Designing gene therapy vectors: avoiding immune responses by using tissue-specific promoters


Attempts to correct genetic disorders by gene therapy have been hindered by various problems including unwanted immune responses against the gene product. It has been shown that immune responses with DNA vaccines after i.m. injection of antigen-encoding plasmid DNA are primed solely by professional antigen-presenting cells (APC), even though myocytes are the primary type of cell transfected. This possibly involves direct transfection of some APC in regional lymph nodes draining the injected muscle. Here we have used plasmid DNA vaccines that express hepatitis B surface antigen (HBsAg) to evaluate the possibility of abrogating these immune responses by use of a tissue-specific promoter that does not drive expression in APC. We show that HBsAg-specific humoral or cell-mediated responses are not induced in mice when the muscle-specific human muscle creatine kinase promoter is used in place of the ubiquitous cytomegaloviral promoter to drive expression of HBsAg. This may have significance in the field of gene therapy where one aims to achieve stable expression of the desired gene product without interference from the host immune response.


Gene therapy is a relatively new and evolving paradigm for treating human illness. At its inception, it was considered that gene therapy would treat genetic disorders, such as inborn errors of metabolism, by expressing a therapeutic gene for augmentation or replacement of a defective homologous gene, in order to correct the disease phenotype. The field of gene therapy now also includes the treatment of chronic infections and cancer by expression of antigenic proteins for the purpose of inducing immune responses that will clear the infection or destroy the neoplastic cells. However, for the purpose of this article, strategies where immune responses are desired will be called DNA-based immunization and the term gene therapy will be restricted to those situations where long-term stable expression of the therapeutic gene product is desired.

With gene therapy, the therapeutic gene product may be seen as foreign by the immune system and resulting immune responses can destroy transfected cells or neutralize the therapeutic gene product. This is particularly true in cases where the genetic abnormality results in a complete absence of the normal structural or functional protein, but it can also occur in cases where a truncated or otherwise abnormal gene product is missing an epitope normally found in the complete protein. Evidence for unwanted immune responses in gene therapy comes from results in animal models. Adenovirus-mediated transfer of a dystrophin mini-gene into dystrophin-deficient mice (mdx mice) is capable of correcting morphological and physiological abnormalities in the muscle fibers. However, dystrophin expression was found to be transient unless immunosuppression was used.1 Similar observations have been made in other disease models such as cystic fibrosis (CF)2 and hemophilia A.3 It is clear that loss of the therapeutic gene expression and transfected cells is immune mediated, since such losses do not occur in immunocompromised animals.456

The knowledge gathered from the field of DNA-based immunization may provide important insight for the field of gene therapy. It was initially thought that the efficacy of DNA vaccines was by virtue of in vivo synthesis of antigen in transfected cells, for example muscle cells after intramuscular (i.m.) injection of plasmid DNA. However, it was subsequently shown that only professional antigen-presenting cells (APC) could actually prime immune responses with DNA vaccines.78910 While direct transfection of APC is not absolutely essential,11 it is probable that some APC are transfected and antigen expression in these APC primes MHC class I and II responses. This may account for the strong immune responses induced by DNA vaccines, especially considering the extremely small quantities of antigen expressed from typical doses of plasmid DNA. Furthermore, most DNA-based immunization strategies use strong viral promoters to ensure high levels of expression. These promoters are ubiquitously expressed and will drive antigen expression in a wide range of cell types, including APC. Therefore by using strategies that avoid gene expression in APC, it may be possible to abrogate unwanted immune responses against a desired gene. We have evaluated this possibility using a modification of our well-established DNA vaccine model whereby expression of HBsAg under the control of a ubiquitously expressed promoter induces strong antibodies against HBsAg (anti-HBs) and cytotoxic T lymphocytes (CTL) responses.1213141516 Herein we show that expression of the same highly antigenic HBsAg protein does not elicit, or even prime, any immune responses when expressed under the control of muscle-specific creatine kinase (MCK) promoters that express poorly or not at all in cells of the lymphocytic lineage.


Tissue specificity of CMV and MCK promoters in vitro

Luciferase reporter expressed under the control of CMV or MCK promoters in the mouse myoblast/myotube (muscle, C2C12) and mouse monocyte/macrophage (RAW264.7) cell lines were used to assess the tissue specificity of these promoters.

When tested in vitro, the CMV promoter drove strong expression of luciferase in both mouse myoblast/myotube (muscle, C2C12) and mouse monocyte/macrophage (RAW264.7) cell lines (assayed 2 days after transfection). A given dose of DNA gave approximately 100-fold higher expression in the C2C12 than RAW264.7 cells (Figure 1).

Figure 1

Luciferase expression in C2C12 cells (murine myoblast; gray bars) and RAW264.7 cells (murine monocyte/macrophage; black bars) 48 h after transfection with pCMV-luc (1 or 5 μg), pMCK-luc (5 or 10 μg) or pΔMCK-luc (5 or 10 μg). The results are expressed as RLU/s/mg protein. The experiments were done in duplicate and each bar represents the average luciferase levels from the two experiments.

In contrast, the pMCK-luc vector gave high luciferase activity in C2C12 cells, but no detectable activity in RAW264.7 cells, indicating a strong muscle tissue-specificity of the MCK promoter/intron region. Luciferase reporter gene expression in muscle cells with 5 μg pMCK-luc was nearly as high as with 1 μg pCMV-luc Figure 1, indicating very strong activity since CMV is one of the strongest promoters known. However, the reduction in luciferase expression seen with pMCK-luc might also be attributed to the decrease in transfection efficiency due to its large size. In order to rectify this, the intron following the MCK promoter region was deleted from the expression vectors (ΔMCK).

The pΔMCK-luc vector gave approximately five-fold higher luciferase levels than pMCK-luc in the muscle cell line, but lost some of its muscle specificity as there were also detectable levels of luciferase in the macrophage cell line.

Activity of CMV and MCK promoters in vivo

Each of pCMV-luc, pMCK-luc and pΔMCK-luc vectors gave high levels of luciferase reporter gene activity in muscle in vivo (Figure 2a). Luciferase levels in TA muscles removed 3 days after injection were significantly higher with 10 μg of pCMV-luc than with 10 μg of either pMCK-luc or pΔMCK-luc (P < 0.001). However, there was no significant difference between levels with 10 μg of pCMV-luc and 100 μg of pΔMCK-luc (P = 0.1742) or 5 μg of pCMV-luc and 10 μg of pMCK-luc (P = 0.8564). Notably, luciferase levels in muscles injected with 5 μg of pCMV-luc were significantly lower than in the muscles injected with either 10 or 100 μg of pΔMCK-luc (P = 0.0335 and P = 0.0013).

Figure 2

(a) Comparison of CMV- and muscle-specific promoters (MCK and ΔMCK) for their ability to express the luciferase reporter gene after direct injection into muscles of mice. TA muscles of BALB/c mice were injected with 5 μg (light gray), 10 μg (dark gray bars) or 100 μg (black bars) of pCMV-luc (left panel), pMCK-luc (middle panel) or pΔMCK-luc (right panel). Muscles were removed 3 days after injection and assayed for luciferase expression. Results are expressed as RLU/sec/muscle. Each bar represents the mean luciferase levels in 10 muscles and T-bars represent the s.e.m. (b) Humoral responses in BALB/c mice injected with 5 μg (light gray), 10 μg (dark gray bars) or 100 μg (black bars) of pCMV*-S (left panel), pMCK-S (middle panel) or pΔMCK-S (right panel). Each bar represents the group mean (n = 5-10) for titers of anti-HBs (total IgG) at 4 weeks after injection, as determined in triplicate by end-point dilution ELISA assay. End-point titers were defined as the highest plasma dilution that resulted in an absorbance value (OD 450) two times greater than that of control non-immune plasma with a cut-off value of 0.05. The numbers above the individual bars indicate the number of animals seroconverted/total number of animals injected.

Expression of HBsAg by CMV and MCK promoters in vitro

Before use in vivo, all of the plasmid constructs, pCMV*-S, pMCK-S and pΔMCK-S, were tested and confirmed for expression of HBsAg in C2C12 cells in vitro (Figure 3). The plasmid pCMV*-S is identical to the plasmid reported earlier as pMAS-S. However, we use the nomenclature pCMV*-S in this manuscript to emphasize the use of CMV promoter in this vector to drive expression of HBsAg.

Figure 3

Expression of HBsAg in C2C12 (murine myoblast cells) transfected with pCMV (upper left panel), pCMV*-S (upper right panel), pMCK-S (lower left panel) or pΔMCK-S (lower right panel). C2C12 cells were transfected with 10 μg of DNA. Expression of HBsAg was detected by immunoassay using mouse anti-HBs polyclonal antibody and fluorescein-labeled sheep anti-mouse IgG (F[ab’]2 fragment). The cells were examined by confocal microscopy for HBsAg expression, which is evident as green fluorescence.

Immune responses against HBsAg in mice

Humoral response:

As shown previously,15 animals injected with the plasmid expressing HBsAg under the control of the CMV promoter (pCMV-S) showed 100% seroconversion (ELISA titers >100) by 2 weeks after injection with the 100-μg dose and by 4 weeks with the 10-μg dose. In contrast, seroconversion was not observed by 4 weeks after injection in animals injected with 10 or 100 μg of the vectors expressing HBsAg under the control of the muscle creatine kinase promoter, both with (pMCK-S) and without the first intron (pΔMCK-S) Figure 2b. Seroconversion was only ever detected in one of six animals injected with 10 μg pMCK-S (end point titer of 200 at 8 weeks). Thus, only one of 32 animals injected with 10 or 100 μg of either pMCK-S (n = 11) or pΔMCK-S (n = 20) seroconverted. When 16 of the non-responding animals were ‘challenged’ with 1 μg of recombinant HBsAg protein at 8 weeks, all mice exhibited a typical primary response, similar to that seen in naive mice receiving HBsAg for the first time. This primary response, which was characterized by no seroconversion by 7 days and only weak responses (mean titers <200) by 14 days, indicated that HBsAg-specific responses had not even been primed by the plasmids driven by the muscle-specific promoters (Figure 4). In contrast, mice that had been primed by injection of pCMV-S at 7 days of age in the presence of maternal anti-HBs antibody, but which had no detectable anti-HBs after the maternal antibodies had waned (by 12 weeks), showed a strong anamnestic response upon subsequent ‘challenge’ with HBsAg. In this case, seroconversion was evident by 3 days and high titers (>1000) were detectable by 7 days Figure 4.

Figure 4

Humoral responses in animals injected (challenged) with HBsAg at 8 or 12 weeks after primary injection with plasmid DNA. In order to investigate the anamnestic humoral immune responses, BALB/c mice that had not seroconverted with previous injection of HBsAg-expressing DNA (pMCK-S, striped bars; or pΔMCK-S, dotted bars) were injected i.m. with 1 μg of recombinant HBsAg at 8 weeks after primary injection. Naive animals (ie no previous treatment) were used as a negative controls (gray bars). As a positive control (black bars), mice that had been immunized with pCMV-S at 7 days of age in the presence of maternal anti-HBs were injected with HBsAg at 12 weeks after primary injection (no detectable maternal antibody was present at the time of HBsAg injection). Each bar represents the mean (n = 5–10 per group) for anti-HBs (total IgG) titers at 3, 7 and 14 days after challenge as determined in triplicate by end-point dilution ELISA assay. End-point titers were defined as the highest plasma dilution that resulted in an absorbance value (OD 450) two times greater than that of control non-immune plasma with a cut-off value of 0.05.

Cell-mediated response:

Splenocytes recovered from mice 4 weeks after i.m. injection with 5, 10 and 100 μg of pCMV*-S showed HBsAg-specific lysis of 72%, 71% and 45%, respectively, at an E:T ratio of 50:1. In contrast, animals injected with 100 μg of either pMCK-S or pΔMCK-S did not show levels of CTL activity above non-injected controls (<20% lysis at an E:T ratio of 50:1) (Figure 5).

Figure 5

Cytotoxic T lymphocyte responses in BALB/c mice injected with pCMV*-S, pMCK-S or pΔMCK-S. Mice were injected with 5, 10 or 100 μg of pCMV*-S (black bar), 100 μg of pMCK-S (striped bar) or 100 μg of pΔMCK-S (dotted bar) and spleens were removed at 4 weeks after injection for CTL assay. Each bar represents the mean % specific lysis from five animals ± s.e.m.


We have evaluated whether immune responses against a foreign protein expressed after i.m. injection of plasmid DNA would be abrogated by use of a tissue-specific promoter that would not be active in APC. Herein we show that expression of HBsAg, a highly antigenic protein, under the control of the human muscle-specific promoter (MCK) with or without the intron region, does not induce humoral or cell-mediated immune responses against HBsAg in mice. Furthermore, it appears that antigen-specific immune responses were not even primed with these vectors, since animals do not show anamnestic antibody responses upon subsequent challenge with recombinant HBsAg. In contrast, all animals injected with the pCMV*-S vector, where the ubiquitous CMV promoter drives expression of HBsAg, developed strong antibody and CTL against HBsAg. This was true even at doses where less gene product would have been produced from the CMV than the MCK plasmids.

There have been several studies on DNA vaccines reporting that only professional APC can prime immune responses with DNA-based immunization.78910 Other studies have shown increased efficacy of DNA vaccines with ex vivo transfection of dendritic cells (DC).1718 The finding that transfected muscle cells cannot prime immune responses was not unexpected since they do not normally express class II MHC molecules or the co-stimulatory molecules necessary to activate T cells.19 However, muscle cells are efficiently transfected by plasmid DNA,2021 and thus they may help boost immune responses once they are primed. Even this does not appear to be necessary since immediate removal of a muscle immediately after i.m. injection of a DNA vaccine does not diminish immune responses.22 These results suggest that some of the plasmid DNA injected i.m. exits the muscle immediately. This might be via the lymphatic system, and indeed it has been shown that after i.m. injection, some cells in the draining lymph nodes are transfected.1723 A study by Loirat et al24 showed that immunizing mice with a DNA vaccine expressing HBsAg under the control of a muscle-specific promoter derived from human desmin gene resulted in both humoral and CTL immune responses against HBsAg. However, the authors in this study have not confirmed the muscle specificity of this promoter and thus it is possible that the HBsAg is also expressed in APC transfected with the DNA, which are then involved in the induction of immune responses. Thus, although the absolute necessity for transfection of APC is still controversial, it is likely that expression of a foreign protein from a DNA vaccine within normal host cells that are not APC will not induce immune responses. Indeed, in the present study we show that expression of antigen in mature muscle fibers does not induce immune responses.

It is possible that the failure to induce immune responses in the present study is simply due to expression of an inadequate quantity of antigen with the MCK promoters compared with the stronger CMV promoter. However, this is unlikely to be the case since i.m. injection of 100 μg of the pΔMCK-luc gave the same level of luciferase activity in muscle as 10 μg of pCMV-luc, yet there were no immune responses with 100 μg of pΔMCK-S and very strong responses with 10 μg of pCMV*-S. Furthermore, we have been able to detect both antibody and CTL responses in mice injected with 5 μg pCMV*-S, even though 5 μg of pCMV-luc gave a considerably lower level of luciferase expression than 10 or 100 μg of pΔMCK-luc. We have previously shown a strong correlation between the level of luciferase expression by pCMV-luc (which indicates the efficiency of muscle cell transfection and promoter activity) and anti-HBs humoral immune responses induced by pCMV-S.25

As mentioned earlier, an important problem for gene therapy applications today is unwanted immune responses against the therapeutic gene product. We have previously shown immune-mediated destruction of virtually all HBsAg-expressing muscle fibers following i.m. injection of HBV DNA vaccines, and this occurs largely between 10 and 20 days after gene transfer.15 Thus the results of this study may have important clinical implications for the gene therapy field. However, it will be necessary to identify tissue-specific promoters that allow high levels of expression in the cells of choice.

Materials and methods

Plasmid construction

Starting plasmids:

A 2.7-kb DNA fragment containing the human muscle creatine kinase (MCK) promoter was provided by Dr Richard J Bartlett (University of Miami). This fragment contains a 5’ BamHI site and 3’ HindIII site before the splice donor site for the first intron. Plasmid p31C, which contains the MCK first intron subcloned into the original TA cloning vector pCRTM2.1 (Invitrogen, Carlsbad, CA, USA) was also provided by Dr Bartlett. The start codon within the MCK first intron on pC31 was mutated to the unique restriction enzyme site SalI. The cloning vector pUK21, which contains a kanamycin resistance gene and multiple cloning sites, was provided by Dr Martin Schleef (Qiagen, Hilden, Germany). The luciferase reporter vector pGL2-Basic was purchased from Promega (Madison, WI, USA). The plasmid pMAS-S (GeneBank accession number AFO53407), expresses the small hepatitis B surface antigen (HBsAg, ay subtype) under the control of the human cytomegalovirus (CMV) major intermediate-early promoter/enhancer region.26 In order to emphasize the presence of the CMV promoter, pMAS-S will be referred to as pCMV*-S throughout this article. E. coli strain DH5α was used as the bacterial host.

Expression vectors with the MCK promoter/intron region:

The 2.7 kb DNA fragment containing the MCK promoter was cloned into pUK21 using the BamHI and HindIII sites, creating pUK21-MCK. p31C was digested with KpnI, blunt-ended, and then digested with SphI to excise the MCK first intron. pUK21-MCK was digested with XhoI, blunt-ended, and then digested with SphI for insertion of the MCK intron. Insertion of the intron fragment into the pUK21-MCK created the plasmid pUK21-MCKIN, which contains the 5.9-kb full-length MCK promoter and the first intron. The 5.9 kb promoter/intron region was then excised from pUK21-MCKIN using EcoRV and SalI sites, blunt-ended, and cloned into pGL2-Basic using blunt-ended HindIII and SacI sites. The resulting plasmid, pMCK-luc, expresses luciferase driven by the full-length MCK promoter. The CMV promoter was removed from pCMV*-S using HpaI and EcoRI sites, blunt-ended, and then replaced with the 5.9-kb blunt-ended MCK promoter/intron region. The resulting plasmid, pMCK-S, expresses the HBsAg under the control of the MCK promoter and first intron.

Expression vectors with the MCK promoter:

The 2.7-kb MCK promoter was excised from pUK21-MCK with EcoRV and HindIII, and then cloned into pGL2-Basic using SmaI and HindIII sites. The resulting plasmid, pΔMCK-luc, expresses luciferase driven by the MCK promoter. The 2.7-kb MCK promoter fragment containing terminal EcoRV and HindIII sites was blunt-ended, then used to replace the CMV promoter of pCMV*-S, creating pΔMCK-S, which expresses HBsAg under the control of the MCK promoter.

DNA production:

All plasmid DNA was produced by growing transformed E. coli cells (DH5-α) in LB media and then purified on Qiagen DNA purification columns (Qiagen). Recovered DNA was redissolved in 0.15 M NaCl and stored at or below −20°C.

In vitro assays of gene expression

Luciferase expression:

C2C12 (murine myoblast, ATCC No. CRL-1772, ATCC, Manassas, VA, USA) and RAW264.7 cells (murine monocyte/macrophage, ATCC No. TIB-71) were plated in six-well plates at 2 × 105 cells/ml. The C2C12 cells were grown in Dulbecco's modified Eagle media (D-MEM) (Gibco-BRL, Burlington, ON, Canada) containing 10% horse serum (Gibco-BRL) in order to enhance myocyte differentiation and myotube formation. The RAW264.7 cells were grown in D-MEM containing 10% fetal bovine serum (Gibco-BRL). The cells were grown to approximately 60% confluency and then transfected with pCMV-luc (1 or 5 μg), pMCK-luc (5 or 10 μg) or pΔMCK-luc (5 or 10 μg) using SuperFect transfection reagent following the protocol provided by the manufacturer (Qiagen). Each transfection was performed in duplicate. Forty-eight hours following the transfection, the cells were isolated and luciferase activity was determined using the Luciferase Assay System according to the manufacturer's suggested protocol (Promega, Madison, WI, USA). Protein content was determined by the microassay method using the BioRad Protein Assay Reagent (Hercules, CA, USA). Levels of luciferase activity were expressed as relative light units (RLU)/s/mg protein.

HBsAg expression:

C2C12 and RAW264.7 cells were grown in six-well plates as described in the previous section with the exception that sterile coverslips (1 oz micro cover glasses; VWR Scientific, Media, PA, USA) were placed on the bottom of the wells. Cells were allowed to grow to approximately 60% confluency and then transfected with pCMV*-S, pMCK-S and pΔMCK-S plasmid DNA (10 μg) using SuperFect as described above. These cell lines were also transfected with pCMV* backbone vector, that is without the S gene, for use as a negative control. Forty-eight hours following transfection, the coverslips were removed and washed three times by immersion for 5 min in 2 ml of sterile PBS. The coverslips were then air-dried and cells were fixed for 2.5 min with ice cold acetone and washed a further three times with 2 ml of sterile PBS. After this, the cells were blocked for 15 min using 60 μl/coverslip of fetal bovine serum to avoid non-specific interactions, and then washed again three times using 2 ml of sterile PBS. The cells were then incubated for 30 min at room temperature (RT) with 60 μl/coverslip of the primary antibody (anti-HBs in pooled plasma from BALB/c mice immunized with recombinant HBsAg; each animal had a total IgG titer >150 000). Following this incubation, the cells were washed three times with sterile PBS. The cells were then incubated for 30 min at RT with 60 μl/coverslip of secondary antibody sheep anti-mouse Ig-Fluorescein F[ab’]2 fragment (Boehringer, Mannheim, Germany). The coverslips were washed a final time and mounted on a glass microscope slide (Frosted microscope slides; Fisher Scientific, Pittsburgh, PA, USA) using Permount (Fisher Scientific, Fair Lawn, NJ, USA). The cells were examined by confocal microscopy for HBsAg expression, which was evident as green fluorescence.

In vivo evaluation of gene transfer and immune responses

Direct gene transfer in muscles of mice:

All experiments were carried out using female BALB/c mice (Charles River, Montreal, Canada) at 6–10 weeks of age. All experimental groups comprised 6–10 animals. Mice were maintained at the animal care facility at Loeb Health Research Institute. The DNA solutions were administered by i.m. injection into the left tibialis anterior (TA) muscles in a total volume of 50 μl as previously described.27

Evaluation of in vivo expression of luciferase:

TA muscles were removed from mice 3 or 6 days after injection of DNA, and luciferase activity was assayed as previously described.27 Luciferase activity was reported as group means ± s.e.m. in relative light units (RLU)/s/muscle.

Evaluation of humoral responses to HBsAg:

Mice were bled at regular time intervals after injection as described elsewhere12 and the plasma was recovered and stored at 4°C until assayed. Antibodies (total IgG) specific to HBsAg (anti-HBs) were detected and quantified by endpoint dilution ELISA assay which was performed in triplicate on samples from individual animals.14 End-point titers were defined as the highest plasma dilution that resulted in an absorbance value (OD 450) two times greater than that of non-immune plasma with a cut-off value of 0.05. These were reported as group mean titers ± s.e.m. Seroconversion was defined as titers >100.

In order to investigate the anamnestic humoral immune responses, animals previously injected with HBsAg-expressing DNA were injected i.m. (into the right TA muscle) with 1 μg of recombinant HBsAg (Genzyme, San Carlos, CA, USA) in a total volume of 50 μl. As a negative control, some naive (ie no previous treatment) animals were injected with HBsAg to determine the characteristics (kinetic and magnitude) of a primary response. As a positive control, mice that had been immunized with pCMV*-S (10 μg in 20 μl bilaterally into the posterior thigh muscles) at 7 days of age in the presence of maternal anti-HBs were, as adults (once maternal antibody had been lost), injected with HBsAg. We have previously shown this to prime a response in most animals, even though no detectable anti-HBs remains once the maternal antibodies diminish.28

Evaluation of CTL responses to HBsAg:

From all experimental groups, five animals were killed and spleens were removed at 4 weeks after injection and used for CTL assay, which were conducted as previously described.29 Briefly, splenocytes were restimulated for 5 days using mouse mastocytoma cells expressing HBsAg (p815-S) and CTL activity was measured by 51Cr release assay. 51Cr-labeled P815-S cells were used as targets. Wild-type P815 cells were used as controls in the 51Cr release assay. The culture medium used for restimulation was RPMI 1640 supplemented with 10% fetal bovine serum (Gibco/BRL), penicillin-streptomycin solution (final concentration of 1000 U/ml and 1 mg/ml, respectively; Sigma, Irvine, UK), 5 × 10-5 M β-mercaptoethanol (Sigma) and 10 U/ml mouse recombinant IL-2 (Sigma). The results are presented as % specific lysis at an effector:target (E:T) ratio of 50:1, which was representative of relative results obtained at other E:T ratios.

Statistical analysis:

Statistical analysis was performed using InStat program (Graph PAD Software, San Diego, CA, USA). The statistical difference between groups was determined by Student's t test (for two groups) or by 1-factor ANOVA followed by Tukey's test (for three or more groups) on raw data or transformed data (log10, for heteroscedastic populations).


  1. 1

    Ohtsuka Y et al. Dystrophin acts as a transplantation rejection antigen in dystrophin-deficient mice: implication for gene therapy J Immunol 1998 160: 4635–4640

  2. 2

    van Ginkel F et al. Adenoviral gene delivery elicits distinct pulmonary-associated T helper cell responses to the vector and to its transgene J Immunol 1997 159: 685–693

  3. 3

    Evans G, Morgan R . Genetic induction of immune tolerance to human clotting factor VIII in a mouse model for hemophilia A Proc Natl Acad Sci USA 1998 95: 5734–5739

  4. 4

    Kass-Eisler A et al. Circumventing the immune response to adenovirus-mediated gene therapy Gene Therapy 1996 3: 154–162

  5. 5

    Wells K et al. Immune responses, not promoter inactivation, are responsible for decreased long-term expression following plasmid gene transfer into skeletal muscle FEBS Lett 1997 407: 164–168

  6. 6

    Petrof BJ . Respiratory muscles as a target for adenoviral-mediated gene therapy Eur Respir J 1998 11: 492–497

  7. 7

    Corr M, Lee D, Carson D, Tighe H . Gene vaccination with naked plasmid DNA: mechanism of CTL priming J Exp Med 1996 184: 1555–1560

  8. 8

    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

  9. 9

    Iwasaki A et al. The dominant role of bone-marrow derived cells in CTL induction following plasmid DNA immunization at different sites J Immunol 1997 159: 11–14

  10. 10

    Ulmer JB et al. Generation of MHC class-I restricted T lymphocytes by expression of a viral protein in muscle cells: antigen presentation by non-muscle cells Immunology 1996 89: 59–67

  11. 11

    Fu T et al. Priming of cytotoxic T lymphocytes by DNA vaccines: requirement for professional antigen presenting cells and evidence for antigen transfer from myocytes Mol Med 1997 3: 362–371

  12. 12

    Davis HL, Michel M-L, Whalen RG . DNA based immunization induces continuous secretion of for hepatitis B surface antigen and high levels of circulating antibody Hum Mol Genet 1993 2: 1847–1851

  13. 13

    Davis HL, Mancini M, Michel M-L, Whalen RG . DNA-mediated immunization to hepatitis B surface antigen: longevity of primary response and effect of boost Vaccine 1996 14: 910–915

  14. 14

    Davis HL et al. CpG DNA is a potent enhancer of specific immunity in mice immunized with recombinant hepatitis B surface antigen J Immunol 1998 160: 870–876

  15. 15

    Davis HL, Brazolot Millan CL, Watkins SC . Immune-mediated destruction of transfected muscle fibers after direct gene transfer with antigen-expressing plasmid DNA Gene Therapy 1997 4: 181–188

  16. 16

    Michel ML et al. DNA-mediated immunization to the hepatitis B surface antigen in mice: Aspects of the humoral response mimic hepatitis B viral infection in humans Proc Natl Acad Sci USA 1995 92: 5307–5311

  17. 17

    Chattergoon MA, Robinason TM, Boyer JD, Weiner DB . Specific immune induction following DNA-based immunization through in vivo transfection and activation of macrophages/antigen-presenting cells J Immunol 1998 160: 5707–5718

  18. 18

    Manickan E et al. Enhancement of immune response to naked DNA vaccine by immunization with transfected dendritic cells J Leuk Biol 1997 61: 125–132

  19. 19

    Hohlfeld R, Engel AG . The immunobiology of muscle Immunol Today 1994 15: 269–274

  20. 20

    Wolff JA et al. Direct gene transfer into mouse muscle in vivo Science 1990 247: 1465–1468

  21. 21

    Wolff JA et al. Long-term persistence of plasmid DNA and foreign gene expression in mouse muscle Hum Mol Genet 1992 1: 363–369

  22. 22

    Torres C, Iwasaki A, Barber B, Robinson H . Differential dependence on target site tissue for gene gun and intramuscular DNA immunizations J Immunol 1997 158: 4529–4532

  23. 23

    Condon C et al. DNA-based immunization by in vivo transfection of dendritic cells Nat Med 1996 2: 1122–1128

  24. 24

    Loirat D et al. Muscle-specific expression of hepatitis B surface antigen: no effect on DNA-raised immune responses Virology 1999 260: 74–83

  25. 25

    Loirat D et al. Muscle-specific expression of hepatitis B surface antigen: no effect on DNA-raised immune responses

  26. 26

    Krieg AM et al. Sequence motifs in adenoviral DNA block immune activation by stimulatory CpG motifs Proc Natl Acad Sci USA 1998 95: 12631–12636

  27. 27

    Davis HL, 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

  28. 28

    Brazolot Millan CL et al. CpG DNA can induce strong Th1 humoral and cell-mediated immune responses against hepatitis B surface antigen in young mice Proc Natl Acad Sci USA 1998 95: 15553–15558

  29. 29

    McCluskie MJ, Davis HL . CpG DNA is a potent enhancer of systemic and mucosal immune responses against hepatitis B surface antigen with intranasal administration to mice J Immunol 1998 161: 4463–4466

Download references


We are most grateful to Amanda Boyd, Lorraine Hamblin and Yu Xu for technical assistance. We also thank Dr Yubo Ren for his help with confocal microscopy and Dr Richard Bartlett for providing plasmids (see Materials and methods).

Author information

Correspondence to RD Weeratna.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Weeratna, R., Wu, T., Efler, S. et al. Designing gene therapy vectors: avoiding immune responses by using tissue-specific promoters. Gene Ther 8, 1872–1878 (2001) doi:10.1038/

Download citation


  • DNA vaccines
  • tissue-specific promoters
  • gene therapy applications

Further reading