Measurement

Our project rests on a single claim: that E. coli Nissle 1917 carrying F. prausnitzii FAAH degrades oleoylethanolamide in the gut lumen. Everything else, the localization variants, the secretion tag, the chassis choice, only matters if that claim can be tested quantitatively.

When we looked for a way to measure it, we found a gap. The established methods for quantifying OEA and other N-acylethanolamines are LC-MS based. They are accurate and they are the right answer for a well-funded lab, but they need instrument access, method development time, and a per-sample cost that puts repeated measurement out of reach for a team like ours. We needed something we could run ourselves, on a plate reader, often enough to compare six constructs against each other.

So our measurement work this season was not a single number. It was building a two-tier measurement strategy that answers two separate questions with two different methods, and designing it so that another team can pick it up and run it.

The two questions are easy to confuse, and conflating them is the most common way an enzyme result goes wrong.

Tier 1, is the enzyme there and is its active site intact? This is a protein question. An enzyme can be expressed, visible on a gel, and completely non-functional because it misfolded, because the active site is blocked, or because the localization tag buried it in an inclusion body.

Tier 2, is the substrate actually being consumed? This is a chemistry question. It is the one that matters for the project claim, and it is the one that cannot be answered by looking at a protein gel.

Our one completed measurement is expression. On September 15 we ran SDS-PAGE on uninduced culture, induced pellet, and induced supernatant for each of three constructs, induced with doxycycline at 1 µg/mL for three hours at 37 °C.

Running all three fractions matters more than it first appears. The uninduced lane is the control that makes the induced lane mean anything, and splitting pellet from supernatant is how a secretion construct gets distinguished from one whose product is stuck inside the cell. For the YebF constructs, which are designed to export FAAH, the supernatant fraction is the result rather than a side observation.

FAAH is a serine hydrolase. Fluorophosphonate probes bind covalently and irreversibly to the catalytic serine of an active serine hydrolase and do not bind if that site is disrupted. A TAMRA-conjugated fluorophosphonate therefore reports active sites rather than protein mass.

Design

1. Incubate cell lysate with the TAMRA-FP probe so the probe labels accessible active serine hydrolase sites.
2. Quench and separate by SDS-PAGE.
3. Image the gel for TAMRA fluorescence first, then stain for total protein.
4. A fluorescent band at the FAAH mass means an intact, accessible active site.

Controls, and why the catalytic mutant is the whole point

The catalytic mutant construct is not a nice-to-have. It is the control that converts this assay from suggestive to interpretable. The mutant is expressed at the same size and carries the same tag, so it appears identically on a Coomassie gel, but its dead active site should give no fluorescent band. A fluorescent band in the wild type and no fluorescent band in the mutant is strong evidence that the signal is FAAH catalysis and not background labelling of a host serine hydrolase.

We have not built the catalytic mutant yet. Running TAMRA-FP without it would produce a picture we could not defend, so the assay waits on the control.

• Positive: wild-type 6xHis-FAAH lysate
• Negative: catalytic mutant lysate
• Second negative: heat-inactivated wild-type lysate
• Background: lysate from cells carrying the empty backbone

This is the measurement the project turns on. FAAH hydrolyses OEA into oleic acid and ethanolamine. Oleic acid is a non-esterified free fatty acid, and NEFA colorimetric kits quantify free fatty acids on a standard plate reader. So rather than trying to measure the disappearance of OEA directly, which needs mass spectrometry, we measure the appearance of its product.

Assay design

1. Reaction. Incubate purified OEA substrate with the enzyme source, either lysate or whole cells, in buffer at 37 °C.
2. Timepoints. Sample at fixed intervals and quench, so the readout is a rate rather than a single endpoint. A single endpoint cannot distinguish a fast enzyme that has run out of substrate from a slow one still working.
3. Detection. Apply the NEFA colorimetric reagents to each quenched sample. The enzymatic cascade in the kit converts free fatty acid into a coloured product.
4. Read. Measure absorbance on a plate reader at the kit’s specified wavelength.
5. Quantify. Convert absorbance to free fatty acid concentration using an oleic acid standard curve run on the same plate.

Calibration standard design

• Prepare an oleic acid dilution series spanning the expected range of product formation, run in triplicate on every plate.
• Run the standard curve on the same plate as the samples, every time. Plate-to-plate and day-to-day variation in a colorimetric assay is large enough that a stored curve is not trustworthy.
• Fit a linear regression and report the R² alongside any derived concentration. Report sample values that fall outside the standard range as out of range rather than extrapolating.

Controls

The no-substrate blank matters more than it sounds. Bacterial lysate and rich medium both contain free fatty acids, and that background is subtracted from every sample, which is why it is read on the same plate rather than assumed.

What this design lets us compare

Once running, the same assay reads out across all our localization variants under identical conditions, which is what makes the comparison meaningful:

• Wild-type FAAH, membrane associated, as the reference
• ΔTm-FAAH, cytoplasmic
• YebF-ΔTm-FAAH, secreted
• OmpA- and OmpX-ΔTm-FAAH, surface displayed
• Catalytic mutant, for each of the above

Measuring whole cells rather than lysate is the version that answers the real question, because a probiotic in the gut cannot lyse itself to reach the substrate. A construct that works on lysate but not on intact cells has failed the test that matters, and running both is how that distinction gets made.

The NEFA route to measuring fatty acid amide hydrolase activity trades sensitivity and specificity for accessibility. LC-MS will beat it on both. What it offers instead is a measurement a high school or small university team can actually run, repeatedly, on equipment they already have, using a commercial kit and a plate reader.

For any future team working on N-acylethanolamine signalling, fatty acid amide hydrolases, or gut-localized enzyme delivery, the design above is a starting point that does not require instrument access. The controls listed are the part we would ask them not to cut, particularly the catalytic mutant, because without it the assay measures free fatty acid rather than FAAH activity.

We want to be clear about what we have and have not measured.

Measured. Induced expression of three constructs by SDS-PAGE, against uninduced controls.

Not yet measured. Protein identity by Western blot. Active site integrity by TAMRA-FP. OEA hydrolysis by NEFA assay. No kinetic constants, no calibration curve, and no activity comparison between localization variants have been produced.

Known limitations of the design.

• The NEFA assay reports total free fatty acid, not oleic acid specifically. The catalytic mutant and empty backbone controls are what separate FAAH-derived signal from background, and without them the number is not interpretable.
• Colorimetric kits have a limited dynamic range. Samples may need dilution, and dilution must be applied to the standards identically.
• Whole-cell measurements confound enzyme activity with substrate accessibility. That is a feature for answering the project question and a problem for extracting true kinetics.