Cellular Immunology 261 (2010) 128–133

Contents lists available at ScienceDirect

Cellular Immunology

journal homepage: www.elsevier .com/ locate/yc imm

Expression and functional analyses of liver expressed antimicrobial peptide-2(LEAP-2) variant forms in human tissues

Alison Howard a,1, Claire Townes a,*,1, Panagiota Milona a, Christopher J. Nile a,Giorgios Michailidis b, Judith Hall a

a Institute for Cell and Molecular Biosciences, University of Newcastle upon Tyne, NE2 4HH, UKb Department of Animal Production, School of Agriculture, Aristotle University of Thessaloniki, Greece

a r t i c l e i n f o

Article history:Received 3 July 2009Accepted 25 November 2009Available online 3 December 2009

Keywords:LEAP-2THP-1 monocytesIron responseEpitheliaInnate immunity

0008-8749/$ - see front matter � 2009 Elsevier Inc. Adoi:10.1016/j.cellimm.2009.11.010

* Corresponding author.E-mail address: c.l.townes@ncl.ac.uk (C. Townes).

1 These authors contributed equally to this work.

a b s t r a c t

The antimicrobial peptide Liver Expressed Antimicrobial Peptide-2 (LEAP-2) is proposed to function aspart of the vertebrate innate immune system. However, the highly conserved nature of the LEAP-2 pep-tide primary structure among vertebrates suggests more fundamental physiological roles. RT-PCR analy-ses confirmed expression of LEAP-2 mRNA variants in human gastro-intestinal (GI) epithelial tissues andTHP-1 monocytes. Three cDNA products indicative of at least three different spliced transcripts wereobserved. Translation of the cDNA sequences supported synthesis of transcripts encoding the secretedLEAP-2 peptide and two variants lacking signal sequences suggesting intracellular localisation. The syn-thesis and cytoplasmic localisation of LEAP-2 peptides in epithelia was supported by immunohistochem-ical analyses. Functional data suggested that LEAP-2 is not involved in the physiological response of GIepithelia to iron, nor is it mitogenic for epithelial cells or chemotactic for THP-1 monocytes. However,changes in the LEAP-2 transcript patterns associated with the challenge of THP-1 monocytes with lipo-polysaccharide (100 ng/ml) were supportive of the peptides having multiple roles in the innate immuneresponse.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

Epithelia comprise cell layers that, in addition to essentialabsorptive and secretory functions, act as physical and biologicalbarriers that protect the host against microbial assault. As part oftheir defence mechanisms epithelial cells synthesise and secrete avariety of small (<10 kDa) cationic antimicrobial peptides (CAMPs),including the well characterized defensins and cathelicidins, whichnot only kill potential pathogens through disruption of their micro-bial membranes, but also promote bacterial clearance through theenhancement, inhibition or complementation of cellular functionssuch as chemotaxis, apoptosis, gene transcription and cytokine pro-duction [1].

While the information relating to the activities of CAMPs in thepost-genomic era has increased dramatically, that related to LiverExpressed Antimicrobial Peptide-2 (LEAP-2), the sole representa-tive of a unique four cysteine (Cys) peptide family, remains lesswell reported. Since its discovery in 2003, LEAP-2, a secreted4 kDa peptide characterized by two intracellular di-sulphidebridges, has been acknowledged as a CAMP functioning as part of

ll rights reserved.

the vertebrate innate immune system [2,3]. Despite its nameLEAP-2, encoded by a gene consisting of three exons and two in-trons, is actually expressed by a plethora of vertebrate epithelialtissues [2,3].

Genomic studies have predicted LEAP-2 peptides to be synthes-ised by a number of species including the mouse, monkey, cow andchicken [2,3]. In addition all the mammalian peptide primarystructures show almost 100% amino acid (aa) homology. Betweenspecies conservation of classical CAMPs is atypical as the primarysequences of traditionally recognised CAMP molecules reflect gen-erally the adaptation of the host species to its specific endemic mi-crobes. This is illustrated particularly well by the defensins whichin vertebrates have evolved in response to microbial challengesthrough gene duplication, exon shuffling and the positive selectionof amino acid mutations [4,5].

This conservation of amino acid sequence suggests that LEAP-2has, in addition to its antimicrobial activities (AMA), other func-tions that are of fundamental physiological importance to mam-mals. The observation that physiologically important moleculespossess AMA is not unique. For example hepcidin, formerly knownas LEAP-1, was identified originally as a CAMP but actually plays akey role in mammalian iron homeostasis [6,7]. Similarly the angio-genins, discovered nearly two decades ago, were characterized ini-tially as having roles in vasculogenesis and tumour cell growth, but

A. Howard et al. / Cellular Immunology 261 (2010) 128–133 129

have been shown to exhibit bacteriocidal and fungicidal activity,suggesting an important role in systemic innate immunity [8].

This study aimed to confirm the presence of multiple LEAP-2transcripts in epithelial and immune cells and to investigatewhether the functions of LEAP-2 extend beyond microbial killing.To explore potential physiological and immunological propertiesof the LEAP-2 peptides, tissues and in vitro cell lines, modellinggastro-intestinal and liver epithelia, and cell monocytes wereemployed.

2. Materials and methods

2.1. Cell lines

The culture and seeding of Caco-2 cells on Transwell (Costar) fil-ter inserts were as previously described [9]. HepG2 cells were cul-tured in six well plastic plates in 5% CO2 at 37 �C in growth mediumconsisting of Eagle’s Minimal Essential Medium containing 1%200 mM glutamine, 100 U/ml penicillin, 100 U/ml streptomycinsolution (Invitrogen) and 10% foetal calf serum.

For the iron (Fe2+) challenge experiments the growth media ofthe Caco-2 cells, cultured as polarised monolayers, and the HepG2cells, cultured as monolayers, was replaced with medium contain-ing 100 lM Fe SO4 for time periods of up to four hours. After theappropriate incubation RNA was extracted from the cells forsemi-quantitative analyses of Hepcidin (LEAP-1) and LEAP-2expression.

THP-1 promonocytic cells (a kind gift of Dr. S. Ali, Newcastle Uni-versity) were cultured in plastic flasks (Corning, UK) at 5% CO2 and37 �C in RPM1-1640 medium (Sigma, Poole, UK) supplemented with1% 200 mM glutamine, 100 U/ml penicillin, 100 U/ml streptomycinsolution (Invitrogen) and 10% foetal calf serum (Sigma). THP-1pro-monocytes were converted to THP-1 monocytes by incubationwith 100 nM 1a,25-dihydroxy-vitamin D3 (Merck Chemicals, Not-tingham, UK) for 48 h. Further differentiation to THP-1 macrophageswas performed by subsequent incubation with 50 ng/ml phorbolmyristate acetate (PMA) for 48 h. Dendritic cells were derived fromprimary monocytes as follows. The primary monocytes were ob-tained from buffy coat fractions of human blood (national Blood Ser-vice, Newcastle, UK). The peripheral blood mononuclear cellpopulation (PBMC) was extracted from the buffy coat fraction usingHistopaque-1077 (Sigma). PBMCs were cultured in complete RPM1-1640 medium for 24 h at 37 �C with 5% CO2. After 24 h the non-adherent cell population was removed by repeated washes inRPM1-1640 medium to reveal adherent primary monocytes. Pri-mary monocytes were converted to dendritic cells by incubationin RPMI medium supplemented with 10 ng/ml granulocyte–macro-phage stimulating factor (GMCSF) and 50 ng/ml interleukin-4 for7 days, changing the medium every other day. Differentiation wasconfirmed by morphological changes and analysis of CD1a, CD11C,CD14, HLA-DR and CD83 by FACS.

2.2. Reverse transcription polymerase chain reaction (RT-PCR)

Total RNA was isolated from the cell lines using the SV TotalRNA Isolation System (Promega) and subjected to reverse tran-scription (RT) using random hexamer primers. To amplify LEAP-2, RT-PCR was performed as previously described using a primerpair designed to exon 1: F-50CAAGATGTGGCACCTCAAAC30 andexon 3: R-50GCATTGTCGGAGGTGACTG30 and which detected allknown variants of LEAP-2 [10]. For each RNA sample a RT negativecontrol was generated in which reverse transcriptase was omitted.PCR conditions included 15 min at 95 �C, 35 cycles of 95 �C for 30 s,58 �C for 30 s and 72 �C for 1 min, followed by a final extensionstep of 10 min at 72 �C. The cDNA products were separated on 1%

agarose ethidium bromide gels and visualised using UV light. AllcDNA products were verified by sequencing. For relative quantifica-tion of template and sample RNA concentrations all reactions weremeasured in the linear exponential phase, 15 cycles for the 18S ribo-somal RNA primer pair and 30 cycles for each antimicrobial primerpair. Product optical density was determined on an AlphaImager1200 gel documentation and analysis system (Flowgen). The Hepci-din and 18S primers were as previously described [10].

2.3. Western analyses

HCT-8 cells were lysed in 0.1% Triton X-100. Samples were sep-arated on an SDS-12.5% polyacrylamide gel and transferred onto apolyvinylidene difluoride membrane (Hybond P; Amersham). Themembrane was incubated overnight at 4 �C in a blocking solutionof phosphate-buffered saline (PBS), 0.1% (vol/vol) Tween 20, and5% (wt/vol) milk powder containing a 1:5000 dilution of anti-LEAP-2 antibody (Phoenix Pharmaceuticals, Germany). The mem-brane was washed in PBS-Tween 20 (0.1%) and incubated with asecondary anti-rabbit IgG antibody at room temperature for a fur-ther 2 h. Following further washing, as above, a chemilumines-cence detection reaction was performed by using ECL Plus(Amersham), and the membrane was exposed to X-ray film accord-ing to the manufacturer’s recommendations.

2.4. Immunofluorescent staining

7 lm frozen sections of the respective tissue were fixed in 2%paraformaldehyde, permeabilised using 0.1% Triton X-100 in PBS,and incubated in 10% goat serum to block any non-specific proteinbinding. Sections were incubated overnight at 4 �C with 1:500 dilu-tion of anti-LEAP-2 antibody (Phoenix Pharmaceuticals). Afterovernight incubation, any unbound primary antibody was re-moved by washing in (phosphate-buffered saline) PBS. For eachstaining protocol negative controls in which the primary antibodywas omitted were performed. The secondary antibody goat anti-rabbit TRITC, (Chemicon), diluted 1:50 in PBS, was incubated for1 h at room temperature and any unbound material removed bywashing. The slides were air-dried for 5 min, mounted with Vecta-shield (Vector Laboratories, Burlingame, USA), and imaged on a Lei-ca TCS LASER scanning confocal microscope.

2.5. Chemotaxis and cell proliferation

The chemotactic potential of LEAP-2 was assessed using themethod of Ali et al. [11]. Essentially synthetic mature LEAP-2 pep-tide (Phoenix Parmaceuticals) was diluted to the appropriate con-centrations (1–500nM) in 1 ml medium, each dilution added to thelower compartments of a 24 microchemotaxis chamber and 3 lmpore membrane inserts added (Corning). After a 30 min incubationat 37 �C, the upper compartment of the filter was loaded with500 ll of THP-1 cells (1 � 106) stimulated overnight with IFNc(300 U/ml). After 90 min at 37 �C the filters were removed andplaced in a 24 well plate containing methanol. After 2 h at�20 �C the methanol was decanted off, the filters washed with dis-tilled water, 0.5 ml of Mayer’s haemotoxylin added to each filterand left for 30 min at room temperature. After removal of the stain,the filters were washed with 1� SCOTTS water and incubated atroom temperature for 5 min before washing with tap water andthen air-drying. Once dry the filters were mounted onto glassslides in DPX and visualised under high power field (HPF). Chemo-taxis was assessed by counting the mean number of migrant cellsper HPF. Negative controls contained only medium in the lowerchamber or THP-1 cells in both chambers. The positive controlused 10 ng/ml FMLP (Formyl Met-Leu-Phe) an activator ofchemotaxis.

Fig. 1A. Expression of LEAP-2 variants in human gastro-intestinal tissues. Samplesfrom a human digestive tissue cDNA panel (Clontech/BD biosciences) wereamplified by PCR with a primer pair designed to detect all known variants ofLEAP-2. The two major cDNA products correspond to the fully spliced (�0.3 kb) andintron 1 (�0.5 kb) retaining variants. Although faint 0.8 kb cDNA bands indicative ofintron 1 and intron 2 retaining LEAP-2 variant are also visible. L – Liver; Ca –Caecum; R – Rectum; IC – Ileo-caecum; DC – Descending colon; I – Ileum; TC –Transverse colon; S – Stomach; AC – Ascending colon; O – Oesophagus.

Fig. 1B. Expression of LEAP-2 cDNA variants in human GI mucosa and immortalisedhuman cell lines. cDNA samples from human intestinal cell lines HCT-8 (H) andCaco-2 (C) and the human liver cell line HepG2 (He) were amplified by PCR with aprimer pair designed to detect all known variants of LEAP-2. In Caco-2 and HepG2cells the major product of �300 bp was confirmed as the fully spliced LEAP-2mRNA. In HCT-8 cells the major product of �500 bp corresponds to the intron 1retaining variant of LEAP-2; the 800 bp product corresponds to an intron 1 andintron 2 containing variant. Caco-2 and HCT-8 samples are shown alongsideamplification products from human duodenal (D) and ileal (I1 and I2) mucosalcDNAs. The positions of the 0.5 and 1kbp markers in a DNA ladder (M) are indicated.ntc – no-template control.

130 A. Howard et al. / Cellular Immunology 261 (2010) 128–133

For the cell proliferation studies HepG2 cells were seeded into 96well plates at a density of 1 � 103/well and allowed to recover for24 h at 37 �C and 5%. The medium was replaced with either freshmedium (control) or fresh medium containing the test concentra-tions of synthetic LEAP-2 (1 nM–100 lM). After incubation periodsof 48, 72 and 96 h respectively cell proliferation was assessed usinga SRB assay. Essentially the cells were fixed for 60 min at 4 �C using50% TCA, washed with tap water, 100 ll of working SRB (Sulforhoda-mine B) solution [100 ll, 5% SRB; 9.9 ml, 1% aceticacid] was added toeach well and left at room temperature for 15 min. The plates wereair-dried, the stain dissolved in 200 ll of 10 mM Tris–HCl and theabsorbance of the solutions read at 490 nm.

3. Results and discussion

3.1. LEAP-2 variant forms and expression patterns

Typical of many antimicrobial peptides synthesised by the epi-thelia, mature LEAP-2 is cleaved from a fully-spliced transcript as a

Fig. 1C. Peptides encoded by LEAP-2 transcripts. Unique N and C termini are highlighte3Cys intra – 8 kDa (http://ca.expasy.org/tools/peptide-mass.html).

77 aa precursor molecule, containing a signal sequence and a 53 aapro-peptide. Purification of a 40 aa LEAP-2 molecule from blood [2]indicates that the secreted LEAP-2 pro-peptide molecule is cleaved,probably by furin [2,3]. However, analysis of the LEAP-2 transcrip-tome indicates that LEAP-2 expression involves more than one tran-script. Previous studies have shown, by RT-PCR analyses, that anLEAP-2 transcript encoding a fully-spliced 77 aa peptide is the pri-mary PCR product present in RNA extracted from human small intes-tine (SI), liver, kidney and bladder. Moreover, an intron 1 retainingtranscript (Int1) was also detected in the liver, kidney, heart and lungand dominated in tissues such as the colon, lung and testis [2].

To investigate the reproducibility of this observation theexpression of LEAP-2 mRNA variants in human gastro-intestinalepithelial tissues was investigated using a primer pair designedto detect all known variants of LEAP-2 and a human digestive sys-tem MTC™ panel (Clontech/BD Biosciences). This panel containednormalised first strand cDNA amplified from colonic, duodenum,ileocaecum, ileum, jejunum, rectum, caecum, stomach, liver andoesophagus RNA purified from the disease free whole tissues ofmale and female donors aged 17–75 years respectively. With theexception of the liver, where only one cDNA product representingthe fully spliced variant was detected, all tissues generated twomajor products, corresponding to the fully spliced (�0.3 kb cDNAproduct) and intron 1 retaining (�0.5 kb cDNA product) mRNAvariants. Faint cDNA products indicative a 0.8 kb transcript werealso visible (Fig. 1A). In contrast to whole tissue sections molecularanalyses of human duodenal and ileal mucosal samples detectedonly the 0.3 kb cDNA band indicative of the secreted matureLEAP-2 peptide, a pattern mirrored in the in vitro cell lines usedto model such tissues (Fig. 1B).

Translation of the 0.3, 0.5 and 0.8 kb cDNA sequences predictsthe synthesis of three putative LEAP-2 peptides. Two peptidesassociated with the 0.3 and 0.5 kb cDNA products are translatedfrom ATG start codons identified in exon 1 and intron 1 respec-tively, and are each predicted to contain four cysteines (4Cys). Theydiffer in that the peptide encoded by the 0.5 kb cDNA band is lack-ing a signal sequence and is predicted to be intracellularly located(Fig. 1C). A third putative peptide encoded by the 0.8 kb cDNAproduct is also translated from the intron 1 ATG start codon andis not predicted to contain a secretion sequence, but contains se-quence originating from intron 2 and is predicted to contain only3Cys. The synthesis of two peptides of a size consistent with the re-sults of the PCR analyses were confirmed by western blot analysesof the human intestinal cell line HCT-8 using an anti-LEAP-2 anti-body (Phoenix Pharmaceuticals, Germany), (Fig. 1D).

These data therefore indicate that in the different gastro-intes-tinal tissues the LEAP-2 gene undergoes alternative splicing to pro-duce an array of distinct but related peptides. Two of thesepeptides are predicted to be intracellular and one secreted. Theexistence of multiple gene transcripts within cells is not unique;three types of mRNA encoding either mitochondrial, non-mito-chondrial or nucleolar localised phospholipid hydroperoxide gluta-thione peroxidises, a family of Cys rich anti-oxidant enzymes, alloriginate from one gene [12]. However why transcripts encodingan intracellular LEAP-2 peptide should dominate in the gastro-intestinal tract whole tissues compared to the mucosa and liveris not known. From our own studies [3] and those of others [13],LEAP-2 expression is induced in gastro-intestinal tissues in re-sponse to microbial infection. Infection is commonly associated

d. Predicted sizes of the mature peptides are, secreted – 4 kDa; 4Cys intra – 7 kDa;

Fig. 1D. Synthesis of LEAP-2 peptides. Western analyses of cell lysates of thehuman colon cell line HCT-8 (A) and THP-1 monocytes (B) using anti-LEAP-2 or (C)anti-LEAP-2 pre-absorbed by incubation with synthetic LEAP-2.

A. Howard et al. / Cellular Immunology 261 (2010) 128–133 131

with cell stress and apoptosis, and key intracellular proteins func-tioning to defend against stress are cysteine rich [14], thus it is fea-sible that the intracellularly located cysteine containing LEAP-2molecules may function in protecting the gut epithelium frommicrobial induced damage.

3.2. LEAP-2, immunohistochemical staining patterns

To further investigate the synthesis and localisation of LEAP-2peptides in epithelial tissues frozen sections of human colon and

Fig. 2. Immunohistochemical analyses of LEAP-2 in (A) human colon and (B) human kidn2. (A) Low power (20�magnification) of a section of human colon showing LEAP-2 localis(63� magnification) of human kidney section showing LEAP-2 in cells lining the proxim

Fig. 3. Fe2+ regulation of Hepcidin (LEAP-1) & LEAP-2 mRNA in HepG2 cells. Expressicontaining 100 lM FeSO4 for 1 (1H), 2 (2H) or 4 (4H) hours was measured by semi-quantin standard medium. Results, shown as per cent of control, are mean ± SEM of three sam

kidney were stained using an anti-LEAP-2 polyclonal antibody.Analyses of the gastro-intestinal tissue (Fig. 2A) indicated thatimmunoreactive LEAP-2 localised specifically to epithelial cells lin-ing the colonic crypts, while analyses of human kidney localisedLEAP-2 staining specifically to the epithelial cells lining the proxi-mal and distal tubules (Fig. 2B). In addition the LEAP-2 stainingpatterns were indicative of whole cell staining supporting the syn-thesis and cytoplasmic localisation of LEAP-2 peptides. In the colonthese data endorsed the molecular LEAP-2 panel expression pro-files, which predicted the major peptide to lack a signal sequenceand to be located intracellularly.

3.3. LEAP-2 and iron

Hepcidin previously known as LEAP-1 is a cysteine rich mole-cule purified from blood that has been shown to have AMA [15].Like LEAP-2, the hepcidin primary sequence is conserved acrossmammalian species and transcripts are found in numerous epithe-lial tissues including the liver and intestine [16,17]. As hepcidin isnow recognised as a peptide hormone with a key physiological rolein iron homeostasis [7,18] we investigated whether LEAP-2 is alsoinvolved in iron metabolism. To achieve this we used in vitro cellmodels including polarised Caco-2 cells, an enterocyte model,

ey. Frozen sections of human colon and kidney were stained as described in Sectioned to cells lining the colonic crypts. (c – crypt lumen, L – gut lumen). (B) High poweral (p) and distal (d) tubules.

on of hepcidin (LEAP-1) and LEAP-2 mRNAs in HepG2 cells cultured in mediumitative RT-PCR, corrected to 18S rRNA expression. Control cells (C) were maintainedples and are in arbitrary units.

Fig. 4. Chemotaxis of THP-1 monocytic cells incubated with synthetic matureLEAP-2 (0–500 nM). Chemotactic control (10) was FMLP (10 ng/ml). Results aremean ± SEM (n = 9).

Fig. 5. Expression of LEAP-2 in immune presenting cells. (A) cDNA samples from pro-monPCR with primer pairs designed to detect all known variants of LEAP-2 (a) 18S (b) and noare indicated. (B) THP-1 monocytes challenged with lipopolysaccharide (100 ng/ml). cDNprimer pair designed to detect all known variants of LEAP-2. Top panel shows LEAP-2 ex(100 ng/ml) treatment. Bottom panel shows 18S control. The positions of the 300, 500 apeptide profiles. Western analyses of THP-1 monocyte cell lysates using LEAP-2 antibodyFollowing probing the filters were stripped with Restore™ western blot stripping buffer (dilution of 1:2000 dilution.

132 A. Howard et al. / Cellular Immunology 261 (2010) 128–133

and HepG2 cells to model liver hepatocytes, (Fig. 1B), and each cellline was incubated with 100 lM Fe2+ for up to four hours. Analysesof the DNAase treated RNA extracted from the cells for hepcidinand LEAP-2 expression indicated up-regulation of hepcidin geneexpression within one hour of treatment, but no up-regulation ofthe LEAP-2 transcripts was detected (Fig. 3). These data indicatedthat LEAP-2 is not involved in the physiological response to iron.

3.4. LEAP-2 and cell proliferation

Hepatocytes are a major source of the secreted LEAP-2 mole-cules and so we investigated, using the HepG2 cells, whether thepeptide was functioning in an autocrine or paracrine fashion as agrowth factor. Addition of synthetic mature LEAP-2 to the cellgrowth medium did not increase hepatocyte cell proliferation atany concentration of LEAP-2 employed. This indicated that themature peptide is not mitogenic for epithelial cells (data notshown).

ocytes (1) monocytes (2) Macrophages (3) and Dendritic cells (4) were amplified byRT control (c). The positions of the 300, 500 and 800 bp markers in a DNA ladder (M)A prepared from THP-1 monocytes challenged with LPS was amplified by PCR with apression at 0.5 h (A & B); 3 h (C & D); 6 h (E & F); 9 h (G & H) 24 h (I & J) after LPSnd 800 bp markers in a DNA ladder (M) are indicated. (C) Immunoblots of LEAP-2before and after challenge with lipopolysaccharide. A: time 0 h, B: 3 h, C: 6 h, D: 9 h.Thermoscientific) for 15 min and probed with anti-a-Tubulin antibody (Abcam) at a

A. Howard et al. / Cellular Immunology 261 (2010) 128–133 133

3.5. LEAP-2 and chemotactic activity

As mentioned previously the fully-spliced LEAP-2 gene tran-scripts are up-regulated in vertebrate tissues in response to micro-bial challenge [3,19,13]. As other CAMPs have been shown, inaddition to their AMAs, to have chemotactic properties [20] weinvestigated whether the secreted mature LEAP-2 peptide mayplay a role in microbial clearance through its ability to attractmonocytes to a site of microbial infection. Increasing concentra-tions of synthetic LEAP-2, analogous to the secreted LEAP-2 pep-tide, did not induce the chemotaxis of THP-1 monocytes (Fig. 4),suggesting that mature LEAP-2 does not function directly to linkthe innate and adaptive immune systems.

Analyses of THP-1 monocytes indicated three cDNA bands (0.8,0.5 and 0.3 kbp) associated with three LEAP-2 immunoreactivepeptides. Interestingly no LEAP-2 transcripts were detected whenmonocytic cells were differentiated in vitro to either macrophagesor dendritic cells (Fig. 5A). Challenging THP-1 monocytic cells with100 ng/ml lipopolysaccharide (LPS) resulted in the loss of the800 bp and 300 bp cDNA products concomitant with changes inthe peptide profile of the cells (Fig. 5B). Although the significanceof this observation is not known it is possible that the LEAP-2 pep-tides function in the maturation and differentiation of monocyticcells, an interesting observation given the proposed developmentalrole of LEAP-2 in the catfish [19]. The retention of the 0.5 kb cDNAproduct encoding the 4Cys intracellular LEAP-2 peptide, compara-ble to that in the GI whole tissues, may indicate a role for this pep-tide in protecting the cell from injury and/or stress.

LEAP-2 gene expression in epithelial cells was shown to beassociated with multiple transcripts encoding at least three differ-ent LEAP-2 peptides of which two are predicted to be intracellu-lar. The roles of these peptides remain to be established althoughour data suggests that functions are not associated with theeither the physiological response to iron, epithelial cell prolifera-tion or cell chemotaxis although the chemotaxis of primary anti-gen presenting cells e.g. leukocytes cannot be excluded. Moreoverroles in cell differentiation and/or cell protection cannot beeliminated.

Acknowledgments

This work was funded by a grant to GM from the European Un-ion and Greek Government. PM was supported by a Greek StateFoundation Scholarship.

References

[1] K.L. Brown, R.E. Hancock, Cationic host defense (antimicrobial) peptides,Current Opinions in Immunology 18 (1) (2006) 24–30.

[2] A. Krause, R. Sillard, B. Kleemeier, E. Kluver, E. Maronde, J.R. Conejo-Garcia, W.G. Forssmann, P. Schulz-Knappe, M.C. Nehls, F. Wattler, S.Wattler, K. Adermann, Isolation and biochemical characterization of LEAP-2, a novel blood peptide expressed in the liver, Protein Science 12 (2003)

143–152.[3] C.L. Townes, G. Michailidis, C.J. Nile, J. Hall, Induction of cationic chicken liver-

expressed antimicrobial peptide 2 in response to Salmonella enterica infection,Infection and Immunity 72 (2004) 6987–6993.

[4] K. Luenser, A. Ludwig, Variability and evolution of bovine beta-defensin genes,Genes and Immunity 6 (2005) 115–122.

[5] C.A. Semple, P. Gautier, K. Taylor, J.R. Dorin, The changing of the guard:molecular diversity and rapid evolution of beta-defensins, Molecular Diversity10 (2006) 575–584.

[6] E. Nemeth, M.S. Tuttle, J. Powelson, M.B. Vaughn, A. Donovan, D.M. Ward, T.Ganz, J. Kaplan, Hepcidin regulates cellular iron efflux by binding to ferroportinand inducing its internalization, Science (New York, NY) 306 (2004) 2090–2093.

[7] G. Nicolas, L. Viatte, M. Bennoun, C. Beaumont, A. Kahn, S. Vaulont, Hepcidin, anew iron regulatory peptide, Blood Cells Molecules and Diseases 29 (2002)327–335.

[8] L.V. Hooper, T.S. Stappenbeck, C.V. Hong, J.I. Gordon, Angiogenins: a new classof microbicidal proteins involved in innate immunity, Nature Immunology 4(2003) 269–273.

[9] K.L. Soole, M.A. Jepson, G.P. Hazlewood, H.J. Gilbert, B.H. Hirst, Epithelialsorting of a glycosylphosphatidylinositol-anchored bacterial protein expressedin polarized renal MDCK and intestinal Caco-2 cells, Journal of Cell Science 108(Pt. 1) (1995) 369–377.

[10] S.L. Ball, G.P. Siou, J.A. Wilson, A. Howard, B.H. Hirst, J. Hall, Expression andimmunolocalisation of antimicrobial peptides within human palatine tonsils,The Journal of Laryngology and Otology 121 (2007) 973–978.

[11] S. Ali, H. Robertson, J.H. Wain, J.D. Isaacs, G. Malik, J.A. Kirby, A non-glycosaminoglycan-binding variant of CC chemokine ligand 7 (monocytechemoattractant protein-3) antagonizes chemokine-mediated inflammation,Journal of Immunology 175 (2005) 1257–1266.

[12] H. Imai, M. Saito, N. Kirai, J. Hasegawa, K. Konishi, H. Hattori, M. Nishimura, S.Naito, Y. Nakagawa, Identification of the positive regulatory and distinct coreregions of promoters, and transcriptional regulation in three types of mousephospholipid hydroperoxide glutathione peroxidase, Journal of Biochemistry140 (2006) 573–590.

[13] Y. Sang, B. Ramanathan, J.E. Minton, C.R. Ross, F. Blecha, Porcine liver-expressed antimicrobial peptides. hepcidin and LEAP-2: cloning and inductionby bacterial infection, Developmental and Comparative Immunology 30 (2006)357–366.

[14] C. Berndt, C.H. Lillig, A. Holmgren, Thiol-based mechanisms of thethioredoxin and glutaredoxin systems: implications for diseases in thecardiovascular system, American Journal of Physiology 292 (2007) H1227–H123615.

[15] A. Krause, S. Neitz, H.J. Magert, A. Schulz, W.G. Forssmann, P. Schulz-Knappe, K.Adermann, LEAP-1, a novel highly disulfide-bonded human peptide, exhibitsantimicrobial activity, FEBS Letters 480 (2000) 147–150.

[16] V. Atanasiu, B. Manolescu, I. Stoian, Hepcidin – central regulator of ironmetabolism, European Journal of Haematology 78 (2007) 1–10.

[17] G. Nicolas, C. Chauvet, L. Viatte, J.L. Danan, X. Bigard, I. Devaux, C. Beaumont, A.Kahn, S. Vaulont, The gene encoding the iron regulatory peptide hepcidin isregulated by anemia, hypoxia, and inflammation, The Journal of ClinicalInvestigation 110 (2002) 1037–1044.

[18] G. Nicolas, M. Bennoun, I. Devaux, C. Beaumont, B. Grandchamp, A. Kahn, S.Vaulont, Lack of hepcidin gene expression and severe tissue iron overload inupstream stimulatory factor 2 (USF2) knockout mice, Proceedings of theNational Academy of Sciences of the United States of America 98 (2001) 8780–8785.

[19] B. Bao, E. Peatman, P. Xu, P. Li, H. Zeng, C. He, Z. Liu, The catfish liver-expressed antimicrobial peptide 2 (LEAP-2) gene is expressed in a widerange of tissues and developmentally regulated, Molecular Immunology 43(2006) 367–377.

[20] D. Yang, O. Chertov, S.N. Bykovskaia, Q. Chen, M.J. Buffo, J. Shogan, M.Anderson, J.M. Schroder, J.M. Wang, O.M. Howard, J.J. Oppenheim, Beta-defensins: linking innate and adaptive immunity through dendritic and T cellCCR6, Science (New York, NY) 286 (1999) 525–528.