Báo cáo khoa học: Apolipoproteins A-I and A-II are potentially important effectors of innate immunity in the teleost fish Cyprinus carpio

doi:10.1111/j.1432-1033.2004.04228.x

Eur. J. Biochem. 271, 2984–2990 (2004) (cid:1) FEBS 2004

Apolipoproteins A-I and A-II are potentially important effectors of innate immunity in the teleost fish Cyprinuscarpio

Margarita I. Concha1, Valerie J. Smith2, Karina Castro1, Adriana Bastı´as1, Alex Romero1 and Rodolfo J. Amthauer1 1Instituto de Bioquı´mica, Facultad de Ciencias, Universidad Austral de Chile, Valdivia, Chile; 2Gatty Marine Laboratory, School of Biology, University of St. Andrews, Fife, UK

shown to posses antimicrobial activity (EC50 ¼ 3–6 lM) against Planococcus citreus. This peptide was also able to potentiate the inhibitory effect of lysozyme in a radial diffusion assay at subinhibitory concentrations of both effectors. Finally, limited proteolysis of HDL-associated apoA-I with chymotrypsin in vitro was shown to generate a major truncated fragment, which indicates that apoA-I peptides liberated in vivo through a regulated proteolysis could also be involved in innate immunity.

Keywords: antimicrobial cationic peptide; carp; HDL; innate immunity; synergism.

We have previously shown that high density lipoprotein is the most abundant protein in the carp plasma and dis- plays bactericidal activity in vitro. Therefore the aim of this study was to analyze the contribution of its principal apolipoproteins, apoA-I and apoA-II, in defense. Both apolipoproteins were isolated by a two step procedure involving affinity and gel filtration chromatography and were shown to display bactericidal and/or bacteriostatic activity in the micromolar range against Gram-positive and Gram-negative bacteria, including some fish patho- gens. In addition, a cationic peptide derived from the C-terminal region of carp apoA-I was synthesized and

This lipoprotein is constituted by two major apolipopro- teins (apoA-I and apoA-II) and corresponds to the most abundant plasma protein in several teleost fish [9,10], with a concentration as high as 1 gÆdL)1 in the carp [11]. Although the main role of HDL and its principal apolipoproteins has long been considered to be its participation in reverse cholesterol transport and its anti- atherogenic effect [12], more recent studies have involved these proteins in other defensive functions in mammals, such as antiviral, antimicrobial and anti-inflammatory activities [13–15].

The innate immune system is essential to prevent infections during the first critical hours and days of exposure to a pathogen. Although innate immunity is not specific to a particular pathogen in the way that the adaptive immune system is, it is of critical relevance in lower vertebrates such as teleost fish, where the acquired immunity is not well developed [1]. Antibacterial proteins and peptides have been recognized as important effectors of the innate immune system in most animals, however, the importance of these molecules in the primary defense of fish has been only recently demonstrated by several studies [2,3]. Most of these antimicrobial macromolecules have been isolated from fish skin that constitutes a first line barrier against microbial invasion. Surprisingly several of these antimicrobial com- pounds seem to correspond to proteins or protein fragments previously considered nonimmune, e.g. histones H1 and H2A [4–7].

In a previous study we demonstrated that high-density lipoprotein (HDL) locally produced in the carp (Cyprinus carpio) epidermis is secreted to the mucus and displays antimicrobial activity against Escherichia coli in vitro [8].

Multiple alignments of apolipoprotein A-I deduced amino acid sequence shows that the primary structure of this protein is poorly conserved among vertebrates, how- ever, the predicted secondary structure of these proteins is surprisingly similar (high content of amphipathic a-helix). Therefore we hypothesized that in spite of the low sequence similarities that exist between mammalian and teleost apolipoprotein A-I, its conserved overall structure would be responsible for preserving these defensive functions through evolution. The aim of this study was to evaluate if the antimicrobial activity observed for carp HDL resides in its major apolipoproteins (apoA-I and apoA-II) and in addition to determine if a synthetic peptide derive from apoA-I sequence could display a similar activity.

Materials and methods

Blood sample collection

Common carp (C. carpio L) were caught in the Cayumapu river (Province of Valdivia, Chile) and maintained in an outdoor tank with running river water. Fish weighing

Correspondence to M. I. Concha, Instituto de Bioquı´ mica, Universi- dad Austral de Chile, Campus Isla Teja, Valdivia, Chile. Fax: + 56 63 221 107, Tel.: + 56 63 221 108, E-mail: mconcha@uach.cl Abbreviations: AMP, antimicrobial peptide; Apo, apolipoprotein; EC50, effective inhibitory concentration; HDL, high-density lipoprotein; MBC, minimal bactericidal concentration; MHB, Mueller–Hinton broth. (Received 13 February 2004, revised 6 April 2004, accepted 25 May 2004)

ApoA-I and apoA-II in innate immunity of the carp (Eur. J. Biochem. 271) 2985

(cid:1) FEBS 2004

selected for immunization after checking their preimmune sera by Western blot. The immunization schedule consisted of one subcutaneous injection of antigen plus Freund’s complete adjuvant, two injections of antigen plus Freund’s incomplete adjuvant spaced by a period of 12 days and a final booster. The ethical guidelines, from the UK Home Office, on animal care were followed.

800–1200 g were acclimatized at 20 ± 2 (cid:2)C with a photo- period of 14-h light : 10-h dark, for at least 3 weeks before they were used. Animals were anesthetized in a bath containing 50 mgÆL)1 of benzocaine and blood samples were collected from the caudal vein in heparinized tubes. The ethical guidelines, from the UK Home Office, on animal care were followed.

Antimicrobial activity assays

Bacterial strains and culture

Field isolates of the salmonid pathogens Yersinia ruckeri and Pseudomonas sp. were kindly provided by M. Fernan- dez (Fundacio´ n Chile, Puerto Montt, Chile) and were typed with Mono-year (BIONOR, Norway) and API 20NE (BioMerioux, France) kits, respectively. Fish bacteria, Planococcus citreus (NCIMB 1493) and E. coli DH5a were grown to logarithmic phase in Mueller–Hinton broth (Merck) at the appropriate temperature (20 (cid:2)C for fish pathogens and P. citreus and 37 (cid:2)C for E. coli).

HDL and apolipoprotein isolation

Carp plasma HDL was purified from fresh plasma samples treated with protease inhibitors (phenylmethanesulfonyl fluoride and benzamidine) by affinity chromatography on Affi-Gel(cid:3) Blue-Gel (Bio-Rad), essentially as described by Amthauer and coworkers [16]. ApoA-I and A-II were isolated from HDL particles according to Amthauer and coworkers [9]. Briefly, HDL was delipidated with eth- anol:ether (3 : 2, v/v) at ) 20 (cid:2)C. One milliliter of the delipidated plasma HDL (5 mgÆml)1) was loaded on a Sephacryl S-200 (Pharmacia) column (100 · 1.5 cm) equil- ibrated with 10 mM Tris/HCl pH 8.6/8 M urea/1 mM EDTA and eluted with the same buffer at a flow rate of 0.3 mLÆmin)1. Fractions corresponding to the three peaks were pooled, exhaustively dialyzed in 5 mM Tris/HCl pH 8.0/0.1 mM EDTA and concentrated 10-fold in a Speed-Vac centrifuge. Prior to its use in antimicrobial assays, protein concentration of apolipoprotein samples was determined by the bicinconinic acid method [17] and its purity and integrity was checked by SDS/PAGE according to Laemmli [18].

Peptide synthesis

A 24-residue peptide derived from the C-terminal sequence of carp apoA-I [AQEFRQSVKSGELRKKMNELGRRR] was produced, using N-(9-fluorenyl)methoxycarbonyl chemistry and purified to > 70% by HPLC (Global Peptide Services LLC, Fort Collins, CO, USA).

Determination of the effective 50% reduction concentration (EC50) of the purified protein/peptide against each of the test bacteria used was performed using the microtiter broth dilution assay [19]. One hundred microliters of each bacterial suspension containing 105 colony-forming units per mL was mixed with serial twofold dilutions of test protein/peptide in 0.2% (w/v) bovine serum albumin in sterile polypropylene 96-well microtiter plates (Corning Costar, Cambridge, UK). The positive control contained bacteria and diluent only. P. citreus, Y. ruckeri and Pseudo- monas sp. were incubated at 20 (cid:2)C and E. coli at 37 (cid:2)C and the attenuance (D) was read at 570 nm using a MRX II microtiter plate reader (Dynex, West Sussex, UK) against a blank comprising diluent only. Values for experimental wells were recorded when the attenuance reached 0.2 in the positive control well. The EC50 was considered to be the lowest concentration of protein that reduces the growth by 50% relative to the control. The minimal bactericidal concentration (MBC) was obtained by plating out the contents of each well showing no visible growth. MBC was taken as the lowest concentration of protein that prevents any residual colony formation after incubation for 24 h. Synergism between hen egg white lysozyme and the carp apoA-I synthetic peptide was assessed using a modified version of the two-layer radial diffusion assay of Lehrer et al. as described by Smith et al. using P. citreus Gram-positive as a test bacteria [20,21]. Briefly, bacteria grown exponen- tially in Mueller–Hinton broth were washed, resuspended in Mueller–Hinton broth (MHB) and adjusted to an attenu- ance at 570 nm of 0.4. An aliquot of 100 lL of the bacterial suspension was mixed with 15 mL of melted sterile Mueller– Hinton agar (0.1· MHB in 1 gÆdL)1 agar), immediately prior to its solidification and poured into a sterile square (100 · 100 mm) Petri dish. Once solidified for 15 min at 4 (cid:2)C, 0.3-mm diameter wells were bored into the agar using a sterile plastic Pasteur pipette. Three microliter aliquots of different combinations of the peptide and lysozyme, each at subinhibitory concentrations were loaded into each well and allowed to diffuse for at least 3 h at 4 (cid:2)C. After the diffusion step, melted top agar (1· MHB in 1 gÆdL)1 agar) was poured onto the dishes and after 20–24 h of incubation at 20 (cid:2)C the diameter of the inhibition halos were measured.

Antiserum preparation

Limited proteolysis and Western blotting

To obtain a limited proteolysis of HDL-associated apoA-I; HDL particles (200 lgÆmL)1 in 100 mM ammonium bicar- bonate buffer) was incubated with bovine pancreas chymo- trypsin at 37 (cid:2)C using a molar ratio of protease to lipoprotein (1 : 100) and taking aliquots each 30 min over 4 h. The reaction was stopped by heating the samples at 100 (cid:2)C for 5 min in sample buffer [62.5 mM Tris/HCl;

Antiserum to apoA-I synthetic peptide coupled to keyhole lymphet hemocyanin was raised in rabbits by the following procedure. Briefly, 4 mg of peptide was dissolved in 1 mL of sterile NaCl/Pi and mixed with 1 mL of 8 mgÆmL)1 hemocyanin. Twenty-five aliquots of 100 mM glutaralde- hyde (20 lL each) were slowly added to the mix while stirring at room temperature for 1 h. The reaction was stopped by the addition of an excess of glycine and diluted aliquots were stored at ) 20 (cid:2)C until used. Rabbits were

2986 M. I. Concha et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

protein bands on SDS/PAGE (lane 1), that correspond to apoA-I and A-II. Delipidation of the concentrated HDL fractions and separation on Sephacryl S-200 gel filtration chromatography resulted in three major peaks. The first peak was shown to contain aggregates of both apolipopro- teins, while peaks 2 and 3 contained isolated apoA-I and apoA-II, respectively (Fig. 1, insert, lanes 2 and 3). The identity of both apolipoproteins was demonstrated not only by their expected molecular mass (27.5 and 12.5 kDa, respectively) but also by Western blot analysis using previously characterized antibodies specific for each apo- lipoprotein (data not shown) [8].

Antimicrobial activity of purified apolipoproteins

2% w/v SDS; 10% v/v glycerol; 5% v/v 2-mercaptoethanol and bromophenol blue). The products of proteolysis were analyzed by Tricine-SDS/PAGE essentially as described by Scha¨ gger and von Jagow [22] and then transferred to nitrocellulose membranes using a semidry blotter unit. Membranes were blocked for 1 h with 1% (w/v) bovine serum albumin in NaCl/Pi buffer and then alternatively incubated a further hour with rabbit anti-carp apoA-I serum diluted 1 : 25 000, rabbit anti-apoA-I synthetic peptide serum diluted 1 : 1500 or with rabbit anti-carp apoA-II diluted 1 : 1000 in the same blocking solution. After several washes with NaCl/Pi, the membranes were incubated for 1 h with a 1 : 2000 dilution of alkaline phosphatase-conjugated goat anti-rabbit IgG (Gibco BRL). The blot was developed by incubating the membranes for 10 min in phosphatase buffer (0.1 M Tris/HCl, pH 9.5; 0.1 M NaCl; 5 mM MgCl2) 0.16 mgÆmL)1 containing of 5-bromo-4-chloroindolyl phosphate and 0.33 mgÆmL)1 nitroblue tetrazolium.

Results

Purification of HDL and apolipoproteins A-I and A-II

As shown in Fig. 1 (insert), the plasma HDL particles isolated by affinity chromatography display essentially two

Quantification of antimicrobial activity using the microtiter broth dilution assay showed that apoA-I is active at submicromolar concentrations, with an EC50 and a MBC of approximately 0.4 lM against P. citreus (Gram-positive) and at micromolar concentrations (2.6–4.0 lM) against two Gram (–) fish pathogens Pseudomonas sp. and Yersinia ruckeri (Table 1). In addition, purified apoA-II also dis- played bacteriostatic activity against the Gram-positive and -negative bacteria at micromolar concentrations (Table 1). These results clearly show that although apoA-I seems to be more active than apoA-II, both major apolipoproteins contribute significantly to the antimicrobial activity dis- played by carp plasma HDL.

Design and evaluation of apoA-I synthetic peptide

Based on our observations and on several studies that have shown that mammalian apoA-I associated to HDL parti- cles suffers limited proteolysis in vitro by several potentially relevant insult-activated proteases (e.g. tryptase, chymase and several matrix metalloproteases) [23,24], we hypothes- ized that during acute inflammation one or more peptides could be released from the HDL particle by proteolysis either from the N- or C-terminal region of apoA-I. In this context, we postulate that these putative peptides could also contribute to the systemic and mucosal innate immunity. Initially we analyzed the carp apoA-I amino acid sequence deduced from the sequence of a partial cDNA clone isolated in a previous study [8] and we found that this sequence is predicted to posses a high content of amphipathic a-helix (Fig. 2B). In particular, a peptide corresponding to the last 24 residues (Fig. 2A) would be a highly cationic helix (net charge + 5) although not amphi- pathic. Thus, this peptide should share some important

Fig. 1. Purification of apolipoproteins A-I and A-II from isolated carp plasma HDL. HDL-associated apolipoproteins were purified by gel filtration chromatography on Sephacryl S-200. Dialyzed and concen- trated fractions of each peak were separated by SDS/PAGE. (Insert) Carp plasma (lane P), lanes 1–3 correspond to peaks 1–3, respectively (50 lg protein per lane). Arrows indicate the migration of carp apoA-I (27.5 kDa) and A-II (12.5), respectively.

Table 1. Bacteriostatic and bactericidal activities of carp apoA-I and A-II. Each value in the table represents the mean ± SE of experiments performed in triplicate. Similar results were obtained with different preparations of apolipoproteins. ND, not determined.

ApoA-I ApoA-II

Bacterium Gram staining MBC (lM) EC50 (lM) EC50 (lM)

Planococcus citreus Pseudomonas sp. Yersinia ruckeri Escherichia coli + – – – 0.3 ± 0.06 2.6 ± 0.01 2.6 ± < 0.001 5.2 ± 0.85 0.4 ± 0.2 ND 4.0 ± 0.5 8.5 ± 0.5 1.8 ± < 0.001 3.5 ± 0.04 3.7 ± 0.15 7.1 ± < 0.001

ApoA-I and apoA-II in innate immunity of the carp (Eur. J. Biochem. 271) 2987

(cid:1) FEBS 2004

[25,26]. Therefore we synthesized this C-terminal peptide and evaluated its antimicrobial activity in vitro. The synthetic peptide was active against P. citreus displaying an EC50 of 3–6 lM.

Considering that other cationic proteins and peptides have been shown to exhibit synergism with hen egg white lysozyme [27–29], we attempted to ascertain if the carp apoA-I peptide would be also able to synergize with lysozyme. As shown in Fig. 3, the synthetic peptide enhanced the activity of lysozyme when both compounds were used at subinhibitory concentrations in a radial diffusion assay against P. citreus. Maximal synergism was observed at concentrations of 6 lgÆmL)1 and 0.8 mM of lysozyme and peptide, respectively (Fig. 3).

Limited proteolysis of HDL-associated apoA-I invitro

Fig. 3. Synergy of the apoA-I synthetic peptide with lysozyme. (A) Bacterial growth in the presence of different combinations of the peptide and lysozyme, each at subinhibitory concentrations, was analyzed by radial diffusion assay using P. citreus as test bacterium. Variable concentrations of lysozyme without peptide (d); plus 0.2 mM (h); 0.4 mM (m); 0.6 mM (e) or 0.8 mM (r) of the synthetic peptide. The experiments were performed in triplicate and the error bars cor- respond to the standard error around the mean. (B) Depicts the increased inhibitory halo observed with increasing concentrations of peptide were used in combination with 6 lgÆmL)1 of lysozyme. Wells 1–5 correspond to the same peptide concentration as in (A), ranging from 0 to 0.8 mM, respectively.

To determine if one or more peptides could be liberated after limited proteolysis of apoA-I in vitro, we analyzed the kinetics of carp HDL digestion with chymotrypsin by SDS/ PAGE (Fig. 4A). Under the conditions used, two major truncated apoA-I fragments were generated; one of them seemed to be short-lived while the third band remained stable far more than 3 h. Duplicate gels were transferred to nitrocellulose membranes and analyzed by Western blot using specific antiserum against the intact carp apoA-I or the C-terminal apoA-I synthetic peptide. As shown on

structural features with a known group of antimicrobial peptides (AMPs) also released by regulated proteolysis from larger precursor polypeptides (e.g. cathelicidins)

Fig. 2. Prediction of a-helicity of apoA-I peptide and three-dimensional model of carp apoA-I. (A) Helical wheel projection of the synthetic peptide performed with ANTHEPROT V.5. program (http:// www.antheprot-pbil.ibcp.fr). A discontinuous line was used to separ- ate the helix in two faces. The preferential localization of the positively charged residues on the upper face of the helix is depicted. The amino acid sequence of the peptide is shown at the top of the figure; basic residues are underlined. (B) The three-dimensional model of the partial carp apoA-I sequence was generated by SWISS-MODEL (http:// www.expasy.org/swissmod/SWISS-MODEL.html) based on the crystallographic data for human apoA-I. The N-terminal residue (N) corresponds to the first residue of the carp apoA-I partial sequence (GenBank accession number AJ308993) and (C) corresponds to the C-terminal residue. Hydrophilic residues are in dark gray and hydrophobic residues in light gray.

2988 M. I. Concha et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

growth of Gram-positive and -negative bacteria, including fish pathogens, at micromolar concentrations. These find- ings indicates that HDL and its apolipoproteins could constitute important effectors in the systemic innate defense mechanisms of the carp, especially taking in consideration that the plasma concentration of HDL-associated apolipo- proteins reaches values as high as 1 gÆdL)1 irrespective, of the acclimatization condition of the fish [11]. Although the relative abundance of HDL varies among different teleosts, it is generally accepted that this lipoprotein is clearly more abundant in fish plasma than in higher vertebrates [10]. This situation probably reflects among other things, the need of teleost fish to rely more on their innate immunity for survival. As we described previously [8], apoA-I and apparently also apoA-II are locally synthesized and secreted in the carp epidermis as a nascent HDL particle. Although as yet we cannot state unequivocally that this particle or even plasma HDL contribute significantly to innate defense, the present study, together with the previous work described by Concha et al. [8], offers promising evidence that they might. Certainly there is no reason why apolipoproteins in the skin secretion should function independently from apolipoproteins and HDL in the plasma. In both, apoA-I and apoA-II are derived from HDL particles. While the size of skin nascent HDL is different from that of plasma HDL, it contains both apoA-I and apoA-II, molecules shown by the present paper to have potent antimicrobial properties in vitro. Work is currently underway to investigate the precise mechanism by which HDL and the associated apolipoproteins act. The results of these studies should help to confirm the biological role of these proteins.

Fig. 4B, the intact apoA-I (band a), an intermediary fragment (band b) and a more stable third band (band c) were recognized by the specific antiserum against carp apoA-I. However, when incubated with the antiserum against the synthetic carp apoA-I peptide, only bands a and b were immunodetected while the third band, which was the most abundant after 30 min of digestion was not detected by this antiserum, indicating that it would correspond to a fragment truncated both at the N- and C-terminal end of the protein. The detection of the larger fragment (band b) of apoA-I in Fig. 4B indicates that it still contains at least part of the epitope(s) recognized by the antipeptide antibodies. Using the same antiserum we could not detect a band in the range of molecular masses expected for the peptide (3 kDa). However with the antiserum against intact apoA-I we observed during the first minutes of digestion a very faint band that could correspond to this peptide (data not shown). In the same experiment, no degradation was observed for apoA-II neither by direct staining (Fig. 4D) nor by Western blot (Fig. 4E), reflecting that in the HDL particle, apoA-II should be much less exposed to the protease than apoA-I.

Fig. 4. Limited proteolysis of HDL-associated apoA-I. (A) Tricine- SDS/PAGE and Coomassie blue staining were used to analyze the progress of HDL-associated apoA-I proteolysis with chymotrypsin. (B,C) Western blot analyses of the gel in (A) immunodetected with a specific anti-apoA-I and anti-peptide serum, respectively. Arrows indicate the different bands of apoA-I: (a) intact form; (b) intermediary fragment and (c) stable truncated apoA-I. Incubation time is shown above each gel. (D) Tricine-SDS/PAGE and Coomassie blue staining of HDL-associated apoA-II incubated with chymotrypsin under the same conditions as in (A). (E) Western blot analysis of the gel in (D) using a specific anti-apoA-II serum.

Discussion

Although the primary structure of apoA-I is poorly conserved among different species, the overall secondary and tertiary structure of HDL-associated apoA-I is remarkably similar, displaying an arrangement of several amphipatic a- helices in a horseshoe-shape structure [12]. In fact, it has been demonstrated that various HDL functions (e.g. activation of lecithin-cholesterol acyltransferase or lipid binding) are dependent on these structural features of apoA-I [31]. In view of the fact that an important group of antimicrobial peptides (cationic peptides) also have a-helical structure, in the present study we demonstrate that a cationic peptide analog to the C-terminus of carp apoA-I exhibits in vitro antimicrobial activity at micromolar concentration. This peptide was susceptible to salt as no activity was detected at 150 mM NaCl. This is a rather common feature among antimicrobial peptides, for example magainins and cecropins which also correspond to a-helical peptides are inhibited at 100 mM NaCl [19]. In the particular case of carp apoA-I, it could be argued that if a C-terminal peptide would be released in vivo it would be expected to be more active in a low-salt environment like the mucus of this freshwater fish than in its blood stream. Another interesting feature of this C-terminal peptide is its ability to synergize with lysozyme. Synergy of several antimicrobial peptides and proteins with lysozyme has been previously described [27,28]. In teleosts, the blood and skin mucosa are particularly rich in lysozymes [32,33], so as HDL is also very abundant in these tissues it could assist in pathogen killing. At this point we cannot assure that such a synergism observed in vitro would be physiologically relevant, neither can we rule out a possible synergism between intact apolipoproteins and lysozyme.

Although there are a few studies of mammalian HDL and its principal apolipoproteins A-I and A-II in antimicrobial or antiviral activities in vitro [13,14,30], these proteins have not been yet recognized as important effectors in innate immunity. Moreover, only recently we reported that this defensive function could also be relevant for teleost fish [8]. Here we clearly demonstrate the important contribution of both apolipoproteins A-I and A-II in the in vitro anti- microbial activity of carp HDL. Both proteins inhibit the

ApoA-I and apoA-II in innate immunity of the carp (Eur. J. Biochem. 271) 2989

(cid:1) FEBS 2004

References

Once the antimicrobial mechanism of action of apolipopro- teins has been established it would be very interesting to evaluate these and several other possibilities of synergistic and additive effects between effectors.

1. Magor, B.G. & Magor, K.E. (2001) Evolution of effectors and receptors of innate immunity. Dev. Comp. Immunol. 25, 651– 682.

2. Fernandes, J.M., Kemp, G.D., Molle, G. & Smith, V.J. (2002) Antimicrobial properties of histone H2A from skin secretions of rainbow trout, Oncorhynchus mykiss. Biochem. J. 368, 611–620. 3. Fernandes, J.M. & Smith, V.J. (2002) A novel antimicrobial function for a ribosomal peptide from rainbow trout skin. Bio- chem. Biophys. Res. Commun. 296, 167–171.

4. Richards, R.C., O’Neil, D.B., Thibault, P. & Ewart, K.V. (2001) Histone H1: an antimicrobial protein of Atlantic salmon (Salmo salar). Biochem. Biophys. Res. Commun. 284, 549–555.

5. Cho, J.H., Park, I.Y., Kim, M.S. & Kim, S.C. (2002) Matrix metalloproteinase 2 is involved in the regulation of the anti- microbial peptide parasin I production in catfish skin mucosa. FEBS Lett. 531, 459–463.

6. Birkemo, G.A., Luders, T., Andersen, O., Nes, I.F. & Nissen- Meyer, J. (2003) Hipposin, a histone-derived antimicrobial peptide in Atlantic halibut (Hippoglossus hipoglossus L.). Biochim. Bio- phys. Acta 1646, 207–215.

7. Fernandes, J.M., Molle, G., Kemp, G.D. & Smith, V.J. (2004) Isolation and characterization of oncorhyncin II, a histone H1-derived antimicrobial peptide from skin secretions of trout, Oncorhynchus mykiss. Dev. Comp. Immunol. 28, 127–138.

8. Concha, M.I., Molina, S., Oyarzu´ n, C., Villanueva, J. & Amthauer, R. (2003) Local expression of apolipoprotein A-I gene and a possible role for HDL in primary defence in the carp skin. Fish Shellfish Immunol. 14, 259–273.

9. Amthauer, R., Villanueva, J., Vera, M.I., Concha, M.I. & Krauskopf, M. (1989) Characterization of the major plasma apolipoproteins of the high density lipoprotein in the carp (Cyprinus carpio). Comp. Biochem. Physiol. 92B, 787–793. 10. Babin, P.J. & Vernier, J.M. (1989) Plasma lipoproteins in fish. J. Lipid Res. 30, 467–489.

The above results raise the possibility that not only the intact apolipoproteins but also putative fragments derived from their limited proteolysis could participate in innate defense. Additional support for this idea comes from previous studies and our results that show that HDL- associated apoA-I is susceptible to limited proteolysis by physiologically relevant proteases, such as those liberated by neutrophils and mast cells after an insult [23,24]. In the present study, chymotrypsin was used because it has the same specificity than chymase, a protease released by mast cells, which has previously been shown to produce human apoA-I truncated either at the N- or at the C-terminus [23]. In this same study it was also demonstrated that apoA-II is resistant to degradation under the conditions used. Our results are in close agreement with these data as following the digestion with chymotrypsin, a stable apoA-I fragment that seems to lack both N- and C-termini, is generated. We also observed negligible degradation of apoA-II associated to HDL. Although we could not detect the C-terminal peptide released from apoA-I by Western blot utilizing the specific antipeptide serum, it must be considered that under the in vitro conditions of protease digestion, the peptide could be very short-lived and therefore extremely hard to detect. Based on these preliminary results we postulate that besides the constitutive contribution of HDL and its apolipoproteins in teleost fish innate immunity, an additional mechanism might involve the release of one or more antimicrobial peptides by limited proteolysis of HDL-associated apoA-I possibly triggered by one or more insult-regulated proteases, e.g. elastase or chymase. Such a mechanism has already been described for another nonimmune protein, histone H2A, in catfish skin, where a complex cascade of injury-induced proteases is involved in the regulation of the AMP parasin I production [5]. Therefore further studies will attempt to evaluate the presence of the peptide in the mucus and plasma of pathogen-challenged fish.

11. Krauskopf, M., Amthauer, R., Araya, A., Concha, M.I., Leon, G., Rios, L., Vera, M. & Villanueva, J. (1988) Temperature acclimatization of the carp. Cellular and molecular aspects of the compensatory response. Arch. Biol. Med. Exp. 21, 151–157. 12. Bolanos-Garcia, V.M. & Miguel, R.N. (2003) On the structure and function of apolipoproteins: more than a family of lipid- binding proteins. Prog. Biophys. Mol. Biol. 83, 47–68.

13. Srinivas, R.V., Venkatachalapathi, Y.V., Rui, Z., Owens, R.J., Gupta, K.B., Srinivas, S.K., Anantharamaiah, G.M., Segrest, J.P. & Compans, R.W. (1991) Inhibition of virus-induced cell fusion by apolipoprotein A-I and its amphipathic peptide analogs. J. Cell. Biochem. 45, 224–237.

Given that anti-inflammatory, antiviral, antibacterial activities have been reported for mammalian HDL and its apolipoproteins [13–15,30], the findings described in the present study showing antimicrobial activity for teleost apolipoproteins A-I and A-II and for a synthetic peptide derived from apoA-I, further confirm the multifunctionality of these proteins. Moreover the synergism observed between the apoA-I synthetic peptide and lysozyme suggests that a mechanism involving the regulated release of peptides from the HDL-associated apoA-I present in plasma and mucus could be very important in the context of innate defense in fish.

14. Tada, N., Sakamoto, T., Kamgami, A., Mochizuki, K. & Kur- osaka, K. (1993) Antimicrobial activity of lipoprotein particles containing apolipoprotein A-1. Mol. Cell. Biochem. 119, 171–178. 15. Burger, D. & Dayer, J.M. (2002) High-density lipoprotein-asso- ciated apolipoprotein A-I: the missing link between infection and chronic inflammation? Autoimmun. Rev. 1, 111–117.

Acknowledgements

16. Amthauer, R., Concha, M.I., Villanueva, J. & Krauskopf, M. (1988) Interaction of Cibacron Blue and anilinonaphthalene sul- phonate with lipoproteins provides a new means for simple iso- lation of these plasma proteins. Biochem. Biophys. Res. Commun. 154, 752–757.

17. Smith, P.K., Krohn, R.I., Hermanson, G.T., Mallia, A.K., Gartner, F.H., Provenzano, M.D., Fujimoto, E.K., Goeke, N.M., Olson, B.J. & Klenk, D.C. (1985) Measurement of protein using bicinchoninc acid. Anal. Biochem. 150, 76–85.

18. Laemmli, UK. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680– 685. This research was supported by grant (S-2002–11) from the Direccio´ n de Investigacio´ n y Desarrollo, Universidad Austral de Chile. We are also grateful for grants MECESUP AUS 0006 and AUS 0005 that supported the research visits of M.I.C. to the Gatty Marine Laboratory University of St. Andrews, Scotland, UK and of V.J.S. to the Institute of Biochemistry, Faculty of Sciences, Universidad Austral de Chile, Valdivia, Chile.

2990 M. I. Concha et al. (Eur. J. Biochem. 271)

(cid:1) FEBS 2004

highly effective against infections with antibiotic-resistant bacteria. Infect. Immun. 69, 1469–1476.

27. Singh, P.K., Tack, B.F., McCray, P.B. & Welsh, M.J. (2000) Synergistic and additive killing by antimicrobial factors found in human airway surface liquid. Am. J. Physiol. Lung Cell. Mol. Physiol. 279, L799–L805. 19. Friedrich, C., Scott, M.G., Karunaratne, N., Yan, H. & Hancock, R.E.W. (1999) Salt resistant a-helical cationic antimi- crobial peptides. Antimicrob. Agents. Chemother. 43, 1542–1548. 20. Lehrer, R.I., Rosenman, M., Harwig, S.S., Jackson, R. & Eisenhauer, P. (1991) Ultrasensitive assays for endogenous poly- peptides. J. Immunol. Methods 137, 167–173.

21. Smith, V.J., Fernandes, J.M., Jones, S.J., Kemp, G.D. & Tatner, M.F. (2000) Antibacterial proteins in rainbow trout, Oncor- hynchus mykiss. Fish Shellfish Immunol. 10, 243–260. 28. Patrzykat, A., Zhang, L., Mendoza, V., Iwama, G.K. & Hancock, R.E.W. (2001) Synergy of histone-derived peptides of coho salmon with lysozyme and flounder pleurocidin. Antimicrob. Agents Chemother. 45, 1337–1342.

29. Risso, A. (2000) Leukocyte antimicrobial peptides: multifunc- tional effector molecules of innate immunity. J. Leukoc. Biol. 68, 785–792. 22. Scha¨ gger, H. & von Jagow, G. (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal. Biochem. 166, 368–397.

30. Motizuki, M., Satoh, T., Takei, T., Itoh, T., Yokota, S., Kojima, S., Miura, K., Samejima, T. & Tsurugi, K. (2002) Structure- activity analysis of an antimicrobial peptide derived from bovine apolipoprotein A-II. J. Biochem. 132, 115–119. 23. Lee, M., Kovanen, P.T., Tedeschi, G., Oungre, E., Franceschini, G. & Calabresi, I. (2003) Apolipoprotein composition and particle size affect HDL degradation by chymase: effect on cellular cho- lesterol efflux. J. Lipid Res. 44, 539–546.

24. Eberini, I., Calabresi, L., Wait, R., Tedeschi, G., Pirillo, A., Puglisi, L., Sirtori, C.R. & Gianazza, E. (2002) Macrophage metalloproteinases degrade high-density-lipoprotein-associated apolipoprotein A-I at both N- and C-termini. Biochem. J. 362, 627–634. 31. McManus, D.C., Scott, B.R., Frank, P.G., Franklin, V., Schultz, J.R. & Marcel, Y.L. (2000) Distinct central amphipathic alpha-helices in apolipoprotein A-I contribute to the in vivo maturation of high density lipoprotein by either activating lecithin-cholesterol acyltransferase or binding lipids. J. Biol. Chem. 275, 5043–5051.

32. Yano, T. (1996) Non-specific immune system: humoral defense. In The Fish Immune System (Iwama G., Nakanishi T., eds), pp. 106– 159. Academic Press, San Diego, CA, USA.

25. Sørensen, O.E., Follin, P., Johnsen, A.H., Calafat, J., Tjabringa, G.S., Hiemstra, P.S. & Borregard, N. (2001) Human cathelicidin, hCAP-18, is processed to the antimicrobial peptide LL-37 by extracellular cleavage with proteinase 3. Blood 97, 3951–3959. 26. Nibbering, P.H., Ravensbergen, E., Welling, M.M., van Berkel, L.A., van Berkel, P.H.C., Pauwels, E.K.J. & Nuijens, J.H. (2001) Human lactoferrin and peptides derived from its N terminus are 33. Fernandes, J.M.O., Kemp, G.D. & Smith, V.J. (2004) Two novel muramidases from skin mucus of rainbow trout. Comp. Biochem. Physiol. B. Biochem. Mol. Biol. 138, 53–64.