Porcine β-defensin 2 displays broad antimicrobial activity against pathogenic intestinal bacteria

Molecular Immunology 45 (2008) 386–394

Porcine ␤-defensin 2 displays broad antimicrobial activity against pathogenic intestinal bacteria Edwin J.A. Veldhuizen a,∗ , Mari¨ella Rijnders a , Erwin A. Claassen b , Albert van Dijk a , Henk P. Haagsman a a

Department of Infectious Diseases and Immunology, Faculty of Veterinary Medicine, Utrecht University, Utrecht, The Netherlands b Infection Biology, Division of Infectious Diseases, Animal Sciences Group, Wageningen UR, Lelystad, The Netherlands Received 11 April 2007; received in revised form 5 June 2007; accepted 7 June 2007 Available online 20 July 2007

Abstract Defensins are small antimicrobial peptides that play an important role in the innate immune system of mammals. Here, we describe the antimicrobial activity of pBD-2, a recently discovered new porcine defensin that is produced in the intestine. A synthetic peptide corresponding to the mature protein showed high antimicrobial activity against a broad range of pathogenic bacteria, while it only showed limited hemolytic activity against porcine red blood cells. Highest activity was observed against Salmonella typhimurium, Listeria monocytogenes and Erysipelothrix rhusiopathiae. pBD-2 (4–8 ␮M) killed these pathogens within 3 h. The activity of pBD-2 against S. typhimurium was studied in more detail. At the minimum bactericidal concentration (MBC) of pBD-2, complete killing of S. typhimurium was relatively fast with no viable bacteria left after 90 min. However, antimicrobial activity of pBD-2 was decreased at higher ionic strengths with no residual activity at 150 mM NaCl. Transmission electron microscopy of pBD-2 treated S. typhimurium indicated that relatively low doses of pBD-2 caused a retraction of the cytoplasmic membrane, while pBD-2 concentrations close to the MBC led to cytoplasm leakage and complete lysis of bacterial cells. Considering the site of production and the activity, pBD-2 may be an important defense molecule for intestinal health. © 2007 Elsevier Ltd. All rights reserved. Keywords: Antimicrobial peptide; Innate immunity; Pig

1. Introduction For many decades, sub-therapeutic levels of antibiotics have been added to pig feed as growth promoters. The growing occurrence of antibiotic resistance in microorganisms has raised major concerns about this treatment and eventually has led to a complete ban on the use of antimicrobial growth promoters in 2006 in the European Union. As a consequence, other methods to suppress microbial outgrowth and infection of pigs are needed. One of the alternatives might be the stimulation of the innate immune system of pigs by dietary modulation. Therefore, identification and characterization of new intestinal innate immune effector molecules is required.

∗ Corresponding author at: Department of Infectious Diseases and Immunology, Faculty of Veterinary Medicine, Utrecht University, P.O. Box 80.175, 3508 TD Utrecht, The Netherlands. Tel.: +31 30 2535361; fax: +31 30 2532365. E-mail address: [email protected] (E.J.A. Veldhuizen).

0161-5890/$ – see front matter © 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2007.06.001

Innate immunity is the first line of host defense against invading pathogens. Part of the innate immune response is the secretion of antimicrobial peptides (AMPs). Hundreds of AMPs have already been described in vertebrates, invertebrates, plants and fungi (http://www.cnbi2.com/cgi-bin/amp.pl). AMPs are classified based on structural and sequence homology, with defensins comprising one of the major subclasses of the family of AMPs. Increasing evidence suggests the importance of defensins in the immune response. An upregulation of intestinal defensins is observed as an early response to bacterial infections of the intestine (Zilbauer et al., 2005). In addition, increasing evidence suggests that decreased levels or malfunctioning of defensins in the intestine might lead to disturbance of the intestinal homeostasis resulting in chronic diseases such as Crohn’s disease and ulcerative colitis in human (Wehkamp et al., 2005a,b). Based on the spatial distribution of their six-cysteine residues and the connectivity of the disulfide bonds, defensins can be classified into ␣-, ␤- and ␪-defensins (reviewed in Lehrer,

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

2004; Selsted and Ouellette, 2005). In pig, the information on defensin activity and function is very scarce. Neither ␣- nor ␪defensins have been described and only one ␤-defensin, porcine ␤-defensin 1 (pBD-1) has been studied up to date on a protein level (Shi et al., 1999). Several potentially new porcine ␤-defensins were recently described, based on sequence homology with existing ␤-defensins (Sang et al., 2006), including pBD-2, which is thought to be the porcine orthologue of human ␤-defensin 1. Recently, we showed that the pBD-2 gene is expressed in the intestine (Veldhuizen et al., 2007) and that this expression is upregulated in the porcine ileum epithelial cell line IPI-2I upon infection with the pathogen Salmonella typhimurium DT104 (Veldhuizen et al., 2006). Therefore, this defensin might be a good first candidate for stimulation to increase intestinal health. In this report we describe the antimicrobial activity of pBD2 against a broad range of bacteria that are linked to diarrhea, enterocolitis and related intestinal diseases in pig. In addition, killing kinetics and the effect of pBD-2 on morphology of S. typhimurium were studied using transmission electron microscopy. Finally, the influence of ionic strength on pBD-2 activity and the antiviral and hemolytic activity are described. 2. Materials and methods 2.1. Peptide synthesis A 37 amino acid residues peptide corresponding to the deduced mature pBD-2 peptide was synthesized by Genosphere Biotechnologies (Paris, France) using Fmoc solid-phase synthesis on a Symphony synthesizer (Protein Technology Inc., Tucson, AZ). The peptide was purified by Reversed Phase HPLC on a Zorbax C8 column (Agilent Technologies, Palo Alto, CA) eluted with a 20 min linear gradient of 0–100% acetonitrile in 0.1% (w/v) trifluoroacetic acid. Finally, RP-HPLC purified peptide was dissolved in water and characterized by mass spectrometry.

387

pellets were diluted in 2× assay medium (10 mM phosphate, pH 7.0, 1/100 TSB). Initial concentrations of bacteria were determined by measuring the optical density at 620 nm. To determine cell viability, 100 ␮l of 10-fold serial dilutions in peptone physiological salt solution were transferred onto Trypticase Soy Agar (TSA; Oxoid Limited) plates and colonies were counted after 24 h of incubation. Final dilutions were prepared in 2× assay buffer to reach a cell density of ∼2 × 106 CFU/ml for the antimicrobial assays. The antimicrobial activity of pBD-2 was determined using colony counting assays. Twenty-five microliters of bacterial culture was mixed with 25 ␮l of pBD-2 solution (0–256 ␮g/ml) in polypropylene 96-well microtiter plates and incubated for 3 h at 37 ◦ C. After the incubation period, 200 ␮l of TSB medium was added, further diluted 10- to 1000-fold in TSB, and transferred onto TSA plates after which colonies were counted after 24 h incubation at 37 ◦ C. Incubation steps were performed anaerobically for the antimicrobial assay against C. perfringens. 2.3. Effect of ionic strength The effect of ionic strength on the antibacterial activity of pBD-2 against S. typhimurium DT104 was studied in colony count assays in the presence of 20 or 150 mM NaCl (final concentration). Bacteria were grown and treated as described above except that bacterial pellets were subdivided and further diluted in 2× assay buffer containing 0, 40 or 300 mM NaCl. The antibacterial activity of pBD-2 was assessed as described above. 2.4. Kill-curve studies One hundred microliters of 2 × 106 CFU/ml S. typhimurium was mixed with 100 ␮l of 32 or 128 ␮g/ml pBD-2 (final concentrations 1 × 106 CFU/ml S. typhimurium, 16 or 64 ␮g/ml pBD-2). At various time points, 25 ␮l was taken from the mixture, serially diluted in TSB, and plated out on TSA plates. Plates were incubated o/n at 37 ◦ C after which bacterial colonies were counted.

2.2. Antimicrobial activity 2.5. Hemolytic activity of pBD-2 Antimicrobial activity of pBD-2 was tested against Escherichia coli ATCC 25922 (E. coli), Clostridium perfringens ATCC 12915 (C. perfringens), and clinical isolates of Salmonella enterica serovar Typhimurium DT104 (S. typhimurium), Pseudomonas aeruginosa ATCC 27853 (P. aeruginosa), Staphylococcus aureus ATCC 29213 (S. aureus), Listeria monocytogenes (L. monocytogenes), Yersinia enterocolitica 0:2 (Y. enterocolitica) and Erysipelothrix rhusiopathiae (E. rhusiopathiae). All bacteria were grown and maintained in Trypton Soya broth (TSB, Oxoid Limited) under aerobic conditions, except for C. perfringens, which was grown under an anaerobic atmosphere. Bacteria were cultured to mid-logarithmic phase by transferring 100 ␮l of stationary phase suspension into 50 ␮l TSB medium, followed by incubation and shaking for 3–7 h (depending on the bacteria used) at 37 ◦ C. Mid-logarithmic phase cultures were centrifuged for 10 min at 900 × g, and the bacterial

The release of hemoglobin from porcine erythrocytes was used as a measure for the hemolytic activity of pBD-2, as described by Kl¨uver et al. (2005). In short, 9 ml whole porcine blood was centrifuged at 1800 × g for 10 min at RT. The pellet was washed three times with assay medium (3 g/l TSB, 287 mM glucose) and resuspended in assay medium (in a 100fold diluted concentration of erythrocytes compared to blood). Subsequently, aliquots of 75 ␮l were added to 75 ␮l peptide solutions (final concentration 0–128 ␮g/ml pBD-2) in polypropylene 96-well microtiter plates and the mixture was incubated for 1 h at 37 ◦ C. After incubation the plate was centrifuged for 5 min at 1800 × g and 100 ␮l supernatant of each well was transferred to a new 96-well plate. Extinction was measured at 450 nm and the percentage hemolysis was calculated by comparison with the control samples containing no peptide or 1% Tween-20. Additionally, similar experiments were performed in a second

388

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

test medium (5 mM phosphate, pH 7.4, 5 mM glucose, 154 mM NaCl, final concentration), which is a better reflection of the physiological environment of erythrocytes. 2.6. Transmission electron microscopy Mid-logarithmic phase S. typhimurium cells (∼2 × 108 CFU/ml) were treated with pBD-2 (final concentration: 0–4 ␮g/ml) for 30 min at 37 ◦ C. After treatment, bacterial pellets were prefixed in Karnovsky’s reagent (2% paraformaldehyde, 2.5% glutaraldehyde, 0.25 mM CaCl2 , 0.5 mM MgCl2 in 80 mM sodium cacodylate buffer, pH 7.4) and postfixed in 2% osmiumtetroxide buffered in 0.1 M cacodylate (pH 7.4) for 2 h. The samples were blockstained with 2% uranyl acetate for 1 h and subsequently dehydrated in acetone (50, 70, 80, 96 and 100%). The cells were immersed in acetone/durcupan resin (1:1) overnight, immersed in pure durcupan resin (Fluka, Buchs, Switzerland) for 2 h and embedded in durcupan resin at 60 ◦ C. Ultra-thin sections (thickness, 50 nm), were stained with lead citrate and examined in a CM10 electron microscope (Philips, Amsterdam, The Netherlands). 2.7. Antiviral activity of pBD-2 A North American strain of porcine reproductive and respiratory syndrome virus (PRRSV, 1 × 104.5 TCID50 /ml) was preincubated with pBD-2 (0, 4, 8, 16, 32 and 64 ␮g/ml) for 1 h at 37 ◦ C. After treatment the complete mixture was transferred to 70% confluent precultured monolayers of MA-104 cells for titration. Monolayers were inspected daily for signs of cytopathic effect (CPE). After an incubation of 5 days at 37 ◦ C 5% CO2 , the cells were fixed and an immuno peroxidase monolayer assay was performed using a PRRSV specific monoclonal antibody. 3. Results 3.1. Characterization of the pBD-2 peptide The amino acid sequence of pBD-2 has the highest sequence similarity (60% identity) with human ␤-defensin 1 (hBD-1) and is considered the porcine orthologue of this defensin (Sang et al., 2006). The synthetic mature pBD-2 peptide was produced starting at Asp33 based on the known protein sequence of mature hBD-1 (Fig. 1). Similar cleavage sites of the propeptide are found in the primate orthologues Pan troglodytes (chimpanzee) BD-1 and Hylobates concolor (crested gibbon) BD-1, but also in bovine neutrophil ␤-defensin-1. Reverse phase-

Fig. 1. Sequence alignment of pBD-2 and its orthologue hBD-1. Secondary structure of hBD-1 is shown. Asterisks indicate identical amino acids, conserved amino acid residues are indicated by dots.

Fig. 2. Reverse phase HPLC chromatogram of synthetic pBD-2. The peptide eluted at 11 min using a 0–100% acetonitrile gradient (5%/min) in water.

HPLC (Fig. 2) of the produced peptide indicated >95% purity and mass spectrometry showed a molecular mass of 4084.39, which is in close agreement with the calculated mass of 4086. Circular dichroism of the peptide indicated that secondary structures expected in defensins were present. pBD-2 dissolved in water contained 7% helical and 33% ␤-structures, while in the structure promoting solvent trifluoroethanol, the helical content rose to 20%, while the content of ␤-structures was lowered to 22%. 3.2. Antimicrobial activity Incubations of peptide and bacteria were performed in 5 mM phosphate, pH 7.0, 1/200 TSB. Using these conditions, bacteria were metabolically active as shown by their ability to multiply 2–4 times during the 3 h incubation period (data not shown). In addition, the pH was not affected by this growth due to the buffering capacity of this medium (data not shown). The survival of a wide range of bacteria treated with pBD-2 was determined using colony counting. The data are depicted in Fig. 3 and show that pBD-2 was active against both gram-positive and gram-negative bacteria. Survival of all tested bacteria was affected by pBD-2. The most susceptible bacteria were L. monocytogenes, E. rhusiopathiae and S. typhimurium, where a concentration of 16 or 32 ␮g/ml pBD-2 (∼4–8 ␮M) led to a decline of surviving bacteria below the detection limit of 100 CFU/ml. Higher peptide concentrations of 64–128 ␮g/ml pBD-2 were needed to reduce survival of S. aureus, C. perfringens, E. coli, P. aeruginosa, and Y. enterocolitica by 3 or 4 log units. 3.3. Effect of ionic strength on antimicrobial activity of pBD-2 The antimicrobial activity of pBD-2 against S. typhimurium was determined in the presence of 0, 20 and 150 mM NaCl. This salt has an inhibiting effect on the activity as is shown in Fig. 4. At 20 mM NaCl the concentration of pBD-2 had to be doubled to reach an approximately 3 log units decrease in surviving S. typhimurium compared to the salt-free control. At 150 mM NaCl, antimicrobial activity of pBD-2 was completely abolished.

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

389

Fig. 3. Concentration-dependent inhibition of the growth of bacteria by synthetic pBD-2. In colony counting assays, bacteria were incubated for 3 h with various concentrations of pBD-2 and were counted after 24 h incubation on agar media at 37 ◦ C. Data represent mean ± S.D.

3.4. Hemolytic activity Erythrocytes were collected from fresh porcine blood and incubated with synthetic pBD-2 in two different incubation media. The release of hemoglobin was measured as an indication for the hemolytic activity of pBD-2. The test was performed in two different test media leading to a different outcome at higher pBD-2 concentrations (Fig. 5). In the medium containing 287 mM glucose and 1/10 TSB, less than 20% hemolysis was observed up to 64 ␮g/ml. At 128 ␮g/ml, pBD-2 hemolysis increased to 50%. At physiological conditions in medium

containing 154 mM NaCl, no hemolysis was observed at any pBD-2 concentration, indicating that NaCl, besides antimicrobial activity, also inhibits the hemolytic activity of pBD-2 in vitro. 3.5. Killing kinetics of pBD-2 Kill-curve studies were performed to determine the rate of decrease in viable S. typhimurium upon incubation with pBD-2 at two different concentrations (Fig. 6). At 16 ␮g/ml a decrease below the detection limit was accomplished after 90 min. At a

390

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

Fig. 4. Effect of ionic strength on the antimicrobial activity of pBD-2 against S. typhimurium. Bacteria (1 × 106 CFU/ml) were incubated with various concentrations of pBD-2 in the presence of 0 (triangles), 20 (circles) and 150 mM NaCl (squares). Viability of bacteria was determined using colony count assays. Data represent mean ± S.D.

Fig. 5. Hemolytic activity of pBD-2. Freshly isolated porcine red blood cells were incubated with various concentrations of pBD-2 (0–128 ␮g/ml) in 5 mM phosphate, pH 7.4, 5 mM glucose, 154 mM NaCl (open circles), or 3 g/l TSB, 287 mM glucose (closed circles). Release of hemoglobin, as a measure of hemolysis, was measured at 450 nm. Release of hemoglobin upon addition of Tween-20 was set at 100%. Data represent mean ± S.D.

concentration of 64 ␮g/ml, a reduction below the detection limit of surviving S. typhimurium was already observed after 20 min. These results show that loss of viability of S. typhimurium is pBD-2 concentration dependent and relatively fast. 3.6. Effects of pBD-2 treatment on the ultrastructure of S. typhimurium To further elucidate the nature of the killing mechanism(s) used by pBD-2, S. typhimurium was treated for 30 min with 0–64 ␮g/ml pBD-2, and the bacteria were analyzed by trans-

Fig. 6. Killing kinetics of pBD-2 against S. typhimurium. Bacteria (1 × 106 CFU/ml) were treated with 16 (closed circles) or 64 ␮g/ml pBD-2 (open circles). Samples were taken after various incubation times and viable S. typhimurium were determined using colony count assays. Data represent mean ± S.D.

mission electron microscopy. Compared to the control, pBD-2 treatment resulted in clear morphological changes (Fig. 7). The most obvious change was a retraction of the cytoplasm from the membrane, and an increase in the number of bacteria that contained vacuoles in their cytoplasm. Other morphological changes that were observed were leakage of cytoplasm and irregular septa of dividing cells. Finally, a ghost-like appearance and complete lysis of bacteria were observed at the higher defensin concentrations. The morphological effects were analyzed by counting the number of cells with retracting cytoplasm, vacuoles and the number of lysed cells (Table 1). At higher concentrations of pBD-2, less dividing cells were observed. A clear increase in the number of vacuoles and retraction of cytoplasm was seen at 16 ␮g/ml pBD-2 and higher, while an increase in the number of lysed cells was only observed at 64 ␮g/ml, the highest concentration tested. These results indicate that formation of vacuoles and/or retracting cytoplasm precedes complete lysis of the bacterial cell. 3.7. Antiviral activity of pBD-2 against PRRSV In addition to the antibacterial activity, antiviral activity of pBD-2 was assessed against PRRSV virus. Virus titers after incubation with pBD-2 were determined using fivefold dilutions of the virus and pBD-2 mixture on MA-104 cells. In four of the six performed experiments, lower virus titers (5× or 25×) were observed after incubation with 64 ␮g/ml pBD-2. No effect was seen for lower concentrations of pBD-2 (results not shown).

Table 1 Quantification of morphological changes of S. typhimurium upon incubation with pBD-2 pBD-2 (␮g/ml)

Total cells counted

Dividing cells (% of total)

Vacuoles (% of total)

Retracting cytoplasm (% of total)

Lysed cells (% of total)

0 2 8 16 32 64

408 321 386 421 294 276

6.4 9.0 6.7 6.2 3.4 1.8

10.0 20.2 25.9 37.1 18.7 24.3

1.2 3.4 7.8 2.9 8.5 8.3

4.4 9.0 9.6 7.8 9.9 33

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

391

Fig. 7. Morphological changes of S. typhimurium upon incubation with pBD-2. S. typhimurium (1 × 108 CFU/ml), incubated for 30 min with increasing concentrations of pBD-2 showed dose-dependent changes in ultrastructure, not observed in control bacteria. Shown are (A) undamaged control cells, (B) vacuole formation, (C) retraction of cytoplasmic membrane, (D) damage at septa of dividing bacteria and (E) ‘ghost-like’ appearance of S. typhimurium cells. Bars represent 1 ␮m.

392

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

4. Discussion In the present report we describe the antibacterial, antiviral and hemolytic activity of synthetic pBD-2 in vitro. Synthetic pBD-2 peptide inhibited growth of all bacteria tested, had a very limited effect on reproduction of PRRSV and showed no hemolytic activity under physiological conditions at the concentrations tested. It is unclear to which extent the synthetic peptide used in this study resembles the mature peptide in vivo, since natural pBD-2 has not been purified yet. The N-terminus of the synthetic peptide was based on the known sequence of the mature form of the human homologue hBD-1 and some of its orthologues. However, post-translational N-terminal processing of ␤-defensins is tissue specific as shown by the presence of multiple alternatively processed hBD-1 forms in urine and plasma (Bensch et al., 1995; Valore et al., 1998). Similarly, it is possible that pBD-2 is differently processed in vivo, possibly depending on the site of synthesis. In addition, the synthetic peptide was produced in its reduced form without disulphide linkages. The effect of disulphide linkages on defensin activity has been best determined for hBD-3 (Hoover et al., 2003; Kl¨uver et al., 2005; Wu et al., 2003). It was shown for this defensin that incorrect disulfide bonding had an effect on the chemotactic properties of hBD-3. However, no effect was observed on the antimicrobial spectrum or the activity of the peptide. This lack of effect of disulfide linkages on antimicrobial activity was observed in bovine defensin as well (Hoover et al., 2003; Kl¨uver et al., 2005; Mandal et al., 2002; Wu et al., 2003). The solution structure of the pBD-2 orthologue hBD-1 has been solved (Hoover et al., 2001) and shows a typical ␤defensin shape. In addition, the crystal structures of several hBD-1 mutants with point mutations in the surface exposed area of the peptide were determined (Pazgier et al., 2007). This study showed that none of the point mutations led to structural differences indicating that these amino acids are not involved in the three-dimensional fold of the peptide. Assuming a similar structure of pBD-2, comparison of the primary sequence shows that all the major differences in the primary sequence are on the surface of the molecule. Amino acid mutations leading to local charge differences are the two positively charged lysines at positions 7 and 8 that are not present in hBD-1, while arginine 29 is absent from pBD-2. Finally, a negatively charged glutamic acid at position 24 is present in pBD-2 instead of glutamine. These changes would lead to a more hydrophilic end of the starting helix, while loop 3 would have a more hydrophobic nature. The functional consequences of the differences in the amino acid sequence of pBD-2 compared to hBD-1 (or other defensins) are hard to predict. In our study, a 4 log units decrease in viable S. typhimurium was observed at 16 ␮g/ml (∼4 ␮M) while the least susceptible bacteria P. aeruginosa and C. perfringens required 128 ␮g/ml (∼30 ␮M) for a 3 log decrease in survival. Although reported hBD-1 activity cannot be truly compared to our own findings due to different experimental set-ups and the use of different activity assignments, it is clear that antimicrobial activity of both orthologues are comparable (Chen et al., 2005; Goldman et al., 1997; Jia et al., 2001; Pazgier et al., 2006; Valore et

al., 1998). Furthermore, the observed antimicrobial activity of pBD-2 is in a similar range compared to defensins of other species (Aono et al., 2006; Sang et al., 2005; Thouzeau et al., 2003). In analogy with these, usually better-studied peptides, it is likely that pBD-2 can fulfill an important antimicrobial role in the pig. As is described for most defensins and other AMPs (Bals et al., 1998a,b; Porter et al., 1997; Zucht et al., 1998), sodium chloride inhibited the antibacterial activity of pBD-2. Tomita et al. (2000) reported similar effects of other salts, with a higher inhibiting activity of divalent ions, indicating that the ionic strength and not the specific ions are the cause of the inhibition. Disruption of the initial interaction between the negatively charged membrane of bacteria and the positively charged peptide is thought to be the cause of this dependency of ionic strength on activity. In order to elucidate the mechanism involved in pBD-2 mediated killing, the killing kinetics of pBD-2 against S. typhimurium and the morphological changes of S. typhimurium treated with pBD-2 were examined. Within 90 min all bacteria were killed at 16 ␮g/ml pBD-2 and at higher doses even faster complete killing was observed. This resembles killing kinetics of other defensins (Sahl et al., 2005; van Dijk et al., 2007). With the use of transmission electron microscopy, the effect of pBD-2 on the ultrastructure of S. typhimurium was visualized. Vacuole formation, and retraction of the cytoplasmic membrane were the main effects that could be detected at lower pBD-2 concentrations. Whether these are different morphological changes is not clear, because a local retraction of the cytoplasm could show up as a vacuole in the image, depending on the angle at which the bacteria were cut. At higher PBD-2 concentrations, an increase in ghost-like and lysed cells were observed. The semiquantitative analysis of these morphological changes indicated that vacuole formation and retraction of the cytoplasm precedes cytoplasm leakage and lysis of the bacterial cell. In some studies, other effects such as granulation of intracellular material was described for defensin treated bacteria (Lee et al., 2004; Shimoda et al., 1995; van Dijk et al., 2007), but these effects were not observed in this study. Whether this is due to different experimental conditions or implies a different mode of action of pBD-2 is unclear. Both the antiviral and the hemolytic activity of pBD-2 were very low compared to the antibacterial activity. The difference in membrane composition of bacterial membranes and the host or viral membranes, which are derived from the host, is likely the cause of this specificity of pBD-2. Higher cholesterol levels and a lack of negatively charged phospholipids in the outer leaflet of mammalian membranes inhibit the binding of many antimicrobial peptides (Ishitsuka et al., 2006; Matsuzaki et al., 1995). In addition to direct effects on the viral envelope, antiviral activity of AMPs could also include blocking of viral entry through binding to cellular components of the host or through binding of viral glycoproteins. Furthermore, the cell-to-cell spread of viruses could be affected by AMPs, but so far no correlation between structure or charge of the peptide and antiviral activity has been detected (Jenssen et al., 2006). The effect of pBD-2 on these processes needs further investigation.

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

In conclusion, we determined, for the first time, the antimicrobial activity of pBD-2 against common pathogens that are found in the intestine of the pig. The peptide has antimicrobial activity against a broad range of bacteria, while hemolytic activity of pBD-2 is very low at physiological concentrations. These results, combined with the described inducibility of the peptide in a porcine cell line (Veldhuizen et al., 2006), indicate that pBD-2 could play an important role in the innate immune response against these pathogens in the intestine of pigs. Stimulation of pBD-2 production in the intestine through, for example, dietary modulation could provide a means to increase intestinal health of pigs, thereby reducing the occurrence of infections in pig herds, and the subsequent use of antibiotic treatment. Acknowledgments The authors would like to thank Ton Ultee of the Center for Cell Imaging (Department of Biochemistry & Cell Biology, Faculty of Veterinary Medicine, Utrecht University) for his assistance with the transmission electron microscope; and Julliette Ketelaar for her excellent work in the antiviral activity assays. References Aono, S., Li, C., Zhang, G., Kemppainen, R.J., Gard, J., Lu, W., Hu, X., Schwartz, D.D., Morrison, E.E., Dykstra, C., Shi, J., 2006. Molecular and functional characterization of bovine beta-defensin-1. Vet. Immunol. Immunopathol. 113, 181–190. Bals, R., Goldman, M.J., Wilson, J.M., 1998a. Mouse beta-defensin-1 is a saltsensitive antimicrobial peptide present in epithelia of the lung and urogenital tract. Infect. Immun. 66, 1225–1232. Bals, R., Wang, X., Wu, Z., Freeman, T., Bafna, V., Zasloff, M., Wilson, J.M., 1998b. Human beta-defensin 2 is a salt-sensitive peptide antibiotic expressed in human lung. J. Clin. Invest. 102, 874–880. Bensch, K.W., Raida, M., M¨agert, H.J., Schulz-Knappe, P., Forssmann, W.G., 1995. hBD-1: a novel beta-defensin from human plasma. FEBS Lett. 368, 331–335. Chen, X., Niyonsaba, F., Ushio, H., Okuda, D., Nagaoka, I., Ikeda, S., Okumura, K., Ogawa, H., 2005. Synergistic effect of antibacterial agents human betadefensins, cathelicidin LL-37 and lysozyme against Staphylococcus aureus and Escherichia coli. J. Dermatol. Sci. 40, 123–132. Goldman, M.J., Anderson, G.M., Stolzenberg, E.D., Kari, U.P., Zasloff, M., Wilson, J.M., 1997. Human beta-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis. Cell 88, 553–560. Hoover, D.M., Chertov, O., Lubkowski, J., 2001. The structure of human betadefensin-1: new insights into structural properties of beta-defensins. J. Biol. Chem. 276, 39021–39026. Hoover, D.M., Wu, Z., Tucker, K., Lu, W., Lubkowski, J., 2003. Antimicrobial characterization of human beta-defensin 3 derivatives. Antimicrob. Agents Chemother. 47, 2804–2809. Ishitsuka, Y., Pham, D.S., Waring, A.J., Lehrer, R.I., Lee, K.Y., 2006. Insertion selectivity of antimicrobial peptide protegrin-1 into lipid monolayers: effect of head group electrostatics and tail group packing. Biochim. Biophys. Acta 1758, 1450–1460. Jenssen, H., Hamill, P., Hancock, R.E., 2006. Peptide antimicrobial agents. Clin. Microbiol. Rev. 19, 491–511. Jia, H.P., Starner, T., Ackermann, M., Kirby, P., Tack, B.F., McCray Jr., P.B., 2001. Abundant human beta-defensin-1 expression in milk and mammary gland epithelium. J. Pediatr. 138, 109–112. Kl¨uver, E., Schulz-Maronde, S., Scheid, S., Meyer, B., Forssmann, W.G., Adermann, K., 2005. Structure–activity relation of human beta-defensin

393

3: influence of disulfide bonds and cysteine substitution on antimicrobial activity and cytotoxicity. Biochemistry 44, 9804–9816. Lee, H.Y., Andalibi, A., Webster, P., Moon, S.K., Teufert, K., Kang, S.H., Li, J.D., Nagura, M., Ganz, T., Lim, D.J., 2004. Antimicrobial activity of innate immune molecules against Streptococcus pneumoniae, Moraxella catarrhalis and nontypeable Haemophilus influenza. BMC Infect. Dis. 4, 12. Lehrer, R.I., 2004. Primate defensins. Nat. Rev. Microbiol. 2, 727–738. Mandal, M., Jagannadham, M.V., Nagaraj, R., 2002. Antibacterial activities and conformations of bovine beta-defensin BNBD-12 and analogs:structural and disulfide bridge requirements for activity. Peptides 23, 413–418. Matsuzaki, K., Sugishita, K., Fujii, N., Miyajima, K., 1995. Molecular basis for membrane selectivity of an antimicrobial peptide, magainin 2. Biochemistry 34, 3423–3429. Pazgier, M., Hoover, D.M., Yang, D., Lu, W., Lubkowski, J., 2006. Human beta-defensins. Cell. Mol. Life Sci. 63, 1294–1313. Pazgier, M., Prahl, A., Hoover, D.M., Lubkowski, J., 2007. Studies of the biological properties of human beta-defensin 1. J. Biol. Chem. 282, 1819– 1829. Porter, E.M., van Dam, E., Valore, E.V., Ganz, T., 1997. Broad-spectrum antimicrobial activity of human intestinal defensin 5. Infect. Immun. 65, 2396–2401. Sahl, H.G., Pag, U., Bonness, S., Wagner, S., Antcheva, N., Tossi, A., 2005. Mammalian defensins: structures and mechanism of antibiotic activity. J. Leukoc. Biol. 77, 466–475. Sang, Y., Ortega, M.T., Blecha, F., Prakash, O., Melgarejo, T., 2005. Molecular cloning and characterization of three beta-defensins from canine testes. Infect. Immun. 73, 2611–2620. Sang, Y., Patil, A.A., Zhang, G., Ross, C.R., Blecha, F., 2006. Bioinformatic and expression analysis of novel porcine beta-defensins. Mamm. Genome 17, 332–339. Selsted, M.E., Ouellette, A.J., 2005. Mammalian defensins in the antimicrobial immune response. Nat. Immunol. 6, 551–557. Shi, J., Zhang, G., Wu, H., Ross, C., Blecha, F., Ganz, T., 1999. Porcine epithelial beta-defensin 1 is expressed in the dorsal tongue at antimicrobial concentrations. Infect. Immun. 67, 3121–3127. Shimoda, M., Ohki, K., Shimamoto, Y., Kohashi, O., 1995. Morphology of defensin-treated Staphylococcus aureus. Infect. Immun. 63, 2886–2891. Thouzeau, C., Le Maho, Y., Froget, G., Sabatier, L., Le Bohec, C., Hoffmann, J.A., Bulet, P., 2003. Spheniscins, avian beta-defensins in preserved stomach contents of the king penguin, Aptenodytes patagonicus. J. Biol. Chem. 278, 51053–51058. Tomita, T., Hitomi, S., Nagase, T., Matsui, H., Matsuse, T., Kimura, S., Ouchi, Y., 2000. Effect of ions on antibacterial activity of human beta defensin 2. Microbiol. Immunol. 44, 749–754. Valore, E.V., Park, C.H., Quayle, A.J., Wiles, K.R., McCray Jr., P.B., Ganz, T., 1998. Human beta-defensin-1: an antimicrobial peptide of urogenital tissues. J. Clin. Invest. 101, 1633–1642. van Dijk, A., Veldhuizen, E.J., Kalkhove, S.I., Tjeerdsma-van Bokhoven, J.L., Romijn, R.A., Haagsman, H.P., 2007. The ␤-defensin gallinacin-6 is expressed in the chicken digestive tract and has antimicrobial activity against food-borne pathogens. Antimicrob. Agents Chemother. 51, 912–922. Veldhuizen, E.J.A., Hendriks, H.G., Hogenkamp, A., van Dijk, A., Gaastra, W., Tooten, P.C., Haagsman, H.P., 2006. Differential regulation of porcine betadefensins 1 and 2 upon Salmonella infection in the intestinal epithelial cell line IPI-2I. Vet. Immunol. Immunopathol. 114, 94–102. Veldhuizen, E.J.A., van Dijk, A., Tersteeg, M.H., Kalkhove, S.I., van der Meulen, J., Niewold, T.A., Haagsman, H.P., 2007. Expression of ␤-defensins pBD-1 and pBD-2 along the small intestinal tract of the pig: lack of upregulation in vivo upon Salmonella typhimurium infection. Mol. Immunol. 44, 276–283. Wehkamp, J., Fellermann, K., Herrlinger, K.R., Bevins, C.L., Stange, E.F., 2005a. Mechanisms of disease: defensins in gastrointestinal diseases. Nat. Clin. Pract. Gastroenterol. Hepatol. 2, 406–415. Wehkamp, J., Salzman, N.H., Porter, E., Nuding, S., Weichenthal, M., Petras, R.E., Shen, B., Schaeffeler, E., Schwab, M., Linzmeier, R., Feathers, R.W., Chu, H., Lima Jr., H., Fellermann, K., Ganz, T., Stange, E.F., Bevins, C.L., 2005b. Reduced Paneth cell alpha-defensins in ileal Crohn’s disease. Proc. Natl. Acad. Sci. U.S.A. 102, 18129–18134.

394

E.J.A. Veldhuizen et al. / Molecular Immunology 45 (2008) 386–394

Wu, Z., Hoover, D.M., Yang, D., Boul`egue, C., Santamaria, F., Oppenheim, J.J., Lubkowski, J., Lu, W., 2003. Engineering disulfide bridges to dissect antimicrobial and chemotactic activities of human beta-defensin 3. Proc. Natl. Acad. Sci. U.S.A. 100, 8880–8885. Zilbauer, M., Dorrell, N., Boughan, P.K., Harris, A., Wren, B.W., Klein, N.J., Bajaj-Elliott, M., 2005. Intestinal innate immunity to Campylobacter jejuni

results in induction of bactericidal human beta-defensins 2 and 3. Infect. Immun. 73, 7281–7289. Zucht, H.D., Grabowsky, J., Schrader, M., Liepke, C., J¨urgens, M., SchulzKnappe, P., Forssmann, W.G., 1998. Human beta-defensin-1: a urinary peptide present in variant molecular forms and its putative functional implication. Eur. J. Med. Res. 3, 315–323.