Axenic cultivation and partial characterization of ... - LabRam

103

Axenic cultivation and partial characterization of Leishmania braziliensis amastigote-like stages J. M. F. B A L A N C O", E. M. F. P R A L", S.   S I L V A#, A. T. B I J O V S KY", R. A. M O R T A R A# and S. C. A L F I E R I"* " Departamento de Parasitologia, Instituto de CieW ncias BiomeU dicas, Universidade de Sah o Paulo, Av. Prof. Lineu Prestes 1374, CEP 05508-900, Sah o Paulo, S.P., Brasil # Departamento de Microbiologia, Imunologia e Parasitologia, Escola Paulista de Medicina and Centro de Microscopia EletroW nica, Universidade Federal de Sah o Paulo, R. Botucatu 862, CEP 04062-040, Sah o Paulo, S.P., Brasil (Received 14 May 1997 ; revised 12 September 1997 ; accepted 12 September 1997)

 Leishmania braziliensis strain M2903 was adapted for growth and serially maintained as amastigotes at 34 °C in modified UM-54 medium, with growth curves exhibiting typical log and stationary phases. In late passages, amastigote growth took place in the absence of supplementary haemin and was unaffected when the initial medium pH was adjusted between 5±4 and 6±3. In contrast to promastigotes, which were elongated and exhibited very long free flagella endowed with the paraflagellar rod (PFR), axenic amastigotes were rounded to ovoid and displayed a short flagellum restricted to the pocket area. The absence of PFR in axenic amastigotes was confirmed in Western blots and confocal immunofluorescence microscopy, by lack of reactivity with mAb 1B10. The antibody, which specifically labelled the paraflagellar structure, recognized a 70}72 kDa doublet in Trypanosoma cruzi epimastigotes and two 70}74 kDa related proteins in L. braziliensis promastigotes. Surface "#&I-labelling experiments identified promastigote-specific components (" 100, 74, 45}47 and 28 kDa) and at least 1, a 76 kDa polypeptide was specific for the amastigote stage. While axenic amastigotes were agglutinated by both peanut (PNA) and Lens culinaris (LCA) agglutinins, respectively at 50 and 12±5 µg}ml, promastigotes were not agglutinated by PNA and agglutinated in the presence of LCA at concentrations of 100 µg}ml and higher. Axenic amastigotes infected rat bone marrow-derived macrophages and were avidly taken up by J774 cells, from which numerous organisms, able to proliferate at 34 °C in UM-54 medium, could be recovered 48 h later. Key words : Protozoa, Leishmania braziliensis, amastigote-like forms, axenic amastigotes.

 Trypanosomatid flagellates of the genus Leishmania exist as promastigotes within the alimentary tract of blood-feeding Phlebotomine sandflies, and as intracellular amastigotes within phagolysosomes of mammalian macrophages (see Molyneaux & KillickKendrick, 1987). Promastigotes can be easily maintained at temperatures below 28 °C in axenic, commercially available growth media (Chang, Nacy & Pearson, 1986) but amastigotes have to be isolated from either animal lesions or macrophage cultures. These procedures do not assure that preparations are free from host-derived contaminants, such as organelles and macromolecules. Indeed, studies with L. mexicana have clearly demonstrated the presence of host immunoglobulins at the surface of lesion amastigotes (Peters et al. 1995). Contamination of parasite preparations can be avoided by using amastigotes from axenic cultures. * Corresponding author : Departamento de Parasitologia, Instituto de Cie# ncias Biome! dicas, Universidade de Sa4 o Paulo, Av. Prof. Lineu Prestes 1374, CEP 05508-900, Sa4 o Paulo, SP, Brasil. Tel : 55 11 818 7263, Fax : 55 11 818 7417. E-mail : salfieri!biomed.icb2.usp.br Parasitology (1998), 116, 103–113.

Printed in the United Kingdom

Different Leishmania strains have been adapted for growth as amastigotes at temperatures above 28 °C in cell-free media (reviewed by Pan et al. 1993 ; Hodgkinson et al. 1996). Despite the different conditions that have been used, in all instances the differentiation of promastigotes into amastigotes was achieved by increasing the temperature and}or dropping the medium pH. As reviewed elsewhere (Zilberstein & Shapira, 1994), alterations of both temperature and pH trigger changes in gene expression, leading to the appearance of new stages. Axenic amastigotes have been shown to be morphologically and biochemically similar to lesion-derived parasites (Pan, 1984 ; Pan & Pan, 1986 ; Eperon & McMahon-Pratt, 1989 a ; Rainey et al. 1991 ; Bates et al. 1992 ; Pan et al. 1993 ; Pral et al. 1993), infective to animals and macrophage cultures (Pan & Honigberg, 1985 ; Pan & Pan, 1986 ; Eperon & McMahon-Pratt, 1989 a ; Pan et al. 1993), and to display amastigote-specific surface antigens (Pan & McMahon-Pratt, 1988 ; Eperon & McMahon-Pratt, 1989 b ; Pan et al. 1993 ; Hodgkinson et al. 1996). Axenic amastigote cultures so far described have been established and serially maintained under conditions that differ as to the composition and pH # 1998 Cambridge University Press

J. M. F. Balanco and others

of culture media and the temperature for optimal growth (Pan et al. 1993). While some species, for instance L. mexicana and L. donovani, seem to be more easily adapted, others are not, as judged by the single report describing the establishment at 28 °C in Schneider’s Drosophila medium of an axenic amastigote culture of isolate M5052 of L. braziliensis (Eperon & McMahon-Pratt, 1989 a). By employing the strategy of Eperon & McMahon-Pratt (1989 a) and gradually increasing the temperature, we have successfully established at 34 °C in modified UM-54 medium an axenic amastigote culture of isolate M2903 of L. braziliensis. Here we report the adaptation procedure, the requirements for propagation of amastigotes in culture, and provide morphological, biological and biochemical evidence indicating that a high degree of parasite differentiation was achieved in our in vitro culture system.

   The parasites Leishmania (Viannia) braziliensis (MHOM}BR} 75}M2903), here referred as L. braziliensis, was isolated in 1975 from a cutaneous lesion of a man who contracted the infection in the forested area of Serra dos Caraja! s, Para! State, Brazil (Dr Jeffrey J. Shaw, personal communication). The endemic area has been described in a publication of Lainson et al. (1973). Promastigotes, provided by Dr J. Shaw (Instituto Evandro Chagas, Bele! m do Para! , Brasil) several years ago, were maintained by weekly passages at 22 °C in DL-15 medium (Pral et al. 1993), which consists of a 1 : 1 mixture of Dulbecco’s modified Eagle’s medium (DMEM) and Leibovitz L-15 medium (both from Sigma Chemical Company, St Louis, USA), supplemented with 10 % (v}v) inactivated foetal calf serum (FCS), 25 µg}ml gentamicin sulfate, 20 µg}ml haemin, and 15 m Hepes to a final pH of 7±2. Trypanosoma cruzi epimastigotes (CL strain) were cultivated in liver infusion tryptose (LIT) medium (Camargo, 1964).

Axenic cultivation of amastigotes The axenic culture of amastigotes was initiated from stationary-phase promastigotes. Parasites were washed twice by centrifugation (10 min at 1600 g) in Ca#+, Mg#+-free Dulbecco’s buffered saline (PBS) and were resuspended at 2¬10'}ml in modified UM-54 medium (Pan et al. 1993 ; Pral et al. 1993), containing powder for 1 litre of 199 medium with Hanks’ salts (Sigma Chemical Company, St Louis, USA), 2±5 g}l glucose, 5 g}l trypticase (tryptic soy broth, Pasteur Diagnostics, France), 0±75 g}l glutamine, 20 mg}l haemin, 25 µg}ml gentamicin

104

sulfate, 20 % (v}v) of pre-tested inactivated FCS (Bruch Laboratories, Sa4 o Paulo, Brasil), and 25 m of either Hepes (N-[2-hydroxyethyl] piperazine-N «[2-ethanesulfonic acid]) or Mes (2-[N-morpholino] ethanesulfonic acid) to a final pH of 6±3. Parasites were progressively adapted to grow at elevated temperatures (details are given in the Results section) and were serially maintained at 34 °C. Growth was monitored by scoring cell numbers in Neubauer haemocytometers under a phase-contrast light microscope. Macrophage cultures and infection Rat bone marrow-derived macrophages were prepared as described (Alfieri, Zilberfarb & Rabinovitch, 1987). Marrow cell suspensions obtained from Swiss rats and placed in 75 cm# tissue culture flasks (2±5¬10& cells}cm#) were incubated at 37 °C in 5 % CO atmosphere in macrophage growth me# dium, containing 10 % FCS, 10 % L-cell conditioned medium (LCM), and 50 µg}ml gentamicin sulfate in DMEM. On days 5 or 6 the monolayers were gently washed with Hanks’ balanced salt solution buffered with 10 m Hepes (HBSS–Hepes), pH 7±3, and the cells were scraped and washed by centrifugation in HBSS–Hepes. Macrophages resuspended in growth medium were allowed to adhere to 12 mm diameter glass coverslips (1¬10& cells}cover-slip), which were then transferred into 16 mm diameter wells of 24-well plates (Falcon), containing 0±5 ml of complete DMEM. One day later, cells were washed, placed in 0±5 ml of DMEM containing 0±25 % LCM, and were infected with either stationary promastigotes or axenic amastigotes (5 : 1 and 10 : 1 parasite-to-cell ratios). Infected cultures were transferred to 34 °C and were fixed and stained 24 and 48 h later to determine the percentage of infected cells. Macrophages containing at least 1 parasite were scored as infected. The murine macrophage J774 cell line was cultivated at 37 °C in 25 cm# tissue culture flasks in DMEMcontaining10 %FCS.Semi-confluentmonolayers (approximately 2¬10' cells) were infected with 3¬10( axenic amastigotes and further incubated at 34 °C for 48 h. At the end of the incubation period, infected cells were washed with HBSS–Hepes, pH 7±3, and intracellular parasites were recovered as described (Antoine et al. 1987), by selective host cell lysis. Briefly, infected cells in 25 cm# tissue-culture flasks were incubated for 20–25 min with 2±0 ml of pre-warmed 0±01 % sodium dodecyl sulfate (SDS) prepared in HBSS–Hepes. After the addition of 2±0 ml of HBSS–Hepes containing 30 % FCS, cells were scraped, and lysates were passaged 10 times through a 25-gauge needle. Parasites recovered were counted in a haemocytometer.

Axenic Leishmania braziliensis amastigotes

Monoclonal antibodies Monoclonal antibodies were produced by immunizing BALB}c mice with T. cruzi metacyclic trypomastigotes cytoskeletons (Mortara, 1989) denatured with 0±1 % SDS and emulsified in complete Freund’s adjuvant. Splenocytes of immunized mice were fused with SP }0.Ag13 myeloma cells # (Shulman, Wilde & Ko$ hler, 1978) and positive hybrids were screened by immunofluorescence on formaldehyde-fixed parasites or cytoskeletons. Immunoglobulin isotyping was done with the Gibco–BRL hybridoma isotyping kit. MAb 1B10, an IgG1, gave intense flagellar staining on T. cruzi flagellated forms and was used in the present study. Light microscopy For Giemsa-staining, parasites were smeared on glass slides, air-dried and fixed for 10 min with methanol. Infected macrophage cultures were washed twice with HBSS–Hepes prior to the fixation procedure. After Giemsa-staining (30 min in phosphate buffer, pH 7±2), preparations were examined under a light microscope (Carl Zeiss, Germany). Nomarski differential interference microscopy (Olympus Tokyo, Japan) of 1±25 % (v}v) glutaraldehyde-fixed parasites was also used to characterize the morphology of axenic L. braziliensis amastigotes. For immunofluorescence, parasites were fixed with methanol and incubated with mAb 1B10 followed by a mixture of 4,6-diamino-2-phenylenileneindol (DAPI, Molecular Probes, Eugene, OR, EUA) and rabbit, FITC-labelled anti-mouse IgG (Sigma Chemical Company, St Louis, USA). Observations were made on a BioRad 1024-UV confocal system attached to a Zeiss Axiovert 100 microscope, using a 40¬N.A. 1.2 Plan-Apochromatic (DIC) water immersion objective. Images were collected by Kalman averaging at least 15 frames (512¬512 pixels), using an aperture (pinhole) of 2±0 mm maximum. The collected DIC images were sharpened with a minimum setting using BioRad Lasersharp 1024 software version 2.1a. Fluorescence images were generated by dye-sublimation on a Codonics NP1600 printer.

105

Cytoskeletons of live parasites attached to Carboncoated Formvar Ni grids were generated by gentle lysis (Sherwin & Gull, 1989), and fixed with glutaraldehyde. Following a blocking step (Beesley et al. 1984) and incubations with mAb 1B10 and protein A coupled to 10 nm gold particles (Electron Microscopy Sciences, Fort Washington, PA, EUA), samples were stained with 0±7 % aurothioglucose (Sherwin & Gull, 1989) and observed on a Jeol 1200 EX II electron microscope.

Western blots Promastigotes and axenic amastigotes, both cultivated in modified UM-54 medium (promastigotes at pH 7±2, amastigotes at pH 6±3), and Trypanosoma cruzi epimastigotes grown in LIT medium were harvested at late log phase, washed and lysed at 6¬10) organisms}ml of 0±1 % Triton X-100 containing a mixture of proteinase inhibitors (20 µg}ml antipain, 20 µ E-64, 5 m o-phenanthroline and 5 m PMSF). After treatment with sample buffer containing 2-mercaptoethanol and boiling, 7±5 µg of protein of each sample were applied onto the lanes of 10 % resolving SDS–polyacrylamide gels (Alfieri, Balanco & Pral, 1995). After electrophoresis at constant 80 V, one half of the gel was fixed and stained with Coomassie brilliant blue R-250, and the other half was transferred (100 V for 2 h at 4 °C) onto a 0±45 µm nitrocellulose membrane. After transfer, membranes were blocked 30 min at room temperature with PBS containing 0±05 % (v}v) Tween-20 and 5 % non-fat dry milk and next incubated for 1 h at room temperature with mAb 1B10 (1 : 100 dilution in PBS–0±05 % Tween 20). After washing (3 times for 10 min) in PBS–Tween solution, nitrocellulose membranes were incubated for 1 h at room temperature with a 1 : 1000 dilution of an anti-mouse IgG coupled to peroxidase (Sigma Chemical Company, St Louis, USA). Membranes were washed again and incubated for 1 min with the ECL (enhanced chemiluminescence) detection reagent kit RPN 2105 (Amersham, UK). Radioautography (5–15 min exposure) was performed on Hyperfilm-ECL (Amersham). Protein was determined according to Bradford (1976), using bovine serum albumin as standard.

Electron microscopy Parasites were harvested by centrifugation, washed twice in HBSS–Hepes and fixed for 2 h at 4 °C in 1 % (v}v) glutaraldehyde, 1 % (w}v) paraformaldehyde in 0±1  cacodylate buffer, pH 7±2. Cells were pelleted, post-fixed with 1 % (w}v) osmium tetroxide for 1 h at room temperature, processed in ethanolpropylene oxide series and embedded in Spurr. Multiple ultrathin sections stained with uranyl acetate and lead citrate were examined on a JEOL 100 CX electron microscope at 80 kV.

Surface radioiodination Stationary phase promastigotes and axenic amastigotes were washed twice in ice-cold PBS (10 min at 1600 g, in a Sorvall model RT 6000B centrifuge). In 1 ml of ice-cold PBS, 10) cells were transferred to tubes coated with 10 µg of Iodogen (Pierce Chemical Co., USA) and labelled on ice for 10 min with 100 µCi of carrier-free Na"#&I (Amersham, UK). Labelled cells were washed 3 times by centrifugation


106

10 lm

1·0 lm

Fig. 1. Morphology of Leishmania braziliensis promastigotes. (A) Nomarski differential microscopy of glutaraldehyde-fixed parasites displaying extremely long free flagella. (B) Transmission electron microscopy of promastigotes, in which the PFR (arrowheads) is evident (K, kinetoplast ; N, nucleus ; F, flagellum ; L, lipid droplets).

in PBS and the final pellets were lysed by adding 450 µl of 0±1 % (v}v) Triton X-100 containing the proteinase inhibitors described above. After treatment with sample buffer containing 2-mercaptoethanol and boiling, samples were applied onto 10 % resolving SDS–acrylamide gels (Alfieri et al. 1995). After electrophoresis at constant 20 mA and staining with Coomassie Blue R-250, gels were dried and radioiodinated bands were detected after exposure to Kodak XOMAT AR film at ®70 °C, using an intensifying screen. Lectin-mediated agglutination Concanavalin A (Con A), Lens culinaris (LCA) and peanut (PNA) agglutinins were purchased from Sigma Chemical Company (St Louis, USA). Stock

Fig. 2. Growth of axenic Leishmania braziliensis amastigotes in modified UM-54 medium, pH 6±3. Growth curves obtained on 2 different occasions at 33 °C ((A) subculture 50 ; (B) subculture 108) and in the eighth subculture after transfer to 34 °C (C).

solutions (Con A and LCA : 1 mg}ml ; PNA : 2 mg}ml) prepared in PBS were diluted in PBS in the presence or absence of 0±1  competing sugars (mannose for Con A and LCA ; -galactose for PNA), and 50 µl aliquots were distributed into the wells of 96-well flat-bottomed microtitre plates. After the addition of parasite suspensions, final lectin concentrations ranged from 200 to 12±5 µg}ml. Promastigotes and axenic amastigotes, both cultivated in modified UM-54 medium (promastigotes at pH 7±2 ; amastigotes at pH 6±3) were harvested and washed on day 5, and were resuspended at 1¬10) organisms}ml in PBS. Aliquots of 50 µl were distributed into the wells of the microtitre plates. After 1 h incubation at 22 °C (promastigotes) or 34 °C (amastigote assay), plates were gently shaken and scored microscopically for agglutination as described (Jacobson & Schnur, 1990), by 2 independent observers.

 Establishment of the axenic culture of L. braziliensis amastigotes When serially maintained at 22 °C in DL-15 medium, the population of L. braziliensis promastigotes was relatively homogeneous. Both mid-log and stationary-phase parasites exhibited an elongated morphology, the presence of extremely long free flagella being noticeable (Fig. 1 A). Electron micrographs of stationary-phase promastigotes (Fig. 1 B and results not shown) revealed in the cytoplasm abundant ribosomes, small mitochondria and several profiles of endoplasmic reticulum and Golgi cisternae. Lipid droplets and several electron-dense bodies were frequently observed. The paraflagellar rod (PFR), which sometimes expanded and exhibited


10 lm

107

10 lm

0·5 lm

0·5 lm

Fig. 3. Morphology of axenic Leishmania braziliensis amastigotes. (A) Light micrograph of a Giemsa-stained preparation of axenic amastigotes (K, kinetoplast ; N, nucleus). (B) Nomarski differential microscopy of glutaraldehyde-fixed amastigotes ; 2 dividing stages are indicated by arrows. (C and D) Transmission electron microscopy of axenic amastigotes. N, nucleus ; K, kinetoplast ; F, flagellum ; G, Golgi apparatus with numerous associated vesicles.

membranous structures at the exit of the pocket area, was evident in the extracellular portion of the flagella (Fig. 1 B). Stationary-phase promastigotes were used to begin the adaptation of the parasites to elevated temperatures, assuming that this condition would favour promastigote-to-amastigote differentiation. After washing, parasites were resuspended in modified UM-54 medium, pH 6±3, at 2¬10'}ml, and were incubated for 1 week at 28 °C. In the next 3 subcultures, the organisms, at 2¬10'}ml, were incubated for 1 week first at 30 °C, and then at 31 and 32 °C. During this adaptation period, in which a considerable loss of viability was observed, parasites became immotile and underwent pronounced morphological changes, such as rounding up, shortening

and loss of the extracellular flagellum. These changes took place asynchronically, the result being the presence in culture of distinct morphological types, including few but still motile promastigotes, intermediate forms and amastigote-like stages. The number of amastigote stages progressively increased, while that of promastigotes and intermediate forms diminished after cultures had been subpassaged at least 10 times at 32 °C. By this time, when the size of initial inocula was fixed at 1¬10'}ml, parasites could be transferred to and serially maintained at 33 and 34 °C essentially as amastigotes. Growth rates were similar at the 2 temperatures (Fig. 2), and did not vary with long-term culture. Growth curves of axenic amastigotes showed typical log and stationary phases (Fig. 2 and results


108

area. The absence of the PFR in amastigote-like stages was additionally confirmed by lack of reactivity of mAb 1B10 in both Western blots (Fig. 4, lane 3) and confocal immunofluorescence microscopy (Fig. 5). In Western blots, the antibody recognized a 70}72 kDa doublet in T. cruzi epimastigotes (Fig. 4, lane 1), and cross-reacted with two, 70 and 74 kDa proteins of L. braziliensis promastigotes (Fig. 4, lane 2). Immunoelectron microscopy of the cytoskeletons of flagellated forms of L. braziliensis and T. cruzi indicated positive reaction of the antibody with the PFR (Fig. 6 and results not shown).

Surface labelling and lectin-mediated agglutination Fig. 4. Western blot analysis of the reactivity of mAb 1B10. After electrophoresis of total lysates of Trypanosoma cruzi epimastigotes (1) and Leishmania braziliensis promastigotes (2) and axenic amastigotes (3), the gel was transferred onto a 0±45 µm nitrocellulose membrane which was then incubated with mAb 1B10 and anti-mouse IgG coupled to peroxidase as described in the Materials and Methods section. The blot was developed by chemiluminescence followed by radioautography. The positions of molecular weight markers are indicated.

not shown). Parasite population doubled every 10 h, reaching densities varying from 5¬10( to 8¬10(}ml on day 4, at late log phase. In several consecutive passages, parasite growth was shown not to require addition of haemin in the culture medium and was unaffected when initial medium pH was adjusted within the range 5±4 to 6±3 (results not shown). None of the modifications introduced, including the replacement of Hepes by Mes, which is a more suitable buffer for the pH range assayed, seemed to interfere with the amastigote morphology. The latter was retained after numerous (more than 150) subcultures at pH 6±3.

Morphology of axenic amastigotes By Giemsa staining and Nomarski differential interference microscopy, amastigote-like stages were round-shaped and lacked an extracellular flagellum (Fig. 3 A and B). Transmission electron microscopy (Fig. 3 C and D) revealed the presence of numerous ribosomes, prominent mitochondria with scarce cristae enveloping the kinetoplast and a welldeveloped Golgi apparatus with numerous associated vesicles (Fig. 3 C). Occasionally, lipid droplets and electron-dense bodies were observed in the cytoplasm. The short flagellum lacked the PFR and was normally restricted to the pocket area (Fig. 3 C and D) ; sometimes it expanded at the tip (Fig. 3 D) but without extending beyond the limits of the pocket

That axenic amastigotes and promastigotes differentially express surface components as indicated by surface "#&I-labelling experiments and lectin-mediated agglutination assays. Despite the difficulty in labelling the surface components of axenic L. braziliensis amastigotes parallel experiments with promastigotes and axenic amastigotes indicated the differential expression of certain membrane polypeptides (Fig. 7). The major "#&I-labelled bands detected in stationary promastigotes were a high molecular weight (" 100 kDa) component, and polypeptides migrating at 74, 65, 60, 54, 45}47, 40, 34, 28 and 26 kDa. The " 100, 74, 45}47, and 28 kDa bands were absent, and the others were only faintly detected in the axenic amastigotes. In addition, axenic amastigotes clearly displayed a stage-specific, 76 kDa "#&I-labelled polypeptide (Fig. 7). As indicated in Table 1, both axenic amastigotes and stationary promastigotes were similarly agglutinated by Con A, but axenic amastigotes were selectively agglutinated by PNA and reacted with LCA at a minimal effective concentration of 12±5 µg}ml. Agglutination of promastigotes by LCA did take place, but with the lectin at concentrations of 100 µg}ml and higher. Confirming the sugarbinding specificities, lectin-mediated agglutination was completely impaired by 50 m concentrations of the known competitor sugars -galactose (PNA) and -mannose (Con A and LCA) (Table 1).

Infectivity of axenic amastigotes The infectivity of axenic L. braziliensis amastigotes to rat bone marrow-derived macrophages was also demonstrated. At multiplicities of 5 and 10 amastigotes per adherent cell, respectively, 46±2 and 62±6 % of macrophages were infected within 24 h (Table 2). By 48 h, the percentage of infected cells dropped to 22±4 and 34±0 %, respectively. Within the same time period, 38±2 and 21±9 % respectively of marrow macrophages were infected when 10 stationary promastigotes were given per adherent cell. In all


109

Fig. 5. Confocal immunofluorescence microscopy of Leishmania braziliensis promastigotes (A, B) and axenic amastigotes (C, D) and Trypanosoma cruzi epimastigotes}trypomastigotes (E, F), showing that labelling with mAb 1B10 is associated with the flagella. Magnifications in microns ( µm) are indicated on the DIC Nomarski images (A, C, E).

instances, infected cells contained few parasites lodged in small parasitophorous vacuoles, each containing 1 or 2 amastigotes ; only rarely were numerous parasites lodged within the same host cell. Axenic amastigotes were avidly taken up by J774

cells, from which numerous organisms were recovered 48 h p.i. by lysing the host cells with 0±01 % SDS. In 3 experiments performed, 40±0, 56±6, and 30±6 % of the parasites initially given to cell monolayers could be recovered after the lysis procedure.


110

375 nm

Fig. 6. Immunoelectron microscopy of Leishmania braziliensis promastigotes. Section of the extracellular portion of a flagellum, showing the PFR labelling with mAb 1B10 (arrows).

Fig. 7. Surface "#&I-labelling of Leishmania braziliensis promastigotes (P) and axenic amastigotes (A). After labelling, cells were washed and lysed and processed for electrophoresis and radioautography as described in the Materials and Methods section. (A) Coomassie blue staining ; (B) radioautograph of gel (A), in which the "#&I-labelled promastigote components and the 76 kDa, amastigote-specific polypeptide are indicated. The equivalent of 4±5¬10' parasites was loaded onto each lane. The positions of molecular weight markers are given.

When placed at 34 °C in modified UM-54 medium, isolated amastigotes rapidly started to multiply (Fig. 8), a result confirming that viability was retained within the parasitophorous vacuoles of the host cells.  This paper reports the establishment of an axenic culture and the continuous propagation at 34 °C of amastigote-like stages of strain M 2903 of L. brazili-

ensis, an agent of mucocutaneous leishmaniasis. The strategy used was similar to that employed by Eperon & McMahon-Pratt (1989 a) for isolate M5052 of L. braziliensis in that the temperature was gradually changed, but differed with respect to the culture medium used and the temperature required for optimal growth. In our hands, modified UM-54 permitted both the adaptation and the subcultivation of amastigotes. When placed in Schneider’s Drosophila medium (available from Sigma Chemical Company), cultures could be initially maintained, but parasite numbers significantly diminished, the cultures being lost after 5–8 subcultures. In contrast to axenic amastigotes of isolate M5052, which could not be kept at temperatures above 28 °C (Eperon & McMahon-Pratt, 1989 a), those of isolate M2903 could be serially maintained at 34 °C. This finding again emphasizes that optimal conditions for continuous propagation of axenic amastigotes vary and have to be determined for each Leishmania species} isolate (Pan et al. 1993). Axenic L. braziliensis amastigotes were morphologically characterized and shown to be similar to intracellular amastigotes and very different from promastigotes. Parasites were round to ovoid, aflagellate, and lacked the PFR. The absence of PFR in axenic amastigotes was clearly confirmed in Western blots and confocal immunofluorescence by the absence of reactivity with the PFR-specific monoclonal antibody 1B10. It has been pointed out by others (Pan et al. 1993) that although amastigote-like stages initially predominate when the temperature is increased, promastigotes may become frequent as the organisms adapt to a new temperature. This was also the case with L. braziliensis M2903, but only in the initial subcultures at 32 and 33 °C ; thereafter parasites retained the amastigote morphology, an observation indicating that a high level of differentiation was achieved in our in vitro culture system. The effect of medium pH on both growth and morphology of axenic amastigotes was not examined in detail, but no differences became apparent when the initial medium pH was adjusted between 5±4 and


111

Table 1. Agglutination of axenic amastigotes and stationary promastigotes of Leishmania braziliensis by lectins

Lectin* LCA Amastigotes Promastigotes PNA Amastigotes Promastigotes Con A Amastigotes Promastigotes

200 µg}ml

100 µg}ml

50 µg}ml

25 µg}ml

200 µg}ml 50 m 12±5 µg}ml -mannose

200 µg}ml 50 m -galactose

†

0

0

0

0 0

Not done Not done

0

0

0

0 0

0 0

Not done Not done

0 0

0 0

Not done Not done

* Lectin-mediated agglutination was carried out at 22 °C (promastigotes) or 34 °C (amastigotes) as described in the Materials and Methods section, with the lectins at the concentrations indicated. † : Strong (parasites completely agglutinated) ; : moderate, with several rosettes ; : weak, with few rosettes ; 0 : negative.

Table 2. Infectivity of axenic amastigotes and promastigotes of Leishmania braziliensis to rat bone marrow macrophages (Macrophage cultures were infected at the multiplicities indicated and were fixed and stained at the times indicated.)

Developmental stage

% Infection* Parasite : host cell ratio 24 h p.i.

48 h p.i.

Axenic amastigotes Axenic amastigotes Promastigotes

5:1 10 : 1 10 : 1

22±4³5±6 (2) 34±0³9±9 (4) 21±9³5±2 (4)

46±2³10±8 (4) 62±6³6±7 (4) 38±2³11±0 (3)

* The percentages of infected cells³.. (number of separate experiments) are given.

Fig. 8. Viability of parasites recovered from J774 cells 48 h after infection with axenic Leishmania braziliensis amastigotes. Parasites released by host cell lysis were counted and placed in modified UM-54 medium and incubated at 34 °C. Represented are the growth curves obtained in 3 separate experiments.

6±3. The final medium pH was measured in cultures initiated at pH 6±3 and was shown to drop to approximately 5±5 on day 7. These results suggest

that amastigotes of L. braziliensis are similar to those of L. mexicana and L. amazonensis, known to require a pH in the acidic range (Pan et al. 1993 ; Bates et al. 1992 ; Hodgkinson et al. 1996). We have also shown that axenic amastigotes and promastigotes differentially express surface components. This was indicated by (a) the detection of stage-specific "#&I-labelled polypeptides, including a 76 kDa band specifically detected in the amastigote stages, and (b) the lectin-binding capacities, here inferred from agglutination assays. To assure an adequate comparison between the 2 developmental stages, and taking into account the possibility that medium components may regulate carbohydrate surface configurations (Jacobson & Schnur, 1990), in the agglutination assays both axenized amastigotes and promastigotes were grown in modified UM-54 medium (amastigotes at pH 6±3 ; promastigotes at pH 7±2). In general, agglutination profiles of promastigotes were in agreement with previous findings (Almeida et al. 1993 ; Jaffe & McMahon-Pratt, 1988) : parasites strongly agglutinated by Con A, in a mannose-dependent manner, did not react with PNA, and were induced to partially agglutinate by


high concentrations (equal and above 100 µg}ml) of LCA. In contrast, axenized amastigotes agglutinated with lower concentrations of LCA (12±5 µg}ml) and were selectively agglutinated by PNA. These results indicate different sugar-binding capacities, and suggest expression of different membrane glycoconjugates in the 2 developmental stages. The infectivity of axenic L. braziliensis amastigotes was also examined. Both stationary promastigotes and axenic amastigotes were shown to infect rat bone marrow-derived macrophages, but the percentages of infected cells were much lower than those obtained with L. mexicana and L. amazonensis lesion amastigotes (Alfieri et al. 1987 ; Bates et al. 1992), and significantly lower when compared to values obtained 24 h after infection of J774 cells with axenic amastigotes of L. braziliensis M5052 (Eperon & McMahon-Pratt, 1989 a). In addition, with both axenic amastigotes and promastigotes, the percentage of infected cells markedly decreased 48 h p.i. ; at a multiplicity of 10 : 1, values dropped from 62±6 to 34±0 % and 38±2 to 21±9 %, respectively for axenic amastigotes and promastigotes. The reductions in infected cell numbers were attributed to (a) loss of infectivity, probably associated with the long-term culture of L. braziliensis M2903 (see below), and (b) continued division of marrow macrophages at 34 °C, due to incomplete removal of LCM from the culture medium. Indeed, higher values were obtained in a single time-point experiment (results not shown), in which parasites were given at a multiplicity of 5 : 1 to mouse resident peritoneal macrophages. In this case, after infection with axenic amastigotes and stationary promastigotes, respectively 70±0 (.. ¯³6±8) and 72±2 (.. ¯³10±0) % of cells were infected at the end of 96 h (average of quadruplicate cover-slips). Finally, direct demonstration that some intracellular parasites resisted the host’s microbicidal mechanisms was the recovery of viable organisms 48 h after infection of J774 cells. Despite the severity of human infection caused by species such as L. braziliensis, leishmanias belonging to the Viannia subgenus remain poorly understood. Since convenient animal models are presently unavailable, most Viannia isolates, including M2903, have been maintained for several years as promastigotes. This raises relevant questions concerning not only the original biological and biochemical properties of those isolates but also the changes resulting from their long-term culture. For instance, it is well known that promastigotes of several Leishmania species become less infective when serially maintained in culture (Chang et al. 1986 ; Grimaldi & Gesh, 1993). Loss of infectivity has been associated with production of fewer metacyclic forms or metacyclic membrane components (Alexander & Russell, 1992). Cryopreservation of infective promastigotes, use of lesion amastigotes or freshly differentiated promastigotes, although not always easily applicable

112

to the Viannia group, are the known ways to preserve the original characteristics of Leishmania isolates. We have shown here that after infection of J774 cells with axenic amastigotes parasites could be recovered and subcultivated axenically as amastigotes. If cycling of axenic amastigotes in macrophage cell lines proves to preserve the biological and biochemical properties of the parasites, it is expected that the use of axenic amastigote culture systems combined with periodic cycles within macrophages will provide an alternative and excellent model to study the Viannia species. The authors thank Dr Michel Rabinovitch for critical reading of the manuscript, and Cassiano Pereira Nunes for help in final artwork. This work was supported by Brazilian funds from Fundaça4 o de Amparo a' Pesquisa do Estado de Sa4 o Paulo (FAPESP) and Conselho Nacional de Desenvolvimento Cientı! fico e Tecnolo! gico (CNPq).

 , .  , . . (1992). The interaction of Leishmania species with macrophages. Advances in Parasitology 31, 175–254. , . ., , .  , . (1987). Destruction of Leishmania mexicana amazonensis amastigotes by leucine methyl ester : protection by other amino acid esters. Parasitology 95, 31–41. , . ., , . . .  , . . . (1995). A radioiodinated peptidyl diazomethane detects similar cysteine proteinases in amastigotes and promastigotes of Leishmania (L.) mexicana and L. (L.) amazonensis. Parasitology Research 81, 240–244. , . ., , . . .,  ! , . ., , . ., , . .  , . . (1993). Metacyclogenesis of Leishmania (Viannia) braziliensis in vitro : evidence that lentil lectin is a marker of complement resistance and enhanced infectivity. Transactions of the Royal Society of Tropical Medicine and Hygiene 87, 325–329. , . ., , ., , .  , . (1987). Leishmania mexicana : a cytochemical and quantitative study of lysosomal enzymes in infected rat bone marrow-derived macrophages. Experimental Parasitology 64, 485–498. , . ., , . ., , .  , . . (1992). Axenic cultivation and characterization of Leishmania mexicana amastigote-like forms. Parasitology 105, 193–202. , . ., , . . ., , . .  , . . (1984). Immunocytochemical labelling of Bacterioides nosodus pili using an immunogold technique. Journal of General Microbiology 130, 1481–1487. , . . (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Analytical Biochemistry 72, 248–254. , . . (1964). Growth and differentiation in Trypanosoma cruzi. I. Origin of metacyclic trypanosomes in liquid media. Revista do Instituto de Medicina Tropical de Sah o Paulo 6, 93–100.

Axenic Leishmania braziliensis amastigotes , . ., , . .  , . . (1986). Intracellular parasitism of macrophages in leishmaniasis : in vitro systems and their applications. Methods in Enzymology 132, 603–626. , .  -, . (1989 a). I. Extracellular cultivation and morphological characterization of amastigote-like forms of Leishmania panamensis and L. braziliensis. Journal of Protozoology 36, 502–510. , .  -, . (1989 b). Extracellular amastigote-like forms of Leishmania panamensis and L. braziliensis. II. Stage- and species-specific monoclonal antibodies. Journal of Protozoology 36, 510–518. , . .  , . . (1993). Leishmaniases of the new world : current concepts and implications for future research. Clinical Microbiology Reviews 6, 230–250. , . ., , ., , . .  -, . (1996). Leishmania amazonensis : cultivation and characterization of axenic amastigotelike organisms. Experimental Parasitology 83, 94–105. , . .  , . . (1990). Changing surface carbohydrate configurations during the growth of Leishmania major. Journal of Parasitology 76, 218–224. , . .  -, . (1988). The identification of membrane glycoconjugates in Leishmania species. Journal of Parasitology 74, 548–561. , ., , . ., , . .  , . (1973). Leishmaniasis in Brazil. IX. Considerations on the Leishmania braziliensis complex :–importance of sandflies of the genus Psychodopygus (Mangabeira) in the transmission of L. braziliensis braziliensis in north Brazil. Transactions of the Royal Society of Tropical Medicine and Hygiene 67, 184–196. , . .  -, . (1987). Morphology, ultrastructure and life cycles. In The Leishmaniases in Biology and Medicine (ed. Peters, W. & Killick-Kendrick, R.), vol. 1, pp. 121–176. Academic Press, New York. , . . (1989). Studies on trypanosomatid actin. I. Immunochemical and biochemical identification. Journal of Protozoology 36, 8–13. , . . (1984). Leishmania mexicana : serial cultivation of intracellular stages in a cell-free medium. Experimental Parasitology 58, 72–80.

113 , . .  , . . (1985). Leishmania mexicana pifanoi : in vivo and in vitro interactions between amastigotes and macrophages. Zeitschrift fuX r Parasitenkunde 71, 3–13. , . .  , . . (1986). Leishmania mexicana : comparative fine structure of amastigotes and promastigotes in vitro and in vivo. Experimental Parasitology 62, 254–265. , . .  -, . (1988). Monoclonal antibodies specific for the amastigote stage of Leishmania pifanoi. I. Characterization of antigens associated with stage- and species-specific determinants. Journal of Immunology 140, 2406–2414. , . ., , . ., , ., , ., , ., -, .  -, . (1993). Developmental life cycle of Leishmania – cultivation and characterization of cultured extracellular amastigotes. Journal of Eukaryotic Microbiology 40, 213–223. , ., , ., , . ., , .  , . (1995). The role of macrophage receptors in adhesion and uptake of Leishmania mexicana amastigotes. Journal of Cell Science 108, 3715–3724. , . . ., , . ., , . . .  , . . (1993). Leishmania mexicana : proteinase activities and megasomes in axenically cultivated amastigote-like forms. Experimental Parasitology 77, 62–73. , . ., , . ., -, .  , . . (1991). Biochemical and molecular characterization of Leishmania pifanoi amastigotes in continuous axenic culture. Molecular and Biochemical Parasitology 49, 111–118. , .  , . (1989). The cell division cycle of Trypanosoma brucei brucei : timing of event markers and cytoskeletal modulations. Philosophical Transactions of the Royal Society of London, B 323, 573–588. , ., , . .  $ , . (1978). A better cell line for making hybridomas secreting specific antibodies. Nature, London 276, 269–270. , .  , . (1994). The role of pH and temperature in the development of Leishmania parasites. Annual Review of Microbiology 48, 449–470.