- Email: [email protected]
Electron microscopic investigation of intracellular events after ingestion of Rhodococcus equi by foal alveolar macrophages
.Veterinary Microbiology, 14 (1987) 295-305 Elsevier Science Publishers B.V., Amsterdam - - Printed in The Netherlands
295
Electron Microscopic Investigation of Intracellular Events after Ingestion of Rhodococcus equi by Foal Alveolar Macrophages M.C. Z I N K la,J.A. Y A G E R
I,J.F. P R E S C O T T
2 and M.A. F E R N A N D O
l
Department of Patholog~ and Veterinary Microbiology and Irnnrnunology~,Ontario Veterinary College, Universityof Guelph, Guelph, Ontario NIG 2 W1 (Canada)
ABSTRACT Zink, M.C., Yager, J.A., Prescott, J.F. and Fernando, M.A., 1987. Electron microscopic investigation of intracellular events after ingestion of Rhodococeus equi by foal alveolar macrophages. Vet Microbiol., 14: 295-305. It has been suggested that R. equi causes pulmonary disease in foals by persisting within the lung as a facultative intmcellular parasite of alveolar macrophages. This paper describes an uRra-
structural study of the intracellular events after ingestion ofR. equi by fos~ alveolar macrophages, in an attempt to determine the mechanism of intracellular survival of R. equi. Secondary lysoeomes of alveolar macrophages recovered from foals by bronchoalveoler lavage were labelled with electron-dense ferritin, and the ceils were challenged with either viable or formalin-ktIed R. equi. After 0-, 3-, 8- or 24-h incubation, the cells were fixed and processed for electron microscopy. There was no evidence of phagosome-lysosome fusion after ingestion of either viable or non-viable R. equi by foal alveolar macrophages. Rhodococcus equi persisted and multiplied within dilated phegosomes, which were often lined by elongate microvilleus structures. After 24-h incubation, 75% of the ingested bacteria were still structurally intact. Macrophages with ingested viable R. equ/were irreversibly d~maged and released intracellular bacteria into the surrounding medium. These data confirm that R~ equi is a facultative intracellular parasite of foal alveolar macrophages and is able to persist and multiply within the phagosome, apparently inhibiting phagosome-lysosome fusion by some as yet unknown mechanism.
INTRODUCTION
Several investigators have suggested that Rhodococcua equi is a facultative intracellular parasite of the foal alveolar macrophage (Wilson, 1955; Knight, 1969; Elissalde et al., 1980; Johnson et al., 1983). Circumstantial evidence for this hypothesis includes the granulomatous nature of the pulmonary lesions of R. equi infection (Johnson et al., 1983), the tendency for horses exposed to R. •Author to whom correspondence should be addressed at: Department of Neurology, Meyer 6-181, Johns Hopkins University School of Medicine, 600 N. Wolfe St., Baltimore, MD 21205, U.S.A.
0378-1135/87/$03.50
© 1987 Elsevier Science Publishers B.V.
296
equi to develop cell-mediated immune responses (Wilson, 1955; Prescott, et al., 1979; Ellenberger et al., 1984) and similarity in bacterial cell wall structure of R. equi to well recognized intracellular parasites from the Mycobacterium, Nocardia and Bacterionema genera (Brennan and Lehane, 1971; Goodfellow et al., 1976; Stackebrandt et al., 1980). Further evidence has been provided by studies on the bactericidal function of foal alveolar macrophages, in which it was shown that ~ 75% of R. equi remain viable 4 h after ingestion by foal alveolar macrophages (Zink et al., 1985). Thus alveolar macrophages, which function both as the first line of defense against inhaled bacteria and as effector cells for pulmonary cell-mediated immunity (Hocking and Golde, 1979), are probably of central importance in the pathogenesis of R. equi pneumonia in foals. The purpose of this study was to characterize the intracellular events occurring within foal alveolar macrophages during the first 24 h after ingestion of R. equi in order to identify the mechanism of intracellular survival of this bacterium. MATERIALSAND METHODS
Collection and labelling of alveolar macrophages Bronchoalveolar lavage was performed on foals under general anesthetic and the recovered cells were washed and counted as described elsewhere (Zink and Johnson, 1984). The secondary lysosomes of the alveolar macrophages were labelled with electron-dense ferritin (Sigma Chemical Co., St. Louis, MO) by incubating the cells in RPMI-1640 containing 2.5% fetal bovine serum (FBS) with or without 20 mg m - 1 ferritin for 4 h. Cell suspensions were then washed 3 times with R P M I containing 2.5% FBS, incubated for a further 3 h to allow complete uptake of residual ferritin and resuspended at a final concentration of 2.0 × 10e cells m l - 1.
Preparation of bacteria Viable and non-viable bacterialsuspensions were prepared by culturingR. equi on Mueller-Hinton agar for 36 h then suspending the bacteria in phosphate-buffered saline (PBS) with or without 0.5% formalin for 1 h. After 4 washes in PBS, the bacteria were suspended at a concentration of 2.0X 107 ml-i in R P M I containing 10% normal horse serum.
Electron microscopy Suspensions (4 ml) of ferritin-labeUedor unlabelledmacrophages werecentrifugedat 400 × g for 10 rain and the supernatant was replacedwith 4 ml viable
297 or formalin-killed R. equi suspension, resulting in a final bacteria:macrophage ratio of 10:1. The bacteria-cell suspensions were incubated for 4 h at 37 °C to allow ingestion of R. equi by the alveolar macrophages, after which the cells were washed four times by centrifuging at 280 Xg for 6 min at 4 ° C. The purpose of these washes was to augment phagocytosis and to separate the non-ingested bacteria from the cells containing ingested bacteria (Territo and Cline, 1977; Tomita and Kanegasaki, 1982). The final cell suspension was incubated in RPMI-1640 with 10% horse serum and 2 ttg m l - 1gentamicin ( Sigma Chemical Co., St. Louis, MO ), a concentration previously determined to be bacteristatic, but not bactericidal to R. equi. At 0, 3, 8 and 24 h (To-24) after this ingestion period, aliquots of each suspension were removed, pelleted and fixed in 2.5% glutaraldehyde for 3 h. After three 10-min washes in phosphate buffer, the pellet was fixed in 2% osmium tetroxide for 1 h, dehydrated through a series of graded acetone solutions and embedded in Spurr embedding medium ( Spurr, 1969). Ultrathin sections were cut and stained with uranyl acetate and lead citrate. Serial sections representing levels throughout each pellet were examined. Each cell was examined for the presence of intracellular bacteria, the occurrence of phagosome-lysosome fusion and ultrastructural evidence of damage to the macrophage and/or the bacteria. Ultrastructural criteria for damage to the bacteria were as follows: distortion or loss of continuity of the cell wall, contraction of the cytoplasmic contents away from the bacterial cell wall, disorganization of the cytoplasm and/or formation of apatite crystals in the cytoplasm (Armstrong and Hart, lb71; Hard, 1972; Davis-Scibienski and Beaman, 1980; Cheville, 1983). RESULTS
Ultrastructural morphology of cells Unchallenged (control) alveolar macrophages (Fig. 1) averaged 12 tim in diameter and had eccentric nuclei and abundant cytoplasm with many elongate surface extensions (filopodia). Cytoplasmic oganelles included abundant, moderately electron-dense lysosomes, mitochondria, a Golgi apparatus with associated small vesicles, small arrays of microfilaments, a few cisternae of rough endoplasmic reticulum and variable numbers of lipid vacuoles and undigested cellular debris. In samples treated with ferritin, the majority of alveolar macrophages had ferritin-labelled lysosomes. Ferritin was recognizable as small, roughly polygonal, highly electron-dense granules ~ 6-8 ttm in diameter. Ferritin was frequently seen adjacent to and within residual bodies and myelin bodies (Fig. 2), but was never seen in organelles other than lysosomes.
298
Fig. I. An unlabelled,unchallenged foalalveolarmacrophage with numerous filopodia(f),mitochondria (m), lysosomes (ly),moderate amounts of endoplasmic reticulum (rer),vacuoles (v) and a residualbody (rb). To. Bar represents 1 #m.
Fig. 2. Ferritin-labeUedmyelinbody (rob) in the cytoplasmofa foal alveolarmacrophageadjacent to a labelledlysosome (l). To. Bar represents 0.1/~m.
Ultrastructural characteristics of R. equi Rhodococcus equi were coccoid to rod shaped and measured ~ 0.5 × 1.0-2.0 /~m ( Fig. sule. The apposed, densities
3 ). ExtraceUular bacteria were enclosed in a prominent fibrillar capcell wall was uniformly electron dense and was surrounded by a closely fine, electron-dense membrane. The cytoplasm contained granular and a pale, central nuclear area. A central mesosome consisting of
299
Fig. 3. An extracellular R. equi demonstrates the thick capsule (c), electron-dense cell wall (cw), central nucleoid (n) and the mesosome (m) which are characteristic of this bacterium. To. Bar represents 0.1/~rn.
parallel layers of membranous structures was often present. There was no difference in the ultrastructural appearance of live and formalin-killed bacteria.
Ultrastructural events after ingestion of R. equi The majority of ingested bacteria at all time periods was present within "loose" phagosomes in which the phagosomal wall was widely separated from the bacteria (Fig. 4). Capsular material was no longer visible on bacteria fixed at T3-T24. Beginning at Ts, the phagosomal membrane had elongate invaginations of the cytoplasm into the phagosome (Fig. 5). In some cases, especially after 24-h incubation when macrophages were showing evidence of degeneration, R. equi were seen free in the cytosol and not surrounded by a phagosomal membrane ( Table I). The majority of bacteria which had been viable when inoculated remained morphologically intact, although with increased incubation time the number of viable intracellular bacteria decreased (Table I). After 24-h incubation, 75% of the intracellular bacteria still appeared to be morphologically intact. There was no apparent difference between viable and non-viable bacterial inocula with respect to the percentage of morphologically degenerate intracellular bacteria identified at Ts and Te4. Despite the relatively common occurrence of fusion of ferritin-labelled secondary lysosomes with myelin bodies and other cellular residua, ferritin was never identified within phagosomes containing R. equi in viable macrophages. Occasionally ferritin-labelled lysosomes were present close to phagosomes containing bacteria. Ferritin was seen occasionally in the cytosol and in R. equi-containing phagosomes in degenerating cells in which there was deterio-
300
Fig. 4. Several R. equi present within phagosomes (p) of an alveolar macrophage. Some bacteria have cross walls, indicating cellular division (arrowheads). Note the absence of ferritin in phagosomes. 7"24.Bar represents 1/zm. Fig. 5. Two phagosomes (p) contain several degenerate R. equi (b). The phagosomes have elongated cytoplasmic invaginations resembling microvilli (m). T24. Bar represents 1/lm.
ration of lysosomal membranes with leakage of ferritin to surrounding structures. In general, cells with ingested R. equi contained fewer ferritin-labelled lysosomes than cells which had not undergone phagocytosis. At Ts, bacteria could be seen multiplying intracellularly within phagosomes in cells inoculated with viable R. equi. By T24, intracellularly multiplying bacteria were numerous (Fig. 4 ). Extracellularly multiplying bacteria were rarely seen and when present, they were intimately associated with cellular debris, suggesting that these bacteria had been recently expelled from degenerate cells. Alveolar macrophages which had been inoculated with viable R. equi began degenerating at Ts. By T24, the majority of these cells were showing evidence TABLE I The intracellular location and extent of degradation of viable and non viable inocular of R. equi 8 and 24 hours after ingestion by foal alveolar macrophages Time postinfection
Viability of inoculum
Number of bacterial profiles examined
Bacteria within phagosomes
Bacteria free in cytosol
Non-degenerate bacteria
Degenerate bacteria
live dead live dead
50 50 200 200
48 49 179 200
2 1 21 0
47 42 149 151
3 8 51 49
(h) 8 8 24 24
301
Fig.6. A degeneratealveolarmacrophage releasingbacteriathrough a broken cytoplasmicmembrane (pro).Remnants ofcellularorganeUes,probablydegeneratemitochondria (cr),arepresent. The majorityofbacteriaappear viable.T24.Bar represents1/~m. of advanced degeneration, including uniformly dense nuclear chromatin and clearing of the cytoplasm with small clumps of flocculent debris and broken, vacuolated mitochondria. With destruction of the cytoplasmic membrane, the cytoplasmic contents, including large numbers of bacteria, were released into
the surrounding medium (Fig. 6). DISCUSSION The ultrastructural evidence illustrated here confirms that R. equi behaves as a facultative intracellular parasite, surviving and multiplying in the foal alveolar macrophage. Inside the macrophage, R. equi resided within "loose" membrane-bound phagocytic vacuoles which, after prolonged incubation, became crenate and often developed elongate, villous invaginations. Other intracellular bacteria which reside within "loose" phagosomes include Propionobacterium aches ( Pringle et al., 1982), LegioneUapneumophila ( Horwitz, 1983; Oldham and Rodgers, 1985) and some Escherichia coli ( Rollag and Hovig, 1984). Phagosomes containing Corynebacterium pseudotuberculosis have crenated membranes, similar to those seen in this study (Hard, 1972; Tashjian and Campbell, 1983). Microvillous structures line the phagosomes of a number of protozoal parasites such as Toxoplasma gondii (Sheffield and Melton, 1968) and it has been postulated that these structures facilitate the transfer of cytoplasmic factors required for the growth ofintracellular parasites (Jones and Hirsch, 1972). Although the significance of loosely or tightly apposed pha-
302 gosomes is not clear,an association has been made between the occurrence of loose phagosomes and inhibition of phagosome-lysosome fusion (Hart and Young, 1975, 1978). Despite the relativelyc o m m o n occurrence of fusion of ferritin-labelledlysosomes with phagosomal bodies containing substances other than R. equi for myelin bodies, there was no evidence that lysosomes fused with plmgosomes containing R. equi. Other bacteria which inhibit phagosome,lysosome fusion include Mycobacterium tuberculosis (Armstrong and Hart, 1.971), Mycob~terium microti (Hart et al., 1972), Nocard/a as~ero/des (Davis-Scibienski and Beaman, 1980) and L. pneumophila (Horwitg, !983). The mechanism ofphagosome-lysosome fusion is poorly understood, but a number of substances have been identified which prevent phagosome-lysosome fusion by modifying the lysosomal membrane. Polyanionic molecules such as sulfafides (present as a cell wall component of virulent, but not avirulent M. tuberculosis) and surarain, a lysosomotropic drug structurally similar to sulfatides (Goren, 1977) are potent inhibitors of phagosome-lysosome fusion. That there was no difference in the frequency of phagosome-lysosome fusion with viable or non-viable R. equi inocula suggests that the inhibition of phagosome-lysosome fusion in these cells was not dependent on the continued biosynthesis of a membraneactive product by R. equi ( Elsbach, 1980). Instead, R. equi may have inhibited phagosome-lysosome fusion passively, by virtue of a structural component which was preserved upon formalin treatment. Without a detailed biochemical analysis of the structural components of R. equi, one can only speculate that the biochemical nature of either the capsule or the cell wall of R. equi may influence the interaction of the phagosomal wall with the lysosome. It is possible that R. equi utilizes a similar mechanism of inhibition of phagosome-lysosome fusion as M. tuberculosis, with which it shares cell wall structural similarities (Brennan and Lehane, 1971; Goodfellow et al., 1976; Stackebrandt et al., 1980). There was good correlation between the percentage of bacteria considered to be viable by ultrastructural morphological criteria and in vitro assays of the bactericidal ability of foal alveolar macrophages (Zink et al., 1985) which determined that ~ 60-75% of ingested R. equi remain viable after incubation for 24 h. This study demonstrated that, in addition to surviving within alveolar macrophages, R. equi was capable of extensive multiplication within the cells' phagosomes. The earliest evidence of bacterial multiplication was seen 8 h after macrophages ingested viable bacteria. At the same time, these cells began to show morphological evidence of degeneration. By 24-h post-inoculation, most macrophages contained multiplying bacteria and the majority of these cells had undergone irreversible cell damage. In contrast, alveolar macrophages which ingested formalin-killed R. equi were still viable after 24-h incubation, although some showed evidence of reversible cellular injury such as glycogen
303 accumulation and autophagic vacuoles. These data suggest that viable intracellular R. equi may have been synthesizing a substance which contributed to accelerated cell death. Rhodococcus equi has always been considered remarkable in its lack of biochemical activity (Elissalde et al., 1980; Mutimer and Woolcock, 1981; Prescott et al., 1982), but the majority of strains produce equi factors which interact with. the phospholipase D of Corynebacterium pseudotuberculosis (a partial hemolysin) to produce an area of complete hemolysis on sheep blood agar (Prescott et al., 1982). These equi factors, a cholesterol oxidase and a phospholipase, appear to exert their effects on components within the cell membrane of erythrocytes, causing lysis (Linder and Bernheimer, 1982). The significance of these factors with respect to the virulence of this organism is still under investigation. It is known that sera from the majority of foals with naturally occurring R. equi pneumonia contain antibodies to equi factors (Prescott et al., 1984). It is possible that the action of equi factors on the lysosomal membrane may contribute to the cellular degeneration which occurs after foal alveolar macrophages ingest viable R. equi. The results of this study have confirmed that R. equi is a facultative intracellular parasite. This bacterium is ingested by the alveolar macrophage and persists and multiplies within the phagosome, apparently inhibiting phagosome-lysosome fusion by some as yet unknown mechanism. Cells with viable intracellular R. equi undergo irreversible degenerative changes, eventually releasing the bacteria into the surrounding milieu.
ACKNOWLEDGMENTS This research was funded by grants from the Natural Sciences and Engineering Research Council, the Canadian Veterinary Research Trust Fund and the Ontario Racing Commission. M.C. Zink is the recipient of a Medical Research Council Fellowship.
REFERENCES Armstrong, JA. and Hart, P.D., 1971. Response of cultured macrophagesto Mycobacterium tuberculosis, with observationson fusion of lysosomeswith phagosomes.J. Exp. Med., 134: 713-740. Brennan, P.J, and Lehane,D.P., 1971.The phospholipidsof Corynebacteria~Lipids, 6: 401-409. Cheville,N.F., 1983. Cell Pathology.IowaState UniversityPress, Ames,IA, pp. 479-487. Davis-Scibienski, C. and Beaman, B.L., 1980. Interaction of Nocardia a.steroides with rabbit alveolarmacroph~ges:effectof phase of growthand viabilityon phagosome-lysosomefusion. Infect. Immml.,29: 24-29. Elissalde, G.S., Renshaw, H.W. and Walberg, J.A., 1980. Corynebacterium equi - an interhost reviewwith emphasison the foal. Comp. Immun.Microbiol.Infect.Dis., 3: 433-445.
304 Ellenberger, M.A., Kaeberle, M.L. and Roth, J.A., 1984. Equine ceU-mediated immune response to Rhodococcus (Corynebacterium) equi. Am. J. Vet. IRes., 45: 2424-2427. Elsbach, P., 1980. Degradation of microorganisms by phagocytic cells. Rev. Infect. Dis., 2: 106-128. Goodfellow, M., Collins, M.D. and Minnikin, D.E., 1976. Thin-layer chromatographic analysis of mycolic acid and other long-chain components of whole organism methanolysates of coryneform and related taxa. J. Gen. Microbiol., 96: 351-358. Goren, M.B., 1977. Phagocyte lysosomes: Interactions with infectious agents, phagosomes and experimental perturbations in function. Annu. Rev. Microbiol., 31: 507-533. Hard, G.C., 1972. Examination by electron microscopy of the interaction between peritoneal phagocytes and Corynebacterium ov/s. J. Med. Microbiol., 5: 483-491. Hart, P.D. and Young, M.R., 1975. Interference with normal phagosome-lysosome fusion in macrophages, using ingested yeast cells and suramin. Nature, London, 258: 47-49. Hart, P.D. and Young, M.R., 1978. Manipulations of the phagosome-lysosome fusion response in cultured macrophages. Enhancement of fusion by chloroquine and other amines. Exp. Cell Res., 114: 486-490. Hart, P.D., Armstrong, J.A., Brown, C.A. and Draper, P., 1972. Ultrastructural study of the behaviour of macrophages toward parasitic mycobacteria. Infect. Immun., 5: 803-807. Hocking, W.G. and Golde, D.W., 1979. The pulmonary-alveolar macrophage (Part 1). N. Engl. J. Med., 301: 580-587. Horwitz, M.A., 1983. The Legionnaires' disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J. Exp. Med., 158: 2108-2126. Johnson, J.A., Prescott, J.F. and Markham, R.J.F., 1983. The pathology of experimental Corynebacterium equi infection in foals following intrabronchial challenge. Vet. Pathol., 20: 440-449. Jones, T.C. and Hirsch, J.G., 1972. The interaction between Toxoplasma gondii and mammalian cells.II.The absence of lysosomal fusion with phagocytic vacuoles containing livingparasites. J. Exp. Med., 136: 1173-1194. Knight, H.D., 1969. Corynebacterial infections in the horse: Problems of prevention. J. Am. Vet. Med. Assoc., 155: 446-452. Linder, R. and Bernheimer, A.W., 1982. Enzymatic oxidation of membrane cholesterolin relation to lysis of sheep erythrocytes by corynebacterial enzymes. Arch. Biochem. Biophys., 213: 395-404. Mutimer, M.D. and Woolcock, J.B., 1981. Some problems associated with the identificationof Corynebacterium equi. Vet. Microbiol., 6: 331-338. Oldham, L.J. and Rodgers, F.G., 1985. Adhesion, penetration and intracellularreplicationof Legionellapneumophila: An in vitro model of pathogenesis. J. Gen. Microbiol., 131: 697-706. Prescott, J.F.,Markham, R.J.F. and Johnson, J.A., 1979. Cellularand humoral immune response of foalsto vaccination with Corynebacterium equi Magnusson. Can. J. Comp. Med., 43: 356-364. Prescott, J.F., Lastra, M. and Barksdale, L., 1982. Equi factors in the identificationof Corynebacterium equi Magnusson. J. Clin. Microbiol., 16: 988-990. Prescott, J.F., Coshan-Gauthier, R. and Barksdale, L., 1984. Antibody to equi factor(s) in the diagnosis of Corynebacterium equi pneumonia of foals.Can. J. Comp. Med., 48: 370-373. Pringle, A.T., Cummins, C.S., Bishop, B.F. and Viers, V.S., 1982. Fate of vaccines of Propionibacterium aches after phagocytosis by murine macrophages. Infect. Immun., 38: 371-374. Rollag, H. and Hovig, T., 1984. Phagocytosis of non-opsonized Escherichia coli by mouse peritoneal macrophages. An electron microscopic study. Zentralbl. Bakteriol. Mikrobiol. Hyg. Reihe A, 257: 93-107. Sheffield, H.G. and Melton, M.L., 1968. The fine structure and reproduction of Toxoplasma gondii. J. Parasitol., 54: 209-226. Spurr, A.R., 1969. A low viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. IRes., 26: 31-43.
305 Stackebrandt, E., Lewis, B.J. and Woese, C.R., 1980. The phylogenetic structure of the Coryneform group of bacteria. Zentralbl.Bakteriol. Mikrobiol. Hyg., I Abt. Orig. CI: 137-149. Tashjian, J.J. and Campbell, S.G., 1983. Interaction between caprine macrophages and Corynebacteriumpseudotuberculosis:An electron microscopic study. Am. J. Vet. Res., 44: 690-693. Territo, M.C. and Cline, M.J., 1977. Monocyte function in man. J. Immunol., 118: 187-192. Tomita, T. and Kanegasaki, S., 1982. Enhanced phagocytic response of macrophages to bacteria by physical impact caused by bacterialmotilityor centrifugation.Infect.Immun., 38: 865-870. Wilson, M.M., 1955. A study of Corynebacterium equi infection in a stud of thoroughbred horses in Victoria.Aust. Vet. J., 31: 175-181. Zink, M.C. and Johnson, J.A., 1984. Cellular constituents of normal foal bronchoalveolar lavage fluid during postnatal maturation. Am. J. Vet. ires., 45: 893-897. Zink, M.C., Yager, J.A., Prescott, J.F. and Wilkie, B.N., 1985. In vitro phagocytosis and killing of Corynebacterium equi by foal alveolar macrophages during postnatal maturation. Am. J. Vet. Res., 46: 2171-2174.
295
Electron Microscopic Investigation of Intracellular Events after Ingestion of Rhodococcus equi by Foal Alveolar Macrophages M.C. Z I N K la,J.A. Y A G E R
I,J.F. P R E S C O T T
2 and M.A. F E R N A N D O
l
Department of Patholog~ and Veterinary Microbiology and Irnnrnunology~,Ontario Veterinary College, Universityof Guelph, Guelph, Ontario NIG 2 W1 (Canada)
ABSTRACT Zink, M.C., Yager, J.A., Prescott, J.F. and Fernando, M.A., 1987. Electron microscopic investigation of intracellular events after ingestion of Rhodococeus equi by foal alveolar macrophages. Vet Microbiol., 14: 295-305. It has been suggested that R. equi causes pulmonary disease in foals by persisting within the lung as a facultative intmcellular parasite of alveolar macrophages. This paper describes an uRra-
structural study of the intracellular events after ingestion ofR. equi by fos~ alveolar macrophages, in an attempt to determine the mechanism of intracellular survival of R. equi. Secondary lysoeomes of alveolar macrophages recovered from foals by bronchoalveoler lavage were labelled with electron-dense ferritin, and the ceils were challenged with either viable or formalin-ktIed R. equi. After 0-, 3-, 8- or 24-h incubation, the cells were fixed and processed for electron microscopy. There was no evidence of phagosome-lysosome fusion after ingestion of either viable or non-viable R. equi by foal alveolar macrophages. Rhodococcus equi persisted and multiplied within dilated phegosomes, which were often lined by elongate microvilleus structures. After 24-h incubation, 75% of the ingested bacteria were still structurally intact. Macrophages with ingested viable R. equ/were irreversibly d~maged and released intracellular bacteria into the surrounding medium. These data confirm that R~ equi is a facultative intracellular parasite of foal alveolar macrophages and is able to persist and multiply within the phagosome, apparently inhibiting phagosome-lysosome fusion by some as yet unknown mechanism.
INTRODUCTION
Several investigators have suggested that Rhodococcua equi is a facultative intracellular parasite of the foal alveolar macrophage (Wilson, 1955; Knight, 1969; Elissalde et al., 1980; Johnson et al., 1983). Circumstantial evidence for this hypothesis includes the granulomatous nature of the pulmonary lesions of R. equi infection (Johnson et al., 1983), the tendency for horses exposed to R. •Author to whom correspondence should be addressed at: Department of Neurology, Meyer 6-181, Johns Hopkins University School of Medicine, 600 N. Wolfe St., Baltimore, MD 21205, U.S.A.
0378-1135/87/$03.50
© 1987 Elsevier Science Publishers B.V.
296
equi to develop cell-mediated immune responses (Wilson, 1955; Prescott, et al., 1979; Ellenberger et al., 1984) and similarity in bacterial cell wall structure of R. equi to well recognized intracellular parasites from the Mycobacterium, Nocardia and Bacterionema genera (Brennan and Lehane, 1971; Goodfellow et al., 1976; Stackebrandt et al., 1980). Further evidence has been provided by studies on the bactericidal function of foal alveolar macrophages, in which it was shown that ~ 75% of R. equi remain viable 4 h after ingestion by foal alveolar macrophages (Zink et al., 1985). Thus alveolar macrophages, which function both as the first line of defense against inhaled bacteria and as effector cells for pulmonary cell-mediated immunity (Hocking and Golde, 1979), are probably of central importance in the pathogenesis of R. equi pneumonia in foals. The purpose of this study was to characterize the intracellular events occurring within foal alveolar macrophages during the first 24 h after ingestion of R. equi in order to identify the mechanism of intracellular survival of this bacterium. MATERIALSAND METHODS
Collection and labelling of alveolar macrophages Bronchoalveolar lavage was performed on foals under general anesthetic and the recovered cells were washed and counted as described elsewhere (Zink and Johnson, 1984). The secondary lysosomes of the alveolar macrophages were labelled with electron-dense ferritin (Sigma Chemical Co., St. Louis, MO) by incubating the cells in RPMI-1640 containing 2.5% fetal bovine serum (FBS) with or without 20 mg m - 1 ferritin for 4 h. Cell suspensions were then washed 3 times with R P M I containing 2.5% FBS, incubated for a further 3 h to allow complete uptake of residual ferritin and resuspended at a final concentration of 2.0 × 10e cells m l - 1.
Preparation of bacteria Viable and non-viable bacterialsuspensions were prepared by culturingR. equi on Mueller-Hinton agar for 36 h then suspending the bacteria in phosphate-buffered saline (PBS) with or without 0.5% formalin for 1 h. After 4 washes in PBS, the bacteria were suspended at a concentration of 2.0X 107 ml-i in R P M I containing 10% normal horse serum.
Electron microscopy Suspensions (4 ml) of ferritin-labeUedor unlabelledmacrophages werecentrifugedat 400 × g for 10 rain and the supernatant was replacedwith 4 ml viable
297 or formalin-killed R. equi suspension, resulting in a final bacteria:macrophage ratio of 10:1. The bacteria-cell suspensions were incubated for 4 h at 37 °C to allow ingestion of R. equi by the alveolar macrophages, after which the cells were washed four times by centrifuging at 280 Xg for 6 min at 4 ° C. The purpose of these washes was to augment phagocytosis and to separate the non-ingested bacteria from the cells containing ingested bacteria (Territo and Cline, 1977; Tomita and Kanegasaki, 1982). The final cell suspension was incubated in RPMI-1640 with 10% horse serum and 2 ttg m l - 1gentamicin ( Sigma Chemical Co., St. Louis, MO ), a concentration previously determined to be bacteristatic, but not bactericidal to R. equi. At 0, 3, 8 and 24 h (To-24) after this ingestion period, aliquots of each suspension were removed, pelleted and fixed in 2.5% glutaraldehyde for 3 h. After three 10-min washes in phosphate buffer, the pellet was fixed in 2% osmium tetroxide for 1 h, dehydrated through a series of graded acetone solutions and embedded in Spurr embedding medium ( Spurr, 1969). Ultrathin sections were cut and stained with uranyl acetate and lead citrate. Serial sections representing levels throughout each pellet were examined. Each cell was examined for the presence of intracellular bacteria, the occurrence of phagosome-lysosome fusion and ultrastructural evidence of damage to the macrophage and/or the bacteria. Ultrastructural criteria for damage to the bacteria were as follows: distortion or loss of continuity of the cell wall, contraction of the cytoplasmic contents away from the bacterial cell wall, disorganization of the cytoplasm and/or formation of apatite crystals in the cytoplasm (Armstrong and Hart, lb71; Hard, 1972; Davis-Scibienski and Beaman, 1980; Cheville, 1983). RESULTS
Ultrastructural morphology of cells Unchallenged (control) alveolar macrophages (Fig. 1) averaged 12 tim in diameter and had eccentric nuclei and abundant cytoplasm with many elongate surface extensions (filopodia). Cytoplasmic oganelles included abundant, moderately electron-dense lysosomes, mitochondria, a Golgi apparatus with associated small vesicles, small arrays of microfilaments, a few cisternae of rough endoplasmic reticulum and variable numbers of lipid vacuoles and undigested cellular debris. In samples treated with ferritin, the majority of alveolar macrophages had ferritin-labelled lysosomes. Ferritin was recognizable as small, roughly polygonal, highly electron-dense granules ~ 6-8 ttm in diameter. Ferritin was frequently seen adjacent to and within residual bodies and myelin bodies (Fig. 2), but was never seen in organelles other than lysosomes.
298
Fig. I. An unlabelled,unchallenged foalalveolarmacrophage with numerous filopodia(f),mitochondria (m), lysosomes (ly),moderate amounts of endoplasmic reticulum (rer),vacuoles (v) and a residualbody (rb). To. Bar represents 1 #m.
Fig. 2. Ferritin-labeUedmyelinbody (rob) in the cytoplasmofa foal alveolarmacrophageadjacent to a labelledlysosome (l). To. Bar represents 0.1/~m.
Ultrastructural characteristics of R. equi Rhodococcus equi were coccoid to rod shaped and measured ~ 0.5 × 1.0-2.0 /~m ( Fig. sule. The apposed, densities
3 ). ExtraceUular bacteria were enclosed in a prominent fibrillar capcell wall was uniformly electron dense and was surrounded by a closely fine, electron-dense membrane. The cytoplasm contained granular and a pale, central nuclear area. A central mesosome consisting of
299
Fig. 3. An extracellular R. equi demonstrates the thick capsule (c), electron-dense cell wall (cw), central nucleoid (n) and the mesosome (m) which are characteristic of this bacterium. To. Bar represents 0.1/~rn.
parallel layers of membranous structures was often present. There was no difference in the ultrastructural appearance of live and formalin-killed bacteria.
Ultrastructural events after ingestion of R. equi The majority of ingested bacteria at all time periods was present within "loose" phagosomes in which the phagosomal wall was widely separated from the bacteria (Fig. 4). Capsular material was no longer visible on bacteria fixed at T3-T24. Beginning at Ts, the phagosomal membrane had elongate invaginations of the cytoplasm into the phagosome (Fig. 5). In some cases, especially after 24-h incubation when macrophages were showing evidence of degeneration, R. equi were seen free in the cytosol and not surrounded by a phagosomal membrane ( Table I). The majority of bacteria which had been viable when inoculated remained morphologically intact, although with increased incubation time the number of viable intracellular bacteria decreased (Table I). After 24-h incubation, 75% of the intracellular bacteria still appeared to be morphologically intact. There was no apparent difference between viable and non-viable bacterial inocula with respect to the percentage of morphologically degenerate intracellular bacteria identified at Ts and Te4. Despite the relatively common occurrence of fusion of ferritin-labelled secondary lysosomes with myelin bodies and other cellular residua, ferritin was never identified within phagosomes containing R. equi in viable macrophages. Occasionally ferritin-labelled lysosomes were present close to phagosomes containing bacteria. Ferritin was seen occasionally in the cytosol and in R. equi-containing phagosomes in degenerating cells in which there was deterio-
300
Fig. 4. Several R. equi present within phagosomes (p) of an alveolar macrophage. Some bacteria have cross walls, indicating cellular division (arrowheads). Note the absence of ferritin in phagosomes. 7"24.Bar represents 1/zm. Fig. 5. Two phagosomes (p) contain several degenerate R. equi (b). The phagosomes have elongated cytoplasmic invaginations resembling microvilli (m). T24. Bar represents 1/lm.
ration of lysosomal membranes with leakage of ferritin to surrounding structures. In general, cells with ingested R. equi contained fewer ferritin-labelled lysosomes than cells which had not undergone phagocytosis. At Ts, bacteria could be seen multiplying intracellularly within phagosomes in cells inoculated with viable R. equi. By T24, intracellularly multiplying bacteria were numerous (Fig. 4 ). Extracellularly multiplying bacteria were rarely seen and when present, they were intimately associated with cellular debris, suggesting that these bacteria had been recently expelled from degenerate cells. Alveolar macrophages which had been inoculated with viable R. equi began degenerating at Ts. By T24, the majority of these cells were showing evidence TABLE I The intracellular location and extent of degradation of viable and non viable inocular of R. equi 8 and 24 hours after ingestion by foal alveolar macrophages Time postinfection
Viability of inoculum
Number of bacterial profiles examined
Bacteria within phagosomes
Bacteria free in cytosol
Non-degenerate bacteria
Degenerate bacteria
live dead live dead
50 50 200 200
48 49 179 200
2 1 21 0
47 42 149 151
3 8 51 49
(h) 8 8 24 24
301
Fig.6. A degeneratealveolarmacrophage releasingbacteriathrough a broken cytoplasmicmembrane (pro).Remnants ofcellularorganeUes,probablydegeneratemitochondria (cr),arepresent. The majorityofbacteriaappear viable.T24.Bar represents1/~m. of advanced degeneration, including uniformly dense nuclear chromatin and clearing of the cytoplasm with small clumps of flocculent debris and broken, vacuolated mitochondria. With destruction of the cytoplasmic membrane, the cytoplasmic contents, including large numbers of bacteria, were released into
the surrounding medium (Fig. 6). DISCUSSION The ultrastructural evidence illustrated here confirms that R. equi behaves as a facultative intracellular parasite, surviving and multiplying in the foal alveolar macrophage. Inside the macrophage, R. equi resided within "loose" membrane-bound phagocytic vacuoles which, after prolonged incubation, became crenate and often developed elongate, villous invaginations. Other intracellular bacteria which reside within "loose" phagosomes include Propionobacterium aches ( Pringle et al., 1982), LegioneUapneumophila ( Horwitz, 1983; Oldham and Rodgers, 1985) and some Escherichia coli ( Rollag and Hovig, 1984). Phagosomes containing Corynebacterium pseudotuberculosis have crenated membranes, similar to those seen in this study (Hard, 1972; Tashjian and Campbell, 1983). Microvillous structures line the phagosomes of a number of protozoal parasites such as Toxoplasma gondii (Sheffield and Melton, 1968) and it has been postulated that these structures facilitate the transfer of cytoplasmic factors required for the growth ofintracellular parasites (Jones and Hirsch, 1972). Although the significance of loosely or tightly apposed pha-
302 gosomes is not clear,an association has been made between the occurrence of loose phagosomes and inhibition of phagosome-lysosome fusion (Hart and Young, 1975, 1978). Despite the relativelyc o m m o n occurrence of fusion of ferritin-labelledlysosomes with phagosomal bodies containing substances other than R. equi for myelin bodies, there was no evidence that lysosomes fused with plmgosomes containing R. equi. Other bacteria which inhibit phagosome,lysosome fusion include Mycobacterium tuberculosis (Armstrong and Hart, 1.971), Mycob~terium microti (Hart et al., 1972), Nocard/a as~ero/des (Davis-Scibienski and Beaman, 1980) and L. pneumophila (Horwitg, !983). The mechanism ofphagosome-lysosome fusion is poorly understood, but a number of substances have been identified which prevent phagosome-lysosome fusion by modifying the lysosomal membrane. Polyanionic molecules such as sulfafides (present as a cell wall component of virulent, but not avirulent M. tuberculosis) and surarain, a lysosomotropic drug structurally similar to sulfatides (Goren, 1977) are potent inhibitors of phagosome-lysosome fusion. That there was no difference in the frequency of phagosome-lysosome fusion with viable or non-viable R. equi inocula suggests that the inhibition of phagosome-lysosome fusion in these cells was not dependent on the continued biosynthesis of a membraneactive product by R. equi ( Elsbach, 1980). Instead, R. equi may have inhibited phagosome-lysosome fusion passively, by virtue of a structural component which was preserved upon formalin treatment. Without a detailed biochemical analysis of the structural components of R. equi, one can only speculate that the biochemical nature of either the capsule or the cell wall of R. equi may influence the interaction of the phagosomal wall with the lysosome. It is possible that R. equi utilizes a similar mechanism of inhibition of phagosome-lysosome fusion as M. tuberculosis, with which it shares cell wall structural similarities (Brennan and Lehane, 1971; Goodfellow et al., 1976; Stackebrandt et al., 1980). There was good correlation between the percentage of bacteria considered to be viable by ultrastructural morphological criteria and in vitro assays of the bactericidal ability of foal alveolar macrophages (Zink et al., 1985) which determined that ~ 60-75% of ingested R. equi remain viable after incubation for 24 h. This study demonstrated that, in addition to surviving within alveolar macrophages, R. equi was capable of extensive multiplication within the cells' phagosomes. The earliest evidence of bacterial multiplication was seen 8 h after macrophages ingested viable bacteria. At the same time, these cells began to show morphological evidence of degeneration. By 24-h post-inoculation, most macrophages contained multiplying bacteria and the majority of these cells had undergone irreversible cell damage. In contrast, alveolar macrophages which ingested formalin-killed R. equi were still viable after 24-h incubation, although some showed evidence of reversible cellular injury such as glycogen
303 accumulation and autophagic vacuoles. These data suggest that viable intracellular R. equi may have been synthesizing a substance which contributed to accelerated cell death. Rhodococcus equi has always been considered remarkable in its lack of biochemical activity (Elissalde et al., 1980; Mutimer and Woolcock, 1981; Prescott et al., 1982), but the majority of strains produce equi factors which interact with. the phospholipase D of Corynebacterium pseudotuberculosis (a partial hemolysin) to produce an area of complete hemolysis on sheep blood agar (Prescott et al., 1982). These equi factors, a cholesterol oxidase and a phospholipase, appear to exert their effects on components within the cell membrane of erythrocytes, causing lysis (Linder and Bernheimer, 1982). The significance of these factors with respect to the virulence of this organism is still under investigation. It is known that sera from the majority of foals with naturally occurring R. equi pneumonia contain antibodies to equi factors (Prescott et al., 1984). It is possible that the action of equi factors on the lysosomal membrane may contribute to the cellular degeneration which occurs after foal alveolar macrophages ingest viable R. equi. The results of this study have confirmed that R. equi is a facultative intracellular parasite. This bacterium is ingested by the alveolar macrophage and persists and multiplies within the phagosome, apparently inhibiting phagosome-lysosome fusion by some as yet unknown mechanism. Cells with viable intracellular R. equi undergo irreversible degenerative changes, eventually releasing the bacteria into the surrounding milieu.
ACKNOWLEDGMENTS This research was funded by grants from the Natural Sciences and Engineering Research Council, the Canadian Veterinary Research Trust Fund and the Ontario Racing Commission. M.C. Zink is the recipient of a Medical Research Council Fellowship.
REFERENCES Armstrong, JA. and Hart, P.D., 1971. Response of cultured macrophagesto Mycobacterium tuberculosis, with observationson fusion of lysosomeswith phagosomes.J. Exp. Med., 134: 713-740. Brennan, P.J, and Lehane,D.P., 1971.The phospholipidsof Corynebacteria~Lipids, 6: 401-409. Cheville,N.F., 1983. Cell Pathology.IowaState UniversityPress, Ames,IA, pp. 479-487. Davis-Scibienski, C. and Beaman, B.L., 1980. Interaction of Nocardia a.steroides with rabbit alveolarmacroph~ges:effectof phase of growthand viabilityon phagosome-lysosomefusion. Infect. Immml.,29: 24-29. Elissalde, G.S., Renshaw, H.W. and Walberg, J.A., 1980. Corynebacterium equi - an interhost reviewwith emphasison the foal. Comp. Immun.Microbiol.Infect.Dis., 3: 433-445.
304 Ellenberger, M.A., Kaeberle, M.L. and Roth, J.A., 1984. Equine ceU-mediated immune response to Rhodococcus (Corynebacterium) equi. Am. J. Vet. IRes., 45: 2424-2427. Elsbach, P., 1980. Degradation of microorganisms by phagocytic cells. Rev. Infect. Dis., 2: 106-128. Goodfellow, M., Collins, M.D. and Minnikin, D.E., 1976. Thin-layer chromatographic analysis of mycolic acid and other long-chain components of whole organism methanolysates of coryneform and related taxa. J. Gen. Microbiol., 96: 351-358. Goren, M.B., 1977. Phagocyte lysosomes: Interactions with infectious agents, phagosomes and experimental perturbations in function. Annu. Rev. Microbiol., 31: 507-533. Hard, G.C., 1972. Examination by electron microscopy of the interaction between peritoneal phagocytes and Corynebacterium ov/s. J. Med. Microbiol., 5: 483-491. Hart, P.D. and Young, M.R., 1975. Interference with normal phagosome-lysosome fusion in macrophages, using ingested yeast cells and suramin. Nature, London, 258: 47-49. Hart, P.D. and Young, M.R., 1978. Manipulations of the phagosome-lysosome fusion response in cultured macrophages. Enhancement of fusion by chloroquine and other amines. Exp. Cell Res., 114: 486-490. Hart, P.D., Armstrong, J.A., Brown, C.A. and Draper, P., 1972. Ultrastructural study of the behaviour of macrophages toward parasitic mycobacteria. Infect. Immun., 5: 803-807. Hocking, W.G. and Golde, D.W., 1979. The pulmonary-alveolar macrophage (Part 1). N. Engl. J. Med., 301: 580-587. Horwitz, M.A., 1983. The Legionnaires' disease bacterium (Legionella pneumophila) inhibits phagosome-lysosome fusion in human monocytes. J. Exp. Med., 158: 2108-2126. Johnson, J.A., Prescott, J.F. and Markham, R.J.F., 1983. The pathology of experimental Corynebacterium equi infection in foals following intrabronchial challenge. Vet. Pathol., 20: 440-449. Jones, T.C. and Hirsch, J.G., 1972. The interaction between Toxoplasma gondii and mammalian cells.II.The absence of lysosomal fusion with phagocytic vacuoles containing livingparasites. J. Exp. Med., 136: 1173-1194. Knight, H.D., 1969. Corynebacterial infections in the horse: Problems of prevention. J. Am. Vet. Med. Assoc., 155: 446-452. Linder, R. and Bernheimer, A.W., 1982. Enzymatic oxidation of membrane cholesterolin relation to lysis of sheep erythrocytes by corynebacterial enzymes. Arch. Biochem. Biophys., 213: 395-404. Mutimer, M.D. and Woolcock, J.B., 1981. Some problems associated with the identificationof Corynebacterium equi. Vet. Microbiol., 6: 331-338. Oldham, L.J. and Rodgers, F.G., 1985. Adhesion, penetration and intracellularreplicationof Legionellapneumophila: An in vitro model of pathogenesis. J. Gen. Microbiol., 131: 697-706. Prescott, J.F.,Markham, R.J.F. and Johnson, J.A., 1979. Cellularand humoral immune response of foalsto vaccination with Corynebacterium equi Magnusson. Can. J. Comp. Med., 43: 356-364. Prescott, J.F., Lastra, M. and Barksdale, L., 1982. Equi factors in the identificationof Corynebacterium equi Magnusson. J. Clin. Microbiol., 16: 988-990. Prescott, J.F., Coshan-Gauthier, R. and Barksdale, L., 1984. Antibody to equi factor(s) in the diagnosis of Corynebacterium equi pneumonia of foals.Can. J. Comp. Med., 48: 370-373. Pringle, A.T., Cummins, C.S., Bishop, B.F. and Viers, V.S., 1982. Fate of vaccines of Propionibacterium aches after phagocytosis by murine macrophages. Infect. Immun., 38: 371-374. Rollag, H. and Hovig, T., 1984. Phagocytosis of non-opsonized Escherichia coli by mouse peritoneal macrophages. An electron microscopic study. Zentralbl. Bakteriol. Mikrobiol. Hyg. Reihe A, 257: 93-107. Sheffield, H.G. and Melton, M.L., 1968. The fine structure and reproduction of Toxoplasma gondii. J. Parasitol., 54: 209-226. Spurr, A.R., 1969. A low viscosity epoxy resin embedding medium for electron microscopy. J. Ultrastruct. IRes., 26: 31-43.
305 Stackebrandt, E., Lewis, B.J. and Woese, C.R., 1980. The phylogenetic structure of the Coryneform group of bacteria. Zentralbl.Bakteriol. Mikrobiol. Hyg., I Abt. Orig. CI: 137-149. Tashjian, J.J. and Campbell, S.G., 1983. Interaction between caprine macrophages and Corynebacteriumpseudotuberculosis:An electron microscopic study. Am. J. Vet. Res., 44: 690-693. Territo, M.C. and Cline, M.J., 1977. Monocyte function in man. J. Immunol., 118: 187-192. Tomita, T. and Kanegasaki, S., 1982. Enhanced phagocytic response of macrophages to bacteria by physical impact caused by bacterialmotilityor centrifugation.Infect.Immun., 38: 865-870. Wilson, M.M., 1955. A study of Corynebacterium equi infection in a stud of thoroughbred horses in Victoria.Aust. Vet. J., 31: 175-181. Zink, M.C. and Johnson, J.A., 1984. Cellular constituents of normal foal bronchoalveolar lavage fluid during postnatal maturation. Am. J. Vet. ires., 45: 893-897. Zink, M.C., Yager, J.A., Prescott, J.F. and Wilkie, B.N., 1985. In vitro phagocytosis and killing of Corynebacterium equi by foal alveolar macrophages during postnatal maturation. Am. J. Vet. Res., 46: 2171-2174.