• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of molcellbPermissionsJournals.ASM.orgJournalMCB ArticleJournal InfoAuthorsReviewers
Mol Cell Biol. Jan 2004; 24(1): 172–185.
PMCID: PMC303351

The Homeobox Gene Caudal Regulates Constitutive Local Expression of Antimicrobial Peptide Genes in Drosophila Epithelia

Abstract

In Drosophila melanogaster, although the NF-κB transcription factors play a pivotal role in the inducible expression of innate immune genes, such as antimicrobial peptide genes, the exact regulatory mechanism of the tissue-specific constitutive expression of these genes in barrier epithelia is largely unknown. Here, we show that the Drosophila homeobox gene product Caudal functions as the innate immune transcription modulator that is responsible for the constitutive local expression of antimicrobial peptides cecropin and drosomycin in a tissue-specific manner. These results suggest that certain epithelial tissues have evolved a unique constitutive innate immune strategy by recruiting a developmental “master control” gene.

The innate immune system is an essential means of host defense in all eukaryotes, and this system also plays an instructive role in the induction of adaptive immunity in vertebrates (32). The production of antimicrobial peptides (AMPs) is the key feature of innate immunity aimed at neutralizing microbial infections in all multicellular organisms inhabiting various microbial environments (17). In the past decade, our understanding of innate immune signaling pathways leading to AMP expression in Drosophila melanogaster has dramatically increased. Most studies focused on the inducible systemic AMP gene expression observed in response to bacterial injection in the hemocoel. Genetic evidence from Drosophila demonstrates the existence of at least two distinct regulatory mechanisms for AMP synthesis in systemic innate immunity: the Toll pathway and the immune deficiency (IMD) pathway. The Toll pathway, primarily involved in drosomycin (Drs) antifungal peptide expression, requires a hemolymph serine proteinase cascade for its activation. This cascade, initiated by soluble pattern recognition proteins, is required for the processing of the Toll ligand, spaetzle, and for the subsequent activation of the p65-like Rel protein, Dif (18, 25, 29, 33, 34, 44). The IMD pathway is more specifically implicated in the expression of antibacterial peptide genes (such as Cecropin [Cec] and Diptericin [Dipt]) than the Toll pathway and requires the sequential activation of membrane peptidoglycan recognition protein receptor, IMD, TAK1, dFADD, Dredd, IκB kinase, and the p105-like Rel protein, Relish (6, 11, 12, 14, 20, 27, 28, 31, 37, 42, 45, 47, 52). In addition to these NF-κB signaling pathways, the c-Jun NH2-terminal kinase (JNK) and JAK-STAT pathways are also involved in other immune functions, such as cytoskeletal remodeling for wound healing in the case of the JNK pathway (3).

In Drosophila, all known AMP genes are synthesized by the fat body, a functional homologue of the mammalian liver, during a systemic immune response (15, 50). However, it is believed that the first line of defense of the organism consists of the local expression of AMPs in barrier epithelia (referred to as local innate immunity), which are in direct contact with microorganisms (5, 50, 53). The in vivo monitoring of AMP expression in transgenic flies, expressing green fluorescence protein (GFP), revealed the existence of two distinct types of local innate immunity: the so called inducible local AMP gene expression and constitutive local AMP gene expression. In inducible local AMP gene expression, most barrier epithelia express at least one AMP in an inducible tissue-specific manner, primarily through the IMD pathway (10, 38, 51). For example, Drs and Dipt are induced in the tracheae and the gut, respectively, via the IMD pathway in response to local infection by bacteria such as Erwinia carotovora (10, 51). In the midgut and the proventriculus, Cec expression is normally absent but is rapidly induced in response to local infection (51). This inducible local immunity is activated by natural local infection but not by bacterial injection into the hemocoel, used for the initiation of systemic immunity. The other important form of local AMP gene expression is the constitutive form. In this case, the AMP gene is expressed constitutively in a defined tissue and its expression is not up-regulated during microbial infection (10, 51). To date, the regulatory signaling pathway(s) controlling constitutive local AMP gene expression is unknown. The most intense constitutive expression of Cec is found in the reproductive organs, such as the male ejaculatory duct (51). For Drs, the strongest constitutive expression is found in the salivary glands and the female reproductive organs (51). Furthermore, it was demonstrated that this type of Drs expression was independent of NF-κB pathways (Toll and IMD pathways) (10, 51).

The high complexity of AMP regulation indicates that the gene promoters must be regulated by different types of trans-activators. The κB sites (found in all known AMPs) and Rel family transcription factors (Dif and Relish) are essential for inducible expression of all AMPs during systemic immune and inducible local immune responses. As for Cec, in addition to a κB site, a GATA site is necessary for fat body-specific and immune-inducible expression in vivo (38, 40, 48). As mentioned above, the molecular mechanism of infection-independent constitutive local expression of AMPs in the barrier epithelia that are in direct or indirect contact with the external environment is presently unknown. We suspect that other important cis elements and trans-activators are involved in the regulation of AMPs in certain epithelia. In this study, we provide evidence that Caudal (Cad), in addition to its role as a homeotic transcription factor for anteroposterior body axis formation, is involved in the constitutive expression of a subset of AMP genes in epithelia.

MATERIALS AND METHODS

Fly strains.

Flies were maintained on standard cornmeal-agar medium at 25°C. OregonR flies were used as the standard wild-type strain. The Cec-GFP- and Drs-GFP-expressing flies were obtained from B. Lemaitre (CNRS, Gif-sur-Yvette, France) (51). The RelishE20-expressing mutant flies were obtained from D. Hultmark (University of Umeå, Umeå, Sweden) (14). The c729-GAL4-expressing line was obtained from Y. Engström (University of Stockholm, Stockholm, Sweden) (40). The Yolk-GAL4-expressing line was obtained from D. Ferrandon (Institut de Biologie Moléculaire et Cellulaire, Strasbourg, France) (11). The lines expressing HS-GAL4, Da-GAL4, and cad3 mutant [b1 pr1 cad3/In(2LR)Gla, wgGla-1] were obtained from the Bloomington Stock Center.

Electrophoretic mobility gel shift assay (EMSA).

The recombinant glutathione S-transferase (GST)-Cad DNA binding domain (amino acids 273 to 427) was expressed in Escherichia coli and purified by glutathione-Sepharose affinity chromatography according to the vendor's directions (Amersham Pharmacia Biotech). Nuclear extracts were prepared as described previously (20) from untreated and heat-shocked UAS-Cad/+; HS-GAL4/+-expressing transgenic flies. The binding reaction was performed for 30 min at room temperature by mixing 1 ng of purified 32P-labeled probe, 10 μg of nuclear extracts (or 50 ng of GST-Cad), and 300 ng of poly(dI-dC) in the presence of a protease inhibitor cocktail (Complete; Roche Molecular Biochemicals) as described previously (20). Supershift analysis was performed by mixing nuclear extract with antiserum against the activation domain of Cad (amino acids 1 to 272) or preimmune serum for 15 min at room temperature prior to adding the 32P-labeled probes.

Real-time quantitative PCR analysis.

The total RNA was extracted with an RNAzol reagent. The first cDNA was synthesized by using a first cDNA synthesis kit (Roche) according to the manufacturer's instruction. Fluorescence real-time PCR was performed with double-stranded DNA dye SYBR Green (Perkin-Elmer) to quantify the amount of gene expression. Primer pairs for Cec (sense, 5′-ATG AAC TTC TAC AAC ATC TTC G-3′; antisense, 5′-GGC AGT TGC GGC GAC ATT GGC G-3′), Drs (sense, 5′-GCA GAT CAA GTA CTT GTT CGC CC-3′; antisense, 5′-CTT CGC ACC AGC ACT TCA GAC TGG-3′), Cad (sense, 5′-CCA TCG AAG CCG CCA TAC T-3′; antisense, 5′-TTT GCC TGG TTG TGG TTG TG-3′), glucose dehydrogenase (sense, 5′-GGA AGC CGC CGC GTA TTG TG-3′; antisense, 5′-GAT TCT CCG GAC CCG TGT TCT GC-3′), B52 (sense, 5′-CAC CGG ACC GCA ATA ACG AGA GCA-3′; antisense, 5′-GAC GAG GCC CGA CAG TGG TGG ATT-3′), and the control Rac2 (sense, 5′-CAG ACG ATC GAG AAG CTG AAG G-3′; antisense, 5′-GTG CCG CTT GGG TCC TCG AAC G-3′) were used to detect the target gene transcripts. SYBR Green analysis was performed on an ABI PRISM 7700 system (PE Applied Biosystems) according to the manufacturer's instruction. All samples were analyzed in triplicate, and the levels of messages detected were normalized relative to the control Rac2 values. The normalized data were used to quantify the relative levels of a given mRNA according to the ΔCt analysis (26).

Reporter gene assay.

Drosophila immunocompetent Schneider cells (ATCC CRL-1963) were maintained in Schneider medium (Sigma) as previously described (13). Transient transfections were carried out by the calcium phosphate method (9). All transfection mixtures contained 100 ng of pPacPL-LacZ as an internal standard, 3 μg of pPacPL-Cad, and 100 ng of the luciferase reporter constructs (Drs-luciferase construct [a 2.4-kb upstream fragment of the Drs promoter] or the Cec-luciferase reporter construct [bp −751 to +71]). In addition, various Drs and Cec reporter constructs with deletions and point mutations were also cotransfected. At 48 h after transfection, luciferase activity was measured according to the manufacturer's instructions (Promega). Unlike in vivo experiments, these experiments required, for unknown reasons, incubation of the Schneider cells with lipopolysaccharide (10 μg/ml) for 6 h in order to conduct the Cad-induced Cec reporter assay. Luciferase activity was normalized with respect to β-galactosidase activity to correct for variations in the transfection efficiency.

In vivo detection of reporter transgenes.

GFP reporter-expressing flies and dissected organs were examined under a stereofluorescence microscope (Leica; MZFLIII). Histochemical analysis of β-galactosidase expression was performed as previously described (23). In the bacterial-challenge experiment, the flies were pricked with a fine needle previously dipped into a concentrated culture of Escherichia coli and Micrococcus luteus. Quantitative analysis of the GFP reporter was performed with protein extract (100 μg) of the ejaculatory ducts by using the spectrofluorophotometer (excitation wavelength, 488 nm; emission wavelength, 507 nm) according to the manufacturer's instructions (Shimadzu; RF-5301PC).

Plasmids and the generation of transgenic animals.

The promoter region (3.1 kb) of Cad was generated by PCR and subcloned into the pCaSpeR-AUG-β-gal vector to obtain the Cad-LacZ construct. The open reading frame of the Cad cDNA was subcloned into the pUAST vector (4) to obtain the upstream activation sequence (UAS)-Cad construct. To obtain the UAS-Cad-RNAi construct, a 510-bp cDNA fragment encoding amino acids 24 to 193 of Cad was amplified by PCR and the head-to-tail inverted repeats were subcloned into the pUAST vector. Mutations of Cad binding sites (S2 and S5; see Fig. Fig.1)1) on the Cec A1 gene promoter construct (bp −751 to +71) were created by site-directed mutagenesis, and subsequently the Drs-GFP plasmid PJM802 was replaced with the NheI-SpeI fragment of the S2- and S5-mutated Cec A1 promoter to yield the Cecmut-GFP construct. To yield the Drsmut-GFP construct, mutations of Cad binding sites (S1 to S4; see Fig. Fig.2)2) on the Drs promoter construct were created by site-directed mutagenesis by using the Drs-GFP plasmid PJM802. To create the Cad repressor construct (pUAST-Cad-En) (19), the Drosophila En repressor domain (amino acids 1 to 296 of the Drosophila En protein) was amplified and inserted in front of the DNA binding domain (amino acids 273 to 427) of Cad to allow the production of an in-frame N-terminal fusion of the En repressor domain to the DNA binding domain of Cad. These constructs were then used to generate transgenic animals by P-element-mediated transformation (43). The construct was injected into w1118-expressing embryos.

FIG. 1.
Identification of Caudal protein DNA recognition elements in the Cec promoter. (A) Schematic structures of the transfected reporter plasmids are shown on the left. T bars, standard deviations (SD) of at least five independent experiments. Normalized luciferase ...
FIG. 2.
Identification of Caudal protein DNA recognition elements in the Drs promoter. (A) Schematic diagram of the various Drs deletion constructs used in this study. The expression levels of the various Drs mutant constructs in the Schneider cells overexpressing ...

RESULTS

Identification of CDREs in Cec and Drs promoters.

We have been interested in novel transcription factors involved in the innate immune response, and we performed the in silico identification of putative genomic binding sites of AMP genes and their transcription factors by using the MatInd and MatInspector systems (41). In this analysis, we found several cis elements (such as the κB motif, the GATA motif, and Cad binding motifs) commonly found in the promoter regions of all known AMPs. Transcription factors resulting from this analysis were systematically tested for their capacity to induce AMP genes in the immunocompetent Schneider cell line SL2 (13). In Schneider cells stably expressing Cad, the expression of all seven AMP genes was greatly enhanced, suggesting that Cad is a potential transcription regulator (data not shown). This result prompted our in-depth investigation into the in vivo role of Cad using two representative AMP genes (IMD pathway-controlled Cec and Toll pathway-controlled Drs). Because the Cad gene product contains a homeodomain, which indicates that the protein has a DNA-binding capability, we examined the cis elements responsible for Cad-induced Cec and Drs expression. To identify the cis elements responsible for Cad-induced Cec and Drs expression, we performed a luciferase reporter assay of various mutant constructs having deletions in the Cec promoter region in Drosophila Schneider cells. Cad-induced luciferase activity in cells transfected with the plasmid with a deletion from −751 to −484 bp was found to be almost invariant compared with that in cells transfected with the wild-type construct (Fig. (Fig.1A)1A) . However, luciferase activity remained at the basal level in cells transfected with the plasmid having a deletion from bp −751 to −377 (Fig. (Fig.1A).1A). These results suggest that the region from bp −484 to −377 of the Cec promoter is a candidate region for Cad-protein DNA recognition elements (CDREs). For Drs, we could also identify the region covering bp −1082 to −1008 as a candidate region for CDREs of Drs (Fig. (Fig.2A)2A) . Based on these results, we identified six putative binding sites (S1 to S6) with the consensus Cad binding motif (T>C/A>G)TTT(A>G>C)(T>G/A/C)(G>T/C/A)(A>G/T/C) (2) in the promoter region of Cec and Drs (Fig. (Fig.1B1B and and2B).2B). To determine whether Cad possesses a DNA binding capability with these putative binding sites of the Cec promoter, we performed DNA-binding experiments with the recombinant Cad protein using wild-type probes and various mutant probes (Fig. (Fig.1B).1B). The results showed that GST-Cad is able to bind to two Cad binding motifs, at the S2 and S5 sites. To further confirm this Cad binding activity, we performed an EMSA using a nuclear extract from Cad-expressing transgenic flies. Using CDREs as probes, we observed a faint nuclear Cad binding activity in the nuclear extract from control transgenic flies (UAS-Cad; Heat Shock [HS]-GAL4, without heat shock), which was greatly enhanced in the nuclear extract of Cad-expressing transgenic flies (UAS-Cad; HS-GAL4, with heat shock) (Fig. (Fig.1B).1B). Furthermore, an immune serum directed against Cad supershifted the protein-DNA complex, whereas the addition of preimmune serum did not result in the formation of a CDRE-protein-antibody complex (Fig. (Fig.1B).1B). Luciferase reporter analysis with a plasmid carrying double mutations in the putative binding sites (S2 and S5) revealed that these sites are essential for Cad-mediated Cec promoter regulation (Fig. (Fig.1C).1C). We also employed similar methods to identify the CDREs for Cad-mediated Drs promoter regulation. The luciferase reporter assay with plasmids carrying point mutations in the putative CDREs together with the EMSA and supershift assay revealed that Cad is capable of directly regulating the expression of Drs via four CDREs (S1 to S4) found in its promoter (Fig. 2B and C). These results demonstrate the involvement of Cad in the regulation of AMP genes, providing yet another function for this homeotic transcriptional regulator, well known for its key role in anteroposterior patterning of the embryo (8, 35, 36).

Mutations affecting CDREs do not abolish inducible systemic AMP expression.

The above results were obtained from in vitro-cultured cells. To analyze the contribution of CDREs to AMP gene expression in vivo, we generated GFP reporter-expressing transgenic flies carrying the Drs and Cec promoter, in which the CDREs (at the S1 to S4 sites for the Drs promoter and at the S2 and S5 sites for the Cec promoter) were mutated (Drsmut-GFP and Cecmut-GFP, respectively). We compared these reporter-expressing transgenic flies with transgenic flies carrying wild-type promoters: the Drs promoter fused to GFP (Drs-GFP) and the Cec A1 promoter fused to GFP (Cec-GFP). To investigate whether CDREs are involved in the systemic expression of AMP genes after septic injury, Drs-GFP- and Drsmut-GFP-expressing flies were pricked with a bacterium-soaked needle. The result showed that mutation in the CDREs does not affect the systemic immune response: a strong diffuse fat body-derived fluorescence was observed in both lines of transgenic flies (Fig. (Fig.3A).3A). No difference in fluorescence intensity between transgenic flies carrying Drs-GFP and those carrying Drsmut-GFP was observed. Similar results were obtained with both Cec-GFP- and Cecmut-GFP-expressing transgenic flies (data not shown). These results clearly indicate that the CDREs, in contrast to κB sites, are not required for the inducible expression of these genes.

FIG. 3.
Caudal protein DNA recognition elements are required for the constitutive local expression of Drs. (A) Cad protein DNA recognition elements are not involved in inducible systemic innate immunity. Transgenic flies carrying the wild-type Drs promoter fused ...

Mutations affecting CDREs abolish constitutive local AMP expression.

Previous studies have shown that, in addition to systemic expression of AMPs by the fat body, several epithelia can express AMPs (local expression of AMP genes) (5, 10, 38, 51). As occurs in the case of vertebrate epithelia, insect epithelial tissue specifically produces various AMPs that help maintain a steady state of natural microflora (5, 10, 16, 24, 38, 51). Some epithelial tissues constitutively express AMP genes even in the absence of infection. For Drs, such constitutive local Drs expression is mainly detected in the salivary glands and in the female reproductive organs (10, 51). Interestingly, Drs expression in these epithelial tissues is not dependent on the known NF-κB pathways (Toll and IMD pathways) (10, 51). As CDREs are not involved NF-κB-dependent inducible AMP expression, we questioned whether CDREs are involved in constitutive local Drs expression in these epithelia. Our results concerning in vivo Cad expression using Cad-LacZ-expressing transgenic flies showed that a high level of expression of LacZ is present in various Drs-expressing epithelial tissues, including the salivary glands and the spermathecas and seminal receptacles (Fig. (Fig.3B3B and and3C).3C). We also checked the levels of endogenous Cad expression in salivary glands and spermathecas and seminal receptacles with regard to that in the intestine. Real-time PCR analysis showed that the expression levels of the salivary glands and of the spermathecas and seminal receptacles reached 88 and 22% of the intestinal Cad mRNA level, respectively (data not shown). This result and the existence of CDREs in the Drs promoter prompted us to further investigate whether CDREs are implicated in the constitutive expression of Drs in these tissues. The result showed that, in contrast to Drs expression in the salivary glands of transgenic flies carrying Drs-GFP, Drs expression in the salivary glands in our transgenic flies (12 independent transgenic lines) carrying Drsmut-GFP was almost completely absent (Fig. (Fig.3D).3D). However, strong constitutive Drs expression in the female reproductive organs in the Drsmut-GFP-expressing flies and also in the Drsmut-GFP-expressing flies under a Relish−/− genetic background was not affected (Fig. (Fig.3E).3E). This reporter analysis using GFP-expressing transgenic flies clearly showed that the CDREs are absolutely necessary for the constitutive expression of Drs in the salivary glands, but not in other Drs-expressing tissues, such as female reproductive organs.

To determine whether CDREs are implicated in the local constitutive expression of AMP genes other than Drs, we tested their role in Cec expression. In Drosophila, the strongest local constitutive expression of Cec is observed in the male ejaculatory duct in unchallenged adults (51). The inducible local expression of Cec in midgut, proventriculus, and Malpighian tubules has been shown to be under the control of the IMD pathway-activated p105-like NF-κB, Relish (38, 51). However, the involvement of the IMD-Relish pathway in the expression of Cec in the ejaculatory duct is unknown. Therefore, we first investigated whether the high constitutive expression of Cec in the male ejaculatory duct is controlled by Relish. We analyzed the expression of Cec by using transgenic flies carrying Cec-GFP in a Relish−/− genetic background. The result showed the .constitutive expression of Cec-GFP in the male ejaculatory duct was not significantly affected in Relish mutant flies (Fig. (Fig.4A).4A). This result suggested that the Cec expression in the ejaculatory duct is constitutive and that a transcription factor other than Relish is involved in this organ. To see whether Cad is normally expressed in the ejaculatory duct, we first examined in vivo Cad expression. The result showed that the Cad reporter is detected in this tissue by using transgenic flies carrying Cad-LacZ (Fig. (Fig.4B).4B). We further checked the level of endogenous Cad expression in the ejaculatory duct with regard to that of the intestine. Real-time PCR analysis showed that endogenous Cad expression in the ejaculatory duct reached 45% of intestinal Cad mRNA level (data not shown).

FIG. 4.
Caudal protein DNA recognition elements are required for the constitutive local expression of Cec. (A) Relish is not required for Cec expression in the ejaculatory duct. Shown is constitutive Cec-GFP reporter activity in the ejaculatory ducts of the wild ...

To investigate whether the CDREs found in the Cec promoter are essential for the high constitutive expression level of Cec in the male ejaculatory duct, we examined Cec expression in transgenic flies carrying Cec-GFP and Cecmut-GFP. Male transgenic flies carrying Cec-GFP exhibited strong constitutive Cec expression in the abdominal region, whereas Cec expression was severely impaired in flies carrying Cecmut-GFP (Fig. (Fig.4C).4C). Upon dissection, we observed that the Cec reporter activities in the ejaculatory ducts of male transgenic flies carrying Cecmut-GFP were significantly reduced (Fig. (Fig.4D).4D). Quantitative analysis of GFP reporter expression showed that transgenic flies carrying Cecmut-GFP gained only 20% of the reporter activity in the ejaculatory ducts gained with transgenic flies carrying Cec-GFP (Fig. (Fig.4E).4E). All six independent transgenic fly lines carrying Cecmut-GFP displayed reduced GFP reporter activity, showing that CDREs are needed for full activation of the Cec promoter. However, we still detected a low level of residual GFP expression in the ejaculatory ducts of flies carrying Cecmut-GFP, which suggested that other regulatory elements may also be involved in AMP signaling in this tissue.

Altogether our results demonstrate that CDREs are obligatory for the constitutive local expression of Cec and Drs in a subset of epithelial tissues where the expression of these genes is NF-κB independent.

Cad regulates the constitutive expression of Cec and Drs in a subset of epithelial tissues.

The results in the previous section suggest that Cad is involved in the regulation of constitutive local expression of Cec and Drs through CDREs. To test this hypothesis, we first examined whether the constitutive local expression of Cec and Drs is affected in a heterozygote Cad mutant (30), because homozygous expression of mutant Cad is embryonically lethal. We analyzed the expression of these genes by using transgenic flies carrying Cec-GFP or Drs-GFP in a Cad+/− genetic background. The result showed that the constitutive expression of Cec-GFP in the male ejaculatory duct and Drs-GFP expression in the salivary glands in heterozygote Cad mutant flies were not significantly affected (Fig. (Fig.5A).5A). We then performed RNA interference (RNAi) experiment by generating transgenic flies carrying the UAS-Cad-RNAi construct in order to mimic the loss-of-function mutation. Using this method, we achieved a partial decrease of Cad activity in the ejaculatory duct after introducing Cad-RNAi using a line ubiquitously expressing Daughterless (Da)-GAL4. Real-time PCR analysis showed an endogenous Cad mRNA reduction of 37% and an endogenous Cec mRNA reduction of 40% (Fig. (Fig.5B).5B). No Cad-RNAi effect was observed by using a control female fat body-specific Yolk-GAL4 driver (11), which is not expressed in the ejaculatory duct (Fig. (Fig.5B).5B). To see whether the general physiological functions of the ejaculatory duct were affected by the expression of Cad-RNAi, the expression of the ejaculatory duct-specific gene Gld (encoding glucose dehydrogenase) was examined (46). The result showed that the expression of Gld in the ejaculatory ducts of flies carrying UAS-Cad-RNAi and Da-GAL4 was not affected (Fig. (Fig.5B).5B). As Cad is known as a developmental gene, we set out to compare levels of expression of Cec and Cad in the male reproductive organs of two different developmental stages (larvae and adults). The result showed that a similar levels of Cad mRNA were detected in both the larval genital disk and the adult ejaculatory duct (Fig. (Fig.5C).5C). However, the level of Cec mRNA is much more higher in the adult ejaculatory duct than in the larval genital disk (Fig. (Fig.5C).5C). To completely inhibit endogenous Cad activity in the adult ejaculatory duct, we generated a dominant-negative construct of Cad by using the domain-swapping method. Accordingly, we removed the transcriptional activation domain of Cad and replaced it with the strong transcriptional repressor domain of the Drosophila engrailed (En) protein (Cad-En) (19). When this Cad repressor construct was cotransfected with Cad, Cad-induced Cec reporter activity in Drosophila immunocompetent Schneider cells was completely inhibited, confirming the dominant-negative effect of the Cad-En fusion protein (Fig. (Fig.5D).5D). To inhibit endogenous Cad activity in vivo, we generated transgenic flies carrying UAS-Cad-En. These transgenic flies were crossed with flies carrying the HS-GAL4 driver in order to inhibit endogenous Cad activity, in the presence of a Cec-GFP insertion. As was confirmed by using flies carrying UAS-EGFP, the adult flies carrying HS-GAL4 expressed GAL4 ubiquitously, including expression in the ejaculatory duct after heat shock treatment (data not shown). The result showed that strong expression of the Cec-GFP reporter in the ejaculatory ducts of Cec-GFP/UAS-Cad-En; HS-GAL4/+-expressing flies was severely reduced when the flies were subjected to heat shock treatment, compared with that in control flies (untreated Cec-GFP/UAS-Cad-En; HS-GAL4/+-expressing flies or treated Cec-GFP/+; HS-GAL4/+-expressing flies) (Fig. (Fig.5E).5E). These findings agree with the results for Cecmut-GFP-expressing flies presented in the previous sections and demonstrate that Cad activity is important for constitutive Cec expression in the ejaculatory duct in a Relish-independent manner.

FIG. 5.
Caudal regulates the constitutive expression of Cec through Caudal protein DNA recognition elements. (A) Constitutive Cec and Drs expression is not affected in a heterozygote Cad mutant. Constitutive Cec-GFP reporter activity in the ejaculatory ducts ...

Previous studies demonstrated that constitutive Drs-GFP expression is mainly observed in the adult salivary glands (10, 51). When we compared levels of expression of Drs and Cad in the salivary glands of flies at different developmental stages, we found that Drs and Cad expression in the adult salivary glands is higher than that in the larvae and prepupae (Fig. (Fig.6A).6A). Thus, we wanted to confirm the role of Cad in Drs expression in the adult salivary glands. When Cad-RNAi was introduced in lines expressing GAL4 in the salivary glands (UAS-Cad-RNAi combined with the c729-GAL4 and the Drs-GFP reporter gene), expression of the Drs-GFP reporter in the salivary glands of these flies overexpressing the Cad-RNAi construct was nearly abolished (Fig. (Fig.6B).6B). The control flies carrying c729-GAL4 alone showed a normal Drs-GFP activity in this tissue (Fig. (Fig.6B).6B). Similar results were obtained with transgenic flies carrying UAS-Cad-RNAi and with other GAL4-expressing lines (such as lines carrying Da-GAL4 or HS-GAL4) (data not shown). In this condition, consistent with the result for transgenic flies carrying Drsmut-GFP (Fig. (Fig.3E),3E), Cad inhibition by RNAi does not have an effect on Drs expression in the female reproductive organs (Fig. (Fig.6B).6B). Real-time PCR analysis showed that, in the salivary glands of Cad-RNAi-expressing flies, endogenous Drs transcripts reached ~25% of the control level and that endogenous Cad transcripts reached ~35% of the control level (Fig. (Fig.6C).6C). No Cad-RNAi effect was observed by using a control female fat body-specific Yolk-GAL4 driver (Fig. (Fig.6C).6C). To see whether the condition of Cad-RNAi could influence the expression of other genes unrelated to innate immunity, we examined the expression of B52 (21), which is known to be expressed in the salivary glands and some other tissues. The result showed that the expression of B52 in the salivary glands of flies carrying UAS-Cad-RNAi and Da-GAL4 was not affected (Fig. (Fig.6C).6C). Altogether, these results demonstrate that full Cad activity is essential for constitutive Drs expression in the local epithelial tissues such as the salivary glands.

FIG. 6.
Caudal regulates the constitutive expression of Drs through Caudal protein DNA recognition elements. (A) Endogenous Drs and Cad expression in the salivary glands of different developmental stages. Quantitative real-time PCR analysis of Drs and Cad gene ...

DISCUSSION

The homeobox transcription factor Cad was originally found to regulate the anteroposterior body axis of Drosophila (8, 35, 36). During embryogenesis, the Cad expression level is tightly regulated in response to developmental signals. For example, a high level of Cad is needed in posterior structures to activate the segmentation genes fushi tarazu and spalt, which are involved in terminal specification (8, 22), whereas the development of the anterior part of the embryo is associated with a low Cad expression level (8, 22, 35). During postembryogenesis, Cad expression is primarily restricted to the intestine and to the Malpighian tubules and gonads (35). Cad expression in postembryonic life is known to be restricted to organs that display cell renewal or remodeling, such as the intestine (35). Our results show that the Cad-LacZ reporter (Fig. (Fig.3B3B and and4B)4B) and endogenous Cad mRNA (Fig. (Fig.55 and and6)6) are also expressed in the salivary glands and ejaculatory duct, where AMP expression is constitutive. Vertebrate Cad homologues are well known to participate in early embryogenesis, the development of the intestine, and colon tumorigenesis (7, 49). However, apart from their developmental roles, the physiological functions and target genes of the Cad homeobox gene family are unknown. The observation that Cad regulates AMP gene expression in a subset of epithelia indicates a new function for this trans-activator in the local defense against microbial infection and/or maintenance of microbial flora. At present, the real in vivo function of AMP gene expression in local epithelia in Drosophila is not known (10, 51). In the local-infection experiment, we could not observe the enhanced mortality in the Cad-RNAi-expressing flies following short-term (1 h) bacterial feeding (J.-H. Ryu and W.-J. Lee, unpublished data). However, although local AMP expression is not directly related to the rate of survival of infection, the locally secreted AMPs may help to prevent the onset of infections.

It is well known that the Toll/NF-κB signaling pathway for dorsoventral body axis formation mainly regulates the inducible expression of the Drs gene during the systemic immune response (17, 25). Interestingly, this pathway has been well conserved during evolution and assists NF-κB activation via Toll-like receptors in the human innate immune system (1, 18). Our results show that, in the local epithelial immune system, NF-κB-independent, constitutive expression of Drs and Cec in the barrier epithelial tissues is mainly controlled by the homeobox gene Cad, a master controller of anteroposterior body axis formation. The developmental genes involved in specification of the fly body plan (dorsoventral and anteroposterior body axes) have been recruited for this evolutionally ancient first line of defense. Our results together with those of others further demonstrate a link between development and immunity.

The involvement of Cad in the constitutive local innate immunity illustrates the complexity of the tissue-specific regulation of AMP expression in Drosophila. To better visualize the complexity and dynamic of the innate immune response in Drosophila, we constructed a comprehensive scheme (Fig. (Fig.7).7). Experimental infection such as septic injury rapidly induces various AMPs, mainly in the fat body (known as systemic immunity), via two different NF-κB pathways (Toll and IMD pathways), whereas natural infection, such as local bacterial infection, activates the expression of AMPs via the IMD pathway only in a subset of epithelial tissues (known as inducible local innate immunity) (10, 38, 51). These two inducible innate immune systems in Drosophila are rather distinct because septic injury cannot activate the inducible local immune system (51). The third type of AMP regulation is the constitutive local expression of AMPs in an NF-κB-independent manner in several epithelia (10, 51). This type of strategy is believed to be very ancient in evolution and may be very efficient in certain epithelia by avoiding chronic NF-κB activation where the contact with microbes is continuous.

FIG. 7.
Tissue-specific regulation of antimicrobial peptide genes in Drosophila. This model is based on previous studies (10, 38, 51) and the present study. Note that different types of infection can activate distinct innate immune pathways (systemic immunity ...

In this study, we show that that Cad is capable of directly regulating Cec and Drs via CDREs found in the their promoters in Drosophila Schneider cells (Fig. (Fig.11 and and2).2). Furthermore, Cad binds in vitro to the CDREs found upstream of AMP genes in a gel shift assay (Fig. (Fig.11 and and2).2). These results demonstrate that Cad is a direct trans-activator of AMP genes. The in vivo reporter analysis demonstrates that mutations affecting CDREs do not abolish the inducible systemic Drs expression in the fat body (Fig. (Fig.3A).3A). These results clearly indicate that the CDREs, in contrast to κB sites, are not required for inducible Drs expression in the fat body during a systemic immune response. In addition to the fat body, the trachea is involved in inducible Drs expression. This tissue, in which Drs expression is normally absent but rapidly induced in response to local infection by Erwinia carotovora, is known to be involved in inducible local immunity (51). Surprisingly, even though there is no appreciable role for CDREs in the fat body, we found that all 12 independent Drsmut-GFP-expressing fly lines (larvae and adults) exhibited spontaneous constitutive expression of Drs reporter activity in the trachea in the absence of local infection (J.-H. Ryu and W.-J. Lee, unpublished data). One may speculate that CDREs can also act as negative cis elements in some epithelial tissues such as the trachea, where they can maintain the silencing of Drs expression, and that this depends on the specific cell type. Further studies will be needed to understand the complete tissue-specific Cad signaling pathway for AMP regulation in all epithelial tissues. In contrast to κB-dependent inducible AMP expression, the constitutive local innate immunity employs Cad for the expression of AMPs through CDRE motifs rather than κB motifs (Fig. (Fig.33 and and4).4). Interestingly, for salivary glands, overexpression of the Cad-RNAi construct is sufficient to severely reduce Drs expression, indicating that constitutive local expression of Drs in salivary glands is greatly dependent on Cad (Fig. (Fig.6).6). For the ejaculatory duct, although partial reduction of Cad modestly reduces Cec expression (Fig. (Fig.5B),5B), we can detect only minor expression (~20%) of the Cec reporter in flies carrying Cecmut-GFP (Fig. 4D and E), as well as flies overexpressing the dominant-negative form of Cad (Fig. (Fig.5E).5E). This also indicates that Cec expression in this tissue is largely dependent on Cad. Interestingly, our study showed that not all constitutive local expression is dependent on Cad. The constitutive local expression in the female reproductive organs is completely CDRE independent (Fig. (Fig.3E3E and and6B),6B), suggesting the existence of yet another unknown signaling pathway(s). Recently, Drosophila Toll-9, one of the Toll-related receptors, was found to trigger the constitutive expression of Drs in cultured cells (39). It is possible that Toll-9 may control constitutive Drs expression in certain epithelia. The studies on the in vivo function of Toll-9 should elucidate this issue.

The expression of various AMPs in analogous human epithelial tissues suggests that epithelial innate immunity is well conserved and that the careful regulation of AMP levels may be needed to maintain homeostasis in these tissues from Drosophila to humans (24). The presence of human Cad homologues, CDXs, raises interesting questions concerning their putative role(s) in human epithelial innate immune gene regulation.

Acknowledgments

We thank Bruno Lemaitre, Dan Hultmark, and Ylva Engström for fly strains and plasmid constructs.

This work was supported by KOSEF (R2-2000-000-26-0, R2-2001-000-35-0, and R2-2002-000-58-0) (W.-J.L.) and the Center for Cell Signaling Research and 21C Frontier Functional Human Genome Project (F-4-09) (M.-A.Y. and W.-J.L.) from the Ministry of Science and Technology of Korea. J.-H. Ryu and K.-B. Nam were supported by a postdoctoral fellowship from the Brain Korea 21 project of the Korea Ministry of Education. P.T.B. received support from Institut Pasteur.

REFERENCES

1. Akira, S., K. Takeda, and T. Kaisho. 2001. Toll-like receptors: critical proteins linking innate and acquired immunity. Nat. Immunol. 2:675-680. [PubMed]
2. Berman, B. P., Y. Nibu, B. D. Pfeiffer, P. Tomancak, S. E. Celniker, M. Levine, G. M. Rubin, and M. B. Eisen. 2002. Exploiting transcription factor binding site clustering to identify cis-regulatory modules involved in pattern formation in the Drosophila genome. Proc. Natl. Acad. Sci. USA 99:757-762. [PMC free article] [PubMed]
3. Boutros, M., H. Agaisse, and N. Perrimon. 2002. Sequential activation of signaling pathways during innate immune responses in Drosophila. Dev. Cell 3:711-722. [PubMed]
4. Brand, A. H., and N. Perrimon. 1993. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118:401-415. [PubMed]
5. Brey, P. T., W. J. Lee, M. Yamakawa, Y. Koizumi, S. Perrot, M. Francois, and M. Ashida. 1993. Role of the integument in insect immunity: epicuticular abrasion and induction of cecropin synthesis in cuticular epithelial cells. Proc. Natl. Acad. Sci. USA 90:6275-6279. [PMC free article] [PubMed]
6. Choe, K. M., T. Werner, S. Stoven, D. Hultmark, and K. V. Anderson. 2002. Requirement for a peptidoglycan recognition protein (PGRP) in Relish activation and antibacterial immune responses in Drosophila. Science 296:359-362. [PubMed]
7. Clatworthy, J. P., and V. Subramanian. 2001. Stem cells and the regulation of proliferation, differentiation and patterning in the intestinal epithelium: emerging insights from gene expression patterns, transgenic and gene ablation studies. Mech. Dev. 101:3-9. [PubMed]
8. Dearolf, C. R., J. Topol, and C. S. Parker. 1989. The caudal gene product is a direct activator of fushi tarazu transcription during Drosophila embryogenesis. Nature 341:340-343. [PubMed]
9. Di Nocera, P. P., and I. B. Dawid. 1983. Transient expression of genes introduced into cultured cells of Drosophila. Proc. Natl. Acad. Sci. USA 80:7095-7098. [PMC free article] [PubMed]
10. Ferrandon, D., A. C. Jung, M. Criqui, B. Lemaitre, S. Uttenweiler-Joseph, L. Michaut, J. Reichhart, and J. A. Hoffmann. 1998. A drosomycin-GFP reporter transgene reveals a local immune response in Drosophila that is not dependent on the Toll pathway. EMBO J. 17:1217-1227. [PMC free article] [PubMed]
11. Georgel, P., S. Naitza, C. Kappler, D. Ferrandon, D. Zachary, C. Swimmer, C. Kopczynski, G. Duyk, J. M. Reichhart, and J. A. Hoffmann. 2001. Drosophila immune deficiency (IMD) is a death domain protein that activates antibacterial defense and can promote apoptosis. Dev. Cell 1:503-514. [PubMed]
12. Gottar, M., V. Gobert, T. Michel, M. Belvin, G. Duyk, J. A. Hoffmann, D. Ferrandon, and J. Royet. 2002. The Drosophila immune response against gram-negative bacteria is mediated by a peptidoglycan recognition protein. Nature 416:640-644. [PubMed]
13. Han, S.-J., K.-Y. Choi, P. T. Brey, and W.-J. Lee. 1998. Molecular cloning and characterization of a Drosophila p38 mitogen-activated protein kinase. J. Biol. Chem. 273:369-374. [PubMed]
14. Hedengren, M., B. Asling, M. S. Dushay, I. Ando, S. Ekengren, M. Wihlborg, and D. Hultmark. 1999. Relish, a central factor in the control of humoral but not cellular immunity in Drosophila. Mol. Cell 4:827-837. [PubMed]
15. Hoffmann, J. A., and J. M. Reichhart. 2002. Drosophila innate immunity: an evolutionary perspective. Nat. Immunol. 3:121-126. [PubMed]
16. Huttner, K. M., and C. L. Bevins. 1999. Antimicrobial peptides as mediators of epithelial host defense. Pediatr. Res. 45:785-794. [PubMed]
17. Imler, J. L., and J. A. Hoffmann. 2000. Signaling mechanisms in the antimicrobial host defense of Drosophila. Curr. Opin. Microbiol. 3:16-22. [PubMed]
18. Ip, Y. T., M. Reach, Y. Engstrom, L. Kadalayil, H. Cai, S. Gonzalez-Crespo, K. Tatei, and M. Levine. 1993. Dif, a dorsal-related gene that mediates an immune response in Drosophila. Cell 75:753-763. [PubMed]
19. Isaacs, H. V., M. E. Pownall, and J. M. Slack. 1998. Regulation of Hox gene expression and posterior development by the Xenopus caudal homologue Xcad3. EMBO J. 17:3413-3427. [PMC free article] [PubMed]
20. Kim, Y.-S., S.-J. Han, J.-H. Ryu, K.-H. Choi, Y.-S. Hong, Y.-H. Chung, S. Perrot, A. Raibaud, P. T. Brey, and W.-J. Lee. 2000. Lipopolysaccharide-activated kinase, an essential component for the induction of the antimicrobial peptide genes in Drosophila melanogaster cells. J. Biol. Chem. 275:2071-2079. [PubMed]
21. Kraus, M. E., and J. T. Lis. 1994. The concentration of B52, an essential splicing factor and regulator of splice site choice in vitro, is critical for Drosophila development. Mol. Cell. Biol. 14:5360-5370. [PMC free article] [PubMed]
22. Kuhnlein, R. P., G. Bronner, H. Taubert, and R. Schuh. 1997. Regulation of Drosophila spalt gene expression. Mech. Dev. 66:107-118. [PubMed]
23. Kwon, E. J., H. S. Park, Y. S. Kim, E. J. Oh, Y. Nishida, A. Matsukage, M. A. Yoo, and M. Yamaguchi. 2000. Transcriptional regulation of the Drosophila raf proto-oncogene by Drosophila STAT during development and in immune response. J. Biol. Chem. 275:19824-19830. [PubMed]
24. Lehrer, R. I., and T. Ganz. 1999. Antimicrobial peptides in mammalian and insect host defence. Curr. Opin. Immunol. 11:23-27. [PubMed]
25. Lemaitre, B., E. Nicolas, L. Michaut, J. M. Reichhart, and J. A. Hoffmann. 1996. The dorsoventral regulatory gene cassette spatzle/Toll/cactus controls the potent antifungal response in Drosophila adults. Cell 86:973-983. [PubMed]
26. Leulier, F., C. Parquet, S. Pili-Floury, J. H. Ryu, M. Caroff, W. J. Lee, D. Mengin-Lecreulx, and B. Lemaitre. 2003. The Drosophila immune system detects bacteria through specific peptidoglycan recognition. Nat. Immunol. 4:478-484. [PubMed]
27. Leulier, F., A. Rodriguez, R. S. Khush, J. M. Abrams, and B. Lemaitre. 2000. The Drosophila caspase Dredd is required to resist gram-negative bacterial infection. EMBO Rep. 1:353-358. [PMC free article] [PubMed]
28. Leulier, F., S. Vidal, K. Saigo, R. Ueda, and B. Lemaitre. 2002. Inducible expression of double-stranded RNA reveals a role for dFADD in the regulation of the antibacterial response in Drosophila adults. Curr. Biol. 12:996-1000. [PubMed]
29. Ligoxygakis, P., N. Pelte, J. A. Hoffmann, and J. M. Reichhart. 2002. Activation of Drosophila Toll during fungal infection by a blood serine protease. Science 297:114-116. [PubMed]
30. Liu, S., and J. Jack. 1992. Regulatory interactions and role in cell type specification of the Malpighian tubules by the cut, Kruppel, and caudal genes of Drosophila. Dev. Biol. 150:133-143. [PubMed]
31. Lu, Y., L. P. Wu, and K. V. Anderson. 2001. The antibacterial arm of the Drosophila innate immune response requires an IκB kinase. Genes Dev. 15:104-110. [PMC free article] [PubMed]
32. Medzhitov, R., and C. A. Janeway, Jr. 1997. Innate immunity: the virtues of a nonclonal system of recognition. Cell 91:295-298. [PubMed]
33. Meng, X., B. S. Khanuja, and Y. T. Ip. 1999. Toll receptor-mediated Drosophila immune response requires Dif, an NF-κB factor. Genes Dev. 13:792-797. [PMC free article] [PubMed]
34. Michel, T., J. M. Reichhart, J. A. Hoffmann, and J. Royet. 2001. Drosophila Toll is activated by gram-positive bacteria through a circulating peptidoglycan recognition protein. Nature 414:756-759. [PubMed]
35. Mlodzik, M., and W. J. Gehring. 1987. Expression of the caudal gene in the germ line of Drosophila: formation of an RNA and protein gradient during early embryogenesis. Cell 48:465-478. [PubMed]
36. Moreno, E., and G. Morata. 1999. Caudal is the Hox gene that specifies the most posterior Drosophila segment. Nature 400:873-877. [PubMed]
37. Naitza, S., C. Rosse, C. Kappler, P. Georgel, M. Belvin, D. Gubb, J. Camonis, J. A. Hoffmann, and J. M. Reichhart. 2002. The Drosophila immune defense against gram-negative infection requires the death protein dFADD. Immunity 17:575-581. [PubMed]
38. Onfelt Tingvall, T., E. Roos, and Y. Engstrom. 2001. The imd gene is required for local cecropin expression in Drosophila barrier epithelia. EMBO Rep. 2:239-243. [PMC free article] [PubMed]
39. Ooi, J. Y., Y. Yagi, X. Hu, and Y. T. Ip. 2002. The Drosophila Toll-9 activates a constitutive antimicrobial defense. EMBO Rep. 3:82-87. [PMC free article] [PubMed]
40. Petersen, U. M., L. Kadalayil, K. P. Rehorn, D. K. Hoshizaki, R. Reuter, and Y. Engstrom. 1999. Serpent regulates Drosophila immunity genes in the larval fat body through an essential GATA motif. EMBO J. 18:4013-4022. [PMC free article] [PubMed]
41. Quandt, K., K. Frech, H. Karas, E. Wingender, and T. Werner. 1995. MatInd and MatInspector: new fast and versatile tools for detection of consensus matches in nucleotide sequence data. Nucleic Acids Res. 23:4878-4884. [PMC free article] [PubMed]
42. Ramet, M., P. Manfruelli, A. Pearson, B. Mathey-Prevot, and R. A. Ezekowitz. 2002. Functional genomic analysis of phagocytosis and identification of a Drosophila receptor for E. coli. Nature 416:644-648. [PubMed]
43. Rubin, G. M., and A. C. Spradling. 1982. Genetic transformation of Drosophila with transposable element vectors. Science 218:348-353. [PubMed]
44. Rutschmann, S., A. C. Jung, C. Hetru, J. M. Reichhart, J. A. Hoffmann, and D. Ferrandon. 2000. The Rel protein DIF mediates the antifungal but not the antibacterial host defense in Drosophila. Immunity 12:569-580. [PubMed]
45. Rutschmann, S., A. C. Jung, R. Zhou, N. Silverman, J. A. Hoffmann, and D. Ferrandon. 2000. Role of Drosophila IKK gamma in a toll-independent antibacterial immune response. Nat. Immunol. 1:342-347. [PubMed]
46. Schiff, N. M., Y. Feng, J. A. Quine, P. A. Krasney, and D. R. Cavener. 1992. Evolution of the expression of the Gld gene in the reproductive tract of Drosophila. Mol. Biol. Evol. 9:1029-1049. [PubMed]
47. Silverman, N., R. Zhou, S. Stoven, N. Pandey, D. Hultmark, and T. Maniatis. 2000. A Drosophila IκB kinase complex required for Relish cleavage and antibacterial immunity. Genes Dev. 14:2461-2471. [PMC free article] [PubMed]
48. Tingvall, T. O., E. Roos, and Y. Engstrom. 2001. The GATA factor Serpent is required for the onset of the humoral immune response in Drosophila embryos. Proc. Natl. Acad. Sci. USA 98:3884-3888. [PMC free article] [PubMed]
49. Traber, P. G. 1999. Transcriptional regulation in intestinal development. Implications for colorectal cancer. Adv. Exp. Med. Biol. 470:1-14. [PubMed]
50. Tzou, P., E. De Gregorio, and B. Lemaitre. 2002. How Drosophila combats microbial infection: a model to study innate immunity and host-pathogen interactions. Curr. Opin. Microbiol. 5:102-110. [PubMed]
51. Tzou, P., S. Ohresser, D. Ferrandon, M. Capovilla, J. M. Reichhart, B. Lemaitre, J. A. Hoffmann, and J. L. Imler. 2000. Tissue-specific inducible expression of antimicrobial peptide genes in Drosophila surface epithelia. Immunity 13:737-748. [PubMed]
52. Vidal, S., R. S. Khush, F. Leulier, P. Tzou, M. Nakamura, and B. Lemaitre. 2001. Mutations in the Drosophila dTAK1 gene reveal a conserved function for MAPKKKs in the control of rel/NF-κB-dependent innate immune responses. Genes Dev. 15:1900-1912. [PMC free article] [PubMed]
53. Young, D., T. Hussell, and G. Dougan. 2002. Chronic bacterial infections: living with unwanted guests. Nat. Immunol. 3:1026-1032. [PubMed]

Articles from Molecular and Cellular Biology are provided here courtesy of American Society for Microbiology (ASM)
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...