86
Use of Restriction Endonuclease Analysis of Plasmids and Pulsed-Field Gel
Electrophoresis to Investigate Outbreaks of Methicillin-Resistant Staphylococcus
aureus Infection
Peter Yuk-Fong Liu, Zhi-Yuan Shi, Yeu-Jun Lau,
Bor-Shen Hu, Jainn-Ming Shyr, Wen-Shih Tsai,
Yu-Hui Lin, and Ching-Yu Tseng
From the Section of Infectious Diseases, Taichung Veterans General
Hospital, Taichung, Taiwan, Republic of China
We used restriction endonuclease analysis of plasmids (REAP) and pulsed-field gel electrophoresis
(PFGE) to investigate c1usterings of methicillin-resistant Staphylococcus aureus (MRSA) infections
in our orthopedic unit, neurosurgery unit, intensive care unit, and burn unit. Fourteen different
strain types were identified by REAP and 10 different strain types were identified by PFGE among
2S MRSA isolates collected during these incidents of infection. Though neither technique was clearly
superior to the other for typing MRSA isolates, REAP is recommended as a relatively simple and
reproducible technique for the preliminary investigation of MRSA infection outbreaks in clinical
settings.
There has been a steady increase in the incidence of nosocomial infections caused by methicillin-resistant Staphylococcus
aureus (MRSA) in recent years. Reducing the number of
MRSA infections by detecting and eradicating the sources of
the organisms or by interrupting their path of transmission to
the patient is an important goal and requires the ability to
distinguish individual strains. The recent development of molecular genetic techniques has reduced the dependence on detecting phenotypes for epidemiological studies. Restriction endonuclease analysis of whole plasmid DNA (REAP) has been
shown to be superior to bacteriophage typing and other traditional typing techniques [1-5]. Zuccarelli et al. [5] identified
a rich diversity of plasmids among MRSA isolates with their
finding of37 different profiles among 120 independent isolates,
as shown by EcoR! digestion of plasmid DNA. Plasmid digestion profiles of MRSA organisms isolated repeatedly from nine
of 10 patients were also stable over a period of up to 3 months.
Recently, a new molecular technique, pulsed-field gel electrophoresis (PFGE) of genomic DNA, has been widely used
for typing a variety of microorganisms and has been found to
have high discriminatory power and reproducibility [6]. This
technique has been applied to the investigation of the epidemiology of MRSA [7-11]. Its discriminatory power is higher
than that of ribotyping and equivalent to or higher than that of
random amplified polymorphic DNA assay [7, 9, 11]. Four
clusterings of MRSA infections at Taichung Veterans General
Received 17 April 1995; revised 6 September 1995.
Financial support: This work was supported in part by a research grant from
Taichung Veterans General Hospital, Taichung, Taiwan, R.O.C.
Reprints or correspondence: Dr. Peter Yuk-Fong Liu, Section of Infectious
Diseases, Taichung Veterans General Hospital, 160 Taichung Harbour Road,
Section 3, Taichung, Taiwan, R.O.C.
Clinical Infectious Diseases 1996;22:86-90
© 1996 by The University of Chicago. All rights reserved.
1058-4838/96/2201-0015$02.00
Hospital, a 1,200-bed teaching hospital, allowed us the opportunity to compare the use of REAP and PFGE in characterizing
the isolates that caused the infections.
Materials and Methods
Background. From September to November 1994, we observed the sudden emergence of MRSA isolated from patients
hospitalized in the orthopedic unit, neurosurgery unit, intensive
care unit, and burn unit. Since the observed rate of clustering
cases exceeded the predicted 95% confidence interval [12], an
outbreak investigation was instituted. Hospital charts of all
patients infected with MRSA were reviewed, and cases involving an onset ~ 72 hours after admission were considered to
be of nosocomial origin [13]. In addition, surveillance culture
specimens were collected from the anterior nares of hospital
personnel who cared for these patients. Outbreak control measures instituted by the infection control team included the following: reinforcing handwashing protocols, isolating infected
or colonized patients in the same room, and treating colonized
personnel with a regimen of topical mupirocin for 7 days. With
the institution of these measures, only sporadic cases were seen
in these units during the following 3 months.
Bacterial isolates. The S. aureus isolates obtained were
screened for methicillin resistance by the standard disk diffusion method, with use of Mueller-Hinton agar and a 1-J-tg oxacillin disk after incubation for 24 hours at 35°C [13]. MRSA
isolates from 25 patients and staff were identified and formed
the basis for this study. All isolates were frozen at -70°C until
typed. The stability of the REAP type of each isolate was
checked by the performance of analysis within 1 week of collection and at the end of the study (following up to 4 months
of storage on agar slants) after five subcultures.
Preparation ofplasmid DNA. Plasmid DNA was prepared
by a modification of the alkaline lysis method of Bimboim
and Doly [14], including extra steps to remove contaminating
cm
Typing of MRSA by REAP and PFGE
1996;22 (January)
87
Table 1. Characteristics of methicillin-resistant S. aureus isolated in four incidents of clustering of infections at a hospital in Taiwan in 1994.
Isolate no.
Incident 1 (orthopedic unit)
Sl
Source/specimen
Date of
isolation
(mo/d)
Patient l/blood
10/14
S2
Patient 2/blood
10/11
S3
Patient 3/blood
9/21
Clinical significance
Surgical wound
sepsis
Surgical wound
sepsis
Surgical wound
sepsis
Surgical wound
Surgical wound
Peritonitis
infection with
A
2
B
infection with
3
B
4
1
No plasmid
C
A
B
Central venous catheter
infection with bacteremia
Surgical wound infection
Pneumonia
4
B
5
6
A
D
7
1
E
A
8
I
A
Patient 4/wound swab
Patient 5/wound swab
Patient 6/ascites
9/29
11/1
10/7
Patient 7/blood
11/15
S8
S9
Incident 3 (intensive care unit)
S10
Sl1
Patient 8/wound swab
Patient 91sputum
11/10
11/16
Patient 10/blood
Patient 11/blood
11/15
11/16
Patient 12/wound swab
Patient 13/blood
11/16
11/15
Patient 14/blood
11/16
Pneumonia with sepsis
Central venous catheter
infection with bacteremia
Surgical wound infection
Surgical wound infection with
sepsis
Pneumonia with sepsis
Patient 15/wound swab
Patient 16/wound swab
Patient 17/wound swab
Patient 18/wound swab
Patient 19/urine
Patient 20/wound swab
Nurse 1/nasal swab
Nurse 2/nasal swab
Nurse 3/nasal swab
Doctor I/nasal swab
Doctor 2/nasal swab
11/9
11/15
11/25
11/25
11/21
11/25
12/1
12/1
12/1
12/1
12/1
Bum wound infection
Bum wound infection
Bum wound infection
Bum wound infection
Urinary tract infection
Bum wound infection
Colonization
Colonization
Colonization
Colonization
Colonization
S14
Incident 4 (bum unit)
S15
S16
S17
SI8
S19
S20
S21
S22
S23
S24
S25
PFGE pattern
(digestion with Sma I)t
infection with
S4
S5
S6
Incident 2 (neurosurgery unit)
S7
S12
S13
REAP DNA profile
(with use of EcoRl)*
infection
infection
F
A
1
1
1
9
9
10
9
11
12
13
14
A
A
A
A
A
A
A
G
H
I
J
* REAP
t
= restriction endonuclease analysis of plasmids; pattern was determined by digestion with EcoRl.
PFGE = pulsed-field gel electrophoresis; pattern was determined by digestion with Sma 1.
chromosomal DNA [15]. Stated briefly, the method involved
the following steps. Cells were grown overnight on a blood
agar plate at 37°C. A rice grain-size pellet of growth was
suspended in 100 J-LL of 25% sucrose-TE buffer (l0 mM Tris
and 1 mM disodium EDTA [pH, 8.0]) containing lysozyme
(0.5 mg/mL; Sigma Chemical, St. Louis) and lysostaphin (50
J-Lg/mL; Sigma) and was incubated for 30 minutes at 37°C. To
this, 200 J-LL of 0.2 M NaOH-I % (wtIvol) sodium dodecyl
sulphate was added. After 1 hour at 56°C, 150 J-LL of 3 M
potassium acetate (pH, 4.8) was added and the lysate was
incubated for 10 minutes on ice. The mixture was centrifuged
at 13,OOOg for 5 minutes. After an extraction with 400 J.lL of
phenol-chloroform (1: 1), plasmid DNA was precipitated with
3 volumes of cold ethanol. The resulting precipitate was dissolved in 30 J-LL of EcoRI reaction buffer (GIBCO-BRL, Life
Technologies, Gaithersburg, MD).
REAP. For restriction enzyme analysis of plasmid DNA,
20 U of EcoRI and I J-LL of RNase A (1 mg/ml.; Sigma) were
added to the above plasmid preparation, and the mixture was
incubated at 37°C overnight. Digested samples and a HindIII
digest of lambda phage DNA (GIBCO-BRL, Life Technologies) were electrophoresed on a 0.8% (wt/vol) agarose gel
(SeaKem GTG agarose; FMC Bioproducts, Rockland, ME) in
0.5 X TBE buffer (90 mM Tris, 90 mM boric acid, and 2 mM
EDTA [pH, 8.3]) at 90 V for 3 hours. The gel was stained
with ethidium bromide and photographed with ultraviolet light.
The REAP profiles were considered to be identical if they had
an identical number of bands as well as no more than 5%
variance in the molecular size of each compared band [1]. The
difference of any faint band was ignored [15].
PFGE. PFGE was performed with a contour-clamped homogeneous electric field (CHEF-DRII) apparatus from Bio-
88
Liu et al.
Rad Laboratories (Richmond, CA), as described previously
[10, 11]. The bacterial suspension was prepared by the
scratching of bacterial colonies directly from overnight-incubated culture on blood agar and was adjusted to a concentration
of 1 X 109 cfulmL in SE buffer (75 mM NaCl and 25 mM
EDTA [pH, 7.5]) with a VITEK colorimeter (Hach Co., Loveland, CO). This bacterial suspension was then mixed with an
equal volume of 2% low-melting agarose (Bio-Rad Laboratories) and allowed to solidify in a 100-ttL plug mold (BioRad Laboratories). The block was incubated overnight at 37°C
in 2 mL oflysis solution (10 mM Tris-HCI [pH, 7.6]; 100 rnM
EDTA; 100 mM NaCI; 0.5% Brij-58; 0.2% sodium deoxycholate; 0.5% sodium lauryl sarcosine; lysozyme [0.5 mg/mL];
and lysostaphin [50 ttg/mL]). Following this step, the lysis
buffer was replaced by 2 mL of proteolysis buffer (1% sodium
lauryl sarcosine, 0.5 MEDTA [pH, 9.5], and proteinase K [500
ttg/mL; Sigma]), and this solution was incubated for 2 days at
56°C; it underwent gentle shaking and a change of proteolysis
buffer at 24 hours. Lysed bacterial material and proteinase K
activity were then eliminated by three washes for 10 minutes
(at 4°C) in TE buffer (10 mM Tris-HCI [pH, 7.5] and 10
mM EDTA) containing 1 mM phenylmethylsulfonyl fluoride
(Sigma).
Before DNA digestion, the agarose plugs were washed three
times with TE buffer. A slice of each plug (2.5 mm) was cut
and incubated overnight with 20 U of Sma I (GIBCO-BRL,
Life Technologies), with use of the buffers and the reaction
conditions recommended by the manufacturer. DNA was electrophoresed in 1.2% SeaKem GTG agarose (FMC) at 6 V/cm
for 23 hours; the pulse time was increased from 5 to 40 seconds.
A lambda ladder (Bio-Rad Laboratories) was used as the molecular weight marker. The gel was stained for 30 minutes with
ethidium bromide (5 j.lg/mL), washed in distilled water for 3
hours, and photographed under ultraviolet light. The relation
between two given isolates was estimated by calculation of
Dice's coefficient of similarity: 2 X number of matching band!
total number of bands in both strains. Isolates were considered
to be different if the value was <0.90 [16,17].
Results and Discussion
Table 1 depicts the typing results for the 25 isolates. Fourteen
REAP DNA profiles (figure 1) and 10 PFGE patterns (figure
Kb
23.1
9.4
6.6
4.4
em 1996;22
(January)
2) were identified. According to the present analysis, although
there was clustering of cases infected by the same strain of
MRSA in the bum unit, no definite outbreak was noted in
other units. It has been suggested that PFGE assay is the most
discriminative technique in comparison with the other molecular typing methods, such as ribotyping and fingerprinting by
PCR [7-9]. In this study we found that REAP DNA profiling
is also very discriminative and may increase the discrimination
afforded by PFGE. In incident 4, isolates S15-S21 were not
differentiated by PFGE, but three different patterns were identified by REAP. Isolates S15- S17 were found to be identical
by REAP. Their epidemiological relationship was confirmed
by the fact that they were isolated from patients who were
cared for by the same group of nurses. On the other hand,
isolates S18-S20 were isolated from patients who were cared
for by another group of nurses, including nurse 1. It is possible
that nurse 1 carried this second epidemic strain and transmitted
it to patient 18 and patient 19 or vice versa. The REAP type
of isolate S20 was different from that of S18 and S19. As we
reviewed the medical history of the patient who carried this
strain, we found MRSA had been isolated from this patient
before he was admitted to the bum unit. Thus, it is possible
that isolate S20 was epidemiologically unrelated to isolates
S18 and S19.
However, we found that isolates S1 and S5 (isolated from
the orthopedic unit), isolates Sll, S13, and S14 (from the
intensive care unit), and isolates S15, S16, and S17 (from the
bum unit) were identical by REAP and PFGE assay. Such
epidemiological transmission pathways are difficult to trace,
but we found that operations had been performed on patient 5
and patient 13 by the same surgical team who performed an
operation for patient 15. It is possible that cross-infection occurred among these patients and that they carried the epidemic
strain back to their units and were the source of dissemination
in those units. It might be argued that the differences of REAP
patterns among the clinical MRSA isolates may be due to the
instability ofplasmids. However, when we compared the results
of REAP performed within 1 week of collection and those
performed at the end of the study (after up to 4 months of
storage) after five subcultures, the REAP DNA profiles were
exactly the same. Diversity and stability ofREAP DNA profiles
among isolates of MRSA has also been noted by other authors
Figure 1. Agarose gel electrophoresis of plasmid DNA from 25
MRSA isolates after digestion with
EcoRl. Lane M, HindIII restriction
fragments of phage lambda; lanes
1-25 are restriction fragment patterns of EcoRl-digested total plasmid DNA from isolates SI-S25,
respectively.
Typing of MRSA by REAP and PFGE
CID 1996;22 (January)
89
M 1 2 3 4 5 6 7 8 9 10 11 12 I3 14 15 16 17 18 19 20 21 22 23 24 25
Kb
Figure 2. PFGE of Sma I -digestedgenomic DNA from 25clinical isolates of MRSA. Lane M is
a lambda ladder(Bio-Rad), which
served as the molecular size
marker; lanes 1-25 are DNA digests of isolates Sl-S25, respectively.
388.0
291.0
145.5
48.5
[4, 5]. The problem with REAP is that it cannot be applied to
isolates devoid of plasmid DNA, such as isolate S6.
PFGE has been proposed as a good typing technique for
epidemiological studies ofmost species. Profile types generated
by this technique appear stable and reproducible, even after 40
subcultures [10]. Chromosomal stability makes this technique
suitable for the long-term follow-up of epidemic strains of
MRSA. In contrast to other bacterial species, the polymorphism
provided by PFGE is relatively low [8, 9] because of a high
degree of genetic relatedness between MRSA strains [18]. In
the study of Prevost et al. [9], 26 different fingerprints among
239 MRSA isolates from 142 patients were identified by this
technique. However, 51% of the isolates belonged to the same
PFGE type. Discrimination has been considered statistically
adequate only when the most common type occurs in < 5% of
isolates [19, 20]. To increase the discriminatory power of this
technique for typing MRSA, combining it with other typing
methods, such as REAP, is helpful [10, 17, 20-22].
As shown in this study, the combination ofPFGE and REAP
can provide a greater discriminatory capacity between MRSA
isolates. In contrast with the finding of Prevost et al. [10], the
discriminatory power of REAP in this study seems to be superior to that ofPFGE. However, the number of isolates included
in this study is small, and we agree with Tenover et al. [21],
who stated that no single technique was clearly superior to
others for typing MRSA isolates and that a combination of
different typing techniques is sometimes necessary.
In conclusion, although plasmids are mobile and unstable
genetic elements, REAP is still useful for delineation of the
relationships of MRSA isolates during the acute outbreak stage.
Although we cannot prove that the discriminatory power of
REAP is clearly superior to that of PFGE, the simplicity and
speed of this technique make it well-suited for the preliminary
study of MRSA infection outbreaks in clinical settings.
Acknowledgments
The authors thank: Meei-Fang Liu and Su-Fen Lee for their
technicalassistanceand Hsin-JyurRehn, Meei-Rumg Liu, and YuMei Huang for their help in investigating the outbreak.
References
1. Hartstein AI, Morthland VH, Eng S, Archer GL, Schoenknecht FD, Rashad
AL. Restriction enzyme analysis of plasmid DNA and bacteriophage
typing of paired Staphylococcus aureus blood culture isolates. J Clin
Microbiol 1989;27:1874-9.
2. Licitra CM, Brooks RG, Terry PM, Shaw KJ, Hare RS. Use of plasmid
analysis and determination of aminoglycoside-modifying enzymes to
characterize isolates from an outbreak of methicillin-resistant Staphylococcus aureus. J Clin Microbiol 1989;27:2535-8.
3. Rhinehart E, Shlaes DM, Keys TF, et al. Nosocomial clonal dissemination
of methicillin-resistant Staphylococcus aureus: elucidation by plasmid
analysis. Arch Intern Med 1987; 147:521-4.
4. Trilla A, Nettleman MD, Hollis RJ, Fredrickson M, Wenzel RP, Pfaller
MA. Restriction endonuclease analysis of plasmid DNA from methicillin-resistant Staphylococcus aureus: clinical application over a threeyear period. Infect Control Hosp Epidemiol 1993; 14:29-35.
5. Zuccarelli AJ, Roy I, Harding GP, Couperus J1. Diversity and stability of
restriction enzyme profiles of plasmid DNA from methicillin-resistant
Staphylococcus aureus. J Clin Microbiol 1990;28:97-102.
6. Goering RY. Molecular epidemiology of nosocomial infection: analysis
of chromosomal restriction fragment patterns by pulsed-field gel electrophoresis. Infect Control Hosp EpidemioI1993; 14:595-600.
7. van Belkum A, Bax R, Prevost G. Comparison offour genotyping assays
for epidemiological study of methicillin-resistant Staphylococcus
aureus. Eur J Clin Microbiol Infect Dis 1994; 13:420-4.
8. Ichiyama S, Ohta M, Shimokata K, Kato N, Takeuchi 1. Genomic DNA
fingerprinting by pulsed-field gel electrophoresis as an epidemiological
marker for the study of nosocomial infections caused by methicillinresistant Staphylococcus aureus. J Clin Microbioll991;29:2690-5.
9. Prevost G, Jaulhac B, Piemont Y. DNA fingerprinting by pulsed-field gel
electrophoresis is more effective than ribotyping in distinguishing
among methicillin-resistant Staphylococcus aureus isolates. J Clin Microbiol 1992; 30:967 - 73.
90
Liu et al.
10. Prevost G, Pottecher B, Dahlet M, Bientz M, Mantz JM, Piemont Y. Pulsed
field gel electrophoresis as a new epidemiological tool for monitoring
methicillin-resistant Staphylococcus aureus in an intensive care unit. J
Hosp Infect 1991; 17:255-69.
11. Saulnier P, Bourneix C, Prevost G, Andremont A. Random amplified
polymorphic DNA assay is less discriminant than pulsed-field gel electrophoresis for typing strains of methicillin-resistant Staphylococcus
aureus. J Clin Microbioll993;31:982-5.
12. Doebbeling BN. Epidemics: identification and management. In: Wenzel
RP, ed. Prevention and control of nosocomial infections. 2nd ed. Baltimore: Williams & Wilkins, 1993:177-205.
13. National Committee for Clinical Laboratory Standards. Performance standards for antimicrobial disk susceptibility tests. 5th ed. Villanova, Pennsylvania: National Committee for Clinical Laboratory Standards, 1993.
14. Birnboim HC, Doly 1. A rapid alkaline extraction procedure for screening
recombinant plasmid DNA. Nucleic Acid Res 1979;7:1513-23.
15. Woodford N, Johnson AP, Threlfall EJ. Extraction and fingerprinting of
bacterial plasmids. In: Chart H, ed. Methods in practical laboratory
bacteriology. Boca Raton, Florida: CRC Press, 1994; 93-105.
16. Dice LR. Measures of the amount of ecological association between species. Ecology 1945;26:297-302.
CID 1996;22 (January)
17. Branchini MLM, Morthland VH, Tresoldi AT, von Nowakonsky A, Dias
MBS, Pfaller MA. Application of genomic DNA subtyping by pulsed
field gel electrophoresis and restriction enzyme analysis of plasmid
DNA to characterize methicillin-resistant Staphylococcus aureus from
two nosocomial outbreaks. Diagn Microbiol Infect Dis 1993; 17:
275-81.
18. Kreiswirth B, Kornblum J, Arbeit RD, et al. Evidence for a clonal origin
of methicillin resistance in Staphylococcus aureus. Science 1993;
259:227-30.
19. Hunter PRo Reproducibility and indices of discriminatory power of microbial typing methods. J Clin MicrobioI1990;28:1903-5.
20. Mulligan ME, Arbeit RD. Epidemiologic and clinical utility of typing
systems for differentiating among strains ofmethicillin-resistant Staphylococcus aureus. Infect Control Hosp Epidemio11991; 12:20-8.
21. Tenover FC, Arbeit R, Archer G, et al. Comparison of traditional and
molecular methods of typing isolates of Staphylococcus aureus. J Clin
Microbioll994;32:407-15.
22. Sabria-Leal M, Morthland VH, Pedro-Botet ML, et al. Molecular epidemiology for local outbreaks of methicillin-resistant Staphylococcus aureus
(MRSA). The need for several methods. Eur J Epidemiol 1994; 10:
325-30.