Determination of kinetic parameters and structure–activity relationships of ginsenosides as inhibitors of cyclin-dependent kinase 5/p25 using ultra-pressure liquid chromatography with triple quadrupole tandem mass spectrometry


  • Ning Chen,

    1. Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun Center of Mass Spectrometry, Changchun, China
    2. Graduate School of Chinese Academy of Sciences, Beijing, China
    Search for more papers by this author
  • Hongmei Yang,

    1. Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun Center of Mass Spectrometry, Changchun, China
    Search for more papers by this author
  • Jun Niu,

    1. Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun Center of Mass Spectrometry, Changchun, China
    Search for more papers by this author
  • Shuying Liu

    Corresponding author
    1. Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun Center of Mass Spectrometry, Changchun, China
    2. Changchun University of Chinese Medicine, Changchun, China
    • Correspondence to: S.-Y. Liu, Changchun Institute of Applied Chemistry, Chinese Academy of Sciences, Changchun Center of Mass Spectrometry, 5625# Renmin Street, Changchun 130022, China.


    Search for more papers by this author



Cyclin-dependent kinase 5 (cdk5) is a serine/threonine kinase that is reported to play an important role in the pathogenesis of Alzheimer's disease. Ginsenosides have beneficial effects on Alzheimer's disease in both in vivo and in vitro experiments, but the precise mechanisms are not yet entirely clear.


In the present study, an ultrahigh-pressure liquid chromatography (UPLC) and triple quadrupole mass spectrometry (TQMS) assay was developed to study the activities of cdk5 for the first time.


The calibration curves showed a good linear behavior over the range 0.04 μM to 10 μM (y = 0.934x + 0.045, R2 = 0.995) with product phosphorylated peptide (PKpTPKKAKKV). The screening results suggested that the inhibition activities of ginsenosides are related to their chemical structures.


The developed UPLC/TQMS-based method for determination of an inhibitor of cdk5/p25 is sensitive and reliable. The effect of ginsenosides on Alzheimer's disease may be involved with the regulation of activities of cdk5/p25. Copyright © 2013 John Wiley & Sons, Ltd.

The prominent neuropathological characteristics of Alzheimer's disease (AD) are neurofibrillary tangles (NFT) and senile plaques.[1-3] The main components of neurofibrillary tangles are abnormally hyperphosphorylated forms of the microtubule-binding protein tau, while brain tau has little or no phosphorylation under normal conditions.[4-11] Thus, it is important to identify the kinase responsible for phosphorylation of tau. Cyclin-dependent kinase 5 (cdk5) phosphorylates microtubule-associated proteins, including tau, Munc18, Map2, and DARPP32.[3, 12-14] Dysregulation of cdk5 is reported to play an important role in the pathogenesis of AD.[15] Although in vivo AD hyperphosphorylation of tau may result from the cooperative action of other kinases, in vitro evidence suggested that cdk5-mediated tau phosphorylation may be the critical step.[16] Cdk5 is activated by binding to one of its two activators, p35 or p39. The active form of kinase is a heterodimer of cdk5 and 25 kDa protein p25, which is a truncated form of p35.[17] Many studies have indicated that inhibition of cdk5/p25 may prevent initiation of tau hyperphosphorylation and in turn should preserve destabilization of microtubules and delay the onset of dementia in AD and other related tauopathies.[16]

In recent years, there has been increasing interest in kinases for finding new drug targets and it is estimated that about 30% of all drug discovery projects in the pharmaceutical industry are currently developing protein kinase inhibitors.[18] So, many biochemical and chromatographic assays have been developed to evaluate the enzyme activity and screen the enzyme inhibitors.[19, 20] For cdk5/p25, the widely used detection techniques are radioactive labeling methods[21, 22] and luminescent kinase methods.[23] However, the above methods are usually accompanied with some disadvantages, for instance high costs, time consuming, and they need of radio-isotope labeling substrates or a secondary reaction coupled to the enzymatic conversion.[24] Therefore, the demand for a label-free, non-radioactive assay scheme is high. Mass-spectrometry-based methods as a sensitive, rapid and accurate tool have been successfully and widely used to screen enzyme inhibitors.[22, 25-27] Up to now, no report has been found in the literature using a mass-spectrometry-based method to quantify the cdk5/p25 activities.

In the present study, a new screening method based on ultrahigh-pressure liquid chromatography (UPLC) and triple quadrupole mass spectrometry (TQMS) was developed. It can accurately quantify the amount of analytes by using a multiple reaction monitoring (MRM) function, which monitors the ratio of mass and charge (m/z) of precursor ions and diagnostic product ions.[25] During this process, the synthetic peptide PKTPKKAKKL (an excellent substrate for cdk5/p25[28, 29]) was applied in the enzymatic reaction. Compared with other existing analytical methods, the UPLC/TQMS method can detect the products or substrates directly and quantitatively, avoiding modification of substrates or secondary enzymatic reactions that are irrelevant to target enzyme reaction.[30]

In this study, the experimental conditions including pH value, temperature, reaction time, and the concentration of reagents such as peptide (PKTPKKAKKL) and adenosine-triphosphate (ATP), were first optimized. Subsequently, the validated method was used to evaluate the inhibitory activity of ginsenosides.

Ginseng has been a well-known herbal medicine in China, Korea, Russia, and Japan for thousands of years. In traditional Chinese medicine, ginseng has been used primarily as a treatment for weakness and fatigue. Nowadays, research studies have demonstrated that ginseng has pharmacological effects on the central nervous system as well as cardiovascular, immune, and endocrine systems.[31] Ginsenosides or ginseng saponins are the major active components of ginseng. Recently, increasing evidence of the beneficial effects of ginsenosides on AD has emerged from both in vivo and in vitro experiments.[32]



Cdk5/p25, adenosine-triphosphate (ATP), dithiothreitol (DTT), and MgCl2·6H2O were purchased from Sigma (St. Louis, MO, USA). Tris base was acquired from Promega (Madison, WI, USA). The peptides were synthesized by HD Biosciences Co., Ltd. (Shanghai, China). Acetonitrile and methanol (HPLC grade) were obtained from Fisher Chemical Company (Loughborough, UK). Ginsenosides were purchased from Jilin University (Changchun, China). All reagents were purchased in the highest purity and used without further purification. Ultrapure water (specific conductivity, 18.2 MΩ/cm) was produced by a MilliQ device (Millipore, Milford, MA, USA).

The enzymatic reaction assay

The assay was carried out in 50 μL of solution containing 50 mM Tris, at pH 7.4, 0.5 mM DTT, 5 mM MgCl2, 5 μM peptide (PKTPKKAKKL), and 20 nM cdk5/p25 at 30 °C. Reactions were initiated by the addition of ATP. After incubation for 30 min, the reaction was quenched with 50 μL acetonitrile containing 1 μM internal standard (IS), then centrifuged at 10000 g for 10 min and subjected to UPLC/TQMS analysis.

The peptide PKpTPKKAKKV (for clarity, “p” is used prior to T to indicate phosphorylation in that particular threonine residue) was used as the IS because its structure and ionization efficiency are very similar to the product peptide PKpTPKKAKKL.


Aliquots (5 mL each) of each enzymatic reaction solution were analyzed using an ACQUITYTM UPLC system (Waters Corp., Milford, MA, USA) with a thermostated autosampler at 4 °C. The separation was carried out on an Agilent SB-C18 column (100 mm × 3.0 mm i.d., 1.8 µm; Agilent Technologies, Santa Clara, CA, USA) maintained at 25 °C. The gradient solvent system consisted of solvent A (acetonitrile) and solvent B (0.5% acetic acid in water, v/v). The UPLC conditions were as follows: 4% B for 1 min, followed by a gradient to 8% B in 4 min, then a linear gradient progressing from 8% to 60% B in 10 min. The early eluent (before 1.2 min) was sent to the waste to remove the additives. The flow rate was kept constant at 0.2 mL/min, and 5 μL of sample was injected.

Mass spectrometric detection was carried out on a Xevo TQ mass spectrometer (Waters Corp.) with an electrospray ionization (ESI) source. The ESI source was operated in positive ionization mode. Quantification analysis was performed in multiple reaction monitoring (MRM) mode. The MRM mode following the ion transitions with appropriate instrumental parameters is described in Table 1. Source voltage was 3 kV, source temperature 150 °C, desolvation temperature 350 °C, desolvation gas flow 800 L/h, and the collision gas argon was kept at a pressure of 1.4 × 10−3 mbar.

Table 1. Ion transitions for the product and internal standard (IS), and the instrumental parameters for their MS/MS detection in MRM mode
PeptideIon charge statesIon transitionCone voltage (V)Collision energy (eV)
  • *

    Used for quantification; the others were used for peptide identification.

Product[M + 2H]2+610.3 > 83.9*3870
[M + 3H]3+610.3 > 129.03842
407.2 > 83.9*2236
407.2 > 129.02228
IS[M + 2H]2+603.3 > 83.9*3462
603.3 > 129.03438
[M + 3H]3+402.6 > 83.9*2446
402.6 > 129.02428

Assay validation

The intra-day precision and accuracy were investigated by determining QC samples at three different concentrations (five replicates for each concentration level). The inter-day precision and accuracy were evaluated by analyzing five replicates for each concentration level on three separate days. The concentrations were calculated using calibration curves obtained daily. The accuracy of the method at each QC concentration was described as relative error (RE) and the precision was expressed as the relative standard deviation (RSD). The suitability of the precision and accuracy was assessed by the following criteria: the RSD should not exceed 15% and the accuracy should be within 15% of the actual values for QC samples.

The linear relationship of the method was evaluated by preparing eight different concentrations of samples in enzymatic reaction buffer solution. Replicate samples of each concentration were prepared and analyzed. The calibration curves were established by plotting peak area ratios of the product to the internal standard, versus the respective true standard concentration.

The limit of quantification (LOQ) is defined as the lowest concentration giving a signal-to-noise ratio of at least 10-fold.

Application to analysis of ginsenosides

The validated UPLC/TQMS method was applied to quantify the inhibitory activity of ginsenosides. To perform the inhibition assay, we also estimated the effect of the content of methanol on the cdk5/p25 activity prior to evaluation of inhibitors. The results showed that less than 1% methanol in the reaction mixture did not affect the enzyme activity under the assay conditions used. The enzyme reaction mixture was pre-incubated in water bath at 30 °C for 3 min, and then ATP was added to the tube to initiate the reaction. The final reaction conditions were in 50 μL containing 50 mM Tris, at pH 7.4, 0.5 mM DTT, 5 mM MgCl2, 5 μM peptide (PKTPKKAKKL), 5 μM ATP, 10 μM tested compound, and 20 nM cdk5/p25. The concentration of substrate was chosen to be close to the Km values (5 μM ATP, 5 μM peptide). Enzyme reactions were incubated at 30 °C for 30 min. After incubation, the reactions were stopped by the addition of 50 μL acetonitrile containing 1 μM IS (PKpTPKKAKKV). The samples were centrifuged at 10000 g for 10 min. An amount of 5 μL of the supernatant was injected.


Optimization of MS method

The activities of the enzyme were characterized by monitoring the formation of its phosphorylated peptide (PKpTPKKAKKL). Preliminary experiments indicated that the phosphorylated peptide was more facilitated to ionization than the substrate peptide. Therefore, the phosphorylated peptide is used to evaluate the kinase activities. Through chromatographic separation, the product and IS could be identified and quantified (Fig. 1). Because peptides are easily ionized to multiply charged ions, the double-charged and triple-charged species were observed predominantly in this study. The results show that the prominent peaks are at m/z 610.3 and m/z 407.2, which correspond to the +2 and +3 charge states of the product, respectively. The prominent peaks were m/z 603.3 ([M + 2H]2+) and m/z 402.6 ([M + 3H]3+) for the IS. The Intellistart function was used to optimize the cone voltages and the collision energies to find the most specific and sensitive detection parameters for each peptide in MRM mode. Every peptide solution (2 μM) was directly injected into the mass spectrometer with the syringe, at the rate of 20 μL/min. The instrumental parameters are listed in Table 1.

Figure 1.

UPLC/MS chromatograms of product (PKpTPKKAKKL) and internal standard (PKpTPKKAKKV). (A) Total ion chromatogram (TIC) showing the separation of the two peptides. (B) Extracted chromatogram of product (MRM mode: 610.3 > 83.9). (C) Extracted chromatogram of internal standard (MRM mode: 603.3 > 83.9).

UPLC/TQMS method development

In this study, different mobile phase compositions were employed to develop a suitable and robust UPLC/TQMS method for the separation of IS and product. When water and methanol were used as the mobile phase, the IS and product could not be separated very well. However, the best separation of these compounds in the samples was achieved when water and acetonitrile were employed as the mobile phase.

The sensitivity and accuracy of the mass signal can be greatly compromised by the presence of kinase reaction additives. In order to determine the compounds by ESI-MS without performing sample pretreatment, an online desalting system was adopted in this study. A switching valve on the Waters UPLC/TQMS system was successfully combined with efficient LC separation to completely remove non-volatile additives, such as Tris-HCl, DTT, and MgCl2, from the kinase reaction mixtures prior to analysis by MS.[30] The early eluent (before 1.2 min, Fig. 1) was switched to the waste to remove the additives.

Precision and accuracy

The intra-day and inter-day precisions ranged from 2.56% to 3.92% and 3.48% to 8.99%, respectively, and the accuracy ranged from −2.44% to 2.0% and −6.47% to −1.17%, respectively, which successfully met the requirements in the US Food and Drug Adminstration Guidance.[19]

Linearity and LOQ

The calibration curves showed a good linear behavior over the range 0.04 μM to 10 μM (y = 0.934x + 0.045, R2 = 0.995). The LOQ was 0.04 μM for the product. The LOQ is appropriate for quantitative detection of analytes in the enzymatic studies.

Influence of temperature on the enzymatic reaction

The effect of temperature on the enzymatic reaction was investigated by varying the temperature in the range of 20–40 °C. A temperature of 30 °C gave the highest response for the product.

Effect of pH value on the enzymatic reaction

The effect of pH value on the enzymatic reaction was investigated by varying the pH in the range of 6.5 to 8.0. The optimum pH value of 7.5 was chosen for the reaction.

Optimization of kinase concentration and reaction time

To ensure the utilization of initial rate velocities for the calculation of enzyme constants according to steady-state kinetics, the optimum reaction time and kinase concentration should be within the linear region of the formation of the product.[33] Two reaction progress curves were generated by monitoring the product as a function of time and as a function of enzyme concentration. The reaction time of 30 min (Fig. 2) and optimum kinase concentration of 20 nM (Fig. 3) were selected.

Figure 2.

Progress curve of product formation for cdk5/p25 as a function of reaction time with the inset showing the linear range (10 to 60 min) when 20 nM cdk5/p25 and 10 μM peptide (PKTPKKAKKL) were used.

Figure 3.

Progress curve of product formation for cdk5/p25 as a function of cdk5/p25 concentration. Samples were incubated with the indicated amounts of cdk5/p25, 10 μM ATP and 10 μM peptide (PKTPKKAKKL) for 30 min at 30 °C under the conditions described in the Experimental section.

Determination of kinetic constants of the substrates

Kinetic studies on enzymes are important because they can provide essential information about how an enzyme will behave or respond in a given situation. In this study, a two-substrate enzymatic reaction was studied using UPLC/MS. One substrate was peptide (PKTPKKAKKL) and the other ATP. The peptide concentration range of 1–50 μM was used to determine its Km, while ATP concentration was held at 200 μM. The ATP concentration range of 1–50 μM was used to determine its Km while peptide concentration was held at 100 μM. Figures 4(a) and 5(a) present the Michaelis–Menten plots of peptide and ATP, respectively. The corresponding Lineweaver–Burk plots (Figs. 4(b) and 5(b)) showed excellent linearity. The Km and Vmax values were determined using the Michaelis–Menten method. The values of Km and Vmax for the peptide were determined to be 2.851 ± 0.116 and 128.200 ± 1.216 μM/min/mg and for ATP were 5.222 ± 0.374 and 190.800 ± 3.946 μM/min/mg. The results were consistent with the result reported by Beaudette et al.[34]

Figure 4.

(A) Michaelis–Menten and (B) Lineweaver–Burk plots for the peptide (PKTPKKAKKL). The concentration of peptide was varied from 1 to 50 μM, and the concentration of ATP was held constant at 200 μM.

Figure 5.

(A) Michaelis–Menten and (B) Lineweaver–Burk plots for ATP. The concentration of ATP was varied from 1 to 50 μM, and the concentration of peptide (PKTPKKAKKL) was held constant at 100 μM.

Validation of the screening method using known inhibitors

Before using the UPLC/TQMS method to screen inhibitors, the method was evaluated by roscovitine, which was known to be the potent and selective inhibitor of the cyclin-dependent kinase 5.[23, 34] Figure 6 shows the inhibition effect of roscovitine, and the half maximal inhibitory concentration (IC50) was calculated using the Graphpad Prism software. The IC50 was determined to be 0.25 μM, which agreed well with the previously reported value (0.2 μM[35]). The result showed that the UPLC/TQMS assay could be used to screen inhibitors of cdk5/p25.

Figure 6.

Inhibitory curve of roscovitine to cdk5/p25 illustrated by the UPLC/TQMS assay developed.

Application to analysis of ginsenosides

The validated method was applied to evaluate the inhibitory activity of ginsenosides. The basic structures of ginsenosides are similar, and most ginsenosides share a dammarane steroid nucleus with 17 carbon atoms arranged in four rings. Ginsenosides differ from one another by sugar type, number, and linkage position.[31, 36] Most ginsenosides can be divided into a protopanaxadiol (PPD) group and a protopanaxatriol (PPT) group. The PPD group has sugar moieties attached to the β-OH at C-3 and/or C-20; the PPT group has sugar moieties attached to the α-OH at C-6 and/or β-OH at C-20 (Fig. 7).[36] As summarized in Table 2, the inhibition activities of ginsenosides are related to their chemical structures. The results suggested that the PPT group ginsenosides (protopanaxatriol, Rg1, Rg2, Re, Rf, F1, and Rh1) exerted more potent activities than the PPD group ginsenosides (Rb1, Rb2, Rb3, Rc, Rd, Rg3, Rh2, F2, and protopanaxadiol). Taking these results into consideration, one can conclude that the sugar substituents or hydroxyl groups at C-6 are very important for the inhibitory activity of PPT group ginsenosides.

Figure 7.

The chemical structures of ginsenosides. Glc, β-D-glucopyranosyl; Arap, α-L-arabinopyranosyl; Araf, α-L-arabinofuranosyl; Xyl, β-D-xylopyranosyl; Rha, α-L-rhamnose.

Table 2. IC50 values, Km values of ATP and inhibition modes of ginsenosides and rescovitine on CDK5/p25 by theUPLC-TQMS assay. All of the IC50 values were calculated by the software Graphpad Prism. All assays were carried out in triplicate
GinsenosidesIC50 (μM)Km (μM)Inhibition mode
  1. km, while inhibtitor's concentration at IC50 values.

  2. ND, not detected.


The number of sugar moieties within a ginsenoside has an inverse impact on the kinase.[36, 37] The activity of the PPT group ginsenosides has been demonstrated to be in the order: Rg2 ~ Re ~ Rf < Rg1 < Rh1 < F1 < protopanaxatriol, indicating that reducing the number of sugar moieties increases the potency of ginsenosides.[35] Among these ginsenosides, protopanaxatriol shows the most potent activity on the cdk5/p25. Previous studies have demonstrated the similar results of PPT and PPD group ginsenosides.[38] However, this phenomenon was not seen in the PPD group ginsenosides in our experiments. The results suggested that the existence of sugar moieties could attenuate the activities of PPT group ginsenosides, probably because the sugar moieties increase the steric hindrance for these molecules to bind to their targets.[38]

We also used the assay to investigate the action mechanism of inhibitors. Ginsenosides Rd, F2, protopanaxadiol, Rb3, Rb2, Rg3, Rc, and Rh2 showed little activity and were not investigated further. We found that the other ginsenosides (Table 2) displayed non-competitive behavior (the Km values of ATP remained consistent), while roscovitine was a ATP-competitive inhibitor (the Km value of ATP increased).[39, 40] The compounds (ginsenosides and roscovitine) were not similar in structure, which could explain the different inhibitory modes.


A new UPLC/TQMS method has been developed. The MS-based assay offers excellent accuracy and reproducibility, and is suited for those enzymes with no chromophore substrate or product. This study presents a detailed kinetic analysis for the two substrates of cdk5/p25 (peptide and ATP). In addition, the UPLC/TQMS-based assay directly detects the native product, which is more accurate than the analysis of quenched samples. The assay is validated by the known inhibitor, roscosvitine, and the IC50 is 0.25 μM, which agreed well with a previously reported value. Aside from the kinetic analysis, the UPLC/TQMS assay can be used to screen the potential inhibitors of cdk5/p25.

Sixteen ginsenosides were identified to be able to inhibit the activity of cdk5/p25. The screening results demonstrated that the effects of PPT group ginsenosides were more potent than those of PPD group ginsenosides. Moreover, there was a possible structure–function relationship that the activity will decrease with increasing number of sugar moieties of PPT group ginsenosides. Consequently, this study opens the way to the study of the structure–function relationship of kinase reactions by the UPLC/TQMS method. This method is also an efficient way to determine the inhibition mode of cdk5/p25 inhibitors. There is therefore no doubt that the information obtained will be useful for the future study of the inhibition mechanism of ginsenosides in Alzheimer's disease.


This work was supported by the National Science and Technology Ministry (ID:2011BAI03B01), Jilin Province Science and Technology Department (YYZX201131), and China Postdoctoral Science Foundation funded project (2012 M511355).