Skip to main content
NCBI home page
Infection and Immunity logoLink to Infection and Immunity
. 1998 May;66(5):2245–2255. doi: 10.1128/iai.66.5.2245-2255.1998

The Legionella pneumophila icmGCDJBF Genes Are Required for Killing of Human Macrophages

Mary Purcell 1, Howard A Shuman 1,*
PMCID: PMC108188  PMID: 9573114

Abstract

Previously, a collection of mutants of Legionella pneumophila that had lost the ability to multiply within and kill human macrophages was generated by Tn903dIIlacZ transposon mutagenesis and classified into DNA hybridization groups. A subset of these mutants was complemented by a plasmid, pMW100, containing a 13.5-kb genomic DNA insert. This plasmid restored the ability to multiply within and produce cytopathic effects on human macrophages to members of DNA hybridization groups II, IV, VI, and XVII. A region of the genomic insert of pMW100 was sequenced, and eight potential genes were identified and named icmE, icmG, icmC, icmD, icmJ, icmB, icmF, and tphA. None of the genes encode potential protein products with significant homology to previously characterized proteins, except for tphA, whose product has significant homology to a family of metabolite/H+ symport proteins from gram-negative bacteria. The positions of the Tn903dIIlacZ insertions within the genes were determined by nucleotide sequencing. No Tn903dIIlacZ insertions mapped to icmG, icmJ, or tphA; therefore, these loci were mutated to test whether they were required for macrophage killing. Complementation analysis was used to evaluate the roles of the potential gene products and provide information on the organization of transcriptional units within the region. The results indicate that all identified open reading frames except tphA are required for killing of human macrophages.

Legionella pneumophila is a gram-negative, broad-host-range, facultative intracellular bacterium that is capable of infecting, multiplying within, and killing human monocytes and alveolar macrophages (16). Inhalation of L. pneumophila can cause either a severe pneumonia called Legionnaires’ disease or a milder, self-limiting infection called Pontiac fever (11, 19). The bacteria bind to the surface of the host cell and are internalized by coiling phagocytosis (15). Once internalized, the bacteria reside within a phagosomal compartment that does not acidify and does not fuse with host lysosomes (13, 14). Mitochondria, smooth vesicles, and rough endoplasmic reticulum are recruited to the periphery of this internal compartment, and the bacteria are able to multiply within this Legionella-specific phagosome (13, 31). The bacteria destroy the host cell and are able to infect neighboring cells to initiate multiple rounds of infection.

The genes that encode and/or regulate this complex intracellular infection pathway are not well characterized. Few L. pneumophila genes have been identified that are specifically required for intracellular growth within and killing of human macrophages. The mip gene encodes a 24-kDa surface protein that has peptidyl-prolyl-cis/trans isomerase activity and is homologous to FK506 binding proteins (10). The Mip protein has been shown to be responsible for the efficient initiation of intracellular infection of macrophages, and mip mutants are less virulent than wild-type L. pneumophila but retain the ability to replicate within and kill host cells (5, 6). The dotA gene encodes a 1,048-amino-acid inner membrane protein (2, 27). Strains with mutations in dotA fail to inhibit phagosome-lysosome fusion and also fail to recruit host cell organelles to the periphery of the Legionella-specific phagosome (1, 2). These mutants are also unable to multiply in macrophages. Mutations in the genes of the icmA locus were shown to be completely defective for intracellular multiplication and host cell killing (4). Although the genes described above are important for L. pneumophila intracellular infection, their precise function is not known.

Previously, a collection of mutants defective for the ability to kill HL-60-derived macrophages was generated by Tn903dIIlacZ mutagenesis of L. pneumophila (28). These Mak− mutants were grouped into 16 DNA hybridization groups based on the location of the Tn903dIIlacZ insertions on EcoRI fragments in the mutant chromosomes. Mutants within group I were complemented for macrophage killing and intracellular multiplication by the previously isolated icmA/dotA locus. More recently, mutants within group III were complemented by a region that contains the icmT, icmS, icmR, icmQ, icmP, and icmO genes for macrophage killing and intracellular multiplication (30). Also, a mutant within group IX was complemented for macrophage killing by the icmE gene (29).

The large number of Mak− DNA hybridization groups indicated that several genes might be required to kill macrophages. To isolate and characterize these genes, we used a plaque assay to identify regions of DNA that would restore to the Mak− mutants the ability to replicate within and kill host cells. We identified a locus from the wild-type L. pneumophila chromosome that complements Mak− mutants from DNA hybridization groups II, IV, and VI and two previously ungrouped mutants. The nucleotide sequence of this locus was determined, and the translated open reading frames (ORFs) were found to have no homology to previously characterized proteins, except for one that was found to have homology to transport proteins from gram-negative bacteria. The positions of the Tn903dIIlacZ insertions within the ORFs were mapped. Genetic complementation analysis was used to determine which genes are responsible for macrophage killing and to organize the region into potential transcriptional units.

MATERIALS AND METHODS

Media and reagents.

L. pneumophila was grown in AYE broth (16a) and on ABCYE agar plates (9a). For electroporation, bacteria were grown in AYE medium lacking bovine serum albumin. Escherichia coli was grown in Luria-Bertani (LB) broth or on LB agar plates as described previously (24). All reagents and chemicals were obtained from Fisher Scientific (Springfield, N.J.). Fetal calf serum was obtained from Sigma-Aldrich Corp. (St. Louis, Mo.). RPMI 1640 medium was purchased from JRH Biosciences (Lenexa, Kans.). Cellgro l-glutamine (Gln) was purchased from Mediatech Inc. (Herndon, Va.). Normal human serum (NHS) was obtained from healthy volunteers. Bacto Agar was purchased from Difco Laboratories, Detroit, Mich. Restriction enzymes and Vent polymerase were supplied by New England Biolabs, Inc. (Beverly, Mass.). Taq polymerase was supplied by Perkin-Elmer Corp. (Foster City, Calif.). T4 polynucleotide kinase was supplied by USB Specialty Biochemicals (Cleveland, Ohio). Antibiotics for L. pneumophila selection were used at the following concentrations: kanamycin, 50 μg/ml; streptomycin, 50 μg/ml; chloramphenicol, 5 μg/ml; gentamicin, 5 μg/ml. Antibiotics for E. coli selection were used at the following concentrations: kanamycin, 50 μg/ml; ampicillin, 100 μg/ml; chloramphenicol, 25 μg/ml; gentamicin, 5 μg/ml.

Bacterial strains, plasmids, bacterial mating of plasmids, and DNA manipulations.

The bacterial strains and plasmids used in this work are described in Tables 1 and 2, respectively. E. coli DH5α and XL1-Blue were used for propagation of plasmids. Bacterial mating of plasmids was performed as previously described (28). Isolation of plasmid DNA, chromosomal DNA preparation, DNA cloning techniques, and Southern analysis were performed as previously described (21).

TABLE 1.

Bacterial strains used in this study

Strain Genotype and features Reference or source
L. pneumophila
 JR32 NaCl-sensitive isolate of AM511 28
 LELA1205 JR32 icmD1205::Tn903dIIlacZ, DNA hybridization group II 28
 LELA1506 JR32 icmB1506::Tn903dIIlacZ, DNA hybridization group II 28
 LELA1984 JR32 icmE1984::Tn903dIIlacZ, DNA hybridization group II 28
 LELA1996 JR32 icmB1996::Tn903dIIlacZ, DNA hybridization group II 28
 LELA2517 JR32 icmC2517::Tn903dIIlacZ, DNA hybridization group II 28
 LELA3244 JR32 icmD3244::Tn903dIIlacZ, DNA hybridization group II 28
 LELA3379 JR32 icmB3379::Tn903dIIlacZ, DNA hybridization group II 28
 LELA3393 JR32 icmB3393::Tn903dIIlacZ, DNA hybridization group II 28
 LELA3563 JR32 icmB3563::Tn903dIIlacZ, DNA hybridization group II 28
 LELA3896 JR32 icmB3896::Tn903dIIlacZ, DNA hybridization group II 28
 LELA1223 JR32 icmB1223::Tn903dIIlacZ, DNA hybridization group IV 28
 LELA1566 JR32 icmB1566::Tn903dIIlacZ, DNA hybridization group IV 28
 LELA3150 JR32 icmB3150::Tn903dIIlacZ, DNA hybridization group IV 28
 LELA3323 JR32 icmB3323::Tn903dIIlacZ, DNA hybridization group IV 28
 LELA1012 JR32 icmB1012::Tn903dIIlacZ, DNA hybridization group VI 28
 LELA2947 JR32 icmB2947::Tn903dIIlacZ, DNA hybridization group VI 28
 LELA1275 JR32 icmF1275::Tn903dIIlacZ, DNA hybridization group XVII 28
 LELA1718 JR32 icmF1718::Tn903dIIlacZ, DNA hybridization group XVII 28
 MW635 JR32 icmG::Kmr This study
 MW645 JR32 icmC::Kmr This study
 MW656 JR32 icmJ::Kmr This study
 MW627 JR32 tphA::Kmr This study
 Philadelphia-1 Virulent, serogroup 1 16
E. coli
 DH5α F−endA1 hsdRI7 (r−k m+k) supE44 thi-1 λ− recA1 relA1 Δ(argF-lacZYA) U169 φ80dlacZΔM15 deoR gyrA96 Nalr 33
 LE392 F− e14− (McrA−) hsdR514 (r−k m+k) supE44 supF58 lacY1 or Δ(lacIZY)6 galK2 galT22 metB1 trpR55 3
 LW211 LE392 with an integrated RP4 (Δbla tetA::Mu) transferred by P1 transduction from SM10; Tra+ Kmr Apr Tcs 28
 MW67 DH5α containing a L. pneumophila Philadelphia-1 chromosomal DNA library consisting of partial EcoRI fragments (10–20 kb) in pMMB207 This study
Open in a new tab

TABLE 2.

Plasmids used in this study

Plasmid Genotype and features Reference or source
pAB13 Tn903dIIlacZ-containing EcoRI fragment from LELA1012 in pBSK KS+ 28
pAB14 Tn903dIIlacZ-containing EcoRI fragment from LELA1566 in pBSK KS+ 28
pBluescript II KS (+) oriR (f1) MCS oriR (ColE1), Apr Stratagene
pBC SK (+) oriR (f1) MCS oriR (ColE1), Cmr Stratagene
pBR322 oriR (ColE1), Apr Tcr New England Biolabs
pCR2.1 oriR(f1) MCS oriR (ColE1), Apr Kmr Invitrogen
pMMB207 oriR (RSF1010), Cmr 25
pMMB207αb pMMB207 MCS, Cmr 30
pMMB207αb::Gm pMMB207 MCS, Gmr This study
pMW100 icmEGCDJB-, tphA-, and icmF-containing fragment in pMMB207 This study
pMW150 Tn903dIIlacZ-containing EcoRI fragment from LELA3563 in pMMB207 This study
pMW152 Tn903dIIlacZ-containing EcoRI fragment from LELA1223 in pMMB207 This study
pMW318 Tn903dIIlacZ-containing EcoRI fragment from LELA1996 in pMMB207 This study
pMW320 Tn903dIIlacZ-containing EcoRI fragment from LELA1506 in pMMB207 This study
pMW420 1,864-bp icmJ-containing fragment in pBSK KS+ This study
pMW424 2,892-bp icmGCD-containing fragment in pBSK KS+ This study
pMW525 3,257-bp HindIII fragment from pMW100 in pBSK KS+ This study
pMW528 icmB-containing fragment in pBSK KS+ This study
pMW560 icmB-containing fragment in pBC SK+ This study
pMW564 897-bp icmC-containing fragment in pCR2.1 This study
pMW565 811-bp icmD-containing fragment in pCR2.1 This study
pMW566 1,495-bp icmG-containing fragment in pCR2.1 This study
pMW576 2,004-bp tphA-containing fragment in pUC18 This study
pMW582 pMW424 with a KpnI site at bp 2993 This study
pMW584 2,540-bp icmGCD-containing fragment in pUC18 This study
pMW587 pMW420 containing ΔicmJ with a SalI site at bp 2993 This study
pMW589 pMW576 containing ΔtphA with a SalI site at bp 6862 This study
pMW591 pMW584 containing ΔicmG with a SalI site at bp 995 This study
pMW593 pMW584 containing ΔicmC with a SalI site at bp 1831 This study
pMW596 pMW589 with Km cassette inserted in ΔtphA This study
pMW598 pMW591 with Km cassette inserted in ΔicmG This study
pMW600 pMW593 with Km cassette inserted in ΔicmC This study
pMW602 1,331-bp HindIII-BamHI fragment from pMW587 in pUC18 This study
pMW604 2,540-bp icmGCD-containing fragment in pMMB207 This study
pMW606 pMW602 with Kmr cassette inserted in ΔicmJ This study
pMW616 ΔtphA::Kmr in pLAW344 This study
pMW618 ΔicmG::Kmr in pLAW344 This study
pMW620 ΔicmC::Kmr in pLAW344 This study
pMW622 ΔicmJ::Kmr in pLAW344 This study
pMW680 1,864-bp icmJ-containing fragment in pBC, opposite orientation to PlacUV5 This study
pMW681 1,864-bp icmJ-containing fragment in pBC, same orientation to PlacUV5 This study
pMW728 897-bp icmC-containing fragment in pMMB207αb, same orientation to Ptac This study
pMW730 897-bp icmC-containing fragment in pMMB207αb, opposite orientation to Ptac This study
pMW734 811-bp icmD-containing fragment in pMMB207αb, same orientation to Ptac This study
pMW736 811-bp icmD-containing fragment in pMMB207αb, opposite orientation to Ptac This study
pMW741 1,495-bp icmG-containing fragment in pMMB207αb, same orientation to Ptac This study
pMW743 1,495-bp icmG-containing fragment in pMMB207αb, opposite orientation to Ptac This study
pMW790 1,894-icmJ-containing fragment in pMMB207αb::Gm, opposite orientation to Ptac This study
pLB41 HindIII fragment containing Gmr in pBR322 L. Babiss and D. Figurski
pUC18 oriR(ColE1) MCS, Apr New England Biolabs
Open in a new tab

Cell culture and L. pneumophila growth within and cytotoxicity for HL-60 cells.

The human leukemia cell line HL-60 was used for all tissue culture studies (8). HL-60 cells were maintained in RPMI 1640 medium supplemented with 2 mM l-glutamine (Gln) and 10% fetal calf serum at 37°C under 5% CO2–95% air. HL-60 cells were differentiated into macrophages by incubation for 48 h with 10 ng of phorbol 12-myristate 13-acetate per ml in RPMI 1640–2 mM Gln–10% NHS. HL-60-derived macrophages were washed twice with RPMI 1640–2 mM Gln and incubated with RPMI 1640–2 mM Gln–10% NHS prior to infection with L. pneumophila. The HL-60 cell plaque assay and growth assay to measure L. pneumophila multiplication within HL-60 cells were performed as previously described (22). The degree of cytotoxicity of strains of L. pneumophila for HL-60-derived macrophages was determined by measuring the level of cell survival after infection by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay. The MTT assay was performed as described by Marra et al. (23) with 4.0 × 105 differentiated HL-60 cells infected with strains of L. pneumophila at a multiplicity of infection ranging from 10 to 106 CFU/ml. The growth assay and MTT assay for each strain were performed in separate experiments at least twice, and concordant results were obtained.

DNA sequencing and sequence analysis.

The double-stranded nucleotide sequence of 11,249 bp of the approximately 13,500 bp on the genomic insert of pMW100 was determined by the dideoxy chain termination reaction of Sanger et al. with the Sequenase kit version 2.0 (USB Specialty Biochemicals, Cleveland, Ohio). Some sequence was generated by the DNA Synthesis and Sequencing Facility of the Comprehensive Cancer Center, College of Physicians and Surgeons of Columbia University. ORFs were identified by using the MacDNAsis V 2.0 program. The nucleotide sequence and amino acid sequences identified were compared to the GenBank/EMBL and SwissProt databases by using the programs BLASTX and BLASTP (18) and TFASTA. Promoter searches were performed by using the MacTargsearch program (12). Terminator sequence searches and motif searches were performed by using the Sequence Analysis Software Package, version 7 (Genetics Computer Group, Inc. [GCG]). The Psort (26) program was used to identify potential transmembrane domains and to predict the cellular location of each potential protein product. Tn903dIIlacZ fusions into the L. pneumophila genomic DNA were determined by using synthetic primers corresponding to the 5′ lacZ coding region from nucleotides 56 to 41 (5′-CCCAGTCACGACGTTG-3′) or the 3′ Kmr end from nucleotides 4286 to 4305 (5′-CCAACCGCTGTTTGGTCTGC-3′) of Tn903dIIlacZ (9).

Determination of the Tn903dIIlacZ insertion sites.

The majority of Tn903dIIlacZ insertions in the Mak− L. pneumophila strains were identified by inverse PCR. Genomic DNA was isolated as previously described (21) and digested to completion with the restriction enzymes DraI and ScaI. The restriction digests were precipitated with ethanol. The DNA concentration was determined, and the restriction digest was ligated with T4 DNA ligase overnight at 16°C at a DNA concentration of <2.0 μg/ml to favor the formation of monomeric circles. The ligation reaction was stopped by heating to 68°C for 15 min and isolated by precipitation with ethanol. The entire ligation mix was added to a PCR mixture containing 300 ng each of the lacZ primer and Kmr primer corresponding to the 5′ and 3′ ends of the Tn903dIIlacZ, respectively, 200 μM each deoxynucleoside triphosphate, PCR buffer lacking MgCl2, 3 mM MgCl2, and 2.5 U of Taq polymerase. The inverse PCR products were amplified at 94°C for 4 min, followed by 35 cycles of 94°C for 45 s, 55°C for 45 s, and 72°C for 2 min, and ending with 72°C for 7 min. The inverse PCR products were visualized on a 0.7% agarose gel and cloned into the PCR cloning vector pCR2.1 (Invitrogen Corp.). The T7 or reverse primers corresponding to the polylinker cloning site were used to sequence the inverse PCR products to identify the Tn903dIIlacZ-Legionella DNA fusions.

The Tn903dIIlacZ insertion locations in LELA2947, LELA3150, and LELA3323 were identified by PCR. The bacteria were grown on ABCYE plates containing kanamycin until single colonies were formed. One colony was added to a PCR mixture as described above. LELA3150 and LELA3323 were amplified with the Km primer of the Tn903dIIlacZ and a primer corresponding to nucleotides 8299 to 8284 (5′-CGCTTCACCTGGCCTC-3′) of pMW100. LELA2947 was amplified by using the Km primer and a primer corresponding to nucleotides 6861 to 6845 (5′-GGCTCCGGCTTGGTACC-3′) of pMW100. The Km primer was used to sequence the fusions to the Tn903dIIlacZ insertions.

The Tn903dIIlacZ EcoRI fragments from LELA1223, LELA1506, LELA1996, LELA3563 were isolated from the chromosome as described below. Genomic DNA from these strains was isolated and digested to completion with EcoRI. The restriction digest was visualized on a 0.7% low-melting-temperature agarose gel, and the region of the gel containing the correct size of the EcoRI fragment plus the Tn903dIIlacZ insertion (in kilobases) was cut out of the gel. The DNA was isolated and ligated to pMMB207 digested with EcoRI. The ligation was used to transform DH5α, and the transformation mixture was plated on LB agar plates containing kanamycin (Tn903dIIlacZ), chloramphenicol (pMMB207), and 5-bromo-4-chloro-3-indolyl-β-d-galactopyranoside (X-Gal) (Tn903dIIlacZ). The plasmid DNA was digested with EcoRI to verify the insert and then sequenced with the lacZ primer to identify the Tn903dIIlacZ location. The plasmids pAB13 (LELA1012) and pAB14 (LELA1566) were sequenced with the Km primer to identify the Tn903dIIlacZ-Legionella DNA fusions (28).

Construction of plasmids for complementation.

Plasmid pMMB207αb::Gm was constructed by cloning the 2,118-bp HincII Gmr fragment from pLB41 into pMMB207αb digested with DraI. Plasmid pMW560 (containing the icmB gene) was constructed by first subcloning the HindIII 3,257-bp fragment (nucleotides 3344 to 6601) from pMW100 to pBSK KS+ to form pMW525. Plasmid pMW525 was digested with HincII (which cuts at nucleotide 5333) and KpnI (polylinker), and the 1,512-bp HincII-KpnI fragment from pMW100 (nucleotides 5333 to 6845) was cloned to create pMW528 and to complete the icmB gene. The 3,501-bp KpnI-XbaI fragment from pMW528 was subcloned to pBC SK+ digested with KpnI and XbaI to generate pMW560. Plasmid pMW604 (icmGCD) was constructed by subcloning the EcoRI (polylinker)-KpnI (which cuts at nucleotide 2992) 2,540-bp fragment from pMW582 (described below) to pMMB207 digested with EcoRI and KpnI. Plasmids pMW680 (icmJ in the opposite direction to PlacUV5) and pMW681 (icmJ in the same direction to PlacUV5) were constructed by ligating the 1,864-bp PstI fragment (nucleotides 2429 to 4293) from pMW100 to pBC SK+ digested with PstI. Plasmid pMW790 (icmJ in the opposite direction to Ptac) was constructed by ligating the same 1,864-bp PstI fragment into pMM207αb::Gm digested with PstI. Plasmids pMW728 and pMW730 were constructed in two steps. First, a PCR under the amplification conditions described above amplified an 897-bp fragment containing the entire icmC gene from the L. pneumophila chromosome. The primers corresponded to nucleotides 1505 to 1520 (5′-GTGTCTCGGCCAATAG-3′) and 2402 to 2387 (5′-GCCAAACAAGCTGCGC-3′) of pMW100. The isolated PCR product was subcloned into pCR2.1 to generate pMW564. The PCR product was sequenced to ensure that no errors occurred in the PCR amplification. pMW564 was digested with EcoRI, and the 897-bp insert was subcloned to pMMB207αb digested with EcoRI to generate pMW728 (icmC in the same direction to Ptac) and pMW730 (icmC in the opposite direction to Ptac). Plasmids pMW734 (icmD in the same direction to Ptac) and pMW736 (icmD in the opposite direction to Ptac) were constructed as described above. The primers corresponded to nucleotides 2181 to 2196 (5′-GCGCGTTCCGCGTCGC-3′) and 2992 to 2977 (5′-CGCCAGGAACCTGGTG-3′) of pMW100. The isolated 811-bp PCR product was subcloned into pCR2.1 to generate pMW565 and then into pMMB207αb digested with EcoRI to generate pMW734 and pMW736. The plasmids pMW741 (icmG in the same direction to Ptac) and pMW743 (icmG in the opposite direction to Ptac) were generated as described above with primers corresponding to nucleotides 563 to 579 (5′-CACGGCAAGAACAGCCC-3′) and 2058 to 2043 (5′-CACCTCCTGAGTAGGC-3′) of the pMW100 sequence. The isolated 1,495-bp PCR product was subcloned into pCR2.1 to generate pMW566 and then into pMMB207αb digested with EcoRI to generate pMW741 and pMW743.

Construction of plasmids for allelic exchange.

Deletions were generated as described by Imai et al. (17). First, pMW424 was constructed by cloning the 2,892-bp HindIII fragment from pMW100 (nucleotides 452 to 3344) into pBSK KS+ digested with HindIII. Plasmid pMW424 was amplified in a PCR with using Vent polymerase, an annealing temperature of 72°C, and amplification at 75°C for 6 min. The primers used were generated with half of a KpnI site (5′-ACC-3′) at each 5′ end to generate a full KpnI site after self-ligation. The primers corresponded to nucleotides 2992 to 2973 (5′-ACCCGCCAGGAACCTGGTGAAGC-3′) and 3210 to 3229 (5′-ACCGGTGGTTATGGTGGAGGTAC-3′) of pMW100. The PCR products were visualized on a 0.7% low-melting-temperature agarose gel, and the expected 5,641-bp product was isolated and self ligated in a reaction mixture that contained 1 U of T4 polynucleotide kinase. The ligation was transformed to DH5α, and the transformants were screened for the presence of the KpnI site generated by PCR. The resulting plasmid, pMW582, contained the complete ORFs of icmGCD. The 2,540-bp HindIII-KpnI insert from pMW582 was subcloned to pUC18 digested with HindIII and KpnI to generate pMW584. To delete the icmG gene, the exact procedure as described above was performed on pMW584 with the following revisions. The primers used contained one half of a SalI site (5′-GAC-3′) at each 5′ end and corresponded to nucleotides 994 to 976 (5′-GACCCGATTCACCAGCCTGATCC-3′) and 1505 to 1527 (5′-GACGTGTCTCGGCCAATAGTTCAAGC-3′) of pMW100. The 4,703-bp PCR product was isolated and ligated, screening for the presence of the SalI site generated. The icmG deletion plasmid was named pMW591. To delete the icmC gene, pMW584 was used in the procedure described above. Each primer contained half of a SalI site and corresponded to nucleotides 1830 to 1811 (5′-GACGCCTTTGAACAGGCTCCAAC-3′) and 2181 to 2201 (5′-GACGCGCGTTCCGCGTCGCAGGGG-3′) of pMW100. The 4,862-bp PCR product was isolated and ligated, and the icmC deletion plasmid was named pMW593. To delete the icmJ gene, a plasmid was constructed by cloning the 1,864-bp PstI fragment from pMW100 (nucleotides 2429 to 4293) in pBSK KS+ digested with PstI. This plasmid, pMW420, was used in a PCR as described above. Each primer contained half of a SalI site and corresponded to nucleotides 2992 to 2973 (5′-GACCGCCAGGAACCTGGTGAAGC-3′) and 3526 to 3546 (5′-GACGGGCTGCCAGTGCTTTAGAAG-3′) of pMW100. The 4,298-bp product was isolated and ligated, and the icmJ deletion plasmid was named pMW587. The 1,331-bp HindIII-BamHI insert of pMW587 was subcloned to pUC18 digested with HindIII and BamHI to generate pMW602. To construct the tphA deletion, the 2,004-bp PstI-BamHI fragment of pMW100 (nucleotides 6369 to 8373) was cloned to pUC18 digested with PstI and BamHI to generate pMW576. Plasmid pMW576 was used in the deletion PCR as described above; each primer contained half of a SalI site and corresponded to nucleotides 6861 to 6839 (5′-GACGGCTCCGGCTTGGTACCTTTTCC-3′) and 7936 to 7956 (5′-GACGG AGCAAAACTGGCATAGAGC-3′) of pMW100. The 3,622-bp product was isolated and ligated, and the tphA deletion plasmid was named pMW589. Plasmids pMW591 (ΔicmG), pMW593 (ΔicmC), pMW602 (ΔicmJ), and pMW589 (ΔtphA) were digested with SalI, and a Kmr cassette (Pharmacia Biotech) was cloned into the site. The resulting plasmids, pMW598 (ΔicmG::Km), pMW600 (ΔicmC::Km), pMW606 (ΔicmJ::Km), and pMW596 (ΔtphA::Km), were digested with PvuII, and the Kmr fragments were ligated to pLAW344 digested with EcoRV to generate pMW618, pMW620, pMW622, and pMW616, respectively. These plasmids were used for allelic exchange. Allelic exchange was performed as previously described (32).

Nucleotide sequence accession number.

Sequence data of the partial icmE gene and the complete icmGCDJB, tphA, and icmF genes was assigned accession no. Y14948 in the EMBL nucleotide sequence database.

RESULTS

Isolation of a complementing Mak+ library clone.

To complement the macrophage-killing defect of the Mak− mutants, a library of wild-type L. pneumophila DNA was constructed in the IncQ cloning vector pMMB207. pMMB207 contains an RSF1010 origin of replication and is stably maintained in L. pneumophila (25). The library was electroporated into one mutant of group II, LELA3896. The pool of transformed bacteria was then tested in a plaque assay on HL-60 cells. The plaque assay selects for bacteria that are able to infect a macrophage monolayer, multiply intracellularly, kill the host cell, and infect neighboring cells to continue multiple rounds of infection. This technique is valuable since it selects for the few Mak+ transformants within the large pool of Mak− bacteria. Most Mak− mutants do not form plaques at a detectable frequency, and strain LELA3896 containing the vector pMMB207 was unable to form plaques (data not shown). One plaque was isolated from the library pool in LELA3896, and the Mak+ bacteria were rescued, purified from the plaque, and retested in a modified plaque assay. In this modified assay, the bacteria were inoculated directly through the agarose and into the macrophage monolayer with a sterile toothpick. The purified transformants retained the ability to form plaques surrounding the site of inoculation in this assay (data not shown).

To show that the Mak+ phenotype was linked to the genomic DNA insert on the library clone, the complementing plasmid, pMW100, was isolated from the Mak+ bacteria and reintroduced into the original Mak− mutant strain. The transformants were tested in the modified plaque assay and were found to be able to form plaques. To find if a mutation occurred in the LELA3896 chromosome that restored the macrophage-killing phenotype in the original isolate, the complemented bacteria were rescued from the macrophage monolayer and isolated on ABCYE plates lacking chloramphenicol. The bacteria were then passed two more times on ABCYE plates to cure the library clone from the bacteria. These Cms cured bacteria were tested in the modified plaque assay and were unable to form plaques. Therefore, the Mak+ phenotype is linked to the genomic insert on pMW100 and is not due to a reversion or recombination event in the chromosome of LELA3896.

Growth within and killing of differentiated HL-60 cells.

To determine the capacity of pMW100 to complement LELA3896 for intracellular multiplication, growth assays were performed with HL-60 cells. As shown in Fig. 1A, the wild-type strain JR32 multiplied >100-fold by the fourth day following infection. Strain JR32 containing the vector pMMB207 or plasmid pMW100 showed the same pattern of growth as JR32 (data not shown). The Mak− mutant LELA3896 did not multiply detectably. LELA3896 carrying the vector pMMB207 was also unable to multiply within HL-60 cells. LELA3896 containing plasmid pMW100 multiplied >100-fold by the fourth day following infection, comparable to the wild-type strain. Therefore, the genomic information on pMW100 was sufficient to enable the Mak− mutant LELA3896 to survive and multiply within HL-60 cells.

FIG. 1.

FIG. 1

Open in a new tab

(A) Intracellular multiplication of strain LELA3896 within HL-60-derived macrophages. Differentiated HL-60 cells (1.5 × 106 per well) were infected with approximately 104 bacteria on day 0. The number of CFU of each strain per milliliter was determined daily for 5 days, in triplicate, on ABCYE plates. ▪, JR32; □, LELA3896; •, LELA3896(pMMB207); ○, LELA3896(pMW100). Error bars indicate the standard error of the mean. (B) Cytotoxicity of strain LELA3896 for HL-60-derived macrophages. Differentiated HL-60 cells (4 × 105 per well) were infected with the indicated strains of bacteria at concentrations ranging from 10 to 106 CFU/ml. After 6 days, the remaining viable macrophages in the monolayer were quantitated by adding the dye MTT and measuring the absorbance at 570 nm (A570). ▪, JR32(pMMB207); □, JR32(pMW100); •, LELA3896; ○, LELA3896(pMMB207); ▴, LELA3896(pMW100). Values are the average of three determinations. Error bars indicate the standard error of the mean.

A cytotoxicity assay was performed on HL-60 cells to quantitate the ability of pMW100 to complement LELA3896 for the killing of macrophages. As shown in Fig. 1B, the wild-type strain JR32 containing the vector pMMB207 was able to kill the HL-60 cell monolayer with increasing numbers of input bacteria. JR32 containing plasmid pMW100 showed a similar level of macrophage killing. Strains LELA3896 and LELA3896 carrying the vector pMMB207 did not kill macrophages even with a high multiplicity of infection. Strain LELA3896 containing pMW100 was able to kill the macrophage monolayer at a level comparable to that for wild-type bacteria. Therefore, pMW100 complemented LELA3896 for the ability to both replicate within and kill macrophages and fully restores the Mak+ phenotype to this mutant.

Plasmid pMW100 complements members of DNA hybridization groups II, IV, and VI.

To test if other Mak− mutants were complemented by pMW100, the plasmid was transferred into each of the 55 Mak− LELA strains by bacterial mating. Each pMW100-containing mutant was tested in the modified plaque assay on HL-60 cells. All the mutants in Mak− group II were restored for the ability to produce plaques on HL-60 cells except for strains LELA1984 and LELA2517. All members of group IV and group VI were also complemented by pMW100. Every member of DNA hybridization groups I, III, VII, VIII, IX, X, and XII was not complemented for its Mak− defect when transformed with pMW100 (data not shown). The mutants representing DNA hybridization groups XI, XIII, XIV, XV, and XVI and the six ungrouped mutants were unable to be tested by this assay. These strains retain some ability to kill macrophages, and this residual activity is sufficient to enable the mutants alone to form plaques on macrophage monolayers (data not shown).

Restriction map of pMW100 and Mak− Tn903dIIlacZ insertions on pMW100.

An EcoRI restriction map of pMW100 is shown in Fig. 2A. Four EcoRI fragments were found: 5.4, 0.6, 2.0, and 5.5 kb. The Ptac promoter of the vector pMMB207 directs transcription to the left into the 5.5-kb EcoRI fragment. Southern hybridization analysis showed that these four EcoRI fragments were contiguous in the wild-type L. pneumophila chromosome (data not shown).

FIG. 2.

FIG. 2

Open in a new tab

(A) EcoRI restriction map of the genomic insert on plasmid pMW100. EcoRI restriction sites are labeled R. The approximate size of each EcoRI restriction fragment is shown in kilobases. Ptac of the pMMB207 vector directs transcription to the left into the 5.5-kb EcoRI fragment. The dotted line at the 3′-end of the 5.5-kb EcoRI fragment represents the region not sequenced. Mak− DNA hybridization groups represented on the genomic insert of pMW100 as shown by Southern analysis are shown above the EcoRI fragments. (B) ORFs present on the genomic insert of pMW100 and locations of Tn903dIIlacZ insertions (▹) and deletion substitutions (▪) in the Mak− mutants. Triangles on top of one another indicate that the LELA strains contain a Tn903dIIlacZ insertion at the same nucleotide. Vertical numbers indicate the strain number. ORFs shown were identified by nucleotide sequencing and MacDNAsis analysis. Solid arrows indicate the direction of transcription of each ORF. The location with respect to each EcoRI restriction site and size of each ORF is approximate. icmE is represented as a partial ORF, as indicated by a dotted 5′ end.

To determine if any Mak− mutants contained Tn903dIIlacZ insertions in the genomic insert represented by pMW100, the Mak− mutant genomes were digested with EcoRI and probed with each pMW100 EcoRI fragment in a Southern analysis (data not shown). A Mak− mutant that contained a Tn903dIIlacZ insertion within the genomic EcoRI fragment used as a probe contained a hybridizing band 4.4 kb larger (the size of the Tn903dIIlacZ insertion) than wild-type genomic DNA. As shown in Fig. 2A, every mutant in Mak− groups II, VI, and IV contained Tn903dIIlacZ insertions in pMW100 EcoRI fragments of 5.4, 0.6, and 2.0 kb, respectively. Ungrouped mutants LELA1275 and LELA1718 contained insertions in the 5.5-kb EcoRI fragment, creating a new Mak− DNA hybridization group, XVII.

Since the mutants in Mak− group XVII were able to form visible plaques on HL-60 cell monolayers, a cytotoxicity assay was performed on these two mutants transformed with pMW100 to see if the plasmid complemented the strains for killing macrophages. As shown in Fig. 3, strains LELA1275 and LELA1718 containing the vector pMMB207 had a partial or Mak± phenotype. Plasmid pMW100 restored macrophage killing to strain LELA1275 approximately fourfold. However, 103-fold more bacteria were required to produce a similar level of cytopathic effects to that produced by the wild-type strain. LELA1718 containing pMW100 was restored for killing approximately 10-fold to a level similar to that for the wild-type strain JR32. Therefore, plasmid pMW100 partially complemented these two mutants for the ability to kill macrophages.

FIG. 3.

FIG. 3

Open in a new tab

Cytotoxicity of the icmF mutant strains for HL-60-derived macrophages. Differentiated HL-60 cells (4 × 105 cells per well) were infected with the indicated strains of bacteria at concentrations ranging from 10 to 106 CFU/ml. After 6 days, the remaining viable macrophages in the monolayer were quantitated by adding the dye MTT and measuring the A570. (A) icmF mutant strain LELA1275. ▪, JR32(pMMB207); □, LELA1275(pMMB207); •, LELA1275(pMW100). (B) icmF mutant strain LELA1718. ▪, JR32(pMMB207); □, LELA1718(pMMB207); •, LELA1718(pMW100). Values are the average of three determinations. Error bars indicate the standard error of the mean.

Sequence analysis of the genomic insert on pMW100 and identification of potential ORFs.

The double-stranded nucleotide sequence of 11,249 bp of the genomic insert present on pMW100 was determined. A total of eight potential ORFs were identified. As described below, seven of the ORFs were shown to be required for macrophage killing and were named icmE, icmG, icmC, icmD, icmJ, icmB, and icmF (for intracellular multiplication). Only the 3′ end of the icmE ORF was present on pMW100; therefore, icmE was not characterized further. The eighth ORF, tphA, was shown to be dispensable for killing macrophages (see below) and was named according to its homology to transport protein homologs. The location and approximate size of each ORF are shown in Fig. 2B. Ribosome binding sites were identified, and these precede icmC, icmD, icmJ, icmB, and icmF at an appropriate distance from the putative initiation codons. The icmG and tphA genes did not appear to contain recognizable ribosome binding sites. The MacTargsearch program did not identify E. coli-like consensus promoter sequences upstream of any of the ORFs (12). One possible reason is that L. pneumophila chromosomal DNA has a low G+C content (39%) compared to E. coli (51%). A conserved sequence was found in the upstream regions of the icmB and icmF genes that might serve as a promoter or a recognition site for a transcription factor (30). A GCG search to identify transcriptional termination sequences identified a rho-independent termination sequence 47 bp downstream of the termination codon of icmD characterized by a GC-rich inverted repeat followed by a run of five T’s. Approximately 12 bp 5′ to this termination sequence, a 10-bp inverted repeat was found. The combination of this inverted repeat and the rho-independent termination sequence may indicate a termination of transcription after icmD.

Each ORF was analyzed by a variety of techniques, and the results are summarized in Table 3. To determine if the potential products of the identified ORFs have homology to previously identified proteins, the GenBank/EMBL and SwissProt databases were searched. All ORFs, except for tphA, show no homology to previously described genes or proteins at the nucleotide and amino acid levels. TphA contains 30% identity over 424 amino acids to the ProP protein of E. coli encoding the (proline/betaine:H+/Na+) symport protein (28a). Motif searches of each protein product identified a consensus ATP/GTP binding site motif A in the products of the icmB and icmF genes. The Psort program predicted that all the ORF products are located in the bacterial inner membrane and identified a number of potential transmembrane domains within each amino acid sequence (26). The hydropathic profile of each ORF product is shown on Kyte-Doolittle plots (20) in Fig. 4. The Psort program also located an uncleavable N-terminal signal sequence in icmC. The icmD gene product appears to have an N-terminal signal sequence with predicted cleavage after the alanine at amino acid 33, even though the Psort program predicted an inner membrane location for this protein.

TABLE 3.

Characteristics of predicted ORFs

Potential ORF identified No. of amino acidsa Mol mass (kDa)b Predicated cellular locationc Predicted motifsd Predicted homologye
icmE
icmG 269 29.7 IM (1 TM) None None
icmC 194 20.3 IM (4 TM) None None
icmD 132 13.5 IM (1 TM) None None
icmJ 212 24.2 IM (1 TM) None None
icmB 1,009 112.1 IM (1 TM) ATP/GTP binding site None
tphA 418 46.4 IM (12 TM) Sugar transport protein ProP
icmF 973 110.7 IM (2 TM) ATP/GTP binding site None
Open in a new tab
a

The number of amino acids encoded by each complete ORF is reported. icmE is partial on pMW100; therefore the total number of amino acids is not reported. 

b

The predicted molecular mass of each complete ORF is shown. The molecular mass of the partial icmE cannot be determined. 

c

The location of each ORF was predicted by the Psort program (26). IM designates an inner membrane location. The number of transmembrane domains (TM) as determined by the Psort program is shown in parentheses. 

d

Motif searches were performed with the GCG motif search program. 

e

Homology searches were performed at the amino acid and nucleotide levels by using the TFASTA, BLASTP, and BLASTX (18) programs. 

FIG. 4.

FIG. 4

Open in a new tab

Representation of predicted transmembrane domains in the icm gene products. The hydropathic profiles of the predicted Icm proteins are shown according to the Kyte-Doolittle scale with a window size of 20 amino acids. The transmembrane domains as identified by the Psort program are represented as bars above the graphs. The vertical axis represents the hydrophobic value as measured on the Kyte-Doolittle scale, and the horizontal axis represents the number of amino acids in the icm gene products.

Mapping of the Tn903dIIlacZ insertions within the identified ORFs.

A variety of techniques was used to identify the precise location of each Tn903dIIlacZ insertion. Previously, the EcoRI fragments containing the Tn903dIIlacZ fusions from a number of the Mak− mutant chromosomes were cloned. To identify the locations of additional Tn903dIIlacZ insertions, the genomic EcoRI fragments containing the Tn903dIIlacZ insertions from a number of Mak− mutants were subcloned on EcoRI fragments into the vector pMMB207 to generate plasmids pMW150, pMW152, pMW318, and pMW320. These plasmids and the previously constructed plasmids pAB13 and pAB14 were sequenced to identify the fusion junctions to the Tn903dIIlacZ insertions. Inverse PCR was used to identify and confirm the remaining Tn903dIIlacZ insertion sites (see Materials and Methods). The insertions in LELA2947 (group VI) and LELA 3150 and LELA3323 (group IV) could not be cloned from their respective chromosomes and were not amplified by inverse PCR. Therefore, direct PCR was used to amplify a genomic fragment containing the region flanking the Tn903dIIlacZ insertions. The majority of Tn903dIIlacZ insertions were mapped in icmB; therefore, PCRs were performed with the Km primer of the Tn903dIIlacZ and primers downstream of the predicted sites of insertions in the icmB gene (see Materials and Methods). A genomic fragment of the predicted size was amplified for all three mutant strains (data not shown), and the PCR products were cloned in the pCR2.1 vector and sequenced as described in Materials and Methods. Figure 2B summarizes the results of the map locations of the Tn903dIIlacZ insertions within the ORFs identified. No insertions were found in icmG, icmJ, and tphA.

Complementation analysis of icmB and icmJ.

To examine the role of icmB in macrophage killing, the entire gene was subcloned onto pBC SK+ in the same direction as the vector PlacUV5. This plasmid, pMW560, was used to transform four different icmB mutants (LELA1012, LELA1223, LELA3393, and LELA3896), and the transformants were tested for macrophage killing by the cytotoxicity assay. As shown in Fig. 5, the plasmids pMW560 and pMW100 complemented each icmB mutant strain to a cytotoxicity level similar to that of wild-type bacteria. However, strain LELA1012 required approximately 100-fold more bacteria to produce a similar degree of cytopathic effects to that of JR32. The icmB gene is therefore expressed from the pMW560 plasmid, and mutations in icmB could be complemented. These results confirm that icmB plays a role in macrophage killing.

FIG. 5.

FIG. 5

Open in a new tab

Cytotoxicity of the icmB mutant strains containing different plasmids for HL-60-derived macrophages. Differentiated HL-60 cells (4 × 105 per well) were infected with the indicated strains of bacteria at concentrations ranging from 10 to 106 CFU/ml. After 6 days, the remaining viable macrophages in the monolayer were quantitated by adding the dye MTT and measuring the A570. (A) icmB mutant strain LELA3393. ▪, JR32(pMMB207); □, LELA3393(pBC SK+); •, LELA3393(pMW100); ○, LELA3393(pMW560). (B) icmB mutant strain LELA1012. ▪, JR32(pMMB207); □, LELA1012(pBC SK+); •, LELA1012(pMW100); ○, LELA1012(pMW560). (C) icmB mutant strain LELA1223. ▪, JR32(pMMB207); □, LELA1223(pBC SK+); •, LELA1223(pMW100); ○, LELA1223(pMW560). (D) icmB mutant strain LELA3896. ▪, JR32(pMMB207); □, LELA3896(pBC SK+); •, LELA3896(pMW100); ○, LELA3896(pMW560). Values are the average of three determinations. Error bars indicate the standard error of the mean.

No Tn903dIIlacZ insertions were mapped to icmJ; therefore, a mutation within this gene was constructed by allelic exchange. Figure 6A shows that the mutant strain MW656 containing the vector pMMB207αb was completely unable to kill macrophages. To test if icmJ itself plays a role in macrophage killing or if the icmJ mutation is polar on expression of icmB, complementation tests were performed. Strain MW656 was transformed with plasmid pMW560 (icmB), and this strain was completely unable to kill macrophages. Therefore, the icmJ gene is required for macrophage killing and polarity alone on expression of the icmB gene is not sufficient to produce the Mak− phenotype observed. Plasmid pMW100 (all ORFs) restored the ability to kill macrophages to the icmJ mutant at a level comparable to the wild-type strain JR32. Plasmids pMW680 and pMW681 (icmJ in the opposite and same orientation with respect to PlacUV5, respectively) partially complemented this mutant. Approximately 103-fold more bacteria were required to observe the same level of cytotoxicity as that of JR32. These results confirm that icmJ plays a role in macrophage killing. The partial complementation observed may have been due to polarity on expression of the downstream icmB gene. To test this possibility, plasmids pMW560 (icmB) and pMW790 (icmJ in the opposite orientation to Ptac) were both transformed into the icmJ mutant strain MW656, and this strain exhibited a wild-type level of macrophage killing. These results are consistent with the idea that the icmJ::Km mutant decreases the expression of icmB and suggest that icmJ and icmB may form a transcriptional unit.

FIG. 6.

FIG. 6

Open in a new tab

Cytotoxicity assay of L. pneumophila strains containing different plasmids for HL-60-derived macrophages. Differentiated HL-60 cells (4 × 105 cells per well) were infected with the indicated strains of bacteria at concentrations ranging from 10 to 106 CFU/ml. After 6 days, the remaining viable macrophages in the monolayer were quantitated by adding the dye MTT and measuring the A570. (A) icmJ allelic exchange strain MW656. ▪, JR32(pMMB207); □, MW656(pMMB207αb); •, MW656(pMW100); ○, MW656(pMW680); Inline graphic, MW656(pMW681); ▵, MW656(pMW560); ▾▴, MW656(pMW560, pMW790). (B) icmD mutant strain LELA1205. ▪, JR32(pMMB207); □, LELA1205(pMMB207αb); •, LELA1205(pMW100); ○, LELA1205(pMW734); Inline graphic, LELA1205(pMW736). (C) icmC allelic exchange strain MW645. ▪, JR32(pMMB207); □, MW645(pMMB207αb); •, MW645(pMW100); ○, MW645(pMW604); Inline graphic, MW645(pMW728); ▵, MW645(pMW730); ▾▴, MW645(pMW734). (D) icmG allelic exchange strain MW635. ▪, JR32(pMMB207); □, MW635(pMMB207αb); •, MW635(pMW100); ○, MW635(pMW741); Inline graphic, MW635(pMW743). Values are the average of three determinations. Error bars indicate the standard error of the mean.

Complementation analysis of icmD, icmC, and icmG.

A potential transcriptional terminator was identified downstream from the termination codon of icmD. To test if transcription terminates after the icmD gene or if icmJ and icmB are encoded in the same transcript, genetic complementation was performed to determine whether mutations in icmD decrease the expression of icmJ and icmB. As shown in Fig. 6B, the icmD mutant strain LELA1205 containing the vector pMMB207αb was completely defective for killing macrophages. The icmD mutant was complemented for macrophage killing to a level similar to wild-type bacteria by plasmids pMW100 (all ORFs), pMW734 (icmD same orientation to Ptac), and pMW736 (icmD opposite orientation to Ptac). These results confirmed that icmD is essential for macrophage killing and suggest that the insertion in icmD is not polar on expression of the downstream genes icmJ and icmB.

The original icmC mutant strain LELA2517 was not complemented by pMW100 for macrophage killing. The lack of complementation may have been due to a second mutation(s) elsewhere in the chromosome of this strain, or the Tn903dIIlacZ mutation may have been dominant. Therefore, a new icmC mutant was constructed by allelic exchange. As shown in Fig. 6C, the icmC mutant strain MW645 containing the vector pMMB207αb was completely defective for the ability to kill macrophages. To determine if icmC plays a role in macrophage killing or if the icmC mutation is polar on expression of icmD, strain MW645 was transformed with plasmid pMW734 (icmD in the same orientation to Ptac). This strain was unable to kill macrophages. Therefore, the icmC gene is required for macrophage killing, and polarity on expression of the icmD gene alone is not sufficient to produce the macrophage-killing defect observed. Strain MW645 was complemented equally by pMW100 (all ORFs) and pMW604 (icmGCD) to a similar level of cytopathic effects to that of the wild-type strain. However, icmC in the same (pMW728) and opposite orientation (pMW730) with respect to Ptac on the vector showed a partial complementation phenotype. High MOIs were required for macrophage killing. The icmC mutation is probably polar on expression of icmD because a plasmid containing icmGCD showed better complementation of this mutant than did plasmids containing icmC only. Therefore, icmC may be part of an operon with icmD.

No Tn903dIIlacZ insertions were mapped to icmG. To determine if icmG is required for macrophage killing, an icmG mutant was constructed by allelic exchange. The complementation analysis for this mutant is shown in Fig. 6D. The icmG mutant strain MW635 containing the vector pMMB207αb showed a partial macrophage-killing defect. The mutant required 100-fold more bacteria to kill the macrophage monolayer as efficiently as the wild-type strain. The partial Mak− phenotype might be due to the disruption of icmG and/or polarity on the expression of icmC and icmD. Plasmids pMW100 (all ORFs), pMW741 (icmG in the same orientation to Ptac), and pMW743 (icmG in the opposite orientation to Ptac) all complemented the icmG mutant strain equally, similar to the wild-type level of macrophage killing. Therefore, the expression of downstream genes is probably not affected by the icmG gene disruption and icmG does play a role in macrophage killing but is not absolutely required.

Complementation analysis of icmF and tphA.

As described above, icmF mutants were complemented by pMW100; however, LELA1275 was only partially complemented. The direction of transcription on pMW100 from the Ptac promoter is the same as for icmF, so icmF RNA is probably expressed from Ptac. Addition of 1 mM IPTG to the ABCYE agar plates during growth of the LELA1275 strain or to the RPMI 1640 tissue culture media in the cytotoxicity assays did not increase the level of complementation (data not shown). One possible explanation for the lack of complete complementation is that the transposon insertion in this strain is polar on the expression of tphA and that the tphA protein product is required for macrophage killing. To test this possibility, a tphA::Km mutant was constructed by allelic exchange. The mutant strain specifically lacking TphA killed macrophages in a manner identical to the wild-type strain (data not shown). Therefore, even if the insertions in icmF are polar on expression of the tphA protein product, this polarity has no consequence on macrophage killing. The exact reason why LELA1275 cannot be complemented to a wild-type level is not completely understood. However, a second icmF mutant, LELA1718, is completely complemented, confirming that icmF plays a role in host cell killing.

DISCUSSION

L. pneumophila belongs in a class of intracellular pathogens that subvert host cell defenses to form a specialized phagosome. The mechanisms that control phagosome formation, trafficking of the phagosome within the host cell, and subsequent bacterial replication are not well understood. To identify genes that allow L. pneumophila to establish its specialized replicative niche, Tn903dIIlacZ mutagenesis was performed on the L. pneumophila chromosome to isolate Mak− mutants that are unable to multiply within and kill human macrophages.

In this report, we describe a locus isolated from a wild-type L. pneumophila library that is able to complement a subset of the Mak− mutants for their macrophage-killing defect. Genetic complementation analysis of the identified ORFs was used to identify which genes play a role in macrophage killing. This analysis showed that icmC, icmD, icmJ, and icmB are completely required to kill macrophages and that icmG and icmF are partially required. The gene products of icmG and icmF may play accessory roles that are somewhat dispensable for host cell killing. The tphA gene was shown to be completely dispensable for macrophage killing. Either the function of the tphA gene product is not required for killing of the host cell, or another family member of the metabolite/H+ symport proteins may compensate for the defect in the tphA gene. Mutations in all the icm genes were complemented for macrophage killing by various plasmids, and the complementation was to a level similar to that for the wild-type strain JR32. For two mutants, the complementation was 100-fold (LELA1012) and 1,000-fold (LELA1275) less than for the wild-type strain. The chromosomal mutations within the icm genes in these strains may be partially dominant.

The identified icm genes certainly play a role in host cell killing, but the exact function of the icm gene products is not known. The icm genes encode potential protein products that have no homology to any previously identified proteins. All the potential icm gene products described in this report are predicted to be associated with the bacterial inner membrane. Recent evidence has shown that DotA, an L. pneumophila protein required for host cell killing, is also a cytoplasmic membrane protein (27). When the Psort analysis and the Kyte-Doolittle plots of the predicted icm gene products are compared, the assignment of the locations of some of the proteins is shown to be ambiguous. For example, icmB is predicted to be associated with the bacterial inner membrane. However, the calculated Psort score of 0.111 is quite low. The Kyte-Doolittle plot clearly shows characteristics of a cytoplasmic protein with no dramatic hydrophobic regions identified in the protein. Further analysis of the icm gene products is required to determine the exact location of the proteins within the bacterial cell.

Recently, the icm genes described in this study were shown to be contiguous to other icm genes required for the killing of human macrophages by the L. pneumophila bacteria. The icmTSRQPO-lphA-icmMLKE genes were shown to be linked to the icmGCDJB-tphA-icmF genes (29, 30). Therefore, the locus described in this study is part of a 22-kb region of the L. pneumophila chromosome that contains a number of contiguous genes, most of which have been shown to play a role in intracellular infection by the bacteria. Most of the icm genes in this 22-kb region are predicted to be located in the bacterial inner membrane, periplasm, and outer membrane. The majority of these icm gene products do not contain homology to previously identified proteins. Four icm genes (icmP, icmO, icmL, and icmE) encode protein products that showed significant homology to plasmid genes involved in conjugation.

Since the identified genes are clustered and mutations within these contiguous genes show similar killing phenotypes, it is possible that the icm proteins associate with one another or form a multiprotein membrane complex that may play a role in the establishment of infection in the host cell. A mutation in any one of the icm genes may disrupt the interaction between the Icm proteins and have a detrimental effect on complex formation and subsequent activity. It is possible that such a complex imports material such as nutrients for bacterial multiplication. By-products of metabolism may also pass into or out of the cell through this complex. Also, the Icm proteins may be used to export material into the replicative phagosome to modify the phagosomal membrane so that fusion to lysosomes does not occur.

A few of the icm genes have been studied more closely to examine their role in inhibition of phagosome-lysosome fusion. Strains containing Tn903dIIlacZ insertions in dotA and icmX (4), icmR (30), icmE (29), and icmB (see above) were tested for their ability to prevent phagosome-lysosome fusion. In contrast to the phagosomes containing wild-type bacteria, 70 to 80% of phagosomes containing the mutant bacteria fused with host cell lysosomes as quickly as 30 min after infection. These observations are similar to those made with heat-killed bacteria (32a). However, it is difficult to differentiate whether mutations in these genes result in a direct defect in inhibition of phagosome-lysosome fusion or if some other defect in an earlier step of the infection pathway leading to phagosome-lysosome fusion. Such early steps may include the proper sorting of plasma membrane proteins (7) or signaling events during phagosome formation by the bacteria. Further characterization of the function of the icm gene products will enhance our understanding of the mechanisms by which intracellular organisms establish a productive infection and parasitize their host cell.

ACKNOWLEDGMENTS

We thank Gil Segal and Laura Hales for critical analysis of the manuscript. We thank Milton H. Saier, Jr., and Lawrence Wiater for sharing unpublished results. We are grateful to Steven D. Goodman for the MacTargsearch program. We are grateful to Mrs. Carmen Rodriguez for excellent technical assistance during this work.

This work is supported by NIH grant AI23549.

REFERENCES

  • 1.Berger K H, Isberg R R. Two distinct defects in intracellular growth complemented by a single genetic locus in Legionella pneumophila. Mol Microbiol. 1993;7:7–19. doi: 10.1111/j.1365-2958.1993.tb01092.x. [DOI] [PubMed] [Google Scholar]
  • 2.Berger K H, Merriam J J, Isberg R R. Altered intracellular targeting properties associated with mutations in the Legionella dotA gene. Mol Microbiol. 1994;14:809–822. doi: 10.1111/j.1365-2958.1994.tb01317.x. [DOI] [PubMed] [Google Scholar]
  • 3.Borck K, Beggs J S, Brammar W J, Hopkins A S, Murry N E. The construction in vitro of transducing derivatives of phage λ. Mol Gen Genet. 1976;146:199–207. doi: 10.1007/BF00268089. [DOI] [PubMed] [Google Scholar]
  • 4.Brand B C, Sadosky A B, Shuman H A. The Legionella pneumophila icm locus: a set of genes required for intracellular multiplication in human macrophages. Mol Microbiol. 1994;14:797–808. doi: 10.1111/j.1365-2958.1994.tb01316.x. [DOI] [PubMed] [Google Scholar]
  • 5.Cianciotto N P, Eisenstein B I, Mody C H, Engleberg N C. A mutation in the mip gene results in an attenuation of Legionella pneumophila virulence. J Infect Dis. 1990;162:121–126. doi: 10.1093/infdis/162.1.121. [DOI] [PubMed] [Google Scholar]
  • 6.Cianciotto N P, Eisenstein B I, Mody C H, Toews G B, Engleberg N C. A Legionella pneumophila gene encoding a species-specific surface protein potentiates the initiation of intracellular infection. Infect Immun. 1989;57:1255–1262. doi: 10.1128/iai.57.4.1255-1262.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Clemens D L, Horwitz M A. Membrane sorting during phagocytosis: exclusion of MHC molecules but not complement receptor CR3 during conventional and coiling phagocytosis. J Exp Med. 1992;175:1317–1326. doi: 10.1084/jem.175.5.1317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Collins S J, Ruscetti F W, Gallagher R E, Gallo R C. Terminal differentiation of human promyelocytic leukemia cells induced by dimethyl sulfoxide and other polar compounds. Proc Natl Acad Sci USA. 1978;75:2458–2462. doi: 10.1073/pnas.75.5.2458. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Derbyshire K M. An IS903-based vector for transposon mutagenesis and the isolation of gene fusions. Gene. 1995;165:143–144. doi: 10.1016/0378-1119(95)00512-5. [DOI] [PubMed] [Google Scholar]
  • 9a.Feeley J C, Gibson R J, Gorman G W, Langford N C, Rasheed J K, Mackel D C, Baine W B. Charcoal-yeast extract agar: primary isolation medium for Legionella pneumophila. J Clin Microbiol. 1979;10:437–441. doi: 10.1128/jcm.10.4.437-441.1979. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Fischer G, Bang H, Ludwig B, Mann K, Hacker J. Mip protein of Legionella pneumophila exhibits peptidyl-prolyl-cis/trans isomerase (PPIase) activity. Mol Microbiol. 1992;6:1375–1383. doi: 10.1111/j.1365-2958.1992.tb00858.x. [DOI] [PubMed] [Google Scholar]
  • 11.Fraser D W, Tsai T R, Orenstin W, Parken W E, Beechan H J, Sharrar R G, Harris J, Mallison G F, Martin S M, McDade J E, Shepard C C, Brachman P S. Legionnaires’ disease: description of an epidemic of pneumonia. N Engl J Med. 1977;297:1189–1197. doi: 10.1056/NEJM197712012972201. [DOI] [PubMed] [Google Scholar]
  • 12.Goodrich J A, Schwartz M L, McClure W R. Searching for and predicting the activity of sites for DNA binding proteins: compilation and analysis of the binding sites for Escherichia coli integration host factor (IHF) Nucleic Acids Res. 1990;18:4993–5000. doi: 10.1093/nar/18.17.4993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Horwitz M A. Formation of a novel phagosome by the Legionnaires’ disease bacterium (Legionella pneumophila) in human monocytes. J Exp Med. 1983;158:1319–1331. doi: 10.1084/jem.158.4.1319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Horwitz M A. The Legionnaires’ disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J Exp Med. 1983;158:2108–2126. doi: 10.1084/jem.158.6.2108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Horwitz M A. Phagocytosis of the Legionnaires’ disease bacterium (Legionella pneumophila) occurs by a novel mechanism: engulfment within a pseudopod coil. Cell. 1984;36:27–33. doi: 10.1016/0092-8674(84)90070-9. [DOI] [PubMed] [Google Scholar]
  • 16.Horwitz M A, Silverstein S C. Legionnaires’ disease bacterium (Legionella pneumophila) multiplies intracellularly in human monocytes. J Clin Invest. 1980;60:441–450. doi: 10.1172/JCI109874. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16a.Horwitz M A, Silverstein S C. Intracellular multiplication of Legionnaires’ disease bacteria (Legionella pneumophila) in human monocytes is reversibly inhibited by erythromycin and rifampin. J Clin Invest. 1983;71:15–26. doi: 10.1172/JCI110744. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Imai Y, Matsushima Y, Sugimura T, Terada M. A simple and rapid method for generating a deletion by PCR. Nucleic Acids Res. 1991;19:2785. doi: 10.1093/nar/19.10.2785. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Karlin S, Altschul S F. Methods for assessing the statistical significance of molecular sequence features by using general scoring schemes. Proc Natl Acad Sci USA. 1990;87:2264–2268. doi: 10.1073/pnas.87.6.2264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Kaufmann A F, McDade J E, Patton C M, Bennett J V, Skaliy P, Feeley J C, Anderson D C, Potter M E, Newhouse V F, Gregg M B, Brachman P S. Pontiac fever: isolation of the etiologic agent (Legionella pneumophila) and demonstration of its mode of transmission. Am J Epidemiol. 1981;114:337–347. doi: 10.1093/oxfordjournals.aje.a113200. [DOI] [PubMed] [Google Scholar]
  • 20.Kyte J, Doolittle R F. A simple method for displaying the hydropathic character of a protein. J Mol Biol. 1982;157:105–132. doi: 10.1016/0022-2836(82)90515-0. [DOI] [PubMed] [Google Scholar]
  • 21.Maniatis T, Fritsch E F, Sambrook J. Molecular cloning: a laboratory manual. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1982. [Google Scholar]
  • 22.Marra A, Blander S J, Horwitz M A, Shuman H A. Identification of a Legionella pneumophila locus required for intracellular multiplication in human macrophages. Proc Natl Acad Sci USA. 1992;89:9607–9611. doi: 10.1073/pnas.89.20.9607. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Marra A, Horwitz M A, Shuman H A. The HL-60 model for the interaction of human macrophages with the Legionnaires’ disease bacterium. J Immunol. 1990;144:2738–2744. [PubMed] [Google Scholar]
  • 24.Miller J H. Experiments in molecular biology. Cold Spring Harbor, N.Y: Cold Spring Harbor Laboratory; 1972. [Google Scholar]
  • 25.Morales V M, Backman A, Bagdasarian M. A series of wide-host-range low-copy-number vectors that allow direct screening for recombinants. Gene. 1991;97:39–47. doi: 10.1016/0378-1119(91)90007-x. [DOI] [PubMed] [Google Scholar]
  • 26.Nakai K, Kanehisa M. Expert system for prediction of protein localization sites in gram negative bacteria. Protein Struct Funct Genet. 1991;11:95–110. doi: 10.1002/prot.340110203. [DOI] [PubMed] [Google Scholar]
  • 27.Roy C R, Isberg R R. Topology of Legionella pneumophila DotA: an inner membrane protein required for replication in macrophages. Infect Immun. 1997;65:571–578. doi: 10.1128/iai.65.2.571-578.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Sadosky A B, Wiater L A, Shuman H A. Identification of Legionella pneumophila genes required for growth within and killing of human macrophages. Infect Immun. 1993;61:5361–5373. doi: 10.1128/iai.61.12.5361-5373.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28a.Saier, M. H., Jr. Personal communication.
  • 29.Segal G, Purcell M, Shuman H A. Host cell killing and bacterial conjugation require overlapping sets of genes within a 22-kb region of the Legionella pneumophila genome. Proc Natl Acad Sci USA. 1998;95:1669–1674. doi: 10.1073/pnas.95.4.1669. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Segal G, Shuman H A. Characterization of a new region required for macrophage killing by Legionella pneumophila. Infect Immun. 1997;65:5057–5066. doi: 10.1128/iai.65.12.5057-5066.1997. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Swanson M S, Isberg R R. Association of Legionella pneumophila with the macrophage endoplasmic reticulum. Infect Immun. 1995;63:3609–3620. doi: 10.1128/iai.63.9.3609-3620.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Wiater L A, Sadosky A B, Shuman H A. Mutagenesis of Legionella pneumophila using Tn903dIIlacZ: identification of a growth-phase-regulated pigmentation gene. Mol Microbiol. 1994;11:641–653. doi: 10.1111/j.1365-2958.1994.tb00343.x. [DOI] [PubMed] [Google Scholar]
  • 32a.Wiater, L. A., K. Dunn, F. R. Maxfield, and H. A. Shuman. Unpublished data.
  • 33.Woodcock D M, Crowther P J, Doherty J, Jefferson S, DeCruz E, Nayer-Weidner M, Smith S S, Michael M Z, Graham M W. Quantitation evaluation of Escherichia coli for tolerance of cytosine methylation in plasmid and phage recombination. Nucleic Acids Res. 1989;9:3469–3478. doi: 10.1093/nar/17.9.3469. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Infection and Immunity are provided here courtesy of American Society for Microbiology (ASM)

RESOURCES