Nine chemically defined inoculation diluents, with compositions ranging from. 0.85% NaCl to 350/,o marine salts, were used to evaluate the influence of diluent.
APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Aug. 1982, p. 423-427
Vol. 44, No. 2
0099-2240/82/080423-05$02.00/0
Diluent Composition for Use of API 20E in Characterizing Marine and Estuarine Bacteria M. T. MAcDONELL,t* F. L. SINGLETON, AND M. A. HOOD: Department of Biology, The University of West Florida, Pensacola, Florida 32504
Received 9 February 1982/Accepted 27 April 1982
Nine chemically defined inoculation diluents, with compositions ranging from 0.85% NaCl to 350/,o marine salts, were used to evaluate the influence of diluent composition on the biochemical profiles of 30 marine and estuarine bacterial strains, including species of Vibrio, Aeromonas, Allomonas, and Photobacterium. Results demonstrated that a 200/oo marine salts diluent enabled the characterization of halophilic strains normally nonreactive by the API 20E system. Furthermore, the use of 200/oo marine salts showed that certain environmental isolates, identifiable as Vibrio parahaemolyticus by the recommended clinical inoculation procedure, were Vibrio vulnificus. An analysis of the profiles provided by the nine diluents indicates that the API 20E system, modified by the use of a diluent composed of 200/oo marine salts and incubated at 22°C, can provide a reliable tool for the rapid characterization of marine and estuarine bacterial isolates. Several commercial systems for the identification of the Enterobacteriaceae have been available for more than a decade. Generally, these systems consist of a series of biochemical tests contained in miniaturized compartments which are each inoculated with a suspension of bacterial cells in 0.85% NaCl. The impact of such devices has been significant, especially in the clinical diagnostic field for which they were designed (3). One such system, the API 20E system (Analytab Products, Plainview, N.Y.) designed for the identification of gram-negative enteric bacteria, consists of a series of cupules containing appropriate dehydrated media and yields a total of 20 different biochemical tests. These biochemical tests, along with an independent cytochrome oxidase determination, generate a seven-digit numerical profile (or profiles) characteristic of each species (3). A directory of profile numbers, the Analytical Profile Index (Analytab Products), provides an adjunct for species identification. Robertson and MacLowry (17) applied a computer diagnostic model to estimate the relative accuracy of the various identifications included in the API data set and confirmed that the API Analytical Profile Index is accurate in identifying 99.4% of the isolates listed. Numerous other investigations confirming the accuracy and reliability of the API 20E system for the identification of the bacterial t Present address: Department of Microbiology, University of Maryland, Coilege Park, MD 20742. t Present address: Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston, MA 02115.
species included in the API data set have been reported previously (1, 2, 8, 22). Although designed primarily for clinical application, the API 20E system has also been used to a limited degree in the fields of aquatic microbiology and marine microbiology (6, 10, 13, 18, 19). The inclusion of the profiles of approximately 40 groups and species of bacteria in addition to the Enterobacteriaceae in the API data set has promoted the use of this system in areas other than the clinical field. Among the non-Enterobacteriaceae for which profiles have been included in the data base are Vibrio cholerae, Vibrio parahaemolyticus, Vibrio alginolyticus, "Lac' halophilic vibrio" (i.e., Vibrio vul-
nificus), and Aeromonas hydrophila, all of
which are estuarine and aquatic denizens, but also potentially pathogenic to humans. With the current renaissance of the study of the Vibrionaceae, the genus Vibrio has been, and is, undergoing extensive reevaluation (9, 12, 14). This, along with the abolition of the genus Beneckea (4), has resulted in a doubling of the number of species recognized as belonging to the genus Vibrio (4, 9). The use of the API 20E system to generate numerical profiles for other Vibrio species may give investigators a means of phenotypic characterization and comparison with V. cholerae. When a bacterial strain is to be characterized by the API 20E system, the recommended procedure is to prepare a cell suspension in 0.85% saline and to inoculate the various biochemical test cupules. Although such a protocol permits proper identification of enteric bacteria and 423
424
MAcDONELL, SINGLETON, AND HOOD
closely related bacteria, there is no evidence to indicate that environmental isolates, especially those from the marine environment, can be processed in the same manner as clinical isolates. Indeed, there is evidence to the contrary. Seidler et al. (19) and Davis and Sizemore (7) have reported the occurrence of marine isolates which are identifiable as A. hydrophila by the API data set but which have, upon further phenotypic characterization and DNA base ratio determinations, been shown to be, in fact, members of the genus Vibrio. Accounts of the clinical misdiagnosis of V. vulnificus ("Lac' halophilic vibrio") in human infections as V. parahaemolyticus are cited by Blake in a review of pathogenic vibrios (5). Whether the API 20E system can provide a valid phenotypic profile of halophilic and haloduric bacteria depends upon whether the enzymatic reactions of the particular biochemical pathways are able to function in a medium virtually free of cations other than Na+ and in which Na+ itself is present at only a fraction of the concentration which the organism would encounter in the environment. The purpose of this study was, therefore, to evaluate the applicability of the API 20E system (in conjunction with selected diluents) for use in identifying marine and estuarine bacterial isolates. MATERIALS AND METHODS Bacterial strains and identification system. Thirty selected bacterial strains representing the Enterobacteriaceae, Vibrionaceae, and Pseudomonadaceae were used to evaluate the API 20E system and the influence of diluent composition. Of the bacterial strains tested, 27 were estuarine isolates, identified by the protocols recommended by Fumiss et al. (9) and by Hugh and Gilardi (11) for the Vibrionaceae and Pseudomonadaceae, respectively. o-Nitrophenyl-p-Dgalactopyranosidase (ONPG) was assayed in the presence of toluene (15). Three strains, Serratia marcescens ATCC 8100, Vibrio metschnikovii PICC 61, and Vibrio anguillarum PICC 62, were obtained from stock culture collections (American Type Culture Collection, Rockville, Md., and Presque Isle Culture Collection, Presque Isle, Pa.). All cultures were maintained at room temperature (22°C) on seawater complete agar (16) with the salinity adjusted to 150/00 and incubated in the API 20E system for 24, 36, and 48 h, and the reactions were recorded. Duplicate API 20E strips were inoculated and incubated simultaneously. Chemicals and marine salts mix. All chemicals were Merck Blue Label (Merck and Co., Inc., Rahway, N.J.) or when supplied by another manufacturer of comparable purity (analytical grade). Marine salts mix (Instant Ocean) was purchased from Aquarium Systems, Mentor, Ohio. Double-distilled deionized water was used in all diluents. Diluent compositions. The nine chemically defined diluents which were used consisted of three groups. Group 1 was composed of three NaCi solutions: 0.85, 2.0, and 3.5%. Group 2 (marine salts) was composed of solutions of marine salts adjusted to salinities of 8.5,
APPL. ENVIRON. MICROBIOL. 20, and 350/O,* Group 3 represented three major cations in concentrations approximately equivalent to the concentrations of those cations encountered in seawater of a salinity of 20o/oo. These were (i) 15.49 g of NaCl per liter-0.434 g of KCI per liter, (ii) 15.49 g of NaCl per liter-2.89 g of MgCl2 per liter, and (iii) 15.49 g of NaCl per liter-0.434 g of KCI per liter-2.89 g of MgCl2 per liter. The API 20E strips were inoculated according to the prescribed technique of the manufacturer. Controls were established by filling the cupules of each of nine API 20E strips with only the diluent to detect diluent-reagent interactions. Scoring method and data analysis. All data, a total of 10,800 observations (30 strains x 20 biochemical tests x 9 diluents x 2 parallel trials), were stored on an IBM (White Plains, N.Y.) diskette and processed with an SAS (SAS Institute, Inc., Cary, N.C.) general linear model on an IBM System 370 series 3033 computer. The ideal diluent(s) for each species and for each biochemical test was determined by use of a binary scoring method in which a positive result was assigned a value of 1.0 and a negative result was assigned a value of 0.0, regardless of whether a trait was characteristically positive or negative for a particular strain. This ensured that all observations were equally weighted. The determination of an optimal diluent(s) and a minimal diluent(s) for each species was made by calculating the mean score (i.e., percent positive) of all 20 tests for each strain for each diluent and selecting the largest and smallest values. The determination of an optimal diluent(s) and a minimal diluent(s) for each biochemical test was made by calculating the mean score (i.e., percent positive) of all 30 strains for each biochemical test for each diluent and selecting the largest and smallest values.
RESULTS AND DISCUSSION A series of API 20E systems was inoculated with cell suspensions of environmental and reference bacterial strains prepared with diluents of different ionic composition. To evaluate which diluent was the most appropriate for each strain tested (regardless of the biochemical test), the proportion of the total number of biochemical tests which were determined to be positive for each strain for each diluent was calculated, and these values were compared (Table 1). Not only was there variation in the proportion of positive tests per strain within each group of diluents, but there were also variations among the three diluent groups. The largest proportion of positive tests was consistently detected in the presence of the marine salts diluent group. The marine salts diluent prepared with the salinity adjusted to 20O/00 yielded the largest fraction of positive tests. Conversely, the 0.85% NaCl diluent (diluent recommended by the manufacturer for clinical diagnosis) consistently yielded the smallest proportion of positive tests. To evaluate which diluent was the most appropriate for each individual biochemical test (regardless of strain), the proportion of the total
425
CHARACTERIZING MARINE BACTERIA WITH API 20E
VOL. 44, 1982
TABLE 1. Influence of diluent on percentage of positive reactions for each species
Organism
Aeromonas sp ...................... Aeromonas hydrophila (2) ........... Allomonas sp. strain a .............. Allomonas sp. strain b ..............
Vibrio cholerae (4) .................. V. parahaemolyticus ................ V. vulnificus (4) .................... V. harveyi (4) ...................... V. fischeri ......................... V. mimicus ... ....................... V. metschnikovii .................... V. anguillarum ..................... Photobacterium phosphoreum (2) .... Pseudomonas sp. (2) ................ Pseudomonas putrifaciens (3) ........ Serratia marcescens ........ ........
Reaction' with following diluent: Major cation group Marine salts group NaCI NaCI KCI0.85% 2.0% 3.5% 8.5-/° KCIb MgClIN MgCIKd 200/,,, 350/
NaCI group
0.25 0.45 0.15 0.30 0.40 0.35 0.35 0.33 0.35 0.20 0.20 0.35 0.08 0.05 0.20 0.45
NaCi
0.20 0.38 0.20 0.40 0.35 0.40 0.30 0.45
0.40 0.45 0.36 0.33 0.45 0.40 0.20 0.20 0.25 0.05 0.15 0.35
0.38 0.40 0.40 0.33 0.45 0.40 0.20 0.20 0.25 0.05 0.15 0.35
0.25 0.43 0.40 0.45 0.49 0.55 0.40 0.36 0.40 0.40 0.25 0.35 0.23 0.03 0.23 0.55
0.45 0.35 0.45 0.40 0.50 0.55 0.49 0.46 0.50 0.50 0.30 0.55 0.45 0 0.25 0.50
0.35 0.50 0.55 0.40 0.41 0.50 0.43 0.44 0.60 0.60 0.20 0.35 0.475 0 0.22 0.40
0.30 0.40
0.30 0.35
0.45 0.25 0.31 0.45 0.39 0.36 0.45 0.40 0.10
0.40 0.30 0.41 0.45 0.36 0.31 0.45 0.45 0.15
0.25 0.25 0.03 0.17 0.45
0.45
0.35 0 0.08 0.40
0.35 0.48 0.40 0.40 0.41 0.55 0.39 0.33 0.50 0.40 0.25 0.45 0.30 0.03 0.17 0.50
Diluent performance'
1 0 5 0 2 8 1 1 0 4 1 3 1 0 2 3 Minimal ......................... 6 a For example, 0.25 = 25% (i.e., total number of positive reactions/total number of tests). b Diluent composed of 15.49 g of NaCl per liter and 0.434 g of KCI per liter. c Diluent composed of 15.49 g of NaCl per liter and 2.89 g of MgCl2 per liter. d Diluent composed of 15.49 g of NaCl per liter, 0.434 g of KCI per liter, and 2.89 g of MgCl2 per liter. 'Values indicate the number of times that the diluent proved to be the best (optimal) or worst (minimal)
Optimal ......................... 1
performer.
number of strains producing a positive reaction per test for each diluent was calculated, and these values were compared (Table 2). Variations in the proportion of positive strains per test within each group of diluents, as well as variations among the three diluent groups, were detected. Also, interactions between the diluent(s) and several of the biochemical tests were detected. As in the case of the optimal diluent(s) for each strain, the largest proportion of positive strains was consistently detected in the presence of the marine salts diluent with the salinity adjusted to 20°o4,. The 0.85% NaCl diluent yielded the smallest proportion of positive results. Interactions between diluent(s) and biochemical tests (false-negatives or false-positives) were observed in the case of the 350/.0 marine salts diluent. These interactions were particularly pronounced for the following biochemical tests: (i) tryptophan deaminase, for which all strains are negative; (ii) indole, for which one strain, S. marcescens, is negative; (iii) Voges-Proskauer reaction, for which only six strains (20%) are positive; and (iv) gelatinase, for which several strains are negative. Also, there were detectable interactions between the indole test and both the 3.5% NaCl diluent and the two NaCl-MgCl2 diluents. Spe-
cific reasons for these interactions are unknown. None of the three interacting diluents was considered in performance comparisons. The biochemical profiles of some bacterial strains, V. cholerae and A. hydrophila among others, were found to be unaffected by the diluent composition (Table 1). For this reason, it may be assumed that for certain estuarine isolates the selection of a diluent composed of either physiological-strength saline (0.85% NaCI) or 20O/0, marine salts may be arbitrary. The biochemical profiles of other strains, however, were markedly influenced by the diluent composition. This was apparent in the 0.85% NaCl diluent systems.
Indications of the influence of the diluent composition on the biochemical characterization of marine bacteria are apparent in the profiles of Photobacterium phosphoreum and Allomonas spp. (12) (Table 3). P. phosphoreum is nonreactive in the API 20E system in the presence of the 0.85% NaCl diluent. However, 7 of the 20 tests were positive in the presence of a 20O/0, marine salts diluent. Likewise, six positive biochemical traits of Allomonas spp. were inapparent in the absence of marine salts. The phenotypic characterization of V. vulnificus in the presence of 0.85% NaCl was identical
MAcDONELL, SINGLETON, AND HOOD
426
APPL. ENVIRON. MICROBIOL.
percentage of positive reactions for each biochemical test Reaction' with following diluent: NaCl group Marine salts group Major cation group
TABLE 2. Effect of diluent
on
Biochemical test
3.5%
8.5-/,,,
200/.,
0.56
0.71
0.53
0.67
0.72
0.52
0.50
0.62
0.11
0
0.19
0.29
0.17
0.10
0.21
0.19
0.85%
2.0o
ONPG ...................
0.48
Arginine dihydrolase
0.14
.......
Lysine decarboxylase
350/
NaCI
NaCl
NaCI
KCd-
0.52
0.67
0.57
0.48
0.52
0.67
0.05
0.57
0.52
....
0.62
0.67
0.57
0.57
0.67
0.61
0.57
0.64
0.62
...................
Ornithine
...... decarboxylase
0.38
0.22
0.14
0.57
0.71
0.72
0.29
0.21
0.48
H2S ...................
0.14
0
0
0.19
0.14
0.28
0.14
0
0.19
Urease
0
0
0
0
0.05
0.11
0.05
0.07
0.05
0
0
0
0
0
0.06
0
0
0
Indole ...................
0.76
0.89
1.0
0.76
0.76
1.0
0.71
1.0
0.76
Voges-Proskauer
0
0
0
0.19
0.19
0.39
0
0.07
0.10
Citrate
...................
Tryptophan deaminase Gelatinase
.....
..........
................
0.71
0.89
0.71
0.86
0.90
1.0
0.89
0.86
0.90
..................
0.57
0.89
0.57
0.71
0.76
0.94
0.71
0.86
0.71
Mannitol ..................
0.52
0.89
0.71
0.71
0.76
0.89
0.71
0.86
0.76
Inositol ...................
0.09
0.22
0.29
0.05
0.09
0
0.05
0.07
0.05
0.09
0.22
0.29
0.10
0.14
0.11
0.10
0.14
0.10
0
0
0
0
0
0
0
0
0
Glucose
Sorbitol
..................
Rhamnose
................
Saccharose
0.48
0.22
0.43
0.48
0.57
0.44
0.48
0.36
0.48
Melibiose .................
0
0
0
0
0.05
0.11
0
0
0.48
Amygdalin
................
0.38
0.67
0.57
0.62
0.57
0.78
0.57
0.79
0.57
................
0.19
0.22
0.29
0.33
0.28
0.17
0.29
0.43
0.33
1
8
NEe
3
8
NE
1
NE
4
Minimal.8
5
NE
4
0
NE
5
NE
1
Arabinose
...............
Diluent performance Optimal
................
For example, 0.48 48% (i.e., total number of positive reactions/total number of species). b Diluent composed of 15.49 g of NaCl per liter and 0.434 g of KCI per liter. C Diluent composed of 15.49 g of NaCl per liter and 2.89 g of MgCl2 per liter. d Diluent composed of 15.49 g of NaCl per liter, 0.434 g of KCI per liter, and 2.89 g of MgCl2 ' NE, Not evaluated (see text). a
=
per
liter.
TABLE 3. Phenotypic profiles of A. hydrophila, V. cholerae, V. vulnificus, P. phosphoreum, and Allomonas spp. in the presence of 0.85% NaCl and 20O/O. marine salts diluents Phenotypic profile' V. cholerae V. vulnificus P. phosphoreum Allomonas spp. A. hydrophila Biochemical test (WFM504) (WFM401) (1B39) (WF110r) (WF8AIIIO) 20°/00 0.85% 0.85% 2000 0.85% 0.85% 200/,,, 0.85% Marine Marine Marine NaCI Marine NaCI NaCI NaCI Marine NaCl salts salts salts salts salts W + + + + ONPG ................... + + + + + Arginine dihydrolase ...... + + + + + W Lysine decarboxylase ..... + + + + Ornithine decarboxylase... W + W + + Citrate .................. +
200/,..
200/,,
H2S
-
-
.Urease -I-ITryptophan deaminase
-
Indole ..................
+
.....................
-
-
-
-
-
-
-
-
+
+
-
+
+ Voges-Proskauer .........--
-
+
-
+
+
-
+
+
+
+
+
Gelatinase ...............
+
+
+
W
+
+
-
Glucose .................
+
+
+
+
+
+
-
+
Mannitol ................. + + + + + + + Inositol. ................Sorbitol. ................Rhamnose ............... + + + Saccharose .............. + + + + Melibiose.W + + + + + Amygdalin ............... + + Arabinose ............... + + a +, Positive; -, negative; W, weak or delayed positive.
VOL. 44, 1982
CHARACTERIZING MARINE BACTERIA WITH API 20E
to that expected of V. parahemolyticus (Table 3). Indeed, the API numerical profile generated from these results (4146106) is listed in the API Analytical Profile Index as "V. parahaemolyticus-excellent identification." Such misidentification by the API 20E system is accounted for by the fact that V. vulnificus and V. parahaemolyticus are differentiated by the ONPG reaction.
4.
V. vulnificus yields a positive ONPG reaction in
6.
the API 20E system in the presence of marine salts, but not in the presence of 0.85% NaCl. The observed variations in phenotypic characterization cannot be directly attributed to either the presence or the absence of particular ions or salinity. Thus, it appears that the success or failure of a particular diluent is dependent upon a complex relationship between the type and the concentration of ions present as well as the overall salinity. Naturally, a bacterial strain incapable of a particular enzymatic process will not acquire that capability as a result of the substitution of one ionic environment for another. However, an organism genetically competent for a particular enzymatic pathway may be limited in its expression should ions required by a particular enzyme, coenzyme, or second messenger(s) (20, 21) be absent or in prohibitively low concentrations. For these reasons and on the basis of the results presented herein, the use of a diluent composed of marine salts adjusted to a salinity of 200/oo is recommended when using the API 20E system to rapidly characterize isolates of estuarine and marine origins. The use of diluents of higher salinities for use with halophiles may not be judicious in light of the interactions between high salinities and test reagents. Further study to determine the nature of the specific interactions must be recommended before an attempt is made to use salinities higher than 20°o4,. Consideration must also be given to the practice of referring to the Analytical Profile Index for identification of marine and estuarine environmental isolates, since this data base was not developed for these bacteria. This is especially true if a marine salts diluent is substituted, since by doing so one changes the basic premise upon which the API data set was compiled. ACKNOWLEDGMENT We acknowledge Analytab Products for supplying the API 20E system used in this study. LITERATURE CITED 1. Aldridge, K. E., B. B. Gardner, S. J. Clark, and J. M. Matsen. 1978. Comparison of Micro-ID, API 20E, and conventional media systems in identification of Enterobacteriaceae. J. Clin. Microbiol. 7:507-513. 2. Aldridge, K. E., and R. L. Hodges. 1981. Correlation studies of Entero-Set 20, API 20E, and conventional
3.
5.
7.
8.
9. 10.
11.
12.
13.
14.
15. 16.
17.
18.
427
media systems for Enterobacteriaceae identification. J. Clin. Microbiol. 13:120-125. Analytab Products. 1980. The analytical profile index. Analytab Products, Plainview, N.Y. Baumann, P., L. Baumann, S. S. Bang, and M. J. Woolkalis. 1980. Reevaluation of the taxonomy of Vibrio, Beneckea, and Photobacterium: abolition of the genus Beneckea. Curr. Microbiol. 4:127-132. Blake, P. A. 1980. Diseases of human (other than cholera) caused by vibrios. Annu. Rev. Microbiol. 34:341-367. Colwell, R. R., J. Kaper and S. W. Joseph. 1977. Vibrio cholerae, Vibrio parahaemolyticus, and other vibrios: occurrence and distribution in Chesapeake Bay. Science 196:394-396. Davis, J. W., and R. K. Sizemore. 1981. Nonselectivity of Rimler-Shotts medium for Aeromonas hydrophila in estuarine environments. AppI. Environ. Microbiol. 42:544545. Edberg, S. C., B. Atkinson, C. Chambers, M. H. Moore, L. Palumbo, C. F. Zorzon, and J. M. Singer. 1979. Clinical evaluation of the Micro-ID, API 20E, and conventional media systems for identification of Enterobacteriaceae. J. Clin. Microbiol. 10:161-167. Furlss, A. L., J. V. Lee, and T. J. Donovan. 1978. The vibrios. Public Health Laboratory Services Board, Her Majesty's Stationery Office, London. Hood, M. A., G. E. Ness, and G. E. Rodrick. 1981. Isolation of Vibrio cholerae serotype 01 from the eastern oyster, Crassostrea virginica. Appl. Environ. Microbiol. 41:559-560. Hugh, R., and G. L. Gilardi, 1974. Pseudomonas, p. 250269. In E. H. Lennette, E. H. Spaulding, and J. P. Truant (ed.), Manual of clinical microbiology, 2nd ed. American Society for Microbiology, Washington, D.C. Kalina, G. P., A. G. Somova, L. S. Podosinnikova, and T. I. Grafova. 1978. Allomonas, a new group of microorganisms of the family Vibrionaceae. Z. Mikrobiol. Epidemiol. Immunobiol. 57:40-46. Kaper, J., H. Lockman, R. R. Colwell, and S. W. Joseph. 1979. Ecology, serology, and enterotoxin production in Vibrio cholerae in Chesapeake Bay. Appl. Environ. Microbiol. 37:91-103. Kaper, J., R. J. Seidler, H. Lockman, and R. R. Colwell. 1979. Medium for the presumptive identification of Aeromonas hydrophila and Enterobacteriaceae. AppI. Environ. Microbiol. 38:1023-1026. Lennette, E. H., E. H. Spaulding, and J. P. Truant (ed.). 1974. Manual of clinical microbiology, 2nd ed. American Society for Microbiology, Washington, D.C. Nealon, K. H. 1978. Isolation, identification and manipulation of luminous bacteria. Methods Enzymol. 42:153166. Robertson, E. A., and J. D. MacLowry. 1974. Mathematical analysis of the API Enteric 20 profile register using a computer diagnostic model. Appl. Microbiol. 28:691-695. Schmidt, U., H. Chmel, and C. Cobbs. 1979. Vibrio alginolyticus infections in humans. J. Clin. Microbiol.
10:666-668. 19. Seider, R. J., D. A. Allen, R. R. Colwell, S. W. Joseph, and 0. P. Daily. 1980. Biochemical characteristics and virulence of environmental group F bacteria isolated in the United States. Appl. Environ. Microbiol. 40:715-720. 20. Shamoo, A. E., and D. Goldstein. 1977. Isolation of inophores from ion transport systems and their role in energy transduction. Biochim. Biophys. Acta 472:13-53. 21. Stryer, L. 1981. Biochemistry. W. H. Freeman and Co., San Francisco. 22. Washington, J. A., II, P. K. W. Yu, and W. J. Martin. 1971. Evaluation of accuracy of multitest micromethod system for the identification of Enterobacteriaceae. AppI. Microbiol. 22:267-269.
