Volatile organic compound (VOC) profile and plantlet growth of Aeollanthus suaveolens under conventional and alternative membrane systems

In conventional in vitro culture, plantlets are kept in closed containers to avoid contamination and drying of the explant. Ventilation inside the containers improves plant growth and affects secondary metabolism, as it modifies the microenvironmental conditions. Therefore, the objective of this study was to evaluate the effect of the use of ventilation systems on the growth characteristics and production of photosynthetic pigments and volatile organic compounds (VOCs) in Aeollanthus suaveolens Mart. ex Spreng. (Lamiaceae) cultured in vitro. Nodal segments containing one pair of leaves were cultured in a conventional system (NMS) and in natural ventilation systems with one (AMS1), two (AMS2) and four (AMS4) porous membranes. At 40 days, the plantlets were evaluated for growth, VOC concentration, and photosynthetic pigment production. The number of porous membranes used in the vial cap affected growth, photosynthetic pigments and VOCs. A higher number of porous membranes (AMS4) led to greater dry weight accumulation, increased production of photosynthetic pigments, and enhanced synthesis of (Z)-β-farnesene. Lower growth and fewer photosynthetic pigments, and increase linalool acetate synthesis were observed in the culture without the use of porous membranes (NMS). The leaf area of plantlets cultivated with the use of four membranes was 3.8 times greater than that of plantlets cultivated without the use of membranes. For the photoautotrophic cultivation of A. suaveolens in vitro, the use of natural ventilation with four membranes is recommended because it promotes better growth, increases the production of photosynthetic pigments and is superior to the conventional sealed system. Alternative membrane systems led to differentiation in the contents of the main volatile organic compound and improve anatomical characteristics, accumulation of photosynthetic pigments and dry weight.


Introduction
The species Aeollanthus suaveolens Mart. ex Spreng. (Lamiaceae), an aromatic plant of African origin, was introduced in Brazil for medicinal purposes (Ferreira et al. 2017). In the Amazon region, it is popularly known as "Catinga de mulata" and is used in scent baths, in religious rituals, and in homemade perfumes (Harley 2012). In addition, it is used in folk medicine to combat fever, headache, skin, and eye disorders, and as a sedative and anticonvulsant (Simionatto et al. 2007). Studies have shown that crude extracts obtained from A. suaveolens have antibacterial activity (Veloso et al. 2020) and that the essential oil has potential sedative-hypnotic effects when used as a nanoemulsion (Ferraz et al. 2020) and has antimicrobial activity (Ngo-Mback et al. 2019).
Tissue culture techniques are important biotechnological tools for the multiplication and analysis of medicinal plants (Siahsar et al. 2011). In conventional micropropagation, the culture container is usually sealed to prevent contamination and dehydration of the explants and culture medium (Schuelter et al. 2015). However, the cap or the type of seal can limit gas exchange between the in vitro and ex vitro environments, causing morphophysiological disorders (Saldanha et al. 2012;Lazzarini et al. 2019). The microenvironment formed inside the vial in conventional culture shows a high relative humidity, ethylene accumulation, and a low carbon dioxide (CO 2 ) concentration (Chen 2006;Schuelter et al. 2015). Another important characteristic of in vitro culture is the presence of sucrose in the culture medium, which is considered the main or only source of carbon and energy for plant growth (Xiao et al. 2011). According to Kozai and Xiao (2008), the presence of sucrose in the culture medium increases the risk of microbial contamination.
Plantlets grown in conventional micropropagation systems may have anatomical and physiological abnormalities. Some of these problems include reduced photosynthetic activity, a low chlorophyll content, abnormal stomatal opening, lack of a leaf cuticle layer, and abnormal leaf parenchyma and xylem . These morphophysiological changes cause high plant mortality during acclimatization (Alvarez et al. 2012). The use of a natural ventilation system in in vitro culture is an alternative to reduce these problems.
The use of a natural ventilation system has great advantages for micropropagation. Such systems decrease relative humidity, increase aeration, and maintain adequate CO 2 concentrations to stimulate photosynthesis (Saldanha et al. 2012;Moreira et al. 2013). The use of natural ventilation systems in in vitro culture has shown positive responses in terms of the growth and accumulation of secondary metabolites in some species: Lippia dulcis (Trev.) (Rocha et al. 2022); Lippia gracilis Schauer (Verbenaceae) (Lazzarini et al. 2019) and Plectranthus amboinicus (Lour.) Spreng. (Lamiaceae) (Silva et al. 2017).
The objective of this study was to evaluate the effects of natural ventilation systems on the growth, production of photosynthetic pigments, and volatile organic compounds (VOCs) in A. suaveolens plantlets grown in vitro.

In vitro establishment of plant material
Voucher specimens are deposited in the herbarium of EMBRAPA/Eastern Amazon in Belém, Pará, Brazil, under record number IAN 184,689. The A. suaveolens seeds obtained from mother plants grown in a greenhouse were used in this study. The seeds were treated with sodium hypochlorite (1% active chlorine) for 5 min under constant stirring. Subsequently, the seeds were washed with autoclaved distilled water three times in a laminar flow hood. The seeds were then inoculated in test tubes containing 15 mL of MS medium (Murashige and Skoog 1962) supplemented with 30 g L −1 sucrose and 6 g L −1 agar, and the pH was adjusted to 5.7 ± 0.1 before autoclaving at 121 °C at 0.1 atm for 20 min. The tubes were kept in a growth room under cold-white fluorescent lamps (OSRAM®, Brazil) with an intensity of 39 µmol m −2 s −1 , 16 h photoperiod and temperature of 25 ± 1 °C. After 30 days, the grown plantlets were multiplied in vials containing 40 mL of MS medium.

Alternative membrane system (AMS)
Nodal segments (± 1 cm) with one pair of leaves were cultured in vitro in a conventional sealed system (no-membrane system-NMS) or alternative membrane systems (AMSs). For NMS, the vials were sealed with polypropylene rigid closure, as conventional systems. In the AMSs, holes of 10 mm in diameter were made in the caps of the vials and then covered with membranes. Porous membranes were prepared using one layer of polytetrafluoroethylene film (Amanco®) and three layers of microporous tape (Cremer®), similar to manufactured by Saldanha et al. (2012) (Fig. 1). The explants were removed from plantlets established in vitro and inoculated in 250-mL vials containing 45 mL of MS medium without sucrose supplemented with 5.5 g L −1 agar (Himedia®, type I), and the pH was adjusted to 5.7 ± 0.1 before autoclaving at 121 °C at 0.1 atm for 20 min. After inoculation, the tubes were placed in a growth room under cold-white fluorescent lamps (OSRAM®, Brazil) with an intensity of 39 μmol m −2 s −1 , 16 h photoperiod and temperature of 25 ± 1 °C. The natural ventilation systems were established by placing polytetrafluoroethylene (PTFE) membranes on the caps of the culture vials. The membranes were mounted according to Saldanha et al. (2012).
The experiment was conducted in a completely randomized design totaling four treatments: NMS and AMSs with one (AMS1), two (AMS2) and four (AMS4) porous membranes. Each treatment had five replicates, with four vials per replicate, and three explants per vial.
After 40 days, the plantlets were evaluated for the shoot number (SN), leaf number (LN), root number (RN), shoot length (SL, cm), length of longest root (LLR, cm), total Fig. 1 Scheme of the vials with used membranes and Aeollanthus suaveolens plantlets grown in MS medium without sucrose addition in a conventional system (NMS) and in natural ventilation systems with one (AMS1), two (AMS2) and four (AMS4) porous membranes leaf area (TLA, cm 2 ), leaf dry weight (LDW, mg), stem dry weight (SDW, mg), root dry weight (RDW, mg), total dry weight (TDW, mg), concentration of photosynthetic pigments, and presence of VOCs. To determine the dry weight, the leaves, stems and roots were placed in paper bags and dried in a forced air oven at 35 °C to constant weight. TLA was measured in WinFOLIA™ software using an Epson Perfection V700 photo scanner to scan all leaves present in five plantlets from each treatment. The leaf area ratio (LAR = TLA/TDW), specific leaf area (SLA = TLA/LDW), leaf weight ratio (LWR = LDW/LDW + SDW), and specific leaf weight (SLW = LDW/TLA) were calculated according to Benincasa (2003). In addition, a ratio was established between the parameters obtained for AMS4 and NMS (AMS4/NMS).

Quantification of photosynthetic pigments
The extraction of the pigments chlorophyll a and b, total chlorophyll (a + b) and carotenoids was performed according to the method described by Costa et al. (2014a) with modifications to the amount of plant material and the use of dimethyl sulfoxide (DMSO) not saturated with CaCO 3 as a solvent. A leaf fresh weight of 50 mg was used without the midrib. Next, the plant material was incubated in 10 mL of DMSO in Falcon tubes wrapped in aluminum foil and placed in an oven at 65 °C for 24 h. After this period, 3 mL aliquots of the solutions were transferred to a 3 cm 3 quartz cuvette to perform spectral absorbance readings in a Tecan Infinite M200 PRO microplate reader operated with the software I-control® version 3.37 at 480, 649 and 665 nm wavelengths for carotenoids and chlorophyll b and a, respectively. The blank contained DMSO only. All readings were performed in triplicate.
From the readings obtained, the concentrations of chlorophyll a and b and carotenoids were determined using the equations proposed by Wellburn (1994): chlorophyll a = (12 .47 × H 665 ) − (3.62 × H 649 ); chlorophyll b = (25.06 × A 649 ) − (6.5 × A 665 ); and carotenoids: (1000 × A 480 − 1.29 × C a − 53 .78 × C b )/220. Total chlorophyll was calculated by summing the results of the equations for chlorophyll a and b, with the results expressed in mg/g.

Analysis of volatile compounds by headspace GC/MS
For the analysis of VOCs, dry leaves of A. suaveolens from each treatment were used. Individual 90 mg aliquots (leaf dry weight) were added in triplicate to 20 mL headspace vials and sealed with silicone/PTFE caps until analysis.
The static headspace technique was used in the extraction of VOCs from A. suaveolens. For this purpose, a Combi PAL Autosampler System automatic headspace extractor/ sampler (CTC Analytics AG, Switzerland) coupled to the gas chromatography/mass spectrometry (GC/MS) system was used. The extraction parameters applied were as follows: sample incubation temperature of 110 °C for 30 min, syringe temperature of 120 °C, gas-phase automatic injection and 500 μL injection volume. The volatile fraction was analyzed in an Agilent® 7890A GC system coupled to an Agilent® MSD 5975C mass selective detector (Agilent Technologies, California, USA) operated with electron impact ionization at 70 eV in scanning mode, a mass acquisition range of 40-400 m/z and a speed of 1.0 scan/s. An HP-5MS fused silica capillary column (30 m length × 0.25 mm internal diameter × 0.25 μm film thickness) (California, USA) was used. Helium gas was used as the carrier gas at a flow rate of 1.0 mL/min; the injector and mass spectrometer transfer line temperatures were kept at 230 °C and 240 °C, respectively. The initial oven temperature was 60 °C, followed by a temperature ramp of 3 °C/min to 230 °C, followed by a ramp of 10 °C min − 1 to 250 °C, and a hold for 1 min. The injection was performed in split mode at an injection ratio of 1:20. The concentrations of the constituents present in the volatile fraction were expressed as the percent normalized area of the chromatographic peaks.
The volatile fraction constituents were identified by comparing their linear retention indices relative to those of a coinjected standard solution of n-alkanes (C 8 -C 20 , Sigma-Aldrich®, St. Louis, USA) and by comparing the mass spectra with those from the NIST library and from Adams (2007). The retention index was calculated using the equation proposed by van den Dool and Kratz (1963), and for the attributions, the retention indices in Adams (2007) were consulted.

Statistical analysis
The data were subjected to analysis of variance by the F-test, and the means were compared by the Scott-Knott test at 5% probability. The statistical program SISVAR® (Ferreira 2011) was used for data processing. Principal component analysis (PCA) was used to study the influence of different AMSs on the volatile constituents and growth parameters of A. suaveolens. PCA was performed in Statistica® version 13.4 (StatSoft, Tulsa, OK, USA).

Growth analysis
The number of porous membranes influenced the growth of A. suaveolens cultured in vitro in MS medium without sucrose supplementation (photoautotrophic). Plantlets grown in the AMSs showed better growth variable responses than those grown in NMS. AMS4 had the highest means for all growth variables analyzed. Conversely, plantlets grown in NMS showed the lowest values for all evaluated parameters (Table 1, Fig. 2). The SL and LLR values obtained for AMS4 were 2.7-fold and 1.8-fold greater, respectively, than those of plantlets grown in NMS (Table 1). The plantlets cultured with ventilation were much more vigorous (Fig. 1).
The AMS4/NMS ratio showed that there was a gain in dry weight of between 4.5 and 6.1 times in AMS4 compared to that in NMS. This finding may be related to the photoautotrophic conditions promoted by the greater number of porous membranes, which allowed greater nutrient absorption due to greater evapotranspiration of the culture medium and greater gas exchange (Fig. 1). The poor growth of the plantlets grown in NMS is explained by the absence of sucrose in the culture medium and low gas exchange.
Plantlets grown with membranes showed higher dry weight accumulation, indicating the importance of gas exchange for in vitro morphogenesis in this species. This result may be related to the increased photosynthetic rate caused by the greater availability of CO 2 in AMSs than in NMS. According to Arigita et al. (2010), gas exchange between the external and internal environment in a natural ventilation system favors the effective absorption of nutrients from the culture medium by the plantlets, increasing their growth.
Several studies have reported that increased gas exchange in the vials used for in vitro propagation of Table 1 In vitro growth of Aeollanthus suaveolens grown for 40 days in natural ventilation systems with one (AMS1), two (AMS2) and four (AMS4) porous membranes and in a conventional system (NMS) SN shoot number, SL shoot length, LN leaf number, RN root number, LLR length of the longest root, AMS4/ NMS ratio between the values for AMS4 and NMS, CV coefficient of variation Means (± standard deviation) in the same row followed by the same letter do not differ significantly according to the Scott-Knott test at the 5% probability level  Saldanha et al. (2012), it was found that AMSs added 30 g L −1 sucrose improved the in vitro growth of Pfaffia glomerata (Spreng.) Pedersen. AMSs with one and two membranes containing 30 g L −1 sucrose optimized the in vitro growth of apical and nodal segments of Plectranthus amboinicus (Lour.) Spreng (Silva et al. 2017). An AMS with 30 g L −1 sucrose was also beneficial for the in vitro propagation of Azadirachta indica A. Juss (Rodrigues et al. 2012). Lazzarini et al. (2019) also reported that an AMS with four porous membranes added 15 g L −1 sucrose led to better growth and increased leaf area in Lippia gracilis.
Regarding TLA, the AMS4 treatment was superior to all other cultivation systems (Table 2). It is believed that the AMS4 treatment favored greater entry of CO 2 and, consequently, greater photosynthetic capacity, leading to greater plantlet growth. According to Souza et al. (2014), the higher the leaf area is, the higher the photosynthetic rate of the plants will be. The LAR, SLA and LWR were higher in NMS and AM1.
LAR is the ratio between the leaf area responsible for the interception of light energy and CO 2 and the TDW of the plant. LAR declines as the plant grows due to self-shading, with a tendency to decrease the useful leaf area (Benincasa 2003). The plantlets grown in the NMS treatment had a higher LAR due to lower TDW accumulation, and those grown in the AMS4 treatment had a lower LAR, suggesting that the plantlets grown in the AMS4 treatment were more efficient in dry weight production. Lazzarini et al. (2019) worked with explants with and without leaves and observed a lower LAR in explants with leaves, which were more efficient in dry weight accumulation. The same was observed by Souza et al. (2014) and Chagas et al. (2013) in Rosmarinus officinalis L. and Mentha arvensis L., respectively, in a greenhouse.
SLA is expressed as the ratio between TLA and LDW. It is considered an important factor from a physiological standpoint because it describes the allocation of leaf biomass per unit area (Scalon et al. 2017). Plantlets cultured in the NMS and AMS1 treatments had a higher SLA, indicating that a larger leaf area was needed to accumulate dry weight. The SLA was lower with a higher number of membranes (AMS2 and AMS4), indicating that the plantlets were more efficient in biomass allocation. According to Barreiro et al. (2006), decreases in SLA indicate an increase in leaf thickness resulting from the increase in the number and size of plant cells. SLW is also indicative of leaf thickness (Benincasa 2003). The A. suaveolens plantlets cultivated with greater aeration (AMS2 and AMS4) had a higher SLW, which indicates that they had a greater leaf thickness (Table 2).
LWR represents the fraction of dry weight that is not exported from the leaves to other plant parts. The NMS and AMS1 treatments showed a small difference in LWR from the other ventilation systems (Table 2). Thus, it can be inferred that the export of dry weight from the leaf to other parts of the plantlet was balanced and that under lower ventilation, export was lower. Lazzarini et al. (2019) reported no difference in LWR between ventilation systems, except when the presence or absence of leaves on the initial explant differed.
In young plantlets, which consist mostly of leaves, the LWR values are high and decrease over time because other plant parts emerge and grow from the material exported from the leaves. LWR is thus an important growth index for plants whose leaves are of economic interest (Costa et al. 2014b).

Quantification of photosynthetic pigments
The increase in gas exchange increased the production of photosynthetic pigments in A. suaveolens plantlets ( Table 3). The AMS2 and AMS4 treatments showed the maximum accumulation of chlorophyll a, total chlorophylls, and carotenoids. AMS1 had the highest concentration of chlorophyll b, and the other treatments (NMS, AMS2, and AMS4) did not differ significantly from each other. The photosynthetic activity of plantlets may vary according to genotype and Table 2 Total leaf area (TLA), leaf area ratio (LAR), specific leaf area (SLA), leaf weight ratio (LWR) and specific leaf weight (SLW) of Aeollanthus suaveolens grown in vitro at 40 days in different culti-vation systems (a conventional system (NMS) and natural ventilation systems (AMSs) with 1, 2 and 4 porous membranes) CV coefficient of variation Means (± standard deviation) in the same row followed by the same letter do not differ significantly according to the Scott-Knott test at the 5% probability level In vitro culture may result in some species having weak photosynthetic activity and others having higher photosynthetic activity. Acacia suaveolens was found to be a species with high photosynthetic activity. Other studies also reported increased photosynthetic activity with ventilation inside the vials for Plectranthus amboinicus (Lour.) Spreng (Silva et al. 2017) and Juglans regia L. ). According to Saldanha et al. (2014), the increase in CO 2 concentration inside the vials during in vitro growth leads to high photosynthetic rates in plants grown under photoautotrophic conditions. However, other species, such as Lippia gracilis (Lazzarini et al. 2019), Musa cavendishii (Schoh et al. Table 3 Concentration of photosynthetic pigments of Aeollanthus suaveolens plantlets grown in vitro for 40 days in different culture systems: a conventional system (NMS) and natural ventilation systems with one (AMS1), two (AMS2) and four (AMS4) porous membranes CV coefficient of variation, FW fresh weight Means (± standard deviation) in the same row followed by the same letter do not differ significantly according to the Scott-Knott test at the 5% probability level 1989), and Prunus avium (Righetti et al. 1990) showed weak photosynthetic activity. Lazzarini et al. (2019) reported that photosynthetic pigments decreased with the increase in the number of porous membranes in the culture vial. The CO 2 content in the culture vial and the light intensity are the limiting factors of photosynthesis in plantlets grown in vitro (Kozai et al. 1990;Kozai 2010). AMSs maintain an adequate CO 2 concentration within the culture vial to stimulate photosynthesis (Saldanha et al. 2012). According to Zobayed et al. (2001), it is important for air exchange to occur between the environment and the vial so that plantlets grown in vitro develop the ability to build their photosynthetic apparatus and grow independent of exogenous carbon sources.

Effect of the number of membranes on VOCs by headspace GC/MS
The number of porous membranes affected the VOC concentrations of A. suaveolens grown in vitro (Table 4). Headspace GC/MS analysis identified 18 compounds with a total content above 96%. Five major compounds were identified, and their sum ranged from 88.38 to 90.52%. The compounds were linalool (9.78 to 11.10%), linalool acetate (26.64 to 32.31%), α-santalene (6.57 to 8.25%), (Z)-β-farnesene (38.65 to 42.05%), and massoia lactone (2.01 to 2.44%). Regardless of the treatment, the VOCs identified were grouped into four main classes: hydrocarbon monoterpenes, oxygenated monoterpenes, hydrocarbon sesquiterpenes, and lactones (Table 4). Oxygenated monoterpenes tended to decrease with an increase in the number of membranes, and the opposite was observed for hydrocarbon sesquiterpenes.
PCA was performed to understand the relationship among VOCs, dry weight accumulation and photosynthetic pigments and how these parameters vary according to the different number of porous membranes in the caps of the culture vials. For the analysis of variations in composition and treatment, 17 compounds and four treatments were subjected to PCA. The two principal components (PCs) together explained 95.44% of the total variance, with the first principal component (PC1) explaining 78.53% and the second principal component (PC2) explaining 16.91% (Fig. 3). The score plot revealed three distinct clusters (Fig. 3). The results indicated that A. suaveolens grown with greater ventilation inside the vial (AMS4 and AMS2) had greater synthesis of (Z)-β-farnesene and that samples grown with lower ventilation had greater synthesis of linalool acetate. In AMS1, there was greater synthesis of linalool and massoia lactone. There was also a strong correlation between the AMS4 treatment and all evaluated parameters, including photosynthetic pigments, dry weight production, TLA, NL and LLR and SL. The number, content and profile of VOCs and dry weight production in Lippia gracilis grown in vitro also varied according to different natural ventilation systems (Lazzarini et al. 2019). Silva et al. (2017) reported that the highest carvacrol content and growth parameters of Plectranthus amboinicus were obtained when the plantlets were cultivated with one or two porous membranes.
PCA also confirmed that (Z)-β-farnesene were negatively correlated with linalool acetate, linalool and massoia Spreng. under different membrane systems. NMS no-membrane system, AMS alternative membrane system with 1, 2 or 4 filter membranes lactone. In addition, all growth parameters studied showed a negative correlation with lower ventilation. In the present study, the number of membranes showed a positive and significant correlation with all studied dry weight production and VOC characteristics of A. suaveolens. Thus, improved ventilation with alternative membranes may produce betterquality plantlets.

Conclusion
The micropropagation system under natural ventilation with four membranes promotes better in vitro growth of Aeollanthus suaveolens, increases the production of photosynthetic pigments, and is superior to the conventional sealed system. The use of different ventilation systems led to differentiation in the contents of the main VOCs. For the in vitro culture of A. suaveolens, it is recommended that explants be cultivated with greater ventilation in the culture vial.