Chlamydia species-dependent differences in the growth requirement for lysosomes

Abstract

Genome reduction is a hallmark of obligate intracellular pathogens such as Chlamydia, where adaptation to intracellular growth has resulted in the elimination of genes encoding biosynthetic enzymes. Accordingly, chlamydiae rely heavily on the host cell for nutrients yet their specific source is unclear. Interestingly, chlamydiae grow within a pathogen-defined vacuole that is in close apposition to lysosomes. Metabolically-labeled uninfected host cell proteins were provided as an exogenous nutrient source to chlamydiae-infected cells, and uptake and subsequent labeling of chlamydiae suggested lysosomal degradation as a source of amino acids for the pathogen. Indeed, Bafilomycin A1 (BafA1), an inhibitor of the vacuolar H(+)/ATPase that blocks lysosomal acidification and functions, impairs the growth of C. trachomatis and C. pneumoniae, and these effects are especially profound in C. pneumoniae. BafA1 induced the marked accumulation of material within the lysosomal lumen, which was due to the inhibition of proteolytic activities, and this response inhibits chlamydiae rather than changes in lysosomal acidification per se, as cathepsin inhibitors also inhibit the growth of chlamydiae. Finally, the addition of cycloheximide, an inhibitor of eukaryotic protein synthesis, compromises the ability of lysosomal inhibitors to block chlamydial growth, suggesting chlamydiae directly access free amino acids in the host cytosol as a preferred source of these nutrients. Thus, chlamydiae co-opt the functions of lysosomes to acquire essential amino acids.

Figure 1. The chlamydial inclusion is in close apposition to lysosomes throughout its developmental cycle.

Cells were transfected with LAMP1-YFP (green) and were infected with Chlamydia trachomatis serovar L2 (red) or left uninfected (UI). At indicated times post-infection, cells were fixed and stained for chlamydiae. Arrowhead indicates an early chlamydial inclusion.

Figure 2. Lysosomal degradation products are transferred to chlamydiae.

AHA-labeled protein was fed to infected cells, and cells were subsequently fixed and processed for confocal microscopy. Chlamydiae were counter-stained with an antibody to facilitate identification of AHA label within chlamydiae. For Ctr, arrowheads indicate individual organisms positive for labeling with AHA whereas for Cpn, arrowheads indicate inclusions with multiple organisms. Note the intensity of the stain in the zoomed insets and the more uniform staining of Cpn organisms compared to Ctr. Bright staining on the cell surface is AHA-labeled protein lysate that was not endocytosed.

Figure 3. Bafilomycin A1 (BafA1) has selective effects on the host cell.

(A) LysoTracker™ Red staining of untreated (UTD) or 100 nM BafA1-treated HEp-2 cells for the indicated times. Bar represents 10 µm. (B) Cell viability was measured by flow cytometry of Annexin/PI stained cells. Viability of uninfected and Cpn-infected cells +/− 100 nM BafA1 (top) and of uninfected and Ctr-infected cells +/− 100 nM BafA1 (bottom). (C) BafA1 does not disrupt uptake of fluorescent sphingomyelin by chlamydiae during BafA1 treatment. Cells were infected with Ctr L2 and pulsed with fluorescent ceramide, which is converted to sphingomyelin in the Golgi and subsequently transported by the exocytic pathway. Images were acquired at 630× magnification on an epifluorescent microscope. (D) Ultrastructural analysis of untreated and BafA1-treated uninfected HEp-2 cells. Note the presence of enlarged vesicular structures that contain undigested materials. Scale bar equals 1 µm.

Figure 4. Impairing lysosomal functions blocks growth of Chlamydia.

(A) IFU recovery of Ctr serovar L2 from cells treated with the indicated doses of BafA1. (B) 100 nM BafA1 was added to Cpn or Ctr at different times relative to infection: toi  =  time of infection, pi = 24 hr post-infection, pt = 24 hr pre-treatment. For IFU recovery experiments, data were pooled from at least three independent experiments with the average and standard deviation shown in percentage of the untreated control. Two-tailed Student's t tests assuming equal variance were performed for each experiment and p values combined using Fisher's Test for Combining p values. * indicates p<0.0001 for treatments compared to untreated control. (C) Ultrastructural analysis of untreated and BafA1 (100 nM)-treated Cpn- or Ctr-infected cells. Arrow indicates lysosome with undegraded material. Bars for cells infected with Cpn only (A and B) represent 2 µm whereas all others represent 1 µm. (D) BafA1 does not induce an anti-chlamydial effector. Cells were infected first with Cpn and then with Ctr to determine if Ctr could protect Cpn during BafA1 treatment, which would suggest a Ctr secreted factor in allowing growth. Arrow indicates Cpn inclusion whereas arrowhead indicates Ctr inclusion. The bottom panel shows a cell co-infected with Cpn- and Ctr L2 and treated with BafA1, with only a distinguishable Ctr inclusion present.

Figure 5. Cathepsin proteolytic activity is required for optimal growth of Chlamydia.

(A) LysoTracker™ Red staining of untreated or 75 µM Cathepsin Inhibitor III-treated HEp-2 cells for the indicated times. Bar represents 10 µm. (B) Cathepsin Inhibitor III compromises chlamydial growth. This cathepsin inhibitor was titrated for its effect and was used at 40 µM. For IFU recovery experiments, data were pooled from at least three independent experiments with the average and standard deviation shown in percentage of the untreated control. Two-tailed Student's t tests assuming equal variance were performed for each experiment and p values combined using Fisher's Test for Combining p values. * indicates p<0.0001 for treatments compared to untreated control.

Figure 6. TFEB knockdown and sub-inhibitory doses of BafA1 block chlamydial growth.

(A) Schematic of the role of TFEB in controlling lysosomal function. (B) Fold change in TFEB expression normalized to endogenous control 18S rRNA expression in un-transfected (UTD), negative control siRNA transfected (Neg), mock-transfected (Mock), or TFEB siRNA transfected cells. Data represent 3 pooled experiments performed in triplicate. (C) Measurement of IFU recovery during transfection in the presence and absence of a sub-inhibitory dose (5 nM) of BafA1. Data represent at least 3 pooled experiments performed in triplicate. * indicates p<0.02 vs. all conditions.

Figure 7. Autophagy is not a required source of nutrients for Chlamydia.

(A) Western blot analyses demonstrated a marked accumulation of LC3 in BafA1-treated (100 nM) cells. Equal loading was confirmed by immunoblot with anti-actin antibody. (B) BafA1 treatment induced the accumulation of GFP-LC3 localized to vesicular structures in HEp-2 cells. (C) IFUs recovered from Ctr L2-infected, paired Atg7 ^+/+ or Atg7 ^−/− MEFs that were left untreated (UTD) or were treated with BafA1 (50 nM). Data were pooled from three experiments performed in triplicate with the mean and standard deviation shown. Two-tailed Student's t tests assuming equal variance were performed for each experiment and p values were combined using Fisher's Test for Combining p values (* indicates p<0.005 for comparisons of treated and untreated cells). There was no significant difference in IFU recovery between Atg7 ^+/+ and Atg7 ^−/− MEFs.

Figure 8. Cycloheximide rescues Chlamydia from the growth inhibitory effects of BafA1 and Cathepsin Inhibitors.

IFUs recovered from (A) Cpn- or (B) Ctr L2-infected cells treated with BafA1 (100 nM) or Cathepsin Inhibitor III (40 µM) +/− cycloheximide (1 µg/ml) are shown. Data were pooled from at least three experiments performed in triplicate, with the mean and standard deviation shown as the percentage of the untreated control. Two-tailed Student's t tests assuming equal variance were performed for each experiment and p values combined using Fisher's Test for Combining p values (* indicates p<0.05 for cycloheximide treatments compared to relevant sample). (C) Immunofluorescence images of Cpn and Ctr inclusion in HEp-2 monolayers that were either untreated, BafA1-treated, or BafA1/cycloheximide-treated. Note the complete absence of mature Cpn inclusions in BafA1-treated cells.

Figure 9. Lysosomally digested material promotes chlamydial growth in amino acid-free medium.

(A) IFUs recovered from Cpn-infected cells grown in the indicated media are shown. (B) IFUs from Ctr L2-infected cells grown in the indicated media are shown. For (A) and (B), data were pooled from at least three experiments performed in triplicate with the mean and standard deviation shown as the percentage of the untreated control. Two-tailed Student's t tests assuming equal variance were performed for each experiment and p values combined using Fisher's Test for Combining p values (* indicates p<0.0001 for comparisons to HBSS alone; + indicates that two of three experiments testing the effects of exogenous BSA in Ctr infected cells were not significant). (C) Cell survival is not compromised by amino acid or serum deprivation. Viable cell numbers in the different growth media were determined at the indicated intervals by trypan blue exclusion. Data were pooled from two experiments with the mean and standard deviation shown. HBSS = Hank's Balanced Salt Solution. BSA = 5% BSA in HBSS. iIMDM = incomplete IMDM (lacking serum). (D–F) Amino acid preferences of Ctr and Cpn as revealed by labeling with AHA amino acid or AHA-labeled protein lysate fed to infected cells and imaged by confocal microscopy. (D) Pulse-labeling of chlamydiae with free AHA amino acid in the presence of cycloheximide. Note the weak labeling of Cpn compared to Ctr. (E) Infected cells were fed AHA-labeled protein cell lysate in the absence of cycloheximide. Note the comparative levels of stain between species. (F) Cells infected with both Cpn and Ctr were either pulse-labeled with free AHA amino acid or fed AHA-labeled protein cell lysate, as above, to directly compare the relative labeling efficiencies between the species. For panels (D–F), organisms were counterstained with antibody to facilitate visualization of organisms.

Figure 10. Chlamydial acquisition of amino acids.

The data support a model where under normal growth conditions, chlamydiae compete with the host cell for free cytosolic pools of amino acids as well as those produced from lysosomal proteolytic degradation. EBs come pre-equipped with both amino acid and oligopeptide transporters. Early after infection and differentiation, the RB acquires amino acids likely using oligopeptide transporters. At later times, Ctr successfully competes with the host cell using its amino acid transporters whereas Cpn continues to rely on oligopeptide production from the lysosome. BafA1 abolishes lysosomal degradation of proteins. Cycloheximide blocks host cell usage of amino acids, thus increasing pools of free amino acids and ameliorating the effects of disrupting lysosomal-mediated protein degradation.