA), and the expression vector pCMV (7, 2 kb) from Clontech (Heidelberg, Germany, Cat No. 6177-1; GenBank Accession No.: U02451). The expression of the lacZ gene is driven by the Moloney murine leukemia virus LTR promoter in the BAG vector, which also contains a neomycin resistance gene under the control of a SV40 promoter. In the pCMV vector lacZ expression is regulated by the CMV immediate–early gene promoter. Plasmids were grown in Escherichia coli and purified using the Qiagen Mega Prep Kit (Qiagen, Hilden, Germany). The DNA, with a 260:280 ratio from 1.8 to 2.0, was dissolved in water for storing. For in vitro injections, the DNA was freshly diluted and adjusted to 1 g/l in 1 × PBS final concentration. Mice Pathogen-free female DBA/2 mice were purchased from IFFA Credo (Lyon, France) and used at 6–12 weeks of age. Cell lines ESbL-lacZ, a highly metastatic lacZ-transfected lymphoma,28 P815, a mastocytoma, and its lacZ-transfected variant P13.137 were cultivated in RPMI-1640 medium, as described earlier.38 Peptide The synthetic nonamer TPHPARIGL represents the naturally processed H-2Ld-restricted epitope of -gal spanning amino acids 876–884.39 It was synthesized by R Pipkorn (DKFZ). Immunizations with DNA and tumor cells Fifty micrograms of DNA, dissolved in PBS, were injected either i.m. or i.d. into the shaved flank by using 27-gauge needles, or into the ear pinna of anesthetized animals using 32-guage needles. For i.m. injections a volume of 100 l was used, whereas for i.d. and i.e. 50 l volumes were administered. Anesthesia was carried out by i.p. injection of Rompun- (Parke Davis, Berlin, ¨ Germany) Ketanest- (Bayer, Leverkusen, Germany) PBS (1:1:3). For immunization with tumor cells, a subtumorigenic dose of 5 × 104 ESb or ESbL-lacZ cells in 50 l PBS was injected i.e. ELISA Blood samples were obtained from the tail vein of mice every second week. Sera were prepared and stored at 795 Intra-pinna DNA vaccination ¨ P Forg et al 796 −20°C until they were analyzed for anti--gal antibodies in ELISA. For ELISA, 96-well plates were coated overnight with purified -gal (Sigma, Deisenhofen, Germany) in 0.05 m carbonate buffer at 4°C. After washing with 0.05% Tween-20/PBS the plates were blocked with 0.2% gelatine in PBS for 1 h at 37°C. Sera were serially diluted in Tween/PBS and applied to the plates for 1 h at room temperature. After further washing steps, bound antibodies were detected with a peroxidase-conjugated goat anti-mouse IgG immunoglobulin (Dianova, Hamburg, Germany). The plates were washed and finally developed with ortho-phenyl-diamine substrate buffer and read at OD 492 nm. As a control, a monoclonal mouse anti--gal antibody (Boehringer Mannheim, Mannheim, Germany) was applied in different concentrations. Effector cells Animals were immunized with 50 g lacZ DNA i.m., i.d. or i.e., respectively. Nine to 28 days later, 2 × 107 spleen cells were restimulated in vitro for 5 days in RPMI medium containing 10% FCS and 0.5 g/ml TPHPARIGL. For the cell-based induction of lacZ immunity, 5 × 104 live ESbL-lacZ in PBS were inoculated i.e. and splenocytes were restimulated in vitro after 9 days. 51 Cr release assay The effector cells were tested for their cytotoxic activity in a standard 4 h 51Cr release assay at different effector:target ratios against 5 × 103 51Cr-labeled P13.1 and P815 cells, respectively, as described earlier.21,38 The amount of 51Cr released was measured in a gamma counter and the percentage of lysis was calculated from the formula: [(experimental cpm − spontaneous cpm)/ (maximum cpm − spontaneous cpm)] × 100. X-Gal staining Organs were removed from the animals and stained immediately by washing in PBS and incubating in fixation solution (2% formaldehyde/0.2% glutaraldehyde in PBS) on ice for 1 h. For this purpose, ear pinnae were peeled to separate the dorsal and ventral sides. After washing in PBS, organs were incubated in X-gal solution (5 mm K3Fe(CN)6, 5 mm K4Fe(CN)6, X 3 H2O), 2 mm MgCl2, 0.01% sodiumdesoxycholate, 0.02% NP40, 1 mg/ ml X-gal (Boehringer Mannheim) at 37°C for 3–16 h. Lymph nodes were snap-frozen in liquid nitrogen and cut with a cryostat (Leica, Bensheim, Germany), where on average every sixth piece was taken for analysis. After fixation for 10 min on ice and washing in PBS the slices were stained in X-gal solution as described. Acknowledgements The financial support of SmithKline Beecham Biologicals ¨ for Petra Forg is gratefully acknowledged. References 1 Wolff JA et al. Direct gene transfer into mouse muscle in vivo. Science 1990; 247: 1465–1468. 2 Pardoll DM, Beckerleg AM. Exposing the immunology of naked DNA vaccines. Immunity 1995; 3: 165–169. 3 Ulmer JB et al. Heterologous protection against influenza by injection of DNA encoding a viral protein. Science 1993; 259: 1745–1749. 4 Wang B et al. Gene inoculation generates immune responses against human immunodeficiency virus type 1. Proc Natl Acad Sci USA 1993; 90: 4156–4160. 5 Davis HL, Michel ML, Whalen RG. DNA-based immunization induces continuous secretion of hepatitis B surface antigen and high levels of circulating antibody. Hum Mol Genet 1993; 2: 1847–1851. 6 Sedegah M, Hedstrom R, Hobart P, Hoffman SL. Protection against malaria by immunization with plasmid DNA encoding circumsporozoite protein. Proc Natl Acad Sci USA 1994; 91: 9866–9870. 7 Conry RM et al. Immune response to a carcinoembryonic antigen polynucleotide vaccine. Cancer Res 1994; 54: 1164–1168. 8 Wang B et al. Immunization by direct DNA inoculation induces rejection of tumor cell challenge. Hum Gene Ther 1995; 6: 407– 418. 9 Raz E et al. Systemic immunological effects of cytokine genes injected into skeletal muscle. Proc Natl Acad Sci USA 1993; 90: 4523–4527. 10 Davis H, Whalen RG, Demeneix BA. Direct gene transfer into skeletal muscle in vivo: factors affecting efficiency of transfer and stability of expression. Hum Gene Ther 1993; 4: 151–159. 11 Sikes ML, O’Malley BW Jr, Finegold MJ, Ledley FD. In vivo gene transfer into rabbit thyroid follicular cells by direct DNA injection. Hum Gene Ther 1994; 5: 837–844. 12 Tang D, DeVit M, Johnston SA. Genetic immunization is a simple method for elicting an immune response. Nature 1992; 356: 152–154. 13 Rax E et al. Intradermal gene immunization: the possible role of DNA uptake in the induction of cellular immunity to viruses. Proc Natl Acad Sci USA 1994; 91: 9519–9523. 14 Hengge UR et al. Cytokine gene expression in epidermis with biological effects following injection of naked DNA. Nat Genet 1995; 10: 161–166. ¨ 15 Ciernik IF, Krayenbuhl BH, Carbone DP. Puncture-mediated gene transfer to the skin. Hum Gene Ther 1996; 7: 893–899. 16 Ulmer JB et al. Generation of MHC class I-restricted cytotoxic T lymphocytes by expression of a viral protein in muscle cells: antigen presentation by non-muscle cells. Immunology 1996; 89: 59–67. 17 Corr M, Lee DJ, Carson DA, Tighe H. Gene vaccination with naked plasmid DNA: mechanism of CTL priming. J Exp Med 1996; 184: 1555–1560. 18 Doe B et al. Induction of cytotoxic T lymphocytes by intramuscular immunization with plasmid DNA is facilitated by bone marrow-derived cells. Proc Natl Acad Sci USA 1996; 93: 8578–8583. 19 Schirrmacher V et al. Antigenic variation in cancer metastasis: immune escape versus immune control. Cancer Metastas Rev 1982; 1: 241–274. 20 Schirrmacher V, Jurianz K, Griesbach A. Intra-pinna induction of specific antitumor immune T cell functions: effect of ear resection after antigen application. Int J Oncol 1997; 11: 227–233. ¨ 21 Kruger A, Schirrmacher V, von Hoegen P. Scattered micrometastases visualized at the single-cell level: detection and re-isolation of lacZ-labeled metastasized lymphoma cells. Int J Cancer 1994; 58: 275–284. ¨ 22 Schirrmacher V, Schild HJ, Guckel B, von Hoegen P. Tumorspecific CTL response requiring interactions of four different cell types and dual recognition of MHC class I and class II-restricted tumor antigens. Immunol Cell Biol 1992; 71: 311–326. 23 Hengge U, Walker PS, Vogel JC. Expression of naked DNA in human, pig, and mouse skin. J Clin Invest 1996; 97: 2911–2916. 24 Watanabe A et al. Induction of antibodies to a kappa V region by gene immunization. J Immunol 1993; 151: 2871–2876. 25 Ertl HCJ, Xiang ZQ. Genetic immunization. Virol Immunol 1996; 9: 1–9. 26 Jurianz K, von Hoegen P, Schirrmacher V. Superiority of the ear pinna over a subcutaneous tumour inoculation site for induction of a Th1 type cytokine response. Cancer Immun Immunother 1998; 45: 327–333. Intra-pinna DNA vaccination ¨ P Forg et al 27 Condon C et al. DNA-based immunization by in vivo transfection of dendritic cells. Nature Med 1996; 2: 1122–1128. 28 Cheng L, Ziegelhoffer PR, Yang NS. In vivo promoter activity and transgene expression in mammalian somatic tissues evaluated by using particle bombardment. Proc Natl Acad Sci USA 1993; 90: 4455–4459. 29 Schirrmacher V, Zangemeister-Wittke U. ␥-Irradiation suppresses T cell-mediated protective immunity against a metastatic tumor in the afferent phase of the immune response but enhances it in the efferent phase when given before immune cell transfer. Int J Oncol 1994; 4: 335–346. ¨ 30 Bohm W et al. DNA vector constructs that prime hepatitis B surface antigen-specific cytotoxic T lymphocyte and antibody responses in mice after intramuscular injection. J Immunol Meth 1996; 193: 29–40. 31 Macatonia S et al. Localization of antigen on lymph node dendritic cells after exposure to the contact sensitizer fluorescein isothiocyanate. J Exp Med 1987; 166: 1654–1667. 32 Donnelly JJ, Ulmer JB, Liu MA. Immunization with DNA. J Immunol Meth 1994; 176: 145–152. 33 Pertmer TM, Roberts TR, Haynes JR. Influenza virus nucleoprotein-specific immunoglobulin G subclass and cytokine responses 34 35 36 37 38 39 elicited by DNA vaccination are dependent on the route of vector DNA delivery. J Virol 1996; 70: 6119–6125. Raz E et al. Preferential induction of a Th1 immune response and inhibition of specific IgE antibody formation by plasmid DNA immunization. Proc Natl Acad Sci USA 1996; 93: 5141–5145. Fynan EF et al. DNA vaccines: protective immunizations by parenteral, mucosal, and gene-gun inoculations. Proc Natl Acad Sci USA 1993; 90: 11478–11482. Price J, Turner D, Cepko CL. Lineage analysis in the vertebrate nervous system by retrovirus-mediated gene transfer. Proc Natl Acad Sci USA 1987; 84: 156–160. Carbone FR, Bevan J. Class I-restricted processing and presentation of exogenous cell-associated antigen in vivo. J Exp Med 1990; 171: 377–387. von Hoegen P, Weber E, Schirrmacher V. Modification of tumor cells by a low dose of Newcastle disease virus. Augmentation of the tumor-specific T cell response in the absence of an antiviral response. Eur J Immunol 1988; 18: 1159–1166. Gavin MA et al. Alkali hydrolysis of recombinant proteins allows for the rapid identification of class I MHC-restricted CTL epitopes. J Immunol 1993; 151: 3971–3980. 797