Comparison of protein patterns between invasive and non-invasive ovine strains of Chlamydia psittaci

Research in Veterinary Science 1989, 46, 40-42

Comparison of protein patterns between invasive and non-invasive ovine strains of Chlamydia psittaci D. BUZONI-GATEL, K. LAYACHl, G. DUBRAY, A. RODOLAKIS, INRA, Station de Pathologie de la Reproduction, 37380 Nouzilly, France Protein patterns displayed in sodium dodecyl sulphate (SDS) electrophoresis by invasive and noninvasive strains of Chlamydia psittaci showed three constant differences, the most distinctive being a band at 90 Kd from invasive strains. In comparison with other methods so far described, this method provides a more efficient means of differentiating between invasive and non-invasive strains. Furthermore, it may lead to the development of improved methods of diagnosis.

CHLAMYDIA psittaci accounts for three main conditions in sheep: abortion, polyarthritis and conjunctivitis. C psittaci isolates have been divided into two serotypes on the basis of a plaque reduction test (Schachter et a! 1974, 1975). Serotype I is associated with abortions and inapparent infections of the intestinal tract whereas serotype 2 is associated with polyarthritis and conjunctivitis. This classification does not a!low the differentiation, within the first group, between the pathogenic strains which cause abortion and non-pathogenic intestinal strains isolated from

the faeces of apparently healthy ewes. Such a differentiation is crucial for a better understanding of the epidemiology of chlamydia! abortion and could lead to improved preventive methods. An in vivo method to differentiate these strains has been proposed (Buzoni-Gatel and Rodolakis 1983, Rodolakis et al 1989). This method makes use of the fact that most strains isolated from abortions are invasive in mice following subcutaneous inoculation, whereas in the same model strains isolated from faeces of apparently healthy ewes are not invasive. The present report shows a difference in protein patterns, in polyacrylamide gel electrophoresis analysis, between invasive and non-invasive strains. Materials and methods

Organisms and growth conditions The origin of the strains and number of egg passages to which each had been subjected are listed in Table 1. All strains isolated at the Station de Patho-

TABLE 1: Origin of chlamydial strains

Strains

Geographical origin

Ovine abortion AB7 AB4 AB10 S26/3 H574 A22

Aveyron (F) Indre & Loire (F) Indre & Loire IF) Scotland Scotland Scotland

From' P. Faye

PR PR I. D. Aitken I. D. Aitken I. D. Aitken

Thermosensitive Ivaccinal) 1B Mutant of AB7 1H Mutant of AB7

PR PR

Intestinal (from faeces of healthy sheep or goats) iB1 Indre & Loire IF) PR iB2 Indre & Loire IF) PR iB3 Indre & Loire IFI iB5 Marne I FI PR MO 907 USA J. Storz

Number of egg passages Invasivenesst

2 2 2 3 5 40

+++ +++

2 2

+ +

+ + +

3 3 4 ?

+

PR Strains isolated at the Station de Pathologie de la Reproduction, INRA, Nouzilly • Faye et al (19721, McClenaghan et al (1984), Storz (1963) t Invasiveness for mice after footpad inoculation. - non-invasive, + invasive, + + + highly invasive strains (Rodolakis et al 1989) F France

40

Protein patterns of Chlamydia psittaci logie de la Reproduction had been cloned by plaque purification and their invasiveness for mice (Table I) tested by determination of splenic infection after subcutaneous inoculation into the footpads (BuzoniGatel and Rodolakis 1983, Rodolakis et alI989). All the strains used were propagated in McCoy cells as previously described (Rodolakis 1983) and stored in phosphate glutamine sucrose buffer at -70°C.

Chlamydial purification Chlamydiae were purified on Renografin as described by Caldwell et al (1981), using 150mM Trispotassium chloride buffer [pH 7· 5]. The purified elementary bodies were stored at - 70°C. The purity of elementary body preparations was determined by electron microscopy.

Lysis of chlamydiae Purified elementary bodies were suspended in an equal volume of phosphate buffered saline containing 4 per cent sodium dodecyl sulphate (50S) and 3 mM EOTA, incubated for I hour at 37°C and then centrifuged for 40 minutes at 11,000 g. The soluble supernatants were analysed by 50S-PAGE and proteins were estimated by the bicinchroninic acid (BCA) protein assay (Pierce; Rockford). The protein concentration was adjusted to 5 IJg 10IJI- 1 sample.

~.

~ I \J~

,

i

t

.....

, .. N

~-

2

,l

. ~ ., ..

3

4

5

6

7

8

9

10

<11194K

~

!

11 KIT

FIG 1: Comparing the polypeptide patterns of invasive and noninvasive strains of C psitteci. Invasive strains: Lane: 1 (lHI; 2 (526); 4 (AB4); 6 (1BI; 8 (M0907); 11 (AB7), Non-invasive strains: Lane: 3 (iBl); 5 OB2); 7 (iB51; 9 (AB10); 10 (iB3)., (Marker kit lane 12). The differences between the invasive and non-invasive strains are indicated by the arrows, The protein profiles of the invasive strains H574 and A22 (Table 1) are not shown, but do not differ from those of other invasive strains

41

SDS-PAGE

Electrophoresis was carried out in 8 to 20 per cent gradient acrylamide slab gel as described by Laemmli (1970). Before electrophoresis, the samples were mixed with an equal volume of solubilising solution containing 2 per cent 50S, 5 per cent fJ-mercaptoethanol, 2·5 per cent glycerol and 0·05 per cent bromophenol blue, and boiled for 10 minutes. The molecular mass of polypeptide bands was estimated by using low molecular mass markers (Pharmacia): phosphorylase 94 Kd, bovine serum albumin 68 Kd, ovalbumin 43 Kd, carbonic anhydrase 30 Kd, soy bean trypsin inhibitor 21 Kd and lactalbumin 14'4 Kd. Polyacrylamide gel electrophoresis in Tris glycine buffer (pH 8' 6) containing 0'1 per cent 50S was carried out at a constant current of 20 rnA for one hour (stacking gel) and then at 60 rnA for five hours (separating gel). The protein bands were revealed by silver staining (Oakley et al 1980). Results and discussion The comparison of polypeptide patterns revealed three constant differences between invasive and noninvasive strains, regardless of the origin of the strains. The major difference was the presence of a 90 Kd band in lysate from invasive strains, which was lacking in that from non-invasive strains. In addition, a single band located at about 78 Kd was observed from invasive strains, whereas at the same location lysate from non-invasive strains gave two bands. Also, lysate from invasive strains gave a faint band located at about 96 Kd, which did not occur with that from the non-invasive strains (Fig I). The results demonstrate a close correlation between the invasiveness of ovine C psittaci in the mouse model and the protein .profile in 50S polyacrylamide gel electrophoresis. The main distinctive feature was the presence in Iysates from all the invasive strains tested of the 90 Kd band which was lacking in all the non-invasive strains. This band could therefore be regarded as a virulence marker for invasive strains. Several other proteins also differed between the two types of strain but their significance is not clear because of the apparent differences in their concentration between individual strains. The isolates that the authors have studied originated from a variety of unrelated sources. It is therefore likely that the findings are representative of all strains within serotype I. With further confirmation, it is possible that the association of the 90 Kd protein with invasive strains of C psittaci may lead to the development of new diagnostic methods, using purified antibodies against this protein. Such methods would extend the discriminating ability offered by the conventional methods of typing

42

D. Buzoni-Gatel, K. Layachi, G. Dubray, A. Rodolakis

(Spears and Storz 1979a,b, Eb and Orfila 1982, Johnson 1983, McClenaghan et al 1984, PerezMartinez and Storz 1985) and improve the understanding of the epidemiology of the disease. In addition, the role of the proteins associated with invasiveness could be examined for their involvement in pathogenesis and for their immunogenic properties. Such investigations could open the field to the development of new vaccines against chlamydial abortion using purified proteins. Acknowledgements

We thank C. Limouzin, F. Bernard and G. Bezard for technical help and J. De Rycke for linguistic corrections. K.L. is supported by the Conseil Regional de la region Centre. References BUZONI-GATEL, D. & RODOLAKIS, A. (1983) Annales de Microbiologie 134A, 91-99 CALDWELL, H. D., KROMHOUT, J.&SCHACHTER, J. (1981) Infection and Immunity 31, 1161-1176 EB, F. & ORFILA, J. (1982) Infection and Immunity 37,1289-1291

FAYE, P., CHARTON, A., MAGE, c., BERNARD, C. & LELAYEC, C. (1972) Bulletin de l'Academie Veterinaire de France 45, 169-173 JOHNSON, F. W. A. (1983)Some Diseases of Emerging Importance to Community Trade. Eds J. R. Walton, E. G. White & S. A. Hall. EEC Seminar (1982) Scarborough, UK. p 100 LAEMMLI, U. K. (1970) Nature 227,680-685 McCLENAGHAN, M., HERRING, A. J. & AITKEN, I. D. (1984) Infection and Immunity 45, 384-389 OAKLEY, B. R., KIRCH, D. R. & MORRIS, N. R. (1980) Analytical Biochemistry lOS, 361-363 PEREZ-MARTINEZ, J. A. & STORZ, J. (1985) Infection and Immunity 50, 905-910 RODOLAKIS, A. (1983) Infection and Immunity 42,525-530 RODOLAKIS, A., BERNARD, F. & LANTIER, F. (1989) Research in Veterinary Science 46,34-39 SCHACHTER, J., BANKS, J., SUGG, N., SUNG, M., STORZ, J. & MEYER, K. F. (1974) Infection and Immunity 9,92-94 SCHACHTER, J., BANKS, J., SUGG, N., SUNG, M., STORZ, J. & MEYER, K. F. (1975) Infection and Immunity 11, 904-907 SPEARS, P. & STORZ, J. (l979a) Infection and Immunity 24, 224-232 SPEARS, P. & STORZ, J. (I 979b) The Journal of Infectious Diseases 140, 959-967 STORZ, J. (1963) Cornell Veterinarian 53,469-480

Received February 4, 1987 Accepted February 17, 1988