Veterinary Microbiology, 27 ( 1991 ) 3 9 - 4 7 Elsevier Science Publishers B.V., A m s t e r d a m
39
Comparison of Haemophilusparagallinarum isolates by restriction endonuclease analysis of chromosomal DNA P.J. Blackall a, L.E. Eaves a and C.J. Morrow b aAnirnal Research Institute, 665 Fairfield Road, Yeerongpilly 4105, Australia bVictorian Institute of Anirnal Sciences, Attwood 3049, Australia (Accepted 19 September 1990)
ABSTRACT Blackall, P.J., Eaves, L.E. and Morrow, C.J., 1991. Comparison of Haernophilus paragallinarum isolates by restriction endonuclease analysis of chromosomal DNA. Vet. Microbiol., 27: 39-47. Chromosomal DNA from Haernophilus paragallinarum was examined by restriction endonuclease analysis (REA) using the enzymes BamHI, EcoRI, HindlII, or Sinai. The enzyme SmaI had no apparent effect upon the DNA from eight representative H. paragallinarum isolates. The remaining enzymes cut the H. paragallinarurn DNA to varying degrees, with the most useful patterns for distinguishing isolates being given by HindllI. The REA patterns given by HindlII were stable under both in vitro and in vivo conditions. The use of the enzyme HindlII showed that eight Australian isolates of H. paragallinarum were genetically similar. In contrast, 14 isolates of H. paragallinarurn from outside Australia were markedly different from each other and the Australian isolates. A plasmid of approximately 6 kilobase pairs in size was found in one isolate ofH. paragallinarum.
INTRODUCTION
Haemophilus paragallinarum is the causative agent of infectious coryza, an upper respiratory tract disease of chickens (Yamamoto, 1984). Recently, we have shown that restriction endonuclease analysis (REA) of chromosomal DNA using the enzyme HindlII can be a useful tool in studying the epidemiology of infectious coryza outbreaks (Blackall et al., 1990b ). Our epidemiological study appears to have been the only reported application of REA to H. paragallinarum. In contrast REA has been used extensively with other avian bacterial pathogens, for example Mycoplasma gallisepticum (Kleven et al., 1988a,b). In this paper, we evaluate the use of different restriction endonuclease enzymes, the stability of REA patterns and the ability of REA to distinguish isolates of H. paragallinarum from diverse sources. 0378-1135/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
4(I
~, i I~l \< Ix \11 I [ \i
MATERIALS A N D METH~ )I)S
Lyo]utes T h e origin o f the isolates o f H . paragal[inarum u s e d a r e d e s c r i b e d i n T a b l e s o f these isolates h a v e b e e n p r e v i o u s l y p h e n o t y p i c a l l y c h a r a c t e r i z e d in this l a b o r a t o r y ( B l a c k a l l e t al., 1 9 8 9 ; B l a c k a l l e t al., 1 9 9 0 a ) a n d t h e k n o w n p r o p e r t i e s are s h o w n i n T a b l e s 1 a n d 2. 1 a n d 2. M a n y
Media Test medium agar (Rimler, 1979) supplemented with 5% (v/v) oleic albumin complex, 1% (v/v) chicken serum and 0.0025% (w/v) reduced nicotinamide adenine dinucleotide was used for the propagation of the isolates. This medium was termed TM/SN. A broth version of TM/SN, termed TMB, was used for the growth of cells for DNA extraction. DNA extraction Chromosomal DNA was extracted and purified by the method of Marshall et al. ( 1981 ) except that a ribonuclease treatment step was included. Plasmid DNA was extracted and purified by the technique described by Roberts et al. (1986). Plasrnid characterisation The identification of plasmid DNA bands, as either open circular forms or covalently closed circular forms, was performed using UV nicking and two dimensional agarose gel electrophoresis (Hintermann et al., 1981). This FABLEI
List of Australian isolates of tt. para;,allinaru,t used Strain
Kume serovaP
Biochemical biovar 2
HP14 HPI5
A-4 A-4 C-2 C-4 (?-2 C-2 A-4
1 1 l I I Ill
H P31 HP60 HP74 HP112 HPI79 HP183 HP186 HPI92
C-2 (7-2 (?-2
1 11 1 I1
Antimicrobial biovar:
I1
I IV
1
~Scrotyped by the Kume haemagglutinin scheme (Kume el al., 1983) as modified by Blackall ctal. ( 1990a ). aAs defined previously (Blackall et al.. 1989 ).
41
COMPARISON OF tfAEMOPHILUSPARAGALLINARUMISOLATES TABLE 2 List o f H . paragallinarum from outside Australia used in this study
Study code
Original
Country ~
Source 2
Kume serovar 3
Biochemical biovar 4
Antimicrobial biovar 4
USA USA USA FRG FRG Japan Brazil RSA Japan FRG Taiwan Brazil USA Brazil FRG
Rimler Rimler Rimler Rimler Hinz Rimler Kume Kume Kume Kume Rimler Rimler Rimler Rimler Hinz
A-1 B-1 C-2 A-2 A-2 A-I A-3 C-3 C-I B-1 9 9 9 ? 9
I IV V I I I 95 I I I 9 ? ? 9 v
II I I II II I 9 I I 1| 9 9 9 9 ?
code HP1 HP2 HP3 HP21 HP90 HPI37 HP144 H P 145 HP146 HP147 HP250 HP253 HP254 HP255 HP257
0083 0222
Modesto ATCC29545 ATCC29545 221 E3C SA-3 H-18 IPDH2671 TW-1 N2A1 0211 PI-E1 IPDHI676
~Aust = Australia; F R G = Federal Republic o f Germany; RSA-- Republic of South Africa; USA = United States o f America. 2 H i n z = D r . K-H. Hinz, Klinik Cfir Gefliigel, H a n n o v e r , F R G ; K u m e = D r . K. K u m e , T h e Kitasato Institute, Tokyo, Japan; R i m l e r = Dr. R.B. Rimler, National A n i m a l Disease Center, Ames, IA. ~Serotyped by the K u m e h a e m a g g l u t i n i n s c h e m e ( K u m e et al., 1983) as modified by Blackall et al. (1990a) 4As defined previously (Blackall et al., 1989). 5? = u n k n o w n .
technique utilises the fact that UV irradiation introduces single strand nicks in ethidium bromide stained DNA, converting some of the covalently closed circular forms to open circular forms. This means that in the second dimension electrophoresis, a band that originally contained only covalently closed circular forms yields two bands, the newer and slower migrating band corresponding to the open circular form.
REA The methods for REA, gel electrophoresis and photography were as previously described (Blackall et al., 1990b). When evaluating restriction enzymes, two control samples, one containing 0.5 #g 2 DNA and the other 0.5 /zg 2 D N A plus H. paragallinarum isolate HP 15 DNA, were used. All restriction digests were performed using the conditions and buffers recommended by the enzyme manufacturer (Boehringer, Mannheim).
"4"
lil
I]I -M k \ l I E l \ I
R ESI~JLTS
Evaluation q/enzymes Digestion of H. paragallinarum DNA with the enzyme Sinai did not yield any fragments that could be visualised in the electrophoresis system we used. However, digestion with enzymes BamHBI, EcoRI and HindIII did yield fragment patterns of varying complexity. Digestion with BarnHI resulted in only a small n u m b e r of mainly small molecular size fragments ( 10 to 20 resolvable bands) while digestion with EcoRI resulted in many more bands ( 30 to 40 ). However, use of EcoRI appeared to result in only partial digestion in some cases. Digestion with HindIII resulted in the most fragments (Fig. 1 ) and so the most suitable enzyme for comparing the patterns of different isolates. A careful examination of the patterns generated by the HindIII revealed that the four Australian isolates (HP15, 31, 192 and !83) gave very similar patterns while the non-Australian isolates ( HP3, 21,250 and 253 ) were clearly different from each other and the Australian isolates (Fig. 1 ). A similar con-
Fig. 1. Hindlll restriction endonuclease patterns of selectyed 11. paragallinarum isolates. Lane 1,2 DNA cut by HindIII, lane 2, 2 D N A plus HPI5" lane 3, HP15,1ane 4, HP31: lane 5, H P 1 9 2 lane 6, H P I 8 3 ; lane 7, HP3: lane 8, HP21" lane 9, HP250; lane 10, HP253. The arrow m lane q indicates a low molecular size b a n d subsequently shown to be a plasmid.
COMPARISON OF HAEMOPHILUS PARAGALLINARUM ISOLATES
43
clusion could be drawn from the BamHI digestions while the quality of the EcoRI digestions was insufficient to allow a confident comparison.
Plasmid characterisation A prominent low molecular size band was detected in all digestions of D N A from H. paragallinarum isolate H P 2 5 0 (arrow in lane 9 of Fig. 1 ). Following the use of an isolation technique specific for plasmid D N A (Roberts et al., 1986), two bands (one of approximately 6 kilobase pairs and the other approximately 12 kilobase pairs) could be detected. Following irradiation with U V and electrophoresis in a second dimension, the 6 kilobase pair band was found to yield two bands; one co-migrated with the 12 kilobase pair band. This suggests that the two bands represent different forms of the one plasmid with the 12 kilobase band representing the open circular form and the 6 kilobase band the covalently closed circular form. The plasmid, named p250, was not cut by Sma 1, BamHI or EcoRI. Following digestion of p250 with HindlII, two fragments were detected, one of approximately 4.5 kilobases and the other 1.5 kilobases.
Fig. 2. HindllI restriction endonuclease patterns of selected H. paragallinarum isolates illustrating the stability of the patterns under both laboratory and field conditions. Lane 1, molecular weight markers, size given in kilobases; lane 2, HP 14; lane 3, HP 15; lanes 4 and 5, HP31; lane 6, HP21; lane 7, HP90; lane 8, HP 192; lane 9, HP 183. See text for full discussion.
~'~
!'l
I~i \~ } ' , . \ i ) I I
\[
Stability o tRl~\4 paller,s. Fig. 2 illustrates the stability o f H i n d l l l REA patterns. Lanes 2 and 3 show ti. paragallinarum H P I 4 and H P 15, which were obtained from the sinus and trachea of the same chicken. Lanes 4 and 5 show II. paraga//inarum ttP31, with the D N A having been extracted on two separate occasions. Lanes 6 and 7 show H. paragallinarum HP21 and HP90, duplicate isolates of the same strain, ATCC29545, obtained from different laboratories. HP21 was obtained in 1979 from Dr. R. Rimler (National Animal Disease Center, Ames) who had originally obtained the strain from Dr, K.-H. Hinz (Klinik fiir Geflfigel, Hannover, F R G ) . Isolate H P 9 0 was obtained in 1985 direct from Dr. K.-H. Hinz. Isolates HP21 and H P 9 0 have been maintained separately at this Institute since their arrival. Lanes 8 and 9 show ti. paraA,a/linarum HP192 and HP183, which are isolates from chickens on different farms suffering related outbreaks of infectious coryza (Blackall et al., 1990b). In all cases, the pairs of isolates shown in Fig. 2 have the same HindIII pattern.
Fig. 3. ttindlll restriction endonculease patterns of Australian isolates of H. paragu[/inarum. Lane 1 molecular weight markers, size given in kilobases; lane 2, HPI4: lane 3. HP31: lane 4, HP60:lane 5, HP74: lane 6, HP112:lane 7. Hl'179:lanc 8, HP186: lane 9, HPI92.
COMPARISONOF HAEMOPHILUS PARAGALLINARL~I ISOLATES
45
Fig. 4. H i n d l l I restriction endonuclease patterns of isolates of H. paragallinarurn from outside Australia. Lane 1 molecular weight markers, size given in kilobases; lane 2, HP 1; lane 3, HP2; lane 4, HP3; lane 5, HP21; lane 6, HP137, lane 7, HP144; lane 8, HP145; lane 9, HP146; lane 10, HP147; lane 11, HP250; lane 12, HP253; lane 13, HP254; lane 14, HP255; lane 15, HP257.
Variations of REA patterns Figures 3 and 4 show the HindlII REA patterns of selected Australian and non-Australian H. paragallinarum isolates, respectively. The Australian H. paragallinarum isolates were very similar, although minor variations were observed among the isolates. In contrast, the non-Australian isolates were markedly different, both from each other and from the Australian isolates. DISCUSSION
In this study we have established that REA is a suitable technique for comparing the genetic relatedness ofH. paragallinarum isolates. In the initial work, we examined the suitability of four restriction enzymes (BamHI, EcoR1, HindlII and Sinai) using eight H. paragallinarum isolates from diverse sources. SmaI did not generate fragments detectable in the electrophoresis system used, while BamHI, EcoRI and HindlII all generated fragments patterns of varying complexity. All of the enzymes digested control 2 DNA to completion even when it was mixed with H. paragallinarum chromosomal DNA. This retention of enzyme activity in the presence of the H. paragalli-
Harum DNA indicates that there were no inhibitory factors present in the preparations. Within the genus Haemophilus, the restriction enzymes previously found to generate satisfactory banding patterns have been BamHI and Hindlll for Actinobacillus (Haernophilus) pleuropneumoniae ( Maclnnes et al., 1990 ), and BamHI for tt. parasuis and 'H. somnus' (McGillivery et al., 1986: Smart et ai., 1988). While we found that BamHI, EcoRI and HindIII are suitable for use with H. paragallinarum, HindIII generated the most DNA fragments, allowing the easiest recognition of genetic differences amongst the isolates. Plasmids have not previously been reported in H. paragallinarum. In a previous study using specific plasmid isolation techniques, we failed to detect any plasmids in 75 isolates o f H . paragallinarurn, despite the fact that 20 of those isolates showed resistance to streptomycin a n d / o r tetracycline ( Blackall, 1988 ). Further studies to determine if any function can be associated with plasmid p250 are required. Previously, we showed that ltindIII REA patterns are stable under field conditions (Blackall et al., 1990b ). The results of the present study have confirmed this finding. As well, we have shown that REA patterns are stable, despite extensive subculturing, under laboratory conditions. Australian isolates of tt. paragallinarum showed only minor differences in their HindllI REA patterns despite the fact that the isolates possessed a considerable range in phenotypic properties (Table 1 ), representing all the Kume serovars and biochemical and antimicrobial biovars known to occur in Australian isolates of H. paragallinarum (Blackall et al.. 1989). In contrast, selected non-Australian isolates, which represented a similar range of phenotypic variation (Table 2 ), showed considerable differences in their tlindlll REA patterns. Our finding that the HindIII REA patterns of Australian H. paragallinarum isolates are very similar indicates that there is little genetic variation in these isolates. It is possible that all Australian H. paragallinarum isolates have evolved from only a few original strains that were introduced into this country. The physical isolation of Australia and, in recent times, strict quarantine regulations have meant reduced opportunities for the importation of live chickens into Australia and hence limited the genetic pool of II. paragallinarum. Our results suggest that in countries like Australia, with a history of biological isolation, bacterial poultry pathogens may show very limited genetic variation. As methods to detect chickens harbouring H. paragallinarum in the absence of clinical disease have not been established, it is possible that H. paragallinarum and other poultry pathogens, not regarded as 'exotic agents', may enter Australia if quarantine regulations are altered and live bird importations allowed. Such importations could result in a widened genetic pool and the establishment of strains from new serovars or possessing greater virulence.
COMPARISON OF flAEMOPHILUS PARAGALLINARUM ISOLATES
47
ACKNOWLEDGMENTS
This work was partly supported by grants from the Australian Chicken Meat and Egg Industry Research Councils. The technical assistance of S. Brice and D.G. Rogers is gratefully acknowledged.
REFERENCES Blackall, P.J., 1988. Antimicrobial drug resistance and the occurrence of plasmids in Haemophilus paragallinarurn. Avian Dis., 32: 742-747. Blackall, P.J., Eaves, L.E. and Rogers, D.G., 1989. Biotyping of Haemophilus paragallinarum using hemagglutinin serotyping, carbohydrate fermentation patterns, and antimicrobial drug resistance patterns. Avian Dis., 33:491-496. Blackall, P.J., Eaves, L.E. and Rogers, D.G., 1990a. Proposal of a new serovar and altered nomenclature for Haemophilus paragallinarum in the Kume hemagglutinin scheme. J. Clin. Microbiol., 28:1185-1187. Blackall, P.J., Morrow, C.J., Maclnnes, A., Eaves, L.E. and Rogers, D.G., 1990b. Epidemiological studies on infectious coryza outbreaks in northern New South Wales, Australia using serotyping, biotyping and restriction endonuclease analysis of chromosomal DNA. Avian Dis., 34: 267-276. Hintermann, G., Fischer, H.-M., Crameri, R. and Hurter, R., 1981. Simple procedure for distinguishing CCC, OC and L forms of plasmid DNA by agarose gel electrophoresis. Plasmid, 5: 371-373. Kleven, S.H., Browning, G.F., Bulach, D.M., Ghiocas, E., Morrow, C.J. and Whithear, K.G., 1988a. Examination ofMycoplasma gallisepticum strains using restriction endonuclease DNA analysis and DNA-DNA hybridisation. Avian Pathol., 17: 559-570. Kleven, S.H., Morrow, C.J. and Whithear, K.G., 1988b. Comparison of Mycoplasma gallisepticum strains by hemagglutination-inhibition and restriction endonuclease analysis. Avian Dis., 32: 731-741. Kume, K., Sawata, A., Nakai, T. and Matsumoto, M., 1983. Serologxcal classification of Haemophilus paragallinarum with a hemagglutinin system. J. Clin. Microbiol., 17: 958-964. Maclnnes, J.I., Borr, J.D., Massoudi, M. and Rosendal, S., 1990. Analysis of Southern Ontario Actinobacillus (Haemophilus) pleuropneumoniae isolates by restriction endonuclease fingerprinting. Can. J. Vet. Res., 54: 244-250. Marshall, R.B., Wilton, B.E. and Robinson, A.J., 1981. Identification of leptospira serovars by restriction-endonuclease analysis. J. Med. Microbiol., 14:163-166. McGillivery, D.J., Webber, J.J. and Dean, H.F., 1986. Characterisation ofHistophilus ovis and related organisms by restriction endonuclease analysis. Aust. Vet. J., 63: 389-393. Rimler, R.B., 1979. Studies of the pathogenic avian haemophili. Avian Dis., 23:1006-1018. Roberts, I., Holmes, W.M. and Hyleman, P.B., 1986. Modified plasmid isolation method for CTostridium perfringens and Clostridium absonum. Appl. Environ. Microbiol., 52: 197-199. Smart, N.L., Miniats, O.P. and Maclnnes, J.I., 1988. Analysis of Haemophilus parasuis isolates from southern Ontario swine by restriction endonuclease fingerprinting. Can. J. Vet. Res., 52:319-324. Yamamoto, R., 1984. Infectious coryza. In: M.S. Hofstad, H.J. Barnes, B.W. Calnek, W.M. Reid and H.W. Yoder, Jr. (Editors), Diseases of Poultry. 8th Edn. Iowa State University Press, Ames, IA, pp. 178-186.
39
Comparison of Haemophilusparagallinarum isolates by restriction endonuclease analysis of chromosomal DNA P.J. Blackall a, L.E. Eaves a and C.J. Morrow b aAnirnal Research Institute, 665 Fairfield Road, Yeerongpilly 4105, Australia bVictorian Institute of Anirnal Sciences, Attwood 3049, Australia (Accepted 19 September 1990)
ABSTRACT Blackall, P.J., Eaves, L.E. and Morrow, C.J., 1991. Comparison of Haernophilus paragallinarum isolates by restriction endonuclease analysis of chromosomal DNA. Vet. Microbiol., 27: 39-47. Chromosomal DNA from Haernophilus paragallinarum was examined by restriction endonuclease analysis (REA) using the enzymes BamHI, EcoRI, HindlII, or Sinai. The enzyme SmaI had no apparent effect upon the DNA from eight representative H. paragallinarum isolates. The remaining enzymes cut the H. paragallinarurn DNA to varying degrees, with the most useful patterns for distinguishing isolates being given by HindllI. The REA patterns given by HindlII were stable under both in vitro and in vivo conditions. The use of the enzyme HindlII showed that eight Australian isolates of H. paragallinarum were genetically similar. In contrast, 14 isolates of H. paragallinarurn from outside Australia were markedly different from each other and the Australian isolates. A plasmid of approximately 6 kilobase pairs in size was found in one isolate ofH. paragallinarum.
INTRODUCTION
Haemophilus paragallinarum is the causative agent of infectious coryza, an upper respiratory tract disease of chickens (Yamamoto, 1984). Recently, we have shown that restriction endonuclease analysis (REA) of chromosomal DNA using the enzyme HindlII can be a useful tool in studying the epidemiology of infectious coryza outbreaks (Blackall et al., 1990b ). Our epidemiological study appears to have been the only reported application of REA to H. paragallinarum. In contrast REA has been used extensively with other avian bacterial pathogens, for example Mycoplasma gallisepticum (Kleven et al., 1988a,b). In this paper, we evaluate the use of different restriction endonuclease enzymes, the stability of REA patterns and the ability of REA to distinguish isolates of H. paragallinarum from diverse sources. 0378-1135/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
4(I
~, i I~l \< Ix \11 I [ \i
MATERIALS A N D METH~ )I)S
Lyo]utes T h e origin o f the isolates o f H . paragal[inarum u s e d a r e d e s c r i b e d i n T a b l e s o f these isolates h a v e b e e n p r e v i o u s l y p h e n o t y p i c a l l y c h a r a c t e r i z e d in this l a b o r a t o r y ( B l a c k a l l e t al., 1 9 8 9 ; B l a c k a l l e t al., 1 9 9 0 a ) a n d t h e k n o w n p r o p e r t i e s are s h o w n i n T a b l e s 1 a n d 2. 1 a n d 2. M a n y
Media Test medium agar (Rimler, 1979) supplemented with 5% (v/v) oleic albumin complex, 1% (v/v) chicken serum and 0.0025% (w/v) reduced nicotinamide adenine dinucleotide was used for the propagation of the isolates. This medium was termed TM/SN. A broth version of TM/SN, termed TMB, was used for the growth of cells for DNA extraction. DNA extraction Chromosomal DNA was extracted and purified by the method of Marshall et al. ( 1981 ) except that a ribonuclease treatment step was included. Plasmid DNA was extracted and purified by the technique described by Roberts et al. (1986). Plasrnid characterisation The identification of plasmid DNA bands, as either open circular forms or covalently closed circular forms, was performed using UV nicking and two dimensional agarose gel electrophoresis (Hintermann et al., 1981). This FABLEI
List of Australian isolates of tt. para;,allinaru,t used Strain
Kume serovaP
Biochemical biovar 2
HP14 HPI5
A-4 A-4 C-2 C-4 (?-2 C-2 A-4
1 1 l I I Ill
H P31 HP60 HP74 HP112 HPI79 HP183 HP186 HPI92
C-2 (7-2 (?-2
1 11 1 I1
Antimicrobial biovar:
I1
I IV
1
~Scrotyped by the Kume haemagglutinin scheme (Kume el al., 1983) as modified by Blackall ctal. ( 1990a ). aAs defined previously (Blackall et al.. 1989 ).
41
COMPARISON OF tfAEMOPHILUSPARAGALLINARUMISOLATES TABLE 2 List o f H . paragallinarum from outside Australia used in this study
Study code
Original
Country ~
Source 2
Kume serovar 3
Biochemical biovar 4
Antimicrobial biovar 4
USA USA USA FRG FRG Japan Brazil RSA Japan FRG Taiwan Brazil USA Brazil FRG
Rimler Rimler Rimler Rimler Hinz Rimler Kume Kume Kume Kume Rimler Rimler Rimler Rimler Hinz
A-1 B-1 C-2 A-2 A-2 A-I A-3 C-3 C-I B-1 9 9 9 ? 9
I IV V I I I 95 I I I 9 ? ? 9 v
II I I II II I 9 I I 1| 9 9 9 9 ?
code HP1 HP2 HP3 HP21 HP90 HPI37 HP144 H P 145 HP146 HP147 HP250 HP253 HP254 HP255 HP257
0083 0222
Modesto ATCC29545 ATCC29545 221 E3C SA-3 H-18 IPDH2671 TW-1 N2A1 0211 PI-E1 IPDHI676
~Aust = Australia; F R G = Federal Republic o f Germany; RSA-- Republic of South Africa; USA = United States o f America. 2 H i n z = D r . K-H. Hinz, Klinik Cfir Gefliigel, H a n n o v e r , F R G ; K u m e = D r . K. K u m e , T h e Kitasato Institute, Tokyo, Japan; R i m l e r = Dr. R.B. Rimler, National A n i m a l Disease Center, Ames, IA. ~Serotyped by the K u m e h a e m a g g l u t i n i n s c h e m e ( K u m e et al., 1983) as modified by Blackall et al. (1990a) 4As defined previously (Blackall et al., 1989). 5? = u n k n o w n .
technique utilises the fact that UV irradiation introduces single strand nicks in ethidium bromide stained DNA, converting some of the covalently closed circular forms to open circular forms. This means that in the second dimension electrophoresis, a band that originally contained only covalently closed circular forms yields two bands, the newer and slower migrating band corresponding to the open circular form.
REA The methods for REA, gel electrophoresis and photography were as previously described (Blackall et al., 1990b). When evaluating restriction enzymes, two control samples, one containing 0.5 #g 2 DNA and the other 0.5 /zg 2 D N A plus H. paragallinarum isolate HP 15 DNA, were used. All restriction digests were performed using the conditions and buffers recommended by the enzyme manufacturer (Boehringer, Mannheim).
"4"
lil
I]I -M k \ l I E l \ I
R ESI~JLTS
Evaluation q/enzymes Digestion of H. paragallinarum DNA with the enzyme Sinai did not yield any fragments that could be visualised in the electrophoresis system we used. However, digestion with enzymes BamHBI, EcoRI and HindIII did yield fragment patterns of varying complexity. Digestion with BarnHI resulted in only a small n u m b e r of mainly small molecular size fragments ( 10 to 20 resolvable bands) while digestion with EcoRI resulted in many more bands ( 30 to 40 ). However, use of EcoRI appeared to result in only partial digestion in some cases. Digestion with HindIII resulted in the most fragments (Fig. 1 ) and so the most suitable enzyme for comparing the patterns of different isolates. A careful examination of the patterns generated by the HindIII revealed that the four Australian isolates (HP15, 31, 192 and !83) gave very similar patterns while the non-Australian isolates ( HP3, 21,250 and 253 ) were clearly different from each other and the Australian isolates (Fig. 1 ). A similar con-
Fig. 1. Hindlll restriction endonuclease patterns of selectyed 11. paragallinarum isolates. Lane 1,2 DNA cut by HindIII, lane 2, 2 D N A plus HPI5" lane 3, HP15,1ane 4, HP31: lane 5, H P 1 9 2 lane 6, H P I 8 3 ; lane 7, HP3: lane 8, HP21" lane 9, HP250; lane 10, HP253. The arrow m lane q indicates a low molecular size b a n d subsequently shown to be a plasmid.
COMPARISON OF HAEMOPHILUS PARAGALLINARUM ISOLATES
43
clusion could be drawn from the BamHI digestions while the quality of the EcoRI digestions was insufficient to allow a confident comparison.
Plasmid characterisation A prominent low molecular size band was detected in all digestions of D N A from H. paragallinarum isolate H P 2 5 0 (arrow in lane 9 of Fig. 1 ). Following the use of an isolation technique specific for plasmid D N A (Roberts et al., 1986), two bands (one of approximately 6 kilobase pairs and the other approximately 12 kilobase pairs) could be detected. Following irradiation with U V and electrophoresis in a second dimension, the 6 kilobase pair band was found to yield two bands; one co-migrated with the 12 kilobase pair band. This suggests that the two bands represent different forms of the one plasmid with the 12 kilobase band representing the open circular form and the 6 kilobase band the covalently closed circular form. The plasmid, named p250, was not cut by Sma 1, BamHI or EcoRI. Following digestion of p250 with HindlII, two fragments were detected, one of approximately 4.5 kilobases and the other 1.5 kilobases.
Fig. 2. HindllI restriction endonuclease patterns of selected H. paragallinarum isolates illustrating the stability of the patterns under both laboratory and field conditions. Lane 1, molecular weight markers, size given in kilobases; lane 2, HP 14; lane 3, HP 15; lanes 4 and 5, HP31; lane 6, HP21; lane 7, HP90; lane 8, HP 192; lane 9, HP 183. See text for full discussion.
~'~
!'l
I~i \~ } ' , . \ i ) I I
\[
Stability o tRl~\4 paller,s. Fig. 2 illustrates the stability o f H i n d l l l REA patterns. Lanes 2 and 3 show ti. paragallinarum H P I 4 and H P 15, which were obtained from the sinus and trachea of the same chicken. Lanes 4 and 5 show II. paraga//inarum ttP31, with the D N A having been extracted on two separate occasions. Lanes 6 and 7 show H. paragallinarum HP21 and HP90, duplicate isolates of the same strain, ATCC29545, obtained from different laboratories. HP21 was obtained in 1979 from Dr. R. Rimler (National Animal Disease Center, Ames) who had originally obtained the strain from Dr, K.-H. Hinz (Klinik fiir Geflfigel, Hannover, F R G ) . Isolate H P 9 0 was obtained in 1985 direct from Dr. K.-H. Hinz. Isolates HP21 and H P 9 0 have been maintained separately at this Institute since their arrival. Lanes 8 and 9 show ti. paraA,a/linarum HP192 and HP183, which are isolates from chickens on different farms suffering related outbreaks of infectious coryza (Blackall et al., 1990b). In all cases, the pairs of isolates shown in Fig. 2 have the same HindIII pattern.
Fig. 3. ttindlll restriction endonculease patterns of Australian isolates of H. paragu[/inarum. Lane 1 molecular weight markers, size given in kilobases; lane 2, HPI4: lane 3. HP31: lane 4, HP60:lane 5, HP74: lane 6, HP112:lane 7. Hl'179:lanc 8, HP186: lane 9, HPI92.
COMPARISONOF HAEMOPHILUS PARAGALLINARL~I ISOLATES
45
Fig. 4. H i n d l l I restriction endonuclease patterns of isolates of H. paragallinarurn from outside Australia. Lane 1 molecular weight markers, size given in kilobases; lane 2, HP 1; lane 3, HP2; lane 4, HP3; lane 5, HP21; lane 6, HP137, lane 7, HP144; lane 8, HP145; lane 9, HP146; lane 10, HP147; lane 11, HP250; lane 12, HP253; lane 13, HP254; lane 14, HP255; lane 15, HP257.
Variations of REA patterns Figures 3 and 4 show the HindlII REA patterns of selected Australian and non-Australian H. paragallinarum isolates, respectively. The Australian H. paragallinarum isolates were very similar, although minor variations were observed among the isolates. In contrast, the non-Australian isolates were markedly different, both from each other and from the Australian isolates. DISCUSSION
In this study we have established that REA is a suitable technique for comparing the genetic relatedness ofH. paragallinarum isolates. In the initial work, we examined the suitability of four restriction enzymes (BamHI, EcoR1, HindlII and Sinai) using eight H. paragallinarum isolates from diverse sources. SmaI did not generate fragments detectable in the electrophoresis system used, while BamHI, EcoRI and HindlII all generated fragments patterns of varying complexity. All of the enzymes digested control 2 DNA to completion even when it was mixed with H. paragallinarum chromosomal DNA. This retention of enzyme activity in the presence of the H. paragalli-
Harum DNA indicates that there were no inhibitory factors present in the preparations. Within the genus Haemophilus, the restriction enzymes previously found to generate satisfactory banding patterns have been BamHI and Hindlll for Actinobacillus (Haernophilus) pleuropneumoniae ( Maclnnes et al., 1990 ), and BamHI for tt. parasuis and 'H. somnus' (McGillivery et al., 1986: Smart et ai., 1988). While we found that BamHI, EcoRI and HindIII are suitable for use with H. paragallinarum, HindIII generated the most DNA fragments, allowing the easiest recognition of genetic differences amongst the isolates. Plasmids have not previously been reported in H. paragallinarum. In a previous study using specific plasmid isolation techniques, we failed to detect any plasmids in 75 isolates o f H . paragallinarurn, despite the fact that 20 of those isolates showed resistance to streptomycin a n d / o r tetracycline ( Blackall, 1988 ). Further studies to determine if any function can be associated with plasmid p250 are required. Previously, we showed that ltindIII REA patterns are stable under field conditions (Blackall et al., 1990b ). The results of the present study have confirmed this finding. As well, we have shown that REA patterns are stable, despite extensive subculturing, under laboratory conditions. Australian isolates of tt. paragallinarum showed only minor differences in their HindllI REA patterns despite the fact that the isolates possessed a considerable range in phenotypic properties (Table 1 ), representing all the Kume serovars and biochemical and antimicrobial biovars known to occur in Australian isolates of H. paragallinarum (Blackall et al.. 1989). In contrast, selected non-Australian isolates, which represented a similar range of phenotypic variation (Table 2 ), showed considerable differences in their tlindlll REA patterns. Our finding that the HindIII REA patterns of Australian H. paragallinarum isolates are very similar indicates that there is little genetic variation in these isolates. It is possible that all Australian H. paragallinarum isolates have evolved from only a few original strains that were introduced into this country. The physical isolation of Australia and, in recent times, strict quarantine regulations have meant reduced opportunities for the importation of live chickens into Australia and hence limited the genetic pool of II. paragallinarum. Our results suggest that in countries like Australia, with a history of biological isolation, bacterial poultry pathogens may show very limited genetic variation. As methods to detect chickens harbouring H. paragallinarum in the absence of clinical disease have not been established, it is possible that H. paragallinarum and other poultry pathogens, not regarded as 'exotic agents', may enter Australia if quarantine regulations are altered and live bird importations allowed. Such importations could result in a widened genetic pool and the establishment of strains from new serovars or possessing greater virulence.
COMPARISON OF flAEMOPHILUS PARAGALLINARUM ISOLATES
47
ACKNOWLEDGMENTS
This work was partly supported by grants from the Australian Chicken Meat and Egg Industry Research Councils. The technical assistance of S. Brice and D.G. Rogers is gratefully acknowledged.
REFERENCES Blackall, P.J., 1988. Antimicrobial drug resistance and the occurrence of plasmids in Haemophilus paragallinarurn. Avian Dis., 32: 742-747. Blackall, P.J., Eaves, L.E. and Rogers, D.G., 1989. Biotyping of Haemophilus paragallinarum using hemagglutinin serotyping, carbohydrate fermentation patterns, and antimicrobial drug resistance patterns. Avian Dis., 33:491-496. Blackall, P.J., Eaves, L.E. and Rogers, D.G., 1990a. Proposal of a new serovar and altered nomenclature for Haemophilus paragallinarum in the Kume hemagglutinin scheme. J. Clin. Microbiol., 28:1185-1187. Blackall, P.J., Morrow, C.J., Maclnnes, A., Eaves, L.E. and Rogers, D.G., 1990b. Epidemiological studies on infectious coryza outbreaks in northern New South Wales, Australia using serotyping, biotyping and restriction endonuclease analysis of chromosomal DNA. Avian Dis., 34: 267-276. Hintermann, G., Fischer, H.-M., Crameri, R. and Hurter, R., 1981. Simple procedure for distinguishing CCC, OC and L forms of plasmid DNA by agarose gel electrophoresis. Plasmid, 5: 371-373. Kleven, S.H., Browning, G.F., Bulach, D.M., Ghiocas, E., Morrow, C.J. and Whithear, K.G., 1988a. Examination ofMycoplasma gallisepticum strains using restriction endonuclease DNA analysis and DNA-DNA hybridisation. Avian Pathol., 17: 559-570. Kleven, S.H., Morrow, C.J. and Whithear, K.G., 1988b. Comparison of Mycoplasma gallisepticum strains by hemagglutination-inhibition and restriction endonuclease analysis. Avian Dis., 32: 731-741. Kume, K., Sawata, A., Nakai, T. and Matsumoto, M., 1983. Serologxcal classification of Haemophilus paragallinarum with a hemagglutinin system. J. Clin. Microbiol., 17: 958-964. Maclnnes, J.I., Borr, J.D., Massoudi, M. and Rosendal, S., 1990. Analysis of Southern Ontario Actinobacillus (Haemophilus) pleuropneumoniae isolates by restriction endonuclease fingerprinting. Can. J. Vet. Res., 54: 244-250. Marshall, R.B., Wilton, B.E. and Robinson, A.J., 1981. Identification of leptospira serovars by restriction-endonuclease analysis. J. Med. Microbiol., 14:163-166. McGillivery, D.J., Webber, J.J. and Dean, H.F., 1986. Characterisation ofHistophilus ovis and related organisms by restriction endonuclease analysis. Aust. Vet. J., 63: 389-393. Rimler, R.B., 1979. Studies of the pathogenic avian haemophili. Avian Dis., 23:1006-1018. Roberts, I., Holmes, W.M. and Hyleman, P.B., 1986. Modified plasmid isolation method for CTostridium perfringens and Clostridium absonum. Appl. Environ. Microbiol., 52: 197-199. Smart, N.L., Miniats, O.P. and Maclnnes, J.I., 1988. Analysis of Haemophilus parasuis isolates from southern Ontario swine by restriction endonuclease fingerprinting. Can. J. Vet. Res., 52:319-324. Yamamoto, R., 1984. Infectious coryza. In: M.S. Hofstad, H.J. Barnes, B.W. Calnek, W.M. Reid and H.W. Yoder, Jr. (Editors), Diseases of Poultry. 8th Edn. Iowa State University Press, Ames, IA, pp. 178-186.
