Initial screening. (i) Pathological and serological findings.

No anti-M. gallisepticum antibody was detected in the serum of any bird at the time of inoculation. More serum samples were anti-M. gallisepticum antibody positive at 4 weeks than at 2 weeks after infection, but no antibody response was detected in uninoculated in-contact birds in any group (Table 3). Fewer birds had air sac lesions at 4 weeks than at 2 weeks after inoculation, and more severe air sac lesions were generally observed at 2 weeks after infection (scores of 0.5 in group A, 1.0 to 2.0 in group B, and 0.5 to 2.5 in group C). The severity of the lesions did not correlate closely with the serology results, with some birds having high rapid serum agglutination (RSA) scores but no detectable air sac lesions.

TABLE 3

Serology, air sac lesions, and mutants targeted for further investigation detected in each group in initial and confirmatory screens

Screen and group	No. of birds positive for the following/no. examined:							Mutant tag(s) detected (no. of times detected)
	Antibody against M. gallisepticum				Air sac lesions
	Wk 0	Wk 2	Wk 4	Control	Wk 2	Wk 4	Control	Air sacs	Trachea
Initial screen
Aa	0/30	7/9	8/9	0/8	2/10	0/9	0/8	oppD1-20 (2)b	oppD1-20 (6), oppC1-24 (6), oppD1-26 (1) MGA_0588-18 (1)
B	0/31	6/10	9/11	0/10	2/10	1/11	0/10	oppD1-20 (1)	oppD1-20 (6)
Cc	0/29	5/10	9/10	0/7	4/10	2/10	0/7	oppD1-20 (7), MGA_0588-18 (1)	oppD1-20 (10), oppC2-24 (5), oppB1a-36 (1), MGA_0588-18 (1)
Confirmatory screend
A	0/20	15/19			3/19			MGA_1102-03 (1)	MGA_1102-03 (2)
B	0/20	14/19			1/19				MGA_1102-03 (1), MGA_1079-22 (2)

Open in a separate window

^aTwo inoculated and two in-contact birds in group A died prior to necropsy.

^bIncluding recovery from one bird in the in-contact group at 4 weeks after inoculation.

^cTwo in-contact birds in group C died prior to necropsy.

^dOne bird each in groups A and B died before necropsy. The pool for group A contained mutant 03-1, and the pool for group B contained mutant 22-1.

(ii) Identification of recoverable ST mutants.

The rates of recovery of ST mutants from birds at 14 and 28 days after inoculation are summarized in Tables 3 and ​and4.4. The oppD₁ ST mutant 20 (mutants 20-1, -2, and -3) was the mutant most frequently recovered from the air sacs, as well as from the tracheas, and was isolated at a high frequency from all three groups. It was also recovered from the air sacs of one in-contact control bird in group A 4 weeks after inoculation. In contrast, oppD₁ ST mutant 26 (mutants 26-1, -2, and -3) was recovered only from the trachea of one bird in group A. Six and five isolations of oppC₁ ST mutant 24 (mutants 24-1 and -3) were made from the tracheas of birds in groups A and C, respectively, while only one isolation of the oppB_1a ST mutant 36 (mutant 36-3) was made, and it was from the trachea of a bird in group C. MGA_0588 ST mutant 18 was reisolated only from the air sacs of one bird in group C (mutant 18-3) and from the trachea of one bird each in groups A and C (mutants 18-1 and -3, respectively). Sixteen ST mutants, including ST mutants 03 (mutants 03-1, -2, and -3; MGA_1102 interrupted) and 22 (mutants 22-1, -2, and -3; MGA_1079 interrupted), were unable to be recovered from any bird in any of the groups (Table 3).

TABLE 4

All mutants detected in each experimental group in initial and confirmatory screens^a

Screen and group	Signature tags detected (no. of times detected)
	Air sacs	Trachea
Initial screen
Ab	16-1 (1),c 20-1 (2)c	07-1 (3), 11-1 (1), 12-1 (4), 14-1 (1), 18-1 (1), 19-1 (2), 20-1 (6), 21-1 (1), 24-1 (6), 26-1 (1), 27-1 (2), 31-1 (1), 32-1 (1), 35-1 (1), 38-1 (1)
B	12-2 (1), 20-2 (1)	07-2 (1), 11-2 (1), 12-2 (3), 14-2 (1), 16-2 (1), 20-2 (6), 27-2 (1)
Cd	11-2 (1), 12-2 (1), 16-2 (3), 18-2 (1), 20-2 (7), 27-2 (1)	01-3 (1), 02-3 (8), 11-3 (1), 12-3 (5), 13-3 (1), 14-3 (1), 16-3 (3), 18-3 (1), 20-3 (10), 24-3 (5), 27-3 (6), 36-3 (1), 38-3 (2)
Confirmatory screene
A	03-1 (1), 09-1 (1), 10-1 (2), 15-1 (1), 23-1 (2)	02-2 (4), 03-1 (2), 04-3 (1), 09-1 (3), 10-1 (10), 15-1 (1), 17-2 (12), 23-1 (2), 25-1 (1), 34-1 (1)
B	09-3 (2), 17-3 (1), 23-1 (2)	02-2 (1), 03-1 (1), 09-3 (14), 10-1 (5), 15-1 (2), 17-3 (10), 23-1 (15), 25-1 (3), 34-1 (2)

Open in a separate window

^aBirds in group A were inoculated with a pool containing ST mutants 02-2, 03-1, 04-1, 06-1, 09-1, 10-1, 15-1, and 17-2; birds in group B were inoculated with a pool containing ST mutants 04-3, 09-3, 17-3, 22-1, 23-1, 25-1, 33-1 and 34-1. The following tags were not detected in the initial screen: 02-2, 03, 04-3, 09, 10, 15, 17, 22, 23, 25, and 34. The following mutants have been described previously (5): 02-1; 04-1; 06-1, -2, and -3; 28-1, -2, and -3; 33-1, -2, and -3; and 39-1, -2, and -3.

^bOne inoculated and two in-contact birds in group A died prior to necropsy.

^cIncluding recovery from one bird in the in-contact group at 4 weeks after inoculation.

^dTwo in-contact birds in group C died prior to necropsy.

^eOne bird each in groups A and B died before necropsy.

Confirmatory screening. (i) Pathological and serological assessments.

No anti-M. gallisepticum antibody was detected prior to inoculation in any bird (Table 3). Severe air sac lesions (lesion score, 2.5) were seen in one bird, and mild lesions (lesion scores, 0.5 and 1.0) were seen in another two chickens in group A. One bird in group B had mild air sac lesions (lesion score, 1.0). At 2 weeks after inoculation, 15/19 chickens in group A and 14/19 birds in group B had detectable antibody against M. gallisepticum (Table 3).

(ii) Reisolation of ST mutants.

Birds in group A were infected with a pool containing MGA_1102 ST mutant 03-1, and a pool that contained MGA_1079 ST mutant 22-1 was used to inoculate the birds in group B. A total of 11 ST mutants, including ST mutant 03-1, were reisolated from 16 chickens in group A, and 10 mutants, including ST mutants 22-1 and 03-1, were reisolated from 18 birds in group B (Table 3).

Infectivity and virulence analysis of selected ST mutants. (i) Clinical signs and postmortem examination.

The prevalence and severity of lesions in chickens infected with individual mutants are shown in Table 5. Air sac lesions were not seen in the uninfected control birds (group 1) or in birds exposed to aerosols of MGA_1079 ST mutant 22-1 (group 5). Mild lesions (lesion score, 0.25) were observed in one bird inoculated with MGA_1102 ST mutant 03-1 (group 2). Of the 20 birds infected with oppD₁ ST mutant 26-1 (group 6), four had mild lesions (lesion scores, 0.5 to 1.0), while lesions were seen only in the abdominal air sacs of 5 birds exposed to MGA_0588 ST mutant 18-1 (group 3). Six of 20 birds in the group infected with oppD₁ ST mutant 20-1 (group 4) had mild to severe lesions (lesion scores, 0.5 to 2.5), while mild to severe lesions (lesion scores, 0.5 to 3.0) were seen in 11/18 birds infected with the virulent Ap3AS strain (group 7).

(ii) Antibody responses and reisolation of ST mutants.

No anti-M. gallisepticum antibody was detected at the time of infection in the serum of any birds (Table 5). Two weeks after exposure, antibody responses were not detected in any of the birds in group 5 (MGA_1079 mutant 22-1 infected) or 6 (oppD₁ mutant 26-1 infected), while a response was detectable in only one bird in group 2 (MGA_1102 mutant 03-1 infected). In contrast, strong antibody responses against M. gallisepticum were detectable in all the birds in groups 3 (MGA_0588 mutant 18-1 infected) and 4 (oppD₁ mutant 20-1 infected). The uninfected control birds (group 1) did not develop any antibody against M. gallisepticum over the course of the experiment, while all birds infected with the wild-type Ap3AS strain (group 7) had strong antibody responses against M. gallisepticum.

M. gallisepticum was not isolated on mycoplasma agar (MA) plates inoculated with swab specimens of the air sacs of any birds in group 2 (MGA_1102 mutant 03-1 infected) or 5 (MGA_1079 mutant 22-1 infected), but it was isolated from the tracheas of two birds in group 2 (MGA_1102 mutant 03-1 infected) and one bird in group 5 (MGA_1079 mutant 22-1 infected). In group 6 (oppD₁ mutant 26-1 infected), M. gallisepticum was isolated from the air sacs of one bird and the tracheas of six birds. M. gallisepticum was also isolated from the tracheas of 17/20 birds in group 3 (MGA_0588 mutant 18-1 infected) and of all the birds in group 4 (oppD₁ ST mutant 20-1 infected) and also from swab specimens of the air sacs of 5/20 birds in group 3 and 7/20 birds in group 4. In the positive-control group (group 7), M. gallisepticum was isolated from the air sacs of 10/18 birds and the tracheas of 18/18 birds (Table 5).

M. gallisepticum was recovered in mycoplasma broth (MB) inoculated with swab specimens of the air sacs of one bird in group 6 (oppD₁ mutant 26-1 infected), seven birds in group 3 (MGA_0588 mutant 18-1 infected), and eight birds in group 4 (oppD₁ mutant 20-1 infected). M. gallisepticum was recovered from the tracheas of birds in all six of the groups exposed to ST mutants, with the number of colonized birds ranging from 3 in group 5 (MGA_1079 mutant 22-1 infected) to 20 in group 4 (oppD₁ mutant 20-1 infected). The identity of the recovered M. gallisepticum ST mutants was confirmed using the unique signature tags that they carried (Table 5).

(iii) Tracheal lesions and mucosal thicknesses.

Median tracheal lesion scores are shown in Table 6. The scores of birds in all the mutant-infected groups differed significantly from those of birds in the wild-type-infected group (group 7) (P < 0.0001). There was no significant difference between the lesions scores for the uninfected group (group 1) and those for the mutant-infected groups, with the exception of those for the upper tracheas of the birds infected with MGA_1102 mutant 03-1. The upper tracheal lesion scores of the birds infected with MGA_1102 mutant 03-1 (group 2) did not differ from those of birds in groups 3, 5, and 6 but did differ significantly from those of birds in group 4 (oppD₁ mutant 20-1 infected).

Mean tracheal mucosal thicknesses are shown in Table 6. The mean tracheal mucosal thicknesses of the wild-type-infected birds were significantly different from those of both the uninfected and the mutant-infected birds (P < 0.0001). The mean mucosal thicknesses in the middle and lower tracheas did not differ significantly between the uninfected group and any of the mutant-infected groups (groups 2 to 6). However, they were significantly different between the mutant-infected groups (groups 2 to 6) and the wild-type-infected group (group 7). While there was no significant difference between the tracheal mucosal thicknesses in the upper tracheas of the uninfected controls and birds in groups 3, 4, 5, and 6, the mean upper tracheal mucosal thickness in the group infected with MGA_1102 mutant 03-1 was significantly greater than that in the uninfected birds and birds infected with oppD₁ mutant 20-1. In summary, the wild-type-infected controls had significantly more severe tracheal lesions, as assessed by histological lesion score or mucosal thickness, than any of the groups infected with the ST mutants. Group 2 (MGA_1102 mutant 03-1 infected) was the most severely affected among the groups infected with the ST mutants.

Screening of ST mutants in infected birds.

All infection experiments were approved by the University of Melbourne Animal Ethics Committee. The experimental design has been described previously (5). Briefly, in the initial screening, 102 ST mutants labeled with 34 different signature tags were cultured and combined into three pools, each of which contained 34 different mutants with distinct tags (Tables 1 and ​and2),2), and each pool was used to infect 20 specific-pathogen-free (SPF) chickens by aerosol (31), with an additional 10 uninfected chickens being included with each group to assess the transmissibility of the mutants. Necropsies were conducted on 10 inoculated birds at 14 days after infection and on the remaining 20 birds at 28 days after infection. Gross air sac lesions were assessed for severity as described previously (32). Swab specimens of the trachea and air sacs of each bird were assessed by culture in mycoplasma broth and on mycoplasma agar, as described previously (5), and the signature tags present were amplified by PCR using the P2/P4 primer pair if the broth culture showed signs of growth of M. gallisepticum. The tags present were identified by dot blotting, as described previously (5). Blood samples were collected from all birds before inoculation and before euthanasia, and the serum was assessed for antibody against M. gallisepticum using a commercial agglutination antigen (Intervet International) in the rapid serum agglutination (RSA) test (33).

Dilutions of the ST mutants that were not reisolated in the initial screening experiment were subjected to confirmatory screening to further reduce the number of candidate ST mutants for definitive testing. The mutants were separated into two pools, with the mutants in each pool being distinguishable from each other by their distinct tags. Each pool contained a similar concentration of each mutant, and each pool was used to infect 20 birds, as described above. In-contact controls were not included, but all 40 birds were placed into the same positive-pressure fiberglass isolator so that each group could be used as an in-contact group for the other. Necropsies and sample collection were performed as described above at 14 days after inoculation. Dot blot hybridization and ST mutant-specific PCRs were performed to identify the ST mutants recovered.

Comparison of the levels of persistence and virulence of selected mutants.

Several mutants were detected infrequently in the second screening experiment, suggesting that the genes that were interrupted in these mutants may have a significant influence on virulence. Two different oppD₁ ST mutants were recovered from birds at distinct rates, suggesting that the different insertion sites may have different impacts on persistence in vivo and potentially on pathogenicity. To assess this and to assess the persistence and virulence of three mutants with insertions in hypothetical genes, the pathogenicities and rates of colonization of five distinct mutants were compared. Seven groups of 20 4-week-old SPF chickens were housed separately. Group 1 was exposed to an aerosol of mycoplasma broth only, group 2 was exposed to MGA_1102 ST mutant 03-1, group 3 was exposed to MGA_0588 ST mutant 18-1, group 4 was exposed to oppD₁ ST mutant 20-1, group 5 was exposed to MGA_1079 ST mutant 22-1, group 6 was exposed to oppD₁ ST mutant 26-1, and the 7th group was exposed to wild-type strain Ap3AS. All birds were subjected to necropsies at 14 days after infection, with samples being collected as described above. Sections of the upper, middle, and lower trachea were also taken from each bird.

Dot blot hybridization and specific PCRs were used to identify ST mutants reisolated in broth cultures. Because strain Ap3AS did not have a specific sequence tag, dot blot hybridization could not be used for cultures from the positive-control group. A pair of PCR primers, STM13-KE-complement-1-Rev and STM13-KF-C, which yielded a 400-bp product, were used to confirm the presence of Ap3AS (Table 7) (5). Sections of the trachea were processed and stained for histological examination and scored for mucosal lesions, and the mucosal thickness was measured (32). Statistical analyses were conducted using Kruskal-Wallis tests to compare median tracheal lesion scores, with the Dunn post hoc test being used to assess the significance of differences between pairs of groups, and an analysis of variance and Tukey's test were used to evaluate the mean tracheal mucosal thicknesses. Differences in the proportions of birds with lesions, of birds with serum antibody against M. gallisepticum, and of birds positive by culture for M. gallisepticum were assessed using Fisher's exact test (34). A probability of less than or equal to 0.05 was regarded as significant.

This work was partially supported by funding from the Australian Egg Corporation Ltd. and the Australian Poultry CRC.

We thank Chris Minion, Amir Hadji Noormohammadi, Cheryl Colson, and Tony Belfiore for their assistance with these studies.