Coxiella burnetii Exhibits Morphological Change and Delays Phagolysosomal Fusion after Internalization by J774A.1 Cells

Abstract

Coxiella burnetii, the etiological agent of Q fever, is an obligate intracellular bacterium proliferating within the harsh environment of the phagolysosome. Mechanisms controlling trafficking to, and survival of pathogens within, the phagolysosome are unknown. Two distinct morphological variants have been implicated as playing a role in C. burnetii survival. The dormant small-cell variant (SCV) is resistant to extracellular stresses and the more metabolically active large-cell variant (LCV) is sensitive to environmental stresses. To document changes in the ratio of SCVs to LCVs in response to environment, a protein specific to SCV, ScvA, was quantitated. During the first 2 h after internalization of C. burnetii by J774A.1 cells, the level of ScvA decreased, indicating a change from a population containing primarily SCVs to one containing primarily LCVs. In vitro experiments showed that 2 h of incubation at pH 5.5 caused a significant decrease in ScvA in contrast to incubation at pH 4.5. Measuring in vitro internalization of [³⁵S]methionine-[³⁵S]cysteine in response to pH, we found the uptake to be optimal at pH 5.5. To explore the possibility that after uptake C. burnetii was able to delay phagolysosomal fusion, we used thorium dioxide and acid phosphatase to label phagolysosomes during infection of J774A.1 cells. We determined that viable C. burnetii was able to delay phagolysosomal fusion. This is the first time that a delay in phagolysosomal fusion has been shown to be a part of the infection process of this pathogenic microorganism.

Coxiella burnetii, the causative agent of Q fever, is an obligate intracellular organism that replicates within the phagolysosome of host cells. The phagolysosome is a harsh environment where C. burnetii is exposed to degradative proteases, reactive oxygen species, and a pH below 4.8 (13, 19). Despite the inability of C. burnetii to replicate under any known in vitro conditions, some metabolic processes can be supported in vitro where the pH appears to be a critical factor. Thus, C. burnetii can transport and incorporate glucose, glutamate (11), and proline (14) and synthesize nucleic acids (9) and proteins (28) at a mildly acidic, but not at a neutral, pH.

By 48 h postinfection, vacuoles containing C. burnetii (VCB) appear to be typical phagolysosomes. Antibodies to lysosomal membrane proteins label the membranes of VCB (13). The lysosomal enzyme, acid phosphatase, as well as fluid phase markers used to label lysosomes, are found inside VCB (1, 8). There is evidence to suggest, however, that C. burnetii is able to modify its environment. The respiratory burst that accompanies phagolysosomal fusion is significantly reduced during C. burnetii infection of J774A.1 cells (6).

C. burnetii has two cell variants, the large-cell variant (LCV) and the small-cell variant (SCV), both of which are infectious (30). The morphological differences between these two variants have been carefully described (5, 8, 20, 22). The LCV has greater metabolic activity and is more sensitive to environmental stresses than the SCV, while the environmentally stable SCV has a thicker peptidoglycan layer, has more condensed nuclear material and, as the name implies, is smaller in size. On the basis of these studies and the evidence that C. burnetii in the extracellular environment can remain infectious for more than a year (31), it has been suggested that the infective variant in natural aerosols is primarily the SCV and that infection is initiated when phagocytic cells internalize C. burnetii contained in inhaled aerosols. In this model, exposure to the phagolysosomal environment activates C. burnetii metabolism and replication ensues. Based on electron microscopic studies, it has been proposed that intracellular C. burnetii goes through a typical bacterial growth cycle, with an increase in the relative number of LCVs as the population enters log phase (20). Then, as the stationary phase is approached, there is an increase in the number of SCVs, and LCVs occasionally divide asymmetrically producing a spore-like form (21). The bacteria are released from the host cell as a result of host cell lysis or possibly exocytosis, and these “naturally released” C. burnetii infect other host cells.

The present study considers this model by exploiting the recent discovery of an SCV-specific C. burnetii protein, ScvA. It has been shown that when C. burnetii variants are separated on a density gradient that antibody to ScvA binds only to the more dense SCV (12). Using naturally released C. burnetii to infect host cells, we found that the transition from SCV to LCV takes place immediately following uptake and that in vitro this transition takes place most rapidly at a pH higher than that expected in the phagolysosome. In addition, C. burnetii is able to delay phagolysosomal fusion, perhaps to facilitate this transition from SCV to LCV.

MATERIALS AND METHODS

Organisms and cell lines.

J774A.1 cells, a continuous murine macrophage-like cell line (TIB-67; American Type Culture Collection, Rockville, Md.), were used as host cells in all experiments. Cells were maintained in RPMI 1640 medium (Gibco BRL, Grand Island, N.Y.) supplemented with 10% heat-inactivated fetal bovine serum (FBS) and 100 U of penicillin and 100 U of streptomycin per ml at 37°C and 5% CO₂. Nine Mile Phase I C. burnetii (27) was propagated in persistently infected J774A.1 cells maintained as described above without antibiotics. C. burnetii cells were harvested from the supernatants of cultures by differential centrifugation, first at 550 × g for 10 min to remove cell debris and then at 15,000 × g for 1 h to pellet bacteria. The bacterial numbers were determined with a Klett-Summerson photoelectric colorimeter (Klett Manufacturing Co., Inc., New York, N.Y.) using a no. 42 filter (23). Quantitation with a photoelectric colorimeter is an approximation due to variable amounts of cell debris in the final pellet. C. burnetii were inactivated by a 12-h incubation with vigorous shaking in 2% paraformaldehyde–2% glutaraldehyde in phosphate-buffered saline (PBS; 0.1 M NaPO₄, 0.1 M NaCl [pH 7.4]). Salmonella enterica serovar Enteritidis, used as a biodegradable particle, was grown in Luria-Bertani broth and inactivated by incubation with gentamicin at 250 μg/ml, in addition to 25 μg of gentamicin per ml added to the tissue culture media to ensure inactivation.

Electron microscopy.

A total of 5 × 10⁵ J774A.1 cells were seeded onto 15-mm Thermonox coverslips (Fisher Scientific, Santa Barbara, Calif.) in the wells of a 24-well Falcon plate (VWR Scientific Products, Seattle, Wash.) and allowed to adhere for 12 h. Media were removed, and J774A.1 cells were incubated with C. burnetii at an approximate multiplicity of infection (MOI) of 1,000:1 in RPMI 1640–10% FBS. After 15 min C. burnetii was removed by washing three times with RPMI 1640 prewarmed to 37°C. Cells were fixed at 0, 15, 30, 60, and 120 min in Immunofix {0.5% gluteraldehyde, 2% paraformaldehyde, 0.1 M PIPES [piperazine-N,N′-bis(2-ethanesulfonic acid)]} for 1 h at 25°C, after which fixative was replaced with 0.1 M PIPES buffer. Thermonox coverslips were transferred to glass vials for further processing. Samples were dyhydrated through an ethanol series and infiltrated with L. R. White resin. To embed the monolayer, Thermonox coverslips were inverted on 7- by 7-mm Histoprep disposable base molds (Fischer, Pittsburgh, Pa.) and cured overnight at 60°C. The Thermonox coverslips were removed from the hardened resin by heating, and J774A.1 cells were preserved as a monolayer on the surface of the resin block. Thin sections were taken from the block face and mounted on Formvar-coated nickel grids for immunogold labeling.

Immunogold labeling of ScvA was conducted by blocking samples for 1 h in TBST (10 mM Tris, 250 mM NaCl, 0.3% Tween 20 [pH 8.2])–1% bovine serum albumin (BSA), followed by a 1 h of incubation in rabbit anti-ScvA polyclonal sera (kindly provided by R. Heinzen) diluted 1:25 in TBST-BSA. Samples were then washed three times for 15 min in TBST-BSA and incubated for 1 h with goat anti-rabbit antibody which was conjugated to 10- or 20-nm gold particles. Samples were again washed twice with TBST-BSA, twice with TBST, and once with double-distilled H₂O. Sections were stained with uranyl acetate-KMnO₄ for 10 min and examined with a JEOL 1200 EX transmission electron microscope.

To label phagolysosomes with thorium dioxide, J774A.1 cells were seeded at 5 × 10⁵ cells on 15-mm Thermonox coverslips as described above. Cells were incubated for 6 h in thorium dioxide (kindly provided by O. Baca) diluted in RPMI 1640–10% fetal calf serum to label the lysosomes. This was followed by a 12-h chase period to allow internalization of all thorium dioxide. These cells were incubated with Nine Mile Phase I C. burnetii or inactivated C. burnetii at an MOI of approximately 1,000:1 for 1 h or inactivated serovar Enteritidis at 100:1 for 30 min. Inocula were removed by washing three times with prewarmed RPMI 1640, and the media were replaced.

At specified time points, inoculated J774A.1 cells were fixed for electron microscopy in 4% paraformaldehyde–2% glutaraldehyde–0.1 M sodium cacodylate at 4°C for 12 h. Fixative was washed off with 0.1 M cacodylate buffer (pH 7.5) and Thermonox coverslips were processed as described above. Samples were embedded in Epon, inverted on Histoprep disposable base molds, and cured overnight at 70°C. Thin sections were mounted on Formvar-coated copper grids, stained with 2% uranyl acetate and lead citrate, and examined with a Hitachi 600 transmission electron microscope.

To label phagolysosomes with acid phosphatase stain, J774A.1 cells were seeded on Thermonox coverslips as described above. J774A.1 cells were inoculated with viable or inactivated C. burnetii at an MOI of approximately 1,000:1 or 0.8-μm-diameter latex beads (Sigma, St. Louis, Mo.). Inocula were removed by washing three times with warmed RPMI. At each time point samples were fixed in 1.5% glutaraldehyde–0.1 M sodium cacodylate buffer [pH 7.4] for 15 min. Fixative was removed, and samples were washed three times for 2 min each time in Tris-malate buffer (0.2 M Tris, 0.2 M maleic acid, with 0.8 g of NaOH/100 ml added [pH was adjusted to 5.0 with 10 N NaOH]). Samples were incubated in the reaction mix (one part 1.25% β-glycerophosphate [pH 5.0], one part H₂O, one part Tris-malate buffer, and two parts 0.2% lead nitrate [added drop by drop to avoid precipitation] mixed immediately before use) for 1 h with shaking at 37°C. Samples were then washed three times for 2 min in Tris-malate buffer and fixed for 12 h at 4°C in 3% glutaraldehyde–1% sucrose–0.2 M sodium cacodylate buffer (pH 7.4) (7). Samples were prepared and embedded for electron microscopy as above.

Activation and metabolic labeling.

Activation of C. burnetii was achieved by incubating 10⁹ C. burnetii in 0.5 ml in acid activation buffer (32 mM KPO₄, 15 mM NaCl, 152.2 mM KCl, 100 mM glycine, 5 mM l-glutamate, 20 μM l-proline), adjusted to pH 7.0, 5.5, or 4.5 at 37°C for 2 h. To verify that the pH did not change during the incubations, the pH was monitored immediately before and after all incubations. Pelleted C. burnetii were embedded in 10 μl of 4% low-melting-point agarose and processed for immunolocalization and electron microscopy as described above.

Approximately 5 × 10⁹ C. burnetii were incubated in 250 μl of acid activation buffer at pH 4.5, 5.5, or 7.0 with 50 μCi of [³⁵S]methionine-[³⁵S]cysteine (Express; NEN, Boston, Mass.) and 50 μg of cyclohexamide per ml to prevent incorporation by residual host cells. For each experiment the harvested C. burnetii cells were divided equally into three samples (approximately 10⁹ C. burnetii cells/sample). After a 16-h incubation, C. burnetii cells were washed three times in PBS to remove the extracellular [³⁵S]methionine-[³⁵S]cysteine and denatured by boiling in Laemmli sample buffer. The counts per minute (cpm) in the denatured samples, from three independent experiments, were determined by scintillation counting. To visualize incorporation, proteins were resolved by sodium dodecyl sulfate-polyacrylamide electrophoresis (SDS-PAGE) (17) using a minigel apparatus (Bio-Rad, Hercules, Calif.). Gels were loaded with equal numbers of bacteria per lane which was confirmed by staining with Coomassie brilliant blue R250 (Bio-Rad). Autoradiographs of dried gels were analyzed with an Image Master scanning densitometer and software (Pharmacia Biotech, Uppsala, Sweden).

RESULTS

Kinetics of ScvA degradation following internalization.

J774A.1 cells infected with a mixture of SCV and LCV C. burnetii were examined at 2 and 6 h postinfection by transmission electron microscopy (TEM). The morphology of the C. burnetii in the inoculum, as well as samples viewed at 2 h postinfection, indicate that these organisms are primarily SCVs. By 6 h postinfection, internalized C. burnetii cells appear to be primarily LCVs (Fig. 1). Infected host cells and the inoculum were prepared for immunoelectron microscopy, and the sections were incubated with anti-ScvA polyclonal sera followed by colloidal gold labeling. To quantitate the apparent decrease in SCVs following internalization, we monitored the decrease in ScvA, a protein specific to the SCVs of C. burnetii, using anti-ScvA polyclonal sera and colloidal gold. There was a reduction in the ScvA/SCV ratio from an average of 4.1 to 2.2 ScvA/SCV during the 2 h immediately after internalization. In addition, we found that the percentage of LCV C. burnetii, as defined by the complete lack of gold label, increased from 25% in the inoculum to 83% during this time period (Fig. 2).

FIG. 1.

(A) TEM image showing a typical vacuole in J774A.1 cells infected with C. burnetii at 2 h postinfection. (B) J774A.1 cells infected with C. burnetii, in the same experiment as panel A, at 6 h postinfection. At least 20 vacuoles were examined at each time point. Bar, 100 nm.

FIG. 2.

In vivo experiments show the decrease in ScvA label after infection with viable C. burnetii. ScvA, a small-cell-specific protein, was labeled with anti-ScvA polyclonal sera. The percentage of LCV (□, as defined by the absence of any anti-ScvA polyclonal serum label) increases over the 2 h, while at the same time the average number of ScvA/SCV (●) decreases. The data were obtained from a minimum of 50 C. burnetii organisms for the inocula and at each time point shown.

Effect of pH on ScvA.

When C. burnetii is internalized the endosome is acidified by vacuole ATPases to pH 5.5 (32) and, after phagolysosomal fusion, the pH drops still further to pH 4.5 (19, 24). The pH is one element of the vacuolar environment that could initiate the decrease in ScvA; therefore, we monitored the effect of pH on ScvA in vitro. Identical samples of C. burnetii were incubated at 37°C for 2 h in an acid activation buffer with the pH adjusted to 7.0, 5.5, or 4.5. The samples, as well as an untreated control, were then fixed for immunoelectron microscopy, and ScvA was labeled with colloidal gold. The average number of gold particles (ScvA label)/C. burnetii cell was determined for each treatment in four independent experiments. The number of gold particles/C. burnetii incubated at a pH of 7.0 or 4.5 was found to be not statistically different from untreated controls. The decrease in ScvA label seen with incubation at pH 5.5 was statistically significant in all experiments (P = 0.001), with an average decrease of 42% after 2 h (Fig. 3). This decrease is less than the decrease in ScvA label seen in vivo during the same time period (an average decrease in ScvA of 71%). This suggests that either some of the SCV are degraded in vivo or that the decrease in ScvA is not only a response to pH.

FIG. 3.

C. burnetii were incubated for 2 h in acid activation buffers at pH 7.0, 5.5, or 4.5. Samples were labeled with anti-ScvA polyclonal serum and examined by electron microscopy. The ScvA/C. burnetii ratio was compared to that found in an untreated control. At least 200 C. burnetii organisms were examined for each treatment in four independent experiments, with error bars indicating the standard deviation. Only treatment at pH 5.5 yielded a ScvA/C. burnetii ratio that was significantly different as compared by Student t tests, with a reduction of 42% and a 95% confidence interval (P ≤ 0.001).

Effect of pH on internalization of [³⁵S]methionine-[³⁵S]cysteine.

To further explore the effect of pH, C. burnetii was incubated for 16 h in acid activation buffers containing cyclohexamide and [³⁵S]methionine-[³⁵S]cysteine under the same conditions as described above. The ³⁵S internalized by C. burnetii was determined for three experiments (Fig. 4). There was significantly more ³⁵S in organisms incubated at pH 5.5 than at pH 4.5 or 7.0. However, at pH 4.5, the estimated pH of the phagolysosome, there was 2.5 times as much ³⁵S as in organisms incubated at pH 7.0. Proteins were resolved by electrophoresis, with an equal number of organisms being loaded onto each lane (Fig. 5). Proteins labeled with [³⁵S]methionine-[³⁵S]cysteine during acid activation were visualized by autoradiography. Although there were no obvious differences in protein expression, incubation at pH 5.5 resulted in an increase in protein synthesis by C. burnetii. This increased synthesis and the observed decrease in ScvA in response to pH may be necessary for the establishment of infection.

FIG. 4.

Approximately 5 × 10⁹ C. burnetii were incubated for 16 h in acid activation buffers at a pH of 7.0, 5.5, or 4.5 with [³⁵S]methionine-[³⁵S]cysteine and cycloheximide. Internalization of labeled amino acids was determined by scintillation counting. Bars indicate the average cpm internalized during incubation for three experiments, with error bars indicating the standard deviation. Internalized cpm for each treatment were significantly different from the other two by Student t tests, with a confidence interval of 95% (P ≤ 0.05).

FIG. 5.

Approximately 10⁹ C. burnetii were incubated in acid activation buffers with pH set to 7.0, 5.5, or 4.5 for 16 h with [³⁵S]methionine-[³⁵S]cysteine and cycloheximide. C. burnetii organisms were denatured in Laemmli buffer, and proteins were separated by SDS-PAGE. (A) Coomassie blue-stained image with equal numbers of C. burnetii per lane. (B) Autoradiograph of the same gel. Samples incubated at pH 4.5 (lane 1), 5.5 (lane 2), 7.0 (lane 3) are shown. (C) Comparison of proteins synthesized during incubations at pH 4.5 and 5.5. Autoradiographs were exposed for different lengths of time. Lane 4, containing sample incubated at pH 4.5, was exposed for 72 h, and lane 5, containing sample incubated at pH 5.5, was exposed for 24 h.

Phagolysosomal fusion.

Most intracellular pathogens are able to modify their intracellular environments. The possibility that C. burnetii is also able to affect phagolysosomal fusion was investigated here. Two methods, thorium dioxide labeling and acid phosphatase staining, were used to visualize phagolysosomal fusion, and both techniques yielded very similar results. Viable and formaldehyde-inactivated C. burnetii were used in both of these experiments. There are limitations to both of these methods. During thorium dioxide labeling, an extended chase period after labeling is necessary for complete internalization of label in order to avoid false-positive results found when extracellular label is taken up with the inoculum. During the chase period, new lysosomes may form that do not contain the marker. Acid phosphatase staining is occasionally variable, resulting in little or no staining of some cells (7). Using thorium dioxide or acid phosphatase, we obtained maxima of 70 and 90% labeled VCB, respectively.

Lysosomes of J774A.1 cells were labeled with thorium dioxide, a fluid-phase marker, and then labeled cells were inoculated with viable or inactivated C. burnetii or with inactivated Salmonella serovar Enteritidis. The thorium dioxide is delivered to vacuoles when phagolysosomal fusion occurs (Fig. 6A and B). By 1 h postinoculation, thorium dioxide was found in 27% of the vacuoles containing viable C. burnetii and 53% of the vacuoles containing inactivated C. burnetii. The percentage of labeled vacuoles containing viable C. burnetii began to increase at 6 h postinoculation, and by 24 h it had reached 51%, while 66% was seen for inactivated C. burnetii. The percentage of labeled VCB was compared to the labeling of vacuoles containing inactivated serovar Enteritidis, a biodegradable control. At 1 h postinoculation 60% of the vacuoles that contained serovar Enteritidis also contained thorium dioxide, and by 2 h postinoculation the serovar Enteritidis was completely degraded, indicating a rapid rate of phagolysosomal fusion and that the degradative processes of the J774A.1 cells were intact (Fig. 7A).

FIG. 6.

(A) TEM image showing J774A.1 cells labeled with thorium dioxide are shown at 2 h postinfection with viable C. burnetii. Two SCVs of C. burnetii, indicated by small arrows, are visible within unfused vacuoles. Larger arrows indicate lysosomes labeled with thorium. Bar, 500 nm. (B) Thorium dioxide is visible within a vacuole containing an LCV of C. burnetii 6 h after infection. A small arrow indicates C. burnetii, and larger arrows indicate the thorium dioxide label. Bar, 500 nm. (C) Acid phosphatase stain showing lysosomes wrapping around C. burnetii at 4 h postinfection. Bar, 100 nm.

FIG. 7.

Results of two experiments to determine the rate of phagolysosomal fusion. (A) The lysosomes of J774A.1 cells were labeled with the fluid-phase marker, thorium dioxide, and subsequently inoculated with inactivated serovar Enteritidis (○), inactivated C. burnetii (▵), or viable C. burnetii (▴). Vacuoles containing both inoculant and thorium dioxide were considered to be fused. When cells were inoculated at 0 h, the percentage was assumed to be zero. (B) Phagolysosomal fusion visualized by histochemical staining of the lysosomal enzyme, acid phosphatase. J774A.1 cells were inoculated with latex beads (○), inactivated C. burnetii (▵), or viable C. burnetii (▴), and the percent fusion was determined as described above. At least 50 vacuoles were counted for each sample at each time point. Fusion was determined at 24 h postinfection with thorium dioxide label, and it was found that the fusion of vacuoles containing viable C. burnetii was slightly less than that seen for vacuoles containing inactivated C. burnetii.

The percentage of fused vacuoles was also monitored by determining the presence of acid phosphatase. After inoculation of J774A.1 cells with viable or inactivated C. burnetii or with latex beads, cells were stained for acid phosphatase, a lysosomal enzyme that is delivered to phagosomes when fusion takes place (Fig. 6C). Similar to results seen with thorium labeling, at 1 h postinoculation 31% VCB were labeled, while 52% of the vacuoles containing inactivated C. burnetii were labeled for acid phosphatase (Fig. 7B). At 1.5 h postinoculation, 70% of vacuoles containing latex beads, a nondegradable control, were labeled, and this percentage remained constant throughout the remainder of the experiment. Taken together, the results of both of these methods clearly demonstrate that vacuoles containing viable C. burnetii exhibit significantly reduced phagolysosomal fusion during the first 6 h postinfection.

DISCUSSION

Although the ability of C. burnetii to proliferate within the harsh environment of the phagolysosome is well established, little is known about processes that allow this intracellular parasite to colonize this niche. The present study examines C. burnetii immediately after its internalization by the mouse macrophage cell line J774A.1 and correlates pH to changes observed in this pathogen. Previous investigators have shown that while phagolysosomal fusion is necessary for C. burnetii proliferation, the parasitopherous phagolysosome may be modified, as demonstrated by a reduced oxidative burst (6). We show here for the first time that phagolysosomal fusion is delayed by viable C. burnetii but not by inactivated organisms. This raises the question: if C. burnetii requires the phagolysosomal environment for proliferation, why would phagolysosomal fusion be delayed?

We found that when J774A.1 cells were infected with a mixture of LCVs and SCVs, within 6 h postinfection many vacuoles contained primarily LCVs, suggesting that this transition begins immediately after internalization during the time that phagolysosomal fusion is delayed. The transition from SCV to LCV observed by electron microscopy was quantitated using serum to an SCV-specific protein, ScvA. A decrease in ScvA was clearly demonstrated in vivo and in vitro in response to pH. ScvA has been shown to bind DNA, and it has been proposed that it might play a part in the DNA condensation and protection in the SCV (12). During the transition from SCV to LCV, DNA condensation and protection would no longer be necessary.

Several in vitro studies using C. burnetii mechanically released from an intracellular environment have established its acidophilic nature. The incorporation of glutamate and the incorporation and catabolism of glucose are highest at a pH close to 4.5, while the catabolism and transport of glutamate and the transport of glucose are maximal at pH 3 (11). Proline transport is also maximal at pH 3 (14). Because it has been shown that naturally released C. burnetii differs from mechanically released organisms in DNA and protein synthesis (9, 28), we have used naturally released organisms in this study to mimic the population of C. burnetii available to infect cells in vivo. In vitro incorporation assays were carried out to compare C. burnetii metabolism at the endosomal pH of 5.5 (32) and at the phagolysosomal pH of 4.5 (19). We found that pH 5.5 is optimum for in vitro internalization of [³⁵S]methionine-[³⁵S]cysteine by C. burnetii. To mimic the transition from the endosomal pH to the phagolysosomal pH, C. burnetii was activated for 1 or 3 h at either pH 5.5 or 4.5 and then labeled with [³⁵S]methionine-[³⁵S]cysteine at pH 4.5 for 2 h. There was no difference in incorporation or in proteins expressed between samples activated at pH 5.5 and those activated at pH 4.5 (data not shown). These results may reflect the limitations of this in vitro system or it is possible that, in vivo, C. burnetii protein synthesis is reduced at pH 4.5. The in vitro pH optimum for metabolic activities found here introduces the possibility that C. burnetii may initially prefer a slightly less acidic environment.

Morphological differences between C. burnetii LCV and SCV are suggestive of the morphological differences seen in other gram-negative bacteria, when organisms in log phase are compared to those in stationary phase. In both cases there is an increase in lipopolysaccharide (4, 15), a thickening of the peptidoglycan layer (3, 18), an increase in the protein cross-linking of the outer membrane to the peptidoglycan (3, 29), and a decrease in outer membrane proteins (2, 3). These comparisons suggest that the SCV may be similar to other gram-negative bacteria in the stationary phase and that the LCV morphology may be similar to that of gram-negative bacteria in the log phase. In addition to these morphological differences it has been shown that, when exposed to adverse conditions such as starvation, E. coli (26), Vibrio cholerae (10), Salmonella serovar Enteritidis (26), Campylobacter jejuni (25), Legionella pneumophila (16) and other organisms require a resuscitation period before they resume proliferation. Release into the extracellular environment by lysed host cells may impose stresses on C. burnetii such that it also requires resuscitation prior to proliferation in the phagolysosome.

This study reports for the first time a delay in phagolysosomal fusion induced by viable C. burnetii upon infection of J774A.1 cells. Accompanying this delay is a transition from a population of mixed LCVs and SCVs to one that is composed largely of LCVs. These observations suggest that the current model of the C. burnetii developmental cycle should be modified to one in which the transition from SCV to LCV begins in the endosome prior to phagolysosomal fusion. A more complete understanding of the C. burnetii developmental cycle will lead to the development of a means to interrupt this cycle and the subsequent control of C. burnetii infection.