LC-MS/MS based metabolite profiling and lipase enzyme inhibitory activity of Kaempferia angustifolia Rosc. with different extracting solvents

2.1 Chemicals and reagents

The materials used were methanol, acetone, water grade LC-MS, dipotassium hydrogen phosphate, potassium dihydrogen phosphate, sodium chloride, sodium hydroxide, acetonitrile, lipase from porcine pancreas, and p-nitrophenyl butyrate (pNPB) were purchased from Merck (Darmstadt, Germany). The UHPLC-Q Exactive Plus Orbitrap HRMS (Thermo Fisher, Waltham, USA) was used for metabolite profiling and the microplate reader (Epoch BioTek, Winooski, USA) was used for in vitro enzymatic activity measurement.

2.2 Sample preparation and extraction

Kaempferia angustifolia was obtained from the Tropical Biopharmaca Research Center (TropBRC), IPB University, Bogor, West Java, Indonesia. The specimen has been identified with a voucher specimen (BMK0433012018) by Mr. Taufik Ridwan, a botanist from TropBRC. Before being used for extraction, we dried and pulverized the rhizomes of K. angustofolia. The dried sample was macerated for about 12 h with a ratio of 1:15 in ethanol p.a., 70% ethanol, 50% ethanol, 30% ethanol, and water. The filtrate was concentrated using a rotary evaporator to produce the extract. Then each sample extract was analyzed for the LC-MS/MS analysis and inhibition of lipase enzyme activity.

2.3 LC-MS/MS analysis

The metabolite profiles were analyzed using UHPLC-Q Orbitrap HRMS. The column used was Accucore C₁₈ (100 × 2.1 mm), with a particle size of 1.5 μm. The column was eluted using 0.1% formic acid in water (A) and 0.1% formic acid in acetonitrile (B), with gradient conditions of 0–2 min 5% B, 2–3 min 5–25% B, 3–14 min 25–100% B; 14–19 min 100% B, and 19–23 min 5% B with a flow rate of 0.2 mL/min. The injection volume was 5 μL. The ionization source is ESI with positive and negative ionization modes. The mass scan range was set to m/z 100–1500. The mass spectrometry was conditioned: Capillary temperature 320 °C, spray voltage 3.2 kV, S lens RF level 50, sheath gas and aux gas flow rates 12 and 3, respectively. The power resolution was set to 70,000 FWHM. Accurately weighed 10 mg extract was dissolved in 5 mL methanol and sonicated for 30 min. The solution was filtered using 0.2 μm PTFE membrane and analyzed by UHPLC-Q Orbitrap HRMS.

2.4 In vitro lipase inhibition assay

Enzymatic inhibition assay was performed according to Chedda et al. (2016). Briefly, pNPB was used as a substrate in the enzymatic reaction. The test tube contained 100 μL of phosphate buffer (pH 7.2), 50 μL of pancreatic lipase, 25 μL of pNPB, and 25 μL of K. angustifolia extract (100 µg/mL) in total volume of 200 μL. The blank tube for 100% activity was also prepared by replacing the 25 μL of K. angustifolia extract with 100 μL of phosphate buffer (pH 7.2). All tubes were incubated at 37 °C for 30 min. Orlistat was used as an inhibitor for positive control. para-nitrophenol (pNP) produced from the enzymatic reaction was measured spectrophotometrically by measuring the absorbances at 400 nm using a microplate reader. The percentage of inhibitory activity was calculated using the following formula: $$\frac{(\mathrm{A}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}-\mathrm{A}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{s}\mathrm{a}\mathrm{m}\mathrm{p}\mathrm{l}\mathrm{e})}{\mathrm{A}\mathrm{b}\mathrm{s}\mathrm{o}\mathrm{r}\mathrm{b}\mathrm{a}\mathrm{n}\mathrm{c}\mathrm{e}\mathrm{o}\mathrm{f}\mathrm{b}\mathrm{l}\mathrm{a}\mathrm{n}\mathrm{k}}\times 100\%$$

2.5 Data analysis

The metabolite profiles were processed using the Compounds Discoverer 3.2 (Thermo Fisher, Waltham, USA) with in-house databases of metabolites collected from the Kaempferia genus. The steps in identifying the metabolites began with selecting the spectra, aligning the retention times, detecting the unknown compounds, grouping the unknown compounds, predicting the compositions, searching the mass list, filling the gaps, normalizing the areas and marking the background compounds. The MS2 was confirmed to identify the metabolites.

PLS-DA was performed to determine the pattern of extract grouping based on the extracting solvents using area data of the successfully identified compounds or metabolites. The PLS-DA analysis used The Unscrambler X software version 10.1 (Camo, Oslo, Norway). The success of the extract grouping was seen in the number of key components involved, the total variation represented, and the score plot's visualization.

Percentage of inhibition lipase enzyme activity was presented as mean ± standard error for at least three independent experiments. Statistical comparison for inhibition lipase enzyme activity was carried out using one-way analysis of variance (ANOVA) followed by the Tukey test. We used a significant difference at the 95% confidence level (p < 0.05) was used.

3.1 Carboxylic acids

There are eight carboxylic acids identified in the K. angustifolia extract, namely citric acid (1), trans-aconitic acid (2), 2-isopropylmalic acid (4), phenyllactate (6), azelaic acid (9), 4-hydroxybenzoic acid (11), benzoic acid (17), gemfibrozil (23). Fragmentation of the carboxylic acids mostly by releasing one or two H₂O molecules characterized by the ion fragments [M − H − 18]⁻ and CO₂, at m/z 129 [M − H − CO₂]⁻ and 111 [M − H − CO₂ − H₂O]⁻ for trans-aconitic acid, m/z 147 [M − H − H₂O]⁻ and 103 [M − H − H₂O − CO₂]⁻ for phenyllactate. Compounds with m/z 176 identified as 2-isopropylmalic acid provide fragments at m/z 157 [M − H − H₂O]⁻, 131 [M − H − H₂O − 26]⁻, and 113 [M − H − 2H₂O − 26]⁻.

Azelaic acid with [M − H]⁻ at m/z 187 is fragmented by releasing one CO₂ molecule at [M − H − 44]⁻, one H₂O molecule at [M − H − 44–18]⁻, and one CO molecule at [M − H − 44–18 − 28]⁻. The compound (1) is citric acid, as indicated by fragments at m/z 173, 147, 129, 111, and 87. The compound (11) at m/z 137 is identified as a 4-hydroxybenzoic acid for fragments at m/z 108, 93, and 65. Compound (17) at m/z 123 is known as benzoic acid, and compound (23) is identified as gemfibrozil-produced fragments at m/z 233, 191, 105, and 71.

3.2 Terpenoids

Terpenoids are predominant in the extract (Elshamy et al., 2019), and are divided into three types, namely monoterpenoids (2 compounds), diterpenoids (3 compounds), and sesquiterpenoids (12 compounds). The monoterpenoids are camphor (37) at m/z 153 which produces fragments at m/z 135 [M + H − H₂O]⁺, 109 [M + H − H₂O − C₂H₂]⁺, 95 [M + H − H₂O − C₂H₂ − CH₂]⁺ and camphene (42) at m/z 137 with fragments at m/z 95 [M + H − 42]⁺ and [M + H − 42 − CH₂]⁺ (Turek and Stinting, 2011). The three diterpenoids are retinoic acid (14), traversianal (27), and retinal (40). Retinoic acid with [M + H]⁺ at m/z 301 fragmented at m/z 283 [M + H − H₂O]⁺, 255 [M + H − H₂O − CO]⁺, 227 [M + H − H₂O − CO − 2CH₂]⁺, 213 [M + H − H₂O − CO − 3CH₂]⁺, 185 [M + H − H₂O − CO − 5CH₂]⁺, 157 [M + H − H₂O − CO − 7CH₂]⁺, and 143 [M + H − H₂O − CO − 8CH₂]⁺. Compounds (27) at m/z 317 are recognized as traversianal with fragments at m/z 299 [M + H − H₂O]⁺, 271 [M + H − H₂ − CO]⁺, and 253 [M + H − 2H₂O − CO]⁺. Retinal with [M + H]⁺ at m/z 285 produces fragments at m/z 285, 215, 201, 159, 143, and 83.

In addition, 12 sesquiterpenoids are identified in the extract. Furanogermenone (12) at m/z 233 [M + H]⁺ is identified at retention times of 7.71 min with fragmentation at m/z 175, 147, 119, 105, and 91. The compound (13) is identified as curzeone at m/z 229 [M + H]⁺ with fragments at m/z 211 [M + H − H₂O]⁺. The compound (16) at m/z 253 [M + H]⁺ is known as zedoarondiol with fragments at m/z 235 [M + H − H₂O]⁺, 175 [M + H − H₂O − 60]⁺ showing the opening of cyclopentane and loss of H₂O molecules, and 147 [M + H − H₂O − 60 − 2CH₂]⁺. Compound (20) at m/z 237 [M + H]⁺ is identified as neocurdione with a fragment releasing one H₂O molecule at m/z 219 [M + H − 18]⁺. The compound (21) at m/z 239 [M + H]⁺ is recognized as a culmorin with fragments at m/z 221, 203, 177, 163, 161, 149, 147, 137, 133, 121, 107, 95, and 81 (Hao et al., 2019).

The compound (22) is known as zederone at m/z 247 [M + H]⁺ fragmented at m/z 229 [M + H − H₂O]⁺ and 183 [M + H − H₂O − C₂H₆O]⁺. Compounds (24) is caryophyllene oxide at m/z 221 [M + H]⁺ produces fragments at m/z 175, 161, 147, and 95 (Turek et al., 2011). The compound (26) is curcumenol at m/z 235 [M + H]⁺ and the fragments at m/z 217 [M + H − H₂O]⁺. Compound (28), furanodiene at m/z 217 [M + H]⁺ has fragments at m/z 157 and 119. From the compound (36) cis-nerolidol at m/z 223 [M + H]⁺ there are fragments at m/z 73 [M + H − 150]⁺ and compound (39) α-cyperone at m/z 219 [M + H]⁺ has fragments at m/z 163. The compound (41) at m/z 213 [M + H]⁺ is pyrocurzerenone as indicated by fragments at m/z 198 [M + H − CH₃]⁺ and 128 [M + H − 85]⁺ (Hao et al., 2019).

3.3 Phenolic acids

A total of 2 phenolic acids were identified in K. angustifolia extract and 2 simple phenolic compounds. The first compound identified as phenolic acid is o-coumaric acid (3) at m/z 165 [M + H]⁺ with fragments at m/z 147 [M + H − H₂O]⁺, 123 [M + H − C₃H₆]⁺, 95 [M + H − C₃H₆ − CO]⁺. The second compound (10) is octyl gallate at m/z 281 [M − H]⁻ fragmented at m/z 237 [M − H − CO₂]⁻, 139 [M − H − CO₂ − 98]⁻, and 125 [M − H − CO₂ − 139]. The simple phenolic compounds identified are 2,4,5-trimethoxybenzaldehyde (m/z 197 [M + H]⁺) and 4-tert-butylcatechol (m/z 167 [M + H]⁺).

3.4 Flavonoids

There are two flavonoid compounds identified and belong to the flavanones. Pinocembrine (25) at m/z 255 [M − H]⁻ is identified at a retention time of 10.16 min with fragmentation at m/z 213 [M + H − C₃H₆]⁺, 211 [M + H − C₂H₄O]⁺, 151 [M + H − C₈H₈]⁺, and 145 [M + H − C₇H₁₀O]⁺. Alpinetin (33) at m/z 271 [M + H]⁺ has fragments at m/z 253 [M + H − H₂O]⁺, 229 [M + H − C₃H₆]⁺, 167 [M + H − H₂O − C₅H₁₁O]⁺, and 131 [M + H − H₂O − C₅H₁₄O₃]⁺.

3.5 Other groups of compound

Some other groups of compounds in the extract are aldehyde (benzaldehyde), ketone (dihydrojasmone), fatty acid (eicosapentaenoic acid), furan (zedoarofuran), anthraquinone (eleutherin), coumarin (3,4-dihydrocoumarin), lactone (curcumenolactone A), chalcone (2′-hydroxy-2,4,4′-trimethoxychalcone), vinilogous acid (hulupinic acid and benzyl salicylate), and steroid (norethindrone). Compound (31) is benzaldehyde at m/z 107 [M + H]⁺. The dihydrojasmone (29) is identified at m/z 167 [M + H]⁺ with fragments at m/z 123, 111, 107, 97, 93, and 81. The compound (32) at m/z 303 [M + H]⁺ is recognized as eicosapentaenoic acid and has fragments at m/z 285, 267, 229, 163, and 109.

Zedoarofuran (8) is identified at m/z 265 [M + H]⁺ with fragments at m/z 247 [M + H − H₂O]⁺, 229 [M + H − 2H₂O]⁺, and 105 [M + H − 2H₂O − C₈H₁₂O]⁺ (Hao et al., 2019). The compound (19) at m/z 273 [M + H]⁺ is known as eleutherin with its fragments at m/z 255, 199, and 185. Compound (30) at m/z 149 [M + H]⁺ is 3,4-dihydrocoumarin with fragmentation releasing CO [M + H − 28]⁺ and CH₂ [M + H − CO − 14]⁺. Curcumenolactone A (18) is identified at m/z 249 [M + H]⁺ producing fragments at m/z 203, 143, and 105 (Hao et al., 2019). The compound (38) 2′-hydroxy-2,4,4′-trimethoxychalcone is at m/z 315 [M + H]⁺ giving fragments at m/z 273, 191, 190, and 151.

Hulupunic acid (15) at m/z 265 [M + H]⁺ produces fragments at m/z 247 [M + H − H₂O]⁺, 219 [M + H − H₂O − CO]⁺, 177 [M + H − H₂O − CO − C₃H₆]⁺, and 145 [M + H − H₂O − CO − C₄H₁₀O]⁺. The compound (35) at m/z 229 [M + H]⁺ is identified as benzyl salicylate with its fragments at m/z 151, 105, 95, 77, and 53. The compound (34) norethindrone in at m/z 299 [M + H]⁺ with fragments at m/z 281, 241, 229, 199, 185, 171, 83, and 55.

2-acetylrotepoxide A (43) pada m/z 427 [M + H]⁺ produce a fragmentation at m/z 105 [M + H − C₁₆H₁₆O₈]⁺, follow by releasing CO molecule at m/z 77 [M + H − C₁₆H₁₆O₈ − CO]⁺. (-) 6-acetylzeylenol (44) at m/z 427 [M + H]⁺ fragmented at m/z 105 [M + H − C₁₆H₁₇O₇]⁺ and follow with the releasing CO molecule at m/z 77 [M + H − C₁₆H₁₇O₇ − CO]⁺. 5,7-dihydroxyflavanone (45), a compound of the flavonoid group, fragmented at m/z 105 after retro diels alder reaction (RDA). Zeylenol (46) was identified with m/z 385 and fragmented at m/z 263, 123, and 105.

9-octadecenamide (47) at m/z 282 was fragmented with releasing NH₂ at m/z 265 [M + H − NH₂]⁺. Benzylbenzoat (48) was identified at m/z 213 and gave fragmentation by releasing the C₇H₇O molecule at m/z 105 [M + H − C₇H₇O]⁺. Benzylalcohol (49) at m/z 109 produces a fragment at m/z 91 by releasing H₂O [M + H − H₂O]⁺. Bis (2-ethylhexyl) phthalate (50) at m/z 391 produces a fragment at m/z 391, 167, 149, dan 71. Propylparaben (51) was identified at a retention time of 12.435 min at m/z 181. Sandaracopimaradiene (52) was identified in 18.422 min at m/z 273[M + H]⁺. The stigmasterol (53) was identified with m/z 413, 107, dan 81.

3.6 Classification of Kaempferia angustifolia extracts

K. angustifolia extracts with different extracting solvents are classified based on the area of the identified compound. The variables used in this PLS-DA analysis were the area of 53 identified compounds. Each point on the score plot represents a single sample, and the samples indicating similarities were grouped (Liu et al., 2016). Fig. 2 shows the plot score, explaining that 99% of the total variation (PC1 = 97%, and PC2 = 2%). Samples with the same label or color are grouped in adjacent positions (96% ethanol and water extracts). Some single samples do not represent a good grouping (30%, 50%, and 70% EtOH extracts).

The resulting score plot (Fig. 1) explains that the extracts from the solvent composition can be differentiated into three groups: Group 1 (EtOH), Group 2 (50% EtOH), and group 3 (30, 70% EtOH, and water). The 96% and 50% EtOH extracts can be classified well, while water, 30%, and 70% EtOH do not group well. It indicates that the 96% and 50 ethanol extracts have different metabolite compositions, while those of water, 30%, and 70% EtOH extracts have similar metabolite content as the compositions are close together.

The PCA biplots can explain the compounds that play a role in or influence the sample grouping patterns (Fig. 3). Each point on the PLS-DA loading plot is a component variable that contributes significantly to the differences between groups, which is the furthest component of the main group (Taha et al., 2020). Several variables or compounds affect the samples' grouping pattern differentiation based on the combined plot between the score plot and loading (biplot). The biplots (Fig. 2) show that some compounds can be used as markers in grouping K. angustifolia extracts based on extracting solvents. The compounds that play a role in the grouping pattern are α-cyperone, neocurdione, 2′-hydroxy-2,4,4′-trimethoxychalcone (Fig. 4).

Close

Fig. 3

Plot score of PLS-DA of the five K. rotunda extracts.

Export to PPT

Close

Fig. 4

Biplot of PLS-DA of the five K. rotunda extracts.

Export to PPT

3.7 In vitro pancreatic lipase inhibition activity

Inhibition of water extract and ethanol extract of K. angustifolia on pancreatic lipase enzyme activity was measured in vitro and the results were compared with Orlistat as a positive control. As shown in Fig. 5, Orlistat produced a high percentage inhibition value of about 72.54%, indicating strong inhibitory activity towards pancreatic lipase. For the sample extracts with a 100 µg/mL concentration, 50% ethanol extract of K. angustifolia gave the highest inhibition, about 59.82%, compared to the other extracts. These results show that K. angustifolia extracted with 50% ethanol contains chemical compounds that have the potency of pancreatic lipase inhibitors.

Close

Fig. 5

Pancreatic lipase inhibition activity of K. angustifolia extracts. Numbers followed by different letters show a significant difference (p < 0.05).

Export to PPT

Several compounds identified in the water and EtOH extracts of K. angustifolia, namely citric acid, 4-hydroxybenzoic acid, benzoic acid, pinocembrine, propylparaben, and stigmasterol, have been reported to have inhibitory activity against lipase enzymes (Ekinci et al., 2015, Ong et al., 2016, Liu et al., 2020, Sari et al., 2021). Other compounds such as 2,4,5-trimethoxybenzaldehyde, azelaic acid, octyl gallate, retinoic acid, and benzyl benzoate are also reported to have potency as antiobesity agents but through other mechanisms other than lipase inhibitors (Ha et al., 2004, Taha et al., 2008, Neels 2013, Wang & Kuo 2014, Khairudin et al., 2018, Lee et al., 2021).