From bioactive compounds to volatile profiles: a multidimensional characterization of Indonesian stingless bee honeys

Abstract

BACKGROUND

Stingless bee honeys are drawing increasing attention as ingredients for functional foods and health‐oriented products because of their distinctive sensory characteristics and bioactive potential. In this study, honeys collected from nine stingless bee species reared in West Sumatra, Indonesia, were comprehensively characterized using physicochemical indices, antioxidant assays [DPPH (i.e. 2,2‐diphenyl‐1‐picrylhydrazyl) and ferric reducing antioxidant power], microbiological screening, volatile profiling [gas chromatography‐mass spectrometry (GC‐MS)] and Fourier transform infrared (FTIR) fingerprinting.

RESULTS

Marked between‐sample variability was observed across key quality attributes, including pH (2.80–3.68), Brix (49.83–61.25), viscosity (23.36–175.22 cP) and color parameters. FTIR spectra were consistent with carbohydrate‐rich matrices and exhibited carbonyl‐related bands. GC‐MS profiling identified linalool oxide isomers and junenol among the predominant volatiles. To the best of our knowledge, junenol has not previously been reported in stingless bee honey and may represent a potential regional chemical marker for Indonesian stingless bee honeys. Lactic acid bacteria were detected in selected samples, whereas yeast and mold were not detected. Antioxidant activities were comparatively low, which may reflect local environmental and ecosystem‐related factors.

CONCLUSION

The results provide a multi‐parameter baseline for stingless bee honeys produced within a shared ecosystem in West Sumatra and demonstrate the value of integrating conventional chemical indices with FTIR and volatile fingerprints for quality assessment. This combined approach may also support future authentication and origin‐tracing frameworks for Indonesian stingless bee honeys. © 2026 Society of Chemical Industry.

INTRODUCTION

Indonesia is one of the most biodiverse countries in Southeast Asia, with remarkable richness in both flora and fauna. Bees represent an important component of Indonesia's faunal biodiversity and are broadly classified into two main groups: stingless bees (Meliponini) and honeybees (Apis spp.). Stingless bees are generally smaller than honeybees and possess distinct biological and behavioral characteristics. In Indonesia, there are 46 species of stingless bees out of approximately 500 species found worldwide. ¹ The advantages of stingless bees include their ability to be cultivated around home gardens, their resistance to diseases, and their non‐threatening nature to humans. Furthermore, stingless bees are known to have various health benefits. Several studies have shown that the products produced by stingless bees exhibit anti‐inflammatory properties, ² antioxidant and antibacterial effects, ³ and anticancer and anti‐asthma properties. ⁴ The products derived from stingless bees are increasingly being utilized as health supplements, cosmetics, antioxidants, wound healing remedies, and raw materials in the food and beverage industry. ⁵

One of the areas currently focusing on developing stingless bee farming is Sijunjung Regency, West Sumatra Province, Indonesia. Based on observations, there are nine species of stingless bees being cultivated in this region: Geniotrigona thoracica, Heterotrigona itama, Homotrigona fimbriata, Lophotrigona canifrons, Tetrigona binghami, Tetrigona apicalis, Tetragonula minangkabau, Tetragonula testaceitarsis and Tetragonula laeviceps. Each species has distinct characteristics, including size, body color and honey productivity.

Stingless bees produce various economically valuable products, such as honey, pollen, propolis and geopropolis.^{6, 7} Honey is the primary product harvested and has significant market value in the stingless bee production chain. ⁸ Honey from stingless bees is known for its more liquid texture compared to honey from honeybees, as a result of its higher water content, and it has a sweet–sour taste. ⁹ This characteristic is attributed to the high levels of organic acids in stingless bee honey. Several factors influence the quality of honey, including the species of bees, the type of flowers visited, climate, weather and harvest time. These factors can result in variations in nutritional composition, antioxidant content and the microbiological profile of the honey. Based on these considerations, further research is warranted to elucidate how stingless bee species reared in Sijunjung, Indonesia, influence the physicochemical characteristics, bioactive composition, microbiological attributes and volatile profiles of the honeys they produce, as well as to determine whether species‐specific or regional markers may support future quality assessment and authentication.

MATERIALS AND METHODS

Honey sample

The honey sampling was conducted in Lubuak Tarok Village, Sijunjung district, West Sumatra Province, Indonesia (0.8540°S, 101.0178°E, elevation 267.7 m) on 15 September 2024, with a temperature of 26 °C and humidity of 90%. The honey sampling locations are shown in Fig. 1, which visually represents the sampling points. Samples were taken from nine species of stingless bees, namely: H1:T. minangkabau, H2:T. binghami, H3:H. itama, H4:H. fimbriata, H5: T. apicalis, H6: T. testaceitarsis, H7:G. thoracica, H8:T. laeviceps and H9:L. canifrons. Around the bee location, there are 27 types of plants used by bees as nectar sources, including: bridal tears (Antigonon leptopus), santos lemon (Xanthostemon chrysanthus), calliandra (Calliandra calothyrsus), dombeya (Dombeya rotundifolia), cacao (Theobroma cacao L.), creeping foxglove (Asystasia gangetica ssp. micrantha), eucalyptus (Melaleuca cajuputi), mango (Mangifera indica), candlenut (Aleurites moluccana), rubber (Hevea brasiliensis), jackfruit (Artocarpus heterophyllus), banana (Musa paradisiaca), jengkol (Pithecellobium jiringa), petai (Parkia speciosa), langsat (Lansium domesticum), coffee (Coffea arabica), clove (Syzygium aromaticum L.), passion fruit (Passiflora edulis var. flavicarpa), starfruit (Averrhoa bilimbi L.), mangosteen (Garcinia mangostana), rambutan (Nephelium lappaceum L.), avocado (Persea americana), durian (Durio zibethinus), sapodilla (Manilkara zapota), orange jasmine (Murraya paniculata), matoa (Pometia pinnata) and soursop (Annona muricata).

Figure 1.

Map of research location stingless bee, Sijunjung regency, West Sumatra, Indonesia. RBI map and Google satellite imagery.

Honey harvesting was performed using complete protective beekeeping equipment, and the honey was collected with a BE 08 honey pump. Immediately after extraction, the honey was filtered through a nylon cloth with a density of 120 mesh, which effectively removed coarse impurities such as propolis fragments, comb particles, and wood residues introduced during harvesting. The filtered honey was transferred into sterile 100‐mL polypropylene vials, tightly sealed and reinforced with duct tape to prevent leakage during transport.

All samples were harvested on the same day and transported to Turkey under ambient conditions, requiring approximately 2 days. Upon arrival, the samples were immediately stored at 4 °C for 5 days before analysis, resulting in a total holding period of 7 days from harvesting to laboratory testing.

For sample extraction, two solvent systems were used: distilled water and 80:20 (v/v) methanol:water. One gram of honey was mixed with 10 mL of each solvent in separate tubes, vortexed for 1 min, and incubated in a shaker (SI‐600; LAB Companion, Korea) at 100 rpm and 30 °C for 30 min. The solutions were then filtered through a 0.45‐μm syringe filter and the filtrates were stored at 4 °C until further analysis.

Physicochemical, bioactivity and microbiological analyses

Brix content

A hand refractometer was used to determine the total soluble solids in the honey samples. A small amount of each honey sample was applied to the refractometer's glass prism, and the reading was recorded at room temperature. The results were expressed in °Brix.

pH

The pH test was conducted by dissolving 1 g of honey in 10 mL of distilled water. The pH was then measured using a pH meter (Orion star A211; Thermo Fisher Scientific, Waltham, MA, USA). Before the measurement, the pH meter was calibrated at pH 4 and 7. After each measurement, the device was cleaned to ensure accurate results.

Viscosity

Honey viscosity values were determined using a rheometer (DVNext Rheometer; AMETEK Brookfield Middleboro, MA, USA). The samples were sheared using a LV‐3C spindle and the speed was 10 rpm. The samples were stored in a jar at room temperature, and the sample temperatures were measured as 25 ± 1 °C. All measurements were three repetitions.

Color analysis

Color analysis was performed using a Precision Colorimeter NR110 (3NH Technology Co., Ltd, Shenzhen, China). The instrument was calibrated using black and white reference surfaces before measurement. The colorimeter was then positioned vertically and directly above the honey sample for testing. Each measurement was repeated five times to ensure accuracy and representativeness. Four key parameters were recorded: L* (lightness), indicating the brightness; a* (redness), representing the chromaticity along the green–red axis; b* (yellowness), indicating the chromaticity along the blue–yellow axis; and C* (chroma), which measures the strength or saturation of the color.

Electrical conductivity (EC)

EC was measured according to the International Honey Commission method. The moisture content of each honey sample was first determined to calculate its dry matter fraction. A 20:80 (w/v) honey:water solution was then prepared by adjusting the mass of honey according to its dry matter content and diluting the sample with distilled water to a final volume of 100 mL. Electrical conductivity was recorded at 20 °C using a EC M170 Bench meter (Milwaukee Instruments, Inc., Rocky Mount, NC, USA).

Hydroxymethylfurfural (HMF)

HMF contents of the honey samples were determined by using HPLC (Shimadzu, Kyoto, Japan). One g of honey was mixed with 9 mL of distilled water and vortexed for 1 min. Then the samples were centrifuged at 7500 × g for 5 min at 4 °C (Hettich, Kirchlengern, Germany). After that, the supernatant was filtered using a 0.45‐μm syringe filter and injected to HPLC equipped with a C18 column (5.0 μm 4.4 × 250 mm) and a photodiode array detector set at 284 nm. The injection volume was 20 μL and the column temperature was set as 35 °C. The mobile phase was distilled water including 1 g per 100 g acetic acid and acetonitrile (95:5 v/v). To determine the HMF of the samples, a calibration curve prepared by HMF standard (Sigma‐Aldrich, St Louis, MO, USA) was used to calculate the HMF level of the honey samples. All measurements were performed with two repetitions.

Total phenolic content (TPC)

The analysis was performed using the Folin–Ciocalteu method with minor modifications based on the method proposed by Singleton and Rossi. ¹⁰ A 200‐μL sample was placed in a test tube, followed by the addition of 1800 μL of distilled water. Next, 1 mL of a 1:10 diluted Folin reagent (Merck, Darmstadt, Germany) was added. Then, 2 mL of a 2 g per 100 g Na₂CO₃ (Merck) solution was added to the mixture and thoroughly mixed. The mixture was vortexed thoroughly and incubated in the dark at room temperature for 2 h. After incubation, the absorbance of the solution was measured at 765 nm using a C‐7100 spectrophotometer (Peak Instruments, Houston, TX, USA). The TPC of the samples was determined using a gallic acid standard curve prepared under the same conditions and expressed as mg gallic acid equivalents (GAE) g^–1. The measurement was repeated three times for replication.

Total flavonoid content (TFC)

In this study, the TFC of stingless bee honey was determined according to a previously described method. ¹¹ First, a 0.5‐mL honey sample was taken and diluted with 2 mL of distilled water. To this, 150 μL of 5% sodium nitrite (NaNO₂; Merck) was added, and the mixture was allowed to stand for 5 min. Subsequently, 150 μL of 10% aluminum chloride hexahydrate (AlCl₃·6H₂O; Merck) was added, followed by a 6‐min incubation period. Afterward, 1 mL of 1 m sodium hydroxide (NaOH; Merck) and 1.2 mL of distilled water were added. The mixture was vortexed to ensure homogeneity, and the absorbance was measured at 510 nm using a C‐7100 spectrophotometer (Peak Instruments). The results were expressed in terms of rutin hydrate equivalents (Sigma‐Aldrich) as mg retinol equivalents (RE) g^–1. The measurement was performed in triplicate.

2,2‐Diphenyl‐1‐picrylhydrazyl (DPPH) radical scavenging activity

The antioxidant activity of the sample was determined using the DPPH radical scavenging assay according to a previously described method. ¹² A volume of 100 μL of the sample was taken, and 3900 μL of a 0.2 mm DPPH radical solution (Merck) was added. The 0.2 mm DPPH solution was prepared by dissolving 3.96 mg of DPPH radical in 100 mL of methanol. The mixture was vortexed and incubated in the dark at room temperature for 30 min. The absorbance was measured at 517 nm using a C‐7100 spectrophotometer (Peak Instruments), which was calibrated with methanol.

Ferric reducing antioxidant power (FRAP)

The FRAP assay was used to assess the antioxidant activity of the samples. Reagents were prepared as follows: a 300 mm acetate buffer (pH 3.6) was made by dissolving 3.1 g of sodium acetate in distilled water, adding 16 mL of acetic acid, and adjusting the volume to 1000 mL with distilled water. A 10 mm 2,4,6‐Tris(2‐pyridyl)‐S‐triazine (TPTZ) solution was prepared by dissolving 0.031 g of TPTZ in 10 mL of 40 mm HCl, and a 20 mm FeCl₃·6H₂O solution was made by dissolving 0.054 g of FeCl₃·6H₂O in 10 mL of distilled water. The FRAP working solution was prepared by mixing 25 mL of acetate buffer (Merck), 2.5 mL of TPTZ solution (Merck) and 2.5 mL of FeCl₃·6H₂O solution (Merck), and was kept protected from light. For the assay, 0.1 mL of sample extract was added to 2.9 mL of the working solution and incubated for 30 min at room temperature in the dark. Absorbance was measured at 593 nm using a C‐7100 spectrophotometer (Peak Instruments), with distilled water as the blank. A calibration curve was constructed using ascorbic acid.

LAB, mold and yeast enumeration

A one‐gram sample was taken from the honeys brought to the laboratory under sterile conditions and homogenized with 9 mL of Maximum Recovery Diluent (Merck) solution for 5 min. Detailed dilutions up to 10⁻⁵ were then prepared. For the enumeration of lactic acid bacteria (LAB) from the prepared dilutions, samples were incubated in an anaerobic environment at 30 °C on De Man–Rogosa–Sharpe agar (Merck). For mold and yeast enumeration, samples were incubated in an aerobic environment at 25 °C on Dichloran Rose‐Bengal Chloramphenicol agar (Merck).

Fourier transform infrared (FTIR) analysis

The IR spectrum of honey samples was monitored using a 400 FT‐IR Spectrometer Spotlight 400 Imaging System (PerkinElmer, Springfield, IL, USA) with the wavenumbers ranging from 400 to 4000 cm⁻¹.

Volatile composition using a gas chromatography‐mass spectrometry (GC‐MS) system

Volatile components were identified using GC‐MS (QP2010 SE; Shimadzu). The membrane of the screw cap was pierced with a divinylbenzene/carboxen/polydimethylsiloxane fiber needle, and honey (2.5 g) was exposed to the gas phase in the headspace. After exposure for 50 min at room temperature (20 ± 1 °C), the fiber was placed in the injection port of the gas chromatograph for 10 min. The device was equipped with an HP 5‐ms capillary column (30 m × 0.25 mm inner diameter, 0.25 μm film thickness) (Agilent Technologies, Santa Clara, CA, USA) with electronic pressure control and a split/splitless injector and operated in splitless mode at 220 °C. The helium flow rate through the column was 1 mL min^–1 in constant flow mode, the initial column temperature was 40 °C, and it was increased to 180 °C at a rate of 3 °C min⁻¹. Microstrip silicon detector acquisition parameters were set as follows: transfer line temperature 280 °C, MS source temperature 230 °C, and MS quad temperature 150 °C. Electron‐induced mass spectra were obtained at 70 eV ionization energy. Detection was performed in full scan mode. After integration, the fraction of separated components in the total ion current was calculated. The W10N14(https://www.sisweb.com/software/wiley-ffnsc.htm) and FFNSC(https://www.agilent.com/en/product/gas‐chromatography‐mass‐spectrometry‐gc‐ms/gc‐ms‐application‐solutions/gc‐ms‐libraries) libraries were used to identify volatile components.

Statistical analysis

The collected data were averaged, and the standard deviation was calculated. Subsequently, analysis of variance was performed at a significance level of P < 0.05. If a significant effect was observed on the measured variables, further analysis was conducted using Duncan's multiple range test (DMRT). All statistical analyses were carried out using SPSS, version 22 (IBM Corp., Armonk, NY, USA). Principal component analysis (PCA) techniques were used to classify the samples according to the physicochemical and bioactive parameters. PCA was performed using XLSTAT Software (Addinsoft, New York, NY, USA).

RESULTS AND DISCUSSION

Physical analyses

The physical analyses (pH, °Brix and viscosity) of the honeys produced by the nine stingless bee species from Sijunjung, Indonesia are presented in Table 1. The color values are presented in Table 2.

Table 1.

Results of the physical analyses of stingless bee honeys in Sijunjung, Indonesia

	Physical analyses
Honey	pH	°Brix	Viscosity (cP)
H1	2.92 ± 0.01 d	55.63 ± 0.02 c	69.46 ± 0.09 d
H2	2.80 ± 0.01 a	56.31 ± 0.02 d	30.62 ± 0.19 b
H3	2.96 ± 0.01 e	59.53 ± 0.02 g	72.95 ± 0.45 e
H4	2.91 ± 0.01 d	54.55 ± 0.05 b	73.25 ± 0.04 e
H5	2.85 ± 0.01 c	49.83 ± 0.03 a	23.36 ± 0.02 a
H6	3.01 ± 0.01 f	59.08 ± 0.01 f	50.20 ± 0.36 c
H7	2.83 ± 0.01 b	56.67 ± 0.01 e	50.45 ± 0.71 c
H8	3.31 ± 0.01 g	60.95 ± 0.02 h	175.20 ± 0.17 g
H9	3.68 ± 0.00 h	61.25 ± 0.03 i	172.80 ± 0.67 f
Mean ± SD	3.03 ± 0.01	57.09 ± 0.02	79.81 ± 0.30
Minimum value	2.80	49.83	23.36
Maximum value	3.68	61.25	175.2

Values with different lowercase letters in the same column are significantly different (P < 0.05) according to DMRT.

Table 2.

Color parameters of stingless bee honeys from Sijunjung, Indonesia

Honey	Color
	L*	a*	b*	C*
H1	56.74 ± 6.19 c	3.11 ± 0.68 b	18.97 ± 2.29 abc	19.03 ± 1.92 ab
H2	53.02 ± 1.85 ab	2.07 ± 0.46 a	18.5 ± 0.78 ab	18.60 ± 0.82 ab
H3	53.59 ± 0.90 abc	3.13 ± 0.33 b	20.53 ± 0.75 def	20.79 ± 0.71 cd
H4	52.98 ± 1.49 ab	3.74 ± 0.22 c	21.92 ± 1.05 f	22.24 ± 1.03 e
H5	54.14 ± 0.89 abc	3.27 ± 0.23 bc	20.52 ± 0.96 ef	20.78 ± 0.94 cd
H6	56.61 ± 1.15 bc	1.89 ± 0.39 a	17.64 ± 0.80 a	17.75 ± 0.77 a
H7	55.18 ± 2.38 abc	2.38 ± 0.19 a	20.36 ± 0.97 cde	20.50 ± 0.96 cd
H8	52.34 ± 1.64 a	3.48 ± 0.25 bc	20.79 ± 0.53 ef	21.09 ± 0.53 de
H9	52.95 ± 1.31 ab	3.40 ± 0.41 bc	19.19 ± 0.58 bcd	19.49 ± 0.57 bc
Mean ± SD	54.17 ± 1.97	2.94 ± 0.35	19.82 ± 0.97	20.03 ± 0.91
Minimum value	52.34	1.89	18.5	17.75
Maximum value	56.74	3.74	21.92	22.24

Note: Values with different lowercase letters in the same column are significantly different (P < 0.05) according to DMRT; L (lightness), a (redness), b* (yellowness), and c* (chroma).

pH

The pH values of honey obtained in this study ranged from 2.80 to 3.68, with the lowest pH found in honey from H2 and the highest pH in honey from H9. Currently, Indonesia does not have a set minimum and maximum pH standard for stingless bee honey, either in national regulations or in standards set by the International Honey Commission. However, based on the standards used in Malaysia, the pH range for stingless bee honey is between 2.5 and 3.8. ¹³ The pH values observed in this study fall within a relatively low range and are generally consistent with previous reports on stingless bee honey, with reported pH values ranging from 3.32 to 3.36, ⁷ 3.40 to 3.70 ¹⁴ and 3.44 to 3.71 ¹⁵ across different stingless bee species. These findings are consistent with the well‐established characteristic that stingless bee honey generally exhibits higher acidity than honey produced by Apis species.

However, the pH values obtained in the present study were lower than those reported in previous studies. Reported pH values ranged from 3.82 to 3.95 in some stingless bee honeys ¹⁶ while values of 4.16, 4.28, 4.48, 4.52, and 4.21 were reported for Tetragonisca fiebrigi, Tetragona clavipes, Scaptotrigona bicunctata, Melipona scutellaris, and Melipona rufiventris mondury, respectively ¹⁷ .The highest reported value was 5.19 for Melipona mondury. ¹⁵

The notably lower pH observed in this study is likely influenced by fermentation activity mediated by naturally occurring LAB. During fermentation, LAB convert glucose into organic acids, predominantly lactic acid, resulting in increased acidity and a corresponding reduction in pH. This interpretation is supported by the confirmed presence of LAB in the analyzed samples and by previous findings demonstrating that glucose degradation into lactic acid by LAB plays a central role in pH reduction during Kelulut honey fermentation. ¹⁸

Beyond microbial activity, pH variation in stingless bee honey is also influenced by species‐specific differences and nectar sources.^{19, 20, 21} In addition, gluconic acid formed through the oxidation of glucose by glucose oxidase contributes substantially to the acidic environment of stingless bee honey, with higher gluconic acid levels being associated with lower pH values. ²²

Brix

°Brix in honey represents the total dissolved solids, predominantly sugars, present in the honey solution. ¹⁸ In this study, °Brix values ranged from 49.83 to 61.25, with the highest value observed in sample H9 (61.25) and the lowest in sample H5 (49.83). According to the Indonesian standard for stingless bee honey (SNI 8664:2018), the minimum acceptable °Brix value is 55. Based on this criterion, samples H4 (54.55) and H5 (49.83) did not meet the required quality standard for this parameter.

Compared with previous studies, the °Brix values obtained in this study were generally lower. Earlier investigations reported substantially higher °Brix values for various stingless bee species.^{22, 23} These findings indicate that stingless bee honey commonly exhibits higher total dissolved solids than those observed in the present study.

The relatively low °Brix values recorded are most likely attributable to the high moisture content of the honey, which dilutes the solution and reduces the concentration of total dissolved solids. This explanation is consistent with previous reports demonstrating a direct inverse relationship between moisture content and °Brix values as a result of dilution effects. ¹⁵ In addition to moisture content, °Brix values are influenced by multiple factors, including bee species, floral origin, geographical location, processing practices and storage conditions.^{24, 25} Furthermore, prolonged storage may lead to a decline in °Brix values as a result of fermentation processes, during which sugars are metabolized into alcohol and carbon dioxide.^{26, 27}

Viscosity

Viscosity is a test used to measure the thickness of honey. In Indonesia, no standard for the viscosity of honey from stingless bees has been established. In the present study, the viscosity of the honey ranged from 23.36 to 175.20 cP. The viscosity values obtained in this study were lower than those reported in other studies, ²⁸ where Melipona honey viscosities ranged from 259 cP to 6769 cP in rural areas and from 244 cP to 7545 cP in urban environments. Similarly, other work reported a viscosity of 596.33 cP for Trigona honey. ²⁹

The relatively low viscosity observed in this study can primarily be attributed to the higher moisture content and lower sugar concentration of the honey. ²⁹ This explanation is supported by previous findings indicating that stingless bee honey typically exhibits reduced viscosity because of its elevated water content and lower levels of polysaccharides and sugars. ³⁰ Honey viscosity is influenced by multiple factors, including nectar source, moisture content, temperature, fructose and glucose composition, crystallization behavior and overall chemical composition. ³¹ In addition to water content, thermal treatment has been shown to reduce honey viscosity. ²⁸

Color

Honey color is widely recognized as a quality attribute by consumers, although it does not always accurately reflect its internal chemical composition.^{31, 32} The samples analyzed in this study exhibited a distinctly light color profile, characterized by high L* values and moderate a* to b* chromatic intensities.

The light color observed is likely influenced by the nectar sources available at the collection site. Vegetation surrounding the sampling area primarily consisted of fruit trees and smallholder plantation crops located near residential and community farmland. These plants generally yield nectar with lower mineral and pigment content compared to darker honey producing species such as H. brasiliensis. ³³ Reduced mineral and chromophore contribution from these nectar sources may therefore diminish pigmentation intensity, resulting in a lighter honey hue.

A notable observation was that the honey samples retained their light coloration despite their low pH values. This phenomenon may be associated with the limited progression of non‐enzymatic browning reactions during storage and processing, since such reactions are generally promoted by heat exposure and prolonged storage conditions.^{34, 35} The stability of the light color is further corroborated by the exceptionally low HMF content (2.47–12.78 mg kg⁻¹). Given that the honey was analyzed 1 week after harvest, transported for 2 days, and stored at 4 °C for 5 days prior to testing, the low HMF levels strongly indicate that the samples remained fresh and did not undergo thermal degradation or heat‐related browning processes.

Chemical analyses

The chemical analyses of the honeys included electrical conductivity (EC) and HMF measurements are presented in Table 3. The bioactivity analyses included TPC and TFC are presented in Table 4.

Table 3.

Electrical conductivity (EC) and HMF analyses of stingless bee honeys in Sijunjung, Indonesia

Honey	Chemical analyses
	EC (mS cm−1)	HMF (mg kg−1)
H1	0.223 ± 0.64 a	4.13 ± 0.78 a
H2	0.299 ± 0.10 f	4.26 ± 0.71 a
H3	0.263 ± 0.10 d	3.30 ± 1.02 a
H4	0.751 ± 0.10 i	4.31 ± 0.93 a
H5	0.471 ± 0.10 h	4.04 ± 0.44 a
H6	0.291 ± 0.15 e	2.47 ± 0.06 a
H7	0.301 ± 0.21 g	2.62 ± 0.59 a
H8	0.259 ± 0.10 c	12.78 ± 2.40 b
H9	0.253 ± 0.38 b	2.84 ± 0.54 a
Mean ± SD	0.325 ± 0.40	4.52 ± 0.83
Minimum value	0.223	2.47
Maximum value	0.751	12.78

Note: Values with different lowercase letters in the same column are significantly different (P < 0.05) according to DMRT; EC (electrical conductivity) and HMF (hydroxymethylfurfural).

Table 4.

Bioactivity analyses of stingless bee honeys in Sijunjung, Indonesia

Honey	Bioactivity analyses
	TPC (mg GAE g–1)		TFC (mg RE g–1)
	Methanol (80%)	Water	Methanol (80%)	Water
H1	1.80 ± 0.33 c	1.86 ± 0.02 d	0.15 ± 0.14 a	0.08 ± 0.06 a
H2	1.73 ± 0.04 c	1.64 ± 0.00 c	0.06 ± 0.00 a	0.04 ± 0.02 a
H3	1.00 ± 0.03 a	1.08 ± 0.02 a	0.07 ± 0.09 a	0.06 ± 0.05 a
H4	2.28 ± 0.03 d	2.75 ± 0.09 g	0.28 ± 0.12 b	0.27 ± 0.07 d
H5	1.59 ± 0.08 c	1.44 ± 0.03 b	0.08 ± 0.04 a	0.07 ± 0.01 a
H6	1.76 ± 0.06 c	1.99 ± 0.01 f	0.08 ± 0.00 a	0.09 ± 0.03 ab
H7	1.35 ± 0.09 b	1.37 ± 0.04 b	0.07 ± 0.02 a	0.18 ± 0.00 c
H8	2.10 ± 0.07 d	2.33 ± 0.06 e	0.11 ± 0.00 a	0.15 ± 0.03 bc
H9	3.12 ± 0.17 e	2.81 ± 0.17 g	0.31 ± 0.02 b	0.09 ± 0.02 bc
Mean ± SD	1.85 ± 0.10	1.61 ± 0.05	0.13 ± 0.05	0.11 ± 0.03
Minimum value	1.00	1.08	0.06	0.04
Maximum value	3.12	2.81	0.31	0.27

Note: Values with different lowercase letters in the same column are significantly different (P < 0.05) according to DMRT; TPC (total phenolic content), TFC (total flavonoid content), GAE (gallic acid equivalent) and RE (rutin equivalent).

EC

EC is an important parameter used to describe the levels of minerals, salts, organic acids and proteins in honey, and is therefore widely applied as an indicator of honey quality and botanical origin. ³⁶ In this study, the EC values of honey produced by nine stingless bee species ranged from 0.223 to 0.751 mS cm⁻¹, with the highest value observed in sample H4 (0.751 mS cm⁻¹) and the lowest in H1 (0.223 mS cm⁻¹). These values are consistent with previous reports for several stingless bee species, such as S. bicunctata (0.285 mS cm⁻¹) ³⁰ and Hypotrigona sp. (0.303 mS cm⁻¹). ²¹ Similar ranges have also been reported for freshly harvested and commercial stingless bee honey (0.33–0.65 mS cm⁻¹), ³⁷ Melipona sp. (0.392 mS cm⁻¹) ³⁸ and Melipona beecheii (0.58 mS cm⁻¹). ³⁹ In addition, EC values ranging from 0.270 to 0.700 mS cm⁻¹ have been reported for Melipona bicolor, Melipona marginata and S. bicunctata. ¹⁵ In contrast, substantially higher EC values have been reported for other stingless bee honeys, including M. mondury, Melipona angustula, Melipona quadrifasciata and Trigona clavipes, ¹⁵ as well as H. fimbriata, with EC values reaching up to 2.6 mS cm⁻¹. ¹⁴

As a result of the restricted foraging activity caused by persistent rainfall, bees predominantly collected nectar from plantation crops and fruit trees, which constitute the dominant vegetation in the study area. Under low‐mineral soil conditions, these plant types tend to produce nectar with lower mineral concentrations compared to forest vegetation or plants growing in mineral‐rich soils. Consequently, the honey produced exhibited relatively low EC values. This interpretation is consistent with previous studies,^{15, 19} which demonstrated that soil mineral composition and dominant floral sources are major determinants of EC variability in stingless bee honey.

This cause–effect relationship is further supported by the color characteristics of the honey samples. Darker honey is generally associated with higher mineral content, ⁴⁰ whereas the honey analyzed in this study exhibited a relatively light color, indicating low mineral levels. This observation is fully consistent with the low EC values recorded and reinforces the interpretation that edaphic conditions, high rainfall and limited nectar diversity collectively contributed to the EC profile of stingless bee honey in this region.

HMF

HMF is widely recognized as a key indicator of honey freshness and thermal or storage‐induced degradation. In this study, the HMF content of nine stingless bee honey samples from Sijunjung ranged from 2.47 to 12.78 mg kg⁻¹, which is well below the maximum limits established by the Indonesian National Standard (40 mg kg⁻¹) and the International Honey Commission (60 mg kg⁻¹). These values reflect the freshness of the samples, which were analyzed within 1 week after harvest, stored at 4 °C, and not subjected to any heating treatment.

The low HMF levels are consistent with the relatively low total sugar content (°Brix) observed in the samples. Because HMF is formed through the degradation of reducing sugars, lower sugar concentrations tend to slow its formation. This supports the biochemical stability of the samples and aligns with the principle that higher glucose and fructose contents promote more rapid Maillard and caramelization reactions.

Comparative data from previous studies show that the HMF values obtained here fall within the lower range commonly reported for stingless bee honey. Previous reports ¹⁵ documented similarly low HMF concentrations across several stingless bee honey types, typically ranging from a few mg kg⁻¹ to approximately 10 mg kg⁻¹. By contrast, higher HMF values reported in other studies are generally associated with heat exposure or prolonged storage conditions.

Environmental factors may also contribute to the low HMF levels observed. The collection sites are characterized by low mineral soils with pale yellow, sandy textures, coupled with high rainfall. These conditions limit nectar diversity, leading to a floral source profile dominated by fruit and plantation crops. Such ecological characteristics can influence nectar sugar composition and subsequently reduce the potential for HMF formation. Although low pH is generally considered favorable for HMF formation, ⁴⁰ the samples in this study exhibited low HMF levels despite their relatively low pH values. This may be attributed to the optimal handling and storage conditions applied in the present study, including short storage duration, cold storage at 4 °C, absence of heating and relatively low sugar content, all of which may slow the pathways leading to HMF formation. Overall, the combination of low HMF, low °Brix, optimal storage conditions and the distinctive environmental characteristics of the region strongly suggests that the Sijunjung stingless bee honey samples remained fresh, thermally stable and of high quality.

TPC and TFC

From Table 4, it can be observed that the use of water and methanol:water solution (80:20, v/v) for extracting stingless bee honey yielded results that were not significantly different, except for sample H9, which showed a notable difference. The TPC of the methanol:water extract (80:20, v/v) ranged from 1.00 mg GAE g^–1 to 3.12 mg GAE g^–1, with the highest TPC observed in H9 and the lowest in H3, whereas the water solvent ranged from 1.08 mg GAE g^–1 to 2.81 mg GAE g^–1. This difference is attributed to the properties of each solvent. The methanol:water solution (80:20, v/v), being semi‐polar, is more effective in dissolving slightly polar or more hydrophobic phenolic compounds. This solvent can extract compounds like flavonoids and stilbenoids, which are more soluble in organic solvents. Conversely, water, being highly polar, is better suited for dissolving highly hydrophilic phenolics, such as phenolic acids and compounds with hydroxyl groups attached to aromatic rings.

These values are higher than previous reports, ⁴¹ which documented TPC levels of 0.65–2.30 mg GAE 100 g^–1 in T. laeviceps honey collected from several regions in Indonesia. Comparable ranges have also been reported in other stingless bee honeys, ⁴² with values between 1.30 and 66.00 mg GAE 100 g^–1, whereas other studies^{15, 43} reported phenolic contents reaching up to approximately 98 mg GAE 100 g^–1. The comparatively higher TPC values observed in the present study are most likely attributable to variations in bee species, botanical origin and habitat conditions, as previously highlighted. ⁴⁴ In addition, geographical variation has been shown to significantly influence phenolic composition and antioxidant properties of stingless bee honey. ⁴¹

In the present study, TFC in methanolic extracts ranged from 0.06 to 0.31 mg RE g^–1, while water extracts ranged from 0.04 to 0.27 mg RE g^–1. These values are generally lower than those reported in previous studies on stingless bee honey from different regions,^{45, 46, 47, 48} which documented considerably higher flavonoid levels across various bee species and geographical origins. Nevertheless, variation in TFC is strongly influenced by botanical origin, bee species and environmental conditions. ⁴⁸ Regional comparisons have demonstrated that climatic conditions may significantly influence flavonoid accumulation, with higher TFC often reported in hotter environments compared to cooler and wetter regions. ⁴⁶ Therefore, the relatively lower TFC observed in the present study may be associated with high rainfall conditions, limited foraging activity, and the dominance of plantation crops and fruit vegetation in the study area.

Antioxidant activity

The antioxidant activity of honey was analyzed using two methods: the DPPH assay and the FRAP assay. The DPPH assay was used to measure the ability of honey to scavenge free radicals, whereas the FRAP assay evaluated its capacity to reduce ferric ions to ferrous ions. The results are shown in Table 5.

Table 5.

Antioxidant activity of stingless bee honeys in Sijunjung, Indonesia

Honey	Antioxidant properties
	DPPH (% inhibition)		FRAP (mg AAE kg–1)
	Methanol (80%)	Water	Methanol (80%)	Water
H1	18.52 ± 0.69 abc	16.10 ± 0.36 a	460.92 ± 55.38 bc	493.81 ± 45.75 bc
H2	17.36 ± 0.16 bc	16.92 ± 0.49 a	392.55 ± 17.45 b	389.97 ± 73.89 ab
H3	18.1 0 ± 0.48 abc	16.52 ± 0.39 a	255.16 ± 19.38 a	315.79 ± 75.76 a
H4	20.7 ± 0.55 abcd	19.80 ± 0.20 cd	609.27 ± 61.47 d	615.72 ± 27.09 cd
H5	21.69 ± 0.43 cd	20.80 ± 1.03 d	616.36 ± 56.16 d	809.86 ± 117.64 e
H6	20.56 ± 1.49 abcd	18.28 ± 0.12 b	447.37 ± 61.43 bc	506.71 ± 46.45 bc
H7	16.94 ± 0.24 a	16.57 ± 0.32 a	297.09 ± 27.52 a	388.68 ± 70.95 ab
H8	21.45 ± 1.64 bcd	18.75 ± 0.45 bc	537.03 ± 91.05 cd	716.34 ± 164.78 de
H9	22.84 ± 6.04 d	18.54 ± 1.60 bc	418.99 ± 45.97 b	485.43 ± 42.45 bc
Mean ± SD	20.00 ± 1.30	18.03 ± 0.55	448.30 ± 124.96	524.26 ± 164.34
Minimum value	17.36	16.10	255.16	315.79
Maximum value	22.84	20.80	616.36	809.86

Note: Values with different lowercase letters in the same column are significantly different (P < 0.05) according to DMRT.

DPPH

Antioxidant capacity values can vary depending on the analytical method employed, as different assays respond to different radical species. In the present study, antioxidant activity was evaluated using the DPPH radical scavenging assay. The DPPH inhibition values ranged from 17.36–22.84% for methanol:water solution (80:20, v/v) and 16.10–20.80% for aqueous extracts, indicating that methanol provided more efficient extraction of antioxidant compounds.

The DPPH radical‐scavenging activity observed in this study is comparable to previously reported values for several stingless bee species, including G. thoracica, T. laeviceps, L. terminata and H. itama. ²¹ However, the values observed in the present study remained lower than those reported in other studies, where broader ranges of DPPH scavenging activity were documented across different Meliponini species.^{3, 7} Stingless bee honey has generally been reported to exhibit higher antioxidant capacity than honey from Apis species because of its richer phenolic composition and bioactive compound profile. ⁴⁹

The relatively lower antioxidant activity observed in this study is likely associated with the lower TPC and TFC, which are influenced by high moisture content, harvesting under high humidity conditions (90%) and persistent rainfall. These environmental conditions can reduce nectar quality and bioactive compound accumulation. Previous studies have demonstrated a strong positive correlation between phenolic and flavonoid content and DPPH radical scavenging activity.^{50, 51}

In addition, botanical origin, geographical conditions and bee species are known to significantly influence antioxidant activity in stingless bee honey. ¹⁵ Even within the same geographical location, differences in bee species can result in significant variation in DPPH inhibition values. This observation is consistent with previous findings showing that antioxidant activity in T. laeviceps honey is strongly associated with its phenolic and flavonoid composition and influenced by meliponiculture location. ⁵²

FRAP

The FRAP values of the stingless bee honey samples showed considerable variation among species and extraction solvents. In methanolic extracts, FRAP values ranged from 255.16 to 616.36 mg ascorbic acid equivalents (AAE) kg^–1, whereas aqueous extracts exhibited slightly broader values ranging from 315.79 to 809.86 mg AAE kg^–1. The highest ferric reducing antioxidant power was observed in sample H5, particularly in the water extract, indicating the presence of highly water‐soluble antioxidant constituents. In general, aqueous extracts demonstrated higher FRAP values than methanolic extracts, suggesting that hydrophilic antioxidant compounds contributed substantially to the reducing capacity of the honey samples. A previous study has documented comparable FRAP values in Kelulut honey produced by Heterotrigona itama bees (41.95 mg AAE 100 g^–1). ⁵³

The variation in FRAP values is strongly associated with the total phenolic and flavonoid content of honey, as these compounds function as major antioxidant constituents. ⁵⁴ Previous studies have demonstrated that higher FRAP values are directly correlated with higher TPC.^{3, 51} Processing and storage conditions may also contribute to differences in FRAP values because thermal treatment has been shown to increase FRAP through the formation of Maillard reaction products with antioxidant activity. ⁵⁵

FTIR spectroscopy

FTIR spectroscopy analysis of nine stingless bee honey samples (H1–H9) revealed spectral characteristics that were very similar, indicating comparable chemical compositions across all samples, as shown in Fig. 2. All samples exhibited strong absorption bands in the range 3198–3239 cm⁻¹, characteristic of O—H stretching vibrations, indicating the presence of hydroxyl groups.

Figure 2.

FTIR spectra of honey samples from stingless bees. (H1:T. minangkabau, H2:T. binghami, H3:H. itama, H4:H. fimbriata, H5:T. apicalis, H6:T. testaceitarsis, H7:G. thoracica, H8:T. laeviceps and H9:L. canifrons)

These groups could originate from water content or carbohydrate structures within the honey. These findings are consistent with a previous study as well as a peak at 3242 cm⁻¹. ⁵⁶ The FTIR spectrum also revealed other functional groups contributing to the chemical composition of honey. Absorption bands in the range of 2870–2933 cm⁻¹ indicated C—H stretching vibrations, which are commonly associated with carbohydrates and other aliphatic compounds. A weak absorption band at 2934 cm⁻¹ has also been reported, potentially associated with the presence of carboxylic acids or other organic compounds. ⁵⁶ The variability in carboxylic acid content in honey is influenced by botanical origin, geographical location and environmental conditions. ⁵⁷

A strong absorption band at 1629–1640 cm⁻¹ can be attributed to C=O stretching vibrations, indicating the presence of carbonyl groups in the honey samples. This peak suggests contributions from phenolic compounds, proteins, or bound water. The FTIR spectra also revealed peaks associated with sugars and polysaccharides. Several samples showed absorption bands at 1416.6 cm⁻¹, 1341.5 cm⁻¹ and 1252.3 cm⁻¹, indicating the presence of C—H and C—O bonds typical of carbohydrate structures. Major absorption bands corresponding to C—C and C–O–C linkages in polysaccharides were identified in the ranges of 818.09–820.96 cm⁻¹ and 769.20–777.83 cm⁻¹, confirming the presence of oligosaccharides and polysaccharides that contribute to honey texture and stability. Additional bands at 915.86–920.14 cm⁻¹ and 861.22 cm⁻¹ (observed in sample H3) further support the presence of polysaccharide structures. The band at 1252.3 cm⁻¹ in sample H6 also indicates C–O–C stretching in polysaccharides. Furthermore, low‐frequency absorptions at 582.28 cm⁻¹ and 531.96 cm⁻¹ correspond to C—C or C—O vibrations, reflecting the presence of more complex carbohydrate structures.

Other studies have reported strong C—O and C—C stretching vibrations in the 1000–1200 cm⁻¹ region, associated with sucrose, glucose and fructose, which are the primary sugars responsible for the characteristic sweetness of honey.^{56, 58}

Microbiological analyses

Microbiological analysis was conducted to assess the quality of honey based on the presence of microorganisms, including the total count of LAB, as well as the growth of yeast and molds. The results of the microbiological analysis can be seen in Table 6. Honey quality is influenced by various factors, including the presence of microorganisms such as LAB, molds and yeasts. Honey‐associated LAB are known to produce antimicrobial compounds such as organic acids, hydrogen peroxide and antimicrobial peptides, thereby contributing to the antimicrobial properties of honey. ⁵⁹ Meanwhile, molds and yeasts are among the pathogenic agents in honey that can contribute to quality deterioration. Various microorganisms have been reported in honey including molds, yeasts and bacteria such as Alternaria alternata, Aspergillus niger, Aspergillus proliferans, Aspergillus spelunceus, Chaetomium globosum, Cladosporium cladosporioides, Daldinia concentrica, Emericella discophora, Emericella qinqixianii, Penicillium corylophilum, Penicillium decumbens and Penicillium echinulatum, as well as bacterial genera including Bacillus sp. (particularly Bacillus cereus and Bacillus pumilus), Clostridium, Escherichia coli, Salmonella sp., Enterococcus sp., Staphylococcus sp. and Pseudomonas spp. have also been identified.^{60, 61} Lactic acid bacteria were detected only in H4, H7, and H8 samples, with counts ranging from 2.79 to 3.22 log CFU mL⁻¹. No yeast or mold growth was detected in any honey sample indicating that the honeys were in good hygienic condition and presented an unfavorable environment for fungal proliferation. Fresh honey may contain up to 10⁸ CFU g^–1 of LAB. ⁶² Variations in LAB abundance are influenced by multiple factors, including bee species, nectar source, harvesting and storage conditions, and environmental or climatic factors. Temperature‐related changes in nectar microbial abundance and sugar composition may also contribute to differences in the microbial profile of bee products.^{63, 64, 65}

Table 6.

Microbiological analyses of stingless bee honeys in Sijunjung, Indonesia

Honey	Lactic acid bacteria (log CFU mL−1)	Yeast (log CFU L−1)	Mold (log CFU mL−1)
H1	ND	ND	ND
H2	ND	ND	ND
H3	ND	ND	ND
H4	2.77 ± 0.17	ND	ND
H5	ND	ND	ND
H6	ND	ND	ND
H7	2.95 ± 0.62	ND	ND
H8	3.10 ± 0.32	ND	ND
H9	ND	ND	ND

Note: Results are expressed as mean ± standard deviation (log CFU mL⁻¹) calculated from countable dilution plates. ND = not detected.

Stingless bee honey is known to possess antimicrobial activity against both Gram‐positive and Gram‐negative bacteria, including Staphylococcus aureus and Pseudomonas aeruginosa. ⁶⁶ The flavonoid compounds may contribute to the antimicrobial activity of stingless bee honey because their presence has been associated with higher antimicrobial activity and regulation of hive microbiota. ⁶⁷

Volatile compound composition

Analysis of the volatile compound profiles of nine different stingless bee species revealed significant differences in chemical composition between species, despite being kept in the same region and utilizing similar nectar sources (Fig. 3). This variation is considered to be a result of differences in the physiology and metabolism of each species, as well as the tendency of bees to prefer different nectar sources. The species T. minangkabau (H1) produced a total of 18 volatile compounds with five main compounds: linalool oxide (cis) (33.12%), benzyl alcohol (15.71%), linalool oxide (trans) (11.58%), phenethyl alcohol (5.37%) and isopropyl tetradecanoate (5.23%). Tetrigona binghami (H2) produced twenty five compounds with the three main compounds being 1‐octanol (34.66%), junenol (26.58%) and linalool oxide (cis) (8.24%).

Figure 3.

Relative abundance of volatile compounds of stingless bee honeys (%).

Tetrigona apicalis (H5) showed high compound diversity with a total of 29 compounds, with the main ones being junenol (27.06%), linalool oxide (cis) (21.11%), 1‐octanol (19.92%), octyl acetate (9.29%) and linalool oxide (trans) (8.08%). In total, 16 volatile compounds were detected in T. testaceitarsis (H6), the most dominant of which were phenethyl alcohol (50.32%), linalool oxide (cis) (22.41%) and linalool oxide (trans) (6.97%). Geniotrigona thoracica (H7) produced 15 compounds, mainly phenethyl alcohol (21.21%), linalool oxide (cis) (22.92%), linalool oxide (trans) (8.55%), phenylacetaldehyde (16.91%) and benzaldehyde (15.15%). In total, 16 compounds were identified in T. laeviceps (H8) and the dominant compounds were phenethyl alcohol (33.72%), linalool oxide (cis) (14.83%), junenol (13.6%), benzeneacetaldehyde (6.77%), benzyl alcohol (5.95%) and linalool oxide (trans) (5.53%). Finally, 16 volatile compounds were detected in L. canifrons (H9) and the six most dominant compounds were phenethyl alcohol (25.04%), junenol (19.26%), linalool oxide (cis) (12.39%), benzyl alcohol (9.8%), 5‐[(2′‐hydroxyethoxy) carbonyl] pentane (5.77%) and phenylacetaldehyde (5.66%).

Both trans‐ and cis‐linalool oxide, which are linalool derivatives, were found in all honey samples. cis‐Linalool oxide is formed from 6,7‐linalool via epoxylinalool. This conversion requires the action of an enzyme secreted by bees, which is likely a hydroxylase. Some linalool derivatives, such as linalool oxide, can also be produced during honey storage. ⁶⁸ Linalool and its derivatives have also been widely reported as common volatile components in Apis mellifera honeys.^{69, 70} In the present study, junenol was detected in all samples except H1. This compound has also been reported as one of the major constituents of Brazilian propolis. ⁷¹ Furthermore, junenol has been identified as the dominant compound (13–52%) in Acritopappus confertus, a plant known for its wound‐healing properties and proposed as a chemical marker for this species. ⁷² Previous studies on the volatile profile of stingless bee honey have not reported the presence of junenol. Therefore, the detection of junenol in the present study represents a potentially novel finding in stingless bee honey. This compound is likely associated with the unique botanical origin of honey from West Sumatra, Indonesia, and may serve as a potential regional marker for Indonesian stingless bee honey.

Principal Component Analysis

Figure 4 shows the PCA biplot diagram based on the physicochemical and bioactive properties of honey samples.

Figure 4.

Principal component analysis biplot diagram based on physicochemical and bioactive properties of stingless bee honeys.

PCA biplot analysis demonstrated that the samples were distinctly separated in terms of physicochemical and bioactive properties. The first two principal components (PCs) explained approximately 67.6% of the total variance (PC1: 40.87%; PC2: 26.71%) and the samples showed a strong separation, particularly along PC1. The clustering of vectors for parameters such as pH, viscosity, TPC, DPPH radical scavenging activity and HMF in the positive direction of PC1 suggests that these variables are positively correlated with each other and that PC1 represents these characteristics. Accordingly, it is understood that samples located on the positive side of PC1 (e.g. H8 and H9) tend to have higher values for the aforementioned parameters, whereas samples on the negative side of PC1 (e.g., H1, H2, H3, H6 and H7) exhibit a lower profile for these characteristics. The PC2 axis is characterized particularly by the °Brix parameter, and it has been observed that samples on the positive side of PC2 are more closely related to °Brix. In contrast, the positioning of EC and some color/antioxidant indicators (e.g. variables aligned with FRAP) on the negative side of PC2 suggests that these variables may show an inverse trend with °Brix along the PC2 component. Overall, PCA revealed that the differences between samples were primarily explained by phenolic content/antioxidant capacity and certain physicochemical properties (pH, viscosity and HMF).

CONCLUSIONS

This study provides an integrated characterization of stingless bee honeys produced within a shared ecosystem, combining physicochemical quality indices with antioxidant/bioactivity metrics and fingerprinting/profiling tools (e.g. FTIR and volatile profiling). Beyond reporting compositional values, our results indicate that the overall chemical and functional signature of these honeys is sufficiently distinctive to support species‐associated differentiation, particularly when multiple analytical blocks are interpreted together rather than in isolation.

From an authentication perspective, the combined use of routine parameters (e.g. moisture, acidity, EC and HMF) together with spectral/volatile fingerprints offers a pragmatic framework for building reference profiles that can help distinguish stingless bee honeys from conventional Apis honeys and from potential adulterated products. For quality control, the dataset highlights which parameters are most informative for screening freshness/handling and for setting preliminary specification ranges in regional stingless bee honeys.

Regarding functional food development, the observed antioxidant‐related properties and bioactive‐associated signatures support the potential of these honeys as ingredients for value‐added formulations, at the same time as emphasizing the importance of standardized quality criteria to ensure batch‐to‐batch consistency.

Finally, we note that the lack of melissopalynological confirmation limits definitive separation of floral and bee‐species effects; therefore, our conclusions are framed as species‐associated patterns under a shared environment. Future studies incorporating pollen analysis and expanded sampling will enable more robust attribution and facilitate the development of validated multivariate classification models for routine authentication.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.