Biosensors and Bioelectronics 231 (2023) 115300

Contents lists available at ScienceDirect

Biosensors and Bioelectronics

journal homepage: www.elsevier.com/locate/bios

Non-invasive in-vivo glucose-based stress monitoring in plants

Sammy A. Perdomo a, b, 1, Ernesto De la Paz b, 1, Rafael Del Caño b, c, 1, Sumeyye Seker b,
Tamoghna Saha b, Joseph Wang b, **, Andres Jaramillo-Botero a, d, *
a
Omicas Alliance, Pontificia Universidad Javeriana, Cali, 760031, Colombia
b
Department of Nanoengineering, University of California, San Diego, San Diego, CA, 92093, United States
c
Department of Physical Chemistry and Applied Thermodynamics, University of Cordoba, E- 14014, Spain
d
Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, 91125, United States

A R T I C L E I N F O A B S T R A C T

Keywords: Plant stress responses involve a suite of genetically encoded mechanisms triggered by real-time interactions with
Glucose biosensor their surrounding environment. Although sophisticated regulatory networks maintain proper homeostasis to
Non-invasive detection prevent damage, the tolerance thresholds to these stresses vary significantly among organisms. Current plant
In-vivo monitoring
phenotyping techniques and observables must be better suited to characterize the real-time metabolic response
Plant stress
Precision agriculture
to stresses. This impedes practical agronomic intervention to avoid irreversible damage and limits our ability to
breed improved plant organisms. Here, we introduce a sensitive, wearable electrochemical glucose-selective
sensing platform that addresses these problems. Glucose is a primary plant metabolite, a source of energy pro­
duced during photosynthesis, and a critical molecular modulator of various cellular processes ranging from
germination to senescence. The wearable-like technology integrates a reverse iontophoresis glucose extraction
capability with an enzymatic glucose biosensor that offers a sensitivity of 22.7 nA/(μM⋅cm2), a limit of detection
(LOD) of 9.4 μM, and a limit of quantification (LOQ) of 28.5 μM. The system’s performance was validated by
subjecting three different plant models (sweet pepper, gerbera, and romaine lettuce) to low-light and low-high
temperature stresses and demonstrating critical differential physiological responses associated with their glucose
metabolism. This technology enables non-invasive, non-destructive, real-time, in-situ, and in-vivo identification of
early stress response in plants and provides a unique tool for timely agronomic management of crops and
improving breeding strategies based on the dynamics of genome-metabolome-phenome relationships.

1. Introduction et al., 2018) and affects their overall growth and development. Early
diagnosis of stress conditions will help to reduce overall plant damage,
Ensuring sufficient food for an increasing global population requires improve crop yields through effective protection protocols, and support
improving crop yields and optimizing the use of our natural resources the effective administration of nutrients, pesticides, and water.
(Pereira, 2017). However, climate changes, uncontrolled pests, and Traditional precision agriculture techniques rely on monitoring
improper use of pesticides limit these actions, becoming the most sig­ continuum-level atmospheric, soil, and plant variables to mitigate a
nificant causes of crop losses worldwide (Arora, 2019; Giraldo et al., posteriori the deleterious impact of stress on crops (Balasundram et al.,
2019; Jhariya et al., 2019; Sharma et al., 2017). Climate change alone 2020; Sishodia et al., 2020). Unfortunately, they do not capture the early
induces different abiotic stresses on plants (H. C. Zhang et al., 2022) metabolic response to such stresses (Kim and Lee, 2022). Existing
(drought (Farooq et al., 2012; Salehi-Lisar and Bakhshayeshan-Agdam, techniques for quantifying metabolites involve ex-vivo tissue extraction
2016), heat (Hassan et al., 2021), cold (Thakur et al., 2010), salinity and offline, offsite analysis (Jones and Kinghorn, 2012), which convey a
(Isayenkov and Maathuis, 2019), floodings (Tewari and Mishra, 2018), mere snapshot of the plant’s state when the tissue was extracted. Un­
light (Singh and Thakur, 2018; Yang et al., 2019), heavy metals (Ghori derstanding the role of metabolites in time for specific genetic makeups
et al., 2019)), which in turn alters their phenotypic expression (Singh and different phenotypic conditions would contribute significantly to

* Corresponding author. Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA, 91125, United States.
** Corresponding author.
E-mail addresses: josephwang@ucsd.edu (J. Wang), ajaramil@caltech.edu (A. Jaramillo-Botero).
1
Equal Contributions.

https://doi.org/10.1016/j.bios.2023.115300
Received 11 January 2023; Received in revised form 11 March 2023; Accepted 5 April 2023
Available online 8 April 2023
0956-5663/© 2023 Elsevier B.V. All rights reserved.
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

our understanding of plant physiology and developing new varieties altering the plant’s natural response.
with improved traits and tolerance to stress (Francini and Sebastiani, Sugars, such as glucose, fructose, and galactose, are some of the
2019; Kumar, 2020). primary metabolites that play a critical role as signaling molecules to
Recently, wearable sensors have emerged as an alternative for real- control metabolic and physiological functions required for the growth
time and in-situ plant health monitoring (Giraldo et al., 2019; Lee and development of a plant (Cho et al., 2018). Changes in environmental
et al., 2021). They are well-suited for non-invasive and continuous conditions lead to changes in these metabolic processes, hence fluctua­
detection of biomarkers, even at ultra-low concentrations, and offer a tions in sugar concentrations. Glucose, a direct byproduct of plant
wide range of applications (Kim et al., 2019; Lin et al., 2021; Xu et al., photosynthesis (Fig. 1SI), is required to perform different biological
2021; Ye et al., 2020). Furthermore, they can be manufactured to processes essential for the plant. A plant organism’s photo spectra,
possess high flexibility and stretchability while providing high selec­ resource use efficiency, and other physiological phenomena correlate
tivity and sensitivity toward the biomarker of interest, thereby offering directly with its glucose production. Unfortunately, current technologies
accurate real-time detection capabilities to favor sustainable agriculture for sugar level estimation in plants involve tissue extraction, trained
frameworks (Forzato et al., 2020; Giraldo et al., 2019; Jaramillo-Botero personnel, and highly complex and expensive instruments and pro­
et al., 2022; Li et al., 2021; Liu, 2022; Qu et al., 2021). cesses, which limits their use for on-field application (Deuschle et al.,
Up to now, novel biosensors have been proposed for the monitoring 2006; Li et al., 2018; Zhu et al., 2017). In addition, no single technology
of plant environmental conditions and growth (Lu et al., 2020; Nassar has been demonstrated, until now, for non-invasive, non-destructive,
et al., 2018; Zhao et al., 2019), plant diseases (Kim et al., 2020), pesti­ real-time, in-situ and in-vivo monitoring of glucose in plants (Diacci et al.,
cide management (Zhao et al., 2020) and plant stress (Barbosa et al., 2020, 2021; Giné Bordonaba and Terry, 2009; Wu et al., 2022; Zheng
2022; Hossain et al., 2021; Koman et al., 2017; C. H. Zhang et al., 2022). et al., 2022).
However, despite the tremendous progress in materials and techniques, Here, we report a novel non-invasive, non-destructive, and real-time
wearable sensors have yet to be demonstrated for the early detection of wearable electrochemical biosensor for in-situ and in-vivo detection of
stresses from real-time plant metabolite signatures without the need to glucose extracted from plant’s leaves through reverse iontophoresis (RI)
perform an invasive and destructive sample extraction that ends up (Fig. 1). The sensor is held on the leaf surface using a sandwich

Fig. 1. Conceptual illustration of the wearable elec­

trochemical biosensor for plant glucose monitoring.
a.) Magnetic clamp assembly on the leaf to secure the
sensor. b.) Cross-section illustration of a leaf with
clamped sensor inducing a glucose gradient by RI. c.)
Clamped sensor mounted on a leaf. d.) Sensor system
components and assembly. e.) Glucose recognition
mechanism via the GOx enzymatic reactions. f.)
Conceptual influence of light and temperature
stresses on glucose production (photosynthesis) and
its quantification by chronoamperometry. Glucose
concentration is determined by the increment of
current (red curve) with respect to the baseline (dark
curve) of the electrode.

2
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

configuration of barium ferrite magnets (with 25 mm diameter and 10 g 10 min at 85 ◦ C. Afterward, working and counter electrode patterns
weight), which facilitates its use and convenient removal (Fig. 1a). The were printed with Prussian Blue carbon ink, and curing was used under
sensor is comprised of two major parts: cathode and anode (Fig. 1b). The the same conditions.
cathode magnet consists of the measurement electrode (glucose-selec­
tive) and the negative terminal of the iontophoretic system. For its part, 2.3. Iontophoretic electrode fabrication
the anode magnet only contains the positive terminal for the ionto­
phoretic application. A negligible current density (0.2 mA/cm2) is The iontophoretic electrodes (A = 2.65 cm2) were designed on
applied between anode and cathode, through the iontophoresis elec­ AutoCAD 2018 software (Autodesk, San Rafael, CA, USA) and fabricated
trodes, to trigger the molecular flow of plant fluids from the leaves to the using the Cricut Explore Air 2 machine. First, three-layers screen
measurement electrode. Neutral glucose molecules flow from the anode printing was performed with Ag/AgCl ink on a PET substrate using the
to the cathode in the direction of the ionic current induced by the layer-by-layer technique, and curing was applied at 85 ◦ C for 10 min at
externally applied iontophoretic field. Fluid samples containing glucose the end of each layer. After all these processes, the iontophoretic elec­
are extracted within 10 min. The sandwich-type configuration allows trodes were cut by positioning the substrate in the Cricut Explore ma­
the sensor to be placed on the leaf surface without causing any damage, chine. These electrodes were then stored in a Petri dish under room
since the magnets do not have a strong enough magnetic field to exert conditions to be used both for the anode and cathode.
high pressure on it. In addition, it protects the sensing area from envi­
ronmental exposure (Fig. 1c). The overall system integrates a three- 2.4. Chemical modification
electrode screen-printed electrochemical cell configuration (with an
enzyme-functionalized working or sensing electrode, counter, and The enzyme-based modification applied to the Prussian Blue WE
reference electrode) for glucose sensing, two Ag electrodes (anode and transducer area (0.07 cm2) of the electrodes was carried out with a
cathode) for iontophoretic extraction of plant fluid, two agarose and one single-layer drop-casting method. First, stock enzyme solution was
PVA hydrogels, medical tape, and two magnetic holders (Fig. 1d). The prepared, dissolving 40 mg/mL GOx and 10 mg/mL BSA in 0.1 M PBS
agarose hydrogels serve as a protective physical barrier between the leaf (pH 7.4). Then, the drop-casting solution was formed by mixing the
and the iontophoretic electrodes to prevent any damage to the leaf. In stock enzyme stock solution, 0.1% Nafion, and Glutaraldehyde 1 wt % in
contrast, the PVA hydrogel acts as a reservoir and transport scaffold for H2O in a ratio of 1:1:0,33 respectively. Therefore, the modification
moving the extracted glucose to the sensing electrode surface. The process was completed by taking 3 μL for each WE from this mixed so­
sensing working electrode is modified with the glucose oxidase (GOx) lution. After completing all these procedures, the electrodes were stored
enzyme. GOx catalyzes glucose oxidation to gluconic acid and oxygen at 4 ◦ C overnight to allow them to dry. The dried electrodes continued to
reduction to hydrogen peroxide (H2O2). Thus, glucose can be quantified be stored under the same conditions before testing.
based on either the detection of the H2O2 produced or the oxygen
consumed. We use chronoamperometric detection of the H2O2 byprod­ 2.5. Iontophoretic hydrogel preparation
ucts (Fig. 1e). The integrated setup was used to study plant glucose
levels under different light and temperature conditions (Fig. 1f). Agarose hydrogels were created by dissolving 3% (w/v) agarose in
0.1 M PBS buffer. For this dissolution to occur, a magnetic bead was
2. Materials and methods placed in the solution, and then, this solution was left on a hot plate at
150 ◦ C with a constant rotation speed of 100 rpm until it became
2.1. Chemicals and instrumentation transparent. A mold consisting of circular discs with a diameter of 3 cm
and a thickness of 500 μm was used to give the desired shape to the
Glucose Oxidase (GOx) from Aspergillus niger, Type X–S (EC solution. Next, 600 μL of agarose gel solution was transferred for each
1.1.3.4), Nafion perfluorinated resin solution, glutaraldehyde solution circular disc to these cavities. The gels that hardened at room temper­
50%, bovine serum albumin (BSA), potassium hydroxide (KOH), su­ ature were then stored at 4 ◦ C for future use.
crose, fructose, polyvinyl alcohol (PVA -molecular weight ~ 89,000), D
(+) glucose, agarose type IV, phosphate buffer (PBS) solution (pH 7.4, 2.6. PVA gel preparation
1M) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Ag/AgCl
ink (E2141) and Prussian Blue conductive carbon (C2070424P2) were PVA porous hydrogel membrane was produced by adapting from our
purchased from Ercon Inc (Wareham, MA, USA) and Gwent Group (UK), previous studies (X. C. Chen et al., 2019; Sempionatto et al., 2021). First,
respectively. Medical tape (# 1521 - Double Sided Transparent Poly­ 0.1 g/mL PVA (molecular weight ~ 89,000) and 0.2 g/mL KOH stock
ethylene) used to assemble the designed electrochemical system was solutions were prepared with 1:10 and 1:5 wt ratios in deionized (DI)
purchased from 3M (Maplewood, MN, USA). All amperometric and water, respectively. The PVA solution prepared in a beaker was then
iontophoretic measurements were performed using Palmsens4 heated in a microwave on high power for 30 s to ensure that all the PVA
(PalmSens). powder dissolved in water to become sticky. Throughout this process,
the solution was stirred by pausing the microwave at intervals of
2.2. Three-electrode system fabrication approximately 5 s. Then, 10 g of the PVA solution was transferred to
another beaker, and 14 g of KOH solution was carefully added. Finally,
Glucose sensors were fabricated using a screen-printing process with 2.6 g of sucrose solution dissolved in 2 mL of DI water using an ultra­
a semiautomatic MMP-SPM printer (Speedline Technologies, Franklin, sonic bath sonicator (for 10 min at 27 ◦ C) was added to the same beaker
MA, USA). The metallic steel stencils compatible with this printer were to form a hydrogel precursor as a porogenic agent (Park et al., 2010).
manufactured by engraving the electrode design drawn on AutoCAD Afterward, the solution in the beaker was mixed thoroughly, and 15 g
2018 software (Autodesk, San Rafael, CA, USA) onto 12 in. × 12 in. and was taken from it to transfer to a glass Petri dish with a diameter of about
75 μm thick stencils by Metal Etch Services, San Marcos, CA, USA. Thus, 9 cm. To ensure excess water removal and cross-linking, the Petri dish
the three-electrode system consists of a working electrode (WE) with 3 was left in a vacuum desiccator overnight until the PVA compound
mm diameter, an Ag/AgCl reference electrode (RE), and a counter became two-thirds of its original weight. Then, the cross-linked porous
electrode (CE) fabricated using stencils with two different designs with PVA gel was immersed in a PBS buffer (0.1 M) to remove the sugar and
the layer-by-layer screen printing method. In this context, Ag/AgCl KOH excess and kept until neutral pH (~7.4). When the desired pH
patterns were printed on the polyethylene terephthalate (PET) substrate value was reached, the PVA gel (approximately 1 mm thick) was cut into
as interconnection and reference electrode, then curing was applied for circular discs of 1 cm diameter and stored in PBS (0.1M) solution at

3
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

room temperature to be ready for use. Before each use, the PVA gels 2.9. On-plant evaluation
removed from this PBS (0.1 M) solution were dried with tissue wipes to
remove excess water from their surfaces and then positioned on the The cathode and anode terminals of the sensor were positioned on
glucose sensors. the abaxial and adaxial planes of the leaf surface (Sánchez et al., 2017),
respectively, in a sandwich-like format via magnetic holders. This
2.7. Plant species arrangement reduces the intensity of current necessary to favor the
extraction of fluids from the plant, as there is less effective resistance
Mature (>180 days) Capsicum annuum L. (sweet pepper), Gerbera between both sensor terminals. Only one designated leaf was used in
jamesonii (gerbera), and Lactuca sativa L. (romaine lettuce) plants were each study; no experiment was performed on more than one leaf during
purchased in healthy conditions at a local greenhouse. We chose these a stress experiment. For the stress studies, light bulbs with the same
plants because they have a leaf surface area >7 cm2, which fits the characteristics as the study by (Cioć et al., 2021; Pawłowska et al., 2018)
sensing platform comfortably. (fluorescent lamp: 10 W/21397, 400 nm–700 nm, Goodwill; blue:
9W/A19, 430 nm, Gonhom; red: 9W/A19, 670 nm, Gonhom) were used
2.8. Sensor configuration, assembly, and mounting on plant leaves and placed at 30 cm from the top of the plant to ensure constant and
equally distributed irradiance (Miao et al., 2019). The temperature was
The sensor assembly is attached to a leaf surface using two comple­ controlled with an infrared thermometer (Etekcity, Lasergrip 1080). The
mentary magnets, each embedded in a plastic support to facilitate electrochemical experiments on these leaves were based on three basic
handling. First, we attached a double-sided medical tape onto each of steps: Pre-RI current measurement with chronoamperometric (CA)
the exposed magnetic surfaces (Fig. 2a). We then attached the three- scans, application of RI, and post-RI current measurement with CA
electrode glucose biosensor to one of the taped magnet surfaces, as scans. After the three-step experiment, the used biosensor was disposed
shown in Fig. 2b, and the complementary RI electrodes onto each taped off, and a new biosensor was used for the subsequent experiments. The
magnet surface, as shown in Fig. 2c (the RI electrode should not interfere CA measurements were carried out through a PalmSens4 electro­
with the electrochemical sensor’s electrodes). Next, we applied agarose chemical interface by applying a constant reduction potential of − 0.15 V
hydrogels over both iontophoretic electrode surfaces (Fig. 2d) and the (vs Ag/AgCl) for 60 s. In the first step, to stabilize the glucose sensors
PVA hydrogel over the sensing zone of the biosensor (Fig. 2e). The setup before RI application, 15 repetitive CA scans were performed until the
is then ready to be transferred onto a leaf in a sandwiched configuration, current signal change became constant; thus, the data from the last scan
with the leaf between the two magnets (Fig. 2f). It is important to note was used to compare the pre-and post-RI glucose measurement. In the
that the electrode terminals have been designed long enough so that second step, glucose extraction from the plants was carried out with the
they are exposed outside the area of the magnets and thus facilitate their RI technique by connecting the cathode and anode silver electrodes to an
connection to the potentiostat. Autolab potentiostat (PGSTAT101), performing chronopotentiometric
measurements for 10 min with 0.2 mA/cm2 current density. In the last
step, at the end of RI, five repetitive CA scans were carried out; in the
same way, the previous scan was evaluated for glucose measurement.

Fig. 2. Assembly and transfer of the reverse iontophoresis-enabled electrochemical glucose monitoring patch onto a plant leaf. a-c.) Attachment of the glucose
biosensor and iontophoretic electrodes onto the magnetic holder assembly. d-e.) Application of agarose and PVA hydrogels onto the iontophoretic electrodes and the
glucose sensor, respectively. f.) Sensor placement on the plant leaf surface, sandwiched between the magnets.

4
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

Moreover, the three-step experiment was carried out without incremental additions of both interferent molecules showed a negligible
applying RI (second step) to validate the post-RI data. After all these (decreasing) response, as depicted in Fig. 3c.
processes were completed, the sandwich-like biosensor was removed Considering the application of glucose oxidase (GOx) enzymes as a
without damaging the leaves; then, the leaves used in the experiments recognition element, we performed a temperature calibration over the
were cleaned with distilled water. Finally, the current difference (ΔI = I expected operational range of the sensor, 10–40 ◦ C (Fig. 3b). At low
pre-IR – I post-IR) was calculated with the CA data by considering the temperatures, the enzymatic activity was not altered, i.e., the rate of the
values at 60 s of the last scans to compare the before (I pre-IR) and after enzymatic reaction remains almost constant for this temperature range
(I post-IR) RI current measurements. Therefore, glucose was detected on (10–25 ◦ C) due to the enzyme stability on electrode surface provided by
the leaves based on this current difference (ΔI). glutaraldehyde and Nafion coating (Belyad et al., 2018; Sari et al., 2012;
The statistical analysis was carried out using R software (http: Sungur and Numanoğlu, 2006). In contrast, at high temperatures, an
//www.r-project.org). To explore the significance level between the expected increase in the enzymatic reaction rates with glucose leads to a
stress-induced changes in glucose concentration, we used a T-test with rise in the amperometric signal of the sensor. Given that the effect of
respect to previously obtained data at a confidence interval of 95%. temperature remains linear across the operational range, we use the
Significant codes for the two-tailed P values were related as: P > 0.05 temperature-corrected slope difference between the lines in Fig. 3b to
“NS”, P < 0.05 “*”, P < 0.01 “**”, and P < 0.001 “***”. adjust our sensor’s calibration curve by a constant factor given by the
temperature. This effect of temperature on the sensor can be observed by
3. Results studying its chemical kinetics through the Arrhenius equation, as shown
in Fig. 2SI-a. From the slopes obtained for each calibration curve at a
3.1. In-vitro electrochemical characterization given temperature (10, 25, and 40 ◦ C) and the ordinates at the origin,
Fig. 2SI-b, and 2SI-c, respectively, we derive a general
To evaluate the performance and accuracy of our glucose sensor, we temperature-adjusted glucose concentration (Gc in μM) equation in the
first assessed the sensor through an extensive in-vitro characterization. range of interest (10–40 ◦ C),
The screen-printed electrode was tested for a range of glucose concen­
Is + 0.0256 − 0.00017033 • T
trations between 20 and 80 μM in 20 μM increments using PBS (pH 7.4, Gc (T) =
0.0015 + 0.00004566 • T
0.1 M) as buffer medium (Fig. 3a). The results indicated a linear
response in current vs. concentration as depicted in Fig. 3b, with a Where Is is the sensor signal (in μA) obtained after the RI, and T is the
sensitivity of 22.7 nA/(μM⋅cm2), a limit of detection (LOD) of 9.4 μM and temperature (◦ C). The surface plot in Fig. 2SI-d depicts the temperature-
a limit of quantification (LOQ) of 28.5 μM. LOD=3.3σ /S, and LOQ=10σ / adjusted Gc as a function of T and for a resulting measured value of Is.
S, where σ is the standard deviation of the response and S is the slope of
the calibration curve shown by the dark blue line in Fig. 3b. Negative
control experiments were carried out to determine the sensor’s selec­ 3.2. In-vivo enzymatic and reverse iontophoresis evaluation
tivity to glucose, using fructose and sucrose at different concentrations.
These sugars are also vital metabolites in plants’ physiology. The Reverse iontophoresis for non-destructive metabolite extraction and

Fig. 3. In-vitro glucose sensor characterization. a.) Amperometric response of the sensor for different glucose concentrations (0–80 μM). b.) Effect of temperature on
glucose sensor calibration curves at 25 ◦ C (dark blue), 10 ◦ C (light blue), and 40 ◦ C (red) (LOD = 9.4 μM). c.) Amperometric response of the sensor to potential
interferents, such as sucrose and fructose.

5
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

identification in plants has been studied before (González-Sánchez et al.,

2015; Sánchez et al., 2017). However, the extraction time required to
obtain useable volume samples for electrochemical analysis has been
reported to be more than an hour (>1 h), which would make this
technique unattractive for sensing in plants given that the temporal
resolution of measurement must be less than the biological process of
interest (C. X. Chen et al., 2019; Hu et al., 2018; Pham et al., 2022; Wang
et al., 2019). To address this issue, an agarose hydrogel was incorpo­
rated between the iontophoresis electrodes and the plant’s leaf surface
to enhance electron transfer across the interface. In contrast, the porous
PVA hydrogel membrane improved the diffusion and absorption of
glucose into the working electrode. This sensor design was adapted
based on our previous work (De la Paz et al., 2021). The magnetic
sandwich setup maintained constant pressure on all surfaces. As a result,
the total RI time was lowered to 10 min.
To validate the in-vivo performance of the sensor as a wearable de­
vice for plants, enzymatic and RI control measurements were performed
onto the leaf of the plant (Fig. 3SI). In the first experiment, an initial
patch was assembled using a non-modified working electrode (without
GOx enzyme) and subsequently mounted on a gerbera leaf. After per­
forming an amperometric stabilization step to establish a baseline, RI
was applied for 10 min, followed by a sensing step. As shown in Fig. 3SI-
a, no glucose was detected from the leaf after the post-RI amperometry Fig. 4. Low-light stress evaluation in Gerbera. Gerbera glucose response over
stage, confirming the need for the enzymatic bioreceptor to enable the five days of darkness, watering at a rate of 400 mL/week with a constant
successful quantification of glucose. In the second experiment, a stabi­ temperature of 25 ◦ C. Statistical analysis was performed using a T-test with
lization step was applied to establish a baseline followed by a waiting respect to the previously obtained data. Significant codes: P > 0.5 “NS”, P <
time of 10 min. The current curves in the post-stabilization step showed 0.001 “***”.
negligible glucose detection, confirming the non-existing extraction of
glucose from the leaf without the RI (Fig. 3SI-b). Finally, a third patch existing glucose to perform vital biological functions (e.g., respiration).
with the enzyme-functionalized working electrode was assembled, fol­ Based on the statistical analysis performed using the T-test, there are
lowed by an amperometric stabilization measurement, and 10 min of RI. significant differences (P < 0.001) in the measured glucose concentra­
The resulting increase in the current response showed that the combi­ tion for days 1, 2, 4, 5. Interestingly, for day 3, the difference was not
nation of both enzymatic bioreceptor and RI steps enables the extraction significant (NS), and the concentration change remained almost con­
and subsequent quantification of glucose (Fig. 3SI-c). stant (within 5 μM), which provides evidence of the plant’s effort to
To gauge the effect of RI on the plant’s natural glucose production, conserve the glucose reservoirs, or that other carbohydrates are being
we performed one glucose measurement on a gerbera plant leaf under consumed to reduce the total energy consumption (Scialdone and
our ‘normal’ laboratory conditions, then placed the whole plant in a Howard, 2015). Once glucose levels are entirely depleted, after eight
dark box at the same temperature for 12 h, and then re-exposed it to the days, the plant enters an early senescence stage (Asim et al., 2022;
same ‘normal’ laboratory lighting conditions before performing three Durgud et al., 2018). For ii), the photosynthetic efficiency of three
consecutive measurements at 30-min intervals (Fig. 4SI). Our ‘normal’ different plant specimens (sweet pepper, romaine lettuce, and gerbera)
laboratory conditions in this work correspond to a white light radiated was studied under three different light wavelengths (fluorescent, blue,
from a Philips 32 W fluorescent lamp with an illuminance of 2600 lu­ and red). Again, each plant was placed inside a box without light after
mens, reference F32T8/TL941, and a temperature of 25 ± 0.5 ◦ C. We obtaining the glucose concentration baseline measured under our
found that the plant had recovered its original glucose production level ‘normal’ laboratory conditions. After 12 h in the dark, glucose concen­
after the first 30 min of exposure to light, with little change after that tration was measured, and the fluorescent light source was turned on.
(<2% deviation). We confirmed that the plant could still upregulate its After 1 h of exposure, glucose was measured, and the light source was
glucose production by exposing it to daylight and performing one final turned off. The process was repeated 12 h later to evaluate blue and red
measurement. sequentially under the same conditions. Glucose concentrations for each
plant under the described light conditions are reported in Fig. 5a.
3.3. Glucose monitoring under light stress The results showed that fluorescent light significantly affected
glucose production compared to blue and red sources (Fig. 5b–d). Given
To investigate the light effect on photosynthesis production (Rehman that fluorescent light covers the entire visible spectrum, specific fre­
et al., 2017), we monitored glucose production under two stress sce­ quency bands of photosynthetic pigments can receive more energy (Luo
narios: i) no light for five days and ii) 1-h exposure to three different et al., 2020). Our statistical analysis shows significant differences (P <
radiation wavelengths (fluorescent or FL, blue or B, and red or R). For i), 0.001) in the increase of glucose concentration caused by each exposure
the gerbera plant was placed inside a box (with no light) after measuring to light in the three plant models. Each glucose measurement performed
the glucose concentration under our ‘normal’ laboratory conditions (day without light, dark 1, dark 2, and dark 3 bars in Fig. 5b–d, displayed a
0). Glucose measurements were performed each day at the same time different sugar level after the light stimulus, which is confirmed by the
(18:00 PT), the plant was watered with 400 mL/week, and the tem­ significant differences (P < 0.001) in glucose concentration obtained
perature inside the box was controlled at a constant 25 ◦ C. Gerbera’s with the T-test; for this reason, each dark measurement was taken as a
glucose response during the 5-day darkness period is presented in Fig. 4. reference to calculate the overall glucose concentration increment or
As observed in Fig. 4, after exposing the plant to complete darkness photosynthetic rate caused by each light effect (Fig. 5e). With fluores­
for 120 h (day 5), its glucose concentration was reduced by 85% of its cent light exposure, sweet pepper and gerbera showed a higher photo­
initial value (day 0). This behavior was expected, considering the synthetic rate (>75%) than romaine lettuce, assuming glucose
absence of light hinders the synthesis of glucose molecules (Poór et al., production and photosynthetic efficiency are proportional. This indi­
2018; Seluzicki et al., 2017) and induces the gradual consumption of cated a more sensitive photosynthetic system over full spectra (Cioć

6
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

Fig. 5. Low-light stress analysis. a.) Implemented

protocol for glucose measurement under low-light
stress for each plant. Glucose is measured before
and after each light stimulus. Light effect on the
photosynthetic response is evaluated for b.) sweet
pepper, c.) gerbera and d.) romaine lettuce. e.) The
calculated rate of photosynthesis for sweet pepper,
gerbera, and romaine lettuce for each light wave­
length source. Statistical analysis was performed
using a T-test with respect to the previously obtained
data. Significant codes: P < 0.05 “*”, P < 0.01 “**”, P
< 0.001 “***”.

Fig. 6. Temperature stress response for our three plant models. a.) Schematic protocol used in this experiment. Glucose response under temperature stress for b.)
sweet pepper. c.) romaine lettuce, and d.) gerbera. Statistical analysis was performed using a T-test with respect to the previously obtained data. Significant codes: P
< 0.01 “**”, P < 0.001 “***”.

7
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

et al., 2021; Massa et al., 2015). In the case of blue and red lights, the Kim, 2021; Zhu, 2016). Different studies have proven the critical role of
highest photosynthesis rate obtained was for gerbera (Meng et al., 2019) primary (Saddhe et al., 2021) and secondary (Banerjee and Roychoud­
and sweet pepper, respectively. While romaine lettuce expressed minor hury, 2019; Šamec et al., 2021; Sharma et al., 2019) metabolites in the
differences between red and blue lights, having a slightly higher regulatory events in response to specific stresses and other vital bio­
photosynthetic rate with the latter (Johkan et al., 2010; Muneer et al., logical processes that lead to growth and development.
2014). Here, we introduced a novel leaf-based wearable electrochemical
biosensor for non-invasive, non-destructive, real time, in-situ and in-vivo
3.4. Glucose monitoring under low-and-high temperature stress glucose monitoring in plants. The biosensor provides a sensitivity of
22.7 nA/(μM⋅cm2) and a LOD of 9.4 μM. The combination of an enzy­
Climate change, including global warming, hurts agricultural pro­ matic bioreceptor (GOx) and a low reduction potential step (− 0.15 V)
duction. All biological and chemical processes involved in plant growth enabled higher selectivity toward glucose detection and negligible
and development depend on temperature (Kai and Iba, 2014; response to interferent molecules, such as sucrose and fructose. Three
Źróbek-Sokolnik, 2012). Therefore, uncontrolled temperature changes key innovations are contributed in this work, namely: i) non-invasive
can directly affect a plant’s health. To evaluate the effect of temperature analyte detection capabilities via short 10 min RI to extract the inter­
on our three plant models, the glucose production at high (40 ◦ C) and stitial fluid from plant leaves, ii) a wearable technology designed for
low (10 ◦ C) temperatures were further investigated (Fig. 6a). easy placement and removal without harming the plant, and iii) in-situ
Each plant was placed inside a dark room, with controlled environ­ and in-vivo quantification of glucose, which can be adapted to fit a wide
mental temperature, and watered regularly (400 mL/week). The starting variety of other analytes or plants for in-field measurements. RI allows
temperature was fixed to 25 ◦ C. After 12 h, the glucose concentration the extraction of plant fluids that contain important metabolites of in­
was measured, followed by an increase in the temperature to 40 ◦ C. After terest in plant physiology (Sánchez et al., 2017). Previous works had
45 min of exposure, the glucose was measured (under the same settings). demonstrated the stress effect on glucose production (Ahmad, 2019;
After concluding the sensing step, the plant was left at room temperature Martínez-Noël and Tognetti, 2018; Saddhe et al., 2021; Sami et al.,
(25 ◦ C). To avoid cumulative stress on the plant, a 24-h (one-day) re­ 2016) over relatively short exposures, <1 h. Here we reduce the sam­
covery step was applied to allow the plant specimen to recover under the pling period to <10 min, thus increasing resolution. The sandwich-based
basal conditions. The same procedure was repeated with a lower tem­ configuration evaluated three different plant models over different light
perature (10 ◦ C), having measured the glucose production at 25 ◦ C and temperature scenarios. Light stress studies revealed that white
before the temperature treatment. Fig. 6 shows the glucose response for fluorescent sources lead to the highest photosynthetic activity on all the
each plant under the described temperature stress scenario. Glucose plant models used in this study. Under prolonged periods of deprived
values were corrected using the results from our in-vitro calibration light, plants appear to regulate glucose consumption to avoid rapid
curves at each specific temperature (Fig. 3b). depletion of the available reservoirs. High temperatures (40 ◦ C) cause a
Plants affected by high-temperature stress induce stomatal closure to fast reduction of glucose levels, which is directly associated with leaf
avoid extreme water loss. Consequently, the photosynthetic rate is water loss and stomatal closure. In contrast, cold environments lead to
reduced (Sami et al., 2016) due to ribulose 1,5-bisphosphate carboxylase glucose accumulation, aimed at protecting cells and replenishing
(Rubisco) inhibition (Bita and Gerats, 2013) and intracellular CO2 glucose reserves.
reduction. Leaf water loss caused by heat stress reduces intracellular The key benefits of our proposed sensor are compared in Table II
chlorophyll and carbohydrate content (Hasanuzzaman et al., 2013). against other glucose-selective plant biosensors.
Therefore, each plant specimen in Fig. 6b–d showed a glucose concen­ Our wearable technology can contribute to elucidating vital regula­
tration decrease, with high significant differences (P < 0.001), after tory mechanisms that govern plant adaptability and tolerance to
exposure to higher temperatures (40 ◦ C). From Table 1, it can be different biotic and abiotic conditions without the need for complex and
deduced that gerbera has the highest tolerance to temperature changes, expensive protocols and instruments for on-field measurements. It can
which is also corroborated by the lowest percentage of changes depicted also improve our existing crop breeding optimization strategies based on
in Fig. 6d. On the other hand, cold-temperature stress induces sugar high-resolution phenotyping. It is a proof-of-concept technology that
accumulation in plants (Aslam et al., 2022; Martínez-Noël and Tognetti, can be extended to quantify other metabolites. Further work needs to be
2018). To ensure that all vital biological processes continue working at done, beyond the scope of this manuscript, to determine the effect of RI
low temperatures, sugars act as osmoprotectants (Sami et al., 2016; on the analyte’s bioavailability and dynamic concentration in the plant.
Sharma et al., 2020) to keep cells from damage and provide acclimati­
zation through lipid bilayer contact (Gangola and Ramadoss, 2018). 5. Conclusions
From Fig. 6b and d, at low temperatures (10 ◦ C), an increase in glucose
concentration, with high significant differences (P < 0.001) and (P < We have demonstrated a novel non-invasive, real-time, in-situ and in-
0.01), was observed for sweet pepper and gerbera plants but not for vivo approach for glucose monitoring in plants. The integration of a
romaine lettuce (Fig. 6c). This was consistent with romaine lettuce, glucose-selective sensor with the reverse iontophoresis method allowed
having the highest tolerance to low temperature, as shown in Table 1. us to perform a clean and fast (<10 min) extraction of the samples for
analysis without the need to cause any damage to the plant. Two mag­
4. Discussion netic holders made up of measurement and iontophoresis electrodes,
allow easy positioning and removal of the sensor on the leaves, and at
Plants’ response to biotic and abiotic stresses involves numerous the same time protect the sensing area from any environmental expo­
genetically-encoded mechanisms that remain unclear to date (Naing and sure. The sensing performance of the device was tested by monitoring
glucose concentration in three different plant species subjected to light
Table 1 and temperature stresses. The obtained results allowed us to charac­
Temperature tolerance for the studied plant species. terize for the first time, quantitatively and directly, the behavior of this
sugar on plants subjected to various stress conditions. Since plant fluids
Plant Specie Temperature (◦ C)
are enriched with a vast amount of primary and secondary metabolites,
Minimum Optimal Maximum
the proposed sensor concept can be easily adapted to monitor any other
Sweet Pepper (Saha et al., 2010) 18 25 32 analyte of interest simply by changing the selectivity of the working
Gerbera (Li et al., 2019) 16 21 35 electrode. Future improvements to the sensor include the inclusion of an
Romaine Lettuce (Dufault et al., 2009) 8 20 25
integrated electronic platform that reduces the size factor of the

8
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

Table II
Advantages of the proposed sensor compared with other glucose-selective plant biosensors.
Glucose Detection Our Work Diacci et al. Wu et al. (2022) Giné Bordonaba and Terry Zheng et al. Diacci et al. (2020)
(2021) (2009) (2022)

in-vivo (on plant YES YES NO NO YES NO

evaluation)
Non-invasive sample YES NO NO NO NO NO
extraction
Sensor placement and Magnetic Stem wound NO NO Microneedle NO
removal patch
1
LOD 9.4 μM NA 13 μMol/L 0.01 mg glucose g− FW NA NA
Sensitivity 22.7 nA/(μM⋅cm2) NA 11.64 μA L/(mmol⋅cm2) NA 1.22 nA/mM NA
Measurement time <10 min <10 min <1 min <1 min <1 min <1 min
Requires trained personal NO YES YES YES NO YES
Requires specialized NO NO YES YES NO YES
laboratories
Negative controls to other YES (fructose, YES (sucrose) YES (ascorbic acid, YES (phenolics, organic NO YES (enzyme
sugars sucrose) fructose) acids) omitted)
Portable YES YES NO NO YES NO
Detection mechanism Enzymatic Enzymatic Enzymatic Enzymatic Enzymatic Enzymatic
Measurement method CA OECT DPV CA CA OECT
Live plant stress evaluation Light and NO NO NO NO NO
temperature

NA: not available, CA: chronoamperometry, DPV: differential pulse voltammetry OECT: organic electrochemical transistor.

measurement transducer and at the same time the exploration of new J.W. acknowledges support from the UCSD Center of Wearable Sensors
materials that allow for increasing the flexibility and long-term stability (CWS). R.D.C. acknowledges support from “Plan Propio de Investigación
of the system. de la Universidad de Córdoba 2021”.

CRediT authorship contribution statement Appendix A. Supplementary data

Sammy A. Perdomo: Formal analysis, Writing – original draft, Supplementary data to this article can be found online at https://doi.
designed the experiments, conducted the experiments, analyzed and org/10.1016/j.bios.2023.115300.
interpreted the result, co-wrote the manuscript. Ernesto De la Paz:
Formal analysis, Writing – original draft, designed the experiments, References
conducted the experiments, analyzed and interpreted the result, co-
Ahmad, I.Z., 2019. Role of sugars in abiotic stress signaling in plants. In: Plant Signaling
wrote the manuscript. Rafael Del Caño: Formal analysis, Writing –
Molecules. Woodhead Publishing, pp. 207–217.
original draft, designed the experiments, conducted the experiments, Arora, N.K., 2019. Environmental Sustainability 2, 95–96.
analyzed and interpreted the result, co-wrote the manuscript. Sumeyye Asim, M., Zhang, Y., Sun, Y., Guo, M., Khan, R., Wang, X.L., Hussain, Q., Shi, Y., 2022.
Seker: Formal analysis, Writing – original draft, conducted the experi­ Crit Rev Biotechnol 1–19.
Aslam, M., Fakher, B., Ashraf, M.A., Cheng, Y., Wang, B., Qin, Y., 2022. Agronomy 12,
ments, co-wrote the manuscript. Tamoghna Saha: Formal analysis, 702.
Writing – original draft, analyzed and interpreted the result, co-wrote Balasundram, Siva K., Golhani, K., Shamshiri, R.R., Vadamalai, Ganesan, 2020. Precision
the manuscript, All authors discussed the results and revised the Agriculture Technologies for Management of Plant Diseases. Springer, Cham,
pp. 259–278.
manuscript. Joseph Wang: Supervision, conceived and supervised the Banerjee, A., Roychoudhury, A., 2019. The regulatory signaling of gibberellin
research. Andres Jaramillo-Botero: Supervision, Formal analysis, metabolism and its crosstalk with phytohormones in response to plant abiotic
Writing – original draft, conceived and supervised the research, stresses. In: Plant Signaling Molecules. Woodhead Publishing, pp. 333–339.
Barbosa, J.A., Freitas, V.M.S., Vidotto, L.H.B., Schleder, G.R., de Oliveira, R.A.G., da
designed the experiments, analyzed and interpreted the result, co-wrote Rocha, J.F., Kubota, L.T., Vieira, L.C.S., Tolentino, H.C.N., Neckel, I.T., Gobbi, A.L.,
the manuscript. Santhiago, M., Lima, R.S., 2022. ACS Appl. Mater. Interfaces 14, 22989–23001.
Belyad, F., Karkhanei, A.A., Raheb, J., 2018. EuPA Open Proteom 19, 1–5.
Bita, C.E., Gerats, T., 2013. Front. Plant Sci. 4, 273.
Declaration of competing interest Chen, C., Feng, S., Zhou, M., Ji, C., Que, L., Wang, W., 2019. Biosens. Bioelectron. 140,
111342.
Chen, X., Yin, Lu Lv, Jian, Gross, Andrew, J., Le, Minh, Gutierrez, Nathaniel Georg,
The authors declare that they have no known competing financial Li, Yang, Jeerapan, Itthipon, Giroud, Fabien, Berezovska, Anastasiia, O’Reilly, R.K.,
interests or personal relationships that could have appeared to influence Xu, Sheng, Cosnier, Serge, Wang, Joseph, 2019. Adv. Funct. Mater. 29, 1905785.
the work reported in this paper Cho, L.-H., Pasriga, R., Yoon, J., Jeon, J.-S., An, G., 2018. J. Plant Biol. 61, 121–130.
Cioć, M., Tokarz, K., Dziurka, M., Pawłowska, B., 2021. Biology 10, 1035.
De la Paz, E., Barfidokht, A., Rios, S., Brown, C., Chao, E., Wang, J., 2021. Anal. Chem.
Data availability 93, 12767–12775.
Deuschle, K., Chaudhuri, B., Okumoto, S., Lager, I., Lalonde, S., Frommer, W.B., 2006.
Plant Cell 18, 2314–2325.
Data will be made available on request.
Diacci, C., Lee, J.W., Janson, P., Dufil, G., Méhes, G., Berggren, M., Simon, D.T.,
Stavrinidou, E., 2020. Adv Mater Technol 5, 1900262.
Acknowledgments Diacci, C., Abedi, T., Lee, J.W., Gabrielsson, E.O., Berggren, M., Simon, D.T., Niittylä, T.,
Stavrinidou, E., 2021. iScience 24, 101966.
Dufault, R.J., Ward, B., Hassell, R.L., 2009. Sci. Hortic. 120, 452–459.
Partial support was provided by the OMICAS Alliance (Optimización Durgud, M., Gupta, S., Ivanov, I., Omidbakhshfard, M.A., Benina, M., Alseekh, S.,
Multiescala In-silico de Cultivos Agrícolas Sostenibles: Infraestructura y Staykov, N., Hauenstein, M., Dijkwel, P.P., Hörtensteiner, S., Toneva, V.,
Brotman, Y., Fernie, A.R., Mueller-Roeber, B., Gechev, T.S., 2018. Plant Physiol. 177,
validación en Arroz y Caña de Azúcar), sponsored within the Colombian
1319–1338.
Scientific Ecosystem by The World Bank, the Colombian Ministry of Farooq, M., Hussain, M., Wahid, A., Siddique, K.H.M., 2012. Drought stress in plants: an
Science, Technology and Innovation, the Colombian Ministry of Edu­ overview. In: Plant Responses to Drought Stress. Springer Berlin Heidelberg, Berlin,
Heidelberg, pp. 1–33.
cation, the Colombian Ministry of Industry and Tourism, and ICETEX
Forzato, C., Vida, V., Berti, F., 2020. Biosensors 10, 105.
under GRANT ID: FP44842–217–2018, OMICAS Award ID: 792-61187.

9
S.A. Perdomo et al. Biosensors and Bioelectronics 231 (2023) 115300

Francini, A., Sebastiani, L., 2019. Horticulturae 5, 67–70. Pereira, L.S., 2017. Water Resour. Manag. 31, 2985–2999.
Gangola, M.P., Ramadoss, B.R., 2018. Sugars play a critical role in abiotic stress Pham, J., Forget, A., Bridonneau, N., Mattana, G., Stavrinidou, E., Zrig, S., Piro, B.,
tolerance in plants. In: Biochemical, Physiological and Molecular Avenues for Noel, V., 2022. Electrochem. Commun. 137, 107270.
Combating Abiotic Stress Tolerance in Plants. Academic Press, pp. 17–38. Poór, P., Takács, Z., Patyi, G., Borbély, P., Bencsik, O., Szekeres, A., Tari, I., 2018. South
Ghori, N.-H., Ghori, T., Hayat, M.Q., Imadi, S.R., Gul, A., Altay, V., Ozturk, M., 2019. Int. Afr. J. Bot. 118, 98–104.
J. Environ. Sci. Technol. 16, 1807–1828. Qu, C., Sun, X., Sun, W., Cao, L., Wang, X., He, Z., 2021. Small 17, 2104482.
Giné Bordonaba, J., Terry, L.A., 2009. J. Agric. Food Chem. 57, 8220–8226. Rehman, M., Ullah, S., Bao, Y., Wang, B., Peng, D., Liu, L., 2017. Environ. Sci. Pollut. Res.
Giraldo, J.P., Wu, H., Newkirk, G.M., Kruss, S., 2019. Nat. Nanotechnol. 14, 541–553. 24, 24743–24752.
González-Sánchez, M.I., Lee, P.T., Guy, R.H., Compton, R.G., 2015. Chem. Commun. 51, Saddhe, A.A., Manuka, R., Penna, S., 2021. Physiol. Plantarum 171, 739–755.
16534–16536. Saha, S.R., Hossain, M.M., Rahman, M.M., Kuo, C.G., Abdullah, S., 2010. Bangladesh J.
Hasanuzzaman, M., Nahar, K., Alam, M.M., Roychowdhury, R., Fujita, M., 2013. Int. J. Agric. Res. 35, 525–534.
Mol. Sci. 14, 9643–9684. Salehi-Lisar, S.Y., Bakhshayeshan-Agdam, H., 2016. Drought stress in plants: causes,
Hassan, M.U., Chattha, M.U., Khan, I., Chattha, M.B., Barbanti, L., Aamer, M., Iqbal, M. consequences, and tolerance. Drought Stress Tolerance in Plants 1, 1–16. Springer
M., Nawaz, M., Mahmood, A., Ali, A., Aslam, M.T., 2021. Plant Biosystems - An International Publishing, Cham.
International Journal Dealing with all Aspects of Plant Biology 155, 211–234. Šamec, D., Karalija, E., Šola, I., Vujčić Bok, V., Salopek-Sondi, B., 2021. Plants 10, 118.
Hossain, N.I., Noushin, T., Tabassum, S., 2021. Leaf-FIT: a wearable leaf sensor for in-situ Sami, F., Yusuf, M., Faizan, M., Faraz, A., Hayat, S., 2016. Plant Physiol. Biochem. 109,
and real-time monitoring of plant phytohormones. In: IEEE Sensors. IEEE, pp. 1–4. 54–61.
Hu, Y., Zhao, J., Li, H., Wang, X., Hou, P., Wang, C., Li, A., Chen, L., 2018. RSC Adv. 8, Sánchez, M.I.G., McCullagh, J., Guy, R.H., Compton, R.G., 2017. Phytochem. Anal. 28,
23404–23410. 195–201.
Isayenkov, S.V., Maathuis, F.J.M., 2019. Front. Plant Sci. 10, 1–11. Sari, N., Antepli, E., Nartop, D., Yetim, N.K., 2012. Int. J. Mol. Sci. 13, 11870–11880.
Jaramillo-Botero, A., Colorado, J., Quimbaya, M., Rebolledo, M.C., Lorieux, M., Ghneim- Scialdone, A., Howard, M., 2015. Front. Plant Sci. 6, 1–7.
Herrera, T., Arango, C.A., Tobón, L.E., Finke, J., Rocha, C., Muñoz, F., Riascos, J.J., Seluzicki, A., Burko, Y., Chory, J., 2017. Plant Cell Environ. 40, 2487–2501.
Silva, F., Chirinda, N., Caccamo, M., Vandepoele, K., Goddard, W.A., 2022. Front. Sempionatto, J.R., Moon, J.-M., Wang, J., 2021. ACS Sens. 6, 1875–1883.
Plant Sci. 13, 1–20. Sharma, S., Kooner, R., Arora, R., 2017. Insect pests and crop losses. In: Breeding Insect
Jhariya, M.K., Banerjee, A., Meena, R.S., Yadav, D.K. (Eds.), 2019. Sustainable Resistant Crops for Sustainable Agriculture. Springer Singapore, Singapore,
Agriculture, Forest and Environmental Management, Sustainable Agriculture, Forest pp. 45–66.
and Environmental Management. Springer Singapore, Singapore. Sharma, A., Shahzad, B., Kumar, V., Kohli, S.K., Sidhu, G.P.S., Bali, A.S., Handa, N.,
Johkan, M., Shoji, K., Goto, F., Hashida, S., Yoshihara, T., 2010. Hortscience 45, Kapoor, D., Bhardwaj, R., Zheng, B., 2019. Biomolecules 9, 285.
1809–1814. Sharma, P., Arora, P., Kapoor, D., Khanna, K., Atri, P., Bali, R.S., Kaur, R., Walia, A.,
Jones, W.P., Kinghorn, A.D., 2012. Extraction of plant secondary metabolites. In: Bhardwaj, R., 2020. The role of sugars in improving plant abiotic stress tolerance. In:
Methods in Molecular Biology. Humana Press, pp. 341–366. Improving Abiotic Stress Tolerance in Plants. CRC Press, pp. 31–48.
Kai, H., Iba, K., 2014. Temperature stress in plants. In: ELS. Wiley & Sons. Singh, J., Thakur, J.K., 2018. Photosynthesis and abiotic stress in plants. In: Biotic and
Kim, M.-Y., Lee, K.H., 2022. Front. Chem. 10, 1–14. Abiotic Stress Tolerance in Plants. Springer Singapore, Singapore, pp. 27–46.
Kim, J., Campbell, A.S., de Ávila, B.E.-F., Wang, J., 2019. Nat. Biotechnol. 37, 389–406. Singh, B., Mishra, S., Bohra, A., Joshi, R., Siddique, K.H.M., 2018. Crop phenomics for
Kim, J.J., Fan, R., Allison, L.K., Andrew, T.L., 2020. Sci. Adv. 6, 3296–3307. abiotic stress tolerance in crop plants. In: Biochemical, Physiological and Molecular
Koman, V.B., Lew, T.T.S., Wong, M.H., Kwak, S.Y., Giraldo, J.P., Strano, M.S., 2017. Lab Avenues for Combating Abiotic Stress Tolerance in Plants. Elsevier, pp. 277–296.
Chip 17, 4015–4024. Sishodia, R.P., Ray, R.L., Singh, S.K., 2020. Remote Sens (Basel) 12, 3136–3166.
Kumar, S., 2020. International Journal of Food Science and Agriculture 4, 367–378. Sungur, S., Numanoğlu, Y., 2006. Artificial cells. Blood Substitutes, and Biotechnology
Lee, G., Wei, Q., Zhu, Y., 2021. Adv. Funct. Mater. 31, 1–14. 34, 41–54.
Li, J., Wu, H., Santana, I., Fahlgren, M., Giraldo, J.P., 2018. ACS Appl. Mater. Interfaces Tewari, S., Mishra, A., 2018. Flooding stress in plants and approaches to overcome. In:
10, 28279–28289. Plant Metabolites and Regulation under Environmental Stress. Elsevier,
Li, F., Li, S., Shan, Q., 2019. Hortscience 54, 1164–1167. pp. 355–366.
Li, Z., Zhou, J., Dong, T., Xu, Y., Shang, Y., 2021. Biosens. Bioelectron. 182, 113105. Thakur, P., Kumar, S., Malik, J.A., Berger, J.D., Nayyar, H., 2010. Environ. Exp. Bot. 67,
Lin, Y., Bariya, M., Javey, A., 2021. Adv. Funct. Mater. 31, 2008087. 429–443.
Liu, J., 2022. Smart-agriculture: wearable devices for plant protection. In: Wearable Wang, H.R., Bi, X.M., Fang, Z.J., Yang, H., Gu, H.Y., Sun, L.J., Bao, N., 2019. Sensor.
Physical, Chemical and Biological Sensors. Elsevier, pp. 275–294. Actuator. B Chem. 286, 104–110.
Lu, Y., Xu, K., Zhang, L., Deguchi, M., Shishido, H., Arie, T., Pan, R., Hayashi, A., Wu, B., Xu, H., Shi, Y., Yao, Z., Yu, J., Zhou, H., Li, Y., Chen, Q., Long, Y., 2022. Food
Shen, L., Akita, S., Takei, K., 2020. ACS Nano 14, 10966–10975. Qual Saf 6, 1–11.
Luo, Y., Zhao, S., Tang, J., Zhu, H., Wei, H., Cui, W., Wang, M., Guo, P., 2020. J. Plant Xu, J., Fang, Y., Chen, J., 2021. Biosensors 11, 1–21.
Nutr. 43, 2876–2884. Yang, B., Tang, J., Yu, Z., Khare, T., Srivastav, A., Datir, S., Kumar, V., 2019. J. Plant
Martínez-Noël, G.M.A., Tognetti, J.A., 2018. Sugar signaling under abiotic stress in Growth Regul. 38, 1489–1506.
plants. In: Plant Metabolites and Regulation under Environmental Stress. Academic Ye, S., Feng, S., Huang, L., Bian, S., 2020. Biosensors 10, 1–34.
Press, pp. 397–406. Zhang, C., Zhang, Chi, Wu, X., Ping, J., Ying, Y., 2022. npj Flex Electron 6, 43.
Massa, G., Graham, T., Haire, T., Flemming, C., Newsham, G., Wheeler, R., 2015. Zhang, H., Zhu, J., Gong, Z., Zhu, J.-K., 2022. Nat. Rev. Genet. 23, 104–119.
Hortscience 50, 501–506. Zhao, Y., Gao, S., Zhu, J., Li, J., Xu, H., Xu, K., Cheng, H., Huang, X., 2019. ACS Omega 4,
Meng, X., Wang, Z., He, S., Shi, L., Song, Y., Lou, X., He, D., 2019. Horticultural Science 9522–9530.
and Technology 37, 473–489. Zhao, F., He, J., Li, X., Bai, Y., Ying, Y., Ping, J., 2020. Biosens. Bioelectron. 170, 112636.
Miao, Y., Chen, Q., Qu, M., Gao, L., Hou, L., 2019. Sci. Hortic. 257, 108680. Zheng, L., Zhu, D., Wang, W., Liu, J., Thng, S.T.G., Chen, P., 2022. Sensor. Actuator. B
Muneer, S., Kim, E., Park, J., Lee, J., 2014. Int. J. Mol. Sci. 15, 4657–4670. Chem. 372, 132626.
Naing, A.H., Kim, C.K., 2021. Physiol. Plantarum 172, 1711–1723. Zhu, J.K., 2016. Cell 167, 313–324.
Nassar, J.M., Khan, S.M., Villalva, D.R., Nour, M.M., Almuslem, A.S., Hussain, M.M., Zhu, Q., Wang, L., Dong, Q., Chang, S., Wen, K., Jia, S., Chu, Z., Wang, H., Gao, P.,
2018. npj Flex Electron 2, 24. Zhao, H., Han, S., Wang, Y., 2017. J. Plant Physiol. 215, 65–72.
Park, J.H., Chung, B.G., Lee, W.G., Kim, J., Brigham, M.D., Shim, J., Lee, S., Hwang, C. Źróbek-Sokolnik, A., 2012. Temperature stress and responses of plants. In:
M., Durmus, N.G., Demirci, U., Khademhosseini, A., 2010. Biotechnol. Bioeng. 106, Environmental Adaptations and Stress Tolerance of Plants in the Era of Climate
138–148. Change. Springer New York, New York, NY, pp. 113–134.
Pawłowska, B., Żupnik, M., Szewczyk-Taranek, B., Cioć, M., 2018. Hortic Environ
Biotechnol 59, 115–123.

10