- Email: [email protected]
Rapid identification of enterococci by pyrrolidonyl aminopeptidase activity (PYRase)
DIAGNMICROBIOLINFECTDIS 1987;6:283-286
283
Rapid Identification of Enterococci by Pyrrolidonyl Aminopeptidase Activity (PYRase) M.J. Mitchell, P.S. Conville, and V.J. Gill
Group A streptococci and enterococci can be differentiated from other streptococci by their ability to cleave pyrrolidonyl [3-naphthylamide (PYBose). We evaluated two PYBase systems [Strep-A-Chek (SAC), E-Y Laboratories, San Mateo, CA; Strep-A-Fluor (SAF), BioSpec, Inc., Dublin, CA) for the presumptive identification of enterococci. Initially, 40 enterococcal and 21 nonenterococcal streptococci were tested, retrospectively. A prospective comparison of SAC and SAF to bile---esculin reaction (BE) was then incorporated into our routine procedure for the identification of non-13-hemolytic streptococcal colonies from cultures. All isolates were speciated using standard biochemical tests. We encountered 85 enterococcal and 26 nonenterococcal isolates. Tests were performed on colonies from primary plates whenever possible ~77 isolates). Sensitivity and specificity of both SAC and SAF were greater than 96% in identifying enterococci from routine cultures. These PYBase tests were cost-effective, easily adaptable to work-flow, and yielded results within 30 min. Thus, PYBase testing appears to be a reasonable alternative for the identification of enterococci in the clinical laboratory.
INTRODUCTION Enterococci are frequently encountered in the diagnostic microbiology laboratory and are involved in a variety of types of infection, including urinary tract infection, w o u n d infections, and intraabdominal abscesses. The importance of identifying enterococci lies in their increased resistance to antibiotic agents, compared with other types of streptococci. Godsey et al. (1981) have suggested the usefulness of pyrrolidonyl aminopeptidase activity in the identification of enterococci. We evaluated two commercially available kits, which detect PYRase, for the identification of enterococci in our laboratory. These kits measure preformed enzyme and yield results within 30 rain. The study was designed to incorporate the tests in the normal workflow of the microbiology laboratory. MATERIALS AND METHODS The evaluation was conducted in two phases. During the retrospective phase, laboratory stock cultures of previously speciated, non-G-hemolytic streptococci and isolates of several other genera were tested. During the prospective phase, PYRase testing was added to the routine identification schema for all suspected enterococcal
From the Microbiology Service, Clinical Pathology Department, Clinical Center, National Institutes of Health, Bethesda, MD. Address reprint requests to: M.J. Mitchell, Clinical Microbiology, Bldg. 10, Rm. 2C385, National Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892. Received July 2, 1986; revised and accepted December 2, 1986. © 1987 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Avenue, New York, NY 10017
0732-8893/87/$3.50
Z84
M.J. Mitchell, P.S. Conville, and V.J. Gill
isolates from routine, aerobic cultures. A RIM Series Bile Esculin Test (BE) (ABL, Austin, TX), Strep-A-Fluor (SAF) (BioSpec, Inc. Dublin, CA), and Strep-A-Chek (SAC) (E-Y Labs, San Mateo, CA), were inoculated with non-~-hemolytic colonies having streptococcal Gram stain and/or colony morphology. Tests were inoculated and interpreted according to the manufacturers' instructions. If there was insufficient growth on the primary plates to complete the above tests, a bile esculin slant (NIH Media Service) and a 5% sheep blood agar plate (GIBCO Laboratories, Chagrin Falls, OH) were inoculated. In these cases, PYRase tests were performed on growth from the isolation plates after 18-24 hr incubation. All isolates evaluated during the prospective phase were subsequently speciated using the 20S strip (API, Plainview, NY) plus additional biochemical testing according to Facklam (1977), when necessary. RESULTS Table 1 shows PYRase activity of laboratory stock strains of streptococci. The three enterococcal isolates, which were initially negative in SAC, were positive on repeat testing. Initial negative results might have resulted from difficulty transferring growth onto the test pad. In order to determine PYRase activity of organisms whose colonies might appear similar to enterococci in young cultures, a number of nonstreptococcal TABLE 1. PYRase Activity of Laboratory Stock Isolates Positive (n} Species
Strains tested (n}
SAC
SAF
GROUP D STREPTOCOCCI
Enterococci: S. ovium S. durans
2 2 30
1~ 2 30
2 2 29
5
3a
5
1
0
0
3 2
0 0
0 0
S. mitis
1
0
0
S. morbillorum S. m u t a n s S. salivarius
2 3 3
0 0 0
0 0 0
S. sanguis I
2
0
0
S. sanguis II
3
0
0
S. faecalis S. faecium
Nonenterococci S. bovis NON-GROUP D STREPTOCOCCI
S. anginosus-constellatus S. intermedius
NON-STREPTOCOCCI
Staphylococcus aureus Staphylococcus, coagulase negative Micrococcus spp. Listeria monocytogenes Yeast Corynebacterium spp. Aerococcus spp.
°Initially negative isolate(s) all positive on repeat testing.
39 10 1
18 (46%) 4 (40%) 1
6
0
8 9 2
0 2 (22%} 1
Rapid Identification of Enterococci
285
TABLE 2. PYRase Activity of Organisms Tested in the Routine Laboratory Positive (n} Isolate
Strains isolated (n)
SAC
SAF
BE
5
3°
4°
4
2
2
2
0
75
74a
74a
74
3
3
3
3
85
82
83
81
5 1
0 0
0 0
0 0
ENTEROCOCCI
S. S. S. S.
avium durans faecalis faecium
Total NON-GROUP D STREPTOCOCCI
S. agalactiae S. anginosus-constellatus S. intermedius S. mitis S. salivarius S. sanguis II Streptococcus, spp. b
Total
15
0
0
4
2 1 1
0 0 0
0 0 0
0 0 0
1
0
0
0
26
0
0
4
°Initially negative isolate(s) all positive on repeat testing. ba-hereolytic Streptococcus was not further identifiable by routine biochemical testing. isolates were tested. The PYRase activity was tested by SAC only, and these results are also shown in Table 1. The PYRase activity of suspected enterococcal isolates encountered in the diagnostic laboratory was tested prospectively between March and July, 1985. Tests were performed on 85 enterococcal and 26 nonenterococcal strains of streptococci identified in our laboratory during this period. Tests were performed on colonies from primary plates in 77 of the 85 enterococcal isolates identified. The species encountered and PYRase activities are shown in Table 2. Initial testing of two Streptococcus a v i u m isolates and one S t r e p t o c o c c u s f a e c a l i s were negative in SAC and one isolate each of S. a v i u m and S. f a e c a l i s were negative in SAF. All of these isolates gave positive reactions when retested from purity check plates. DISCUSSION Accurate detection of enterococci is important because of their pathogenic potential and increased resistance to antibiotics. Rapid methods yielding preliminary or presumptive identification of enterococci might allow early initiation or adjustment of antibiotic therapy. Enzymatic hydrolysis of L-pyrrolidonyl aminopeptides has been show to be a characteristic of enterococci, and its detection could provide such a method for their rapid identification. Several retrospective studies have reported this reaction to have good sensitivity and specificity in the identification of enterococcal strains that had been isolated from clinical cultures. Facklam et al. (1982) incorporated L-pyrrolidonyl-j3-naphthylamide into agar and measured hydrolysis after overnight incubation. Bosley et al. (1983) used an L-pyrrolidonyl-~-naphthlamide containing broth, w h i c h required 4 hr of incubation. Recently, several kits that measure L-pyrrolydonyl aminopeptidase activity by preformed enzyme have become commercially available. These tests yield results
286
M.J. Mitchell, P.S. Conville, and V.J. Gill
within 30 min. Ellner et al. (1985) have reported PYRase activity of enterococcal isolates using one of these kits. In the present study, we wished to compare the performance of two other kits in a clinical laboratory. Sensitivity and specificity of results obtained by technologists and the impact on work-flow were evaluated. Initially, we tested 39 enterococcal and 22 nonenterococcal laboratory stock cultures. Both kits yielded good sensitivity and specificity in the identification of enterococci. A number of nonstreptococcal isolates were also tested for PYRase activity at this time. A significant number of staphylococcal and diphtheroid strains also showed PYRase activity. Previously reported data have also shown PYRase activity in nonstreptococcal isolates (Ellner et al., 1985; Oberhofer, 1986). Furthermore, group A ~-hemolytic streptococci are PYRase positive; therefore, attention to hemolytic reaction and colony and Gram stain morphology is critically important in the selection of colonies for testing and the interpretations of results. During the second phase of the study, SAF and SAC were used to identify enterococci prospectively as they were encountered by technologists during work-up of cultures. Positive results in both kits were 100% specific for enterococci. The sensitivity of detection of enterococci by SAC and SAF was 96.5% and 97.6%, respectively. False negative reactions occured only in tests performed from primary plates. All strains gave positive reactions when retested from purity check plates. In contrast, the bile-esculin reaction was only 95% sensitive or specific as a test for the presumptive identification of group D streptococci during this study period. Inclusion of 6.5% NaC1 tolerance would have only resulted in the incorrect identification of four strains of Streptococcus intermedius as Streptococcus bavis. During the prospective phase of this study, no nonenterococcal group D streptococci were encountered. Several restrospective studies (Bosley et al., 1983; Oberhofer, 1986) have shown nonenterococcal group D streptococcal strains to be consistently PYRase negative. Identification of enterococci by SAF and SAC was comparable with classical schemes. PYRase testing was possible on growth from primary plates in 77 of 85 strains of enterococci encountered and results were available within 30 min from the time a colony was detected. Tests from both kits were simple to perform and interpret without special training. Strips could be inoculated, set aside, and then batched for later addition of test reagent. Testing was easily incorporated into routine work-flow and was readily accepted by the technologists. The current cost per test is about 30¢ for SAC and 70¢ for SAF. In conclusion, PYRase activity as measured by the SAC or SAF kits is an accurate, cost-effective method for the preliminary or presumptive identification of enterococci, when used in conjunction with Gram stain and colony morphology. REFERENCES Bosley GS, Facklam RR, Grossman D (1983) Rapid identification of enterococci. J Clin Microbiol 18:1275. Ellner PD, Williams DA, Hosmer ME, Cohenford MA (1985) Preliminary evaluation of a rapid colorimetric method for the presumptive identification of Group A streptococci and enterococci, l Clin Microbiol 22:880. Facklam RR (1977) Physiological differentiation of viridans streptococci. ] Clin Microbiol 5:184. Facklam RR, Thacker LG, Fox B, Eriquez L (1982) Presumptive identification of streptococci with a new test system. J Clin Microbiol 15:987. Godsey J, Schulman R, Eriquez L (1981) Annual meeting of the American Society of Microbiologists, abstr C 84, p. 276. Oberhofer TR (1985) Value of the L-pyrrolidonyl-[3-naphthylamide hydrolysis test for identification of select Gram-positive cocci. Diagn Microbiol Infect Dis 4:43.
283
Rapid Identification of Enterococci by Pyrrolidonyl Aminopeptidase Activity (PYRase) M.J. Mitchell, P.S. Conville, and V.J. Gill
Group A streptococci and enterococci can be differentiated from other streptococci by their ability to cleave pyrrolidonyl [3-naphthylamide (PYBose). We evaluated two PYBase systems [Strep-A-Chek (SAC), E-Y Laboratories, San Mateo, CA; Strep-A-Fluor (SAF), BioSpec, Inc., Dublin, CA) for the presumptive identification of enterococci. Initially, 40 enterococcal and 21 nonenterococcal streptococci were tested, retrospectively. A prospective comparison of SAC and SAF to bile---esculin reaction (BE) was then incorporated into our routine procedure for the identification of non-13-hemolytic streptococcal colonies from cultures. All isolates were speciated using standard biochemical tests. We encountered 85 enterococcal and 26 nonenterococcal isolates. Tests were performed on colonies from primary plates whenever possible ~77 isolates). Sensitivity and specificity of both SAC and SAF were greater than 96% in identifying enterococci from routine cultures. These PYBase tests were cost-effective, easily adaptable to work-flow, and yielded results within 30 min. Thus, PYBase testing appears to be a reasonable alternative for the identification of enterococci in the clinical laboratory.
INTRODUCTION Enterococci are frequently encountered in the diagnostic microbiology laboratory and are involved in a variety of types of infection, including urinary tract infection, w o u n d infections, and intraabdominal abscesses. The importance of identifying enterococci lies in their increased resistance to antibiotic agents, compared with other types of streptococci. Godsey et al. (1981) have suggested the usefulness of pyrrolidonyl aminopeptidase activity in the identification of enterococci. We evaluated two commercially available kits, which detect PYRase, for the identification of enterococci in our laboratory. These kits measure preformed enzyme and yield results within 30 rain. The study was designed to incorporate the tests in the normal workflow of the microbiology laboratory. MATERIALS AND METHODS The evaluation was conducted in two phases. During the retrospective phase, laboratory stock cultures of previously speciated, non-G-hemolytic streptococci and isolates of several other genera were tested. During the prospective phase, PYRase testing was added to the routine identification schema for all suspected enterococcal
From the Microbiology Service, Clinical Pathology Department, Clinical Center, National Institutes of Health, Bethesda, MD. Address reprint requests to: M.J. Mitchell, Clinical Microbiology, Bldg. 10, Rm. 2C385, National Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892. Received July 2, 1986; revised and accepted December 2, 1986. © 1987 Elsevier Science Publishing Co., Inc. 52 Vanderbilt Avenue, New York, NY 10017
0732-8893/87/$3.50
Z84
M.J. Mitchell, P.S. Conville, and V.J. Gill
isolates from routine, aerobic cultures. A RIM Series Bile Esculin Test (BE) (ABL, Austin, TX), Strep-A-Fluor (SAF) (BioSpec, Inc. Dublin, CA), and Strep-A-Chek (SAC) (E-Y Labs, San Mateo, CA), were inoculated with non-~-hemolytic colonies having streptococcal Gram stain and/or colony morphology. Tests were inoculated and interpreted according to the manufacturers' instructions. If there was insufficient growth on the primary plates to complete the above tests, a bile esculin slant (NIH Media Service) and a 5% sheep blood agar plate (GIBCO Laboratories, Chagrin Falls, OH) were inoculated. In these cases, PYRase tests were performed on growth from the isolation plates after 18-24 hr incubation. All isolates evaluated during the prospective phase were subsequently speciated using the 20S strip (API, Plainview, NY) plus additional biochemical testing according to Facklam (1977), when necessary. RESULTS Table 1 shows PYRase activity of laboratory stock strains of streptococci. The three enterococcal isolates, which were initially negative in SAC, were positive on repeat testing. Initial negative results might have resulted from difficulty transferring growth onto the test pad. In order to determine PYRase activity of organisms whose colonies might appear similar to enterococci in young cultures, a number of nonstreptococcal TABLE 1. PYRase Activity of Laboratory Stock Isolates Positive (n} Species
Strains tested (n}
SAC
SAF
GROUP D STREPTOCOCCI
Enterococci: S. ovium S. durans
2 2 30
1~ 2 30
2 2 29
5
3a
5
1
0
0
3 2
0 0
0 0
S. mitis
1
0
0
S. morbillorum S. m u t a n s S. salivarius
2 3 3
0 0 0
0 0 0
S. sanguis I
2
0
0
S. sanguis II
3
0
0
S. faecalis S. faecium
Nonenterococci S. bovis NON-GROUP D STREPTOCOCCI
S. anginosus-constellatus S. intermedius
NON-STREPTOCOCCI
Staphylococcus aureus Staphylococcus, coagulase negative Micrococcus spp. Listeria monocytogenes Yeast Corynebacterium spp. Aerococcus spp.
°Initially negative isolate(s) all positive on repeat testing.
39 10 1
18 (46%) 4 (40%) 1
6
0
8 9 2
0 2 (22%} 1
Rapid Identification of Enterococci
285
TABLE 2. PYRase Activity of Organisms Tested in the Routine Laboratory Positive (n} Isolate
Strains isolated (n)
SAC
SAF
BE
5
3°
4°
4
2
2
2
0
75
74a
74a
74
3
3
3
3
85
82
83
81
5 1
0 0
0 0
0 0
ENTEROCOCCI
S. S. S. S.
avium durans faecalis faecium
Total NON-GROUP D STREPTOCOCCI
S. agalactiae S. anginosus-constellatus S. intermedius S. mitis S. salivarius S. sanguis II Streptococcus, spp. b
Total
15
0
0
4
2 1 1
0 0 0
0 0 0
0 0 0
1
0
0
0
26
0
0
4
°Initially negative isolate(s) all positive on repeat testing. ba-hereolytic Streptococcus was not further identifiable by routine biochemical testing. isolates were tested. The PYRase activity was tested by SAC only, and these results are also shown in Table 1. The PYRase activity of suspected enterococcal isolates encountered in the diagnostic laboratory was tested prospectively between March and July, 1985. Tests were performed on 85 enterococcal and 26 nonenterococcal strains of streptococci identified in our laboratory during this period. Tests were performed on colonies from primary plates in 77 of the 85 enterococcal isolates identified. The species encountered and PYRase activities are shown in Table 2. Initial testing of two Streptococcus a v i u m isolates and one S t r e p t o c o c c u s f a e c a l i s were negative in SAC and one isolate each of S. a v i u m and S. f a e c a l i s were negative in SAF. All of these isolates gave positive reactions when retested from purity check plates. DISCUSSION Accurate detection of enterococci is important because of their pathogenic potential and increased resistance to antibiotics. Rapid methods yielding preliminary or presumptive identification of enterococci might allow early initiation or adjustment of antibiotic therapy. Enzymatic hydrolysis of L-pyrrolidonyl aminopeptides has been show to be a characteristic of enterococci, and its detection could provide such a method for their rapid identification. Several retrospective studies have reported this reaction to have good sensitivity and specificity in the identification of enterococcal strains that had been isolated from clinical cultures. Facklam et al. (1982) incorporated L-pyrrolidonyl-j3-naphthylamide into agar and measured hydrolysis after overnight incubation. Bosley et al. (1983) used an L-pyrrolidonyl-~-naphthlamide containing broth, w h i c h required 4 hr of incubation. Recently, several kits that measure L-pyrrolydonyl aminopeptidase activity by preformed enzyme have become commercially available. These tests yield results
286
M.J. Mitchell, P.S. Conville, and V.J. Gill
within 30 min. Ellner et al. (1985) have reported PYRase activity of enterococcal isolates using one of these kits. In the present study, we wished to compare the performance of two other kits in a clinical laboratory. Sensitivity and specificity of results obtained by technologists and the impact on work-flow were evaluated. Initially, we tested 39 enterococcal and 22 nonenterococcal laboratory stock cultures. Both kits yielded good sensitivity and specificity in the identification of enterococci. A number of nonstreptococcal isolates were also tested for PYRase activity at this time. A significant number of staphylococcal and diphtheroid strains also showed PYRase activity. Previously reported data have also shown PYRase activity in nonstreptococcal isolates (Ellner et al., 1985; Oberhofer, 1986). Furthermore, group A ~-hemolytic streptococci are PYRase positive; therefore, attention to hemolytic reaction and colony and Gram stain morphology is critically important in the selection of colonies for testing and the interpretations of results. During the second phase of the study, SAF and SAC were used to identify enterococci prospectively as they were encountered by technologists during work-up of cultures. Positive results in both kits were 100% specific for enterococci. The sensitivity of detection of enterococci by SAC and SAF was 96.5% and 97.6%, respectively. False negative reactions occured only in tests performed from primary plates. All strains gave positive reactions when retested from purity check plates. In contrast, the bile-esculin reaction was only 95% sensitive or specific as a test for the presumptive identification of group D streptococci during this study period. Inclusion of 6.5% NaC1 tolerance would have only resulted in the incorrect identification of four strains of Streptococcus intermedius as Streptococcus bavis. During the prospective phase of this study, no nonenterococcal group D streptococci were encountered. Several restrospective studies (Bosley et al., 1983; Oberhofer, 1986) have shown nonenterococcal group D streptococcal strains to be consistently PYRase negative. Identification of enterococci by SAF and SAC was comparable with classical schemes. PYRase testing was possible on growth from primary plates in 77 of 85 strains of enterococci encountered and results were available within 30 min from the time a colony was detected. Tests from both kits were simple to perform and interpret without special training. Strips could be inoculated, set aside, and then batched for later addition of test reagent. Testing was easily incorporated into routine work-flow and was readily accepted by the technologists. The current cost per test is about 30¢ for SAC and 70¢ for SAF. In conclusion, PYRase activity as measured by the SAC or SAF kits is an accurate, cost-effective method for the preliminary or presumptive identification of enterococci, when used in conjunction with Gram stain and colony morphology. REFERENCES Bosley GS, Facklam RR, Grossman D (1983) Rapid identification of enterococci. J Clin Microbiol 18:1275. Ellner PD, Williams DA, Hosmer ME, Cohenford MA (1985) Preliminary evaluation of a rapid colorimetric method for the presumptive identification of Group A streptococci and enterococci, l Clin Microbiol 22:880. Facklam RR (1977) Physiological differentiation of viridans streptococci. ] Clin Microbiol 5:184. Facklam RR, Thacker LG, Fox B, Eriquez L (1982) Presumptive identification of streptococci with a new test system. J Clin Microbiol 15:987. Godsey J, Schulman R, Eriquez L (1981) Annual meeting of the American Society of Microbiologists, abstr C 84, p. 276. Oberhofer TR (1985) Value of the L-pyrrolidonyl-[3-naphthylamide hydrolysis test for identification of select Gram-positive cocci. Diagn Microbiol Infect Dis 4:43.