Inﬂuence of Drought Stress on Growth and Essential Oil Yield of Ocimum Species

: A pot experiment was conducted to assess the effect of drought stress on growth and secondary compound accumulation of Ocimum species, in a semi-controlled greenhouse during the year 2020. The experiment was performed as a factorial that was based on a randomized complete design with three species of basil namely O. basilicum ‘Genovese’, O. x africanum , and O. americanum, and three levels of water supply (70, 50, and 30% soil water capacity-SWC) that were replicated twice. The results revealed that severe drought conditions (30% SWC) signiﬁcantly lowered the relative water content and, as a consequence, plants were shorter, narrower in the canopy, and produced smaller leaves which, in turn, resulted in a 50% fresh and dry herb yield loss. Furthermore, severe drought stress negatively affected the essential oil content (except O. x africanum where no change was seen), essential oil yield, and the antioxidant capacity. However, slight enhancements of glandular hair density were observed in the drought treatment than in the control. Regardless of the soil water capacity variation, the major compound of the essential oil and total polyphenol content remained unchanged. Besides drought, morphological and chemical variations were also detected among the Ocimum species. Sweet basil cultivar Genovese produced the maximum biomass (344.7 g/plant) whereas O. x africanum accumulated higher essential oil (2.79%). The major compounds that were identiﬁed were linalool in O. basilicum ; 1, 8-cineole, and camphor in O. x africanum; and O. americanum had more neral and geranial. In conclusion, a higher water supply is recommended for higher biomass and essential oil yield production in the tested basil species.


Introduction
The genus Ocimum L. belongs to the family Lamiaceae. It exhibits large morphological groups, comprising of 30 to 160 species owing to the ease of cross-pollination which has led to a large number of subspecies and varieties. Ocimum species are annual and perennial herbs/shrubs that are indigenous to Africa, Asia, Central, and South America, but extensively disseminated worldwide [1][2][3]. These highly aromatic plants of the genus have long been established as economically important medicinal plants due to their essential oils that have medicinal, culinary, and perfumery applications [4]. In medicinal and aromatic plants, growth and essential oil production are influenced by various environmental factors, drought stress being one [5][6][7]. Apart from environmental conditions other multiple factors such as; geographical area of cultivation [8], cultivation cycle [9,10], cultivation condition [11], harvest year [12,13], cultivars or varieties [14][15][16], age and part of the plant part [17][18][19], salinity [20], essential oil extraction methods [21,22], and storage conditions [23,24], can also influence essential oil production and its composition. Moreover, a significant reduction of plant growth under drought stress is a well-known and widely reported phenomenon, Ocimum species are not exceptions. Thus, the effect of drought stress on fresh and dry herb weights of basil has been reported by many scholars. Water shortage conditions provoked lower shoot fresh (−48.3%) and dry weight (−50.6%) compared with non-stressed sweet basil plants [25]. A 34% dry mass reduction due to drought stress was also reported [26]. Furthermore, a reduction of the total herb fresh and dry weights of American basil due to drought stress were observed [27]. Similarly, several authors have reported a significant reduction of yield in sweet basil due to drought stress [27][28][29][30][31]. In addition, water supply also modifies polyphenol as well as the antioxidant capacities of plants. In line with that, a moderate water deficit enhanced the antioxidant capacity and the total polyphenol content in medicinal and aromatic plants [5,32]. However, its effect on essential oil content, essential oil composition, antioxidant activities, and polyphenol content often remains controversial. On one hand, increased essential oil content and enhanced antioxidant activities under drought treatment was reported on O. basilicum, O. americanum, and O. x africanum [5,33,34]. On the other hand, a review on 'whether drought stress increases the biosynthesis and accumulation of plant volatiles,' concluded that the available scientific evidence was not adequate to generalize [35]. Thus, further holistic studies to determine the optimal water supply for different species are paramount. The effect of drought stress on the growth characteristics and essential oil content of sweet basil (O. basilicum) is relatively well studied, however, its effect on different species of basil is not well known. Hence, the objective of this study was to determine the effect of drought stress on the growth and essential oil yield of three Ocimum species under semi-controlled greenhouse conditions and reveal the differences between the species.

Experimental Site Description
A pot experiment was carried out under greenhouse conditions at the Experimental and Research farm of Hungarian University of Agriculture and Life Sciences, Budapest-Soroksár (Hungary) in 2020. The mean air temperature during the experimental period was 26.08 • C. The characteristics of the soil media mixture are summarized in Table 1.

Plant Material and Treatments
The experiment was laid out as a two-factor randomized block design with two replications ( Figure 1). A total of three basil species namely O. basilicum 'Genoves', O. x africanum, and O. americanum, and three levels of water supply [70% (control), 50% (moderate drought stress), and 30% (severe drought stress) soil water capacities] were used as treatments. The seeds of the Ocimum species were selected from the gene bank of the Medicinal and Aromatic Plants department. The seeds were sown into seed trays (27 × 57 cm) in the greenhouse in the middle of March. Seedlings that were two-months-old, that developed three leaves, were transplanted into 12 Litre pots. The pots were filled with 7 kg of sandyloam soil and peat moss mixture (1:1 v/v ratio). A total one plant per pot and of 10 plants per treatment were used for the experiment. The drought stress treatment was initiated 10 days after transplanting and lasted for 43 days afterwards. The soil water capacity (SWC) was determined using the modified gravimetric method of Reynolds [36] based on the water holding capacity of the soil. The SWC of the respective treatments were maintained three times per week. Every second day the weight of each of the pots was measured to determine the amount of moisture that was lost. After calculating the moisture loss of the respective treatments, additional water was added to maintain the moisture content of the treatments. The amount of water that was added per pot depended on the moisture loss and the treatment. maintain the moisture content of the treatments. The amount of water that was added per pot depended on the moisture loss and the treatment. The pots were kept under greenhouse conditions to exclude natural precipitation. Weeds were removed weekly; no fertilizer or plant protection chemicals were used.

Data Collection
The relative water content (RWC): 6-12 (depending on the species) fully developed leaves from three sample plants per treatment were randomly sampled from the third and fourth internode of the plant top and replicated twice. The leaf parts were weighed to determine the fresh weight (FW) and then soaked in distilled water for 24 h. After that period, excess surface water of the leaf parts was removed by paper towels and the turgid weight (TW) was determined. After drying at 105 °C until a constant weight, the dry weight (DW) of the leaf parts was determined. The RWC was then calculated according to the formula below [37,38]: RWC (%) = (FW−DW)/TW−DW) × 100 Chlorophyll content (SPAD value): The chlorophyll content of the leaves as indicated by the quantification of green color intensity was measured with a handheld SPAD meter (SPAD-502Plus Konica Minolta Inc., Osaka, Japan). The readings were taken at the third internodes from a fully developed leaf before harvesting. To calculate the mean, five plants per treatment (2 leaves per plant and 5 readings per each leaf) were taken.
Plant growth measurements: The plant height (cm), canopy diameter (cm), root fresh weight (g/plant), shoot fresh weight (g/plant), and shoot dry weights (g/plant) (after drying at room temperature in shadow) were measured from five sample plants per treatment at harvesting. The leaf area (cm 2 ): was measured by tracing the leaves over a square paper, and the grids that were covered by the leaf were counted to give the area. A set of 20 randomly selected leaves (2 leaves per plant and 10 leaves per treatment) were used to calculate the mean.
Glandular hair density: The glandular hair density was measured according to Radácsi et al. [16]. The samples were taken from the leaf blade at the 3rd internode from the top. Circles of 4.0 mm diameter were cut out from the central part of the leaf blade excluding the main vein. Then, the number of glandular peltate hairs on the abaxial surface of these blade samples was counted under a stereomicroscope (type BMS 74959). A total of 20 replicates per treatment were carried out.
Essential oil content determination: A total of ten sample plants were harvested per treatment and dried in well-ventilated rooms for two weeks. Then, a bulk sample of the dried leaves and inflorescences (excluding the stem) were used to measure the essential oil content in six replications. A total of 20 grams of dried material from each sample was hydro-distilled in a Clevenger-type apparatus using 500 mL of distilled water for 2 h, according to the method that was recommended by the VII Hungarian Pharmacopoeia The pots were kept under greenhouse conditions to exclude natural precipitation. Weeds were removed weekly; no fertilizer or plant protection chemicals were used.

Data Collection
The relative water content (RWC): 6-12 (depending on the species) fully developed leaves from three sample plants per treatment were randomly sampled from the third and fourth internode of the plant top and replicated twice. The leaf parts were weighed to determine the fresh weight (FW) and then soaked in distilled water for 24 h. After that period, excess surface water of the leaf parts was removed by paper towels and the turgid weight (TW) was determined. After drying at 105 • C until a constant weight, the dry weight (DW) of the leaf parts was determined. The RWC was then calculated according to the formula below [37,38]: RWC (%) = (FW−DW)/TW−DW) × 100.
Chlorophyll content (SPAD value): The chlorophyll content of the leaves as indicated by the quantification of green color intensity was measured with a handheld SPAD meter (SPAD-502Plus Konica Minolta Inc., Osaka, Japan). The readings were taken at the third internodes from a fully developed leaf before harvesting. To calculate the mean, five plants per treatment (2 leaves per plant and 5 readings per each leaf) were taken.
Plant growth measurements: The plant height (cm), canopy diameter (cm), root fresh weight (g/plant), shoot fresh weight (g/plant), and shoot dry weights (g/plant) (after drying at room temperature in shadow) were measured from five sample plants per treatment at harvesting. The leaf area (cm 2 ): was measured by tracing the leaves over a square paper, and the grids that were covered by the leaf were counted to give the area. A set of 20 randomly selected leaves (2 leaves per plant and 10 leaves per treatment) were used to calculate the mean.
Glandular hair density: The glandular hair density was measured according to Radácsi et al. [16]. The samples were taken from the leaf blade at the 3rd internode from the top. Circles of 4.0 mm diameter were cut out from the central part of the leaf blade excluding the main vein. Then, the number of glandular peltate hairs on the abaxial surface of these blade samples was counted under a stereomicroscope (type BMS 74959). A total of 20 replicates per treatment were carried out.
Essential oil content determination: A total of ten sample plants were harvested per treatment and dried in well-ventilated rooms for two weeks. Then, a bulk sample of the dried leaves and inflorescences (excluding the stem) were used to measure the essential oil content in six replications. A total of 20 grams of dried material from each sample was hydro-distilled in a Clevenger-type apparatus using 500 mL of distilled water for 2 h, according to the method that was recommended by the VII Hungarian Pharmacopoeia [39]. The oils were collected; traces of water were removed and stored in an airtight vial in a refrigerator at 4 • C for a week before analysis.
Total polyphenol content: The total amount of phenolic compounds in each extract was determined using the Folin-Ciocalteu method [40] with slight modifications. A total of 0.5 g of dried and powdered plant material was extracted by 50 mL of boiling distilled water and was allowed to stand for 24 h at room temperature. Then, the extracts were filtered (filter paper description: pore size of 10-12 µ; weight 85 g/m 2 ; thickness: 200 µm) and stored in a freezer until the measurements were taken. During the measurements, 40 µL of the test sample and 460 µL of distilled water were placed into a test tube and then mixed with 2.5 mL Folin-Ciocalteau's reagent (10 v/v%). After 1 min of incubation, 2 mL of sodium carbonate (0.7 M) was added. Then, the mixture was kept in hot water (50 • C) for 5 min and the absorbance was measured at the wavelength of λ = 760 nm with a Thermo Evolution 201 spectrophotometer (Thermo Fisher Scientific, Inc., Waltham, MA, USA). Gallic acid (0.3 M) was used as a chemical standard for calibration. The total phenolic content of the samples was expressed in the gallic acid equivalent was calculated on the dry weight basis of the extract (mg GAE/g DW). The measurements were done in 6 replications.
Antioxidant capacity: The FRAP assay was performed with slight modifications [41]. The same extract mentioned that is above was used for antioxidant capacity too. The FRAP reagent was prepared that contained sodium acetate buffer (pH 3.6), TPTZ (2, 4, 6-tripiridil-s-triazine) in HCl and FeCl 3 ·6H 2 O solution (20 mmol/L), in proportion 10:1:1 (v/v/v), respectively. Then, a 10 µL of the test sample was added to 1.5 mL of acting FRAP reagent and 40 µL distilled water, and the absorbance was recorded at 593 nm after 5 min using the spectrophotometer that was mentioned above. A blank contained distilled water instead of extract. The FRAP values of the samples were calculated from the standard curve equation and expressed as mg ascorbic acid equivalent (AAE)/g of dry extract.
Essential oil composition: The GC-MS method was used to determine the EO composition. GC analysis was carried out using an Agilent Technologies 6890 N (Agilent Technologies, Inc., USA) instrument that was equipped with an HP-5MS capillary column (30 m × 0.25 mm, 0.25 µm film thickness), with the following temperature program: initial temperature 60 • C, heating at a rate of 3 • C/min up to 240 • C; the final temperature was maintained for 5 min; injector and detector temperatures: 250 • C; carrier gas: helium (constant flow rate: 1 mL/min); and a split ratio: 30:1, injection volume 0.2 µL (1%, n-hexane). The proportions of the individual compounds were expressed as total area percentages. The same equipment was used to identify the components with an Agilent Technologies MS 5975 detector (Agilent Technologies, Inc., Waltham, MA, USA). The ionization energy was 70 eV. The mass spectra were recorded in full scan mode, which revealed the total ion current (TIC) chromatograms. A mixture of aliphatic hydrocarbons (C9-C23) in n-hexane was injected to calculate the linear retention indices using the generalized equation of [42]. The linear retention indices (LRI) and mass spectra were compared with commercial ones (NIST, Wiley) and a homemade library mass spectra was built up from data that were obtained from standard (Sigma/Aldrich) pure compounds. The GC samples were repeated three times.

Data Analysis
The data were evaluated using a two-way analysis of variance (ANOVA). A Shapiro-Wilk's test and Levene's test were used for checking the normality of distribution and the homogeneity of the variances, respectively. Significant mean differences were examined with Tukey HSD at p < 0.05. All statistical analysis was performed using IBM SPSS 25.

Relative Water Content and Chlorophyll Content (SPAD Value)
The relative water content and chlorophyll content of the basil species were significantly modified with water supply (

Plant Height and Canopy Diameter
The water availability strongly limited the growth and development of plants. The statistical analysis revealed that the plant height and canopy diameter were significantly different (p < 0.01) among the basil species and drought treatments. Plants under the control treatment were taller and wider in the canopy as indicated in Figure 2 and Table 3. Basil plants that were grown under severe drought conditions were 7.3 cm shorter and 10.4 cm narrower. Besides drought, a variation among the species was also detected. O. basilicum cultivar 'Genovese' was found to be 11.05 cm taller on average over O. x africanum and O. americanum. Unlike the two upright species, O. americanum has a spreading growth habit. Thus, it has a wider diameter.

Relative Water Content and Chlorophyll Content (SPAD Value)
The relative water content and chlorophyll content of the basil species were significantly modified with water supply (Table 2). Hence, compared to the control treatment, severe drought conditions lowered the RWC of the O. basilicum 'Genovese' leaves by 13.8% as well as 16.31% RWC drop in O. x africanum while the highest reduction (28.84%) was observed in the O. americanum leaves. Regarding the chlorophyll content, the severe drought treatment caused a 28.2% and 13.3% greater SPAD value compared to the control and moderate drought treatment, respectively.

Plant Height and Canopy Diameter
The water availability strongly limited the growth and development of plants. The statistical analysis revealed that the plant height and canopy diameter were significantly different (p < 0.01) among the basil species and drought treatments. Plants under the control treatment were taller and wider in the canopy as indicated in Figure 2 and Table  3. Basil plants that were grown under severe drought conditions were 7.3 cm shorter and 10.4 cm narrower. Besides drought, a variation among the species was also detected. O. basilicum cultivar 'Genovese' was found to be 11.05 cm taller on average over O. x africanum and O. americanum. Unlike the two upright species, O. americanum has a spreading growth habit. Thus, it has a wider diameter.

Leaf Area and Root Fresh Weight
As an adaptive response to drought, Ocimum plants significantly modified their leaf area and fresh root weight ( Table 4). As a result, severely stressed plants reduced their leaf area and root fresh weight in contrast to the control plants. A higher water supply on average (mean of three species) resulted in a 2-fold larger leaf area and 2.31-fold heavier fresh root weight compared to the severe drought condition. The sweet basil cultivar had

Fresh and Dry Herb Weight
The fresh and dry herb weight was found to be statistically significant between the drought stress treatments and the Ocimum species (Table 5). Severe drought conditions resulted in a 50.15 to 55.86% yield loss range of the basil species as compared to the control. Likewise, moderate drought conditions also showed a 19.9%, 36.5%, and 12.9% fresh weight reduction in O. basilicum 'Genovese, O. x africanum, and O. americanum, respectively. The same reduction trend was observed on the dry weight as well. The fresh and dry weight reduction could be partly due to a restricted water supply. A significant positive correlation was observed between the fresh weight and the RWC (r = 0.72, p < 0.01). The Genovese cultivar produced the maximum fresh and dry weight followed by O. americanum under higher water supply (70% SWC).

Essential Oil Content and Essential Oil Yield
Essential oil content and essential oil yield were found to be significantly different (p < 0.05) among the Ocimum species and drought treatments ( Table 6). The effect of water supply was not the same for each species. On one hand, severe drought conditions had a negative influence on the EO content of O. basilicum 'Genovese' and O. americanum. On the other hand, no change was detected in the case of O. x africanum. Therefore, the effect of drought on the EO content is species-dependent. However, the essential oil yield was uniformly lower both under the moderate and severe drought conditions. O. x africanum produced significantly higher essential oil yield.

Glandular Hair
Essential oil is accumulated in the glandular hairs of basil plants. The glandular hair density was slightly affected by the soil water capacity among the investigated basil species ( Table 7). The severe drought conditions increased the glandular hair density in O. x africanum and O. americanum. Whereas the Genovese cultivar showed higher glandular hair under the moderate drought conditions. Even though moderate and severe drought conditions have a positive effect on the glandular hair density, its effect on the glandular hair number was negative, partly due to a smaller leaf area production in response to water shortage. As a result, the severe drought treatment produced a 27.7 to 54.5% lower glandular hair number per leaf. Moreover, significant positive correlations were observed between the leaf area and the glandular hair numbers of O. basilicum 'Genovese (r = 0.57, mboxemphp < 0.01), O. x africanum (r = 0.63, p < 0.01), and O. americanum (r = 0.77, p < 0.01). Differences were also detected between the species; O. x africanum produced 4-fold and 2.7-fold higher glandular hair density compared to O. basilicum 'Genovese' and O. americanum, respectively.

Total Polyphenol Content and Antioxidant Capacity
No significant change was detected between the different soil water capacities on the total polyphenol content (TPC) except for O. basilicum 'Genovese' where severe drought stress enhanced the TPC by 12.33% compared to the control (Table 8). In general, basil species showed different response patterns in TPC as the soil water capacity changed. A notable difference was observed among the soil water capacity treatments and the basil species on AOC. Accordingly, the higher the soil water content, the higher AOC that was accumulated in all species uniformly. Under the control treatment, O. x africanum produced more TPC and AOC.

Essential Oil Composition
The water supply did not significantly (p > 0.05) affect the EO composition ratio except for the camphor ratio of O. x africanum (Table 9)

Discussion
Drought stress is one of the most important abiotic stresses which can affect the growth and accumulation of bioactive compounds in medicinal and aromatic plants. These changes are mainly related to altered metabolic functions, physiological, and morphological characteristics. Thus, this study aimed to describe the drought-triggered physiological, morphological, and biochemical changes on selected Ocimum species. The measured relative water content (RWC) to evaluate the water status of Ocimum species showed a reduction under the drought-stressed conditions. Similar findings were reported on sweet basil [25,26,28], lemon thyme [43], and summer savory [44]. The reduction of RWC resulted in turgor loss which strongly influences the plant growth and biomass production through its effect on cell expansion [45]. A higher SPAD value was observed in the severe drought intensity. Previous studies also showed a higher SPAD value under water-deficit conditions in two Plantago species [46], summer savory [44], and Jerusalem artichoke [47]. The SPAD value is based on the light absorbance of leaves and thus turgor, leaf thickness, or leaf hairiness might influence the results. A higher SPAD value does not necessarily mean an increase in the chlorophyll content [16].
Drought stress leads to turgor loss, trim down in photoassimilation, and metabolites that are required for cell division. As a consequence, impaired mitosis, cell elongation, and expansion result in reduced growth [48,49]. Likewise, we observed a significant reduction in all the measured production parameters (plant height, canopy diameter, leaf area, fresh root weight, and biomass production) under drought stress uniformly across all the tested species. Numerous scholars also reported the negative impact of drought stress on the production of sweet basil [16,25,28,29,50], O. x africanum [33], and O. americanum [27].
The influence of drought stress on secondary compound accumulation is highly debatable and a lot of contradictory results are reported. In this experiment, we observed that, regardless of the drought intensity, the total polyphenol content (O. x africanum and O. americanum), essential oil content of O. x africanum, and the essential oil composition ratios of the major compounds remained unchanged, whereas drought stress led to lower EO yield and antioxidant capacity in all species. Drought stress slightly enhanced the glandular hair densities of the basil species and the TPC of O. basilicum 'Genovese'. Other scholars, however, have stated that trichome density is mainly determined by ontogenesis and species [51,52]. We observed that water supply can also significantly influence the glandular hair density. On one hand, no changes in the essential oil content on five Lamiaceae species (Lavandula latifolia, Mentha × piperita, Salvia lavandulifolia, Thymus capitatus, and Thymus mastichina) were demonstrated [53]. On the other hand, a higher EO content in a drought condition was observed on sweet basil [7,26,54,55]. Moreover, a water-deficit also increased the essential oil content of O. x africanum while the essential oil production tended to be lower in the drought treatments [33]. Although a slight essential oil content enhancement was reported by many scholars, drought stress significantly influences the essential oil production as a result of lower biomass production in water-deficit treatments.
The genus Ocimum is characterized by great variability in both morphology and chemical composition due to polyploidy, aneuploidy, and inter-and intraspecific hybridizations [15,56,57]. The studied species also showed variation in morphology and chemical compositions under optimum water supply. The sweet basil cultivar Genovese had an upright growth, large leaf area, higher biomass yield, and more linalool. O. x africanum produced more essential oil, glandular hair density, TPC, and AOC. The African basil accumulated more 1,8-cineole and camphor. Whereas the third species, O. 10mericanum, had a spreading growth habit, narrow leaves, and higher ratios of neral and geranial.

Conclusions
In the present experiment, the physiological, morphological, and secondary compound accumulation of three Ocimum species were evaluated under drought stress treatments. It is concluded that among the investigated water supplies, the 70% SWC increased the biomass production, essential oil yield, glandular hair number, and the antioxidant capacity.
The EO composition and TPC remained unchanged regardless of the soil water capacity. It was also observed that each species responds to drought intensity differently. The impact of drought was severe on the biomass production of O. x africanum, while the accumulation of the secondary compound in O. americanum was highly affected under the severe drought conditions. Hence, the general conclusion that states that drought stress can increase secondary compound accumulation in medicinal and aromatic plants is misguided. Beyond drought, the three species vary morphologically and in their chemical composition. To sum up, among the water supply treatments, 70% SWC showed up as an optimum, which is recommended for growth and essential oil production. In addition, taking into account the specific requirements of each species is paramount.

Data Availability Statement:
The data presented in this study are available within the article.