FEMS Microbiology Lelters 93 (1~92) 240-254 © 1992 Federation of Eur~pean Microbiological Societies 0378-1tl97/92/505.(~1 Published by Elsevier
249
FEMSLE (1491)7
Isolation and characterization of a lipopolysaccharide mutant of Legionellapneurnophila Clifford S. Mintz and Chang Hua Zou Department of Microbiolo~' and bnmuno[ogy, Unit'ersity of Miami SchtxJI of Medicine, Mi.mi, FL, USA Received 2 January 1992 Accepted 18 March 1902
Key words: Legionella pneumophila; Lipopolysaccharide mutant; Intraeellular growth
1. SUMMARY A mutant unable to bind a monoclonal antibody (mAb 1E6) directed against serogroup l lipopolysaccharide (LPS) was isolated from L. pneumophila strain Philadelphia-1. SDS-PAGE analysis of isolated LPS from the mutant and wild type revealed that there were no obvious structural differences between the two LPS. The results from Western-blot experiments showed that the mutant LPS was unable to bind mAb 1E6 but retained the ability to bind polyclonal serogroup 1 antibodies. Loss of the LPS epitope recognized by mAb 1E6 did not alter the ability of the mutant to multiply in human monocyte-like U937 cells. Also, the mutant, like wild type, was resistant to killing by normal human serum. These results show that a minor change in the antigenic composition of serogroup 1 LPS has no effect on the virulence properties of strain Philadelphia-l. Additionally, this mutant may be useful for
Corre.~pondence to: C.S. Minlz, Department of Microbiology and Immunology, University of Miami School of Medicine. P.O. Box 016960 {R-138), Miami, Florida 33101. USA.
molecular genetic analysis of serogroup 1 LPS biosynthesis and assembly. 2. INTRODUCTION
Legionella pneumophila is a Gram-negative intracellular pathogen capable of entering and growing in alveolar macrophages and monocytes [1]. Recent evidence [2,3] suggests that opsonic complement (C) component C3 and corresponding phagocyte C receptors CR1 and CR3 facilitate the uptake of L. pneumophila by monocytes. Although it was demonstrated that incubation of L. pneumophila in normal human serum (NHS) results in the activation of C and the deposition of C3 on the Legionella cell surface [3], the bacterial ligand responsible for C activation remains to be identified. Recently, we showed that iipopolysaccharide (LPS) isolated from L. pneumophila strain Philadelphia 1 can activate C in NHS (C. Mintz, unpublished results). This suggested that LPS plays an important role in the uptake of L. pneumophila by mononuclear phagocytes. The isolation of mutants that produce altered LPS has contributed to the understanding of the role of LPS in the pathogenesis of a variety of
25t~ Gram-negative infections. In general, structural changes in LPS significantly alter the interaction between Gram-negative pathogens and targethost cells. For example, rough mutants of Sahnom'lla typhimum#n arc phagocytosed and killed more efficiently than wild type by mononuclear phagocytes [4,5]. Therefore, in order to more clearly define the role of LPS in the L. pneumophila-monocyte interaction, wc sought to isolate LPS mutants from L. imemnophila, in this paper, we report the isolation and characterization of a mutant from L. pneumopMa strain Philadelphia-I that produces an antigenically altered LPS.
3. MATERIALS AND METHODS
3.1. Bacterial strains and growth conditions L. pmmmophila strain Philadelphia-1 is a virulent clinical isolate that produces serogroup 1 LPS. Strain AM511 is a streptomycin-resistant, restriction minus isogeneie mutant of strain Philadelphia-I [6]. L. pnettmopMa was routinely cultivated on ACES-buffered charcoal-yeast-extract platcs (ABCYE) or in albumin yeast extract broth as previously described [7]. When necessary antibiotics were added at the following concentrations: kanamycin (Km), 25/zg/ml and streptomycin (Sin), 51)/ag/ml. 3.9... Mutagenesis of strain AM5I I with bacteriophage Mu Bacteriophage Mu was introduced into L. pneumophila strain AM511 according to the methods of Mintz and Shuman [7]. Plasmid pRK212 (which carries Mu cts62)was transferred by conjugation from E. coil into L. pneumophila strain AM511 at 30°C as previously described [7]. Transconjugants were removed from selection plates, combined together, and grown overnight at 30°C in AYE broth. After incubation, the culture ~vas serially diluted and plated for single colonies on ABCYE agar that contained Km and Sin. These colonies were replica plated onto nitrocellulose filters (MSI) and tested for the ability to bind monoclonal antibody (mAb) IE6 in a
colony immunoblot assay described by Spinola et al. [8].
3.3. Sot ahem hybridization Chromosomal DNA was isolated from mutant and wild-type legionellae, digested with the restriction endonuclease Bglll and electrophoresed thlough 0.8% agarose gels. The DNA was transferred to nitrocellulose paper and probed with ~"P-labeled Mu-containing plasmid DNA as previously described [7]. 3.4. Isolation and characterization of L. pneumophila LPS LPS was isolated from L. pneumophila by the EDTA-Folch method described by Otten et ai. [9]. Isolated LPS was analyzed by SDS-PAGE in 10% polyacrylamide gels (BioRad) followed by silver staining [9]. In Western-blot experiments, LPS was electrophoresed in 11)% acrylamide gels and transferred to nitrocellulose paper (MSI) at constant voltage (225 V) for 40 rain at 4°C in a BioRad Transblot apparatus. After transfer, blots were incubated with mAb I E6 (1:500 dilution) or monospecific serogroup 1 antiserum (1:1000 dilution) in Tris-buffered saline (TBS, 50 mM Tris. HCI, 150 mM NaCl, pH 7.5) containing 5% nonfat milk (Carnation) for 1 h at room temperature. The blots were washed several times in TBS and incubated with horse radish peroxidase-conjugated goat-anti-mouse or goat-anti-rabbit immunoglobulin (1:1000 dilution, CappeD. Color was developed by immersing blots in TBSmethanol (5: 1, vol/vol) containing 0.05% 4chloro-l-naphthol (Sigma) and 0.015% hydrogen peroxide. 3.5. Serum bactericidal assays The ability of L. pneumophila to resist killing by normal human serum (NHS)was determined by incubating 10I' legionellae in pooled 50% NHS for I h at 37°C. At 0 and 60 rain, serial dilutions of the mixture were made and bacteria were plated in duplicate on ABCYE agar plates for enumeration of viable legionellae. Percentage of serum resistance was calculated by dividing the number of bacteria at zero time by the number of bacteria obtained at 60 rain multiplied by 100.
25t
3.6. Infection of U937 cell monolayers The human monocyte-likc U937 cell line was infected with legionellae as previously described by King et al. [10]. At daily intervals, samples were removed from the infected monolayers contained in 24-well tissue-culture dishes (Falcon), serially diluted in phosphate-buffered saline and plated on ABCYE agar to determine the number of viable legionellae per ml.
4. RESULTS AND DISCUSSION
L. pneumophila LPS has a structural motif similar to enterobacterial LPS, i.e. it contains lipid A, a core oligosaccharide and a possible O-polysaccharide designated the serogroup antigen [9,11]. It has been clearly established that the LPS serogroup antigens define the serogroup specificity of L. pneumophila [12]. Following mutagenesis with bacteriophage Mu, we isolated a mutant from strain AM511, that was unable to bind a serogroup 1 LPS specific mAb (1E6). The mutant, subsequently designated strain CS280, was found after screening approximately 2500 colonies. LPS was isolated from strains AM5II and CS280 and analyzed by SDS-PAGE followed by silver staining. There was no discernable difference in the electrophoretic profiles of the mutant and wild-type LPS (data not shown). Both LPS exhibited the ladder-like banding pattern characteristic of L. pneumophila LPS [9,12]. This suggested that there were no obvious structural differences between the two LPS. The results from Western-blot experiments revealed that isolated CS280 LPS did not bind mAb 1E6. In contrast, LPS from strain CS280 was reactive with polyclonal serogroup 1 serum (Fig. 1). Wild-type LPS bound both mAb 1E6 and serogroup 1 serum. These results indicated that the mutant LPS was structurally similar to wild-type LPS but lacked the epitope recognized by mAb 1E6, Interestingly, Southern-hybridization experiments revealed that strain CS280 did not contain a chromosomal Mu insertion (data not shown). "Ibis suggested that the inability of strain CS280 to bind mAb 1E6 may have resulted from a sponta-
A
B ': :i::
1234
123
Fig. 1. Western-bh~t analysis eft wild-type and mutant LPS. LPS was etectropht~resed in 10% SDS-PAGE gels. transferred to nitrocellulose and probed with mAb left (A) ¢~r pCJlycklnal scrogroup I antiserum (BJ. A. Lanes: i. Philadelphia.l I,?S; 2, AMS11 LPS; 3, CS28{I LPS; 4, Bk×~minglon-2 LPS. B, Lanes: 1, AMSil LPS: 2. CS28~) LPS; 3, BlCmmingt~m-2 LPS. Strain Bloomington-2 LPS ~'as included as a negative control, This strain produces serogr~up 3 LPS which &~es n~t bind rnAb I Eft or polyclonal serogroup 1 antiserum,
neous mutation in a gene(s) involved in serogroup 1 LPS biosynthesis rather than from an insertion mutation caused by bacteriophage Mu. Alternatively, a DNA rearrangement or deletion within a serogroup t LPS gene (s) caused by the imprecise excision of Mu from the CS280 chromosome could also explain the inability of the mutant to bind mAb IE6. A common plciotropic effect of UPS mutations is an increased susceptibility of LPS mutants, as compared with wild type, to C-mediated killing in NHS [4]. To determine if loss of the LPS epitope recognized by mAb IE6 affected the serum resistance of strain CS280, we performed serum bactericidal assays with strains AM511 and CS280. The results from these experiments showed that there was no significant difference in the ability of strains AM51I and CS280 to resist killing during incubation in 50% NHS. The percent resistance to serum of strain AM511 was determined to be 65% whereas that of strain CS280 was 48%. This indicated that loss of the mAb 1E6-specific LPS epitope did not significantly altcr the serum resistant phenotype of strain CS280. To determine if the LPS epitope recognized by
252 8-
7-
=J IlL O O
sO
4
3
0
,
-
,
1
2
3
DAYS POST INFECTION
Fig. 2. infection of U937 cells with strains AM51} and CS280. U937 cell monolayers were infected with strains AM51| and CS2gOas described in MATERI,M2,iANDMETHODS.Each ooin~ represents the mean_-L-SEfor three separate U937 cell cultures. Closed triangles represent strain C5280, open tri.':ngfes strain AM51I.
that the inability of CS280 LPS to bind mAb 1E6 resulted from a conformational change in the mutant LPS. Alternatively, there may be differences in the carbohydrate content of CS280 LPS as compared with wild type. We are currently analyzing the chemical composition and structure of CS280 and AM511 LPS to explore this possibility. To our knowledge, we are the first to report the isolation of an LPS mutant from L. pneumophila. Recently, using strain CS280 in complementation studies, we identified recombinant cosmids from libraries containing Philadelphia-1 genomic DNA fragments that restore the ability of the mutant to bind mAb I E6 (C. Mintz, unpublished results), This demonstrates that strain CS280 will be useful for the identification of genes involved in the biosynthesis and assembly of the L. pneumophila serogroup 1 LPS antigen.
ACKNOWLEDGEMENTS
mAb IE6 contributed to the intracellular growth of L. pneumophila, we compared strain CS280 with strain AM511 for the ability to grow in monocyte-like U937 cells. The results from these experiments indicated that loss of the mAb IE6. specific epitope from serogroup 1 LPS did not interfere with the intracellular multiplication of strain CS280 in U937 cells (Fig. 2). Recently, we determined that strain CS280 is also unaltered in its ability to grow intracellularly in the amoeba Hartmannella t'ermiformis (C. Mintz, unpublished results). it is welI-documented that enterobaeterial LPS mutants that produce LPS devoid of O-antigen or LPS molecules that contain incomplete core structures are less virulent than wild type [4,5], As previously mentioned, SDS-PAGE analysis indicated that the LPS produced by strain CS280 was structurally indistinguishable from wild type, Therefore, in the absence of any obvious structural defects, it was not surprising that strain CS280 was unaltered in its ability to resist killing by NHS or to multiply in U937 ceils. It is possible
We thank William Johnson for his helpful discussions concerning L. pneumophila LPS and supplying us with mAb 1E6. This work was supported by a grant from the American Lung Association of Florida awarded to C. Mintz.
REFERENCES [1] Horwitz, M.A. ,'i981,')J. Clin. Micro. 66, 441-450. [2] Payne, N.R. and Ho,rwilz, M.A. (1987)J. Exp. Mcd. 166, 1377-1389. [3] Bellinger-Kawahara, C. and Horwitz, M.A. (lYe) J. Exp. Med. 172, 1201-1210. [4] Makela, P.K. and Stocker, B.A.D. (1984) In: Handbook of endotoxins (Proctor, R.A., Ed.), Volume 1, Chemistry of endotoxin, (Riet~hel, E.Th,, Ed,), pp. 5'.1-119, Elsevier, Amsterdam. [5] Friedberg, D. and Shilo, M. (1970) Infect. Immun. 2, 279-285. [6] Marra, A. and Shuman, H.A. (1989) J. BacterioL 171, 2238 -224)}. [71 Mintz, C.S. and Shuman, H.A. (lq87) Prec. Natl. Acad. Sci. USA 84, 4645-4669, [8] Spinola, S.M., KwaiL Y.A,, Lessc, A.J., Campagnari,
253 A.A. and Apicell:l, M.A. {19~1) Infect. Immun. 58, 1551tISM. [g] Olten, S., i~er, S., Johnson, W. :rod Montgomery. R. (19861 J. Bacteriot. 167. 893-*01}4. [10] King. C.II., Fields, B.S., Shotls, E.B. and White, E.II. lltJgl) ]hi'cot. lmmun. 59, "/58-763.
[I 11 Sonc,;.~)n, A., Jantzcn, [-,, Bryn, K., ~rstm. L. and F.ng. J. I ]lJ~9) Arch. Micmbiot. 153, 72-7~. [12] Conk, n. J.W. and Ash~nrth, LA.E. (19861 J. tly~. ('~,mb. 96, 31)-4~.
249
FEMSLE (1491)7
Isolation and characterization of a lipopolysaccharide mutant of Legionellapneurnophila Clifford S. Mintz and Chang Hua Zou Department of Microbiolo~' and bnmuno[ogy, Unit'ersity of Miami SchtxJI of Medicine, Mi.mi, FL, USA Received 2 January 1992 Accepted 18 March 1902
Key words: Legionella pneumophila; Lipopolysaccharide mutant; Intraeellular growth
1. SUMMARY A mutant unable to bind a monoclonal antibody (mAb 1E6) directed against serogroup l lipopolysaccharide (LPS) was isolated from L. pneumophila strain Philadelphia-1. SDS-PAGE analysis of isolated LPS from the mutant and wild type revealed that there were no obvious structural differences between the two LPS. The results from Western-blot experiments showed that the mutant LPS was unable to bind mAb 1E6 but retained the ability to bind polyclonal serogroup 1 antibodies. Loss of the LPS epitope recognized by mAb 1E6 did not alter the ability of the mutant to multiply in human monocyte-like U937 cells. Also, the mutant, like wild type, was resistant to killing by normal human serum. These results show that a minor change in the antigenic composition of serogroup 1 LPS has no effect on the virulence properties of strain Philadelphia-l. Additionally, this mutant may be useful for
Corre.~pondence to: C.S. Minlz, Department of Microbiology and Immunology, University of Miami School of Medicine. P.O. Box 016960 {R-138), Miami, Florida 33101. USA.
molecular genetic analysis of serogroup 1 LPS biosynthesis and assembly. 2. INTRODUCTION
Legionella pneumophila is a Gram-negative intracellular pathogen capable of entering and growing in alveolar macrophages and monocytes [1]. Recent evidence [2,3] suggests that opsonic complement (C) component C3 and corresponding phagocyte C receptors CR1 and CR3 facilitate the uptake of L. pneumophila by monocytes. Although it was demonstrated that incubation of L. pneumophila in normal human serum (NHS) results in the activation of C and the deposition of C3 on the Legionella cell surface [3], the bacterial ligand responsible for C activation remains to be identified. Recently, we showed that iipopolysaccharide (LPS) isolated from L. pneumophila strain Philadelphia 1 can activate C in NHS (C. Mintz, unpublished results). This suggested that LPS plays an important role in the uptake of L. pneumophila by mononuclear phagocytes. The isolation of mutants that produce altered LPS has contributed to the understanding of the role of LPS in the pathogenesis of a variety of
25t~ Gram-negative infections. In general, structural changes in LPS significantly alter the interaction between Gram-negative pathogens and targethost cells. For example, rough mutants of Sahnom'lla typhimum#n arc phagocytosed and killed more efficiently than wild type by mononuclear phagocytes [4,5]. Therefore, in order to more clearly define the role of LPS in the L. pneumophila-monocyte interaction, wc sought to isolate LPS mutants from L. imemnophila, in this paper, we report the isolation and characterization of a mutant from L. pneumopMa strain Philadelphia-I that produces an antigenically altered LPS.
3. MATERIALS AND METHODS
3.1. Bacterial strains and growth conditions L. pmmmophila strain Philadelphia-1 is a virulent clinical isolate that produces serogroup 1 LPS. Strain AM511 is a streptomycin-resistant, restriction minus isogeneie mutant of strain Philadelphia-I [6]. L. pnettmopMa was routinely cultivated on ACES-buffered charcoal-yeast-extract platcs (ABCYE) or in albumin yeast extract broth as previously described [7]. When necessary antibiotics were added at the following concentrations: kanamycin (Km), 25/zg/ml and streptomycin (Sin), 51)/ag/ml. 3.9... Mutagenesis of strain AM5I I with bacteriophage Mu Bacteriophage Mu was introduced into L. pneumophila strain AM511 according to the methods of Mintz and Shuman [7]. Plasmid pRK212 (which carries Mu cts62)was transferred by conjugation from E. coil into L. pneumophila strain AM511 at 30°C as previously described [7]. Transconjugants were removed from selection plates, combined together, and grown overnight at 30°C in AYE broth. After incubation, the culture ~vas serially diluted and plated for single colonies on ABCYE agar that contained Km and Sin. These colonies were replica plated onto nitrocellulose filters (MSI) and tested for the ability to bind monoclonal antibody (mAb) IE6 in a
colony immunoblot assay described by Spinola et al. [8].
3.3. Sot ahem hybridization Chromosomal DNA was isolated from mutant and wild-type legionellae, digested with the restriction endonuclease Bglll and electrophoresed thlough 0.8% agarose gels. The DNA was transferred to nitrocellulose paper and probed with ~"P-labeled Mu-containing plasmid DNA as previously described [7]. 3.4. Isolation and characterization of L. pneumophila LPS LPS was isolated from L. pneumophila by the EDTA-Folch method described by Otten et ai. [9]. Isolated LPS was analyzed by SDS-PAGE in 10% polyacrylamide gels (BioRad) followed by silver staining [9]. In Western-blot experiments, LPS was electrophoresed in 11)% acrylamide gels and transferred to nitrocellulose paper (MSI) at constant voltage (225 V) for 40 rain at 4°C in a BioRad Transblot apparatus. After transfer, blots were incubated with mAb I E6 (1:500 dilution) or monospecific serogroup 1 antiserum (1:1000 dilution) in Tris-buffered saline (TBS, 50 mM Tris. HCI, 150 mM NaCl, pH 7.5) containing 5% nonfat milk (Carnation) for 1 h at room temperature. The blots were washed several times in TBS and incubated with horse radish peroxidase-conjugated goat-anti-mouse or goat-anti-rabbit immunoglobulin (1:1000 dilution, CappeD. Color was developed by immersing blots in TBSmethanol (5: 1, vol/vol) containing 0.05% 4chloro-l-naphthol (Sigma) and 0.015% hydrogen peroxide. 3.5. Serum bactericidal assays The ability of L. pneumophila to resist killing by normal human serum (NHS)was determined by incubating 10I' legionellae in pooled 50% NHS for I h at 37°C. At 0 and 60 rain, serial dilutions of the mixture were made and bacteria were plated in duplicate on ABCYE agar plates for enumeration of viable legionellae. Percentage of serum resistance was calculated by dividing the number of bacteria at zero time by the number of bacteria obtained at 60 rain multiplied by 100.
25t
3.6. Infection of U937 cell monolayers The human monocyte-likc U937 cell line was infected with legionellae as previously described by King et al. [10]. At daily intervals, samples were removed from the infected monolayers contained in 24-well tissue-culture dishes (Falcon), serially diluted in phosphate-buffered saline and plated on ABCYE agar to determine the number of viable legionellae per ml.
4. RESULTS AND DISCUSSION
L. pneumophila LPS has a structural motif similar to enterobacterial LPS, i.e. it contains lipid A, a core oligosaccharide and a possible O-polysaccharide designated the serogroup antigen [9,11]. It has been clearly established that the LPS serogroup antigens define the serogroup specificity of L. pneumophila [12]. Following mutagenesis with bacteriophage Mu, we isolated a mutant from strain AM511, that was unable to bind a serogroup 1 LPS specific mAb (1E6). The mutant, subsequently designated strain CS280, was found after screening approximately 2500 colonies. LPS was isolated from strains AM5II and CS280 and analyzed by SDS-PAGE followed by silver staining. There was no discernable difference in the electrophoretic profiles of the mutant and wild-type LPS (data not shown). Both LPS exhibited the ladder-like banding pattern characteristic of L. pneumophila LPS [9,12]. This suggested that there were no obvious structural differences between the two LPS. The results from Western-blot experiments revealed that isolated CS280 LPS did not bind mAb 1E6. In contrast, LPS from strain CS280 was reactive with polyclonal serogroup 1 serum (Fig. 1). Wild-type LPS bound both mAb 1E6 and serogroup 1 serum. These results indicated that the mutant LPS was structurally similar to wild-type LPS but lacked the epitope recognized by mAb 1E6, Interestingly, Southern-hybridization experiments revealed that strain CS280 did not contain a chromosomal Mu insertion (data not shown). "Ibis suggested that the inability of strain CS280 to bind mAb 1E6 may have resulted from a sponta-
A
B ': :i::
1234
123
Fig. 1. Western-bh~t analysis eft wild-type and mutant LPS. LPS was etectropht~resed in 10% SDS-PAGE gels. transferred to nitrocellulose and probed with mAb left (A) ¢~r pCJlycklnal scrogroup I antiserum (BJ. A. Lanes: i. Philadelphia.l I,?S; 2, AMS11 LPS; 3, CS28{I LPS; 4, Bk×~minglon-2 LPS. B, Lanes: 1, AMSil LPS: 2. CS28~) LPS; 3, BlCmmingt~m-2 LPS. Strain Bloomington-2 LPS ~'as included as a negative control, This strain produces serogr~up 3 LPS which &~es n~t bind rnAb I Eft or polyclonal serogroup 1 antiserum,
neous mutation in a gene(s) involved in serogroup 1 LPS biosynthesis rather than from an insertion mutation caused by bacteriophage Mu. Alternatively, a DNA rearrangement or deletion within a serogroup t LPS gene (s) caused by the imprecise excision of Mu from the CS280 chromosome could also explain the inability of the mutant to bind mAb IE6. A common plciotropic effect of UPS mutations is an increased susceptibility of LPS mutants, as compared with wild type, to C-mediated killing in NHS [4]. To determine if loss of the LPS epitope recognized by mAb IE6 affected the serum resistance of strain CS280, we performed serum bactericidal assays with strains AM511 and CS280. The results from these experiments showed that there was no significant difference in the ability of strains AM51I and CS280 to resist killing during incubation in 50% NHS. The percent resistance to serum of strain AM511 was determined to be 65% whereas that of strain CS280 was 48%. This indicated that loss of the mAb 1E6-specific LPS epitope did not significantly altcr the serum resistant phenotype of strain CS280. To determine if the LPS epitope recognized by
252 8-
7-
=J IlL O O
sO
4
3
0
,
-
,
1
2
3
DAYS POST INFECTION
Fig. 2. infection of U937 cells with strains AM51} and CS280. U937 cell monolayers were infected with strains AM51| and CS2gOas described in MATERI,M2,iANDMETHODS.Each ooin~ represents the mean_-L-SEfor three separate U937 cell cultures. Closed triangles represent strain C5280, open tri.':ngfes strain AM51I.
that the inability of CS280 LPS to bind mAb 1E6 resulted from a conformational change in the mutant LPS. Alternatively, there may be differences in the carbohydrate content of CS280 LPS as compared with wild type. We are currently analyzing the chemical composition and structure of CS280 and AM511 LPS to explore this possibility. To our knowledge, we are the first to report the isolation of an LPS mutant from L. pneumophila. Recently, using strain CS280 in complementation studies, we identified recombinant cosmids from libraries containing Philadelphia-1 genomic DNA fragments that restore the ability of the mutant to bind mAb I E6 (C. Mintz, unpublished results), This demonstrates that strain CS280 will be useful for the identification of genes involved in the biosynthesis and assembly of the L. pneumophila serogroup 1 LPS antigen.
ACKNOWLEDGEMENTS
mAb IE6 contributed to the intracellular growth of L. pneumophila, we compared strain CS280 with strain AM511 for the ability to grow in monocyte-like U937 cells. The results from these experiments indicated that loss of the mAb IE6. specific epitope from serogroup 1 LPS did not interfere with the intracellular multiplication of strain CS280 in U937 cells (Fig. 2). Recently, we determined that strain CS280 is also unaltered in its ability to grow intracellularly in the amoeba Hartmannella t'ermiformis (C. Mintz, unpublished results). it is welI-documented that enterobaeterial LPS mutants that produce LPS devoid of O-antigen or LPS molecules that contain incomplete core structures are less virulent than wild type [4,5], As previously mentioned, SDS-PAGE analysis indicated that the LPS produced by strain CS280 was structurally indistinguishable from wild type, Therefore, in the absence of any obvious structural defects, it was not surprising that strain CS280 was unaltered in its ability to resist killing by NHS or to multiply in U937 ceils. It is possible
We thank William Johnson for his helpful discussions concerning L. pneumophila LPS and supplying us with mAb 1E6. This work was supported by a grant from the American Lung Association of Florida awarded to C. Mintz.
REFERENCES [1] Horwitz, M.A. ,'i981,')J. Clin. Micro. 66, 441-450. [2] Payne, N.R. and Ho,rwilz, M.A. (1987)J. Exp. Mcd. 166, 1377-1389. [3] Bellinger-Kawahara, C. and Horwitz, M.A. (lYe) J. Exp. Med. 172, 1201-1210. [4] Makela, P.K. and Stocker, B.A.D. (1984) In: Handbook of endotoxins (Proctor, R.A., Ed.), Volume 1, Chemistry of endotoxin, (Riet~hel, E.Th,, Ed,), pp. 5'.1-119, Elsevier, Amsterdam. [5] Friedberg, D. and Shilo, M. (1970) Infect. Immun. 2, 279-285. [6] Marra, A. and Shuman, H.A. (1989) J. BacterioL 171, 2238 -224)}. [71 Mintz, C.S. and Shuman, H.A. (lq87) Prec. Natl. Acad. Sci. USA 84, 4645-4669, [8] Spinola, S.M., KwaiL Y.A,, Lessc, A.J., Campagnari,
253 A.A. and Apicell:l, M.A. {19~1) Infect. Immun. 58, 1551tISM. [g] Olten, S., i~er, S., Johnson, W. :rod Montgomery. R. (19861 J. Bacteriot. 167. 893-*01}4. [10] King. C.II., Fields, B.S., Shotls, E.B. and White, E.II. lltJgl) ]hi'cot. lmmun. 59, "/58-763.
[I 11 Sonc,;.~)n, A., Jantzcn, [-,, Bryn, K., ~rstm. L. and F.ng. J. I ]lJ~9) Arch. Micmbiot. 153, 72-7~. [12] Conk, n. J.W. and Ash~nrth, LA.E. (19861 J. tly~. ('~,mb. 96, 31)-4~.
