Abstract
Interest in the human and experimental animal metabolism of substrates containing an odd number of carbons capable of fueling the tricarboxylic acid cycle such as heptanoic acid has motivated us to develop and validate a selective and specific liquid chromatography tandem mass spectrometric method for the simultaneous, quantitative determination of the ketone body byproducts 3-hydroxypentanoic acid and 3-oxopentanoic acid in plasma. Human plasma samples were protein-precipitated with methanol containing 0.2% formic acid. Chromatographic resolution was achieved on a Phenomenex Luna C18 column using gradient elution with mobile phases of water containing 0.1% formic acid and methanol containing 0.1% formic acid at 0.3 mL/min flow rate. The retention times of 3-hydroxypentanoic acid, 3-oxopentanoic acid and sulbactam (internal standard) were 3.85, 4.23, and 5.11 min, respectively. Validation was conducted in accordance with United States Food and Drug Administration guidance. The validated range of 3-hydroxypentanoic acid was 0.078–5 μg/mL and 0.156–10 μg/mL for 3-oxopentanoic acid. The method was accurate and precise over this range and exhibited 10-fold dilution integrity in human plasma. Recovery >88% was achieved for analytes and internal standard. There was no matrix effect observed in human plasma. Both 3-hydroxypentanoic acid and 3-oxopentanoic acid were stable across conditions including autosampler, benchtop and freeze-thaw, as well as demonstrated long-term stability at –80 °C. The method was applied to the measurement of 3-hydroxypentanoic acid and 3-oxopentanoic acid concentrations in plasma from subjects receiving the triglyceride triheptanoin (as a source of heptanoate) for the experimental treatment of glucose transporter type I deficiency (G1D) syndrome.
Keywords: BHP, BKP, 3-hydroxypentanoic acid, 3-oxopentanoic acid, β-hydroxy valeric acid, plasma, LC-MS method
1. Introduction
Interest in the metabolic fate of fuel molecules containing an odd number of carbons such as seven-carbon heptanoic acid stems from their anaplerotic potential, since they can refill natural tricarboxylic acid (TCA) cycle precursors lost in the course of metabolism. After consumption in triglyceride form (usually as a conjugate of three heptanoates per glycerol or triheptanoin), heptanoic acid gives rise to 5-carbon ketone bodies following hepatic metabolism [1]. These ketone bodies readily enter the brain and are avidly metabolized into TCA cycle precursors and their derivative neurotransmitters glutamate and GABA, thus providing a means for the therapeutic modification of disorders or states characterized by reduced anaplerosis [2] or by net brain carbon deficit such as glucose transporter type I deficiency syndrome (G1D) [3–4]. The structures of these 5-carbon ketones 3-oxopentanoic acid and 3-hydroxypentanoic acid are presented in Figure 1. Of note, these compounds are referred to in several ways including 3-oxopentanoate, β-ketopentanoate and 3-oxovaleric acid for 3-oxopentanoic acid and 3-hydroxypentanoate, β-hydroxy pentanoate and 3-hydroxyvaleric acid for 3-hydroxypentanoic acid.
Fig. 1.
Structural representation of (a) 3-Hydroxypentanoic acid, (b) 3-Oxopentanoic acid and (c) Sulbactam [internal standard].
To the best of our knowledge, there are no reported methods for the quantitative determination of 3-hydroxypentanoic acid and 3-oxopentanoic acid in human plasma using LC-MS/MS. There are two published LC-MS/MS methods for the quantitative estimation of similar short- or medium-chain fatty acids [5–6]; however, neither method included either 3-hydroxypentanoic acid or 3-oxopentanoic acid. Alternative analytical procedures have been developed for the analysis of 3-hydroxypentanoic acid and 3-oxopentanoic acid, but these methods typically include laborious derivatization procedures and are generally not sufficiently sensitive for clinical investigation. Specifically, Leclerc and coworkers studied the metabolism of R-β-hydroxypentanoate and β-ketopentanoate in dogs with quantitative analysis by GC-MS following derivatization [7]. Kinman and colleagues used derivatization with sodium borodeuteride followed by GC-MS for quantitation of 3-hydroxypentanoic acid and 3-oxopentanoic acid in rat blood [4]. Given the paucity of LC-MS/MS based methods and the complex and laborious sample preparation procedures described in the literature, it would be desirable to develop a robust, sensitive bioanalytical method that employed a simpler sample preparation procedure.
The objective of this study was to develop and validate such a direct LC-MS/MS method for the simultaneous analysis of 3-hydroxypentanoic acid and 3-oxopentanoic acid from human plasma samples. The developed method was then used to quantitate 3-hydroxypentanoic acid and 3-oxopentanoic acid in plasma samples from our human clinical study of G1D subjects at the University of Texas Southwestern Medical Center (ClinicalTrials.gov Identifiers: NCT03041363 and NCT03301532).
2. Experimental
2.1. Chemicals and reagents
Analytical standards, 3-hydroxypentanoic acid (purity 95%; Fig. 1a), 3-oxopentanoic acid (purity 97%; Fig. 1b) were purchased from Toronto Research Chemicals (Toronto, ON, Canada) and sulbactam (Internal Standard (IS); purity ≥ 99.7%; Fig. 1c) was purchased from Sigma-Aldrich (Saint Louis, MO, USA). Methanol (Optima LC-MS grade), water (Optima LC-MS grade), and formic acid (Optima LC-MS grade) were purchased from Fisher Chemicals (Hampton, NH, USA). Human plasma with potassium ethylenediaminetetraacetic acid (K2EDTA) as the anticoagulant was procured from BioIVT (Westbury, NY, USA). All chemicals and reagents used without further purification.
2.2. Chromatography
A Shimadzu Nexera X2 series liquid chromatography system (Shimadzu Corporation, Nakagyo-ku, Kyoto, Japan) consisting of a control module (CBM-20A) equipped with degasser (DGU-20A), two binary pumps (LC-30AD), an auto sampler (SIL-30AC) and column oven (CTO-30A) was employed. A Phenomenex Luna 3μm C18 (2) 100°A, 150 × 2 mm (Phenomenex, Redwood City, CA, USA) column was used for chromatographic separation. Gradient-based elution was employed using 2 mobile phases. Mobile phase A (MP-A) consisted of water with 0.1% formic acid and mobile phase B (MP-B) consisted of methanol with 0.1% formic acid. The chromatographic flow rate was 0.3 mL/min, and the gradient elution program started with 10% of MP-B, which maintained for 1 min, then MP-B was ramped to 70% at 5 min, and additionally increased to 90% MP-B by 5.2 min where it was maintained at 90% MP-B for the next two minutes. The mobile phase composition was then reverted to 10% MP-B at 7.5 min where it was maintained until the end of the chromatographic run at 10 min. The final 2.5 minutes of time at 10% MP-B was used to equilibrate the column. The retention times of 3-hydroxypentanoic acid, 3-oxopentanoic acid and sulbactam were 4.23, 3.85 and 5.11 min, respectively. A 5μL sample injection volume was used.
2.3. Mass spectrometry
The mass spectrometric system consisted of a SCIEX QTRAP® 6500+ mass spectrometer (Redwood City, CA, USA) equipped with a Turboionspray™ ionization source. The source was operated with negative ion electrospray ionization. The following mass spectrometer parameters employed: curtain gas = 45 psi; nebulizer gas = 50 psi; auxiliary gas = 50 psi; CAD gas = high; and Ion spray voltage = −4500 V. The compound dependent parameters employed for 3-hydroxypentanoic acid, 3-oxopentanoic acid and sulbactam, respectively, were: declustering potential (DP) = −68, −60, −30 V; collision energy = −15, −15, −18 V; and collision cell exit potential (CXP) = −8, −8, −10 V. An entrance potential (EP) = −10 V was used for all compounds. Detection of the analytes was carried out in the multiple-reaction monitoring mode (MRM) by monitoring the transition pairs of m/z 117 (precursor) to m/z 59 (product) for 3-hydroxypentanoic acid (Fig. 2a) and m/z 115.1 (precursor) to m/z 71 (product) for 3-oxopentanoic acid (Fig. 2b). The sulbactam MRM transition included a precursor ion of m/z 232 and a product ion of m/z 140.2. The data obtained was processed using Analyst software™ (version 1.7.1).
Fig. 2.
Mass Spectra of (a) 3-Hydroxypentanoic acid and (b) 3-Oxopentanoic acid
2.4. Standard solutions of analyte and internal standard
Separate preparations of primary stock solutions of 3-hydroxypentanoic acid and 3-oxopentanoic acid were used for preparation of the calibration curve (CC) and quality control samples (QC). Primary stock solutions were prepared by weighing the analyte of interest and diluting it in the appropriate solvent. Primary stock solutions of 3-hydroxypentanoic acid were prepared at a 1 mg/mL concentration in dimethyl sulfoxide (DMSO), and primary stock solutions of 3-oxopentanoic acid were prepared at 1 mg/mL in methanol and stored at –20 °C. Working stock solution dilutions were prepared by diluting the primary stock solutions using methanol: water (1:1, v/v). Working stock solutions were prepared at 50 μg/mL of 3-hydroxypentanoic acid and 100 μg/mL of 3-oxopentanoic acid and stored at –20 °C. Working stock solutions were prepared fresh every 30 days.
The primary stock solution of sulbactam (IS) was prepared by weighing sulbactam and diluting in water to 1 mg/mL. The IS working solution (10 μg/mL concentration) was prepared by mixing 10 μL of the primary stock solution with 990 μL of methanol: water (1:1, v/v) and stored at –20 °C. Working IS stock solutions were prepared fresh every 30 days.
2.5. Calibration and quality control samples
Calibration curves in human plasma were validated over the range of 0.078 to 5 μg/mL for 3-hydroxypentanoic acid and 0.156 to 10 μg/mL for 3-oxopentanoic acid. Calibration curve standards were prepared in plasma by adding 180 μL of working stock solutions (100 μg/mL of 3-oxopentanoic acid and 50 μg/mL of 3-hydroxypentanoic acid) into 1620 μL of plasma to yield solutions of 5 μg/mL of 3-hydroxypentanoic acid and 10 μg/mL of 3-oxopentanoic acid which served as the upper limit of quantification (ULOQ) standard. This standard was serially diluted using appropriate volumes of blank plasma to yield standards with concentrations of 0.078, 0.156, 0.312, 0.625, 1.25, 2.5 and 5 μg/mL for 3-hydroxypentanoic acid and 0.156, 0.312, 0.625, 1.25, 2.5, 5 and 10 μg/mL for 3-oxopentanoic acid.
Quality Control standards were prepared at concentrations of 0.078, 0.225, 2.0, and 4.0 μg/mL and 0.156, 0.450, 4.0, and 8.0 μg/mL for 3-hydroxypentanoic acid and 3-oxopentanoic acid, respectively. QC standards were prepared by adding 144 μL of the working stock solutions (100 μg/mL of 3-oxopentanoic acid and 50 μg/mL of 3-hydroxypentanoic acid) into 1656 μL of plasma to yield the high QC standard (HQC). The medium QC standard (MQC), low QC standard (LQC), and lower limit of quantitation (LLOQ) standards were prepared by serial dilution of the HQC standard with blank plasma.
2.6. Sample preparation
Samples were prepared using protein precipitation with methanol containing 0.2% of formic acid. Additionally, IS was added to the protein precipitation reagent. A 0.2 μg/mL solution of IS in protein precipitation reagent was prepared by adding 240 μL of 10 μg/mL IS solution to 12 mL of methanol containing 0.2% of formic acid.
A 50 μL sample (e.g., CC standard, QC standard, or unknown sample) was mixed with 100 μL of protein precipitation reagent. The resultant samples were vortexed for 30 seconds and centrifuged for 5 min at 14300×g and 15 °C. A 100 μL aliquot of clear supernatant was placed into an autosampler vial with the autosampler maintained at 15 °C and 5 μL aliquot of the sample was injected onto the column.
2.7. Method validation
This method validated in human plasma as per the Food and Drug Administration’s (FDA) Guidance for Industry entitled “Bioanalytical Method Validation” [9].
2.7.1. Specificity and selectivity
Specificity of the method was determined by analyzing six different batches (i.e., 6 different sources) of human plasma to demonstrate the lack of chromatographic interference at the retention time of 3-hydroxypentanoic acid, 3-oxopentanoic acid and IS. The acceptance criterion for specificity/selectivity was that the blanks from six lots of triheptanoin-untreated human plasma should not exhibit a 3-hydroxypentanoic acid and 3-oxopentanoic acid response greater than 20% of the LLOQ response in that same matrix.
2.7.2. Recovery
The recovery of 3-hydroxypentanoic acid and 3-oxopentanoic acid following protein precipitation was evaluated by comparing pre- and post-protein precipitation standards. A solution was prepared containing 40 μg/mL of 3-hydroxypentanoic acid and 80 μg/mL 3-oxopentanoic acid in methanol:water (1:1, v/v). Further dilutions were made with same solvent to yield 20 and 2.25 μg/mL of 3-hydroxypentanoic acid and 40 and 4.5 μg/mL for 3-oxopentanoic acid. These stocks used to prepare two sets of samples: (1) standard sample set to be processed as described in Section 2.6; and (2) a post-protein precipitation set where protein precipitation solvent was added to blank plasma, the suspension was vortexed for 30 seconds and subsequently the analytes were added. The recovery of 3-hydroxypentanoic acid and 3-oxopentanoic acid and IS were determined by comparing the response of test samples (n=3) with the mean response of post-protein precipitation prepared plasma samples. Recoveries were determined at 0.225, 2, 4 for 3-hydroxypentanoic acid and 0.450, 4 and 8 μg/mL concentrations for 3-oxopentanoic acid. Recovery of IS was determined at 0.2 μg/mL.
2.7.3. Matrix effect
The matrix effect on 3-hydroxypentanoic acid and 3-oxopentanoic acid was determined at three concentrations of 0.225, 2, 4 for 3-hydroxypentanoic acid and 0.450, 4 and 8 μg/mL concentrations for 3-oxopentanoic acid. To evaluate matrix effect a total of 12 blank plasma samples were utilized. To these samples protein precipitation solvent was added, the samples were vortexed for 30 seconds and subsequently 3-hydroxypentanoic acid and 3-oxopentanoic acid were added (3 replicates at each concentration level) to yield the desired concentrations. The matrix effect determined by comparison of the mean peak area of these samples where analyte was added following protein precipitation to neat standards of analyte (i.e., no matrix). The matrix effect on the IS was also evaluated in the same manner at a concentration of 0.2 μg/mL.
2.7.4. Calibration curve and linearity
Calibration curves were prepared at the concentrations of 0.078, 0.156, 0.312, 0.625, 1.25, 2.5 and 5 μg/mL for 3-hydroxypentanoic acid and 0.156, 0.312, 0.625, 1.25, 2.5, 5 and 10 μg/mL for 3-oxopentanoic acid. Calibration curves were acquired by plotting the peak area ratio of analyte/IS (relative response) against the nominal concentration of calibration standards. A 1/X2 weighting factor was employed. The acceptance criteria for each back-calculated standard concentration was less than ±15% deviation from the nominal value except at LLOQ, which was set at ±20%. Additionally, at least 75% of the calibration standards were expected to meet the above criteria.
2.7.5. Precision and accuracy
The accuracy and precision of the method was determined using 3 batches of QC standards prepared at 4 concentrations and 6 replicates per concentration. The concentrations of QC standards were 0.078 (LLOQ), 0.225 (LQC), 2 (MQC) and 4 μg/mL (HQC) for 3-hydroxypentanoic acid and 0.156 (LLOQ), 0.450 (LQC), 4 (MQC) and 8 μg/mL (HQC) for 3-oxopentanoic acid. Both intra- and inter-day precision and accuracy were assessed over 3 different days.
The acceptance criteria for intra-day (n=6) and inter-day (n=18) accuracy included that the back-calculated concentrations of the QC standards at each concentration should be within ±15% (percent relative error) of their nominal value except LLOQ QC concentration, where it should be within ±20%. The acceptance criteria for intra-day (n=6) and inter-day (n=18) precision included that the percent relative standard deviations of the QC standards at each concentration should be within <15% except LLOQ QC concentration, where it should be within <20%.
2.7.6. Stability experiments
All stability experiments were performed in plasma. The stability tests were conducted using 6 replicates at 0.225 μg/mL (LQC) and 4 μg/mL (HQC) for 3-hydroxypentanoic acid and 0.450 μg/mL (LQC) and 8 μg/mL (HQC) for 3-oxopentanoic acid. Stability samples processed as described under sample preparation. Samples were considered stable if assay values were within the acceptable limits of accuracy (i.e., ±15% relative error) and precision (i.e., <15% relative standard deviation) against a freshly prepared calibration curve.
Autosampler stability was determined by analysis of processed samples after storage in the autosampler for a period of 30 h. These samples were analysed against a freshly prepared calibration curve. Bench-top stability of 3-hydroxypentanoic acid and 3-oxopentanoic acid in plasma was assessed by evaluation of 6 replicates of LQC and HQC standards which had been maintained 6 hours under laboratory ambient conditions and were subsequently extracted and analyzed against a freshly prepared calibration curve. The freeze-thaw stability of 3-hydroxypentanoic acid and 3-oxopentanoic acid in plasma was assessed following three repeated freeze-thaw cycles (stored at −80 ± 10°C between cycles), subsequent sample preparation and comparison against a freshly prepared calibration curve. Long-term stability storage (−80 ± 10 °C) stability of 3-hydroxypentanoic acid and 3-oxopentanoic acid in plasma was determined by analysing QC standards after storage for 27 days.
2.7.7. Dilution integrity
To test the ability of diluted samples to yield accurate results (dilution integrity) 3-hydroxypentanoic acid and 3-oxopentanoic acid were added to blank human plasma at 5 and 10 μg/mL, respectively. Subsequently this QC standard was further 10-fold diluted using blank plasma matrix. Six (6) replicates of the diluted samples were processed, analysed, and compared versus a fresh calibration curve. The acceptance criteria for dilution integrity was that the concentrations (after incorporation of the dilution factor) were within ±15% of the nominal concentrations and had a percent CV of not more than 15% within the replicates.
2.8. Human study
Six (6) subjects with G1D were enrolled in a clinical study of triheptanoin (three in NCT03041363 and three in NCT03301532) approved by the Institutional Review Board of UT Southwestern Medical Center (Dallas, TX, USA). Eight to twelve (8–12) hours prior to triheptanoin administration, the subjects had displayed no analytical deviations from normal values for cholesterol, triglycerides, or free fatty acids. Following a 6-hour fasting period, 3 of the subjects (NCT03041363) ingested two triheptanoin doses separated by ~4 hours equivalent to 10% of their habitual daily caloric consumption per dose (using 9 Kcal per gram as the caloric value of triheptanoin). The remainder 3 subjects (NCT03301532) also followed a 6-hour fasting period and then ingested four triheptanoin doses separated by ~3 hours equivalent to 10% of their habitual daily caloric consumption per dose. For all subjects, 5 mL of venous blood were obtained at the times indicated in Figure 6, centrifuged to separate the plasma. and stored at −80 °C until analysis.
Fig. 6.
3-oxopentanoic acid and 3-hydroxypentanoic acid concentrations in human subjects following oral triheptanoin consumption.
3. Results and Discussion
There have been no reported LC-MS/MS methods validated in human plasma for the simultaneous quantitative estimation of 3-hydroxypentanoic acid and 3-oxopentanoic acid. These molecules are ketone bodies which ionize well by negative electrospray ionization and a tandem mass spectrometer allows for selective and quantitative concentration determinations. Internal standard selection was difficult due to the lack of availability of stable isotope labelled standards. Initially 3-hydroxyhexanoic acid was assessed as a potential internal standard; however, significant and variable endogenous levels precluded its use. Many additional compounds were screened such as fumaric acid, succinic acid, citric acid, lactic acid, glucose, 2-hydroxy glutaric acid, tazobactam, and ceftriaxone. Nevertheless, all of these compounds imposed limitations on retention time, peak shape or recovery efficiency. Through additional screening we ultimately found that sulbactam produced appropriate recovery efficiency and chromatography for this method.
The approach to sample preparation also required optimization. Initially, liquid-liquid extraction was assessed in various solvents such as ethyl acetate, dichloromethane, tert-butyl methyl ether, hexane and toluene. However, none of them yielded >50% recoveries. Subsequently, protein precipitation was investigated using methanol, acetonitrile as well as combinations of these solvents and formic acid. Methanol with 0.2% formic acid proved optimal with respect to recovery efficiency, peak shape and baseline noise.
The selected method was then subjected to the validation parameters set forth by the U.S. FDA for bioanalytical method validation. Specific results of the method validation are presented in Section 3.1 to Section 3.8. The method met all validation criteria and was sufficiently accurate, precise, selective and sensitive and robust for bioanalytical analysis. This method was subsequently used to analyze human clinical samples from a time course of repeated triheptanoin administration which is presented in Section 3.9.
3.1. Method development
Initially the mass spectrometric parameters were optimized for 3-hydroxypentanoic acid and 3-oxopentanoic were optimized using neat solutions prepared at 0.5 μg/mL in methanol:water (50:50, v/v) containing 0.1% formic acid. Both analytes demonstrated a robust signal in negative electrospray ionization mode. The optimal multiple reaction monitoring (MRM) transitions were also determined.
The composition of the chromatographic mobile phase was assessed including evaluation of various buffers and additives including formic acid, ammonium acetate, ammonium bicarbonate and ammonium formate in water with acetonitrile and/or methanol in varying proportions. Additionally, numerous column chemistries were evaluated such as C18, F5, PS-C18 columns.
Retention time and chromatographic peak shape were favorable for 3-hydroxypentanoic acid, 3-oxopentanoic acid and the IS on a Phenomenex Luna 3μm, C18(2), 100Å, 150×2 mm column. From a mobile phase perspective, gradient elution using the binary pair of 0.1% formic acid in water and 0.1% formic acid in methanol achieved optimal peak configuration, retention time and stable response for analytes. The optimal flow rate for peak configuration was 0.3mL/min. The resultant retention times for 3-hydroxypentanoic acid, 3-oxopentanoic acid and IS were 4.23, 3.85, and 5.11 min, respectively with a total run time of 10 min. The 10-minute run time allowed for re-equilibration of column between runs. To prevent unnecessary ions from entering the mass spectrometer the flow was diverted to waste collection from 0–1 min and from 8–10 min.
3.2. Specificity and selectivity
To demonstrate specificity and selectivity of 3-hydroxypentanoic acid a number of representative chromatograms are presented including a double blank (Fig 3a), LLOQ (Fig 3b), LQC standard (Fig 3c), and a human clinical study sample (Subject A, sample 9). To demonstrate specificity and selectivity of 3-oxopentanoic acid a number of representative chromatograms are presented including (Fig 4a), LLOQ (Fig 4b), LQC standard (Fig 4c), and a human clinical study sample (Subject A, sample 9). A representative MRM chromatogram for IS 0.2 μg/mL in human plasma presented Fig 5.
Fig. 3.
Representative MRM chromatograms of 3-Hydroxypentanoic acid (a) human double blank plasma (i.e., blank plasma prepared without analyte or IS) (b) human plasma spiked at LLOQ (0.078 μg/mL) (c) LQC sample at 0.225 μg/mL in plasma (d) human clinical study sample Subject A, sample 9
Fig. 4.
Representative MRM chromatograms of 3-Oxopentanoic acid (a) human double blank plasma (i.e., blank plasma prepared without analyte or IS) (b) human plasma spiked at LLOQ (0.156 μg/mL) (c) LQC sample at 0.450 μg/mL in plasma (d) human clinical study sample Subject A, sample 9
Fig. 5.
Typical MRM chromatogram for sulbactam (internal standard) at 0.2 μg/mL in plasma.
It is noteworthy that a peak was observed in 3-oxopentanoic acid MRM channel which is adjacent to the primary peak; however, it did not interfere with the analyte peak integration. Additionally, there was a small interfering peak from endogenous matrix components observed at the retention time of 3-hydroxypentanoic acid. The signal-to-noise of the peak was less than 20% of the 3-hydroxypentanoic acid response at LLOQ; therefore, it does not interfere with quantitation. There were no significant interfering peaks observed for the IS.
3.3. Recovery
The determined recoveries, mean (range), in human plasma for 3-hydroxypentanoic acid at 0.225, 2, 4 μg/mL and 3-oxopentanoic acid at 0.450, 4, 8 μg/mL were 88.2% (86.3–89.4%), 93.7% (89.0–96.5%), 94.0% (92.1–97.4%) and 98.0% (95.8–99.5%), 108.6% (103.2–114.6%), 109.5% (107.1–113.3%), respectively. The recovery of the IS in human plasma was 99.4% (98.8–99.6 %).
3.4. Matrix effect
The average matrix effect accuracy, mean (range), in human plasma at 3 concentrations in triplicate (n=9) was determined to be 87.8% (85.4–96.1%) for 3-hydroxypentanoic acid and 90.1% (86.3–95.7%) for 3-oxopentanoic acid. The average matrix effect accuracy, mean (range), for the IS was 98.6% (97.7–99.2%).
3.5. Calibration curve or linearity
A weighting factor of 1/X2 provided the best fit calibration line. All calibration curves for both analytes which were prepared for the validation had a correlation coefficient (‘r’ value) ≥0.998. The lowest quantifiable concentration with a %RE of ±20% and %RSD <20% was the LLOQ. This was 0.078 μg/mL for 3-hydroxypentanoic acid and 0.156 μg/mL for 3-oxopentanoic acid in human plasma. The accuracy (%RE) of human plasma CC standards (7 non-zero concentration in each across 4 calibration curves) for 3-hydroxypentanoic acid and 3-oxopentanoic acid varied between −6.2 to 6.6%. The precision (%RSD) across all 4 calibration curves ranged between 0.9 to 4.8%.
3.6. Precision and accuracy
The determined intra-day accuracy (%RE) for 3-hydroxypentanoic acid (n=6 on 3 days) varied between −12.7 to 5.9% and the precision (%RSD) ranged between 1.6 to 6.6%. Inter-day accuracy of 3-hydroxypentanoic acid (n=18) varied between −0.6 to 0.4% and the precision ranged between 4.0 to 6.2% (Table 1). The determined intra-day accuracy for 3-oxopentanoic acid (n=6 on 3 days) varied between −7.1 to 12.0 % and the precision ranged between 1.8 to 7.2%. Inter-day accuracy of 3-oxopentanoic acid (n=18) varied between −1.8 to 4.7% and the precision ranged between 4.3 to 8.6% (Table 2).
Table 1.
Intraday and interday precision and accuracy of 3-hydroxypentanoic acid in human plasma (intraday (n=6), interday (n=18))
| Quality Control Standards, Nominal Concentrations (μg/mL) | ||||||
|---|---|---|---|---|---|---|
|
|
||||||
| Parameter | Run | n | 0.078 (LLOQ) | 0.225 (LQC) | 2.00 (MQC) | 4.00 (HQC) |
|
| ||||||
| Mean (μg/mL) | 1 | 6 | 0.08 | 0.22 | 1.9 | 3.5 |
| Accuracy (%RE)a | 1 | 6 | −3.8 | −2.8 | −6.2 | −13 |
| Precision (%RSD)b | 1 | 6 | 6.6 | 5.3 | 2.7 | 2.6 |
| Mean (μg/mL) | 2 | 6 | 0.08 | 0.23 | 2.0 | 3.8 |
| Accuracy (%RE)a | 2 | 6 | −3.2 | 1.4 | 0.5 | −4.1 |
| Precision (%RSD)b | 2 | 6 | 6.6 | 2.2 | 3.1 | 4.3 |
| Mean (μg/mL) | 3 | 6 | 0.08 | 0.23 | 2.12 | 4.0 |
| Accuracy (%RE)a | 3 | 6 | −1.9 | 2.7 | 5.9 | −1.0 |
| Precision (%RSD)b | 3 | 6 | 5.3 | 1.6 | 2.7 | 2.0 |
| Mean (μg/mL) | 1,2,3 | 18 | 0.08 | 0.23 | 2.0 | 3.8 |
| Accuracy (%RE)a | 1,2,3 | 18 | −3.0 | 0.4 | 0.1 | −6.0 |
| Precision (%RSD)b | 1,2,3 | 18 | 5.9 | 4.0 | 5.7 | 6.2 |
= Percent Relative Error
= Percent Relative Standard Deviation
Table 2.
Intraday and Interday precision and accuracy of 3-oxopentanoic acid in human plasma (intraday (n=6), interday (n=18))
| Quality Control Standards, Nominal Concentrations (μg/mL) | ||||||
|---|---|---|---|---|---|---|
|
|
||||||
| Parameter | Run | n | 0.156 (LLOQ) | 0.450 (LQC) | 4.00 (MQC) | 8.00 (HQC) |
|
| ||||||
| Mean (μg/mL) | 1 | 6 | 0.17 | 0.45 | 3.9 | 7.8 |
| Accuracy (%RE)a | 1 | 6 | 6.2 | −0.6 | −2.0 | −2.9 |
| Precision (%RSD)b | 1 | 6 | 7.2 | 3.8 | 3.2 | 4.0 |
| Mean (μg/mL) | 2 | 6 | 0.15 | 0.43 | 4.1 | 8.4 |
| Accuracy (%RE)a | 2 | 6 | −7.1 | −5.5 | 2.6 | 5.0 |
| Precision (%RSD)b | 2 | 6 | 7.2 | 3.4 | 4.0 | 4.6 |
| Mean (μg/mL) | 3 | 6 | 0.15 | 0.45 | 4.4 | 9.0 |
| Accuracy (%RE)a | 3 | 6 | −1.2 | 0.8 | 9.7 | 12 |
| Precision (%RSD)b | 3 | 6 | 6.5 | 3.0 | 2.3 | 1.8 |
| Mean (μg/mL) | 1,2,3 | 18 | 0.15 | 0.44 | 4.1 | 8.4 |
| Accuracy (%RE)a | 1,2,3 | 18 | −0.7 | −1.8 | 3.4 | 4.7 |
| Precision (%RSD)b | 1,2,3 | 18 | 8.6 | 4.3 | 5.7 | 6.9 |
= Percent Relative Error
= Percent Relative Standard Deviation
3.7. Stability studies
The stability of 3-hydroxypentanoic acid and 3-oxopentanoic acid in human plasma met criteria for processed sample (autosampler) for 30 hours (15 °C); benchtop stability for 6 hours (ambient conditions); and freeze-thaw stability for 3 repeated freeze-thaw cycles (stored at −80 ± 10°C between cycles). Long-term freezer (−80 ± 10 °C) stability was verified following 27 days for human plasma (Table 3).
Table 3.
Stability studies of 3-hydroxypentanoic acid and 3-oxopentanoic acid in human plasma.
| 3-Hydroxypentanoic Acid Quality Control Nominal Concentrations (μg/mL) | 3-Oxopentanoic Acid Quality Control Nominal Concentrations (μg/mL) | |||||
|---|---|---|---|---|---|---|
|
|
||||||
| Stability Assessment (Conditions) | Parameter | n | 0.225 (LQC) | 4.00 (HQC) | 0.450 (LQC) | 8.00 (HQC) |
|
| ||||||
| Autosampler (30 hours) | Mean (μg/mL) | 6 | 0.25 | 3.90 | 0.47 | 8.3 |
| Accuracy (%RE)a | 6 | 9.9 | −2.5 | 4.2 | 3.4 | |
| Precision (%RSD)b | 6 | 2.1 | 4.1 | 5.7 | 3.9 | |
| Benchtop (6 hours) | Mean (μg/mL) | 6 | 0.23 | 3.8 | 0.45 | 8.4 |
| Accuracy (%RE)a | 6 | 2.7 | −5.0 | −0.4 | 4.5 | |
| Precision (%RSD)b | 6 | 2.2 | 4.2 | 2.5 | 4.8 | |
| Freeze Thaw (3 cycles) | Mean (μg/mL) | 6 | 0.22 | 3.8 | 0.40 | 8.5 |
| Accuracy (%RE)a | 6 | −1.3 | −4.8 | −11 | 6.8 | |
| Precision (%RSD)b | 6 | 1.7 | 2.9 | 2.1 | 2.7 | |
| Long Term (27 days at −80 °C) | Mean (μg/mL) | 6 | 0.24 | 3.6 | 0.42 | 8.7 |
| Accuracy (%RE)a | 6 | 7.8 | −8.9 | −6.6 | 9.3 | |
| Precision (%RSD)b | 6 | 1.7 | 2.4 | 8.7 | 6.8 | |
= Percent Relative Error
= Percent Relative Standard Deviation
3.8. Dilution integrity
A 10-fold dilution integrity was established in plasma for 3-hydroxypentanoic acid and 3-oxopentanoic acid. Specifically, dilution integrity was confirmed with all diluted samples meeting the %RE criterion of ±15% and a %RSD of not more than a 15% deviation.
3.9. Human study results
Pre- and post-triheptanoin concentrations of 3-hydroxypentanoic acid and 3-oxopentanoic acid in human subjects who consumed this food supplement for the first time is depicted in Fig. 6. The plasma concentrations for the subjects gradually increased with time after administration and reached as high as 15.1 μg/mL for 3-hydroxypentanoic acid and 39.3 μg/mL for 3-oxopentanoic acid. These values are consistent with metabolism of triheptanoin as reflected in rat studies [10]. For example, the highest concentration of total C5-ketone bodies achieved during intraduodenal infusion at 40% of the caloric requirement was ~5.5 μg/mL.
4. Conclusion
A method for simultaneous determination of 3-hydroxypentanoic acid and 3-ketopentanoic acid in human plasma has been developed and validated. This is the first reported liquid chromatography tandem mass spectrometry method. The method is adequately sensitive with lower limits of quantitation of 0.078 for 3-hydroxypentanoic acid and 0.156 μg/mL for 3-ketopentanoic acid. Through validation, the method proved to be accurate, precise, selective and sensitive for absolute determination of 3-hydroxypentanoic acid and 3-ketopentanoic acid in human plasma as employed to analyze samples from a human clinical study of subjects with G1D.
HIGHLIGHTS.
This study provides insight into animal metabolism of substrates containing an odd number of carbons capable of fueling the tricarboxylic acid cycle such as heptanoic acid.
The bioanalytical method developed is the first LC-MS/MS method to simultaneously analyze 3-hydroxypentanoic acid and 3-oxopentanoic acid in human plasma.
The method has a simple and efficient sample preparation procedure employing protein precipitation, and a 10 min run time which makes it employable for high throughput analyses.
The method was validated per US FDA regulatory guidelines for analysis of 3-hydroxypentanoic acid and 3-oxopentanoic acid in human plasma.
The method was demonstrated to be highly sensitive, specific, precise, accurate and no significant matrix effect was observed over the ranges of 0.078–5 μg/mL for 3hydroxypentanoic acid was and 0.156–10 μg/mL for 3-oxopentanoic acid.
Acknowledgements
The generous support of the Glut1 Deficiency Foundation is gratefully acknowledged, as is that of NIH grants NS094257 and NS102588. None of the acknowledged persons or institutions participated in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript; or decision to submit the manuscript for publication.
Footnotes
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- [1].Marin-Valencia I, Good LB, Ma Q, Malloy CR, Pascual JM, Heptanoate as a neural fuel: energetic and neurotransmitter precursors in normal and glucose transporter I-deficient (G1D) brain. J Cereb Blood Flow Metab. 33 (2013) 175–182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [2].Marin-Valencia I, Roe CR, Pascual JM, Pyruvate carboxylase deficiency: mechanisms, mimics and anaplerosis. Mol Genet Metab. 101 (2010) 9–17. [DOI] [PubMed] [Google Scholar]
- [3].Pascual JM, Ronen GM, Glucose Transporter Type I Deficiency (G1D) at 25 (1990–2015): Presumptions, Facts, and the Lives of Persons With This Rare Disease. Pediatr Neurol. 53 (2015) 379–393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [4].Pascual JM, Liu P, Mao D, I Kelly D, Hernandez A, Sheng M, Good LB, Ma Q, Marin-Valencia I, Zhang X, Park JY, Hynan LS, Stavinoha P, Roe CR, Lu H, Triheptanoin for glucose transporter type I deficiency (G1D): modulation of human ictogenesis, cerebral metabolic rate, and cognitive indices by a food supplement. JAMA Neurol. 71 (2014) 1255–1265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [5].Zhang Q, Ford LA, Goodman KD, Freed TA, Hauser DM, Conner JK, Vroom KE, Toal DR, LC-MS/MS method for quantitation of seven biomarkers in human plasma for the assessment of insulin resistance and impaired glucose tolerance. J Chromatogr B Analyt Technol Biomed Life Sci. 1038 (2016) 101–108. [DOI] [PubMed] [Google Scholar]
- [6].Jaochico A, Sangaraju D, Shahidi-Latham SK. A rapid derivatization based LC-MS/MS method for quantitation of short chain fatty acids in human plasma and urine. Bioanalysis. 11 (2019) 741–753. [DOI] [PubMed] [Google Scholar]
- [7].Leclerc J, Des Rosiers C, Montgomery JA, Brunet J, Ste-Marie L, Reider MW, Fernandez CA, Powers L, David F, Brunengraber H, Metabolism of R-beta-hydroxypentanoate and of beta-ketopentanoate in conscious dogs. Am J Physiol. 268 (1995) 446–452. [DOI] [PubMed] [Google Scholar]
- [8].Kinman RP, Kasumov T, Jobbins KA, Thomas KR, Adams JE, Brunengraber LN, Kutz G, Brewer WU, Roe CR, Brunengraber H, Parenteral and enteral metabolism of anaplerotic triheptanoin in normal rats. Am J Physiol Endocrinol Metab. 291 (2006) 860–866. [DOI] [PubMed] [Google Scholar]
- [9].United States Food and Drug Administration, Guidance for Industry, Bioanalytical Method Validation. (2018) 1–41. [Google Scholar]
- [10].Gu L, Zhang GF, Kombu RS, Allen F, Kutz G, Brewer WU, Roe CR, Brunengraber H, Parenteral and enteral metabolism of anaplerotic triheptanoin in normal rats. II. Effects on lipolysis, glucose production, and liver acyl-CoA profile. Am J Physiol Endocrinol Metab. 298 (2010) 362–371. [DOI] [PMC free article] [PubMed] [Google Scholar]