Chlamydophila abortus is a leading cause of ovine abortion in Ireland and the United Kingdom. In order to emulate as closely as possible the naturally occurring disease process, pregnant ewes were inoculated with 2 × 106 inclusion forming units of C. abortus. In this way, sera from a clinically relevant experimental infection were used to identify immunologically relevant proteins of C. abortus that are expressed during ovine infection.
C. abortus has a biphasic life cycle comprising elementary and reticulate bodies. Elementary bodies are the extracellular infectious form compared to the metabolically active intracellular reticulate bodies. The phenotypic differences between purified EB and purified RB were apparent by transmission electron microscopy and confirmed by differential protein expression by 1-D SDS-PAGE. As expected, several proteins of EB were reactive with sera from infected ewes compared to RB. Results highlight the specific host interactions with infectious EB during disease processes, compared with the paucity of reactivity of sera with intracellular RB.
All experimentally infected pregnant ewes had seroconverted by day 14, and all reacted specifically with antigens from purified EB of 40 and 90 kDa, compared to negative controls (Fig. ). The 40-kDa antigen was also evident in 2-D immunoblots in several isoforms and, when excised from protein gels, was subsequently identified as the major outer membrane protein (MOMP). This provides an excellent positive control for our experimental design, as MOMP is a well-characterized outer membrane protein from C. abortus
and is also currently used in diagnostic assays (52
). While a 90-kDa antigen is apparent on 2-D immunoblots of EB separated over a pH of 3 to 10, it was not detected on immunoblots separated over a pH range of 4 to 7. Previous studies have identified 90-kDa antigens as the polymorphic outer membrane proteins (12
). However, while not identified at 90 kDa, the polymorphic outer membrane protein CAB776, which has a predicted molecular mass of 163 kDa, was reactive with sera from experimentally infected ewes and identified at an apparent molecular mass of 50 kDa; this is likely a POMP breakdown product and is supported by the mass spectrometry results which showed that the identified peptides were derived from the C terminus of the POMP (see Table S1 in the supplemental material).
Interestingly, sera from fetuses of experimentally infected ewes failed to react with the 40- and 90-kDa antigens in 1-D immunoblots, even at day 43 postinfection. The most immunogenic protein detected at day 43 postinfection was at an apparent molecular mass of 26 kDa. Similarly, sera from 75% and 93.8% of infected ewes were reactive with a 26- and 28-kDa antigen, respectively. No fetal antibodies specific for EB were detected at 30 dpi. The failure of the fetuses to respond serologically to C. abortus
until day 35 is consistent with the time required for the organism to spread from the infected trophoblast to the fetal tissues and the development of lesions of multifocal necrosis in the major fetal organs (37
The 26- to 32-kDa antigens were clearly detected on 2-D immunoblots with both ewe and fetal sera. In contrast to the MOMP, which was readily identified on corresponding protein gels, increased resolution and protein amounts of EB were required on 2-D protein gels in order to detect proteins of similar masses and pI values which aligned with these antigens. When corresponding protein spots were excised for identification by mass spectrometry, several proteins were identified. However, it is of note that the macrophage infectivity potentiator (MIP) protein was consistently identified, with amino acid coverage ranging from 14 to more than 68%. Further, this suggests that MIP is highly immunogenic, given its reactivity with ovine sera, relative to the amount present. MIP is a lipoprotein that belongs to the FKBP-type PPIase family and has been identified in other chlamydial species, including C. trachomatis
and C. caviae
, on both inner and outer membranes of EB (22
). Further, the recombinant MIP of C. trachomatis
induced release of the proinflammatory cytokines interleukin-1β (IL-1β), tumor necrosis factor alpha (TNF-α), IL-6, and IL-8 in human monocytes/macrophages through Toll-like receptor 2 (TLR2)/TLR1/TLR6 and CD14, suggesting a significant role for MIP in the pathogenesis of C. trachomatis
-induced inflammatory responses (3
). MIP has also been identified in other intracellular bacteria, including Legionella pneumophila
, in which it is reported to facilitate dissemination within host tissues, as demonstrated by transmigration of lung epithelial cells due to a serine protease activity and specific interaction with collagen (50
Additional antigens of EB which are reactive with ovine sera and identified by mass spectrometry include proteins with functions in cell shape, metabolism, lipid synthesis, and transport. The 60-kDa cysteine-rich outer membrane protein (OmcB) maintains the structure of the outer membrane (7
) while a putative UDP-N
-acetylglucosamine acyltransferase facilitates synthesis of lipid A. A putative lipoprotein was also identified and had 70%, 63%, and 44% similarity with hypothetical proteins of Chlamydophila felis
, C. caviae
, and C. pneumoniae
, respectively, but low similarity (26% and 22%) with proteins from Chlamydia muridarum
and C. trachomatis
). A putative inorganic pyrophosphatase, an enzyme that catalyzes the formation of orthophosphate from pyrophosphate, is highly conserved in both Chlamydia
species, with similarities of between 82% and 93% (45
). A putative elongation factor Ts (EF-Ts) is essential for cell viability in C. muridarum
). An ABC transporter was identified and was 66% similar to C. trachomatis
TroA, a protein of the chlamydial envelope which is reactive with sera from human patients suffering from infection (2
). ABC transporter proteins may be suitable targets for the development of antibacterial vaccines, postinfection therapies, or the development of novel antimicrobials which exploit ABC transport systems (10
Experimental infection of pregnant ewes with C. abortus
emulates clinically relevant disease processes. Alternative routes of infection, including intranasal and oral challenges, have recently been validated and may facilitate the identification of additional antigens expressed during infection (13
). This might explain why the infected fetus reacts with different antigens compared to the ewe, since it was infected in utero
. The identification of the macrophage infectivity potentiator protein as being highly reactive with sera from both infected ewes and fetuses relative to other proteins suggests that it plays a significant role during pathogenesis and is worthy of further investigation as a diagnostic antigen.