Temporal Tracking of Metabolomic Shifts in In Vitro-Cultivated Kiwano Plants: A GC-MS, LC-HRMS-MS, and In Silico Candida spp. Protein and Enzyme Study

Abstract

Cucumis metuliferus E. Mey or kiwano/African horned melon is a good source of bioactive compounds of various pharmacological and industrial importance. This study investigated metabolomic shifts in in vitro cultivated kiwano plants over ten weeks of maturity time through GC-MS and LC-HRMS-MS untargeted analysis of volatile and non-volatile metabolites. Furthermore, in silico screening of the highly abundant volatile compounds from each sample was performed against three different proteins and enzymes of Candida spp. These results obtained from GC-MS and LC-HRMS-MS analysis highlight the potential of in vitro culture for enhancing the biosynthetic potential of C. metuliferus for sustainable and controlled production of target metabolites. Furthermore, this work also highlights the potential inhibitory properties of abundant volatile compounds in each stage of maturation period of C. metuliferus, providing a platform for further exploration of the therapeutic applications of C. metuliferus metabolites against Candida spp.

1. Introduction

A continuous search for products rich in biomolecules and secondary metabolites that can be used in the food industry and in industries related to human health has become one of the new global targets. Plant-derived natural products have become a focus of scientific research with the aim of isolating and identifying these types of chemical compounds in larger quantities to determine their nutritional or bioactive potential.

In vitro plant tissue culture methods are essential for producing plant-derived metabolites with significant commercial and biological value [1]. Many plants grown in vitro show a better proportion of metabolites compared to those grown in a greenhouse or field. A good example is Ziziphora persica, which produces a higher amount of salicylic acid compared to control when grown in vitro [2]. Many commercial laboratories and national institutes worldwide use in vitro culture systems for rapid plant multiplication, germplasm conservation, elimination of pathogens, genetic manipulations, and for secondary metabolite production [3].

Cucumis metuliferus E. Mey—commonly known as African horned melon, jelly melon (England), and bitter wild cucumber (South Africa) [4]—belongs to the family Cucurbitaceae with about 120 genera and 800 species [5]. It is usually a climbing plant but often also a hanging plant with stems up to 5 m long on wooden rhizomes. The kiwano fruit is round ellipsoid–cylindrical in shape. The cortex (exocarp) is spotted green (unripe) or orange (ripe). The kiwano is covered with conical protrusions that have sharp spines at their tips. The endocarp or interior of the kiwano fruit is a green, sticky mass in which hundreds of white seeds are embedded [6]. The herb and roots of this plant are used in traditional medicine for treatment of gonorrhoea, post-partum pains, headache [7], and asthma [8]. A large number of phytochemicals and nutrients contained in kiwano fruit can contribute to both the pharmaceutical and food industries, as it is well known for its antiviral [9], antimicrobial [10], anti-ulcer [11], and hypoglycemic [12] activities. The taste of kiwano could be significantly improved under greenhouse conditions and with good fertilization and irrigation [13].

Previous chemical studies have shown that kiwano fruits and leaves contain alkaloids, flavonoids, saponins, triterpenoids, cardiac glycosides, and volatile oils. However, while the leaves contain all these classes of compounds and steroids, they lack triterpenoids and cardiac glycosides [14], a class of animal and plant-derived compounds used in treatment of heart failure [15]. Flavonoids (catechin, epicatechin, rutin, quercetin, quercetin-3-galactoside, kampferol-3-glucoside, kaempferol), phenolic acids (gallic acid, neochlorogenic acid, chlorogenic acid, caffeic acid, p-coumaric acid, trans-ferulic acid), triterpenoids (oleanolic acid, and ursolic acid), and procyanidin B2 and A2 are present in fresh juices of C. metuliferus, with catechin, which is a compound with antibacterial, antioxidant, and hepatoprotective properties [16]. It presents the largest concentration of 928.74 ± 0.9 mg kg⁻¹ compared to all quantified compounds [17]. Flavonoid isovitexin-2″-O-glucoside, besides the above-mentioned flavonoids, is present in the leaf [18]. Vitamins like thiamine (vit B1), riboflavin (vit B2), niacin (vit B3), pantothenic acid (B5), vitamin B6, and folate (vit B9) are present in the raw horned melon fruit, with the highest concentration of niacin being 0.565 mg per 100 g of raw horned melon fruit [19]. The fruit also contains measurable amounts of citric and malic acids, which contribute to its antioxidant potential and make kiwano a valuable natural source of bioactive compounds [20].

A total of 20 different volatile compounds and 17 amino acids are detected in the seeds of kiwano. It is noted that oleic acid is the compound with the largest peak area (22.25%) among all volatile compounds, and glutamic acid is the most prominent compound among all detected non-volatile compounds, with a concentration of 13.1 mg per 100 mg of protein [21]. Myristic, palmitic, stearic, and arachidic acids along C16:1n-9, C18:1n-9, C18:1n-7, C8:2n-6, and C18:3n-3 fatty acids are present in the seed, with the highest concentration of C18:2n-6. α-tocopherol, γ-tocopherol, and β-carotene are also present in the seed, with the concentration of γ-tocopherol being the highest among these three compounds [22], making kiwano seed a rich source of γ-tocopherol (vitamin E), a compound with antioxidant, anticancer, and anti-inflammatory properties [23].

It has previously been shown that C. metuliferus could be a good source of biologically active secondary metabolites. Cultivation of this plant in vitro should aim to improve the overall yield of potentially biologically active compounds. GC-MS, an efficient method of separation and identification of compounds, is used in this study. This method uses precise separation with fast and highly accurate identification of volatile compounds, and it is highly sensitive. Furthermore, we also use LC-MS-Orbitrap, a method ideal for non-targeted and unknown analysis with strong sensitivity and wide dynamic range and excellent performance with complex samples.

The objective of this study was to track how volatile and non-volatile metabolites change throughout a ten-week period that captures the full early in vitro development of Cucumis metuliferus, from germination to the formation of a young plant, thereby avoiding artefacts associated with overgrown or stressed cultures and providing a realistic view of natural metabolic dynamics. Additionally, this study aimed to perform molecular docking analyses of the most abundant volatile compounds with key Candida spp. enzymes and proteins.

2. Materials and Methods

2.1. Collection of Seed Material

Ripe fruits of the C. metuliferus (Cucurbitaceae) cultivated for experimental use only were collected from a local grower in Bogatić, Serbia during September 2022 and kept refrigerated until analysis. The pulp and seeds were separated from the peel, and the seeds were air-dried and stored until further use.

2.2. Plant Material and In Vitro Culture Initiation

C. metuliferus seeds were surface sterilized in a 3% solution of commercial bleach for 10 min and rinsed five times using sterile deionized water and subsequently transferred to Petri dishes containing 20 mL of half-strength Murashige and Skoog medium at pH 5.8, supplemented with 20 g l-1 sucrose, 7 g l-1 agar (Torlak, Belgrade, Serbia). Seeds were kept in a growth chamber under long-day conditions (16 h light/8 h dark regime) at 25 ± 2 °C and relative humidity of 60–70%. After five days, germinated seedlings were aseptically transferred to 350 mL glass jars closed with polycarbonate caps, each containing 70 mL of the same medium. Aerial parts of plants were collected every two weeks over a period of ten weeks, and samples were frozen in liquid nitrogen and stored at −4 °C until analysis. Samples were marked according to time intervals of collection: Week 2 (T1), Week 4 (T2), Week 6 (T3), Week 8 (T4), and Week 10 (T5).

A ten-week window captures the complete early in vitro development of Cucumis metuliferus from germination through progressively advanced seedling and juvenile stages to a fully formed young plant. The tracking of metabolites across such natural stages avoids artefacts of overgrown or stressed cultures and provides a realistic view of how volatile and non-volatile compounds change during early growth.

2.3. Preparation of the Samples for GC-MS and LC-MS

Samples were powdered using mortar and pestle and extracted for 10 min using 99.8% methanol (w:v = 1:10). The suspension was centrifuged, and supernatants were filtered through 0.2 µm filters into the HPLC vial and stored at 4 °C until use. Samples prepared this way were firstly analyzed by GC-MS, and later by the HPLC-Orbitrap.

2.4. GC-MS Analysis of Volatile Compounds

The method for identification of volatile compounds was previously described [24]. An Agilent 8890 gas chromatography (GC) system with 5977B GC/MSD (Agilent Technologies, Santa Clara, CA, USA) coupled with the Centri sample extraction and enrichment platform (Markes International Ltd., Bridgend, UK) was used to profile the volatile compounds in C. metuliferus samples grown in vitro. Chromatographic separations were carried out using He (99.999%, The Linde Group, Dublin, Ireland) as a carrier gas at a flow rate of 1.6 mL min⁻¹ on an HP-5MS column (30 m × 0.25 mm, 0.25 mm film thickness) (Agilent Technologies, CA, USA). The detector temperature was set to 270 °C, and the transfer line was heated to 280 °C. The EI source’s temperature was set to 280 °C, and mass spectra were obtained in positive EI mode (+70 eV). The temperature of the column was held isothermally at 240 °C for the next 10 min after being linearly programmed from 40 to 300 °C at a rate of 20 °C min⁻¹. A split mode (20:1) injection of 1 mL of methanol extract was performed with a split flow of 24 mL min⁻¹. SCAN mode was used for the analyses, which tracked the compounds between 45 and 500 amu. The components of the reaction mixtures were determined by comparing their retention periods and mass spectra to those of the corresponding standards as well as to the NIST05 library.

2.5. LC-HRMS-MS Analysis of Non-Volatile Compounds

The method used for this type of analysis was previously described [25]. The extract’s metabolic profile was determined using an LC-HRMS-MS (Thermo Scientific™ Vanquish™ Core HPLC system coupled with an Orbitrap Exploris 120 mass spectrometer, San Jose, CA, USA). The Hypersil GOLD^TM C18 analytical column (50 × 2.1 mm, 1.9 μm particle size) was part of the liquid chromatography system. The flow rate was steady at 300 µL/min, and the injection volume was 5 µL. Ultrapure water + 0.1% formic acid (A) and acetonitrile (MS grade) + 0.1% formic acid (B) were used to elute the compounds of interest: 5% B for the first minute, 5–95% B for 1–10 min, 95% B for 10–12 min, and 5% B until 15 min. An ESI source running in negative ionization mode was installed in the Orbitrap Exploris 120 mass spectrometer. While data-dependent MS² experiments were carried out at an Orbitrap resolution of 15,000 FWHM with a normalized collision energy of CID set to 35%, full-scan MS experiments were monitored from 100 to 1500 m/z with an Orbitrap resolution of 60,000 FWHM. Based on their chromatographic behavior and literature data that offered a preliminary identification, compounds were identified. The molecular formulas of compounds of interest were obtained by using full-scan MS analysis to determine the monoisotopic mass of unknown compounds. The compounds’ chemical structures were tentatively identified using the fragmentation pathway that was produced by high-resolution MS² fragmentation. Chemical formulas or suggested chemical structures were entered into the scientific SciFinfer database (https://scifinder-n.cas.org/; accessed on 15 September 2025) to conduct the literature search. These were then connected to LC-MS or other spectroscopic information on Cucurbitaceae species. The Xcalibur^® data system (Thermo Finnigan, San Jose, CA, USA) was used to collect the data.

2.6. Molecular Docking Studies

A cellular receptor in complex with FK506 from Candida glabrata (FKBP12-FK506 complex, PDB ID: 5HUA), enzyme phosphomannose isomerase in complex with inhibitor from Candida albicans (PDB ID: 5NW7), and sterol 14-alpha demethylase in complex with a tetrazole-based antifungal drug candidate from Candida albicans (CYP51-VT1 complex PDB ID: 5TZ1) were the target enzymes and protein, whose crystal structures were sourced from the Protein Data Bank (https://www.rcsb.org/, accessed on 6 October 2025). The specific protein ligand complexes’ interactions were also investigated in detail. Preparation of the enzymes, protein, and ligands for docking and visualization of the protein–ligand interactions were carried out according to the procedure given in the previous work [26] using Data Warrior [27], OpenBabel GUI [28], AutoDockTools 1.5.7 [29], AutoDock Vina [30], and Biovia DS Visualizer 4.5 (Dassault Systèmes Biovia Software Inc., Wien, Austria). Grid box sizes for all three enzymes were the same with 40 Å for x, y, and z. In the GC-MS chromatograms, phytol showed the largest peak in T1 and 2,5-diaminopentanoic acid in T2 and T3. For the docking analysis, the compound with the second largest peak area was adenine, and it was selected from T3. The T4 and T5 samples yielded 1-Azabicyclo[3.1.0]hexane and octadec-9-enamide, respectively, as representative compounds. The co-crystallized ligands of each enzyme and protein were used to define the active sites. The ligands with the lowest binding energy demonstrated the highest binding affinity to the enzyme and protein active site, according to the estimation of each ligand’s binding energy.

2.7. Statistical Data Analysis

To differentiate between samples, hierarchical cluster analysis (HCA) plots were constructed in Morpheus software (https://software.broadinstitute.org/morpheus; accessed on 15 September 2025), based on the Spearman method of cluster agglomeration, adopting the average linkage method.

3. Results and Discussion

3.1. GC-MS Analysis of Volatile Compound Analysis of In Vitro Plant Tissue Culture of C. metuliferus

Table 1. Compound names, retention times (t_R, min), molecular formulas, and presence in the samples of C. metuliferus in vitro cultures.

Compound Name	tR, min	Molecular Formula	T1	T2	T3	T4	T5
2-methoxyethyl 2-propenoate	3.94	C6H10O3					+
Methyl 2-methylbutanoate	4.31	C6H12O2	+		+	+	+
Methyl 2-oxopropanoate	4.54	C4H6O3	+			+	+
4H-1,2,4-triazole-3-carbonitrile	5.00	C3H2N4	+		+
Methyl pyrazine	5.00	C5H6N2		+		+	+
Propoxybenzene	5.17	C9H12O			+
2-methyl-1H-pyrrole	5.20	C5H7N					+
2,5-dimethyl-1H-pyrrole	5.36	C6H9N			+		+
Dimethylbenzene	5.59	C8H10	+			+
3-methoxy-2,2-dimethylcyclopropane carboxylic acid	5.82	C7H12O3					+
2-butoxyethanol	6.05	C6H14O2	+	+	+	+	+
Butyrolactone	6.16	C4H6O2	+		+	+	+
2,3-dimethylpyrazine	6.21	C6H8N2					+
2,3-dihydro-4H-pyran-4-one	6.26	C5H6O2			+	+	+
3-ethoxypentane	6.59	C7H16O	+	+	+	+	+
Benzaldehyde	6.69	C7H6O		+	+	+	+
2,6-dimethyloct-2-ene	6.76	C10H20	+
2,4-dihydroxy-2,5-dimethylfuran-3-one	6.87	C6H8O4		+	+		+
5-methylnon-4-ene	6.88	C10H20	+	+	+		+
2-propen-1-ol	6.95	C3H6O					+
1H-pyrazol-5-amine	7.02	C3H5N3		+	+	+	+
Decane	7.04	C10H22	+	+	+	+	+
2,3,5-trimethylpyrazine	7.11	C7H10N2		+	+		+
2-ethyl-1-hexanol	7.33	C8H18O					+
2,4-dioxohexahydro-1,3,5-triazine	7.39	C3H5N3O2					+
1-phenyl-1,2-propanediol	7.41	C9H12O2			+		+
Benzeneacetaldehyde	7.51	C8H8O					+
1-(3-methyl-2H-pyrazol-4-yl)ethanone	7.53	C6H8N2O					+
2,4,5-trihydroxypyrimidine	7.56	C4H4N2O3				+	+
6,6-dimethylundecane	7.65	C13H28					+
Pyrrolidin-2-one	7.69	C4H7NO		+	+	+	+
prop-2-yn-1-yl heptan-2-ylcarbamate	7.77	C11H19NO2					+
3-hydroxy-4-methylbenzaldehyde	7.82	C8H8O2			+	+	+
1-Azabicyclo[3.1.0]hexane	7.86	C5H9N	+	+	+	+	+
2,6-diethylpyrazine	7.89	C8H12N2					+
Methoxyphenol	7.93	C7H8O2	+			+
Undecane	7.96	C11H24	+		+	+
2-methyl-6-(1-propenyl)pyrazine	8.00	C8H10N2					+
2,3-butanedione monoxime	8.04	C4H7NO2					+
1-phenylethanol	8.13	C8H10O					+
2,3-dihydro-1H-Isoindole	8.18	C8H9N					+
3,5-dihydroxy-6-methyl-2,3-dihydropyran-4-one	8.38	C6H8O4		+	+	+	+
Butanoic acid	8.48	C4H8O2	+
4-vinyl-1H-imidazole	8.55	C5H6N2					+
1H-Tetrazole	8.70	CH2N4			+		+
Benzene-1,2-diol (catechol)	8.74	C6H6O2	+				+
2,3-dihydro-1-benzofuran	8.91	C8H8O		+	+	+	+
Oxan-3-one	9.01	C5H8O2	+
Pyrrolidine	9.08	C4H9N	+			+	+
1-(1-butoxypropan-2-yloxy)propan-2-ol	9.11	C10H22O3	+	+			+
2-(2-hydroxypropoxy)propan-1-ol	9.11	C6H14O3		+	+	+	+
ethyl 2-amino-4-methylpentanoate	9.22	C8H17NO2					+
3-ethenyl-4-methylpyrrole-2,5-dione	9.23	C7H7NO2			+		+
2-[(2-cyanoacetyl)oxy]ethyl 2-methylprop-2-enoate	9.56	C9H11NO4			+	+	+
m-aminophenylacetylene	9.57	C8H7N			+		+
2,5-diaminopentanoic acid (ornithine)	9.62	C5H12N2O2	+	+	+	+	+
2-methoxy-4-vinylphenol	9.69	C9H10O2				+	+
3-aminopiperidine-2,6-dione	9.71	C5H8N2O2	+	+	+	+	+
1H-Imidazole-2-carboxaldehyde	9.83	C4H4N2O					+
3,5-dimethyl-1H-pyrazole-4-carbaldehyde	9.96	C6H8N2O				+
Methyl 5-oxo-prolinate	10.07	C6H9NO3			+		+
Piperidine-2,6-dione (glutarimide)	10.18	C5H7NO2					+
Pyridine-4-carboxamide (Isonicotinamide)	10.20	C6H6N2O		+		+	+
1-ethyl-1H-indole	10.23	C10H11N					+
4-(dimethylamino)benzonitrile	10.23	C9H10N2				+
6-acetamido-N-acetyl-2-amino-N-(naphthalen-2-yl)hexanamide	10.31	C20H25N3O3					+
Isoindole-1,3-dione (phthalimide)	10.69	C8H5NO2			+	+
N,N’-(2-hydroxytrimethylene)diphthalimide	10.70	C19H14N2O5	+
Ethyl tridec-2-yn-1-yl terephthalate	10.87	C23H32O4					+
3,5-dihydroxybenzaldehyde	10.93	C7H6O3					+
3-Hydroxybenzaldehyde oxime	10.93	C7H7NO2					+
(2,4-ditert-butylphenyl) 5-hydroxypentanoate	10.97	C19H30O3	+	+		+	+
1-(2-hydroxy-4,5-dimethylphenyl)ethanone	11.09	C10H12O2		+			+
1H-pyrrole-2-carboxamide	11.10	C5H6N2O			+
ethyl 2-amino(N-dimethylaminomethylene)-3-phenylpropanoate	11.39	C14H20N2O2	+
2-methylnaphthalen-1-amine	11.46	C11H11N					+
2,2,4-trimethylpentane-1,3-diyl bis(2-methylpropanoate)	11.52	C16H30O4					+
2-methyl-4-pyridinamine 1-oxide	11.53	C6H8N2O			+		+
3-acetyl-4-hydroxy-6-methyl-2-pyridone	11.58	C8H9NO3					+
1-(1-piperidino)cyclohexene	11.71	C11H19N					+
7H-Purin-6-amine (adenine)	12.30	C5H5N5	+	+	+	+	+
2-anilino-N,N-dimethylacetamide	12.31	C10H14N2O			+
2,5-diallyldecahydroquinoline	12.31	C15H25N					+
2-(1H-Indol-3-yl)ethanamine (tryptamine)	12.41	C10H12N2			+
nonyl 2-((methoxycarbonyl)amino)pentanoate	12.42	C16H31NO4					+
bis(2-formylphenyl) 2,2′-oxydiacetate	12.74	C18H14O7					+
7,11,15-trimethyl-3-methylidenehexadec-1-ene (neophytadiene)	12.82	C20H38	+	+	+		+
3,7,11,15-tetramethylhexadec-2-ene	12.85	C20H40	+				+
1,1′-methylenediazetidine	12.85	C7H14N2			+
cyclohexyl (4-methylpentyl)phthalate	13.02	C18H24O4				+
1-O-(2-methylpropyl) 2-O-octan-4-yl benzene-1,2-dicarboxylate	13.03	C20H30O4	+
butyl pentan-2-yl phthalate	13.03	C17H24O4					+
Methyl hexadecanoate	13.25	C17H34O2	+			+	+
7,9-ditert-butyl-1-oxaspiro[4.5]deca-6,9-diene-2,8-dione	13.32	C17H24O3		+			+
Hexadecanoic acid (palmitic acid)	13.41	C16H32O2	+	+	+		+
Dibutyl phthalate	13.50	C16H22O4					+
Methyl 2-methylhexadecanoate	13.59	C18H36O2					+
cyclohexane-1,4-dimethanol diacetate	14.13	C12H20O4			+	+	+
3,7,11,15-tetramethylhexadec-2-en-1-ol (phytol)	14.19	C20H40O	+	+	+	+	+
Methyl stearate	14.23	C19H38O2					+
Hexadecanamide	14.49	C16H33NO				+	+
Nonanamide	14.49	C9H19NO		+
Bis(2-ethylhexyl) 2-butenedioate	14.69	C20H36O4	+	+	+	+	+
5,7-dimethylpyrimido-[3,4-a]indole	14.79	C13H12N2				+	+
2-(dimethylamino)ethyl ethyl carbonate	14.94	C7H15NO3					+
Glycidyl tetradecanoate	15.00	C17H32O3	+		+	+
Octadec-9-enamide	15.31	C18H35NO	+	+	+	+	+
Heptadeca-1,8,11,14-tetraene	15.33	C17H28	+
Tricyclo[4.3.1.0(2,5)]decane	15.79	C10H16	+				+
2-hydroxy-1-(hydroxymethyl)ethyl hexadecanoate	15.89	C19H38O4	+			+	+
Octadec-17-en-14-yn-1-ol	16.70	C18H32O					+
2,3-Dihydroxypropyl octadecanoate	16.75	C21H42O4					+
2,6,10,15,19,23-hexamethyltetracosa-2,6,10,14,18,22-hexaene (supraene)	17.31	C30H50					+
2-(4-methylphenyl)indolizine	19.35	C15H13N					+
2,5,7,8-tetramethyl-2-(4,8,12-trimethyltridecyl)chroman-6-ol (α-tocopherol)	19.45	C29H50O2					+

‘+’ symbol represents presence of compounds in the samples.

lists 115 volatile compounds detected by GC-MS analysis during the ten-week germination period. Figure S1 shows GC-MS chromatograms of samples according to time intervals of collection (T1–T5). Table S1 shows the peak areas of all compounds in triplicate, together with a comparison of T1 with all other sample collection times (T2–T5) using Student’s t-test. In the same table, p values for intervals that are statistically significantly different are marked. Phytol, the common component of plant oils with application in the fragrance and production industries [31], can be seen as a component of vitamin E and K [32]. Substances such as esters can also serve as a good source of starting substances in the cosmetic industry [33], and they are also found in this stage of plant development. The nitrogen-based compound 1-Azabicyclo[3.1.0]hexane is known for its biological effects against bacteria, fungi, and tumors [34]; it and octadec-9-enamide were present in the all samples, while compound 2,5-diaminopentanoic acid (an intermediate in the urea cycle and the production of citrulline, proline and polyamines [35]), which belongs to the same class of molecules, was also present in all samples. Several compounds like 2,6-dimethyloct-2-ene, butanoic acid, oxan-3-one, N,N′-(2-hydroxytrimethylene)diphthalimide, ethyl 2-amino(N-dimethylaminomethylene)-3-phenylpropanoate, 1-O-(2-methylpropyl) 2-O-octan-4-yl benzene-1,2-dicarboxylate, and heptadeca-1,8,11,14-tetraene were present only in T1. Figure 1 shows the fluctuations in the concentrations of the compounds present in all samples.

Looking at T2, we can note the presence of 28 volatile compounds, one of which appears only in this stage: nonanamide. The analysis presented in Table 1 reveals the presence of various compounds in the sample, including a fatty acid (hexadecanoic acid), a hydrocarbon (decane), and ester derivatives such as (2,4-ditert-butylphenyl) 5-hydroxypentanoate. Additionally, nitrogen-based compounds like 1H-pyrazol-5-amine, and 2,5-diaminopentanoic acid suggest potential applications in the pharmaceutical and cosmetic industries due to their significant concentrations. Compounds such as 2,3-dihydro-1-benzofuran and benzaldehyde may contribute to flavor or fragrance production and could also have medicinal applications, owing to their aromatic properties.

The presence of 43 volatile compounds is identified in T3, encompassing a wide variety of compound classes. These include lactones (e.g., butyrolactone), fatty acids (e.g., hexadecenoic acid), amines (e.g., 1H-pyrazol-5-amine), and aromatic compounds such as 3-hydroxy-4-methylbenzaldehyde.

T4 reveals the presence of 41 compounds (Table 1), with 3 compounds present only in this sample: 3,5-dimethyl-1H-pyrazole-4-carbaldehyde, 4-(dimethylamino)benzonitrile, and cyclohexyl (4-methylpentyl)phthalate.

A total of 92 compounds were found in the last sample (T5), and 42 compounds were only found in this sample (Table 1). This sample contains a various class of compound like diterpenes, fatty acid derivatives, esters, aromatic compounds, and many others. It also contains some compounds with biological properties like α-tocopherol, commonly known as vitamin E, a powerful antioxidant that plays a key role in protecting cells from oxidative stress [36].

The five developmental stages (T1–T5) of Cucumis metuliferus show significant changes in the profile of volatile metabolites during germination. T1, among all the early samples, contains the highest number of stage-specific metabolites, such as 2,6-dimethyloct-2-ene and butanoic acid, but also universally present molecules like phytol and bioactive nitrogen compounds. T2 shows reduced diversity with a total of 28 volatiles, of which only 1, nonanamide, was stage-specific, though it contained key industrial fatty acids, hydrocarbons, and aromatic molecules. The chemical complexity increases anew in T3, where 43 volatiles are identified from lactones, fatty acids, amines, and aromatics, pointing out metabolic expansion in this stage. T4 presents 41 compounds, including 3 exclusive molecules not detected in any other stages, which would suggest a transitional shift in metabolic pathways. The most chemically diverse stage is T5, with 92 volatiles—42 unique—representing diterpenes, fatty acid derivatives, esters, and additional aromatic structures. Notably, T5 also includes α-tocopherol (vitamin E), which indicates an enhancement of antioxidant metabolism in later development. Overall, the comparison demonstrates progress from early-stage specificity to late-stage chemical diversification, reflecting dynamic metabolic reprogramming throughout germination.

It is important to mention that some compounds found in this study have also been reported previously: decane, undecane, octadec-9-enamide [21], palmitic acid, and 2,5,7,8-tetramethyl-2-(4,8,12-trimethyltridecyl)chroman-6-ol (α-tocopherol) [22].

3.2. LC-MS-Orbitrap Analysis of In Vitro Plant Tissue Culture of C. metuliferus

The results of the LC-MS-Orbitrap analysis of the chemical compounds provide significant information on the composition of different metabolites of these compounds in various samples. There are 10 hydroxybenzoic acids, 5 hydroxycinnamic acids, 8 flavonoid glycosides, and 7 flavonoid aglycones among the 30 compounds that were found and are shown in

Table 2. Compound names, retention times (t_R, min), molecular formulas, MS data (calculated and exact masses, as well as mean mass accuracy—Δ ppm), and major MS² fragment ions of the compounds present in the five samples of C. metuliferus in vitro cultures.

Compound Name	tR, min	Molecular Formula, [M—H]—	Calculated Mass, [M—H]—	Exact Mass, [M—H]—	Δ ppm	MS2 Fragments, (% Base Peak)
Hydroxybenzoic acids
Dihydroxybenzoic acid hexoside	0.66	C13H15O9—	315.07216	315.07278	−1.99	108.02177(49), 109.02966(34), 152.01178(100), 153.0195(50), 315.07251(35)
Gallic acid pentoside	0.68	C12H13O9—	301.05651	301.05723	−2.39	125.02458(43), 149.99619(5), 168.00658(100), 169.01462(10), 301.05676(82)
Dihydroxybenzoic acid pentosyl hexoside	0.72	C18H23O13—	447.11442	447.11485	−0.98	108.02205(8), 109.02945(13), 152.01176(100), 153.01964(5), 315.07217(5), 447.11554(67)
Vanillic acid hexoside	0.80	C14H17O9—	329.08781	329.08829	−1.48	167.03508(100)
Dihydroxybenzoic acid pentoside	0.85	C12H13O8—	285.06159	285.06199	−1.41	109.02968(26), 152.01163(19), 153.01958(100), 285.06195(32)
Dihydroxybenzoic acid pentosyl pentoside	1.14	C17H21O12—	417.10385	417.10411	−0.63	108.02173(8), 109.02964(18), 152.01166(100), 241.07196(13), 417.10410(73)
Hydroxybenzoic acid hexoside	1.17	C13H15O8—	299.07724	299.07760	−1.21	93.03463(23), 137.02451(100)
Dihydroxybenzoic acid	2.68	C7H5O4—	153.01933	153.01951	−1.14	109.02969(78), 153.01941(100)
Vanillic acid	4.30	C8H7O4—	167.03498	167.03516	−1.06	108.02194(46), 123.04408(5), 152.01152(28), 167.03532(100)
Hydroxybenzoic acid	5.45	C7H5O3—	137.02442	137.02455	−0.99	93.03468(100), 137.02461(51)
Hydroxycinnamic acids
Ferulic acid hexoside	4.05	C16H19O9—	355.10346	355.10350	−0.14	134.03749(27), 149.06081(22), 175.04056(67), 191.07166(45), 193.05077(100), 235.06145(68)
Coumaric acid	4.15	C9H7O3—	163.04007	163.04024	−1.07	119.05038(100), 163.04027(11)
Ferulic acid	5.06	C10H9O4—	193.05063	193.05079	−0.81	134.03755(100), 149.06088(9), 178.02788(15), 193.05122(8)
Caffeic acid	5.69	C9H7O4—	179.03498	179.03519	−1.19	135.00087(100), 179.03587(21)
Methoxycinnamic acid	6.45	C10H9O3—	177.05572	177.05590	−1.03	145.02975(33), 162.03249(10), 177.05592(100)
Flavonoid glycosides
Isoorientin 2″-O-hexoside	4.49	C27H29O16—	609.14611	609.14663	−0.85	298.04852(74), 309.04111(69), 327.05148(48), 339.05124(24), 357.06216(32), 369.06125(13), 429.08282(40), 489.10437(100), 609.14557(11)
Isovitexin 2″-O-hexoside-7-O-hexoside	4.87	C33H39O20—	755.20402	755.20477	−1.00	293.04572(100), 311.05673(17), 323.05719(7), 341.06674(15), 413.08716(44), 635.16211(6)
Isovitexin 2″-O-hexoside	4.90	C27H29O15—	593.15119	593.15175	−0.94	293.04587(100), 311.05649(10), 323.05624(7), 341.06699(7), 413.08817(34), 473.10788(3)
Isovitexin 2″-O-hexoside-7-O-pentoside	4.92	C32H37O19—	725.19345	725.19370	−0.34	293.04623(100), 311.05737(22), 323.05554(14), 341.06769(18), 413.08798(42)
Apigenin 6-C-hexoside (Isovitexin)	5.08	C21H19O10—	431.09837	431.09864	−0.63	283.06143(15), 311.05634(100), 323.05588(7), 341.06659(39), 353.06714(4)
Kaempferol 3,7-di-O-rhamnoside	5.13	C27H29O14—	577.15628	577.15711	−1.43	283.02441(35), 284.03177(6), 285.04074(100), 430.09116(30), 431.09875(24)
Kaempferol 3-O-hexoside	5.46	C21H19O11—	447.09329	447.09370	−0.92	227.03535(4), 255.03033(11), 284.03290(100), 285.04065(27), 327.05139(2), 447.09360(18)
Apigenin 7-O-(6”-pentosyl)-hexoside	5.52	C26H27O14—	563.14063	563.14169	−1.88	269.04575(100), 563.14227(4)
Flavonoid aglycones
Patuletin	6.32	C16H11O8—	331.04594	331.04676	−2.49	165.99101(21), 181.01465(11), 287.01797(15), 316.02267(100), 331.04724(14)
Apigenin	6.72	C15H9O5—	269.04555	269.04585	−1.14	151.00407(3), 225.05812(2), 269.04584(100)
Naringenin	6.72	C15H11O5—	271.06120	271.06156	−1.33	107.01396(13), 119.05036(42), 151.00381(100), 177.01939(10), 271.06140(47)
Chrysoeriol	6.76	C16H11O6—	299.05611	299.05667	−1.87	284.03284(100), 285.03745(4)
Luteolin	6.82	C15H9O6—	285.04046	285.04086	−1.38	285.04089(100)
Cirsimaritin	7.39	C17H13O6—	313.07176	313.07209	−1.04	283.02496(100), 297.04062(16), 298.04840(80), 313.07242(23)
Biochanin	7.91	C16H11O5—	283.06120	283.06175	−1.96	268.03793(91), 283.06165(100)

, as well as their abundances in all 5 samples compared in Figure 2. Table S2 shows the obtained peak areas of all detected compounds in triplicate, along with a comparison of T1 with all other sample collection times (T2-T5) using Student’s t-test. In the same table, p values for intervals that are statistically significantly different are marked.

If we compare the mean values (Table S2) of all compounds, we can say that T3 shows the highest concentrations of hydroxybenzoic acids, especially the dihydroxybenzoic acid derivatives, which are the most abundant. Particularly significant are the high levels of dihydroxybenzoic acid pentosyl hexoside and dihydroxybenzoic acid pentosyl pentoside in T1; a similar but less pronounced signature is also noted in T4 and T5 in the form of vanillic acid and vanillic acid hexoside, respectively.

In contrast, T1 contains the highest amount of coumaric acid. However, as an example, T4 has the highest levels of methoxycinnamic acid, ferulic acid, caffeic acid and ferulic acid hexoside, so the substances are likely to occur with the highest levels in T4. Biological/environmental variables that influence the concentration of these compounds in various samples may be reflected by these changes in concentration profiles.

The distribution of flavonoid glycosides is novel, since T5 contains the highest amount of seven flavonoid glycosides (isoorientin 2′-O-hexoside, isovitexin 2′-O-hexoside-7-O-hexoside, isovitexin 2′-O-hexoside, isovitexin 2′-O-hexoside-7-O-pentoside, apigenin 6-C-hexoside, kaempferol 3-O-hexoside and apigenin 7-O-(6″-pentosyl)hexoside), which may be responsible for the particular chemical profile of this sample. The T4 contents of some flavonoids i.e., kaempferol 3,7-di-O-rhamnoside hexoside, increased. The variability in levels of flavonoid glycosides across samples leads us to the potential conclusion that this is based on sample type; they may be involved in distinct functions that contribute to the biochemical diversity observed.

On the other hand, the concentrations of flavonoid aglycones are relatively constant in all samples except for the second, in which they are lower. Apigenin and naringenin, compounds with antifungal activity against C. albicans and C. parapsilosis [37,38], are highest in T1, and aglycone compounds patuletin and luteolin are highest in T4. T3 contains chrysoeriol, while T5 has the highest concentrations of the other flavonoid aglycones, biochanin and cirsimaritin.

The LC-MS-Orbitrap analysis identified 30 metabolites in 5 samples, which included hydroxybenzoic acids, hydroxycinnamic acids, flavonoid glycosides, and flavonoid aglycones. Obtained chromatograms from LC-MS analysis are presented in Figure S2. Thus, T3 had the higher amount of hydroxybenzoic acids, and T1 was rich in coumaric acid. The maximum flavonoid glycosides were in T5, contributing to its unique chemical profile, while T4 had higher levels of some flavonoids such as its kaempferol derivatives. The flavonoid aglycones remained stable, except for marked variations in T1, T3, T4, and T5.

These differences reflect stage-specific metabolic patterns and suggest potential biological and pharmacological activities, highlighting the importance of sample-specific profiling. In addition, the various concentrations of flavonoids, hydroxybenzoic acid, hydroxycinnamic acid, and other metabolites also contain evidence of possible biological and pharmacological activities of the samples. These findings emphasize the importance of sample-specific profiles and pave the way for further studies into the functional properties of these compounds.

It is important to mention that some compounds found in this study have also been reported previously: coumaric acid, caffeic acid, ferulic acid, kaempferol-3-glucoside [17], and isovitexin 2″-O-glucoside [18].

3.3. Molecular Modeling

In the case of the first tested protein, a cellular receptor in complex with FK506 from Candida glabrata with the six compounds and 8-deethyl-8-[but-3-enyl]-ascomycin as the control, the best result was for 3,7,11,15-tetramethylhexadec-2-en-1-ol, with a binding affinity energy of −6.5 kcal/mol (

Table 3. Docking score values of the compounds to FKBP12.

Compounds	Affinity (kcal/mol)
3,7,11,15-tetramethylhexadec-2-en-1-ol (phytol))	−6.5
2,5-diaminopentanoic acid (ornithine)	−4.3
7H-Purin-6-amine (adenine)	−4.6
1-Azabicyclo[3.1.0]hexane	−3.8
Octadec-9-enamide	−6.2
8-deethyl-8-[but-3-enyl]-ascomycin (control)	−10.5

). This compound formed ten alkyl and pi-alkyl bounds and one hydrogen bond in the site where the known inhibitor 8-deethyl-8-[but-3-enyl]-ascomycin binds (Figure 3). Compared to the control molecule, this molecule did not bind better than the inhibitor (control), which has a binding affinity of −10.5 kcal/mol.

Furthermore, compound 7H-Purin-6-amine, also known as adenine, had better results in binding affinity compared to the enzyme’s known inhibitor (

Table 4. Docking score values of the compounds to phosphomannose isomerase.

Compounds	Affinity (kcal/mol)
3,7,11,15-tetramethylhexadec-2-en-1-ol (phytol))	−4.6
2,5-diaminopentanoic acid (ornithine)	−5.2
7H-Purin-6-amine (adenine)	−5.9
1-Azabicyclo[3.1.0]hexane	−3.8
Octadec-9-enamide	−4.4
Inhibitor (control) *	−5.2

* [(2~{R},3~{R},4~{S})-5-diazanyl-2,3,4-tris(oxidanyl)-5-oxidanylidene-pentyl] dihydrogen phosphate.

). Adenine bound to the enzyme active site via five hydrogen bonds (Tyr16, Asp300, Lys 310, Arg304, and Gln111) and two pi-anion and pi-alkyl bounds, with Val302 and 48Glu presenting a binding affinity of −5.9 kcal/mol (Figure 4). In this case, one compound among all tested compounds gave better results compared to the control compound.

The last enzyme, CYP51, was tested with the six compounds and with the inhibitor. One compound, 3,7,11,15-tetramethylhexadec-2-en-1-ol, had better binding affinity compared to other tested compounds, with a binding energy of −8.0 kcal/mol (

Table 5. Docking score values of the compounds to sterol 14-alpha demethylase.

Compounds	Affinity (kcal/mol)
3,7,11,15-tetramethylhexadec-2-en-1-ol (phytol))	−8.0
2,5-diaminopentanoic acid (ornithine)	−4.8
7H-Purin-6-amine (adenine)	−5.5
1-Azabicyclo[3.1.0]hexane	−3.7
Octadec-9-enamide	−7.2
Inhibitor (control) *	−12.6

* (R)-2-(2,4-Difluorophenyl)-1,1-difluoro-3-(1H-tetrazol-1-yl)-1-(5-(4-(2,2,2-trifluoroethoxy)phenyl)pyridin-2-yl)propan-2-ol.

). This compound bound to the active site of the enzyme via two hydrogen bonds (Ser336 and His335) and six alkyl and pi-alkyl bonds (Phe191, Phe338, Leu79, Tyr76, Leu334, and HEM) (Figure 5).

Molecular docking analyses of the dominant compounds identified from C. metuliferus tissue culture revealed moderate to strong binding affinities toward key Candida spp. enzymes and receptors. Among the tested molecules, phytol exhibited the highest interaction energy with FKBP12 (−6.5 kcal/mol) and CYP51 (−8.0 kcal/mol), forming multiple hydrophobic and hydrogen interactions within the active sites. Adenine demonstrated notable affinity toward phosphomannose isomerase (−5.9 kcal/mol), surpassing the known inhibitor. These results suggest that the selected metabolites possess potential antifungal relevance, particularly through interactions with targets involved in Candida virulence and ergosterol biosynthesis.

These docking results are intended as a first step toward more detailed computational and experimental studies in the future.

4. Conclusions

This study revealed dynamic changes in the abundance and diversity of metabolites over developmental stages, identifying 115 volatile and 30 non-volatile compounds in Cucumis metuliferus tissue cultures over a ten-week period. Furthermore, GC-MS analysis was better in detecting small volatile compounds, compared to compounds detected using LC-HRMS-MS. The presence of dihydroxybenzoic acid derivatives, flavonoid glycosides, and fatty acid derivatives in specific developmental stages suggests that precise harvesting times could be optimized to maximize yields of target compounds, and the observed fluctuations in bioactive metabolites, such as phytol, octadec-9-enamide, and α-tocopherol, highlight the potential of kiwano tissue cultures as a sustainable source of antioxidants, antimicrobials, and other therapeutic agents. The data also suggests that developmental factors significantly influence metabolite profiles.

The docking studies show that certain metabolites, including phytol and adenine, have a binding affinity toward Candida spp. enzymes, hence supporting their antifungal potential. Overall, the results presented here establish C. metuliferus in vitro cultures as a sustainable source of bioactive compounds with potential pharmaceutical, nutraceutical, and cosmetic applications, providing a basis for future biological testing and structure–activity relationship studies.