YUE JialiCHENG WeiWEI ShutongLIU GuilinZHOU MeichenLV Zhihuaand YU Mingming*

Development and Validation of UHPLC-MS/MS Method for Quantifying of Agarotriose:An Application for Pharmacokinetic,Tissue Distribution,and Excretion Studies in Rats

YUE Jiali1), CHENG Wei1), WEI Shutong1), LIU Guilin1), ZHOU Meichen1), LV Zhihua1),2),3),4), and YU Mingming1),2),3),4), *

A sensitive, rapid, and robust ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/ MS) method was established for the first time to quantify agarotriose (A3) in rat plasma, tissues, urine, and feces. A3 and stachyose (internal standard) were separated by a BEH amide column at 65℃ under the mobile phase of 10mmolL?1ammonium acetate-acetonitrile (42:58,/) with 350μLmin?1. The acquisition of transitions was carried out in multiple reaction monitoring (MRM) pattern operating with positive ionization at/509.16>329.15 for A3 and/689.15>527.11 for stachyose. The linearity ranges of A3 were 10 to 5000nmolL?1for plasma, 20 to 10000nmolL?1for tissues, and 40 to 20000nmolL?1for urine and feces. The accuracy and precision ranged from 90.9% to 111.6% and 0.7% to 10.1%, respectively. The stability was between 86.1% and 102.5%. The extraction recovery was consistent and reproducible. The matrix effect ranged from 1.5% to 11.4%. The pharmacokinetic, tissue distribution, and excretion studies were successfully conducted with the validated method. Results showed that A3 could be absorbed by rats, and the absolute bioavailability was 6.7%. Furthermore, it was rapidly distributed in rat tissues and mainly eliminatedfeces excretion (67.0%) after oral administration. For intravenous bolus, 85.5% was recovered, and renal excretion was the primary pathway (77.6%) for cumulative recovery.

agarotriose; UHPLC-MS/MS; pharmacokinetic; tissue distribution; excretion

1 Introduction

Agarose, a neutral linear phyto-polysaccharide, is derived from marine red algae such as Pterocladiaceae, Gelidiellaceae, Gracilariaceae, and Gelidiaceae(Lee., 2017) by degradation (Kazlowski., 2008; Kim., 2013; Kazlowski., 2015; Gao., 2019), consisting of D-galactose and 3,6-anhydro-L-galactose units alternately linked by β-1,4- and α-1,3-glycosidic bonds (Marinho-Soriano and Bourret, 2005). Agarotriose (A3) is an oligosaccharide monomer with low degrees of poly- merization obtained by separation and purification from agarose extracts (Fig.1). Studies have shown that A3 has a good pharmacological activity in inhibited intestinal in- flammation by inducing heme oxygenase-1 expression (Higashimura., 2013) and prebiotic effects (Yun., 2021a). Generally, the physiological effects of drugs depend on their ability to be absorbed by the organism and accumulated at the target site. Therefore, a detailed understanding of pharmacokinetic, tissue distribution, and excretion is critical for illustrating biological disposal(Zhang., 2017). However, in spite of the pharmacological activities of A3 have been revealed, the disposal processis still unclear.

Compared to published methods of A3, such as high- performance liquid chromatography (HPLC) (Kim., 2013), gas chromatography-mass spectrometry (GC-MS) (Lee., 2014), and liquid chromatography hybrid ion trap time-of-flight mass spectrometry (LC/MS-IT-TOF) (Kim., 2017; Yun., 2021b), ultra-high performance liquid chromatography-tandem mass spectrometry (UHPLC-MS/MS) is a natural and universal choice for quantifying of oligosaccharides in biological matrix owing to the advanced chromatographic separation, superior sensitivity, improved ionization and ion activation (Volpi and Linhardt, 2010; Kim., 2013; Kailemia., 2014). In the drug development process, a sensitive and rapid bioanalytical method is indispensable. However, at present, there is no quantitative analysis method for A3 in the biological matrix such as plasma, tissues, urine, and feces. Hence, in order to investigate the pharmacokinetic, tissue distribution, and excretion of A3 in rats, it is necessary to develop and validate a robust UHPLC-MS/MS method.

Fig.1 Chemical structure of agarotriose.

In the present study, a sensitive, rapid, and robust UHP LC-MS/MS method was established to quantify of A3 in rat plasma, tissues, urine, and feces and successfully applied to the disposition for the first time. Furthermore, the analysis time is only 3.5min, while effectively avoiding the phenomenon of anomeric oligosaccharides.

2 Materials and Methods

2.1 Chemicals

TheA3 (purity ≥98%) was provided by BZ Oligo Biotech (Qingdao, China). The stachyose (purity≥98%), used as internal standard (IS), heparin sodium (185 USP units mg?1), and LC-MS grade of ammonium acetate were accessed from Aladdin (Shanghai, China). The LC- MS grade of acetonitrile and water were obtained from Merck (Darmstadt, Germany). The physiological saline solution was obtained from Huaren Pharmaceutical (Qingdao, China). The Oasis PRiME HLB 96-well μElution Plates with sorbent of 3mgwell?1were supplied by Waters Corporation (Massachusetts, USA). The metabolic cages were purchased from Tecniplast (Milan, Italy).

2.2 Animals

The male pathogen-free Sprague-Dawley (SD) rats, weighting 200±20g, were provided by Pengyue Experimental Animal Breeding Center (Jinan, China) (SCXK- 2019003). Animals were kept in controlled environment with 22±2℃, 60%±5% relative humidity, and 12h light/ dark cycles for 7 d to acclimatize. During the period, water and food were unconstrained. Animals were subjected to the experiment after fasting overnight. After the investigation, animals were sacrificed by over-anesthetized with chloral hydrate. The procedure was reviewed and approved by the Laboratory Animal Ethics Committee of School of Medicine and Pharmacy Ocean University of China (#3 2019).

2.3 UHPLC-MS/MS Conditions

The chromatographic separation of A3 and IS was exe- cuted on Dionex UltiMate 3000 system using a Waters BEH amide column (2.1mm×150mm, 1.7μm) protect- ed by a guard cartridge (2.1mm×5mm, 1.7μm) at 65℃. The temperatures of autosampler and column oven were controlled at 4℃ and 65℃, respectively. The mobile phase was 10mmolL?1ammonium acetate-acetonitrile (42:58,/) with 350μLmin?1. The data were collected for 3.5min. The injection volume was 2μL. The acetonitrile- water (10:90) was introduced as the needle wash solution.

The mass detection was performed on the Thermo TSQ QUANTIVE tandem mass spectrometer equipped with a heated-electrospray ionization (H-ESI) source. The acquisition of transitions was done in multiple reaction monitoring (MRM) pattern operating in positive ionization with [M+Na]+. The typical mass spectra and fragmentation of A3 were present in Fig.2. The ion spray voltage was 5500 V. The temperatures of ion transfer tube and vaporizer were maintained at 350℃ and 275℃, respectively. The sheath gas, auxiliary gas, and collision in- duced dissociation gas were set at 35Arb, 10Arb, and 2 mTor, respectively. The dwell time was 100 ms. The ion transitions at509.16 > 329.15 for the A3 and at/689.15 > 527.11 for IS were selected, using collision energy of 26.97 and 33.61V, respectively. The radio frequency Lens for A3 and IS were 181 and 228V, respectively.

Fig.2 Precursor and product mass spectra of agarotriose.

2.4 Sample Treatment

The biological samples were removed from ?40℃ refrigerator on ice to thaw. The tissues, urine, and feces were homogenized with acetonitrile-water (50:50,/) solution at a ratio of 1:5 (/), 1:10 (/), and 1:20 (/), respectively.The cleanup of plasma samples was conducted by a single-step protein precipitation. Specifically, 100μL plasma was mixed with 300μL acetonitrile containing 1μmolL?1IS, vortexed for 60s and centrifuged at 18000for 20min at 4℃. Subsequently, 300μL supernatant was transferred in a centrifuge tube and evaporated under a vacuum concentrator at 3000rmin?1for 40℃ at 4h. The residue was reconstituted by 50μL mobile phase for the UHPLC-MS/MS analysis after centrifuged twice at 18000for 20min at 4℃. For tissues, urine, and feces, 300μL acetonitrile containing 1μmolL?1IS was added into 100μL homogenate followed by vortex-mixed for 60s and centrifuged twice at 18000for 20min at 4℃. After that, 300μL of the supernatant was performed into Oasis PRiME HLB 96-well μElution Plates for solid phase extraction (SPE). Subsequently, the filtrate was subjected to the quantitative analysis.

2.5 Preparation of Calibration Concentrators and Quality Control Samples

The stock solutions of A3 (3mmolL?1) and IS (2mmolL?1) were prepared by acetonitrile-water (50:50,/). The working solutions of A3 (300μmolL?1) were diluted by the blank biomatrix to construct concentrator levels and quality control samples (QCs). For plasma, the concentrator levels were 5000, 2000, 1000, 500, 100, 50, 20, and 10nmolL?1. For tissue homogenates, the concentrator levels were 10000, 5000, 1000, 500, 100, 40, and 20nmolL?1. For urine and feces homogenates, the concentrator levels were 20000, 15000, 5000, 1000, 200, 80, and 40nmolL?1. Correspondingly, the QCs were prepared at concentrations of 4000, 200, and 25nmolL?1for plasma, 8000, 400, and 50nmolL?1for tissues, 16000, 2000, and 100nmolL?1for urine and feces as high quality control (HQC), middle quality control (MQC), and low quality control (LQC), respectively. The lower limit of quantitation (LLOQ) of A3 in plasma, tissues, urine, and feces were 10, 20, and 40nmolL?1, respectively. For concentrators and QCs, the pre-treatment was the same as the samples.

2.6 Bioanalytical Method Validation

The proposed method was validated in accordance with the U.S. Food and Drug Administration of Bioanalytical Method Validation Guidance for Industry (USFDA, 2018).

2.6.1 Calibration curve

The calibration curve was assessed by concentrators using a weighted (1/2) least-squares linear regression (Sonawane., 2019). The square of correlation coefficient (2) was considered acceptable under higher than 0.99. The sensitivity was defined by LLOQ.

2.6.2 Selectivity and carryover

The selectivity was evaluated by contrasting the signal intensity in blank and LLOD to investigate potential interferences of the biological matrix with A3 and IS. The carryover was performed by analyzing a blank sample after the highest calibrator. The response in blank should not surpass 20% of LLOQ and 5% of IS.

2.6.3 Accuracy and precision

The accuracy and precision of within-run and between-runs were carried out using four QC levels (LLOQ, LQC, MQC, and HQC) with three independent analytical runs. The accuracy and precision were expressed by the ratio of measured concentration to nominal concentration (% normal) and the coefficient of variation (% CV), respectively. The accuracy should be 85% to 115% (80% to 120% for LLOQ). The precision should not surpass± 15% (±20% for LLOQ).

2.6.4 Extraction recovery and matrix effect

The extraction recovery of LQC, MQC, and HQC was assessed by contrasting the concentration of spiked into A3 before extraction with those of the spiked into A3 after extraction. The matrix effect of LQC and HQC was calculated by comparing the concentration of the spiked into A3 after extraction with those of the spiked directly in a neat solution. The recovery should be consistent and reproducible. The CV of matrix effect should not exceed ±15%.

2.6.5 Stability

The stability was carried out at room temperature, autosampler (4℃), freeze-thaw (3 times), and long-term stability (?40℃, 60d) using LQC and HQC. The criteria for stability (% normal) was 85%–115%.

2.7 Pharmacokinetic Study

Six healthy rats were selected and divided into two groups at random. The jugular venous catheterization was performed per the previous protocol (Harms and Ojeda, 1974; Thrivikraman., 2002). The pharmacokinetic study was conducted on the third day after the rat model established. The doses of A3 for a single oral administration and intravenous bolus were 50mgkg?1and 5mgkg?1, respectively. The blood (300μL) was collected and transferred to heparin-coated tubes at 0.083, 0.167, 0.25, 0.5, 1, 2, 3, 4, and 5h following oral administration, respectively. For intravenous bolus, the blood (300μL) was drawn at 0.033, 0.083, 0.25, 0.5, 0.75, 1, 1.5, 2, and 3h after the dosing, respectively. The samples were centrifuged immediately at 3500for 10min at 4℃. Subsequently, 100 μL of supernatant was transferred and maintained at ?40 ℃ until analysis. The formula for calculating the absolute bioavailability (abs) of A3 was as follows:

where,, po, and iv represent the area under the plasma concentration-time curve, dose administered, oral administration, and intravenous administration, respectively.

2.8 Tissue Distribution Study

Nine healthy rats were selected and divided into three groups randomly. Rats were administered a single oral administration (50mgkg?1) by gavage. The tissues, including heart, liver, spleen, lung, kidney, stomach, duodenum, jejunum, ileum, cecum, colon, muscle, testis, and brain, were dissected at 0.25, 0.5, and 1.5h after dosing, respectively. Tissues were washed with physiological sa- line solution for 3 times to clear up the residual blood or contents, then blotted on filter paper and weighed, and kept at ?40℃ until further analysis.

2.9 Excretion Study

Six healthy rats were randomly selected and divided into two groups. The excretion studies were carried out under a single oral administration (50mgkg?1) and intravenous bolus (5mgkg?1), respectively. The samples of urine and feces were collected at 0–4, 4–8, 8–12, 12–24, 24–36, and 36–48h post dose. The corresponding excretion volume of urine was recorded, and the feces were weighed after freeze-dried. The samples were kept at ?40℃ until analysis.

2.10 Data Analysis

The data of mass spectra were acquired and post- processed by the Xcalibur 4.1. The statistics were completed using Microsoft Excel 2019. The pharmacokinetic parameters were estimated by Phoenix WinNonlin 8.1 (Pharsight Corporation, USA) using non-compartmental analysis. The graphics drawing was done through GraphPad Prism 9.0 (GraphPad software Inc., USA). The data were expressed in arithmetic mean±standard deviation (SD).

3 Results and Discussion

3.1 Method Development

During the pre-treatment of biomatrix samples, the protein precipitation combined with SPE can effectively remove the influence of interfering substances on A3 in tissues, urine, and feces samples. Protein precipitation, however, was sufficient to purify plasma. In chromatographic separation, the detection abilities of different mobile phases (such as acetonitrile, methanol, and additives) and columns (such as amino column, amide column, and HILIC column) for A3 were investigated. It has been demonstrated that the maximum analytical performance can be obtained by using the amide column under the mobile phase of 10mmolL?1ammonium acetate-aceto- nitrile (42:58,/). Meanwhile, the column temperature of 65℃ can effectively eliminate the anomeric phenomenon in the presence of oligosaccharides. Furthermore, in the positive ionization mode, the oxygen on the ether bond can form a stable sodium adduct during the ionization process owing to the anhydro-galactose structure in A3, which can generate the highest mass response.

3.2 Method Validation

3.2.1 Calibration curve

The calibration curves of A3 in plasma, tissues, urine, and feces were=2.675e-004+1.999e-004(2=0.9962),=3.437e-003+9.696e-005(2=0.9957),=4.537e-002 +4.275e-004(2=0.9929), and=2.981e-002+9.337e- 005(2=0.9931), respectively, whereis the peak area ratio of A3 to IS andis the nominal concentration. The calibration curves displayed excellent linearity in plasma, tissues, urine, and feces over a certain range from 10 to 5000nmolL?1, 20 to 10000nmolL?1, 40 to 20000nmolL?1, and 40 to 20000nmolL?1, respectively. The LLOQ of A3 in plasma, tissues, urine, and feces were 10, 20, 40, and 40nmolL?1with signal-to-noise (/)≥10, respectively.

3.2.2 Selectivity and carryover

As displayed in Fig.3, the responses of the A3 in the blank biological samples were lower than 20% of LLOQ and 5% of IS, demonstrating that the interference of plasma, tissues, urine, and feces to A3 was within the specified range. The carryover of A3 in the blank plasma, tissues, urine, and feces following the highest concentrator were 4.2%, 7.8%, 4.9%, and 8.9% of LLOQ, respectively, and 0.6%, 0.3%, 0.5%, and 0.5% of LLOQ for IS, respectively, indicating that no significant carryover effect in this method.

3.2.3 Accuracy and precision

The accuracy and precision of the method in within- run and between-runs were all present in Table 1. The within-run accuracy of A3 in plasma, tissues, urine, and feces rangedfrom 92.1% to 111.6%, and the precision was between 0.7% and 7.6%. Correspondingly, the between-runs accuracy of A3 ranged from 90.9% to 106.4%, and the precision was between 3.7% and 10.1%.

3.2.4 Extraction recovery and matrix effect

As shown in Table 2, the extraction recovery of A3 in different biological samples was consistent and reproducible. The CV of A3 ranged from 1.5% to 11.4%, implying there was essentially no matrix effect in this method.

Fig.3 Extracted chromatograms of A3 and IS in blank biological matrix and LLOQ. A3, agarotriose; LLOQ, lower limit of quantitation; IS, internal standard; A, B, C, and D represent the plasma, tissues, urine, and feces, respectively.

3.2.5 Stability

The stability of A3 was summarized in Table 3. The accuracy of A3 in plasma, tissues, urine, and feces ranged from 86.6% to 97.0%, 86.1% to 98.4%, 88.5% to 97.2%, and 90.7% to 102.5%, respectively. The results showed that A3 was stable during sample disposals.

3.3 Application

The developed and validated UHPLC-MS/MS method for A3 was applied to the pharmacokinetic, tissue distribution, and excretion studies in rats.

Table 1 Within-run and between-runs accuracy and precision of A3 in biological matrix of rats with four QC levels

Notes: A3, agarotriose; CV, coefficient of variation; QCs, quality control samples.

Table 2 Extraction recovery and matrix effect of A3 in biological matrix of rats (n=6)

Notes: -, undetected; A3, agarotriose; CV, coefficient of variation; QCs, quality control samples; SD, standard deviation.

Table 3 Stability of A3 in biological matrix of rats with LQC and HQC under different conditions (Mean±SD, n=6)

Notes: A3, agarotriose; QCs, quality control samples; LQC, low quality control samples; HQC, high quality control samples; RT, room temperature; SD, standard deviation.

3.3.1 Pharmacokinetic study

The main pharmacokinetic parameters of A3 after oral administration and intravenous bolus were summarized in Table 4, and the mean plasma concentration-time curves were shown in Fig.4. The results showed that after oral administration of 50mgkg?1, A3 was quickly absorbed, reaching amaxof 929±177ngmL?1in plasma with amaxof 0.5±0.03h, and the elimination half-life (1/2) and mean residence time (MRT) were 0.85±0.02 and 1.4±0.26h, respectively. The1/2and MRT for intravenous bolus were 0.63±0.16 and 0.34±0.02h, respectively, indicating rapid elimination of A3 in rats. The AUC(0-t)for oral administration and intravenous bolus were 2068 ± 410 and 3066±170hngmL?1, respectively. Based on the formula ofabs, theabsof A3 was 6.7%.

Table 4 Pharmacokinetic parameters of A3 following a single oral administration (50mgkg?1) and intravenous bolus (5mgkg?1) in rats (Mean±SD, n=3)

Notes: A3, agarotriose; SD, standard deviation;max, maximum plasma concentration;max, time tomax;1/2, elimination half- life; AUC, area under the plasma concentration-time curve; MRT, mean residence time.

3.3.2 Tissue distribution study

The tissue distribution study of A3 at 0.25, 0.5, and 1.5 h after a single oral administration (50mgkg?1) to rats was present in Fig.5. After oral administration for 0.25h, A3 could be rapidly distributed in tissues, among which the stomach and small intestine were the most, indicating that A3 was absorbed from the gastrointestinal tract (GIT). The duodenum and jejunum were the main parts of A3 absorption in the small intestine. After 0.5h, the concentration of A3 was increased in the stomach, kidney, duodenum, and jejunum. The concentration of A3 decreased in tissues by 1.5h post dose. During the entire period, A3 was not detected in the brain, implying that it cannot pass the blood-brain barrier. Since A3 was most absorbed in the duodenum and jejunum, which was consistent with the duodenum and jejunum containing the abundant brush borders under normal physiological conditions. It further explained that the absorption of A3 in rats was mainly concentrated in the upper segment of GIT. Additionally, the small intestine may be the primary binding site of A3 because of its higher enrichment, which acts on intestinal inflammation (Higashimura., 2013).

3.3.3 Excretion study

The urinary and fecal cumulative excretion curves of A3 following a single oral administration (50mgkg?1) and intravenous bolus (5mgkg?1) were presented in Fig.6. After 12h post dose, the cumulative excretion of A3 in urine and feces reached a plateau. As shown in Fig.6A, the cumulative excretion of A3 accounted for 77.1% of the oral administration, whereas 67.0% was recovered from feces in 48h,indicating that A3 was mainly elimi- nated by feces excretion.After intravenous bolus in rats (Fig.6B), a total of 85.5% was recovered, and the cumulative recovery in urine was 77.6%, implying that A3 was excreted by urine primarily.It may be associated with the excellent water solubility of A3 (Paradkar, 2008).

Fig.6 Urinary and fecal cumulative excretion curves of A3 following a single oral administration (50mgkg?1) and intravenous bolus (5mgkg?1) in rats (Mean±SD, n=3). A3, agarotriose; SD, standard deviation; A, oral administration; B, intravenous bolus.

4 Conclusion

In summary, a sensitive, rapid, and robust UHPLC- MS/MS method was established and validated to quantify A3 in biological matrix of rats.In addition, the pharmacokinetic, tissue distribution, and excretion studies were successfully conducted with this method for the first time.This study may clarify the biological disposal of A3 in rats. Furthermore, it can provide a valuable reference for establishing bioanalytical methods of other oligosaccharides and pharmacokinetic studies. Since A3 can be absorbed by the GIT of rats, the intestinal transport mechanism study by Caco-2 cells monolayer requires further study.

Acknowledgements

This study was funded by the Fundamental Research Funds for the Central Universities (Nos. 201912008, 2019 64019), and the Natural Science Foundation of Shandong Province (No. ZR2019BC025).

Gao, B., Li, L., Wu, H., Zhu, D., Jin, M., Qu, W.,., 2019. A novel strategy for efficient agaro-oligosaccharide production based on the enzymatic degradation of crude agarose inWPAGA1.10: 1231.

Harms, P. G., and Ojeda, S. R., 1974. A rapid and simple procedure for chronic cannulation of the rat jugular vein., 36: 391-392.

Higashimura, Y., Naito, Y., Takagi, T., Mizushima, K., Hirai, Y., Harusato, A.,., 2013. Oligosaccharides from agar inhibit murine intestinal inflammation through the induction of heme oxygenase-1 expression., 48: 897- 909.

Kailemia, M. J., Ruhaak, L. R., Lebrilla, C. B., and Amster, I. J., 2014. Oligosaccharide analysis by mass spectrometry: A review of recent developments., 86: 196- 212.

Kazlowski, B., Pan, C. L., and Ko, Y. T., 2008. Separation and quantification of neoagaro- and agaro-oligosaccharide products generated from agarose digestion by beta-agarase and HCl in liquid chromatography systems., 343: 2443-2450.

Kazlowski, B., Pan, C. L., and Ko, Y. T., 2015. Monitoring and preparation of neoagaro- and agaro-oligosaccharide products by high performance anion exchange chromatography systems., 122: 351-358.

Kim, H. T., Yun, E. J., Wang, D., Chung, J. H., Choi, I. G., and Kim, K. H., 2013. High temperature and low acid pretreatment and agarase treatment of agarose for the production of sugar and ethanol from red seaweed biomass., 136: 582-587.

Kim, J. H., Yun, E. J., Yu, S., Kim, K. H., and Kang, N. J., 2017. Different levels of skin whitening activity among 3,6-an-hydro-l-galactose, agarooligosaccharides, and neoagarooligosaccharides., 15: 321.

Lee, C. H., Kim, H. T., Yun, E. J., Lee, A. R., Kim, S. R., Kim, J. H.,., 2014. A novel agarolytic beta-galactosidase acts on agarooligosaccharides for complete hydrolysis of agarose into monomers., 80: 5965-5973.

Lee, W. K., Lim, Y. Y., Leow, A. T., Namasivayam, P., Ong Abdullah, J., and Ho, C. L., 2017. Biosynthesis of agar in red seaweeds: A review., 164: 23-30.

Marinho-Soriano, E., and Bourret, E., 2005. Polysaccharides from the red seaweed(Gracilariales, Rhodophyta)., 96: 379-382.

Paradkar, A. and Bakliwal, B., 2008. Biopharmaceutics & pharmacokinetics. In:. 2: 1-32.

Sonawane, S. S., Chhajed, S. S., Attar, S. S., and Kshirsagar, S. J., 2019. An approach to select linear regression model in bioanalytical method validation., 10: 1.

Thrivikraman, K. V., Huot, R. L., and Plotsky, P. M., 2002. Ju- gular vein catheterization for repeated blood sampling in the unrestrained conscious rat., 10: 84- 94.

Volpi, N., and Linhardt, R. J., 2010. High-performance liquid chromatography-mass spectrometry for mapping and sequen- cing glycosaminoglycan-derived oligosaccharides., 5: 993-1004.

Yun, E. J., Yu, S., Kim, Y. A., Liu, J. J., Kang, N. J., Jin, Y. S.,., 2021a.prebiotic and anti-colon cancer activities of agar-derived sugars from red seaweeds., 19: 213.

Yun, E. J., Yu, S., Park, N. J., Cho, Y., Han, N. R., Jin, Y. S.,., 2021b. Metabolic and enzymatic elucidation of coope- rative degradation of red seaweed agarose by two human gut bacteria., 11: 13955.

Zhang, Q., Guan, J., Li, S., Zhao, Y., and Yu, Z., 2017. Application of an UHPLC-MS/MS method to tissue distribution and excretion study of 2-(2-hydroxypropanamido) benzoic acid in rats., 1070: 54-61.

(September 6, 2022;

November 30, 2022;

December 28, 2022)

? Ocean University of China, Science Press and Springer-Verlag GmbH Germany 2023

. E-mail: yumingming@ouc.edu.cn

(Edited by Ji Dechun)