This investigation has confirmed that an APP response occurs in sheep experimentally-infected with C. pseudotuberculosis
, and extends the previous report of an increase in the concentration of serum Hp in CLA [9
] by demonstrating that increases also occur in the concentrations of SAA and AGP. The result of both the profile and AUC of the responses shows a significant increase in Hp, SAA and AGP in CLA. The profiles of the protein response of Hp and SAA were similar, although the fall in SAA after day 7 p.i. was more rapid than that of Hp. In contrast, the profile of the increase in the concentration of AGP was substantially different from the other APP with a more gradual increase followed by a slower decline in the serum concentrations of Hp and SAA. This difference in the profile of the APP during disease progression corresponds to previous studies in cattle, in which AGP was found to be elevated in a greater proportion of clinical cases involving chronic disease in comparison to conditions involving acute conditions [19
]. The acute phase reaction of Hp in CLA has been described previously [9
], but the reaction of SAA and the continued production of AGP during the transition to chronic stages of the infection are novel observations. The evidence here further supports the concept that raised AGP could be a biomarker of chronic conditions in sheep.
Most investigations of the acute phase protein response, and its control by pro-inflammatory cytokines, focus on the immediate or 'acute phase' response to the infection. CLA is known to develop as a chronic infection and the most interesting finding of the study is the extension of the AGP response into the later phase of the disease while the Hp and SAA responses fall more rapidly. The phases of infection during CLA have been described as an initial phase (day 1–4 p.i.) characterised by recruitment of neutrophils to the inoculation site and the draining lymph node, an amplification phase (day 5–10 p.i.) characterised by the development of pyrogranuloma and a stabilisation phase characterised by maturation and persistence of the pyrogranuloma [4
]. On the basis of the previous study on Hp response [9
] and with the results presented here, the initial and amplification phases would correspond to the acute phase of the infection with raised levels of Hp and SAA. The stabilisation phase, when Hp and SAA concentrations return towards the baseline while AGP is maintained at a significantly higher concentration than in controls would correspond to the period when the acute phase of CLA is transformed into a chronic condition.
The mechanism of the switch between acute phase and the chronic phase of infections has been a relatively neglected area of investigation, and the experimental CLA model used in this study would be a means for further investigation of this aspect of the innate immune system. It is possible that alterations in expression of cytokines at different stages of the infection could relate to the differing profile of APP responses. In a study of cytokine expression, following infection with wild type C. pseudotuberculosis
it was found that at the site of inoculation, the inflammatory cytokines TNF-α and IL1-β were up regulated on day 7 p.i., whereas on day 28 p.i IL1-β and IL-6 showed increased expression while the expression of TNF-α had returned to baseline. In contrast, T-cell associated cytokines such as IL2, IL4 and IFN-γ were up regulated at the draining lymph node [4
]. The contrasting AGP and Hp/SAA responses could be indicative of an alteration in the output of cytokine from these different sites.
Production of IFN-γ by leukocytes in response to excretory-secretory products of C. pseudotuberculosis
was able to differentiate goats into high and low responding groups [5
]. It would be of interest to determine if the variation found between individuals in their responses of Hp, SAA and AGP were also different between such groups. However the limited numbers in this study precluded assessment of such a relationship which would need larger experimental animal groups to provide a definitive answer. It could be possible that the use of the IFN-γ stimulation test for the presence of CLA [5
] could be enhanced by simultaneous assessment of the Hp, SAA and AGP concentrations.
There has recently been growing evidence that the APP can be synthesised by non-hepatic tissue, such as lung [20
] and mammary gland tissue [22
], and activated neutrophils [23
] and alveolar macrophages [24
]; in the latter, differing AGP glycan isoforms were found during an acute phase response. It is possible that the extended acute phase response of AGP in serum in CLA could be a result of such non-hepatic synthesis of AGP, but it would require further investigation to assess this possibility and to identify the source of the increased production of this protein.
The increase in antibody titre against PLD was in agreement with previous studies of antibody response to experimental infection with C. pseudotuberculosis
], though with other experimental approaches specific antibody can be demonstrated as early as 5 days after infection [5
]. As would be expected with mediators of the innate and acquired immune systems, the APP responses were earlier than that of the specific antibody response to PLD. The concentrations of Hp and SAA were returning towards those in the control animals by the time the IgG levels were raised (day 11 p.i.) although the mean AGP concentration continued to rise after the IgG response was initiated and was consistent with a switch from the innate to the acquired immune responses. Of great interest would be an investigation to determine if the response of all or of any of the acute phase proteins in the immediate p.i. period, as measured by AUC, would correlate to the later IgG response in individual animals. The number of animals was not sufficient to examine this possibility in the present study.
The APP are valuable markers of disease in man and in animals [11
] but in veterinary medicine are generally regarded as markers of the 'acute phase' of infection or immunity. The findings here demonstrate that AGP responds in different ways to the progress of infection compared to Hp and SAA and therefore the difference in response profile indicates that investigation of a range of APP could provide additional diagnostic information on the progress of disease. Feline AGP is known to be a differential test in the diagnosis of feline infectious peritonitis [25
] and it could be that ovine AGP may serve the same purpose in the diagnosis and prognosis of CLA in sheep.