MS/MS Response Factors and the use of calibrated P/IS mixtures

The analyte-to-internal standard MS/MS response factor (hereafter called the RF) is given by:

 

                           

 

 

where MRM intensity is the area under the MRM peak for the analyte or internal standard (provided by your MSMS software).

 

If you injected into the MSMS a mixture that truly had equal moles of analyte and internal standard, you might expect the ratio of MRM intensities to be unity (especially for internal standards that are deuterated forms of the products), but it may be different from unity because your quadrupole mass filters are not perfect, and there could be an isotope effect on the parent ion fragmentation (in the case of isotopic internal standards).

 

For example, if you inject a mixture of analyte and internal standard with a true mole ratio of say 0.9 and you observe a ratio of MRM intensities of 1.1, then  RF = 1.1/0.9 = 1.22

 

RF is used to obtain the µmoles of analyte in your sample as follows:

 

 

 

 

For example, if (MRM intensity for analyte)/(MRM intensity for internal standard) = 1.0 and if you added 1 µmole of your internal standard and if RF = 0.9 then µmole of analyte in the sample is:            

                                                                                           

 (1 x 1) / 0.9  = 1.11 µmole

 

     Why do we care about RF?  If you use a single MSMS instrument and you report your results as percent of daily mean enzyme activity, you don't care about RF since it cancels out of the percent of the daily mean. If on the other hand, you report your results as absolute µmole/hr/L, you need to know RF to be accurate in the absolute value.

A second point is that RF may be different on different MSMS instruments, and this would create problems even when you report the percent of daily mean if the daily mean is calculated from data obtained on multiple machines on the same day.

RF on any given MS/MS instrument may change slowly over time.  We don't really have much data on this, and the drift, if any, is expected to be slow and slight.

 

       GelbChem provides vials of products (P) and internal standards (IS) for each lysosomal enzyme assay in which the P/IS mole ratio is accurately known (i.e. the truth). Below we explain how we can know this, but first, we give the suggested use of these P/IS vials. The idea is that you use this P/IS mixture to obtain the RF using the formula at the top of this document and the certified P/IS mole ratio that is provided with the vials.  You can do this analysis for each MS/MS instrument and compare the two RF values, and you can use each RF value to obtain the true µmole/h/L for data collected on each machine. If the RF values are the same on both machines and you are reporting your results as percent of daily mean (or medium) you don't need to use the RF value in your calculation, but if the RF values are different you need to correct the data on one machine before you compute the daily mean using data from both machines.

Perhaps you can measure RF say once per month per machine.  If RF is not changing over several months, maybe you can measure RF every 6-12 months for example.  It is a good idea to measure RF after a preventive maintenance or major instrument repair.

 

        For most lysosomal enzyme assays 2-3 ng of the internal standard is injected to the MS/MS instrument.  Each vial of P/IS provided by GelbChem contains about 300 ng of P and 300 ng of IS. Add 1.5 ml of acetonitrile/water (1/1) to the vial and place it into the autosampler on your MS/MS instrument and inject 10 µL. Measure the MRM intensity for P and IS, and calculate RF according to the equation at the top of this document. Then store the vial in the freezer for later use.  Note that solvent loss due to evaporation is not a factor since it won't change the P/IS mole ratio.  Also, the exact molar concentration of P and IS in the stock solution is not important, only the P/IS mole ratio, which is provided in the certificate from GelbChem. If there is some chemical decomposition of P and IS in these vials it is of no concern since P and IS would decompose at the same rate (they are the same chemicals but IS has deuterium); thus, the P/IS mole ratio will not change.  With 1.5 mL of P/IS solution, you have enough for 150 injections, or 12 years if you measure RF once per month.

 

One final and important note:

      You may think that you don't need to worry about RF because you are using PerkinElmer or CDC QC DBS standards to check the stability of your assays over time.  Time will tell if this is adequate, but please note that PerkinElmer and the CDC are NOT checking their RF when they test their lots of QC DBS.  If their RF values change slowly over time, they will not know this.  They will just assume that their RF did not change and indicate on the package of the QC DBS what the measured µmole/h/L values are.  This could lead to an unknown change in the truthfulness of these µmole/h/L values.  This is a potentially troubling scenario, but the good news is that with MS/MS there is an easy solution, which is to keep track of your RF values.

 

 

 How do we measure the true P/IS mole ratio in the certified samples sent to you?

 

       It is NOT a good idea to measure out a known weight of P and IS since you should not assume that P and IS are PURE BY WEIGHT.  Only if they are pure by weight you can calculate the moles of P and IS from the amount weighed.  Be careful with purity by weight statements from vendors and elsewhere.  For example, a lipid may be determined to be free of other lipids when analyzed say by thin layer chromatography or HPLC, but this type of analysis cannot detect impurities such as water and other weight contributors that are not observed. 

     The best method to measure the actual moles of a compound is quantitative nuclear magnetic resonance (qNMR).  This technique gives sample signals that are proportional to the moles of hydrogen atoms attached to the molecule of interest.   Each of the different hydrogen atoms in the molecule gives different NMR signals (some give the same but most do not, and this is well understood based on the chemical structure of the molecule).  One can add an internal standard to the qNMR sample, and the mole amount of this compound is known a priori because it comes as a certified qNMR internal standard from a qualified qNMR vendor. Suppose your analyte of interest gives a qNMR signal of 100 for one of its hydrogen atoms and the analyte has only one of these types of hydrogen atoms.  Suppose 50 µmole of internal standard added to the sample of analyte gives a qNMR signal of 200 for one of its hydrogen atoms, and it has only one of these types of hydrogens. Thus, the µmole of analyte in the qNMR sample is given by:

 

(100/200) x 50 µmole = 25 µmole

 

Now, let's think a bit more.  What if the qNMR signal seen for the analyte is due to the analyte together with an impurity in the analyte that also contributes to the qNMR signal? Thus the signal intensity would overestimate the amount of analyte in the sample. Actually, the qNMR spectrum for the analyte gives a signal for each type of hydrogen atom in the molecule, and the number of each type is known from the structure of the analyte. For example, suppose the analyte structure dictates that there are 2 A-type hydrogen atoms and 3 B-type hydrogen atoms. Then the qNMR signal of A divided by that of B should be 2-to-3 if the A and B signals are not contaminated by impurities. But wait, the skeptic may suggest that both qNMR signals are contaminated by impurities to the same extent. Well, you can play this game for several different types of hydrogen atoms in the analyte, and if all the ratios agree, you can pretty safely assume that there are no contributing impurities.

 

What we did to certify the P/IS vials that you have received is the following:

 

1.  We obtained the qNMR of P and looked at the ratio of signals to be sure the signals are free of contributions from contaminants. We do not need an internal standard for this.

 

2.  We did the same thing for IS.

 

3. We mix P and IS in some approximate mole ratio, we don't care yet about the exact ratio.

 

4.  We look at the peak from the hydrogen atoms that are common to P and IS.  For example, if both compounds have a methyl group we integrate the peak due to the methyl group contributed by both IS and P.  Suppose the IS has a benzene ring with deuteriums but the P has these as hydrogen atoms. The deuteriums are invisible in the NMR spectrum. From the integration of the benzene hydrogens (coming from P-only)  compared to the integration of the hydrogens that come from both P and IS (methyl in this example), we can get the exact P/IS mole ratio. Note that nowhere in this process do we need a qNMR internal standard. A qNMR internal standard is needed only if we want to certify the true mole amount of some compound in the vial, not if we just specify the P/IS ratio.

1.png
1.png