Functional understanding of secondary cell wall cellulose synthases in Populus trichocarpa via the Cas9/gRNA‐induced gene knockouts

Summary Plant cellulose is synthesized by a large plasma membrane‐localized cellulose synthase (CesA) complex. However, an overall functional determination of secondary cell wall (SCW) CesAs is still lacking in trees, especially one based on gene knockouts. Here, the Cas9/gRNA‐induced knockouts of PtrCesA4, 7A, 7B, 8A and 8B genes were produced in Populus trichocarpa. Based on anatomical, immunohistochemical and wood composition evidence, we gained a comprehensive understanding of five SCW PtrCesAs at the genetic level. Complete loss of PtrCesA4, 7A/B or 8A/B led to similar morphological abnormalities, indicating similar and nonredundant genetic functions. The absence of the gelatinous (G) layer, one‐layer‐walled fibres and a 90% decrease in cellulose in these mutant woods revealed that the three classes of SCW PtrCesAs are essential for multilayered SCW structure and wood G‐fibre. In addition, the mutant primary and secondary phloem fibres lost the n(G + L)‐ and G‐layers and retained the thicker S‐layers (L, lignified; S, secondary). Together with polysaccharide immunolocalization data, these findings suggest differences in the role of SCW PtrCesAs‐synthesized cellulose in wood and phloem fibre wall structures. Overall, this functional understanding of the SCW PtrCesAs provides further insights into the impact of lacking cellulose biosynthesis on growth, SCW, wood G‐fibre and phloem fibre wall structures in the tree.


Supplemental methods
Methods S1 Analysis of putative Cas9/gRNA off-target sites.

Methods S2 RNA extraction and RT-PCR analysis.
Methods S3 Scanning electron microscopy (SEM) of leaf epidermal cells.
The peptides selected for antibody production are highlighted in yellow. (c) The SCW PtrCesA from xylem protein extracts in the wild-type Populus was specifically identified with the corresponding antibody by western blot analysis. The full sizes of the detected PtrCesAs (indicated with asterisks) were consistent with the predicted molecular weights.   Values are means ± SD (n = 3). Asterisks denote significant difference from WT by Student's t-test: **P < 0.01, ***P < 0.001.

Fig. S13 Induction of tension wood (TW) in WT and ptrcesa mutants under gravistimulation.
(a-f) WT,ptrcesa8a,ptrcesa8b,ptrces8ab,ptrces7ab, and ptrcesa4 mutants straight grown for 4 months in a greenhouse were inclined by a 45˚ angle from the vertical direction to induce TW for 10 days. Scanning electron microscopy images were taken from the crosssections of the 16 th internode of each sample. Scale bars: 1mm.

Fig. S14 Immunolocalization of crystalline cellulose in tension and opposite wood (TW,
OW) sides in WT, ptrcesa4, 7ab, and 8ab mutants. The 8μm transverse cross-sections of the 16 th internode from each sample were incubated with CBM3a-6×His protein and anti-His antibody. CBM3a binds crystalline cellulose in plant cell wall. Signals (red) were detected with Alexa Fluor 633 goat anti-rat IgG. Arrowheads indicate G-layers of TW fibers in the WT. Scale bars: 10 μm.

Fig. S15 Immunolocalization of the xylan in tension and opposite wood (TW, OW) sides in
WT, ptrcesa4, 7ab, and 8ab mutants. The 8μm transverse cross-sections of the 16 th internode from each sample were incubated with the LM10 antibody. LM10 binds the xylan in plant cell wall. Signals (red) were detected with Alexa Fluor 633 goat anti-rat IgG. Scale bars: 10 μm.

Fig. S17 Immunolocalization of the mannan in tension and opposite wood (TW, OW) sides
in WT, ptrcesa4, 7ab, and 8ab mutants. The 8μm transverse cross-sections of the 16 th internode from each sample were incubated with the LM21 antibody. LM21 binds the mannan in plant cell wall. Signals (red) were detected with Alexa Fluor 633 goat anti-rat IgG. Scale bars: 10 μm.
Fig. S18 Immunolocalization of crystalline cellulose in phloem fibres of WT, ptrcesa4, 7ab, and 8ab mutants. Primary and secondary phloem fibers (PPF, SPF) in the 8μm transverse crosssections of the 20 th internode from each sample were incubated with CBM3a-6×His protein and anti-His antibody. Signals (red) were detected with Alexa Fluor 633 goat anti-rat IgG. CBM3a binds crystalline cellulose in plant cell wall. Scale bars: 10 μm.
Methods S1 Analysis of putative Cas9/gRNA off-target sites. Putative gRNA off-target sites were identified by a BLASTN search using a target sequence in the P. trichocarpa genome. The final off-target sites were selected according to the following criteria: two to three mismatches in the PAM-proximal region, four to five mismatches within the PAM-proximal region, or one to two mismatches in the PAM. Specific primers (Table S1) flanking the off-target sites were used to amplify the corresponding loci, and the PCR amplicons were sequenced as described above.
Methods S2 RNA extraction and RT-PCR analysis. Total RNA was isolated using the pBIOZOL Plant Total RNA Extraction Reagent (Bio-Flux, China). First-strand cDNA synthesis was performed with 2 μg RNA using the PrimeScript RT Reagent Kit with gDNA Eraser (TaKaRa, China) according to the manufacturer's procedure. Reverse transcription PCR (RT-PCR) was employed to analyze the expression of PtrCesA4, 7A, 7B, 8A and 8B genes in the ptrcesa mutants. RT-PCR analysis was performed for three biological replicates each sample.
The PtrActin2 gene was used as a reference control. The PCR conditions were set as follows: 3 min at 95°C; 28 cycles of 30 s at 95°C, 30 s at 62°C and 30 s at 72°C; and 7 min at 72°C.

Methods S3 Scanning electron microscopy (SEM) of leaf epidermal cells. Fresh 6 th leaves
below the terminal buds of 3-month-old mutant and wild-type young trees were immediately frozen in liquid nitrogen and stored until observation. These samples were laid on a copper sheet (1.5 cm × 2 cm), transferred to an SEM chamber and scanned to produce micrographs. These images were used to determine the shapes and sizes of the epidermal cells.
Methods S4 Wood fibre and vessel cell length analysis. The lengths of the wood fibres and vessel cells were determined according to the previously described method (Lautner et al., 2007) with minor modifications. The 20 th stem internodes of 6-month-old wild-type and mutant trees were peeled, cut into small pieces, and incubated in a maceration solution (10% HNO3 and 10% CrO 3 in v/v, 1:1) for 2-4 h at 60 °C. Thereafter, each stem internode was rinsed lightly with water, and the libriform fibres and vessel cells were disaggregated into the water by oscillation.
After staining with 0.1% acid magenta, images were taken under a light microscope (Olympus, BX43). The lengths of the wood fibres and vessel cells were measured using ImageJ software.

gCHLI1b-off-1 C C T A A C A G T C G T G G A C T T C C C C G CHLI2 No gCHLI1b-off-2 A T T A G A A T T G G T G G A C T T T C C G G Potri.018G015300 No gCHLI1b-off-3 G T A A A C C G T G G T G G A C T T T C T G A Potri.013G004900 No
Nucleotides in green shading are the same with the designed gRNAs target sites and mismatched nucleotides are without shading. The PAM regions are highlighted in red shading. Potential off-target sites in protein coding gene regions are annotated with gene ID and the others located in noncoding regions are detailed with position information on chromosomes.    2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

gCesA7A/B-3 T T T T G T G A C T T C A A C T T T A A T G G
PtrCesA7A -

T T T T G T G A C T T C A A C T T T A A T G G
PtrCesA7B -    T0  T1-1 T1-2 T2-1 T2-2  T1-1 T1-2 T1-