Yes, we can provide data sheets containing both ATP and Staurosporine IC50(and/or Ki) values on request. Overall these datasets show a good correlation with recombinant data, however there are overall shifts in potency (weaker binding in the KiNativ™ assay) for Staurosporine, possibly due to general protein binding (see “What is the protein concentration in your assay…”). There are also some kinases that show very different behavior in the KiNativ™ assay compared to the recombinant enzymes.

From the large number of compounds we’ve evaluated, it would be fair to conclude that it is most common to see a generalized decrease in potency in the KiNativ™ assay relative to recombinant assays, however the magnitude of this shift can vary from compound to compound, even within the same structural series. As mentioned above, these shifts can be due to generalized protein binding (lowering the effective compound concentration), or in some cases they can result from distinct behavior of the native kinase relative to the recombinant enzymes. We have observed clear examples of both of these phenomena. In general, the potency determined in the KiNativ™ assay shows a much better correlation with cellular/in vivo efficacy than data from recombinant assays.

Yes and no. We have observed true activation/inactivation of several kinases in multiple systems throughout our studies using the KiNativ™ platform. However, in some studies, we have convincing data that for some kinases our probes are NOT sensitive to activity/phosphorylation changes. However, it is highly unlikely that the probes will be sensitive to all activity altering changes that can occur with kinases. Kinase “activation” can occur through multiple routes including phosphorylation, dephosphorylation, acylation, altered localization, protein-protein interaction, proteolysis etc… Whether these changes will affect the affinity of the kinase for ATP/Probe or the structure of the ATP site in a way which can be detected with our probes is a unique property of each kinase and activation mechanism. There is clear evidence in the literature suggesting that for some kinases, activation events lead to dramatic shifts in ATP affinity, while for other kinases, there is no ATP affinity change in the active vs. inactive state. Overall, our approach is not a “magic bullet” for comprehensively monitoring kinase pathway activation, however, it is clearly a valid approach for these types of studies and will likely lead to many new discoveries regarding kinase activation mechanisms.

Our probes contain an ATP moiety and utilize ATP binding energy of the kinase to achieve selective labeling. If an allosteric inhibitor affects the structure of the ATP site and/or ATP affinity it is likely that some change in probe labeling will be seen. However this really depends on the nature of the allosteric inhibitor. We have currently only evaluated two allosteric inhibitors (GNF2/ABL and KL3-75/AKT) and in both of these cases, there was no effect on probe labeling. Our method does provide a rather unique ability, however, to explore the mechanism of allostery, and thus may prove useful for detailed studies. We can determine ATP Ki values in the presence/absence of allosteric inhibitor for example. Or we can determine the ability of an ATP-competitive inhibitor to bind to a kinase in the presence/absence of an allosteric inhibitor.

Yes, we have extensive experience running the KiNativ™ platform in both tissues and cell lines. We have evaluated nearly every tissue that is available and the method works well in all tissues examined..

The KiNativ™ method will perform equally well in any organism with a well annotated genome (human, mouse, rat, dog, drosophila, C. elegans, E. coli etc…). Even for species with weak genomic information, such as monkey and hamster, we have obtained very satisfactory results by searching in the databases from similar organisms (human database for monkey, rodent database for hamster). We have validated the method in every clinically relevant species and nearly every tissue imagineable.

No, the probes are not cell permeable, so we have to lyse cells prior to adding probe. However, we can evaluate lysates from cells that have previously been treated with a compound (see “Can I treat live cells with my compounds and then analyze…”).

Yes, we have validated protocols for quickly washing away free compound and then lysing cells in a minimal volume to best preserve the cellular compound levels (overall ~4X dilution from cellular concentration of compound). When live cells are treated, there are possibilities of indirect changes in activity/expression that are not a result of direct compound/probe competition. Also, cell permeability of the compound will be reflected in the overall level of inhibition. Generally this method is best used as a followup to standard lysate treatment studies, so that it is possible to interpret the level of cell permeability (based on inhibition shift) and also whether any indirect activation/inactivation events have occurred.

There are three options for evaluating new cell lines. The best option depends on your budget and goals:

1. If you have a project where it is important to have the most comprehensive coverage of the specific set of kinases expressed in a particular tissue or cell line, we can generate a custom MS protocol or “target list” that will cover all probe-addressable kinases in that cell line/tissue. This process involves a series of exhaustive MS runs to identify all kinases present, followed by computational/bioinformatics work assemble the actual MS program. There is a one-time setup fee for this “target list generation”. To achieve best coverage, we always evaluate the new proteome with both an ATP and ADP based probe (see FAQ about differences between probes) and two “target lists” or “panels” are constructed in this process. After paying the initial setup fee, all subsequent profiling in your custom proteome is charged at our standard rates.

2. If you are most interested in monitoring a particular kinase or set of kinases in a cell line/tissue of your choosing, and the kinase(s) of interest are already listed in one of our standard panels, we can run the MS program corresponding to any of our standard panels in any proteome with no setup fee. In this case, the customer must provide the cell/tissue(s) they would like to analyze. It is also important to realize that we will only be able to see the kinases that are present in the cell/tissue being analyzed, so we can’t guarantee which kinases that are typically seen in the selected “panel” will actually be detected in your specific proteome.

3. If you have a project where it is important to comprehensively cover the specific complement of kinases in a cell line/tissue of your choosing, but do not have the budget to set up a custom target list, we can perform “data dependent” analysis (see FAQ on the difference between “data dependent” and “targeted” analysis) of your cell line/tissue in the presence/absence of compound/treatment. This method typically enables evaluation of approximately 80% of the kinases that would be monitored in option #1, however multiple follow-up analyses are typically required to achieve reliable quantitation on all available kinases. Because of the necessity of follow up (targeted) runs, the per compound costs of this method are higher than the per-compound costs after the setup fee in option #1.

For each treatment group (i.e. 1 compound dose x 1 panel) we prepare duplicate samples for MS analysis, starting with 0.5 ml of lysate per sample (1 ml total for duplicates). Duplicate control, untreated, samples are required for every ~3 treatment groups. We have done analyses at protein concentrations as low as 1 mg/ml with good results, however this is very cell line/tissue dependent. If possible we prefer to work in the 5-10 mg/ml range. Cell lines typically yield ~40 mg of total protein per 1 ml cell pellet. Tissues average ~30 mg of protein per g wet weight. If you are doing an analysis in one of our standard panels, we provide the cell lysates at no cost to you. If you are planning a project any other cell line/tissue, you will have to provide the cell pellet/tissue (or we may be able to obtain it for an additional charge).

No, our “one panel” pricing is the minimum unit price. Our method is an “all or none” analysis. In order to evaluate 1 kinase or 120 kinases, we have to prepare the same amount of cell lysate, and carry it through the same protocol, including a full MS analysis (i.e. full length gradient on LCMC/MS). We are currently evaluating other analysis methods that may allow for more cost effective analysis of smaller sets of kinases, but for now, our minimum kinase set is ~100 kinases or one “panel”.

Our “standard panels” cover the kinases that we are able to evaluate in HL60, PC3 and HeLa cells (plus Jurkat’s for lipid kinases). There are many kinases that our probes can lable that are not expressed in these lines. We have evaluated a large number of cell lines and tissues from humans as well as several other species, and overall we have confirmed that our probes label ~80% of the known kinases. If you have a particular kinase(s) you are interested in evaluating with our technology, let us know and we can check our database and tell you if we have seen it and what cell line/tissue could be used to evaluate your kinase in.

This is not a problem, but it’s best to let us know so we can plan the experiment accordingly. Our standard pre-incubation is 10 minutes (compound + proteome prior to probe addition). If you feel your compounds will not reach equilibrium in this time, let us know and we can extend the pre-incubation time.

We return data in the form of Microsoft Excel files. Other formats (pdf) can be provided upon request. Sample data files (ATP and/or Staurosporine IC50 values) can be provided upon request. Due to the nature of the platform, the data reporting files are somewhat different than what is typically provided from recombinant enzyme panels. We provide an explanation of the data format when data is returned.

For compound/inhibitor experiments untreated control samples are evaluated. We do not evaluate positive control samples with each set of samples. The MS data is extremely rich and there are numerous data features that are used to verify the quality of each run. Because all data for a given panel is collected from each sample/MS run, the data is highly internally controlled.

The simple answer to this question is that we use specialized Mass Spectrometry protocols for evaluating kinases that provide the most comprehensive and quantitative data. When we want to globally evaluate other ATPase targets of the probes, we run a more generalized MS protocol that allows the evaluation of a broader range of probe targets (with the trade off of reduced sensitivity and less precise quantitation).

More specifically, for kinases, we have determined a priori which kinases are present, and which ions in the LCMS/MS runs are the most reliable to monitor for those kinases. We know where they migrate and what the ion mass/charge ratio is. All of this information is programmed into the mass spec instrument so that it can collect fragmentation spectra from these “targeted” ions in the regions where they are known to elute. When we collect data in this way, we are able to use fragmentation spectra (MS/MS scans) for quantitation. Sensitivity is improved 10-50X over parent ion (MS scan) based quantitation methods, and reliability is dramatically improved since the fragment spectra are effectively a third level of separation that cofirms the sequence of the peptide being analyzed. The only trade-off with this approach is that due to data collection speed limitations, we cannot collect data for more than ~140 targets in a single MS run. This approach is very similar to “multiple reaction monitoring” (MRM) approaches that are very commonly used for quantitative drug metabolism and pharmokinetic studies and are gaining poplularity for proteomics quantitation.

Currently, we have not set up any “targeted” MS protocols for non-kinase ATPases, so to evaluate probe targets beyond the protein and lipid kinases, we use a more traditional proteomics approach on the MS instrument called “data-dependent analysis”. In this method, the MS instrument collects fragment (MS/MS) spectra “on the fly” based on what ions it sees in prior parent ion (MS) scans. There is a cycle of scans starting with an MS scan, and then up to ~20 MS/MS scans that are determined based on the ions seen in the original MS scan. In this method, fragmentation spectra (MS/MS) for more than 1000 distinct peptides can be collected in a single run. However, because these fragment spectra are collected more randomly across the true peak, they cannot be used for quantitation. Therefore quantitation must be based primarily on the signals in parent ion (MS) scans. Noise levels in MS scans are much higher than MS/MS scans and there is a much higher chance that two parent ions with the same mass/charge will co-migrate (and are indistinguishable at the MS level). So the quantitative sensitivity, precision and reliability in this approach is lower than in the targeted approach. However it is a good, quick way to look across a broad range of possible targets for potential hits which can be confirmed by additional targeted runs.

It is worth noting that we can set up “targeted” MS programs for any probe targets. We chose protein kinases initially because the market for protein kinase profiling is larger than any other probe target class. We next set up methods for lipid kinases based on customer input and requests. If there is an ATPase (or other nucleotide binding protein) class that you are interest in, let us know as we plan to continually increase the space that we can cover with targeted MS methods.

These two terms refer to the mode that the MS instrument is operating in when we collect data. These two data collection options serve different purposes and each has it’s advantages and disadvantages.

“Targeted” analysis is the most sensitive and quantitative method we employ and it is how we evaluate any pre-fabricated “panel” for protein or lipid kinases. In order to run the MS instrument in targeted mode, however, there is a considerable amount of knowledge required of the system being analyzed. More specifically, to perform “targeted” analysis for kinases, we have to determine, a priori, which kinases are present in the sample being analyzed, and which ions in the LCMS/MS runs will be the most reliable to monitor for each of those kinases. All of this information must be determined by rather exhaustive sample analysis and bioinformatics work but this work only has to be done one time per panel. The migration time and mass/charge information for each ion is programmed into the MS instrument so that it can collect fragmentation spectra (MS/MS spectra) from these “targeted” ions in the regions where they are known to elute. When we collect data in this way, we are able to use fragmentation spectra (MS/MS scans) for quantitation. Sensitivity is improved 10-50X over parent ion (MS scan) based quantitation methods due to dramatic reduction in noise levels, and reliability is improved since the fragment spectra are effectively a third level of resolution that confirms the sequence of the peptide being analyzed. Aside from the up-front work, the only trade-off with this approach is that due to data collection rate limitations, we cannot collect data for more than ~140 targets in a single MS run. This approach is very similar to “multiple reaction monitoring” (MRM) approaches that are very commonly used for quantitative drug metabolism and pharmokinetic studies and are gaining poplularity for proteomics quantitation.

“Data dependent” refers to a method where the MS instrument collects fragment (MS/MS) spectra “on the fly” based on what ions it sees in prior parent ion (MS) scans. There MS instrument performs a repeating cycle of scans starting with an MS scan, and then up to ~20 MS/MS scans, each of which is determined based on the ions seen in the original MS scan. In this method, fragmentation spectra (MS/MS) for more than 1000 distinct peptides can be collected in a single run. However, because these fragment spectra are collected more randomly for each ion, MS/MS (fragment) spectra cannot be used for quantitation. And in many cases, low level ions may only be detected in a small fraction of runs. Therefore quantitation must be based primarily on the signals in parent ion (MS) scans. Noise levels in MS scans are much higher than MS/MS scans and there is a much higher chance that two parent ions with the same mass/charge will co-migrate (and are indistinguishable at the MS level). So the quantitative sensitivity, precision and reliability in this approach is lower than in the targeted approach. However it is a good, quick way to look across a broad range of possible targets for potential hits which can be confirmed by additional targeted runs.

For our standard protocol we lyse cells (resting, non-stimulated) in the presence of detergent (0.1% Triton X-100), remove insoluble material by centrifugation, and then gel-filter the lysates to remove endogenous nucleotides. We do not include phosphatase inhibitors as we have not found that they make a significant difference in the resulting profile (at least for resting cells). The final buffer the lysates are gel filtered into contains Hepes pH 7.8, 150 mM NaCl, 20 mM MnCl2, and 0.1 % Triton X-100.

The KiNativ™ method can be used in a wide variety of conditions, including soluble (no detergent) lysates, solublized membrane fractions (detergents present), intact membranes (no detergent), and solubilized or non-solubilized whole cell extracts. Most standard buffers are compatible as are most buffer additives (phosphatase or protease inhibitors, reducing agents etc…). Please let us know if you would like to perform experiment using specialized lysate or buffer conditions.

Our standard cell lysate preps are labeled at a concentration of 5-10 mg/ml. We have observed several cases where compounds that exhibit high protein binding will appear to be much weaker in the KiNativ™ assay than they will in recombinant assays lacking adequate carrier protein. To avoid running a compound at a dose that is too low to see any inhibition, it is best to confirm that your compound is still active in recombinant assays in the presence of comparable carrier protein to our conditions (~1% albumin is equivalent to our conditions). It is also important to realize that cell based assays conducted in the absence of serum will not account for high protein binding properties of a compound since the compound can accumulate in the cells until it’s free concentration inside the cell is close to it’s free concentration outside the cell. Our data in general, correlates better with cell based assays performed in complete media (with serum).

Our probes are irreversible, yet we are able to determine very accurate binding constants of reversible inhibitors. We have optimized our probe labeling conditions to ensure that in the majority of cases, the conditions will allow for direct binding constant determination. To elaborate, in the presence of a reversible inhibitor, the labeling rate of the probe will be reduced directly in proportion to the fraction of the target bound by the reversible inhibitor at equilibrium. In order to determine an accurate binding constant from a single kinetic (time) point, the probe must still be in a roughly linear phase of labeling (less than ~50% fractional labeling of the target will provide data within ~2-fold of the true labeling rate decrease). We have found that the labeling rates of kinases with our probes fall within a fairly narrow range and thus a single probe concentration/time point can achieve roughly linear kinetics with nearly all kinases. For a small number of targets, we have determined that the labeling kinetics are beyond the linear phase and through probe dose/time course studies, we have determined the correction factor required to convert the “IC50” values back to Ki values. These correction factors are a property of only the probe and the target (kinase) and thus do not need to be determined in each subsequent experiment.

We have found that the competition from endogenous nucleotides (ATP mainly) varies depending on the cell/tissue source. This may be due to varing level/activity of nucleotide hydrolases. Our standard lysis protocol includes a gel-filtration step to remove any endogenous nucleotides or other small molecule competitors, and this generally improves data consistency. If necessary, we can analyze non gel-filtered lysates, and the data quality is usually acceptable.

There is no fundamental difference in the behavior of the two probes, however we use both to maximize our coverage of kinases. Both probes are acyl-phosphates of nucleotides with the acyl-phosphate (lysine reactive) group attached the terminal phosphate of the nucleotide. As the name implies the ATP probe contains ATP with the acyl-phosphate attached to the g-phosphate, while the ADP probe contains ADP with the acyl group off the b-phosphate. Functionally the probes behave very similarly towards kinases with 80-90% of kinases showing robust labeling by both ATP and ADP probe versions. However there are a small number of kinases that label much better with one probe or the other.

We perform all assays in duplicate. In most cases quadruplicate control (untreated) samples are evaluated which are run at the beginning and end of each set of experiments.