Morphologic and molecular evaluation of Chlamydia trachomatis growth in human endocervix reveals distinct growth patterns

Abstract

In vitro models of Chlamydia trachomatis growth have long been studied to predict growth in vivo. Alternative or persistent growth modes in vitro have been shown to occur under the influence of numerous stressors but have not been studied in vivo. Here, we report the development of methods for sampling human infections from the endocervix in a manner that permits a multifaceted analysis of the bacteria, host and the endocervical environment. Our approach permits evaluating total bacterial load, transcriptional patterns, morphology by immunofluorescence and electron microscopy, and levels of cytokines and nutrients in the infection microenvironment. By applying this approach to two pilot patients with disparate infections, we have determined that their contrasting growth patterns correlate with strikingly distinct transcriptional biomarkers, and are associated with differences in local levels of IFNγ. Our multifaceted approach will be useful to dissect infections in the human host and be useful in identifying patients at risk for chronic disease. Importantly, the molecular and morphological analyses described here indicate that persistent growth forms can be isolated from the human endocervix when the infection microenvironment resembles the in vitro model of IFNγ-induced persistence.

Keywords: Chlamydia trachomatis; bacterial persistence; endocervix; human; indole; interferon gamma.

Figure 1

Identification of chlamydial inclusions in endocervical cells retrieved by cytobrush from C. trachomatis-infected women. Cytobrush specimens immediately placed in (A) Surepath or (B,C) Cytolite were processed as described in the methods, stained with anti-chlamydial LPS-FITC (green), Evans blue (red) and DAPI (blue) and visualized by fluorescent deconvolution microscopy. Note the morphology and staining pattern in (A) suggests a single inclusion that is wrapped around the nucleus. Scale bar is 5 μm.

Figure 2

Distribution of chlamydial LPS, OmcA, and CT223 in C. trachomatis-infected epithelial cells harvested from human endocervix. Cervical cells were dual-labeled with anti-chlamydial LPS (green) and (A) OmcA (red) or (B) CT223 (red), counterstained with DAPI (blue) and visualized by fluorescence deconvolution microscopy, as described in the methods. Colocalization of antigen is visualized in yellow. Scale bar is 5 μm. Image (C) is a 3-dimensional maximum intensity projection extracted from a deconvolved stack of an infected cell in image (B). Note the irregular “dashed line” membranous staining of CT223. Grid is 1 μm.

Figure 3

Ultrastructural images of inclusions from the cervix of a single C. trachomatis patient. (A) An inclusion-containing electron micrograph from Patient 1, processed and analyzed as described in the Materials and Methods section. The entire micrograph is shown in the large image on the left, with the scale bar indicating a distance of 1 μm. The smaller image on the right is a magnified view of the features indicated by the green arrow. For this image, the scale bar indicates a distance of 0.5 μm. Mature and immature EBs can be observed in the magnified view. Quantification of the inclusion and its contents is indicated below the image. The inclusion area was 13.23 μm². Forty EBs, RBs, and atypical RBs were counted as particles within the inclusion, with an average particle area of 0.14 μm². The plot indicates the frequency distribution of particles sized by their cross-sectional areas. The anticipated cross-sectional areas for EBs and RBs are indicated in the frequency plot. The plot indicates the frequency distribution of particles sized by their areas. (B) A second inclusion-containing electron micrograph from Patient 1, processed and analyzed as described above. The smaller image on the right indicates a magnified view of the area highlighted by the green arrowhead on the larger image. Two EBs and two RBs can be observed in the magnified view. Quantification of inclusion area, particle area, and particle area frequency distribution indicates the characteristics of this inclusion to be very similar to those in (A).

Figure 4

Ultrastructural images from the cervix of a second C. trachomatis infected patient. (A) An inclusion-containing electron micrograph from Patient 2, processed and analyzed as described for Figure 3. The entire micrograph is shown in the large image on the left, with the scale bar indicating a distance of 1 μm. The smaller images on the right are magnified views of the features indicated by the green (upper image) and red (lower image) arrows. For these images, the scale bar indicates a distance of 0.5 μm. The area indicated by the green arrowhead contains an EB, and an RB. The area indicated by the red arrowhead contains an RB undergoing binary fission. Quantification of the inclusion and its contents is shown below the images. The inclusion area was 26.57 μm². Sixty-two EBs, RBs, and atypical RBs were counted as particles within the inclusion, with an average particle area of 0.21 μm². The plot indicates the frequency distribution of particles sized by their cross-sectional areas. The anticipated cross-sectional areas for EBs and RBs are indicated in the frequency plot. (B) A second inclusion-containing electron micrograph from Patient 2. The smaller image on the right indicates a magnified view of the area highlighted by the green arrowhead on the larger image. A RB and an EB-sized particle can be observed in the magnified view. Quantification of inclusion area, particle area, and particle area frequency distribution indicates the characteristics of this inclusion to differ significantly the inclusion shown in (A) or Figure 3. Very few particles with the area of an EB were observed. The average area, 0.68 μm², was skewed toward the anticipated area of a RB, with several larger particles also observed. (C) A third inclusion-containing electron micrograph from Patient 2. Three inclusions were observed in this micrograph, indicated by the large black arrows, at least two of which are within the same cell. The largest inclusion contained particles with the characteristic area of RBs and EBs, as highlighted by the green arrow, and displayed in the upper image on the right. The two smaller images contained particles with areas corresponding to those of RBs, as highlighted by the red arrow, and displayed in the lower image on the right. Quantification of the three inclusions is indicated below. The largest inclusion had an area of 26.68 μm², whereas the smaller inclusions had areas of 5–6 μm². The average area of particles within these three inclusions was similar to that observed in (B), and distinct from the observations in (A). (D) A fourth inclusion-containing micrograph from Patient 2. The characteristics of this inclusion are close to those observed for the inclusions in (B,C). The green arrowhead indicates particles with the area of EBs (upper magnified image on the right), while the red arrowhead indicates a particle with the area of an RB (lower magnified image on the right). Similar to the inclusions seen in (B,C), the particles within this inclusion were skewed toward the size of RBs, with an average area of 0.58 μm².

Figure 5

Examples of RB binary fission detected in electron micrographs from Patient 2. Three types of binary fission were observed, including, septum formation where the plane of division, indicated by a green arrowhead, is predicted to result in equal division (A–C), or unequal division (D–F), or where multiple septum formation events, indicated by the red arrowheads, were apparently present (G–H). The scale bar indicates a distance of 0.5 μm.