A Novel Murine β-Defensin Expressed in Tongue, Esophagus, and Trachea*

β-Defensins are broad spectrum antimicrobial peptides expressed at epithelial surfaces. Two human β-defensins, HBD-1 and HBD-2, have been identified. In the lung, HBD-2 is an inducible product of airway epithelia and may play a role in innate mucosal defenses. We recently characterized rat homologs (RBD-1, RBD-2) of the human genes and used these sequences to identify novel mouse genes. Mouse β-defensin-4 (MBD-4) was amplified from lung cDNA using polymerase chain reaction primers designed from conserved sequences of RBD-2 and HBD-2. A full-length cDNA was cloned which encodes a putative peptide with the sequence MRIHYLLFTFLLVLLSPLAAFTQIINNPITCMTNGAICWGPCPTAFRQIGNCGHFKVRCCKIR. The peptide shares ∼40% identity with HBD-2. MBD-4 mRNA was expressed in the esophagus, tongue, and trachea but not in any of 20 other tissues surveyed. Cloning of the genomic sequence of MBD-4 revealed two nearly (>99%) identical sequences encoding MBD-4 and the presence of numerous additional highly similar genomic sequences. Radiation hybrid mapping localized this gene to a region of chromosome 8 near several other defensins, MBD-2, MBD-3, and α-defensins (cryptdins)-3 and -17, consistent with a gene cluster. Our genomic cloning and mapping data suggest that there is a large β-defensin gene family in mice. Identification of murine β-defensins provides an opportunity to understand further the role of these peptides in host defense through animal model studies and the generation of β-defensin-deficient animals by gene targeting.

Defensins are broad spectrum, cationic antimicrobial peptides expressed in phagocytic leukocytes and epithelial cells at mucosal surfaces (1). Their antibiotic activity and widespread expression suggest that they participate in innate immune responses in a variety of tissues. In addition, recent studies suggest that defensins may also signal to the adaptive immune system by acting as chemokines for T-cells and dendritic cells (2,3). Defensins have been classified into two families designated ␣ and ␤, based on distinctive, although similar, tridisulfide linkages in the processed peptides. In humans, six ␣-defensins, HD 1 -1 through HD-6, and two ␤-defensins, HBD-1 and HBD-2, have been identified to date. HD-1 through HD-4 are expressed in neutrophils, whereas HD-5 and HD-6 are expressed in epithelial cells of the intestinal and reproductive tract (1). HBD-1 is constitutively expressed in the kidney, urogenital tract, and several other epithelia (4 -8). In contrast, HBD-2 is present in skin, oral mucosa, lung, and other sites, and its expression is remarkably inducible in response to bacteria or proinflammatory cytokines (9 -12). The genes encoding HDs and HBDs are clustered in a contiguous segment of chromosome 8p23, and a current hypothesis holds that ␤-defensins may predate the ␣-defensins (13).
Through their presence as constitutive or inducible components of mucosal secretions, the many defensins are thought to contribute to the antimicrobial activity present at surfaces that frequently encounter bacteria (14). One example is the lung, where a very large surface area frequently comes in contact with microbes, yet the intrapulmonary airways and alveoli normally remain sterile. In the disease cystic fibrosis, chronic bacterial infection develops in the airways and is the chief cause of morbidity and mortality (15). One hypothesis for the origin of this persistent infection is that mutations in cystic fibrosis transmembrane conductance regulator cause alterations in normal ion transport across epithelia, which in turn elevates the NaCl concentration of the surface lining liquid and impairs the function of innate host defense factors, including ␤-defensins, expressed at this site (8, 16 -18).
To understand better the role ␤-defensins play in host defense it would be advantageous to find homologous genes in species of small laboratory animals. To date three ␤-defensin genes have been described in mice (19 -21) and two in rats (22). We hypothesized that a larger ␤-defensin gene family exists in mice and proceeded with gene discovery experiments. Here, we report the identification of a novel mouse ␤-defensin, MBD-4, encoding a peptide homologous to known ␤-defensins. MBD-4 is expressed in the esophagus, tongue, and trachea. Our data support the notion that there is a large family of ␤-defensin genes in mice encoding peptides involved in mucosal defenses. Identification of murine ␤-defensin homologs of human genes may further our understanding of their function through in vivo studies and the generation of ␤-defensin-deficient animals by gene targeting.

General Methodology
Oligonucleotide probes were end labeled to a specific activity of about 10 7 disintegrations/min/pmol using [␥-32 P]ATP as described (23). Purified plasmid DNA was sequenced from both strands using the dideoxy termination method (24).

Animals
All tissue samples were obtained from either adult FvB or C57BL/6 mice. For bacterial exposure experiments, animals were inoculated with 10 6 colony-forming units of Pseudomonas aeruginosa PAO1 intratracheally. All animal procedures were approved by the Animal Care Use and Review Committee of the University of Iowa or the Cleveland Clinic Foundation.

Southern Blot
A genomic Southern blot containing 8 g of genomic DNA from mouse, rat, cow, and human origin was digested to completion with EcoRI restriction enzymes (Bios Evo Blot). The blot was incubated with an oligonucleotide probe, RBD-2-6a (GGCGACAGCAGCACCAGGAGA-AATGAGAAGAGAAGGTAATGGATCCTCAT) based on sequence from the RBD-2 cDNA (22). The hybridization conditions were 35% (v/v) formamide, 5 ϫ SSC, 5 ϫ Denhardt's, 1% (w/v) SDS at 42°C overnight, and the high stringency wash conditions were 2 ϫ SSC, 0.1% SDS at 55°C for 30 min (25). The filter was then subjected to autoradiography at Ϫ70°C in the presence of a Cronex Lightning Plus intensifying screen (PerkinElmer Life Sciences) for 3 weeks.

Cloning of the MBD-4 cDNA
RT-PCR was conducted using mouse lung cDNA as a template and various oligonucleotide primer sets that were selected based on conserved regions of the HBD-2 and RBD-2 sequences (9,22). One set of primers (forward: ATGAGGATCCATTACCTTCTCTTCTCATTTCTC; reverse: GGTTAGTGTACGGACTGGTTTCCTCCGCATACG) generated a PCR product consistent with a partial ␤-defensin sequence. This was confirmed by DNA sequencing. The 5Ј-RACE and 3Ј-RACE techniques (26) were then used to isolate the full-length cDNA for a novel mouse ␤-defensin (Marathon cDNA template, CLONTECH).

Cloning of MBD-4 Genomic Sequences
A mouse genomic library (m129/SvJ) in Lambda Fix II (Stratagene, La Jolla, CA) was plated at a density of 5 ϫ 10 5 clones/150-mm plate (four genome equivalents). Plaque lifts were performed in duplicate using nylon membranes (Colony/Plaque Screen, PerkinElmer Life Sciences). The oligonucleotide RBD-2-6a was used as a hybridization probe under the same hybridization and wash conditions as used for the Southern blot analysis. Positives on duplicate filters were plaque purified, and then phage DNA was isolated using a polyethylene glycol precipitation/DNA adsorption protocol (Qiagen lambda midi kit, Valencia, CA). Isolated phage DNA was digested with restriction endonucleases and analyzed by Southern blot with MBD-4-specific oligonucleotide probes. Fragments of interest were subcloned into pBluescript II s/k (Stratagene).
Additional genomic sequences related to MBD-4 were identified via a sequential nested anchored PCR strategy using two antisense oligonucleotides from exon 1 of MBD-4 (ACAGCAGCACCAGGAGAAATGTG and CCAGGAGAAATGTGAAGAGAAGG) and two oligonucleotides (GTAATACGACTCACTATAGGGC and ACTATAGGGCACGCGTGGT) corresponding to the sequence of an anchored primer. The PCR template was mouse genomic DNA that had been first digested to completion with either DraI or SspI restriction enzyme and then modified by ligation to an adaptor oligonucleotide (GTAATACGACTCACT-ATAGGGCACGCGTGGTCGACGGCCCGGGCTGGT) containing a sequence complementary to the aforementioned anchor sequence (Genomewalker, CLONTECH Laboratories). The PCR was carried out using conditions recommended by the reagent supplier. The PCR products of interest were identified by Southern blot hybridization analysis, subcloned into a pBluescript II s/k plasmid vector, and sequenced in their entirety. The sequence of several independent clones was nearly identical to the 5Ј-flanking region observed in the phage clone isolated as described above, but each contained several nucleotide differences from the phage sequence. To understand if these sequences represented alternative alleles or additional genes highly similar to MBD-4, an oligonucleotide (CGGTTTGTTGAAGTTCTGTGATT) from the 5Ј-flanking region 1.5 kb upstream of the transcription start site together with another partially degenerate oligonucleotide (TTGTSMA-TCTTCATGGAGGAGCAAATT) common to exon 2 of both MBD-3 and MBD-4 were used in a PCR with m129/FvB mouse genomic DNA as a template. A single product of expected size (4.1 kb) was identified and then subcloned into pBluescript II s/k for sequence analysis.

Chromosomal Mapping of Mouse Defensins Using Radiation Hybrids
A sequence tagged site (STS) specific for each of five mouse defensin genes, MBD-1 (Defb1), MBD-2 (Defb2), MBD-3 (Defb3), cryptdin-3 (Defcr3), and cryptdin-17 (Defcr17), was designed from their respective cDNA sequences found in GenBank (Table I). The STS specific for FIG. 1. Southern blot. A radiolabeled oligonucleotide probe derived from the 5Ј-end of the RBD-2 sequence was hybridized against mouse, rat, bovine, and human genomic DNA. Under high stringency conditions, seven distinct bands (asterisks) were detected in mouse, suggesting the presence of a family of genes with sequence homology to RBD-2. cagcatcagtggcctcagta MBD-4(Defb4) was designed from cDNA sequence obtained in this study (GenBank AF155882). The STSs were mapped with the T31 mouse radiation hybrid panel (27). The panel was obtained from Research Genetics (Huntsville, AL) and contains 102 DNA samples along with mouse and hamster genomic DNA controls. Standard PCR conditions were used as described above. Each PCR experiment was performed in duplicate, and the amplification results for each STS were submitted to the Radiation Hybrid Mapper server at MIT and at the Jackson Laboratory. A strong signal in both trials was submitted as a "1," no signal in both trials was submitted as a "0," and weak or conflicting signals were submitted as a "2." The servers returned the MIT framework marker that was most tightly linked to the subject STS when the Logarithm of the ODds score exceeded 3 for the MIT server and 6 for the Jackson Lab server (Table I).
RT-PCR-RT-PCR was used to screen tissues for expression of MBD-4 and MBD-3. 1 g of total RNA from each sample was reverse transcribed by random hexamer primers using SuperScript (Life Technologies, Inc.). First strand cDNA was amplified by PCR. As a control for amplification, a pair of glyceraldehyde-3-phosphate dehydrogenase primers (forward: AGACAGCCGCATCTTCTTGT; reverse: CTTGCCG-TGGGTAGAGTCAT) was used in each PCR. The primer set used to amplify MBD-4 consisted of AACATGCATGACCAATGGAG (forward) and TCATCTTGCTGGTTCTTCATCT (reverse) and amplified a 134-bp product. The primer set for MBD-3 consisted of GCTAGGGAGCACTT-GTTTGC (forward) and TTGTTTGAGGAAAGGAGGCA (reverse) and amplified a product of 220 bp. Each MBD-4 and MBD-3 reaction contained approximately 1.25 pM primers, 3 mM Mg 2ϩ , and 5 l of the RT reaction product for a total volume of 20 l. After an initial denaturing step (95°C for 3 min), 30 cycles of denaturing (94°C for 30 s), annealing (60°C for 30 s), and extending (72°C for 30 s), followed by 5 min at 72°C for elongation were conducted. Aliquots of the MBD-4 and MBD-3 PCR products were separated on a 2% agarose gel and visualized with ethidium bromide.
RNase Protection Assay-Total RNA was isolated from trachea, lung, and tongue using the acid guanidinium thiocyanate-phenol-chloroform method of Chomczynski and Sacchi (28). Ribonuclease protection assays were used to quantitate the mRNA expression of MBD-4 in tissues (5). [␣-32 P]dUTP antisense riboprobes for MBD-4 were prepared using the MAXIscript in vitro transcription kit following the manufacturer's instructions (Ambion, Austin, TX). 1 ng of each probe was used to generate protected fragments of 148 and 80 bp for MBD-4 and 18 S, respectively. Unprotected riboprobe sizes were 327 bp and 128 bp for MBD-4 and 18 S, respectively. Hybridization was performed using 20 g of total RNA following the manufacturer's instructions in the Hybspeed RPA kit (Ambion). RNA-RNA hybrids were digested with RNase and protected fragments visualized by polyacrylamide gel electrophoresis and autoradiography (5).

RESULTS
Southern Blot-Using an oligonucleotide probe (RBD-2-6a) from the 5Ј-portion of the recently characterized RBD-2 cDNA (22) in Southern blot analysis, seven discrete bands in mouse genomic DNA were identified (Fig. 1). Using probes from bovine and human ␤-defensins in similar Southern blot analysis failed to detect any discrete bands in either rat or mouse genomic DNA. 2 This observation suggests that rodent ␤-defensin genes have diverged significantly from those in these two other mammalian species.
Cloning and Sequencing of the MBD-4 cDNA-A PCR strategy was used to clone a novel murine ␤-defensin sequence with homology to RBD-2 and HBD-2. Several PCR primer pairs were designed from the RBD-2 cDNA sequence based on highly conserved portions of the HBD-2 (9) and RBD-2 (22) amino acid sequences. One such primer pair amplified a product of the predicted size from mouse lung cDNA (see "Experimental Procedures"). The complete 327-bp cDNA sequence was obtained by RACE (Fig. 2). The predicted 192-bp open reading frame codes for a 63-amino acid peptide. The putative peptide sequence shares approximately 40, 62, and 75% identity with HBD-2, MBD-3, and RBD-2, respectively. Blast search of the EST data base confirmed that this was a unique sequence with similarity to three other mouse ␤-defensins (MBD-1, MBD-2 (Defb-2), and MBD-3). In addition, the presumed translation product contains the conserved 6-cysteine motif characteristic of a ␤-defensin. We conclude that we cloned a new mouse ␤-defensin gene, MBD-4. This sequence was entered into the GenBank data base under the accession number AF155882.
Genomic Organization of MBD-4 -To initiate characterization of the mouse sequences identified in Fig. 2, RBD-2-6a was used as a probe to screen a mouse genomic library under the same experimental conditions. Cloning at the genomic level offers potential advantage of identifying genes that might be 2 F. Barahmand-pour and C. L. Bevins, unpublished observations. inducible or present in unexpected tissues or stages of development. This strategy has provided a useful approach to identify several defensins (29 -32). From approximately four mouse genome equivalents of recombinant lambda phage clones, we detected 69 corresponding signals on duplicate filters. 28 clones were purified and then categorized by a combination of restric-tion enzyme, hybridization and partial sequence analysis. All genomic clones had inserts in the range of ϳ10 -15 kb, and some appeared to contain more than one restriction fragment that contained a ␤-defensin-related sequence (data not shown). Hybridization properties to a panel of oligonucleotides were used to segregate these clones into several groups. One clone, SW30 -52, from a group that appeared to contain the MBD-4 sequence, was selected for in-depth characterization. XbaI and SacI restriction fragments that contained the MBD-4-like sequence within this clone were isolated, and the nucleotide sequence was determined (Fig. 3). Sequence analysis revealed two exons that corresponded exactly to the MBD-4 cDNA sequence, separated by an intron of 2.4 kb. The sequence was submitted to GenBank with the accession number AF287475. A second MBD-4 genomic sequence was also cloned using a combination of PCR strategies as described under "Experimental Procedures." This sequence contained 29 nucleotide differences in the intron and 5Ј-flanking regions and 2 nucleotide differences in the coding region (Fig. 3). This sequence was submitted to GenBank with the accession number AF288371. It is likely that this sequence represents a second highly similar gene, as opposed to a second allele, given that both were identified in genomic DNA from inbred 129 mouse strains and that multiple bands of similar intensity were observed by Southern blot analysis (Fig. 1).
The MBD-4 5Ј-flanking sequence exhibited no consensus sequences for NFB binding. A comparison of the proximal 5Јflanking region of the MBD-3 and MBD-4 genes shows remarkable similarity except for the omission of 26 bp of sequence in the MBD-4 gene which encompasses the NFB consensus binding site. However, similar to MBD-3, several NF interleukin-6 and interferon-␥ consensus sequences were present.
MBD-4 Localizes to the Defensin Locus on Chromosome 8 -Previously, individual defensin and cryptdin genes were localized to mouse chromosome 8 (19,21,(33)(34)(35)(36), but comprehensive mapping of this gene family has not been performed. We used the T31 mouse radiation hybrid panel (27) to map MBD-4 (Defb4) along with the previously identified MBD-1, -2, and -3 (Defb1-3) and two representatives of the cryptdin family, cryptdins -3 and -17(Defcr3 and Defcr17). The overall retention pattern for these six STSs was very similar, and all but Defb1 (see below) map in a narrow interval between markers D8Mit159 and D8Mit257. This interval is near the centromere of chromosome 8 and is consistent with previous studies that localized individual defensin genes by fluorescent in situ hybridization or by genetic linkage. Based on the radiation hybrid data, the order of the defensin genes on mouse chromosome 8 is CEN-D8Mit159/Defb3-Defb4-D8Mit170-Defcr3-Defb2-Defcr17-D8Mit257-TEL (Fig. 4). The defensin gene Defb1 could not be localized anywhere in the mouse genome with a Logarithm of the ODds Ͼ3. Although all of the cell lines that amplified with the Defb1 STS also amplified with other defensin STSs, the retention frequency was decreased 3-fold (data not shown). Similarly, the retention frequency was reduced when the human homolog of the Defb1 gene (HBD-1) was mapped by radiation hybrid mapping (37), suggesting that a conserved DNA sequence(s) may colocalize with the Defb1/HBD1 locus in both genomes, which causes this region to be genetically unstable. These results are consistent with the mouse ␤and ␣-defensin genes existing as a cluster on chromosome 8.

MBD-4 Expression Is Restricted to Specific
Tissues-Northern analysis was used to screen for the tissue expression pattern of MBD-4. These results are shown in Fig. 5A. For comparison, the same Northern blot filter was screened for MBD-1, MBD-2, MBD-3, and cryptdin expression. MBD-4 mRNA expression was restricted to a limited number of tissues. Positive signals were readily detected in tongue, esophagus, and trachea. In contrast, the signals for MBD-1, MBD-2, and MBD-3 were overall much less intense in the same tissues. MBD-1 was detectable in kidney, consistent with previous studies (19 -21). For MBD-2, no hybridization signals were detected by Northern analysis. The MBD-3 mRNA was weakly detectable in the tongue. The cryptdin signal was strong in the small intestine samples, consistent with published data (38). We also used a sensitive RT-PCR method to confirm the tissues expressing MBD-3 and MBD-4. As shown in Fig. 5B, although a wide array of tissues was analyzed, MBD-4 mRNA expression was again detected only in tongue, esophagus, and trachea. MBD-3 expression was also detected in the tongue, esophagus, and trachea.
Regulation of MBD-4 Expression-An important feature of the innate immune system is that it is either ever ready or rapidly able to respond to microbial challenges from the environment. In this regard, epithelial ␤-defensins described to date have been characterized as constitutively expressed or inducible. For example, the expression of bovine TAP, a prototypical epithelial ␤-defensin, is markedly induced by proinflammatory stimuli such as bacterial lipopolysaccharide, tumor necrosis factor-␣, or interleukin-1 (39,40). In contrast, HBD-1 is constitutively expressed at several epithelial sites, and expression does not change in response to proinflammatory stimuli (4,6,10,12). To test whether MBD-4 expression in the trachea is inducible, mice were challenged intratracheally with P. aeruginosa PAO1 (10 6 colony-forming units in 100 l). Control animals received phosphate-buffered saline in a similar volume. In previous studies we have shown that such a chal-lenge with PAO1 stimulates a robust pulmonary inflammatory response (18). Animals were sacrificed 24 h after the bacterial challenge, and tissues were evaluated for MBD-4 expression by a sensitive ribonuclease protection assay. As shown in Fig. 6, intratracheal bacterial challenge had no effect on the abundance of MBD-4 mRNA in the trachea, suggesting that its expression at this site is constitutive. DISCUSSION In this work we report the discovery of a novel mouse ␤-defensin with an interesting pattern of tissue expression. MBD-4 mRNA expression was observed in tongue, esophagus, and trachea but not in several other mucosal tissues where defensins have been identified in other species. The genomic organization of the MBD-4 gene is similar to other ␤-defensins reported, and evidence for two nearly identical versions of this gene was obtained. Radiation hybrid mapping confirmed that MBD-4 is part of the mouse defensin locus, and for the first time, we colocalized the ␤-defensin and ␣-defensin (cryptdin) gene families to a narrow region of mouse chromosome 8. This information, coupled with our Southern blot and genomic cloning data, supports the notion that the mouse defensin family exists as a gene cluster and suggests that there are likely to be several other related family members.
We found that the MBD-4 genomic sequence contains two exons and a single intron, similar to other mammalian ␤-defensins (14). The putative MBD-4 peptide sequence contains the conserved 6-cysteine motif characteristic of the ␤-defensins and is homologous with other epithelial defensins. In contrast to the inducible ␤-defensins TAP, HBD-2, and MBD-3 (9,21,40), the MBD-4 5Ј-flanking sequence exhibited no consensus sequences for NFB binding. A comparison of the proximal 5Ј-flanking region of the MBD-3 and MBD-4 genes shows remarkable similarity except for the omission of 26 bp of sequence in the MBD-4 gene which encompasses the NFB consensus binding site. However, similar to MBD-3, several NF interleukin-6 and interferon-␥ consensus sequences are present in the MBD-4 promoter region, suggesting the possibility for regulation at the transcriptional level resulting in inducible expression. Following intratracheal challenge with bacteria, we found no evidence that MBD-4 expression increased. Perhaps MBD-4 expression is constitutive as reported for HBD-1 and the recently described porcine ␤-defensin-1 (41). However, this experiment does not rule out the possibility that there may be other stimuli that influence MBD-4 expression. Similarly, it is possible that mouse strain or housing conditions could influence the response to an inflammatory challenge. Further studies are needed to address these possibilities.
We used radiation hybrid mapping to localize MBD-2, MBD-3, MBD-4, and the ␣-defensins (cryptdins)-3 and -17 to a contiguous region on mouse chromosome 8. The data for MBD-1 were noninformative by this technique, but others have mapped the MBD-1 gene to chromosome 8 near the ␣-defensin locus (19,35). These data confirm and extend previous reports linking the mouse defensin locus to chromosome 8p (19,20,35), the region syntenic to the human locus on 8p23 (13,37,42,44). Importantly, we were able to demonstrate simultaneously, using a single technique, that the mouse ␤and ␣-defensins exist as a gene cluster, similar to that described in humans (13). These data from the mouse add further support to the speculation that these genes arose from the duplication of an ancestral gene (13,42,45,46).
The tissue distribution of MBD-4 mRNA expression is uniquely restricted. Although clearly species to species variations exist, these data support the speculation that most mucosal surfaces express a specific repertoire of ␤-defensins. The abundant expression of MBD-4 in the tongue is reminiscent of FIG. 6. Effect of bacterial challenge on MBD-4 expression in mouse trachea. Mice received P. aeruginosa PAO1 intratracheally as described under "Experimental Procedures." 24 h later tissues from the tongue, lung, and trachea were obtained, and RNA was extracted and used in a ribonuclease protection assay. A representative result from individual animals is shown from six control and six PAO1 animals. The MBD-4 signal was detected readily in the tongue, with a less abundant signal in trachea, and no signal detected in lung. There was no appreciable change in mRNA expression following the bacterial challenge. The 18 S RNA subunit serves as a control. related gene products identified in cows (47), pigs (48), and humans (12). Similarly, the bovine ␤-defensins TAP and LAP (47,49), pig ␤-defensin-1 (48), sheep ␤-defensin-1 (43), and HBD-1 and HBD-2 (4,5,8,10,11) are expressed in tracheal epithelia. Few studies have documented the presence of ␤-defensins in the esophagus, but there is evidence of such expression in cows (47), pigs (48), and sheep (43). The restricted pattern of expression exhibited by MBD-4 suggests that the peptide product may have effects that are best suited for function at these mucosal surfaces. Further studies of the peptide product are needed to address this possibility.
Our Northern blot data show a more limited distribution of MBD-3 expression than that reported by Bals et al. (21). The high similarity of nucleotide sequence for MBD-3 and MBD-4 suggests that cross-hybridization is very likely when fulllength cDNA probes are used. Highly similar cDNA sequences for various ␤-defensin cDNA were also detected in cattle and prompted the development of a strategy of using oligonucleotide probes to assess effectively tissue-specific expression of related ␤-defensins (23). Although other explanations are possible for the discrepant findings regarding MBD-3 expression, our identification of at least one, and possibly other, ␤-defensin(s) with similar nucleotide sequence suggests that tissue expression of related ␤-defensins should be assessed using highly discriminating methodology.
Our Southern blot analysis suggests that the mouse defensin locus encodes several ␤-defensin genes. All known mouse ␤-defensin genes map to chromosome 8, but some are quite divergent at the nucleotide and amino acid level. This diversity makes hybridization and antibody based approaches to gene discovery more difficult. A genomic sequencing approach is an alternative approach to identify other members of the gene family in mice. Further studies are needed to identify the additional ␤-defensins and determine the structure of the mouse defensin gene cluster. Such information will allow a better understanding of their role in host defense through focused animal model studies and the generation of ␤-defensin deficient animals by gene targeting.