assay · intermediate · partly written
Enzyme activity assay
An enzyme activity assay measures how fast a purified enzyme makes product, then asks how a compound slows that rate under conditions the experimenter chose.
Also called: enzyme kinetics assay, kinase assay, enzymatic assay, biochemical inhibition assay
In short
An enzyme assay does not measure binding. It measures how fast product appears, then asks how a compound slows that. A hit is a slower rate, under conditions you chose. IC50 moves with substrate concentration; Ki does not. A hit is a statement about a rate, and the conditions are part of the result. 123
Reported, not yet replicated
His line: inhibition means an enzyme assay
He said: "If it is inhibition, then maybe you have an enzyme assay, and if it is activation, maybe you have something else, right? You have some pipeline."
The first half is right for a target that is an enzyme: the rest of this record is what that assay is. The second half, what the equivalent is for activation, is his open question and is not answered here; see open_questions. 4
Amit's hypothesis
The curve, and why only the first 10% counts more
Leonor Michaelis and Maud Menten proposed the model in 1913, building on Victor Henri. Reaction speed rises with substrate, then flattens: v = Vmax[S] / (Km + [S]). Vmax is the ceiling rate for a given enzyme amount; kcat is turnover per enzyme per second (Vmax = kcat x enzyme). Km is the substrate concentration at half Vmax; it is often mistaken for affinity but is not. The assumptions are enzyme far below substrate, steady state, and initial rates measured.
Plot product against time: early on it is a straight line, then it bends as substrate runs out or product builds up. Initial velocity means staying where less than 10% of substrate is used or product formed; outside it, the guidance says the kinetic treatment is invalid. In practice, run time courses at three or four enzyme concentrations and keep substrate in large excess (ratios typically above 100).
Km is measured as initial velocity at substrate from 0.2 to 5 Km, six concentrations first and then 8 or more, fitted by nonlinear regression; linearised plots should not be used, because inverting the data amplifies error. For kinases, ATP Km is measured with the other substrate saturating. 523
Reported, not yet replicated
IC50 versus Ki, and why screens sit near Km more
IC50 is the compound concentration giving 50% inhibition under your conditions. For simple competitive inhibition Ki = IC50 / (1 + [S]/Km) (Cheng-Prusoff). Higher substrate or more conversion raises IC50. Worked consequence, the atlas's arithmetic: at [S] = Km, IC50 is 2 times Ki; at 5 times Km, 6 times Ki.
A competitive inhibitor and substrate fight for the same pocket, so flooding the assay with substrate hides the inhibitor. That is why a substrate concentration around or below Km is ideal for finding competitive inhibitors.
Cheng-Prusoff converts one number to the other only for simple competitive inhibition; applied to non-competitive or mixed inhibition it gives a wrong Ki, so the mode must be found first. 321
Reported, not yet replicated
Mode of inhibition more
Hold the compound near its IC50, vary the substrate, and watch which parameters move.
Competitive: Vmax unchanged, Km rises. Uncompetitive: Vmax falls, Km falls. Non-competitive: Vmax falls, Km unchanged. Mixed: Vmax falls, Km rises or falls.
The guidance's example is a kinase inhibitor competitive with ATP and non-competitive with the peptide substrate: the mode depends on which substrate you vary. 62
Reported, not yet replicated
Time dependence and tight binders deep
Some inhibitors act slowly: a weak complex then tightens, so the apparent Ki drifts with time. Some are so tight that their concentration approaches the enzyme's, which breaks the standard curve; the lowest measurable IC50 is half the enzyme concentration. Covalent inhibitors make IC50 invalid; the rate constant kobs/[I] is used instead. Substrate depletion complicates slow-binding analysis when substrate is below Km, and numerical fitting is needed when inhibitor and enzyme concentrations are comparable. Residence time, the lifetime of the drug-target complex, is argued to govern in vivo activity more than affinity; that is one author's model, from an Epizyme scientist, and only the abstract was read. 6278
Reported, not yet replicated
A real programme: the STK33 screen deep
The STK33 primary screen (PubChem AID 2661) used purified STK33 at 50 nM, myelin basic protein at 0.5 mg/mL and 2.5 nL of 10 mM compound in a 1536-well plate. Steps: enzyme and compound for 15 minutes; add ATP, 60 minutes; ADP-Glo Reagent I (stops the reaction, destroys leftover ATP) 40 minutes; Reagent II (turns ADP back into ATP for the luciferase reaction) 30 minutes; read on a ViewLux. The final compound concentration is about 10 uM if 2.5 nL goes into about 2.5 uL (the atlas's arithmetic from stated volumes; the record does not say).
The campaign: 321,908 PubChem substances in the single-point screen, 235 actives, retest at 8 concentrations from 0.1 to 12 uM (AID 2821), then chemistry to ML281 at 14 nM IC50, about 550-fold over Aurora B and over 700-fold over PKA. In the paper the kinetic mode, Km(ATP) and Z-prime are not reported; the assay details are in supporting information that was not opened. 9101112
Reported, not yet replicated
Where it breaks, and how practitioners check more
Past initial velocity: the rate is no longer proportional to enzyme and the Michaelis maths is invalid. Check: time courses at several enzyme levels.
High substrate: competitive inhibitors vanish and IC50 inflates. Check: screen at or below Km and report [S]/Km.
Cheng-Prusoff misapplied: wrong Ki for non-competitive or mixed modes. Check: determine the mode first.
Tight binders: IC50 floors at half the enzyme concentration. Check: lower the enzyme or use a quadratic fit.
Slow or irreversible binders: IC50 depends on incubation time. Check: vary pre-incubation; use kobs/[I] for covalent inhibitors.
Readout artefacts: in 3,048 Lilly biochemical assays, PAINS-matching compounds were active 13.8% of the time versus 9.6% overall; the alerts mostly flag AlphaScreen interference; and 41 of 61 alerts showed at least 1.5-fold enrichment for high Hill slope (above 2.0), linked to aggregation. The paper recommends confirming with a biophysical assay and mechanism studies.
Shared luciferase: in the STK33 screen every biochemical step used the one luciferase chain, so a luciferase inhibitor would lower light like a kinase inhibitor. This is the report's concern, not a documented ADP-Glo finding.
Enzyme instability: curves flatten during the assay; check with progress curves. Background: EDTA is not recommended as the background control in kinase validation; use enzyme-free or substrate-free wells. 2313791
Reported, not yet replicated
Cost and throughput, with basis more
One academic core rate sheet (file name suggests 2020-05-27; date not shown in the text; reagents supplied by the PI's lab) prices a primary biochemical screen of 25,000 compounds at $7,041 internal and $13,377.90 external, about $0.28 per compound internal (the atlas's division), and a 60-compound, 8-point, n=2 dose-response confirmation at $1,837 internal. These are academic core rates, not industry cost, and likely higher now. The incubation for one STK33 plate sums to 145 minutes. No opened page lists a price for a Km or mode-of-inhibition study. 1491
Reported, not yet replicated
Question answered
Readout physical
Light, absorbance or fluorescence from a product (or a coupled product) in a well, over time or at one fixed time. In the STK33 screen: purified kinase, substrate and ATP; ADP-Glo Reagent I stops the reaction and destroys leftover ATP, Reagent II turns ADP back into ATP for a luciferase light reaction, and the plate is read on a ViewLux. Less light means less ADP means less kinase activity. This is an endpoint, one reading after a fixed time, so initial velocity has to be proven beforehand in assay development. 9121
Reported, not yet replicated
Not measured
Whether the compound enters or survives in a cell, or whether the target matters in a cell; where it binds (mode of inhibition is a pattern in rates, not a structure); true affinity from a single-substrate-level IC50; selectivity, unless counter-screened (STK33 used PKA and Aurora B); and anything about a different protein form, since the enzyme is purified in buffer, not the complex in a cell. 111
Reported, not yet replicated
Setting
Where in cascade
Sample prep
Purified enzyme and compound in a well. In the STK33 screen: purified STK33 at 50 nM, myelin basic protein at 0.5 mg/mL as substrate, and 2.5 nL of 10 mM compound in a 1536-well plate; enzyme and compound were pre-incubated for 15 minutes before ATP. 9
Reported, not yet replicated
Conditions
Three choices decide what the number means: read only the first 10% of the reaction (initial velocity), set substrate near or below Km, and keep enzyme far below substrate. Break any and the Michaelis-Menten treatment no longer applies. In the STK33 screen the ATP stock was 250 uM, "about 1.5 x Km" per the record; the final well concentration is not stated. Controls: DMSO neutral wells on every plate (the last two columns got buffer instead of enzyme, not labelled as controls in the record), data normalised to plate medians; active meant 25% or more inhibition; mostly singlicate (635 compounds in duplicate). 291
Reported, not yet replicated
Throughput
The STK33 library was 321,811 compounds in 1536-well format (321,908 PubChem substances in the primary screen record), giving 13 submicromolar hits in the paper and 235 actives at the 25% cutoff in the screen record. The class "high" is the atlas's own sorting from this scale (inferred); no per-day figure was found. One academic core prices a 25,000-compound primary biochemical screen. 11914
Our inference
Time per run
Cost
Per sample
0.28 to 0.54 USD per compound, single-concentration primary biochemical screen (MSU Assay Development and Drug Repurposing Core recharge sheet: 25,000-compound primary biochemical screen at $7,041 internal and $13,377.90 external, reagents supplied by the PI's lab; per-compound figures are the atlas's own division. Academic core rate, not an industry cost. This is an academic core-facility recharge rate, not a commercial or contract-lab price; no commercial price was found.; as of 2020 (file name suggests 2020-05-27; date not shown in text; likely higher now))141
Per campaign
7,041 to 13,377.9 USD per 25,000-compound primary biochemical screen (Same MSU core rate sheet: internal rate (low) and external rate (high); reagents supplied by the PI's lab. The whole-campaign cost of a screen of the STK33 size was not found. This is an academic core-facility recharge rate, not a commercial or contract-lab price; no commercial price was found.; as of 2020 (file name suggests 2020-05-27; date not shown in text; likely higher now))14
Engineerable variants
Which substrate sits near Km
Choosing ATP near Km favours ATP-competitive hits; this decides what mechanism the screen can find. The guidance says a substrate concentration around or below the Km is ideal for finding competitive inhibitors, and higher substrate makes them harder to see (the second sentence of reasoning, that substrate floods the pocket, is the report's own). 21
Reported, not yet replicated
Enzyme concentration and time
Lower enzyme extends the linear range; time courses at three or four enzyme concentrations prove initial velocity. 2
Reported, not yet replicated
Pre-incubation
The STK33 screen used 15 minutes of enzyme plus compound before ATP, which lets slow binders act; the record does not say why. Varying it is the test for time dependence (the report's reasoning). 91
Our inference
Readout
Coupled luminescence (ADP-Glo) works up to 1 mM ATP and down to 5 uL per reaction. Other readouts exist but none was opened. 12
Reported, not yet replicated
Concentration series and fit
The guidance asks for 10 or more concentrations at 3-fold or half-log spacing, maximum inhibition above 50%, and 95% confidence limits within 2 to 5-fold. A standard IC50 curve is used, or a numerical fit when inhibitor and enzyme concentrations are comparable. 27
Reported, not yet replicated
Use cases
STK33 screen to the probe ML281 (Broad MLPCN)
A single-concentration luciferase-coupled kinase screen of 321,908 PubChem substances (321,808 compounds; the probe paper says 321,811 compounds) gave 235 actives, retested at 8 concentrations from 0.1 to 12 uM (active if IC50 within a log of the top dose), then chemistry to ML281 at 14 nM IC50, about 550-fold over Aurora B and over 700-fold over PKA. At 1 uM against 83 kinases only FLT3 and KDR were inhibited 25% or more besides STK33. The enzyme assay delivered exactly what it measures, a potent selective inhibitor of the enzyme; the cell result (no effect on viability up to 10 uM) is a separate lesson covered in the screening-cascade report. 910111
Reported, not yet replicated
ML role
Does
No machine-learning method specific to enzyme activity assays was found in the opened reports. The report's only parallel is its own comparison: IC50 versus Ki has the same shape as a docking score versus a measured affinity, both being numbers that hold only for the setup that made them. 1
Our inference
Cannot
Remove the conditions from the number: an IC50 connotes potency only together with [S]/Km and incubation time, so a model trained on IC50 labels from different assays mixes quantities that are not comparable. This follows from the guidance and the Cheng-Prusoff relation; no source tests it on a model. 231
Our inference
Data availability
PubChem BioAssay AIDs 2661 and 2821 (STK33) public
AID 2661 is the Broad STK33 single-concentration kinase screen with its reagent concentrations, incubation steps, 25% inhibition cutoff and substance and active counts; AID 2821 is the 8-concentration retest. 91011
Caveats. One target and one programme. Z-prime, Km(ATP) and the mode of inhibition for ML281 are not in the paper text read; the record states the ATP stock but not the final well concentration.
Reported, not yet replicated
Academic assay core (MSU Assay Development and Drug Repurposing Core) in-house
A published recharge sheet prices a 25,000-compound primary biochemical screen at $7,041 internal and $13,377.90 external, and a 60-compound 8-point dose-response confirmation at $1,837 internal. 14
Caveats. File name suggests 2020-05-27, date not shown in the text; reagents supplied by the PI's lab; academic recharge, likely higher now. No price for a Km or mode-of-inhibition study was found.
Reported, not yet replicated
Teach this
In one sentence
An enzyme assay reads how fast a purified enzyme makes product and how a compound slows it, so a hit is a statement about a rate under conditions you picked. 12
Reported, not yet replicated
The five-minute version
A plate of tiny reactions, each holding enzyme, substrate, ATP and one compound. The reaction runs inside the linear range that assay development already proved; chemistry turns leftover ADP into light; the plate reader reports one number per well; a compound scoring 25% inhibition or more in the STK33 screen is a hit. A retest in a dose series turns a hit into an IC50 under those ATP conditions. Only a substrate-variation series says whether the inhibitor is competitive, and only then can Cheng-Prusoff give a Ki. 921
Reported, not yet replicated
What a listener will ask
Why is substrate set near Km and not high?
Because a competitive inhibitor and substrate fight for the same pocket; flooding substrate hides the inhibitor and inflates its IC50, so the guidance calls a concentration around or below Km ideal. 2
Reported, not yet replicated
Why only the first part of the reaction?
Initial velocity means less than 10% of substrate used; beyond it the rate is no longer proportional to enzyme and the guidance says the kinetic treatment is invalid. 2
Reported, not yet replicated
If the enzyme assay shows a potent inhibitor, does it work in cells?
Not necessarily. The STK33 probe ML281 was a potent selective enzyme inhibitor with an IC50 of 14 nM, yet it did not meaningfully reduce cell viability up to 10 uM. 11
Reported, not yet replicated
The common wrong idea
A lower IC50 from an enzyme assay means the compound binds the enzyme more tightly.
An IC50 connotes potency only together with [S]/Km and incubation time; raise the substrate and a competitive inhibitor's IC50 rises while its Ki does not.
The one citation to hand over: 2
Still open here
- Amit asked what the equivalent is for activation. The opened reports do not cover activator assays; the only related finding is that qHTS on pyruvate kinase found activators as well as inhibitors (abstract only). Which assay formats confirm an activator, and what mode-of-action analysis replaces inhibition mode, is unanswered.
- No price for a Km or mode-of-inhibition study was found in any opened page, so only the primary-screen price from one academic core sheet is given.
- Time per run is given only as the incubation sum of one protocol; the full time from plate to result including reading and analysis was not found.
- Capital: no source on funding or company activity around enzyme assays was opened, so the capital field is unwritten.
- before_after is unwritten: the sources give the 1913 origin of the Michaelis-Menten model but not the four-step practice shift this field asks for.
- For ML281 the kinetic mode, Km(ATP) and Z-prime are not reported in the paper's text; the supporting information was not opened.
- Luciferase interference specific to ADP-Glo was not sourced: every biochemical step in the STK33 screen used one luciferase chain, so a luciferase inhibitor would lower light like a kinase inhibitor, but no opened source documents that for ADP-Glo. Treated as an unconfirmed concern.
- A rule of thumb for when to use the quadratic (Morrison) fit (Km under about 5 times the enzyme concentration) was seen only in a search summary of a primer whose page returned 404. Unconfirmed, not stated as a claim.
- How Km measured under assay conditions differs from Km in a cell, where ATP is already present at a concentration the screen never set, is a question the report raises without a source.
- No machine-learning method specific to enzyme assays was found in the opened reports.
- The Assay Guidance Manual text read is a 2008 archive; the current chapter (blocked in this run) may differ.
- The report was written on 4 distinct searches, below its floor of 8, so coverage is thin; readouts other than ADP-Glo were not opened.
Not yet written: Before after, Capital.
Linked from
- Dose-response assay
- Reporter gene assay
- Selectivity panel
- Surface plasmon resonance (SPR)
- Target engagement in cells (CETSA, NanoBRET)
- Western blot
- Dose response
- High-throughput screening
- Hit discovery
- Medicinal chemistry: potency, selectivity, ADMET
- The assay-error ceiling
- A hit series that binds and then fails in cells
Sources
- HowStuffWorks: enzyme kinetics assays (our own earlier write-up, unpublished). Kept in the atlas repository; not published.
- Assay Guidance Manual section 4, 2008 web archive (NCGC). NCGC via UNT web archive (2008). web, not peer reviewed; opened 2026-11-17 all records citing this
- IC50 (Wikipedia). Wikipedia. web, not peer reviewed; opened 2026-11-17 all records citing this
- Amit's dictation of 2026-11-17: what the atlas must contain. Kept in the atlas repository; not published.
- Michaelis-Menten kinetics (Wikipedia). Wikipedia. web, not peer reviewed; opened 2026-11-17 all records citing this
- Enzyme inhibitor (Wikipedia). Wikipedia. web, not peer reviewed; opened 2026-11-17 all records citing this
- Waley. Waley 1993 on slow- and tight-binding inhibition (abstract). PubMed Central (1993). paper; opened 2026-11-17 all records citing this
- Copeland. Copeland 2016 on residence time (abstract via Europe PMC). Nature Reviews Drug Discovery (2016). review; opened 2026-11-17 all records citing this
- PubChem BioAssay AID 2661: STK33 primary screen (ADP-Glo). PubChem. database, not peer reviewed; opened 2026-11-17 all records citing this
- PubChem BioAssay AID 2821: STK33 inhibitor retest, 8-point dose response. PubChem (2010). database, not peer reviewed; opened 2026-11-17 all records citing this
- Weiwer et al.. Weiwer et al. 2012: the STK33 probe ML281 (PMC full text). PubMed Central (2012). paper; opened 2026-11-17 all records citing this
- Promega TM313: ADP-Glo kinase assay protocol (revised 7/23). Promega (2023). vendor-page, not peer reviewed; opened 2026-11-17 all records citing this
- Lilly 2018 analysis of PAINS in biochemical assays (PMC full text). PubMed Central (2018). paper; opened 2026-11-17 all records citing this
- MSU Assay Development and Drug Repurposing Core recharge rates (file dated 2020-05-27). Michigan State University (2020). report, not peer reviewed; opened 2026-11-17 all records citing this
Where it sits on the map
History of this record (7)
This record has been changed 7 times, on the dates below. The note kept against each change is written for the people who maintain the atlas, in their own shorthand, so it is not shown here; it is in the record itself.
- Throughput class
- high
- Emerging
- no
- Status
- partly written
- Last updated
- 2026-11-19
- Record
- content/assays/enzyme-activity-assay.json