3) Supplemental data

Supplemental Figure 1. Fold basal expression of the 40S gene in the latex compared to in the inner bark tissues, from control and stimulated trees.

cDNA isolated from both latex and inner bark samples of control and 24 h Ethrel®- stimulated trees, were assayed by qPCR for 40S gene expression. The CT values for each sample series are shown in the table above.

Assuming a calculated average 87% 40S gene amplification efficiency (E=1.94) for both control and stimulated trees, and choosing the inner bark samples as the calibrator, the average relative expression of 40S gene in latex vs. inner bark samples is calculated as follows: Ratio control (test/calibrator) = ECT(calibrator)-CT(test) = 1.94(24.81-17.9) = 95 and ratio stimulated = 1.94(24.89-17.93) = 98.5

Therefore, the expression of 40S gene is about 100-fold higher in latex cell than in inner bark samples of the control and stimulated trees. Since Ethrel® stimulation had no significant effect on 40S gene expression in both tissues, this gene could be chosen as a reference gene for the study of differential aquaporins gene expression in response to Ethrel®.

Supplemental Figure 2. The nucleotide sequence of the amplified 5’upstream region of HbPIP2;1.

The translational start site is labeled +1. The putative CAAT box and ATG codon are underlined, and the putative TATA are pink boxed. Putative regulatory motifs in the HbPIP2;1 promoter region are boxed and labeled under the sequences. These motifs include a dehydration responsive element (CBF), G-box (CACGTG), GATA box (GATA), ABA responsive elements (ABREs), CuRE-core/cupper responsive element (GTAC), GT1 consensus (GRWAAW), GT1-core (GGTTAA), MYB1AT, light-induced responsive element (SORLIP1AT), sulfur responsive element (SuRE-core), MYCATRD22 (CACATG) and another light responsive element (Inr).

Supplemental Figure 3. The nucleotide sequence of the amplified 5’upstream region of HbTIP1;1.

The translational start site is labeled +1. The putative CAAT box and ATG codon are underlined, and the putative TATA are pink boxed. Putative regulatory motifs in the HbTIP1;1 promoter region are boxed and labeled above the sequences. These motifs include GATA box (GATA), Dof1 protein binding site, W-box, MYB1AT, Aux/IAA responsive element, ABRE (ACGTG)/ G-box (CACGTG), dehydration responsive element (CBF), ERE/ethylene responsive element (AWTTCAAA), CuRE-Core/cupper and SuRE-Core responsive copper and sulfur responding elements, respectively.

Supplemental Figure 4. Changes in the plasmalemma and tonoplast water conductance of the rubber tree liber cells and laticifers due to HbPIP2;1 and HbTIP1;1 aquaporin genes regulation, in response to ethylene. In the rubber tree trunk, the liber cells are interconnected through a symplastic network. As they are devoid of plasmodesmata, the laticifers can only exchange water via the aquaporin pathway. Further, the turgor of liber tissues and the fluidity of the latex are known to control latex flow, hence rubber yield.

1) In control virgin trees (A and B): Before the first tapping (A), plasmalemma water conductance of the liber cells is limited, due to relatively low HbPIP2;1 expression. Thus the symplastic pathway is the main route for liber cell-to-cell water exchanges, which are facilitated by their tonoplast high water conductance, due to high HbTIP1;1 expression. The laticifers have limited water exchanges due to low HbPIP2;1 expression. Upon tapping (B), the drop in the turgor of the laticifers together with the consecutive latex flow, induce a demand for water that can be supplied from the surrounding liber tissues. Due to their high tonoplast water conductance, the liber cells release their vacuolar water, inducing a decrease in the turgor of all the liber tissues , thus limiting latex flow and hence yield.

2) In Ethrel-treated trees (C and D): Before the first tapping (C), the increase in the HbPIP2;1 expression in both liber cells and the laticifers results in facilitated water exchanges, leading to dilution of latex. Even though the liber cells tonoplast conductance is lowered due to HbTIP1;1 down-regulation, the liber cell-to-cell water exchanges remain active, due to the additive function of their symplastic and more efficient PIP2 routes. Upon tapping (D): Due to upregulation of HbPIP2;1 in the laticifers and their surrounding liber tissues, the latter can easily fuel the demand for water by the laticifers, without releasing their vacuolar water (very low tonoplast water conductance), thus maintaining their high pressure turgor.

The combination of latex dilution (increased fluidity) and the maintenance of high turgor in the surrounding liber cells, results in a faster and prolonged latex flow, thus in a hormone-induced increase in latex yield.

La: laticifers; PC: parenchyma cells; VR: vascular ray cells; blue arrows: symplastic pathway; red arrows: HbPIP2;1; pink arrows: HbTIP1;1; solid/dot yellow lines: plasmalemma with low/high water conductance; solid/dot green lines: tonoplast with low/high water conductance. The arrows weight accounts for the intensity of the phenomenon. Blue background: apoplastic water.

Supplemental Table I.Gene specific primer pairs used for real-time PCR experiments.

Supplemental Figure 5. Determination of the liber microsomal protein concentrations

From 1 g of the inner bark tissues, the microsomal pellet obtained was finally dissolved in 35 µL membrane buffer (330 mM sucrose, 200 mM DTT, 25 mM Tris, pH 8.5) according to Bots et al. method (2005). Using the Bio-Rad kit, 10-µL membrane protein concentration measured at 595 nm absorbance was at the limit of detection but almost equal for all samples (# 0.11 µg µL-1). The protein concentrations being too low to be precisely quantified, thus equal 12-µl volume (# 1.3 µg protein) of each sample were loaded for SDS-PAGE. It is possible that the additional effects of some of the dissolving buffer reagents such as DTT, Tris and sucrose may interfere with coloration and lead to under-estimation of all samples protein concentration.

1