Abstract

   Chronic cerebral ischemia hypoperfusion plays a role in the initiation
   and progression of vascular dementia, which causes changes in
   metabolites. Currently, there is no standard treatment to treat,
   prevent and reduce the severity of this condition. Thai herbal Yahom
   no.20 (YHF20) is indicated for fatigue and dizziness. The components of
   YHF20 have been found to have pharmacological effects related to the
   pathology of chronic cerebral ischemia hypoperfusion. This study aimed
   to investigate metabolomic changes after YHF20 administration in a rat
   model of permanent bilateral common carotid artery occlusion (2-VO)
   induced chronic cerebral ischemia hypoperfusion, and to explore its
   impact on spatial learning and memory. Albino Wistar rats were randomly
   allocated to 5 groups; sham, 2-VO, 2-VO+ 100 mg/kg YHF20,
   2-VO+300 mg/kg YHF20, and 2-VO+1000 mg/kg YHF20. The rats were
   administered YHF20 daily by oral gavage for 56 days after 2-VO
   induction. Plasma was collected weekly for metabolome change analysis
   using LC-MS/QTof and toxicity study. The rats were evaluated for
   spatial learning and memory using the Morris water maze. The results
   showed that 78 known metabolites and 10 tentative pathways altered
   after chronic cerebral hypoperfusion, although it was not able to
   determine the effect on memory and learning behaviors of rats.
   Glutathione and glutathione metabolism might be metabolite-pathway that
   were the affect after YHF20 administration in cerebral ischemic
   condition. The 4 known metabolites may be the metabolites from the
   constituents of YHF20 could be considered and confirmed for quality
   control purpose. In conclusion, YHF20 administration might contribute
   to metabolic changes related to cerebral ischemia condition without the
   effect on spatial learning and memory, including hepatotoxicity and
   nephrotoxicity after 56 days of treatment. Alterations in the potential
   metabolites may provide data support for elucidating dementia
   pathogenesis and selecting pathways for intervention.

   Keywords: YAHOM20, YHF20, Cerebral ischemia, Metabolomics, Cognitive
   impairment, Thai herbal formula

1. Introduction

   Dementia is characterized by multiple cognitive deficits, including
   memory loss, decline in thinking skills, and emotional and language
   problems. The most common types of dementia are Alzheimer's disease
   (AD) and vascular dementia (VaD), respectively. VaD is a
   pathophysiological condition resulting in a sudden and progressive
   decline in cognitive and memory function caused by inadequate blood
   flow to the brain [[37]1]. Etiology of VaD is unclear. Previous studies
   showed the chronic reduction of cerebral blood flow, which is found in
   aging and neurodegenerative disorders, was associated with cognitive
   deficit [[38]2,[39]3]. athologies such as neuronal death (mostly
   apoptosis), inflammation, and oxidative stress occur when there is the
   chronic reduction of cerebral blood flow [[40]4]. The evidence suggests
   that chronic cerebral hypoperfusion is essential in the initiation and
   progression of VaD. Many animal models have been created to simulate
   cerebral blood flow impairment conditions, which are manipulated in
   acute or slow, complete or incomplete, focal or global, and permanent
   or transient reductions. The experimental ischemic models comprise
   focal ischemic, global ischemic [[41]5,[42]6], and hypoperfusion models
   [[43]5]. This study will focus on learning and memory impairment caused
   by chronic cerebral hypoperfusion. Memory loss in dementia was
   associated with explicit memory. The encoding and storages of explicit
   information are due to the activity of the prefrontal cortex and
   hippocampus. The prefrontal cortex is associated with working memory
   that could change to short-term memory. When we need to recall the
   information, we could change short-term memory to long-term memory. The
   transformation mechanism was called the consolidation process in which
   the hippocampus is an important region that plays a critical role in
   this process. Hippocampal-dependent learning and memory assessment has
   many tests such as radial arm maze, Barnes maze, Morris water maze,
   contextual fear conditioning test, object location memory test, etc
   [[44]7]. The Morris water maze is widely used for evaluating behavioral
   and cognitive abilities in rodents because it is a well-validated test
   that assesses spatial learning and memory.

   YHF20 is a herbal formulation that contains 25 components ([45]Table
   S1) as follows Amomum testaceum Rid., Aquilaria malaccensis Lam.,
   Euphorbia antiquorum L., Syzygium aromaticum (L.) Merr. & L.M. Perry,
   Artemisia pallens Wall. Ex DC., Angelica sinensis (Oliv.) Diels.,
   Terminalia chebula Retz. Var. chebula, Ligusticum sinense Oliv. Cv.
   Chuanxiong, Mimusops elengi L., Dracaena Loureirin Gagnep., Myristica
   fragrans Houtt., Alyxia reinwardtii Blume, Glycyrrhiza glabra L.,
   Nelumbo nucifera Gaertn., Mesua ferrea L., Vetiveria zizanioides (L.)
   Nash ex Small, Urceola rosea (Hook. & Arn.) D.J.Middleton, Cinnamomum
   bejolghota (Buch.-Ham.) Sweet, Mammea siamensis Kosterm., Cinnamomum
   verum J.S. Presl., Cinnamomum loureirii Nees., Sodium borate and Borneo
   Camphor; their pictures are shown in Supplementary Material, [46]Fig
   S1. The proportions of components are equal. YHF20 is often used for
   the relief of symptoms like nausea, vomiting, dizziness, fatigue, and
   insufficient sleep. Many components of YHF20 have been reported for
   pharmacological activities related to the hemodynamic system [[47][8],
   [48][9], [49][10], [50][11], [51][12], [52][13], [53][14]],
   vasodilatation [[54]13,[55]15], antioxidation [[56][16], [57][17],
   [58][18], [59][19], [60][20], [61][21], [62][22], [63][23]], and
   anti-inflammation [[64][24], [65][25], [66][26], [67][27], [68][28],
   [69][29], [70][30], [71][31]] ([72]Table S1). Thus, the present study
   was designed to investigate metabolomics changes after YHF20
   administration in ischemic rats and to explore its impact on spatial
   learning and memory.

2. Materials and methods

2.1. Materials and reagents

   YHF20 powder was obtained from Herbal Medicines and Products
   Manufacturing Unit, manufactured under Good Manufacturing Practice
   (GMP) PIC/s by Ayurved Siriraj, Center of Applied Thai Traditional
   Medicine, Faculty of Medicine Siriraj Hospital, Mahidol University,
   Thailand. YHF20 was verified for quality control using LC-MS by
   comparing the chromatogram fingerprint of each production batch from
   manufacturing unit. The chromatogram of YHF20 is shown in Supplementary
   Material, [73]Fig S2. Alanine aminotransferase (ALT), aspartate
   aminotransferase (AST), urea/BUN-UV, and creatinine were purchased from
   Biosystems (Spain). All chemicals used were optima® LC-MS grade in
   metabolic profile analysis. For the animal study, YHF20 was suspended
   in sterile water.

2.2. Experimental animals

   Seventy-eight adult (7 weeks) male Wistar rats (weight between 200 and
   250 g) were used in all experiments. All animals were obtained from
   Nomura Siam International Co, Ltd., Pathumwan, Bangkok, Thailand. They
   were maintained in the 12-h light/dark cycle and were housed in
   individual ventilated cage with Ad libitum standard diet and sterile
   distilled water at Siriraj Laboratory Animal Research and Care Center.
   They were acclimatized for 2 weeks before the experiments.

2.3. Establishment of permanent bilateral common carotid occlusion (pBCCAO)

   The rats were anesthetized by intramuscular injection of ketamine
   (60 mg/kg) and xylazine (6.0 mg/kg). The unconscious rats were placed
   onto a controlled heating pad and then body temperature was monitored.
   The rats were incised at the midline of the neck region. Both common
   carotid arteries were dissected and carefully separated from the vagus
   nerve. Both common carotid arteries were ligated with silk suture
   before suturing the incision. The suture area was applied with
   antibiotic ointment and xylocaine 2% gel, respectively. The rats were
   subcutaneously injected with carprofen 5 mg/kg 2 times to relief pain.
   Post-surgery, rats were kept under observation and a heating lamp until
   recovery from anesthesia. The rats were returned to the cage when they
   fully regained consciousness.

2.4. Animal grouping and drug administration

   The rats were randomly allocated into 5 different groups which were the
   sham-operated group (sham), the vehicle-treated bilateral common
   carotid artery-occluded group (2-VO + vehicle), the 100 mg/kg of YHF20
   vehicle-treated bilateral common carotid artery-occluded group
   (2-VO+100 mg/kg YHF20), the 300 mg/kg of YHF20 vehicle-treated
   bilateral common carotid artery-occluded group (2-VO+300 mg/kg YHF20),
   and the 1000 mg/kg of YHF20 vehicle-treated bilateral common carotid
   artery-occluded group (2-VO+1000 mg/kg YHF20). The YHF20 doses were
   converted from clinical doses by the equivalent dose ratio between body
   surface areas of human and animals. Each group daily received sterile
   water or YHF20 by oral gavage for 8 weeks ([74]Fig. 1). Then, the first
   day of administration started within 24 h after pBCCAO.

Fig. 1.

   [75]Fig. 1
   [76]Open in a new tab

   Experimental protocol.

2.5. Blood collection and preparation

   The rats blood were collected in heparin tubes from the lateral tail
   vein using a butterfly needle (23G) at day 1, 9, 16, 23, 30, 37, 44,
   51, and 58. Blood samples were centrifuged (8000 g for 10 min at 4 °C)
   to separate plasma. All plasma samples were stored at −80 °C within 4 h
   and were used for metabolic profiles analysis and biochemical analysis
   ([77]Fig. 1).

2.6. Sample preparation for metabolic profiles analysis

   Plasma was thawed at room temperature until completely melt. It was
   centrifuged at 4000 g 10 min 4 °C. Supernatant was divided as sample,
   pooled qc, and pooled group. Then, MeOH (700 μL) and internal standard
   (25 μL) were pipetted into plasma, mixed using vortex for 10 s, and
   continued mixed using a shaker at 2000 rpm for 15 min. After that,
   samples were stored at −80 °C for 60 min. Then, samples were
   centrifuged at 18,000 g for 10 min at 4 °C. The supernatant was
   pipetted as samples (300 μL) and technical QC (TQC). The supernatant
   was concentrated by SpeedVac concentrator until samples were completely
   dry and were stored at −80 °C until analysis. Samples were
   reconstituted with Milli-Q water (100 mL) and were centrifuged 15,800 g
   at 4 °C for 15 times. In all steps, samples were placed on the cold
   rack (∼4 °C) except SpeedVac concentrating. The supernatant was used to
   determine the metabolic profiles using LC-MS-QtoF.

2.7. Metabolic profile analysis

   The chromatographic separation was performed with AQUITY UPLC HSS T3
   (1.8 μm, 2.1 × 100 mm) column with guard column coupling binary solvent
   system and auto-sampler. The column temperature was maintained at
   40 °C. The sample temperature was 4 °C. The injection volume was 5 μL.
   The mobile phase was consisted of 0.1% Formic acid in Milli-Q water (A)
   and 0.1% Formic acid in MeOH (B) at a flow rate of 0.4 mL/min. The
   condition was linear-gradient (curve 6) elution as follows: kept 100%A
   0%B at 0–1 min for equilibration (1 min), in next changed to 0%A 100%B
   at 1–16 min (15 min) for analysis and maintained to washing column at
   16–20 min (4 min), recondition to 100%A 0%B at 20–22 min (2 min), and
   maintained at 22–24 min (2 min). The total run time was 24 min per
   analysis. Mass spectrometry detection was performed on SYNAPT G2-Si
   HDMS QTOF (Waters Corp., MA, USA) equipped with electrospray ionization
   (ESI) interference. The analysis was in positive mode and negative mode
   using Mse analysis mode. Acquisition time was 24 min in the resolution
   analyzer mode. The mass range was 100–1200 Da with continuum scanning
   (0.5 s). Low collision energy for precursor analysis was set at 4 V but
   high collision energy was performed at 20–40 V for fragment ions. The
   parameters were capillary voltage of 3 kV, sampling cone voltage of
   40 V, source offset of 80 V, source temperature of 150 °C, and
   desolvation temperature of 500 °C. The cone and desolvation gas flow
   were 50 L/h and 1000 L/h, respectively. Cone gas and collision gas were
   nitrogen and argon, respectively. Trap collision energy and transfer
   collision energy were off. Leucine enkephalin was used as a lock spray
   at a concentration of 200 pg/mL to ensure mass accuracy and
   reproducibility. This establishment was referenced in both positive
   mode at m/z 556.2771 [M+H]^+ and negative mode at m/z 554.2615 [M H]^-.
   The mass window was allowed to deviate 0.3 Da. The infusion flow rate
   of the lock spray was 10 μL/min. The scan time was every 60 s.
   MassLynx™ V4.1 software was used to operate the system. The data were
   analyzed by Progenesis QI (Waters Corp., Milford, USA), MarkerLynx™ XS
   (Waters Corp., Milford, USA), and MassFragment™ software (Waters Corp.,
   Milford, USA). The significantly changed metabolites were filtered out
   based on VIP values (VIP>1), ANOVA (P < 0.05) and max fold change (≥2).
   The filtered metabolites were selected under special conditions; match
   score ≥40, mass error (ppm) close to 0 and highest mass fraction score.
   The chemical structures of important metabolites were then identified
   according to online databases such as the Human Metabolome Database
   ([78]www.hmdb.ca), Metlin ([79]www.metlin.scripps.edu) and the MassBank
   ([80]www.massbank.jp) using the data of accurate masses and MS/MS
   fragments.

2.8. Morris water maze test

   The Morris water maze was the behavioral test for spatial learning and
   memory after YHF20 or vehicle administrations ([81]Fig. 1). The
   matt-black circle pool was 2.14 m in diameter and filled with water
   (20 ± 2 °C) to a depth of 0.25 m. The pool area was divided into 4
   quadrants. The escape platform was a clear glass cylinder 9 cm in
   diameter. Visual cues were placed on the basin's edge, above the
   surface of the quadrant's four corners. Each one of them had a unique
   color and shape. In the experimental area, the least amount of light
   was allowed for monitoring and recording rat's behavior by a video
   camera above the pool [[82]32]. SMART video tracking software v3.0
   (Panlab, Spain) was used to analyze for speed and retention time.

2.9. Hepatotoxicity and nephrotoxicity analysis

   The kidney and liver functions were assessed by BUN, creatinine (Cr),
   aspartate aminotransferase (AST), alanine aminotransferase (ALT), and
   alkaline phosphatase (ALP)-AMP, for acute (7 days) and subacute (28
   days) drug toxicity study using automated biochemistry analyzer A15
   (BioSystems, Barcelona, Spain).

2.10. Statistical analysis

   All values in both the text and figures were expressed as
   mean ± standard deviation (SD) for n observations. When the data were
   normally distributed, one-way analysis of variance with Dunnett's post
   hoc test was performed using SPSS v.25 (IBM® SPSS® statistics, New
   York, USA), and p-values below 0.05 were considered to be significant.
   If the data were not normally distributed, Kruskal–Wallis H test or
   Friedman test were performed. p-values below 0.05 were considered to be
   significant. Graphs were plotted with GraphPad prism v.9 (Dotmatics,
   Boston, MA, USA).

3. Results

3.1. Identification of tentative biomarkers and pathway after BCCAO-induced
ischemia

   PCA and OPLS-DA were applied to compare metabolic changes after
   ischemic-induced surgery in all groups at T0 compared to the 2-VO group
   at T1-T8. As demonstrated by the PCA score plot in [83]Fig. 2A–B, the
   difference between T0 of all groups versus 2-VO at T1-T8 was unclear in
   both modes. The R2X [[84]1], R2X [[85]2], R2X (cum) and Q2 (cum) were
   0.64, 0.07, 0.93 and 0.85 in positive mode and were 0.22, 0.13, 0.82
   and 0.59 in negative mode. Although, a clear separation was observed in
   the OPLS-DA score plot ([86]Fig. 2C–D). The R2Y (cum) and Q2 (cum) were
   0.75, 0.59 in positive mode and 0.90, 0.82 in negative mode. No
   overfitting was observed. The cutoff points were ±0.01 in S-plot
   ([87]Fig. 2E–F). Subsequently, a total of 78 potential biomarkers
   including 39 changed metabolites in positive and 39 in negative mode
   were identified (Supplementary material, [88]Table S2). There are 28
   potential biomarkers which were found involved in various pathways
   using HMDB or Pubchem or KEGG database ([89]Table 1).

Fig. 2.

   [90]Fig. 2
   [91]Open in a new tab

   Metabolic changes after ischemic-induced surgery at T0 compared to the
   2-VO group at T1-T8; (A) PCA score plot in positive mode, (B) PCA score
   plot in negative mode, (C) OPLS-DA score plot in positive mode, (D)
   OPLS-DA score plot in negative mode, (E) S-plot in positive mode and
   (F) S-plot in negative mode.

Table 1.

   Known-tentative metabolic change that found in database.
   No. Metabolite Formula Mass error (ppm) Database
     __________________________________________________________________

   Class
   HMDB PubChem KEGG
   1 l-Acetylcarnitine C9H17NO4 −2.24 HMDB0000201 7,045,767 [92]C02571
   Fatty Acyls
   2 Glutaurine C7H14N2O6S 2.47 HMDB0004195 68,759 [93]C05844 Carboxylic
   acids and derivatives
   3 ELLIPTICINE C17H14N2 3.10 HMDB0034310 – – Vinylogous acids
   4 Genistein C15H10O5 −1.36 HMDB0003217 5,280,961 [94]C06563
   Isoflavonoids
   5 Naringenin C15H12O5 −3.98 HMDB0002670 439,246 [95]C00509 Flavonoids
   6 O-oleoylcarnitine C25H47NO4 0.20 HMDB0005065 46,907,933 – Fatty Acyls
   7 cinitapride C21H30N4O4 1.93 HMDB0015698 68,867 – Benzene and
   substituted derivatives
   8 Ricinoleic Acid C18H34O3 0.54 HMDB0034297 643,684 [96]C08365 Fatty
   Acyls
   9 3-dehydroecdysone C27H42O6 −3.86 METPA1794 – [97]C02513 –
   10 1-Tetradecanol C14H30O −1.79 HMDB0011638 8209 – Fatty Acyls
   11 Platelet-activating factor C26H54NO7P −1.09 METPA0517 – [98]C04598 –
   12 1-Hydroxy-1-phenyl-3-hexadecanone C22H36O2 −0.27 HMDB0035676
   131,751,841 – Fatty Acyls
   13 FF-MAS C29H46O −1.25 HMDB0001023 443,212 [99]C11455 Steroids and
   steroid derivatives
   14 bilirubin C33H36N4O6 0.78 HMDB0000054 21,252,250 [100]C00486
   Tetrapyrroles and derivatives
   15 Azithromycin C38H72N2O12 0.06 HMDB0014352 447,043 [101]C06838
   Organooxygen compounds
   16 1-Palmitoyl-2-docosahexaenoyl-sn-glycero-3-phosphocholine C46H80NO8P
   1.97 HMDB0007991 6,441,886 [102]C00157 Glycerophospholipids
   17 PC(18:2 (9Z,12Z)/P-16:0) C42H80NO7P 0.28 HMDB0008159 53,478,799
   [103]C00157 Glycerophospholipids
   18 PC C46H84NO8P 3.17 HMDB0000095 8117 [104]C00055 Pyrimidine
   nucleotides
   19 Glutathione C20H32N6O12S2 −2.13 HMDB0062697 745 [105]C00051
   Carboxylic acids and derivatives
   20 Carthamone C21H20O11 −1.51 HMDB0037377 12,305,281 [106]C08598
   Cinnamic acids and derivatives
   21 Baicalin C21H18O11 −2.00 HMDB0041832 13,654,833 [107]C10025
   Flavonoids
   22 glycochenodeoxycholic acid 7-sulfate C26H43NO8S −2.70 HMDB0002496
   11,954,205 [108]C01324 Homogeneous halogens
   23 Glycodeoxycholic acid C26H43NO5 −4.44 HMDB0000631 3,035,026
   [109]C05464 Steroids and steroid derivatives
   24 Isodeoxycholic acid C24H40O4 −3.85 HMDB0002536 164,672 [110]C17661
   Steroids and steroid derivatives
   25 ruscogenin C27H42O4 −0.76 – 441,893 [111]C08909 steroidal saponins
   26 Cyclopentadecanolide C15H28O2 −2.82 HMDB0034455 235,414 – Macrolides
   and analogues
   27 4,6-Nonadecanedione C19H36O2 −4.12 HMDB0035575 91,713,156 –
   Organooxygen compounds
   28 1D-myo-Inositol 1,2-cyclic phosphate C6H11O8P −2.89 HMDB0001125
   122,331 [112]C04299 Organic phosphoric acids and derivatives
   [113]Open in a new tab

   28 metabolites were analyzed the potential pathways using MetaboAnalyst
   5.0 ([114]https://www.metaboanalyst.ca). The algorithm for pathways was
   hypergeometric test and relative-betweenness centrality selected for
   pathway enrichment analysis and topology analysis, respectively. The
   Homo sapiens from KEGG library was selected for this analysis. The
   overview of the matched pathway that arranged according to the scores
   from enrichment analysis (y axis) and from topology analysis (x axis)
   was shown in [115]Fig. 3. The pathways have impact as glutathione
   metabolism, glycerophospholipid metabolism, ether lipid metabolism,
   steroid biosynthesis, porphyrin and chlorophyll metabolism, and
   pyrimidine metabolism. Glutathione metabolism has the highest impact
   that mainly the pathway associated with cerebral ischemia and/or
   dementia. The 78 tentative metabolites and 10 tentative pathways were
   used to identify the ischemic condition.

Fig. 3.

   [116]Fig. 3
   [117]Open in a new tab

   The overview of pathway analysis from metabolic change after the
   BCCAO-induced ischemia.

3.2. Metabolic changes after YHF20 administration

   The tentative metabolites were analyzed the metabolic changes. The
   response of tentative metabolites was visualized using heatmap
   ([118]Fig. 4). The results showed 7 known-tentative metabolites tended
   to change after pBCCAO-induced ischemia such glutathione, glutaurine,
   phosphatidylcholine (PC(16:0/16:0) or PC(18:2 (9Z, 12Z)/P-16:0)),
   FF-MAS, PC, platelet-activating factor and bilirubin ([119]Fig. 5A–G)
   but not statistically significant difference between times.
   PC(16:0/16:0) tended to increase in 2-VO group at T1 and had decreasing
   trend in YHF20-treated group ([120]Fig. 5B).
   N-Carbamimidoyl-4-hydrazinobenzenesulfonamide, that only significant
   increased in 2-VO group at T1 (7 days after pBCCAO-induced ischemia)
   compare with sham group, showed decreasing trend in YHF20-treated group
   ([121]Fig. 5I).

Fig. 4.

   [122]Fig. 4
   [123]Open in a new tab

   The heatmap analysis of 78 differential metabolites identified between
   sham group, 2-VO group and 2-VO + YHF20-treated group. The vertical
   lines represent differential metabolites, and the horizontal lines
   represent group and time. The blue band indicates a decreased level of
   metabolite, and the red band indicates an increased level of
   metabolite.

Fig. 5.

   [124]Fig. 5
   [125]Open in a new tab

   Known metabolic change level; (A) Glutaurine (B) PC(16:0/16:0) (C)
   Platelet-activating factor (D) FF-MAS (E) PC (F) Bilirubin (G)
   Glutathione (H) N-Carbamimidoyl-4-hydrazinobenzenesulfonamide (I)
   N-Carbamimidoyl-4-hydrazinobenzenesulfonamide at T1. *p < 0.05 compared
   with 2-VO group; **p < 0.01 compared with 2-VO group.

3.3. Effect of YHF20 on spatial learning and memory performance

   In the visible platform trial, the results showed a statistically
   significant difference in escape latency between the sham group and the
   2-VO + YHF20 100 mg/kg group (p = 0.03). The swimming speed, distance
   to target and resting time were no statistically significant difference
   between all groups (p = 0.82, 0.19, 0.99, respectively) ([126]Fig.
   6A–D). These results indicated that visual and motor performances were
   not affected by pBCCAO. In the hidden platform trial, the results were
   a statistically significant between 2-VO and 2-VO + YHF20 100 mg/kg
   group in escape latency on day 1 (p = 0.02). The results showed a
   statistically significantly different between day 1 and day 5 in the
   sham group and all various doses of the YHF20-treated group, while the
   2-VO group showed a statistically difference between day1 and day 3
   ([127]Fig. 6E). It indicates that the pBCCAO could not yet produce the
   impairment in spatial learning and memory which was induced by chronic
   cerebral ischemia hypoperfusion under this number of rats. Spatial
   memory retention was no statistically significant difference in mean
   retention time (p = 0.38) ([128]Fig. 6F), swimming distance in the
   target quadrant (p = 0.16) ([129]Fig. 6G) and latency to 1st entire
   target quadrant (p = 0.58) ([130]Fig. 6H).

Fig. 6.

   [131]Fig. 6
   [132]Open in a new tab

   Spatial learning and memory performance after YHF20 administration; (A)
   escape latency, (B) mean swimming speed without resting time, (C)
   distance to target, (D) resting time, (E) escape latency of hidden
   platform trial, (F) retention time in target quadrant, (G) distance in
   target quadrant, (H) latency to 1st entired target quadrant. ^p < 0.05
   compared with sham group; *p < 0.05 compared with 2-VO group in the
   same day; #p < 0.05 compared with day1 in the same group.

3.4. Toxicity after YHF20 administration

   The hepatotoxicity and nephrotoxicity were not significantly different
   from sham group (p > 0.05) (Supplementary Material, [133]Fig. S3). The
   weight change showed statistically significant difference between sham
   and 2-VO group on day 49, 56 and 71 (p = 0.049, 0.006, 0.007) and
   significant difference between sham and 2-VO + YHF20 100 mg/kg group on
   day 56 and 71 (p = 0.046, 0.006). (Supplementary Material, [134]Fig.
   S3).

4. Discussion

   In the present study, a non-targeted metabolomic approach based on
   LC-MS was performed to explore the characteristics of blood metabolism
   in rats with dementia. Cerebral ischemia hypoperfusion was induced by
   pBCCAO before 8 weeks of YHF20 administration. In metabolic profiling
   analysis, we identified 78 differentially expressed metabolites. Some
   metabolites had been reported in the progression of dementia such as
   l-Acetylcarnitine, Glutaurine, O-oleoyl carnitine [[135]33,[136]34].
   The tentative metabolites were found in 2-VO group but not sham group
   may be the potential biomarkers of cerebral ischemia hypoperfusion
   especially N-Carbamimidoyl-4-hydrazinobenzenesulfonamide. However,
   there is no report about N-Carbamimidoyl-4-hydrazinobenzenesulfonamide
   pharmacological activity related to cerebral ischemia. The identified
   potential biomarkers were screened according to the FDR of
   significance. These biomarkers by pathway topology analysis were
   primarily involved in the metabolism of taurine and hypotaurine
   metabolism, alpha-linolenic acid metabolism, ether lipid metabolism,
   glutathione metabolism, porphyrin and chlorophyll metabolism,
   arachidonic acid metabolism, glycerophospholipid metabolism, pyrimidine
   metabolism, steroid biosynthesis as many previous studies [[137][35],
   [138][36], [139][37], [140][38]]. Therefore, our results revealed that
   cerebral ischemia hypoperfusion could cause cognitive impairment by
   affecting a variety of metabolic pathways mainly in glutathione
   metabolism. Glutathione metabolism links to glutathione metabolite that
   play role in antioxidant defense and regulation of cellular events
   [[141]39]. Glutathione can suppress cerebral infarct volume and cell
   death after ischemic injury [[142]40]. Nevertheless, this study was a
   non-target metabolomics study. The results were potential biomarkers
   that have not been studied for pharmacological activity. In the future,
   extraction methods should be developed to quantify these compounds or
   analyze their activity as target metabolomics studies such as the
   effect of YHF20 on glutathione level.

   Spatial learning and memory could be determined by many tasks, such as
   the spatial object recognition (SOR) test, Radial arm maze, Y-maze,
   Barnes maze and Morris water maze (MWM). This study chose the MWM to
   evaluate the effect of YHF20 on spatial learning and memory. The
   initial assessment involved evaluating sensorimotor performance in the
   visible platform trial that monitored escape latency and swimming
   speed. Swimming speed of 2-VO group and 2-VO + YHF20-treated groups
   were not different from sham group indicated the normal motor
   performance. However, it was observed that the 2-VO+100 mg/kg YHF20
   group exhibited high escape latency and a greater distance from the
   target, suggesting thigmotaxis behavior which is often associated with
   anxiety or stress in rats, leading to spending more time along the pool
   wall [[143]18,[144]41].

   During the acquisition trial, the 2-VO group showed less decreases of
   escape latency as compared to the sham group but was not statistically
   significant when compared within same day ([145]Fig. 6E). The wide
   standard deviation might cause the non-significant different results
   and a greater number of rats might be needed to cover the behavioral
   variation. The minimum number of rats was 10, indicating the difference
   between the sham group and the 2-VO group [[146]42]. Further study may
   increase number of rats in experimental period and add a positive
   control to confirm the success of treatment in this ischemic or
   learning and memory deficit model during the method validation. In
   addition, cognitive deficit should be confirmed the occurrence of the
   cerebral ischemia hypoperfusion by measuring the numbers of surviving
   hippocampal neurons (CA1, CA3 and dentate gyrus) and grading the white
   matter injury.

   Therefore, the effect of YHF20 was considered to be compared by the
   intragroup change values. YHF20 has been used in Thai traditional
   medicine. Some components had been reporting pharmacological effects
   but there was no evidence the effect of YHF20 against the cerebral
   ischemia. In this study, sham group and YHF20-treated groups showed a
   statistically significant change in escape latency between day 1 and
   day 5 in the acquisition trial. All YHF20-treated groups tended to
   cause more decreases in escape latency than the 2-VO group on day 5.
   After the competition of the acquisition trial, the retention of
   spatial memory was assessed in probe trial. The data showed no
   difference of retention time in target quadrant between any groups and
   sham group tended to use more time to 1st entired the target quadrant.
   However, 2-VO+300 mg/kg YHF20 group tended to be more distant in target
   quadrant than other groups that indicated rats might remember the
   platform position.

   Another potential issue with the MWM is that the period after 2-VO also
   plays a role in evaluating spatial learning and memory impairment.
   Various times after the 2-VO procedure were chosen from 30 to 90 days
   [[147][43], [148][44], [149][45]]. The previous study showed
   arteriogenesis in the vertebrobasilar tree to recover cerebral blood
   flow and signs of optic tract degeneration after 3 months [[150]46] and
   reported the significant different escape latency at 8 weeks from 1 to
   3 weeks [[151]47]. Likewise, this study chose 2 months after 2-VO
   corresponded to YHF20 was usually consecutively used for 1–3 months ago
   in clinical practice.

   YHF20 is a herbal formula regularly used in the Ayurved Siriraj Clinic
   of Thai traditional medicine to improve symptoms resulting from wind
   element obstruction in the upper region of the body, such as nausea,
   vomiting, dizziness and exhaustion. According to Thai traditional
   medicine theory, YHF20 has a gentle taste that can regulate wind
   element to restore normal blood flow. Cerebral blood flow deficiency
   for brain metabolism leads to cerebral ischemia and the development of
   vascular dementia. Many compounds have been reported in the component
   of YHF20 which relate to attenuate the cerebral ischemic injury such as
   antioxidation, anti-inflammatory effect, anti-ischemic effect. This
   study found 4 tentative metabolites may be the metabolites of YHF20's
   constituents including genistein, naringenin, baicalin and ruscogenin.
   Genistein downregulate the expression of nod-like receptor protein 3
   (NLRP3) inflammasome, decreased inflammatory factor release and
   regulate the mitochondria-dependent apoptosis pathways [[152][48],
   [153][49], [154][50]]. Naringenin can inhibit oxidative stress and
   inflammation [[155]51,[156]52]. Baicalin can attenuate ischemia-induced
   neuronal cell damage via anti-oxidative and anti-apoptotic effect
   [[157]53,[158]54]. Ruscogenin also suppress thiredoxin-interactive
   protein (TXNIP) and NLRP3 inflammasome activation. Therefore, the
   effect YHF20 on vascular system are required in the future
   investigation.

   This study had limitations that could be addressed in future research.
   First, the cerebral ischemia condition confirmation might directly
   measure using histological staining [[159]55,[160]56],
   Immunofluorescence analysis [[161][55], [162][56], [163][57]],
   brain-derived neurotrophic factor (BDNF) level [[164]58,[165]59]. Then,
   this can be indirectly measure by inflammatory cytokines detection
   [[166]60,[167]61] and oxidative stress assessment [[168]56,[169]61].
   Second, increasing of the sample size to at least 10 will improve the
   accuracy of the behavioral result and avoid erroneous conclusions.
   However, this study is the first to examine how YHF20 affects rats
   after neural damage. It was discovered that YHF20 can change
   metabolites associated with pathways which occur during cerebral
   ischemia. Furthermore, these findings can serve as the groundwork for
   evidence-based medicine in future studies. These entails developing a
   method for determining a targeted metabolic change and performing in
   vitro tests using anti-inflammatory, antioxidant evaluations to provide
   information for investigating the metabolic pathways of YHF20.

5. Conclusion

   In this study, the permanent bilateral common carotid artery occlusion
   for 2 months was not interfering the learning and memory performance
   (n = 6–11). Although, the administration of the various dose of YHF20
   has not been able to conclude the effect on spatial learning and memory
   impairment but might contribute in metabolic changes related to
   cerebral ischemia condition. Glutathione and glutathione metabolism may
   be the metabolic path affected by YHF20. The 4 identified metabolites
   which might be the metabolites from the components of YHF20 can be
   taken into consideration and verified for quality control. No toxicity
   was observed after the YHF20 administration for 56 days.

Ethics statement

   All experimental protocols were approved by the Siriraj Animal Care and
   Use Committee (SiACUC), Faculty of Medicine Siriraj Hospital, Mahidol
   University. (COA No. 002/2559).

Funding statement

   This research was made possible through the generous financial support
   of the Siriraj Research Development Fund, grant number R016032004,
   Faculty of Medicine Siriraj Hospital, Mahidol University, Thailand.

Data availability statement

   Data included in article/supplementary material/referenced in article.

CRediT authorship contribution statement

   Pravit Akarasereenont: Writing – review & editing, Methodology,
   Conceptualization. Saracha Pattanapholkornsakul: Writing – original
   draft, Methodology, Formal analysis, Data curation. Suveerawan
   Limsuvan: Writing – original draft, Data curation. Dollaporn
   Mamaethong: Writing – original draft, Data curation. Suksalin
   Booranasubkajorn: Writing – review & editing, Formal analysis, Data
   curation. Narawut Pakaprot: Writing – review & editing, Methodology,
   Conceptualization. Pinpat Tripatara: Writing – review & editing,
   Methodology, Conceptualization. Kajee Pilakasiri: Writing – review &
   editing, Methodology, Conceptualization.

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.

Acknowledgement