Veterinary Microbiology, 30 (1992) 165-177 Elsevier Science Publishers B.V., Amsterdam
165
Antigenic and morphological differentiation of placental and intestinal isolates of Chlamydia psittaci of ovine origin Peter C. Griffiths a'~, Helen L. Philips b, Michael Dawson a and Michael J. Clarkson b aMinistry of Agriculture, Fisheriesand Food, VirologyDepartment, Central VeterinaryLaboratory, New Haw, Weybridge,Surrey KT15 3NB, UK bDepartment of Veterinary Clinical Science, Leahurst Field Station, Universityof Liverpool, Chester High Road, Neston, Wirral, MerseysideL64 7TE, UK (Accepted 5 August 1991 )
ABSTRACT Griffiths, P.C., Philips, H.L., Dawson, M. and Clarkson, M.J., 1992. Antigenic and morphological differentiation of placental and intestinal isolates of Chlamydia psittaci of ovine origin. Vet. Microbiol., 30: 165-177. Ewe placental and lamb intestinal isolates of Chlamydia psittaci recovered from flocks affected with ovine enzootic abortion were examined by inclusion morphology, indirect immunofluorescence (IIF) and immunoblot analysis. Chlamydiae recovered from the faeces of sheep from two flocks free of clinical disease were also examined. In cell culture ovine abortion (OA) and intestinal isolates were distinguishable by inclusion development and morphology. Similarly, in two-way IIF tests with one week mouse antisera isolates fell into two distinct groups: abortion or intestinal. Immunoblotting with convalescent sheep abortion antiserum identified 30 out of at least 40 silver staining polypeptides as antigenic both in OA and intestinal isolates. The serum produced a similar reaction pattern to the resolved proteins of each OA isolate, indicating a higher degree of antigenic conservation among these isolates. Considerable cross reactivity between the OA and intestinal isolates was identified, but the serum also showed apparent molecular weight differences between antigens of the two types in the 87116 kDa, 38-44 kDa and 26-28 kDa regions. Furthermore, the immunoblotting analysis revealed heterogeneity among the intestinal isolates, particularly in antigens between 87-116 kDa and 38-44 kDa.
INTRODUCTION
The species Chlamydia psittaci consists o f a heterogeneous group o f obligate intracellular bacteria which is associated with a wide spectrum o f diseases and inapparent infections in avian and mammalian hosts, including man (Moulder et al., 1984). In Great Britain ovine enzootic abortion is a major ~Author for correspondence.
0378-1135/92/$05.00
© 1992 Elsevier Science Publishers B.V. All rights reserved.
166
P.O. GRIFF1THS ET AL.
cause of reproductive failure in sheep. Inapparent intestinal infections with chlamydiae are also common, both in enzootic abortion-affected and unaffected flocks. In the genus Chlamydia serotyping of the human and mouse pathogen C. trachomatis is well established and based on detection of subspecies variation of major outer membrane protein (MOMP) using a microimmunofluorescence (MIF) test (Wang and Grayston, 1982). In contrast, attempts to type mammalian isolates of C. psittaci have produced apparently contradictory results with regard to ovine placental and intestinal isolates. In comparisons that included the MO-907 faecal isolate (from an apparently healthy sheep) investigators using plaque reduction (Schachter et al., 1974), MIF (Schachter et al., 1975; Eb and Orfila, 1982 ), inclusion morphology and the effects of cytoactive agents (Spears and Storz, 1979 ) did not differentiate it from ovine abortion (OA) isolates. However, there are reports in which other ovine faecal isolates have been distinguished from OA isolates by serum neutralisation (Wilson and Dungworth, 1963 ), inclusion morphology (Johnson, 1984) and MIF (Perez-Martinez and Storz, 1985 ). Differentiation has also been reported in mouse infectivity assays in which OA and intestinal isolates were graded on their invasiveness (ability to spread after footpad inoculation) (Buzoni-Gatel and Rodolakis, 1983; Rodolakis et al., 1989 ). Buzoni-Gatel et al. ( 1989 ) have linked differences in invasiveness to differences in the polypeptide profiles of purified OA and intestinal isolates. However, the profiles shown in Buzoni-Gatel et al. (1989) differ from recent data for OA isolates (McClenaghan et al., 1991 ). At present the principle method for the routine serodiagnosis of OA in sheep flocks is a microplate modification of the complement fixation test (CFT; Stamp et al., 1952 ). However, since the test is broadly specific and can detect antibodies to inapparent chlamydial infections, interpretation of flock status can be problematic. Furthermore, the significance of intestinal infections, and their role, if any, in the epidemiology of enzootic abortion remains poorly understood. In this study chlamydiae were isolated from ewe placentae and lamb intestine and compared by inclusion morphology, immunofluorescence and immunoblotting. MATERIALS AND METHODS
Chlamydiae Ewe placental and lamb intestinal isolates recovered from five farms are detailed in Table 1. The prototype OA strain A22 was kindly supplied by Dr. I. D. Aitken (Moredun Research Institute, Edinburgh). OA isolate BS was of placental origin and recovered by one of us (M.D.) at the Central Veterinary Laboratory.
PLACENTAL AND INTESTINAL ISOLATESOF CHLAMYDIA PSITTACI
167
TABLE 1 Isolation of chlamydiae from farms Farm
Ewe placental isolate
Lamb intestinal isolate
1
+,F
+,TI3
2 3 4 5
+,T14 + ,T35" -* -*
+,T25 + +,T22 +,T23
+ culture positive; - c u l t u r e negative; *normal lambings.
Primary isolation methods Placental and intestinal isolates were recovered from small pieces of ewe placentae or from 0.1-0.25 g of faeces taken from the recta of lambs as described by Johnson et al. ( 1983 ). During the isolation procedure in McCoy cell monolayer cultures chlamydial infectivity was determined by methylene blue staining (Johnson et al., 1978).
Propagation of chlamydiae in eggs Six day old embryonated eggs were inoculated with 5 X 105 inclusion forming units of chlamydiae in 100/d of phosphate buffered saline (PBS), pH 7.2, with vancomycin and streptomycin at 1 mg m l - 1 each. Infected yolk sacs were homogenised in sucrose phosphate glutamate (SPG) m e d i u m (Bovarnick et al., 1950) supplemented with antibiotics (Johnson et al., 1983 ), centrifuged ( 1000 g at 4 ° C for 15 min ), and the midlayers stored at - 70 ° C.
Chlamydial antigens for indirect immunofluorescence A modified version of R i c h m o n d and Caul ( 1975 ) was followed (Johnson, 1986 ). Briefly, stock cultures of infected yolk sac material were used to inoculate McCoy cell monolayers, with centrifugation, to produce infection in nearly every cell. The next day cells were trypsinised, adjusted to 5 × 105 m l - ~, and 20/zl drops of the suspension added to wells of Teflon-coated slides (Flow Laboratories Ltd. ). After cell attachment, m e d i u m containing cycloheximide (at 1/tg ml -~ ) was added, slides were incubated (37°C, 5% CO2), and fixed in cold 80% acetone when chlamydial inclusions had reached maturity.
Production of antisera in mice Swiss Webster mice received single intraperitoneal inoculations of chlamydia-infected yolk sac (200/tl, diluted 50% in SPG). After one week antisera were collected and stored at - 20 ° C.
168
P.o.GRIFF1THSETAL.
Indirect immunofluorescence Murine antisera were diluted two-fold (16-256) in PBS, pH 7.2, and 10 #1 added to slide wells. After incubation (37°C for 15 min) and three brief washes in PBS, 10 #1 fluorescein isothiocyanate-conjugated goat anti-mouse polyvalent immunoglobulins (Nordic Immunological Laboratories Ltd. ) were added. After incubation (as before), washing ( 3 X 5 min) and mounting (in PBS: 10% glycerol) slides were examined with a Leitz Orthomat microscope at mag. X 400. The endpoint titre was recorded as the highest serum dilution that gave maximal fluorescence of chlamydial inclusions.
Preparation of purified elementary bodies (EBs) EBs were purified essentially as described by McClenaghan et al. (1984), except that homogenised infected (and uninfected) yolk sacs were treated with ethylenimine (Bahnemann, 1975 ) prior to purification in Urografin 370 (Schering AG, Germany ).
SDS-PAGE Samples containing about 10 #g of purified EBs or control YS proteins, estimated by the method of Bradford (1976), were resolved by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were solubilised by heating at 100°C for 10 min in a solution of 2% (w/v) SDS, 5% 2-mercaptoethanol, 3% (w/v) sucrose, 0.002% (w/v) bromophenol blue, 50 mM Tris (pH 6.8). Electrophoresis was carried out with 12.5% (w/ v) polyacrylamide gels ( 1.5 m m thick) in the discontinuous buffer system of Laemmli (1970) at 25 mA in the 5% stacking gel and 35 mA thereafter. Gels were either silver stained (Merril et al., 1981 ) or blotted. The apparent molecular weights (MWs) of silver stained and immunoblotted antigens were estimated from plots of MW versus -log Rf of MW marker proteins: phosphorylase b (97.4), bovine serum albumin (68), ovalbumin (43), carbonic anhydrase (29), fl-lactoglobulin (18.4), lysozyme ( 14.3 ) (Life Technologies Ltd. ), expressed in kiloDaltons (kDa).
Immunoblotting Following SDS-PAGE resolved proteins were electrophoretically transferred from gels to polyvinylidene difluoride (PVDF) membranes (Millipore, UK) (Towbin et al., 1979 ). Blotted chlamydial and control YS proteins were probed with a convalescent abortion serum (at 1:150) from a ewe experimentally challenged with OA isolate BS (kindly supplied by Dr. A.J. Wilsmore, Royal Veterinary College, Potters Bar). Antigens were also probed with a negative control serum (at 1:150) from a two week old colostrumdeprived lamb. Immunoreactivity was visualised by an alkaline phosphatase system (Pluzek and Ramlau, 1988 ).
PLACENTAL AND INTESTINAL ISOLATES OF CHLAMYDIA PSITTACI
169
RESULTS
Inclusion morphology In McCoy cell culture OA isolates (F, T14, T35, A22 and BS) typically developed compact, deeply methylene blue staining inclusions (isolate T14, Fig. la), which were just visible after 41 h and reached maturity around 72 h. Intestinal chlamydiae (T13, T25, T22 and T23) produced diffuse, less deeply staining inclusions (isolate T25, Fig. lb), which developed more rapidly in culture than OA isolates; they were visible after 24 h and matured within 48 h.
Indirect immunofluorescence OA isolates F, T 14, T35 and A22, and intestinal isolates T 13, T25, T22 and T23, were tested against homologous and heterologous antisera raised in mice (Table 2). Antisera to OA isolates gave high titres (32 to 128) against OA isolates. Antisera against the intestinal isolates reacted with most (but not all) of the intestinal isolates, producing lower titres (16 to 64). Some low titred ( 16 ) cross reactivity was observed in antisera to OA isolates F and T 14 with individual intestinal isolates (T22 and T23 ), and in antiserum to intestinal isolate T22 which reacted with OA isolate A22. Apart from these minor cross reactions, isolates fell into two groups: abortion and intestinal, in which the latter exhibited a greater degree of antigenic variability.
SDS-PAGE At least forty components of each chlamydial isolate were detected by silver staining (Fig. 2). Each isolate possessed a densely staining major band at about 40 kDa. Typically in OA isolates this band was seen to migrate faster than the corresponding densely stained band in intestinal isolates.
Antigenic analysis A colostrum-deprived lamb serum did not react with either purified chlamydial or control YS antigens (data not shown). The results of probing eight isolates and YS with a convalescent sheep abortion serum raised experimentally against OA isolate BS are presented in Fig. 3. At least 30 antigens possessing determinants common to both types were detected in each isolate by the serum. The reactivity patterns produced against OA isolates F, T 14, T35 and A22 (and BS, data not shown) were similar; major antigens recognised had apparent MWs of 94, 90, 87, 72, 61, 58.5, 55, 50, 49, 46, 38, 35, 33, 32, 30, 28, 19, and 16 kDa + 3 kDa. The patterns obtained with intestinal isolates T13, T25, T22 and T23 revealed variations between OA and intestinal isolates, and among intestinal isolates, in three main regions: 87-I 16, 38-44 and 26-28 kDa. Reactive an-
170
P.C. GRIFFITHSET AL.
Fig. 1. Methylene blue stained inclusions (arrowed) of Chlamydia psittaci in monolayer cultures of cycloheximide-treated McCoy cells: (a) ewe abortion isolate T14 at 72 h post infection; (b) lamb intestinal isolate T25 at 48 h post infection. (mag. X 400).
PLACENTALAND INTESTINAL ISOLATESOF CHLAMYDIA PSITTACI
171
TABLE 2 Summary of two-way immunofluorescence tests with mouse antisera against ovine abortion ( a - d ) and intestinal ( e - h ) isolates Isolate
a b c d e f g h
Antiserum
F TI4 T35 A22 T13 T25 T22 T23
F
T14
T35
A22
64 32 32 32
64 128 64 64
64 64 64 32
32 64 64 64
16 16
TI3
64 64 32 64
T25
T22
T23
16 16 64 32 32
32 32
32 32
Titres expressed as reciprocal values.
tigens which appeared to be c o m m o n to intestinal isolates had apparent MWs of 62, 57, 47, 26, (25) and 15.5_+3 kDa. In the 87-116 kDa region, OA isolates possessed a triplet of highly reactive antigens of 97, 90 and 87 kDa. Intestinal isolates exhibited a similar set of reactive antigens, but with higher MWs than in the OA isolates. Inter-isolate variation of these antigens was seen, in T13 (110, 104, 100 kDa), in T25 (116, 106, 102 kDa), in T22 (112, 105, 97 kDa) and in T23 (116, 99, 94 kDa). A dominant and broad reactive band at 38 kDa, which corresponded to the densely stained major band in Fig. 2, was detected in each OA isolate, and identified as M O M P (Salari and Ward, 1981 ). Dominant antigens corresponding to the major band of the intestinal isolates in Fig. 2 were detected by the serum in only two intestinal isolates, T 13 (41 kDa) and T22 (40 kDa), with associated bands at 42 and 39 kDa, respectively. In the two other intestinal isolates, T23 and T25, the serum did not react with the dense band shown in Fig. 2, but bands at 44, 42 and 37 kDa in T23, and 44 and 37 kDa in T25 were detected. Furthermore, a band at 57 kDa was particularly reactive in isolates T23 and T25 compared with T13 and T22 but was not detected in the OA isolates. The most i m m u n o d o m i n a n t antigen c o m m o n to all the intestinal isolates was at 26 kDa; it reacted particularly strongly in isolates T l 3, T25 and T23, but was not seen in the OA isolates, although a faintly reactive band at 25.5 kDa was detected in isolates F and T14. The serum reacted strongly with a 28 kDa antigen in the OA isolates, producing a weaker reaction to a similar band in the intestinal isolates (except possibly T 13 ). Antigens of 49 and 50 kDa recognised in all the OA isolates were not detected in the intestinal isolates.
172
P.C. GRIFFITHSETAL.
F
TI4
1"35
A22
BS
TI3
T25
T22
T23
YS
18 14.
Fig. 2. Polypeptide profiles of purified elementary bodies of Chlamydia psittaci ovine abortion isolates F, T 14, T35, A22 and BS, lamb intestinal isolates T 13, T25, T22 and T23, and purified yolk sac (YS). The positions of molecular weight markers (in kDa) are indicated on the left. The gel was silver stained by the method of Merril et al. ( 1981 ). DISCUSSION
In cell culture the inclusion development and morphology of the OA isolates were similar to the biotype 1 strains of Spears and Storz (1979), whereas
173
PLACENTAL AND INTESTINAL ISOLATES OF CHLAMYD1A PSITTAC1
. . . . 116 9 4 ~ _ 97-4-
-
87 . . . . 72- . . . . 6861 . . . . .
-94
-
....
62
55 . . . . . 50 . . . . . 49-
....
. . . .
43_
MOMP__
47
.... 41 -----40
29_ 28 . . . . . -
--
--26
....
25
....
15.5
18.416
.....
~
~
~
14.313-
Isolate
-F
TI4
T35
A22
TI3
T25
T22
T23
YS
Fig. 3. Immunoblotting of the resolved proteins of purified elementary bodies of Chlamydia psittaci ovine abortion isolates F, T14, T35 and A22, lamb intestinal isolates T13, T25, T22 and T23, and purified yolk sac (YS), with a convalescent sheep abortion antiserum raised experimentally against OA isolate BS. The positions of molecular weight markers (in kDa) are indicated on the left. Estimated MWs of chlamydial proteins are denoted by dotted lines. inclusions of the intestinal isolates exhibited different characteristics, confirming the findings of Johnson (1984) and Anderson and Baxter (1986). Apart from minor cross reactions, two-way IIF tests with one week mouse sera divided isolates into two distinct groups: abortion or intestinal (Table
174
P.C. GRIFFITHS ET AL.
2), which agrees with the indirect MIF results of Perez-Martinez and Storz ( 1985 ). The slight variations that we found in titres among isolates of both groups, may reflect antibody recognition of sub-type antigenic differences (Anderson, 1987). In comparison, preliminary IIF tests with sheep sera have shown only one-way distinction between isolates, in which intestinal isolate sera reacted only with intestinal isolates and post-abortion sera reacted with both groups of isolates (C. Venables and M. Dawson, personal communication, 1990). In silver stained gels of resolved EB proteins it was possible, on a limited basis, to distinguish between the intestinal and OA isolates, since MOMPs of the latter were seen to migrate faster than a similarly dense band in the intestinal isolates. In a previous study, Buzoni-Gatel et al. (1989) reported SDSPAGE profiles for OA and intestinal isolates in which M O M P was apparently absent, which contrasts with the profiles presented here. Variations in the virulence of OA isolates for sheep (Aitken et al., 1981 ) and mice (Johnson and Clarkson, 1986) have been reported, but no clear differences have been detected in culture (Anderson, 1986 ) or DNA restriction endonuclease analysis (McClenaghan et al., 1984). Inter-type variation among OA isolates was not detected by examination of silver stained gels, or by immunoblotting in the present study. A convalescent OA serum from an experimentally infected ewe produced similar reaction patterns to the SDS-stable and reduced antigens of four OA isolates (Fig. 3 ), indicating substantial antigenic conservation among this group compared with intestinal isolates. Considerable cross reactivity between OA and intestinal isolates was shown, although a number of c o m m o n epitopes appear to be located on protein antigens of different size, or possibly charge, in the two types. Serum reaction patterns showed differences between the two types in three main regions: 87-116 kDa, 38-44 kDa and 26-28 kDa, and revealed heterogeneity among the intestinal isolate group particularly in antigens between 87-116 kDa and 38-44 kDa. Interestingly, a highly immun o d o m i n a n t antigen c o m m o n to the intestinal isolates (26 kDa) was not detected in the OA isolates in which M O M P (38 kDa) was i m m u n o d o m i n a n t . In the 38-44 kDa region sheep antibodies identified dominant antigens, possibly MOMP, in only two intestinal isolates, T 13 and T22, indicating that the other two intestinal isolates, T25 and T23, may be more antigenically distinct. Immunoblot probing of the OA and intestinal isolates was also performed with one week (IgM) mouse antiserum to OA isolate BS, which unlike the post-abortion (IgG) sheep serum did not distinguish between the intestinal isolates and was only weakly reactive (P.C. Griffiths, unpublished observations, 1991 ). It is likely that the differentiating properties of the mouse serum in the IIF test were not apparent in immunoblotting due to the degradation of epitopes recognised in the IIF. In immunoblot data of OA isolate $26/3 EBs probed with homologous post-
PLACENTAL AND INTESTINAL ISOLATES OF CHLAMYDIA PSITTACI
175
abortion sheep serum, Huang et al. (1990) described a characteristic spectrum of 12-14 immunoreactive bands, including estimated antigens of 60, 56-58, 50, 45, 40, 32, 30 and 18 kDa, which are comparable with antigens detected in this study by post-abortion serum (to OA isolate BS) in four heterologous OA isolates (F, T14, T35 and A22; Fig. 3) and in isolate BS (data not shown). However, immunoblotting results on ovine intestinal isolates have not been previously published. It is possible that the antigenic differences detected between the OA and intestinal isolates, and the apparent heterogeneity among the latter group reported here, may be a reflection of chlamydial adaptation to different intracellular environments, in which the pressures necessary for the selection of variants among abortion organisms may be limited. Progress in differentiating ovine (and bovine) pathogenic and intestinal isolates by polymerase chain reaction (Saiki et al., 1985) amplification of part of the MOMP gene was recently reported from this laboratory (Hewinson et al., 1991 ). Future investigations with specific antibody and DNA probes will aim to reveal differences between abortion or intestinal organisms, which may facilitate the development of more discriminatory routine diagnostic tests, and lead to an improved understanding of the epidemiology of ovine chlamydial infections. ACKNOWLEDGEMENTS
We gratefully acknowledge the excellent assistance of Angus D. MacDougall, and more recently, Adrian Buckle, at the CVL. We would also like to thank Rex Cooper (CVL) and Gary Hynes (Liverpool) for the photography. The investigations at the University of Liverpool were supported by a project grant awarded by the Agricultural and Food Research Council. REFERENCES Aitken, I.D., Robinson, G.W. and Anderson, I.E., 1981. Enzootic abortion of ewes: experimental infection. Proc. Sheep Vet. Soc., 5: 53-60. Anderson, I.E., 1986. Comparison of five ovine isolates of Chlamydiapsittaci: an evaluation of three culture treatments. Med. Lab. Sci., 43:241-248. Anderson, I.E. and Baxter, T.A., 1986. Chlamydia psittaci: Inclusion morphology in cell culture and virulence in mice ofovine isolates. Vet. Rec., 119: 453-454. Anderson, I.E., 1987. Comparison of some ovine Chlamydia psittaci isolates by indirect immunofluorescence. Vet. Microbiol., 13: 69-74. Bahnemann, H.G., 1975. Binary ethylenimine as an inactivant for foot-and-mouth disease virus and its application for vaccine production. Arch. Virol., 47; 47-56. Bovarnick, M.R., Miller, J.C. and Snyder, J.C., 1950. The influence of certain salts, amino acids, sugars and proteins on the stability of rickettsiae. J. Bacteriol., 59: 509-522. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem., 72: 255-260.
176
P.C. GRIFFITHS ET AL.
Buzoni-Gatel, D. and Rodolakis, A., 1983. A mouse model to compare virulence of abortive and intestinal ovine strains of Chlamydia psittaci: influence of the route of inoculation. Ann. Microbiol., 134:91-99. Buzoni-Gatel, D., Layachi, K., Dubray, G. and Rodolakis, A., 1989. Comparison of protein patterns between invasive and non-invasive ovine strains of Chlarnydia psittaci. Res. Vet. Sci., 46: 40-42. Eb, F. and Orfila, J., 1982. Serotyping of Chlamydia psittaci by the microimmunofluorescence test: isolates ofovine origin. Infect. Immun., 37: 1289-1291. Hewinson, R.G., Griffiths, P.C., Rankin, S.E.S., Dawson, M. and Woodward, M.J., 1991. Towards a differential polymerase chain reaction test for Chlamydia psittaci. Vet. Rec., 128: 381-382. Huang, H.-S., Tan, T.W., Buxton, D., Anderson, I.E. and Herring, A.J., 1990. Antibody response of the ovine lymph node to experimental infection with an ovine abortion strain of Chlamydia psittaci. Vet. Microbiol., 21: 345-351. Johnson, F.W.A., Chancerelle, L.Y.J. and Hobson, D., 1978. An improved method of demonstrating the growth of chlamydiae in tissue culture. Med. Lab. Sci., 35: 67-74. Johnson, F.W.A., Clarkson, M.J. and Spencer, W.N., 1983. Direct isolation of the agent ofenzootic abortion of ewes (Chlamydia psittaci) in cell cultures. Vet. Rec., 113:413-414. Johnson, F.W.A., 1984. Abortion - - continuing flock problem: enteric infections in sheep associated with enzootic abortion (Chlamydia psittaci). Irish Vet. News (December): 10-15. Johnson, F.W.A., 1986. Studies on Chlamydia psittaci associated ovine foetopathy in the United Kingdom. PhD. thesis, University of Liverpool. Johnson, F.W.A and Clarkson, M.J., 1986. Ovine abortion isolates: antigenic variation detected by mouse infection. In: I.D. Aitken (Editor), Chlamydial diseases of ruminants, Commission of the European Community, Luxembourg, pp. 129-132. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London), 227: 680-685. McClenaghan, M., Herring, A.J. and Aitken, I.D., 1984. Comparison of Chlamydia psittaci isolates by DNA restriction endonuclease analysis. Infect. Immun., 45: 384-389. McClenaghan, M., Inglis, N.F. and Herring, A.J., 1991. Comparison of isolates of Chlamydia psittaci of ovine, avian and feline origin by analysis of polypeptide profiles from purified elementary bodies. Vet. Microbiol., 26: 269-278. Merril, C.R., Goldman, D., Sedman, S.A. and Ebert, M.H., 1981. Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science, 211: 1437-1438. Moulder, J.W., Hatch, T.P., Kuo, C.-C., Schachter, J. and Storz, J., 1984. In: N.R. Krieg (Editor), Bergey's Manual of Systematic Bacteriology, Vol. 1, Williams and Wilkins, Baltimore, Md., pp 729-739. Perez-Martinez, J.A. and Storz, J., 1985. Antigenic diversity of Chlamydia psittaci of mammalian origin determined by microimmunofluorescence. Infect. Immun., 50:905-910. Pluzec, K.-J. and Ramlau, J., 1988. Alkaline phosphatase labelled reagents. In: O.J. Bjerrum and N.H.H. Heegaard (Editors), Handbook of Immunoblotting of Proteins, Vol. 1, CRC Press, Inc., Boca Raton, Florida, pp. 177-188. Richmond, S.R. and Caul, E.O., 1975. Fluorescent antibody studies on chlamydial infections. J. Clin. Microbiol., 1: 345-352. Rodolakis, A., Bernard, F. and Lantier, F., 1989. Mouse models for evaluation of virulence of Chlarnydia psittaci isolated from ruminants. Res. Vet. Sci., 46: 34-39. Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Arnheim, N., 1985. Enzymatic amplification of B globin genomic sequences and restriction site analysis for diagnosis of sickle cell anaemia. Science, 230:1350-1354.
PLACENTALANDINTESTINALISOLATESOF CHLAMYDIAPSITTACI
177
Salari, S.H. and Ward, M.E., 1981. Polypeptide composition of Chlamydia trachomatis. J. Gen. Microbiol., 123: 197-207. Schachter, J., Banks, J., Sugg, N., Sung, M., Storz, J. and Meyer, K.F., 1974. Serotyping of Chlamydia I. Isolates of ovine origin. Infect. Immun., 9: 92-94. Schachter, J., Banks, J., Sugg, N., Sung, M., Storz, J. and Meyer, K.F., 1975. Serotyping of Chlamydia II: isolates of bovine origin. Infect. Immun., 1 l: 904-907. Spears, P. and Storz, J., 1979. Biotyping of C. psittaci based on inclusion morphology and response to diethylaminoethyl-dextran and cycloheximide. Infect. Immun., 24: 224-232. Stamp, J.T., Watt, J.A.A. and Cockburn, R.B., 1952. Enzootic abortion in ewes. Complement fixation test. J. Comp. Pathol, 62: 93-101. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Nat. Acad. Sci., 76: 4350-4354. Wang, S.-P. and Grayston, J.T., 1982. Microimmunifluorescence antibody responses in Chlamydia trachomatis infection, a review. In: P.A. Mardh, K.K. Holmes, J.D. Oriel, P. Piot and J. Schachter (Editors), Chlamydial Infections. Elsevier, Amsterdam, pp. 301-316. Wilson, M.R. and Dungworth, D.L., 1963. Psittacosis lymphogranuloma-venereum group viruses in sheep. Comparison between a faecal and an enzootic abortion strain. J. Comp. Pathol., 73" 277-284.
165
Antigenic and morphological differentiation of placental and intestinal isolates of Chlamydia psittaci of ovine origin Peter C. Griffiths a'~, Helen L. Philips b, Michael Dawson a and Michael J. Clarkson b aMinistry of Agriculture, Fisheriesand Food, VirologyDepartment, Central VeterinaryLaboratory, New Haw, Weybridge,Surrey KT15 3NB, UK bDepartment of Veterinary Clinical Science, Leahurst Field Station, Universityof Liverpool, Chester High Road, Neston, Wirral, MerseysideL64 7TE, UK (Accepted 5 August 1991 )
ABSTRACT Griffiths, P.C., Philips, H.L., Dawson, M. and Clarkson, M.J., 1992. Antigenic and morphological differentiation of placental and intestinal isolates of Chlamydia psittaci of ovine origin. Vet. Microbiol., 30: 165-177. Ewe placental and lamb intestinal isolates of Chlamydia psittaci recovered from flocks affected with ovine enzootic abortion were examined by inclusion morphology, indirect immunofluorescence (IIF) and immunoblot analysis. Chlamydiae recovered from the faeces of sheep from two flocks free of clinical disease were also examined. In cell culture ovine abortion (OA) and intestinal isolates were distinguishable by inclusion development and morphology. Similarly, in two-way IIF tests with one week mouse antisera isolates fell into two distinct groups: abortion or intestinal. Immunoblotting with convalescent sheep abortion antiserum identified 30 out of at least 40 silver staining polypeptides as antigenic both in OA and intestinal isolates. The serum produced a similar reaction pattern to the resolved proteins of each OA isolate, indicating a higher degree of antigenic conservation among these isolates. Considerable cross reactivity between the OA and intestinal isolates was identified, but the serum also showed apparent molecular weight differences between antigens of the two types in the 87116 kDa, 38-44 kDa and 26-28 kDa regions. Furthermore, the immunoblotting analysis revealed heterogeneity among the intestinal isolates, particularly in antigens between 87-116 kDa and 38-44 kDa.
INTRODUCTION
The species Chlamydia psittaci consists o f a heterogeneous group o f obligate intracellular bacteria which is associated with a wide spectrum o f diseases and inapparent infections in avian and mammalian hosts, including man (Moulder et al., 1984). In Great Britain ovine enzootic abortion is a major ~Author for correspondence.
0378-1135/92/$05.00
© 1992 Elsevier Science Publishers B.V. All rights reserved.
166
P.O. GRIFF1THS ET AL.
cause of reproductive failure in sheep. Inapparent intestinal infections with chlamydiae are also common, both in enzootic abortion-affected and unaffected flocks. In the genus Chlamydia serotyping of the human and mouse pathogen C. trachomatis is well established and based on detection of subspecies variation of major outer membrane protein (MOMP) using a microimmunofluorescence (MIF) test (Wang and Grayston, 1982). In contrast, attempts to type mammalian isolates of C. psittaci have produced apparently contradictory results with regard to ovine placental and intestinal isolates. In comparisons that included the MO-907 faecal isolate (from an apparently healthy sheep) investigators using plaque reduction (Schachter et al., 1974), MIF (Schachter et al., 1975; Eb and Orfila, 1982 ), inclusion morphology and the effects of cytoactive agents (Spears and Storz, 1979 ) did not differentiate it from ovine abortion (OA) isolates. However, there are reports in which other ovine faecal isolates have been distinguished from OA isolates by serum neutralisation (Wilson and Dungworth, 1963 ), inclusion morphology (Johnson, 1984) and MIF (Perez-Martinez and Storz, 1985 ). Differentiation has also been reported in mouse infectivity assays in which OA and intestinal isolates were graded on their invasiveness (ability to spread after footpad inoculation) (Buzoni-Gatel and Rodolakis, 1983; Rodolakis et al., 1989 ). Buzoni-Gatel et al. ( 1989 ) have linked differences in invasiveness to differences in the polypeptide profiles of purified OA and intestinal isolates. However, the profiles shown in Buzoni-Gatel et al. (1989) differ from recent data for OA isolates (McClenaghan et al., 1991 ). At present the principle method for the routine serodiagnosis of OA in sheep flocks is a microplate modification of the complement fixation test (CFT; Stamp et al., 1952 ). However, since the test is broadly specific and can detect antibodies to inapparent chlamydial infections, interpretation of flock status can be problematic. Furthermore, the significance of intestinal infections, and their role, if any, in the epidemiology of enzootic abortion remains poorly understood. In this study chlamydiae were isolated from ewe placentae and lamb intestine and compared by inclusion morphology, immunofluorescence and immunoblotting. MATERIALS AND METHODS
Chlamydiae Ewe placental and lamb intestinal isolates recovered from five farms are detailed in Table 1. The prototype OA strain A22 was kindly supplied by Dr. I. D. Aitken (Moredun Research Institute, Edinburgh). OA isolate BS was of placental origin and recovered by one of us (M.D.) at the Central Veterinary Laboratory.
PLACENTAL AND INTESTINAL ISOLATESOF CHLAMYDIA PSITTACI
167
TABLE 1 Isolation of chlamydiae from farms Farm
Ewe placental isolate
Lamb intestinal isolate
1
+,F
+,TI3
2 3 4 5
+,T14 + ,T35" -* -*
+,T25 + +,T22 +,T23
+ culture positive; - c u l t u r e negative; *normal lambings.
Primary isolation methods Placental and intestinal isolates were recovered from small pieces of ewe placentae or from 0.1-0.25 g of faeces taken from the recta of lambs as described by Johnson et al. ( 1983 ). During the isolation procedure in McCoy cell monolayer cultures chlamydial infectivity was determined by methylene blue staining (Johnson et al., 1978).
Propagation of chlamydiae in eggs Six day old embryonated eggs were inoculated with 5 X 105 inclusion forming units of chlamydiae in 100/d of phosphate buffered saline (PBS), pH 7.2, with vancomycin and streptomycin at 1 mg m l - 1 each. Infected yolk sacs were homogenised in sucrose phosphate glutamate (SPG) m e d i u m (Bovarnick et al., 1950) supplemented with antibiotics (Johnson et al., 1983 ), centrifuged ( 1000 g at 4 ° C for 15 min ), and the midlayers stored at - 70 ° C.
Chlamydial antigens for indirect immunofluorescence A modified version of R i c h m o n d and Caul ( 1975 ) was followed (Johnson, 1986 ). Briefly, stock cultures of infected yolk sac material were used to inoculate McCoy cell monolayers, with centrifugation, to produce infection in nearly every cell. The next day cells were trypsinised, adjusted to 5 × 105 m l - ~, and 20/zl drops of the suspension added to wells of Teflon-coated slides (Flow Laboratories Ltd. ). After cell attachment, m e d i u m containing cycloheximide (at 1/tg ml -~ ) was added, slides were incubated (37°C, 5% CO2), and fixed in cold 80% acetone when chlamydial inclusions had reached maturity.
Production of antisera in mice Swiss Webster mice received single intraperitoneal inoculations of chlamydia-infected yolk sac (200/tl, diluted 50% in SPG). After one week antisera were collected and stored at - 20 ° C.
168
P.o.GRIFF1THSETAL.
Indirect immunofluorescence Murine antisera were diluted two-fold (16-256) in PBS, pH 7.2, and 10 #1 added to slide wells. After incubation (37°C for 15 min) and three brief washes in PBS, 10 #1 fluorescein isothiocyanate-conjugated goat anti-mouse polyvalent immunoglobulins (Nordic Immunological Laboratories Ltd. ) were added. After incubation (as before), washing ( 3 X 5 min) and mounting (in PBS: 10% glycerol) slides were examined with a Leitz Orthomat microscope at mag. X 400. The endpoint titre was recorded as the highest serum dilution that gave maximal fluorescence of chlamydial inclusions.
Preparation of purified elementary bodies (EBs) EBs were purified essentially as described by McClenaghan et al. (1984), except that homogenised infected (and uninfected) yolk sacs were treated with ethylenimine (Bahnemann, 1975 ) prior to purification in Urografin 370 (Schering AG, Germany ).
SDS-PAGE Samples containing about 10 #g of purified EBs or control YS proteins, estimated by the method of Bradford (1976), were resolved by sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE). Proteins were solubilised by heating at 100°C for 10 min in a solution of 2% (w/v) SDS, 5% 2-mercaptoethanol, 3% (w/v) sucrose, 0.002% (w/v) bromophenol blue, 50 mM Tris (pH 6.8). Electrophoresis was carried out with 12.5% (w/ v) polyacrylamide gels ( 1.5 m m thick) in the discontinuous buffer system of Laemmli (1970) at 25 mA in the 5% stacking gel and 35 mA thereafter. Gels were either silver stained (Merril et al., 1981 ) or blotted. The apparent molecular weights (MWs) of silver stained and immunoblotted antigens were estimated from plots of MW versus -log Rf of MW marker proteins: phosphorylase b (97.4), bovine serum albumin (68), ovalbumin (43), carbonic anhydrase (29), fl-lactoglobulin (18.4), lysozyme ( 14.3 ) (Life Technologies Ltd. ), expressed in kiloDaltons (kDa).
Immunoblotting Following SDS-PAGE resolved proteins were electrophoretically transferred from gels to polyvinylidene difluoride (PVDF) membranes (Millipore, UK) (Towbin et al., 1979 ). Blotted chlamydial and control YS proteins were probed with a convalescent abortion serum (at 1:150) from a ewe experimentally challenged with OA isolate BS (kindly supplied by Dr. A.J. Wilsmore, Royal Veterinary College, Potters Bar). Antigens were also probed with a negative control serum (at 1:150) from a two week old colostrumdeprived lamb. Immunoreactivity was visualised by an alkaline phosphatase system (Pluzek and Ramlau, 1988 ).
PLACENTAL AND INTESTINAL ISOLATES OF CHLAMYDIA PSITTACI
169
RESULTS
Inclusion morphology In McCoy cell culture OA isolates (F, T14, T35, A22 and BS) typically developed compact, deeply methylene blue staining inclusions (isolate T14, Fig. la), which were just visible after 41 h and reached maturity around 72 h. Intestinal chlamydiae (T13, T25, T22 and T23) produced diffuse, less deeply staining inclusions (isolate T25, Fig. lb), which developed more rapidly in culture than OA isolates; they were visible after 24 h and matured within 48 h.
Indirect immunofluorescence OA isolates F, T 14, T35 and A22, and intestinal isolates T 13, T25, T22 and T23, were tested against homologous and heterologous antisera raised in mice (Table 2). Antisera to OA isolates gave high titres (32 to 128) against OA isolates. Antisera against the intestinal isolates reacted with most (but not all) of the intestinal isolates, producing lower titres (16 to 64). Some low titred ( 16 ) cross reactivity was observed in antisera to OA isolates F and T 14 with individual intestinal isolates (T22 and T23 ), and in antiserum to intestinal isolate T22 which reacted with OA isolate A22. Apart from these minor cross reactions, isolates fell into two groups: abortion and intestinal, in which the latter exhibited a greater degree of antigenic variability.
SDS-PAGE At least forty components of each chlamydial isolate were detected by silver staining (Fig. 2). Each isolate possessed a densely staining major band at about 40 kDa. Typically in OA isolates this band was seen to migrate faster than the corresponding densely stained band in intestinal isolates.
Antigenic analysis A colostrum-deprived lamb serum did not react with either purified chlamydial or control YS antigens (data not shown). The results of probing eight isolates and YS with a convalescent sheep abortion serum raised experimentally against OA isolate BS are presented in Fig. 3. At least 30 antigens possessing determinants common to both types were detected in each isolate by the serum. The reactivity patterns produced against OA isolates F, T 14, T35 and A22 (and BS, data not shown) were similar; major antigens recognised had apparent MWs of 94, 90, 87, 72, 61, 58.5, 55, 50, 49, 46, 38, 35, 33, 32, 30, 28, 19, and 16 kDa + 3 kDa. The patterns obtained with intestinal isolates T13, T25, T22 and T23 revealed variations between OA and intestinal isolates, and among intestinal isolates, in three main regions: 87-I 16, 38-44 and 26-28 kDa. Reactive an-
170
P.C. GRIFFITHSET AL.
Fig. 1. Methylene blue stained inclusions (arrowed) of Chlamydia psittaci in monolayer cultures of cycloheximide-treated McCoy cells: (a) ewe abortion isolate T14 at 72 h post infection; (b) lamb intestinal isolate T25 at 48 h post infection. (mag. X 400).
PLACENTALAND INTESTINAL ISOLATESOF CHLAMYDIA PSITTACI
171
TABLE 2 Summary of two-way immunofluorescence tests with mouse antisera against ovine abortion ( a - d ) and intestinal ( e - h ) isolates Isolate
a b c d e f g h
Antiserum
F TI4 T35 A22 T13 T25 T22 T23
F
T14
T35
A22
64 32 32 32
64 128 64 64
64 64 64 32
32 64 64 64
16 16
TI3
64 64 32 64
T25
T22
T23
16 16 64 32 32
32 32
32 32
Titres expressed as reciprocal values.
tigens which appeared to be c o m m o n to intestinal isolates had apparent MWs of 62, 57, 47, 26, (25) and 15.5_+3 kDa. In the 87-116 kDa region, OA isolates possessed a triplet of highly reactive antigens of 97, 90 and 87 kDa. Intestinal isolates exhibited a similar set of reactive antigens, but with higher MWs than in the OA isolates. Inter-isolate variation of these antigens was seen, in T13 (110, 104, 100 kDa), in T25 (116, 106, 102 kDa), in T22 (112, 105, 97 kDa) and in T23 (116, 99, 94 kDa). A dominant and broad reactive band at 38 kDa, which corresponded to the densely stained major band in Fig. 2, was detected in each OA isolate, and identified as M O M P (Salari and Ward, 1981 ). Dominant antigens corresponding to the major band of the intestinal isolates in Fig. 2 were detected by the serum in only two intestinal isolates, T 13 (41 kDa) and T22 (40 kDa), with associated bands at 42 and 39 kDa, respectively. In the two other intestinal isolates, T23 and T25, the serum did not react with the dense band shown in Fig. 2, but bands at 44, 42 and 37 kDa in T23, and 44 and 37 kDa in T25 were detected. Furthermore, a band at 57 kDa was particularly reactive in isolates T23 and T25 compared with T13 and T22 but was not detected in the OA isolates. The most i m m u n o d o m i n a n t antigen c o m m o n to all the intestinal isolates was at 26 kDa; it reacted particularly strongly in isolates T l 3, T25 and T23, but was not seen in the OA isolates, although a faintly reactive band at 25.5 kDa was detected in isolates F and T14. The serum reacted strongly with a 28 kDa antigen in the OA isolates, producing a weaker reaction to a similar band in the intestinal isolates (except possibly T 13 ). Antigens of 49 and 50 kDa recognised in all the OA isolates were not detected in the intestinal isolates.
172
P.C. GRIFFITHSETAL.
F
TI4
1"35
A22
BS
TI3
T25
T22
T23
YS
18 14.
Fig. 2. Polypeptide profiles of purified elementary bodies of Chlamydia psittaci ovine abortion isolates F, T 14, T35, A22 and BS, lamb intestinal isolates T 13, T25, T22 and T23, and purified yolk sac (YS). The positions of molecular weight markers (in kDa) are indicated on the left. The gel was silver stained by the method of Merril et al. ( 1981 ). DISCUSSION
In cell culture the inclusion development and morphology of the OA isolates were similar to the biotype 1 strains of Spears and Storz (1979), whereas
173
PLACENTAL AND INTESTINAL ISOLATES OF CHLAMYD1A PSITTAC1
. . . . 116 9 4 ~ _ 97-4-
-
87 . . . . 72- . . . . 6861 . . . . .
-94
-
....
62
55 . . . . . 50 . . . . . 49-
....
. . . .
43_
MOMP__
47
.... 41 -----40
29_ 28 . . . . . -
--
--26
....
25
....
15.5
18.416
.....
~
~
~
14.313-
Isolate
-F
TI4
T35
A22
TI3
T25
T22
T23
YS
Fig. 3. Immunoblotting of the resolved proteins of purified elementary bodies of Chlamydia psittaci ovine abortion isolates F, T14, T35 and A22, lamb intestinal isolates T13, T25, T22 and T23, and purified yolk sac (YS), with a convalescent sheep abortion antiserum raised experimentally against OA isolate BS. The positions of molecular weight markers (in kDa) are indicated on the left. Estimated MWs of chlamydial proteins are denoted by dotted lines. inclusions of the intestinal isolates exhibited different characteristics, confirming the findings of Johnson (1984) and Anderson and Baxter (1986). Apart from minor cross reactions, two-way IIF tests with one week mouse sera divided isolates into two distinct groups: abortion or intestinal (Table
174
P.C. GRIFFITHS ET AL.
2), which agrees with the indirect MIF results of Perez-Martinez and Storz ( 1985 ). The slight variations that we found in titres among isolates of both groups, may reflect antibody recognition of sub-type antigenic differences (Anderson, 1987). In comparison, preliminary IIF tests with sheep sera have shown only one-way distinction between isolates, in which intestinal isolate sera reacted only with intestinal isolates and post-abortion sera reacted with both groups of isolates (C. Venables and M. Dawson, personal communication, 1990). In silver stained gels of resolved EB proteins it was possible, on a limited basis, to distinguish between the intestinal and OA isolates, since MOMPs of the latter were seen to migrate faster than a similarly dense band in the intestinal isolates. In a previous study, Buzoni-Gatel et al. (1989) reported SDSPAGE profiles for OA and intestinal isolates in which M O M P was apparently absent, which contrasts with the profiles presented here. Variations in the virulence of OA isolates for sheep (Aitken et al., 1981 ) and mice (Johnson and Clarkson, 1986) have been reported, but no clear differences have been detected in culture (Anderson, 1986 ) or DNA restriction endonuclease analysis (McClenaghan et al., 1984). Inter-type variation among OA isolates was not detected by examination of silver stained gels, or by immunoblotting in the present study. A convalescent OA serum from an experimentally infected ewe produced similar reaction patterns to the SDS-stable and reduced antigens of four OA isolates (Fig. 3 ), indicating substantial antigenic conservation among this group compared with intestinal isolates. Considerable cross reactivity between OA and intestinal isolates was shown, although a number of c o m m o n epitopes appear to be located on protein antigens of different size, or possibly charge, in the two types. Serum reaction patterns showed differences between the two types in three main regions: 87-116 kDa, 38-44 kDa and 26-28 kDa, and revealed heterogeneity among the intestinal isolate group particularly in antigens between 87-116 kDa and 38-44 kDa. Interestingly, a highly immun o d o m i n a n t antigen c o m m o n to the intestinal isolates (26 kDa) was not detected in the OA isolates in which M O M P (38 kDa) was i m m u n o d o m i n a n t . In the 38-44 kDa region sheep antibodies identified dominant antigens, possibly MOMP, in only two intestinal isolates, T 13 and T22, indicating that the other two intestinal isolates, T25 and T23, may be more antigenically distinct. Immunoblot probing of the OA and intestinal isolates was also performed with one week (IgM) mouse antiserum to OA isolate BS, which unlike the post-abortion (IgG) sheep serum did not distinguish between the intestinal isolates and was only weakly reactive (P.C. Griffiths, unpublished observations, 1991 ). It is likely that the differentiating properties of the mouse serum in the IIF test were not apparent in immunoblotting due to the degradation of epitopes recognised in the IIF. In immunoblot data of OA isolate $26/3 EBs probed with homologous post-
PLACENTAL AND INTESTINAL ISOLATES OF CHLAMYDIA PSITTACI
175
abortion sheep serum, Huang et al. (1990) described a characteristic spectrum of 12-14 immunoreactive bands, including estimated antigens of 60, 56-58, 50, 45, 40, 32, 30 and 18 kDa, which are comparable with antigens detected in this study by post-abortion serum (to OA isolate BS) in four heterologous OA isolates (F, T14, T35 and A22; Fig. 3) and in isolate BS (data not shown). However, immunoblotting results on ovine intestinal isolates have not been previously published. It is possible that the antigenic differences detected between the OA and intestinal isolates, and the apparent heterogeneity among the latter group reported here, may be a reflection of chlamydial adaptation to different intracellular environments, in which the pressures necessary for the selection of variants among abortion organisms may be limited. Progress in differentiating ovine (and bovine) pathogenic and intestinal isolates by polymerase chain reaction (Saiki et al., 1985) amplification of part of the MOMP gene was recently reported from this laboratory (Hewinson et al., 1991 ). Future investigations with specific antibody and DNA probes will aim to reveal differences between abortion or intestinal organisms, which may facilitate the development of more discriminatory routine diagnostic tests, and lead to an improved understanding of the epidemiology of ovine chlamydial infections. ACKNOWLEDGEMENTS
We gratefully acknowledge the excellent assistance of Angus D. MacDougall, and more recently, Adrian Buckle, at the CVL. We would also like to thank Rex Cooper (CVL) and Gary Hynes (Liverpool) for the photography. The investigations at the University of Liverpool were supported by a project grant awarded by the Agricultural and Food Research Council. REFERENCES Aitken, I.D., Robinson, G.W. and Anderson, I.E., 1981. Enzootic abortion of ewes: experimental infection. Proc. Sheep Vet. Soc., 5: 53-60. Anderson, I.E., 1986. Comparison of five ovine isolates of Chlamydiapsittaci: an evaluation of three culture treatments. Med. Lab. Sci., 43:241-248. Anderson, I.E. and Baxter, T.A., 1986. Chlamydia psittaci: Inclusion morphology in cell culture and virulence in mice ofovine isolates. Vet. Rec., 119: 453-454. Anderson, I.E., 1987. Comparison of some ovine Chlamydia psittaci isolates by indirect immunofluorescence. Vet. Microbiol., 13: 69-74. Bahnemann, H.G., 1975. Binary ethylenimine as an inactivant for foot-and-mouth disease virus and its application for vaccine production. Arch. Virol., 47; 47-56. Bovarnick, M.R., Miller, J.C. and Snyder, J.C., 1950. The influence of certain salts, amino acids, sugars and proteins on the stability of rickettsiae. J. Bacteriol., 59: 509-522. Bradford, M.M., 1976. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem., 72: 255-260.
176
P.C. GRIFFITHS ET AL.
Buzoni-Gatel, D. and Rodolakis, A., 1983. A mouse model to compare virulence of abortive and intestinal ovine strains of Chlamydia psittaci: influence of the route of inoculation. Ann. Microbiol., 134:91-99. Buzoni-Gatel, D., Layachi, K., Dubray, G. and Rodolakis, A., 1989. Comparison of protein patterns between invasive and non-invasive ovine strains of Chlarnydia psittaci. Res. Vet. Sci., 46: 40-42. Eb, F. and Orfila, J., 1982. Serotyping of Chlamydia psittaci by the microimmunofluorescence test: isolates ofovine origin. Infect. Immun., 37: 1289-1291. Hewinson, R.G., Griffiths, P.C., Rankin, S.E.S., Dawson, M. and Woodward, M.J., 1991. Towards a differential polymerase chain reaction test for Chlamydia psittaci. Vet. Rec., 128: 381-382. Huang, H.-S., Tan, T.W., Buxton, D., Anderson, I.E. and Herring, A.J., 1990. Antibody response of the ovine lymph node to experimental infection with an ovine abortion strain of Chlamydia psittaci. Vet. Microbiol., 21: 345-351. Johnson, F.W.A., Chancerelle, L.Y.J. and Hobson, D., 1978. An improved method of demonstrating the growth of chlamydiae in tissue culture. Med. Lab. Sci., 35: 67-74. Johnson, F.W.A., Clarkson, M.J. and Spencer, W.N., 1983. Direct isolation of the agent ofenzootic abortion of ewes (Chlamydia psittaci) in cell cultures. Vet. Rec., 113:413-414. Johnson, F.W.A., 1984. Abortion - - continuing flock problem: enteric infections in sheep associated with enzootic abortion (Chlamydia psittaci). Irish Vet. News (December): 10-15. Johnson, F.W.A., 1986. Studies on Chlamydia psittaci associated ovine foetopathy in the United Kingdom. PhD. thesis, University of Liverpool. Johnson, F.W.A and Clarkson, M.J., 1986. Ovine abortion isolates: antigenic variation detected by mouse infection. In: I.D. Aitken (Editor), Chlamydial diseases of ruminants, Commission of the European Community, Luxembourg, pp. 129-132. Laemmli, U.K., 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London), 227: 680-685. McClenaghan, M., Herring, A.J. and Aitken, I.D., 1984. Comparison of Chlamydia psittaci isolates by DNA restriction endonuclease analysis. Infect. Immun., 45: 384-389. McClenaghan, M., Inglis, N.F. and Herring, A.J., 1991. Comparison of isolates of Chlamydia psittaci of ovine, avian and feline origin by analysis of polypeptide profiles from purified elementary bodies. Vet. Microbiol., 26: 269-278. Merril, C.R., Goldman, D., Sedman, S.A. and Ebert, M.H., 1981. Ultrasensitive stain for proteins in polyacrylamide gels shows regional variation in cerebrospinal fluid proteins. Science, 211: 1437-1438. Moulder, J.W., Hatch, T.P., Kuo, C.-C., Schachter, J. and Storz, J., 1984. In: N.R. Krieg (Editor), Bergey's Manual of Systematic Bacteriology, Vol. 1, Williams and Wilkins, Baltimore, Md., pp 729-739. Perez-Martinez, J.A. and Storz, J., 1985. Antigenic diversity of Chlamydia psittaci of mammalian origin determined by microimmunofluorescence. Infect. Immun., 50:905-910. Pluzec, K.-J. and Ramlau, J., 1988. Alkaline phosphatase labelled reagents. In: O.J. Bjerrum and N.H.H. Heegaard (Editors), Handbook of Immunoblotting of Proteins, Vol. 1, CRC Press, Inc., Boca Raton, Florida, pp. 177-188. Richmond, S.R. and Caul, E.O., 1975. Fluorescent antibody studies on chlamydial infections. J. Clin. Microbiol., 1: 345-352. Rodolakis, A., Bernard, F. and Lantier, F., 1989. Mouse models for evaluation of virulence of Chlarnydia psittaci isolated from ruminants. Res. Vet. Sci., 46: 34-39. Saiki, R.K., Scharf, S., Faloona, F., Mullis, K.B., Horn, G.T., Erlich, H.A. and Arnheim, N., 1985. Enzymatic amplification of B globin genomic sequences and restriction site analysis for diagnosis of sickle cell anaemia. Science, 230:1350-1354.
PLACENTALANDINTESTINALISOLATESOF CHLAMYDIAPSITTACI
177
Salari, S.H. and Ward, M.E., 1981. Polypeptide composition of Chlamydia trachomatis. J. Gen. Microbiol., 123: 197-207. Schachter, J., Banks, J., Sugg, N., Sung, M., Storz, J. and Meyer, K.F., 1974. Serotyping of Chlamydia I. Isolates of ovine origin. Infect. Immun., 9: 92-94. Schachter, J., Banks, J., Sugg, N., Sung, M., Storz, J. and Meyer, K.F., 1975. Serotyping of Chlamydia II: isolates of bovine origin. Infect. Immun., 1 l: 904-907. Spears, P. and Storz, J., 1979. Biotyping of C. psittaci based on inclusion morphology and response to diethylaminoethyl-dextran and cycloheximide. Infect. Immun., 24: 224-232. Stamp, J.T., Watt, J.A.A. and Cockburn, R.B., 1952. Enzootic abortion in ewes. Complement fixation test. J. Comp. Pathol, 62: 93-101. Towbin, H., Staehelin, T. and Gordon, J., 1979. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc. Nat. Acad. Sci., 76: 4350-4354. Wang, S.-P. and Grayston, J.T., 1982. Microimmunifluorescence antibody responses in Chlamydia trachomatis infection, a review. In: P.A. Mardh, K.K. Holmes, J.D. Oriel, P. Piot and J. Schachter (Editors), Chlamydial Infections. Elsevier, Amsterdam, pp. 301-316. Wilson, M.R. and Dungworth, D.L., 1963. Psittacosis lymphogranuloma-venereum group viruses in sheep. Comparison between a faecal and an enzootic abortion strain. J. Comp. Pathol., 73" 277-284.
