Technical Note pubs.acs.org/ac
Dried Saliva Spot (DSS) as a Convenient and Reliable Sampling for Bioanalysis: An Application for the Diagnosis of Diabetes Mellitus Masahiro Numako,† Takahiro Takayama,† Ichiro Noge,‡ Yutaka Kitagawa,§ Kenichiro Todoroki,† Hajime Mizuno,† Jun Zhe Min,† and Toshimasa Toyo’oka*,† †
Laboratory of Analytical and Bio-Analytical Chemistry, School of Pharmaceutical Sciences, University of Shizuoka, 52-1 Yada, Suruga-ku, Shizuoka 422-8526, Japan ‡ Department of Pharmaceuticals, Numazu City Hospital, Harunoki, Higashi-Shiizi, Numazu 410-0302, Japan § Department of Clinical and Molecular Endocrinology, Numazu City Hospital, Harunoki, Higashi-Shiizi, Numazu 401-0302, Japan S Supporting Information *
ABSTRACT: This paper proposes the dried saliva spot (DSS) as a convenient sampling technique for bioanalysis. The analytical method with the DSS was used for the determination of D,L-lactic acid (D,L-LA) and the D/L ratio of diabetic patients and prediabetic persons for the simple screening of the disease. The D,L-LA in the DSS was labeled with a chiral reagent (DMT-3(S)-Apy) for carboxylic acids and determined by UPLCESI-MS/MS. The limits of detection (signal-to-noise ratio (S/N) = 3) for the DSS analysis were on the amol level (∼30 amol). Because good stability, recovery, accuracy, and precision of the D,L-LA for the DSS method was also obtained from the proposed procedure, the DSS method was applied to the determination of the D- and L-isomers of LA of diabetic patients, and prediabetic and healthy persons. The D/L-LA ratio by the present DSS method and the HbA1c value in blood were well-correlated to the serious diabetic patients, whereas the relation in the prediabetic persons was not very good. The reason seems to be due to the rough saliva sampling, and not to the DSS method, because strict regulation was not requested for the prediabetic and healthy persons. In order to have a successful DSS analysis, the stability of the target molecule, the detection sensitivity to the target molecule, and the validated determination method are important.
D
during the diabetic stage.11−15 Based on this background, we have developed a new methodology using LC-MS for the simultaneous determination of L- and D-LA. During the course of our study, a highly sensitive and specific determination method using a chiral labeling reagent has been reported in previous papers.16−19 The methods using UPLC-ESI-MS/MS were applied to the simultaneous determination of L- and D-LA in the saliva of diabetic patients and healthy volunteers. The results show that the ratio of D-LA to L-LA in diabetic saliva was significantly higher than that in the saliva of the healthy persons. The method using DMT-3(S)-Apy seems to be an excellent one in terms of sensitivity, specificity, and separation efficiency of the resulting diastereomers. In this study, the saliva poured into a tube without a sample collection device was used. Although the sampling is very easy, the transport of the sample tubes to the analytical facility may be complicated for the general public. Therefore, another sampling technique is required to accelerate the further use of the method. Consequently, a dried saliva spot (DSS) using filter paper
iabetes mellitus is one of the interesting diseases, because of its increasing worldwide prevalence, not only in the European−American, but also Asian countries.1,2 Diabetes in humans is a multifactorial disorder based on environmental factors and genetic background. In many cases, diabetes is asymptomatic for a long period and the patient is not aware of the disease.3,4 Type II diabetes is generally dependent on lifestyle, such as exercise, sleep, and meals. The crisis of diabetes is avoidable by early detection before serious symptoms arise. Therefore, the diagnosis during the predisease state is important to avoid the crisis and/or complications with other diseases.5 Lactic acid (LA), which is one of the biologically active carboxylic acids, ubiquitously occurs in living organisms. The LLA, which is an abundant enantiomer in mammals, is produced from pyruvic acid under anaerobic conditions. Its optical isomer, D-LA, also exists in living systems, even though its amount is relatively low, compared to L-LA. The endogenous DLA is produced from methyglyoxal mediated by glyoxalases I and II in the metabolic pathway of glucose.6,7 The D-LA concentration is significantly increased in the serum of diabetic animals and patients.8−10 This has been considered to be due to the accelerated production of its precursor, i.e., methyglyoxal, which causes the patient to develop diabetic complications © 2015 American Chemical Society
Received: October 23, 2015 Accepted: December 2, 2015 Published: December 2, 2015 635
DOI: 10.1021/acs.analchem.5b04059 Anal. Chem. 2016, 88, 635−639
Technical Note
Analytical Chemistry was investigated in this study. As a similar technique using filter paper, the dried blood spot (DBS) is well-known and applied to the diagnosis of various diseases.20−27 The diagnostic analysis of the inborn error of metabolism, such as phenylketonuria and tyrosinuria, is classically and extensively used.20−27 Over the past decade, DBS has become increasingly popular not only for the determination of diseases, but also therapeutic drug monitoring and pharmacokinetic studies.20,21,28 For instance, the analytical methods for several drugs (e.g., tacrolimus and cyclosporine A) by DBS have been successfully developed.29,30 The main advantage of the DBS method is that the subject can perform the sampling at home and/or can send it directly to the analytical facility. This similar technique, which is patientfriendly, seems to be also adaptable to DSS. As the DSS method, Abdel-Rehim et al.31 determined lidocaine (a local anesthetic drug) by liquid chromatography−mass spectroscopy (LC-MS). However, the determination of endogenous biological component by the DSS sampling method is not mentioned in this paper. Based on the observation, we investigated the DSS sampling method for the determination of biological components. The proposed method using DSS biosampling is applied to the determination of both enantiomers of LA and the D/L ratios in the saliva of diabetic patients, prediabetic persons, and healthy volunteers.
Derivatization of LA with DMT-3(S)-Apy and Determination by UPLC-MS/MS. The freshly prepared solutions of 20 μL EDC in methanol (20 mM), 20 μL HOAt in methanol (20 mM) and 10 μL DMT-3(S)-Apy (30 mM) in acetonitrile/ methanol (1:1) containing 1% TEA were added to 50 μL of the acetonitrile extraction solution of the saliva. The solution was vigorously mixed and reacted at 60 °C for 90 min. After the reaction, 300 μL of water was added to the tube. The solution was then centrifuged at 3000 rpm for 5 min, and an aliquot (5 μL) was subjected to UPLC-ESI-MS/MS. The selected reaction monitoring (SRM) chromatogram (precursor ion m/ z 298.2 → product ions m/z 209.3 and 226.3) was used for the determination of the resulting diastereomers. DSS Validation. Calibration Curve. The fixed concentrations of both enantiomers of D,L-LA and D,L-LA-d3 (5.0− 1000 μM) were prepared by sequential dilutions of the stock solutions. Three microliters (3 μL) each of the working solutions (15.0−3000 pmol) was spotted on the 5 mm ϕ filter paper punched from the DMPK-C card. The filter paper was dried by cool air from a hairdryer and placed into a tube. Three microliters (3 μL) of each working solution of D,L-LA was also spotted on the 5 mm ⌀ filter paper containing 3 μL of saliva from healthy persons. The filter paper was similarly dried and placed in another tube. A 50 μL aliquot of acetonitrile was added to the tube. The solution was reacted and determined using the procedures previously described in the section entitled “Derivatization of LA with DMT-3(S)-Apy and Determination by UPLC-MS/MS”. The calibration curves were developed by plotting the peak area of the diastereomers (y) versus the concentration of LAs (x, ng/mL), using linear regression 1/x weighting (n = 3). The limit of detection (LOD) (signal-to-noise ratio (S/N) of 3) on the SRM chromatogram was calculated from the peak intensity of the low concentration of D,L-LA-d3 versus the baseline noise. The limit of quantification (LOQ) was also calculated from the signal-to-noise ratio of 10 (S/N = 10). Intraday and Interday Stability. The precisions of the intraday and interday assays were determined by the repeated measurement (n = 5) of the DSS (3 μL saliva in 5 mm ϕ filter paper), obtained from healthy volunteers, on 1 day and over 14 consecutive days, respectively. The pretreatment, derivatization and UPLC-ESI-MS/MS determination were the same as those done for the “calibration curve”. The precision was evaluated as the relative standard deviation (RSD, %). The D,L-LA amount, the D/L ratio and their RSDs (%) were calculated from the intensities of the corresponding diastereomers. The intraday and interday precisions of the D,L-LA-d3 spiked in the 5 mm ϕ filter paper containing 3 μL saliva were also determined. Recovery. The amounts of D,L-LA (30−1200 pmol) in 3 μL of acetonitrile were spiked on the 5 mm ϕ filter paper containing 3 μL saliva, obtained from the healthy volunteers (n = 5). The 5 mm ⌀ filter paper containing 3 μL saliva without spiking the D,L-LA was also tested as the standard sample. The DSS with and without spiking the D,L-LA were then pretreated, derivatized, and subjected to UPLC-ESI-MS/MS, according to the procedure used for the “calibration curve”. The amounts of the D,L-LA in the filter were determined from each calibration curve. The recovery percentage was defined as F/(F0 + A) × 100 (%), where F is the amount of LA in the spiked sample, F0 the amount of the LA in the unspiked sample, and A the spiked amount. The amounts of D,L-LA-d3 (3, 30, 300, 3000 pmol) were separately spiked on the 5 mm ⌀ filter paper containing 3 μL
■
EXPERIMENTAL SECTION Materials and Chemicals. 1-Ethyl-3-(3-(dimethylamino)propyl)-carbodiimide hydrochloride (EDC) and 1-hydroxy-7azabenzotriazole (HOAt) were purchased from Tokyo Kasei (Tokyo, Japan). The D,L-lactic acids (D,L-LA) (Sigma−Aldrich; St. Louis, MO, USA) and the sodium salt of D,L-lactic acid3,3,3-d3 (D,L-LA-d3) (C/D/N Isotopes; Quebec, Canada) were used. DMT-3(S)-Apy was synthesized in our laboratory according to a previous paper.19 The FTA DMPK-C card for DBS (GE Healthcare Co., U.K.) was used as the filter paper for the DSS sampling. Acetonitrile (CH3CN), methanol (MeOH), and formic acid (HCOOH) (LC-MS grade) were purchased from Kanto Chemicals. Deionized and distilled water (H2O) was used throughout the study; this water was obtained using an Aquarius PWU-200 automatic water distillation apparatus (Advantec, Tokyo, Japan). All other reagents and solvents were of analytical grade. A 10 μmol/mL solution of LAs (D-LA, L-LA, D,L-LA, and D,LLA-d3) in water was prepared as the stock solution. The working solutions were prepared by sequential dilutions with acetonitrile. UPLC-ESI-MS/MS Analysis. The UPLC-ESI-MS/MS analysis was performed using an ACQUITY ultraperformance liquid chromatograph (UPLC I-class, Waters) that was connected to a Xevo TQ-S triple quadrupole-mass spectrometer (Waters, Milford, MA). An ACQUITY UPLC BEH C18 column (1.7 μm, 100 mm × 2.1 mm i.d.; Waters) was used at the flow rate of 0.4 mL/min and 40 °C. The diastereomers of the LAs labeled with DMT-3(S)-Apy were analyzed by UPLCESI-MS/MS in the positive-ion mode. The separation and detection conditions were as follows: mobile phase, acetonitrile−water (7:93) containing 0.1% (v/v) HCOOH; capillary voltage, 3.00 kV; cone voltage, 50 V; desolvation gas flow, 1000 L/h; cone gas flow, 150 L/h; nebuliser gas flow, 7.0 L/h; collision gas (Ar) flow, 0.15 mL/min; collision energy, 20 eV; collision cell exit potential, 5 V; and desolvation temperature, 500 °C. Analytical software (MassLynx, version 4.1) was used for the system control and data processing. 636
DOI: 10.1021/acs.analchem.5b04059 Anal. Chem. 2016, 88, 635−639
Technical Note
Analytical Chemistry
performed before the real sample analysis. In the validation test, the exact saliva volume was used for the calculation of various parameters, such as RSD (%) and recovery (%), although the volume on the filter paper is essentially different for each real sample. A good linearity of the calibration curves was obtained from the D,L-LA and D,L-LA-d3 spiked in the 5 mm ϕ filter paper (see Figure S1 in the Supporting Information). The slopes of the curves were almost the same for all the enantiomers. Thus, the reactivity of DMT-3(S)-Apy to LA seems to be almost equivalent for both enantiomers. However, the y-axis value of the L-LA curve spiked in saliva was relatively higher than that of the other curves. This is due to the L-LA ubiquitously existing in the saliva. The limit of detection (LOD) and the limit of quantification (LOQ) of the LA spiked on the filter paper were calculated as 30 amol (S/N = 3) and 100 amol (S/N = 10), respectively, based on the SRM chromatogram. The highly sensitive detection obtained from the UPLC-MS/MS analysis seems to be good enough for the determination of the trace volume of the saliva spot. Next, the recovery, stability, accuracy, and precision of the DSS analysis were tested. The RSD (%) of the interday and intraday determinations of the D,L-LA and D,L-LA-d3 by the present DSS analysis were 3.92%). Based on these results, both enantiomers of LA in the DSS at room temperature seem to be stable for at least 14 days. The percentage recovery of LA in the DSS was tested next. From a preliminary experiment, relatively large amounts of LLA and trace amounts of D-LA were present in the saliva of healthy persons. Therefore, the percentage recovery of the different amounts of D,L-LA spiked in the 5 mm ⌀ filter paper was determined. A good recovery (99.5%−104.6%) and precision (RSD of >6.71%) were obtained from the tested concentrations (see Table S2 in the Supporting Information). However, good recovery and precision at low LA concentrations in the DSS could not be guaranteed. Therefore, the recovery using the isotopic variants was also tested over a wide concentration range (3−3000 pmol/spot). A similar recovery (97.5%−103.8%) was observed from the trace amounts of D,LLA-d3 spiked in the DSS (see Table S2). Therefore, a quantitative recovery of D,L-LA from the DSS seems to be possible. The matrix effect by the DSS was finally tested. The matrix factor (MF) was determined using the isotopic variants (D,L-
■
RESULTS AND DISCUSSION The DSS biosampling seems to be convenient for general subjects who cannot go to the hospital. In order to allow the extensive use of the DSS as a sampling method, the exact results using a trace sampling volume are essentially required, the same as the conventional and straightforward saliva analysis. Therefore, the validation tests affecting the DSS method were 637
DOI: 10.1021/acs.analchem.5b04059 Anal. Chem. 2016, 88, 635−639
Technical Note
Analytical Chemistry LA-d3) spiked in the DSS, because the saliva ubiquitously contains large amounts of L-LA and small amounts of D-LA, as previously described, and thus an incorrect result might be obtained from the addition of the D,L-LA. Therefore, the MF value at different concentrations of D,L-LA-d3 in the DSS analysis was determined. Based on the MF value (102.2%− 106.2%), the influence of the saliva matrix including the filter paper was negligible (Table S3 in the Supporting Information). These validation data suggest that the determination using the DSS sampling is applicable for the diagnosis of diabetes. The DSS analysis seems to provide an alternative option for the early screening of diseases, such as diabetes, because of the noninvasive and sampling ease. The DSS use is excellent, in terms of cost-effectiveness, straightforwardness, storage, and shipping ease. However, a trace volume analysis of
