Intensity Intensity Intensity Intensity Intensity Intensity 150 50 150 0 10 20 50 C 150 0 10 20 50 D 0 10 20 Distance (μm) 50 20 40 E 50 F 0 10 20 50 0 15 30 Distance (μm) Supplemental Figure 1: Co-localization of HDC1 with HD6 and HD19 within nuclei of transiently expressing tobacco epidermis cells. High-magnification images of nuclei in tobacco (N. benthamiana) epidermal leaf cells after transient expression of GFP-HDC1 and RFP-HD6 or RFP-HD19. Each row contains the following images from left to right: bright field, GFP fluorescence, RFP fluorescence, GFP/RFP overlay, quantitative comparison GFP and RFP signals along line scan (arrows in overlay images). HDC1 co-localizes with HD6 (-C) and HD19 (D-F) in the entire nucleus (, D), in distinct speckles (-E) or in the nucleolus (C, F). Scale bar is 10 µm. 1
HDC1 mrn /Tub9 1 2 2 3 3 4 4 1 C WT hdc1-1 Pair 1 Pair 2 Pair 3 Pair 4 Tub 9 HDC1-GFP HDC1 Rubisco D 4000 3000 hdc1-1 WT GFP-HDC1. thaliana N. tabacum 2000 0 0 WT 35S Ubi10 HDC1 HDC1 Supplemental Fig. S2 Supplemental Figure 2: Confirmation of hdc1-1 knockout and HDC1 over-expressing lines : Position of T-DN and primer pairs in the genomic DN of hdc1-1 knockout line (GI-Kat 054G03). Numbers indicate position of primer pairs used for genotyping. : HDC1 mrn in wildtype and hdc1-1 as determined by RT-PCR using the primer pairs indicated in. Tubulin 9 (Tub 9) was used as a loading control. C: Western blot with anti-hdc1 detects HDC1 in. thaliana wildtype but not in hdc1-1. Detection of the larger HDC1-GFP fusion protein transiently expressed in tobacco is shown for comparison. Rubisco (loading control) was detected by Ponceau staining. D: HDC1 mrn levels (relative to Tub 9) in two lines overexpressing HDC1 under control of 35-S or Ubiquitin-10 promoters. sterisks indicate significant differences (p < 0.001) to wild type (WT). 2
HDC1 mrn /RPII Germination (%) Salk_ 150126 1 1 2 2 Sail_1263_E05 C WT Salk_150126 Sail_1263_E05 1 1 2 2 4 4 80 0.012 Primer pair 1 0.06 Primer pair2 60 40 0.008 0.04 20 0.004 0.02 0 0 0.2 0.3 0 WT Salk 150126 Sail 1263 0 WT Salk 150126 Sail 1263 [] /mm Supplemental Fig. S3 Supplemental Figure 3: Salk150126 and Sail1263E05 are not hdc1 knockouts : Position of T-DN and primer pairs in the genomic DN for Salk_150126 and Sail_1263_E05 lines. : HDC1 mrn levels in wildtype, Salk_150126 and Sail_1263_E05 using the primer pairs indicated in. RpII is RN polymeraseii (loading control). sterisks indicate significant differences to the wild type (p<0.05). C: Germination rates of. thaliana wildtype (black), Salk_150126 (grey stripes) and Sail_1263_E05 (light grey stripes) on agar containing different concentrations of. ars are means +/- SE of at least 3 plates containing at least 50 seeds each. Note that neither of the lines shows hypersensitivity. 3
mrn /TU9 Control + 0.3 mm WT KO OX 2 DG 6 DG Supplemental Fig. S4 Supplemental Figure 4: Without HDC1 does not alter germination or progression of seedlings to vegetative stage : ppearance of young seedlings on day 6 after sowing. Wildtype (upper third of plate), hdc1-1 (centre) and OX (lower) seeds were imbibed and allowed to germinate on half strength Murashige Skoog medium without (control) or with 0.3 µm added. Pictures were taken on the same day as germination rate was scored. Note that without, number and size of seedlings is similar for all lines. : Transcript levels for embryogenesis related genes I3, FUS3 and LEC1 in wildtype (WT, black), hdc1-1 knockout (KO, white) and HDC1 over-expressing (OX, grey) seedlings 2-6 days after germination (DG). ars represent means of 4 technical qpcr replicates with mrn pooled from 50 seedlings. sterisk indicates significant difference to wildtype (p < 0.05). 4
Plant weight /g [] /mm C FW DW FW DW Supplemental Fig. S5 Supplemental Figure 5: HD6 over-expression does not affect germination or growth. : Germination rates of imbibed wildtype (black), 35S:HDC1 (light grey) and 35S:HD6 (dark grey) seeds. Germination rates in % reflect the number of seedlings that had developed cotyledons on day 6, normalized to the total number of seeds plated out. ars are means ± SE of 3 plates containing 50 seeds each. sterisks indicate significant differences (p < 0.05) to wildtype. : Transcript levels of HD6 in wildtype and 35S::HD6 lines. C: Shoot weights (FW: fresh weight, DW: dry weight of 5-weeks old plants). ars are means of 8 plants. 5
HDC1-OX1 HDC1-OX2 Wildtype hdc1-1 Supplemental fig. S6 Supplemental Figure 6: Root and leaf growth phenotypes in young plants. Leaf surface areas and main root length of 2-weeks old wildtype (black), hdc1-1 (white) and HDC1-OX (grey). For leaf surface measurements plants were grown on soil. fter 2 weeks, leaves were removed in order of appearance and analyzed with Image J. For root measurements plants were germinated and grown on vertical agar plates containing half-strength MS medium. ars are means ± SE of 3 plants for leaf surface data and of at least 8 plants for root data. sterisks indicate significant differences (p < 0.05) to wildtype. 6
Plant weight /g FW DW FW DW Supplemental Fig. 7 Supplemental Figure 7: HD6 knockdown affects plant growth without delaying leaf development : Fresh and dry weights of 4-weeks old wildtype (Col-DR5, black) and hda6-knockdown (axe1-5, white dotted) plants. : Leaf numbers in wildtype and axe1-5 mutants. ars are means ± SE of 5 plants. sterisk indicates significant difference (p < 0.05) to wildtype. 7
[] /%WT+ [] ng/mg DW - + - + - + - + - + - + - + - + - + - Control + 150 mm NaCl for 24 h - + - + - + Supplemental Figure 8: HDC1 has a small effect on content after salt treatment. : Shoot content of wildtype (WT, black), hdc1-1 knockout (KO, white) and HDC1 over-expressing (OX, grey). Plants were grown for 4 weeks in short day conditions and subjected (+) or not (-) to 150 mm NaCl for 24 h in hydroponics. bsolute results from three independently treated plant batches (5 plants each) are shown. : Relative change of content in hdc1-1 and HDC1-overexpressing plants compared to wildtype. content was normalized to the content of salt-treated wildtype plants in the same batch. ars are means ± SE of 3 independently treated plant batches (same as in ). Differences were significant with P-values of p = 0.3 (OX/WT, ) or p = 0.14 (KO/ WT). 8
ChIP DN /CT2 /Input 1 3 OX WT KO OX WT KO Supplemental Figure 9: HDC1 alters H3K9K14 acetylation levels in 1 but not 3 Relative amounts of DN associated with acetylated H3K9/K14 for 1 and 3 determined by ChIPqPCR in wildtype (WT, black), hdc1-1 knockout (KO, white) and HDC1 over-expressing (OX, grey) plants. Leaf tissue was pooled from 4-weeks old plants grown in 3 independent batches 12 plants each. Chromatin extracted and immunoprecipitated with anti-h3k9k14c. qpcr-amplified ChIP-DN was normalized to actin 2 and to input DN (chromatin before immunoprecipitation). ars are means of 4 technical qpcr-replicates ± SE. sterisks indicate significant differences to the wild type (p < 0.05). 9
HDC1 Control Gene X No expression Stress induced Low sensitivity hdc1 No expression induced High sensitivity HDC1 Gene Y Low expression repressed High sensitivity hdc1 High expression repressed High sensitivity Supplemental Figure 10: Scenarios for the effects of HDC1 on -dependent gene expression : In the case of -inducible gene X, HDC1 decreases sensitivity without altering transcription in control conditions. No transcription factors are present in control conditions and hence the gene remains inactive independent of its de-acetylation state. Under stress the gene is transcribed through the action of an -induced activator (e.g. transcription factor). How much of the activator is required for gene activity depends on the histone acetylation status of the gene. Thus, to obtain the same effect, more stimulus () is required when HDC1 is overexpressed (hypo-sensitivity) and less stimulus is required when HDC1 is knocked out (hyper-sensitivity). : In the case of -repressed gene Y, HDC1 decreases constitutive transcript levels with or without altering -sensitivity. In this case transcription factors are present in control conditions and have to overcome the repressive effect of deacetylation. HDC1 therefore directly determines constitutive expression levels. Stress induces a repressor that effectively blocks transcription. Whether knockout/overexpression of HDC1 alters stress-sensitivity will depend on the level at which the repressor operates. In the scenario shown here the -induced repressor acts downstream of transcription factor activity and is therefore independent of HDC1 levels. 10
% Control % Control Water-limited Shoot Rosette FW DW Salt-stressed Root Shoot HDC1-OX1 HDC1-OX2 Wildtype hdc1-1 FW DW FW DW Supplemental Figure 11: Effects of HDC1 expression on stress-sensitivity during vegetative growth. : Rosette diameter, fresh weight (FW) and dry weight (DW) at the end of the water-limiting experiment (day 40) relative to control (well-watered). For labelling of lines see legend box in figure. : Root and shoot fresh weight (FW) and dry weight (DW) of plants at the end of the salt stress experiment (day 6) relative to control. Due to low root DW, plants had to be pooled for determination of DW. 11
Supplemental Table 1: Primers for genotyping gdn
Supplemental Table 1 cont.