Dipartimento di
* Scienze e Tecnologie Biomediche and
Ricerche Mediche e Morfologiche, Università di Udine, Italy; and
Laboratorio Nazionale CIB, AREA Science Park, Trieste, Italy
Correspondence: Margherita Zanetti, Dept. Biomedical Sciences and Technology, University of Udine, P.le Kolbe 4, I-33100 Udine, Italy. E-mail: zanetti{at}icgeb.trieste.it
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: Bac5 • neutrophil • gene expression
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A variety of mammalian antimicrobial peptides belong to the cathelicidin protein family [8
9
10
11
]. Bac5 was the first reported member of this family [8
, 12
]. Its sequence was deduced from bovine bone marrow (BM) cells mRNAs [12
], and early studies of this molecule contributed substantially to elucidate the biosynthesis and maturation process of cathelicidins in myeloid-derived cells [13
, 14
]. Bac5 is actively synthesized during myelopoiesis but not in mature resting neutrophils [13
] and is stored in the cytoplasmic neutrophil granules as a propeptide (proBac5), in which the C-terminal antimicrobial domain is joined to a conserved, cathelin-like propiece [13
]. The antimicrobial domain displays a proline-rich cationic sequence of 43 amino acid residues and is liberated from the propeptide coincident with neutrophil degranulation [14
, 15
]. It is remarkable that whereas the propiece is fairly conserved in all cathelicidin members, the peptide regions are highly heterogeneous and in addition to proline-rich peptides, may include 
-helical, tryptophan-rich, and disulfide-bridged peptides displaying distinct spectra of antimicrobial activity [8
9
10
11
].
All known cathelicidins were originally identified in myeloid-derived cells. However, expression studies of the human LL-37 [16 ] have indicated that this gene is also expressed in mucosal epithelia and in other nonmyeloid cells [5 , 17 18 19 ]. In addition, the human LL-37 [16 ] and the murine cathelin-related antimicrobial peptide (CRAMP) [20 ] cathelicidin genes are up-regulated in skin in response to inflammatory and infectious stimuli [6 , 21 ]. These studies thus suggest a more widespread pattern of cathelicidin expression than previously recognized.
To gain a better insight into the regulated expression of the Bac5 gene, we have extended here the analysis of the expression of this gene to normal and inflamed nonmyeloid tissues that are known to be sites of antimicrobial peptide synthesis. We find that the expression of the Bac5 gene is induced in neutrophils accumulated at inflammation sites in infected lung tissue, and we analyze the expression of this gene in primary cultures of peripheral neutrophils in response to bacteria and lipopolysaccharide (LPS), providing evidence for de novo synthesis of Bac5 in mature neutrophils.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell isolation and stimulation
BM cells were isolated from bovine ribs as described [13
]. Polymorphonuclear leukocytes (PMN) and mononuclear cells (MNC) were purified from whole blood of freshly killed cows. After differential centrifugation, PMN were isolated by hypotonic lysis of the supernatant, whereas the MNC fraction was obtained from the interface buffy coat after lysis with isotonic NH4Cl of the contaminating erythrocytes [13
]. As assessed by Wright staining, PMN preparations contained >97% granulocytes (92% neutrophils and 5% other granulocytes, mainly eosinophils) and <2% MNC. Promyelocytes and myelocytes were absent in all the neutrophil preparations used, and metamyelocytes and band cells ranged from 0 to 1.4% of the total cells. Less than 5% of granulocytes was present in the MNC preparations. Cell viability of both cell fractions exceeded 98% as determined by the trypan blue dye exclusion after overnight culture. PMN and MNC were plated at 5 x 106 cells/ml into 10-cm tissue-culture Petri dishes and maintained in RPMI 1640 supplemented with 100 units/ml penicillin (Gibco-BRL, Grand Island, NY), 100 µg/ml streptomycin (Gibco-BRL), and 2 mM L-glutamine (Gibco-BRL) in the presence of 5% heat-inactivated fetal bovine serum (FBS; Sigma Chemical Co., St. Louis, MO) at 37°C in a humidified atmosphere with 5% CO2. Cells were stimulated by addition of serum-opsonized, heat-inactivated Escherichia coli to the culture medium at a 10:1 ratio bacteria-to-neutrophils or by purified LPS from E. coli serotype 026:B6 (Sigma Chemical Co.) at 10, 100, and 1000 ng/ml. When indicated, polymyxin B (PB) at 1 µg/ml was preincubated with LPS (100 ng/ml) for 30 min at room temperature before addition to the cell culture.
Reverse transcriptase-polymerase chain reaction (RT-PCR) and Southern blot analysis
Total RNA was isolated from homogenized tissue and cell samples as described [22
]. RT was performed by using 1 µg DNase-treated, total RNA primed with oligo(dT) adaptor primer (Amersham Biosciences, Little Chalfont, UK) in a final volume of 20 µl containing 0.5 mM dNTPs (Amersham Biosciences), 10 mM dithiothreitol, 40 units RNaseOUT inhibitor (Gibco-BRL), and 200 units SUPERSCRIPT II RT (Gibco-BRL). The reaction was carried out at 42°C for 2 h. PCR amplification was performed in a total volume of 25 µl including 0.6 units Amplitaq Gold DNA polymerase (Perkin Elmer, Foster City, CA), 2.5 mM MgCl2, 0.2 mM dNTPs, 0.5 µM each sense and antisense primer pairs, and 5 µl 1:10 diluted RT sample. After 10 min at 94°C, 25 or 30 cycles were performed with the following program: 30 s at 94°C, 40 s at the specific annealing temperature, and 1 min at 72°C. PCR products were analyzed in 1.5% agarose gel, and their sizes were compared with a 100-bp DNA ladder (New England Biolabs, Beverly, MA). Primer sequences were designed to specifically amplify the transcript regions of bovine cathelicidins, interleukin-8 (IL-8), and glyceraldehyde 3-phosphate dehydrogenase (GAPDH). For some cathelicidins, a sense primer (5'-GTGGAATTCAGGGTGAAGGAGAC-3') based on the conserved proregion of bovine cathelicidins and sequence-specific antisense primers were used as follows: Bac5: 5'-CTCAGGGCAAAAGAGAT-3'; Bac7: 5'-AGTGCTAACCTTGATGTT-3'; indolicidin: 5'-TCTGAACAAATCAGACACTTA-3'; BMAP-27: 5'-CGGAATTCACCCCAAATGGAGTA-3'; and BMAP-28: 5'-AATTGGGCCATACTTCTTCC-3'. For the cyclic dodecapeptide amplification, the sense and antisense primers were, respectively: 5'-GATGATGAAGACCCAGACAG-3' and 5'-GACGAATTCGAAAACCCTTAGGACTC-3'; for BMAP-34: 5'-ACCGAATTCAGCTACAGGGAGGCCGT-3' (sense) and 5'-ACCTGATCCTAAGGACT-TT-3' (antisense); for IL-8: 5'-GTGTGAAGCTGCAGTTCTG-3' (sense) and 5'-TTCTGCACCCACTTTTCC-3' (antisense); and for GAPDH: 5'-CATGCCATCACTGCCACCC-3' (sense) and 5'-ACCTGGTCCTCAGTGTAGC-3' (anti-sense). Bovine ß-defensin (BBD)-1S and BBD-2A primer sequences were designed to amplify a 163-bp-long region that is highly conserved among ß-defensins as reported [23
]. The identity of amplified products was confirmed by DNA sequencing with the BigDye Terminator kit (PE Applied Biosystems, Foster City, CA). For Southern analysis, RT-PCR amplified fragments run on a 1.5% agarose gel were blotted onto GeneScreen Plus nylon membrane (NEN Life Sciences, Boston, MA). Filters were prehybridized 1 h at 50°C in 0.5 M NaH2PO4, pH 7.2, containing 5% sodium dodecyl sulfate (SDS), 1 mM EDTA, and 100 µg/ml denatured salmon sperm DNA. The hybridization was carried out overnight at 50°C using a [32P] end-labeled probe corresponding to nt 429–449 of Bac5 cDNA [12
]. The membranes were washed twice for 5 min with 0.1% SDS in 2 x saline sodium citrate (SSC; 30 mM tri-sodium citrate, pH 7.0, 0.3 M NaCl) and twice with 0.1% SDS in 0.1 x SSC at 50°C for 15 min and were then rinsed in 2 x SSC and exposed to X-ray film (Kodak).
Northern blot analysis
Total RNA (7 µg) was size-separated on a 1% denaturing agarose gel, denaturated by incubation in 8 mM NaOH for 20 min, and then transferred to a GeneScreen Plus nylon membrane (NEN Life Sciences) in 3 M NaCl, 8 mM NaOH. The filter was prehybridized in ULTRAhyb solution (Ambion, Austin, TX) for 30 min at 42°C and then hybridized for 6–24 h at the same temperature with a [32P]-labeled fragment corresponding to 3' ends Bac5 cDNA coding for the antimicrobial domain [12
], to IL-8, and to ß-actin cDNAs. After hybridization, the membranes were washed twice for 5 min with 0.1% SDS in 2 x SSC and twice for 15 min with 0.1% SDS in 0.1 x SSC at 42°C, rinsed in 2 x SSC, and then exposed to X-ray film (Kodak).
Preparation of RNA probes and in situ hybridization
Digoxigenin (DIG)-labeled RNA probes were based on the 3' end of Bac5 cDNA [12
] coding for the antimicrobial domain, cloned in pBluescript-II SK. After linearization of template DNA with the appropriate restriction enzyme, in vitro transcription was performed by using T7 or T3 RNA polymerase according to the instructions of the DIG RNA labeling kit (Boehringer Mannheim, Mannheim, Germany).
Lung tissue (7-µm-thick sections) was mounted on slides and deparaffinized at 60°C for 15 min. Following dehydration, sections were treated with 1 µg/ml proteinase K and fixed in 4% paraformaldehyde in PBS. Acetylation was performed by incubating slides twice for 5 min with 0.25% acetic anhydride in 0.1 M triethanolamine, pH 8.0. Sections were prehybridized for 90 min at 50°C in 20 mM Tris-HCl, pH 7.5, containing 1x Denhardt’s solution, 300 mM NaCl, 100 mM dithiothreitol, 1 mM EDTA, 0.5 mg/ml salmon sperm DNA, 0.5 mg/ml polyadenylic acid (Boehringer Mannheim), and 50% formamide. Hybridization was performed overnight at 50°C in the hybridization solution plus 200 ng/ml DIG-labeled antisense or sense Bac5 probe. Slides were then washed three times for 20 min at 55°C with 50% formamide in 2 x SSC containing 0.1% Tween-20 (SSCT), 20 min at 55°C with 2 x SSCT, and if necessary, 20 min at 60°C in 0.2 x SSCT. Detection of the DIG-labeled probe was carried out by incubating sections with alkaline phosphatase-conjugated anti-DIG antibodies (Boehringer Mannheim) diluted 1:200 in PBS containing 0.1% Tween-20 and 10% FBS. Slides were rinsed four times in PBS with 0.1% Tween-20 for 10 min and incubated in a chromogen solution of nitro-blue tetrazolium/5-bromo-4-cloro-3-indolyl-phosphate plus 1 mM Levamisole. Nuclear Fast Red (BDH Chemicals, Poole, UK) was used as counterstain. Control of specificity was achieved by hybridizing with the sense riboprobe and by treating sections with 20 µg/ml RNase A at 37°C for 1 h before hybridization with the antisense probe. Hematoxylin/eosin and Giemsa-staining reagents were from Sigma Chemical Co.
Metabolic labeling, immunoprecipitation, and Western blotting
Peripheral blood neutrophils (2x107 cells/ml in 3-cm tissue-culture dishes) were maintained in RPMI 1640 in the presence of 5% heat-inactivated FBS and stimulated by addition of heat-inactivated E. coli opsonized with autologous serum at a 1:10 cells-to-bacteria ratio. The cell-free media were discarded after 6 h and replaced with Cys-free RPMI 1640 containing 5% FBS, 125 µCi/ml [35S]Cys (Amersham Biosciences), E. coli at 1:10 cell-to-bacteria ratio, and 1 mM phenylmethylsulfonyl fluoride. Incubations at 37°C were prolonged for further 6, 16, and 22 h. At each time point, cell-free media were collected, and cells were washed twice with ice-cold PBS. Cell pellets were solubilized at 4 x 107 cell equivalents/ml in lysis buffer (20 mM Tris-HCl, pH 8.2, 1.5% SDS) as described [13
]. Bac5 was immunoprecipitated from cell extracts and cell-free media using affinity-purified Bac5 antibodies and protein A-Sepharose [13
]. Protein recovered by immunoprecipitation was analyzed by a 15% polyacrylamide gel followed by Western blotting using Bac5-specific antibodies or fluorography [13
]. Exposure times at -80°C ranged from 15 to 21 days.
Sequencing and analysis of 5' flanking region of the gene
The sequence of the Bac5 gene was obtained from the genomic clone 
Cl1. This was identified by screening of a DASH II bovine genomic library with the conserved 5' cDNA of the bovine cathelicidins [24
]. Phage DNA was directly sequenced on both strands, and the 5' flanking region of the gene was analyzed for the presence of transcription factors binding sites using the MatInspector software [25
].
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (43K): [in a new window] |
Figure 1. (A) RT-PCR analysis of airway and intestine samples from healthy cows. (B) RT-PCR analysis of normal (Control) and infected cow lung. RT products were amplified using Bac5-, BBD-, or GAPDH-specific primer pairs, as indicated. The size of the amplified fragments is indicated on the left.
|
 View larger version (77K): [in a new window] |
Figure 2. In situ hybridization of inflamed bovine pulmonary tissue. Lung sections hybridized with antisense (A, C, F) and sense (B, D) Bac5 riboprobes. (E) Hematoxylin/eosin-stained serial section. Original magnification: A, B, E, F, 20x; C, D, 100x.
|
As shown in the Northern blot in Figure 3A , the Bac5 gene was not expressed at detectable levels in unstimulated neutrophils but was markedly induced after 20 h incubation of the cells with E. coli or LPS. The expression levels increased in a dose-dependent manner in the presence of increasing amounts of LPS and were found decreased when LPS was added in combination with polymyxin B (Fig. 3B) , a molecule which is known to bind LPS and neutralize its effects. The IL-8 gene was also strongly induced after a 20-h incubation of neutrophils with E. coli or E. coli LPS (Fig. 3A) , in good agreement with published results indicating expression of this gene in stimulated neutrophils [27 , 28 ]. Unlike Bac5, a basal expression of the IL-8 gene was also detected in the absence of added stimuli (Fig. 3A) .
 View larger version (24K): [in a new window] |
Figure 3. (A) Northern blot analysis of RNA from bovine PMN of a single adult animal. Peripheral blood neutrophils were cultured for up to 20 h in RPMI supplemented with 5% FBS in the absence (-) or the presence (+) of E. coli or 100 ng/ml E. coli LPS. Filters were sequentially hybridized under high-stringency conditions with Bac5 and IL-8 cDNA probes. Hybridization with ß-actin probe served as a control of the amount and quality of the RNA present in each lane. (B) Induction of Bac5 mRNA in peripheral neutrophils after 20 h incubation with increasing amounts of E. coli LPS in the absence (-) or the presence (+) of 1 µg/ml PB.
|
 View larger version (54K): [in a new window] |
Figure 4. RT-PCR from total RNA isolated from bovine PMN and MNC and Southern analysis of amplified Bac5 products (top row). Cells were purified from a single adult animal and incubated for 20 h in RPMI supplemented with 5% FBS in the presence (+) or the absence (-) of heat-killed, opsonized E. coli or 100 ng/ml E. coli LPS. Bac5- or GAPDH-specific primer pairs were used for PCR amplification. Southern blot analysis was carried out using a Bac5-specific probe.
|
 View larger version (24K): [in a new window] |
Figure 5. RT-PCR from total RNA isolated from bovine PMN cultured for the indicated time lengths in RPMI supplemented with 5% FBS, in the absence (-) or the presence (+) of heat-killed, opsonized E. coli (A) or 100 ng/ml E. coli LPS (B). Primer pairs specific for Bac5, IL-8, or GAPDH cDNAs were used for PCR amplification. Southern analysis of amplified products (A, top row) was carried out using a Bac5-specific probe.
|
 View larger version (61K): [in a new window] |
Figure 6. (A) Western blot of products recovered by immunoprecipitation from cell-free media of neutrophils incubated for 6 h in the absence (a, d) or presence (b, e) of E. coli and metabolically labeled with [35S]Cys for 6 h (a, b) and 16 h (d, e). Western blotting of SDS-lysed neutrophils (c). Bac5-specific antibodies were used for immunoprecipitation and Western analyses. The apparent molecular weight of the heavy and light immunoglobulin G chains and of proBac5 is indicated on the left. (B) Fluorographic analysis of Bac5-related products recovered by immunoprecipitation from cell-free media of neutrophils incubated for 6 h in the absence (a', d') or presence (b', e') of E. coli and biosynthetically labeled for 6 h (a', b') and 16 h (d', e').
|
B (NF-B), nuclear factor-interleukin-6 (NF-IL-6) (Fig. 7
).
 View larger version (54K): [in a new window] |
Figure 7. Nucleotide sequence of the 5' flanking region of the Bac5 gene (uppercase). The coding sequence is in lowercase, and putative transcription factor binding sites are underlined.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The antimicrobial peptides belonging to the cathelicidin family [8 9 10 11 ] are neutrophil-associated molecules, and some, i.e., LL-37 [16 ] and CRAMP [20 ], are also found in nonmyeloid tissues, particularly skin and other specialized epithelia [6 , 17 18 19 20 21 ].
Bac5 is a cathelicidin family member, which is stored in fairly large amounts (2 µg/106 cells) in the bovine neutrophil granules [14 ] and is processed to a mature, proline-rich, antimicrobial peptide in activated neutrophils [14 ]. We have investigated here the expression of this gene in epithelial tissues such as airways and intestine, which are common sites of antimicrobial peptide synthesis. Results obtained using PCR, in situ hybridization, and immunohistochemistry concur to demonstrate that unlike the human LL-37, the Bac5 gene is not expressed in these sites and is not induced in lung epithelium in response to infectious stimuli. Several other cathelicidins, however, are present in the bovine neutrophils in addition to Bac5 [9 ], and we cannot rule out that some of these have a different pattern of expression. An interesting result of this analysis is the detection of a Bac5-corresponding mRNA in the neutrophils massively recruited at inflammation sites in lung tissue. This result highlights the potential for de novo synthesis of Bac5 in fully differentiated neutrophils. To date, no other antimicrobial peptide genes have been reported to be induced in mature neutrophils of adult mammals. All previous work considered expression of the granule-associated antimicrobial peptides to be restricted to specific stages of cell differentiation [13 , 31 , 32 ], despite the known ability of neutrophils to synthesize cytosolic components [33 ] and a recent published report on the induction of the granule-associated enzyme proteinase 3 in cytokine-primed neutrophils [34 ].
Consistent with the in situ results suggesting induction of Bac5 in activated neutrophils, we demonstrate that E. coli and purified E. coli LPS trigger expression of this gene in primary cultures of peripheral neutrophils at the mRNA and protein levels. The kinetics of Bac5 induction in stimulated neutrophils suggests that this is a relatively late response (16–20 h post-stimulation), and the same is suggested by the detection of a radiolabeled product with a molecular weight compatible with that of proBac5 at correspondingly late times in the cell-free media of activated neutrophils. No labeled Bac5-related products are recognized in cell extracts of stimulated neutrophils. This result suggests that the sorting machinery for storage in the cytoplasmic granules is not active. The newly synthesized polypeptide is thus redirected to the constitutive secretory pathway and accumulates in the extracellular medium. The long cell incubation times in these experiments provide a likely explanation for the presence of lower molecular weight (10–11 kDa) proteolytic fragments in the incubation media of stimulated cells. Whether the newly synthesized propeptide is processed to the mature antimicrobial peptide, however, remains open to future studies.
To obtain a preliminary insight into possible regulatory mechanisms for gene induction in the mature neutrophils, we have inspected the 5' flanking region of the Bac5 gene and found several putative consensus sequences for binding to transcription factors involved in granulocyte gene regulation. Particularly, consensus sites for NF-B and NF-IL-6 are present in this sequence and appear ubiquitous in cathelicidin gene promoters [34
35
36
]. Their engagement in LPS-mediated induction of antimicrobial genes is conserved and widespread, as evidenced by studies on the transcriptional regulation of insect [3
], amphibian [37
], and mammalian antimicrobial genes [4
]. In mammals, NF-B appears to be essential and sufficient for full LPS responsiveness of the human ß-defensin-2 gene in RAW264.7 cells [38
], and NF-IL-6 and NF-B are required for LPS induction of the bovine tracheal antimicrobial peptide and human ß-defensin-2 genes in epithelial cells [39
, 40
]. It is interesting that LPS has also been shown to up-regulate the porcine cathelicidin genes PR-39 and protegrin in immature BM cells [36
], and although the mechanism has not been completely described, a role for NF-IL-6 and NF-B together with other transcription factors has been proposed in this case [36
]. The kinetics of induction of the two genes differs markedly from that observed for Bac5. The levels of PR-39 and protegrin mRNAs in marrow cells increase rapidly, peak at 6 h, and are decreased 12 h after LPS stimulation [36
], whereas we find here that the level of Bac5 mRNA in the mature neutrophils increases at approximately 16 h and is maintained high at 24 h after LPS stimulation. This distinct kinetics suggests that somewhat different mechanisms govern the cathelicidin induction in response to LPS in immature and mature neutrophils, likely involving different modules/combinations of transcription factors. Dissecting the pathways of induction at the molecular level will lead to a better understanding of these mechanisms.
A late boost of Bac5 expression in activated neutrophils could be consistent with the need to maintain a significant level of antimicrobial defense at sites of infection many hours after degranulation, although it is difficult to explain why this applies to only Bac5 out of all the bovine cathelicidins. An alternative explanation is the involvement of the newly synthesized Bac5 in other inflammatory responses. For instance, a signaling role would not require large amounts of the peptide but would require that the peptide be present in the inflammatory environment at the right time. The delayed Bac5 expression would seem to indicate that its presence is required for events whose timing is not coincident with degranulation. In this respect, it is interesting to note that the storage peptide (proBac5) may be largely depleted at the time of de novo synthesis, as suggested by Western analysis of neutrophil extracts after 22 h stimulation with E. coli (results not shown). The proposed hypothesis of a signaling role finds its rationale in the ability of other cathelicidin family members to mediate various host responses. For example, LL-37 is chemotactic for neutrophils, monocytes, and lymphocytes through interaction with formyl peptide-like receptor-1 [41 ], and the proline-rich PR-39 interacts with SH3 domains of cytoplasmic proteins and modulates expression of proteoglycans as part of wound repair in mammalian cells [42 , 43 ]. Preliminary results (L. Tomasinsig, unpublished data) indicate that Bac5 peptide is also capable of translocating across the plasma membrane of mammalian cells without causing membrane damage, and this further supports the potential for this peptide to interact with intracellular targets and possibly trigger cellular responses.
Received June 28, 2002; revised August 3, 2002; accepted August 5, 2002.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  M. Murakami, B. Lopez-Garcia, M. Braff, R. A. Dorschner, and R. L. Gallo Postsecretory Processing Generates Multiple Cathelicidins for Enhanced Topical Antimicrobial Defense J. Immunol., March 1, 2004; 172(5): 3070 - 3077. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Zanetti Cathelicidins, multifunctional peptides of the innate immunity J. Leukoc. Biol., January 1, 2004; 75(1): 39 - 48. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
