Cross-Resistance Studies and Antibiotic Identification'E. 0. STAPLEY

Merek Sharp & Dohrne Research Laboratories, Rahway, New Jersey

Received for publication March 28, 19.58

Considerable attention has been directed over thelast decade toward methods for the rapid identificationof antibiotics in culture broths. The need for such pro-cedures is obvious since extraction and purificationof antibiotic substances involves much expenditure oftime and effort and thus should be reserved for newand potentially interesting materials. Numerous re-ports on the use of paper-strip chromatography haveappeared; however, very little information concerningpractical application of antibiotic-resistant cultures forthis purpose has been disseminated. Stansly (1946)and Eisman et al. (1946) independently made the sug-gestion that antibiotic-resistant cultures would be avaluable tool for evaluation of crude antibiotics. Crossresistance may be defined as the relationship betweentwo antibiotics which is revealed by the increase inresistance to one antibiotic which develops when atest microorganism becomes specifically resistant to asecond antibiotic. The complexity of the interrelation-ships of antibiotics from the standpoint of developmentof resistant cultures has been studied in detail bySzybalski (1953, 1954) and Szybalski and Bryson (1952,1954). Since cross-resistance testing may be an im-portant tool for the identification of unknown crudeantibiotics, we consider it worthwhile to detail some ofour observations and experiences with antibiotic identi-ficationl.

METHODS AND RESULTS

All of the resistant cultures described in this reporthave been derived from a single sensitive parent,Escherichia coli W. The medium employed in this workconsisted of Difco nutrient broth plus 0.2 per cent yeastextract, with 1.5 per cent agar added when solid me-dium was required. Except in the case of disc-assayplates to be described later, the organisms were culti-vated at 37 C in tubes of liquid medium.

Produtction of resistant cultures. The simplest methodof obtaining a resistant culture is to select it by directisolatioin from the inhibition zone produced by an anti-biotic-soaked disc on a seeded agar plate. This tech-nique can only be applied in the case of antibiotics,such as grisein, which elicit, in a single step, a highorder of resistance.

1 Presented at the 57th General Meeting of the Society ofAmericaIn Bacteriologists, Detroit, Michigan, April 2k to May2, 1957.

392

Serial passage of E. coli through sets of nutrientbroth tubes fortified with graded concentrations ofantibiotic has been a useful procedure for selection ofresistant strains. By means of serial tube concentrationsit is possible to select strains of 100-fold or greater re-sistance over the parent type, even with materials suchas polymyxin B to which a high degree of resistanceis not easily developed. The advantages of the, tubemethod are obvious. One selects from the entireinoculum population and there is opportunity for de-bilitated cultures, which might not form visible colQnies,to develop a slight turbidity. The method, however,suffers from three chief disadvantages. The first of theseis that a large quantity of medium and of antibiotic isrequired, coupled with considerable time and work forthe development of each highly resistant culture. Thesecond important drawback is that the actual numberof live cells present in the inoculum for each tube of aconcentration series may be quite small. The thirddisadvantage is the fact that only a single concentra-tion of antibiotic, which may not be at the criticallevel for selection of resistance, is present in each tubeof a series.To realize the advantages of tube concentration

series while eliminating the disadvantages of this pro-

Figutre 1. A modified chemostat for continuous culture. (1)Reservoir of sterile medium; (2) side arm for addition of in-oculumn and antibiotic solution; (3) capillary tube for sterileair; (4) growth vessel (suspended in 37 C water bath, notshown); (5) capillary siphon tube for overflow; (6) graduatedvessel to receive effluent; (7) four steel rollers which "milk"the tubing by slowly revolving; (8) a 400-1 gear ratio reductiongear; (9) a zero to full speed torque converter; (10) a TA0 hpmotor.

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANTIBIOTIC IDENTIFICATION

cedure, a technique of continuous culture was employed.A modified chemostat has been assembled (figure 1)which involves a reservoir of sterile medium, a closedsystem pump, and a growth vessel (resting in a 37 Cwater bath) which contains a constant volume ofapproximately 15 ml of medium stirred by bubbles ofsterile air. The entire apparatus, exclusive of the pump,can be sterilized by autoclaving as a unit. Inoculationand addition of sterile antibiotic to the growing cultureis accomplished via a side arm on the growth vessel.Once an actively growing culture has been established,a concentration of antibiotic is added periodically at alevel above that which will inhibit growth of the culture.As fresh medium enters the growth vessel, the concen-tration of antibiotic drops very slowly permitting themost resistant portion of the population to grow andestablish itself as the dominant population. A sampleof theJ effluent obtained daily from the growth vesselis used to inoculate an assay plate for a test of anti-

TABLE 1Antibiotic resistance developed in Escherichia coli W

Maximum Conc Permitting Growth (ug per ml)*Antibiotic

E. coli XV' strain Fold increase

Streptomycini......... 1.0 1600 1600Streptothricin ......... 1.0 200 200Cycloserine ........... 16.0 2000 125Pleocidin ............. 0.8 2000 2500Chloramphenicol ...... 2.0 500 250Chlortetracycline ...... 0.4 100 250Oxytetracycline ....... 0.4 150 375Neomycin 0.4 >200 500Tetracycline .......... 0.8 200 250Viomycin ............. 8.0 > 1000 125Polymyxin B 0.8 400 500

* The inhibitory level of each antibiotic for the sensitivend resistant strains was determiined by tube dilution assay.

biotic resistance and is streaked out for a check of cul-ture identity and possible isolation. Flow rates of 1.5to 3.0 ml per hr have been used. With this technique, itis possible to select a resistant culture in a short timewith relatively small quantities of antibiotic and verylittle effort.To carry out cross-resistance tests on solid medium

by the disc-assay agar-plate technique, it is necessary,in most instances, to attain a resistance level of 100-fold or greater. Examples of the degree of resistancedeveloped to soine of the well-knowin antibiotics arepresenited in table 1.

MIaintenance of resistant cultures. Once a sufficientdegree of resistance has been attained, the culture ispurified by single-colony isolation. Storage of resistantstrains proved to be a problem in many cases, since re-version to greater sensitivity occurred readily. Tocircumvent this problem, it has been necessary to lyo-philize the fresh isolates. Each time a culture is re-quired for testing purposes it is cultured in nutrientbroth directly from a newly opened lyophilizationtube, thus minimizing any tendency for change in re-sistance levels or cross-resistance characteristics whichwould result from serial passage.

Cross-resistance testing. Cross resistanice is me.Asuredby comparison of inhibition zone diameters producedby 18 min paper discs soaked with the antibiotic andincubated on plates seeded with E. coli WV and theresistanit strains of E. coli W. A standard aImount of a24 hr broth culture, adjusted to a standard opticaldensity, is used to seed the assay plate medium at, 49 C.I'lates are poured with only 5 ml of seeded agar andincubated at, 25 C, except for temperature-dependentstrains which re(juire incubationi at 37 C. Thin agar,low incubation temperature, and light inoculum com-bine to provide a very sensitive assay. A difference of5 to 6 imm in zone diameter is considered indicative of

Figure 2. Photograph of comparisons of sensitive and resistant cultures in cross-resistance assay. A disc-plate assay of five anti-biotic broths on the sensitive parent Escherichia coli W (left) and antibiotic-resistant strain (right). The resistant strain is partiallyresistant to two of these products and completely resistant to the other three antibiotics. The resistant colonies present in two ofthe zones on each plate constitute a useful characteristic for presumptive classification of grisein-like antibiotics.

19581 393

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from

E. O. STAPLEY

cross resistance. In the case of some of the more highlyresistant cultures, a difference of 10 mm or more can bedemonstrated with the specific antibiotic to which thetest strain is resistant. The results of such a comparisonare presented in figure 2.

There is no special advantage to the use of a par-ticular size paper disc. Table 2 presents a comparisonof the use of 13 mm discs with 6.5 mm discs. In general,the inhibition zone minus the disc size is roughly equaland the same conclusion can be drawn from either data.The use of double discs, one on top of another, did notresult in increased zone sizes. It, has been possible toobtain increased zone sizes by prolonged diffusion time,storage 24 to 48 hr at refrigerator temperature priorto incubation. It is also possible to obtain cross-re-sistance tests of antibiotics produced by culturesgrowing oIn a nutrient agar in l'etri plates using thetechnique described by Harris and Ruger (1953). Pllugsof the agar can be cut with a cork borer and used inplace of paper discs. This may be an advantage in thecase of antibiotic-producing cultures which do notreadily produce antibiotic activity in shake flaskculture.

Resistant cultures compared to sensitive parent. Thenature of the resistance manifested by the antibiotic-resistant cultures must be viewed as quantitative inthe sense that it usually represents a shift in the in-hibition curve so that the culture is inhibited bya higherconcentration of the antibiotic. In some instances, thelevel of resistance may be so high that, for practicalpurposes, the resistant culture is indifferent to thepresence of the antibiotic. In other cases a "qualitative"difference in response has been noted, as evidenced by achange in the slope of the inhibition curve. Resistantcultures may show a changed response to other struc-turally related or unrelated antibiotics so that theyappear to be resistant to them or, in some cases, hyper-sensitive. These different reactions are demonistratedin response of a number of antibiotic resistant strains

TABLE 2

Conmparison of two disc sizes for cross-resistance assay

Zone Diameters with Streptomycin(5 pg per ml)

Test Culture 13 mm Disc

mZonmin Zone-disc

Escherichia coli W.. 19.5 6.5E. coli strains resistant to:Streptomycin 0Streptothricin 0Cycloserine ............. 21.0 8.0Pleocidin .............. 0Chlortetracycline ...... 20.3 7.3Viomycin ..... 20.3 7.3Gvrisein ..... 19.0 6.0

6.5 mm Disc

Zone inmm IZone-disc

__- _--

13.0 6.5

0

0

14.0 7.5I

0

13.0 6.514.0 7.513.0 6.5

of E. coli to chloramphenicol (table 3). If the log of thechloramphenicol concentration is plotted against thediameter of the inhibition zones, a family of curves isproduced. All of the curves except for neomycin andviomycin resistant cultures have a similar slope, with ahigher or lower intercept depending upon the degree ofresistance. Some of these data are presented graphicallyin figure 3 to illustrate that in a plate assay of cross re-sistance, the scoring of a particular result may be de-pendent upon the test concentration. To obtain a validcomparison between two antibiotics by cross resistance,they should be tested at concentrations which produceinhibition zones of similar size against the sensitive-par-ent assay organism.

It has been noted that in many cases the develop-ment of resistance is accompanied by a considerabledebilitation of the culture. Some of the resistant strainsisolated have been extremely weak-growing culturesby comparison with the parent. Most of these weak-growing resistant strains have proved to be tempera-ture-dependent. It is necessary to cultivate such strainsat 37 C. Even when these temperature-dependentstrains are cultivated at 37 C, they tend to revert tosensitive strains, which have a tremendous growth rateadvantage, growing quite rapidly. In practice thisproblem of reversion is avoided by permitting only aminimum amount of growth to take place before usingthe cultures in cross-resistance tests.

Cross-resistance data. The zone diameters from cross-resistance tests are reproducible plus or minus 1 mm or

TABLE 3

Comparison of chloramphenicol resistance by paper disc assay

Test Culture

Escherichia coli W.........E. coli strain resistant to:

Streptothricin ..........

Pleoci din ................Grisein ..................Streptomycin ............

Cycloserine ..............Oxytetracycline ..........

Zone Diameters (mm) with Chlorampheni-col in pg per ml:

10 20 50 100

0

0000

00

13.3

13.013.013.7

0

00

Chloramphenicol 0 0Chlortetracycline 0° 0Tetracycline 0 0

Neomvcin ................19.7 21.3Viomycin .............|.20.0 22 0Polvmvxin B ......... j17.0 I 22.7

19.0

18.020.720.317.0

14.30

000

22.722.726.7

25.0 28.7

23.026.724.723.0

28.731.727.328.3

22.3 26.318.2 22.3

000

24.324.729.3

000

28.726.731.3

The responses of this series of resistanit cultures to chlor-amphenicol can be divided into four groups as above. From topto bottom these are (1) essentially the same as the parent strain(E. coli W), (2) somewhat resistant to the antibiotic, (3) veryresistant, and (4) hypersensitive.

[VOL. 6394

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANTIBIOTIC IDENTIFICATION

less from disc to disc. On the basis of examination ofdata obtained in cross-resistance testing, a difference of5 mm in zone diameter was selected as a reasonableindication of cross resistance. Using this criterion, apattern of cross resistance is established for known anti-biotics. To obtain reproducible results it is essentialthat procedures used be standardized very carefully.Since a wide range of antibiotics is examined in thistechnique, it is not possible to include an internalstandard on each assay plate. For this reason, thetreatment accorded each assay plate must be care-fully controlled to avoid differences in inhibition zonesizes which can be produced by variation in temperatureor length of preincubation period (time lapse before theantibiotic discs are set on the plates). If the plates arerefrigerated immediately after pouring and are placedat incubation temperature as soon as discs are in posi-tion, variations of this sort are held to a minimum. Thebehavior of the resistant cultures can be checked foreach experiment by including a standard series of anti-biotics for which the pattern of response has been wellestablished.

Since there are so many interrelationships betweeniantibiotic resistant cultures, finding that a particularunknown is cross resistant with one of the knowns isnot enough information to identify the material in-volved. There are families of antibiotics which arereciprocally cross resistant, as well as numerous oneswith one-way cross resistances, so that it is necessaryto obtain a cross-resistance spectrum or pattern. Table

100.

NO. ZONES

R -6.50 R I9

z~~~~~~~~~

lo -7

0. -

12 14 16 18 20 22 24 26ZONE DIAMETERS (mm.)

28 30 32

Figutre S. Chloramphenicol resistance by paper-disc assay.Graphs of inhibition zone sizes vs. chloramphenicol concentra-tions with several antibiotic-resistant strains of Escherichiacoli W. This is a graphic representation of the results obtainedby disc-plate assay of chloramphenicol. Reading from left toright the curves represent the responses of strains resistant tooxytetracycline (1R-7); cycloserine (1-3); the sensitive parent,pleocidin (R-4); neomycin (R-8); viomycin (R-10); and poly-myxin B (R-11). The curves for cultures resistant to grisein(R-12), streptomycin (R-1), and streptothricin (R-2) coincidewith that for the parent culture. No inhibition is demon-strated against cultures resistant to chloramphenicol (R-5),chlortetracycline (1-6), or tetracycline (R-9) at these concen-trations.

4 presents a picture of the interrelationships of 12 aniti-biotics with their 12 specifically resistant strains ofE'. coli. In the case of these aintibiotics there are twomain groups. The streptothricin group, to wvhich strep-tomycin and cycloserine are closely related, anid thetetracycline group to which chloramphenicol is closelyrelated. Grisein is the most distinctive of the actinomy-cete products and thus it is easily recognized. PolymyxinB, while it would be easily picked out, shows an unex-pected amount of cross-resistance relationship to thestreptothricin group, all of which is one-way in con-trast to the reciprocal relationships within the twomajor groups. If only these 12 products are considered,it is relatively easy to pick out streptomycin,cycloserine, grisein or polymyxin B, but subdivision ofthe streptothricin or tetracycline groups would be im-possible without careful titration of cross-resistancelevels, anl impractical procedure for routine identifica-tion tests. In table 5, the relationship of 8 other aniti-biotics to these same 12 aintibiotic resistant strains ofA'. coli WV are presented. Three of these products,soframycin, streptini, and streptolin A and B, are re-lated to the streptothricin group; two, hygromycin and

TABLE 4Summnlary of the results of disc plate assay of 12 atntibiotics on

the sensitive parent culture Escherichia coli W and thestrains resistant to each of the antibiotics

Antibiotics*

Test Culture .E

Escherichia coliW-sensitiveparent. -

E. coli resistantto

Streptomycin... +Streptothricin.. 41:Pleocidin ....... +Neomycin ...... +Viomycin ....... +Cycloserine....Chlorampheni-

col ...........Chlortetracy-

cline...-.......-Oxytetracy-cline.

TetracyclineGrisein .........-Polymyxin B.. +

.0

.c0

0.sI.4@

++++

+

.0

._1cv

+

_

0

z._I

+++

4-

E

i:

c

0

>5

u

+

+

.0

oQ

._

z

ECd

zo

+-

+-

++1

.z~

Is

1-10

u

+

+l

r.

_,Xd

c~

(L

+

++c

-4

_4 _-

_-

+

+X++- ++E+

* + = definite cross resistance, i = borderline cross re-

sistance, - = sensitivity equal to or greater than that of theparent strain..

t Streptothricin group.t Tetracycline group.

1958] 395

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from

E3 F. STAPLEY

xanthomycin, to the tetracycline group; netropsin isintermediate; albomycin belongs to the grisein group;and thioaurin is classed by itself.An example of the remarkable specificity which can

be found in particular resistant strains is evidenced bythe fact that, with a reisolated set of resistant cultures,it is possible to distinguish between most of the mem-bers of the closely related tetracycline group (table 6).This set of resistant strains was prepared from theoriginal series of cultures, which had lost some of their

TABLE 5

Cross-resistance patterns of 8 antibiotics obtained with theseries of 1.2 antibiotic-resistant strains of Escherichia

coli IV

Antibiotics*

Escherichia coli WV Resistant to:

Streptomycin ...........

Streptothricin ..........

Pleocidin ...............

Neomvcini ..............

Viomvcin ...............

Cycloserine .............

Chloramphenicol........Chlortetracycline .......

Oxytetracycline...... .. .

Tetracyeline ............

Grisein .................Polymyxin B...........

._

._

0

4+

* +, A:, - as in table 4.

t Streptothricin group.I Tetraeyeline grouip.

= P0._ ;0

_d0

+

-4

A4:

A:I4:

._

S+

+:4i

c

co0

cn

+:

+

4-

_

+4-

._

a.0

+

+.

±V

6.E

.cl

+c.

A:eA2

_: __ _

4__4--7 i4

._.c0

.0-o

+

TABLE 6Cross-resistance patterns of 4 antibiotics of the tetracycline

cross-resistance group obtained with a reisolated set ofantibiotic -resistant strains of Escherichia coli W

Reisolated Strains of Eschlerichia Chloram- Chlor- Oxy- Tetra-coli Resistant to: phenicol tetra- tetra- cycline

Streptomycin .............

Streptothricin ...........

Pleocidin .................Neomycin ................

Viomycin ................. +Cycloserine ...............Chlorampheniicolt..... + + + +Chlortetracyclinet ........ + + + +Oxytetracyclinet.....-..... _ + +Tetracyclinet ....... ...... + + + +Grisein . ....... _Polymyxin B .............

* +, - as in table 4.t It is possible to separate chloramphenicol and chlortetra-

cycline from each other and from tetracycline or oxytetra-cvcline with these cultures.

resistance during serial passage on slants. Only tetra-cycline and oxytetracycline are indistinguishable usingthis series of cultures. These data illustrate the fact thateach set of resistant strains isolated may be expected tohave its own idiosyncrasies especially with regard toincidental or unilateral cross-resistant relationships.The situation with respect to the cross-resistance

interrelationships of various antibiotics is obviouslyquite complex. In some cases, entirely different anti-biotics may appear identical by cross-resistance pattern.This occurs with hygromycin and the tetracyclinegroup. If one is comparing the pure substances, noconfusion arises because the tetracyclines are 10 timesaIs active as hygromycin against E. coli W. Confusionmay result, however, in testing crude materials in whichthe actual concentration is unknown. On the otherhand, it is possible to recognize potentially new materi-als by the fact that they have only a slight amount ofcross resistance. Thus a hypothetical antibiotic whichwvas cross resistanit with only streptomycin must bepresumed different from known materials. From thethe degree of initerrelationship found with antibioticswhich have been isolated to date, it would seem plaus-ible that as yet undiscovered materials might haveincidental or one-way cross resistances with well-known antibiotics. For this reason, care must be ex-ercised in assuming that unknown antibiotics can bescreened only on a cross-resistance basis.Paper strip chromatography. It is desirable, in identi-

fying antibiotics, to have some second method of com-parison which is based on the chemical nature of thesubstances rather than their biological activities. Paperstrip chromatography is an obvious answer to thisrequirement. This method of comparison has the addi-tional merit of permitting the detection and resolutionof mixtures, which cannot be accomplished by cross-resistance testing. In practice, it is possible to resolvemany of the antibiotics previously discussed by theuse of five or six solvent systems, such as employed instudies on albomycin (Stapley and Ormond, 1957).Ascending paper chromatograms permit the calculationof Rf values for comparison with values previouslyestablished for known antibiotics. The resolving powerof most of these systems is limited to antibiotics ofquite different types. An example of what can be ac-complished is provided in figure 4. It is convenient toadd 0.00333 per cent triphenyl tetrazolium chloride tothe agar of the bioautograph plates, as suggested byUsdin et al. (1954), so that the incubated plate is redwith contrasting white spots where the antibiotic ispresent. Incubating the bioautograph plates with thepaper strips in position provides a white backgroundwhich makes the inhibition spots obvious. Sensitivitycan also be increased by the use of a low incubationtemperature as in the case of disc assays. Paper chro-matograms can be examined with an ultravriolet light

3.96 [VOL. 6

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from

ANTIBIOTIC IDENTIFICATION

and a phosphorescent plate for ultraviolet-absorptivespots, a characteristic which facilitates recognition ofantibiotics such as viomycin.The chief disadvantage inherent in paper chromato-

grams, as a primary evaluation tool for antibiotics, isthat the Rf values obtained with crude antibioticsmay not agree with those of purified materials. Asimple partial purification of the antibiotic can over-come this difficulty, provided that the type of materialto be purified is known as in the case where cross-resist-ance data have been obtained. A further complication inthe evaluation of paper chromatogram data is the factthat many solvent systems divide the antibiotics intothe same groups as are found by cross-resistance tests.Thus it is not possible with most paper chromatogramsystems to separate members of the tetracycline familyor closely related members of the streptothricin family.In such cases, special paper chromatogram systems canbe selected for use based upon cross-resistance informa-tion.

DISCUSSION

Because there are limitations to the use of bothcross-resistance and paper chromatogram data, thebest procedure involves the use of both techniques. Asuitable combination of tests was reported in ourstudies on albomycin (Stapley and Ormond, 1957)with cross resistance serving as the fundamental toolfor initial classification. The advantage to the use ofcross resistance as the basis for identification testsis that it very quickly classifies the crude productmaking it a simple matter to recognize the more dis-tinctive antibiotics immediately or to select appropriatesecondary tests to establish the identity of members ofthe cross-resistance groups. Since no instances of the

same antibiotic being produced by cultures representa-tive of more than one of the major taxonomic groupshas been found, it is possible to classify the antibioticsaccording to source (bacteria, actinomycetes, and soforth). It has been noted that the various antibioticsproduce inhibition zones of characteristic appearanceon thin-layer assay plates. In many cases, the appear-ance of the inhibition zone may permit tentative assign-ment of the antibiotic in question to one of three groups:either hazy zones, clear zones, or clear zones with re-sistant colonies. A combination of cross-resistance dataand paper chromatogram data will then suffice toclassify the antibiotic as one of the knowns, closelyrelated to one of the knowns, or an unknown entity.In the operation of a screening program it is possibleto produce resistant cultures to, and obtain paperchromatogram data for, unknowns which are found tobe of no interest; even though pure material is neverobtained. Thus unidentified antibiotics which havebeen found need not be retested repeatedly, reservingmore complete evaluation for potentially new products.

ACKNOWLEDGMENTS

The author is indebted to Dr. I. M. Miller for design-ing the closed system pump and to M\irs. Martha Urand MIiss Joan Przystas for their technical assistance.

SUMMARY

A procedure for the identification of antibiotics basedupon their cross-resistance characteristics, measuredby disc-plate assays, has been described. MIethods forproduction, maintenance, and use of antibiotic re-sistant cultures and a study of the interrelationships ofa number of known antibiotics with these resistantstrains has been reported.

Figure 4. Resolution of three antibiotics by paper strip chromatography. (1) Streptothricin (Rf = 0.0); (2) cycloserine (R,f =

0.23); (3) chloramphenicol (Rf = 0.89); (4) streptothricin + cycloserine; (5) streptothricin + cycloserine + chloramphenicol. Sol-vent system = n-butanol (8): H2((2): acetic acid (1). The paper chromatogram was developed by ascending technique, on Whatmanno. 1 filter paper, at room temperature. The bioautograph plate was seeded with Escherichia coli W and incubated overnight at 25 C

1958] 397

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from

DECKER, KUEHNE, BUCHANAN, AND P'ORTEROt

REFERENCES

EISMAN, P. C., MARSH, W. S., AND MAYER, R. L. 1946 Differ-entiation of antibiotics by resistant strains. Science, 103,673-674.

HARRIS, D. A. AND RUGER, M. L. 1953 Microbiological as-pects of new antibiotics screening. I. A plug test procedure.Antibiotics & Chemotherapy, 3, 265-270.

STANSLY, P. G. 1946 The presumptive identification of anti-biotics. Science, 103, 402-403.

STAPLEY, E. 0. ANI) ORMOND, R. E. 1957 Similaritv of al-bomycin and grisein. Science, 126, 587-589.

SZYBALSKI, W. 1953 Genetic studies on microbial crossresistance to toxic agents. II. Cross resistance of Micro-(ococcus pyogenes var. aureus to 34 antimicrobial drugs.Antibiotics & Chemotherapy, 3, 915-918.

SZYBLASKI, W. 1954 Genetic studies on microbial crossresistance to toxic agents. IV. Cross resistance of Bacillusnmegateriurn to 44 antimicrobial drugs. Appl. Microbiol.,2, 57-63.

SZYBALSKI, W. AND BRYSON, V. 1952 Genetic studies onmicrobial cross resistance to toxic agents. 1. Crossresistance of Escherichia coli to 15 antibiotics. J. Bac-teriol., 64, 489-499.

SZYBALSKI, W. AND BRYSON, V. 1954 Genetic studies onmicrobial cross resistance to toxic agents. III. Crossresistance of Mycobacteriunt ranae to 28 antimicrobialagents. Am. Rev. Tuberc., 69, 267-279.

USDIN, E., SHOCKMAN, G. D., AND TOENNIES, G. 1954 Tet-razolium bioautography. Appl. Microbiol., 2, 29-33.

Design and Evaluation of a Slit-Incubator Sampler'H. NI. DECKER, R. W. KUEHNE, L. M. BUCHAN AN. AND R. 1PORTER&2

United States Armttzy Chemtical Corps, Fort Detrick, Frederick, Maryland

Received for publication April 3, 1958

Research laboratories, hospitals, and civil defenseagencies are in need of a simple sampling device forsensitive vegetative cells that wvill determine changesin air-borne bacterial contamination with a high degreeof efficiency over a considerable period of time. Epi-demiological studies of modes of transmission of air-borne disease would be simplified if more were knownregarding the bacteriological content of the air in localeswhere disease has been transmitted by suspected air-borne routes. This is of extreme importance since severalhospitals are currently reporting epidemics of infectionsby Staphylococcus sp. among patients and staff person-nel. (Fekety et al., 1958; Murray et al., 1958)

Decker and Wilson (1954) have reported previouslyon a slit sampler for collection of air-borne micro-organisms. This earlier sampler possessed inherent ad-vantages over those developed by previous workers.However, its collection efficiency decreased if it wasused continuously for periods of time much longer than1 hr.

Recently, Kuehne and Decker (1957) have reportedon several critical factors affecting the efficiency of airsampling when vegetative cells of microorganisms arecollected for extended periods of time. Factors such asincreasing the volume of air per given area of agar isdetrimental to the viability of Serratia marcescens. De-

' Presented at the 58th General Meeting of the Society ofAmerican Bacteriologists, Chicago, Illinois, April 27 to May 1,1958.

2 1Present address: Air Pollution Engineering Staff, Divisionof Sanitary Engineering Services, Bureau of State Services,U. S. Public Health Service, Washington, D. C.

creasing the agar concentration from 3 per cent to 1.5per cent was also found to support greater growth of thetest organism.As a result of these recent studies on air sampling, a

more efficient slit sampler has been designed, con-structed, and evaluated. This sampler permits continu-ous collection of air-borne vegetative organisms for longperiods with a much higher degree of efficiency byproviding a larger area for collection of microorganismsthan samplers previously developed by the authorsand others. Collection of a much greater number oforganisms with less dehydration of the media andrendering of an accurate time concentration relation-ship results. There is also incorporated within thesampler an incubation chamber which is automaticallyturned on at the completion of sampling, requiringattendance of the sampler only once in a 24-hr period.

MATERIALS AND METHODS

The slit-incubator sampler and its component partsare shown in figures 1 and 2. The sampler is composedof a stainless steel and lucite sampling box 40 in. long,4 in. wide, and 43 in. high, a slit and slit tube, mediaculture tray, rotating drive shaft, electric clock motor,and heating element.The slit and slit tube assembly is threaded into the

sampling box. The slit opening may be adjusted bymoving one of the metal plates which are fastened tothe bottom of the slit tube. The height indicator, avertical metal shaft passing through the sampler top, istemporarily set with its lower end on the media surface.

[VOL. 6398

on August 18, 2020 by guest

http://aem.asm

.org/D

ownloaded from