Inhibition of Multidrug-resistant Acinetobacter baumannii by Nonviral Expression of hCAP-18 in a Bioengineered Human Skin Tissue

original article

© The American Society of Gene Therapy

see page 405

Inhibition of Multidrug-resistant Acinetobacter baumannii by Nonviral Expression of hCAP-18 in a Bioengineered Human Skin Tissue Christina L Thomas-Virnig1,2, John M Centanni2, Colette E Johnston2, Li-Ke He3, Sandy J Schlosser1, Kelly F Van Winkle2, Ruibing Chen4, Angela L Gibson5,6, Andrea Szilagyi3, Lingjun Li4, Ravi Shankar3 and B Lynn Allen-Hoffmann1 1 Department of Pathology and Laboratory Medicine, University of Wisconsin–Madison, Madison, Wisconsin, USA; 2Stratatech Corporation, Madison, Wisconsin, USA; 3Department of Surgery and Cell Biology, Neurobiology and Anatomy, Loyola University Medical Center, Maywood, Illinois, USA; 4 School of Pharmacy and Department of Chemistry, University of Wisconsin–Madison, Madison, Wisconsin, USA; 5Program in Cellular and Molecular Biology, University of Wisconsin–Madison, Madison, Wisconsin, USA; 6Medical Scientist Training Program, University of Wisconsin School of Medicine and Public Health, University of Wisconsin–Madison, Madison, Wisconsin, USA

When skin is compromised, a cascade of signals initiates the rapid repair of the epidermis to prevent fluid loss and provide defense against invading microbes. During this response, keratinocytes produce host defense peptides (HDPs) that have antimicrobial activity against a diverse set of pathogens. Using nonviral vectors we have genetically modified the novel, nontumorigenic, pathogenfree human keratinocyte progenitor cell line (NIKS) to express the human cathelicidin HDP in a tissue-specific manner. NIKS skin tissue that expresses elevated levels of cathelicidin possesses key histological features of normal epidermis and displays enhanced antimicrobial activity against bacteria in vitro. Moreover, in an in vivo infected burn wound model, this tissue results in a two log reduction in a clinical isolate of multidrug-resistant Acinetobacter baumannii. Taken together, these results suggest that this genetically engineered human tissue could be applied to burns and ulcers to counteract ­bacterial contamination and prevent infection. Received 4 April 2008; accepted 25 November 2008; published online 3 February 2009. doi:10.1038/mt.2008.289

Introduction Bacterial contamination is the most common reason for the impairment of wound healing.1 Unresolved bacterial infections inhibit wound closure and destroy underlying tissue due to the potent exotoxins and proteases secreted by pathogenic bacteria.2,3 Under these circumstances, healing is delayed until microbial burden is below an infective level, usually <105 organisms/g of ­tissue.1 Multidrug-resistant nosocomial strains of bacteria such as Acinetobacter baumannii have become a major challenge for the treatment of wounds. An ever rising number of A. baumannii clinical isolates are exhibiting resistance to essentially all commonly

used antibiotics.4,5 The increasing frequency of multidrug-resistant clinical isolates of A. baumannii in the United States underscores the need for novel approaches to supplement the current antimicrobial treatment regimes used in cutaneous wound therapy. Host defense peptides (HDPs) have been shown to play an important role in skin homeostasis and in cutaneous wound healing.6,7 Two main classes of HDPs have been identified in humans: the cathelicidins and the defensins. Thus far, a single cathelicidin, hCAP-18, has been found in humans, which is produced by epithelia in a prepropeptide form. The preprosequence contains a conserved cathelin-like domain, while the carboxy terminus harboring the LL-37 peptide is more variable between species.8 The full-length hCAP-18 is cleaved by skin-derived kallikreins 5 and 7, or by proteinase 3 from white blood cells to release the cathelin domain from the LL-37 containing carboxy terminus.9,10 The mature LL-37 can be further processed to release peptides that possess enhanced antimicrobial activity and can act synergistically with each other.9,11 In addition to its broad range of antimicrobial activity against bacteria, fungi, and viruses,12 LL-37 has been implicated in chemotaxis of immune and endothelial cells,13,14 as well as keratinocyte migration during wound healing.6,15 Systemic use of natural and synthetic HDPs as anti-infectives have been problematic due to induction of toxicities such as mast cell degranulation and programmed cell death.16,17 However, these potent bioactive peptides are ideally suited for topical application to cutaneous wounds. Here, we demonstrate that a temporary skin substitute tissue prepared from NIKS cells genetically modified to produce elevated levels of cathelicidin possesses enhanced antimicrobial activity both in vitro and in vivo. The NIKS cells are a novel, long-lived, nontumorigenic, and pathogen-free strain of human keratinocyte progenitors18 that have completed phase I/II clinical testing. The longevity and chromosomal stability of this cell line permits isolation and expansion of stably transfected clones produced using naked DNA. Because infection contributes significantly to the

The first two authors contributed equally to this work. Correspondence: B. Lynn Allen-Hoffmann, Department of Pathology and Laboratory Medicine, University of Wisconsin–Madison, 5605 Medical Sciences Center, 1300 University Avenue, Madison, Wisconsin 53706-1102, USA. E-mail: [email protected]

562

www.moleculartherapy.org vol. 17 no. 3, 562–569 mar. 2009

© The American Society of Gene Therapy

Engineered Skin Possesses Antimicrobial Activity

morbidity and mortality of burn patients, we sought to enhance the ability of NIKS keratinocytes to defend against microbial infection in a wound environment. We chose cathelicidin (hCAP-18) because of its broad specificity against both Gram-positive and Gram-negative bacteria as well as its wound-healing properties. The hCAP-18 is normally expressed in human keratinocytes during re-epithelialization both ex vivo and in vivo, but is lacking in many cutaneous wounds.6 Furthermore, a recent article noted that RNA expression of hCAP-18 was reduced by tenfold at the burn wound edge compared to both the center of the burn and to unrelated healthy skin.19 Genetically engineered NIKS skin substitute tissue expressing hCAP-18 (NIKShCAP-18) significantly counteracted bacterial infection from a multidrug-resistant nosocomial bacterial strain of A. baumannii in an animal model of infected third-degree cutaneous burn wound demonstrating the potential clinical utility of this antimicrobial bioengineered tissue. This innovative approach enables immediate coverage of cutaneous wounds with a full-thickness skin tissue providing topically delivery of a potent natural anti-infective.

ly-A

2 3

4

p

CAMP

CAM

PG lob

Po

1

INV Pro/hCAP-18/pUB/BS

Ubc

bs d

INV Pr om ote r

Molecular Therapy vol. 17 no. 3 mar. 2009

in

ori

c CAMP fold expression

10,000 8,000 6,000 4,000 2,000 0

hCAP–18

NIKS NIKS Monolayer

d

hCAP–18

NIKS NIKS Tissue

60

Colonies formed 2 in a 40 mm area

A nonviral expression vector containing the human CAMP cDNA (encoding hCAP-18 protein) driven by the epithelial-specific human involucrin promoter was constructed to target CAMP transgene expression to the epidermis. The involucrin promoter is expressed in differentiated nonproliferating, suprabasal keratinocyte layers in skin. Figure 1a depicts the INV-CAMP expression construct. Clones of stably transfected cells were selected by growth in medium containing blasticidin. Bacterial proteins such as neomycin phosphotransferase have been used for almost two decades in gene therapy vectors in clinical trials.20–22 The gene construct containing blasticidin S deaminase used in our studies will be evaluated by the National Institutes of Health Recombinant DNA Advisory Committee prior to the Food and Drug Administration review and approval for clinical use. Independent clones harboring the CAMP construct were obtained and expression of the CAMP transgene was confirmed using semiquantitative reversetranscribed PCR (Figure 1b). In order to quantify CAMP expression levels in stably transfected NIKS clones, total RNA was isolated from monolayer cell cultures and tissue derived from each clone, and relative expression levels were analyzed by quantitative PCR. Unmodified NIKS cells were first compared to the primary keratinocyte cells from which they were derived and found to exhibit comparable CAMP expression (data not shown). NIKShCAP-18, one of the highest expressing clones, exhibited a 250-fold enhancement of CAMP mRNA in monolayer culture and an almost 9,000-fold enhancement of CAMP mRNA in skin substitute tissues relative to unmodified NIKS (Figure 1c). A critically important aspect of any cell to be used in tissue engineering applications is that they must not be tumorigenic. Anchorage-independent growth is highly correlated with tumorigenicity in vivo.23 It has previously been demonstrated that NIKS keratinocytes do not form colonies when analyzed for anchorageindependent growth or tumors when injected into nude and severe combined immunodeficiency mice.18 To eliminate the possibility

b AMP -r

uC

Results Genetic modification and testing of the NIKS keratinocytes

a

50 40 30 20 10 0 SCC4

NIKS

hCAP–18

NIKS

Figure 1 Construction of mammalian cell expression vector and screening by semiquantitative PCR, quantitative PCR (qPCR), and for anchorage-independent growth properties. (a) Diagram of human CAMP expression vector. (b) Reverse-transcribed PCR screen for CAMP transgene expression. The forward primer was designed to anneal to the CAMP coding region and the reverse primer was designed to anneal to the rabbit β-globin poly-A fragment of the vector. Because one primer anneals to the rabbit β-globin fragment, this primer set did not amplify endogenous CAMP mRNA. Gel image was cropped to include the DNA ladder next to samples. Lane 1: clone exhibiting low expression of CAMP. Lane 2: NIKS cells exhibiting no expression of CAMP. Lane 3: clone exhibiting high expression of CAMP. Lane 4: 1 kb molecular weight marker. (c) Representative qPCR displaying fold expression of CAMP in monolayer and tissue formats. Monolayer clone contains ~250-fold more expression while the same clone in 15-day-old tissues contains almost 9,000-fold more expression than the unmodified NIKS control. (d) Anchorage-independent growth assay with SCC4-positive control cells, NIKS, and NIKShCAP-18. Bars represent mean + SEM.

that elevated expression of hCAP-18 increases the tumorigenic potential of NIKS cells, NIKShCAP-18 were subjected to an anchorage-independent growth assay. Figure 1d shows that NIKShCAP-18 does not exhibit anchorage-independent growth unlike the positive control SCC4 cells. To confirm that NIKShCAP-18 keratinocytes were not tumorigenic, cells were injected into the flanks of athymic nude mice. Again, SCC4 cells were used as a positive control for tumor production. NIKS keratinocytes and media-only were used as a negative control. Five athymic nude mice were injected for each cell type as well as the media-only control (data not shown). SCC4injected mice were euthanized after 8 weeks due to tumor size. After 12 weeks, NIKS, NIKShCAP-18 cells as well as media-only controls did not produce tumors in athymic nude mice (Figure 2). 563

© The American Society of Gene Therapy

Engineered Skin Possesses Antimicrobial Activity

a

b

a

c

H&E

LL-37/Hoechst

H&E

LL-37/Hoechst

Neonatal foreskin

SCC4

NIKS

hCAP-18

NIKS

Figure 2 NIKShCAP-18 clones do not form tumors in vivo. (a) Representative SCC4-injected mouse at 8 weeks. (b) Representative NIKS-injected mouse at 12 weeks. (c) Representative NIKShCAP-18-injected mouse 12 weeks. Tumors (white arrows) formed only in SCC4-injected mice.

NIKS cells expressing elevated levels of hCAP-18 form normal skin tissue Localization of hCAP-18 and its processed forms to the differentiated and basal layers of the epidermis of human skin has been shown by multiple groups. To explore the localization of hCAP-18 in samples of intact human skin, both neonatal foreskin and adult skin tissue were immunostained using an antibody that detects both hCAP-18 and the LL-37 peptide (hCAP-18/LL-37). As can be seen in Figure 3a, both neonatal foreskin and adult skin tissue display staining in the basal layers of the epidermis, and therefore are consistent with the pattern of staining reported in the literature. In addition, neonatal foreskin exhibits hCAP-18 staining in the upper differentiated layers of the tissue. The physical and biological barrier provided by intact skin is the primary defense against ubiquitous microorganisms. Therefore, the NIKS clones with enhanced levels of cathelicidin were evaluated for their ability to produce skin tissue with architecture comparable to that of the unmodified NIKS cells. As seen in Figure 3b, tissue produced from the highest expressing genetically modified NIKShCAP-18 transgenic cell line was histologically identical to tissue produced using the unmodified NIKS cells and possesses an epidermal layer similar to native human epidermis. To compare the localization of hCAP-18 and mature LL-37 in the NIKS stratified epithelium to tissue generated from the NIKShCAP-18 transgenic cell line, immunostaining was performed on cryosections of cultured skin tissue. A low level of hCAP-18/ LL-37 was present throughout the epidermis of the NIKS skin tissue including the basal keratinocytes immediately adjacent to the dermis (Figure  3b), consistent with previous reports of endogenous hCAP-18/LL-37 expression in human skin. NIKS skin tissue closely resembles the pattern of staining displayed in neonatal foreskin (Figure  3a). In contrast, robust staining of hCAP-18/ LL-37 was evident throughout the dermal and epidermal compartments of the NIKShCAP-18 skin tissue. Staining in the spinous layer was especially strong, in accord with hCAP-18 being driven in the more differentiated layers of skin by the involucrin promoter. The enhanced staining in the dermis further demonstrates that both hCAP-18 and its mature form LL-37 can diffuse through the NIKShCAP-18 tissue. Cathelicidin expression in NIKS expressing hCAP-18 To further characterize hCAP-18 expression, cell lysates from both NIKS and NIKShCAP-18 tissue were analyzed by immunoblot analysis using the polyclonal antibody described above. Control synthetic LL-37 was detected by immunoblot analysis (Figure 4a). 564

Adult skin

b

NIKS

NIKS

hCAP-18

Figure 3 NIKS keratinocytes overexpressing hCAP-18 form normal differentiated tissue and produce high amounts of hCAP-18/ LL-37 that localizes to both the epidermis and dermis. (a) Neonatal foreskin (top), adult skin tissue from breast (bottom) sections for comparison. Left: hemotoxylin- and eosin (H&E)-stained representative tissue sections. Right: LL-37 immunostaining was detected in green and Hoechst nuclei staining is in blue. (b) Twenty-two-day NIKS tissue (top), NIKShCAP-18 (bottom). Left: H&E-stained representative tissue sections. NIKS has previously been established to form normal skin tissue including basal, spinous, granular, and cornified layers. A high expressing hCAP-18 clone is indistinguishable from NIKS tissue. Right: LL-37 immunostaining was detected in green and Hoechst nuclei staining is in blue. NIKS tissue displays a low level of hCAP-18/LL-37 in the epidermis. NIKShCAP-18 produces high levels of hCAP-18/LL-37 that is found in all layers of the epidermis as well as in the dermis suggesting that the hCAP-18/LL-37 is being secreted outside of the epithelial layers. Bar = 50 µm.

Full-length hCAP-18 and post-translationally cleaved forms, a 4–6 kd LL-37 doublet, were readily detected in cell lysates from the NIKShCAP-18 tissue, but were not observed in tissue lysate made from unmodified NIKS cells (Figure 4a, top). The lower band of the doublet is similar in molecular weight to the LL-37 peptide control. Therefore, the upper band of the doublet likely represents an incomplete post-translational modification of hCAP-18. The full-length hCAP-18 was also present in conditioned media, demonstrating that once it is secreted from keratinocytes, it can permeate both the epidermal and dermal compartments to condition the underlying media (Figure  4a, bottom). The mature 4 kd form (LL-37) was not detected in conditioned medium, but was readily detected in tissue lysates. These results indicate that the mature LL-37 peptide is primarily cell and/or tissue associated. www.moleculartherapy.org vol. 17 no. 3 mar. 2009

© The American Society of Gene Therapy

b

6.5 3.5

Conditioned 20.1 media 14.3

hCAP-18

6.5 3.5 300 250

4 µg PR3

2 µg PR3

1 µg PR3 0.25 µg PR3

No PR3

6.5 3.5

d

100 90 80 70 60 50 40 30 20 10 0 2,597.0

% Intensity

200 150 100 50 NIKS

hCAP-18

Conditioned 20.1 14.3 media

Mature doublet

0

NIKS

NIKS

30

30

LL-37 0.25 mg

NIKShCAP–18

hCAP-18

NIKS

LL-37 0.25 mg

Mature doublet

45

ng/mg of protein

Mature doublet

6.5 3.5

1 µg PR3

hCAP-18

14.3

Cell lysate

0.5 µg PR3

20.1

hCAP-18

20.1 14.3

30

c

No PR3

30

45 Cell lysate

NIKS

NIKShCAP-18 LL-37 0.25 mg

NIKShCAP-18

NIKS

LL-37 0.25 mg

a

Engineered Skin Possesses Antimicrobial Activity

NIKShCAP-18

Mature doublet 6,740.2 38.9

LL-37

5,423.2

4,493.7

3,970.4

5,343.8 6,717.2 Mass (m/z)

8,090.6

9,464.0

Figure 4 NIKS expressing hCAP-18 produces the full-length cathelicidin as well as a mature doublet form. (a) Top immunoblot: cell lysate from NIKS and NIKShCAP-18 tissues. The hCAP-18 (~18 kd) as well as a mature doublet (4–6 kd) can be detected in NIKShCAP-18 tissue. Bottom immunoblot: 24-hour conditioned media from NIKS and NIKShCAP-18 tissues. Immunoblot analysis detects the ~18 kd hCAP-18 in both NIKShCAP-18 cell lysate and conditioned media. Both blots were cropped to exclude duplicate samples. (b) Top immunoblot: cell lysate from day 15 tissue. Bottom immunoblot: 24-hour conditioned media from day 15 tissue. Proteolysis with the PR-3 enzyme cleaves hCAP-18 to mature forms in both cell lysate and ­conditioned media from NIKShCAP-18 tissues. (c) Enzyme-linked immunosorbent assay analysis of tissue lysates indicates that 139-fold more mature hCAP-18/LL-37 is produced by NIKShCAP-18 than in the NIKS. Bars represent mean + SD. (d) Representative MALDI-TOF/TOF mass spectrometry detects a putative LL-37 signal at m/z 4,493 in the NIKShCAP-18 tissue.

Post-translational processing of hCAP-18 produces highly antimicrobial forms that may occur in the proteolytic environment of both burn and chronic wound beds.24–26 To test whether extracellular cleavage of hCAP-18 secreted from NIKShCAP-18 tissue was possible in such a cutaneous wound, both tissue lysates and conditioned media were treated with Proteinase 3 (PR-3), a major enzyme component in wound fluid known to cleave hCAP-18 to its mature LL-37 form.10,27 Figure 4b demonstrates production of the two mature forms of hCAP-18 upon treatment with PR-3. It is known that the secretory signals for granules that contain sequestered HDPs peptide appear to be tightly associated with the later stages of squamous differentiation.28,29 The tissues derived from the NIKShCAP-18 transgenic cell line produced increased levels of both hCAP-18 and LL-37 peptides when compared to tissues derived from the unmodified NIKS cells. Enzyme-linked immunosorbent assay results demonstrate that total hCAP-18/LL-37 was 139-fold greater in lysates of NIKShCAP-18 tissue compared to unmodified NIKS tissue (Figure 4c). NIKShCAP-18 tissue contained ~250 ng hCAP-18/LL-37 per mg tissue protein (estimated at 0.6– 1.7 µmol/l in tissue). To confirm that the doublet detected in cell lysates from the NIKShCAP-18 tissue contained LL-37, matrix-assisted laser desorption/ionization–time of flight/time of flight (MALDI-TOF/TOF) Molecular Therapy vol. 17 no. 3 mar. 2009

mass spectrometry analysis was performed. High-performance liquid chromatography (HPLC) of tissue lysate samples was used to separate the correct peptide fraction for analysis, and was confirmed by immunodot blot (data not shown). MALDI-TOF/TOF mass spectrum reveals the presence of a putative LL-37 peptide at m/z 4,493 in the NIKShCAP-18 tissue (Figure 4d), but not the parental NIKS tissue (data not shown). The higher molecular weight form of the doublet was beyond the size limit of detection for our analysis. Taken together, these findings demonstrate the presence of the mature, correctly processed LL-37 peptide in the genetically modified tissue.

NIKShCAP-18 tissue possesses antimicrobial activity An antibacterial assay was used to evaluate bioactivity of the NIKShCAP-18 tissue. Briefly, tissues were grown in culture for 14 days and conditioned topically with serum-free media for 2 hours. LL-37 has been previously shown to be sequestered in lamellar bodies in the upper layers of squamous epithelia;29 therefore, topical extraction of LL-37 was conducted. The conditioned media was combined in a 96-well microplate with bacterial growth medium containing ~104 colony-forming unit (CFU)/ml Staphylococcus carnosus, a nonpathogenic bacterium sensitive to antimicrobials and closely related to pathogenic strains of Staphylococcus.30 565

© The American Society of Gene Therapy

Engineered Skin Possesses Antimicrobial Activity

a

and NIKShCAP-18 was applied. Seventy-two hours following the treatment, skin tissues were removed and samples of the wound bed tissue were harvested for quantitative bacteriological cultures. NIKShCAP-18 tissue significantly reduced the A. baumannii growth in the burn wound by two logs (3.3 × 104/g tissue) compared to unmodified NIKS tissue which possessed a level of A. baumannii bacteria (3.7 × 106/g tissue) that exceeded the criteria for clinical infection (>105/g tissue) (Figure 5b, left). A second experiment confirmed that NIKShCAP-18 tissue was antimicrobial, reducing the bacterial load of A. baumannii by nearly 3 logs (5.1 × 103 g/tissue) compared to unmodified NIKS tissue (4.6 × 106/g tissue) (Figure 5b, right).

* 110 100

% Bacteria compared to growth control

90 80 70 60 50 40

*

30 20 10 0 NIKShCAP-18

NIKS

b

Experiment 2

***

Experiment 1

**

107

CFU/g tissue

106

105

104

103 NIKS

NIKShCAP-18

NIKS

NIKShCAP-18

Figure 5 Tissue created from NIKS cells expressing hCAP-18 significantly reduces the concentration of Staphylococcus carnosus in vitro and a clinical isolate of the pathogenic bacterium, Acinetobacter baumannii, in an in vivo model of infected burn wounds. (a) Percent inhibition of S. carnosus growth for NIKShCAP-18 tissue as compared to NIKS tissue. *Significant (P value = 0.0039) difference from the control NIKS using an unpaired t-test. (b) Left: first in vivo experiment harvesting the wound bed. Right: second in vivo experiment harvesting the wound bed. CFU/g of bacteria counted after 72 hours exposure to A. baumannii. NIKShCAP-18 reduced the A. baumannii concentration by two logs. **Significant (P value = 0.0286) difference of NIKShCAP-18 (n = 4) from the control NIKS (n = 4) using the Mann–Whitney rank sum test. ***Significant (P value = 0.0127) difference of NIKShCAP-18 (n = 6) from the control NIKS (n = 8) using the Mann–Whitney rank sum test. Bars represent mean + SEM.

Conditioned medium from NIKShCAP-18 tissue inhibited bacterial growth by over 78% (Figure 5a) as compared to the unmodified NIKS confirming the antimicrobial properties of the NIKShCAP-18 tissue.

Infected burn wound model Given that LL-37 has many immunomodulatory, wound healing, and antimicrobial effects that are not measurable in an in vitro system, we used an in vivo murine model of infected burn injury to assess the activity of NIKShCAP-18 tissue. In this model a full-thickness third-degree burn wound is induced by a dorsal scald burn injury.31–33 An inoculum of a multidrug-resistant clinical isolate of A. baumannii (4,000–11,000 CFU) was then applied to the burn wound.34 The clinical isolate of A. baumannii used is resistant to at least 13 known antibiotics including ampicillin, cefazolin, ceftriaxone, cefepime, gentamicin, tobramycin, ciprofloxacin, and amikacin. The following day, eschar from the burn wound was removed and stratified tissue from unmodified NIKS 566

Discussion LL-37 possesses a broad range of antimicrobial activity against bacteria, fungi, and viruses, and has been associated with a variety of therapeutic activities including wound healing.6,12,15 Here, we report the first nonviral tissue-specific genetically engineered human organ produced using naked DNA. The full-thickness human skin substitute was generated with a stably transfected clonal population of NIKShCAP-18, which undergoes squamous differentiation, and expresses ~140-fold more hCAP-18/LL-37 than control tissues. In animal studies, the level of hCAP-18 expression results in a significant reduction of a nosocomial, multidrug-resistant pathogen, A. baumannii, in a wound bed. Our nonviral approach for sustained delivery of hCAP-18 to the wound bed from temporarily applied human skin substitute tissue should also circumvent viral gene transfer issues such as the preexisting immunity exhibited by much of the human population toward adenoviruses. Burn and chronic wounds compromise the skin’s ability to function as a barrier from the environment. Currently, the standard of care for burn wounds uses autologous split thickness skin autografts. Unfortunately, these grafts not only create donor site wounds on the patient, but also in cases of extensive skin loss, are limited by the availability of healthy donor tissue. Autologous monolayer grafts have been used to treat patients who lack sufficient healthy skin for autografting, but these grafts take several weeks to culture before placement onto a wound. Human skin substitutes have many advantages over the current standard of care for large burn trauma wounds and chronic ulcers.35,36 Bacterial contamination is the primary reason for graft failure particularly within the first few days following graft placement on the wound.37 The efficacy for the delivery of hCAP-18 as a cutaneous burn therapy was demonstrated in a recent study in which hCAP-18 was adenovirally expressed in rat burn wounds. The number of Pseudomonas aeruginosa was significantly reduced compared to the carrier control in this model.38 The hCAP-18 expressing human skin substitute described here, created using nonviral methods, can immediately be placed on the wound, reducing the risk of infection from pathogenic bacteria including multidrug-resistant nosocomial strains. Proteolytic activities that are prevalent in many wound environments may enhance the antimicrobial activity of hCAP-18. Recent studies have shown that smaller fragments of hCAP-18, including processed peptides derived from LL-37 and the cathelin-like prodomain, also possess antimicrobial activity.9,11,39 Furthermore, some fragments derived from LL-37 have a lower www.moleculartherapy.org vol. 17 no. 3 mar. 2009

© The American Society of Gene Therapy

minimum inhibitory concentration than intact LL-37. Once liberated from the full-length hCAP-18 protein, the cathelin-like domain exhibits antimicrobial activity against bacteria that are resistant to LL-37, and protects against tissue damage by inhibiting cysteine proteinases.39 Our findings indicate that the wound protease, PR-3, processes hCAP-18 produced by NIKShCAP-18 tissue to a mature doublet containing LL-37. Moreover, an immunoblot of the conditioned media collected topically from NIKShCAP-18 tissue revealed an even smaller processed form of LL-37 for a total of three bands (data not shown). These data suggest that continuous hCAP-18 expression in a human skin substitute could provide antimicrobial activity through processing to smaller peptide forms, in addition to protecting the wound bed from excessive proteolytic activity through the cathelin-like domain. Dramatic killing of the multidrug-resistant bacterium, A. baumannii, was achieved in vivo most likely from hCAP-18 proteolytic cleavage into antimicrobial peptides within an immunologically competent wound bed. LL-37 expression is drastically reduced in chronic cutaneous ulcers, suggesting a potential mechanism for why these wounds fail to heal.6 Enhanced production of cathelicidin using a genetically modified tissue substitute may provide several advantages for wound healing. A report demonstrating LL-37mediated enhancement of defensin production suggests that these HDPs may act synergistically to amplify overall innate immune responses to invading microbes.40 Moreover, there is evidence that LL-37 exhibits wound-healing properties distinct from its antimicrobial function. LL-37 has been detected in healing wounds, and localizes to the leading epithelium migrating over the wound bed. Addition of a neutralizing LL-37 antibody to wounds inhibits reepithelization, suggesting a direct role for LL-37 in keratinocyte growth or migration.6 Others have reported that LL-37 stimulates wound vascularization by promoting endothelial proliferation and the chemotaxis of immune cells.14,41 Furthermore, LL-37 embedded in a PEGT/PBT biopolymer scaffold accelerated neovascularization at the site of a dorsal skin wound in mice.42 Lande et al. have recently linked chronic LL-37-mediated self-DNA presentation to plasmacytoid dendritic cell activation in psoriasis.43 High levels of the active LL-37 peptide (3–10 µmol/l) were required in vitro to stimulate interferon-α production, which is associated with psoriasis. NIKShCAP-18 tissue is estimated to possess 0.6–1.7 µmol/l if all full-length hCAP-18 was converted to active LL-37 peptide, a concentration that is insufficient to stimulate dendritic cell interferon-α production based on Lande et al. Moreover, allogeneic NIKShCAP-18 tissue, as a temporary therapy, would ultimately be replaced by autologous keratinocytes in the wound. Our study demonstrates the potential for NIKS cultured skin substitutes to take an active role in enhancing wound healing by counteracting a major obstacle, infection. HDPs can kill extremely rapidly at concentrations similar to conventional antibiotics. Antibiotics currently in development must both act against bacterial targets, and remain within these microorganisms long enough to kill bacteria before being removed by one of the bacteria’s multidrug efflux systems.5 In contrast to most antibiotics, LL-37 can neutralize lipopolysaccharide on the bacterial outer membrane potentially leading to the attenuation of a detrimental host inflammatory response.44,45 In addition, microorganisms have difficulty acquiring resistance to the HDPs due to their mechanisms of Molecular Therapy vol. 17 no. 3 mar. 2009

Engineered Skin Possesses Antimicrobial Activity

action, making HDPs very attractive for therapeutic use.46,47 In order to gain resistance to HDPs, bacteria must alter the composition of their membranes, an adaptation that reduces virulence and bacterial growth capacity.48 The enhancement of hCAP-18 in skin substitute tissues should circumvent many of these issues as well as alleviate concerns over creation of resistant bacteria. The application of NIKS tissue expressing hCAP-18 to cutaneous wounds will not only provide temporary epithelial coverage and growth factors similar to currently available skin substitute products, but enhanced expression of hCAP-18 should inhibit bacterial growth and facilitate re-epithelialization and subsequent wound closure by the patient’s own keratinocytes. Our studies provide compelling evidence for use of nonviral genetically engineered human skin substitutes to combat multidrug-resistant nosocomial strains that are of great concern to the medical community worldwide.

Materials And Methods Monolayer cell culture and formation of organotypic cultures. NIKS keratinocyte monolayer cell culture was performed as previously described.18 Composite human skin substitutes were prepared using either NIKS or NIKShCAP-18. Human skin substitutes were generated by organotypic culture of keratinocytes on an HA (mixed cellulose esters)-membrane Millicell culturing system (Millipore, Billerica, MA). Briefly, normal human dermal fibroblasts at a density of 3.6 × 104 cells/8-mm well or 5.0 × 106 cells/100-mm well were cultured with type 1 collagen for 4–7 days to form a cellularized dermal matrix. Keratinocytes were seeded onto the surface of the cellularized dermal matrix at a density of 3.75 × 105 cells/8-mm well or 1.3 × 106 cells/100-mm well and maintained at the air/medium interface throughout the stratification process while feeding from below with StrataLife media. Full stratification is typically present by day 14 of organotypic culture. Anchorage-independent growth assay. Anchorage-independent growth

was determined by agar assay. NIKS, NIKShCAP-18, or SCC4 cells were suspended at 5 × 104 cells/ml in 0.3% Noble agar as previously described in ref. 18. Please see Supplementary Materials and Methods for further information. Tumorgenicity in athymic nude mice. NIKS and NIKShCAP-18 were sus-

pended in standard keratinocyte media18 at 5 × 106 cells/100 µl, and SCC4 cells were suspended in standard keratinocyte media at a density of 2.5 × 106/100 µl. One hundred microliter of each cell type and media-only were injected subcutaneously into each flank of five athymic nude mice (for a total of twenty mice) and observed over a period of 8–12 weeks. NIKS and media-only was used as a negative control. SCC4-injected mice developed significant tumors at 8 weeks and had to be euthanized. NIKS and NIKShCAP-18 injected mice were observed for the full 12 weeks. HPLC. Cell lysate from at least two NIKS and NIKShCAP-18 tissues of cell lysate was dried in speed vac and dissolved in water. An aliquot of the concentrated lysate sample was injected onto a Macrosphere C18 column (2.1 mm internal diameter × 250 mm, 300 Å pore size, 5-μm particle size; Alltech Associates, Deerfield, IL) for separation. HPLC separation was performed with a Rainin Dynamax HPLC system (Rainin Instrument, Woburn, MA) equipped with a Dynamax UV-D II absorbance detector. The mobile phases used were (A) deionized water containing 0.1% trifluoroacetic acid and (B) acetonitrile (HPLC-grade; Fisher Scientific, Pittsburgh, PA) containing 0.1% trifluoroacetic acid. The concentrated lysate sample was separated with a gradient of 5–30, 30–65, and 65–95% mobile phase B for 10, 48, and 150 minutes, respectively, at a flow rate of 0.2 ml/min. Fractions were collected every 2 minutes for 50 minutes with a Rainin Dynamax FC-4 fraction collector and dried in speed vac. The fractions were reconstituted in 20 μl water and then analyzed using a model 4800 MALDI TOF/ TOF analyzer (Applied Biosystems, Framingham, MA).

567

© The American Society of Gene Therapy

Engineered Skin Possesses Antimicrobial Activity

Immunodot blot assay. HPLC fractions were assayed for immunore-

To estimate the bacterial inoculum, A. baumannii were grown overnight, spun down, resuspended, and diluted in PBS to an optical density of 0.4. This stock suspension was serially diluted tenfold to a final dilution of 1/100,000. Bacterial colony counts were determined by plating several aliquots of 1/10,000 and 1/100,000 serial dilutions onto Tryptic soy agar plate containing rifampicin (this strain of A. baumannii is resistant to this antibiotic) using a WASP2 spiral plater (Microbiology International, Frederick, MD), and counting the following day. The 1/10,000 dilution was estimated to be ~10,000 CFU bacteria in 200 µl. This inoculum was applied to the burn wound of each mouse. The animals recovered and were returned to animal care. The following day, eschar was removed from the burn wound. A 2.9 × 2.8 cm2 graft of either 23-day-old NIKShCAP-18 tissue or controlunmodified NIKS tissue was applied to the wound bed of the infected animals. The graft was placed on top of the wound bed, dermal side down. The graft was secured into place with Tegaderm and adhesive glue around the dressing. For the first in vivo experiment, 72 hours following the treatment, the NIKS tissue was removed and four 8-mm punch samples of the wound bed were taken for quantitative bacterial culture. For the second in vivo experiment, 72 hours following the treatment, three 8-mm punch samples of the wound bed were taken for quantitative bacterial culture. Viable bacteria from serial dilutions of each sample were counted and expressed as CFU/g of burn wound tissue.

Mass spectrometry. All MALDI-mass spectrometry spectra were obtained with a model 4800 MALDI TOF/TOF analyzer (Applied Biosystems, Framingham, MA) equipped with a 200 Hz, 355 nm Nd:YAG laser. Acquisitions were performed in positive ion linear mode. The matrix used in this work was а-cyano-4-hydroxycinnamic acid dissolved in acetonitrile/ water (70%, vol/vol) yielding a saturated solution at room temperature. Before analysis 0.5 μl of sample solution was spotted on sample plate and mixed well with 0.5 μl matrix solution. The mixture was then allowed to dry before being introduced into the mass spectrometer. Each mass spectrum was generated by averaging 3,000 laser shots over the mass range m/z 1,000–14,000 d. Mass spectra were externally calibrated using 10−5 mol/l cytochrome C applied directly to the MALDI target.

Supplementary Material

activity to rabbit anti-LL-37 polyclonal antibody. A 15 μl solution from each fraction was dried in speed vac and reconstituted in 5 μl 1 mg/ml of bovine serum albumin (RIA grade; Fluka, Buchs, Switzerland) in phosphate-buffered saline (PBS). Fractions were then spotted onto nitrocellulose membrane (Whatman, Sanford, ME) in 1-μl aliquots, which was then dried in 60 °С oven for 1 hour. The membrane was fixed for 16 hours at 37 °С with 2% glutaraldehyde (EM grade; Ted Pella, Redding, CA) vapor in a sealed container. Two percent glutaraldehyde solution in PBS was then added to the membrane and incubated for 1 hour to complete the fixing of the peptides. The blots were washed six times for 15 minutes each in PBS, blocked with 5% nonfat dry milk, 0.1% Triton X-100, 0.001% thimerosal in PBS for 30 minutes, and then incubated with LL-37 ­antibody diluted 1:5,000 in blocking solution for 16 hours at 4 °С. The membrane was rinsed three times using blocking solution to remove unbound antibody. A secondary goat antirabbit antiserum coupled to horseradish peroxidase at a dilution of 1:1,000 in blocking solution was then used to incubate the blot for 1 hour. Again the membrane was rinsed three times in blocking solution for 15 minutes and then visualized by adding the blot to a solution of 0.006% H2O2, 0.03% 3,3′-diaminobenzidine 4-HCl, 0.5% CoCl2 in PBS for 4–5 minutes. The membrane was washed thoroughly in water and dried at room temperature.

Antimicrobial assay. Serum-free keratinocyte medium (Stratatech,

Madison, WI) was applied to the surface of fully stratified human skin substitute tissue created with NIKS or NIKShCAP-18 keratinocytes at day 14. Tissues were incubated for 2 hours at 37 °С in a CO2 incubator to extract protein from tissues. For each of three experiments, at least three tissues of each type were treated and then the topical conditioned media pooled. A solution with 104/ml S. carnosus in 1% Trypticase soy broth plus 10 mmol/l sodium phosphate buffer was combined with the conditioned medium in a 1:1 ratio (total 150 µl) and placed in a 96-well plate in ­triplicate. The plate was incubated at 37 °С while shaking for 1 hour. At the end of 1 hour an ­aliquot of the mixture was taken from each well and either pooled with similar samples or diluted directly from each well and spread on a Trypticase soy agar plate using a WASP2 spiral plater (Microbiology International, Frederick, MD), and incubated at 37 °С overnight. The CFU/ml was calculated according to formulation in the WASP2 user manual and then a final average determined for the samples. Thermal injury and infectious challenge. A full-thickness dorsal scald

burn injury was induced in mice, essentially as described by Walker and Mason49 and in more recent publications.31–33 At the time of injury, the animals were randomized into control, burn, or burn plus infection groups. All mice were anesthetized with 60 mg/kg intraperitoneal sodium pentobarbital. The dorsal hair was clipped and the animals in the latter two groups were placed in a polystyrene template and subjected to a 12% full-thickness scald burn by immersion of the dorsal skin in a 90 °С water bath for 10 seconds. All animals received a 2-ml intraperitoneal injection of 0.9% NaCl for resuscitation. In all experiments mice received a topical inoculation of 4,000–11,000 CFU A. baumannii at the site of the burn.

568

Materials and Methods. 

Acknowledgments We thank Allen Comer and Barry Steiglitz for their editorial assistance with this manuscript, Jens Eickhoff for assistance with statistical analysis and Satoshi Kinoshita for the processing of histological samples. This work was supported by the National Institutes of Health (NIH) SBIR Grants NIH/NIDDK (grant no. R44 DK069924). Please note that a conflict of interest exists for Stratatech affiliated authors.

References

1. Robson, MC (1997). Wound infection. A failure of wound healing caused by an imbalance of bacteria. Surg Clin North Am 77: 637–650. 2. Melstrom, KA Jr., Smith, JW, Gamelli, RL and Shankar, R (2006). New perspectives for a new century: implications of pathogen responses for the future of antimicrobial therapy. J Burn Care Res 27: 251–264. 3. Pollack, M, Taylor, NS and Callahan, LT 3rd (1977). Exotoxin production by clinical isolates of Pseudomonas aeruginosa. Infect Immun 15: 776–780. 4. McGowan, JE Jr. (2006). Resistance in nonfermenting gram-negative bacteria: multidrug resistance to the maximum. Am J Med 119 (6 suppl 1): S29–S36; discussion S62–S70. 5. Rice, LB (2006). Challenges in identifying new antimicrobial agents effective for treating infections with Acinetobacter baumannii and Pseudomonas aeruginosa. Clin Infect Dis 43 (suppl. 2): S100–S105. 6. Heilborn, JD, Nilsson, MF, Kratz, G, Weber, G, Sørensen, O, Borregaard, N et al. (2003). The cathelicidin anti-microbial peptide LL-37 is involved in re-epithelialization of human skin wounds and is lacking in chronic ulcer epithelium. J Invest Dermatol 120: 379–389. 7. Ong, PY, Ohtake, T, Brandt, C, Strickland, I, Boguniewicz, M, Ganz, T et al. (2002). Endogenous antimicrobial peptides and skin infections in atopic dermatitis. N Engl J Med 347: 1151–1160. 8. Zanetti, M, Gennaro, R and Romeo, D (1995). Cathelicidins: a novel protein family with a common proregion and a variable C-terminal antimicrobial domain. FEBS Lett 374: 1–5. 9. Yamasaki, K, Schauber, J, Coda, A, Lin, H, Dorschner, RA, Schechter, NM et al. (2006). Kallikrein-mediated proteolysis regulates the antimicrobial effects of cathelicidins in skin. FASEB J 20: 2068–2080. 10. Sorensen, OE, Follin, P, Johnsen, AH, Calafat, J, Tjabringa, GS, Hiemstra, PS et al. (2001). Human cathelicidin, hCAP-18, is processed to the antimicrobial peptide LL-37 by extracellular cleavage with proteinase 3. Blood 97: 3951–3959. 11. Murakami, M, Lopez-Garcia, B, Braff, M, Dorschner, RA and Gallo, RL (2004). Postsecretory processing generates multiple cathelicidins for enhanced topical antimicrobial defense. J Immunol 172: 3070–3077. 12. Durr, UH, Sudheendra, US and Ramamoorthy, A (2006). LL-37, the only human member of the cathelicidin family of antimicrobial peptides. Biochim Biophys Acta 1758: 1408–1425. 13. Oppenheim, JJ and Yang, D (2005). Alarmins: chemotactic activators of immune responses. Curr Opin Immunol 17: 359–365. 14. De, Y, Chen, Q, Schmidt, AP, Anderson, GM, Wang, JM, Wooters, J et al. (2000). LL-37, the neutrophil granule- and epithelial cell-derived cathelicidin, utilizes formyl peptide receptor-like 1 (FPRL1) as a receptor to chemoattract human peripheral blood neutrophils, monocytes, and T cells. J Exp Med 192: 1069–1074.

www.moleculartherapy.org vol. 17 no. 3 mar. 2009

© The American Society of Gene Therapy

15. Carretero, M, Escamez, MJ, Garcia, M, Duarte, B, Holguin, A, Retamosa, L et al. (2008). In vitro and in vivo wound healing-promoting activities of human cathelicidin LL-37. J Invest Dermatol 128: 223–236. 16. Lau, YF, Bowdish, DM, Cosseau, C, Hancock, RE and Davidson, DJ (2006). Apoptosis of airway epithelial cells: human serum sensitive induction by the cathelicidin LL-37. Am J Respir Cell Mol Biol 34: 399–409. 17. Niyonsaba, F, Someya, A, Hirata, M, Ogawa, H and Nagaoka, I (2001). Evaluation of the effects of peptide antibiotics human beta-defensins-1/-2 and LL-37 on histamine release and prostaglandin D(2) production from mast cells. Eur J Immunol 31: 1066–1075. 18. Allen-Hoffmann, BL, Schlosser, SJ, Ivarie, CA, Sattler, CA, Meisner, LF and O’Connor, SL (2000). Normal growth and differentiation in a spontaneously immortalized neardiploid human keratinocyte cell line, NIKS. J Invest Dermatol 114: 444–455. 19. Kaus, A, Jacobsen, F, Sorkin, M, Rittig, A, Voss, B, Daigeler, A et al. (2008). Host defence peptides in human burns. Burns 34: 32–40. 20. Rosenberg, SA, Aebersold, P, Cornetta, K, Kasid, A, Morgan, RA, Moen, R et al. (1990). Gene transfer into humans—immunotherapy of patients with advanced melanoma, using tumor-infiltrating lymphocytes modified by retroviral gene transduction. N Engl J Med 323: 570–578. 21. Till, BG, Jensen, MC, Wang, J, Chen, EY, Wood, BL, Greisman, HA et al. (2008). Adoptive immunotherapy for indolent non-Hodgkin lymphoma and mantle cell lymphoma using genetically modified autologous CD20-specific T cells. Blood 112: 2261–2271. 22. Heslop, H (2007). Giving Gene Marked (Optional) Epstein Barr Virus (EBV) Specific Cytotoxic T-Cells to Patients With Relapsed EBV-Positive Lymphoma, (ANGEL). Food and Drug Administration. Report no. NCT00058617 . 23. Shin, SI, Freedman, VH, Risser, R and Pollack, R (1975). Tumorigenicity of virustransformed cells in nude mice is correlated specifically with anchorage independent growth in vitro. Proc Natl Acad Sci USA 72: 4435–4439. 24. Neely, AN, Brown, RL, Clendening, CE, Orloff, MM, Gardner, J and Greenhalgh, DG (1997). Proteolytic activity in human burn wounds. Wound Repair Regen 5: 302–309. 25. Prager, MD, Baxter, CR and Hartline, B (1994). Proteolytic activity in burn wound exudates and comparison of fibrin degradation products and protease inhibitors in exudates and sera. J Burn Care Rehabil 15: 130–136. 26. Armstrong, DG and Jude, EB (2002). The role of matrix metalloproteinases in wound healing. J Am Podiatr Med Assoc 92: 12–18. 27. He, Y, Young, PK and Grinnell, F (1998). Identification of proteinase 3 as the major caseinolytic activity in acute human wound fluid. J Invest Dermatol 110: 67–71. 28. Oren, A, Ganz, T, Liu, L and Meerloo, T (2003). In human epidermis, beta-defensin 2 is packaged in lamellar bodies. Exp Mol Pathol 74: 180–182. 29. Braff, MH, Di Nardo, A and Gallo, RL (2005). Keratinocytes store the antimicrobial peptide cathelicidin in lamellar bodies. J Invest Dermatol 124: 394–400. 30. Bera, A, Biswas, R, Herbert, S and Gotz, F (2006). The presence of peptidoglycan O-acetyltransferase in various staphylococcal species correlates with lysozyme resistance and pathogenicity. Infect Immun 74: 4598–4604. 31. Cohen, MJ, Carroll, C, He, LK, Muthu, K, Gamelli, RL, Jones, SB et al. (2007). Severity of burn injury and sepsis determines the cytokine responses of bone marrow progenitor-derived macrophages. J Trauma 62: 858–867.

Molecular Therapy vol. 17 no. 3 mar. 2009

Engineered Skin Possesses Antimicrobial Activity

32. Muthu, K, He, LK, Melstrom, K, Szilagyi, A, Gamelli, RL and Shankar, R (2008). Perturbed bone marrow monocyte development following burn injury and sepsis promote hyporesponsive monocytes. J Burn Care Res 29: 12–21. 33. Cohen, MJ, Shankar, R, Stevenson, J, Fernandez, R, Gamelli, RL and Jones, SB (2004). Bone marrow norepinephrine mediates development of functionally different macrophages after thermal injury and sepsis. Ann Surg 240: 132–141. 34. Shankar, R, He, LK, Szilagyi, A, Muthu, K, Gamelli, RL, Filutowicz, M et al. (2007). A novel antibacterial gene transfer treatment for multidrug-resistant Acinetobacter baumannii-induced burn sepsis. J Burn Care Res 28: 6–12. 35. Andreadis, ST (2007). Gene-modified tissue-engineered skin: the next generation of skin substitutes. Adv Biochem Eng Biotechnol 103: 241–274. 36. MacNeil, S (2007). Progress and opportunities for tissue-engineered skin. Nature 445: 874–880. 37. Horch, RE, Kopp, J, Kneser, U, Beier, J and Bach, AD (2005). Tissue engineering of cultured skin substitutes. J Cell Mol Med 9: 592–608. 38. Jacobsen, F, Mittler, D, Hirsch, T, Gerhards, A, Lehnhardt, M, Voss, B et al. (2005). Transient cutaneous adenoviral gene therapy with human host defense peptide hCAP-18/LL-37 is effective for the treatment of burn wound infections. Gene Ther 12: 1494–1502. 39. Zaiou, M, Nizet, V and Gallo, RL (2003). Antimicrobial and protease inhibitory functions of the human cathelicidin (hCAP18/LL-37) prosequence. J Invest Dermatol 120: 810–816. 40. Carretero, M, Del Rio, M, Garcia, M, Escamez, MJ, Mirones, I, Rivas, L et al. (2004). A cutaneous gene therapy approach to treat infection through keratinocyte-targeted overexpression of antimicrobial peptides. FASEB J 18: 1931–1933. 41. Koczulla, R, von Degenfeld, G, Kupatt, C, Krotz, F, Zahler, S, Gloe, T et al. (2003). An angiogenic role for the human peptide antibiotic LL-37/hCAP-18. J Clin Invest 111: 1665–1672. 42. Steinstraesser, L, Ring, A, Bals, R, Steinau, HU and Langer, S (2006). The human host defense peptide LL37/hCAP accelerates angiogenesis in PEGT/PBT. Ann Plast Surg 56: 93–98. 43. Lande, R, Gregorio, J, Facchinetti, V, Chatterjee, B, Wang, YH, Homey, B et al. (2007). Plasmacytoid dendritic cells sense self-DNA coupled with antimicrobial peptide. Nature 449: 564–569. 44. Larrick, JW, Hirata, M, Balint, RF, Lee, J, Zhong, J and Wright, SC (1995). Human CAP18: a novel antimicrobial lipopolysaccharide-binding protein. Infect Immun 63: 1291–1297. 45. Cirioni, O, Giacometti, A, Ghiselli, R, Bergnach, C, Orlando, F, Silvestri, C et al. (2006). LL-37 protects rats against lethal sepsis caused by gram-negative bacteria. Antimicrob Agents Chemother 50: 1672–1679. 46. Schroder, JM (1999). Epithelial peptide antibiotics. Biochem Pharmacol 57: 121–134. 47. Zasloff, M (2002). Antimicrobial peptides of multicellular organisms. Nature 415: 389–395. 48. Hornef, MW, Wick, MJ, Rhen, M and Normark, S (2002). Bacterial strategies for overcoming host innate and adaptive immune responses. Nat Immunol 3: 1033–1040. 49. Walker, HL and Mason, AD Jr. (1968). A standard animal burn. J Trauma 8: 1049–1051.

569