This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 3.0 License.
Introduction
Infections caused by Gallibacterium anatis (G. anatis) can lead to very serious clinical symptoms. Even though this Gram-negative bacterium exists as part of the commensal bacterial flora in birds, with inappropriate environmental factors or co-infections, its pathogenicity can change (19, 31). Mixed bacterial and viral infections are a factor in the development of gallibacteriosis. Bacteria such as E. coli, Avibacterium paragallinarum, Mycoplasma gallisepticum and M. synoviae and viruses such as Newcastle disease virus and adenoviruses can intensify symptoms or increase the mortality of birds in a flock infected with G. anatis (11, 12, 17, 19, 31, 33). In addition, the influence of other factors such as the strain of bacteria, the age and immune status of the birds or stress can sometimes be decisive in the process of weakening avian health (19, 28). There are two phenotypically different biovars of G. anatis. The distinction is made on the basis of their haemolytic properties: the haemolytica biovar causes β-haemolysis, and the anatis biovar does not. It seems that the haemolytica biovar may be more pathogenic (7, 16, 24, 31). G. anatis spreads in flocks by the respiratory route through direct contact between birds in the flock as well as by the vertical route through infected eggs. Its spread is also related to its habitat. The haemolytica biovar causes a range of symptoms associated with three tracts: oophoritis and salpingitis in the reproductive tract; peritonitis, perihepatitis, liver necrosis and enteritis in the gastrointestinal tract; or air sacculitis and tracheitis in the respiratory tract, in addition to inducing septicaemia in chickens (6, 15, 24, 26, 28, 33). In laying hens, the reproductive organs are chiefly affected, and this bacterium produces lesions including haemorrhagic oophoritis and rupture of ovarian follicles (18, 24, 26).
In cockerels, the bacterium causes inflammation of the epididymides and leads to reduced semen quality (27). Infection with G. anatis decreased the laying rate by 8–10%, lowered productivity, and caused mortality of up to 73% in laying hens subjected to experimental immunosuppression. In young chickens, on the other hand, the lesions were usually systemic (19, 26, 37).
On bovine blood agar, smooth, greyish, opaque, shiny colonies of G. anatis biovar haemolytica produce a wide haemolytic zone. Responsible for this is the cytotoxin GtxA (Gallibacterium toxin A), a two-domained protein with a C- and an N-terminus and one of the virulence factors of G. anatis (20, 36). It has the ability to make pores in the plasma membrane of host cells, which can ultimately lead to their necrosis or apoptosis. Cytotoxin GtxA lyses red blood cells in a wide variety of hosts and is also leukotoxic (20, 28, 36). The distribution of G. anatis in the world is quite wide. It is found in poultry flocks on different continents, from Europe and Asia to North America and Africa, and has been isolated not only from poultry but also from wild birds, calves and even humans (2, 3, 23, 29, 31, 32, 34).
G. anatis strains show multidrug resistance to antibacterial substances, which is of concern because of their wide distribution. Multidrug resistance of bacteria is a growing problem for the poultry industry as well as for public health (1, 2, 17, 19, 31, 35). Recent studies show common resistance in a large number of G. anatis strains to erythromycin and tylosin antibiotics from the macrolide class as well as antibiotics from the tetracycline class (2, 8, 17, 21, 31, 35).
The purpose of this study was to examine the susceptibility of G. anatis biovar haemolytica isolates collected from the respiratory, reproduction and gastrointestinal tracts of chickens to different antibiotics from various classes.
Material and Methods
Sampling procedures
Twenty six G. anatis isolates showing β-haemolysis capability were selected from the sample collection. The isolates were from chicken flocks, collected from the respiratory (n = 8), gastrointestinal (n = 8) and reproductive (n = 10) tracts. Each sample came from a different flock (Table 1).
List of Gallibacterium anatis isolates obtained from flocks of hens
Isolate
Type of flock
Age of birds (weeks)
Tract of sample origin
GA1
layer
28
respiratory
GA2
layer
28
respiratory
GA3
layer
31
respiratory
GA4
layer
30
reproductive
GA5
layer
28
reproductive
GA6
layer
31
reproductive
GA7
layer
31
reproductive
GA8
layer
28
reproductive
GA9
layer
28
reproductive
GA10
layer
30
reproductive
GA11
layer
25
respiratory
GA12
broiler
4
respiratory
GA13
broiler
4
respiratory
GA14
broiler
3
respiratory
GA15
layer
27
reproductive
GA16
layer
29
reproductive
GA17
layer
28
reproductive
GA18
broiler
4
respiratory
GA19
broiler
4
gastrointestinal
GA20
broiler
3
gastrointestinal
GA21
broiler
4
gastrointestinal
GA22
broiler
4
gastrointestinal
GA23
broiler
3
gastrointestinal
GA24
broiler
4
gastrointestinal
GA25
broiler
4
gastrointestinal
GA26
broiler
4
gastrointestinal
Trachea swab samples were brought to the Department of Poultry Diseases at the National Veterinary Research Institute in Poland as part of a routine diagnostic test and monitoring programme. Tissues from the gastrointestinal and reproductive tracts were aseptically obtained from birds sent for diagnostic purposes. All examined birds were floor reared. Some of the birds had respiratory signs in the form of rales and coughing, and some had swollen heads and poorer laying performance.
Isolation and identification of G. anatis.
Samples were cultured onto Columbia agar with 5% sheep’s blood for 24 h at 37 ± 1°C in an atmosphere of 5% CO2. After incubation, three colonies from each plate with morphology characteristic of β-haemolysis were selected and transferred to nutrient agar and incubated for 24 h at 37 ± 1°C. Identification of colonies was performed using matrix-assisted laser desorption/ ionisation-time-of-flight mass spectrometry (MALDI-TOF) and the MALDI Biotyper system with MBP COMPASS 4.1 software (Bruker Daltonics, Bremen, Germany). Bacterial colonies from the agar plate were transferred to the MALDI target plate and mixed with formic acid and α-cyano-4-hydroxycinnamic acid matrix solution. Strains identified as G. anatis by matching them to reference species found in the software’s database were preserved and stored at -20°C for further testing.
Antibiotic resistance
The antimicrobial susceptibility of G. anatis biovar haemolytica was determined by using two methods: disc diffusion and microbroth dilution. With these methods it was possible to test the antibiotic resistance of isolates to 20 antimicrobial substances from 12 different classes of antibiotics. Eight different antibiotics were used in the disc diffusion method (Antimicrobial Susceptibility Testing discs; Oxoid, Basingstoke, UK): florfenicol (30 μg) in the amphenicol class, doxycycline (30 μg) in the tetracycline class, amoxicillin (25 μg) in the β-lactam class, enrofloxacin (5 μg) in the fluoroquinolone class, colistin (50 μg) in the polymyxin class, ceftazidime (30 μg) in the cephalosporin class, and tilmicosin (15 μg) and tylosin (30 μg) in the macrolide class (Table 2). The test applied a bacteria volume of 100 μL at 1.5 × 108 colony-forming units/mL (0.5 McFarland scale) distributed uniformly onto the Columbia agar with 5% sheep’s blood. The inhibition zones were interpreted visually.
Determination of minimum inhibitory concentration (MIC) was performed according to the Clinical and Laboratory Standards Institute standard M31-A2(13) using a commercially prepared dehydrated panel for Enterobacteriaceae (Sensititre EU Surveillance Enterococcus EUVENC AST plate; Thermo Fisher Scientific, Waltham, MA, USA). The plates were incubated for 20–24 h at 37 ± 1°C under aerobic conditions. Antimicrobial susceptibility testing of G. anatis biovar haemolytica was performed from a fresh culture on agar and a suspension prepared at 0.5 McFarland density in 0.9% NaCl (bioMérieux, Marcy-’l’Étoile, France), of which 10 μL was transferred to 11 mL of Mueller–Hinton broth (Thermo Fisher Scientific). The suspension was thoroughly vortexed and then 50 μL of it was added to each well of a plate. The plates contained different concentrations of 12 different antibiotics: gentamicin (8–1,024 mg/L) in the aminoglycoside class, ampicillin (0.5–64 mg/L) in the β-lactam class, chloramphenicol (4–128 mg/L) in the amphenicol class, ciprofloxacin (0.12–16 mg/L) in the fluoroquinolone class, teicoplanin (0.5–64 mg/L) and vancomycin (1–128 mg/L) in the glycopeptide class, daptomycin (0.25–32 mg/L) in the lipopeptide class, erythromycin (1–128 mg/L) in the macrolide class, linezolid (0.5–64 mg/L) in the oxazolidinone class, quinupristin/dalfopristin (0.5–64 mg/L) in the streptogramin class, and tetracycline (1–128 mg/L) and tigecycline (0.03–4 mg/L) in the tetracycline class (Table 2). The plates were incubated for 24h at 37 ± 1°C. The MIC was defined as the lowest concentration preventing visible growth using a plate reader (Sensititre-TREK Vizion Digital MIC Viewing System; Thermo Fisher Scientific). Strains resistant to at least three classes of antimicrobials were identified as multidrug resistant (MDR).
List of antibiotics used to determine the antibiotic susceptibility of Gallibacterium anatis biovar haemolytica isolates
Antibiotic class
Antibiotic
aminoglycoside
gentamicin
β-lactam
amoxicillin
ampicillin
cephalosporin
ceftazidime
amphenicol
florfenicol
chloramphenicol
fluoroquinolone
enrofloxacin
ciprofloxacin
glycopeptide
vancomycin
teicoplanin
lipopeptide
daptomycin
macrolide
erythromycin
tylosin
tilmicosin
oxazolidinone
linezolid
polymyxin
colistin
streptogramin
quinupristin/dalfopristin
tetracycline
tetracycline
doxycycline
tigecycline
DNA extraction
DNA was extracted from nutrient agar plate cultures with a Maxwell RSC Cultured Cells DNA Kit (Promega, Madison, WI, USA), according to the manufacturer’s protocol. The quantity and quality of the DNA was determined using the NanoDrop 1000 system (Thermo Scientific). The tris-ethylenediaminetetraacetic acid used for sample preparation was the negative control. Samples were frozen at -20°C.
PCR
G. anatis isolates were identified by the PCR method described earlier by Bojesen et al. (9). Specific primers designed to detect the 16–23S ribosomal RNA (rRNA) region were used. The mixture and the conditions described earlier were used for the reaction (21). The PCR amplicons were separated by electrophoresis on a 2% agarose E-gel plate (Invitrogen, Carlsbad, CA, USA) containing ethidium bromide, and were visualised by ultraviolet transillumination.
Virulence and resistance genes
Isolates of G. anatis biovar haemolytica were tested for presence of the gyrB, GtxA and flfA virulence genes. All samples were also tested for the presence of the blaROB, aphA, tetB and tetH antibiotic resistance genes. A PCR method was used for both test steps with the starters described in a previous publication (1).
Statistical analysis
Venn diagrams were constructed showing the number of antibiotics resisted by at least 50% of G. anatis isolates from the three chicken anatomical tracts. To determine the statistical significance of antibiotic resistance differences between respiratory, reproductive and gastrointestinal isolates and presence differences of virulence and resistance genes in G. anatis isolates, the Mann–Whitney and the one-way ANOVA tests were used. The value of P <0.05 was considered statistically significant. Statistical analyses were performed using the Social Science Statistics program (www.socscistatistics.com).
Results
Isolation and identification of G. anatis biovar haemolytica.
Each tested sample showing β-haemolysis on Columbia agar was confirmed by MALDI TOF as G. anatis. The bacterium was also identified by PCR by obtaining amplicons of 1,030 base pairs in all samples.
Antibiotic resistance
All isolates showed the highest incidence of resistance to tilmicosin (100%), tylosin (100%) and quinupristin/dalfopristin (100%), followed by resistance to erythromycin (96.2%), tetracycline (96.2%), linezolid (92.3%) and teicoplanin (92.3%). All resistance percentages are given by antibiotic class in Table 3. G. anatis isolates showed the greatest susceptibility to chloramphenicol. A full 100% of isolates were susceptible to this antibiotic. Florfenicol and ceftazidime were resisted by only 15% and 19% of isolates, respectively. All isolates were MDR. Five isolates of G. anatis biovar haemolytica showed multiresistance to 16 antibiotics, and were from the respiratory and reproductive tracts. Resistance to 16, 15, 14 and 13 antibiotics was shown by six, five, six and three G. anatis isolates, respectively. Resistance to 12, 11, 10 and 8 antibiotics was shown by two, one, three and one isolates, respectively, noted predominantly (75%) in isolates from the gastrointestinal tract.
Antibiotic resistance of Gallibacterium anatis biovar haemolytica isolates (disc diffusion and minimum inhibitory concentration methods)
Antibiotic class
Antibiotic
Resistance %
aminoglycoside
gentamicin
69.2
β-lactam
amoxicillin
88.0
ampicillin
73.1
cephalosporin
ceftazidime
19.0
amphenicol
florfenicol
15.0
chloramphenicol
0.0
fluoroquinolone
enrofloxacin
88.0
ciprofloxacin
69.2
glycopeptide
vancomycin
88.5
teicoplanin
92.3
lipopeptide
daptomycin
69.2
erythromycin
96.2
macrolide
tylosin
100.0
tilmicosin
100.0
oxazolidinone
linezolid
92.3
polymyxin
colistin
31.0
streptogramin
quinupristin/dalfopristin
100.0
tetracycline
tetracycline
96.2
doxycycline
46
tigecycline
57.7
Statistically significant differences (P-value <0.05) were found between the resistance of strains isolated from the gastrointestinal tract and other tracts to the antibiotics: ceftazidime, tylosin, colistin, daptomycin, gentamicin, ciprofloxacin and vancomycin. Nine antibiotics were commonly resisted by at least 50% of the group in the cases of all three anatomical-origin isolate groups (Table 4, Fig. 1). G. anatis isolated from the respiratory and reproductive tracts had at least 50% incidence of resistance to four antibiotics. Isolates of G. anatis obtained from the gastrointestinal tract showed at least 50% resistance to three antibiotics – colistin, ceftazidime and florfenicol – which belong to the polymyxin, cephalosporin and amphenicol antibiotic classes, respectively (Tables 3 and 4, Fig. 2).
Fig. 1.
Antibiotic resistance of Gallibacterium anatis biovar haemolytica isolates by the anatomical tracts from which they were isolated (minimum inhibitory concentration (MIC) method) red line – dilution at which resistance started; * – statistically significant differences (P-value <0.05) between strains isolated from the gastrointestinal tract and other tracts
Fig. 2.
Venn diagram of the numbers of antibiotics resisted by at least 50% of isolates from the chicken respiratory, reproductive and gastrointestinal tracts GA – Gallibacterium anatis
Antibiotic resistance of Gallibacterium anatis biovar haemolytica isolates by anatomical tract of sample origin (disc diffusion method)
* – statistically significant differences (P-value <0.05) between strains isolated from the gastrointestinal tract and other tracts
Presence of virulence and resistance genes
The GtxA and gyrB virulence genes were present in 100% of isolates. In contrast, the flfA gene was present in only 19.2% of isolates. There was no statistically significant difference (P-value >0.05) between the presence of the GtxA gene and the presence of the gyrB gene. However, statistically significant differences (P-value <0.05) were found between the presence of the GtxA and gyrB virulence genes and that of the flfA gene (Fig. 3a).
Fig. 3.
Venn diagram of a) virulence genes and b) resistance genes in Gallibacterium anatis biovar haemolytica * – statistically significant differences (P-value < 0.05) between the presence of the GtxA and gyrB virulence genes and presence of the flfA gene ** – statistically significant differences associated with the presence of virulence genes GtxA and gyrB, and the presence of resistance genes tetH, aphA and blaROB (P-value < 0.05)
The tetB and tetH tetracycline resistance genes constituted the largest group of genes isolated from G. anatis biovar haemolytica strains. The tetB gene was detected in all tested isolates, while tetH was noted in 34.6% of isolates, mostly isolated from the gastrointestinal (23%) and the respiratory tract (12%). The tetH gene was not found in G. anatis isolates from the reproductive tract. A low percentage of G. anatis isolates were detected with the aphA gene (3.8%), and none were identified with the blaROB gene. Five isolates (19.2%) had the GtxA and gyrB genes plus the flfA gene. Only in one isolate (3.8%) originating from the gastrointestinal tract was the presence of the aphA gene also found besides tetB and tetH (Fig. 3b). There were significant differences associated with the presence of virulence genes GtxA and gyrB, and the presence of resistance genes tetH, aphA and blaROB (P-value <0.05).
Discussion
Gallibacterium anatis is one of the major poultry pathogens. The haemolytic biovar can be responsible for multiple clinical signs caused by either single infections or mixed infections with other pathogens, leading to serious economic losses in the poultry industry (15, 24, 31). The present study aimed to investigate the antibiotic resistance of isolates of G. anatis biovar haemolytica obtained from chicken respiratory, reproductive and gastrointestinal tracts. We also examined the presence of resistance genes to three groups of antibiotics and the presence of virulence genes.
This bacterium can be isolated from the trachea as well as the cloaca of healthy commercial chickens because it forms part of the normal chicken microflora (19, 24). Infections can be associated with a variety of clinical signs occurring together in mixed infections. There are also reports that describe single G. anatis infections causing disorders in the reproductive tract of chickens (24, 26). In addition to type of strain, route of infection and involvement of secondary factors influencing the progress of G. anatis infection, there are many other factors involved in the development of disease. The age of the bird, its subjection to stress or the action of particular hormones are all host-related factors. Environmental factors that can affect the development of infection are seasonal changes or cold stress (10, 22, 25). Many reports inform of an increase of antibiotic resistance among G. anatis isolates (8, 17, 21, 35). In our study, we used G. anatis biovar haemolytica strains isolated from three chicken anatomical tracts. Antibiotic resistance of respiratory, reproductive and gastrointestinal isolates was tested against 20 different antibiotics from 12 different classes using two methods. The majority of G. anatis isolates showed high resistance to antibiotics of the macrolide class: resistance to erythromycin, tylosin and tilmicosin was found in 96.2%, 100% and 100%, respectively. Our results are in agreement with the results of other authors (1, 14, 29). Isolates from our collection also showed high resistance to tetracycline (96.2%), tigecycline (57.7%) and doxycycline (46%), proving similar to resistance noted in previous studies (1, 14, 15, 21). In addition, high resistance was evident to gentamicin (69.2%), an antibiotic of the aminoglycoside class. Resistance to antibiotics from these three classes is common as the MDR described on animal farms, and is escalating. This is related to the frequent use of these substances in animal production. Tetracycline resistance was not only very common among our isolates but also in other studies (5, 8, 35). High tetracycline resistance has also been linked to the presence of resistance genes. In our study, two different tetracycline resistance genes were found. The tetB gene was detected in all tested isolates, while tetH was in 34.6% of isolates, which is comparable to the results obtained by other authors (1). Most G. anatis isolates having the tetH gene were from the gastrointestinal tract (23%) and the respiratory tract (12%). The presence of tet genes has been reported in G. anatis isolates in other studies, where it appeared to be frequent in isolates implicated in all types of infections (1, 8, 35). The aphA gene, which encodes a protein associated with aminoglycoside resistance, was found in one isolate from our collection. The low percentage prevalence of the gene detected was not, however, to any extent proportionate to the frequency of resistance of the isolates to the aminoglycoside antibiotic gentamicin, which was high at 69.2%. High antibiotic resistance was also found to quinupristin/dalfopristin in the streptogramin class (100%), teicoplanin (92.3%) and vancomycin in the glycopeptide class (88.5%), amoxicillin in the β-lactam class (88%) and linezolid in the oxazolidinone class (92.3%). Resistance to these antibiotics may be a grave problem because they are used in human medicine. The elevated rate of colistin resistance among the isolates (31%) is also alarming. A finding which contrasted with the instances of resistance noted was that all isolates were susceptible to chloramphenicol. Multidrug resistance is a huge problem in veterinary and human medicine. Recent antibiotic susceptibility studies of field strains of G. anatis have shown a high number of MDR strains. Our research shows that a significant proportion of isolates showed resistance to at least eight different antibiotics. Five isolates were resistant to 16 antibiotics, which is very alarming.
Virulence factors are involved in many aspects of host–pathogen relationships, including colonisation, nutrient acquisition, immune evasion and immunosuppression (1, 28, 35). They include toxins, enzymes and adhesion molecules and are very important virulence factors that confer the ability to cause disease and thus determine the pathogenicity of microorganisms. The main feature of G. anatis biovar haemolytica isolates is the ability to form a wide β-haemolytic zone around the colony on plates (15, 24, 35). This trait was observed in all isolates used in this study. The protein responsible is the secreted toxin GtxA, which exhibits haemolytic activity against erythrocytes (20). The toxin is a specific virulence factor of G. anatis. Its presence gives G. anatis a means of adhering to cells and changes cell permeability and expression of inflammatory factors, resulting in cell damage and apoptosis (20, 30, 36). Kristensen et al. (20) in their research concluded that GtxA may represent a new form of RTX-like toxin with immune evasion function. In this study, this virulence gene was detected in all tested isolates. In addition to GtxA toxin production, G. anatis is known to produce a variety of other virulence factors. One of them, the gyrB gene encoding for the B subunit of the DNA gyrase, was also present in all of our isolates (1, 35). G. anatis biovar haemolytica has the ability to adhere to chicken epithelial cells because of its fimbriae-like structures involved in adhesion to host cells. One of the genes encoding F17-like fimbriae that bind receptors containing N-acetyl-D-glucosamine (Glc-NAc) on the host cell surface is flfA (4, 28, 35). Among the isolates tested in this study, the presence of this gene was found in five of them.
Considering the clinical signs of the hens from which the test material was obtained, it can be concluded that biovar haemolytica may have had an effect on the onset of laying-related symptoms. The hens from which G. anatis isolates were obtained from the gastrointestinal tract did not show any clinical signs. This may also explain the differences in these strains’ resistance to that of isolates from the other two tracts.
Conclusion
Our results imply that the high prevalence of MDR isolates is an alarming threat, requiring immediate action to prevent the spread of G. anatis isolates resistant to antibiotics used in human medicine such as vancomycin and colistin. In light of our results as well as reports from other poultry researchers around the world, it seems very important to prevent G. anatis infection more effectively.
, 
