The secondary laticifer differentiation in rubber tree is induced by trichostatin A, an inhibitor of histone acetylation

The secondary laticifer, a specific tissue in the secondary phloem of rubber tree, is differentiated from the vascular cambia. The number of the secondary laticifer in the trunk bark of rubber tree is positively correlated with rubber yield. Although jasmonates have been demonstrated to be crucial in the regulation of secondary laticifer differentiation, the mechanism for the jasmonate-induced secondary laticifer differentiation remains to be elucidated. By using an experimental morphological technique, the present study revealed that trichostatin A (TSA), an inhibitor of histone deacetylation, could induce the secondary laticifer differentiation in a concentrationdependent manner. The results suggest that histone acetylation is essential for the secondary laticifer differentiation in rubber tree.


Introduction
Rubber tree (Hevea brasiliensis) is the main source of natural rubber in the word [1] . The rubber biosynthesis occurs uniquely in laticifer, a specific tissue [2,3] . The number of the secondary laticifer in the trunk bark of rubber tree is positively correlated to the rubber yield. The secondary laticifer is consist of numerous laticifer cells that are differentiated from the fusiform initials of vascular cambia [4] . The secondary laticifer differentiation is induced by such factors as jasmonic acid (JA) [4] , mechanical wounding [5] and coronatine (COR) [6] , an active jasmonate homolog [6] . The coronatine is more effective than jasmonate acid and methyl jasmonate (MeJA) in inducing the secondary laticifer differentiation [6] . The mechanical wounding-induced secondary laticifer differentiation is limited to the wounded area and depends on endogenous jasmonates which acts downstream of reactive oxygen species caused by dehydration of the wounded tissues [7][8][9] . In addition, the induction of secondary laticifer differentiation induced by both the mechanical wounding and methyl jasmonate was influenced by season, which is associated with the content of endogenous cytokinin and the activity of vascular cambia [10] . Here, we reported the effect of trichostatin A (TSA), an inhibitor of histone deacetylation [11] , on induction of the secondary laticifer differentiation, which gives a new clue for understanding the regulation of the secondary laticifer differentiation in rubber tree.

Plant materials
Rubber tree clone CATAS7-33-97 budded on rootstocks were grown at the Experimental Station of the Rubber Research Institute (RRI) of the Chinese Academy of Tropical Agricultural Sciences (CATAS), Danzhou, Hainan Province, China. The plants were pruned each year and epicormic shoots grew from the latent buds on the pruned branches. Each epicormic shoot flushed five to six times a year and includes a series of foliage clusters which were classified by the length of stem. Therefore, each of these morphologically distinct growth increments represents a growth flush and is referred to as an extension unit (EU) [4] . Under natural condition, none of the secondary laticifer appear in the stem of EU1 and EU2 (counted from the top of the shoot) [4,5,10] , which is convenient for distinguishing the induced secondary laticifer.

Treatments
In June, when the epicormic shoots had produced more than two extension units, the epidermis and outer parts of cortex of 0.5 cm Â 1 cm area of stem were scratched with a razor blade at the site 1 cm below the lowest foliage leaf of EU1. The wounded surfaces were (1) directly exposed to air (mechanical wounding), (2) applied with sterile water or 5% dimethyl sulfoxide (DMSO) and wrapped immediately with parafilm, (3) applied with 0.07% methyl jasmonate and wrapped immediately with parafilm, (4) applied with a gradient concentration of coronatine and wrapped immediately with parafilm, and (5) applied with a gradient concentration of trichostatin A and wrapped immediately with parafilm. Coronatine was purchased from Sigma (St. Louis, MO, USA) and trichostatin A was purchased from Selleck (Houston, TX, USA). Samples were collected seven days after treatments. Five replicates were conducted for each treatment. To test the effect of trichostatin A on the secondary laticifer differentiation, the treatments applied were water (negative control) as well as coronatine, mechanical wounding and methyl jasmonate (positive control).

Light microscopy
To eliminate tannin-like substances, which can be mistaken for the rubber inclusion in laticifer, bark samples were fixed in 80% ethanol for 24 h at room temperature, treated with iodine and bromine in glacial acetic acid, and embedded in paraffin after dehydration. Sections (12 mm in thickness) were cut with a microtome and stained with fast green. Laticifer in sections could be recognized since rubber in the laticifer was in brown color due to the iodinebromine treatment. Sections of the bark samples were made for examination under a light microscope (Leica DMLB, Leica Microsystems Wetzlar GmbH, Wetzlar, Germany) and the phloem tissues in bark cross sections were measured.

Statistics
Data collection from five sections of independent shoots was referred to Hao and Wu [4] . Significant deviation analysis was performed by Duncan's test (P < 0.05).

Results
There was no effect of water on induction of the secondary laticifer differentiation 7 d after treatment ( Fig. 1a; Table 1). But one line of the secondary laticifer was induced in the secondary phloem of the stem by mechanical wounding (Fig. 1b) or 0.07% methyl jasmonate (Fig. 1c). Although no secondary laticifer was detected in the stem treated with 100 nmol$L -1 coronatine (Fig. 1d), few of the secondary laticifer cells (0.092 lines) appeared near the vascular cambia in the secondary phloem when the coronatine concentration increased to 1 mmol$L -1 for 7 d (Fig. 1e). The number of the secondary laticifer cells increased noticeably (0.234 lines) when the concentration increased to 2.5 mmol$L -1 (Fig. 1f; Table 1). The secondary laticifer cells exhibit as a continuous line in the secondary phloem of stem bark applied with 10 mmol$L -1 coronatine (Fig. 1g). There was no significant difference in the number of secondary laticifers between the treatments of 10 and 20 mmol$L -1 coronatine (Fig. 1h; Table 1).
Under the conditions when all the mechanical wounding, jasmonates and coronatine could induce the secondary laticifer differentiation, we found the trichosatatin A (TSA), a histone deacetylation inhibitor, had also effect on inducing the secondary laticifer differentiation. Similar to the coronatine effect, TSA induced the secondary laticifer differentiation in a concentrationdependent manner. As the solvent of TSA, 5% dimethyl sulfoxide had no effect on inducing the secondary laticifer differentiation (Fig. 2a). Same was for 1 nmol$L -1 trichostatin A (Fig. 2b). There appeared several laticifer cells upon treatment with 10 nmol$L -1 TSA (Fig. 2c), although they were somewhat difficult to be detected. When TSA concentration was increased to 100 nmol$L -1 , the number of the secondary laticifer cells increased obviously ( Fig. 2d; Table 1). Even more the secondary laticifer cells were detected upon treatment with 1 mmol$L -1 TSA ( Fig. 2e; Table 1). However, there was no difference in the number of the secondary laticifers between treatments of 1 mmol$L -1 and 10 mmol$L -1 trichostatin A (Fig. 2f; Table 1).

Discussion
The effect of linolenic acid and mechanical wounding on inducing the secondary laticifer differentiation may be ascribed to the increased level of endogenous jasmonates [7] . Although the jasmonic acid and methyl jasmonate are not the active form, (+)-7-iso-jasmonoyl-Lisoleucine [8] , they are effective in inducing the secondary laticifer differentiation [4,10] . The coronatine could structurally and functionally mimic the active form of jasmonates [6] . It bonds to COI1, the receptor of active jasmonate, and activates jasmonate signaling [12] . With respect to the applied concentration, coronatine is more effective than jasmonic acid [9,11,[13][14][15] and methyl jasmonate [16] while less effective than TSA in inducing the secondary laticifer differentiation.
Changes in epigenetic modification are involved in regulating the active state of tumor suppressor genes and proto-oncogenes [17] . Therefore, tumor is considered to be a kind of cellular epigenetic disease [18][19][20] . Such histone deacetylation inhibitors as O-benzylhydroxylamine   Fig. 1 Representative photo in Fig. 2 No  hydrochloride (trichostatin A), benzhydroxamic acid derivative (SAHA), cyclic depsipeptide, benzamide derivatives (MS-275 and CI-994) and fatty acid (valproic acid and phenylbutyric acid) can inhibit the cell proliferation or activate cell apoptosis in the case of hepatocellular carcinoma [21] , esophageal squamous cell carcinoma [22] , colon mucosa cancer [23] . So, the histone deacetylation inhibitor (HDACi) [24,25] as well as DNA methyltransferase inhibitors (DNMTs) [26] may serve as a new way of "epigenetic therapy." In the present study, TSA is effective in inducing the secondary laticifer differentiation in rubber tree, suggesting that inhibition of cell proliferation may result in activation of cell differentiation.
Post-translational modifications of histones are involved in modulating dynamic changes in chromatin structure and gene activity [27] . Histone acetylation is generally associated with transcription activation, whereas histone deacetylation results in transcription repression [12,15] . The dynamic changes in deacetylation and acetylation of histones are modulated by histone deacetylase (HDA) and histone acetyltransferase [28] . By modulating the histone acetylation or directly interacting with target proteins, the HDAs have pleiotropic effects on regulating plant growth and development [16][17][18][19] , abiotic stress responses [18,20,26] and photomorphogenesis [24] . The TSAinduced secondary laticifer differentiation suggests that HDA(s) may negatively regulate the jasmonate signaling. It is well known that Jasmonate ZIM-domain(JAZ) proteins act as repressors of jasmonate signaling [29] . Available data show that JAZ proteins (JAZ1, JAZ3, JAZ9) recruit HDA6 as a corepressor to modulate histone acetylation and inhibit jasmonate signaling [30] . In addition, the HDA6 also interacts with COI1 in Arabidopsis [31] . On the other hand, the HDA6 seems to be required for the expression of jasmonate responsive genes in Arabidopsis HDA6 mutant, axe1-5, and HDA6 RNA-interfering plants [25] .

Conclusions
Histone acetylation may be associated with jasmonate signaling in regulating the secondary laticifer differentiation. Elucidating their relationship will deepen our knowledge about the secondary laticifer differentiation in rubber tree.