Introduction

Abstract Introduction Materials and Methods Results Discussion References

Secretory leukoprotease inhibitor (SLPI) is an 11.7-kD mucosal secretory protein that was identified as a potent inhibitor of leukocyte serine proteases (1). Immunocytochemical localization studies in humans revealed its presence in respiratory tissues and in salivary, genital, and lacrimal glands (2). Human SLPI is a nonglycosylated single-chain polypeptide consisting of 107 amino acids (1). A crystallographic evaluation showed a boomerang shape composed of two similar domains (3), and the active inhibitory site for elastase was in the C-terminal domain (4).

The human SLPI gene is distributed over four exons with a length of 2.6 kb (5). Recently, we cloned and characterized the mouse complementary DNA (cDNA) (6). The mouse cDNA encodes a protein of 131 amino acids with a 25-amino acid secretory signal peptide. The overall homology with human SLPI is 68% at the nucleotide level and 60% at the amino-acid level.

SLPI is involved in the protection against neutrophil elastase-induced damage at the sites of inflammation (7). This is supported by the following observations: (1) SLPI accounts for about 90% of the total molar concentration of elastase inhibitors in human bronchial secretions (8); (2) SLPI levels are increased in sera and respiratory epithelial lining fluids in inflammatory lung disorders (9, 10); (3) bacterial pneumonia causes augmented expression of the SLPI gene in the murine lung in vivo (6); and (4) SLPI gene expression in human epithelial cells in vitro is upregulated by inflammatory stimuli such as phorbol ester (11), neutrophil elastase (12), corticosteroids (13), interleukin-1 (IL-1) (14), and tumor necrosis factor- (14).

Although human SLPI has been intensively studied, its precise function in vivo remains unclear. For example, SLPI has been found to be overexpressed in a variety of carcinomas (15) and to exhibit anti-HIV-1 and antibacterial activities independent of its antiproteolytic function (16, 17). Moreover, a report that mouse SLPI works as a phagocyte-derived lipopolysaccharide inhibitor (18) raises the possibility that SLPI may function more broadly than previously considered.

For further investigation such as gene disruption, we have isolated and characterized the mouse Slpi gene. We also determined its chromosomal localization as well as the human homologue by fluorescence in situ hybridization.

	Materials and Methods

Abstract Introduction Materials and Methods Results Discussion References

Screening of a Mouse Genomic Library

The mouse Slpi cDNA was cloned from a mouse lung cDNA library in Uni-ZAP XR (Stratagene, La Jolla, CA) by taking advantage of cross-species homologies, and the insert was subcloned in pBluescript SK() (Stratagene), named pAT4 (6). A fragment was released from the recombinant plasmid (pAT4) by digestion with EcoRI and XhoI, and used as a probe to screen the 129SV mouse genomic library in Lambda Fix II (Stratagene) after labeling with [-³²P]dCTP (DuPont, Wilmington, DE) by the Random Primers DNA Labeling System (GIBCO BRL, Gaithersburg, MD). Screening of the library was conducted according to the manufacturer's instructions. Four positive plaques were obtained from 2 × 10⁶ plaques and confirmed in the secondary and tertiary screenings.

Subcloning and Sequencing

One of the four positive phage clones was amplified, and the DNA was extracted as described (19). The phage DNA digested with restriction enzymes was analyzed by Southern blot hybridization. A 3-kb EcoRI fragment, strongly hybridized with the mouse Slpi cDNA, was inserted into pBluescript II KS(+) (p2.6Eco). The recombinant plasmid p2.6Eco was sequenced by primer walking in both directions. Sequencing was performed using the ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit with AmpliTaq DNA Polymerase, FS (Applied Biosystems, Foster City, CA), and data were processed on an ABI Model 377 DNA sequencer (Applied Biosystems). The sequence data were analyzed using DNASIS-Mac 3.6.1 software (Hitachi Software Engineering, Yokohama, Japan).

Determination of the Transcription Start Site

A primer extension assay was conducted using the AMV Reverse Transcriptase Primer Extension System (Promega, Madison, WI). Briefly, 30-mer oligonucleotide primer P-ext-2, 5'-GCCTCCTTCCACAGTCCAGGGTGCCAGGAT-3', was end-labeled with [-³³P]ATP (DuPont) by T4 polynucleotide kinase. One picomole of radiolabeled primer was incubated with 1 µg of poly(A)⁺ RNA from mouse lung (9 to 10 wk old, male BALB/c) (Clontech, Palo Alto, CA) in AMV primer extension buffer (50 mM Tris-HCl [pH 8.3], 50 mM KCl, 10 mM MgCl₂, 10 mM dithiothreitol, 1 mM deoxynicotinamide triphosphate, and 0.5 mM spermidine) at 58°C for 20 min. Reverse transcription was then performed with AMV reverse transcriptase at 42°C for 30 min, followed by precipitation with ethanol. In the meantime, the p2.6Eco plasmid was sequenced using the dsDNA Cycle Sequencing System (GIBCO BRL) with the same end-labeled P-ext-2 primer. The reverse transcripts were electrophoresed with products of sequencing ractions on an 8% polyacrylamide-7 M urea sequencing gel. The gel was dried and exposed to a Fuji imaging plate (Fuji Photo Film Co., Minamiashigara, Japan). Scanning of the signals was performed using a bio-imaging analyzer system (Fuji Photo Film Co.).

Chromosomal Localization

Fluorescence in situ hybridization (FISH) was performed at Nihon Gene Research Laboratories, Inc. (Sendai, Japan). Briefly, bacterial artificial chromosome DNA containing the complete human SLPI gene and phage DNA containing the complete mouse Slpi gene were labeled by nick-translation with biotin-16-deoxyuridine triphosphate. Metaphase chromosomes prepared from human peripheral blood lymphocyte culture, and mouse embryonal cells were hybridized with the human and mouse genomic probes at 37°C for 16 h in a solution containing 2× standard saline citrate, 50% formamide, 10% dextran-sulfate, and 2 mg/ml bovine serum albumin. The chromosomes were counterstained with 0.5 µg/ml propidium iodide. The chromosome spreads were imaged with a fluorescence microscope (Olympus Corporation, Tokyo, Japan).

	Results

Abstract Introduction Materials and Methods Results Discussion References

Cloning and Nucleotide Sequence of the Murine Slpi Gene

The mouse Slpi gene is encoded by four exons spanning 2,222 base pairs (bp) of genomic DNA (Figure 1). The entire Slpi gene was elucidated by sequencing the 3-kb EcoRI fragment identified as an 0.8-kb 5'-flanking region, the 2,222-bp Slpi gene, and a 0.1-kb 3'-flanking region. The Slpi gene consists of four exons of 105, 162, 150, and 243 bp, and three introns of 575, 425, and 562 bp (Figure 1). The coding sequence completely matches the mouse SLPI amino acid and cDNA sequence (6, 18, 20). The first exon contains a 20-bp untranslated region and an 85-bp coding region (see below for mapping of the transcription start site). The cleavage site between glycine and glycine for removal of the signal peptide is present in the coding region of exon 1. Exon 3 encodes the reactive site for proteases Met⁷³-Met⁷⁴ as well as the stop codon (TGA). Exon 4 consists of an untranslated region encoding the polyadenylation signal. The splicing borders between the exons and introns of the mouse Slpi gene were conserved as predicted by GT-AG rules (21). The sequence for the mouse Slpi gene has been submitted to the GenBank/EMBL Data Libraries (Accession No. AF002719).

View larger version (8K): [in this window] [in a new window]   Figure 1.   Genomic structure of the murine Slpi gene, which consists of four exons and three introns. Boxes represent individual exons 1 to 4, with open, shaded, and filled regions indicating noncoding flanking regions, the leader signal sequence, and the coding regions, respectively, for the mature protein. Lines denote introns. The positions of the initiator element, the termination codon (TGA), and polyadenylation signal (AATAAA) are indicated. Several restriction enzyme sites are indicated with arrows. The arrowhead indicates the cleavage site to remove the signal peptide. The open triangle indicates the position of codon for the reactive site for proteases, Met73- Met74. The lengths of exons and introns (in brackets) are indicated below the scheme.

Determination of the Transcription Start Site

The transcription start site of the mouse Slpi gene was determined in a consensus initiator element by a primer extension assay. Two bands were present in the product of the primer extension assay (Figure 2). These two signals correspond to a T residue (stronger signal) and its adjacent C residue, which represent A and G, respectively, in the Slpi gene. This indicates that the transcription of the mouse Slpi gene starts mostly from the nucleotide indicated by the arrowhead in Figure 2. Moreover, nucleotide sequences around the transcription start site of the mouse Slpi gene, TCA₊₁GAGC, are very similar to the consensus sequence of the initiator element, (C/T)(C/T)A₊₁N(T/A)(C/T)(C/T) (22). Therefore, based on the results of the primer extension assay, the size of the first exon is 105 bp and the 5' noncoding region of the mouse Slpi gene has 20 nucleotides.

View larger version (55K): [in this window] [in a new window]   Figure 2.   Determination of the transcription start site of the murine Slpi gene. The double-stranded sequencing reactions (T, C, G, A) were run in parallel with the primer extension reaction (P). Primer extension assay was conducted with a radiolabeled 30-mer oligonucleotide as described in MATERIALS AND METHODS. The major transcription start site detected is indicated by an arrowhead. The corresponding sequence is shown on the right side. An arrow along the nucleotide sequence indicates the major start site, residue A.

Analysis of the Promoter Region

A number of potential regulatory elements were identified in the 837-bp 5'-flanking region of the mouse Slpi gene (Figure 3). The analysis revealed a TATA box-like sequence at 28 to 23 bp and an inverted GC box sequence at 50 to 41 bp from the transcription start site. These are typical distances from the transcription start site (23). Moreover, the 5'-flanking region contained the following motifs associated with an inflammatory response: seven interferon- response elements (-IRE) (CWKKANNY) (24), three nuclear factor (NF)-IL-6-binding sites (TKNNGNAAK) (25), one NF-B binding site (GGGRHTYYMC) (26), and one H-APF-1 binding site (CTGGRAA) (27). In addition, two AP-2 binding sites (GSSWGSCC and CCCMNSSS) (28, 29), one AP-1 binding site (TGASTMA) (28), one cyclic adenosine monophosphate-responsive element (CGTCA) (30), one HNF-5 binding site (TRTTTGY) (31), one c-Myb binding site (CMGTTR) (32), one GATA-1 binding site (WGATAR) (33), and one Ets-1 binding site (SAGGAAGY) (34) were also found in the 5'-flanking sequence of the mouse Slpi gene.

View larger version (29K): [in this window] [in a new window]   Figure 3.   The nucleotide sequence of the 5' flanking region and first exon of the murine Slpi gene. The numbering is based on the first nucleotide transcribed being +1. The arrows indicate potential regulatory elements in the sense (pointing rightward) and antisense (pointing leftward) strands. The location of the initiator element is double underlined with the transcription start site indicated by an arrow. Bold characters represent the coding sequence in exon 1. The sequence has been deposited in GenBank/EMBL Data Libraries under Accession No. AF002720.

Chromosomal Localization in Human and Mouse

By FISH analyses, the human SLPI gene was mapped to 20q12-13.2, and the mouse homologue to 2H (Figure 4). A total of 130 human metaphase spreads were examined to localize the human SLPI gene. Of 232 hybridization signals recorded, 192 were situated at human chromosome 20q12-13.2 (P < 0.01 by Student's t test), and the remaining signals were randomly distributed (Figures 4a-4c). For the mouse Slpi gene, 131 metaphase cells were examined. These metaphases showed specific signals on one or both chromatids of mouse chromosome 2 in the region 2H; 178 of 236 hybridization signals were at 2H (P < 0.01 by Student's t test) (Figures 4d-4f). Mouse chromosome 2H, thus determined as the mouse Slpi gene locus, has homologous loci on human chromosome 20q12-13 (35).

View larger version (43K): [in this window] [in a new window]   Figure 4.   Chromosomal localization of the human SLPI gene and the mouse homologue by in situ hybridization. The mapping of the human SLPI gene is shown in a-c, and that of the mouse Slpi gene in d-f. (a) and (d): Partial metaphase showing signals on both chromatids of human chromosome 20 (a) and mouse chromosome 2 (d) after hybridization with the human (a) and mouse (d) genomic probe. The arrows indicate fluorescent signals on 20q12-13.2 (a) and 2H (d). (b) and (e): Q-band-like image of the metaphase preparation identical with (a) and (d). (c) and (f ): Ideogram of human chromosome 20 showing the location of the human SLPI gene at 20q12-13.2 (c) and mouse chromosome 2 showing the location of the mouse homologue at 2H (f ). Each filled circle indicates 10 chromosomes with twin signals, and each open circle 10 chromosomes with a single signal. Note that 192 of 232 recorded signals were localized to human chromosome 20q12-13.2 (c), and that 178 of 236 recorded signals to mouse chromosome 2H (f ).

	Discussion

Abstract Introduction Materials and Methods Results Discussion References

This paper describes the genomic structure and the promoter region of the murine Slpi gene, which contains four exons spanning 2,222 bp on mouse chromosome 2H. Additionally, this report assigns the locus of the human homologue to human chromosome 20q12-13.2.

The exon-intron organization of the mouse Slpi gene is strikingly similar to that of the human SLPI gene reported previously (5). Because X-ray diffraction analysis of mature human SLPI revealed a boomerang-shaped molecule comprised of two topologically superimposable domains (Ser¹-Pro⁵⁴ and Asn⁵⁵-Ala¹⁰⁷) (3), mature mouse SLPI was also envisioned as composed of two domains (Gly¹- Arg⁵⁵ and Lys⁵⁶-Met¹⁰⁶) on the basis of the protein homology between human and mouse (6, 18, 20). Consistent with structural conservation at the protein level, the human SLPI and mouse Slpi genes are composed of four exons and have similar genomic organizations. In this context, the exons of both genes encode identical components of the SLPI proteins, namely, signal peptides on exon 1, the N-terminal domain on exon 2, the C-terminal domain on exon 3, and the 3' untranslated region on exon 4.

The function of the human SLPI promoter was previously studied by deletion analysis with the luciferase gene as a reporter (11, 36). In these analyses, the human SLPI promoter containing 115 bp upstream from the transcription start site was sufficient to drive expression of the luciferase gene (11, 36). In agreement with these results, comparison of the mouse Slpi and human SLPI 5'-flanking sequences (37) indicates that only the first 110 bp upstream of the transcription start site are highly conserved (73% identity). Moreover, -IRE (95 bp in the mouse and 94 bp in the human) as well as the TATA box-like sequences (23 bp in the mouse and 22 bp in the human) are at almost identical locations in the conserved promoter regions. A recent study demonstrated that mouse Slpi gene expression was transcriptionally regulated by interferon- in macrophages (18).

Consistent with linkage groups conserved between human and mouse (35), the human SLPI gene was localized on chromosome 20q12-13.2 and the mouse homologue on chromosome 2H. It is noteworthy that an epithelial-specific protease inhibitor known as elafin has been localized to chromosome 20q11.2-13.1 (38). Elafin, although different from SLPI, bears rough similarity to the latter in its primary structure (39) and its genomic organization (40).

In summary, we have presented the first characterization of the mouse Slpi gene. This work provides the framework to initiate targeted disruption of the Slpi gene by homologous recombination in the mouse.