Development of Antibiotic Resistance Amongst Pseudomonas Isolates From Veterinary Clinical Specimens
David H. Lloyd
Department of Veterinary Clinical Sciences, Royal Veterinary
College, University of London, North Mymms, Hertfordshire AL9 7TA
Introduction
Pseudomonas spp. Are ubiquitous organisms which occur in water, soil, and decaying
organic matter. They are able to colonise many clinical environments and components,
including disinfectant solutions. They tend to persist in hospitals where an
exchange can occur between patients and environmental habitats. Normal individuals
are generally resistant to infection but immunocompromised patients, particularly
those treated with antibiotics to which the pseudomonads are resistant, are
susceptible. Such treatment promotes infection with resistant strains of Pseudomonas
(1,2,3) that are more virulent and may become established in hospital or farm
premises. It can be difficult to differentiate these from the multitude of strains
normally present in the environment.
Mechanisms of Resistance
Ps. aeruginosa depends for its resistance to antimicrobials on four broad mechanisms (4)
1. Its cell wall has low permeability
2. It has a large genome with the capacity to express a wide range of resistance
mechanisms. The Ps. aeruginosa genome contains about 6.26 Mbp within 5567 genes
and is thus substantially larger than that of Escherichia coli (4.64 Mbp, 4279
genes) and of Staphylococcus aureus (2.81 Mbp, 2594 genes) (4).
3. Chromosomal mutations can lead to changes in the regulation of resistance
genes.
4. It can acquire resistance genes from other organisms via plasmids, transposons
and bacteriophages.
Amongst clinical isolates a correlation has been shown between resistance to antibiotics and biocides indicating that their use can promote resistance to either class of antimicrobials (5).
Cell wall permeability
Aminoglycosides, quinolones, beta-lactams and polymixins require crossing the bacterial cell wall to reach their targets. Penetration can be reduced by the development of an antibiotic-resistant biofilm with secretion of an anionic exopolysaccharide matrix which binds cationic antibiotics (6); quorum sensing in such biofilms may also be involved in changing bacterial metabolism and reducing sensitivity to antimicrobials. The outer membrane also limits penetration by small hydrophilic antibiotics, which must pass through aqueous channels in porin molecules, and excludes large molecules. Amino glycosides and colistin promote their absorption through the cell wall by binding to the superficial lipopolysaccharide, allowing penetration and then active transport via the cytoplasmic membrane. Resistance to these substances can be associated with over expression of outer membrane protein, which reduces binding of the lipopolysaccharide.
Efflux pumps
Those antibiotic molecules, which pass through the cell wall, may then be removed by efflux pumps. Four different efflux systems dependent on the genes mexAB-oprM (beta lactams and biocides), mexXY-oprM (aminoglycosides), mexCD-oprJ and mexEF-oprN (carbapenems and quinolones) (7) are known to exist allowing extrusion of all classes of antibiotics except the polymixins. Genes for these efflux systems are found in all strains of Ps. aeruginosa but are expressed at relatively low levels, under the control of regulatory genes. Mutations in these regulators can lead to high level expression and confer enhanced antibiotic or biocide resistance.
Inactivation of antibiotics
Beta-lactam antibiotics are inactivated by beta-lactamases. These enzymes are inactivated by inhibitors such as clavulanic acid and tazobactam. However, not all pseudomonad beta-lactamases are affected (8) and inhibitor-resistant forms have now been described (9).
Changes in target enzymes
Mutations, which change the structure of enzymes so that they retain their activity but are not affected by antibiotics, are known to be involved in resistance to fluoroquinolones (e.g mutation of gyrA). Changes in the structure of the penicillin binding proteins can provide resistance to beta-lactams and changes in ribosome structure to streptomycin.
DEVELOPMENT OF RESISTANCE
Surveys allowing assessment of antibiotic sensitivity trends amongst Pseudomonas spp. Have been reported from small animal practice both in America and in Europe.
Prescott et al. (10), reporting from the Veterinary Teaching Hospital at Guelph, examined canine isolates between 1984 and 1998 and found marginal increases in resistance for Ps. aeruginosa from urinary tract infections. Petersen et al. (11) examined submissions to the Michigan State University Animal Health Diagnostic Laboratory of specimens from canine skin and ears over the period 1992-1997 inclusive but found no consistent trends.
In Europe, Normand et al. (12,13) reporting from Glasgow, examined canine and feline bacterial isolates from veterinary community practice and from a small animal hospital between 1989 and 1997. They showed a rising trend in multidrug resistance in Pseudomonas spp. from community practice but not from the veterinary hospital.
Lloyd et al. (14,15) examined sensitivity of Ps. aeruginosa from canine infections (predominantly skin and ears) to enrofloxacin and marbofloxacin and contrasted frequencies in 1992, 1995 and 2003 using both disc and broth macrodilution tube assay techniques. In 1992, 79% of 19 isolates showed disc sensitivity to enrofloxacin whereas in 1995 45% of 49 isolates were sensitive; 90%of isolates were sensitive to marbofloxacin. In 2003, of 41 canine isolates, 14.6% were sensitive to enrofloxacin and 85% were sensitive to marbofloxacin (chi-squared test, p<0.001). These data indicate a substantial decrease in sensitivity to enrofloxacin since the antibiotic became available in the UK.
Low levels of sensitivity to enrofloxacin were also reported in Spain by Martin et al. (16) who tested 23 isolates of Ps. aeruginosa from canine chronic otitis externa and found 52% sensitive to enrofloxacin whereas 91% were sensitive to marbofloxacin. Seol et al. (17) reported on 183 isolates of Ps. aeruginosa from dogs visiting the Veterinary Faculty at the University of Zagreb during the years from 1993 to 2000. The great majority of isolates were from the skin and ears. Almost all of the strains (93%) were sensitive to ciprofloxacin and marbofloxacin whilst 71% were sensitive to enrofloxacin. The authors argued that reduced sensitivity might be due to extensive veterinary use of enrofloxacin whereas marbofloxacin was not available and ciprofloxacin was used very rarely in veterinary practice in Croatia.
EVALUATION AND SIGNIFICANCE OF SENSITIVITY DATA
It is difficult to compare the different studies of antibiotic resistance in Pseudomonas owing to differences between sources of isolates and the sampling and sensitivity tests used. For instance, there is evidence that isolates from the middle ear and those from the horizontal ear canal of dogs with otitis media differ in sensitivity (18). In addition, disc sensitivity tests do not necessarily correlate with dilution methods. However, there would seem to be a convincing trend towards increasing resistance to enrofloxacin amongst isolates from chronic otitis. Tejedor et al. (19) have demonstrated that this resistance is associated with mutation of the gyrA gene coupled with up regulation of efflux pump activity.
The significance of elevated levels of resistance to antibiotic e.g. as indicated by the disc sensitivity test, is that empirical selection of antibiotic for systemic therapy in chronic otitis is unwise. However, where topical therapy is planned, sensitivities assessed with systemic treatment in mind, e.g. disc sensitivity tests, are misleading as much higher levels of antibiotic can be achieved and these will commonly exceed the resistance levels in Ps. aeruginosa are rising is a warning that needs to be heeded. We need to develop rational policies of antibiotic prescription and use which will help to reduce the selection of resistant strains (20,21).
Evidence that resistance to biocides may be correlated with antibiotic resistance in clinical isolates of Ps. aeruginosa (Lambert et al., 2001) and recent data showing that methods used for assessment of antimicrobial synergy between EDTA and antibiotics have been flawed (21) show that we still have a lot to learn about handling these very interesting organisms.
REFERENCES AND FURTHER READING
8909 Iverleigh Court 
Potomac, Maryland 20854
301-983-8387 800-755-4738 Fax 301-365-0191
E-mail dermapet@aol.com