How to Interpret Confusing Results from Water Testing with Coliscan Easygel  -- What causes the entire dish of medium to change to a uniform dark blue or uniform pink color?

Confusing results from a volunteer water monitor using Coliscan® Easygel®

We received an email from a person who is in charge of administering a state program of volunteer water monitoring which has been using Coliscan® Easygel® to test for the presence of E. coli and coliforms in the test samples.  The protocol which the volunteers follow is to collect triplicate water samples at each test site and to plate each one separately by adding 1 mL of the water sample into a bottle of liquid Coliscan® Easygel® , swirl it to mix and then pour the medium/water mix into a pretreated petri dish where it gels.  The dishes are incubated for 24-48 hrs. at 35° C and the results are read.  The normal results will be for the formation of individual bacterial colonies where the E. coli show as dark blue/purple dots and the other coliforms show as pink dots on/in the medium.  These are counted and the respective numbers are calculated for 100 mL of the water that was tested. 

Upon plating samples of the water at different sample amounts, the following sets of dishes resulted.  Set #1 can be compared directly to the original photos from the volunteer and it can be seen that the results are essentially duplicated.  This was done by using a 0.25 mL test sample of the water. (These dishes gelled without any problem.)  When these dishes were viewed under a 10X microscope, myriads of tiny bacterial colonies were seen, and there were some tiny dark blue (E. coli) colonies evident, with more of these in the darker dishes.    

These results from comparing the volunteer’s dishes and those of our 0.25 mL test sample illustrate what Micrology Laboratories has consistently told users of the Coliscan® media. (This is briefly explained in FAQ #7 )   If the medium in a dish turns a uniform pink or blue/purple color, this is indicative of extremely high counts of target bacteria.  When this happens, individual bacterial colonies are not apparent and the entire dish assumes a uniform colored cast.  In order to achieve a dish which contains countable, well separated individual colonies, the tester must use a much smaller inoculum (amount of test sample)  This is best achieved by diluting the sample and then inoculating with portions of the diluted sample.  This is standard microbiological procedure and it is understandable that inexperienced individuals (some volunteer monitors) are not familiar with the concept and how to achieve a proper diluted sample.  D etailed diagrams and instructions on doing dilutions can be found here.

In this specific case, which we are describing here is an excellent illustration of the topic of this note, the three water samples show significant variation.  This may be due to them not being collected right at the same time at the same spot, or it may be due to variable currents in the stream or other difficult to assess factors.  However, even the sample with the lowest counts approaches numbers of bacteria found in raw sewage, and this particular stream is heavily polluted with fecal material.  It is obvious that this water source is being heavily abused in some way and needs attention. 

                                                                                    Set 2

In order for a quantitative count of the target organisms (specifically E. coli in this case for volunteer water monitoring) the samples must be diluted considerably.  Set #2 consists of three photos consisting of dishes inoculated with the samples (three samples taken at the same time from the same site) provided.  As was previously mentioned, these supposedly identical samples varied considerably in the amounts of target organisms each contained.  For Set #2, one mL of sample was dispensed into 99 mL of sterile diluent.  This was shaken 25 times to mix well and 0.25 mL of this dilution mix was added to a bottle of the Coliscan® medium and poured into a pretreated petri dish.  The dishes were placed right side up in a 35° C. incubator for 20 hours and then removed, examined and photographed (See Set 2 above and A, B, C below, click to enlarge).

A

B

C

It is obvious from examining the photos that A contains the least number of blue target colonies (E. coli).  These were counted under a 10X binocular microscope and 241 blue colonies were counted.   Since the sample was diluted 100X (1 mL in 99 mL sterile diluent), if 1 mL of the diluted sample had been in the poured dish, you would have multiplied the number of target colonies times 100 which would have been a total of 24,100 colonies/mL.  However since only 0.25 mL had been in the poured dish, you have to multiply again times 4, which makes the total colonies/ml of original water sample 96,400/mL. (The actual dilution was 1:400.)  To calculate the number of colonies in a 100 mL sample of the original water, you have to multiply the number/ml x 100 which means that the water in the creek contained 9,640,000 E. coli in 100 mL.  This almost approaches the numbers in some raw sewage and is excessive by any measure. 

Remember that dishes B and C contain even higher counts of E. coli, not even countable at the 1:100 dilution that was done.  They would have needed to be diluted at least 1:1000 or better at 1:10000 in order to achieve countable dishes.

So far, only the counts of E. coli have been considered.  However, it is obvious that dishes A and B have a strong pink color and this is due to the huge numbers of other coliform bacteria in the samples.  If you wished to count them, much greater dilutions of the samples would be needed.  But you can see that the E. coli grows and the blue colonies stand out in the pink background.  If you wished to reduce the pink coliform background organisms, one alternative would be to incubate the dishes at 40-42° C.  The higher temperature would still allow the E. coli to grow, but would inhibit the growth of some of the coliform background colonies.

  Finally to emphasize, one important additional point pertains to the size of the colonies growing in the Coliscan® medium.  When there are “wall to wall” huge numbers of colonies that are virtually touching each other, they do not have the space to grow to their normal optimum colony size which would occur if they had plenty of space between colonies.  Therefore, the colonies will be extremely small in comparison to the un-crowded normal colony size.  When huge numbers of bacteria are in a sample, the colonies are so small that they cannot be distinguished without additional microscopic magnification and the entire dish assumes a colored tint (bluish if predominance of glucuronidase positive E. coli or pinkish if predominance of galactosidase positive other coliforms).  When this happens, you should recognize that it is necessary to dilute your test sample significantly (probably at least 1000X and likely 10000X) in order to obtain a countable number of colonies in your dishes.