How to Calculate the Number of Bacteria Using the Defined Substrate Technology (Colilert)

The Defined Substrate Technology (DST) method, commercialized as Colilert and Enterolert by IDEXX Laboratories, provides a standardized, enzyme-based approach for simultaneously detecting and enumerating total coliforms and Escherichia coli (or enterococci) in water samples. Unlike traditional multiple-tube fermentation (MTF) that relies on visual gas production in lactose broth, DST uses chromogenic and fluorogenic substrates that are cleaved only by specific bacterial enzymes—β-galactosidase for coliforms and β-glucuronidase for E. coli—producing a color change (yellow) or fluorescence under UV light. The number of bacteria is calculated using a Most Probable Number (MPN) table derived from the pattern of positive wells in a multi-well tray (e.g., Quanti-Tray/2000). This method is particularly useful for drinking water, recreational water, and wastewater testing because it reduces confirmation time from 72 hours to 18–24 hours, minimizes false positives from non-coliform bacteria, and provides quantitative results with defined statistical confidence intervals [1][3].

At a Glance

Scientific Principle

The Defined Substrate Technology exploits the specificity of bacterial enzymes to detect target organisms without the need for selective enrichment or confirmation steps. The method relies on two key substrates:

• Ortho-nitrophenyl-β-D-galactopyranoside (ONPG): This chromogenic substrate is cleaved by β-galactosidase, an enzyme produced by total coliform bacteria. Cleavage releases ortho-nitrophenol, which turns the medium yellow. The reaction is detectable visually or spectrophotometrically.

• 4-methylumbelliferyl-β-D-glucuronide (MUG): This fluorogenic substrate is cleaved by β-glucuronidase, an enzyme produced specifically by E. coli (and a few other Enterobacteriaceae). Cleavage releases 4-methylumbelliferone, which fluoresces blue-white under long-wave UV light (365 nm).

The specificity of these enzyme-substrate reactions is the foundation of DST. Traditional methods like lactose fermentation can be confounded by non-coliform bacteria that produce gas, requiring time-consuming confirmation steps. In contrast, DST's enzyme targets are highly conserved among coliforms and E. coli, reducing false positives and allowing same-day results [1][3].

The MPN calculation is based on the statistical principle that bacteria are randomly distributed in the sample. When a sample is divided into multiple aliquots (wells), the probability of each well containing at least one viable target organism follows a Poisson distribution. The MPN table provides the most probable number of organisms per 100 mL based on the observed pattern of positive wells, along with 95% confidence intervals.

Materials and Instrumentation

Sample Collection and Preparation

• Sterile sample bottles: Use 120 mL or 250 mL polypropylene or glass bottles with sodium thiosulfate (for chlorinated samples) or without (for non-chlorinated samples). Bottles must be sterile and certified for bacteriological water testing.
• Sample volume: Typically 100 mL for drinking water; smaller volumes (1–10 mL) may be diluted with sterile phosphate-buffered saline (PBS) for high-turbidity samples.
• Dilution water: Sterile PBS (0.1 M, pH 7.2) or Butterfield's phosphate buffer for sample dilution when needed.

DST Reagent and Trays

• Colilert or Enterolert reagent: Lyophilized powder in single-use packets. Each packet is designed for 100 mL of sample. Store at 2–8°C and protect from moisture.
• Quanti-Tray/2000: A 97-well tray (49 large wells + 48 small wells) that provides a dynamic range of 1–2,419 MPN/100 mL. For lower expected counts, Quanti-Tray (51 wells) with a range of 1–200.5 MPN/100 mL may be used.
• Quanti-Tray sealer: A heat-sealing device that seals the tray into individual wells. Ensure the sealer reaches proper temperature (typically 180–200°C) before use.

Incubation and Detection Equipment

• Incubator: Capable of maintaining 35 ± 0.5°C. Use a calibrated thermometer to verify temperature daily.
• UV lamp: Long-wave UV (365 nm) for fluorescence detection. A 6-watt handheld lamp is sufficient. Ensure the lamp is warmed up for at least 5 minutes before reading.
• Timer: For precise incubation timing.
• Positive control: E. coli ATCC 25922 (or equivalent) for Colilert; Enterococcus faecalis ATCC 29212 for Enterolert.
• Negative control: Sterile PBS or deionized water.

Instrumentation Choices

The choice between Colilert-18 (18-hour incubation) and standard Colilert (24-hour incubation) depends on laboratory workflow and regulatory requirements. Colilert-18 is validated for faster results but requires strict temperature control and may be less sensitive for stressed organisms [1]. For recreational water monitoring, Enterolert (using 4-methylumbelliferyl-β-D-glucoside for enterococci) is an alternative that provides comparable sensitivity to membrane filtration methods [2].

Controls

Positive Control

• Prepare a suspension of E. coli ATCC 25922 in sterile PBS at approximately 10–100 CFU/100 mL (based on turbidity or previous enumeration).
• Process the positive control sample exactly as test samples.
• Expected result: Yellow color in at least some wells (coliform positive) and fluorescence in at least some wells (E. coli positive) after incubation.

Negative Control

• Use 100 mL of sterile PBS or deionized water.
• Process identically to test samples.
• Expected result: No yellow color and no fluorescence in any well after incubation.

Matrix Control (Optional)

• For samples with high turbidity, color, or chemical interference, prepare a matrix spike by adding a known number of target organisms to a duplicate sample.
• Compare recovery to the positive control to assess inhibition.

Duplicate Samples

• For quality assurance, analyze at least 10% of samples in duplicate.
• Calculate the relative percent difference (RPD) between duplicates. Acceptable RPD is typically <30% for MPN values above 10.

Conceptual Workflow

Step 1: Sample Preparation

1. Collect water samples in sterile bottles according to standard methods (e.g., EPA Method 9223, ISO 9308-2).
2. For chlorinated samples, ensure sodium thiosulfate is present to neutralize residual chlorine.
3. Transport samples at 4°C and process within 6 hours of collection (or within 24 hours if held at 4°C).
4. If sample turbidity exceeds 1 NTU, dilute the sample 1:10 or 1:100 with sterile PBS to avoid clogging the Quanti-Tray wells.

Step 2: Reagent Addition

1. Aseptically open the Colilert reagent packet and add the entire contents to the 100 mL sample.
2. Cap the bottle and shake vigorously for 30 seconds to dissolve the powder completely.
3. Allow the mixture to stand for 2–3 minutes to ensure complete dissolution and substrate availability.

Step 3: Tray Sealing

1. Place the Quanti-Tray/2000 on a clean, flat surface with the well side up.
2. Pour the entire 100 mL sample-reagent mixture into the tray, avoiding air bubbles.
3. Gently tap the tray to distribute the liquid evenly across all wells.
4. Place the tray in the Quanti-Tray sealer with the foil side facing the heating element.
5. Seal according to manufacturer instructions (typically 2–3 seconds for Quanti-Tray/2000).
6. Inspect the sealed tray for proper sealing—all wells should be individually sealed with no cross-leaks.

Step 4: Incubation

1. Place the sealed tray in the incubator at 35 ± 0.5°C.
2. Incubate for 18 hours (Colilert-18) or 24 hours (standard Colilert).
3. Do not stack trays more than 2 high to ensure uniform temperature distribution.
4. Record the incubation start and end times.

Step 5: Reading Results

1. Remove the tray from the incubator.
2. Coliform detection: Count wells that have turned yellow (any intensity). Yellow indicates β-galactosidase activity.
3. E. coli detection: In a darkened room, shine the UV lamp (365 nm) at a 45-degree angle approximately 5–10 cm from the tray. Count wells that show blue-white fluorescence. Only wells that are both yellow and fluorescent are counted as E. coli positive.
4. Enterococci detection (Enterolert): Count wells that show blue-white fluorescence under UV light. No color change is required.
5. Record the number of positive large wells (49 total) and positive small wells (48 total) separately.

Step 6: MPN Calculation

1. Locate the MPN table provided with the Quanti-Tray/2000 product.
2. Find the row corresponding to the number of positive large wells (0–49).
3. Find the column corresponding to the number of positive small wells (0–48).
4. Read the MPN value per 100 mL and the 95% confidence interval.
5. If the pattern exceeds the table (e.g., all 97 wells positive), report as >2,419 MPN/100 mL and consider dilution.

Quality Checks

Incubation Temperature Verification

• Record incubator temperature at the start and end of incubation.
• Acceptable range: 35 ± 0.5°C. If temperature deviates, note in the laboratory record and consider repeating the test.

Positive and Negative Control Results

• Positive control must show both yellow and fluorescence in at least some wells.
• Negative control must show no yellow and no fluorescence.
• If controls fail, investigate possible contamination, expired reagent, or incubator malfunction.

Duplicate Precision

• For duplicate samples, calculate the RPD: |(MPN1 – MPN2)| / ((MPN1 + MPN2)/2) × 100.
• Acceptable RPD: <30% for MPN >10; <50% for MPN 1–10.

Reagent Lot Verification

• Record the lot number and expiration date of each Colilert packet.
• Do not use expired reagent. If a new lot is introduced, perform a parallel comparison with the previous lot using a positive control.

Result Interpretation

Reading Positive Wells

• Yellow color: Any detectable yellow color in a well, regardless of intensity, is considered positive for total coliforms. Very faint yellow may indicate weak β-galactosidase activity or early growth.
• Fluorescence: Fluorescence must be distinctly brighter than the background. Use a non-fluorescent black surface under the tray to enhance contrast. Compare to a negative control tray if uncertain.
• Ambiguous wells: If a well shows yellow but no fluorescence, it is positive for coliforms but negative for E. coli. If a well shows fluorescence without yellow (rare), it may indicate a non-coliform organism with β-glucuronidase activity (e.g., some Shigella species) or a false positive.

MPN Table Usage

The Quanti-Tray/2000 MPN table is based on a 97-well configuration: 49 large wells (each containing 1.86 mL of sample) and 48 small wells (each containing 0.186 mL). The total sample volume is 100 mL. The MPN is calculated using the formula:

MPN = (ln(N/n)) × (100/V)

Where:

• N = total number of wells (97)
• n = number of negative wells
• V = volume per well (average)

However, the pre-calculated table is more practical for routine use. The table provides MPN values from 1 to 2,419 per 100 mL, with 95% confidence intervals.

Example Calculation

Suppose a Quanti-Tray/2000 shows:

• 12 positive large wells (out of 49)
• 8 positive small wells (out of 48)

Using the MPN table, locate row 12 and column 8. The MPN value might be 98.6 MPN/100 mL with a 95% confidence interval of 52–186 MPN/100 mL.

Reporting Results

• Report as "MPN/100 mL" for total coliforms and "E. coli MPN/100 mL" separately.
• For values below the detection limit (0 positive wells), report as "<1 MPN/100 mL" (for Quanti-Tray/2000).
• For values above the upper limit (all 97 wells positive), report as ">2,419 MPN/100 mL" and indicate the dilution factor if used.

Troubleshooting

Limitations

Method-Specific Limitations

• Non-target enzyme activity: Some non-coliform bacteria (e.g., Aeromonas species, Vibrio species) may produce β-galactosidase and cause false-positive yellow color. Similarly, some Shigella and Salmonella strains produce β-glucuronidase, potentially causing false-positive E. coli fluorescence [3].
• Stressed or injured organisms: Chlorinated or otherwise stressed bacteria may require longer incubation (24 hours) to recover and produce detectable enzyme activity. Colilert-18 may underestimate counts in such samples [1].
• High turbidity: Samples with turbidity >1 NTU may clog the Quanti-Tray wells, leading to uneven distribution and inaccurate MPN values. Dilution is required.
• Chemical interference: High levels of chlorine, heavy metals, or organic compounds may inhibit enzyme activity. Neutralize chlorine with sodium thiosulfate and dilute if necessary.

Statistical Limitations

• MPN confidence intervals are wide: The 95% confidence interval for a single MPN value can span a factor of 3–10. For example, an MPN of 100 may have a confidence interval of 30–300. This is inherent to the MPN method and must be considered when comparing results.
• MPN is not a direct count: Unlike colony-forming units (CFU) on agar plates, MPN is a statistical estimate. It assumes random distribution of bacteria, which may not hold for clumped or aggregated cells.
• Upper and lower limits: The Quanti-Tray/2000 has a range of 1–2,419 MPN/100 mL. Samples outside this range require dilution or concentration, adding uncertainty.

Regulatory Considerations

• Method approval: DST methods are approved by the U.S. EPA (Method 9223) and ISO (ISO 9308-2) for drinking water and recreational water testing. However, some regulatory agencies may require confirmation of positive results by traditional methods [1][3].
• Matrix-specific validation: For non-standard matrices (e.g., wastewater, industrial process water), the method should be validated against a reference method to ensure equivalent performance [1].

Documentation

Required Records

• Sample information: Sample ID, collection date/time, location, sampler name, sample type (drinking water, recreational, etc.)
• Sample preparation: Volume tested, dilution factor (if any), neutralization (if chlorinated)
• Reagent information: Product name, lot number, expiration date
• Incubation details: Start time, end time, incubator temperature (start and end)
• Control results: Positive control MPN, negative control result
• Tray reading: Number of positive large wells, number of positive small wells, MPN value, 95% confidence interval
• Interpretation: Total coliform MPN/100 mL, E. coli MPN/100 mL (or enterococci MPN/100 mL)
• Quality control: Duplicate RPD, control acceptance/rejection

Example Documentation Entry

Sample ID: DW-2024-03-15-001
Sample type: Drinking water (tap)
Collection date: 2024-03-15 09:30
Volume tested: 100 mL (no dilution)
Reagent: Colilert-18, Lot #C2401, Exp. 2025-01
Incubation: 35.2°C, 18 h (09:45 to 03:45 next day)
Positive control: E. coli ATCC 25922 – 12 large wells positive, 8 small wells positive (MPN 98.6)
Negative control: 0 large wells, 0 small wells positive
Test sample: 3 large wells positive, 1 small well positive
MPN (from table): 6.3 MPN/100 mL (95% CI: 1.8–19.0)
Interpretation: Total coliforms detected (6.3 MPN/100 mL); E. coli not detected (<1 MPN/100 mL)
Duplicate RPD: N/A (single analysis)
Technician: J. Smith

Biosafety

Risk Assessment

Environmental water samples may contain pathogenic microorganisms, including enteric pathogens (e.g., Salmonella, Campylobacter, Cryptosporidium). However, routine testing of drinking water and recreational water samples is considered BSL-1 under the CDC/NIH guidelines, provided that samples are handled with standard microbiological practices [6].

Standard Practices

• Hand washing: Wash hands before and after handling samples.
• Personal protective equipment (PPE): Wear lab coat, gloves, and safety glasses.
• Work surface: Use a dedicated biosafety cabinet (BSC) for sample preparation if available; otherwise, work on a disinfected bench surface.
• Decontamination: Disinfect all work surfaces with 70% ethanol or 10% bleach after each use.
• Waste disposal: Autoclave all used Quanti-Trays, sample bottles, and contaminated materials before disposal.

Special Considerations

• Recombinant organisms: The Colilert reagent does not contain recombinant organisms. However, if using recombinant positive controls (e.g., genetically modified E. coli), follow NIH Guidelines for Research Involving Recombinant or Synthetic Nucleic Acid Molecules [7].
• Spills: Cover spills with absorbent paper, apply disinfectant (10% bleach), allow 20 minutes contact time, then clean up wearing gloves.
• Shipping: If samples are shipped to a reference laboratory, follow IATA Dangerous Goods Regulations for Category B infectious substances (UN 3373).

Frequently Asked Questions

1. Can I use Colilert for samples other than water, such as food or environmental swabs?

Colilert is specifically validated for water matrices (drinking water, recreational water, wastewater). For food or surface samples, the sample must first be eluted into a liquid medium (e.g., PBS) and then processed as a water sample. However, the method has not been validated for these matrices, and results should be interpreted with caution. Alternative methods like Petrifilm or SimPlate may be more appropriate for food testing.

2. Why does my MPN value sometimes fall outside the table range?

The Quanti-Tray/2000 MPN table covers 1–2,419 MPN/100 mL. If all 97 wells are positive, the true count exceeds 2,419 MPN/100 mL. In this case, dilute the sample 1:10 or 1:100 with sterile PBS and repeat the test. Multiply the MPN from the table by the dilution factor. If no wells are positive, report as "<1 MPN/100 mL" for the undiluted sample.

3. How do I distinguish between total coliforms and E. coli in the same tray?

Total coliforms are indicated by yellow color (ONPG cleavage). E. coli is indicated by fluorescence (MUG cleavage) in wells that are also yellow. A well that is yellow but not fluorescent contains coliforms other than E. coli. A well that is fluorescent but not yellow (rare) may indicate a non-coliform organism with β-glucuronidase activity or a false positive. Always confirm fluorescence under UV light in a darkened room.

4. What should I do if my positive control fails to show fluorescence?

First, verify the UV lamp is functioning and warmed up. Check that the positive control organism (E. coli ATCC 25922) is viable and not contaminated. Ensure the Colilert reagent is not expired and was properly stored. If all checks pass, repeat the positive control with a fresh reagent packet. If failure persists, contact the manufacturer for technical support.

References and Further Reading

1. Boubetra A, Le Nestour F, Allaert C, Feinberg M. Validation of alternative methods for the analysis of drinking water and their application to Escherichia coli. 2011. PubMed
2. Kinzelman J, Ng C, Jackson E, Gradus S, Bagley R. Enterococci as indicators of Lake Michigan recreational water quality: comparison of two methodologies and their impacts on public health regulatory events. 2003. PubMed
3. Eckner KF. Comparison of membrane filtration and multiple-tube fermentation by the Colilert and Enterolert methods for detection of waterborne coliform bacteria, Escherichia coli, and enterococci used in drinking and bathing water quality monitoring in southern Sweden. 1998. PubMed
4. Cartwright A, Dooley JSG, McGonigle CD, Arnscheidt J. How suitable is freshwater sponge Ephydatia fluviatilis for time-integrated biomonitoring of microbial water quality? 2024. PubMed
5. O'Toole J, Sinclair M, Leder K. Transfer rates of enteric microorganisms in recycled water during machine clothes washing. 2009. PubMed
6. CDC and NIH. Biosafety in Microbiological and Biomedical Laboratories (BMBL), 6th Edition. U.S. Department of Health and Human Services, 2020. CDC
7. National Institutes of Health. NIH Guidelines for Research Involving Recombinant or Synthetic Nucleic Acid Molecules. NIH
8. National Center for Biotechnology Information. NCBI Bookshelf: Molecular Biology and Laboratory Methods. NCBI

Related Articles

• How to Calculate the Number of Bacteria Using the Microcolony Method
• How to Calculate the Number of Bacteria Using the Swab and Rinse Method
• How to Calculate the Number of Bacteria Using the Tape Lift Method
• How to Calculate the Number of Bacteria Using the Agar Overlay Method
• How to Calculate the Number of Bacteria Using the Petrifilm Method
• How to Calculate the Number of Bacteria Using the SimPlate Method