Previous Article | Next Article ![]()
Journal of Clinical Microbiology, February 2005, p. 909-912, Vol. 43, No. 2
0095-1137/05/$08.00+0 doi:10.1128/JCM.43.2.909-912.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Department of Avian Medicine, College of Veterinary Medicine, University of Georgia, Athens, Georgia,1 Division of Avian & Aquatic Diseases, Kimron Veterinary Institute, Bet Dagan, Israel,2 Department of Medical Microbiology and Infection Control, University Hospital Vrije Universiteit, Amsterdam, The Netherlands3
Received 11 June 2004/ Accepted 6 September 2004
|
|
|---|
|
|
|---|
Condemnation of the infected flocks and reduction in feed conversion and egg production are the major factors related with economic losses. However, nonpathogenic species such as M. gallinarum and M. gallinaceum are often isolated and must be differentiated from pathogenic species. Amplified fragment length polymorphism (AFLP) has been extensively tested by Kokotovic et al. for typing mycoplasma species isolated from food animals such as cattle (9), swine (8), and goats (6) and from humans (7). The major advantage of this technique compared to other molecular typing methods is that it requires a relatively small amount of DNA and has great discriminatory power and reproducibility (15). So far, the system has not been evaluated for typing avian mycoplasma species. The objective of this study was to apply AFLP for differentiating avian mycoplasma species.
A total of 44 strains of avian mycoplasma representing eight different species were used in this study (Table 1). They were obtained from the depository at the Poultry Diagnostic and Research Center in Athens, Ga. Mycoplasma strains were cultured in Frey's medium with 12% swine serum (2). Unlike Kokotovic's procedure (7), the DNA template in this study was prepared by a simple boiling method described by Fan et al. (2). Mycoplasma culture (1.5 ml) harvested at late log phase was used for DNA extraction. DNA was stored at 20°C until used.
|
View this table: [in a new window] |
TABLE 1. Mycoplasma strains used in this study
|
Fragment detection was carried on an ABI 310 automatic sequencer (ABI Applied Biosystems, Foster City, Calif.). A mixture consisting of 2.0 µl of PCR products, 12.0 µl of 100% deionized formamide, and 0.5 µl of GeneScan 500 ROX size standard (ABI Applied Biosystems, Foster City, Calif.) was heated to 95°C for 5 min and quickly chilled on ice before electrophoresis on the machine. Fragment size determination and pattern analysis were performed by using GeneScan 3.1 fragment analysis software (ABI Applied Biosystems). ABI chromatograms were converted into schematic gel images with GelCompar II 3.5 (Applied Math Inc., Austin, Texas). Background subtraction and data normalization were subsequently conducted. Cluster analysis was performed using the Pearson correlation and unweighted pair group methods with average linkages.
AFLP analysis in this study provided an optimal separation and a uniform sizing of the amplified fragments. Fragments of between 75 and 500 bp were used in numerical and cluster analysis for species differentiation. We found that M. gallisepticum strains had the highest banding pattern complexity, consisting of about 90 AFLP fragments, while M. meleagridis profiles had the lowest complexity, having approximately only 20 amplified fragments. Both M. gallinarum and M. iowae strains contained about 50 fragments. About 60, 40, and 30 fragments were generated for M. gallinaceum, M. pullorum, and M. synoviae, respectively. M. imitans had 70 fragments in its AFLP banding pattern. Reproducibility of AFLP analysis has been extensively tested by Kokotovic et al. (7). In this study, AFLP procedure was repeated three times on three randomly chosen strains from each species and highly reproducible results were obtained. Minor changes in band intensities existed but were insignificant for determining identity of strains (data not shown).
On the basis of cluster analysis, we chose the 35.0% linkage level (percent homology) as a cutoff value for discriminating mycoplasma strains at the species level (Fig. 1). AFLP data revealed eight distinct groups (I to VIII), each consisting of strains belonging to a single species. All strains within each group had different AFLP patterns, some of which were nearly identical. To validate this clustering method, we calculated error flags (not shown in the dendrogram) representing the value of the mean linkage level plus standard deviation for those groups of strains as indicated in the following paragraphs. The 35.0% cutoff line did not cross any of those error flags.
![]() View larger version (63K): [in a new window] |
FIG. 1. AFLP fingerprints of avian mycoplasma species. The dendrogram was constructed using Pearson correlation and the unweighted pair group method with average linkages. The eight avian mycoplasma species groups generated at a 35.0% linkage level cutoff point are indicated.
|
On the other hand, M. iowae and M. meleagridis strains showed high homogeneity in their AFLP profiles and clustered at linkage levels of 73.2% ± 6.2% (group VII) and 68.0% ± 7.8% (group VIII), respectively. Previous studies revealed that although heterogeneity in serological responses was observed for M. iowae (10), less-variable protein profiles and random amplified polymorphic DNA (RAPD) patterns were obtained for different strains (3, 14). Among pathogenic avian mycoplasmas, M. gallisepticum strains revealed the widest intraspecies heterogeneity by AFLP analysis, with a linkage level of 59.2% ± 2.2% (group I). The genetic variation of this species has been documented by several other molecular typing techniques, such as pulse field gel electrophoresis (12), random amplified polymorphic DNA (2, 12), restriction fragment length polymorphism (13), and Southern blotting (17). M. synoviae strains, with a linkage level of 66.1% ± 3.0% (group V), exhibited more genetic homogeneity than M. gallisepticum, and the same conclusion was made using an rRNA gene hybridization test conducted by Yogev et al. (16).
M. imitans shared many phenotypic properties with M. gallisepticum but had low genetic homology with M. gallisepticum in a DNA-DNA hybridization study (1). In this study, the two species were typed into related groups and linked at the 25.3% homology level.
As determined on the basis of these results, AFLP can be used as an additional confirmatory tool for identification of avian mycoplasma species. This can be achieved by setting up a database for reference strains; such a database was partially created in this study and can be easily expanded. Our follow-up studies will focus on using AFLP for typing avian mycoplasma strains within each species.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»