Xinqiao was isolated from ticks in China and its phase I phenotype was demonstrated in a previous study [13
]. In this current study, C. burnetii
Xinqiao was used to infect BALB/c mice and a large amount of C. burnetii
was found in the spleens and livers of the infected mice by qPCR analysis. The Coxiella load in spleens was significantly higher compared with that in the other organs of the infected mice, indicating that the mouse spleen is the most important organ for C. burnetii
propagation and its Coxiella load may reflect the severity of C. burnetii
infection. The highest level of Coxiella in spleens of the infected mice was found on day 7 pi and then gradually decreased, indicating that the infected mice recovered gradually from the severe infection. These results also indicate that the combination of the sublethal challenge mouse model and the qPCR assay may be a useful and sensitive way to evaluate severity of the infection caused by different C. burnetii
strains and evaluate efficiency of drugs or vaccines against this pathogen.
In order to identify the seroreactive proteins of C. burnetii Xinqiao, the whole cell lysates of the organism was separated by 2-D electrophoresis. Immunoblot analysis using the sera of mice obtained at days 14, 21, and 28 pi, indentified 4, 9, and 14 of the separated proteins, respectively. This indicated that the specific immune responses to C. burnetii developed progressively in the infected mice with additional antigens of C. burentii recognized as the immune response grew further. In addition, 15 of the proteins were recognized by sera from two patients with acute Q fever. Among these seroreactive proteins, 9 proteins were recognized by both the mouse and human sera, indicating that these proteins are able to elicit similar humoral immune responses to C. burnetii infection in both species.
A total of 20 seroreactive proteins were recognized by the positive mouse or human sera by mass spectra of MALDI-TOF-MS. GroEL, a conserved heat shock protein (HspB) [14
], has been reported as a major immunodominant antigen of C. burnetii
]. YbgF, a tol-pal system protein that involved in bacterial outer membrane stability [16
], was found in both phases of C. burnetii
]. GroEL and YbgF were both recognized by the sera of C. burnetii
-infected mice and the Q fever patient sera in this study and have been previously documented as seroreactive antigens using a proteomic approach [7
]. While Com1, Mip, and OmpH were recognized by the sera of C. burnetii
-infected mice but were not recognized by Q fever patient sera. This difference might be due to the fact that mouse and human sera were from different infection stages or there were differences in humoral immune responses to C. burnetii
infection between mice and humans.
Com1 was first identified as an outer membrane-associated seroreactive protein of C. burnetii
by Hendrix and colleagues [17
]. Mip is a cell-surface associated peptidylprolyl-isomerase related to macrophage infectivity potentiator protein [18
] and plays a role in enhancing clearance of bacteria from spleens of infected mice [19
]. OmpH is a putative outer membrane chaperone protein required for efficient release of translocated proteins from the plasma membrane [20
]. The 3 proteins had also been recognized as immunodominant antigens in other studies [7
]. DnaK, a surface-associated protein playing a role in assisting with folding of nascent polypeptide chains [23
], and RplL, a ribosomal protein involved in translation, were previously recognized as seroreactive [9
]. In this study, DnaK and RplL were most seroreactive when probed with the sera of patients with acute Q fever but were nonreactive when probed with the sera of C. burnetii
-infected mice. Additionally, another 13 seroreactive proteins identified in this study were housekeeping enzymes, including FbaA, AtpD, and Tuf2 which are involved in metabolism and biosynthesis. Eight of these proteins were previously identified as seroreactive antigens [7
]. This indicated that metabolic enzymes released from C. burnetii
organisms were exposed to the host immune system and induced a specific antibodies response.
Nineteen of the 20 seroreactive proteins identified in this immunoproteomics study were successfully expressed in E. coli cells and the resultant recombinant proteins were used to fabricate a protein microarray. To evaluate their serodiagnostic potential, the protein microarray was probed with Q fever patient sera. As a result, 7 of the 19 proteins (GroEL, YbgF, RplL, Mip, Com1, OmpH, and Dnak) gave a modest sensitivity of more than 48% when probed with acute late Q fever patient sera. We noted that inconsistency existed between immunoproteomic and microarray data: the reaction of Com1 was stronger than that of Mip, OmpH or YgbF in immunoblot assay, whereas FI value of Mip, OmpH or YgbF was higher than that of Com1 in microarray assay with Q fever sera. The inconsistency might be caused by the fact that the Q fever sera recognized linear epitopes of Coxiella proteins in immunoblot assay whereas they recognized conformational epitopes of recombinant proteins in protein microarray assay.
Our results also showed that the average FI value of the 7 major seroreactive proteins probed with acute late sera were significantly higher than those probed with acute early or normal sera, which is generally in accordance with IgG titers determined in IFA. This result firmly suggests that the 7 major seroreactive proteins are immunodominant antigens of C. burnetii and they have capability to evoke strong humoral immune responses in C. burnetii infection. However, compared to IFA, the lower sensitivity of some individual proteins in microarray assay with Q fever patient sera, especially sera in acute early stage, was observed, which might be due to the fact that there were remarkable variation in immune recognition patterns for Q fever and differences between the two assays in calculating positive values. When the seroreactive proteins were analyzed in combination, 98% of antibody responders to one or more of the 7 major seroreactive proteins could be found among the Q fever patients. The remarkable variation in immune recognition patterns for Q fever requires multi-antigen combination to cover the different antibody responses and thus achieve the highest possible test sensitivity.
YbgF, RplL, Mip, Com1, and OmpH were considered as potential antigens for diagnosis of Q fever by other investigators using in vitro transcription and translation (IVTT)-based microarray of C. burnetii
Nine Mile strain, indicated that Xinqiao strain isolated in China shares these major seroreactive antigens with Nine Mile strain [19
]. Two heat shock proteins GroEL and Dnak were also recognized as major seroreactive antigens in this study. The positive frequencies of GroEL probed with acute early and acute late, and convalescent Q fever patient sera were 84%, 88%, and 83%, respectively, higher than the other major seroreactive proteins, suggesting that GroEL is an excellent molecular marker for Q fever. Additionally, the positive frequencies of YbgF with these Q fever patient sera were 44%, 62%, and 77%, lower than GroEL but higher than the other 5 major seroreactive proteins, indicating that it is a better protein antigen for Q fever diagnosis.
Rickettsial spotted fever caused by tick-borne infection may share similar clinical feature with Q fever. Legionella pneumonia is caused by Legionella pneumophila which is the bacterium closely related to C. burnetii with genomic homology and similar clinical presentations. Pneumonia is the major clinical presentation of acute Q fever and most bacterial pneumonia is caused by S. pneumoniae. These bacterial infections must be distinguished from Q fever using serological or molecular tests. Therefore, the 7 Coxiella proteins were used to fabricate a small microarray for further analysis of specificity with the sera of patients with other infectious diseases. The average FI value of each protein probed with acute late Q fever patient sera was significantly higher than that probed with the sera of patients with one of the three other infectious diseases, which indicated that the major seroreactive proteins of Coxiella can be distinguished from other bacteria in general. YbgF and DnaK displayed no cross-reaction with any of the tested sera, and Com1, Mip, OmpH and GroEL cross-reacted with one or two of the sera of patients with rickettsial spotted fever, Legionella pneumonia or bacterial pneumonia. RplL cross-reacted with two of the Legionella pneumonia patient sera and three of the streptococcal pneumonia patient sera. In this analysis, these Coxiella proteins gave a modest specificity for recognizing of Q fever patient sera, suggesting that they are potential serodiagnostic markers for Q fever.
Notably, GroEL had the highest sensitivity and modest specificity for recognizing of Q fever, which may be the most important antigen used for the diagnosis of Q fever. The antigen combination, GroEL, YbgF and Com1, may give a more authentic specificity. Refinement of antigen combination and the production of fusion molecules comprised of the major seroreactive antigens described herein may lead to improved sensitivity and specificity for the development of a rapid, accurate, and convenient seorodiagnostic test of Q fever.