encountered. Gnezda (1899) first described the oxalic acid test. These have been ably discussed and compared by Fellers and Clough in

Transcription

1 THE PHOTOELECTRIC DETERMINATION OF INDOLE IN BACTERIAL CULTURES ALFRED R. STANLEY AND ROBB SPALDING SPRAY Department of Bacteriology and Public Hygiene, School of Medicine, West Virginia University, Morgantoum Received for publication July 1, 1940 Many colorimetric tests for indole have been described since Gnezda (1899) first described the oxalic acid test. These have been ably discussed and compared by Fellers and Clough in 1925, together with a very complete bibliography. They concluded, as a result of comparative studies, that the Ehrlich test (1901) was the most accurate, reliable and simple for indole and skatole when unmixed. The Steensma vanillin test (1906) was as delicate, though not as reliable because of interfering substances. In 1920 Zoller extracted the culture distillate with iso-amyl or iso-butyl alcohol and then used the Salkowski test (1919) on this extract. Happold and Hoyle in 1925 discussed solvents useful for extracting indole from media and chose petroleum ether (Ligroin, B.P 'C.) as the one most desirable. The rosindole body was then extracted from this with Ehrlich's reagent. The method to be presented here is based on the work of Zoller, but dispensing with the distillation and using Ehrlich's reagent instead of the Salkowski test. It is, we believe, more simple than the method of Happold and Hoyle, as there is but one extraction. Difficulties which they anticipate have not been encountered. METHODS General procedure. This method is based upon the direct extraction of indole from a measured portion of culture medium, 251

2 252 ALFRED R. STANLEY AND ROBB SPALDING SPRAY by means of shaking with iso-amyl alcohol. A quantitative indole reagent is then added to the extract and the resulting color read on a photoelectric calorimeter. From the calorimetric reading, the quantity of indole is read directly from a standard curve, plotted from readings on definite concentrations of c.p. indole which have been similarly treated. Preparation of standard curve. Eleven 6 x j inch Pyrex test tubes, numbered 1 to 11, are set up in a rack, each fitted with a No. 0 rubber stopper. Indole dilutions are prepared as shown in the following table. PREPARATION OF INDOLE DILUTIONS Solution 1. Exactly 250 mgm. c.p. indole' is dissolved in distilled water and made up to 500 ml. in a volumetric flask.2 Solution 2. Twenty milliliters (volumetric pipette) of solution 1 made up to 100 ml. in a volumetric flask. Solution 2 Dtilled H2O Indole in tube ml. ml. mmfr. Tube Tube Tube Tube Tube Solution S. Ten milliliters (volumetric pipette) of solution 1 made up to 100 ml. in a volumetric flask. Solution 3 Distilled H2O Indole in tube ml. ml. mom. Tube Tube Tube Solution 4. Eight milliliters (measuring pipette) of solution 1 made up to 200 ml. in a volumetric flask. Solution 4 Distilled HiO Indole in tube ml. ml. mwm. Tube Tube Tube Eastman No *Kimble Exax Blue Line volumetric flasks, pipettes and mixing graduates were used throughout this work. volumetric pipettes, measuring

3 PHOTOELECTRIC DETERMINATION OF INDOLE This gives a series of tubes each with 5 ml. of solution and containing decreasing amounts of indole in regular steps as shown. Using a pipette made from 7 mm. glass tubing and fitted with a 2 ml. rubber bulb, approximately 2 ml. of iso-amyl alcohols is added to each tube. Rubber stoppers are fitted to the tubes and, after inverting about four times, the alcohol is allowed to form in a layer at the surface. This is drawn off with another pipette and transferred to a mixing graduate numbered to conform to the corresponding indole dilution. A separate rubber stopper and pipette is used for each tube. This extraction is repeated three times, pooling the alcoholic extracts in the corresponding mixing graduate. The volume is brought up to 10 ml. with iso-amyl alcohol, then to 25 ml. with ethyl alcohol (95 per cent), and finally, using a volumetric pipette, 10 ml. of Quantitative Indole Reagent, described below, is added to each graduate and the contents thoroughly mixed. A control solution is prepared from 10 ml. iso-amyl alcohol, 15 ml. ethyl alcohol (95 per cent), and 10 ml. Quantitative Indole Reagent. These solutions are allowed to stand, after thorough mixing, for one hour and then readings are taken on the photoelectric colorimeter.4 The colorimeter is standardized at 100 using the control solution containing no indole but otherwise identical to the other solutions. Readings are then made on the solutions from each cylinder and the results plotted against the respective concentrations of indole. The curve obtained by these methods is shown in chart 1. Preparation of quantitative indole reagent. The amount of p-dimethylaminobenzaldehyde to be used in the procedure is very important, as increased quantities give deeper colors in high indole concentrations. To determine the optimum amount to be used, the following experiments were conducted. The formula for the usual Ehrlich's Reagent (1 gram of p-dimethylaminobenzaldehyde in 95 ml. of 95 per cent ethyl alcohol and 20 ml. of concentrated HCl) was used as a basis. This was termed the X concentration of the aldehyde. This was doubled Eastman No. T18. 4 An Eimer and Amend Photoelectric Colorimeter No was used in our experiments.

4 .5 \ ~~~~~~~No 254 ALFRED R. STANLEY AND ROBB SPALDING SPRAY (2X) and quadrupled (4X). with these reagents are shown in chart 2. filter CD{ 0.40 W.A. 8.0 a 8? ~~6.0-3~~~~~~~~~~ U The comparative curves obtained The 4X concentration gives a wider curve and so was chosen as the standard for further work. Higher concentrations were not tried because of the expense of the reagent. Is *3 CHART.2-.1.,i Colorimlter readings.standard CURVE FOR INDOLE 2.0 Z %A 0 Vi 0 Vi 0 0 Vi I 8 Colorimeter readings CHART 2. Enmxm OF VARTING CONCENTRATION OF ALDEHYDIC Samples of c.p. indole obtained from three sources were compared. The results are shown in chart 3, from which we observe that two check almost identically and the third very poorly. Samples of p-dimethylaminobenzaldehyde from three sources were compared on the chosen sample of indole with the results shown in chart 4. It is readily seen that the same brand of aldehyde must be used in running the culture analysis as that used in preparing the standard curve, and the curve might well be checked with each new lot of aldehyde used. As a result of these experiments the following formula has been

5 PHOTOELECTRIC DETERMINATION OF INDOLE derived. Exactly 35 grams of p-dimethylaminobenzaldehyde5 is dissolved in about 700 ml. of 95 per cent ethyl alcohol in a 1000 ml. volumetric flask. One hundred seventy milliliters of concentrated HCl is added, thoroughly mixed, cooled to 200C. and the volume brought to 1000 ml. with 95 per cent ethyl alcohol. Indole determination in bacterial cultures. The indole is extracted from a measured portion (preferably 2 to 5 ml.) of the culture medium, by the same procedure as is given under the.5 A C ~~~~ - CD Go 0~~~~~~~, I- t I & W & Wssg Colorlueter readings CABT 3. COMPARISON OF INDOLE SAMPLES.5- \ VI li 0OlI i O~l 8 Colorlmeter readings CU1AT 4. COMPARISON OF ALDEHYDE SAMPLES preparation of the standard curve. If less than 5 ml. is used, its volume is brought up to that with distilled water. The extract is made up to volume, the indole reagent added, and a numerical reading is taken on the photoelectric colorimeter. This reading is located on the standard curve and the milligrams of indole per unit of culture medium is either read directly or calculated from the size portion used in the test. The standard curve (chart 1) shows the mgm. of indole per sample and also ' Eastman No

6 256 ALFRED R. STANLEY AND ROBB SPALDING SPRAY the mgm. of indole per 100 ml. of culture medium when a 5 ml. sample is used for the test. The size sample to be tested will depend upon the amount of indole expected. The portion of the curve showing greatest accuracy is below 0.30 mgm. per sample, thus if more than 6.0 mgm. per 100 ml. of medium is expected, a portion smaller than 5 ml. should be used. DISCUSSION We have presented here a photoelectric method for -the quantitative determination of indole which we believe to be as accurate as, and more simple than, any so far reported. This method is less complicated than that of Happold and Hoyle (1934) in that there is only one extraction as compared to their two. No difficulty is experienced in the change of color of the rosindole body by an excess of Ehrlich's aldehyde, as the intensity of the color in standard solutions is measured by the photoelectric colorimeter and this can be compared at any time with other readings taken under the same conditions. In this type of work the same concentration of the aldehyde must be used at all times regardless of the quantity of indole. The presence of peptones does not interfere with this test, as has been shown by preparing the c.p. indole dilutions in nutrient broth instead of distilled water. The readings obtained after extraction by iso-amyl alcohol are the same as when the dilutions are made up in distilled water. These methods have been applied to the study of indole production by Clostridium bifermentans and Clostridium sphenoides. Indole was determined in media containing fermentable carbohydrates in which the acidity ran as high as ph 4.78, as well as in sugar-free media at about neutrality. We believe that these methods have a wide applicability to the quantitative determination of indole production under a variety of experimental conditions. REFERENCES EHELicH, PAUL Ueber die Dimethylaminobenzaldehyde Med. Woche., Berlin, 2, Reaktion.

7 PHOTOELECTRIC DETERMINATION OF INDOLE 257 FELLERS, CARL R., AND CLOUGH, RAY W Indol and skatol determinations in bacterial cultures. J. Bact., 10, GNEZDA, JULIUS tlber neue Reaktionen der Indolbasen und der albuminoiden Korper. Compt. rend., 128, HAPPOLD, F. C., AND HOYLE, L The quantitative determination of indole in bacterial cultures. Biochem. J., 28, SALKOWSxI, E Ueber die Ehrlichsche Reaktion mit p-dimethylaminobenzaldehyde und das Urorosein. Biochem. Z. 97, STTEENSMA, F. A Ueber Farbenreaktionen der Eiweisskorper, des Indols und des Skatols mit aromatischen Aldehyden und Nitriten. Z. physiol. Chem., 47, ZOLLER, HARPER F Quantitative estimation of indole in biological media. J. Biol. Chem., 41, Downloaded from on June 30, 2018 by guest