Journal of Antimicrobial Chemotherapy (1983) 11, 263-269

Gentamicin resistance in Staphylococcus aureus—a new mechanism?

R. R. Cutler*

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014

Department ofMedical Microbiology, The London Hospital Medical College, Turner Street,
London El 2 AD. England

Six multiply-resistant strains of Staphylococcus aureus were cured of their genta-

micin resistance. For three strains, the loss or reduction of aminoglycoside-
modifying enzymes in the sensitive variants was accompanied by an increase in
intracellular gentamicin accumulation, confirming that modification of the anti-
biotic was associated with its impaired uptake.
Resistance in the other three strains, however, did not show this association.
For one strain the loss of gentamicin resistance was associated with a loss of
enzyme production, but both resistant and sensitive variants accumulated PH]
gentamicin at the same rate. The sensitive variants of the two remaining strains
showed no increase in PH] gentamicin accumulation but still produced
aminoglycoside-modifying enzymes, and it is suggested that the broad-spectrum
aminoglycoside resistance of these two strains is due to some hitherto undescribed
mechanisms of resistance rather than an alteration of the ribosomal target.

Introduction
Aminoglycoside resistance of Staphylococcus aureus usually involves plasmid-
mediated enzymic modification of the antibiotic (Le Goffic el al, 1977) or the
impaired accumulation of the aminoglycoside (Hancock, 1981).
Most studies of the mechanisms of aminoglycoside resistance of staphylococci
have emphasised the production of aminoglycoside-modifying enzymes. There have
been few studies of other mechanisms of resistance (Courvalin & Carlier, 1981), even
for the increasingly common multiply-resistant strains.
We therefore measured the uptake of gentamicin and the production of
aminoglycoside-modifying enzymes by six gentamicin-resistant strains of Slaph.
aureus and by their six sensitive variants that had been cured of gentamicin
resistance.

Materials and Methods

Strains and MIC determinations
Six multiply-resistant gentamicin-resistant strains of Staph. aureus were selected
•Present address: Newham District Microbiology Laboratories, St. Andrew's Hospital, Devons Road,
Bow, London E3 3NT, England.
263
0305-7453/83/030263+07 $02.00/0 © 1983 The British Society for Antimicrobial Chemotherapy
264 R. R. Cutler
from 30 isolates from the wound infections of patients at The London Hospital.
Minimum inhibitory concentrations (MICs) of gentamicin, amikacin, kanamycin,
streptomycin and tobramycin were determined by agar dilution. Overnight nutrient-
broth cultures were diluted so that, with a multipoint inoculator, inocula of 104
organisms were applied to DST agar (Oxoid CM 261) plates that contained doubling
dilutions of aminoglycoside. MICs of gentamicin of 8 mg/1 or more were considered
to indicate resistance.
Curing
Gentamicin-resistant strains were cured of antibiotic resistances by treatment with
acridine orange at 42°C. Cured strains, susceptible to gentamicin, were checked for
purity, phage type, coagulase and deoxyribonuclease titres, and for growth rate by

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014

methods that have been described previously (Cutler, 1979).
Enzyme detection
A modification of the method of Benveniste & Davies (1971), as described previously
(Cutler, Farrell & Drasar, 1980), was used to detect the inactivation of gentamicin
by enzymic degradation. Staphylococci were first lysed with lysostaphin (Sigma
Chemicals, U.S.A.) to obtain a cell-free extract of aminoglycoside-modifying
enzyme. The gentamicin was then inactivated in a buffered system consisting of the
cell-free extract and a radioactively labelled compound which provides the group
to be transferred to the gentamicin. The production of aminoglycoside-modifying
enzymes was expressed as the ratio of counts obtained with and without substrate.
Measurement of the inlracellular accumulation ofpHJ gentamicin
Each resistant strain and its sensitive variant were cultured for 18 h at 37°C nutrient
broth (Southern Group). 0 1 ml of this culture was diluted in 90 ml of nutrient broth
(warmed to 37°Q and incubated for a further 3 h by which time the cells were in
the log phase. pH] Gentamicin (Radiochemical Centre, Amersham) with a final
activity of 1-6 uCi/ml was mixed with a solution of gentamicin in nutrient broth to
give 100 ml of broth containing 4 mg/I gentamicin.
Two 10 ml aliquotsof each suspension were removed at 0,5,10,15,20 and 30 min
after the addition of pH] gentamicin. One aliquot was used to determine the number
of colony forming units (cfu) per ml. The second aliquot was washed through a glass-
wool column, then passed through a 2 urn millipore filter (Millipore S.A.) that had
been pre-treated with 2000 mg/1 gentamicin, washed three times, dried, and the PH]
gentamicin content of the filter estimated in a liquid scintillation counter (Intertech-
nique, SL 30). Results were expressed as nmoles of PH] gentamicin per 106 cfu.

Results
Aminoglycoside resistance
The MICs of gentamicin for the resistant strains and their sensitive variants are
shown in Table I. Each of the six gentamicin-resistant strains was also resistant to
at least two other aminoglycosides (Table II).
Curing
The six gentamicin-sensitive variants belonged to the same phage group, and gave
the same coagulase, deoxyribonuclease production and growth rates as their resistant
counterparts. All strains belonged to phage group III.
 Gentamicin resistance in Staphylococcus aureus 265

Table I. The gentamicin susceptibility, gentamicin accumulation, and

production of phosphotransferase for six gentamicin-resistant strains of
staphylococci and their sensitive variants

MIC of
gentamicin Gentamicin Phospho-transferase
(mg/1) accumulation* activity"1"
Strain resistant cured resistant cured resistant cured

A 128 006 1 125 92 1

B 64 0-5 44 244 87 1
C 16 2 1 270 60 38
D 32 0125 96 113 25 1

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014

E 32 1 1 1 28 25
F 16 2 1 1 84 118
nmol labelled gentamicin/10* cfu at 30 min
nmol labelled gentamicin/106 cfu at 0 min
counts/min with substrate
counts/min without substrate

Table II. Sensitivity of six gentamicin-resistant staphylococci to five other

aminoglycosides

MIC (mg/1)
Strain Gentamicin Amikacin Kanamycin Tobramycin Streptomycin

A 128 16 >512 128 4

B 64 16 >512 64 4
C 16 4 128 8 4
D 32 16 >512 32 64
E 32 16 >512 32 16
F 16 8 256 16 >512

Aminoglycoside-modifying enzymes
There was no marked change in the low levels of acetyltransferase enzymes produced
by any of the six strains or their variants nor were any aminoglycoside-nucleotyl-
transferases produced. Table I shows the relative production of aminoglycoside-
phosphotransferase enzymes.

Uptake ofpHJ gentamicin

The six strains were divided into three groups according to their accumulation of
pH] gentamicin. The sensitive variants of the first group, consisting of strains A, B
and C, accumulated more gentamicin than did the corresponding resistant strains
(Table I, Figure 1). The second group consisted of only strain D, and both the resist-
ant strain and its sensitive variant accumulated PH] gentamicin at the same rate
266 R. R. Cutler

300

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014

K> 20 30
Time (mm)
Figure 1. Accumulation of PH] gentamicin by gentamicin-resistant Staph. aureus strains A (•), B (•)
and C (A) and their gentamicin-sensitive variants (O, 0 , A).

200 -

§ 100 -

10 20
Time (min)
Figure 2. Accumulation of PH] gentamicin by gentamicin-resistant Staph. aureus strain D (•) and its
gentamicin-sensitive variant (D).
 Gentamicin resistance in Staphylococcus aureus 267

100

^o^^^ _-o

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014

0 10 20 30
Time (mm)
Figure 3. Accumulation of PH] gentamicin by gentamicin-resistant Staph. aureus strains E (•) and F
(•) and their gentamicin-sensitive variants (0,V).

(Table I, Figure 2). In the third group neither the resistant strains E and F nor their
sensitive variants showed any increase in the accumulation of PH] gentamicin (Table
I, Figure 3).

Discussion
Bacteria can resist aminoglycosides by alteration of the cell's permeability to the
antibiotic, by enzymic modification of the antibiotic (Bryan et at., 1976; Courvalin
& Carlier, 1981) or, less commonly, by alteration of the ribosomal target. However,
there is controversy about the contribution of these mechanisms. Davies & Smith
(1978) pointed out that little is known of how aminoglycoside-modirying enzymes
confer resistance and they suggested that enzymically-modified aminoglycoside
within the cell's periplasmic space blocks any subsequent transport of gentamicin
across the inner cell membrane. In contrast, others (for example, Kallings, 1980;
Shannon & Phillips, 1982) have suggested that enzymic activity, although possibly
decreasing transport, does not necessarily stop it entirely. They further suggest that,
since the ribosomes of some resistant enzyme-producing strains are still susceptible
to binding by unmodified aminoglycoside, and modified aminoglycoside is ineffec-
tive in inhibiting protein synthesis by ribosomes in vitro, enzymes must cause resist-
ance primarily by detoxification of the aminoglycoside.
Our results suggest three mechanisms of gentamicin resistance for the six strains
studied here. The sensitive variants of strains A, B and C showed little change in
acetyltransferase activity, but showed a loss of reduction in their ability to produce
aminoglycoside phosphotransferases, and the accumulation of pH] gentamicin was
significantly higher than for their corresponding resistant strains. The PH] gentami-
cin accumulation by these sensitive variants showed an initial lag phase and a sub-
sequent exponential phase, a phenomenon that has been described by other workers
(Hancock, 1981). For these three strains impaired uptake of gentamicin, whether or
not caused by the antibiotic's modification within the periplasmic space, probably
accounts for their resistance. Strains D,E and F did not, however, fit this well-
described pattern.
268 R. R. Cutler

The sensitive variant of strain D showed a loss of modifying-enzyme production,

but no change in pH] gentamicin accumulation. The rate of accumulation was expo-
nential for both resistant strain D and its sensitive variant. Thus the resistance of
strain D may be due to enzymic detoxification of the gentamicin.
The mechanism of resistance for strains E and F appears to be even more unusual.
Both these strains and their sensitive variants produced modifying enzymes but there
was no increase in PH] gentamicin accumulation, even in the sensitive variants. But
the gentamicin must have reached the ribosomal target in order to kill the sensitive
variants which may have become 'super-sensitive' to small amounts of gentamicin.
Since the sensitive variants of strains E and F still produced modifying enzymes, it
is possible that for these organisms modification of the gentamicin does not lead to

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014

its detoxification. Nor does it seem that the mechanism of resistance for strains E
and F is ribosomal. Ribosomal resistance for gentamicin and aminoglycosides other
than streptomycin has not been found in clinical isolates (Shannon & Phillips,
1982). In addition, ribosomal resistance does not confer cross-resistance to other
aminoglycosides (Davies, 1971), but when strains E and F were cured of gentamicin
resistance they were also cured of tobramycin and kanamycin resistance. We there-
fore conclude that some hitherto undescribed mechanism of resistance must be
involved for strains E and F and we intend to investigate these strains for in-vitro
ribosomal binding properties.

Acknowledgement
I am grateful to Dr M. W. Casewell for help with the preparation of the manuscript.

References
Benveniste, R. & Davies, J. (1971). R-factor-mediated gentamicin resistance: a new enzyme
which modifies aminoglycoside antibiotics. FEBS Letters 14, 293-6.
Bryan, L. E., Haraphongse, R. S. & von den Elzen, H. M. (1976). Gentamicin resistance in
clinical isolates of Pseudomonas aeruginosa associated with diminished gentamicin
accumulation and no detectable enzymatic modification. Journal of Antibiotics 29,
743-53.
Courvalin, P. & Carlier, C. (1981). Resistance towards aminoglycoside-aminocyclitol anti-
biotics in bacteria. Journal of Antimicrobial Chemotherapy 8, Suppl. A. 57-69.
Cutler, R. R. (1979). Relationship between antibiotic resistance, the production of'virulence
factors', and virulence for experimental animals in Staphylococcus aureus. Journal of
Medical Microbiology 12, 55-62.
Cutler, R. R., Farrell, W. & Drasar, F. A. (1980). Antibacterial activity and enzyme stability
of UK 31214, a Kanamycin B derivative. Proceedings of the 1 lth International Congress
of Antibiotic and Antimicrobial Chemotherapy, American Society of Microbiology, pp.
383-5.
Davies, J. (1971). Bacterial resistance to aminoglycoside antibiotics. Journal of Infectious Dis-
eases 124, SuppL I. S7-S10.
Davies, J. & Smith, D. I. (1978). Plasmid-determined resistance to antimicrobial agents.
A nnual Reviews of Microbiology 32, 469-518.
Hancock, R. E. (1981). Aminoglycoside uptake and mode of action, with special reference
to streptomycin and gentamicin. I. Antagonists and mutants. Journal of Antimicrobial
Chemotherapy^, 249-76.
Kallings, L. O. (1980). Resistance factors: Influence of netilmicin activity. Scandinavian
Journal of Infectious Diseases, Suppl. 23, 54-8.
 Gentamicin resistance in Staphylococcus aureus 269

Le Goffic, F., Martel, A., Moreau, N., Capman, M. L, Soussy, C. J. & Duval, J. (1977).
2 ' -O-phosphorylation of gentamicin components by a Staphylococcus aureus strain
carrying a plasmid. Antimicrobial Agents and Chemotherapy 12, 26-30.
Shannon, K. & Phillips, I. (1982). Mechanisms of resistance to aminoglycosides in clinical
isolates. Journal of Antimicrobial Chemotherapy 9, 91-102.

(Man uscript accepted 17 A ugust 1982)

Downloaded from http://jac.oxfordjournals.org/ at York University Libraries on June 7, 2014