robotized time-lapse imaging to assess in-planta uptake of...

Robotized time-lapse imaging to assess in-planta uptake of phenylurea

herbicides and their microbial degradation

Laury Chaerlea,1, Kris Hulsenb,1, Christian Hermansc,d, Reto J. Strasserd, Roland Valckee, Monica Hofteb and Dominique

Van Der Straetena,*

aDepartment of Molecular Genetics; Ghent University, K.L. Ledeganckstraat 35, B-9000 Gent, BelgiumbDepartment of Crop Protection, Laboratory of Phytopathology, Ghent University, Coupure links 653, B-9000 Gent, BelgiumcLaboratoire de Physiologie et de Genetique Moleculaire des Plantes, Universite Libre de Bruxelles, B-1050 Bruxelles, BelgiumdBioenergetics Laboratory, University of Geneva, Chemin des Embrouchis 10, CH-1254 Jussy/Lullier, SwitzerlandeLimburgs Universitair Centrum, Department SBG, Universitaire Campus, B-3590 Diepenbeek, Belgium1These authors contributed equally to this work*Corresponding author, e-mail: [email protected]

Received 15 November 2002; revised 26 February 2003

Two key physiological parameters of plant leaves, photosyn-

thesis and transpiration, can be continuously monitored by,respectively, chlorophyll a fluorescence imaging and thermo-

graphy. These non-contact techniques immediately visualize

any local stress or treatment affecting either photosynthetic

efficiency or water status. Photosystem II-inhibiting herbi-cides, including the phenylurea derivatives diuron and linuron,

cause a marked increase in chlorophyll a fluorescence several

days before appearance of chlorosis. Here, bioprotection

through microbial degradation of linuron in the feedingsolution of common bean plants (Phaseolus vulgaris L.) was

monitored by the absence of an increase in chlorophyll afluorescence in primary leaves. The different treatments and

repeats were imaged sequentially at 2 h intervals using a

robotized system with thermal, fluorescence and video cam-eras. Chlorophyll fluorescence imaging visualized the effect of

linuron transported by the transpiration stream earlier than

thermography. In addition, local effects and transport after

topical application of diuron were recorded presymptomatic-ally in tobacco (Nicotiana tabacum L.) and Arabidopsisthaliana (L.) Heynh. Thermal imaging clearly monitored

localized stomatal closure, coinciding with the first increase

in chlorophyll fluorescence, at the sites of diuron treatment. Inconclusion, the robotized chlorophyll a fluorescence set-up

permits fully reliable, early high-contrast visualization for

bioremediation purposes.

Introduction

Imaging techniques are well established as tools tovisualize heterogeneity in plant physiological parametersat the level of single leaves (Chaerle and Van DerStraeten 2000). Such spatial and temporal variation intranspiration and photosynthetic efficiency are readilyvisualized by non-contact methodologies such as thermalimaging and chlorophyll a fluorescence imaging. A localdecrease in transpiration results in a higher leaf surfacetemperature, which can be quantified on thermal images.Thermography has been used to visualize transpirationalchanges in common bean leaves (Jones 1999). In add-ition, local changes in transpiration were monitored after

tobacco mosaic virus infection (Chaerle et al. 1999) orduring spontaneous cell death (Chaerle et al. 2001).Photosynthetic reactions use light energy absorbed bychlorophyll. Absorbed light energy not used for photo-synthesis is dissipated partly as chlorophyll a fluores-cence (Krause and Weis 1991), but mainly as heat(Demmig-Adams and Adams 1996). Then, a reductionof photosynthetic capacity upon emerging stress usuallyimplies an increase in light and/or heat emission (Busch-mann 1999).

Chlorophyll a fluorescence emission can be measuredwith portable, non-imaging fluorometers (Maxwell and

PHYSIOLOGIA PLANTARUM 118: 613–619. 2003 Copyright# Physiologia Plantarum 2003

Printed in Denmark – all rights reserved ISSN 1399-3054

Abbreviations – PAM, pulse amplitude modulation; FIS, fluorescence imaging system; CCD, charge-coupled device; DCMU, diuron; PPFD,photosynthetic photon flux density; PSII, photosystem 2.

Physiol. Plant. 118, 2003 613

Johnson 2000, Strasser et al. 2000). Such repetitive fluor-escence point measurements are commonly used in plantstress physiology studies (Lichtenthaler 1988, Vidal et al.1995). In contrast, fluorescence imaging set-ups providean immediate overview of the fluorescence emissionpatterns from whole leaves or plants. Chlorophyll afluorescence imaging has been applied to visualize theeffects of diverse biotic and abiotic stresses at early stages(Lichtenthaler and Miehe 1997, Buschmann andLichtenthaler 1998, Ciscato and Valcke 1998).

Phenylurea herbicides such as linuron, diuron, isopro-turon and others are widely used for pre- and post-emergence control of annual grasses and broad-leaved weeds(Tomlin 1997). These compounds block photosyntheticelectron flow between the primary (QA) and secondary(QB) plastoquinone acceptor of photosystem 2 (PSII),by competitive displacement of plastoquinone from theQB-site on the PSII D1 protein (Trebst 1987, Fuerst andNorman 1991). Under the low intensity actinic lightconditions used in this study, the electron acceptor Qaremains mainly oxidized, resulting in a low fluorescenceyield. The (local) presence of DCMU results in the accu-mulation of reduced Qa, provoking a high fluorescenceyield at the affected areas. Photosynthesis being blocked,the absorbed light energy can only be dissipated as fluor-escence and as heat.

The strong increase in chlorophyll a fluorescence aftertreatment with the phenylurea herbicide diuron wasimaged with different set-ups on a variety of plant species(Daley et al. 1989, Lichtenthaler et al. 1997, Lootens andVandecasteele 2000). In addition to their effect on photo-synthesis, methylurea herbicides were reported to inhibitstomatal opening in epidermal strips, presumably bycausing a lack of ATP (Dahse et al. 1990). Stomatalclosure results in lower transpiration and thus higherleaf surface temperature, which is readily visualized bythermography. In addition, the part of the absorbed lightenergy that is dissipated as heat might also increase uponmethylurea herbicide treatment. However, a rise in leafsurface temperature due to heat accumulation is highlyimprobable from a thermodynamical viewpoint, giventhe high surface-to-volume ratio of leaves and the ensu-ing rapid heat dissipation (Breidenbach et al. 1997).

In crops or areas where phenylurea herbicides aresprayed generously, they may have a deleterious effect onthe adjacent ecosystems. The negative influence of thesecompounds is dependent on their half-life in soil, whichranges from 90 to 180days and 38–67 days for diuron andlinuron, respectively (Tomlin 1997). Despite their commontendency to adsorb to soil organic matter, the microbialdegradation of these pesticides is considered to be theprimary mechanism for their dissipation from soil.

Bioremediation strategies focus on increasing thebioremoval capacities of contaminated soils (Andersonet al. 1993, Alexander 1994, Dejonghe et al. 2001). Micro-molar linuron concentrations (0.4–8mM) are known tooccurin field conditions and are often near the detection limit ofconventional analysismethods (HPLC,GC).When selectinga suitable inoculum for the degradation of toxicants the

effect of low substrate concentrations on the survival of theinoculum should first be assessed (Pahm and Alexander1993). We previously showed that a bacterial consortiumwas able to degrade micromolar concentrations of linuron(Hulsen et al. 2002). To monitor linuron degradation, weused a bean plant-microbial bioassay combined with pulseamplitude modulation (PAM) non-imaging fluorometry.The potential of chlorophyll fluorescence imaging to visua-lize the bioprotective effect was illustrated by single time-point measurements of detached leaves.

To ascertain that bacterial pesticide degradation iseffective in preventing deleterious effects on the photo-synthetic system during the whole time-course of plantdevelopment, chlorophyll a fluorescence imaging needsto be carried out at regular intervals, in parallel on thesame set of control, pesticide-treated and pesticide 1

bacteria-treated plants. In contrast with previous singletime-point measurements on detached leaves (Hulsenet al. 2002), leaves of intact treated and control beanplants were monitored under growing conditions at 2 hintervals during 1 week. To realize the visualization ofsuch a batch of plants, a previously described robotizedimaging system, integrated into a growth chamber, wasused (Chaerle et al. 1999). Robotized time-lapse imagingof all leaves on a batch of plants enables direct andreliable comparisons between treatments and repeats.

Stomatal closure upon local methylurea herbicidetreatment is the more likely cause for an increase in leafsurface temperature. Furthermore, methylurea herbi-cides transported to the leaf by the transpiration streamcould also affect the pattern of leaf temperature in adynamic way. This was the incentive to perform thermalmeasurements in parallel with chlorophyll a fluorescenceimaging after phenylurea herbicide treatment.

Materials and methods

Plant material and growth conditions

Common bean (Phaseolus vulgaris L. cv. Prelude) wasgrown according to the protocol of the bioremediationassay, under the conditions of the imaging set-up (seebelow). Tobacco plants (Nicotiana tabacum L. cv. petitHavana line SR1) were grown on potting soil at 216 1�Cand at 606 10% relative humidity, under fluorescenttubes (Philips TLD 33; Koninklijke Philips ElectronicsN.V., Eindhoven, the Netherlands) delivering a PPFD of506 10 mmolm�2 s�1. The photoperiod was 16 h light/8 hdarkness. Tobacco plants used for the diuron assay were10weeks old. Arabidopsis plants (Arabidopsis thaliana(L.) Heynh. Col-0 ecotype) were grown in potting soil ata light intensity of 200mmolm�2 s�1 PPFD, under aphotoperiod of 16 h light/8 h dark. Diuron wastopically applied on 40-day-old-plants.

Treatments

Diuron (3-[3,4-dichlorophenyl]-1,1-dimethylurea; abbrev.DCMU; 2mM solution in 30% ethanol) was applied as

614 Physiol. Plant. 118, 2003

50 ml droplets on the tobacco or Arabidopsis leaf surface.In the case of tobacco, the diuron solution was rinsed offafter 1 h to prevent precipitation, and the leaf wassubsequently blotted dry. In Arabidopsis, the diuronsolution droplet was taken up/evaporated within 20minwithout leaving a precipitate.

Linuron (3-[3,4-dichlorophenyl]-1-methoxy-1-methylurea)was purchased as a commercial suspension formulation(500 gml�1) named Linurex (Protex N.V., Wijnegem,Belgium) which was diluted and sterilized by filtration(0.25 mm pore-size filters) prior to addition to the mediumas an aqueous solution (50mg l�1).

Bioremediation bioassay

A similar plant-microbial system to that previouslydescribed (Hulsen et al. 2002) was used for the biodegrad-ation experiment. Common bean was planted in theuppermost pot containing sterile hydrophilic vermiculiteas a substrate. A 10mm top layer of sterile hydrophobicperlite prevented any airborne microbial/pathogen con-tamination of the plant substrate and the lower jar. Thelower jar contained 650ml of sterile Hoagland’s plantnutrient solution, linuron (1mg l�1) and/or the bacterialinoculum, according to the treatment. A paper wickthrough the bottom of the upper pot ensured the absorp-tion of the solution by the plant roots. The linuron-degrading strain Variovorax paradoxus WDL1, isolatedfrom a linuron-degrading bacterial consortium (W.Dejonghe, unpublished) was maintained on a minimalmedium supplemented with 25mg l�1 linuron as solecarbon and nitrogen source. The inoculum was preparedafter subculturing 10% (v/v) of the initial culture to newflasks, which were incubated for 4 days at 140 r.p.m and28�C. After 4 days the bacterial suspension was centri-fuged at 3000 g for 20min. The supernatant was dis-carded and the cell pellet was re-suspended in sterilewater to give a final cell density of approximately107�108 colony-forming unitsml�1. Next, 3% (v/v)inoculum was added to the lower jar together with thelinuron and the nutrient solution. Four days later thebean plants were seeded in the vermiculite. This periodwas necessary to give the degrading bacteria the oppor-tunity to metabolize the linuron since previous experi-ments indicated an instant uptake of the linuron by theseedlings (data not shown).

Imaging set-up

After seeding of beans, eight pots representing four treat-ments (control, 1mg l�1 linuron, 1mg l�1 linuron1 V.para-doxus WDL1 and control 1 V.paradoxus WDL1) wereplaced in a custom-built plant growth chamber with built-in robotized (XYZ positioning) imaging capability (Chaerleet al. 1999). A mobile chlorophyll a fluorescence imagingsystem (FIS) was integrated into the automated imagingsystem, already equipped with a thermal (FLIR-AgemaTHV-900 LW (FLIR Systems, Portland, OR, USA) –Stirling cooled)-and a colour CCD video camera (Watec

LCL-217HS, Watec America Corporation, Las Vegas, NV,USA). The thermal camera captures images of infrared lightemission in the 8–12mm part of the spectrum; the videocamera is used to capture colour reflectance images. TheFIS consists of an illumination head with small halogenlamps fitted with cut-off low-pass blue filters and a centrallymounted miniature BandW CCD camera equipped with acut-off high-pass red (B1W 092) filter, avoiding capture ofreflected excitation light (M. Ciscato 2000. Thesis, LimburgsUniversitair Centrum, Diepenbeek, Belgium).

At the start of an experiment, selected positions forthermal image capture were memorized into the robotcontroller (teach-in procedure). Subsequently, the systemacquired sequentially the three types of images at theentered positions. Fluorescence and reflectance image-capture co-ordinates were generated automaticallybased on fixed offsets between the side-by-side mountedcameras. Images were captured at intervals of 2 h. Thesequence of imaging positions was chosen as to minimizethe influence of the additional excitation light on neigh-bouring plants. The two primary leaves of each beanplant were imaged separately, 5min apart. Temperaturemeasurements on the presented thermal images werecarried out with the FLIR systems Research software.

Primary bean leaves were imaged 10 days after plant-ing. From this time-point on, the height of the primaryleaves is constant. The trifoliate leaves emerge about5 days later. Differences in height between the primaryleaves of different plants were compensated by adjustingthe Z-axis positioning during teach-in. Increases in leafsize are due to the leaf expansion during growth. Earlierimaging of the emerging primary leaves would need con-tinuous adjustment of the imaging distance to keepfocus.

To permit a time-lapse capture of images under con-stant conditions bean plants were grown under continu-ous light (60mmolm�2 s�1 PPFD) from fluorescent tubes(Radium 21–840 Spectralux Plus, Radium Lampenwerk,Wipperfurth, Germany). In addition, continuous lightminimizes the nyctinastic leaf movements. Fluorescenceimages were captured after 1 s of supplemental illumina-tion with 250mmolm�2 s�1 excitation light from the FIS,without predarkening.

Due to the remaining leafmovement (nyctinastic and leaf-expansion linked), fluorescence intensity and light reflectionstill varied over time. However, this did not hamper thecontinuous visualization of the characteristic high-contrastchlorophyll a fluorescence pattern caused by linuron uptake.If necessary the leaf movements could be further attenuatedby supporting the leaf with fine nylon mesh.

Laboratory chlorophyll a fluorescence kinetics ondark-adapted Arabidopsis plants have been performedwith the Fluorcam M690 (PSI Instruments, Brno, CzechRepublic), by illuminating the entire rosette target withcontinuous actinic orange light provided with 635 nmlight-emitting diode panels that generate a uniform irradi-ance of 300mmol photonsm�2 s�1. Images were capturedon selected time points, 50 s after the start of actinicillumination.


Results

Determination of the efficiency of microbial linuron

degradation by multispectral imaging in common bean

From 10 days after planting, chlorophyll a fluorescence,thermal and colour reflectance image sequences wererecorded from primary bean leaves over a period of2 days at 2 h intervals. Figure 1A shows symptomlesschlorophyll a fluorescence and colour reflectance imagesof a primary bean leaf, 4 h before the appearance of anincrease in chlorophyll a fluorescence at the leaf base.Ten hours later (Fig. 1B), an increase in chlorophyll afluorescence emission was visible along the main and sideveins of the leaf from the plant grown on the linuron-containing solution. The increase was visualized ashigher intensities (tending to white). In the last panel,16 h after its first visualization, the increase in chloro-phyll a fluorescence has spread further towards the apexof the leaf. In all three panels, chlorophyll a fluorescenceimages from control and linuron 1 V. paradoxus WDL1treatment do not show any increase in chlorophyll afluorescence emission. The imaging distance to the leaveswas constant; the increase in leaf size is due to leafexpansion during the 2 day spanning animation.

Visual symptoms of linuron treatment were apparentas yellowing, 5 days after the visualization of the increasein chlorophyll a fluorescence (Fig. 2A). Leaf size wasalso reduced in comparison with control and the linuron1 V. paradoxus WDL1 treatment. At 10 days after thefirst increase (Fig. 2B), the linuron-treated leaf showedextensive necrosis and had lost turgor. The plant grownon a solution containing linuron and linuron-degradingV. paradoxus WDL1 bacteria appeared similar to thecontrol plant. At later time points, the pattern of chloro-phyll a fluorescence emission was also comparable tothe emission of the control (untreated) plant leaf (datanot shown). The phenotype of plants fed with a solutioncontaining V. paradoxus WDL1 only, was comparablewith that of the control plants during the whole timecourse of the experiment, proving the absence of adversebacterial effects on the plants (data not shown).

In the thermal images of bean leaves before theappearance of any effect in the chlorophyll a fluores-cence images, the venation pattern was already charac-terized by a higher temperature (Fig. 3A). The samethermal pattern was seen at the first time-point in controlleaves (data not shown). At the last time point when anincrease in chlorophyll a fluorescence was apparent in awide zone surrounding the venation pattern, thermalimages had a more pronounced venation pattern(Fig. 3C) when compared with Fig. 3A and B, due tothe effect of the herbicide on the tissue adjacent to theveins.

The described difference in chlorophyll a fluorescenceemission can also be observed under low-intensitygrowth chamber light conditions (continuous PPFD of60 mmol photonsm�2 s�1), without using the additionalexcitation light on the FIS (data not shown).

Effects of topical leaf application of diuron in tobacco and

Arabidopsis

Immediately after topical application of diuron dropletson the adaxial side of a tobacco leaf, a local increase inleaf surface temperature at the site of treatment wasvisualized with thermography (Fig. 4A, left column).

Fig. 1. (A)–(C) Chlorophyll a fluorescence imaging of primaryleaves of bean plants. First column: control; second column:1mg l�1 linuron and last column: 1mg l�1 linuron 1 V. paradoxusWDL1. The upper row of each panel shows images of chlorophylla fluorescence under excitation light with a PPFD of250mmolm�2 s�1. Corresponding colour reflectance images areshown in the lower row. (A), Primary leaves 10 days afterplanting; (B), images captured 10 h later; (C), images captured16 h after (A). Leaf sizes increase gradually in function of time dueto growth expansion. Scale bar¼ 10mm (A).


Chlorophyll a fluorescence images show a clear increasein intensity at the treatment sites (Fig. 4A, middlecolumn). The effect of diuron applied on the main veincan be discerned by thermography, although contrast islower in comparison with the fluorescence images. Finedetails of transport along veins present in the fluores-

cence images are not detected with thermal imaging,which reveals a more diffuse signal. Ninety minuteslater (Fig. 4B), diuron was further transported acro-petally along the minor veins, as can be seen on thechlorophyll a fluorescence images. Major sites oftransport along minor veins are hardly visible in thethermal images. The temperature difference betweenuntreated leaf tissue and the diuron-treated regions hasincreased.

In Arabidopsis, diuron was applied to the adaxial sideof a single fully expanded rosette leaf. Fluorescenceimaging visualized the transport throughout the rosette.Increased chlorophyll a fluorescence emission was firstapparent in the youngest upper leaves (Fig. 5B) andthereafter appeared gradually in all petioles and part ofthe leaf blades (Fig. 5C–E).

Additional information on the performed experimentscan be accessed at http://allserv.rug.ac.be/�lchaerle/

Fig. 2. (A), (B) Colour reflectance images of primary bean leaves.The same leaves as in Fig. 1 are imaged: left control, middle 1mg l�1

linuron and right 1mg l�1 linuron1V. paradoxusWDL1. (A), 15daysafter planting; (B), 20 days after planting. The fully expanded leaveswere supported by fine nylon mesh. Scale bar¼ 10mm (A).

Fig. 3. (A)–(C) Thermal and chlorophyll a fluorescence imaging ofprimary bean leaves fed on a 1mg l�1 linuron solution. Left: thermalimages; middle: chlorophyll a fluorescence images; and right: colourreflectance images. The thermal images were saved with atemperature span of 1.5�C. On the thermal scale, white and blackcorrespond, respectively, to temperatures above and below the1.5�C window. (A), The linuron-treated leaf from Figs 1 and 2 at thefirst appearance of the increase in chlorophyll a fluorescence – 4 hlater than the time point in Fig. 1A; panels (B) and (C) show theevolution of the effect, respectively, 12 and 38 h after first symptomappearance. Scale bar¼ 10mm (A).

Fig. 4. (A), (B) Thermal (left column) and chlorophyll afluorescence (middle column) images of a tobacco leaf region aftertopical diuron application. Colour reflectance images are presentedon the right. The thermal images show a temperature range of1.5�C, from 20.1 to 21.6�C. (A), 2 h after droplet application; (B),1.5 h later. Scale bar¼ 10mm (A).

Fig. 5. Chlorophyll a fluorescence imaging of diuron transport inArabidopsis ecotype Col-0. (A), The rosette before treatment; (B),2 h after treatment. Diuron was applied topically on one leaf,discernible in panels (B)–(E) by its overall high chlorophyll afluorescence; (C) 4 h after application of diuron; (D), 6 h afterdiuron application; (E), 10 h after diuron application; (F), grey scalereflectance image of the treated Arabidopsis rosette. Scalebar¼ 10mm (A).


Discussion

Presymptomatic assessment of bacterial herbicide

degradation

Figure 1 illustrates time-lapse visualization of linuron-degradation by the bacterial strain V. paradoxus WDL1.Ten days after exposure of the plants, the first signs oflinuron phytotoxicity appeared in the chlorophyll fluores-cence images of the uninoculated treatment. The spread-ing of the chlorophyll fluorescence emission along theveins and further to the leaf apex and to the intervenalareas is consistent with observations made upon feedingof Xanthium strumarium petioles with diuron (Daley et al.1989). The damage pattern obtained in the fluorescenceimages of the linuron only treatment is closely linked withthe mode of action of the herbicide and the movement ofthe transpiration front through the lamina. The absence ofthese patterns in the inoculated treatment is indicative ofthe degradative capacities of a culture of V. paradoxusWDL1, growing under microbially controlled conditions,when 1mg l�1 of linuron is fed to the Phaseolus plants. Upto the time when visual symptoms appeared and duringthe ensuing necrosis in the bean plant treated solely withlinuron, no increase in fluorescence was observed from theleaves of plants growing on a solution containing linuronand bacteria. Thus, time-lapse monitoring of chlorophylla fluorescence emission from differently treated primarybean leaves clearly proved the efficiency of the bacterialherbicide degradation in the feeding solution.

Until now, very few investigations have reported onpure bacterial isolates able to degrade linuron (Wallnofer1969, Cullington and Walker 1999, Turnbull et al. 2001).In contrast to the latter investigations, the main contri-bution of this plant-microbial bioassay is to highlightand to anticipate a major shortcoming, very often over-looked by conventional degradation studies: how do low(micromolar) and environmentally relevant substrateconcentrations (resulting from field application) affectthe degradative capacities of any microbial inoculumpossibly applicable for bioremediation purposes?

The bioprotective effect obtained with V. paradoxusWDL1 was in agreement with the degradation of higherconcentrations (50mg l�1) of linuron (W. Dejonghe, per-sonal communication), with previous results obtainedwith a very sensitive (nanomolar concentrations –beyond the HPLC detection limit) chlorophyll fluores-cence-based Lemna minor bioassay and with PAM-measurements and single-time point fluorescence imagingusing the bean bioassay (Hulsen and Hofte 2001).

Comparison of thermal and chlorophyll a fluorescence

imaging

Although the venation pattern of bean leaves appearsmore pronounced in Fig. 3B and C, it is clear thatthermal imaging is not optimally suited to follow theevolution of the response to linuron uptake in commonbean, and probably also in leaves from other species. In a

thermal image, the leaf venation pattern has a highertemperature than the rest of the leaf lamina. This is dueto the absence of stomata on leaf epidermis coveringveins, as observed in common bean (personal observa-tion), and is a general characteristic of higher plants(Weyers and Meidner 1990). The lower transpirationalcooling at the veins leads to a higher leaf surface tem-perature. If a compound, transported into the leaf via theveins, inhibits transpiration, an increase in temperatureof the tissue adjacent to the veins is to be expected.However, this affected zone will poorly contrast withthe venation pattern, thus hampering a clear early detec-tion by infrared thermography. In addition, the reso-lution of thermal imaging is inherently lower due to lateralthermal conduction in the leaf tissue. In contrast tothermographic images, chlorophyll a fluorescence imagesare characterized by a dark leaf venation pattern on abrighter leaf lamina. This is a consequence of the lowerchlorophyll contents in the veins. When photosynthesis isinhibited in the zone directly adjacent to the veins, chloro-phyll fluorescence intensity will increase and thus contrastwell both with the veins and the surrounding unaffectedtissue. Thus, high-contrast chlorophyll a fluorescenceimaging is the method of choice for early visualizationof methylurea herbicide-induced effects.

Perspectives for presymptomatic monitoring of stress

induced by the environment

Imaging-aided monitoring of the accumulation ofcompounds interfering with the metabolic processes ofbioindicators offers an alternative to visual assessmentand conventional analytical methods for evaluation ofbioremediation strategies. Because the described robot-ized chlorophyll fluorescence system has been demon-strated to be a very sensitive tool to detect the earliestsymptoms of residual herbicide damage, it is also verysuitable to prove remediation (i.e. breakdown and even-tually absence) of these compounds.

Robotized time-lapse fluorescence imaging of a batchof plants makes direct comparisons between treatmentspossible. This enabled us to evaluate the effectiveness oflinuron degradation by a bacterial inoculum whenapplied in a plant-microbial bioassay. From a practicalviewpoint, this labour-saving methodology allows a con-siderable extension of the experimental set-up (increasednumber of replicates) thereby permitting simultaneousand high-contrast comparison of different treatments.Arabidopsis plants could be used in a similar bioremedi-ation set-up as used by bean, allowing considerable spacesaving. The robotization of this process has a clearpotential for screening purposes.

The ability to monitor plants from the emergence of apresymptomatic effect until the appearance of visualsymptoms permits correlation of the early stress indica-tions with the final damage (Chaerle & Van der Straeten2001). A ‘multispectral’ set-up combining different sen-sors is very convenient to select the most effective ima-ging method to reveal early manifestation of a stress


condition. The combination of imaging procedures withbioindicators has the potential to assess the quality of thegrowth medium and irrigation solution. In laboratoryconditions, when including proper controls and keepingall other factors constant, soil- or irrigation-borne dele-terious effects on crop performance can be highlighted.Thus, monitoring of plant health status using a multi-spectral imaging system offers a wide range of applica-tions in the field of bioremediation.

Acknowledgements – The authors thank Martin vande Ven,Limburgs Universitair Centrum, Belgium for help with the set-upof their FIS system at Ghent University. We acknowledge RonaldMaldonado-Rodriguez, University of Geneva, Switzerland, forassisting C.H. with the chlorophyll a fluorescence imaging ofArabidopsis. This research was supported by grants from the Fundfor Scientific Research (Flanders) (G0015-01 to D.V.D.S. and R.V.,G002798N to M.H.) and the Swiss National Science Foundation (toR.J.S. Nr. 31–570046.99). L.C. is a postdoctoral research assistantof the Fund for Scientific Research (Flanders). C.H. is supported bya grant from the Fonds pour la Formation de la Recherche dansl’Industrie et dans l’Agriculture, Communaute francaise de Belgique(FRIA).

References

Alexander M (1994) Inoculation. In: Alexander M (ed) Biodegrad-ation and Bioremediation. Academic Press, San Diego, CA, pp.226–247

Anderson TA, Guthrie EA, Walton BT (1993) Bioremediation inthe rhizosphere. Environ Sci Technol 27: 2630–2636

Breidenbach RW, Saxton MJ, Hansen LD, Criddle RS (1997) Heatgeneration and dissipation in plants: can the alternative oxidativephosphorylation pathway serve a thermoregulatory role in planttissues other than specialized organs? Plant Physiol 114: 1137–1140

Buschmann C (1999) Thermal dissipation during photosyntheticinduction and subsequent dark recovery as measured by photo-acoustic signals. Photosynthetica 36: 149–161

Buschmann C, Lichtenthaler HK (1998) Principles and character-istics of multi-colour fluorescence imaging of plants. J PlantPhysiol 152: 297–314

Chaerle L, Van Der Straeten D (2000) Imaging techniques and theearly detection of plant stress. Trends Plant Sci 5: 495–501

Chaerle L, Van Der Straeten D (2001) Seeing is believing: imagingtechniques to monitor plant health. Biochim Biophys Acta –Gene Struct Express 1519: 153–166

Chaerle L, De Boever F, Van Montagu M, Van Der Straeten D(2001) Thermographic visualization of cell death in tobacco andArabidopsis. Plant Cell Environ 24: 15–25

Chaerle L, Van Caeneghem W, Messens E, Lambers H, VanMontagu M, Van Der Straeten D (1999) Presymptomaticvisualization of plant–virus interactions by thermography.Nature Biotechnol 17: 813–816

Ciscato M, Valcke R (1998) Chlorophyll fluorescence imaging ofheavy metal treated plants. In: Garab G (ed) Photosynthesis:Mechanisms and Effects, Vol. IV. Kluwer Academic Publishers,Dordrecht, The Netherlands, pp. 2661–2663

Cullington JE, Walker A (1999) Rapid biodegradation of diuronand other phenylurea herbicides by a soil bacterium. Soil BiolBiochem 31: 677–686

Dahse I, Willmer CM, Meidner H (1990) Tentoxin suppressesstomatal opening by inhibiting photophosphorylation. J ExpBot 230: 1109–1113

Daley PF, Raschke K, Ball JT, Berry JA (1989) Topography ofphotosynthetic activity of leaves obtained from video images ofchlorophyll fluorescence. Plant Physiol 90: 1233–1238

Dejonghe W, Boon N, Seghers D, Top EM, Verstraete W (2001)Bioaugmentation of soils by increasing microbial richness: miss-ing links. Environ Microbiol 3: 649–657

Demmig-Adams B, Adams WWI (1996) The role of xanthophyllcycle carotenoids in the protection of photosynthesis. TrendsPlant Sci 1: 21–26

Fuerst EP, Norman MA (1991) Interactions of herbicides withphotosynthetic electron transport. Weed Sci 39: 458–464

Hulsen K, Hofte M (2001) The microbial degradation of nM-/mMconcentrations of linuron estimated by a Lemna minor bioassay.In: Walker A (ed) Pesticide Behaviour in Soils and Water, Sym-posium Proceedings, Vol. 78. British Crop Protection Council(BCPC) Publications, Bracknell, Berks, UK, pp. 157–162.

Hulsen K, Top EM, Hofte M (2002) Chlorophyll fluorescence tomonitor microbial degradation of linuron applied in low con-centrations to bean plants. New Phytol 154: 821–829

Jones HG (1999) Use of thermography for quantitative studies ofspatial and temporal variation of stomatal conductance overleaf surfaces. Plant Cell Environ 22: 1043–1055

Krause GH, Weis E (1991) Chlorophyll fluorescence and photo-synthesis: the basics. Annu Rev Plant Physiol Plant Mol Biol 42:313–349

Lichtenthaler HK (1988) In vivo fluorescence as a tool for stressdetection in plants. In: Lichtenthaler HK (ed) Applicationsof Chlorophyll Fluorescence in Photosynthesis Research,Stress Physiology, Hydrobiology and Remote Sensing.Kluwer Academic Publishers, Dordrecht, The Netherlands,pp. 129–142

Lichtenthaler HK, Miehe JA (1997) Fluorescence imaging as adiagnostic tool for plant stress. Trends Plant Sci 2: 316–320

Lichtenthaler HK, Lang M, Sowinska M, Summ P, Heisel F, MieheJA (1997) Uptake of the herbicide diuron as visualised by thefluorescence imaging technique. Bot Acta 110: 158–163

Lootens P, Vandecasteele P (2000) A cheap chlorophyll a fluores-cence imaging system. Photosynthetica 38: 53–56

Maxwell K, Johnson GN (2000) Chlorophyll fluorescence – a prac-tical guide. J Exp Bot 51: 659–668

Pahm MA, Alexander M (1993) Selecting inocula for the biodegrad-ation of organic-compounds at low concentrations. MicrobialEcol 25: 275–286

Strasser RJ, Srivastava A, Tsimilli-Michael M (2000) The fluores-cence transient as a tool to characterise and screen photosynth-etic samples. In: Mohanty, P, Yunus, M, Pathre, U (eds)Probing Photosynthesis: Mechanism, Regulation and Adapta-tion. Taylor & Francis, London, UK, pp. 443–480

Tomlin CDS (ed) (1997) The Pesticide Manual, 11th edn TheBritish Crop Protection Council, London, UK

Trebst A (1987) The three-dimensional structure of the herbicidebinding niche on the reaction center polypeptides of photosys-tem II. Z Naturforsch 42c: 742–750

Turnbull GA, Ously M, Walker AES, Morgan JAW (2001) Degrad-ation of substituted phenylurea herbicides by Arthrobacterglobiformis strain D47 and characterization of a plasmid-associated hydrolase gene, puhA. Appl Environ Microbiol 67:2270–2275

Vidal D, Martinez-Guijarro J, Simon E (1995) Chlorophyllfluorescence and photosynthetic O2 evolution in Vicia fabaplants treated with methabenzthiazuron. Photosynthetica 31:9–20

Wallnofer P (1969) The decomposition of urea herbicides by Bacil-lus sphaericus, isolated from soil. Weed Res 9: 333–339

Weyers JDB, Meidner H (eds) (1990) Variability in stomatal andepidermal characteristics. In: Methods in Stomatal Research.Longman Scientific and Technical, Harlow, UK, pp. 16–21

Edited by A. J. Stemler


robotized time-lapse imaging to assess in-planta uptake of...

Documents