If MALDI-TOF is capable of measuring the amount of product formed even when the substrate is in excess, then your assay idea may be feasible. A second requirement is that the sensitivity of the product measurement is sufficient to measure it under the initial velocity conditions required for a Michaelis-Menten analysis. Finally, it must be possible to do this at substrate concentrations that are below the Km.

Assuming all that is true, you should prepare samples that vary the time of reaction and the substrate concentration. You may also wish to add enzyme concentration as a variable. After measuring the product concentration for all the samples, you will plot the concentration of product versus time for each enzyme concentration. Measure the initial rate for each curve, i.e. the slope of the linear part of the curve, passing through the origin. Now plot the slope/enzyme concentration versus the substrate concentration and fit to the M-M equation to get the Km.

You will also need a standard curve of the product to find out whether the detection method gives a signal that is directly proportional to the concentration. If not, you can use the standard curve to convert the signal to the product concentration.

Potential pitfalls: 1. The substrate concentration range is too high and saturates the enzyme, so you see no increase in rate when you increase the substrate concentration. 2. The substrate concentration range is too low, so you don't get any curvature in your M-M plot. 3. The enzyme concentration is too low so you don't get measurable product formation. 4. The enzyme concentration is too high so you don't measure initial rates. 5. You can't measure product concentration because the substrate interferes. 6. Substrate inhibition at high substrate concentrations causes your M-M plot to go up then come down again, preventing you from measuring the Km without using an equation that accounts for substrate inhibition.

Using MALDI for finding Km may not be a good idea because MALDI is not quantitative. The ionization of the sample in MALDI varies from one spot to the other depending on various conditions like the type of matrix used, the co crystallization of sample with the matrix , then ionization of sample, etc. So there is large variability in results  from one sample to the other and cannot be accurately measured. LC ESI MS is a better option for all quantitative purpose.

As far as getting similar result in all peptide conc. is concerned. I think you are working with saturated conditions.First try to optimize the time and enzyme conc. where you get 10% of substrate conversion . And then vary the peptide conc. (-+5Km ) .You should get better result. I was facing the same problem before and I am trying to work in similar fashion. I hope it works.

that's not really a good idea because to implement what you want you would have to use labeling stable isotopes like ITRAQ (quantitative method), which is a quite expensive. therefore this methodology is a quite unnecessary for the experiments you would perform. I would advise you use other features on your sample as proteolytic activity, fluorescence or UV absorbance.

I have calculated km by MALDI/TOF MS, using a recent method that we have just published (Ritorto et al, Nature Communications, 2014). However, as MALDI  is a pulsed-laser method, quantitation without an internal standard is quite challenging. We quantified the production of monoUb by DUB and quantify it by spiking into the reaction a known amount of 15N-Ubi (=your internal standard should have the same chemical/physical characteristics of the analyte of interest, only heavier). It works really well. So, even though quantitation in complex samples is not possible by MALDI, in in-vitro conditions (not in cell) is quite robust (and fast!) methodology if an appropriate internal standard is used.

I agree with all the other comments about your biochemical approach and the possibility that is not optimize enough (=saturated conditions).

Hope it helps.

Stella