Parallel, Continuous Monitoring and Quantification of Programmed Cell Death in Plant Tissue

Abstract Accurate quantification of hypersensitive response (HR) programmed cell death is imperative for understanding plant defense mechanisms and developing disease‐resistant crop varieties. Here, a phenotyping platform for rapid, continuous‐time, and quantitative assessment of HR is demonstrated: Parallel Automated Spectroscopy Tool for Electrolyte Leakage (PASTEL). Compared to traditional HR assays, PASTEL significantly improves temporal resolution and has high sensitivity, facilitating detection of microscopic levels of cell death. Validation is performed by transiently expressing the effector protein AVRblb2 in transgenic Nicotiana benthamiana (expressing the corresponding resistance protein Rpi‐blb2) to reliably induce HR. Detection of cell death is achieved at microscopic intensities, where leaf tissue appears healthy to the naked eye one week after infiltration. PASTEL produces large amounts of frequency domain impedance data captured continuously. This data is used to develop supervised machine‐learning (ML) models for classification of HR. Input data (inclusive of the entire tested concentration range) is classified as HR‐positive or negative with 84.1% mean accuracy (F1 score = 0.75) at 1 h and with 87.8% mean accuracy (F1 score = 0.81) at 22 h. With PASTEL and the ML models produced in this work, it is possible to phenotype disease resistance in plants in hours instead of days to weeks.


Parallel, Continuous Monitoring and Quantification of Programmed Cell Death in Plant Tissue
Alexander Silva Pinto Collins,Hasan Kurt,Cian Duggan,Yasin Cotur,Philip Coatsworth,Atharv Naik,Matti Kaisti,Tolga Bozkurt,Firat Güder* Figure S1.Gain factor and Phase Angle System Calibration: a) Correlation plot of measured and actual impedance, measuring resistances in the range 6.6-220 kΩ using the PASTEL system.Impedance calculated using constant gain factor calculated as described in Equation S2 and Equation S3. b) Correlation plot of measured and actual impedance, measuring resistances in the range 6.6-220 kΩ using the PASTEL system.Impedance calculated using 4 th order polynomial mapping function (Equation S4), derived from application of a fitting function performed on data obtained in a).c) Correlation plot of system phase angle against raw magnitude measured, measuring resistances in the range 6.6-220 kΩ using the PASTEL system.d) Correlation plot of measured phase angle against raw magnitude measured, measuring resistances in the range 6.6-220 kΩ using the PASTEL system.Measured Phase angle calculated as described in Equation S5 using a power mapping function derived from application of a fitting function performed on data obtained in c).where  = scale factor,  = conductivity and   = raw magnitude measured by PASTEL.
Constant scale factor can be used at for measurements performed with 10 kHz excitation frequency as system measurements are linear throughout the relevant impedance range at this frequency (Figure S2a).Equation S5: Phase Angle calibration 2 Ø =  −  where Ø is the phase of the unknown impedance,  is the raw phase angle and  is the system phase angle ∇(, ) =       +       where ∇ is the system phase angle,  is the raw magnitude, and   ,   ,   ,   are frequency dependent constants.

Figure S2 .
Figure S2.System Characterization with Potassium Chloride (KCl): a) Impedance measured with PASTEL using range of KCl concentrations (0 -1.0 M) against excitation frequency Data captured at 20 minutes.Data represented as µ ± , with n = 3 independent samples b) Phase angle measured with PASTEL using range of KCl concentrations (0 -1.0 M) against excitation frequency Data captured at 20 minutes.Data represented as µ ± , with n = 3 independent samples.

Figure S3 :
Figure S3: Detailed experimental procedure.1) Preparation of test suspensions for eliciting differing intensities of HR.All suspension have an OD600 = 0.1 to control for any plant response caused by the bacteria itself.2) Infiltration diagram for a single experiment, using two leaves from a single plant.3) Excision of leaf discs.4) Wash step to remove compounds released by excision.5) Transfer of leaf discs to measurement setup.Each well contains the leaf discs from one infiltration patch.

Figure S4 .
Figure S4.Visual HR Symptoms at 1-4 and 7 days after infiltration.Images of leaves agroinfiltrated with an empty vector-carrying bacterial suspension (negative control) and AVRblb2-carrying bacterial suspensions of loading ratios in the range LR= 0.005 -1.0.

Figure S5 .
Figure S5.Flowchart outlining data pre-preprocessing steps and development of machine learning models for classification of HR

Figure S7 .
Figure S7.System electronics schematic of microcontroller, multiplexer and custom printed circuit board.
Propidium Iodide fluorescence imaging.Each datapoint corresponds to an individual z-stack of images on the x-y plane.Quantification performed by projecting maximum intensity across the Z-stack and calculating the percentage of pixels above a fixed threshold intensity.a) One-way ANOVA for 0 hour (24h post infiltration) groups b) Post-hoc Dunnett's test for 0 hour (24h post infiltration) groups c) One-way ANOVA for 24 hour (48h post infiltration) groups d) Post-hoc Dunnett's test for 24 hour (48h post infiltration) groups.(a) One-way ANOVA : 0 hours (24h post infiltration)