Tissue microarray and immunohistochemistry as tools for evaluation of antibodies against Chlamydia-like bacteria

Introduction

Several Chlamydia-like bacteria have recently been identified as potential emerging public health threats or pathogenic agents in animals. Parachlamydia acanthamoebae, Protochlamydia naegleriophila and Simkania negevensis have been reported as possible aetiological agents of pneumonia in humans (Casson et al., 2008a; Corsaro & Greub, 2006; Friedman et al., 2006; Greub et al., 2003). Parachlamydia and related Chlamydia-like organisms have recently been reported in bovine abortion (Borel et al., 2007). Waddlia chondrophila has also been implicated as an abortigenic agent in bovines (Henning et al., 2002; Rurangirwa et al., 1999), and a zoonotic potential for W. chondrophila was suggested by an association of anti-Waddlia antibodies in women and sustained contact with animals (Baud et al., 2007). Infection may also occur following exposure to water colonized with infected free-living amoebae. Thus Parachlamydia acanthamoebae strain BN9 has been identified as a symbiont in free-living amoebae (Amann et al., 1997), whereas Parachlamydia acanthamoebae strain Hall's coccus was identified within an amoeba isolated from a humidifier during investigation of an outbreak. Candidatus Protochlamydia amoebophila UWE25 was originally found in an Acanthamoeba isolate from a soil sample (Collingro et al., 2005). Criblamydia sequanensis was isolated from Seine river water using amoebal co-culture (Thomas et al., 2006) and Neochlamydia hartmannellae was detected in Hartmannella vermiformis isolated from the water conduit system of a dental care unit (Horn et al., 2000).

To define further the possible pathogenic potential of these Chlamydia-like bacteria, new diagnostic tools are needed to demonstrate the agent within tissue lesions. Here, tissue microarray (TMA) technology associated with immunohistochemistry (IHC) was used to determine the specificity of new antisera against various Chlamydia-like bacteria and to investigate the level of cross-reactivity between different Chlamydiales species in formalin-fixed and paraffin-embedded pellets.

Methods

Acanthamoeba castellanii (ATCC 30010) cultures were infected with Parachlamydia acanthamoebae strain Hall's coccus or strain BN9 (ATCC VR-1476), Candidatus Protochlamydia amoebophila strain UWE25 (ATCC PRA-7), Protochlamydia naegleriophila strain Knic, Criblamydia sequanensis strain CRIB-18, S. negevensis (ATCC VR-1471) and W. chondrophila (ATCC VR-1470). H. vermiformis strain BL was infected with N. hartmannellae (ATCC 50802). Uninfected A. castellanii and H. vermiformis cultures were prepared as negative controls.

Vero 76 (ATCC CRL-1587), Caco-2 (ATCC HTB-37) and HEp-2 (ATCC CCL-23) cells were infected with different human and animal chlamydial strains: Chlamydophila pneumoniae strain Kajaaini 6, Chlamydophila abortus strain S26/3 (sheep abortion), Chlamydophila psittaci strain T49/90 (psittacosis agent), Chlamydophila pecorum strain 1710S (swine origin), Chlamydia trachomatis strain LGV 434 and Chlamydia suis strain S45/6 (swine isolate). Uninfected cell cultures of each cell line served as negative controls. Amoebal and cell pellets were fixed in formalin and embedded in paraffin as described elsewhere (Borel et al., 2006).

A cell culture array block including two equally prepared sets of cell and amoebal pellets was created with the TMA Builder (Histopathology; ) according to the manufacturer's instructions. Briefly, the recipient paraffin block with 24 holes of 2 mm diameter each arranged in four columns and six rows was built using the TMA Builder. The whole cell pellets from the donor blocks were punched out with a Paraffin-Punch-Extractor and arrayed in a preformed recipient paraffin block according to the protocol. Four-micrometre sections were cut using a standard microtome.

For IHC, two commercially available chlamydial antibodies were used: a Chlamydiaceae family-specific mouse monoclonal antibody (mAb) directed against the chlamydial LPS (mLPS; clone ACI-P; Progen) and a Chlamydia-specific mouse mAb (IgG1) directed against recombinant Chlamydia trachomatis Hsp60 (clone A57-B9; Milan Analytica).

Polyclonal mouse antisera for the detection of various Chlamydia-like bacteria were produced previously and characterized by immunofluorescence and Western blotting (Casson et al., 2007). Optimization experiments for antigen retrieval and appropriate dilution of antisera for IHC were performed using infected amoebal and cell pellets. Primary antisera were then applied to the TMA as follows: (i) mLPS at a dilution of 1 : 50, (ii) anti-Hsp60 at a dilution of 1 : 1200, (iii) antiserum against Parachlamydia acanthamoebae strain Hall's coccus at a dilution of 1 : 6000, (iv) antiserum against Protochlamydia naegleriophila strain Knic at a dilution of 1 : 20 000 and (v) antisera against Parachlamydia acanthamoebae strain BN9, Criblamydia sequanensis and W. chondrophila at dilutions of 1 : 40 000.

Detection was performed with a detection kit (ChemMate; Dako) according to the manufacturer's instructions. Briefly, paraffin sections were deparaffinized in xylene and rehydrated through graded ethanol to water. Antigen retrieval was performed by 5 min enzyme digestion (Pronase; Dako) for the mLPS antibody, and repeated microwave heating (750 W for 10 min) in citrate buffer, pH 6.0 (Target Retrieval Solution, ChemMate; Dako), for the antisera against Chlamydia-like bacteria. Primary antibodies were incubated for 1 h. For inhibition of endogenous peroxidase activity, the slides were immersed in peroxidase-blocking solution (Dako) for 5 min at room temperature. Negative and positive controls for each section were included as described elsewhere (Borel et al., 2006).

Results And Discussion

The mLPS and anti-Hsp60 antibodies, as well as the antisera against the different Chlamydia-like bacteria, did not exhibit cross-reactivity against uninfected cell or amoebal pellets (Fig. 1a, i). Table 1 shows the antibody titres and cross-reactivities of the different chlamydial antibodies with Chlamydiaceae and Chlamydia-like bacteria, as determined by IHC on the TMA. In general, the antisera exhibited strong reactivity to autologous antigens in amoebal pellets by IHC (Fig. 1b–e). Significant cross-reactivity was observed between closely related species, for instance between Parachlamydia and Protochlamydia strains. In contrast, no or little cross-reactivity was detected between distantly related Chlamydia-like bacteria. Surprisingly, the anti-Waddlia antibody cross-reacted with the Criblamydia sequanensis-infected amoebal pellet (Fig. 1f) but not vice versa. Cross-reactivity of the antiserum against Parachlamydia acanthamoebae strain Hall's coccus with Protochlamydia naegleriophila and N. hartmannellae was observed once, but could not be confirmed in another experiment and was therefore interpreted as questionable. Vero 76, Caco-2 and HEp-2 cells infected with Chlamydiaceae strains remained negative when tested with all polyclonal mouse antisera, indicating no cross-reactivity between any Chlamydiaceae species and the Chlamydia-like bacteria investigated here. The mLPS antibody reacted with the Chlamydiaceae strains as expected (Fig. 1g), but not with Chlamydia-like bacteria. The same was observed for the anti-Hsp60 antibody (Fig. 1h), except for some cross-reactivity to the pellet infected with N. hartmannellae.

(144K): Fig. 1. (a) Uninfected A. castellanii. IHC with mouse antiserum against Parachlamydia acanthamoebae strain Hall's coccus (1 : 6000). (b, c) A. castellanii infected with Parachlamydia acanthamoebae strain Hall's coccus. IHC with mouse antiserum against Parachlamydia acanthamoebae strain Hall's coccus (1 : 6000). (d) A. castellanii infected with Criblamydia sequanensis. IHC with mouse antiserum against Criblamydia sequanensis (1 : 40 000). (e) A. castellanii infected with Protochlamydia naegleriophila strain Knic. IHC with mouse antiserum against Protochlamydia naegleriophila strain Knic (1 : 20 000). (f) A. castellanii infected with Criblamydia sequanensis. IHC with mouse antiserum against W. chondrophila (1 : 40 000). (g) HEp-2 cells infected with Chlamydophila pneumoniae. IHC with the chlamydial mLPS (1 : 50). (h) HEp-2 cells infected with Chlamydophila pneumoniae. IHC with a chlamydial anti-Hsp60 antibody (1 : 1200). (i) Uninfected A. castellanii. IHC with a chlamydial anti-Hsp60 antibody (1 : 1200). In all sections, the secondary antibody was detected using 3-amino-9-ethylcarbazole/peroxidase and counterstained with haematoxylin. Magnifications: x400 (a, b, d–f ); x1000 (c); x200 (g–i).

Table 1. Dilutions and cross-reactivities of different antibodies with Chlamydiaceae and Chlamydia-like bacteria, as determined by IHC on TMA

Antisera against Parachlamydia acanthamoebae strain Hall's coccus and W. chondrophila have already been applied successfully to formalin-fixed and paraffin-embedded bovine placenta specimens and resulted in the first report of Parachlamydia in bovine abortion (Borel et al., 2007). The antibody against Parachlamydia acanthamoebae strain Hall's coccus has also been successfully applied to lung samples from mice experimentally infected with Parachlamydia acanthamoebae: IHC results of lungs correlated well with histopathological lesions and real-time PCR (Casson et al., 2008b).

Reactivity to LPS was not observed in any Chlamydia-like bacteria tested. This suggests that these species do not have a LPS or possess a truncated LPS, as shown for Candidatus Protochlamydia amoebophila UWE25 (Horn et al., 2004). Thus these new agents will not be detected by routine diagnostics when using an antibody directed against chlamydial LPS. Knowledge of the presence of heat-shock proteins in Chlamydia-like bacteria is scarce. Nevertheless, some cross-reactivity of the anti-Hsp60 antibody could be observed in the N. hartmannellae-infected amoebal pellet, but not in H. vermiformis pellets, suggesting the presence of a heat-shock protein-like structure in this species.

In conclusion, TMA technology in combination with IHC is a useful tool for testing the specificity of antibodies for their future use on formalin-fixed and paraffin-embedded tissues. Possible cross-reactivity of antibodies in closely related species should be considered when investigating human or animal tissues with these antisera by IHC.

Acknowledgements

We would like to thank Dr Urs Ziegler and Dr Claudia Dumrese from the Institute of Anatomy, University of Zurich, Switzerland, for providing Chlamydophila pneumoniae-infected and Chlamydia trachomatis-infected HEp-2 cells. We are grateful to the laboratory staff of the Institute of Veterinary Pathology, University of Zurich. We also thank Sebastien Aeby from the Microbiology Institute of the University of Lausanne for technical help. This research was supported by COST Action 855, Animal Chlamydiosis and Zoonotic Implications, Switzerland (SBF no. C05.0141) and by the Swiss National Science Foundation (grant no. 32003B-116445). G. G. is supported by the Leenards Foundation through a career award entitled Bourse Leenards pour la relève académique en médecine clinique à Lausanne.