Natural Etiolation in Spearmint ( Mentha spicata L. var. crispa ) Shoots Emerging From Rhizomes—Plastids, Glandular Hairs, and Phytochemical Profile Authors: Bernadett Szögi‐Tatár, Adrienn Dobi, Andrea Böszörményi, Enkhjin Enkhbileg, Nóra Gampe, Imre Boldizsár, Anna Skribanek, Katalin Solymosi Published: 2026 ABSTRACT ABSTRACTSpearmint ( Mentha spicata L. var. crispa) is a widely cultivated aromatic herb valued both as a culinary ingredient and as a source of bioactive compounds, such as essential oils and phenolics, which contribute to its traditional uses and potential medicinal applications. In this work, we provided the first evidence of natural etiolation with etioplast and prolamellar body formation in leaves and stems of spearmint shoots emerging in the soil from underground rhizomes. Soil‐grown and hydroponically dark‐forced shoots exhibited comparable etiolation symptoms, plastid ultrastructure, and chlorophyll biosynthesis as revealed by 77 K fluorescence emission spectroscopy, establishing laboratory etiolated plants as reliable models for natural skotomorphogenetic processes. Comparison of 14‐day‐old light‐grown and dark‐forced shoots developing from the rhizomes of hydroponically cultured adult plants revealed that the slightly smaller etiolated leaves had significantly smaller peltate glandular hairs than light‐grown ones (diameter: 54–56 μm vs. 59 μm, respectively), but their number and density were not influenced by light and they were always present in much higher numbers on the abaxial side than on the adaxial side of the leaf surfaces. Principal component analysis of the volatile organic compound and phenolic profiles distinctly separated light‐grown and dark‐forced shoots, with fewer compounds being present and in lower relative abundance in etiolated tissues, and more pronounced differences observed in the phenolic composition. These data underscore the value of etiolation and greening research, provide new models for studying development and environmental interactions under naturally light‐deprived conditions, and offer novel insights into primary and secondary metabolism in hidden spearmint organs.Keywords: dark‐forcing, peltate glandular hair, phenolics, prolamellar body, volatile organic compounds Introduction Plants perceive light through several photoreceptors, and their development is influenced in many ways by the absence or presence of light, its quality, and intensity. Seeds of angiosperm plants germinating in the dark undergo skotomorphogenesis, which is characterized by altered hormonal and molecular signaling and regulatory processes in their shoots, resulting in elongated whitish shoots or stem‐related organs, yellow cotyledons and, in dicotyledonous plants with large seeds, often in small yellowish primary leaves (Armarego‐Marriott et al. 2020; Solymosi and Schoefs 2010). The yellow coloration can be explained by the fact that in angiosperms, chlorophyll (Chl) biosynthesis is also light‐dependent because one of its steps, the transformation of protochlorophyllide (Pchlide) into chlorophyllide (Chlide), is catalyzed by the light‐dependent NADPH:protochlorophyllide oxidoreductase (LPOR) enzyme, which requires light for its activity (Solymosi and Mysliwa‐Kurdziel 2021; Wang and Grimm 2021). Consequently, dark‐grown organs contain yellow carotenoids but lack Chl, and due to negative feedback regulation, they only accumulate very low amounts of the Chl precursor, Pchlide. Without Chl and other light‐regulated processes, proplastids of photosynthetic tissues are transformed into etioplasts, and chloroplast development is inhibited (Solymosi and Aronsson 2013). In etioplasts, LPOR dimers helically associated with the external surface of the membranes stabilize a peculiar, highly regular tubuloreticular membrane structure called prolamellar body (PLB), interconnected with lamellar prothylakoid (PT) membranes (Floris and Kühlbrandt 2021; Li et al. 2024; Nguyen et al. 2021). Pchlide pigments are present in several distinct molecular environments within the etioplasts, reflected by slight alterations in their absorption and fluorescence properties (Böddi et al. 1992; Schoefs and Franck 2003). Slight differences were also observed among the organization of these pigments in leaves and stem‐related organs (Böddi et al. 1998; Hassan et al. 2021; Skribanek et al. 2000). However, a major non‐photoactive Pchlide form with fluorescence emission maximum at 633 nm is located in the PTs, and a photoactive Pchlide form with emission at around 655 nm is found in the PLBs (Ryberg and Sundqvist 1982). Upon exposure to light, photoactive Pchlide is converted to Chlide with emission maximum around 688–690 nm, and etioplasts gradually transform into chloroplasts (Pipitone et al. 2021; Solymosi and Aronsson 2013; Solymosi and Mysliwa‐Kurdziel 2021). The above‐described peculiar developmental phenomenon caused by prolonged dark growth, called etiolation, has been widely studied under different conditions, but primarily in dark‐germinated seeds under laboratory conditions (Hernández‐Verdeja et al. 2020; Solymosi and Aronsson 2013; Solymosi and Schoefs 2010). Therefore, studies dealing with etiolated plants have often been criticized because prolonged dark growth is thought to be rare in nature. However, it is evident that for seeds germinating deep in the soil, either under natural or agricultural conditions, hypocotyls, epicotyls, and cotyledons or even primary leaves may get etiolated (Kakuszi et al. 2016, 2017; Solymosi and Aronsson 2013; Solymosi and Schoefs 2010; Vitányi et al. 2013). Etiolation was also reported in several special aboveground structures containing light‐deprived inner tissues like pericarp‐covered cotyledons, avocado fruits, cabbage heads, and tree buds (Solymosi and Aronsson 2013). However, to the best of our knowledge, no detailed analysis was published on shoots developing from rhizomes under fully natural conditions. Light is not only influencing plastid differentiation and Chl biosynthesis but may also affect the production of secondary metabolites, including terpenes or phenolics. For instance, in rosemary plants, we identified important differences in the volatile organic compound (VOC) composition of light‐ or dark‐grown seedlings and shoots with an economically relevant compound (bornyl acetate) having a higher relative abundance in shoots developed in the absence of light (Böszörményi et al. 2020). Similarly, light conditions influenced the essential oil (EO) composition of peppermint (Hubert‐Schöler et al. 2024; Maffei and Scannerini 2000), Mentha arvensis (De Souza et al. 2016; Farooqi et al. 1999), Mentha citrata and Mentha cardiaca (Farooqi et al. 1999), Mentha rotundifolia (Hubert‐Schöler et al. 2024), and both the EO and the phenolic content of spearmint (Ali et al. 2024) or sweet basil (Kivimäenpä et al. 2022). EOs composed of mono‐ and sesquiterpenes are produced by leucoplasts of the secretory cells of capitate and peltate glandular hairs located primarily on leaf surfaces of Lamiaceae plants such as peppermint (Gershenzon et al. 1989; Turner et al. 2000). In addition to developmental stage and age (Maffei et al. 1986, 1989), the light regime also influences the differentiation of the glandular hairs as was shown in sweet basil (Kivimäenpä et al. 2022) and thyme (Yamaura et al. 1989). Marked differences were observed in the number and size of peltate glandular hairs of thyme cotyledons grown in the dark or in the light (Yamaura et al. 1989). Spearmint ( Mentha spicata L. var. crispa) is an important perennial rhizomatous food, aromatic, and medicinal crop used worldwide primarily for its EO content as spice and infusion (Mahendran et al. 2021). Its most valuable compounds are EO with carvone and limonene as major terpenoids (Chrysargyris et al. 2017, 2019) and phenolics (Mahendran and Rahman 2020). As a common means of vegetative reproduction and perennial overwintering, spearmint plants produce new shoots from their rhizomes, which appear etiolated until they reach the soil surface. However, so far, their Chl biosynthesis and plastid development have not been investigated in detail. Also, limited information is available about how the absence of light affects the synthesis of secondary metabolites and the development of glandular hairs. Therefore, in addition to investigating the natural etiolation of the underground shoots of spearmint grown in the soil, we also designed experiments in which we could compare plastid development, Chl biosynthesis, glandular hair development, and secondary metabolite production of light‐grown and dark‐forced shoots of this plant. Materials and Methods Plant Material M. spicata L. var. crispa plants were grown under natural conditions in Dinnyés, Hungary (47°10′35″ N, 18°33′47″ E) in pots. For analysis, they were transported to the laboratory, where the soil was carefully washed out, and their underground shoots were carefully cleaned under dim green light, which had previously been shown not to cause phototransformation of Pchlide and not to affect plastid ultrastructure (data not shown). Samples were taken for 77 K fluorescence spectroscopy and transmission electron microscopy (TEM) from the underground shoot tips with leaves and underground internodal regions as well as aboveground organs (stems and leaves) of these plants. For further experiments, intact adult plants with cleaned roots and rhizomes were placed in glass beakers containing tap water in complete darkness for 2 weeks (dark‐forcing) or in the light under laboratory conditions (22°C, 12:12 h light–dark cycles, 100 μmol m−2 s−1 photon flux density). Water was exchanged every 2–3 days. If present at the beginning of the experiment, young shoots on these plants were labeled or removed so that only the newly formed shoots developed in tap water during the 2 weeks in the light or in the dark were used for further experiments. We analyzed these newly grown young shoots using 77 K fluorescence spectroscopy, scanning electron microscopy (SEM), TEM, gas chromatography (GC), and high‐performance liquid chromatography (HPLC). For GC and HPLC analyses, we used intact shoots (i.e., stems with leaves on them) as we did not separate the leaves due to their small size. In the case of dark‐forced plants, all manipulations were performed under dim green safelight as described above. 77 K Fluorescence Spectroscopy Stems and leaves, as well as small shoot tips covered with leaves, were sampled and put into Pyrex tubes and then immersed in liquid nitrogen until and during the measurements. The 77 K fluorescence emission spectra were recorded using Jobin Yvon Horiba Fluoromax‐3 and Fluoromax‐2 spectrofluorometers over the 580–780 nm range, with excitation at 440 nm or 460 nm. Data were collected for each sample at 0.5 nm intervals with an integration time of 0.1 s. Three spectra were recorded during each measurement and were then averaged. These averaged spectra were then used further for analyses. SPSERV V3.14 program (copyright Csaba Bagyinka, HUN‐REN BRC, Institute of Biophysics) was used to process the measured spectra. First, 5‐ and 3‐point linear smoothing was applied to remove noise from electrical instabilities, light scattering, and liquid nitrogen bubbles. Then, the emission spectra were corrected with a correction curve that compensated for the wavelength‐dependent sensitivity of the detector. In addition, baseline correction was applied, and interference signals were removed where appropriate. Spectra were recorded from at least five different plants; figures show typical spectra. TEM For TEM, 1 × 1 × 2 mm shoot cross sections as well as 2 × 2 mm leaf sections were cut from samples in 2.5% glutaraldehyde under dim green safelight and were stored in this fixative for at least 3 h in the dark. Glutaraldehyde was dissolved in 0.07 M Na–K–phosphate buffer (pH = 7.2). Fixation was followed by rinsing in the above buffer three times for 15 min. Then, 1% osmium tetroxide solution (also dissolved in the above buffer) was used as a second fixative for 2 h to fix the membranes. This was again followed by the same rinsing steps as before. Dehydration of the samples was performed in ethanol at increasing concentrations with 15 min intervals (25%, 50%, 70%, 90%, and 96%) followed by 2 × 20 min in absolute ethanol. The samples were then placed in propylene oxide twice for 20 min and subsequently transferred from pure propylene oxide to a 1:1 mixture of propylene oxide:resin (Durcupan I) for 1 h. The samples were then left in Durcupan I overnight and then in Durcupan II for 3 h. After incubation in Durcupan II, the samples were transferred to silicon molds filled with Durcupan II and oriented manually for sectioning under stereomicroscope. The resin was then polymerized in a humidity‐free thermostat at 62°C for 3 days; 70 nm sections were cut from the blocks using a Reichert Jung ULTRACUT E ultramicrotome with a Diatome diamond knife. Sections were then placed on 200‐mesh copper grids and contrasted with 5% uranyl acetate and then with lead citrate. Grids were then examined on a HITACHI H‐7100 transmission electron microscope at 75 kV accelerating voltage and a JEOL JEM 1011 transmission electron microscope at 80 kV accelerating voltage, using Soft Imaging System's MegaView III and Olympus Morada digital cameras to record images, respectively. ImageJ software (National Institutes of Health) was employed to measure quantitative parameters of the plastids on the micrographs. At least 100 plastids were checked per sample, with 26–35 being randomly sampled for further quantitative analyses. SEM Intact, freshly collected leaves of the young shoots developed from the rhizomes in the dark or in the light in hydroponic system were used for SEM using a HITACHI Tabletop TEM 3000, in low vacuum mode at 5 kV accelerating voltage. Compositional (COMPO) mode of the four‐segmented semiconductor backscattered electron detector was used to record the images. Analyses related to image size were performed in the AxioVision LE64 software (Carl Zeiss GmbH). Photoshop was used to remove the background, that is, the sample holder surface on which the leaves were fixed. This way, the leaves and their margins become more clearly visible. At least 8–8 or 9–9 adaxial and abaxial leaf sides of dark‐forced or light‐grown leaves, respectively, were checked for the quantitative analyses. Solid‐Phase Microextraction (SPME) Pooled samples (0.5–1.0 g) were put into vials (20 mL headspace) sealed with a silicon/PTFE septum prior to SPME‐GC–MS analysis. Sample preparation using the static headspace solid‐phase microextraction (sHS‐SPME) technique was carried out with a CTC Combi PAL automatic multipurpose sampler using a 65 μm StableFlex carboxene/polydimethylsiloxane/divinylbenzene (CAR/PDMS/DVB) SPME fiber (Supelco). After an incubation period of 5 min at 100°C, extraction was performed by exposing the fiber to the headspace of a 20 mL vial containing the sample for 10 min at 100°C. The fiber was then immediately transferred to the injector port of the GC–MS and desorbed for 1 min at 250°C. Injections were made in splitless mode. The SPME fiber was cleaned and conditioned in a Fibre Bakeout Station in pure nitrogen atmosphere at 250°C for 15 min. GC–Mass Spectroscopy (GC–MS) The analyses were carried out with an Agilent 6890N/5973N GC–mass selective detector (MSD) system equipped with a Supelco SLB‐5MS capillary column (30 m × 250 μm × 0.25 μm). The GC oven temperature was programmed to increase from 60°C (3 min isothermal) to 250°C at 8°C/min (1 min isothermal). High‐purity helium (6.0) was used as carrier gas at 1.0 mL/min (37 cm/s) in constant flow mode. The MSD was equipped with a quadrupole mass analyzer and was operated in electron ionization mode at 70 eV in full‐scan mode (41–500 amu at 3.2 scan/s). The data were evaluated using MSD ChemStation D.02.00.275 software (Agilent Technologies). The identification of the compounds was carried out by comparing calculated linear retention indices and recorded spectra with the literature data (Adams 2017), and the NIST 2.0 library (National Institute of Standards and Technology, USA) was also consulted. Relative abundances were calculated by peak area normalization. Ten independent repetitions, each containing different pooled shoot samples, were measured per sample type, that is, both for hydroponically dark‐forced and light‐grown plants. Analytical HPLC Coupled With UV and High‐Resolution Mass Spectrometry Compound separation and detection were performed using a Dionex Ultimate 3000 UHPLC system (SRD‐3400 solvent rack degasser, HPG‐3400RS pump, WPS‐3000TRS autosampler, TCC‐3000RS column thermostat, and 3000RS diode array detector, DAD) coupled to an Orbitrap Q Exactive Focus mass spectrometer equipped with an electrospray ionization source (Thermo Fisher Scientific). Chromatographic separations were carried out on a Kinetex C18 column (75 × 3 mm, 2.6 μm; Phenomenex). The mobile phase consisted of eluent A (0.1% v/v formic acid in water) and eluent B (acetonitrile:0.1% v/v formic acid in water; 80:20, v/v). Gradient elution was applied as follows: 0.0 min, 10% B; 10.0 min, 50% B (linear); 12.0 min, 90% B (linear); 13.5 min, 90% B (isocratic); 14.0 min, 10% B (linear); and 15.5 min, 10% B (isocratic). The flow rate was 0.3 mL/min, the column temperature was maintained at 25°C, and the injection volume ranged from 1.0 to 5.0 μL. The ESI source was operated in polarity switching mode (positive and negative ionization). Instrument parameters were optimized using the manufacturer's software as follows: spray voltage, 3500 V (positive) and 2500 V (negative); capillary temperature, 256°C; sheath gas, auxiliary gas, and spare gas (N2), 47.50, 11.25, and 2.25 arbitrary units, respectively. Full‐scan MS data were acquired over an m/z range of 100–1000 at a resolution of 70,000, while MS/MS spectra were recorded at a resolution of 35,000. Fragmentation was performed using a data‐independent acquisition (DIA) method with isolation windows of 100–300, 295–500, 495–700, 695–850, and 845–1000 m/z. Collision‐induced dissociation (CID) energies were set to 15, 30, and 45 eV, and the resulting MS/MS spectra represent the summed ion intensities across all three collision energies. UV spectra were recorded using the DAD in the 230–600 nm range. Preparation of Extracts for Analysis and Isolation Lyophilized and pulverized shoot tissues (20.0 mg of pooled samples, in five independent repetitions) were extracted with 3.0 mL of 80% (v/v) methanol at 60°C for 30 min in 5 mL screw‐capped vials. The extracts were centrifuged at 2500 × g using a laboratory centrifuge, and the supernatants were analyzed by HPLC–UV–MS/MS. For the isolation of rosmarinic acid (RA), lyophilized and pulverized shoot tissue of field‐grown spearmint (approximately 2 g) was extracted three times with 15 mL of 80% (v/v) methanol at 60°C for 30 min per extraction in screw‐capped vials. After centrifugation, the supernatants were combined and concentrated under reduced pressure using a rotary evaporator. The dried residue was re‐dissolved in 2.0 mL of methanol and centrifuged at 2500 × g prior to preparative HPLC separation. Isolation of RA by Preparative HPLC A Dionex UltiMate 3000 HPLC System (VWD‐3400RS photometer, HPG‐3200BX pump, WPS‐3000TSL autosampler, VF‐F11‐A‐01 Fraction Collector F) (Thermo Fisher Scientific) was used. Detection was at 330 and 336 nm. A Gemini NX‐C18 (250 × 10 mm; 5 μm) (Phenomenex) column was used. Eluents were: eluent A (0.1% v/v formic acid) and eluent B (acetonitrile). The flow rate was 5 mL/min in the gradient program consisting of 0.0 min, 15% B; 5.0 min, 15.0% B (isocratic); 20 min, 50% B (linear gradient); 21.0 min, 58% B (linear gradient); 22.0 min, 90% B (linear gradient); 27.0 min, 90% B (isocratic). The injected amount varied between 100 and 500 μL. Statistical Analysis Data were obtained in at least three independent experiments, with several biological replicates as indicated in the figure and table legends. Data in tables refer to the mean and the standard error of the mean. We performed a normality test on our glandular hair‐related data, and in case of data with normal distribution, we used one‐way ANOVA followed by Tukey's post hoc test (i.e., for analyzing peltate glandular hair diameter) or when standard deviations were not equal, we used Brown–Forsythe and Welch ANOVA followed by Dunnett's post hoc test (for the number of hairs per mm2 entire leaf surface). For data that did not follow a normal distribution, we used Kruskal–Wallis nonparametric ANOVA followed by Dunn's post hoc test (in case of number of hairs per entire leaf surface). In case of VOC analyses, pooled samples were used, and data obtained for each component in light‐grown and dark‐forced samples were compared with t‐test. Similarly, unpaired t‐test was used to compare quantitative data related to ultrastructural analyses. Statistical analyses were done using the software GraphPad InStat. Differences were considered significant at p < 0.05 and are indicated with different letters in the corresponding tables. Principal component analysis (PCA) was applied to the GC–MS‐derived VOC composition data and to the HPLC data to explore multivariate differences among spearmint shoots grown under light and dark conditions. Computations were performed by the SYN‐TAX 2000 software package (Podani 2001). Orbitrap HR‐MS data acquired in positive ionization mode were imported into MZmine 4.9.0 (Schmid et al. 2023) using a threshold of 200,000. From the detected scans, chromatograms were built. The features were filtered for isotopes; peaks longer in duration than 5 min and peaks with a worse RT shape score than 0.94 were discarded. Features of different samples were unified using the join aligner function (max. 10 ppm and 0.1 min differences). Residual features coming from contaminants were manually removed and features were gap‐filled. Altogether 523 features were used for statistical analysis. Results and Discussion Etiolation Symptoms and Plant Morphology Under Natural Conditions Spearmint has a well‐developed underground rhizome system involved in overwintering and clonal spreading. The development of new shoots in spring or throughout the year can be observed either from stem nodes or more typically from the nodes located on the rhizomes. Upon removal of the plants from the soil, new shoots developing underground on the rhizomes could be clearly distinguished (Figure 1). Similarly to the rhizomes, these shoots had a whitish color, but small, decussately arranged yellowish leaf pairs could also be observed on their nodes. This pale coloration of the underground leaves clearly indicated that they developed in the dark and got etiolated. Next, we performed 77 K fluorescence emission spectroscopy to validate this hypothesis and check whether these shoots were fully light‐deprived throughout their development, and to what extent Chl biosynthesis proceeded in them. Well‐developed underground rhizome and root system of spearmint plants (A) with new shoots developing from the rhizomes within the soil. Shoots which have not yet reached the soil surface are etiolated with white stems and small, decussately arranged leaf pairs of yellowish coloration (B). Chlorophyll Biosynthesis in the Underground Shoots of Spearmint Plants Spectra of the underground stems and shoot tips were similar (Figure 2). When excited with 440 nm, they contained a dominating fluorescence emission band at around 630.5–631 nm corresponding to non‐photoactive Pchlide, and a band of lower intensity at around 652–653 nm indicating the presence of photoactive Pchlide (Figure 2A,B). However, the relative intensity of the 653 nm band was higher in the shoot tip surrounded by yellowish leaves (Figure 2A) than in the stem of the underground shoot (Figure 2B). A similar ratio of non‐photoactive to photoactive Pchlide forms was observed in dark‐forced Mentha shoots and stems (Skribanek et al. 2000). Excitation with 460 nm selectively excited long‐wavelength Pchlide forms and also revealed the presence of a Pchlide band at around 640.5–641 nm (Figure 2C,D). In some cases, that is, in shoot tips and leaves located closer to the soil surface, an additional fluorescence peak at 675–680 nm was also observed which probably belongs to Chl pigments (not shown). Literature data often reported similar spectra occasionally containing Chl pigments in dark‐forced stems and stem‐related organs (Skribanek et al. 2000) and in partially etiolated tissues or tissues exposed to very low light (Solymosi and Böddi 2006; Solymosi et al. 2004, 2006, 2007, 2012). Cracks and unevenness in the soil might result in some light reaching shoots close to the soil surface, and according to our light intensity measurements, incident light only decreases to zero at around 1 cm depth in sand, but this value might also depend on the soil type and structure. 77 K fluorescence emission spectra of underground spearmint shoots developing from the rhizome in the soil (as shown in Figure 1). Spectra of the shoot tip covered with leaves (A and C) and stem segments without leaves (i.e., the internodes) (B and D) were recorded with 440 nm (A and B) or 460 nm (C and D) excitation wavelengths. Representative spectra are shown after normalization at their maxima. After measurement at 77 K (solid lines), samples were thawed to 253 K, illuminated for 10 s with white light of 200 μmol m−2 s−1 photon flux density and then 10 s later frozen to 77 K again and re‐measured (broken lines). To prove that the tissues were completely light‐deprived and had therefore accumulated photoactive Pchlide, after measurement at 77 K (−196°C), samples were thawed to 253 K (−20°C), then illuminated for 10 s with white light of 200 μmol m−2 s−1 photon flux density and then immersed in liquid nitrogen before their spectra were recorded again at 77 K. The ~653 nm band disappeared and was replaced by Chlide with fluorescence emission at ~686 nm in the leaves and 675.5 and 682 nm in the stems (Figure 2). Photoactivity of Pchlide is often tested with similar methods; the fluorescence emission of the product depends on many factors but is often modified in samples undergoing freeze–thaw cycles (Kósa et al. 2006). The aboveground stems and the fully developed leaves were green and had 77 K fluorescence emission spectra resembling those of photosynthetically active green leaves with typical Chl emission bands at 685, 695, and 730 nm, respectively (not shown; for similar spectra see later Figure 5). Such bands are characteristic for Chl–protein complexes of PSII and PSI present in green leaves (Kalaji et al. 2017; Lamb et al. 2018). 77 K fluorescence emission spectra of leaves grown on 2‐week‐old shoots emerging from the rhizomes of hydroponically dark‐forced (solid and broken lines) or light‐grown (dotted line) spearmint plants. Excitation wavelength: 440 nm. Representative spectra are shown after normalization at their maxima. After measurement at 77 K (solid lines), samples were thawed to 253 K, illuminated for 10 s with white light of 200 μmol m−2 s−1 photon flux density, and then frozen to 77 K again and re‐measured (broken lines). Plastid Differentiation in Spearmint Shoots Developing Under Natural Conditions TEM analyses were performed to investigate plastid ultrastructure in aboveground and underground stems and leaves. Typical chloroplasts with inversely stained thylakoid membranes occurred in the fully developed green leaves of spearmint plants (Figure 3A) and in the cortex cells of the aboveground green stems as well (Figure 3B). However, in the latter, the number of plastids was relatively low, and the chloroplast stroma was extremely densely stained. The high electron density of the plastid stroma and inverse staining of the thylakoids were reported for light‐grown spearmint leaves earlier (Ounoki et al. 2021). Starch was typically or occasionally present in the leaf and stem chloroplasts, respectively. Plastid ultrastructure in the aboveground and underground spearmint shoots grown from the rhizome under natural conditions. Typical plastid from an aboveground fully developed leaf (A), an aboveground stem (B), a leaf located at the upper part of the shoot apex of a deeply buried underground shoot (C), and a stem of an underground shoot (D). A representative plastid was selected from approximately 100 plastid sections analyzed per sample. Cw: cell wall; S: starch; V: vacuole. Black arrows point to the prolamellar body present in the etioplasts, and asterisks are placed on grana in the chloroplasts. Scale bar: 1 μm. When we studied plastid ultrastructure in the different leaves of the underground shoots located both at the shoot apex (closer to the soil surface) or at a lower node of the stem, similar etioplasts with regular PLBs, relatively well‐developed PTs and starch grains were found (Figure 3C). Stems of the underground shoots developing from the rhizomes also contained etioplasts frequently with well‐developed PLBs and some PTs, which were often enlarged into vesicle‐like structures containing electron‐dense material (Figure 3D). Similar vesicles were described in etioplasts and etio‐chloroplasts of sunflower seedlings (Solymosi et al. 2007). Plastid inclusions, peripheral invaginations of the envelope membrane, and starch grains were observed in these plastids. Similar plastid organization was present in shoots located closer to the soil surface or located relatively deep (approx. 10 cm depth) in the soil. These data clearly indicate that etioplast formation and thus etiolation is a naturally occurring phenomenon in both leaves and stem‐like organs of shoots developing from rhizomes buried deep in the soil. Similarly, in the primary leaves of bean seedlings germinating deep in sand (Whatley 1974) or in vermiculite (Verbelen and De Greef 1979), etioplasts with PLBs were reported under natural light conditions until the organs reached the surface and thus light. Our data confirm the same, that is, natural etiolation in the soil for spearmint leaves and shoots developing not from the seeds but from rhizomes. Morphology of Light‐Grown and Dark‐Forced Hydroponically Grown Spearmint Shoots We have shown that under natural conditions the shoots and leaves developing from the rhizome get etiolated in the soil. Thus, the question could be raised whether these and in general dark‐forced shoots produced similar VOCs and other secondary metabolites, and if yes, in similar composition to light‐grown shoots. However, during the removal of such etiolated shoots from the soil and their washing, most of their glandular hairs and VOC content might also be wiped or scraped off. Also, it is hard to determine the age of the naturally etiolated shoots grown in the soil. Therefore, to investigate the effect of light on leaf development, glandular hair formation, and the biosynthesis of VOCs and phenolics, we have dark‐forced rhizomatous spearmint plants by growing them hydroponically for 2 weeks. The stems of the newly grown etiolated spearmint shoots developed on the rhizomes of the dark‐forced adult plants were elongated and white; the length of the 14‐day‐old shoots varied between 8 and 15 cm (Figure 4A,B). The leaves were very small, scale‐like and had a pale‐yellow color due to the presence of carotenoids and the lack of chlorophyllous pigments. Similar morphology was described in a study investigating 26 dark‐forced stems, including an unspecified mint species (Mentha sp.) (Skribanek et al. 2000). Shoots developed in the dark (A and B) or in the light (C and D) on adult rhizomatous spearmint plants that were transferred from the soil to hydroponic systems either in the dark (dark‐forced—A and B) or in the light (C and D) for 2 weeks. Before transferring these plants into the dark or light, previous similar shoots, if present, were removed, thus those shown in the picture were newly developed under hydroponic conditions. The light‐grown shoots developed during the same time (2 weeks) were much shorter (only 5–7 cm long). Both the shoots and their leaves contained Chl and were therefore green (Figure 4C,D). The leaves were more differentiated, and their lamina were more extended than in dark‐forced shoots. Chlorophyll Biosynthesis and Plastid Development in Light‐Grown and Dark‐Forced Hydroponically Grown Spearmint Shoots To check the native arrangement of the pigments in the different shoots, we recorded their 77 K fluorescence emission spectra (Figure 5). The spectra of the dark‐forced shoots showed an emission maximum at 633 nm corresponding to nonenzyme bound Pchlide, and a smaller peak at 652.5 nm belonging to photoactive Pchlide (Figure 5, solid line). No fluorescence emission characteristic for chlorophyllous pigments was detected, similar to previous data on dark‐forced stems (Skribanek et al. 2000) and to our data obtained in shoots developing in the soil (Figure 2). To check the photoactivity of the bands, after measurement at 77 K, samples were thawed to 253 K, illuminated, and frozen in liquid nitrogen and measured again (Figure 5, broken line). This experiment confirmed the photoactivity of the 652.5 nm band and the non‐photoactive nature of the 633 nm band and also proved that these samples were fully etiolated and developed in the absence of light. Minor amounts of newly formed Chl(ide) bands could be detected at around 691 nm, overlapping with the vibrational side bands of Pchlide pigments. The spectra of the green shoots developed from the rhizome or the roots in the light were like those observed in the case of any photosynthetically active green leaf (Figure 5, dotted line). These contained Chl fluorescence bands at 685 and 694, belonging to core antennas of photosystem II (PSII), and the third one at 733 nm originating from the PSI light‐harvesting complexes (Kalaji et al. 2017; Lamb et al. 2018). Transmission electron microscopic investigations of leaf mesophyll cells confirmed the presence of etioplasts with large and highly regular PLBs in the case of the hydroponically dark‐forced shoots (Figure 6A) and of chloroplasts with grana and stroma thylakoids in light‐grown young shoots (Figure 6B). In both cases, plastids contained starch (not shown in the chosen etioplast image), indicating the availability of storage materials also in shoots developing with heterotrophic metabolism in the dark. Comparison of leaf etioplasts of shoots developing on naturally etiolated (soil‐grown) and hydroponically dark‐forced rhizomes showed similar etioplast organization, with slight differences; for example, in the density of the stroma or the plastid inner membranes and the regularity of the PLBs (Figure 2) as well as in some quantitative parameters (Table 1). PLBs were present in 74% and 92% of the studied etioplast sections in the naturally etiolated, underground leaves and in the hydroponically dark‐forced leaves, respectively. Starch‐containing plastid sections were observed with similar frequency in the two samples (54% and 42% in underground leaves and dark‐forced leaves, respectively). Significantly larger PLBs and starch grains were detected in the hydroponically dark‐forced leaves, while no difference was observed in the diameter of the plastoglobules among the two different samples (Table 1). Transmission electron micrographs of plastids present in the leaf mesophyll cells of shoots emerging from the rhizomes of 2‐week‐old hydroponically dark‐forced (A) or light‐grown (B) spearmint plants. A representative plastid was selected from approximately 100 plastid sections analyzed per sample. Cw: cell wall; S: starch; V: vacuole. Black arrow points at the prolamellar body present in the etioplast, and the asterisk is placed on a granum in the chloroplast. Scale bar: 1 μm. TABLE 1Prevalence of prolamellar bodies (PLBs) and starch in plastid sections, plastoglobule diameter, and PLB and starch area in the leaves of naturally etiolated underground shoots and in hydroponically dark‐forced shoots as calculated based on transmission electron microscopic images in ImageJ.Underground spearmint leafHydroponically dark‐forced leafPrevalence of PLB containing plastid sections74% (n = 35)92% (n = 26)Prevalence of starch‐containing plastid sections54% (n = 35)42% (n = 26)PLB area (nm2)6662 ± 583a (n = 19)9184 ± 795b (n = 23)Starch grain area (nm2)1595 ± 189a (n = 24)3263 ± 297b (n = 20)Plastoglobule diameter52.3 ± 1.3a (n = 187)54.3 ± 1.9a (n = 77) Note: The values represent the percentage of plastid sections containing PLBs and starch, respectively, as well as means ± standard errors for PLB area, starch area, and plastoglobule diameter. Out of the 100 visually checked plastid sections, approximately 25–35 plastid sections were photographed and analyzed per sample type from one or two different plants. Different letters in each row indicate statistically significant differences following independent unpaired t‐test (p ≤ 0.05). Prevalence of prolamellar bodies (PLBs) and starch in plastid sections, plastoglobule diameter, and PLB and starch area in the leaves of naturally etiolated underground shoots and in hydroponically dark‐forced shoots as calculated based on transmission electron microscopic images in ImageJ. Note: The values represent the percentage of plastid sections containing PLBs and starch, respectively, as well as means ± standard errors for PLB area, starch area, and plastoglobule diameter. Out of the 100 visually checked plastid sections, approximately 25–35 plastid sections were photographed and analyzed per sample type from one or two different plants. Different letters in each row indicate statistically significant differences following independent unpaired t‐test (p ≤ 0.05). To the best of our knowledge, such comparisons of fully naturally etiolated tissues and laboratory‐grown dark‐forced tissues were only performed in pea epicotyls (Vitányi et al. 2013), with similar observations and only slight differences in etiolation symptoms and plastid ultrastructure. Due to the slightly different developmental contexts and environments of the naturally or hydroponically etiolated shoots, slight differences of 1–2 nm were observed in the fluorescence emission maxima of their Pchlide forms, that is, 631 vs. 633 nm for the major non‐photoactive Pchlide form. The observed differences in the newly‐produced Chlide forms after illumination were probably associated with the freeze–thaw cycles resulting in disaggregation of the different pigment complexes (Kósa et al. 2006). Glandular Hair Development of Light‐Grown and Dark‐Forced Hydroponically Grown Spearmint Shoots Monoterpene biosynthesis occurs in glandular trichomes located on the leaves of spearmint. Most glandular trichomes start early in leaf development and begin accumulating monoterpenes before the leaves are 5 mm long (Gershenzon et al. 1989). To investigate how the absence of light affected active ingredients produced by secondary metabolism, especially VOC composition, we first analyzed whether the formation, development, and distribution of capitate and peltate glandular hairs were influenced by the absence of light (Figure 7). The leaves developed on the shoots grown on hydroponic systems in the dark were compared with the leaves of light‐grown shoots of the same age (Figure 4). These young developing leaves—both the dark‐forced and the light‐grown ones—contained several nicely recognizable peltate glandular hairs (Figure 7). The developing trichomes and capitate glandular hairs were not well distinguishable at this stage; therefore, their relative abundance could not be determined. Due to the different sizes of the etiolated and light‐grown young leaves (with the former being slightly smaller, i.e., 5–6 mm2 vs. 9–10 mm2, respectively), the different development and different sizes of the cells on these leaves and the uneven distribution of hairs, we decided to calculate the diameter of the peltate glandular hairs and the density of peltate glandular hairs both at the abaxial and adaxial leaf surfaces (Table 2). The latter was expressed both as the number of peltate glandular hairs per entire leaf surface and also as the number of peltate glandular hairs per unit leaf area (mm2) based on data obtained on the entire leaf surface. These latter data (normalized to unit leaf area) have to be treated carefully and only semi‐quantitatively, because evidently, different leaf regions have uneven distribution of glandular hairs; for example, almost no peltate glandular hairs can be observed on vein regions (Figure 7). Representative scanning electron micrographs of the adaxial (A and C) and abaxial (B and D) leaf surfaces of hydroponically light‐grown (A and B) or dark‐forced (C and D) spearmint shoots. Scale bar: 1 mm. TABLE 2Diameter (in μm) and density (number per entire leaf surface and number per mm2 as determined from data calculated for the entire leaf surface) of peltate glandular hairs on the adaxial and abaxial leaf surfaces of light‐grown and dark‐forced spearmint shoots as calculated based on scanning electron microscopic images in AxioVision LE64 (Carl Zeiss GmbH).Light‐grownLight‐grownDark‐forcedDark‐forcedAdaxial sideAbaxial sideAdaxial sideAbaxial sidePeltate glandular hair diameter (μm)58.7 ± 0.5a (n = 178)59.0 ± 0.3a (n = 443)56.0 ± 0.6b (n = 140)54.2 ± 0.2c (n = 930)Peltate glandular hair density (number/entire leaf)20.1 ± 4.4a (n = 9)115.0 ± 10.0b (n = 9)22.9 ± 4.8a (n = 8)140.5 ± 17.7b (n = 8)Peltate glandular hair density (number/mm2)2.2 ± 0.3a (n = 9)14.2 ± 1.7b (n = 9)5.5 ± 1.7a (n = 8)23.1 ± 2.8b (n = 8) Note: The values represent means ± standard errors. Different letters in each row indicate statistically significant differences (p ≤ 0.05). One‐way ANOVA followed by Tukey's post hoc test (peltate glandular hair diameter), Brown–Forsythe and Welch ANOVA followed by Dunnett's post hoc test (in case of number of hairs per mm2 on the entire leaf surface), and Kruskal–Wallis nonparametric ANOVA followed by Dunn's post hoc test (in case of number of hairs per entire leaf surface) were applied. Diameter (in μm) and density (number per entire leaf surface and number per mm2 as determined from data calculated for the entire leaf surface) of peltate glandular hairs on the adaxial and abaxial leaf surfaces of light‐grown and dark‐forced spearmint shoots as calculated based on scanning electron microscopic images in AxioVision LE64 (Carl Zeiss GmbH). Note: The values represent means ± standard errors. Different letters in each row indicate statistically significant differences (p ≤ 0.05). One‐way ANOVA followed by Tukey's post hoc test (peltate glandular hair diameter), Brown–Forsythe and Welch ANOVA followed by Dunnett's post hoc test (in case of number of hairs per mm2 on the entire leaf surface), and Kruskal–Wallis nonparametric ANOVA followed by Dunn's post hoc test (in case of number of hairs per entire leaf surface) were applied. Our data suggest that peltate glandular hairs develop in the dark, but their size is slightly but significantly smaller (head diameter: 54–56 μm) than that of those developing on light‐grown leaves of the same age (head diameter: 59 μm) (Table 2). On light‐grown shoots, the size of the peltate glandular hairs was the same on the adaxial and abaxial sides, while in the case of the dark‐forced leaves, peltate glandular hairs were slightly but significantly smaller on the abaxial side (54 μm). Similarly to these data, the size of peltate glandular hairs was smaller on etiolated cotyledons and increased after illumination in the case of thyme (Yamaura et al. 1989). Such differences may indicate differences in maturation as well as VOC production among the differently developed leaves. However, this is further confirmed by our measurements on the diameter of peltate glandular hairs of fully developed green leaves (68 ± 2 μm; not shown). Similar data were described for peltate glandular hair diameters of fully expanded leaves of two Chinese M. spicata / M. crispata , that is, 84.3 ± 5.4 and 66.7 ± 3.7 μm diameters, respectively, obtained using SEM following fixation, critical point drying, and gold coating (Yu et al. 2018). The observed slight differences may be due to the different cultivars used, to the different studied developmental stages (very young, still expanding leaves in our case and fully developed leaves in the cited paper), or to the differences between SEM sample preparation. Our unfixed and freshly collected leaves used for SEM analyses underwent slight shrinkage and water loss during the recording of the SEM images for a few minutes. However, this phenomenon equally affected the glandular hairs on all samples, making them comparable. Irrespectively of the light regime, peltate glandular hairs were much more frequent on the abaxial side of the leaves (115 or 141 hairs per leaf, and 14 or 23 hairs per mm2 in the case of light‐grown or dark‐forced shoots, respectively) than on the adaxial side (20 or 23 hairs per leaf, and 2 or 6 hairs per mm2 in light‐grown or dark‐forced shoots, respectively) (Table 2). Similar trends were reported for light‐grown spearmint and other Mentha species, with more peltate glandular hairs being observed on the abaxial than on the adaxial leaf sides (Yu et al. 2018). There was no significant difference in the abundance of peltate glandular hairs on the abaxial or adaxial side of light‐grown and dark‐forced spearmint leaves. This indicated that the absence of light only influenced the size of the peltate glandular hairs but not their development. So far, few studies have investigated the effect of light on the development of peltate glandular hairs. They were observed in similar density both in dark‐ and light‐germinated cotyledons of rosemary (Pávai et al. 2025) and thyme (Yamaura et al. 1989), with slightly more being present in the light than in the dark in the case of thyme. However, these data originate not only from different species but also from the cotyledons of much younger seedlings than the shoots investigated in this work; thus, they are not directly comparable with our data. VOC Profile of Light‐Grown and Dark‐Forced Hydroponically Grown Spearmint Shoots Next, we wanted to compare whether the presence or absence of light influenced the VOC composition produced mostly by these peltate glandular hairs in the light‐grown or dark‐forced young shoots. GC–MS analysis revealed that cis‐dihydrocarvone (Figure 8) was the major component in all examined samples, representing approximately half of the total volatile fraction (49.1% or 52.7%) in light‐grown or dark‐forced shoots, respectively, regardless of illumination conditions (Table 3). This VOC profile differs from the typical carvone‐dominated composition reported for most spearmint populations (Lawrence 2006; Tisserand and Young 2014). Although the relative proportion of dihydrocarvone was reportedly increasing with the developmental age of the plants, it still represented only a minor fraction (from 3% to 9%) according to Murray et al. (1972). Therefore, we have also investigated the VOC of adult plants as well as in vitro plants of the same clone (not shown). In all cases, we found that cis‐dihydrocarvone was dominant. The consistently high cis‐dihydrocarvone proportion suggests that the investigated plants may represent a distinct metabolic chemovariant of spearmint characterized by an altered balance within the carvone biosynthetic pathway. Representative GC–MS chromatograms of hydroponically light‐grown (A) and dark‐forced (B) spearmint shoots—for more information on the identified peaks see Table 3. TABLE 3Relative volatile composition of 2‐week‐old hydroponically light‐grown and dark‐forced spearmint shoots before the analyses.Compound nameRIRelative abundance (%)Light‐grownDark‐forcedβ‐Myrcene (1)9801.0 ± 0.5a tr.b Limonene (2)10237.0 ± 0.7a 16.3 ± 4.4a Eucalyptol (3)10240.1 ± 0.1a —a Menthone (4)11519.0 ± 0.8a 6.0 ± 0.8b Isomenthone (5)11601.3 ± 0.2a tr.b α‐Terpineol (6)1190—a 2.2 ± 0.4b cis‐Dihydrocarvone (7)119149.1 ± 1.0a 52.7 ± 3.1a trans‐Dihydrocarvone (8)120014.8 ± 0.5a 14.7 ± 0.5a Pulegone (9)12320.6 ± 0.1a tr.b Carvone (10)12444.5 ± 0.4a 3.0 ± 0.9a β‐Caryophyllene (11)14222.3 ± 0.5a 0.3 ± 0.3b Germacrene D (12)14838.7 ± 1.5a 4.7 ± 0.3b β‐Selinene (13)14981.6 ± 0.5a —b δ‐Cadinene (14)15140.1 ± 0.1a —a Monoterpenes87.395.0Sesquiterpenes12.75.0 Note: Numbers after the compound names correspond to the peak numbers in Figure 8. The values represent means ± standard errors. Different letters indicate significantly different data at p ≤ 0.05 according to t‐test (n = 10).Abbreviations: RI: calculated linear retention index; tr.: compounds detected in very low amounts, < 0.1%. Relative volatile composition of 2‐week‐old hydroponically light‐grown and dark‐forced spearmint shoots before the analyses. Note: Numbers after the compound names correspond to the peak numbers in Figure 8. The values represent means ± standard errors. Different letters indicate significantly different data at p ≤ 0.05 according to t‐test (n = 10). Abbreviations: RI: calculated linear retention index; tr.: compounds detected in very low amounts, < 0.1%. (−)‐Carvone is synthesized from geranyl pyrophosphate via (−)‐limonene and (−)‐trans‐carveol through several enzymatic steps. The C=C bond in (−)‐carvone might be enantioselectively reduced by various reductase enzymes, leading to the formation of dihydrocarvone isomers (Lawrence 2006). The elevated levels of dihydrocarvone detected in the present study may therefore reflect increased activity of such reductive enzymes or a shift in the regulation of monoterpene metabolism. Dihydrocarvones are used as promising insect repellents (Tripathi et al. 2003), antiacaricidal agents (Yang et al. 2014), or potential inhibitors of potato sprout growth (Oosterhaven et al. 1995) and fungal growth (Díaz et al. 2021). Although the dominant compound remained unchanged, illumination conditions affected several minor volatile components (Table 3). Monoterpenes dominated the VOC composition in both treatments (87.3% or 95.0% in light‐grown or dark‐forced shoots, respectively). Light‐grown shoots contained higher levels of sesquiterpenes (12.7%) than dark‐forced shoots (5.0%), including β‐caryophyllene, germacrene D, β‐selinene, and δ‐cadinene (Table 3). Overall, fewer volatile compounds were detected in the shoots grown in the dark compared with those developed under light conditions. β‐Myrcene, eucalyptol, isomenthone, pulegone, β‐selinene, and δ‐cadinene were detected only in light‐grown samples, while α‐terpineol was found only in dark‐forced samples. The presence of α‐terpineol and the higher limonene ratio may indicate that in the dark, limonene is not converted into carvone/dihydrocarvone due to lower enzyme activity but accumulates, and some of it is converted to α‐terpineol (Lawrence 2006). Several, mostly minor, compounds differed significantly between the two illumination regimes. The relative abundance of β‐myrcene, menthone, isomenthone, pulegone, β‐caryophyllene, germacrene D, and β‐selinene was significantly lower in shoots grown in the dark. These differences indicate that reduced illumination affects the biosynthesis of several mono‐ and sesquiterpene constituents of the VOC profile. Such differences may be associated with changes in terpene biosynthesis and carbon allocation under varying light conditions. Similar differences among the VOC compounds of light‐ and dark‐germinated rosemary seedlings were reported by Böszörményi et al. (2020), with some compounds being only detected in the light‐germinated seedlings. Light quality and intensity affect the production of secondary metabolites (Ali et al. 2024), including VOC composition (De Souza et al. 2016; Hubert‐Schöler et al. 2024; Kivimäenpä et al. 2022; Maffei et al. 1999) as well as leaf morphology (Maffei et al. 1999) of several plants belonging to the Lamiaceae family. Further studies involving additional populations are recommended to confirm the underlying metabolic shift and the taxonomic status of this chemovariant. The compositional differences of VOC profiles were further reflected in the multivariate analysis of the GC–MS dataset (Figure 9A). The first two principal components accounted for 70.1% of the total variance, with PC1 explaining 40.6% and PC2 explaining 29.5%. PCA of GC–MS‐derived VOC profiles indicated a clear separation between light‐grown and dark‐forced spearmint shoots, with samples tending to cluster according to illumination regime along the first principal component. The corresponding loading structure suggests that variation in the relative abundance of specific mono‐ and sesquiterpenes may contribute to this pattern. Limonene, α‐terpineol, and cis/trans‐dihydrocarvone appear to be associated with dark‐forced shoots, whereas eucalyptol, β‐myrcene, β‐caryophyllene, β‐selinene, δ‐cadinene, and germacrene D are more closely aligned with light‐grown samples in the multivariate space (Figure 9A). Although PCA is an exploratory method, the observed separations support the compositional differences detected in the univariate statistical analysis. Principal component analysis (PCA) biplot of the volatile organic compound (VOC) profiles (GC–MS, A) and phenolic profiles (HPLC–UV–MS/MS, B) of 2‐week‐old hydroponically light‐grown (L) and dark‐forced (D) spearmint shoots. The first two principal components explained 70.1% of the total variance in the VOC profiles (PC1: 40.6%; PC2: 29.5%) and 93.8% in the phenolic profiles (PC1: 86.5%; PC2: 7.3%). Score plots represent individual biological replicates, while loading vectors indicate the contribution of selected compounds to sample distribution. Phenolic Composition of Light‐Grown and Dark‐Forced Hydroponically Grown Spearmint Shoots Phenolics are important active ingredients in spearmint plants; therefore, we investigated how the presence or absence of light influenced their synthesis. HPLC–UV–HRMS/MS analysis of extracts prepared from spearmint tissues grown under light and dark conditions enabled the tentative identification of seven compounds. Identification was based on molecular formulas calculated from high‐resolution MS data, characteristic MS/MS fragmentation patterns, and UV spectra consistent with previously reported literature data for compounds identified in Mentha species (Figures S1–S9; Table 4) (Ćavar Zeljković et al. 2021; Cirlini et al. 2016). TABLE 4Mass spectrometry (MS) data of compounds 1–7, obtained from the mass spectra (MS1) and tandem mass spectra (MS2) of these compounds, determined by HPLC high‐resolution Orbitrap‐MS and MS/MS using negative and positive electrospray ionization (ESI) modes.No. a Compound nameFormulaMS mode b Detected ionMeasured mass (m/z) c Mass error (ppm)Characteristic fragment ions, d indicated by mass‐to‐charge ratios (m/z) c and ion structures e 1 Chlorogenic acidC16H18O9 +[M + H]+ 355.1013−2.840191.0552, [quinic acid − H]−; 179.0339, [caffeic acid − H]−; 173.0445, [quinic acid – H‐H2O]− −[M − H]− 353.0878−0.100 2 Luteolin rutinosideC27H30O15 +[M + H]+ 595.1644−2.313449.1069, [M + H‐rhamnose]+; 287.0544, [luteolin + H]+ −[M − H]− 593.1511−0.124 3 RhoifolinC27H30O14 +[M + H]+ 579.1686−3.785433.1119, [M + H‐rhamnose]+; 271.0595, [apigenin + H]+ −[M − H]− 577.15640.262 4 DiosminC28H32O15 +[M + H]+ 609.1798−2.604463.1223, [M + H‐rhamnose]+; 301.0700, [diosmetin + H]+ −[M − H]− 607.1667−0.253 5 HesperidinC28H34O15 +[M + H]+ 611.1955−2.531301.0717 [hesperetin − H]− −[M − H]− 609.1824−0.203 6 Rosmarinic acidC18H16O8 +[M + H]+ 361.0906−3.279197.0447, [dihydroxyphenyllactic acid − H]−; 179.0340, [caffeic acid − H]−; 161.0233, [caffeic acid − H‐H2O]− −[M − H]− 359.0772−0.085 7 LinarinC28H32O14 +[M + H]+ 593.1852−2.094447.1276, [M + H‐rhamnose]+; 285.0753, [acacetin + H]+ −[M − H]− 591.1714−0.911 a Compound numbers correspond to their retention order as determined by HPLC–UV–HR–MS analysis (Figure S1). b High‐resolution mass spectrometry (HR‐MS) was performed in both positive (+) and negative (−) ionization modes using polarity switching. c MS1 and MS2 spectra are provided in Figures S2–S8. d Fragmentation was carried out using data‐independent acquisition (DIA) with collision‐induced dissociation (CID) at 15, 30, and 45 eV; the resulting MS/MS spectra represent the summed ion intensities across all three collision energies. e The differences between measured and calculated fragment ion masses (m/z) were < 5.0 ppm. Mass spectrometry (MS) data of compounds 1–7, obtained from the mass spectra (MS1) and tandem mass spectra (MS2) of these compounds, determined by HPLC high‐resolution Orbitrap‐MS and MS/MS using negative and positive electrospray ionization (ESI) modes. Compound numbers correspond to their retention order as determined by HPLC–UV–HR–MS analysis (Figure S1). High‐resolution mass spectrometry (HR‐MS) was performed in both positive (+) and negative (−) ionization modes using polarity switching. MS1 and MS2 spectra are provided in Figures S2–S8. Fragmentation was carried out using data‐independent acquisition (DIA) with collision‐induced dissociation (CID) at 15, 30, and 45 eV; the resulting MS/MS spectra represent the summed ion intensities across all three collision energies. The differences between measured and calculated fragment ion masses (m/z) were < 5.0 ppm. The analyses confirmed the presence of five flavonoid glycosides (compounds 2–5 and 7) and two caffeic acid derivatives (compounds 1 and 6). These included rhamnosyl‐glucosides of four flavones, that is, luteolin rutinoside (2), rhoifolin (3), diosmin (4), and linarin (7), as well as the flavanone hesperidin (5). The two caffeic acid derivatives were identified as chlorogenic acid (caffeoylquinic acid, 1) and RA (caffeoyl‐dihydroxyphenyllactic acid, 6) (Figures S1–S8; Table 4). In agreement with previous reports, RA was the predominant compound in all extracts, with the highest concentration observed in tissues grown under light conditions (47.9 ± 6.7 mg g−1 dry weight) (Table 5). This value is comparable to that reported for field‐grown spearmint leaves (Li et al. 2021) or aboveground organs (Shekarchi et al. 2012). In contrast, significantly lower levels were detected in dark‐forced tissues (32.4 ± 10.1 mg g−1 dry weight) as determined by t‐test (Table 5). TABLE 5Phenolic composition of hydroponically light‐grown (L1–L5) or dark‐forced (D1–D5) spearmint shoots as determined by HPLC–UV.SamplesHPLC–UV signal intensity of peaks a corresponding to the compounds (1–7)Caffeic acid derivativesFlavonoid glycosides 1 6 2 3 4 5 7 L1484577,95114921020725210858583L26019802,2712359341924,724122727,716L33989579,6951773199818,701168621,023L44930736,187244032412 t5389628,351L54859778,7682362396818,622213623,258D11921749,000326167143018808503D21253328,3001936811208035138D31110385,5901961048696504909D41506404,7631545710176772216D51369473,38330882249822126Mean of L1–L54656694,974 c 2085272918,253200621,786Mean of D1–D51432468,207 c 235969378184578 p b 0.001040.037760.000490.007890.004260.086040.00643 Note: Peak areas of detected metabolites are presented as a heat map using column‐wise minimum‐to‐maximum color scaling. Color scale: minimum values are shown indark green, the 50th percentile (median) in yellow, and maximum values in red. Numerical values represent peak areas obtained from chromatograms. a Sum of UV signal intensity across 230–600 nm range. b To compare the peak areas between the samples grown under light and dark conditions, an independent samples t‐test (two‐tailed) was employed. A p value of less than 0.05 was considered statistically significant. c Using the calibration curve of the isolated RA (Figure S9), its concentration was quantified and expressed as mg g−1 dry weight (mean ± SD). Tissues grown under light and dark conditions contained 47.9 ± 6.7 mg g−1 and 32.4 ± 10.1 mg g−1, respectively, the latter being significantly lower based on the t‐test. Phenolic composition of hydroponically light‐grown (L1–L5) or dark‐forced (D1–D5) spearmint shoots as determined by HPLC–UV. Note: Peak areas of detected metabolites are presented as a heat map using column‐wise minimum‐to‐maximum color scaling. Color scale: minimum values are shown indark green, the 50th percentile (median) in yellow, and maximum values in red. Numerical values represent peak areas obtained from chromatograms. Sum of UV signal intensity across 230–600 nm range. To compare the peak areas between the samples grown under light and dark conditions, an independent samples t‐test (two‐tailed) was employed. A p value of less than 0.05 was considered statistically significant. Using the calibration curve of the isolated RA (Figure S9), its concentration was quantified and expressed as mg g−1 dry weight (mean ± SD). Tissues grown under light and dark conditions contained 47.9 ± 6.7 mg g−1 and 32.4 ± 10.1 mg g−1, respectively, the latter being significantly lower based on the t‐test. Similarly, the levels of chlorogenic acid (compound 1) and all flavone glycosides (compounds 2, 3, 4, and 7) were significantly higher in tissues grown under light conditions (Table 5). The only compound whose quantity was not affected by the presence or absence of light was the flavanone, hesperidin (compound 5) (Table 5). The effect of light on metabolite composition was further evaluated using PCA. The analysis of the HPLC–UV data revealed a clear separation between light‐grown and dark‐forced spearmint shoots (Figure 9B). The first principal component (PC1) explained 86.5% of the total variance, representing the main source of variation in the dataset, while the second principal component (PC2) accounted for an additional 7.3%. Together, the first two components captured 93.8% of the total variability. Samples clustered according to illumination regime with no observable overlap. This separation suggests that light availability strongly influences the accumulation of phenolic compounds in developing shoots. While GC–MS analysis revealed that illumination primarily affected minor components, the HPLC–UV data demonstrated a much stronger light‐dependent response in phenolic metabolites. The light‐induced increase in caffeic acid derivatives and flavone glycosides suggests their potential role in photoprotective mechanisms in spearmint. This observation is consistent with previous studies demonstrating the light‐protective functions of these classes of plant metabolites. Caffeic acid and its derivatives, as well as flavonoids, are effective antioxidants that scavenge reactive oxygen species that are often formed by various abiotic stresses, including strong light or UV stress (Agati et al. 2025; Mughal et al. 2024). In spearmint, higher phenolic index and anthocyanin levels were reported following UV‐B elicitation (Ali et al. 2024), indicating the potential influence of light on phenolic biosynthesis in this species. In another study, the RA content of spearmint leaves was shown to be influenced by the developmental stage of the plant (i.e., vegetative vs. flowering stage) and by the photoperiod, with its accumulation being observed in the case of longer photoperiods (Fletcher et al. 2010). To further analyze and compare our HPLC–UV–MS/MS data, we also performed PCA using the built‐in function of MZmine. We used auto‐scaling and zero‐missing‐value imputation as settings, and the features were evaluated by their height. We got a very clear separation of the light and dark‐forced shoots based on this (Figure S10A). In addition to the phenolic secondary metabolites presented above, these analyses included several other primary metabolites (Figure S10B), some of which could be tentatively identified, while others remained unknown. The samples could be differentiated by a sharp division along the x‐axis. Scattering along the y‐axis reflects inter‐individual differences rather than treatment effects. Thus, on the loadings plot, extremities along the x‐axis were investigated. The most descriptive molecules of the samples grown in the dark were amino acids (e.g., glutamic acid) and other small, nitrogen‐containing compounds. Glutamine and glutamate are key metabolites of dark‐grown plants involved in the biosynthesis of other amino acids, in the first steps of the porphyrin/Chl biosynthesis, and in nitrogen metabolism in general, and they can also be formed through nitrogen assimilation in the absence of light (Yoneyama and Suzuki 2020). Because these compounds showed no retention under the previously described chromatographic conditions, they could not be adequately resolved, and their identification therefore remained tentative. In the samples grown in light, the presence of the flavonoids and chlorogenic acid is the most characteristic feature. Besides these phenolic compounds, the derivatives of the plant hormone jasmonic acid (e.g., tuberonic acid) stood out. Jasmonic acid and its derivatives are involved in light signalization processes that drive photomorphogenesis in plants (Kazan and Manners 2011). For example, in various plants, it was shown that tuberonic acid is involved in inhibiting hypocotyl or stolon elongation, which are typical etiolation responses caused by gibberellic acid. This is in line with our observations about the shorter shoots of the light‐grown plants and their photomorphogenesis. Conclusions In this work, we provided evidence for natural etiolation and thus etioplast and PLB formation in both the leaves and the stems of spearmint shoots developing in the soil from underground rhizomes. Soil‐grown and hydroponically dark‐forced shoots developing from the rhizomes had similar etiolation symptoms, plastid ultrastructure and Chl biosynthesis. This indicated that etiolated plants grown under laboratory conditions represent good models for natural etiolation. Our data not only outline the importance of studies dealing with understanding etiolation and the greening of etiolated seedlings (Armarego‐Marriott et al. 2020; Solymosi and Mysliwa‐Kurdziel 2021; Solymosi and Schoefs 2010) but also point to several new research directions. These include the analysis of the latter processes under fully natural conditions, as well as the investigation of how soil‐related stressors, such as high salinity (Ounoki et al. 2021, 2023; Sóti et al. 2023), drought, or heavy metal pollution, influence them, or observations on potential biotic interactions (e.g., with the soil microbiome) of such naturally etiolated, soil‐buried organs. In this respect, it is important to outline the clear difference and usefulness of such plant experimental models when compared to studies carried out on etiolated seedlings developing from seeds: in case of shoots developing in the soil from the rhizome of plants having autotrophic green aboveground shoots, or to a lesser extent even in case of hydroponically dark‐forced plants having large rhizomes and root system, the amount of storage materials available for the etiolated shoots is almost unlimited, thus plant metabolism is not influenced by energy, carbon, nitrogen, or sulfur limitations during their growth. This might explain the slight differences observed. On the other side, it is hard to exactly determine the age and follow the development of naturally etiolated soil‐grown shoots without disturbing the plants. Due to this, but also to the cracks on the soil surface and other factors like naturally infiltrating sunlight, higher biological variability is expected among these shoots, with some already containing traces of Chl pigments close to the soil surface. More importantly, for analyses of VOCs, while removing shoots from the soil or rinsing them with water, some of the peltate glandular hairs might be scrapped off and plants might also be injured. Taken together, both etiolated systems have their own merits and might be used in experiments to provide complementary information in adequate settings. Another interesting aspect related to the etiolation of spearmint shoots is how the absence of light influenced their peltate glandular hair differentiation and secondary metabolism. Comparison of light‐grown and dark‐forced shoots of hydroponically grown adult plants indicated smaller leaves developing on elongated etiolated shoots, with significantly smaller glandular hairs, but similar density to that observed in the light (i.e., with more glandular hairs being present on the abaxial side than on the adaxial side of the leaves). Based on the PCA of both the volatile and phenolic profiles, light‐grown and dark‐forced shoots could be clearly distinguished, with fewer compounds being present and generally at significantly lower relative abundances (e.g., menthone, β‐caryophyllene, germacrene D for the VOCs; caffeic acid derivatives such as RA and chlorogenic acid and the flavone rhamnosyl‐glycosides: luteolin rutinoside, rhoifolin, diosmin, and linarin for phenolics) in dark‐forced shoots. Concerning the VOC profile, only minor differences were observed: α‐terpineol was only detected in dark‐forced shoots, while compounds like β‐myrcene, menthone, isomenthone, pulegone, β‐caryophyllene, germacrene D, and β‐selinene were present in significantly higher amounts in light‐grown samples. On the other side, it is important to mention that a much stronger light‐dependent response was observed in the phenolic metabolism of the leaves than in the biosynthesis of the volatile compounds. Additional MZmine analyses revealed the presence of molecules involved in skotomorphogenesis (glutamine) or photomorphogenesis (tuberonic acid) in dark‐forced and light‐grown seedlings, respectively. Taken together, these data provide novel insights into the primary and secondary metabolism of hidden and less studied organs of spearmint plants and thus represent good models to further investigate plant organ development and differentiation under peculiar natural conditions. Author Contributions K.S. conceived, designed and oversaw the experiments, performed 77 K fluorescence spectroscopy and TEM analyses. B.S.‐T. took care of the plant material and sample preparation and performed SPME‐GC–MS analysis, preparative HPLC and HPLC–UV–HR–MS/MS analyses, RA isolation, data analysis, and part of the statistical analysis. A.S. performed SEM measurements, A.D. took part in 77 K fluorescence spectroscopic analyses, A.B. verified the SPME‐GC–MS analytical methods, and E.E. performed part of the HPLC data analysis and related statistical analysis. I.B. performed and oversaw preparative HPLC and HPLC–UV–HR–MS/MS analysis. N.G. performed the statistical analysis with MZmine software. K.S. and B.S.‐T. wrote the first manuscript draft and finalized it with input from all authors. Funding This work was supported by Nemzeti Kutatási Fejlesztési és Innovációs Hivatal (NKKP Advanced Grant 153509). Supporting information Figure S1: Representative HPLC–UV chromatogram of light‐grown spearmint shoot extract. Figure S2: UV spectrum and mass spectra of chlorogenic acid. Figure S3: UV spectrum and mass spectra of luteolin rutinoside. Figure S4: UV spectrum and mass spectra of rhoifolin. Figure S5: UV spectrum and mass spectra of diosmin. Figure S6: UV spectrum and mass spectra of hesperidin. Figure S7: UV spectrum and mass spectra of rosmarinic acid. Figure S8: UV spectrum and mass spectra of linarin. Figure S9: HPLC–UV calibration curve of rosmarinic acid. Figure S10: Principal component analysis (PCA) biplot (A) and PCA scattergram (B) generated from the HPLC–UV metabolite profiles of spearmint shoots. Acknowledgments This paper is dedicated to the late Prof. Béla Dános (1929–2025)—colleague, mentor, friend, teacher, and expert of pharmacobotany at the ELTE Department of Plant Anatomy. We are grateful to Csilla Gergely (ELTE Department of Plant Anatomy) for her precious assistance with electron microscopic sample preparation and great patience with ultrathin sectioning, and to Viktória Kardos (ELTE Department of Plant Anatomy) for helping us with the experiments in the soil. The authors would like to thank Prof. János Podani (ELTE Department of Plant Systematics, Ecology and Theoretical Biology) for his assistance with the use of the SYN‐TAX software for statistical analyses. We are also grateful to Judit Takács (ELTE Department of Plant Anatomy) for her precious help with spearmint collection. This work was supported by the Bolyai János Research Scholarship of the Hungarian Academy of Sciences to K.S. and the NKKP Advanced grant 153509 of the National Research, Development and Innovation Office in Hungary to K.S. Artificial intelligence tools (ChatGPT and Gemini) were used only in a few cases to improve the English phrasing of complex sentences. Photoshop's built‐in AI tool was used to remove the sample holder visible in the background behind the leaves in the SEM images shown in Figure 7. No further modifications were made, and the authors retain full responsibility for all scientific content, data interpretation, and conclusions, as well as for the overall accuracy and integrity of the work. Szögi‐Tatár, B. , Dobi A., Böszörményi A., et al. 2026. “Natural Etiolation in Spearmint ( Mentha spicata L. var. crispa ) Shoots Emerging From Rhizomes—Plastids, Glandular Hairs, and Phytochemical Profile.” Physiologia Plantarum 178, no. 5: e71109. 10.1111/ppl.71109. Handling Editor: Anja Krieger‐Liszkay Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions. References Associated Data Supplementary Materials Figure S1: Representative HPLC–UV chromatogram of light‐grown spearmint shoot extract. Figure S2: UV spectrum and mass spectra of chlorogenic acid. Figure S3: UV spectrum and mass spectra of luteolin rutinoside. Figure S4: UV spectrum and mass spectra of rhoifolin. Figure S5: UV spectrum and mass spectra of diosmin. Figure S6: UV spectrum and mass spectra of hesperidin. Figure S7: UV spectrum and mass spectra of rosmarinic acid. Figure S8: UV spectrum and mass spectra of linarin. Figure S9: HPLC–UV calibration curve of rosmarinic acid. Figure S10: Principal component analysis (PCA) biplot (A) and PCA scattergram (B) generated from the HPLC–UV metabolite profiles of spearmint shoots. Data Availability Statement The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to privacy or ethical restrictions.