Physiologia Plantarum An International Journal for Plant Biology
Copyright ª Physiologia Plantarum 2008, ISSN 0031-9317
Physiologia Plantarum 135: 1–18. 2009
TECHNICAL FOCUS
Detecting hydrogen peroxide in leaves in vivo – a comparison of methods Iva Sˇnyrychova´a, Ferhan Ayaydinb and E´va Hidegc,* a
Laboratory of Biophysics, Department of Experimental Physics, Faculty of Science, Palacky University Olomouc, Czech Republic Cellular Imaging Laboratory, Biological Research Center, Szeged, Hungary c Institute of Plant Biology, Biological Research Center, Szeged, Hungary b
Correspondence *Corresponding author, e-mail:
[email protected] Received 11 August 2008; revised 17 September 2008 doi: 10.1111/j.1399-3054.2008.01176.x
Four hydrogen peroxide detecting probes, 3,3#-diaminobenzidine (DAB), Amplex Red (AR), Amplex Ultra Red (AUR) and a europium–tetracycline complex (Eu3Tc) were infiltrated into tobacco leaves and tested for sensitivity to light, toxicity, subcellular localization and capacity to detect H2O2 in vivo. In the absence of leaves, in water solutions, AUR was very much sensitive to strong light, AR showed slight light sensitivity, while DAB and Eu3Tc were insensitive to irradiation. When infiltrated into the leaves, the probes decreased the photochemical yield (FPSII) in the following order of effect AR > DAB > AUR > Eu3Tc. With the exception of Eu3Tc, all probes stimulated the build-up of non-photochemical quenching either temporally (DAB, AUR) or permanently (AR), showing that their presence may already limit the photosynthetic capacity of leaves, even in the absence of additional stress. This should be taken into account when using these probes in plant stress experiments. Confocal laser scanning microscopy studies with the three fluorescent H2O2 probes showed that the localizations of Eu3Tc and AUR were mainly intercellular. AR partly penetrated into leaf chloroplasts but probably not into the thylakoid membranes. Photosynthesis-related stress applications of AR seem to be limited by the low availability of internal leaf peroxidases. Applications of AR for kinetic H2O2 measurements would require a co-infiltration of external peroxidase, imposing another artificial modifying factor and thus taking experiments further from ideal, in vivo conditions. Our results suggest that the studied H2O2 probes should be used in leaf studies with caution, carefully balancing benefits and artifacts.
Introduction Hydrogen peroxide (H2O2) is a non-radical reactive oxygen species (ROS), produced in a two-electron reduction of molecular oxygen. Its main source in vivo is the enzymatic or spontaneous dismutation of superox-
ide radicals. Several sites have been recognized as H2O2 sources, including organelles (mitochondria, peroxisomes and chloroplasts), the apoplastic and the plasma membrane as well as cell-wall associated enzymes (various NADPH oxidases and peroxidases). In chloroplasts,
Abbreviations – FPSII, effective photochemical yield of PS II; AR, Amplex Red or Ampliflu Red, 10-acetyl-3,7dihydroxyphenoxazine; AUR, Amplex Ultra Red; CAT, catalase; DAB, 3,3#-diaminobenzidine; DCMU, 3-(3,4-dichlorophenyl)-1,1dimethylurea; Eu3Tc, europium–tetracycline complex; Fo, minimal fluorescence yield; Fm, maximum fluorescence yield; HRP, horseradish-peroxidase; LSM, laser scanning confocal microscopy; PPFD, photosynthetic photon flux density; PS, photosystem; resorufin, 7-hydroxy-3H-phenoxazin-3-one; ROS, reactive oxygen species; UV, ultraviolet.
Physiol. Plant. 135, 2009
1
H2O2 is formed during the photo-reduction of oxygen in the water-water cycle (Asada 1999). The overproduction of H2O2 has been observed in plants exposed to a number of stress conditions and is considered as one of the factors causing oxidative stress. H2O2 is lipid soluble and able to cross biological membranes (Bowler et al. 1992). As a result of this quality, which is unique among ROS, and of its relatively low reactivity as compared with singlet oxygen or oxygen radicals, H2O2 is a potential signal molecule (Laloi et al. 2007, Mittler et al. 2004, Slesak et al. 2007). In plants, H2O2 was shown to be the key regulator in programmed cell death (Bethke and Jones 2001, Desikan et al. 1998), as well as in a number of oxidative stress conditions, including response to pathogens (Levine et al. 1994, Mittler et al. 1999, Tenhaken et al. 1995). Similarities in cellular responses to a variety of abiotic stress conditions such as low (Prasad et al. 1994) or high (Dat et al. 1998) temperature, drought (Dat et al. 2000), osmotic stress (Guan et al. 2000), ultraviolet (UV)B irradiation (A-H-Mackerness et al. 1999), photoinhibition by excess visible light (Hernandez et al. 2004, Karpinski et al. 1999) and ozone (Langebartels et al. 2000) suggest that H2O2 could be a common factor regulating various signaling pathways (Neill et al. 1999). In Arabidopsis, more than one hundred expressed sequence tags were found to be either induced or repressed by H2O2, and many of these were also found in response to stress by UV irradiation or various chemical elicitors (Desikan et al. 2001, Vanderauwera et al. 2005). As part of its putative signaling function, H2O2 has been demonstrated to activate mitogen-activated protein kinases and glutathione S-transferase expression (Desikan et al. 1999) and to participate in the transduction of abscisic acid (ABA) signals (Guan et al. 2000). In order to explore its key function in stress responses, sensitive and selective techniques are needed for measuring H2O2, including its cellular localization and concentration in vivo. Staining with 3,3#-diaminobenzidine (DAB) is a long-established H2O2 detecting method in plant stress physiology. A peroxidase-catalyzed reaction of DAB and H2O2 leads to the formation of an insoluble brown precipitate (Malmgren and Olsson 1977, ThordalChristensen et al. 1997) which can be visualized in leaves and other pigmented plant tissue after the removal of chlorophyll. Consequently, fluorescent probes which can image H2O2 non-destructively are gaining popularity. One of them is 10-acetyl-3,7-dihydroxyphenoxazine [AR, commercially available as Amplex Red from Molecular Probes Invitrogen (Carlsbad, CA) and as Ampliflu Red from Sigma-Aldrich Kft., Budapest, Hungary], a colorless, non-fluorescent derivative of dihydro-resorufin (7hydroxy-3H-phenoxazin-3-one), which produces the highly fluorescent resorufin with H2O2 in a peroxidase2
catalyzed reaction (Zhou et al. 1997). Excitation and emission maxima of this probe (563 and 587 nm, respectively) make it advantageous for plant studies, as these do not overlap with corresponding spectral features of chlorophylls. Amplex Ultra Red (AUR) is sold by Molecular Probes Invitrogen as an improved version of AR with enhanced sensitivity and better stability. However, the chemical modification of AR to AUR (not revealed by the company) has also led to a dramatic change in membrane permeability. Studying cultured tobacco cells, Ashtamker et al. (2007) found that while AR penetrated the cells, AUR was unable to cross the cell membranes and remained in the medium. Resorufinderived probes have been already used in plants, the applications ranging from the detection of H2O2 in plant extracts (Guo and Crawford 2005) and cell cultures (Allan et al. 2006, Ashtamker et al. 2007, Garnier et al. 2006, Janisch and Schempp 2004) to H2O2 localization in pea nodules (Groten et al. 2005), Arabidopsis seedlings (Splivallo et al. 2007) or herbivore-wounded leaves (Maffei et al. 2006). Detecting H2O2 by either DAB or resorufin derivatives depends on peroxidase activity. In vivo the probes usually rely on endogenous peroxidase activity of the studied plant tissue, while in vitro studies this is usually achieved by an exogenous enzyme, for example horseradishperoxidase (HRP) as catalyst. A novel, peroxidase-free method to detect H2O2 utilizes europium-tetracycline [commercialized as europium-tetracycline complex (Eu3Tc) by Active Motif Chromeon, Tegernheim, Germany], a weakly fluorescent dye reactive to H2O2 to form a strongly fluorescent Eu3Tc-H2O2 complex (Wolfbeis et al. 2002). This probe has not been applied in plant studies so far, although its fluorescence emission maximum at 615 nm would also allow H2O2 detection in green leaves without interference with chlorophyll fluorescence. Studies on the role of H2O2 both in oxidative stress and in signaling pathways in leaves need direct and efficient (specific and sensitive) yet non toxic methods of H2O2 detection, which are not influenced either by light itself or by metabolites of steady-state (unstressed) photosynthesis. The aim of the present study was to compare the versatility of the above H2O2 detecting methods in photosynthetic tissue in vivo. We applied DAB, AR, AUR and Eu3Tc assays for H2O2 detection in tobacco leaves.
Materials and methods Chemicals AR was obtained from Sigma-Aldrich. AUR was purchased from Molecular Probes Invitrogen. Stock solutions Physiol. Plant. 135, 2009
(10 mM) of both probes were prepared in dimethylsulfoxide, divided into small aliquots and stored at 220C in the dark. Working solutions were prepared fresh by dilution of stock solution in 10 mM phosphate buffer pH 7.4 and were kept in the dark. Eu3Tc obtained from Active Motif Chromeon GmbH was diluted in 10 mM morpholino propanesulfonic acid (MOPS) buffer (pH 6.9) or distilled water. The stock solution was prepared as 500 mM using the absorption coefficient e401 ¼ 1.88 104 l mol21 cm21 published by Wu et al. (2003), which corresponded to an approximately 10-times more concentrated solution than recommended by the manufacturer. Currently, the molecular mass of Eu3Tc is not available. 3,3#-Diaminobenzidine (DAB) was purchased from Sigma-Aldrich and the solutions were prepared in distilled water. In order to dissolve DAB, the solution was acidified using HCl and after complete dissolution of DAB the pH was re-adjusted to 6–7 using NaOH. This DAB solution was used within 2 h. All other reagents (analytical grade) were obtained from Sigma-Aldrich Corporation (Sigma-Aldrich Kft.). Plant material Tobacco (Nicotiana tabacum, L.) plants were grown in the greenhouse at 22–24C with a natural photoperiod, under daytime irradiation maxima around 180– 200 mmol m22 s21 photosynthetic photon flux density (PPFD). The youngest fully expanded leaves of 3–4 weeks old tobacco plants were excised and used in all leaf experiments. Petioles were kept in water or in wet tissue paper to minimize water loss after excising. Infiltration and stereo microscopy Tobacco leaves or leaf discs were infiltrated with the highly fluorescent resorufin using various methods. Resorufin was prepared by oxidation of 50 mM AR in 10 mM phosphate buffer (pH 7.4) by equimolar amount of H2O2 in the presence of 1 U ml21 HRP. For the infiltration via the transpiration stream, a tobacco leaf was cut under water to prevent embolus air in the vein. The petiole of the leaf was immersed in the solution of resorufin and the leaf was kept in darkness for 3 h, under a fume hood to increase the transpiration rate. For the floating experiments, 1.5 cm diameter discs were cut from tobacco leaves and immediately immersed into the resorufin solution. Floated leaf discs were kept in darkness for 3 h, either in normal atmosphere, or under low pressure provided by a laboratory water suction pump. For the infiltration through the pinhole, the solution was forced into the mesophyll Physiol. Plant. 135, 2009
using a plastic syringe without a needle, through a pinhole made with a sharp pin, as described earlier (Hideg et al. 2002). This direct infiltration did not require further incubation. The distribution of resorufin in leaf samples was checked with Olympus SZX12 Stereo Microscope coupled to F-View II digital camera (Olympus Soft Imaging Solutions GmbH, Mu¨nster, Germany). A 540–580 nm filter was used for excitation and fluorescence was detected above 610 nm. For reference, transmission mode images were also taken from the samples using low intensity visible light. Spectrofluorometry Fluorescence of the probes was measured at room temperature using a Quanta Master QM-1 spectrofluorometer (Photon Technology Inc., Birmingham, NJ). For leaf experiments, the probe was first infiltrated into darkadapted leaf tissue using the pinhole method. A ca. 4 15 mm piece was cut from the infiltrated area and mounted on a special holder, ensuring reproducible positioning of the leaf strip inside the spectrofluorometer. The leaf was fixed with its adaxial surface up, facing both excitation beam and detection pathway at a 45 angle. Crossed polarizer and analyzer were used to shield scattered excitation light. AUR and AR fluorescence was measured in emission scan mode (560–610 nm) using 525 nm excitation. Because of a very small Stoke’s shift of AR and AUR fluorescence and high scattering of the leaf, it was not possible to use excitation wavelengths closer to the maximum of the excitation spectrum. Eu3Tc was excited with 400 nm and its fluorescence was detected between 585 and 635 nm (in emission scan mode) or at 615 nm maximum (in timebased mode). Localization of the fluorescent probes using confocal microscopy Localization of the fluorescent probes in various plant systems was visualized using confocal laser scanning microscope [Olympus FV1000 LSM (laser scanning confocal microscopy), Olympus Life Science Europa GmbH, Hamburg, Germany]. Chloroplasts and thylakoids were identified on the basis of chlorophyll fluorescence (Argon laser excitation: 488 nm, detection: 650–750 nm), the emission of pre-oxidized AR and AUR was detected between 585 and 610 nm using 543 nm HeNe laser excitation. For Eu3Tc, the excitation was set to 405 nm and the fluorescence was detected between 578 and 630 nm. For the localization of the sensors in leaves, tobacco leaves were infiltrated with pre-oxidized 3
forms of AR, AUR and Eu3Tc using the pinhole method and were investigated within 20 min.
leaves, then means and standard deviations were calculated for the graphs.
Light sensitivity of the probes
Photoinhibition and DAB staining
Light sensitivity of the H2O2 probes was tested by illuminating probe solutions in the absence of H2O2 and HRP. Solutions of DAB (1 mg ml21), AR (10 mM), AUR (10 mM) and Eu3Tc (50 mM) were placed in microcentrifuge tubes and illuminated with either 100 or 1200 mmol m22 s21 PPFD (KL-1500 lamp, Schott, Germany) for 30 min. The change in fluorescence (AR, AUR and Eu3Tc) was evaluated by comparing the emission spectra before and after the illumination using a Quanta Master QM-1 spectrofluorometer (Photon Technology Inc.). Conversion of DAB was followed by spectrophotometer.
A KL-1500 lamp (Schott, Germany) equipped with a lightguide was used as a high-light source without additional heat effect. Prior to photoinhibition, both left and right sides of a dark-adapted tobacco leaf were infiltrated with 0.1, 0.5 and 1 mg ml21 DAB then the leaf was cut into halves. The right half was kept in the dark, while the left one was exposed to 1200 mmol m22 s21 PPFD for 30 min. After 30 min, both halves were boiled in ethanol (75C, 10 min) to remove chlorophyll and visualize the brownish polymerization product of the preceding interaction between DAB and H2O2 in the presence of endogenous peroxidases. DAB stained leaves were stored in 96% ethanol and photographed.
Variable chlorophyll fluorescence The effect of the H2O2 probes on the photosynthetic performance of leaves was evaluated on the basis of variable chlorophyll fluorescence imaging (MINI-version of the Imaging-PAM, Heinz Walz GmbH, Effeltrich, Germany). Dark-adapted leaves were infiltrated with various concentrations of DAB, AR, AUR or Eu3Tc using the pin-hole method. The infiltration was performed under dim light and chlorophyll fluorescence parameters were measured within 5 min after the infiltration and then again after 30, 60 and 90 min of dark incubation. Fo, the minimal fluorescence yield of dark-adapted samples, was measured at low frequency of pulse modulated measuring light, while maximal fluorescence yield, Fm, was obtained with the help of a saturation pulse. Using these parameters, maximal photochemical quantum yield of photosystem (PS) II in dark-adapted leaves was calculated as Fv/Fm ¼ (Fm 2 Fo)/Fm. These measurements were followed by a 4-min exposure to 55 mmol m22 s21 blue actinic light, then minimal and maximal fluorescence yields (F, Fm#) were measured again before and after a saturating pulse while keeping the actinic illumination on. The effective photochemical yield of illuminated samples was calculated as FPSII ¼ (Fm# 2 F)/Fm# (Genty et al. 1989) and the non-photochemical quenching was quantified by the parameter non-photochemical quenching (NPQ) ¼ (Fm 2 Fm#)/ Fm# (Bilger and Bjo¨rkman 1990). FPSII and NPQ values were calculated for each point of the imaged 24 36 mm leaf area. To evaluate effects of various chemicals on photosynthesis, FPSII and NPQ images of treated and control leaf areas were compared and the results were quantified by averaging photosynthetic parameters from particular leaf areas. Photosynthesis experiments were repeated at least three times, using newly infiltrated 4
Statistical analysis The experimental data are shown either as mean values SD with minimum three independent repetitions or as representative data. The statistical significance of the differences between the experimental data was evaluated using one-way ANOVA followed by Tukey’s post hoc test.
Results Probe delivery – comparison of infiltration methods Apart from registering the production of H2O2, it is important to study localization and spatial distribution of this ROS in the samples. To achieve this, the H2O2 probe should be present in the sample before the onset of any putative H2O2 elicitor and it should be distributed uniformly. In liquid samples, assay components can be evenly mixed with the cell medium or the buffer containing the protoplasts. Leaves, however, need to be infiltrated. Because all studied H2O2 probes are water soluble, we used one of them to show that its distribution very much depended on the method of delivery. Fig. 1 shows fluorescence images of resorufin (pre-oxidized AR) infiltrated into leaves using various methods. These experiments also demonstrated that the orange-red fluorescence of resorufin is detectable, even at the inner layers of leaf tissue. Natural uptake with transpiration flow through the leaf petiole immersed in the resorufin solution was not suitable, because it delivered most of the probe into vascular tissue (Fig. 1A). Some of the probe was detectable in other leaf tissue as well when the uptake through the petiole was promoted by placing the Physiol. Plant. 135, 2009
Fig. 1. Distribution of resorufin (pre-oxidized AR) in tobacco leaf tissue achieved using various infiltration methods: transpiration uptake through the petiole (A), floating leaf discs on the infiltrating solution under atmospheric (B) or lower (C) pressure or using the pin-hole method (D) as described in Materials and Methods. The upper segment of the leaf in (D) was infiltrated with buffer as a control. The upper row shows images of red resorufin fluorescence, images in the lower row are photos taken in transmission mode. Scale bars: 5 mm.
whole leaf and infiltrating liquid under low pressure, but staining was always more intense in the larger leaf veins than in-between (data not shown). Stress experiments with such leaves would mistakenly conclude a preferential ROS production in vascular tissues. Floating leaf cuttings on infiltration solutions is another possibility. This is illustrated by fluorescence images of two leaf discs which were floated on a water solution of resorufin in the dark (Figs. 1B, C). Again, natural osmotic uptake was insufficient and resulted in a very heterogeneous staining, even though the whole perimeter of the leaf disc was available for the probe to penetrate (Fig. 1B). Identical but pressure assisted infiltration resulted in complete and homogenous distribution of fluorescence (Fig. 1C). This procedure, however, lowered the photochemical yield dramatically (by more than 50%, data not shown) and leaf cuttings with such limited photosynthesis would not be fitting for further stress experiments. Infiltrating a small vascular-tissue-limited segment of the leaf through a pinhole, a method used earlier for singlet oxygen
probes (Hideg et al. 2002) achieved uniform probe distribution over an approx 2 cm2 leaf area (Fig. 1D) with less mechanical injury than cutting out a leaf piece and imposed very little limitation to photosynthesis (
