Evaluation and pathophysiological characterisation of a bovine model of respiratory Chlamydia psittaci infection

Chlamydia (C.) psittaci is a gram-negative, obligate intracellular bacterium, capable of inducing respiratory disease and persistent infection. The host range of this zoonotic pathogen includes not only birds and man, but also various wild and domestic mammals. Knowledge about pathogenesis and functional consequences of C. psittaci infection in the mammalian lung has remained elusive to date. The present project aimed to develop, to evaluate and to characterise a bovine model of experimentally induced respiratory C. psittaci infection which might be beneficial for both human and veterinary medicine. Bovines were chosen as the host because (i) bovine C. psittaci infection closely reflects the situation in a natural host, and (ii) the bovine lung is relevant to model functional consequences of ventilatory disorders due to its segmental anatomy and the lack of collateral airways. Moreover, the pathogenetic potential of C. psittaci for bovines and potential transmission routes were evaluated by assessing clinical and immunological variables of health and lung function as well as the shedding of this potentially zoonotic pathogen. A total of 69 Holstein-Friesian calves aged 6 – 9 weeks were included in four separated studies (STUDY 1 – 4) and were challenged as follows: - Intra-bronchial application of viable C. psittaci, strain DC 15 (n = 35) - Intra-bronchial application of UV-inactivated C. psittaci, strain DC 15 (n = 6) - Intra-bronchial application of cell culture medium (n = 25) - Naïve calves (i.e. sentinels) were socialised with acutely diseased animals due to experimentally induced C. psittaci infection (n =3). Each intra-bronchial application was performed according to a previously developed protocol. In brief, a total volume of 6 mL inoculum/calf was endoscopically administered and distributed in a standardised way into 8 defined bronchi of each animal. In STUDIES 1 & 2, dose response relationships were evaluated up to 3 days post inoculation (dpi). Thus, 14 calves received different doses of viable C. psittaci (106 – 109 inclusion forming units (ifu)/calf). The use of two control groups enabled the separated evaluation of effects mediated by cell culture medium (n = 4) and chlamydial cell components (108 UV-inactivated ifu/calf, n = 6). In STUDIES 3 & 4, time courses and pathophysiological consequences of an acute respiratory C. psittaci infection induced by 108 ifu/calf (n = 21) were assessed in a follow-up study ending at 35 dpi. In addition to the assessment of intra-individual time courses within each calf, a control group was challenged by cell culture medium only (n = 21) for inter-subject comparison. Furthermore, a group of naïve calves (sentinels, n = 3) was socialised with the infected animals, and aimed at evaluating risks and routes of transmission as well as the course of natural infection. In the different studies, systemic host response was assessed by clinical signs, variables of innate immune response, acute phase proteins, and acid-base parameters including electrolytes and metabolites in the peripheral blood. To evaluate pulmonary inflammation, the concentrations of eicosanoids and total protein were measured in broncho-alveolar lavage fluid (BALF) in addition to BALF cytology and pathological as well as histological characterisation of lung lesions. Pulmonary function tests included arterial blood gas analysis, haemoxymetry as well as the assessment of respiratory mechanics, alveolar ventilation and the pattern of breathing. For the latter, non-invasive pulmonary function methods (originally adapted from human medicine and previously evaluated for calves) were used, i.e. impulse oscillometry, volumetric capnography, and helium dilution re-breathing test. Excretion and distribution of the pathogen was assessed by specific nucleic acid-based detection assays (real-time PCR, microarray). The results obtained in the 4 STUDIES can be summarised as follows. Administration of the viable C. psittaci strain succeeded in inducing reproducible infections. Disease outcome was largely dose-dependent. While 106 ifu/calf resulted in mild clinical signs, doses of 107 to 108 ifu/calf induced a moderate respiratory disease, and 109 ifu/calf a severe clinical illness. Also, severity of respiratory and clinical signs, extent and quality of pneumonia, and systemic inflammation increased with increasing challenge doses and resulted in gradually reduced efficacy of pulmonary gas exchange. After inoculation of the finally defined dose of 108 ifu/calf, clinical signs peaked 2 – 3 days post inoculation (dpi). Signs and markers of acute disease subsided considerably, but not completely, within 10 dpi after acute illness. Sentinels acquired the infection but did not develop visible signs of an apparent disease. Thus, two different facets of C. psittaci infection in bovines could be distinguished by the present model: (i) acute clinical disease after experimental challenge and (ii) clinically inconspicuous persistent infection after natural exposure to the pathogen. In both groups, systemic spread and ongoing host-pathogen interactions were detected by chlamydaemia, faecal shedding of the pathogen, and slightly increased levels of monocytes and lipopolysaccharide-binding protein (LBP) in blood, indicating that neither group eliminated the chlamydiae within 5 weeks after exposure. Pathophysiologically, inflammatory cells, mainly neutrophil granulocytes, were recruited into the lung during the acute phase of infection. Pulmonary inflammation resulted in tissue damage, accumulation of detritus and protein rich fluid causing reduced gas exchange, airway obstructions and pulmonary restrictions. Attempts to compensate for alveolar hypoventilation and hypoxaemia included the elevation of both respiratory rate and minute ventilation. Consequences for the acid-base equilibrium were not only determined by pulmonary dysfunctions, compensatory mechanisms or anaerobe metabolism, which are classically assessable by the Hendersen-Hasselbalch approach. A strong ion model revealed mixed acid-base disorders, mediated by metabolic and immunologic influences, i.e. by the reduction of chloride and sodium (assessed by the strong ion difference, SID) and by hypo-albuminaemia and hyper-gammaglobulinaemia (assessed by the sum of non-volatile weak acids, Atot). After experimental challenge, the pathogen was not only detected in lung tissue and blood, but it was also excreted via faeces, nasal excretions and exhaled breath. Despite complete daily cleaning of the animal rooms, contamination sufficed to transmit the infection to naïve sentinels. Although these naturally exposed animals did not develop obvious clinical signs, they became frequent faecal shedders, suggesting a risk of spreading the pathogen under natural conditions. Humoral immune response was generally weak. Only two thirds of experimentally challenged calves developed specific antibodies against C. psittaci detected by immunoblotting. In sentinels, no humoral immune response was observed. In conclusion, the newly introduced large animal model provides a valuable support in elucidating the pathophysiology and complex interactions during pathogenesis of C. psittaci infection in the mammalian lung. In comparative medicine, it can be regarded as a translational model. Pulmonary function tests derived from human medicine and applicable to calves provided comparable data between bovines and humans about pulmonary dysfunction involved in the pathogenesis of this respiratory infection. In veterinary medicine, this biologically relevant model may serve as a suitable basis for studies focusing on vaccine development and treatment evaluation.