Changes in rosewood (Aniba rosaeodora Ducke) essential oil in response to management of commercial plantations in Central Amazonia

Forest Ecology and Management 429 (2018) 143–157

Contents lists available at ScienceDirect

Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco

Changes in rosewood (Aniba rosaeodora Ducke) essential oil in response to management of commercial plantations in Central Amazonia

T

⁎

Pedro Medrado Krainovica, , Danilo Roberti Alves de Almeidab, Valdir Florêncio da Veiga Juniorc, Paulo de Tarso Barbosa Sampaioa National Institute of Amazonian Research – INPA, Av. André Araújo, 2936, Aleixo, CEP 69060-001 Manaus, AM, Brazil University of São Paulo – USP, Forest Sciences Department, Av. Pádua Dias, 11, CEP 13418-900 Piracicaba, SP, Brazil c Federal University of Amazonas – UFAM, Av. General Rodrigo Octávio Jordão Ramos, 3000, Coroado, Manaus, AM, Brazil a

b

A R T I C LE I N FO

A B S T R A C T

Keywords: Forestry Resprouting Endangered species conservation Amazonia

Rosewood essential oil (REO) is an Amazonian industrial crop required by fragrance and cosmetic industries worldwide. This essential oil (EO) is obtained from a singular resource, the endangered tree species Aniba rosaeodora Ducke. The management of this resource influences the chemical composition of the EO, affecting the quality and international price of the product. A systematic study was performed within the rosewood plantations of two major REO producers. Chemical composition (GC–MS) and REO yields were analyzed to identify the best harvesting periods and the potential sustainable use of other plant parts, such as resprouting shoots, to produce the oil. With a large sample and a well-controlled statistical approach, the study’s methodology allowed us to describe the differences in the REO composition between tree parts and between harvest times. REO yield was highest in branches from the first harvest and in resprouting leaves from the second harvest. In the first harvest, α-pinene was found only in REO from branches and leaves, and cyclosativene was sourced only from branches, regardless of the sampling region. Geraniol was detected only in the first-harvest REOs, while myrcenol was found only in second-harvest REOs. The temporal spacing of harvest rotations and the use of different plant parts in extraction are the main management tools determining the variations in REO. Despite higher EO yield in the stem, the management by crown pruning assures sustainable oil production. Greater understanding of these variations may provide opportunities to expand the production chain of globally exported REO.

1. Introduction Rosewood (Aniba rosaeodora Ducke - Lauraceae) is an Amazonian tree threatened with extinction (IBAMA 1992; CITES 2010; IUCN 2014). Since wild populations are under full protection (directive N0 443 12/2014, MMA), commercial rosewood plantations (formed from genetic material of natural populations) are meeting the global cosmetic industry’s demand for the essential oil (EO) derived from this species. These plantations not only regulate the exploitation of rosewood and reduce the pressure on wild populations, but also generate jobs and development in rural areas of the Central Amazon (McEwan et al., 2016). International demand for Brazilian REO has been constant, with untapped growth potential in the world market. Consequently, many producers are interested in commercial rosewood production; however, they encounter a lack of technical information regarding production management (May and Barata, 2004) and a lack of clarity in the standardization established by the current legislation.

⁎

Although the development of technical criteria for rosewood management is recommended by law (IN N0 02/2006, SDS), there is a regrettable lack of studies on the effectiveness of harvesting patterns to optimize REO extraction under commercial production conditions, and on the influence of particular harvesting patterns on the yield and quality of the final product. Above-ground biomass management (ABM) is a harvesting technique whereby trees are either coppiced (cut at 50 cm above the soil) or pollarded (100% of their crowns pruned) and are then allowed to regrow (Sampaio et al., 2005). Managing above-ground biomass does not require intensive preparatory operations (as is the case with seedling planting), and the care requirements are simple and provide quick benefits (Spinelli et al., 2017). Given the difficulties in obtaining rosewood seedlings, regrowth is the best management option for commercial plantations. However, one consequence of this management practice is that the regrowth differs in terms of nutrient status (Krainovic et al., 2017a). Studies on other species have observed that

Corresponding author. E-mail addresses: [email protected], [email protected] (P.M. Krainovic).

https://doi.org/10.1016/j.foreco.2018.07.015 Received 23 May 2018; Received in revised form 4 July 2018; Accepted 4 July 2018 0378-1127/ © 2018 Published by Elsevier B.V.

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

2.2. Field sampling, harvesting and extraction of essential oils

ABM causes differences in physiological mechanisms (Moreira et al., 2012; Shibata et al., 2016; Pausas et al., 2004, 2016). For rosewood, ABM could potentially result in differences in secondary metabolite production, notably terpenes, causing variation in the final product required by industries, though the extent and nature of such changes have not been previously studied. Sequential management (harvest and resprouting) of commercial plantations requires that the characteristics of the regrowth meet the needs of the market. Currently, a diverse chemical composition may diminish the quality of the REO to below the industrial standard for its fixative and perfuming activities, thereby reducing its commercial value. Nevertheless, it could also create new opportunities (new products). The production chain of REO requires a full and integrated understanding of plantation management, from proper tree management to the production of quality REO for the international market. The extent of compositional variability of REO from managed plantations, as well as the factors that can influence this variability, is key information. Although REO consists mostly of low molecular weight monoterpenes and sesquiterpenes—the major constituent is linalool (78–93%) (Chantraine et al., 2009; Krainovic, 2011; Fidelis et al., 2012, 2013)—a range of minor components confer the precise fragrance bouquet unique to REO. Consequently, the conditions under which a rosewood plant is grown must be able to stimulate the redirection of metabolic pathways, resulting in the biosynthesis of different compounds (Morais, 2009). The present study is the first to provide relevant information on the compositional differentiation of REOs cultivated in Central Amazonia and the influence of above-ground biomass management on their chemical composition. The study is based on the largest sample ever used in an analysis of rosewood essential oil production. The study hypotheses were that (1) the cultivation region influences the yield and chemical composition of REO; (2) the use of different tree parts for REO extraction will produce different yields and chemical compositions of the final product; and (3) the use of regenerated biomass will yield REOs with different characteristics than REOs extracted from biomass of the first harvest in a sequential management system. Consequently, the objectives of the current study were to (1) understand the effects of the cultivation region on REO composition and yield; (2) characterize the variation in REOs from different parts of rosewood trees; and (3) investigate the feasibility of sequential harvesting of commercial rosewood, based on yield and chemical composition of the REO.

In each area, samples of plant material (stem, branches and leaves) were collected from 48 trees by cutting the stems of 36 trees at 50 cm above soil level (coppicing), and by pruning 100% of the crown (pollarding) of the other 12 trees. Sampling occurred on two occasions: i) first harvest and ii) second harvest. Stems were then sectioned into discs for the collection of wood samples. Branches and leaves were sampled at the four cardinal points, at the height of the middle third of the tree crown. Sampling this way homogenized the collection in relation to incident sunlight, since the crowns of C 10 and C 12 trees were already touching, while rosewood canopies in C 17 were competing with naturally regenerating vegetation for light. Material was divided into “stem wood”, “leaves”, “branches”, and “branches + leaves” at a 1:1 wt ratio. Material from six trees adjacent to each sampled tree was combined to provide a composite sample for REO extraction. Exactly 12 months after management action, in the same season in each area, a new sampling (second harvest) of all individuals was performed using the same methods, collecting branches and leaves of the resprouting crowns (pollarded trees) and stumps (coppiced trees). We collected a total of 90 samples in the first harvest and 72 samples in the second harvest. Collected samples were stored in paper bags to avoid exposure of the material to light and the consequent loss of volatile constituents. After pre-drying in the shade and at room temperature for 72 h (a duration traditionally used by rosewood oil producers), the samples were ground and then stored in plastic bags in a freezer at −15 °C. To minimize any possible variation, a randomized block was used to establish the extraction sequence. REO was extracted by hydrodistillation using a modified Clevenger apparatus. Samples were weighed for REO yield calculation and then placed with distilled water in a glass flask of known capacity and hydrodistilled for two hours (timed from the first drop of REO at a temperature compatible with the gentle boiling of the material within the flask). All extractions were performed using the same homeothermic blankets and in the shortest time possible (45 days) to reduce variations that could influence distillation results. After each distillation, the Clevenger apparatus was washed with 80% hexane–ethyl acetate solution. Samples of REO + water resulting from hydrodistillation were stored in a freezer at 6 °C for 24 h; they were subsequently dried using anhydrous sodium sulfate (Na2SO4). The dried REOs were stored at a temperature of 6 °C prior to chromatographic analysis. REO yield (Y%) was calculated based on the ratio between the weight of the sample and the weight of the extracted REO:

2. Methods

Y% = (essential oil weight / sample weight ) ∗100

(1)

2.1. Site descriptions and plant materials 2.3. GC-FID analysis The study was conducted in three rural areas. Two of these areas (homogeneous plantations, 10 and 12 years old, hereafter referred to as C 10 and C 12, respectively) are located in the municipality of Maués (350 km by river from Manaus), and the third area (17-year-old plantation in lines traversing the natural vegetation, hereafter C 17) is in the municipality of Novo Aripuanã (469 km by river from Manaus). Both municipalities are in the state of Amazonas, in Central Amazonia (Fittkau et al., 1975), Brazil (Fig. 1). The climate in Maués is hot and humid, with regular and abundant rainfall, an annual rainfall of 2,101 mm, and an annual mean temperature of 27.2 °C. According to Köppen-Geiger, the climate is type Amazonia Af. The soil under rosewood plantations is classified as dystrophic yellow red latosol (Krainovic, 2011). The climate of Novo Aripuanã is also classified by Köppen-Geiger as type Af, hot and humid, with an annual average rainfall of 2,444 mm and an annual mean temperature of 26.9 °C (Kottek et al., 2006; http://en.climate-data.org/). Predominant soils in the region are classified as yellow poor oxisols saturated with oxidized iron and aluminum with low pH (Tanaka and Vieira, 2006).

Analyses were conducted with a flame ionization detector (GC-FID) in a THERMO TRACE 1310 chromatograph, using a capillary column with 100% dimethylpolysiloxane stationary phase (25 m × 0.25 mm, film thickness 0.25 mm) and helium as the carrier gas, at a flow rate of 2 ml/min (model DB-5). The chromatograph was programmed as follows: isothermal at 80 °C for 1 min; 80–150 °C at 12 °C/min; 150–180 °C at a 6 °C/min; isothermal at 180 °C for 1 min; and a total optimal run time of 12.8 min, reducing analysis time of the repetition samples. Samples were injected at a concentration of 5 mg of REO to 1 ml of 80% hexane-ethyl acetate solvent. The injection port was adjusted to 220 °C using a ratio of 1:4, and the detector temperature was set at 240 °C. Given the extreme olfactory and physical characteristics (coloration) of the REOs, chromatographic run optimization was conducted with the aim of minimizing analysis time without loss of chromatogram resolution. Consequently, we tested chromatographic analyses with different rates of temperature variation (°C/min) and ran analyses with and without isotherms (Table A.1). The chosen method was suitable for the essential oils extracted from both the lighter leaf-based material 144

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. 1. Map of study areas showing Brazilian Legal Amazon in South America, the Novo Aripuanã study site just inside Madeira River basin, and the Maués study site, directly linked to main Amazon River channel.

2.5. Statistical methods and analysis

(first harvest and resprouting) and the heavier woody material. A sample of each REO was analyzed using the methodology described by Adams (2007), using a temperature gradient of 60–200 °C with an increase of 2 °C/min. Constituent area percentages were calculated using electronic integration of peak areas present in the chromatograms, without the use of correction factors. The same integration criterion (1300 area unit) was used in all generated chromatograms.

Three categories were created to allow analysis, comparison and discussion of the differences between the REO extracts: 1. Area-specific REOs: the yield and chemical composition of the REOs extracted from each plant part managed in the first harvest of the trees were analyzed according to the region of origin of sample trees; 2. Plant part-specific REOs: REOs extracted from biomass removed in the first harvest were analyzed according to the part of the tree from which they were extracted, and the results were analyzed per region and independent of region; 3. REOs from sequential above-ground management: REOs from each plant part were compared to samples derived from biomass cut in the first harvest and second harvest (resprouting), with data analyzed per region and independent of region. In each of the three categories, the distributions of REO components clearly differentiated plant parts/ regions/harvest times. Consequently, among categories, it was possible to determine differences between REOs, and whether the differences occurred independent of cultivation site, plant part and/or harvest time (first harvest or second harvest). Statistical analyses were performed using either the Tukey test (for parametric data) or Wilcox tests (for non-parametric data). A paired T or Wilcoxon test was used to compare REO yields extracted from the sequential crops (first harvest and resprouting). The exact error probability (p-value) results of the mean tests were presented together with graphs. To analyze REO chemical composition, non-metric dimensional scaling (NMDS) was applied using a Jaccard Similarity Index, based on the presence/absence data of compounds in each sample.

2.4. Identification of components Due to the number of samples, we used a mixed methodology: samples representative of REO variation followed classic methodologies, while most of the samples followed an optimized run that was confirmed with the classic methodologies. To confirm the results, the same samples analyzed using Adams’ methodology were analyzed by mass spectrometry (GC-MS), and then the chromatographic profile obtained from these runs was compared with the chromatographic profile of the optimized runs by overlapping their peaks. To make the identification of some substances more precise, retention indexes were calculated using the Van den Dool and Kratz equation, which relates the retention time of the substances to the retention times of the n-alkanes (C12-C16 homolog series), which aided composition identification and were co-injected into the sample sequence. The retention indices and mass spectra of all detected constituents are defined in the literature (Adams, 2007). A minimum of 90% spectrum library similarity (NIST) was used for positive identification. Finally, the obtained results were confirmed by analyzing data obtained from previous studies on REO chemical composition (Fidelis et al., 2012, 2013). 145

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. 2. Flowchart of the work methodology describing the factors and their respective levels discussed in the present study, as well as the statistical analyses used.

NMDS was complemented by hierarchical grouping analysis, using Euclidean distance as a measure of separation between records, and Ward's method as a grouping medium. This analysis was performed using values for the relative area under the peaks present on the chromatograms, these values being proportional to the concentration of the compound in the total REO sample. Results of the analysis are shown as a dendrogram, which helps identify homogeneous groupings within larger groups and between-group heterogeneity (Young and Hammer, 2000). To validate the robustness of the generated dendrogram, a cophenetic correlation coefficient was calculated for the formed groups. A summary, with factors and their respective evaluated levels, is presented in Fig. 2. All analyses and production of graphic materials were conducted using R software (R Core Team, 2018).

It is important to note that the main factors influencing the rate of secondary metabolite production are reported in studies that were limited to restricted groups of species, predominantly those occurring in temperate regions, many of which are commercially important and may have undergone strong selective pressures (Gobbo-Neto and Lopes, 2007). These factors, therefore, may not be representative of what occurs in wild plants or in other biomes, such as the Amazon. They may not apply to rosewood, which, though cultivated, has little or no level of domestication, since the genetic material of the plantations comes from natural populations. In addition, the genetic variability between the sampled populations, which may have greater or lesser influence on REO variations (Gupta et al., 2017), was not determined in the present study. However, Chantraine et al. (2009) concluded that there is no genetic determinism of specific chemotypes for rosewood.

3. Results and discussion

3.1. Area-specific essential oils - differences associated with cultivation area

Of all 77 components detected in REO, oxygenated sesquiterpenes were numerically the largest group (38.9%). They were followed by oxygenated monoterpenes (24.7%), hydrocarbon sesquiterpenes (23.4%) and monoterpene hydrocarbons (7.8%); the remainder (5.2%) were substances for which a reliable identification could not be obtained. the substances with the highest general averages of area percentage in each group were: Sphatulenol, Linalol, β-Selinene and Camphene respectively for each subtance class (Table A.2). The biosynthetic pathways of terpenes are divided into several phases (Bohlmann et al., 1998), and the yield and proportion of each constituent formed in the secondary metabolism depend on the balance of these reactions. This balance can be influenced by several factors, such as the action of sesquiterpene synthase enzymes, which are known to produce hundreds of C15 hydrocarbon complexes from some simple precursors (Lodewyk et al., 1997). Other factors include heredity, growth stage, the environmental aspects of seasonality, water stress and nutrients (Gupta et al., 2017).

3.1.1. Essential oil yield from different sites in Central Amazonia Tree age appears to have no significant effect on REO yields (amount of REO per amount of biomass) in Central Amazonian rosewood plantations (Krainovic, 2011; Fidelis et al., 2012, 2013). No statistically significant differences were observed between the two sampling regions for REOs from leaves + branches (p = 0.910) and from stem wood (p = 0.126), while the REO yield from leaves and from branches of the Novo Aripuanã region was significantly higher than the yield from those of Maués (p < 0.01 - Fig. 3). The hypothesis that higher levels of soil nutrients would result in higher EO yields was discarded, since leaf samples from Maués had higher element concentrations than those from Novo Aripuanã (Krainovic et al., 2017a), but samples from Novo Aripuanã trees had the highest EO yields. Studies correlating nutrient availability and the EO yields of individual rosewood trees have never been undertaken, but results found in the literature for other species and regions of the world corroborate results from the present study: there is no direct link 146

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. 3. Essential oil yields from leaves, branches, wood stem, and a mix of branches and leaves (B + L), showing the comparison between regions of cultivation by mean test of each compartment.

(Table 1). Eucalyptol is a natural constituent of a variety of aromatic plants with medicinal importance (Caceres et al., 2017). It is also used in the production of fragrances, perfumes and various cosmetics (Surburg and Panten, 2008), imparting a fresh camphor scent. A reduction in rosewood’s constituent fraction of this essential oil is undesirable, given the requirements of the cosmetic and fragrance industries. However, while this substance was absent from the Novo Aripuanã first-harvest sample, it was present in samples of resprouting branches (second harvest) (Table A.3). This result leads us to discard aspects of climate and soil as explanations for the presence or absence of this constituent. This result also coincides with those of Barbosa et al. (2012), who evaluated abiotic factors influencing copaiba oil (Copaifera multijuga Hayne, Fabaceae) composition in the Brazilian Amazon; they concluded that seasonality did not have a significant influence on the chemical composition. Studies in other regions of the world also confirm this finding. Zheljazkov et al. (2009), investigating the influences of sampling region (two regions of Mississippi with different soil granulometry and nutrient availability) on the chemical composition of peppermint EO (Mentha x Piperita hybrid), concluded that concentrations of eucalyptol were not affected by location. Considering that the Maués plantations are much younger than those in Novo Aripuanã, and that eucalyptol was found in resprouting samples from Novo Aripuanã, we suggest that relative age of the source plant material may be a factor for the presence of this substance in the extracted REO. This would agree with the results of Fidelis et al. (2012), who found small differences in the minor constituents in EO from a chronological sequence of rosewood trees. While p-cymene was detectable only in older plant parts from Novo

between the two factors. For example, administration of nitrogen to commercial peppermint (Mentha x piperita) plantations in the United States did not lead to a significant increase in extracted EO yield (Zheljazkov et al., 2009). This was also found by Ucar et al. (2017) for a Rosa damascena hybrid in the Mediterranean Region of Turkey. In the current study, between-region differences in yield may be more strongly linked to the season when the sampling was undertaken—sampling at Maués was performed during the local rainy season (February), while sampling at New Aripuanã was performed in July, when rainfall is scarce (Fig. A1). Several studies report changes in EO yield related to seasonal decreases in the water regime (Gobbo-Neto and Lopes, 2007; Morais, 2009; Prochnow et al., 2017). In the Brazilian Amazon, Taveira et al. (2003) found an increase in the proportions of mono- and sesquiterpene compounds in the EOs of Aniba canelilla (Kunth) Mez (Lauraceae) during the dry season. Furthermore, for rosewood specifically, Chantraine et al. (2009) found a decrease in stem wood REO yield during the rainy season in some of their sampling areas. 3.1.2. Chemical composition of the essential oil from different sites in Central Amazonia Differences in the chemical composition of REOs were observed between the two sampling regions, with those from the Novo Aripuanã region having greater numbers of components. There were also compounds that were only present in samples from one region. For example, eucalyptol was not detected in the branches and leaves from Novo Aripuanã but was present in those from Maués, while p-cymene was present in the stem wood and branches from only Novo Aripuanã

Table 1 Differential substances between plantations for each tree compartment from which the essential oils were extracted. Underlined substances were repeatedly differential across compartments and were considered as differential relative to the region. Comp.

Area

Number of components

OE components that distinguish regions

Stem

Maués Novo Aripuanã

47 51

trans-nerolidol; viridiflorol; α-farnesene; aristolene epoxide; isoaromadendrene epoxide p-cymene; myrcenol; β-citronelol; α-amorphene; NI 1; β-trans-guaiene; cadinene; guaiol; NI 5

Branches

Maués Novo Aripuanã

49 56

eucalyptol; ledene oxide p-cymene; nerol acetate; gurjunene; α-amorphene; NI 1; Y-celinene; β-trans-guaiene; aromadendrene dihydro;

Leaves

Maués

53

Novo Aripuanã

60

camphene; eucalyptol; borneol; β-trans-guaiene; 3-methoxymethoxy 3,7,16,20 tetramethyl eneicosa 1.7.11.15.19 pentaene benzaldehyde; linalyl acetate; trans dihydrocarvone; α-cariofileno; α-humuleno; α-amorphene; γ-celinene; ni 3; aromadendrene dihydro; cadinene; ledol; aristolene epoxide

147

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. 4. (a) Data ordinations in the NMDS plot made with the presence/absence matrix of the substances detected in the chemical composition of essential oils. The colors indicate sampling region: red = Maués site C 10; green = Maués site C 12; blue = Novo Aripuanã site C 17. (b) Dendrogram showing the groupings of chemical compositions of oils from different regions. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

derived samples (1.14%), higher in leaves (1.54%), and highest in stem wood (1.77%). This difference occurred independent of the cultivation region (Fig. 3; values presented are averages between the two regions). Result trends were the same for both regions, and there were no significant differences (p-values < 0.05) in yield between leaves vs branches + leaves mix and stem vs branches + leaves mix of Maués, and between branches vs branches + leaves mix and stem vs leaves of Novo Aripuanã (Table 2). REO yields of stem wood and branches agree with results reported by Chantraine et al. (2009) for rosewood cultivated in French Guiana, though these authors found leaves to have the lowest yield, which was not observed in the current study. The results obtained here also contradict information in the Brazilian legislation (IN N0 02/2006, SDS), which states that rosewood leaves give the highest yields. However, although the legislation is not clear in this regard, there are indications that the data on which the law is based originated from natural populations, suggesting an urgent need to update legal provisions so that

Aripuanã, it was not found in plantation samples from Maués, where the plants are younger. This result agrees with reports of significantly higher p-cymene concentrations in REO extracted from resin of older Protium heptaphyllumi (Aubl.) March. (Burseraceae) than in REO from the resin of younger samples from the same tree. The most likely hypothesis here is that abiotic chemical oxidation is responsible for the conversion of terpinolene to p-cymene (or dehydrogenation of limonene) during natural aging. This would also explain our results. The abiotic factors that influence this oxidation, and how it occurs, remain unknown (Albino et al., 2017). These compositional differences were shown in the NMDS biplot of the presence/absence matrix of chemical constituents. Great differences exist between regions: extract samples from Novo Aripuanã form one distinct group (except for some samples of the stem wood), whereas the extract samples from the two areas in Maués show substantial overlap (Fig. 4a). Cluster analysis confirmed the ordination trends of the NMDS biplot and distinguished two REO groups (Maués and Novo Aripuanã) based on chemical composition (Fig. 4b). The data overlap shows that stem wood-derived REO samples differ less between regions, probably due to the lower metabolic activity of stem wood compared to branches and leaves, which itself is associated with the age of the tissues (Pallardy, 2010; Londero et al., 2011). Our results confirm this assertion, since the average number of recorded substances was lowest in stem wood-derived samples (49), followed by branches (52.5), then leaves (56.6).

Table 2 Mean tests p values of essential oil yield between compartments managed in each rosewood plantation located in the three study areas in the Central Amazon.

3.2. Part-specific oils - differences associated with plant parts managed for extraction of essential oils 3.2.1. Essential oil yield from different plant parts EO yields differed between plant parts, being lowest in branch-

Test categories

Maués

Branches - B + L Leaves - B + L Leaves - Branches Stem - Branches Stem - B + L Stem - Leaves

< 0.01w 0.91w < 0.01w < 0.01w 0.12w 0.03w

T w

148

p value from Tukey test. p value form Wilcox test.

Novo Aripuanã 0.97 T 0.05 T 0.02 T 0.02 T < 0.01 T 0.94 T

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Table 3 Substances that differentiate essential oils extracted from different plant parts collected in the first harvest, regardless of the region of cultivation. Underlined substances were common to all regions and were considered as differential according to the plant part. Reference compartment

Not in stem EO

Not in branch EO

Not in Leaf EO

in branch EO

α-pinene; linalyl acetate; nerol acetate; cyclosativene α-pinene; ledene oxide

–

cyclosativene

α-farnesene; αbisabolol; isoaromadendrene epoxide

–

in leaf EO

they include parameters relevant to rosewood in plantation environments (Krainovic et al., 2017b). Additionally, published results on REO yields show great variation. This, and the variation in the ages, parts and extraction methods used, has led to considerable disagreement regarding the most appropriate EO extraction methodologies (Santana et al., 1997; Chantraine et al., 2009; Fidelis et al., 2012). We emphasize that the time elapsed between the sample material’s collection and the distillation of its mass, used to calculate the yield, can vary considerably according to the degree of humidity. In the present study, the time elapsed between harvesting and distillation followed faithfully what is done in the commercial plantations (72 h of shade drying).

Fig. 5. NMDS plots made with the presence/absence matrix of the substances detected in the chemical composition of essential oils extracted from each compartment: (a) Maués site C 10; (b) Maués site C 12; (c) Novo Aripuanã site C 17 and; (d) dendrogram made with relative area percentage showing the groupings of chemical compositions of oils extracted from different tree compartments.

3.2.2. Chemical composition of essential oil from tree compartments The plant parts used to provide material for REO extraction may have intrinsic differences, such as nutrient status, and may produce different EOs with their own unique characteristics (Krainovic et al., 2017a). The proportions in which these characteristics are mixed can influence the chemical composition of the final EO product. For example, regardless of the region of collection, age of source material, or management regime, detectable amounts of α-pinene were found not in rosewood stems, but rather in its branches and leaves, while cyclosativene was present only in rosewood branches. Table 3 α-Pinene, which is formed during the biosynthesis of limonene (Xu et al., 2017), is the most widespread pinene isomer and is highly desired by the flavors and fragrances industries (Mani et al., 2017). Oxidative metabolization of α-pinene results in other compounds (α-pinene oxide, campholene aldehyde, verbenone and verbenol) that are also important in the chemical and cosmetic industries (Cánepa et al., 2011). Due to their importance, mechanisms through which this terpenoid may be selectively produced, whether by enzymes or microbial transformation, are being studied (Paduch et al., 2016). Therefore, to ensure the presence of α-pinene in REO, a mix of branches, leaves and stem parts should be used to prepare material for distillation. This suggestion contrasts with the results of Chantraine et al. (2009), who concluded from their research in French Guiana that combining rosewood leaves and stem wood material causes a decrease in the quality of the final essential oil product. It is noteworthy that thick branches and thin branches with leaves contribute 34.4% of the mass of a whole rosewood tree (IN N0 02/2006, SDS; IN N0 05/2006, MMA; IN N0 09/2011, IBAMA), and that the proportions of plant parts used in the preparation for REO extraction can be altered, causing great variability. Cyclosativene is a very volatile tetracyclic sesquiterpene that has complex ring systems. It is widely studied due to its antioxidant, anticancer, anti-inflammatory, expectorant and antifungal properties

(Turkez et al., 2015). This sesquiterpene appears as an intermediary in a metabolic reaction chain that participates in the production of other terpenes (Lodewyk et al., 1997). Cyclosativene is one of the major compounds extracted from Luculia pinceana (Rubiaceae Hook.) and has a sweet and pleasant fragrance (Li et al., 2016). In light of these findings in the literature, the presence of this compound in the plant material we managed has important implications for the future creation of fragrances originating from REO, and it underscores the need to consider the diverse chemical composition of raw REO material as potentially promising for the development of new and varied product lines. The ordering of the data on the NMDS biplot distinguished between the REOs derived from different plant parts and confirmed that REO composition varied independently of geographical region. REO extracts from different plant parts presented a compositional uniqueness (with REOs from leaves, branches and resprouting materials being the most distinct) that was similar for both cultivation regions and followed a trend along the gradient from woodiest to least woody material (Fig. 5a–c)—the ordering of the data on the branches + leaves mix was located between the ordinations of the data on the leaf-derived and branch-derived REOs. Cluster analysis distinguished two groups by their chemical composition: one formed by REO extracted from leaves, and another formed by the other plant parts, among which stem woodderived REO was distinguished from that extracted from branches and from the branches + leaves mix (Fig. 5d). The distinction was also observed in the analyses performed separately for each region (Figs. A2–A4). 3.3. Essential oils from sequential above-ground management - differences associated with sequential harvesting Above-ground biomass management provides quick benefits and requires simple care (Spinelli et al., 2017). It is also an alternative to replanting seedlings for each tree removed during harvesting (IN N0

149

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. 6. Box plot and p-values of paired mean tests comparing first harvest (reference) and second harvest (resprouting) by sequential rosewood above-ground mass management. (a) Essential oil yield from leaves of Maués sites. (b) Essential oil yield from leaves of Novo Aripuanã site. (c) Essential oil yield from branches of Maués sites. (d) Essential oil yield from branches of Novo Aripuanã site. (e) Essential oil yield from branches + leaves mix of Maués sites. (e) Essential oil yield from branches + leaves mix of Novo Aripuanã site.

the variation in REO yield and chemical composition during different phases of sequential management of above-ground biomass in commercial plantations. Due to the repetition of results under a variety of conditions, the broad sampling of the present study provides information that can be generalized for the species, regardless of sampling region, tree age and silvicultural treatments. Qualitative and quantitative differences were observed between REOs extracted from biomass before (first harvest) and after management (resprouting). However, no significant quantitative and qualitative differences were observed between the REOs extracted from the resprouting biomass harvested by the two management forms: (i) cutting the stem at 50 cm above the soil

02/2006, SDS, Krainovic et al., 2017a). Replanting is difficult because the acquisition of such seeds is complicated by the conservation status of the species. Thus, mentions in the literature of the vigorous shoot production by cut rosewood trees (Sampaio et al., 2005, 2007) has raised interest in commercial circles, since essential oil production is directly related to biomass production (Sampaio et al., 2005, 2007; Krainovic, 2011; Fidelis et al., 2012; Krainovic et al., 2017b). However, rosewood plantation management using resprouting biomass still needs improvement and further study, especially regarding the changes that can occur in the REO composition as a function of the duration and frequency of inter-harvest intervals. Currently, no study has explored

150

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. 7. Data ordinations in the NMDS plot made with the presence/absence matrix of the substances detected in the chemical composition of essential oils extracted from each compartment collected in each harvest. The colors indicate harvest time: gray = first harvest; black = second harvest.

(coppicing) and (ii) crown pruning (pollarding) (Fig. A8 a and b). This result suggests the possibility of using resprouting material under either type of management, as opposed to first harvest material. 3.3.1. Essential oil yield associated with sequential harvesting Tree compartment maturation induced by sequential management (first harvest and resprouting) increased REO yield from leaves and decreased the yield from resprouted branches (Fig. 6). Overall, while some studies point to an increase in REO yield in young plants (Chantraine et al., 2009), others found no such difference (Fidelis et al., 2012). The reported increases in essential oil yield may often be due to leaf development (Gobbo-Neto and Lopes, 2007), as reported by Zheljazkov et al. (2009) when extracting REO from peppermint; they found a higher yield from material harvested during bud formation rather than during the subsequent flowering period. 3.3.2. Chemical composition of essential oil associated with sequential harvesting In the comparison of first harvest and resprouting, independent of the region or plant part, the NMDS sorting diagrams show where the “harvest” factor (first and second harvest) has been used as a comparative parameter (Fig. 7). The dendrogram generated from the cluster analysis confirmed the results of the NMDS analysis, distinguishing the first harvest-derived REOs from those extracted from resprouting (second harvest) (Fig. 8). One exception was the grouping of the leaf-derived REO from the first harvest with that extracted from resprouting. This result agreed with those presented previously (item 2.2.2), where differences in chemical composition among the REOs followed a trend along the gradient from most woody to least woody material, i.e. young shoots (resprouting), which are tender and green and thus similar to leaf tissues. Schimitberger et al. (2018) comparing chemistry of plant volatile compounds of the unripe and ripe fruit of Schinus terebinthifolia found that only 13 of the 98 substances were common to both samples These results were repeated in separate cluster analysis for each region (Figs. A5 and A6), with the exception of Novo Aripuanã, where grouped REOs from first-harvest branches + leaves were also grouped

Fig. 8. Dendrogram based on the relative areas composition of essential oils, showing the groups among first harvest and second harvest, independent of the cultivation region. Table 4 Substances that differentiate essential oils extracted from each plant part collected in the first and second (resprouting) harvests at commercial rosewood plantations, regardless of the region of cultivation. Underlined substances were common to all regions and were considered as differential according to the harvest time. Compartment

Only present in first harvest

Only present in resprouting

Branches Leaves

α-pinene; geraniol; cyclosativene p-cymene; geraniol; isoaromadendrene epoxide

borneol; myrcenol; NI 4 myrcenol; gurjunene

151

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

geraniol, which explains their presence and absence, respectively, in the first-harvest and resprouting materials. Al-Jaber et al. (2017) studied the chemical composition of EO extracted from different parts of Inula viscosa, and while they found myrcenol in leaf tissue (15.40%), the highest concentration of the compound was found in flowers (51.06%), which are composed of younger tissue and are important in attracting pollinators. Myrcenol is a colorless liquid with a fresh, flowery odor and a slightly lime-like scent. This substance and its derivatives are required by the perfume industry to obtain high-grade citrus and lavender odors and is therefore an important component of fragrances, perfumes and cosmetics. Myrcenol is yet another substance showing differential distribution in tissues collected during short harvesting cycles involving the biomass sprouting from coppiced rosewood.

with resprouting-derived REOs (Fig. A7). This is consistent with the results of Al-Jaber et al. (2017), who studied the aroma emitted by different parts of Inula viscosa (L.) (Asteraceae), producing a dendrogram where the parts of the plants responsible for photosynthesis had similar chemical profiles as the woody material (stem) and floral components (receptacle, flower, petal and pistil) but were grouped separately from them. Differences between the REOs from first-harvest and resprouting materials, independent of the cultivation region and plant part, were the presence of geraniol only in REOs extracted from first-harvest material and the presence of myrcenol only in REOs extracted from resprouting material (Table 4). Low levels (or absence, i.e. substance undetected) of geraniol in juvenile tissues of managed rosewood trees have also been encountered in studies of other species: In a study on fluctuations in geraniol levels during leaf development of lemongrass (Cymbopogon flexuosus Stapf : Poaceae), Ganjewala and Luthra (2009) reported that, during the growth period, the level of geraniol increased from 33 to 91% due to increased activity of an esterase involved in the conversion of geranyl to geraniol. However, the consensus in the literature is that geraniol is a primary product of terpenoid biosynthesis (Surburg and Panten, 2008) and that its synthesis requires, as with all monoterpenes, the synthesis of geranyl diphosphate (GDP). The time necessary for this synthesis to take place will be key to determining the presence of this substance in desirable concentrations and may well be one of the defining factors in determining inter-harvest rotation times for samples based on rosewood regrowth. Geraniol is one of the most important molecules in the flavor and fragrance industries and is a common ingredient in consumer products (Chen and Viljoen, 2010; Surburg and Panten, 2008). The absence of geraniol in EO extracted from resprouting rosewood may represent a serious obstacle to the commercialization of the EO derived from coppiced shoots. Geraniol has a characteristic rose-like odor and is described as “sweet floral rose, citrus with fruity, waxy nuances” (Burdock, 2016, p. 734). It is the raw fragrance material used in 76% of the deodorants in the European market and is included in 41% of household products and 33% of cosmetic formulations based on natural ingredients (Rastogi et al., 2001). Biosynthesis of important substances such as geraniol takes time, which is a key determinant of the length of the sequential harvesting cycle. A harvesting cycle of adequate length may avoid a reduction in quality that would lower the market value of the final REO product. Therefore, knowledge about chemical composition variability—whether due to the particular plant part or proportion of parts used, or due to the use of regrowth—is fundamental to the application of the above-ground biomass management. Myrcenol, present in rosewood branches and leaves, is a tertiary compound (Busmann and Berger, 1994) from a β-myrcene precursor (Behr and Johnen, 2009) that is also considered a precursor of geraniol. Structurally, the fragrances of myrcenol materials and their dihydro and tetrahydro derivatives belong to the terpenes (Surburg and Panten, 2008). The biosynthesis of these compounds originates primarily with

4. Conclusion Central Amazonian REOs from different sites, different plants parts, and different harvesting times vary in yield and quality. Global industries with interests in REO are focused on its chemical composition, with emphasis on the presence of linalool and on the bouquet of fragrances supplied by the minor components. If the REO yield is high, but the content of certain substances is low, the oil will be less valuable to the industry, resulting in a lower price. The proportions of materials to be used in REO extraction should be chosen with the aim of adding value to the final product (e.g., branches have the lowest REO yield, but extracts contain valuable substances unique to branches alone). Therefore, we conclude that adding mixtures of branches and leaves to the stem wood material during the preparation phase of REO extraction will confer the presence of certain substances in commercially valuable proportions, which must be validated by the olfactory characteristics. Further work is required to assay the correct proportions of such mixtures. Regardless of the growing region, there is a quality gradient between the REOs extracted from woody biomass and those from green biomass, and new product lines can be created to exploit these differences. While the compounds and characteristics required by industries are present in the material from both the first and second harvests, their proportions differ. Harvest cycle planning should account for these differences, as they can be used to strengthen the production chain for REO harvested in Central Amazonia and exported to the world. Acknowledgements The authors would like to thank FAPEAM - Amazonas Research Foundation for financial support to carry out the research (grant 016/ 2013-POSGRAD 2013-INPA and PAPAC project to support the publication). Thanks to Carlos Magaldi, Zanone Magaldi and Akira Tanaka for granting access to the study areas. Danilo Roberti Alves de Almeida thanks FAPESP - São Paulo Research Foundation (grant 2016/05219-9) for the financial support.

152

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Appendix See Figs. A1–A8 and Tables A1–A3.

Fig. A1. Characterization of the Maués water regime and temperatures (left); and characterization of the Novo Aripuanã water regime and temperatures (right). Both regions are classified as Amazonia Af climate. Information and graphs available at https://en.climate-data.org/.

Fig. A2. Dendrogram showing the groupings of leaves, branches, and branches + leaves mix (B + L) according to the chemical composition of essential oils derived from the first harvest in the above-ground biomass management of ten-year-old commercial rosewood plantations.

Fig. A3. Dendrogram showing the groupings of leaves, branches, and branches + leaves mix (B + L) according to the chemical composition of essential oils derived from the first harvest in the above-ground biomass management of twelve-year-old commercial rosewood plantations.

153

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. A4. Dendrogram showing the groupings of leaves, branches, and branches + leaves mix (B + L) according to the chemical composition of essential oils derived from the first harvest in the above-ground mass management of seventeen-year-old commercial rosewood plantations.

Fig. A5. Dendrogram showing the groupings of leaves, branches, and stem according to the chemical composition of essential oils derived from the first and second (resprouting) harvests in the above-ground biomass management of ten-year-old commercial rosewood plantations.

Fig. A6. Dendrogram showing the groupings of leaves, branches and stem according to the chemical composition of essential oils measured before (reference) and after (resprouting) the management of the above-ground biomass in twelve-year-old commercial rosewood plantations. 154

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Fig. A7. Dendrogram showing the groupings of leaves, branches, and stem according to the chemical composition of essential oils derived from the first harvest (reference) and second harvest (resprouting) in seventeen-year-old commercial rosewood plantations.

Fig. A8. (a) Data ordinations in the NMDS plot made with the presence/absence matrix of the substances detected in the chemical composition of essential oils extracted from resprouted branches and leaves of trees cut at 50 cm above the soil and of crown-pruned trees. (b) Box plot comparing essential oil yields from leaves and branches from crown-pruning (pollarding) and stem-cutting (coppicing) management.

Table A1 Description of chromatographic analyses with different rates of temperature variation (°C/min), and analyses ran with and without isotherm testing. Runs

Moment

Rate (°C/min)

Temperature (°C)

Hold time (min)

Oven run time (min)

Specific time (min)

1

Initial 1

60 180

0 1

11

10

12

Initial 1 2

80 160 190

1 0 1

13.67

10

12 6

Initial 1

80 200

0 1

12.82

10

11

2

3

155

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

Table A2 Substances with the highest general averages of area percentage in each class of substance found in the extracted essential oils, per plantation, per tree compartment (leaves and branches), and per harvest (first harvest and resprouting). Substance class

Substance

Constituent area percentage of essential oils (%) First harvest

Resprouting

Maués

Monoterpene hydrocarbons Oxygented monoterpenes Sesqiterpene hydrocarbons Oxygenated sesquiterpenes

Camphene Linalool β-Selinene Sphatulenol

Novo Aripuanã

Maués

Novo Aripuanã

Stem

Leaves

Branc.

B+L

Stem

Leaves

Branc.

B+L

Leaves

Branc.

B+L

Leaves

Branc.

B+L

0.114 86.12 0.437 0.112

0.048 81.32 1.409 2.526

0.178 83.88 0.385 0.371

0.434 82.11 0.641 1.010

0.164 81.77 0.564 0.249

– 71.46 2.71 3.161

0.294 81.53 0.615 0.342

– 75.57 1.623 2.031

0.303 83.59 1.416 1.984

1.034 79.49 0.631 1.043

0.710 81.33 1.159 2.068

0.769 73.89 2.196 2.132

1.625 65.08 0.769 0.930

0.971 73.24 1.790 2.158

Table A3 Differential substances found in the extracted essential oils, per plantation, per tree compartment (leaves and branches), and per harvest (first harvest and resprouting), with numbering of detected substances (n peaks). Area

Compartment

Harvest

n peaks

EO components that distinguish harvest times

Maués

Leaves

First harvest

49

Resprouting

56

Branches

First harvest Resprouting

53 57

p-cymene; eucalyptol; geraniol; β-trans-guaiene; 3-methoxymethoxy 3.7.16.20.tetramethyl eneicosa 1.7.11.15.19 pentaene; α-bisabolol; isoaromadendrene epoxide linalyl acetate; myrcenol; 1. p. menthen.9.al; Z- citral; gurjunene; α-caryophyllene; α-amorphene; aromadendrene dihydro α-pinene; eucalyptol; geraniol; cyclosativene borneol; trans-dihydrocarvone; myrcenol; gurjunene; α-amorphene; NI 1; β-trans-guaiene; aromadendrene dihydro; 2.4. diisoprophenyl.1.methylcyclohexane; NI 4; viridiflorol; α-farnesene

Leaves

First harvest

60

Resprouting

60

First harvest Resprouting

56 56

Novo Aripuanã

Branches

α-pinene; benzaldehyde; p-cymene; camphor ou camphene hidrate; geraniol; γ-celinene; aristolene epoxide; isoaromadendrene epoxide canphene; borneol; myrcenol; α-cubene ou β-copaene; gurjunene; NI 1; Z-trans-bergamotol; 2.4. diisoprophenyl.1.methylcyclohexane α-pinene; p-cymene; geraniol; E- citral; nerol; nerol acetate; cyclosativene; γ-celinene benzaldehyde; eucalyptol; borneol; myrcenol; α-caryophyllene; NI 4; 3-methoxymethoxy 3.7.16.20.tetramethyl eneicosa 1.7.11.15.19 pentaene; ledene oxide

05.008. CITES - Convention on International Trade in Endangered Species of Wild Fauna and Flora, 2010. http://www.cites.org/ (accessed 5 January 2017). Fidelis, C.H., Augusto, F., Sampaio, P.T., Krainovic, P.M., Barata, L.E., 2012. Chemical characterization of rosewood (Aniba rosaeodora Ducke) leaf essential oil by comprehensive two-dimensional gas chromatography coupled with quadrupole mass spectrometry. J. Essent. Oil Res. 24 (3), 245–251. Fidelis, C.H., Sampaio, P.T., Krainovic, P.M., Augusto, F., Barata, L.E., 2013. Correlation between maturity of tree and GC× GC–qMS chemical profiles of essential oil from leaves of Aniba rosaeodora Ducke. Microchem. J. 109, 73–77. Fittkau, E.J., Irmler, U., Junk, W.J., Reiss, F., Schmidt, G.W., 1975. Productivity, biomass, and population dynamics in Amazonian water bodies. In: Tropical Ecological Systems. Springer, Berlin, Heidelberg, pp. 289–311. Ganjewala, D., Luthra, R., 2009. Geranyl acetate esterase controls and regulates the level of geraniol in lemongrass (Cymbopogon flexuosus Nees ex Steud.) mutant cv. GRL-1 leaves. Zeitschrift für Naturforschung C 64 (3–4), 251–259. Gobbo-Neto, L., Lopes, N.P., 2007. Plantas medicinais: fatores de influência no conteúdo de metabólitos secundários. Química nova 30 (2), 374. Gupta, A.K., Mishra, R., Singh, A.K., Srivastava, A., Lal, R.K., 2017. Genetic variability and correlations of essential oil yield with agro-economic traits in Mentha species and identification of promising cultivars. Ind. Crops Prod. 95, 726–732. IBAMA. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, 1992. Portaria 06-N de 15 de janeiro de 1992. http://www.mma.gov.br/estruturas/179/_ arquivos/179_05122008033646.pdf (accessed 14 September 2017). IBAMA. Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, 1992. Portaria 37-N de 3 de abril de 1992. http://www.mma.gov.br/estruturas/179/_ arquivos/179_05122008033627.pdf (acessed 14 September 2017). IBAMA - Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis, 2011. Instrução Normativa IBAMA N° 09, de 25 de agosto de 2011. Brasília. http://www. ciflorestas.com.br/arquivos/lei_instrucao_2011_7868.pdf (accessed 14 September 2017). IUCN, The IUCN Red List of Threatened Species, Version 2014.3, 2014. http://www. iucnredlist.org (downloaded on 11 March 2015). Kottek, M., Grieser, J., Beck, C., Rudolf, B., Rubel, F., 2006. World map of the KöppenGeiger climate classification updated. Meteorol. Z. 15 (3), 259–263. Krainovic, P.M., 2011. Plantios de pau-rosa (Aniba rosaeodora ducke) em áreas com histórico de degradação por atividades agrícolas e pecuárias. Master dissertation (in Portuguese). Instituto Nacional de pesquisas da Amazônia. Manaus, Brazil. http:// bdtd.inpa.gov.br/handle/tede/1894 (accessed 5 September 2017). Krainovic, P.M., Almeida, D.R., Desconci, D., Veiga-Junior, V.F., Sampaio, P.T.B., 2017a.

References Adams, R.P., 2007. Identification of Essential Oil Components by Gas Chromatography/ Mass Spectrometry, 4th ed. Allured Publishing Co., Carol Stream, IL. Al-Jaber, H.I., Obeidat, S.M., Al-Qudah, M.A., Abaza, I.F., Lahham, J.N., Abu Zarga, M.H., Afifi, F.U., 2017. Patterns in volatile emission of different organs of Inula Viscosa growing wild in Jordan. J. Essent. Oil Bearing Plants 20 (1), 24–35. https://doi.org/ 10.1080/0972060X.2016.1277167. Albino, R.C., Oliveira, P.C., Prosdocimi, F., da Silva, O.F., Bizzo, H.R., Gama, P.E., de Oliveira, D.R., 2017. Oxidation of monoterpenes in Protium heptaphyllum oleoresins. Phytochemistry 136, 141–146. https://doi.org/10.1016/j.phytochem.2017.01.013. Barbosa, P., Medeiros, R.S., Sampaio, P.T., Vieira, G., Wiedemann, L.S., Veiga-Junior, V.F., 2012. Influence of abiotic factors on the chemical composition of copaiba oil (Copaifera multijuga Hayne): soil composition, seasonality and diameter at breast height. J. Braz. Chem. Soc. 23 (10), 1823–1833. https://doi.org/10.1590/S010350532012005000049. Behr, A., Johnen, L., 2009. Myrcene as a natural base chemical in sustainable chemistry: a critical review. ChemSusChem 2 (12), 1072–1095. https://doi.org/10.1002/cssc. 200900186. Bohlmann, J., Meyer-Gauen, G., Croteau, R., 1998. Plant terpenoid synthases: molecular biology and phylogenetic analysis. Proc. Natl. Acad. Sci. 95 (8), 4126–4133 PubMed ID: 9539701. Burdock, G.A., 2016. Fenaroli's Handbook of Flavor Ingredients. CRC Press ISBN: 978-14200-977-2. Busmann, D., Berger, R.G., 1994. Conversion of myrcene by submerged cultured basidiomycetes. J. Biotechnol. 37 (1), 39–43. https://doi.org/10.1016/0168-1656(94) 90200-3. Caceres, A.I., Liu, B., Jabba, S.V., Achanta, S., Morris, J.B., Jordt, S.E., 2017. Transient receptor potential cation channel subfamily M member 8 channels mediate the antiinflammatory effects of eucalyptol. Br. J. Pharmacol. 174 (9), 867–879. https://doi. org/10.1111/bph.1376. Cánepa, A.L., Herrero, E.R., Crivello, M.E., Eimer, G.A., Casuscelli, S.G., 2011. H2O2 based α-pinene oxidation over Ti-MCM-41. A kinetic study. J. Mol. Catal. A: Chem. 347 (1), 1–7. https://doi.org/10.1016/j.molcata.2011.06.006. Chantraine, J.M., Dhénin, J.M., Moretti, C., 2009. Chemical variability of rosewood (Aniba rosaeodora Ducke) essential oil in french guiana. J. Essent. Oil Res. 21 (6), 486–495. https://doi.org/10.1080/10412905.2009.9700225. Chen, W., Viljoen, A.M., 2010. Geraniol — a review of a commercially important fragrance material. S. Afr. J. Bot. 76 (4), 643–651. https://doi.org/10.1016/j.sajb.2010.

156

Forest Ecology and Management 429 (2018) 143–157

P.M. Krainovic et al.

domestic and occupational products. PMID: 11683833. Contact Dermatitis 45 (4), 221–225. Sampaio, P.T.B., Barbosa, A.P., Vieira, G., Spironello, W.R., Bruno, F.M.S., 2005. Biomassa da rebrota de copas de pau-rosa (Aniba rosaeodora Ducke) em plantios sob sombra parcial em floresta primária. Acta Amazonica 35 (4), 491–494. https://doi. org/10.1590/S0044-59672005000400014. Sampaio, P.T.B., Santos, M.C.dos., Vieira, G., Spironello, W., Useche, F.L., Bruno, F.M.S., 2007. Avaliação da rebrota da copa das ávores de pau-rosa (Aniba rosaeodora Ducke) em sistema de podas sucessivas. Acta Amazonica 37 (1), 55–60. https://doi.org/10. 1590/S0044-59672007000100006. Santana, A., Ohashi, S., De Rosa, L., Green, C.L., 1997. Brazilian rosewood oil: the prospect for sustainable production and oil quality management. Int. J. Aromather. 8 (3), 16–20. https://doi.org/10.1016/S0962-4562(97)80005-7. Secretaria de Estado do Meio Ambiente e Desenvolvimento Sustentável (SDS) 2006. Notícias: Semana do Meio Ambiente. Available online: http://www.sds.am.gov.br/ noticia.php?xcod=2255 (accessed on 1 February 2015). Schimitberger, V.M.B., Pratti, D.L.A., Cavalcanti, L.C., Ramalho, V.F., Costa, A.P.F., Scherer, R., Kuster, R.M., Ramos, A.C., Silva, A.G., 2018. Volatile compounds profile changes from unripe to ripe fruits of Brazilian pepper (Schinus terebinthifolia Raddi). Ind. Crops Prod. 119, 125–131. https://doi.org/10.1016/j.indcrop.2018.04.011. Shibata, R., Kurokawa, H., Shibata, M., Tanaka, H., Iida, S., Masaki, T., Nakashizuka, T., 2016. Relationships between resprouting ability, species traits and resource allocation patterns in woody species in a temperate forest. Funct. Ecol. 30, 1205–1215. https://doi.org/10.1111/1365-2435.12616. Spinelli, R., Pari, L., Aminti, G., Magagnotti, N., Giovannelli, A., 2017. Mortality, resprouting vigor and physiology of coppice stumps after mechanized cutting. Ann. For. Sci. 74 (1), 5. https://doi.org/10.1007/s13595-016-0604-z. Surburg, H., Panten, J., 2008. Common Fragrance and Flavor Materials: Preparation, Properties and Uses. John Wiley & Sons, New Jersey, USA 325p. ISBN: 978-3-52731315-0. Tanaka, A., Vieira, G., 2006. Autoecologia das espécies florestais em regime de plantio de enriquecimento em linha na floresta primária da Amazônia Central. Acta Amazonica 36, 193–204. https://doi.org/10.1590/S0044-59672006000200009. Taveira, F.S.N., De Lima, W.N., Andrade, E.H.A., Maia, J.G.S., 2003. Seasonal essential oil variation of Aniba canelilla. Biochem. Syst. Ecol. 31 (1), 69–75. https://doi.org/10. 1016/S0305-1978(02)00088-1. Turkez, H., Togar, B., Di Stefano, A., Taspınar, N., Sozio, P., 2015. Protective effects of cyclosativene on H2O2 - induced injury in cultured rat primary cerebral cortex cells. Cytotechnology 67 (2), 299–309. https://doi.org/10.1007/s10616-013-9685-9. Ucar, Y., Kazaz, S., Eraslan, F., Baydar, H., 2017. Effects of different irrigation water and nitrogen levels on the water use rose flower yield and oil yield of Rosa damascena. Agric. Water Manag. 182, 94–102. https://doi.org/10.1016/j.agwat.2016.12.004. Xu, J., Ai, Y., Wang, J., Xu, J., Zhang, Y., Yang, D., 2017. Converting S-limonene synthase to pinene or phellandrene synthases reveals the plasticity of the active site. Phytochemistry 137, 34–41. https://doi.org/10.1016/j.phytochem.2017.02.017. Young, F.J., Hammer, R.D., 2000. Defining geographic soil bodies by landscape position, soil taxonomy, and cluster analysis. Soil Sci. Soc. Am. J. 64, 989–998. https://doi. org/10.2136/sssaj2000.643989x. Zheljazkov, V.D., Cerven, V., Cantrell, C.L., Ebelhar, W.M., Horgan, T., 2009. Effect of nitrogen, location, and harvesting stage on peppermint productivity, oil content, and oil composition. HortScience 44 (5), 1267–1270 ISSN: 0018-5345. http://hortsci.ashspublications.org/.

Sequential Management of Commercial Rosewood (Aniba rosaeodora Ducke) Plantations in Central Amazonia: Seeking Sustainable Models for Essential Oil Production. Forests 8 (12), 438. Krainovic, P.M., Almeida, D.R., Sampaio, P.T.B., 2017b. New allometric equations to support sustainable plantation management of rosewood (Aniba rosaeodora Ducke) in the Central Amazon. Forests 8 (9), 327. Li, Y., Ma, H., Wan, Y., Li, T., Liu, X., Sun, Z., Li, Z., 2016. Volatile organic compounds emissions from Luculia pinceana flower and its changes at different stages of flower development. Molecules 21 (4), 531. Lodewyk, M.W., Gutta, P., Tantillo, D.J., 1997. Computational studies on biosynthetic carbocation rearrangements leading to sativene, cyclosativene, R-ylangene, and βylangene. Biochemistry 36, 8340–8348. Londero, E.K., Schumacher, M.V., Szymczak, D.A., Viera, M., 2011. Exportação e reposição nutricional no primeiro desbaste de um povoamento de Pinus taeda L. em área de segunda rotação. Ciência Florestal 21 (3), 487–497. Mani, B.K., Murthy, V., Boland, M., Yee, K., 2017. Analysis of constituents in different Fractions collected during distillation of cardamom oil for flavour and fragrance applications. J. Appl. Pharm. Sci. 7 (01), 177–183. May, P.H., Barata, L.E., 2004. Rosewood exploitation in the Brazilian Amazon: options for sustainable production. Econ. Bot. 58 (2), 257–265. McEwan, A., Magagnotti, N., Spinelli, R., 2016. The effects of number of stems per stool on cutting productivity in coppiced Eucalyptus plantations. Silva Fennica 50. MMA - Ministério do Meio Ambiente, 2006. Instrução normativa N° 05, de maio de 2006. Brasília. http://www.mma.gov.br/estruturas/pnf/_arquivos/in%20mma%2005-06. pdf (accessed 14 September 2017). MMA - Ministério do Meio Ambiente, 2014. Portaria MMA N° 443, de 17 de dezembro de 2014. Brasília. http://cncflora.jbrj.gov.br/portal/static/pdf/portaria_mma_443_ 2014.pdf (accessed 14 September 2017). Morais, L.A.S., 2009. Influência dos fatores abióticos na composição química dos óleos essenciais. Hortic. bras 27 (2), 4050–4063. Moreira, B., Tavsanoglu, Ç., Pausas, J.G., 2012. Local versus regional intraspecific variability in regeneration traits. Oecologia 168 (3), 671–677. https://doi.org/10. 1111/j.1600-0706.2011.20258.x. Paduch, R., Trytek, M., Król, S.K., Kud, J., Frant, M., Kandefer-Szerszeń, M., Fiedurek, J., 2016. Biological activity of terpene compounds produced by biotechnological methods. Pharm. Biol. 54 (6), 1096–1107. https://doi.org/10.3109/13880209.2015. 1103753. Pallardy, S.G., 2010. Physiology of Woody Plants. Academic Press, San Diego, CA, USA ISBN 978-0-12-088765-1. Pausas, J.G., Bradstock, R.A., Keith, D.A., Keeley, J.E., 2004. Plant functional traits in relation to fire in crown-fire ecosystems. Ecology 85, 1085–1100. https://doi.org/10. 1890/02-4094. Pausas, J.G., Pratt, R.B., Keeley, J.E., Jacobsen, A.L., Ramirez, A.R., Vilagrosa, A., Davis, S.D., 2016. Towards understanding resprouting at the global scale. New Phytol. 209, 945–954. https://doi.org/10.1111/nph.13644. Prochnow, D., Altissimo, B.S., da Silva, J.C., Meira, D., Caron, B.O., Heinzmann, B.M., Schmidt, D., 2017. Chemical composition of the essential oil of Aloysia triphylla (L'Hér) Britton due to water deficit and seasonality. Boletín Latinoamericano y del Caribe de Plantas Medicinales y Aromáticas 16 (2), 121–128 ISSN: 0717 7917. R Core Team Development, 2018. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. URL: http:// www.R-project.org/. Rastogi, S.C., Heydorn, S., Johansen, J.D., Basketter, D.A., 2001. Fragrance chemicals in

157