A cutaneous gene therapy approach to treat ... - Semantic Scholar

©2004 FASEB

The FASEB Journal express article 10.1096/fj.04-1515fje. Published online September 29, 2004.

A cutaneous gene therapy approach to treat infection through keratinocyte-targeted overexpression of antimicrobial peptides Marta Carretero,* Marcela del Río,* Marta García,* María José Escámez,* Isabel Mirones,* Luis Rivas,† Cristina Balague,‡ Jose Luis Jorcano,* and Fernando Larcher* *Epithelial Damage, Repair and Tissue Engineering, Ciemat-Fundación Marcelino Botín; † Centro de Investigaciones Biologicas (CSIC), Madrid; and ‡Almirall Prodesfarma, S.A., Barcelona, Spain Corresponding author: Fernando Larcher, CIEMAT, Avenida Complutense 22, Edificio 7, 28040 Madrid, Spain. E-mail: [email protected] ABSTRACT Infection represents a major associated problem in severely burned patients, as it causes skin graft failure and increases the risk of mortality. Topical and systemic antibiotic treatment is limited by the appearance of resistant bacterial strains. Antimicrobial peptides (AMPs) are geneencoded “natural antibiotics” that form part of the innate mechanism of defense and may be active against such antibiotic-resistant microorganisms. Several microbicidal peptides are expressed in human skin under inflammatory conditions, and their function is not only limited to microbial killing but also influences tissue repair and adaptive immunity. Protein delivery through cutaneous gene therapy is a promising therapeutic tool for both skin and nonskin diseases. Here we present a gene transfer approach aimed at delivering antimicrobial peptides from keratinocytes. Adenoviral vectors encoding antimicrobial peptide genes were used to infect human keratinocytes growing either on plastic or as part of cultured skin equivalents. Inhibition of bacterial growth occurred both in conditioned media and in direct contact with AMPs genetransduced keratinocytes. In addition, we showed cooperative effects after transfer of combinations of genes encoding for AMPs with structural differences. Combined cutaneous tissue engineering in conjunction with (microbicidal) gene therapy emerges as a tailored therapeutic approach that is useful for wound coverage and, in this case, concomitantly combating infection. Key words: AMP ● burn wounds ● mulitdrug resistance

S

kin provides a protective barrier against water loss and infection. Massive skin losses occurring after severe burns need to be recovered using autologous skin grafts either from whole donor sites or through cutaneous tissue engineering. A fibrin-based dermoepidermal equivalent developed in our laboratory has been demonstrated to be suitable for the definitive coverage of severely burned patients (1, 2). However, damaged tissue after a burn injury facilitates the multiplication of microorganisms due to the anaerobic and nutrient environment at the wound site. Moreover, generalized immunosuppresion also contributes to Page 1 of 19 (page number not for citation purposes)

microbial growth and dissemination in these patients, infection being the most important cause of skin graft failure and even death. The continuous appearance of multidrug-resistant microbes limits the effectiveness of antibiotic treatment (3), and most of these agents inhibit wound healing at doses that are effective against pathogens and affect viability of keratinocytes in the grafts. For this reason, developing novel therapeutic strategies to control the spread of infection will be important to improve graft take and survival. Antimicrobial peptides are major components of the innate defense system from plants to vertebrates. They are synthesized as pre-propeptides that are processed by proteolytic cleavage releasing the active peptides, which present microbicidal activity against Gram-positive and negative bacteria, yeast, and some enveloped viruses. Several antimicrobial peptides have been isolated from human skin where they play an important role in innate immune defense (4). Among them, HBD-2 and HBD-3, both belonging to the β-defensin family of antimicrobial peptides, and the human cathelicidin LL-37 are expressed in the skin in response to inflammation. All of them show a wide spectrum of activity against pathogens. A synergistic activity between peptides has also been described previously (5). The lack of HBD-2 expression in burned skin and in burn blister fluid might be an important factor influencing microbial growth (6). In addition to the role of LL-37 in anti-microbial defense (7), it has also been demonstrated that this peptide is involved in re-epithelization of skin wounds (8). Additional benefits from these peptides came from their anti-endotoxic activity (9), synergy with classical antibiotics or the cationic innate defense protein lysozyme (10), and action on the immune system as recently described for the involvement of LL-37 in the maturation of dendritic cells quite abundant in epidermis (11). Topical and systemic administration of these peptides is currently being developed by several biotechnological companies (12). Moreover, several reports have pointed out the benefits of an antimicrobial gene therapy for the correction of an AMP defect (13, 14). Protein delivery through cutaneous gene therapy for both systemic and local therapeutic application remains a promising approach for the treatment of pathological skin and nonskin disorders (15, 16) and could be particularly suitable for AMPs delivery in order to overcome problems related to inefficiency and high cost production of standard antimicrobial compounds in the context of skin wounds. In this study, we have demonstrated the efficacy of genetically modified keratinocytes transiently overexpressing various antimicrobial peptides, either alone or in combination, in preventing bacterial growth. In addition, we show efficient antimicrobial activity of AMPs gene-transduced keratinocytes in a skin equivalent setting. MATERIALS AND METHODS Cell culture and preparation of fibrin gels The HaCaT human keratinocyte cell line was grown in Dulbecco’s modified Eagle’s medium with Glutamax (Gibco-BRL, Gaithersburg, MD), supplemented with 10% FCS. Human dermal fibroblasts were obtained according to Meana et al. (1). To generate a dermoepidermal equivalent, 1.5 ml of fibrinogen solution (from cryoprecipitated pig blood) were added to 5 ml of keratinocyte growth medium containing 2.5 × 105 dermal fibroblasts. Immediately after, 0.5 ml Cl2Ca, 0.025 mM with 5.5 IU of bovine thrombin (Sigma-Aldrich Co., St. Louis, Page 2 of 19 (page number not for citation purposes)

MO) was added. The mixture was placed on polycarbonate inserts (4 µM porous) in a 6-well culture plate (Corning Costar Corp., Cambridge, MA) and allowed to solidify at 37°C for 2 h; 106 HaCaT cells were seeded on the fibrin matrix and grown to confluence. Construction of adenoviral vectors Complete sequences of antimicrobial peptides (including signal sequence and pro-region) were employed for the construction of adenoviral vectors. LL-37 cDNA was obtained from G.H. Gudmundsson (Karolinska Institute, Stockholm, Sweden). The cDNAs containing the complete sequences of HBD-2 and HBD-3 were obtained by RT-PCR techniques using total mRNA from human keratinocytes. The following oligonucleotides were employed in the amplification reaction (start codons in bold letters): HBD-2 forward: 5′ CCA GCC ATC AGC CAT GAG GGT 3′; HBD-2 reverse: 5′ TTC TGA ATC CGC ATC AGC CAC A3′; HBD-3 forward: 5′ GAA GCC TAG CAG CTA TGA GG 3′; HBD-3 reverse: 5′ GTC ATG TTT CAG GGT TTT TAT 3′. Sequences were confirmed using standard DNA sequencing methods, and cDNAs were cloned along with an Ires-GFP expression cassette (Clontech, Palo Alto, CA) in pDC315 shuttle vector (Microbix Biosystems Inc., Toronto, Canada). Adenoviral vectors were generated by cotransfection of the shuttle vector along with pBHGlox∆E1,E3Cre in 293 cells (Microbix) (Fig. 1) using the calcium phosphate transfection method. Plaques were isolated and expanded, and viral DNA was extracted and analyzed by restriction enzyme digestion as described previously (17). Ad(GFP) was kindly provided by Dr. David T. Curiel (University of Alabama, Birmingham, AL). Semiquantitative reverse transcriptase/polymerase chain reaction Adenoviral transduced epithelial cells were cultured for 48 h after infection, and total RNA was isolated using the Trizol reagent (Gibco-BRL). Total RNA (1 µg) was reversed transcribed using 0.5 µl SuperScript reverse transcriptase (Invitrogen, Carlsbad, CA) in a Vf = 20 µl. Then, 1 µl of the RT reaction was employed in polymerase chain reaction (PCR) containing the following pairs of specific primers: LL-37 (forward primer: 5′ CCA AAG GAA TGG CCA CTC CC 3′; reverse primer: 5′ CAC ACA CTA GGA CTC TGT CC 3′), HBD-2 (forward primer: 5′ CCA GCC ATC AGC CAT GAG GGT 3′; reverse primer: 5′ TTC TGA ATC CGC ATC AGC CAC A 3′), HBD-3 (forward primer: 5′ GAA GCC TAG CAG CTA TGA GG 3′; reverse primer: 5′ GTC ATG TTT CAG GGT TTT TAT 3′), and GAPDH (forward primer: 5′ ATC TTC CAG GAG CGA GAT CC 3′; reverse primer: 5′ ACC ACT GAC ACG TTG GCA GT 3′). PCR products were subjected to electrophoresis on a 1.5% agarose gel and visualized by ethidium bromide staining. Concentration of supernatants from adenoviral transduced HaCaT cells HaCaT cells were grown to 70% confluence and were then transduced with adenoviral vectors for 6 h at 37°C in a humid atmosphere containing 5% CO2. Medium containing adenoviruses was discarded, and fresh medium was added to transduced cells. Twenty- four hour supernatants were collected and concentrated using Microcon YM-3 devices (Millipore, Billerica, MA) following the recommendations of the manufacturer.

Page 3 of 19 (page number not for citation purposes)

Immunodot blot analysis of concentrated supernatants To measure the concentration of HBD-2, HBD-3, and LL-37 in concentrated supernatants from adenoviral transduced HaCaT cells, quantitative dot-blot analyses were performed. From each supernatant, 1 µl was dotted onto nitrocellulose filters and compared with known amounts of the corresponding synthetic peptides. The membranes were treated with blocking solution (10% nonfat milk in Tris-buffered saline) for 2 h at room temperature. Then, filters where incubated overnight at 4°C either with rabbit anti-HBD-2 antiserum (Alpha Diagnostic International, Lost Lane, San Antonio), rabbit polyclonal human β-defensin-3 affinity purified (Orbigen Inc., San Diego, CA), or monoclonal antibody to human LL37/CAP18 (HyCult Biotechnology b.v., Uden, The Netherlands) diluted in TBS containing 0.1% Tween 20 and 1% nonfat milk. After the membranes were washed three times with 0.1% Tween 20/TBS (TBS-T), hoseradish peroxidaselabeled secondary antibody diluted in TBS-T was incubated with filters for 60 min at room temperature. After being washed with TBS-T, bound antibodies were visualized by using SuperSignal Solution chemoluminiscent substrate (Pierce Biotechnology Inc., Rockford, IL) and exposure to X-ray films. Measurement of antimicrobial activity in concentrated supernatants Colony-forming unit (CFU) assay was performed to determine the antimicrobial activity of the peptides secreted by adenoviral transduced HaCaT cells. Escherichia coli (ATCC 25922), Staphylococcus aureus (ATCC 29213), Pseudomonas aeruginosa (ATCC 27583), and Enterococcus faecalis (ATCC 29212) were employed in these assays. Single colonies of bacteria were inoculated into recommended growth media and cultured overnight at 37°C. An aliquot of this culture was transferred to fresh media and incubated for an additional 3 h at 37°C to obtain mid-logarithmic-phase cells. The organisms were diluted in 10 mM sodium phosphate buffer (pH 7.4) containing 3% (v/v) trypticase soy broth (buffer NPB+3%TSB), and the concentration of CFU per ml-1 was quantified by measuring its absorbance at 600 nm and plating. Concentrated supernatants from adenoviral transduced HaCaT cells (10 µl) were then mixed with 10 µl of bacterial suspension containing 2 x 104-2 x 105 cells/ml and incubated for 2 h at 37°C. Serial dilutions were then plated on Nutrient broth (for E. coli and P. aeruginosa) or Trypticase soy broth (for E. faecalis and S. aureus) agar plates, and the number of CFU was determined the following day. To test for synergistic effects between peptides, we have applied the following criteria: a synergistic activity occurs when the net growth inhibitory effect of antimicrobial peptide combination divided by the sum of the net individual antimicrobial peptide effects results in a quotient >1 (synergism quotient), which suggest a synergistic effect rather than additive (synergism quotient=1). Cell viability after incubation with supernatants was calculated as follows: (cell survival after incubation with peptide containing supernatant)/(cell survival after incubation with GFP control supernatant) x 100. Antimicrobial assays using genetically modified dermo-epidermal equivalents HaCaT cells were grown to confluence on fibroblast-containing fibrin gels and were then transduced using adenoviral vectors for 12 h at 37°C and 5% CO2. Forty-eight hours postinfection, equivalents were raised to the air-liquid interface for an additional 5 days, and 2.4 × 104 E. coli or S. aureus in buffer NPB+ 3% TSB was left to adsorb onto the gels that were then incubated for 24 h at 37°C. Bacteria grown on the surface of the gels were recovered using nutrient broth (E. coli) or tripticase soy broth (S. aureus) medium, and serial dilutions were Page 4 of 19 (page number not for citation purposes)

plated onto appropriate agar plates. After 24 h of incubation at 37°C, the number of colony forming units was determined. XTT colorimetric assay Different numbers of HaCaT cells were plated in 96-well tissue culture plates. After 24 h, cells were transduced with adenoviral vectors for 6 h at 37°C in a humid atmosphere containing 5% CO2. Adenovirus-containing supernatants were then discarded, and 100 µl fresh medium was added to each well. At 24 and 48 h postinfection, cells were incubated with XTT (Boehringer Mannheim Gmbh, Germany) and absorbance at 450 nm (cell number) was measured. Triplicate experiments were performed. Histological and immunohistochemical analysis Equivalents were formalin-fixed and embedded in paraffin. To determine tissue morphology, sections were cut (4-6 µm) and stained with hematoxylin-eosin following a standard procedure. For the immunoperoxidase staining, sections were treated for endogenous peroxidase inactivation. Sections were then blocked and incubated overnight at 4°C with primary antibody anti-keratin 10 (K8.60) (Sigma-Aldrich Co.). The secondary anti mouse biotinilated antibody (Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) was incubated at a final dilution of 1:500 for 45 min at RT. The reaction was visualized with the Avidin-Biotin-Complex (ABC) kit (Vectastain Elite, Vector Laboratories, Inc., Burlingame, CA) by using diaminobenzidine as chromogenic substrate for peroxidase. Sections were counterstained with hematoxylin and dehydrated in water solution containing increasing percentages of ethanol. Finally, the slides were placed for 15 min in histoclear (National diagnostic, Atlanta, GA), mounted, and analyzed by light microscopy. RESULTS AMP overexpression in human keratinocytes The spontaneously immortalized human keratinocyte cell line (HaCaT) was transduced using adenoviral vectors containing the complete sequence of human antimicrobial peptides HBD-2, HBD-3, and LL-37 along with an Ires-GFP expression cassette (Fig. 1). A close to 100% infectivity was achieved in all cases as monitored by green fluorescent protein expression visualized under ultraviolet light at 365 nm (Fig. 2A). Antimicrobial peptide expression was detected by RT-PCR using total mRNA from adenoviral transduced cells (Fig. 2B). Control GFP+ HaCaT cells did not express detectable levels of HBD-3, which was induced in cells transduced with Ad(HBD-3) at 24 h postinfection. HBD-2 and LL-37 were only expressed when Ad(HBD-2) and Ad(LL-37), respectively, were used to infect HaCaT cells either alone or in combination. To assess the amounts of the secreted antimicrobial peptides, concentrated supernatants from adenoviral transduced HaCaT cells were analyzed by immunodot blot along with known concentrations of synthetic peptides. HBD-3 was expressed at ~50 µg/ml whereas LL-37 and HBD-2 were produced at ~25 µg/ml and ~100 µg/ml, respectively (Fig. 2C). These levels are consistent with those previously reported for antimicrobial activity (18–20).


To address whether the secreted peptides are correctly processed, concentrated supernatants from adenoviral transduced HaCaT cells were run on SDS-PAGE along with recombinant mature peptides, followed by immunoblotting with anti-LL-37 and anti-HBD-3 antibodies. At least a fraction of the overexpressed HBD-3 and LL-37 antimicrobial peptides showed identical mobilities to the recombinant mature peptides (data not shown). To demonstrate that the secreted peptides did not produce a reduction in HaCaT cell viability, we performed an XTT assay, which measures metabolization of tetrazolium salt (2,3-bis(2-methoxy4-nitro-5-sulfophenyl)-5-[(phenylamino) carbonyl]-2H-tetrazolium hydroxide). As shown in Fig. 3, neither HBD-3 nor LL-37 altered cell viability compared with control cells (GFP). Moreover, at 48 h postinfection, both peptides appeared to stimulate cell growth, whereas HBD-2 expressing cells showed a slight decrease in cell viability compared with control. Bactericidal activity of AMP transduced HaCaT cells We evaluated the killing activity of HBD-2, HBD-3, and LL-37 antimicrobial peptides expressed in 24 h conditioned media from adenoviral transduced HaCaT cells using a CFU assay as described in Materials and Methods. We used concentrated supernatants to counteract the possible inhibitory effect of salt and FBS on the activity of peptides. The concentration might resemble the situation in vivo, where the local concentration of antimicrobial peptides is augmented by retention between keratinocytes (21). A reduction in E. coli viability ranging from 45 to 67% was observed when HBD-3 containing supernatants were used. In these conditions, HBD-2 and LL-37 yielded 12-27% and 45-79% reduction in bacterial counts, respectively. The co-expression of HBD-2 and LL-37 in the conditioned medium from HaCaT cells produced a strong synergistic activity between these antimicrobial peptides against E. coli (Fig. 4A and B). HBD-3 also efficiently killed S. aureus in this system (49-93% inhibition of bacteria viability), as well as HBD-2 cosecreted along with LL-37 (Fig. 4A). As occurred with E. coli, even when the activity of HBD-2 or LL-37 alone was not detectable, the combination of both dramatically inhibited the bacterial growth (Fig. 4B and data not shown). LL-37 was the most effective in killing P. aeruginosa (53-89% killing activity), but in this case we failed to observe synergistic activity between HBD-2 and LL-37 (Fig. 4A and 4B). All of these peptides showed a significant activity against E. faecalis ranging from 74 to 65% reduction in viability in all cases, and, as occurred with P. aeruginosa, no synergistic activity was observed between LL-37 and HBD-2 (data not shown). The majority of these peptides were still active even when bacterial load was increased by 10fold (input=2000 CFU; Fig. 4C). Antibacterial activity in HaCaT-containing cultured bioengineered skin To determine whether these peptides are active when produced by the epidermal component of a skin equivalent, HaCaT cells were grown to confluence on a fibroblast-containing fibrin matrix and were subsequently transduced with adenoviral vectors containing the different antimicrobial peptides mentioned above. At 48 h postinfection, equivalents were raised to the air-liquid interface for an additional 5 days to achieve a certain degree of epidermal stratification that was confirmed by histological and immunohistochemical analysis (Fig. 5B). A similar degree of Page 6 of 19 (page number not for citation purposes)

differentiation in HaCaT cells growing as an organotypic culture in collagen gels was previously reported (22). To analyze the antimicrobial activity in this system, different bacteria were left to adsorb onto the antimicrobial expressing gels, and after overnight incubation at 37°C, viable bacteria were recovered, serially diluted, plated, and counted (see scheme in Fig. 5A). In this system, E. coli viability was reduced by 28-68% when seeded on LL-37 expressing equivalents. When the epidermal component was provided by HBD-3- transduced HaCaT cells, a 34-72% reduction in bacterial growth was observed (a representative experiment is shown in Fig. 5C). For S. aureus, equivalents transduced with adenoviral vectors containing LL-37 and HBD-3 also showed the best inhibitory activity (20-45% and 20-85%, respectively). Similarly, P. aeruginosa viability was inhibited when seeded on LL-37 (43-53%) and HBD-3 (28-35%) expressing equivalents (Fig. 5C and D, a representative experiment for each bacterial strain is shown). In contrast to the results obtained with the conditioned media, antimicrobial activity in equivalents transduced with both Ad(HBD-2) and Ad(LL-37) against E. coli and S. aureus was similar to that obtained when Ad(LL-37) was used alone (data not shown). To study this apparent discrepancy in the combinatory activity between HBD-2 and LL-37 in the two experimental systems, we carried out expression analysis of the antimicrobial peptides in the adenoviral transduced equivalents. Expression of peptides was confirmed by RT-PCR using total mRNA from the epidermal component of these equivalents. In these conditions, we observed basal expression of HBD-2 and HBD-3 and inducible expression when HaCaT cells overexpressed any of these peptides by adenoviral transfer. That is, equivalents transduced with Ad(HBD-2) expressed high levels of HBD-2 but also increased levels of HBD-3 compared with control. Equivalents infected with Ad(HBD-3) also exhibited augmented HBD-2 expression. Finally, Ad(LL-37) induced high expression levels for LL-37, HBD-2, and HBD-3 (Fig. 6). Thus, LL-37 overexpressing equivalents also co-expressed significant levels of HBD-2 that might cooperate with the cathelicidin reaching the maximum levels of killing activity against E. coli and S. aureus. This fact could explain the lack of any additional bactericidal activity against these bacterial strains when equivalents were transduced with both Ad(HBD-2) and Ad(LL-37) compared with those transduced with Ad(LL-37) alone (data not shown). DISCUSSION Proper treatment of severe burn wounds is critical to avoid microbial multiplication in order to achieve efficient closure of the wound after skin grafting. However, there is an increase in the prevalence of multidrug-resistant microbes that limits the use of topical and systemic antibiotics in the majority of burn units (3). AMPs are presented as an alternative to standard antibiotics to treat such drug-resistant strains. Here we present a novel antimicrobial gene therapy approach based on a genetically modified epidermal component overexpressing different human skin-relevant antimicrobial peptides including defensins HBD-2 and -3 and the cathelicidin LL-37 as a platform to increase the innate immune response of skin to combat infection during the closure of extensive burn wounds. Antibacterial activity of endogenously produced keratinocyte-derived AMPs has been previously reported. Thus, endogenous HBD-2 expressed in the epidermis of a commercial tissueengineered human skin equivalent (Apligraf) prevents the growth of E. coli on its surface Page 7 of 19 (page number not for citation purposes)

although colonies appeared on the dermal surface exposed to bacteria (23). In another report, it was demonstrated that IL-1α or IL-6 treatment of cultured composite keratinocyte grafts stimulates the production of antimicrobial peptides enhancing the antibacterial properties of these grafts and allowing the correct epidermal formation (24). However, both the increased bacterial load in the burn wound beds and the loss of the epidermal source of AMP call for their restoration either through pharmacological or gene therapy approaches. The potential use of antimicrobial gene therapy was suggested in a previous study (14) where E. coli was injected into tumors formed by subcutaneous injection into NOD/SCID mice of HT-1080 cells transduced with a retroviral vector carrying HBD-2. To render high level and transient expression of the AMPs, adenoviral vectors were used in the present study. The adenoviral-mediated gene transfer approach for LL-37 overexpression was previously employed to reverse the bacterial killing defect observed in cystic fibrosis by using a human bronchial xenograft model (13). In this study, conditioned media of AMP genetransduced keratinocytes showed bacteria killing activity against P. aeruginosa, S. aureus, and E. coli, which are major human opportunistic pathogens that cause burn wound infections. In addition to the individual action of each antimicrobial peptide against these bacterial strains, we also observed a synergistic activity between LL-37 and HBD-2 against E. coli and S. aureus. Within this context, it was previously demonstrated that the human cathelicidin LL-37 and HBD2 cooperatively kill S. aureus, the most common pathogen isolated from atopic dermatitis but rarely associated with psoriatic lesions, where significantly higher concentrations of LL-37 and HBD-2 have been observed (5). Similarly, synergy has also been described between two antimicrobial peptides released by human neutrophils, HNP-1 and LL-37, against E. coli and S. aureus (25). In the present work, we failed to observe any synergism between the human cathelicidin and HBD-2 against P. aeruginosa and E. faecalis. Lack of synergism between LL-37 and HNP-1 synthetic peptides against these bacterial strains was also evidenced by a previous report (26). Thus, it appears that the mechanism of synergy observed between the human cathelicidin and peptides of the human α- or β-defensin family might be similar. The overexpression of individual human antimicrobial peptides by keratinocytes seeded on fibroblast-containing fibrin gels also inhibits bacterial growth by almost 50% in most cases. To our surprise, however, the co-infection using adenoviral vectors for HBD-2 and LL-37 did not render additional killing activity compared with that obtained when gels overexpressing LL-37 alone were used. For this reason, we analyzed antimicrobial peptide expression in these equivalents and observed basal expression of HBD-2 in control equivalents (GFP+), which, interestingly, is further augmented in the presence of LL-37. It was previously demonstrated that β-defensin expression can be induced by differentiation (21). In fact, a certain degree of stratification/differentiation achieved in our system as demonstrated by the induction of keratin 10 after 5 days culture in the air-liquid interfase (Fig. 5B) might be responsible for the basal expression of β defensins. However, to our knowledge, this is the first report to show the inducible expression of HBD-2 by the human cathelicidin LL-37. It is tempting to speculate that the presence of NF-κB and AP-1 binding sites in the promoter region of HBD-2 might be involved, as LL-37 is known to activate ERK, p38, and JNK signal transduction pathways in epithelial cells (27). Although a complete bacterial clearance was not achieved in our system, this is in agreement with the natural role of antimicrobial peptides in containing infection until the antigen-specific immune response is armed. Thus, the antimicrobial gene therapy approach proposed here might be considered useful either to prevent infection (a situation in which Page 8 of 19 (page number not for citation purposes)

bacterial load might be low enough to assure antimicrobial peptide effectiveness) or as a treatment of a pre-existing infection when used in combination with other antimicrobial substances (i.e., lysozyme, antibiotics). In contrast with conventional antibiotics, the activity of some of these peptides in wound repair has been previously demonstrated (8). The production of these peptides in this system may be expected to combat bacteria, and, in addition, stimulate cells at the wound margin to cover the lost skin surface, thus accelerating the healing process. Moreover, LL-37 induces an angiogenic response by a direct action on endothelial cells (28). Angiogenesis is an essential process for wound healing and tissue repair. Impaired vascularization of the wound bed in burn patients contributes to infection and graft failure. In addition, the anti-endotoxic properties of this antimicrobial peptide (9) and its involvement in the enhancement of the adaptive immunity by affecting the maturation of dendritic cells are relevant in this context (11). Overall, our results pave the way for alternative therapeutic solutions to the use of synthetic antibiotics in burn patients. An AMP-gene delivery approach through cutaneous gene therapy will be useful to overcome problems concerning high cost production, continuous administration, and adverse side effects. Particularly, LL-37 might be a good candidate for the transient skin gene therapy, not only in the temporary coverage allowing the control of infection at the wound site but also in the definitive coverage, as it exerts several biological activities that might cooperatively contribute to an efficient tissue repair. ACKNOWLEDGMENTS We wish to thank Almudena Holguín, Blanca Duarte, Isabel de los Santos, and Pilar Hernández for excellent technical assistance and Soledad Moreno for help with artwork. M. Carretero is a recipient of a postdoctoral fellowship from Fundación Marcelino Botín. This work was supported by the Spanish Ministry of Science and Technology (SAF 2004-07717). REFERENCES 1.

Meana, A., Iglesias, J., Del Rio, M., Larcher, F., Madrigal, B., Fresno, M. F., Martin, C., San Roman, F., and Tevar, F. (1998) Large surface of cultured human epithelium obtained on a dermal matrix based on live fibroblast-containing fibrin gels. Burns 24, 621–630

2.

Llames, S. G., Del Rio, M., Larcher, F., Garcia, E., Garcia, M., Escamez, M. J., Jorcano, J. L., Holguin, P., and Meana, A. (2004) Human plasma as a dermal scaffold for the generation of a completely autologous bioengineered skin. Transplantation 77, 350–355

3.

Edwards-Jones, V., and Greenwood, J. E. (2003) What's new in burn microbiology? James Laing Memorial Prize Essay 2000. Burns 29, 15–24

4.

Gallo, R. L., Murakami, M., Ohtake, T., and Zaiou, M. (2002) Biology and clinical relevance of naturally occurring antimicrobial peptides. J. Allergy Clin. Immunol. 110, 823– 831

5.

Ong, P. Y., Ohtake, T., Brandt, C., Strickland, I., Boguniewicz, M., Ganz, T., Gallo, R. L., and Leung, D. Y. (2002) Endogenous antimicrobial peptides and skin infections in atopic dermatitis. N. Engl. J. Med. 347, 1151–1160 Page 9 of 19 (page number not for citation purposes)

6.

Ortega, M. R., Ganz, T., and Milner, S. M. (2000) Human beta defensin is absent in burn blister fluid. Burns 26, 724–726

7.

Nizet, V., Ohtake, T., Lauth, X., Trowbridge, J., Rudisill, J., Dorschner, R. A., Pestonjamasp, V., Piraino, J., Huttner, K., and Gallo, R. L. (2001) Innate antimicrobial peptide protects the skin from invasive bacterial infection. Nature 414, 454–457

8.

Heilborn, J. D., Nilsson, M. F., Kratz, G., Weber, G., Sorensen, O., Borregaard, N., and Stahle-Backdahl, M. (2003) The cathelicidin anti-microbial peptide LL-37 is involved in reepithelialization of human skin wounds and is lacking in chronic ulcer epithelium. J. Invest. Dermatol. 120, 379–389

9.

Scott, M. G., Davidson, D. J., Gold, M. R., Bowdish, D., and Hancock, R. E. (2002) The human antimicrobial peptide LL-37 is a multifunctional modulator of innate immune responses. J. Immunol. 169, 3883–3891

10. Hancock, R. E., and Scott, M. G. (2000) The role of antimicrobial peptides in animal defenses. Proc. Natl. Acad. Sci. USA 97, 8856–8861 11. Davidson, D. J., Currie, A. J., Reid, G. S., Bowdish, D. M., MacDonald, K. L., Ma, R. C., Hancock, R. E., and Speert, D. P. (2004) The cationic antimicrobial peptide LL-37 modulates dendritic cell differentiation and dendritic cell-induced T cell polarization. J. Immunol. 172, 1146–1156 12. Marshall, S. H., and Arenas, G. (2003) Antimicrobial peptides: a natural alternative to chemical antibiotics and a potential for applied biotechnology. Electron. J. Biotechnol. 6, 271–284 13. Bals, R., Weiner, D. J., Meegalla, R. L., and Wilson, J. M. (1999) Transfer of a cathelicidin peptide antibiotic gene restores bacterial killing in a cystic fibrosis xenograft model. J. Clin. Invest. 103, 1113–1117 14. Huang, G. T., Zhang, H. B., Kim, D., Liu, L., and Ganz, T. (2002) A model for antimicrobial gene therapy: demonstration of human beta-defensin 2 antimicrobial activities in vivo. Hum. Gene Ther. 13, 2017–2025 15. Spirito, F., Meneguzzi, G., Danos, O., and Mezzina, M. (2001) Cutaneous gene transfer and therapy: the present and the future. J. Gene Med. 3, 21–31 16. Del Rio, M., Larcher, F., and Jorcano, J. L. (2002) Recent advances in gene therapy with skin cells. Eur. Rev. 10, 369–388 17. Graham, F. L., and van der Eb, A. J. (1973) A new technique for the assay of infectivity of human adenovirus 5 DNA. Virology 52, 456–467 18. Harder, J., Bartels, J., Christophers, E., and Schroder, J. M. (2001) Isolation and characterization of human beta-defensin-3, a novel human inducible peptide antibiotic. J. Biol. Chem. 276, 5707–5713


19. Bals, R., Wang, X., Wu, Z., Freeman, T., Bafna, V., Zasloff, M., and Wilson, J. M. (1998) Human beta-defensin 2 is a salt-sensitive peptide antibiotic expressed in human lung. J. Clin. Invest. 102, 874–880 20. Bals, R., Wang, X., Zasloff, M., and Wilson, J. M. (1998) The peptide antibiotic LL37/hCAP-18 is expressed in epithelia of the human lung where it has broad antimicrobial activity at the airway surface. Proc. Natl. Acad. Sci. USA 95, 9541–9546 21. Liu, A. Y., Destoumieux, D., Wong, A. V., Park, C. H., Valore, E. V., Liu, L., and Ganz, T. (2002) Human beta-defensin-2 production in keratinocytes is regulated by interleukin-1, bacteria, and the state of differentiation. J. Invest. Dermatol. 118, 275–281 22. Maas-Szabowski, N., Starker, A., and Fusenig, N. E. (2003) Epidermal tissue regeneration and stromal interaction in HaCaT cells is initiated by TGF-alpha. J. Cell Sci. 116, 2937– 2948 23. Schmid, P., Grenet, O., Medina, J., Chibout, S. D., Osborne, C., and Cox, D. A. (2001) An intrinsic antibiotic mechanism in wounds and tissue-engineered skin. J. Invest. Dermatol. 116, 471–472 24. Erdag, G., and Morgan, J. R. (2002) Interleukin-1alpha and interleukin-6 enhance the antibacterial properties of cultured composite keratinocyte grafts. Ann. Surg. 235, 113–124 25. Nagaoka, I., Hirota, S., Yomogida, S., Ohwada, A., and Hirata, M. (2000) Synergistic actions of antibacterial neutrophil defensins and cathelicidins. Inflamm. Res. 49, 73–79 26. Yan, H., and Hancock, R. E. (2001) Synergistic interactions between mammalian antimicrobial defense peptides. Antimicrob. Agents Chemother. 45, 1558–1560 27. Tjabringa, G. S., Aarbiou, J., Ninaber, D. K., Drijfhout, J. W., Sorensen, O. E., Borregaard, N., Rabe, K. F., and Hiemstra, P. S. (2003) The antimicrobial peptide LL-37 activates innate immunity at the airway epithelial surface by transactivation of the epidermal growth factor receptor. J. Immunol. 171, 6690–6696 28. Koczulla, R., von Degenfeld, G., Kupatt, C., Krotz, F., Zahler, S., Gloe, T., Issbrucker, K., Unterberger, P., Zaiou, M., Lebherz, C., et al. (2003) An angiogenic role for the human peptide antibiotic LL-37/hCAP-18. J. Clin. Invest. 111, 1665–1672 Received March 23, 2004; accepted August 4, 2004.


Fig. 1

Figure 1. Schematic representation of the generation of recombinant adenoviruses by conventional homologous recombination.


Fig. 2

Figure 2. Adenoviral-mediated antimicrobial peptide overexpression in cultured HaCaT cells. A) Direct monitoring of adenoviral infection in HaCaT cells by visualization of GFP expression under UV light (365 nm). B) Expression of antimicrobial peptide transcripts in adenoviral transduced HaCaT cells by RT-PCR. Molecular weight (Mw) marker X174 DNA cleaved with HaeIII. C) Quantitation of antimicrobial peptides by immuno-dot blot assays. Page 13 of 19 (page number not for citation purposes)

Fig. 3

Figure 3. Viability of cells expressing antimicrobial peptides. HaCaT cell viability was quantified using an XTT assay either 24 or 48 h after adenoviral infection. Data are means ± SD of each condition determined in triplicates.


Fig. 4


Fig. 4 (cont)


Fig. 4 (cont)

Figure 4. Antimicrobial activity from adenoviral-transduced HaCaT cell supernatants. A) Bacteria were incubated with cell free supernatants for 2 h at 37°C and inoculated onto agar plates to determine number of CFU. Percent viability was calculated relative to control survival. Error bars are SD from the mean for 3-4 separate experiments. *P < 0.05 compared with control [supernatant from Ad (GFP) transduced HaCaT cells]. B) A representative experiment for each bacterial strain is shown. C) Supernatant containing killing activity after bacterial load was increased by 10x. A representative experiment for each strain is shown.


Fig. 5

Figure 5. Antimicrobial activity of genetically modified HaCaT-containing cultured skin substitutes overexpressing AMPs. A) Schematic representation of experimental setup. B) Histological appearance and immunohistochemical staining of human keratin-10 in a non stratified HaCaT organotypic culture and 5 days after raising to the air-liquid interfase. C) A representative experiment showing S. aureus growth after overnight incubation on the surface of adenoviral transduced cultured skin substitutes. D) Percent viability of bacteria recovered from the surface of genetically modified skin substitutes relative to control equivalents expressing GFP. A representative experiment out of 3 for each bacterial strain is shown.


Fig. 6

Figure 6. Expression of antimicrobial peptide transcripts in adenoviral transduced equivalents by RT-PCR. Molecular weight marker X174 DNA cleaved with HaeIII. Page 19 of 19 (page number not for citation purposes)