Whole-tree Agarwood-Inducing Technique: An Efficient Novel Technique for Producing High-Quality Agarwood in Cultivated Aquilaria sinensis Trees

The agarwood inducer was conveniently injected into the xylem of A. sinensis trees through a transfusion set. Due to water transpiration pull, the inducer was transported to the whole tree in 2 to 3 h and consequently led to internal wounds. Resinous wood formed around the wounds over several months. Interestingly, resinous wood was found in the roots and trunks as well as branches of the tree (Figure 1). An oval-shaped dark area was observed on the cross sections from almost all lateral branches 15 days after initial treatment (Figure 1b). Resinous wood was found inside the living tree trunk four months later (Figure 1c). When the resinous wood was baked with fire, it produced a soft scent and slowly released its aroma. The inducer was also infiltrated down to the root and induced the formation of root resinous wood. The trunk was cut down from 20 cm above the ground, and the root was dug out at harvesting time. Resinous wood was also separated from the white wood of these Aquilaria trees. Around the infusion hole, a unique shape of resinous wood in henna formed, consistent with the root structure (Figure 1d). Fascinatingly, we can found the whole tree stem with systemic production of resinous wood after harvest and process the whole tree (Figure 1e).

Resinous wood formed slowly after treatment by the Agar-Wit, even though without any follow-up treatment. Resin accumulated in the wood over time, making it become darker (Figure 2). A brown area and a thin resinous layer appeared inside the trunk in the first week (Figure 2a). Puce wood and a thick resinous layer were observed all over the trunk after 6 months (Figure 2b). In our previous study, 20 months after Agar-Wit treatment, black resinous wood formed in the whole tree (Figure 2c).

The cross sections of the trees treated by the Agar-Wit with ten different inducers are shown in Figure 3. The whole tree was harvested after 6 months. All of the samples treated with agarwood inducers developed resinous wood and dark areas. However, NK remained as white wood, with no resin formation, identical to CK, indicating that pure water cannot induce resinous wood formation. Therefore, agarwood inducers play an important role in formation of resinous wood.

The TLC fingerprint profiles of ten resinous wood samples A–J treated by Agar-Wit and three infected wood samples (BCD, FI, PTP) induced by the existing methods were analyzed and compared with a pure water negative control (NK) and healthy wood blank control (CK). TLC chromatograms of methanol extracts of all the samples are shown in Figure 4. Five common spots (Rf = 0.10, 0.19, 0.25, 0.31, 0.41) were detected among the samples obtained from inducer-treated and the wild agarwood under UV254nm (Figure 4a); seven common spots (Rf = 0.10, 0.19, 0.25, 0.31, 0.41, 0.53, 0.63) were detected under UV365nm (Figure 4b). These results demonstrate that the resinous wood obtained by Agar-Wit was agarwood [16]. When the TLC plate was stained with 5% vanilin-H2SO4 and heated for 10 min at 100 °C, the spots of Rf =0.31 turned pink (Figure 4c). Empirically, the deeper the pink, the better the corresponding agarwood. As shown in the figures, the pure water negative control and healthy wood blank control exhibited the same results. These results show that the quality of the agarwood induced by the Agar-Wit was similar to that of wild agarwood, and better than that of the agarwood produced by PTP, BCD and FI. Two chromone derivatives were used as standards in TLC fingerprinting. The blue spot of Rf = 0.41 was 6,7-dimethoxy-2-(2-phenylethyl)chromone (Figure 4a). The black spot of Rf = 0.73 was 2-(2-phenylethyl)chromone (Figure 4a). The agarwood samples, including two wild agarwood samples, ten induced agarwood samples by Agar-Wit and three infected wood samples by existing agarwood-inducing methods, contained a relatively high content of 2-(2-phenylethyl)chromone or 6,7-dimethoxy-2-(2-phenylethyl)chromone, except for the control samples of NK and CK.

Alcohol is an ideal solvent for extraction of various chemicals like tannins, resins, etc. Therefore, alcohol soluble extraction is frequently employed to determine the approximate resin content of drug. Generally, high-quality agarwood has high resin content and high alcohol soluble extractive content. The alcohol soluble extractive content of agarwood for medicine is required to be no less than 10% in the Chinese Pharmacopoeia [16]. According to Figure 8, the alcohol soluble extractive contents of the ten samples obtained by the Agar-Wit with ten different agarwood inducers ranged from 11.60% to 18.08%, all surpassing the required 10% standard, and similar to that of wild samples (10.56% and 19.30%). Inducer F and H brought higher alcohol soluble extractive content than other inducers, especially Inducer H which induced the highest and significantly higher content than Inducer C, E and J. Among the three existing methods, only the BCD satisfied the 10% requirement, with its alcohol soluble extractive content getting to 15.15%. For PTP treatment and FI, that value was only 7.61% and 3.15%, respectively, though both treatments had agarwood formation time longer than one year. Their alcohol soluble extractive contents were actually close to that of NK and CK.

Almost all oil from the induced agarwood with the color of yellow to brown and gave out a peculiar aroma similar to that of wild agarwood oil. The wood oil from CK and NK was white. As shown in Table 1, the oil yields of agarwood obtained by the Agar-Wit ranged from 0.14% to 0.247%, showing no difference among them. These values were close to those of the two wild agarwood, which were respectively 0.345% and 0.204%. The oil yield of the agarwood obtained by BCD was 0.199%, and that by the other two existing methods was less than 0.1%, close to those of NK and CK, while far less than that by Agar-Wit.

Neither wild Aquilaria trees nor cultivated Aquilaria trees can form agarwood without wounds caused by certain external factors, such as physical injury, insect attack or bacterial/fungal infection. It will take several years for an Aquilaria tree, which is still alive after wounding, to form agarwood around the wound [14]. Agarwood production occurs naturally to only 10% of Aquilaria trees [20]. To reduce the dependence on wild agarwood, Aquilaria species have been cultivated on a large scale, and wounding methods have been used on healthy trees to produce agarwood [2,3]. To maximize agarwood production and quality in plantations it is important to use an agarwood inducing method that produces a product steadily with high yield and good quality.

Farmers and researchers in different Asian countries have tried several wounding methods to produce agarwood, including axe chopping, nailing, and holing [2,3]. These methods often take a long time, with generally poor yield and quality in the agarwood produced. Concerning the existing agarwood-inducing methods for Aquilaria trees, few reports focused on agarwood yield and quality evaluation [7,21]. To our knowledge, there is no one agarwood-producing method used widely in Asia. In southern China, over 10 million A. sinensis trees have trunk diameter (Ф = 10 cm) sufficient for agarwood-inducing. However, due to the lack of reliable, steady and high yield agarwood-producing method, these trees can’t be ‘changed’ into valuable agarwood. According to our research, the agarwood yield per tree by the BCD method is higher than that by the PTP method as well as by the fungi-inoculation method. By the novel Agar-Wit, the yield per tree reached 6 kg after six months, 4 times higher than by the BCD method, 6 times higher than by the FI, and 28 times higher than by the PTP. All operating steps of the Agar-Wit reported in this paper, including drilling holes and installing transfusion sets, took no longer than 10 minutes per tree, and required no follow-up processing steps. The agarwood inducer was conveniently injected into the xylem of A. sinensis trees through the transfusion sets (Figure 9 and Figure 10a). Due to water transpiration pull, the agarwood inducer was transported to the whole tree and led to wounds in the whole tree. This wounding led to agarwood formation in the whole tree from the trunks to the roots and lateral branches. This is the first report concerning steady, expected and high-yield technique for agarwood production in the Aquilaria genus. The external wounds caused by the existing agarwood-inducing methods, such as PTP method and BCD method, are often detrimental to tree growth, where in some cases cause Aquilaria trees to wither and die, or be blown down by a gale. The novel Agar-Wit is less harmful and Aquilaria trees continued to grow naturally (Figure 1a). Comparing with the CA-Kit [7], Agar-Wit is more advanced and simpler. The CA-Kit should drill many holes in the tree trunk with complex operational approach, from the bottom up to top of the stem. What’s more, agarwood can be induced and formed around the holes with the CA-Kit.

Producing agarwood with similar quality to wild agarwood is another important consideration for an agarwood-inducing method. Up to now, no standard agarwood quality assessment system is available for the whole agarwood industry. According to the previous research [2,3,16,19,22], we believe that the thin layer chromatograph fingerprint profiles, alcohol soluble extractive content, content and compositions of essential oil may form a comprehensive system for agarwood quality evaluation by the Agar-Wit.

The agarwood extracts have been reported by several research teams [23,24,25] to find out the chemical constituents of agarwood. Sesquiterpenes, benzylacetone and chromone derivatives have been reported to be the major constituents of resin [17], with volatile sesquiterpenes regarded as aromatic constituents [26]. In this study, benzylacetone and 16 main sesquiterpenes compounds were chosen as an index for quality, forming the basis of comparison for all agarwood samples listed in Table 3. Benzylacetone and some chromone derivatives are pharmacologically active substances of medicinal agarwood. For example, benzylacetone is used for relieving asthma as antitussive and expectorant [27]. Chromone derivatives have antiallergic activity and show cytotoxicity against human gastric cancer cell line (SGC-7901) in vitro [28]. The agarwood formed after Agar-Wit induction contained characteristic chemical constituents such as benzylacetone, sesquiterpenoids and 2-(2-phenylethyl)chromone derivates, similar to the two samples of wild agarwood.

The test of alcohol soluble extractive content and TLC fingerprint profiles indicates that the agarwood produced by the Agar-Wit could surpass the required standard of medicinal-use agarwood in Chinese Pharmacopoeia [16].

In a previous experiment [29], we obtained two agarwood samples 20 months after Agar-Wit treatment in A. sinensis. They had very high quality similar to wild agarwood in terms of chemical constituents, alcohol soluble extractive content, and essential oil content. With ideal texture and desirable resin lines, they had a strong fragrance resembling the smell of honey or concentrated sugar, and had a bitter and pungent taste after chewing. Thus they were further classified as tagara by agarwood collectors. Their average levels of alcohol soluble extractive content (g/g) were 19.65% and 22.13%, respectively, exceeding that of wild agarwood samples. The contents of essential oil (g/g) were 0.30% and 0.36%, close to or exceeding that of the wild agarwood. The relative percentages of the sesquiterpenes in the essential oil were 75.41% and 70.08%, respectively, approximating the levels found in wild agarwood. Taken together, the agarwood formed 20 months after Agar-Wit treatment had higher quality than the wild agarwood. We suggest that, using our Agar-Wit, qualified agarwood may be produced from Aquilaria trees in 6 months, and high-quality agarwood may be obtainable if prolonging the agarwood-formation time appropriately.

Six months after Agar-Wit treatment, agarwood was formed in A. sinensis trees. With high yield and high quality close to wild agarwood, the induced agarwood could be used as raw material for medicine as well as essential oil extraction. When the agarwood-formation time was prolonged to 12 months or 20 months, higher-quality agarwood was produced. Moreover, agarwood was formed in the whole tree due to water transpiration in Aquilaria trees. Such big-sized and unique-shaped agarwood (Figure 1) could be carved and processed into art work such as sculptures and rosaries, further enhancing the added value of agarwood.

The Agar-Wit can produce high-quality agarwood at a high and steady yield. This efficient candidate technique has been patented in China (CN101755629B, CN101731282B) [15,30] by our laboratory, and PCT applications have been filed for Vietnam, Malaysia, Philippines, India and Indonesia (PCT/CN2012/071599). Its application has also been submitted in Cambodia, Bengal and Burma. So far, the Agar-Wit has been applied on 100,000 A. sinensis trees in Guangdong, Hainan, Guangxi, and Yunnan provinces (Figure 9). It has also been used in some districts of Vietnam, Cambodia, Indonesia and Malaysia. The combination of the Agar-Wit and the big cultivation areas of Aquilaria trees in Asia will not only consecutively supply more high-quality agarwood and essential oils to the international markets, but also protect wild Aquilaria trees.

Compared with high-grade wild agarwood collected from local markets, the Agar-Wit-induced agarwood had some differences in color, unit weight, and aroma. Alcohol soluble extractive content has been regarded as an index for grade partition, between first (25–30%), second (20–25%) and third (15–20%) class agarwood. The agarwood obtained by the Agar-Wit in this study attained a maximum of third-class (Figure 8), so further opportunity exists to refine the method such as prolonging production time, to increasing treatment frequency, and to improving inducers (i.e., by combining inducers with fungi inoculation for example). Moreover, as agarwood has various applications, there may be opportunity to establish different agarwood formation methods for different end uses.

The experiment of this manuscript was carried out in an A. sinensis plantation at Yanfeng Town, Haikou City, Hainan Province, China (19°32′–20°05′N, 110°10′–110°41′E). Seven-year-old A. sinensis trees with similar trunk girth were chosen as experimental materials. Four techniques, including whole-tree agarwood-inducing technique (Agar-Wit), partly-trunk-prunning method (PTP), burning-chisel-drilling method (BCD), and fungi-inoculation method (FI), were applied to induce resin formation (Figure 10).

Agar-wit (Figure 10a): very small holes deep into the xylem were drilled above 50 cm from the ground of the main trunk by an electric drill. The agarwood inducer was slowly injected into the xylem tissues through a transfusion set. Capital letters A, B, C, D, E, F, G, H, I, J respectively represented ten different agarwood inducers. Each agarwood inducer was tested in three trees for the repeating group. The composition of the ten agarwood inducers remains a technical secret. The pure water treatment was taken as negative control (NK), and healthy wood as blank control (CK).

PTP (Figure 10b): cuts of 2–4 cm wide and 3–5 cm deep were sawed along one side of the main trunk of an A. sinensis tree. The first cut was about 50 cm above the ground. The space between every two cuts was about 20 cm.

BCD (Figure 10c): the holes in the trunk from approximately 50 cm above the ground to the top of the trunk were achieved by a burning and red-hot iron drill bit (1.2 cm wide). The holes on the trunk of each tree were approximately 20 cm apart.

FI (Figure 10d): From 50 cm above the ground of a trunk, holes of approximately 8 cm deep were made by a drill. The vertical space between the holes was 20 cm, and in each horizontal line distributed two or three holes. The culture medium Melanotus flavolivens (B. etc) Sing was inserted as the bait into each hole, which was then wrapped by rubberized fabrics.

The Agar-Wit treated A. sinensis trees for the optimization of different agarwood inducers were harvested 6 months later. They were cut down from 20 cm above the ground. The length of the trunk (L) containing agarwood was measured, and the area of discoloration (S) was measured on the transverse plane.

Commercial wild agarwood (WILD1 and WILD2) was purchased from an agarwood market in China, with grade of super A that identified by the dealer (Table 3).

In order to count the agarwood yield of a large number of trees quickly, we have to estimate the yield per tree roughly. To estimate the yield of agarwood produced by Agar-Wit, we cut a trunk into three logs, and sawed off a 2 cm-thick pie respectively in 0 cm, 50 cm and 100 cm from all the felled trees (Figure 11). As agarwood was formed inside the whole tree, to accurately measure the agarwood yield per tree, we had to separate agarwood completely from the 2 cm-thick pie.

Then the resinous wood was separated from the white wood, and dried in a 40 °C oven for 15 days. After weighing the dry weight, the agarwood yield of each tree was calculated by the following formula:where mbasic pie, mmiddle pie and mupper pie represent the dry weight of resinous wood respectively isolated from the three 2 cm-thick cross sections, and L is the length of a trunk containing agarwood. The formula (1) for estimate of agarwood yield per tree is an ideal model. That is, the trunk is supposed as a symmetrical cylinder, and the ratio of resinous part in a single trunk is thought to keep the same through the whole trunk. As for the agarwood produced by the other three existing agarwood-inducing methods, all of the resinous wood was separated from the white and rotted part, weighed and calculated for agarwood yield per tree.

To evaluate agarwood quality, the resinous wood produced by the Agar-Wit and the existing agarwood-inducing methods, as well as the wild agarwood, were chopped into small chips or flakes.

The wood samples of NK and CK were placed in a 40 °C oven immediately after it was sawed off and dried for 15 days. Then they were also chopped into small chips or flakes.

All the agarwood flakes and wood of NK or CK were crushed and filtered (26 meshes). Powder (1 g) was extracted in methanol (25 mL) for 30 min by the ultrasonic method (59 kHz, 500 W, SK8200H, Kunshan, China). The extract was filtered and evaporated to dryness in an 80 °C water bath. The residue was dissolved in methanol and adjusted to 5 mL. 2-(2-Phenylethyl)chromone and 6,7-dimethoxy-2-(2-phenylethyl)chromone isolated from commercial agarwood in our lab, were used as standard constituents. Then, 2 μL of this solution was drawn into a capillary, which was then pressed onto the TLC plate (GF254, 10 × 20 cm, Merck). CHCl3:Et2O (10:1, v/v) was used as the mobile phase. The TLC plate was developed twice and then visualized in UV254nm and UV365nm. The plate was also viewed after being treated with 5% vanillin- sulfuric acid and heated at 105 °C.

Alcohol soluble extractive content of the samples was determined following the procedures given in the Chinese Pharmacopoeia [16]. Coarsely powdered dried sample (2 g) was macerated with alcohol (100 mL) for one hour, then heated under reflux for one hour. After cooling and filtering, and taking precautions against loss of solvent, the filtrate (25 mL) was evaporated to dryness in a tared flat bottomed shallow dish, and dried at 105 °C for 3 h, to constant weight and weighed. The percentage of alcohol soluble extractive was calculated with reference to the dried sample powder. The experiment was repeated twice.

The essential oil was extracted following the procedure given in the Chinese Pharmacopoeia [16]. The determination of essential oil in a sample is made by distilling a sample (50 g) with water (800 mL) in a volatile oil determination apparatus, collecting the distillate in a tube in which the aqueous portion of the distillate is automatically separated and returned to the distilling flask. The extracted essential oil was isolated and dried with anhydrous sodium sulfate, weighed and stored in sealed amber flasks at −20 °C until analysis. Calculate the percentage of essential oil with reference to the dried sample powder.

The main compounds in the essential oil were identified using GC-MS or GC-RI following the reported methods [19].