Plant roots sense soil compaction through restricted ethylene diffusion

Abstract

Ethylene aplenty signals soil compaction
It's tough to drive a spade through compacted soil, and plant roots seem to have the same problem when growing in compacted ground. Pandey et al. found that the problem is not, however, one of physical resistance but rather inhibition of growth through a signaling pathway. The volatile plant hormone ethylene will diffuse through aerated soil, but compacted soil reduces such diffusion, increasing the concentration of ethylene near root tissues. The cellular signaling cascades triggered by too much ethylene stop root growth. Therefore, gaseous diffusion serves as a readout of soil compaction for plant roots growing in search of productive nutrition.
Science , this issue p. 276

Full text

science.sciencemag.org/content/371/6526/276/suppl/DC1
Supplementary Materials for
Plant roots sense soil compaction through restricted ethylene diffusion
Bipin K. Pandey, Guoqiang Huang, Rahul Bhosale, Sjon Hartman, Craig J. Sturrock, Lottie Jose,
Olivier C. Martin, Michal Karady, Laurentius A. C. J. Voesenek, Karin Ljung, Jonathan P. Lynch,
Kathleen M. Brown, William R. Whalley, Sacha J. Mooney, Dabing Zhang*, Malcolm J. Bennett*
*Corresponding author. Email: malcolm.bennett@nottingham.ac.uk (M.J.B.); zhangdb@sjtu.edu.cn (D.Z.)
Published 15 January 2021, Science 371, 276 (2021)
DOI: 10.1126/science.abf3013
This PDF file includes:
Materials and Methods
Figs. S1 to S25
Captions for movies S1 and S2
References
Other supplementary material for this manuscript includes:
Movies S1 and S2
MDAR Reproducibility Checklist (PDF)

2
Materials and Methods
Plant material and growth conditions:
Rice seeds (Nipponbare, osein2 and oseil1, proOsEIL1:OsEIL1-GFP, proOsEIL1:GUS) were
dehusked and surface sterilized with 25 % bleach for 10 minutes, then washed with sterilized water
five to six times. Sterilized seed were pre-germinated on moist sterilized filter paper for 2 days in
the dark at 30°C in a rice growth chamber (12-hour photoperiod and 300 µmol/m2/s light with
70 % relative humidity). After radicle emergence (5-8 mm in length), seedlings were gently placed
in soil for compaction experiments (see below). Arabidopsis seeds (Col-0, etr1, 35S:EIN3-
GFP/ein3eil1, 35S:RAP2.12-GFP, pPCO1:GFP-GUS, pPCO2:GFP-GUS, ein3eil1) were
sterilized with bleach for 2 minutes, then washed with sterilized water 5 times. Seeds were
stratified for 48 hours in a 4°C chamber in sterilized water. Leaf area of top four leaves of Col-0
and ein3eil1 mutants were quantified using Image J.
Soil compaction experimental details:
Loamy sand soil of the Newport series (FAO Eutric Cambisol) (78.7 % sand, 9.4 % silt, 11.9 %
clay and 2.3 % organic matter) and clay soil were collected from the University of Nottingham
farm at Bunny, Nottinghamshire, UK (52·52° N, 1·07° W). Soil was air dried, crushed and then
passed through a sieve with a 2 mm mesh size. To allow packing of soil to higher bulk densities,
the soil was lightly sprayed with sterilized water, mixed thoroughly and stored in dark for three to
four days to room temperature to equilibrate. The moisture content of the damp soil was measured
by drying 10 g of soil until it achieved a constant weight at 45°C in an oven.
The mesocosms were then packed with soil at bulk density (BD) of 1.6 g cm-3 to represent
compacted condition. The columns were saturated and then drained to a notional field capacity,
i.e., the point when free drainage had ceased. Germinated rice seedlings (one seedling per
mesocosm) with equivalent length radicles (see above) were placed on the soil surface and covered
with a 1 cm top layer of loose soil (1.1 BD) in both uncompacted and compacted soil mesocosms.
Seedlings were then placed in a rice growth chamber maintained at 30°C, 12-hour photoperiod and
300 µmol/m2/s light with 70 % relative humidity.

3
X-ray Computed Tomography imaging:
The root systems of five-day old rice seedlings were imaged non-destructively in the soil
mesocosms using a GE Phoenix v|tome|x M 240 kV X-ray tomography system (GE Inspection
Technologies, Wunstorf, Germany) at The Hounsfield Facility, University of Nottingham. Scans
were acquired by collecting 2520 projection images at 140 kV X-ray energy, 200 µA current and
131 ms detector exposure time in FAST mode (5-minute total scan time). Scan resolution was 45
µm. Three-dimensional image reconstruction was performed using Datos|REC software (GE
Inspection Technologies, Wunstorf, Germany). As root length was the key measurement for the
study, a polyline tool in VGStudioMAX V2.2 (Volume graphics GmbH, Germany) was used to
segment the roots from the soil. To verify difference in soil compaction, the defect analysis module
in VGStudioMAX V2.2 was used to quantify soil pore volume and proportion.
Imaging of root tip tissues harvested from compaction experiments:
Rice roots were removed from soil mesocosms and immediately washed with sterilized water in
three consecutive water filled dishes using fine brush strokes to remove soil particles from root
tips. Washed root tips were fixed in freshly prepared 4 % PFA (paraformaldehyde solution) for 1
hours in a vacuum chamber. After fixing, root tips were washed with 1X PBS buffer 3 to 4 times
(2 minutes/wash). Finally, root tips were cleared in ClearSee solution for 10-15 days, then stained
with either Yellow direct 96 (Diphenyl Brilliant Flavine 7GFF Sigma) or Calcofluor dye
(Calcofluor White Stain, Sigma) for 2 hours in a vacuum chamber. After staining, roots were
washed with ClearSee solution prior to confocal microscopy (Leica microsystem, model SP5). For
performing cross-sectional images, root samples comprising root cap, meristematic, elongation
and differentiation zones (~1.5 cm of rice root tip) were embedded in 3.5 % melted agarose. 150
µm thick transverse sections were cut using a Leica Vibrotome, then stained with Calcofluor white
dye for one minute on a glass slide and imaged with a Leica SP5 confocal microscope using the
UV laser.
Ethylene treatment conditions:
Pre-germinated rice seedlings were wrapped in moist paper towel and grown for 2 days in a
chamber at 30°C, 70 % relative humidity and 300 µmol/m2/s light conditions. After 2 days

4
seedlings wrapped in germination paper were moved into a 2.5 L beaker containing 400 ml
sterilized water, then placed in a 10 L gas tight microbiology chamber. Finally, ethylene gas was
injected into the chamber using calibrated syringe and sealed for 2 days using the above growth
conditions.
ACC treatment conditions of rice ethylene reporter:
Pregerminated seedlings of rice ethylene reporter (proOsEIL1:OsEIL1-GFP) were transferred on
½ MS plate containing 100 µM of ACC (1-aminocyclopropane-1-carboxylic acid). The plates were
kept in rice growth chamber at 30°C, 70 % relative humidity and 300 µmol/m2/s light conditions
for 48 hours. After 48 hours of ACC treatment root tips were harvested and imaged under SP8
confocal microscopy.
Arabidopsis soil compaction experimental details:
Arabidopsis seeds were placed on the top of moist soil (sandy loam, see above for details) in
miniaturized 3D printed mesocosms (120 mm height x 30 mm diameter), packed at either 1.1 BD
or 1.4 BD (with the latter topped with 3mm of 1.1 BD to aid seedling establishment). Columns
were designed to open as two semi cylindrical flaps without disrupting the intact root tips in soil.
Columns were grown in an Arabidopsis growth chamber (16-hour light 8-hour night cycle, 22°C
temperature, 100-125 µmol/m2/s light condition and 60 % relative humidity) for 10 days. Intact
Arabidopsis root tips were cleaned using paint brush in sterilized water to avoid drying. Root tips
were directly imaged using confocal microscopy or fixed in 4 % PFA for ClearSee treatment,
staining and confocal microscopy to study anatomical details of root cap, meristematic, elongation
and differentiation zones.
1-Aminocyclopropane-1-carboxylic acid (ACC) measurements
Wildtype (WT) rice seedlings (Nipponbare) were grown in uncompacted (1.1 BD) or compacted
soil (1.6 BD) conditions (as detailed above). After 4 days, root tips (1.5-2.0 cm from the root cap)
were washed and weighed (10 mg fresh weight each in five independent replicates), then snap
frozen in liquid nitrogen. Extraction from snap-frozen plant material was performed by adding 1
ml of H2O:methanol:chloroform (1:2:1) and d4-ACC (Sigma-Aldrich GmbH, Steinheim,

5
Germany) as internal standard. The samples were then homogenized using a bead mill (MixerMill,
Retsch GmbH, Haan, Germany), followed by a 15-minute centrifugation at 19000 rpm at 4°C, then
400 µL supernatant was removed and used for ACC quantification. Further, supernatant was
allowed to evaporate until powder and subsequently derivatized with the AccQ-Tag Ultra kit
(Waters, Milford, MA, USA) (16). ACC quantification was performed using a LC-MS/MS system
comprising a 1260 Infinity II LC System coupled to a 6495 Triple Quad LC/MS System with Jet
Stream and Dual Ion Funnel technologies (Agilent Technologies, Santa Clara, CA, USA). The
samples were injected on Kinetex Biphenyl column (100 x 2.1 mm, 1.7 µm; Phenomenex Inc.,
Torrance, CA, USA), employing a linear gradient (0 to 8.5 min, 8.5 % B; 8.5 to 12 min, 11 % B;
12 to 15 min, 98 % B) of 10 mM formic acid (A) and acetonitrile (B). The multiple reaction
monitoring transitions 272.1>171 and 276.1>171 were used for authentic ACC and d4-ACC
quantification, respectively. The concentration of endogenous ACC in the samples was determined
from the area ratio of endogenous ACC to the corresponding stable isotope-labeled standard (16).
Gas barrier (high vacuum silicone grease) experiment:
Arabidopsis lines expressing reporters 35S:RAP2.12-GFP, pPCO1:GFP-GUS, pPCO2:GFP-GUS
and 35S:EIN3-GFP/ein3eil1 were sterilized using 70 % ethanol for one minute and 25 % sodium
hypochlorite solution for 5 minutes. Seeds were washed with sterilized water six times and kept at
4°C for 2 days, then placed on agar plates containing 0.5XMS growth media and grown vertically
in an Arabidopsis growth room (16-hour light 8-hour night cycle, 22°C temperature, 100-125
µmol/m2/s light condition and 60 % relative humidity). Growing root tip tissues (encompassing
apical meristem and elongation zones) of four days old seedlings were sealed with Dow Corning®
high-vacuum silicone grease (Sigma Z273554) to restrict gas diffusion. After 2 hours the high
vacuum grease was removed from root tips which were washed using sterilized water and then
stained with propidium iodide (10 µg/ml) Sigma P-4170 solution for 3 minutes. Cleared root tips
were then imaged using a Leica SP8 confocal microscope.
Hypoxia treatment under water submerged conditions:
Arabidopsis lines of 35S:RAP2.12-GFP, pPCO1:GFP-GUS and pPCO2:GFP-GUS were directly
sown in sandy loam soil in 3D printed columns having 1.1 soil BD. The columns were grown in

6
an Arabidopsis growth room (16-hour light 8-hour night cycle, 22°C temperature, 100-125
µmol/m2/s light condition and 60 % relative humidity) for 6 days. After 6 days column was filled
with water for overnight. After 12 hours of hypoxia treatment, seedlings were harvested for root
tip clearing and SP8 imaging using confocal microscopy.
GUS histochemical assay:
Rice ethylene transcriptional reporter (proOsEIL1:GUS) seedlings grown in 1.1 and 1.6 BD soils
were removed carefully from the loamy sand and clay soils. Subsequently, the root tips were
harvested and gently washed with sterilized water. These root tips were then incubated in GUS
buffer at 37°C for overnight and the staining was performed as described previously (17). Further,
these root tips were cleared using acidified methanol at 55°C for 15 minutes. Acidified root tips
were neutralized with neutralizing solution (5M NaOH) for 10 minutes and subsequently
dehydrated by 40, 30, 20 and 10 % of ethanol each for 5 minutes. These root tips were fixed in 50
% glycerol and imaged in a brightfield stereo zoom microscope.
Penetrometer resistance and volumetric water content analysis of uncompacted and
compacted soil:
Penetrometer resistance of 1.1, 1.4 and 1.6 BD soil were performed using Instron 5944 (needle
penetrometer). Instrument was equipped with a load frame of 100N (Newton) load cell having 15
cm diameter lower support anvil. To penetrate the soil the 100N force was exerted by a needle
penetrometer (2 mm diameter and 30° cone angle) with a speed of 0.8 mm sec-1. The force was
analyzed as earlier (18) by Instron Bluehill Universal v4.03 software.
Volumetric water content was quantified by weighing the wet soil weight at saturated water level
and at field capacity level in both uncompacted and compacted soil. The moisture content was
evaluated by subtracting the dry soil weight from wet soil weight. Volumetric water content
(g/cm3) was quantified by dividing the soil bulk density values with soil water content. Volumetric
water content analysis was measured as performed earlier (19).
Rice ethylene reporter construct development and transgenic raising:

7
Rice ethylene translational reporter (proOsEIL1:OsEIL1-GFP) was constructed via One Step
Cloning Kit according to manufacturer’s protocol. The EIL1 genomic fragment without stop codon
was amplified by EIL1_F/EIL1_R primers. The PCR product was fused with pCAMBIA1301-
GFP backbone by restriction digestion method using Eco53kI and NruI restriction enzymes. The
homologous recombination reaction was performed at 37°C. The proOsEIL1:OsEIL1-GFP was
sequenced and aligned with the referenced sequence. Agrobacterium mediated rice (Nipponbare
genotype) transformation was performed as described earlier (20).
EIL1 F: GATTACGAATTCGAGGAATTCAATGGTGGTGGTATTGAAGCTG
EIL1 R: CTTGCTCACCATTCGGTAGTACCAATTCGAGCCGTCATTCT
Ethylene diffusion measurement system in uncompacted and compacted soils:
A custom setup was made to measure ethylene diffusion rates across soil samples. The air-tight
measuring setup consists of two glass chambers that are connected by a column containing a soil
sample of either 1.1 g cm-3 or 1.6 g cm-3 bulk density (1 cm thick). Ethylene was injected into the
top compartment at T = 0 to have ~20 µL L−1 initial concentration. Samples (3 mL) were taken
from both compartments over time and ethylene was measured using a gas chromatograph (GC).
For each air sample of 3 mL taken, 3 mL of MQ water was injected in the bottom chamber to
correct for dilution and pressure artifacts. Measurements were continued until ethylene
concentrations were equal in both compartments, with the exception of 1.6 g cm-3 bulk density soil
where no ethylene could be detected after 20 days of injection.
Modeling predicts that ethylene concentrations follow an exponential relaxation law towards
equilibrium:
Given the custom setup consisting of a top compartment separated from a bottom compartment by
the soil column, we can exploit the fact that diffusion through the soil is slow compared to the
diffusion within each separate compartment. We thus model the system by considering that each
compartment separately is in quasi equilibrium, that is the concentration of ethylene therein is
close to being uniform. Furthermore, we can use the steady state approximation for the diffusion
of ethylene through the soil column. Fick’s law then gives the formula for the net flux J (in Moles
per second) transported through the soil column considered as a slab of section ΔS.

8
Mathematically, J is equal to the product of De (an effective diffusion constant which is a
characteristic of the soil and will depend on its compaction level) times the ethylene concentration
gradient (the difference in concentrations between the two compartments divided by the height h
of the column) times the column’s surface area ΔS. This flux also determines the rate of change of
the concentrations in the two compartments. In our experimental apparatus the corresponding
volumes are identical and much larger than that of the soil sample. We can thus neglect the ethylene
in the sample, and then the rate of change of ethylene concentration in each compartment is
proportional to De times the deviation of the concentration C from its equilibrium value Ceq. (Note
that at equilibrium, the concentrations in the top and bottom compartments are equal and thus are
given by the amount of ethylene injected divided by the total volume.) The differential equation
satisfied by the concentration in either compartment is then dC/dt = - K (C- Ceq) where K is equal
to De times a factor depending on the dimensions of the apparatus, here equal to 2 ΔS / (h V) where
V is the volume of each compartment. Note that K corresponds to a rate: the larger De, the faster
the system returns to equilibrium. The solution to the differential equation is simply
!
"
#
$
% !&' (
)
!
"
# % *
$
+ !&'
,
-./0
This prediction of how one approaches equilibrium drove us to perform fits to the ethylene
concentrations as a function of time. The resulting fits to the uncompacted samples are displayed
in fig. S25 (colored curves) and show that the prediction of exponential relaxation is well borne
out. Since the compacted samples did not lead to any measurable equilibration, it was not possible
to perform fits to those cases.
The experimental measurements also included controls, i.e., measurements without any soil
sample. In such a configuration, the quasi equilibrium hypothesis we made above is no longer
valid. Nevertheless, dimensional analysis can be used to obtain a characteristic time scale for
relaxation to equilibrium. Indeed, a freely diffusing molecule has a mean square displacement
growing linearly with time t according to
1"
23
$
4
5
% 67#
where D is the molecule’s diffusion constant. The diffusion constant of ethylene in air is D = 0.2
cm² /sec. In our case the diffusion is constrained by the geometry of the bottles and so the « free

9
diffusion » hypothesis is not completely valid. However, to obtain just an order of magnitude, we
can use the formula above taking for the characteristic length scale the distance between the
injection point in the top bottle and the measuring point in the bottom bottle, Δx=26 cm. This then
predicts a characteristic time for relaxing to equilibrium as the value T = (Δx)² /(2 D), which
numerically is just above 28 minutes. We performed fits to the time series of these control
experiments as shown in the inset of fig. S25. The relaxation times (inverse of relaxation rate)
found in those fits vary between 25.7 and 31.5 minutes. This is certainly closer to the theoretical
result (28 minutes) than could have been demanded by dimensional analysis which only predicts
orders of magnitude, nevertheless it provides compelling evidence that the experiment has no
unanticipated biases.
Modeling provides a qualitative description how compaction level affects ethylene diffusion:
CT imaging of soil as shown in Fig. 1A and B provides insights into the local variations in density
and thus the presence of air pores that may serve as paths for ethylene diffusion. In practice, the
CT analysis leads to a binary representation of the density of matter within a region in the soil.
Specifically, the continuous region of interest is replaced by a regular cubic grid and binary
variables are assigned to each grid site, a value of « 0 » corresponding to having mainly air there
and a value of « 1 » corresponding to having mainly solid or water there. In our case the mesh size
(distance between two adjacent sites) is 45 microns. Based on such CT grids, in the uncompacted
soil samples (density of 1.1 g cm-3), about 5.4 % of the sites are of the air type (value 0). In the
compacted cases (density of 1.6 g cm-3), only about 1.4 % of the sites are of the air type.
Given such a grid representation of the soil, we have modeled diffusion of ethylene at that
microscopic scale as follows. First, each grid site is characterized by some ethylene concentration.
Second, when two neighboring sites have different ethylene concentrations, ethylene can flow
between them according to a diffusion rate as in Fick’s law. If both sites are of the air type, the
corresponding diffusion constant is the same as the one in air (0.2 cm²/sec). If neither sites are of
the air type, we impose a strongly suppressed diffusion by having a tiny diffusion constant (in
practice, 104 times smaller than above). In the case where just one of the two sites are of the air
type, we take the diffusion constant to be the average of the two previous cases. These microscopic
transfers of ethylene have the property that at equilibrium (all flows equal to 0) the concentrations

10
are all identical throughout the sample, a constraint in fact imposed by thermodynamics. Thus,
having equal concentrations at two grid sites does not mean that that they have equal amounts of
ethylene, one site can very well have more free volume to store the gas than the other. We have
extended this framework to allow for diffusion not only from a given site to its 6 nearest neighbors
but also to its neighbors in the larger cube where a unit displacement along each of the x, y and z
axes is allowed. In all cases, the net flow between pairs of sites is given by the diffusion constant
(dependent on the sites of the pair) times the concentration gradient, the gradient being the
difference in concentrations divided by the distance between the two sites.
Although the diffusion constant varies across sites, for a large sample this heterogeneity will
average out and the sample will behave according to Fick’s law (which is macroscopic) with an
effective diffusion constant De. One expects De to grow when the fraction of sites of the air type
rises. To make this quantitative, we have performed simulations to extract the corresponding
relationship (cf. Fig. 3H). Specifically, we generated in silico grids for columns in the shape of
cubes, for a series of values of the fraction of air sites therein. For each grid, we solved numerically
the equations for the steady state flux of ethylene flowing from top to bottom when a vertical
ethylene gradient was imposed as a boundary condition on the grid. The effective diffusion
constant De is then the ratio of that flux and of the macroscopic ethylene gradient imposed. In
practice we measure De in units of D, the diffusion constant of ethylene in air by simply comparing
to the flux arising when all sites in the grid are of the air type. These simulations confirm as
expected the monotonic dependence of De on the fraction of air sites (cf. Fig. 3H). They also show
that this dependence is rather steep for small fractions, and thus specifically that by going from
uncompacted to compacted soil, De can decrease dramatically.
The code for simulation modelling has been provided as additional file 1.

Fig. S1 Microstructure of loamy sand soil cores at different bulk densities.CT images
showing uncompacted (1.1 g cm-3 bulk density [BD]) versus compacted (1.6 BD) soil cores.
Measurable air-filled porosity for each soil core has been indicated.Core diameter, 30 mm.
Scale bars =20 mm.Yellow arrows denote air filled pores in each soil core.Rice root has been
indicated by black arrow.
1.1 BD 1.6 BD
Rice root Rice root
Measurable air-filled porosity
at 45 micron resolution
= 5.4 % (SEM: 0.2)
Measurable air-filled porosity
at 45 micron resolution
= 1.4 % (SEM: 0.3)
11

A1.1 BD
B1.6 BD
Soil pore volume [mm3]Soil pore volume [mm3]Soil pore volume [mm3]Soil pore volume [mm3]
Soil pore volume [mm3]Soil pore volume [mm3]Soil pore volume [mm3]Soil pore volume [mm3]
Fig. S2 Three-dimensional visualization of the soil pore space only.(Aand B) 3D images of air filled soil pores
for a100x100x100 voxel region from 1.1 BD (A) and 1.6 BD (B) soil cores.Four replicates for each bulk densities
(BD) have been shown.
12

Fig. S3 Ethylene treatment induces Arabidopsis ethylene EIN3-GFP reporter.(Aand B)Roots of
Arabidopsis ethylene reporter 35S:EIN3-GFP/ein3eil1were exposed to either 0or 20 ppm C2H4
(ethylene) for 2hours in agas chamber.Untreated root exhibit very faint GFP signal (A) as compared to
treated (B) root tips. (C)Violin plot showing mean GFP signal measured in Aand Broot tips (n= 14) .
Experiments were performed independently three times.*** represents pvalue ≤0.0001.Statistical
analysis was performed using Student’s t-test.. (D)Roots tips of Arabidopsis ethylene reporter 35S:EIN3-
GFP/ein3eil1were covered with HVSG (high vacuum silicon grease) gas barrier for 0hours, (E) 2 hours
and (F)10 hours. (G)Violin plot showing mean GFP signal measured in D, Eand Froots.*** represents
p≤0.0001.Bars =100 µm in Ato Band Dto F.
0ppm C2H420 ppm C2H4
AB
Mean signal intensity
0ppm 20 ppm
C2H4
***
n=14
n=14
C
35S:EIN3-GFP/ein3eil1
0
20
40
60
80
G
0hrs 2hrs 10 hrs
n=12 n=16 n=14
0
5
10
15
Mean signal intensity
***
***
Duration of HVSG application (Gas Barrier)
0hrs 2hrs 10 hrs
D E F
35S:EIN3-GFP/ein3eil1
Duration of HVSG application (Gas Barrier)
13

Fig. S4 Arabidopsis hypoxia reporter pPCO1:GFP-GUS is not induced by soil compaction and silicone
grease treatments.(A)Arabidopsis hypoxia reporter pPCO1:GFP-GUS shows induction when exposed to
submerged water conditions (hypoxia).Seedlings were submerged in water for 12 hrs. (Band C)No GFP
signal was detected in pPCO1:GFP-GUS roots covered with high vacuum silicon grease (HVSG) (compare
Cto B).HVSG which acts as agas barrier was applied on roots of 4days-old seedlings grown on ½MS agar
plates. (D)Submerged pPCO1:GFP roots was used as positive control for soil experiments.Flooding
treatment was given for 12 hrs in 1.1 BD soils. (Eand F)No GFP signal was detected in 10 days-old
pPCO1:GFP-GUS roots grown in 1.4 BD compacted soil (compare Fto E). (G)Table summarizing different
treatments given to pPCO1:GFP-GUS from Ato F. Bars =100 µm in Ato F.
A
Submerged (Positive control)
Positive control for HVSG experiment (submerged for 12 hrs)
B
-
Gas Barrier (HVSG)
Negative control for HVSG application
C
+Gas Barrier (HVSG)
HVSG application
D
Submerged_1.1 BD
Uncompacted soil submerged
for 12 hrs (Positive control for soil experiment)
E
1.1 BD
Uncompacted soil
F
1.4 BD
Compacted soil
Submerged _1.1 BD
(Positive control)
G
Submerged (Positive control)
ABC
FE
D
pPCO1:GFP-GUS
-Gas Barrier (HVSG) + Gas Barrier (HVSG)
1.1 BD 1.1 BD 1.4 BD 1.4 BD
14

Fig. S5 Arabidopsis hypoxia reporter pPCO2:GFP-GUS is not induced by soil compaction and silicone
grease treatments.(A)Arabidopsis hypoxia reporter pPCO2:GFP-GUS shows induction when exposed to
submerged water conditions (hypoxia).Seedlings were submerged in water for 12 hrs. (Band C)No GFP
signal was detected in pPCO2:GFP-GUS roots covered with high vacuum silicon grease (HVSG) (compare
Cto B).HVSG which acts as agas barrier was applied on roots of 4days-old seedlings grown on ½MS agar
plates. (D)Submerged pPCO2:GFP roots was used as positive control for soil experiments.Flooding
treatment was given for 12 hrs in 1.1 BD soils. (Eand F)No GFP signal was detected in 10 days-old
pPCO2:GFP-GUS roots grown in 1.4 BD compacted soil (compare Fto E). (G)Table summarizing different
treatments given to pPCO2:GFP-GUS from Ato F. Bars =100 µm in Ato F.
A
Submerged (Positive control)
Positive control for HVSG experiment (submerged for 12 hrs)
B
-
Gas Barrier (HVSG)
Negative control for HVSG application
C
+Gas Barrier (HVSG)
HVSG application
D
Submerged_1.1 BD
Uncompacted soil submerged
for 12 hrs (Positive control for soil experiment)
E
1.1 BD
Uncompacted soil
F
1.4 BD
Compacted soil
ABC
F
ED
G
Submerged (Positive control) -Gas barrier (HVSG) + Gas barrier (HVSG)
Submerged _1.1 BD
(Positive control)
1.1 BD 1.1 BD 1.4 BD 1.4 BD
pPCO2:GFP-GUS
15

Fig. S6 Arabidopsis hypoxia reporter RAP2.12-GFP is not induced by soil compaction and silicone
grease treatments.(A)Arabidopsis hypoxia reporter RAP2.12-GFP shows induction when exposed to
submerged water conditions (hypoxia).Seedlings were submerged in water for 12 hrs. (Band C)No GFP
signal was detected in RAP2.12-GFP roots covered with high vacuum silicon grease (HVSG) (compare Cto
B).HVSG which acts as agas barrier was applied on roots of 4days-old seedlings grown on ½MS agar
plates. (Dand E)No GFP signal was detected in 10 days-old RAP2.12-GFP roots grown in 1.4 BD
compacted soil (compare Eto D). (F)Table summarizing different treatments given to RAP2.12-GFP from
Ato E. Bars =100 µm in Ato E.
A
Submerged (Positive control)
Positive control for HVSG experiment (submerged for 12 hrs)
B
-
Gas Barrier (HVSG)
Negative control for HVSG application
C
+Gas Barrier (HVSG)
HVSG application
D
1.1 BD
Uncompacted soil
E
1.4 BD
Compacted soil
F
BC
DE
A
Submerged (Positive control) -Gas Barrier (HVSG) + Gas Barrier (HVSG)
1.1 BD 1.1 BD 1.4 BD 1.4 BD
RAP2.12-GFP
100µm 100µm 100µm 100µm 100µm
100µm100µm100µm
100µm
100µm
16

Fig. S7 Compacted soil inhibits primary root length of wildtype rice by reducing epidermal cell
elongation and increasing cortical cell diameter.(A)Representative images of excavated rice primary
roots when grown in 1.1 BD versus 1.6 BD soils. (B)Violin plots showing primary root lengths of WT
(wildtype) grown in uncompacted (1.1 BD) and compacted soil (1.6 BD). (Cand D)Violin plots showing
epidermal cell length and cortical cell diameter in root meristematic zone (MZ), elongation zone (EZ) and
differentiation zone (DZ).Plants were grown in either 1.1 BD or 1.6 BD soils.Measurements were made
using Image Jsoftware.Epidermal cell length and cortical cell diameter was measured in at least 8
independent primary roots.*** shows p≤0.0001.Statistical analysis was performed using Student’s t-test.
Bars= 10 mm in A.
1.1 BD 1.6 BD
B
Primary root length (cm)
***
n=17
n=19
3
4
5
6
7
8
Epidermal cell length(μm)
WT_MZ WT_EZ WT_DZ
n= 40 n= 40
n= 20
n= 20
n= 28
n= 28
***
***
C
1.1 1.6 1.1 1.6 1.1 1.6
0
20
40
60
80
100
Cortical cell diameter (μm)
n= 78 n= 78 n= 78
n= 78 n= 80
n= 78
***
***
D
1.1 1.6 1.1 1.6 1.1 1.6
WT_MZ WT_EZ WT_DZ
0
10
20
40
50
60
30
1.6 BD1.1 BD
A
17

Fig. S8 Ethylene treatment of wildtype rice roots phenocopies effect of soil compaction at
cellular level.(A)Violin plot showing the reduction of primary root length in rice seedlings after
ethylene (C2H4)treatment. (Band C) Violin plots showing quantitative values of epidermal cell
length (B) and cortical cell diameter (C) in ethylene treated or untreated rice roots.Roots were
exposed to 10 ppm ethylene for 48 hours.Root tips were cleared and stained with Yellow 96 dye
and imaged using aLeica SP8confocal microscope.Epidermal cell length and cortical cell
diameter were measured using Image J. ** and *** show p≤0.005 and 0.0001,respectively.
Statistical analysis was performed using Student’s t-test.
Epidermal cell length (µm)
B
***
n=34
n=34
0 ppm 10 ppm
C2H4
Cortical cell diameter (µm)
C
***
n=75
n=50
0 ppm 10 ppm
C2H4
**
Primary root length (mm)
A
0 ppm 10 ppm
n=12
n=12
C2H4
4
5
6
7
8
9
20
40
60
80
30
40
20
10
18

Fig. S9 Rice ethylene insensitive mutants (osein2and oseil1)do not exhibit primary root
inhibition following ethylene treatments.(Aand B)Primary root lengths of 4days-old WT
(wildtype), osein2(A) and oseil1(B) seedlings after 48 hours of 0and 20 ppm ethylene treatments.
Assays were performed independently three times.*** shows pvalue ≤0.0001.Statistical analysis
was performed using Student’s t-test.
0 ppm 20 ppm 0 ppm 20 ppm
Primary root length (cm)
***
WT osein2 WT oseil1
Primary root length (cm)
***
0 ppm 20 ppm 0 ppm 20 ppm
A B
n=10
n=10
n=10 n=10
n=10
n=10
n=10
n=10
4
6
8
10
12
4
6
8
10
12
19

Fig. S10 Penetrometer resistance of compacted soil is higher than uncompacted soil.(Aand B)
Penetrometer resistance of 1.1 g cm-3 bulk density (BD) versus 1.4 BD soils (A) and 1.1 versus 1.6
BD soils (B) used for Arabidopsis and rice soil compaction experiments, respectively. (Cand D)
Bar graphs showing volumetric water content (VWC in gcm-3)of 1.1 and 1.4 BD soils used for
Arabidopsis soil compaction assays following different irrigation treatments;saturated (C) and field
capacity (D). (Eand F)Bar graphs showing volumetric water content (VWC in gcm-3)in 1.1 and
1.6 BD soils used for rice soil compaction assays following different irrigation treatments;saturated
(E) and field capacity (F).The experiment was independently repeated three times with five
replicates *, ** and *** represent pvalue ≤0.05, 0.001 and 0.0001.Statistical analysis was
performed using Student’s t-test.
0
0.1
0.2
0.3
0.4
0.5
VWC_1.1 BD VWC_1.6 BD
VWC (g cm-3)
0
0.1
0.2
0.3
0.4
0.5
VWC_1.1 BD VWC_1.6 BD
VWC (g cm-3)
***
***
F
E
Rice_Saturated Rice_Field capacity
0
100
200
300
400
500
600
1.1 BD 1.6 BD
Mean penetrometer
Resistance (kPa)
0
100
200
300
400
500
600
1.1BD 1.4 BD
Mean penetrometer
Resistance (kPa)
Rice
***
***
A B
0
0.1
0.2
0.3
0.4
0.5
VWC_1.1 BD VWC_1.4 BD
VWC (g cm-3 )
*
0
0.1
0.2
0.3
0.4
0.5
VWC_1.1 BD VWC_1.4 BD
VWC (g cm-3)
**
C D
Arabidopsis_Saturated Arabidopsis_Field capacity
Arabidopsis
20

Fig. S11 The rice osein2mutant does not exhibit enlargement of root cortical cell diameter and
reduction in epidermal cell length when grown in compacted soil.(Ato F)Representative confocal
images showing cortical cell diameter in meristem (MZ), elongation (EZ) and differentiation zones (DZ) of
osein2roots either grown in uncompacted (1.1 BD; A to C) or compacted (1.6 BD; D to F) soils.Cortical
cells have been outlined in red. (Gand H)Violin plots showing cortical cell diameter (G) and epidermal
cell length (H) in MZ, EZ and DZ zones from roots of osein2mutant grown in uncompacted (1.1 BD) or
compacted (1.6 BD) soils.Analysis was performed in at least 6independent primary roots.Significant
differences were evaluated by Student’s t-test.Bars =100 µm in Ato F.
osein2_1.1 BD
DZ
MZ EZ DZ
A B C
F
osein2_1.6 BD
MZ
DEZ
E
osein2_1.1 BD osein2_1.1 BD
osein2_1.6 BD osein2_1.6 BD
1.1 1.6 1.1 1.6 1.1 1.6
G H
Cortical cell thickness (µm)
Epidermal cell length (µm)
n=140 n=140
n=145
n=155
n=75
n=75
n=45 n=45
n=45 n=45
n=45 n=45
osein2_MZ osein2_EZ osein2_DZ
1.1 1.6 1.1 1.6 1.1 1.6
osein2_MZ osein2_EZ osein2_DZ
0
10
20
30
40 80
0
20
40
60
21

Fig. S12 The rice oseil1mutant does not exhibit enlargement of root cortical cell diameter and
reduction in epidermal cell length when grown in compacted soil.(Ato F)Representative confocal
images showing cortical cell diameter in meristem (MZ), elongation (EZ) and differentiation zones (DZ) of
oseil1roots either grown in uncompacted (1.1 BD; A to C) or compacted (1.6 BD; D to F) soils. (Gand H)
Violin plots showing cortical cell diameter (G) and epidermal cell length (H) in MZ, EZ and DZ zones
from roots of oseil1mutant grown in uncompacted (1.1 BD) or compacted (1.6 BD) soils.Analysis was
performed in at least 6independent primary roots.Significant differences were evaluated by Student’s t-
test.Bars =100 µm in Ato F.
oseil1_1.6 BD
oseil1_1.1 BD
AB C
DF
MZ EZ DZ
MZ DZEEZ
Cortical cell thickness (µm)
Epidermal cell length (µm)
GH
n=140 n=140
n=75 n=75 n=60 n=60
n=40 n=40
n=40 n=40
n=40
n=40
1.1 1.6 1.1 1.6 1.1 1.6
oseil1_MZ oseil1_EZ oseil1_DZ
1.1 1.6 1.1 1.6 1.1 1.6
oseil1_MZ oseil1_EZ oseil1_DZ
oseil1_1.1 BD oseil1_1.1 BD
oseil1_1.6 BD oseil1_1.6 BD
100
0
20
40
80
60
120
0
10
20
30
40
22

Fig. S13 Compaction increases root cortical cell thickness and reduces epidermal cell elongation in
wildtype rice.(Ato F)Representative images showing cortical cell expansion in WT (wildtype) roots
grown in uncompacted soil (1.1 BD) (A to C) and compacted soil (D to F).Some cortical cells are marked
with red boundaries for clear observation.Roots were excavated from soil packed columns (both 1.1 and
1.6 BD) and quickly washed with water, fixed in 0.4% para-formaldehyde and cleared using ClearSee
solution.Cleared roots were stained with Calcofluor-White and imaged using Leica SP8multi-photon
confocal microscope.Bars =100 µm.
WT_MZ WT_EZ WT_DZ
1.1 BD1.1 BD 1.1 BD
1.6 BD 1.6 BD
WT_EZ WT_DZ
ABC
F
E
WT_MZD
1.6 BD
23

Fig. S14 Soil compaction responses in wildtype Arabidopsis primary roots are blocked in the
ethylene receptor mutant etr1. (Ato C)Representative confocal images of Col-0 primary root
grown in uncompacted (1.1 BD) (A);and compacted (1.4 BD) (B-C) soil conditions. (Dto F)
Representative confocal images of etr1primary roots grown in 1.1 BD (D) and 1.4 BD (E-F) soil
conditions.For clarity, cortical cells are outlined in red (C and F).Bars =100 µm.
etr1_1.1BD etr1_1.4 BD etr1_1.4 BD
Col-0_1.1 BD Col-0_1.4 BD Col-0_1.4 BD
AB C
DE F
24

10
20
50
60
30
40
Cortical cell diameter (μm)
0
1.1 1.4
WT_MZ
1.1 1.4
WT_EZ
1.1 1.4
WT_DZ
50
100
200
250
150
0
Epidermal cell length(μm)
1.1 1.4
WT_MZ
1.1 1.4
WT_EZ
1.1 1.4
WT_DZ
10
20
30
40
0
1.1 1.4
etr1_MZ
1.1 1.4
etr1_EZ
1.1 1.4
etr1_DZ
Cortical cell diameter (μm)
50
100
200
250
150
0
1.1 1.4
etr1_MZ
1.1 1.4
etr1_EZ
1.1 1.4
etr1_DZ
Epidermal cell length(μm)
AB
CD
***
***
***
***
n= 40 n= 40
n= 40
n= 40
n= 40
n= 40
n= 40 n= 40 n= 40 n= 40
n= 40
n= 40
n= 40 n= 40
n= 40 n= 40 n= 40 n= 40
n= 40 n= 40
n= 40 n= 40
n= 40 n= 40
Fig. S15 Soil compaction responses are blocked in primary roots of Arabidopsis ethylene
receptor mutant etr1. (Ato B)Violin plots showing cortical cell diameter (A) and epidermal cell
length (B) in MZ, EZ and DZ zones from primary roots of Col-0 grown in uncompacted (1.1 BD)
or compacted (1.4 BD) soils. (Cto D)Violin plots showing cortical cell diameter (C) and epidermal
cell length (D) in MZ, EZ and DZ zones from primary roots of ethylene insensitive mutant etr1
grown in uncompacted (1.1 BD) or compacted (1.4 BD) soils.Significant differences were
evaluated by Student’s t-test.*** show pvalue ≤0.0001.
25

Fig. S16 Soil compaction also induces cortical cell expansion and epidermal cell reduction in
wildtype Arabidopsis lateral roots but is blocked in the ethylene receptor etr1mutant.(Aand
B)Representative Col-0 lateral roots showing expanded cortical cells in compacted soil (1.4 BD)
(compare A-B). (Cand D)Lateral roots of Arabidopsis etr1mutants do not display cortical cell
expansion in compacted soils as compared to wildtype [(WT) (compare Dto B)].Bars =100 µm.
WT_1.1 BD WT_1.4 BD etr1_1.4 BD
etr1_1.1 BD
A B C D
26

AB
Cortical cell diameter (μm)
1.1 1.4
10
20
50
60
70
30
40
WT_MZ
n= 40
1.1 1.4
WT_EZ
1.1 1.4
WT_DZ
n= 40 n= 40
n= 50 n= 40 n= 50
Epidermal cell length(μm)
1.1 1.4
WT_MZ
1.1 1.4
WT_EZ
1.1 1.4
WT_DZ
0
50
200
250
300
100
150
1.1 1.4
etr1_MZ
1.1 1.4
etr1_EZ
1.1 1.4
etr1_DZ
Cortical cell diameter (μm)
10
30
40
20
1.1 1.4
etr1_MZ
1.1 1.4
etr1 _EZ
1.1 1.4
etr1 _DZ
Epidermal cell length(μm)
0
50
200
250
100
150
n= 30 n= 30 n= 40 n= 40 n= 60 n= 40
n= 30 n= 30
n= 30 n= 30
n= 30
n= 30
n= 30 n= 30 n= 30
n= 30
n= 30
n= 30
***
*** ***
***
*** ***
CD
Fig. S17 Soil compaction responses are blocked in lateral roots of Arabidopsis ethylene receptor
mutant etr1. (Ato B)Violin plots showing cortical cell diameter (A) and epidermal cell length (B) in
MZ, EZ and DZ zones from lateral roots of Col-0 grown in uncompacted (1.1 BD) or compacted (1.4
BD) soils. (Cto D)Violin plots showing cortical cell diameter (C) and epidermal cell length (D) in MZ,
EZ and DZ zones from lateral roots of ethylene insensitive mutant etr1grown in uncompacted (1.1 BD)
or compacted (1.4 BD) soils.Significant differences were evaluated by Student’s t-test.*** shows p
value ≤0.0001.
27

Fig. S18 Mutants of rice oseil1and osein2exhibit enhanced root and shoot growth in compacted
soils compared to wildtype.(A)Violin plot showing root fresh weight in WT (wildtype), oseil1and
osein2mutants in 1.1 BD and 1.6 BD soils.(B)Violin plot showing the shoot fresh weight in WT,
oseil1and osein2mutants in 1.1 BD and 1.6 BD soils.Experiment was independently repeated three
times.*** shows pvalue ≤0.0001.Significant differences were evaluated by Student’s t-test.
***
*** ***
***
AB
Root fresh weight (mg)
n=8 n=8
n=8
n=8
n=8 n=8
Shoot fresh weight (g)
n=12 n=12
n=12
n=12
n=12
n=12
WT osei l1 osein2 WT oseil1 osein2
1.6 BD
1.1 BD
WT osei l1 osein2 WT oseil1 osein2
1.6 BD
1.1 BD
20
30
40
50
60
70
0
50
100
150
28

Fig. S19 Ethylene insensitive mutant ein3eil1exhibit enhanced leaf expansion in compacted soils
compared to wildtype.(Aand B)Representative aerial images of 10 days-old Col-0 grown in
uncompacted (A) and compacted soil (B). (Cand D)Representative aerial images of 10 days-old
Arabidopsis ein3eil1mutants grown in uncompacted (C) and compacted soil (D). (Eand F)Violin
plots showing the root length (E) and leaf area measurements (F) of Col-0 and ein3eil1double mutant
grown in 1.1 BD and 1.4 BD.*** represents pvalue ≤0.0001.Significant differences were evaluated
by Student’s t-test.Bars = 2 cm in Ato D.
Col-0_1.1 BD Col-0_1.4 BD
ein3eil1_1.1BD ein3eil1_1.4BD
***
***
Root length (cm)
Col-0 ein3eil1 Col-0 ein3eil1
n=10
n=10
n=10
n=10
Col-0 ein3eil1 Col-0 ein3eil1
1.1 BD 1.4 BD
Leaf area(mm2)
n=21 n=21
n=21
n=21
AB
DC
E
F
1.1 BD 1.4 BD
0
2
4
6
8
10
12
0
2
4
6
8
10
12
29

Fig. S20 Ethylene treatment reduces root cap area in wildtype rice.(Ato C)Increasing dosages (0, 10
and 30 ppm) of ethylene causes areduction in root cap area. (D)Violin plot showing quantitative data of root
cap area.Central columella cells are marked with red boundaries and lateral columella cells are marked in
purple.n=8represents the number of root caps analysed for calculating the root cap area using Image J. Bars
=100 µm in Ato C. * and ** represent pvalues ≤0.05 and ≤0.0001,respectively.Significant differences
were evaluated by Student’s t-test.
30 ppm C2H4
10 ppm C2H4
ABC
0 ppm C2H4
D
0 ppm 10 ppm 30 ppm
Root cap area (µm2)
*
**
n=8
n=8
n=8
0
5000
10000
15000
20000
C2H4
30

0μM ACC 100 μM ACC_48 hrs
A B
proOsEIL1:OsEIL1-GFP
1.1 BD 1.6 BD
F
E
Fig. S21 Rice ethylene reporter proOsEIL1:OsEIL1-GFP is induced in compacted clay soil.(Ato
D)proOsEIL1:OsEIL1-GFP translational reporter showing increased expression of GFP in root tip after
treatment with ethylene precursor aminocyclopropane-1-carboxylic acid (ACC) (B and C).100 µM
ACC treatment was given for 48 hrs (B) and 72 hrs (C). (D)Inset shows nuclear signal of OsEIL1-GFP.
(Eand F)Representative images of rice proOsEIL1:OsEIL1-GFP translational reporter showing higher
expression at the root tip in compacted clay soil (1.6 BD) (F) as compared to uncompacted clay soil (1.1
BD) (E).Rice seedlings of proOsEIL1:OsEIL1-GFP were grown in clay soil in uncompacted and
compacted conditions for four days.Root tips were carefully removed from clay soil and thereafter
cleared (prefixed in 4% PFA) and stained with propidium iodide and imaged using Leica SP8confocal
microscopy.Bars =200 µm in Ato D.
31
100 μM ACC_72 hrs
CD

Fig. S22 Rice ethylene reporter proOsEIL1:GUS exhibits higher expression in compacted
clay soil.(Aand B)Representative images of rice proOsEIL1:GUS transcriptional reporter
showing higher expression at the root tip and in the root elongation zone in compacted clay soil
(1.6 BD) (B) as compared to uncompacted clay soil (1.1 BD) (A).Rice seedlings of
proOsEIL1:GUS were grown in clay soil in uncompacted and compacted conditions for four
days.Root tips were washed prior to histochemical GUS staining.
1.1 BD clay soil 1.6 BD clay soil
Root tip Elongation zone Root tip Elongation zone
A B
32

Fig. S23 Rice ethylene reporter proOsEIL1:GUS exhibit higher expression in compacted
sandy loam soil.(Aand B)Representative images showing expression of transcriptional reporter
proOsEIL1:GUS at root tip following ethylene treatment (compare Bto A). 5 days-old rice seedling
expressing proOsEIL1-GUS transcriptional reporter were treated with 30 ppm of ethylene in gas
chamber for two hours. (Cand D)Representative images of proOsEIL1:GUS showing higher
expression at root tip and in elongation zone in compacted soil (1.6 BD) (D) as compared to
uncompacted soil (1.1 BD) (C).Rice seedlings of proOsEIL1:GUS were grown in sandy loam soil
in uncompacted and compacted conditions for four days.
1.1 BD sandy loam soil 1.6 BD sandy loam soil
A B
CD
Root tip Elongation zone Root tip Elongation zone
0ppm 20 ppm
C2H4
33

Fig. S24 Soil compaction does not influence aminocyclopropane-1-carboxylic acid (ACC)
levels in rice root tips.Violin plot showing ACC levels in WT rice (wildtype) root tips grown in
uncompacted (1.1 BD) versus compacted (1.6 BD) soil conditions. 4 days-old WT root tips (1.5
cm) were harvested from compacted and uncompacted soils (see methods for details).10 mg root
tissues were used for ACC estimation using LC-MS/MS.Each replicate contain ~6-8 rice root tips.
Estimation was performed in five independent replicates.Significant differences were evaluated by
Student’s t-test.ns indicates no significant difference.
ACC concentrations
(pg/mg fresh weight)
n=5
n=5
1.1 BD 1.6 BD
60
80
100
120
140
160
ns
34

Fig. S25 Modelling of soil ethylene diffusion measurements reveals exponential return to
equilibrium for hormone concentrations.(A)Given the time series of the 4low compaction
soil samples and of the 2controls (Fig. 3I), we performed fits to aconstant plus an exponentially
decaying law.The cases of the four uncompacted soil samples are represented in color.The fitted
relaxation times for the top and bottom bottles are very similar for agiven sample but vary
significantly across the 4samples, indicating large heterogeneities therein.In contrast, the inset
shows that the two controls give nearly identical curves;furthermore, the corresponding
relaxation times are in good agreement with the theoretical prediction. (B)Custom setup used for
the quantification of ethylene diffusion in uncompacted (1.1 BD) and compacted (1.6 BD) soil
cores.
Injection of ~20 μL L−1 ethylene
(final concentration) in top
compartment at T= 0 min
Measure ethylene
concentration in time until
equilibrium
Soil sample of
different
compaction
A
B
35

36
Movie S1.
CT scan movie showing video of uncompacted soil. Black space is air space filled large pores.
Movie S2.
CT scan movie showing video of compacted soil. Black space is air space filled large pores
which are rarely visible in compact soil as compare to uncompact soil.
References:
1. J. Correa, J. A. Postma, M. Watt, T. Wojciechowski, J. Exp. Bot. 70, 6019-6034 (2019).
2. B. D. Soane, P. S. Blackwell, J. W. Dickson, D. J. Painter, Soil Till. Res. 2, 3-36 (1982).
3. S. Mangalassery, S. Sjögersten, D. L. Sparkes, C. J. Sturrock, J. Craigon, S. J. Mooney,
Sci. Rep. 4, 1-8 (2014).
4. M. A. Hamza, W. K. Anderson, Soil Till. Res. 82, 121-145 (2005).
5. R. Horn, H. Domżżał, A. Słowińska-Jurkiewicz, C. Van Ouwerkerk, Soil Till. Res. 35, 23-
36 (1995).
6. T. Fujikawa, T. Miyazaki, Soil Sci. 170, 892-901 (2005).
7. A. Hussain, C. R. Black, I. B. Taylor, J. A. Roberts, Plant Physiol. 121, 1227-1237 (1999).
8. I. Potocka, J. Szymanowska-Pulka, Ann. Bot. 122, 711-723 (2018).
9. F. An et al., Plant Cell 22, 2384-2401 (2010).
10. D. A. Weits et al., Nature 569, 714-717 (2019).
11. S. Hartman et al., Nature Commun. 10, 1-9 (2019).
12. B. Ma et al., Mol. Plant. 6, 1830-1848 (2013).
13. J. M. Alonso, T. Hirayama, G. Roman, S. Nourizadeh, J. R. Ecker, Science 284, 2148-
2152 (1999).
14. K. D. Montagu, J. P. Conroy, B. J. Atwell, J. Exp. Bot. 52, 2127-2133 (2001).
15. F. J. Romera, E. Alcantara, M. D. De la Guardia, Ann. Bot. 83, 51-55 (1999).
16. L. Lamaire, C. Deleu, E. L. Deunff. J. Exp. Bot. 64, 2725-2737 (2013).
17. P. Mehra P, B. K. Pandey, L. Verma, J. Giri J, Plant Cell Environ. 42, 1167-1179 (2019).
18. W. Zhang, W. Gao, T. Ren, W. R. Whalley, Geoderma 368, 114276 (2020).

References
1. J. Correa, J. A. Postma, M. Watt, T. Wojciechowski, Soil compaction and the architectural
plasticity of root systems. J. Exp. Bot. 70, 6019–6034 (2019). doi:10.1093/jxb/erz383
Medline
2. B. D. Soane, J. W. Dickson, D. J. Campbell, Compaction by agricultural vehicles: A review
III. Incidence and control of compaction in crop production. Soil Tillage Res. 2, 3–36
(1982). doi:10.1016/0167-1987(82)90030-7
3. S. Mangalassery, S. Sjögersten, D. L. Sparkes, C. J. Sturrock, J. Craigon, S. J. Mooney, Sci.
Rep. 4, 1–8 (2014).
4. M. A. Hamza, W. K. Anderson, Soil compaction in cropping systems. Soil Tillage Res. 82,
121–145 (2005). doi:10.1016/j.still.2004.08.009
5. R. Horn, H. Domżżał, A. Słowińska-Jurkiewicz, C. Van Ouwerkerk, Soil compaction
processes and their effects on the structure of arable soils and the environment. Soil
Tillage Res. 35, 23–36 (1995). doi:10.1016/0167-1987(95)00479-C
6. T. Fujikawa, T. Miyazaki, Effects of bulk density and soil type on the gas diffusion coefficient
in repacked and undisturbed soils. Soil Sci. 170, 892–901 (2005).
doi:10.1097/01.ss.0000196771.53574.79
7. A. Hussain, C. R. Black, I. B. Taylor, J. A. Roberts, Soil compaction. A role for ethylene in
regulating leaf expansion and shoot growth in tomato? Plant Physiol. 121, 1227–1238
(1999). doi:10.1104/pp.121.4.1227 Medline
8. I. Potocka, J. Szymanowska-Pulka, Morphological responses of plant roots to mechanical
stress. Ann. Bot. 122, 711–723 (2018). Medline
9. F. An, Q. Zhao, Y. Ji, W. Li, Z. Jiang, X. Yu, C. Zhang, Y. Han, W. He, Y. Liu, S. Zhang, J.
R. Ecker, H. Guo, Ethylene-induced stabilization of ETHYLENE INSENSITIVE3 and
EIN3-LIKE1 is mediated by proteasomal degradation of EIN3 binding F-box 1 and 2 that
requires EIN2 in Arabidopsis. Plant Cell 22, 2384–2401 (2010).
doi:10.1105/tpc.110.076588 Medline
10. D. A. Weits, A. B. Kunkowska, N. C. W. Kamps, K. M. S. Portz, N. K. Packbier, Z. Nemec
Venza, C. Gaillochet, J. U. Lohmann, O. Pedersen, J. T. van Dongen, F. Licausi, An
apical hypoxic niche sets the pace of shoot meristem activity. Nature 569, 714–717
(2019). doi:10.1038/s41586-019-1203-6 Medline
11. S. Hartman, Z. Liu, H. van Veen, J. Vicente, E. Reinen, S. Martopawiro, H. Zhang, N. van
Dongen, F. Bosman, G. W. Bassel, E. J. W. Visser, J. Bailey-Serres, F. L. Theodoulou,
K. H. Hebelstrup, D. J. Gibbs, M. J. Holdsworth, R. Sasidharan, L. A. C. J. Voesenek,
Ethylene-mediated nitric oxide depletion pre-adapts plants to hypoxia stress. Nat.
Commun. 10, 4020 (2019). doi:10.1038/s41467-019-12045-4 Medline

12. B. Ma, S.-J. He, K.-X. Duan, C.-C. Yin, H. Chen, C. Yang, Q. Xiong, Q.-X. Song, X. Lu, H.-
W. Chen, W.-K. Zhang, T.-G. Lu, S.-Y. Chen, J.-S. Zhang, Identification of rice
ethylene-response mutants and characterization of MHZ7/OsEIN2 in distinct ethylene
response and yield trait regulation. Mol. Plant 6, 1830–1848 (2013).
doi:10.1093/mp/sst087 Medline
13. J. M. Alonso, T. Hirayama, G. Roman, S. Nourizadeh, J. R. Ecker, EIN2, a bifunctional
transducer of ethylene and stress responses in Arabidopsis. Science 284, 2148–2152
(1999). doi:10.1126/science.284.5423.2148 Medline
14. K. D. Montagu, J. P. Conroy, B. J. Atwell, The position of localized soil compaction
determines root and subsequent shoot growth responses. J. Exp. Bot. 52, 2127–2133
(2001). doi:10.1093/jexbot/52.364.2127 Medline
15. F. J. Romera, E. Alcantara, M. D. De la Guardia, Ethylene Production by Fe-deficient Roots
and its Involvement in the Regulation of Fe-deficiency Stress Responses by Strategy I
Plants. Ann. Bot. 83, 51–55 (1999). doi:10.1006/anbo.1998.0793
16. L. Lemaire, C. Deleu, E. Le Deunff, Modulation of ethylene biosynthesis by ACC and AIB
reveals a structural and functional relationship between the K15NO3 uptake rate and root
absorbing surfaces. J. Exp. Bot. 64, 2725–2737 (2013). doi:10.1093/jxb/ert124 Medline
17. P. Mehra, B. K. Pandey, L. Verma, J. Giri, A novel glycerophosphodiester phosphodiesterase
improves phosphate deficiency tolerance in rice. Plant Cell Environ. 42, 1167–1179
(2019). doi:10.1111/pce.13459
18. W. Zhang, W. Gao, T. Ren, W. R. Whalley, A novel approach to estimate soil penetrometer
resistance from water content, bulk density, and shear wave velocity: A laboratory study
on a loamy sand soil. Geoderma 368, 114276 (2020).
doi:10.1016/j.geoderma.2020.114276
19. S. R. Tracy, K. R. Daly, C. J. Sturrock, N. M. J. Crout, S. J. Mooney, T. Roose, Three-
dimensional quantification of soil hydraulic properties using X-ray Computed
Tomography and image-based modeling. Water Resour. Res. 51, 1006–1022 (2015).
doi:10.1002/2014WR016020
20. G. Huang, W. Liang, C. J. Sturrock, B. K. Pandey, J. Giri, S. Mairhofer, D. Wang, L. Muller,
H. Tan, L. M. York, J. Yang, Y. Song, Y.-J. Kim, Y. Qiao, J. Xu, S. Kepinski, M. J.
Bennett, D. Zhang, Rice actin binding protein RMD controls crown root angle in
response to external phosphate. Nat. Commun. 9, 2346 (2018). doi:10.1038/s41467-018-
04710-x Medline