Rapid Detection and Quantification of Plant Innate Immunity Response Using Raman Spectroscopy

Abstract

We have developed a rapid Raman spectroscopy-based method for the detection and quantification of early innate immunity responses in Arabidopsis and Choy Sum plants. Arabidopsis plants challenged with flg22 and elf18 elicitors could be differentiated from mock-treated plants by their Raman spectral fingerprints. From the difference Raman spectrum and the value of p at each Raman shift, we derived the Elicitor Response Index (ERI) as a quantitative measure of the response whereby a higher ERI value indicates a more significant elicitor-induced immune response. Among various Raman spectral bands contributing toward the ERI value, the most significant changes were observed in those associated with carotenoids and proteins. To validate these results, we investigated several characterized Arabidopsis pattern-triggered immunity (PTI) mutants. Compared to wild type (WT), positive regulatory mutants had ERI values close to zero, whereas negative regulatory mutants at early time points had higher ERI values. Similar to elicitor treatments, we derived an analogous Infection Response Index (IRI) as a quantitative measure to detect the early PTI response in Arabidopsis and Choy Sum plants infected with bacterial pathogens. The Raman spectral bands contributing toward a high IRI value were largely identical to the ERI Raman spectral bands. Raman spectroscopy is a convenient tool for rapid screening for Arabidopsis PTI mutants and may be suitable for the noninvasive and early diagnosis of pathogen-infected crop plants.

Keywords: Arabidopsis; Raman spectroscopy; carotenoids; elf18; flg22; pathogen-associated molecular pattern (PAMP); plant innate immunity (PTI).

Figure 1

Experimental procedures for the Raman spectroscopic analysis of elicitor-treated plants. (a) A 5-week-old Arabidopsis thaliana plant with leaves numbered according to the developmental stage. Leaf 1 is the first true leaf that started to emerge after the cotyledons. Scale bar = 2 cm. (b) The abaxial (bottom) side of the leaf to be syringe-infiltrated. Leaf was inoculated by infiltration with a blunt syringe. (c) Images of plants 24 h post-infiltration (hpi). Scale bar = 2 cm. (d) Images of callose deposits in aniline-blue-stained leaves that were previously infiltrated with mock (H₂O), 1 μM flg22 or elf18 for 24 h. Scale bar = 20 μm. (e) Schematic of a custom-built NIR Raman spectroscopy used to detect secondary metabolites in plants. An insert shows the infiltrated sites and the leaf discs taken from the proximal region (PR) of the sites for analysis.

Figure 2

Effects of sample location and elicitor treatment time on changes in Raman spectra. (A) The abaxial (bottom) side of a leaf is to be infiltrated with an elicitor. Elicitor (1 μM) was infiltrated with a needleless syringe at two locations (indicated as two small circles) nearest to the petiole. Leaf samples were taken from the PR and DR at 24, 48, and 72 hpi and analyzed by Raman spectroscopy. (B,C) A group of three panels was shown for either the DR or PR sample. Upper panel: mean spectra of control (black) and elicitor-treated samples (red, flg22 and blue, elf18). Middle panel: Difference spectrum between the mean spectrum of the elicitor-treated sample vs. that of the mock-treated sample. Red indicates that the difference is a positive value, whereas blue indicates a negative value. The number inside the panel indicates the average Elicitor Response Index (ERI, ×10⁴) value obtained from four (B) or three (C) independent experiments. Lower panel: t-test was used to evaluate statistically significant differences between elicitor-infiltrated and H₂O-infiltrated (mock) samples and the results were expressed as p-values. Red indicates p < 0.05. (D) Changes in ERI after the time of infiltration. Left, flg22 (red spots), n = 4 independent biological samples; Right, elf18 (blue spots), n = 3 independent biological samples. PR, proximal region; DR, distal region. The results represent the average values of three to four independent biological experiments with SD. Bars with different letters above are significantly different according to Fisher's least significant difference test (p < 0.05).

Figure 3

Raman spectroscopic analysis of wild type (WT) and mutant plants. (A) Raman spectra of WT Col-0 and mutant plants (fls2, efr-2, bak1, bik1, and rbohd) at 24 h after flg22 and elf18 treatment and mock control. Top panel: Images of plants 24 hpi. For each elicitor experiment (flg22, red and elf18, blue), a group of three panels was shown. Scale bar = 2 cm. Upper panel: a mean plot of all 60 spectra, taken from control (black) or elicitor-treated samples (red); Middle panel: the difference spectrum between the mean spectrum of the elicitor-infiltrated sample and that of the H₂O-infiltrated (mock) sample. Red indicates that the difference is a positive value, whereas blue indicates a negative value. The number inside the panel indicates the average ERI (×10⁴) value obtained from four independent experiments. Lower panel: t-test was used to evaluate statistically significant differences between elicitor-infiltrated and H₂O-infiltrated (mock) samples and the results were expressed as p-values. Red indicates p < 0.05. Plants: Col-0, WT A. thaliana; fls2, FLAGELLIN-SENSING2 (At5g46330); efr-2, EF-Tu receptor (At5g20480); bak1, BRASSINOSTEROID INSENSITIVE-ASSOCIATED KINASE1 (At4g33430); bik1, BOTRYTIS-INDUCED KINASE1 (At2g39660); and rbohd, NADPH/respiratory burst oxidase protein D (At5g47910). (B) ERI of WT and mutants treated with either flg22 or elf18. The results show average ERI values along with SD. Bars with different letters above are significantly different according to Fisher's least significant difference test (p < 0.05). Left: flg22 (red spots); Right, elf18 (blue spots). n = 4 independent replicates.

Figure 4

Raman spectroscopic analysis of WT and pub12pub13 and pub25pub26 double mutants. (A) Top panel: images of plants 10 hpi. For each elicitor experiment (flg22, red and elf18, blue), a group of three panels was shown. Raman spectra were acquired at 10 hpi with 1.0 μM flg22 (red peak) and elf18 (blue peak) treatment and mock control (black peak). Upper panel: a mean plot of all 60 spectra, taken from control (black) or the elicitor-treated samples (red, flg22 and blue, elf18); Middle panel: the difference spectrum between the mean spectrum of the elicitor-infiltrated sample and that of the H₂O-infiltrated (mock) sample. Red indicates that the difference is a positive value, whereas blue indicates a negative value. The numbers inside the panel indicate average ERI values. Lower panel: a t-test was used to evaluate statistically significant differences between elicitor-infiltrated and H₂O-infiltrated (mock) samples and the results were expressed as p-value. Red indicates p < 0.05. Plants: Col-0, WT A. thaliana; pub12pub13, U-box E3 ubiquitin ligase PUB12/13 (At2g28830 and At3g46510, respectively); fls2, FLAGELLIN-SENSING2 (At5g46330); pub25pub26, U-box E3 ubiquitin ligase PUB25/26 (At3g19380 and At1g49780, respectively); bik1, BOTRYTIS-INDUCED KINASE1 (At2g39660). Scale bars = 2 cm. (B) ERI of WT and mutants treated with either flg22 or elf18. The results show mean ERI values along with SD. Bars with different letters above are significantly different according to Fisher's least significant difference test (p < 0.05). Left: flg22 (red spots); Right, elf18 (blue spots). n = 4 independent biological replicates. (C) Representative images of callose deposits (bright spots) in the treated samples; Scale bars = 20 μm. (D) Quantification of callose deposits in WT (Col-0), pub12/13, fls2, pub25/26, and bik1 mutant leaves. The number of callose fluorescence spots was used to determine the number of callose deposits. Left: mock (black spots); Middle: flg22 (red spots); Right, elf18 (blue spots). The values shown are mean ± SD of 25–30 surveyed leaf regions. For each experiment, the number of callose deposits was measured in two infiltrated plants with about four to five leaf regions surveyed per plant. The value for each leaf region gives the number of callose deposits per 5,996 μm². This gave a total of about 8–10 data points per experiment. The experiment was repeated with three independent biological samples. The graph shows an average value with n = 25–30. Bars with different letters above are significantly different according to Fisher's least significant difference test (p < 0.05).

Figure 5

Changes in carotenoid levels in Arabidopsis leaf samples treated with an elicitor for 24 h. (A) Raman spectra of carotenoids from the leaves with mock and elicitor treatment for 24 h. Raman shifts at 1,001, 1,151, and 1,521 cm⁻¹ are attributed to carotenoids. n = 3 independent biological samples. (B) Changes of total carotenoids levels in Arabidopsis leaves after treatment with mock control, flg22, and elf18 for 24 h (24 hpi). Carotenoid levels were determined as described in the “Materials and methods” section. The values shown are mean ± SD of the extracted total carotenoids. For each experiment, the quantitation of total carotenoids was measured in three to five infiltrated plants per experiment. The experiment was repeated with four independent biological samples, and this gave a total of about 15–20 data points. ** and *** significantly different from the mock control at p < 0.001 and p < 0.0001 (t-test), respectively.

Figure 6

Raman spectra analysis of plants treated with pathogens. (A,B) Pathogens [Pseudomonas syringae pv. Tomato DC3000 (A); Xanthomonas campestris pv. Campestris (Xcc) (B)] were infiltrated with a needleless syringe into Arabidopsis and Choy Sum. Leaf samples taken from the PRs at 24, 48, and 72 hpi were analyzed by Raman spectroscopy. Upper panel: the mean spectra of control (black) and pathogen-treated samples (red). Middle panel: difference spectrum between the mean spectrum of the pathogen-treated sample vs. that of the mock-treated sample. Red indicates that the difference is a positive value, whereas blue indicates a negative value. The number inside the panel indicates the average Infection Response Index (IRI, ×10⁴) value obtained from the three independent experiments. Lower panel: a t-test was used to evaluate statistically significant differences between pathogen- and mock-infiltrated samples and the results were expressed as p-value. Red indicates p < 0.05. (C,D) Changes in IRI after the time of infiltration. Left, P. syringae pv. Tomato DC3000 (green spots), n = 3 independent biological samples; Right, Xcc (purple spots), n = 3 independent biological samples. The results represent the average values of three independent biological experiments with SD. Bars with different letters above are significantly different according to Fisher's least significant difference test (p < 0.05).