30
Molecular Analysis Provides Evidence for the Endogenous Origin of Bacteremia and Meningitis Due to Enterobacter cloacae in an Infant N. Lambert-Zechovsky, E. Bingen, E. Denamur, N. Brahimi, P. Brun, H. Mathieu, and J. Elion
From the Laboratoire de Microbiologie. Laboratoire de Biochimie Genetique, and Service de Nephrologie, Hiipital Robert Debre, Paris. France
Despite the improved care given to newborns, septicemia remains a serious problem that is associated with a significant mortality rate, especially when complicated by meningitis. The incidence of neonatal septicemia varies from I to 4 per 1,000 live-born infants [I ]. Enterobacter organisms are the fourth most common cause of bacteremia in children [2]. Flynn et al. have reported that most nosocomial infections due to Enterobacter species arise from the patient's own flora, especially intestinal flora [3]. Indeed, organisms from the gastrointestinal tract may translocate to mesenteric lymph nodes or to the bloodstream if bacterial populations reach abnormally high levels in the bowel [4, 5]. However, hard facts are still lacking that definitively demonstrate this mode of bacterial transmission because its study has been somewhat hampered by the absence of a sufficiently discriminating typing system. In this study, we analyzed the restriction fragment length polymorphism (RFLP) of total DNA and of ribosomal DNA (rDNA) regions (ribotyping) to document the occurrence of endogenous, systemic bacteremia and meningitis due to Enterobacter cloacae in a newborn.
Case Report A 2, 150-g male infant was born at 33 weeks of gestation by spontaneous vaginal delivery after an uncomplicated pregnancy. The results ofthe initial examination of the newborn were unremarkable. At 24 days of age, he developed a
Received 26 December 1991; revised 26 February 1992. Reprints or correspondence: Dr. Nicole Lambert-Zechovsky, Laboratoire de Microbiologic, Hopital Robert Debre, 48 Bd Serurier, 75019 Paris. France. Clinical Infectious Diseases
1992;15:30-2
© 1992 by The University of Chicago. All rights reserved.
1058-4838/92/1501-0006$02.00
fever at 38.6°C but had no other clinical symptoms. Cultures of blood and CSF remained sterile. Examination of a gramstained smear of urine revealed numerous gram-negative rods and many polymorphonuclear leukocytes. Culture of urine yielded Escherichia coli (10 6 organisms/ml.). Intravenous therapy with cefotaxime ( 100 mg/[kg . dD and netilmicin (4 mg/[kg· d]) was initiated. The infant became afebrile and did well 2 days after the initiation of treatment. Repeated urine cultures were sterile. Six and 8 days after the beginning of therapy, cultures of stool yielded cefotaxime-resistant E. cloacae at counts of 108 and 109 organisms/g of feces, respectively. The administration of antibiotics was discontinued after 10 days of therapy. Three days later, at 36 days ofage, the infant was lethargic and again became febrile. Two blood cultures and one urine culture yielded E. cloacae that was resistant to cefotaxime. Culture of stool yielded pure growth of the same organism (10 9 organisms/g). CSF contained gram-negative rods and 1,200 nucleated cells/rum! (96% neutrophils, 4% lymphocytes). Other laboratory values for CSF were a glucose level of 11 mg/dL and a protein level of 900 mg/dL. E. cloacae that was resistant to cefotaxime was isolated from the CSF at a count of 106 organisms/ml., On the basis of the results of the antibiotic susceptibility profile, administration of imipenem (60 mg/[kg· dD, ciprofloxacin (20 mg/[kg· dD, and amikacin (15 mg/[kg· dj) was immediately started, resulting in the sterilization of blood, urine, and CSF in 48 hours. The clinical outcome was good, and the infant was discharged after a total of 55 days of hospitalization. A total of 10 isolates of E. cloacae were selected for the study. Five isolates (strains 1-5) were related to the described case. Three of these five isolates were recovered from stool at counts of 108, 109 , and 109 organisms/g of feces on 7 days before, 5 days before, and the day ofthe onset ofbacteremia, respectively; of the other two, one each was isolated from blood and CSF. In addition, four epidemiologically
Downloaded from http://cid.oxfordjournals.org/ at Freie Universitaet Berlin on July 4, 2015
We analyzed the restriction fragment length polymorphism (RFLP) of total DNA and of ribosomal DNA regions (ribotyping) to document the occurrence of endogenous, systemic bacteremia and meningitis due to Enterobacter cloacae in a newborn. Five strains of E. cloacae were isolated from this newborn. Three of these strains were recovered from stool at counts of 108 , 109, and 109 organisms/g offeces, respectively; one strain was isolated from blood; and one strain was isolated from cerebrospinal fluid. In addition, five epidemiologically unrelated strains of E. cloacae were studied for comparison. Our study clearly shows the genetic relatedness of the strains isolated sequentially from cultures of stool, blood, and cerebrospinal fluid. RFLP analysis of total DNA and ribotyping seem particularly well suited to the study of the epidemiology of nosocomial E. cloacae strains.
CID 1992;15 (July)
31
Molecular Analysis of E. cloacae
Discussion Until recently, epidemiological studies of E. cloacae have been based essentially on the study of phenotypic traits such as biochemical profiles, antibiotic resistance, and serological, bacteriocin, and phage typing. However, none of these methods are really satisfactory for the typing of E. cloacae because of insufficient discrimination, poor reproducibility, or low typability [9]. We used the RFLP analysis of total DNA and rONA regions for the study of five E. cloacae strains isolated from an infant who had bacteremia and meningitis. This method allowed us to clearly show the genetic relatedness of the strains isolated sequentially from cultures of stool , blood, and CSF. Strains studied for comparison all had different
2
3
4
5
6
7
8
9
10
11
Kb
- 48.5 - 14.9
-10.6
- 5.6
A
2
3
4
5
6
7
8
9
10
Kb
-10.6
-7.3 - 5.6
- 3.9
B
Figure 1. A. Patterns ofRFLP of total DNA from E. cloacae obtained after BamHI digestion and ethidium bromide staining. Lane t, ATCC 13047; iane 2, strain 1; lane 3, strain 2; lane 4. strain 3; lane 5. strain 4; lane 6. strain 5; lane 7. strain 6; lane 8. strain 7; lane 9. strain 8; lane i 0, strain 9; and lane l l , size markers (Kb = kilobases). Strains 1 to 5 (lanes 2 to 6) show identical patterns. B. Patterns of RFLP of rONA regions from E. cloacae after EcoRI digestion. Strains are the same as those in A.
patterns ofRFLP of DNA. To our knowledge, such results of molecular analysis that strongly support the endogenous nature of systemic bacteremia and meningitis due to E. cloacae have not been previously reported. Six days after the start of cefotaxime therapy, cefotaximeresistant E. cloacae was found in stool from our patient. Emergence of cefotaxime resistance in E. cloacae was previously reported for neonates in intensive care facilities after cefotaxime treatment [10]. Resistance to cefotaxime is believed to result from the derepression of a chromosomal (3lactamase gene [ 11]. In a study by Bryan et al., the intestines
Downloaded from http://cid.oxfordjournals.org/ at Freie Universitaet Berlin on July 4, 2015
unrelated strains of E. cloacae (strains 6-9) and the type strain of the species ATCC 13047 were studied for comparison. Identification of E. cloacae was based on colony morphology and the results of standard biochemical tests included in the API-20E system (API, La Balme-les-Grottes, France). The bacterial count in stool was performed as previously described [6]. Susceptibility testing was performed on plates of Mueller-Hinton agar by the disk diffusion method (Diagnostic Pasteur, Marnes-la-Coquette. France). Total DNA from E. cloacae was prepared by a previously described method [7]. DNA (4 f.Lg) was digested with BamHI and EcoRI restriction enzymes (Boehringer, Mannheim, Germany) according to the manufacturer's specifications and analyzed by electrophoresis on 0.8% ethidium bromidecontaining submarine agarose gels in a 0.04 M Tris-acetate, 0.001 M EDTA buffer. The DNA fragment size marker RaoulI (Appligene, Strasbourg, France) was used. Size-separated restriction fragments were then transferred to a nylon membrane (Gene Screen Plus, New England Nuclear Products, Boston) by the method of Southern. Ribosomal 16 + 23 S RNA from E. coli (Boehringer) was used as a probe and was cold-labeled by random oligopriming with use of a mixture of hexanucleotides (Pharmacia, Uppsala, Sweden) and cloned M-ML V reverse transcriptase (Bethesda Research Laboratories, Gaithersburg, MD) in the presence of 0.35 mM DIG-ll-dUTP (digoxigenin-l l-deoxyuridine 5'-triphosphate, Boehringer) as previously described [8]. All the E. cloacae strains from the newborn and the unrelated strains exhibited identical biochemical patterns and susceptibility profiles. In contrast, digestion with EcoRI and BamHI generated six different patterns of RFLP of total DNA for the 10 studied strains. The same enzymes generated six and five different patterns ofRFLP of rONA, respectively, thus also defining six different ribotypes (figure 1). All the five strains from the newborn shared the same ribotype, which was different from those of the other five studied strains. Thus, each of the strains studied for comparison exhibited a unique ribotype that was clearly distinct from that of the clinically related strains.
32
Lambert-Zechovsky et al.
Acknowledgment The authors thank Michelle Fageon for typing the manuscript. References I. Grauel EL, Halle E, Bollmann R. Buchholz P, Buttenberg S. Neonatal septicaemia-incidence, etiology and outcome. A 6-year analysis. Acta Paediatr Scand Suppl 1989;360: 113-99.
2. Gallagher PG. Enterobacter bacteremia in pediatric patients. Rev Infect Dis 1990;12:808-12. 3. Flynn DM, Weinstein RA, Nathan C, Gaston MA, Kabins SA. Patients' endogenous flora as the source of "nosocomial" Enterobacter in cardiac surgery. 1 Infect Dis 1987; 156:363-8. 4. Berg RD, Garlington AW. Translocation of certain indigenous bacteria from the gastrointestinal tract to the mesenteric lymph nodes and other organs in an gnotobiotic mouse model. Infect Immun 1979;23:403-11. 5. Berg RD. Translocation of indigenous bacteria from the intestinal tract. In: Hentges Df, ed. Human intestinal microflora in health and disease. New York: Academic Press, 1983:333-51. 6. Bingen E, Lambert-Zechovsky N. Technical aspects of quantitative and differential analysis of microbial intestinal ecosystem. Dev Pharmacol Ther 1984;7(suppl I): 134-7. 7. Bingen E, Denamur E, Lambert-Zechovsky N, et al. DNA restriction fragment length polymorphism differentiates crossed from independent infections in nosocomial Xanthomonas maltophilia bacteremia. 1 Clin Microbiol 1991;29: 1348-50. 8. Bingen E, Denamur E, Lambert-Zechovsky N, et al. Analysis of DNA restriction fragment length polymorphism extends the evidence for breast milk transmission in Streptococcus agalactiaelate-onset neonatal infection. 1 Infect Dis 1992;165:569-73. 9. Garaizar 1, Kaufmann ME, Pitt TL. Comparison of ribotyping with conventional methods for the type identification of Enterobacter cloacae. 1 Clin Microbiol 1991;29: 1303-7. 10. Bryan CS, John Jf 1r, Pai MS, Austin TL. Gentamicin vs cefotaxime for therapy of neonatal sepsis. Relationship to drug resistance. Am 1 Dis Child 1985; 139:1086-9. I I. Sanders CC, Sanders WE 1r. Microbial resistance to newer generation tJ-lactam antibiotics: clinical and laboratory implications. 1 Infect Dis 1985; 151:399-406. 12. Forbes KJ, Bruce KD, Jordens rz, Ball A, Pennington TH. Rapid methods in bacterial DNA fingerprinting. 1 Gen Microbiol 1991;137:2051-8. 13. Grimont F, Grimont PAD. Ribosomal ribonucleic acid gene restriction patterns as potential taxonomic tools. Ann Inst Pasteur Microbiol 1986; 137B:165-75. 14. Stull TL, Lipuma 11, Edlind TD. A broad-spectrum probe for molecular epidemiology of bacteria: ribosomal RNA. 1 Infect Dis 1988; 157:280-6.
Downloaded from http://cid.oxfordjournals.org/ at Freie Universitaet Berlin on July 4, 2015
of 71% of the infants who received ampicillin plus cefotaxime were colonized with cefotaxime-resistant E. cloacae. and four of the infants had sepsis caused by this resistant strain [10]. RFLP analysis of total DNA is a rapid and inexpensive method that is easy to perform. In this rather small series, it allowed us to readily compare all our strains. Yet the comparison of numerous strains by this approach is often difficult because ofthe large number ofbands and the poor resolution obtained in the gels. However, improvements in the analysis of the generated patterns have been recently devised that allow a reliable comparison oflarge numbers of isolates [12]. Restriction pattern analysis is also often refined with use of labeled rRNA in probing the ubiquitous and polymorphic rDNA loci [13, 14]. Indeed, ribotyping has already been used in the study of E. cloacae strains and compared with the conventional typing methods [9]. In this study, a nonradioactive probing system was used with great success and several advantages over radioactive probes; these advantages include safety, disposal consideration. and stability. Thus, RFLP analysis of total DNA and ribotyping seem particularly well suited to the study of the epidemiology of nosocomial E. cloacae strains. In our study, digestion with EcoRI generated the same number of ribotypes and patterns ofRFLP of total DNA, thus demonstrating the usefulness of this enzyme in molecular epidemiological studies of E. cloacae strains.
CID 1992; 15 (July)
Molecular Analysis Provides Evidence for the Endogenous Origin of Bacteremia and Meningitis Due to Enterobacter cloacae in an Infant N. Lambert-Zechovsky, E. Bingen, E. Denamur, N. Brahimi, P. Brun, H. Mathieu, and J. Elion
From the Laboratoire de Microbiologie. Laboratoire de Biochimie Genetique, and Service de Nephrologie, Hiipital Robert Debre, Paris. France
Despite the improved care given to newborns, septicemia remains a serious problem that is associated with a significant mortality rate, especially when complicated by meningitis. The incidence of neonatal septicemia varies from I to 4 per 1,000 live-born infants [I ]. Enterobacter organisms are the fourth most common cause of bacteremia in children [2]. Flynn et al. have reported that most nosocomial infections due to Enterobacter species arise from the patient's own flora, especially intestinal flora [3]. Indeed, organisms from the gastrointestinal tract may translocate to mesenteric lymph nodes or to the bloodstream if bacterial populations reach abnormally high levels in the bowel [4, 5]. However, hard facts are still lacking that definitively demonstrate this mode of bacterial transmission because its study has been somewhat hampered by the absence of a sufficiently discriminating typing system. In this study, we analyzed the restriction fragment length polymorphism (RFLP) of total DNA and of ribosomal DNA (rDNA) regions (ribotyping) to document the occurrence of endogenous, systemic bacteremia and meningitis due to Enterobacter cloacae in a newborn.
Case Report A 2, 150-g male infant was born at 33 weeks of gestation by spontaneous vaginal delivery after an uncomplicated pregnancy. The results ofthe initial examination of the newborn were unremarkable. At 24 days of age, he developed a
Received 26 December 1991; revised 26 February 1992. Reprints or correspondence: Dr. Nicole Lambert-Zechovsky, Laboratoire de Microbiologic, Hopital Robert Debre, 48 Bd Serurier, 75019 Paris. France. Clinical Infectious Diseases
1992;15:30-2
© 1992 by The University of Chicago. All rights reserved.
1058-4838/92/1501-0006$02.00
fever at 38.6°C but had no other clinical symptoms. Cultures of blood and CSF remained sterile. Examination of a gramstained smear of urine revealed numerous gram-negative rods and many polymorphonuclear leukocytes. Culture of urine yielded Escherichia coli (10 6 organisms/ml.). Intravenous therapy with cefotaxime ( 100 mg/[kg . dD and netilmicin (4 mg/[kg· d]) was initiated. The infant became afebrile and did well 2 days after the initiation of treatment. Repeated urine cultures were sterile. Six and 8 days after the beginning of therapy, cultures of stool yielded cefotaxime-resistant E. cloacae at counts of 108 and 109 organisms/g of feces, respectively. The administration of antibiotics was discontinued after 10 days of therapy. Three days later, at 36 days ofage, the infant was lethargic and again became febrile. Two blood cultures and one urine culture yielded E. cloacae that was resistant to cefotaxime. Culture of stool yielded pure growth of the same organism (10 9 organisms/g). CSF contained gram-negative rods and 1,200 nucleated cells/rum! (96% neutrophils, 4% lymphocytes). Other laboratory values for CSF were a glucose level of 11 mg/dL and a protein level of 900 mg/dL. E. cloacae that was resistant to cefotaxime was isolated from the CSF at a count of 106 organisms/ml., On the basis of the results of the antibiotic susceptibility profile, administration of imipenem (60 mg/[kg· dD, ciprofloxacin (20 mg/[kg· dD, and amikacin (15 mg/[kg· dj) was immediately started, resulting in the sterilization of blood, urine, and CSF in 48 hours. The clinical outcome was good, and the infant was discharged after a total of 55 days of hospitalization. A total of 10 isolates of E. cloacae were selected for the study. Five isolates (strains 1-5) were related to the described case. Three of these five isolates were recovered from stool at counts of 108, 109 , and 109 organisms/g of feces on 7 days before, 5 days before, and the day ofthe onset ofbacteremia, respectively; of the other two, one each was isolated from blood and CSF. In addition, four epidemiologically
Downloaded from http://cid.oxfordjournals.org/ at Freie Universitaet Berlin on July 4, 2015
We analyzed the restriction fragment length polymorphism (RFLP) of total DNA and of ribosomal DNA regions (ribotyping) to document the occurrence of endogenous, systemic bacteremia and meningitis due to Enterobacter cloacae in a newborn. Five strains of E. cloacae were isolated from this newborn. Three of these strains were recovered from stool at counts of 108 , 109, and 109 organisms/g offeces, respectively; one strain was isolated from blood; and one strain was isolated from cerebrospinal fluid. In addition, five epidemiologically unrelated strains of E. cloacae were studied for comparison. Our study clearly shows the genetic relatedness of the strains isolated sequentially from cultures of stool, blood, and cerebrospinal fluid. RFLP analysis of total DNA and ribotyping seem particularly well suited to the study of the epidemiology of nosocomial E. cloacae strains.
CID 1992;15 (July)
31
Molecular Analysis of E. cloacae
Discussion Until recently, epidemiological studies of E. cloacae have been based essentially on the study of phenotypic traits such as biochemical profiles, antibiotic resistance, and serological, bacteriocin, and phage typing. However, none of these methods are really satisfactory for the typing of E. cloacae because of insufficient discrimination, poor reproducibility, or low typability [9]. We used the RFLP analysis of total DNA and rONA regions for the study of five E. cloacae strains isolated from an infant who had bacteremia and meningitis. This method allowed us to clearly show the genetic relatedness of the strains isolated sequentially from cultures of stool , blood, and CSF. Strains studied for comparison all had different
2
3
4
5
6
7
8
9
10
11
Kb
- 48.5 - 14.9
-10.6
- 5.6
A
2
3
4
5
6
7
8
9
10
Kb
-10.6
-7.3 - 5.6
- 3.9
B
Figure 1. A. Patterns ofRFLP of total DNA from E. cloacae obtained after BamHI digestion and ethidium bromide staining. Lane t, ATCC 13047; iane 2, strain 1; lane 3, strain 2; lane 4. strain 3; lane 5. strain 4; lane 6. strain 5; lane 7. strain 6; lane 8. strain 7; lane 9. strain 8; lane i 0, strain 9; and lane l l , size markers (Kb = kilobases). Strains 1 to 5 (lanes 2 to 6) show identical patterns. B. Patterns of RFLP of rONA regions from E. cloacae after EcoRI digestion. Strains are the same as those in A.
patterns ofRFLP of DNA. To our knowledge, such results of molecular analysis that strongly support the endogenous nature of systemic bacteremia and meningitis due to E. cloacae have not been previously reported. Six days after the start of cefotaxime therapy, cefotaximeresistant E. cloacae was found in stool from our patient. Emergence of cefotaxime resistance in E. cloacae was previously reported for neonates in intensive care facilities after cefotaxime treatment [10]. Resistance to cefotaxime is believed to result from the derepression of a chromosomal (3lactamase gene [ 11]. In a study by Bryan et al., the intestines
Downloaded from http://cid.oxfordjournals.org/ at Freie Universitaet Berlin on July 4, 2015
unrelated strains of E. cloacae (strains 6-9) and the type strain of the species ATCC 13047 were studied for comparison. Identification of E. cloacae was based on colony morphology and the results of standard biochemical tests included in the API-20E system (API, La Balme-les-Grottes, France). The bacterial count in stool was performed as previously described [6]. Susceptibility testing was performed on plates of Mueller-Hinton agar by the disk diffusion method (Diagnostic Pasteur, Marnes-la-Coquette. France). Total DNA from E. cloacae was prepared by a previously described method [7]. DNA (4 f.Lg) was digested with BamHI and EcoRI restriction enzymes (Boehringer, Mannheim, Germany) according to the manufacturer's specifications and analyzed by electrophoresis on 0.8% ethidium bromidecontaining submarine agarose gels in a 0.04 M Tris-acetate, 0.001 M EDTA buffer. The DNA fragment size marker RaoulI (Appligene, Strasbourg, France) was used. Size-separated restriction fragments were then transferred to a nylon membrane (Gene Screen Plus, New England Nuclear Products, Boston) by the method of Southern. Ribosomal 16 + 23 S RNA from E. coli (Boehringer) was used as a probe and was cold-labeled by random oligopriming with use of a mixture of hexanucleotides (Pharmacia, Uppsala, Sweden) and cloned M-ML V reverse transcriptase (Bethesda Research Laboratories, Gaithersburg, MD) in the presence of 0.35 mM DIG-ll-dUTP (digoxigenin-l l-deoxyuridine 5'-triphosphate, Boehringer) as previously described [8]. All the E. cloacae strains from the newborn and the unrelated strains exhibited identical biochemical patterns and susceptibility profiles. In contrast, digestion with EcoRI and BamHI generated six different patterns of RFLP of total DNA for the 10 studied strains. The same enzymes generated six and five different patterns ofRFLP of rONA, respectively, thus also defining six different ribotypes (figure 1). All the five strains from the newborn shared the same ribotype, which was different from those of the other five studied strains. Thus, each of the strains studied for comparison exhibited a unique ribotype that was clearly distinct from that of the clinically related strains.
32
Lambert-Zechovsky et al.
Acknowledgment The authors thank Michelle Fageon for typing the manuscript. References I. Grauel EL, Halle E, Bollmann R. Buchholz P, Buttenberg S. Neonatal septicaemia-incidence, etiology and outcome. A 6-year analysis. Acta Paediatr Scand Suppl 1989;360: 113-99.
2. Gallagher PG. Enterobacter bacteremia in pediatric patients. Rev Infect Dis 1990;12:808-12. 3. Flynn DM, Weinstein RA, Nathan C, Gaston MA, Kabins SA. Patients' endogenous flora as the source of "nosocomial" Enterobacter in cardiac surgery. 1 Infect Dis 1987; 156:363-8. 4. Berg RD, Garlington AW. Translocation of certain indigenous bacteria from the gastrointestinal tract to the mesenteric lymph nodes and other organs in an gnotobiotic mouse model. Infect Immun 1979;23:403-11. 5. Berg RD. Translocation of indigenous bacteria from the intestinal tract. In: Hentges Df, ed. Human intestinal microflora in health and disease. New York: Academic Press, 1983:333-51. 6. Bingen E, Lambert-Zechovsky N. Technical aspects of quantitative and differential analysis of microbial intestinal ecosystem. Dev Pharmacol Ther 1984;7(suppl I): 134-7. 7. Bingen E, Denamur E, Lambert-Zechovsky N, et al. DNA restriction fragment length polymorphism differentiates crossed from independent infections in nosocomial Xanthomonas maltophilia bacteremia. 1 Clin Microbiol 1991;29: 1348-50. 8. Bingen E, Denamur E, Lambert-Zechovsky N, et al. Analysis of DNA restriction fragment length polymorphism extends the evidence for breast milk transmission in Streptococcus agalactiaelate-onset neonatal infection. 1 Infect Dis 1992;165:569-73. 9. Garaizar 1, Kaufmann ME, Pitt TL. Comparison of ribotyping with conventional methods for the type identification of Enterobacter cloacae. 1 Clin Microbiol 1991;29: 1303-7. 10. Bryan CS, John Jf 1r, Pai MS, Austin TL. Gentamicin vs cefotaxime for therapy of neonatal sepsis. Relationship to drug resistance. Am 1 Dis Child 1985; 139:1086-9. I I. Sanders CC, Sanders WE 1r. Microbial resistance to newer generation tJ-lactam antibiotics: clinical and laboratory implications. 1 Infect Dis 1985; 151:399-406. 12. Forbes KJ, Bruce KD, Jordens rz, Ball A, Pennington TH. Rapid methods in bacterial DNA fingerprinting. 1 Gen Microbiol 1991;137:2051-8. 13. Grimont F, Grimont PAD. Ribosomal ribonucleic acid gene restriction patterns as potential taxonomic tools. Ann Inst Pasteur Microbiol 1986; 137B:165-75. 14. Stull TL, Lipuma 11, Edlind TD. A broad-spectrum probe for molecular epidemiology of bacteria: ribosomal RNA. 1 Infect Dis 1988; 157:280-6.
Downloaded from http://cid.oxfordjournals.org/ at Freie Universitaet Berlin on July 4, 2015
of 71% of the infants who received ampicillin plus cefotaxime were colonized with cefotaxime-resistant E. cloacae. and four of the infants had sepsis caused by this resistant strain [10]. RFLP analysis of total DNA is a rapid and inexpensive method that is easy to perform. In this rather small series, it allowed us to readily compare all our strains. Yet the comparison of numerous strains by this approach is often difficult because ofthe large number ofbands and the poor resolution obtained in the gels. However, improvements in the analysis of the generated patterns have been recently devised that allow a reliable comparison oflarge numbers of isolates [12]. Restriction pattern analysis is also often refined with use of labeled rRNA in probing the ubiquitous and polymorphic rDNA loci [13, 14]. Indeed, ribotyping has already been used in the study of E. cloacae strains and compared with the conventional typing methods [9]. In this study, a nonradioactive probing system was used with great success and several advantages over radioactive probes; these advantages include safety, disposal consideration. and stability. Thus, RFLP analysis of total DNA and ribotyping seem particularly well suited to the study of the epidemiology of nosocomial E. cloacae strains. In our study, digestion with EcoRI generated the same number of ribotypes and patterns ofRFLP of total DNA, thus demonstrating the usefulness of this enzyme in molecular epidemiological studies of E. cloacae strains.
CID 1992; 15 (July)
