Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
1
Cuticular Wax Composition is Essential for Plant Recovery Following Drought with
Little Affect under Optimal Conditions
Boaz Negin, Shelly Hen-Avivi, Efrat Almekias-Siegl, Lior Shachar and Asaph Aharoni
Department of Plant and Environmental Sciences, Weizmann Institute of Science, Rehovot,
Israel
Abstract
Despite decades of extensive study, the role of cuticular lipids in sustaining plant fitness is far
from being understood. To answer this fundamental question, we employed genome editing in
tree tobacco (Nicotiana glauca) plants and generated mutations in 16 different cuticular lipids-
related genes. We chose tree tobacco due to the abundant, yet simply composed epicuticular
waxes deposited on its surface. Five out of 9 different mutants that displayed a cuticular lipids-
related phenotype were selected for in depth analysis. They had either reduced total wax load
or complete deficiency in certain wax components. This led to substantial modification in
surface wax crystal structure and to elevated cuticular water loss. Remarkably, under non-
stressed conditions, mutant plants with altered wax composition did not display elevated
transpiration or reduced growth. However, once exposed to drought, plants lacking alkanes
were not able to strongly reduce their transpiration, leading to leaf death and impaired recovery
upon resuscitation, and even to stem cracking, a phenomenon typically found in trees
experiencing drought stress. In contrast, plants deficient in fatty alcohols exhibited an opposite
response, having reduced cuticular water loss and rapid recovery following drought. This
deferential response was part of a larger trend, of no common phenotype connecting plants
with a glossy appearance. We conclude that alkanes are essential under drought response and
much less under normal non-stressed conditions, enabling plants to seal their cuticle upon
stomatal closure, reducing leaf death and facilitating a speedy recovery.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
2
Introduction
Plants aerial organs are exposed to a plethora of conditions that require a barrier to shield their
inner tissues. Such layers protect them from many factors including a desiccating environment,
high radiation in different wavelengths and colonization or feeding by other organisms such as
fungi, bacteria, insects and even grazing mammals. To cope with these, plants developed a
complex surface layer, which at times may also contain specialized components that aid the
plant in its response to the different factors. The cuticle may be imbedded with flavonoids
(Ryan et al., 2001; Ryan et al., 2002) and anthocyanins (Chalker-Scott, 1999) reducing UVB
damage to the mesophyll. Trichomes present on the surface may serve as a structural and
chemical obstacle for plant herbivores (Hanley et al., 2007) and change leaf reflectance,
reducing photoinhibition and UV-B related damage (Steffens and Walters, 1991; Peiffer et al.,
2009; Sonawane et al., 2020; Bickford, 2016). The plant cuticle consists of three core elements
localized beyond the epidermis cells. The cutin polymer composed of an amorphous matrix
largely comprising C16 and C18 fatty acids connected in esteric bonds (Yeats and Rose, 2013),
intracuticular wax embedded in the cutin layer and epicuticular waxes secreted beyond the
cutin. Though it is clear that the cuticle plays a crucial role in preventing non stomatal water
loss, the extent to which epicuticular wax contributes to this varies widely and in some cases
may be negligible compared to the other two elements (Jetter and Riederer, 2016; Zeisler and
Schreiber, 2016; Zeisler-Diehl et al., 2018). This keeps open the question of epicuticular wax’s
function, in cases where it’s contribution to the transpirational barrier in small. A related
question, is to what extent the different wax components contribute to the different functions
of epicuticular waxes.
Decades of extensive research, mainly in the model species Arabidopsis thaliana resulted
in elucidation of most biosynthetic pathways associated with production of the different wax
components (Fig. 1; Lee and Suh, 2015). In Arabidopsis, fatty acids synthesized in plastids are
transported to the endoplasmic reticulum (ER) where they are converted to acyl CoA by LONG
CHAIN ACYL SYNTHETASE (LACS; Schnurr et al., 2004; Lü et al., 2009). The fatty acid
elongase complex composed of four proteins then elongates these C16-C18 precursors. These
include (i) -ketoacyl-CoA synthase (KCS; Millar and Kunst, 1997) having different variants
responsible for elongation of certain acyl substrates (e.g. KCS6 targeted here is responsible for
elongation from 24 to 34 carbons (Millar et al., 1999), (ii) -ketoacyl-CoA reductase (KCR;
Beaudoin et al., 2009), (iii) 3-hydroxyacyl-CoA dehydratase (HCD; Bach et al., 2008) and (vi)
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
3
trans-2,3-enoyl-CoA reductase (ECR; Zheng et al., 2005). These enzymes elongate the Acyl
Figure 1. A schematic model of epicuticular wax biosynthesis and transport pathways.
Not all proteins involved in these pathways are presented and genes corresponding to proteins
marked in red were targeted for editing and exhibited a visual phenotype in this study. The
scheme presents fatty acid elongation and division to the two main wax synthesis pathways –
the alkane and fatty alcohol forming pathways. Genes silenced that are involved in wax
synthesis but not a specific pathway include KCS6 and LACS. Those taking part in alkane
synthesis are CER1, CER3 and MAH1, and fatty alcohol synthetic genes include FAR and
WSD1.
CoAs through a cycle in which malonyl CoA and the acyl are condensed and at its end the acyl
is elongated by two carbons. The outcome of the elongation process are very long chain fatty
(VLCF) acyl CoA of different lengths used in several epicuticular wax biosynthesis pathways.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
4
They can be converted into very long chain fatty acids (VLCFA) or enter two paths, the alkane-
and alcohol- forming pathways. In the alkane-forming pathway, VLCF acyl-CoAs are
decarboxylated to create aldehydes and then alkanes. The exact enzymatic mechanisms taking
place during the conversion of acyl-CoA to alkanes are not clear, though it is known that
ECERIFERUM1 (CER1) and ECERIFERUM 3 (CER3) take part in the first stages of
conversion to alkanes (Aarts et al., 1995; Chen et al., 2003; Bourdenx et al., 2011). Following
the synthesis of alkanes, these can be further modified to form secondary alcohols and ketones,
by the MID-CHAIN ALKANE HYDROXYLASE1 (MAH1); a cytochrome p450 catalyzing
both reactions (Greer et al., 2007). In the primary alcohol forming pathway, VLCF acyl-CoAs
are converted to primary alcohols through the addition of a hydroxyl at the acyls end. This step
is catalyzed by the FATTY ACYL-COA REDUCTASE (FAR) enzyme (Rowland et al., 2006).
An additional stage in the alcohol forming pathway is the conjunction of C16 or C18 fatty acids
to the primary alcohols to create wax esters by the bifunctional wax synthase/ acyl
CoA:diacylglycerol acyltransferase1 (WSD1; Li et al., 2008). Although the presence of
epicuticular wax is almost ubiquitous among plants, there is great diversity between species
and even different organs of the same plant, in chemical structure and form of wax crystals
they produce (Barthlott et al., 1998; Lee and Suh, 2015).
Tree tobacco (Nicotiana glauca) is a perineal shrub originating in South America which has
since spread worldwide. The appearance of its stems and leaves is glaucous, an uncommon
appearance in tobacco species which gave the species its name. This glaucous appearance, is
the result of a high load of epicuticular wax coating the plant’s aerial organs. This wax is
composed almost solely of C31 alkanes with lesser amounts of additional alkanes, fatty alcohols
and aldehydes (Mortimer et al., 2012). Furthermore, N. glauca accumulates substantial
amounts of wax in response to drought, while maintaining its original composition (Cameron
et al., 2006). These characteristics including the availability of its genome sequence (Usadel et
al., 2018), efficient stable transformation and the ease of dissecting its epidermis layer make it
an excellent model plant for studying the yet unresolved role of epicuticular waxes in the plant
life cycle.
In this study, we employed CRISPR-Cas9 based technology to edit 16 cuticular lipids
metabolism genes and generate mutants in N. glauca. Following initial characterization, we
carried out in-depth research on five selected knockout mutants that showed diverse patterns
of wax composition. This mutant set was subjected to a range of examinations to help us better
understand what the contribution of epicuticular wax is to plant fitness, and to attribute this
contribution to specific wax components. We found, that under optimal conditions epicuticular
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
5
wax has little effect on plant fitness. In contrast, following drought the alkane fraction is
essential for plant recovery, whereas deficiency in fatty alcohols did not affect plants negatively
under our experimental conditions. Findings in this study highlight the specific role of plant
epicuticular wax in episodes of drought and the following recovery phase. They further explain
how diversity of wax components and structures contribute to plant fitness under abiotic stress
conditions.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
6
Results
Transcriptomics in Nicotiana glauca epidermis tissues under drought conditions
facilitates the discovery of cuticular lipids-related genes
The most striking feature of N. glauca is the glaucous waxy appearance formed by an extremely
high but very simply composed wax layer covering its above ground organs. Such intense wax
coverage provokes a fundamental question with respect to the high carbon investment in
epicuticular wax production set against its contribution to plant fitness. Which wax component
contributes to fitness and in what manner is a consequent key question. To answer these, it was
first essential to identify cuticular lipids-associated genes active in the N. glauca epidermis
layer. We thus performed transcriptomic analysis of five different shoot tissues including
dissected adaxial and abaxial leaf epidermis, stem epidermis, complete leaves, and whole
stems. Expression profiling was carried out in well-watered plants or those under drought
conditions, as a previous study demonstrated drought-induced wax formation in N. glauca
leaves (Cameron et al., 2006). To detect genes associated with cuticular lipids formation we
next mined the transcriptome dataset for transcripts that were epidermis enriched, drought
induced or a combination of the two conditions. We found that many homologs of cuticular
lipids metabolism genes (e.g. KCS6, CER1, FAR and ABCG32), displayed a similar ‘expression
signature’; high epidermal enrichment and mild drought induction (Fig. 2). Out of tens of genes
putatively associated with cuticular lipids metabolism we selected 16 for further study. Editing
these genes could alter wax and cutin synthesis and transport in a variety of manners resulting
in a set of mutant lines with diverse was compositions (Table S1). This set comprised seven
genes putatively encoding wax biosynthesis enzymes including LACS1, KCS6, CER1, CER3,
MAH1, FAR and WSD1 and three cutin biosynthesis enzymes - Glycerol-3-phosphate 2-O-
acyltransferase 4 (GPAT4), Cytochrome P450 86A22 (CYP86A22) and GDSL-motif
esterase/acyltransferase/lipase (GDSL). Three putative transporters - ATP-BINDING
CASSETTE G11 (ABCG11), ATP-BINDING CASSETTE G32 (ABCG32) and ACYL-COA-
BINDING PROTEIN 1 (ACBP1) and three transcription factor homologs - SHINE1, SHINE3
and MYB96, complemented the set of selected genes. We next generated plant transformation
vectors for editing the 16-gene set through CRISPR-Cas9 technology and introduced them to
N. Glauca (Fig. S2-S10).
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
7
Figure 2. Epidermis enriched and drought-induced expression of N. Glauca genes
showing homology to known cuticular lipids genes from other species. For each sample,
three biological replicates of each tissue, grown either under well-watered conditions or
drought treated, were collected and pooled to reduce variance though not increasing the number
of sequenced samples. The different genes are color coded according to their function;
transcription factors in red (A-C), cutin synthesis genes in orange (D-F), wax synthesis genes
in blue (G-M), and transporter genes in purple (N-P). AD- adaxial epidermis tissue; AB-
abaxial epidermis tissue; L- whole leaf tissue; S- stem epidermis tissue; CS- complete stem;
W- watered plants; D- drought treated plants.
Cuticular lipids mutants display diverse surface and morphological phenotypes
Of the 16 genes targeted, nine exhibited visual surface related phenotypes at the T0 generation
(Fig. 3 and Fig. S1). These included glossy leaves and stems and fused anthers in the kcs6
mutant alleles, glossy leaves and stems in the far and cer1 mutants, glossy leaves alongside
waxy stems in cer3 mutants (Fig. 3). The abcg11 mutant, the only line that never flowered
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
8
Figure 3. Plant surface phenotypes of mutant plants analyzed in this study. (A)
Representative stems of wild type (WT) and the five independent mutant lines investigated in-
depth. (B) Representative leaves of WT and mutants’ leaves. Bar=5cm. (C-H) representative
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
9
whole plants Bar=15cm. C, WT; D, far; E, cer3; F, kcs6; G, cer1; H, abcg32. (I) images of WT
leaves and those of mutants alongside.
(even over two years) possessed small, glossy, crinkled, brittle leaves. Both cyp86a22 and
abcg32 showed elongated and malformed leaves, while gpat4 displayed elongated leaves and
leaf fusions. Finally, the lacs mutants displayed both a glossy phenotype and leaf deformities
(Fig. S1). Sequencing the amplicons covering the targeted editing sites revealed a wide range
of mutations; from single base pair insertions or deletions, deletion of a triplet coding for a
single amino acid to a 200 bp deletion (Fig S2-S10). First generation plants (i.e. T0) had both
homozygous and heterozygous mutations, and frequently carried two different mutations on
their two chromosomes, and not merely two identical mutations.
Surface wax composition and crystal morphology in wax genes mutants
Out of the nine mutants, we focused the study on four wax metabolism genes, as well as the
abcg32 mutant that served as a control as it is affected in cutin monomer transport. We next
extracted leaves’ epicuticular wax and analyzed its composition using gas chromatography -
mass spectrometry (GC-MS). Wax composition in N. Glauca is typically dominated by a C31
alkane (92% according to Mortimer et al., 2012) along with C33 alkanes, C24, C26 and C28 fatty
alcohols and smaller amounts of a C26 aldehydes (as well as trace amounts of other alkanes,
alcohols and aldehydes). Mutations in the different genes had a drastic effect on epicuticular
wax composition (Fig. 4). kcs6 mutant leaves displayed almost complete reduction in all wax
components with a chain length above 26 carbons making them nearly free of alkanes (Fig.
4A-B). The same leaves accumulated shorter chain length waxes such as C18, C20, C22 and C24
alcohols (Fig. 4C-F). Leaves of the cer1 mutants showed massive reduction in abundance of
all alkanes (Fig 4A-B) mirrored by a significant increase in C26 alcohols (Fig. 4E). cer3 mutants
had a significant reduction (approximately 80%) in their C31 wax load. In the same mutant
leaves, the C33 alkane showed a trend of increase that was significant in one independent line
(Fig. 4A-B). The far mutants displayed extremely reduced fatty alcohol content (Fig. 4C-F)
contrasted by a trend of increase in their alkane load (Fig 4A-B). abcg32 mutants’ wax
composition was similar to WT, with only a trend of reduction in C31 alkane load (Fig. 4A) and
a significant increase in C26 alcohols in abcg32-18 and an increase in C28 alcohols in both
abcg32-6 and abcg32-18 (Fig. 4E-F).
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
10
Figure 4. Relative abundance of major wax components in N. glauca leaves,. (A-B) Alkanes
and (C-F) primary alcohols. Numbers indicate carbon chain length. Independent mutant alleles
of the same gene are indicated by the same color. WT, n=6; all other lines n=3. *= p<0.05 and
**= p<0.01 as determined in a student’s t test. Bars represent standard error. Wax components
were extracted in chloroform, derivatized and quantified by GC-MS
Alterations in composition of epicuticular waxes had a strong effect on wax crystal
morphology as visualized using cryo-SEM (Fig. 5). The typical wax crystals of N. Glauca
leaves are of dense rodlets (Fig. 5A). In cer1 leaves, a thin layer of flakes replaced wax rodlets
normally found in WT leaves (Fig. 5B). Leaves of cer3 displayed crystal morphology similar
to that of the WT leaves, however, these crystals were reduced in size and more sparsely
distributed on the leaf surface (Fig. 5C). Leaf wax of kcs6 lost its rodlet-like morphology and
instead appeared as structured lines of vertically positioned membranous platelets (Fig. 5D).
The far leaves exhibited large vertical sharp-edged plates, which were sparsely distributed on
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
11
the leaf (Fig. 5E). Finally, the abcg32 mutant wax crystals appeared normal as expected from
its almost unaltered wax composition (Fig. 5F).
Figure 5. Cryo SEM images of wax crystal dispersion and structure on the adaxial side of
mutant plants. (A) WT leaf; (B) cer1 leaf; (C) cer3 leaf; (D) kcs6 leaf; (E) far leaf; (F) abcg32
leaf. For each line, a magnification of x2500 with a 20µM bar is shown on the left and one
magnified to 25,000 (2µM bar) on the right.
Cutin composition of wax metabolism mutants
Wax metabolism and cutin metabolism share some common enzymes and precursors. We
therefore examined cutin composition in the same lines analyzed for epicuticular wax (Fig. 6).
Here the impact of mutations was far less drastic as compared to their effect on wax
composition – not coming close to a reduction percentage of the stronger wax mutants. This
was not the case for the cer1 lines showing significantly altered load of several cutin
monomers, and reaching a reduction of ~64% in methyl caffeate abundance (Fig. 6A-H).
Furthermore, when we analyzed the data per chemical class we found that cer1 mutants had
significantly lower fatty acid load in two out of three lines (Fig. 6B), but a significantly higher
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
12
levels of -hydroxylated fatty acids (Fig. 6C). Mirroring their wax composition, all three cer1
lines had significantly higher content of very long chain fatty alcohols. Interestingly, cer1 lines
also had reduced total phenol content in their cutin (Fig. 6E).
Figure 6. Relative abundance of cutin and its monomers in mutant lines,. (A-E) Analysis of
main cutin monomer classes. (F-H) Analysis of most abundant cutin monomers. WT, n=6; all
other lines n=3. *= p<0.05 and **= p<0.01 as determined in a student’s t test. Bars represent
standard error. All samples were quantified using GS-MS following delipidation,
transesterification and derivatization.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
13
Reduction in alkane abundance increases cuticular water loss drastically
We next examined how changes in chemical composition affected the physiological state of
mutant plants by analyzing cuticular water loss. Leaves were detached from plants and weighed
every two hours once reaching the linear weight loss stage. Out of the five mutant genotypes,
the cer1 mutants were most strongly affected, showing a three times higher water loss rate (Fig.
7A). The kcs6 mutant leaves also showed a higher water loss rate whereas abcg32 leaves
displayed a slight and yet significant increase. Surprisingly, of the cer3 mutant lines only cer3-
1 had a significantly elevated water loss rate. In contrast, two of three far lines lost water
significantly slower as compared to WT leaves.
Figure 7. Cuticular water loss rate and its correlation to different wax and cutin components.
(A) Cuticular water loss rate. The experiment included three assays in consecutive days each
with its own WT samples (WT1 to WT3). Mutant alleles of the same gene are indicated in the
same color. * = p<0.05; ** = p<0.01 determined in a student’s t test; n=5. Bars represent
standard error. Correlation between water loss rate and (B) sum of all alkane wax components,
(C) sum of very long chain fatty alcohols, (D) sum of all lipidic components of the cutin
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
14
monomers; (E) sum of -hydroxylated fatty acids of the cutin monomers; (F) sum of all
phenols of the cutin monomers. R2 indicated in each correlation graph is generated from
average water loss rate and average abundance of the different wax and cutin components in
in the same independent mutant line, obtained in separate experiments.
Next, we plotted water loss rate against the average abundance of wax and cutin components
in independent mutant lines (Fig 7B-F). We found a correlation with an R2 of 0.786 between
water loss rate and total alkane content. Yet, cer3 leaves lost water at a slower rate than their
alkane content would predict while cer1 plants lost water at a faster rate than that predicted by
the linear regression (Fig. 7B). When cutin components were plotted against water loss, a
positive correlation (R2 = 0.718) was found with total cutin lipids, although the differences in
the cutin component abundance was far smaller between the different lines compared to those
in wax components (Fig. 7D). In contrast, phenols found in the cutin fraction (R2 = 0.616; Fig.
7F), and especially methyl caffeate (R2 = 0.7) were negatively correlated with water loss.
Although the role of the phenolics is not clear to us, these results highlight that alkanes in the
epicuticular wax fraction play an important role in preventing cuticular water loss.
Wax has little effect on water loss pre-drought, but alkanes are essential for recovery
from drought
The major effect on cuticular water loss rates due to changes in epicuticular waxes led us to
hypothesize that under well-watered conditions plants with greatly reduced alkane abundance
would transpire at higher rates and be more susceptible to drought. We tested this hypothesis
using a weighing lysimeter system, in which plants are weighed every three minutes and in
combination with environmental measurements (e.g. temperature, humidity and light intensity)
performed in parallel, transpiration, stomatal conductance (gs), biomass accumulation, water
use efficiency (WUE) and drought response can be extrapolated. The first two experiments
performed with this system in the summer and winter seasons, involved two and three different
mutant alleles of cer3, respectively. To our surprise, in the first experiment (during the
summer), cer3 plants had an almost identical daily transpiration compared to WT ones (Fig
S11A). In the ‘winter’ experiment, we observed a trend of cer3 transpiration starting to be
reduced compared to WT as plants grew and their transpiration rate increased, although this
trend did not reach significance prior to the onset of drought (Fig. S12A). Furthermore, cer3
biomass accumulation was not impaired in both experiments (Fig. S11g, 12G), as was their
WUE (Fig. S11J, S12I). Plants that respond to drying soil by closing their stomata at a higher
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
15
soil water content (“theta point”) are more likely to show drought tolerance. Drought response
in these two experiments was therefore measured by evaluating this theta point. In the
‘summer’ experiment, cer3 lines had a theta point similar to WT (Fig. S11K), whereas in the
‘winter’ experiment, their theta point was at a significantly higher soil water content (Fig. 12J).
The results indicated that cer3 plants (displaying approximately 80% reduction in C31 alkane)
were not negatively affected by their reduced wax load both under well-watered and drought
conditions.
The findings described above prompted us to examine whether plants mutated in additional
genes that severely affect wax composition would behave in a similar manner. To answer this
question, we performed an additional experiment during the spring. Following the previous
experiments we reasoned that during this season we could enjoy the advantages of both the
‘summer’ experiment in which plants grow fast rate and the vapor pressure defecate (VPD)
and light intensity are relatively stable and the ‘winter’ experiment in which plants are less
susceptible to diseases. In this ‘spring’ experiment, we placed three different mutant alleles of
either cer1, kcs6 or far on the lysimetric system alongside three lines of WT plants. Similar to
the cer3 results under well-watered conditions there was no significant difference between the
mutants and WT plants in daily transpiration (Fig. 8A), transpiration rate (Fig. 8D) plant
growth rate (Fig. 8G), and water use efficiency (Fig. 8J).
Once we exposed well-watered plants to drought and recovery (i.e. re-watering), we
detected markedly different response in the wax mutants. The cer1 mutant lines theta point was
significantly lower than WT, while kcs6 and far were similar to WT (Fig. 8K). This led to cer1
plants having a significantly higher transpiration rate throughout long periods of the day, that
was not seen in the other mutant lines (Fig. 8E). Though the uncrease in cer1’s transpiration
rate is significant, the effect of size is relatively small, and reached a ~20% increase in
transpiration rate during the afternoon hours. This was not the case following resuscitation. At
this phase, the daily transpiration of far plants was significantly higher as compared to WT
plants that exhibited significantly greater transpiration than cer1 and kcs6 (Fig. 8C,F). These
differences rose with time up to the sixth day following resuming irrigation; far plants
transpired 533ml a day on average, WT 262ml, kcs6 50ml and cer1 transpired only 36ml on
average during that day (Fig. 8C). These results were mirrored by the transpiration rate, reac
(Fig. 8F), plant growth rate (Fig. 8H) and shoot dry biomass (when the experiment ended) (Fig.
8I). The underlying reason for these drastic differences was evident when we examined the
plants once completing the experiments (Fig. S13). While all far plants retained their original
leaves, some WT plants were left with leaves and others not. In a vast majority of the cer1 and
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
16
kcs6 plants leaves present at the time of drought initiation dried and the transpiration was from
new leaves, which started growing from axillary buds upon resuscitation (Fig. S13).
Figure 8. Multi-parameter analysis of mutant plants assayed in a lysimeteric system and
assayed following full irrigation, drought and recovery. (A) Pre-drought daily transpiration.
(B) Daily transpiration following drought initiation and prior to resuscitation. (C) Daily
transpiration following recovery. Since each individual plant received full irrigation three days
after it transpired below the 20% daily transpiration threshold (see materials and methods), the
days in recovery unlike those pre-drought and during drought are different for each individual
plant, and are averaged according to the number of days past irrigation renewal per plant. (D)
Transpiration rate during day 12 prior to drought induction. Measurements were taken every
3min, though error bars are only shown every 30min. (E) Transpiration rate during the fifth
day of drought. (F) Transpiration rate during the fifth day after plant resuscitation, averaged
from different dates for each plant according to its date of recovery. (G) Plant biomass during
the period prior to drought induction. (H) Plant biomass during the four days before the
experiment ended, at which time all plants had been resuscitated. (I) Dry shoot biomass at the
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
17
experiments end. (J) WUE as calculated pre-drought. (K) “Theta point” volumetric water
content at which plants began reducing transpiration rate in response to drying soil. All data is
the average of three independent mutant alleles used in this experiment, including WT which
had three independent lines grown first in tissue culture prior to the transfer to the greenhouse
as performed with all mutant lines. The following amount of lines were analyzed: WT n=28 to
33; cer1 n=27 to 31; kcs6 n=25 to 29; far n=24 to 26. Different letters indicate significance of
p<0.05 as determined in a Tukey HSD test. Asterisks represent significance of p<0.05 as
determined in a student’s t test. Asterisks were placed above standard error lines (bars), though
points are significant for every point measured between the bars.
Drought followed by resuscitation leads to stem cracking
Towards the completion of the weighing lysimeters experiment, we detected an intriguing
phenotype, which to the best of our knowledge not been reported in the context of wax
deficiency. Sixteen-out of 27 kcs6 lines derived from all three independent mutant alleles
displayed severe cracks along their stems penetrating the xylem and deep into the stem pith
(Fig. 9A-B). Intriguingly, these cracks even led to water dripping freely out of the xylem and
out of the cracks (Fig. 9A). Besides the kcs6 plants, there was one occurrence of stem cracking
in a cer1-1 plant indicating that the occurrence in kcs6 plants was not an isolated event. When
investigating the reasons for the kcs6 stem phenotype we found that the 11 kcs6 plants with
intact stems whose non-cracking phenotype was not due to technical reasons had a similar
biomass and theta point compared to the plants with cracked stems at the onset of drought (Fig.
9E; Fig. 9F). While all 16 cracked plants resuscitated during three days, from the 13th day of
drought to the 15th day, the uncracked plants were resuscitated between the 10th to the the 18th
day in drought. When examining the 11 kcs6 plants with undamaged stems we found that they
reduced their volumetric water content (VWC) at a slower rate than those with cracked stems,
though this difference was not significant on a point measurement basis (Fig. 9G). Thus, as is
seen from the wider range of drought duration prior to recovery, plants that depleted their water
and resuscitated rapidly as well as those that transpired less and were exposed to drought more
gradually did not experience stem cracking.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
18
Figure 9. The stem cracking phenomenon. (A) Images of cracked stems phenotypes in
representative plants from the three kcs6 independent mutant alleles, a cer1 plant displaying a
similar phenotype and a representative WT stem. (B) Images of cut stems showing the radial
formation of the crack through the xylem and into the stem pith cells. (C) Images of whole
plants at the experiments end. (D) Whole plants photographed from above and displaying
differential drought response. (E) Volumetric water content of kcs6 plants in which no stem
cracking was observed compared to those in which stems cracked. Plants included in this
analysis were of similar size at drought onset. (F) Biomass of plants with or without cracked
stems (kcs6) measured during the last irrigated day prior to drought. (G) Average “theta points”
of cracked and uncracked kcs6 plants. cracked n=16; uncracked n=11.
Stomatal aperture and development do not compensate for elevated cuticular water loss
We examined whether the discrepancy between high leaves cuticular water loss and normal
transpiration rate under well-watered conditions was related to a reduced stomatal density, size
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
19
or aperture, leading to the elevated cuticular water loss being compensated for by reduced
stomatal water loss. However, all mutants had similar density (Fig. S14A,D) and sizes (Fig.
S14B,E) compared to WT, and even those insignificant differences in density were
compensated for by larger stomata; i.e. plants with a trend of less dense stomata had a trend of
larger stomata as well. Abaxial stomatal aperture (Fig. S14C) was the only parameter where
we found a significant difference, with cer1 having a larger stomatal aperture. Thus, unlike our
original hypothesis, cuticular water loss was not compensated for by stomatal aperture or
morphology.
The glossy phenotype of wax mutants does not significantly affect light response and plant
fitness
Despite extreme differences in their wax composition, cer1, cer3, kcs6 and far mutants, all
displayed a glossy leaf appearance. To examine if this phenotype affects photosynthetic
efficiency, we examined RUBISCO efficiency (Vcmax) and maximal electron transport rate
(Jmax) by monitoring carbon assimilation as sub-stomatal CO2 concentration was raised (A/Ci
curves). The cer1 mutant lines appeared different from all other mutants and the WT as their
A/Ci curve plateaued much earlier than all other genotypes (Fig. 10A). Although the Jmax of
cer1 plants was not significantly lower (Fig. 10C), their Vcmax was significantly reduced (Fig.
10B). cer1’s wax composition is similar to kcs6, a fact that leads to very similar responses in
Figure 10. Photosynthetic efficiency of WT and
four wax biosynthesis mutant lines as analyzed
by A/Ci curves (A) Average carbon assimilation
with rising Ci concentrations. (B) Vcmax
indicating RUBISCO efficiency. (C) Jmax
indicating maximal electron transport rate.
Asterisks indicate a significance of p<0.05 as
determined in a student’s t test. N=5.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
20
assays where wax composition is the underlying factor. However, cer1 is the only mutant with
altered cutin, leading us to suggest that the differences seen in the A/Ci curves were the result
of altered cutin and not wax load or composition.
We next asked if the glossy phenotypes of wax mutants had a significant impact on light
response and consequently plant fitness. Thus, we exposed plants of each genotype to 12 rising
light intensities and recorded carbon assimilation and stomatal conductance (Fig. 11A and Fig.
11B). No plants reached photoinhibition along these points despite reaching 2200 µE (while
plants were grown at a light intensity of 100µE – 200 µE). Similar to the A/Ci curves results,
cer1 mutants were the only plants displaying an altered phenotype exhibiting reduced carbon
assimilation at five points of the dozen light intensities measured (Fig. 11A). These results
demonstrate that light reflectance by wax crystals does not alter photosynthetic capacity under
non-stressed conditions.
Figure 11. Effects of wax deficiency on gas exchange in response to rising light intensity (A)
Carbon assimilation. )B ( :Stomatal conductance. Asterisks indicate a significance of p<0.05 as
determined in a student’s t test. N=9.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
21
Discussion
Tree tobacco in cuticular lipids research
For many years, cuticular lipids research has been largely performed in Arabidopsis thaliana
(Lee and Suh, 2015). The ease of screening large mutant collections enabled identification of
the majority of eceriferum mutants and their characterization. Advancements in targeted
mutagenesis, predominantly CRISPR-Cas technology, opened the way for exploiting other
species such as tree tobacco in cuticular lipids research. Apart from outstanding advantages
including an extremely high wax load, simplicity of wax composition and induced wax
accumulation following drought, tree tobacco is readily transformed, can grow in a variety of
conditions including natural settings, and recently had its genome sequenced (Usadel et al.,
2018). Furthermore, being a perennial plant, tree tobacco can be explored at both the juvenility
stage as well as upon maturation and bark formation. Hence, several key results reported here
could not have been discovered in Arabidopsis, such as stem cracking and drought experiments
in which plants accumulated tens of grams of biomass and transpired hundreds of milliliters
each day.
Alkanes reduce cuticular water loss, but cutin affects photosynthetic sufficiency and light
response
In this study, we simultaneously characterized a large number of mutants corresponding to
different cuticular lipids genes. By doing so, we filtered out indirect effects not related to
epicuticular wax composition such as changes in cutin composition. Both cer1 and the kcs6
mutants possess an extremely reduced alkane load, yet, cer1 also retains altered cutin
composition, with significantly less phenols and more -hydroxylated fatty acids. Comparing
the phenotype of cer1 to that of kcs6 enabled us to make the distinction. Cuticular water loss
is affected when both genes are mutated, although the rate in cer1 is higher. Here we detected
an additive effect of altered cutin despite the major effect being that of alkane abundance.
Effects that are only present in cer1 and not kcs6 and which we attribute to cutin rather than
epicuticular wax are the reduced Vcmax and lower carbon assimilation with rising light
intensity. In contrast to these, drought response in cer1 and kcs6 plants was nearly identical,
suggesting that alkane abundance is a major factor essential for recovery following drought.
It was previously shown that overall accumulation of epicuticular waxes following exposure
to drought in tree tobacco reduced the rate of cuticular water loss (Cameron et al., 2006).
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
22
However, Cameron et al. (2006) could not determine which of the many wax constituents
underlays this reduction in water loss rate. Here we point to alkanes being the component
preventing cuticular water loss. While the correlation of alkanes with water loss rate is high, it
does not seem to be linear. The cer3 mutants possess ~20% of the WT C31 alkane and had only
a slight increase in water loss rate. This discrepancy could be observed when removing the
cer3 lines from the regression, which raises the R2 from 0.79 to 0.95 value. It therefore seems
that although alkanes reduce cuticular water loss and are essential for recovery following
drought, the threshold amount of these components required for having an effect may be well
under the one present in WT plants.
Alkane abundance is essential for recovery following drought but does not contribute to
fitness under well-watered conditions
Following the results of the cuticular water loss assay performed on detached leaves, we
anticipated that assaying whole plants would mirror this at least partially, and that the mutant
plants would transpire at a higher rate under well-watered conditions. This is in line with many
studies reporting epicuticular wax conferring drought tolerance, achieved in many cases due to
reduced transpiration (reviewed in Xue et al., 2017). Our conclusions from three independent
whole plant growth and drought experiments do not match the common view point that wax
simply confers drought tolerance by reducing transpiration rates. Plants from our mutant
collection, regardless of the metabolic impact of the mutation (whether having no fatty
alcohols, no alkanes or reduced alkane load) all transpire and accumulate biomass at a rate
similar to WT under well-watered conditions. Furthermore, although cer1 responded to
drought by reducing stomatal conductance at a lower soil water content (a phenotype associated
with reduced drought tolerance; Negin and Moshelion, 2017), such a response was not detected
in kcs6 plants. However, this was not the case with the mutants’ recovery dynamics, in which
extreme differences could be seen between plants, clustering according to their wax
composition. We found that the underlying phenotype that determined whether plants would
recover transpiration and growth rapidly following irrigation resumption was leaf survival. We
observed clearly that plants that did not recover, were lines in which the drought caused leaves
to dry out and die (Fig. S13). Far mutant plants in contrast to alkane reduced ones, recovered
at the fastest rate, and lost almost no leaves during the drought treatment.
In our drought assay, soil water content and transpiration were continually monitored and
plants were resuscitated once they had reached the same transpirational threshold. In contrast,
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
23
many studies assay drought by stopping irrigation for a similar duration of time (see for
example Bourdenx et al., 2011). This experimental design causes plants that have higher
transpiration under well-watered conditions to reach drought conditions at an earlier time, and
be exposed to drought conditions for a longer period than their counterparts. Furthermore,
“drought tolerance” attributed to altered drought response may in fact be secondary to
attenuation of other stress conditions. These stress conditions may include high light intensity,
temperature and VPD. For example, plants that are able to disperse light rather than absorb it
may have a lower leaf temperature , which may be compensated for by increasing transpiration
(Richards et al., 1986), making these plants less tolerant to drought induced by similar duration
of irrigation stoppage. Our experimental setup uncoupled well-watered transpiration and
growth from drought response and recovery. This enabled us to point to the ability to strongly
reduce transpiration under drought and prevent leaf death as alkane dependent and essential for
recovery.
Changes in wax composition and the stem cracking phenomena
The phenomena of stem cracking is known in conifers at the end of growing seasons of drought
years, and increases with drought severity (Zeltińš et al., 2018; Cameron, 2019). However, to
the best of our knowledge it has never been linked to alterations in epicuticular wax abundance
or composition. Its appearance in kcs6 and cer1 mutants following recovery indicates that stem
cracking is indeed alkane related. The extensive use of annual plants, mainly Arabidopsis for
molecular research of epicuticular waxes, has left its effect on perennial and woody plants
relatively unexplored. Apart from presenting an intriguing phenotype linked to deficiencies in
epicuticular wax biosynthesis, our study here underscores the use of N. Glauca as a model
system enabling research of wax function in woody species. The exact stage at which stem
cracking took place is not clear to us. Even at the end of drought treatment, we could not detect
cracked stems in the kcs6 plants. However, whether small initial cracking already began during
drought, if a more gradual recovery would have prevented the cracking, and what is the
threshold of drought conditions resulting in cracking, remains to be determined.
Glossy mutants did not share a common phenotype under the examined conditions
Out of the five mutants examined in this study, four display a glossy phenotype. This striking
phenotype appears in plants that lost alkane biosynthesis nearly entirely but also when the C31
alkane is greatly reduced in cer3 mutants, and even in the far mutants that are deficient merely
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
24
in primary alcohols. The far mutants do not exhibit an elevated rate of cuticular water loss.
More strikingly, their recovery following drought is greatly improved compared to WT
indicating that fatty alcohols are completely unnecessary for drought response. This leads to
the question of what is the disadvantage in losing primary alcohols specifically and exhibiting
a glossy appearance? Since the glossy phenotype is dependent on optical parameters, we
examined photosynthetic efficiency as well as light response. In both cases, there was no
correlation between leaf glossiness and these parameters. In eucalyptus species, it was shown
that glaucous leaves had the effect of reducing photosynthesis prior to saturating conditions. In
addition, in these plants, once wax was removed, photoinhibition occurred at high light
intensity while this did not occur in normal plants (Cameron, 1970). Although our initial
hypothesis was similar to the findings in Cameron et al. (1970), both in terms of reaching
photoinhibition at a lower light intensity as well as having higher carbon assimilation at a non
saturating light intensity, this was not the case in the four mutants examined in this study. The
cer1 mutant line even assimilated carbon at lower rates with rising light intensity.
Epicuticular wax has been suggested to play a protective role against a range of stresses.
Drought (Aharoni et al., 2004; Seo et al., 2011; Lee et al., 2014; Lee and Suh, 2015; Xue et al.,
2017), UV radiation (Long et al., 2003), osmotic stress (Liu et al., 2019) and insect herbivory
(reviewed in Eigenbrode and Espelie, 1995) are a few such conditions. In addition, since
epicuticular wax repels water it has been suggested to have an anti-adhesive self-cleaning affect
(also known as the “lotus effect”; Neinhuis and Barthlott, 1997), which causes different
particles including pathogens to be washed away upon leaf wetting. However, under our
experimental conditions, we did not find a common disadvantage in coping with abiotic stress
conditions which is linked to a glossy appearance. Hence, we suggest that glaucous appearance
is likely to bear positive effect in combating biotic stress conditions rather than under abiotic
ones.
Conclusions
The examination of epicuticular wax mutants in N. glauca revealed that while alkane
accumulation is strongly correlated with cuticular water loss, under well-watered conditions
epicuticular wax composition does not affect whole plant water loss or plant growth rate.
Despite this, the presence of alkanes greatly affected the ability of plants to recover following
drought conditions, and its deficiency led to a previously unobserved phenotype of stem
cracking. Furthermore, under our experimental conditions, we could not find a common
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
25
denominator between plants possessing a glossy phenotype suggesting that glaucous
appearance is associated with combating biotic stress conditions. In contrast to this, plants in
which cutin was strongly affected were the only ones whose photosynthetic efficiency and light
response were altered, emphasizing the importance of uncoupling cutin related affects from
those which are epicuticular wax derived.
Materials and methods
Plant material and growth conditions
Seeds of N. glauca were initially collected from plants growing nearby the Weizmann institute
campus in Rehovot, Israel (31.912408, 34.820184). Plants used for transcriptome analysis were
grown in the greenhouse (winter 2013). Plants used in cuticular wax and cutin extraction, SEM,
stomatal analysis, cuticular water loss and photosynthetic efficiency assays were grown in a
greenhouse in two batches during the winter-spring of 2019 and of 2020. Plants used for light
curves were grown in a growth room under a flux of ~150µE, with a temp. of 22˚C and light/
dark period of 16/8 h. Drought experiments were performed in the lysimetric facility in the
Faculty of Agriculture in Rehovot, Israel (31.904134, 34.801060), during 7-9.2018, 1-3.2019
and 3-4.2020.
CRISPR vectors and mutation analysis
Construct assembly was performed using the ‘golden braid’ cloning system (Sarrion-
Perdigones et al., 2013). A codon optimized Streptococcus pyogenes Cas9 (Fauser et al., 2014)
was used, driven by the Solanum lycopersicum UBIQUITIN10 promoter (Dahan-Meir et al.,
2018). All gRNA transcription was driven by the Arabidopsis U6-26 promoter. crRNAs were
planned based on the genome assembly of Usadel et al. (2018) using the Crispr-p software (Lei
et al., 2014). crRNA sequences, gene maps and transformation protocol appear in
supplementary methods and Fig. S2-S10. Mutations were analyzed by DNA extraction and
PCR amplification using oligonucleotides flanking the expected edited region. In plants where
the crRNAs were positioned far from each other on the genome, two PCRs with different
primer sets were performed and PCR products sequenced. For homozygous mutations, NCBI
blast (https://blast.ncbi.nlm.nih.gov/Blast.cgi) was used to compare the mutant’s sequence to
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
26
WT sequence and in heterozygous cases, they were characterized using one of two softwares:
DSDecode (Liu et al., 2015) or CRISP-ID (Dehairs et al., 2016).
Transcriptomics and differential gene analysis
Transcriptome analysis was performed on plants from several tissues under well-watered and
drought conditions (induced by three events of drying and recovery as described in Cameron
et al. (2006). RNA extraction, library preparation and RNA-seq was performed as described in
Hen-Avivi et al. (2016). Since at the time of analysis of the raw reads, the N. glauca genome
was unpublished, we performed de-novo transcriptome assembly based on Hass et al. (2013)
of N. glauca gene expression data published by Long et al. (2016). Based on this assembly,
genes were aligned and their differential expression analyzed. This data was then used to define
conditions of epidermis enrichment and drought induction, and contigs in which at least three
conditions (adaxial epidermis enrichment, abaxial epidermis enrichment and adaxial epidermis
drought induction for example) were searched using BlastX against the NCBI protein sequence
database (NR) to find their closest homologs.
Wax and cutin monomer profiling
For wax extraction, three leaf discs with a diameter of 12mm each were dipped in 4ml
chloroform with an internal standard of 10µg C36 alkane and shaken gently for 15sec. The discs
were then removed and chloroform was evaporated under a nitrogen flow. Samples were then
resuspended in 100µl chloroform to which 20µL pyridine and 20µL BSTFA were added.
Samples were derivatized at 70˚C and injected in a splitless mode to a GC-MS system (Agilent
7890A chromatograph, 5975C mass spectrometer and a 7683 auto sampler) as described in
Cohen et al. (2019). Chromatograms and mass spectra were analyzed using MSD Chemstation
software (Agilent). Identification was based both on fragmentation alignment to the NIST Mass
Spectral Library and in-house retention Indexes. Quantification was performed using the
Chemstation software and was normalized to the C36 alkane internal standard. Cutin extraction
was performed as described in Cohen et al. (2019). Monomer identification and quantification
was performed as in wax components analysis, with normalization performed against a C32
alkane internal standard.
Leaf desiccation assays and whole plant drought trials
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
27
Five leaves from different plants from each independent line were cut in the greenhouse and
immediately inserted to zip-locked bags, brought to the lab and let dry at 22˚C. Leaves were
weighed every two hours for 12 hours. This was done following earlier calibration in which
leaves were weighed every half hour and which showed that following two hours of desiccation
water loss was linear up to 24h. Due to the large number of leaves and need to repeatedly weigh
them at given intervals, the experiment was divided to three different days, each having its own
WT control. Drought trials performed using a lysimetric system (during the summer of 2018,
winter of 2019 and spring of 2020) were performed in either soil (2018, 2019), or sand (2020)
as described in Halperin et al. (2016) and Dalal et al. (2020) (see supplementary methods for
details).
Stomatal analysis and scanning electron microscopy
Stomatal imprints were taken from adaxial and abaxial epidermises of greenhouse grown
plants, as described in Yaaran et al. (2019). Nail polish imprints of dental resin were then
attached to microscope slides and photographed in a Nikon eclipse E800 microscope with a
Nikon Digital sight DS-5Mc camera. Each imprint was photographed in two different locations
at a magnification of x100 for stomatal density and in another two at a magnification of x400
for stomatal aperture and size. Stomata were then quantified and aperture and size was
measured using ImageJ software (https://imagej.nih.gov/ij/). For SEM, small sections were cut
from fresh leaves of WT and representative lines from the five mutant genes and inserted to a
cryo-holder, frozen in liquid nitrogen, coated and photographed using a Zeiss Ultra 55 SEM as
described in Hen-Avivi et al. (2016).
Photosynthetic efficiency and light response
Photosynthetic efficiency was assessed by measuring gas exchange at rising CO2
concentrations (“A/Ci curves”), using a Li-Cor 6800 portable photosynthesis system (LI-COR,
Inc.; Lincoln, NE, USA). Parameters were set to: PAR 1,600, relative humidity 60%, CO2
400PPM and leaves were inserted to the infrared gas analyzer (IRGA) chamber for acclimation.
This stage was performed for at least 15min and was stopped when carbon assimilation ceased
rising exponentially (but no longer than 25min). A/Ci curves were then performed with the
points: 400PPM, 300PPM, 150PPM, 50PPM, 0, 400PPM, 600PPM, 800PPM, 1,000PPM,
1,200PPM and 1,500PPM CO2. Data was analyzed using R (www.r-project.org) and the
‘plantecophys’ package (Duursma, 2015). Light response was measured using the Li-Cor 6800
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
28
portable photosynthesis system. Parameters were similar to those used for A/Ci curves,
excluding light intensity which was raised every 6 min. Plants were first dark acclimated by
covering the leaf on which the measurement would be performed in aluminum foil for 10min,
after which the leaf was inserted to the IRGA chamber. Each light intensity was kept for at
least 6min, and once A and gs were stable a measurement was taken. The light increment
program was as follows: 0, 10µE, 20µE, 50 µE, 100 µE, 200 µE, 400 µE, 800 µE, 1,200 µE,
1,600 µE, 2,000 µE and 2,200 µE.
Statistical analysis
The JMP 14 software (SAS Institute; http://www.jmp.com/en_us/home.html) was used for all
statistical analyses, except for two piece linear curves in the drought experiments in which in-
house statistical tools of the lysimeter system (https://www.plant-ditech.com/) were used to
find the best fitting regression lines. Student’s t test was used when comparing two groups.
Mutants were always compared to WT, except in the spring drought trial where Tukey HSD
test was used and all groups were compared.
Acknowledgments
We thank Prof. Björn Usadel for providing us with access to unpublished N. glauca genomic
and transcriptomic data. These high quality assemblies aided us to a great extent throughout
this study.
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
29
References
Aarts MG, Keijzer CJ, Stiekema WJ, Pereira A (1995) Molecular characterization of the
CER1 gene of Arabidopsis involved in epicuticular wax biosynthesis and pollen fertility.
Plant Cell. 7: 2115-2127
Aharoni A, Dixit S, Jetter R, Thoenes E, van Arkel G, Pereira A (2004) The SHINE clade
of AP2 domain transcription factors activates wax biosynthesis, alters cuticle properties,
and confers drought tolerance when overexpressed in Arabidopsis. Plant Cell 16: 2463–
2480
Bach L, Michaelson L V., Haslam R, Bellec Y, Gissot L, Marion J, Da Costa M, Boutin
JP, Miquel M, Tellier F, et al (2008) The very-long-chain hydroxy fatty acyl-CoA
dehydratase PASTICCINO2 is essential and limiting for plant development. Proc Natl
Acad Sci U S A 105: 14727–14731
Barthlott W, Neinhuis C, Cutler D, Ditsch F, Meusel I, Theisen I, Wilhelmi H (1998)
Classification and terminology of plant epicuticular waxes. Bot J Linn Soc. 126: 237-
260
Beaudoin F, Wu X, Li F, Haslam RP, Markham JE, Zheng H, Napier JA, Kunst L
(2009) Functional characterization of the Arabidopsis β-ketoacyl-coenzyme a reductase
candidates of the fatty acid elongase. Plant Physiol 150: 1174–1191
Bickford CP (2016) Ecophysiology of leaf trichomes. Funct Plant Biol 43: 807–814
Bourdenx B, Bernard A, Domergue F, Pascal S, Leger A, Roby D, Pervent M, Vile D,
Haslam RP, Napier JA, et al (2011) Overexpression of Arabidopsis ECERIFERUM1
promotes wax very-long-chain alkane biosynthesis and influences plant response to
biotic and abiotic stresses. Plant Physiol 156: 29–45
Cameron AD (2019) Mitigating the risk of drought-induced stem cracks in conifers in a
changing climate. Scand J For Res 34: 667–672
Cameron KD, Teece MA, Smart LB (2006) Increased accumulation of cuticular wax and
expression of lipid transfer protein in response to periodic drying events in leaves of tree
tobacco. Plant Physiol 140: 176–183
Cameron RJ (1970) Light intensity and the growth of Eucalyptus seedlings. II.* The effect
of cuticular waxes on light absorption in leaves of Eucalyptus species. Aust J Bot 18:
275–284
Chalker-Scott L (1999) Environmental significance of anthocyanins in plant stress
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
30
responses. Photochem Photobiol 70: 1–9
Chen X, Goodwin SM, Boroff VL, Liu X, Jenks MA (2003) Cloning and characterization
of the WAX2 gene of Arabidopsis involved in cuticle membrane and wax production.
Plant Cell 15: 1170–1185
Cohen H, Dong Y, Szymanski J, Lashbrooke J, Meir S, Almekias-Siegl E, Zeisler-Diehl
VV, Schreiber L, Aharoni A (2019) A multilevel study of melon fruit reticulation
provides insight into skin ligno-suberization hallmarks. Plant Physiol 179: 1486–1501
Dahan-Meir T, Filler-Hayut S, Melamed-Bessudo C, Bocobza S, Czosnek H, Aharoni A,
Levy AA (2018) Efficient in planta gene targeting in tomato using geminiviral replicons
and the CRISPR/Cas9 system. Plant J 95: 5–16
Dalal A, Shenhar I, Bourstein R, Mayo A, Grunwald Y, Averbuch N, Attia Z, Wallach
R, Moshelion M (2020) A High-Throughput Gravimetric Phenotyping Platform for
Real-Time Physiological Screening of Plant–Environment Dynamic Responses.
bioRxiv. doi: 10.1101/2020.01.30.927517
Dehairs J, Talebi A, Cherifi Y, Swinnen J V. (2016) CRISP-ID: Decoding CRISPR
mediated indels by Sanger sequencing. Sci Rep 6: 1–5
Duursma RA (2015) Plantecophys - An R package for analysing and modelling leaf gas
exchange data. PLoS One 10: e0143346
Eigenbrode SD, Espelie KE (1995) Effects of plant epicuticular lipids on insect herbivores.
Annu Rev Entomol 40: 171–194
Fauser F, Schiml S, Puchta H (2014) Both CRISPR/Cas-based nucleases and nickases can
be used efficiently for genome engineering in Arabidopsis thaliana. Plant J 79: 348–359
Greer S, Wen M, Bird D, Wu X, Samuels L, Kunst L, Jetter R (2007) The cytochrome
P450 enzyme CYP96A15 is the midchain alkane hydroxylase responsible for formation
of secondary alcohols and ketones in stem cuticular wax of Arabidopsis. Plant Physiol
145: 653–667
Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, Couger MB,
Eccles D, Li B, Lieber M, et al (2013) De novo transcript sequence reconstruction from
RNA-seq using the Trinity platform for reference generation and analysis. Nat Protoc 8:
1494–1512
Halperin O, Gebremedhin A, Wallach R, Moshelion M (2016) High‐throughput
physiological phenotyping and screening system for the characterization of plant–
environment interactions. Plant J. 89: 839-850
Hanley ME, Lamont BB, Fairbanks MM, Rafferty CM (2007) Plant structural traits and
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
31
their role in anti-herbivore defence. Perspect Plant Ecol Evol Syst 8: 157–178
Hen-Avivi S, Savin O, Racovita RC, Lee W-S, Adamski NM, Malitsky S, Almekias-Siegl
E, Levy M, Vautrin S, Bergès H, et al (2016) A Metabolic Gene Cluster in the Wheat
W1 and the Barley Cer-cqu Loci Determines b-Diketone Biosynthesis and
Glaucousness. Plant Cell 28: 1440-1460
Jetter R, Riederer M (2016) Localization of the transpiration barrier in the epi- and
intracuticular waxes of eight plant species: Water transport resistances are associated
with fatty acyl rather than alicyclic components. Plant Physiol 170: 921–934
Lee SB, Kim H, Kim RJ, Suh MC (2014) Overexpression of Arabidopsis MYB96 confers
drought resistance in Camelina sativa via cuticular wax accumulation. Plant Cell Rep
33: 1535–1546
Lee SB, Suh MC (2015) Advances in the understanding of cuticular waxes in Arabidopsis
thaliana and crop species. Plant Cell Rep 34: 557–572
Lei Y, Lu L, Liu HY, Li S, Xing F, Chen LL (2014) CRISPR-P: A web tool for synthetic
single-guide RNA design of CRISPR-system in plants. Mol Plant 7: 1494–1496
Li F, Wu X, Lam P, Bird D, Zheng H, Samuels L, Jetter R, Kunst L (2008) Identification
of the wax ester synthase/acyl-coenzyme a:diacylglycerol acyltransferase WSD1
required for stem wax ester biosynthesis in Arabidopsis. Plant Physiol 148: 97–107
Liu N, Chen J, Wang T, Li Q, Cui P, Jia C, Hong Y (2019) Overexpression of WAX
INDUCER1/SHINE1 gene enhances wax accumulation under osmotic stress and oil
synthesis in Brassica napus. Int J Mol Sci. 20: 4435
Liu W, Xie X, Ma X, Li J, Chen J, Liu YG (2015) DSDecode: A web-based tool for
decoding of sequencing chromatograms for genotyping of targeted mutations. Mol Plant
8: 1431–1433
Long LM, Prinal Patel H, Cory WC, Stapleton AE (2003) The maize epicuticular wax
layer provides UV protection. Funct Plant Biol 30: 75–81
Long N, Ren X, Xiang Z, Wan W, Dong Y (2016) Sequencing and characterization of leaf
transcriptomes of six diploid Nicotiana species. J Biol Res 23: 6
Lü S, Song T, Kosma DK, Parsons EP, Rowland O, Jenks MA (2009) Arabidopsis CER8
encodes LONG-CHAIN ACYL-COA SYNTHETASE 1 (LACS1) that has overlapping
functions with LACS2 in plant wax and cutin synthesis. Plant J 59: 553–564
Millar AA, Clemens S, Zachgo S, Giblin EM, Taylor DC, Kunst L, Michael Giblin E,
Taylor DC, Kunst L (1999) CUT1, an Arabidopsis gene required for cuticular wax
biosynthesis and pollen fertility, encodes a very-long-chain fatty acid condensing
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
32
enzyme. Plant Cell. 5: 825-838
Millar AA, Kunst L (1997) Very-long-chain fatty acid biosynthesis is controlled through the
expression and specificity of the condensing enzyme. Plant J 12: 121–131
Mortimer CL, Bramley PM, Fraser PD (2012) The identification and rapid extraction of
hydrocarbons from Nicotiana glauca: A potential advanced renewable biofuel source.
Phytochem Lett 5: 455–458
Negin B, Moshelion M (2017) The advantages of functional phenotyping in pre-field
screening for drought-tolerant crops. Funct Plant Biol 44: 107–118
Neinhuis C, Barthlott W (1997) Characterization and distribution of water-repellent, self-
cleaning plant surfaces. Ann Bot 79: 667–677
Richards R., Rawson H., Johnson D. (1986) Glaucousness in wheat: Its development and
effect on water-use efficiency, gas exchange and photosynthetic tissue temperatures*.
Funct Plant Biol 13: 465
Rowland O, Zheng H, Hepworth SR, Lam P, Jetter R, Kunst L (2006) CER4 encodes an
alcohol-forming fatty acyl-coenzyme A reductase involved in cuticular wax production
in Arabidopsis. Plant Physiol 142: 866–77
Ryan KG, Swinny EE, Markham KR, Winefield C (2002) Flavonoid gene expression and
UV photoprotection in transgenic and mutant Petunia leaves. Phytochemistry. 59: 23-32
Ryan KG, Swinny EE, Winefield C, Markham KR (2001) Flavonoids and UV
photoprotection in Arabidopsis mutants. Zeitschrift fur Naturforsch - Sect C J Biosci 56:
745–754
Sarrion-Perdigones A, Vazquez-Vilar M, Palací J, Castelijns B, Forment J, Ziarsolo P,
Blanca J, Granell A, Orzaez D (2013) Goldenbraid 2.0: A comprehensive DNA
assembly framework for plant synthetic biology. Plant Physiol 162: 1618–1631
Schnurr J, Shockey J, Browse J (2004) The Acyl-CoA synthetase encoded by LACS2 Is
essential for normal cuticle development in Arabidopsis. Plant Cell 16: 629–642
Seo PJ, Lee SB, Suh MC, Park MJ, Park CM (2011) The MYB96 transcription factor
regulates cuticular wax biosynthesis under drought conditions in Arabidopsis. Plant Cell
23: 1138–1152
Usadel B, Tohge T, Scossa F, Sierro N, Schmidt M, Vogel A, Bolger A, Kozlo A, Enfissi
EM, Morrel K, et al (2018) The genome and metabolome of the tobacco tree, Nicotiana
glauca : a potential renewable feedstock for the bioeconomy. bioRxiv 351429
Xue D, Zhang X, Lu X, Chen G, Chen ZH (2017) Molecular and evolutionary mechanisms
of cuticular wax for plant drought tolerance. Front Plant Sci. 8: 621
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
33
Yaaran A, Negin B, Moshelion M (2019) Role of guard-cell ABA in determining steady-
state stomatal aperture and prompt vapor-pressure-deficit response. Plant Sci 281: 31–40
Yeats TH, Rose JKC (2013) The formation and function of plant cuticles. Plant Physiol 163:
5–20
Zeisler-Diehl V, Müller Y, Schreiber L (2018) Epicuticular wax on leaf cuticles does not
establish the transpiration barrier, which is essentially formed by intracuticular wax. J
Plant Physiol 227: 66–74
Zeisler V, Schreiber L (2016) Epicuticular wax on cherry laurel (Prunus laurocerasus)
leaves does not constitute the cuticular transpiration barrier. Planta 243: 65–81
Zeltińš P, Katrevičs J, Gailis A, Maaten T, Baders E, Jansons A (2018) Effect of stem
diameter, genetics, and wood properties on stem cracking in Norway spruce. Forests 9:
546
Zheng H, Rowland O, Kunst L (2005) Disruptions of the Arabidopsis enoyl-CoA reductase
gene reveal an essential role for very-long-chain fatty acid synthesis in cell expansion
during plant morphogenesis. Plant Cell 17: 1467–1481
.CC-BY-NC-ND 4.0 International licensemade available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is
The copyright holder for this preprintthis version posted June 9, 2021. ; https://doi.org/10.1101/2021.06.08.447487doi: bioRxiv preprint
