When an Enzyme Endpoint Assay Gives Two Different Answers

Using 1/t in min−1: 20°C 0.105; 30°C 0.238; 37°C 0.476; 45°C 0.625; 55°C 0.833; 65°C 0.263; 75°C approximately 0. The rate rises with temperature to a maximum at 55°C in this data set, then falls sharply. The 65°C slowing and the complete failure at 75°C are consistent with thermal denaturation of the enzyme: above a critical temperature the tertiary structure unfolds, the active site loses its shape and substrate binding stops. The boiled-enzyme control confirms that the activity is protein-based, because boiling abolishes it.

However, 55°C cannot be accepted as the optimum from this experiment. The 45°C duplicate gave 2.9 min, a rate of 0.345 min−1, which is lower than the 37°C value of 0.476 min−1. That inconsistency means the temperature series is not reliable. The admitted failure to equilibrate the tubes means the enzyme was acting at an unknown, changing temperature during the early part of each run, and the effect would be largest for the short, fast runs at 45–55°C. The apparent 55°C peak may therefore be an artefact. The safest conclusion is that activity rises from 20°C to about 37–45°C and is abolished by 75°C, but the exact optimum needs a better design.

The iodine endpoint records the time at which all detectable starch has gone, not the rate at which starch is being hydrolysed at the start. As starch is depleted the reaction slows, so the endpoint is influenced by the starting starch concentration and by the subjective judgement of when the last trace of blue disappears. It is a threshold measurement, not a velocity, and it cannot distinguish a small change in rate from a change in the detection threshold.

An improved method would measure the initial linear phase. Use a colorimeter at about 620–680 nm to follow the starch–iodine complex, or use a DNS reducing-sugar assay to measure maltose production. Take readings at fixed intervals, plot absorbance or product concentration against time, and calculate the slope over the first 10–20% of substrate consumption, where the rate is approximately constant. Pre-equilibrate substrate and enzyme separately at each temperature for at least five minutes, then mix and start the clock. Standardise enzyme dilution and protein concentration, use the same starch batch, and replicate each temperature.

Controls should include a boiled-enzyme tube, a no-enzyme blank, a no-substrate blank, and a zero-time reading. Randomise the order of temperatures and use a calibrated thermometer in the bath, because a two-fold discrepancy such as the 45°C duplicate is too large to dismiss as pipetting error alone.

The most likely diagnosis is a pre-equilibration error. If the tube was still warming from the previous bath when enzyme was added, the enzyme acted at a lower temperature at the start of the run. Because the 45°C run was short, that early period is a large fraction of the total time, so the measured endpoint would vary according to how far the tube had warmed. Other contributors include inconsistent drop size on the spot plate, timing of the 30 s checks, variation in salivary amylase concentration between samples, and starch batch differences. The 1.6 versus 2.9 min difference is nearly two-fold, which points to a systematic temperature effect rather than simple pipetting noise.

To decide between a 37°C and a 55°C optimum, repeat the whole series with independent tubes pre-equilibrated at each temperature, and measure initial rates rather than endpoints. Include 37, 45, 50, 55, 60 and 65°C with at least three replicates, use a calibrated thermometer, and standardise the enzyme. If the initial rates show a broad plateau between 45°C and 55°C, the precise optimum matters less than the denaturation threshold. Human salivary amylase is generally expected to work best near body temperature, so a sharp 55°C peak in endpoint data should be treated with suspicion until the method artefact is excluded.