Without more details, it's difficult to provide advice, but I'll try anyway. If you are using a coupled enzyme assay (for example, I have often used pyruvate kinase and lactic dehydrogenase with phosphoenolpyruvate and NADH to measure ADP production by some other enzyme, using the 340 nm absorbance decrease when NADH is converted to NAD), make sure that your enzyme reaction of interest is rate-limiting, not the coupler reaction. Two things should be true: (1) Increasing the amount of coupler enzyme(s) has no effect on the rate. (2) The initial rate is directly proportional to the concentration of the enzyme of interest in the range at which you are working.

Also, make sure that the rate you are measuring is actually due to the enzyme of interest, not to background activity of the coupler. For example, if you are measuring ATP hydrolysis by the above-mentioned coupler, but the ATP you are using has a substantial contamination with ADP, increasing the ATP concentration in the assay will proportionately increase the ADP concentration, which gives a background signal. You should always subtract the rate from a background control in which you leave out the enzyme of interest but keep everything else the same.

Make sure the enzyme of interest does not have the same activity as the coupler to any significant extent by leaving out the coupling enzyme but keeping everything else the same.

Finally, check that your substrate stock solution has the concentration you think it has, by whatever means is available. As an example, you can check the concentration of an ATP solution by measuring its absorbance at 259 nm in a spectrophotometer with a quartz cuvette, using the Beer-Lambert law and the published extinction coefficient to calculate the concentration.

Hello Dr. Shapiro,

The assay I use is not coupled and is very basic. Both my substrate and its product absorb at 260 nm, but the product's extinction coefficient is lower than the substrate, which allows me to observe the reaction progress by monitoring the decrease in absorbance at 260 nm over time by using a plate reader. I have verified Vi to be directly proportional to enzyme concentration and I have also independently verified substrate concentration using Beer-Lambert. Are there any other common mistakes you can think of that would cause this? Similar to what you are alluding to, the velocity vs substrate concentration curve is so strangely linear that I feel like I may be monitoring something other than my actual reaction, but I feel like I have checked all the possibilities.

Hi Brian,

Can you increase the substrate concentration high enough to see curvature in the rate v. [S] plot? Have you tested the purity of your substrate?

Hi Dr Shapiro,

Thanks for your response. Unfortunately I can only test up to about 5X the expected Km or so because the baseline becomes saturated after that point and i don't trust the absorbance readings. I've plotted increasing amounts of substrate versus absorbance for our plate reader and found that after 2.0 AU the readings aren't reliable. I have not independently checked the substrate purity, but it is commercially bought and is about 95% pure based on the HPLC profile they provided.

You could overcome the absorbance limitation by shifting to a longer wavelength at which the substrate has a lower extinction coefficient.

Dr. Shapiro,

I ran the reaction at 280 nm and went to about 10 X Km, still with no flattening of the velocity curve. In fact, the velocity curve is so bizarrely linear that I must be making some massive mistake. My plot almost looks like a plot of absorbance vs concentration. If each reaction has a different baseline, as in this assay, is there some sort of normalization I need to be doing? I attached a picture of my velocity versus substrate concentration curve so you can see what I mean. It's almost a perfect straight line, which is not something you often see in biology. This is highly reproducible as well. 

Taken at face value, your plot is consistent with a Km that is considerably higher than the concentration range tested (There is a hint of nonlinearity near the origin suggestive of cooperativity, but this may be experimental error.) This would not be a problem except that it doesn't agree with the literature. What might be different between your work and the published work - the buffer composition, the protein, the source of the substrate?

Hi Dr. Shapiro,

Sorry for taking up so much of your time, but I feel like i can see the problem, but not what's causing it. It very much seems to me like I am failing to do some sort of normalization. The near perfect linearity I get is almost reminiscent of what you would see if you plotted Absorbance versus Concentration. Even more strange, is that my reaction velocities, once i put them in comparable units (nmol min-1 mg-1) are way higher than in the literature. For example vmax in the literature is about 180 nmol min-1 mg-1 while my lowest velocity is 217 and the highest is around 2000, yet still not shouldering. The graph I sent you yesterday just shows velocity in AU/second. It almost seems obvious to my that I am failing to normalize some part of the data, but I cannot find anything wrong with my calculations. 

If in your assay the absorbances of the substrate and product are directly proportional to their concentrations, according to the Beer-Lambert law, then there shouldn't be any need for normalization. The change in absorbance should be directly proportional to  product formation.

Since your rate does not reach Vmax at any point so far tested, it is reasonable that you calculate rates that are higher than the literature report, since that was based on Vmax occurring at a lower substrate concentration.

If you conclude in the end that there is nothing wrong with your data, and everything you are doing is the same as in the literature report, you come to the conclusion that there is a difference between your enzyme preparation and the earlier one. It could be a difference in the presence of a cofactor or an allosteric effector, for example.

Dr. Shapiro,

Thank you for all the help. If I could just ask one last question. Is it possible that the fact that my enzyme is a hexamer in its active state could account for the differences I'm seeing? Perhaps I need to use more enzyme to ensure the hexamer is favored. I am using ug of enzyme though, which I assume should be plenty.

Hi Brian,

If the concentration of enzyme is below the dissociation constant (Kd) of the hexamer, then the hexamer will dissociate into smaller oligomers and/or monomers. This could have an effect on the activity of the enzyme and its substrate affinity, especially if the substrate binding site is at the interfaces between monomers. If your assay has sufficient dynamic range, you could try measuring Vmax and Km as a function of enzyme concentration.

Hi Brian,

hopefully you already solved the problem. At the moment I'm also struggling with a way too high Km. I already figured out, that the substrate is causing background activity and the commonly used coupling enzyme preparation contains a competitive inhibitor. However, after circumventing these effects the estimated Km is still about 10-20 fold higher as expected. Another commercially available enzyme, which catalyzes the same reaction behaves as expected under the conditions applied. The sequencing results show no mismatches. The dissociation, mentioned by Dr. Shapiro, and the possibility of translational errors of the expression host are the last ideas I have.