Blue-green laser-induced fluorescencefrom intact leaves: actinic light sensitivity andsubcellular origins

Marinella Broglia

Remote sensing of the health status of vegetation should be possible by using UV laser-inducedfluorescence; nevertheless, the molecular origin of the leaf blue-green fluorescence emission is stillunknown. In order to investigate possible relations of this fluorescence to the photosynthetic apparatus,we looked for its intensity changes after the addition of actinic light. The lack of any changes outside thechlorophyll fluorescence bands (Kautsky effect) was further investigated by collecting spectra from cell,protoplast, and chloroplast suspensions. These spectra led us to ascribe most of the blue-greenlaser-induced fluorescence that is detectable on a leaf by UV laser excitation to extrachloroplasticcompartments. In active chloroplast suspensions blue fluorescence from photosynthetically reducednicotinamide adenine dinucleotide phosphate (NADPH) has been detected and should be characterized bytime-resolved fluorescence techniques.

Key words: Plant stress remote sensing, laser-induced fluorescence of vegetation, leaf blue-greenfluorescence, NADPH fluorescence.

Introduction

Light-induced fluorescence emission is one importantdeexcitation mechanism that, with photosynthesis,shares the light energy transformation in leaves.'The in vivo chlorophyll fluorescence (680-740 nm),excited by different visible lights, has been exten-sively studied, and a good knowledge of its correlationto the mechanism of photosynthesis has been ob-tained. 2 -5

Laser-induced fluorescence (LIF) has been pro-posed as a technique for early remote sensing ofstress conditions in plants.6 Several advantages couldbe obtained by using pulsed UW lasers as excitationsources: (a) the use of this excitation spectral rangeis recommended by the international eye-safe regula-tion in remote sensing; (b) powerful and reliable lasersources are available in this spectral range; (c) effi-cient time discrimination from ambient light andcontinuous fluorescence can be obtained from thehigh repetition rate pulsed excitation; (d) fluores-cence emission in a spectral range as wide as possible

The author is with Progretto Biotecnologie, Ente per le NuoveTecnologie l'Energia e 'Ambiente, Centro Ricerche Energia-Casaccia, P.O. Box 2400, Rome 00100, Italy.

Received 10 December 1991.0003-6935/93/030334-05$05.00/0.3 1993 Optical Society of America.

can be collected, increasing the information contentof the LIF technique.7

A large number of correlations have been obtainedbetween LIF spectra and different plant types (mono-cots, dicots, hardwoods, and algae), their nutritionaldeficiencies and their stress conditions6-9; neverthe-less, a full exploitation of the technique requiresknowledge of the molecular origin of all fluorescencebands.

In this work we used an excimer laser and collectedspectra from leaves of different species and in differ-ent stress conditions. In order to investigate the ori-gin of the blue-green fluorescence, we used the samelaser (at lower power) as a probe, observing fluores-cence changes induced by actinic light on the photo-synthetic apparatus.10 The topology of the emissionwas tentatively assigned by LIF spectra of cell, proto-plast, and chloroplast suspensions obtained from thesame leaves. By using an excimer pumped dye laser,typical fluorescence increases from reduced nicotin-amide adenine dinucleotide phosphate (NADPH) weredetected in spinach chloroplast suspensions.

Methods and Materials

Instrumental Apparatus

Fluorescence on intact leaves, cell, protoplast, andchloroplast suspensions was excited by a XeCl exci-mer laser source (Fig. 1). This source emits UV light

334 APPLIED OPTICS / Vol. 32, No. 3 / 20 January 1993

Fig. 1. Experimental setup: BS, beam splitter; C, chopper; D,diaphragm; F, filter; L, lens; PMT, photomultiplier tube; T, target,A/D, analog-to-digital.

pulses (308 nm) with a half-duration of 8 ns, and itwas run at a pulse repetition rate of 10 Hz. Duringall the measurements the energy output of the laserwas 15 mJ/pulse, and its long-term stability wasmonitored by a vacuum photodiode (500-ps rise time)for measurement normalization.

The laser beam, spatially filtered and focused, wasdirected onto the target, where the beam cross-sectional area was 3 mm x 8 mm. Light intensityon the target has been varied by quartz neutral-density filters.

Emitted fluorescence was collected and spectrallyanalyzed by a scanning double monochromator ([4.0) and a red sensitive photomultiplier tube (Ha-mamatsu R666S). Diffused laser light was cut off infront of the monochromator by a long-wavelengthbandpass filter (> 400 nm). In all the measure-ments the photomultiplier response versus fluores-cence intensity was checked for linearity and, whennecessary, we used glass neutral-density filters.

Electric signals from the photomultiplier tube andfrom the vacuum photodiode were processed by aboxcar averager (15-ns gate, 0.3-s response timeconstant) and, through an analog-to-digital con-verter, entered into a computer.

The comprehensive spectral response of our opticaland photosensitive elements (lenses, filters, mirrors,gratings, photocathode) was measured by using acalibrated quartz-halogen-tungsten lamp. Theemission spectrum of the lamp was collected by theLIF detection system, equipped with a microamperom-eter, and stored in the computer for spectral correc-tion.

In order to investigate actinic effects on the differ-ent LIF bands we used two kinds of actinic light: aHe-Ne laser (632.8 nm, 1 mW) and a high-pressureXe arc lamp (75 W) filtered by a short-wavelengthbandpass filter (< 480 nm). Both sources were mod-ulated at 2 kHz and irradiated the leaf target with across-sectional area of 0.5 cm2. High frequency-modulated signals from the photomultiplier tubehave been processed by a lock-in amplifier. In thesemeasurements the UV laser light was attenuated(1-10%) by quartz neutral-density filters. Outputs

from both integrators (boxcar and lock-in) were re-corded directly.

Fluorescence spectra from cell, protoplast, andchloroplast suspensions (all at 15 pug/mL of chloro-phyll) were obtained by using the same apparatus,replacing the leaf target with a laser-grade quartzcuvette, magnetically stirred; fluorescence was col-lected at 900.

For NADPH fluorescence measurements we usedan excimer-pumped dye laser to produce the excita-tion wavelength (340 nm) and a 250-W halogen lamp,filtered by a long-wavelength bandpass filter (> 600nm), as an actinic light source, directed on the cuvettetop by a fiber-optic cable. The same apparatus asbefore, equipped with a photomultiplier tube that ismore sensitive in the blue spectral region (Philips56DUVP), was used to detect the LIF at 460 nm.

Biological Material

Most of the measurements on intact leaves weremade on attached leaves. We used mainly olive(Olea Europea L., cv Ascolana) clonal plants thatwere propagated by cutting. The plants were grownfor 4 months in Levington compost and Agriperlite(1/1 vol./vol.) and supplemented fortnightly withHoagland's medium (Sigma H2395).

Measurements under stress conditions were madeon detached leaves: some leaves were left to dry outin Petri dishes until loss of turgor was noticed;photosynthesis was inhibited by putting leaves, for afew seconds, in a vacuum vessel in contact with asmall quantity of a 10-5 M solution of 3-(3,4-dichloro-phenyl)-1,1-dimethylurea (DCMU, Sigma D-7763).

Cells were obtained from leaf blades digested over-night in 1.2% cellulase and 0.8% pectase (Ozonuka,Yakult Pharmaceutical Industry Co., Ltd., Japan),purified on mannitol 0.8-0.6-M gradient (500 g x 5min) and suspended in 0.4-M mannitol, 125-mMCaCl2 , 5-mM KCl (5.6 pH). Protoplasts were ob-tained as described by Canlas et al.11 Final suspen-sions were obtained by using high purity reagents.

Intact chloroplasts were prepared in a Percollmedium gradient.12 Since the final suspension me-dium exhibits a strong blue fluorescence, immediatelybefore the LIF measurements were made chloro-plasts were washed and suspended in isotonic su-crose.

Spinach chloroplasts, both from the market andfrom freshly harvested spinach, were extracted accord-ing to Walker,13 washed in a cation-free medium,14

and resuspended in a conventional assay medium forC02-dependent oxygen evolution measurements. 13

This final suspension was checked for actinic increaseof blue fluorescence.

Results and Discussions

In order to check our apparatus we looked for work-ing conditions comparable with those that Chappelleand co-workers used, except that the use excitation at337 nm.6 We collected all our LIF spectra with theaim of investigating intraspectrum peak ratios, andwith respect to these differences only they can be

20 January 1993 / Vol. 32, No. 3 / APPLIED OPTICS 335

considemaximi

LIFdiffererevergrecompleleavession isloss of IchloropDCMUThis fuagreemthat h;Detaileitative lthe scol

As ftcernedthe leaicence e

It is paper)noticedwet puHash <tion suIobserveeven beimental

Fluoiis consicence ntissuespounds,the useBlue-grinae byuted tothat enchlorop

1.0

0. 8

0. 6

0. 4

0. 2

Fig. 2. Ascolana)(d) DCMLis 5 x 10of five mepower) rasystem an

red (we normalized each spectrum to its own We found a decrease of laser-induced green fluores-aum). cence and an increase of chlorophyll fluorescence inspectra obtained from intact leaves under Vicia Faba when epidermal tissue had been removed:it conditions are shown in Fig. 2. Woody this makes it evident that the epidermis is both aneen dicot plants showed the highest spectral important protection against UV radiation and atity; four major bands were obtained in olive possible topological source of green fluorescence.it 440, 530, 680, and 740 nm; blue/red emis- Most of the components that we have cited here werehigher in olive leaves than in medica leaves; recently discussed as Lichtenthaler et al.2 0 as possibleurgor enhances green emission as well as red blue emitters.'hyll bands with respect to the blue band; The hypothesis that some important links of theinfiltration strongly increases the red bands. photosynthetic chain could be identified in the blue-1 set of observations appears to be in good green fluorescence signal seems even more exciting:ent with similar experimental measurements in fluoesence of them snthetingave been reported by Chappelle et al.6 -9 in that case a new probe of the photosynthetic processI discussions of these spectra and their quanti- and apparatus could become availablerelations to stress detections are not within Indeed, far into the 1950's, blue fluorescence in-pe of this paper. creases caused by actinic light on algae and photosyn-ar as the blue-green spectral region is con- thetic bacteria have been related to the bound re-many molecular components and changes in duced phosphopyridine nucleotide (i.e., NADPH) and[ structure can cause and modulate fluores- used as a powerful tool to investigate mechanismsmission. and kinetics of photosynthesis.vell known that a cellulosic matrix (e.g., white Some identification attempts of blue-green fluores-fluoresces in the blue-green region, and we cence emitters in leaves were made by Chappelle andthat the emission intensity doubles when a co-workers9 on the basis of LIF emission of purified

re cellulosic matrix (Wathman n. 41 Fast, plant components in solution (NADPH, trans-beta-0.01%) is allowed to dry out. This observa- carotene, riboflavin, chlorophyll a, vitamin K). Re-

ggests that some fluorescence changes that are cently, these authors claimed to have identified!d on a leaf in water-stress conditions could NADPH, beta-carotene, and riboflavin as principaldue to some structural changes: more exper- fluorescence emitters in the blue-green spectral1 work is required to prove this point. range.22

'escence of plant tissues in this spectral region In order to investigate possible relations betweendered a serious problem in immunofluores- blue-green fluorescence excited at 308 nm and photo-Licroscopy investigation of lignified tissues and synthesis, we observed the time behavior of LIF atwith vacuolar deposits of phenolic com- different wavelengths in the presence of actinic lightand in some cases it has been eliminated by (Fig. 3). In these measurements the UV laser beam

of strong reducing substances, e.g., NaBH4.16 was used as a probe of the leaf intrinsic fluorescenceeen fluorescence detected in cell walls of Gram- changes that are induced by actinic light,O insensitiveUV fluorescence microscopy has been attrib- to light variations (background, actinic, fluorescence)bound ferulic acid.'7 Oxidized chlorophylls longer than 15 ns, and unsynchronized with the laser

it at 540 nm are released to vacuoles from pulse. By sinusoidal modulation of actinic light (2lasts.*8"9 kHz), simultaneous detection of variable fluorescence

induced by this light source was also possible, insensi-tive to variations with a different frequency or varia-tions shifted in phase with respect to the excitation

if 'X~t Ok / \ 4\ source.(a)^\ ( \ Jo/, \\'As we can see from Fig. 3, the actinic capacity of our

light sources was always confirmed by a typicalvariable fluorescence (Kautsky effect) on the chloro-phyll bands, as detected by the lock-in amplifier[traces (a) and (c)]; our laser probe alone (308 nm, 8ns, 10 Hz), although 10 times more powerful than

______________________________________ our actinic sources, was unable to produce similar400 45 0 50 0 550 6 I 0 6I O 70 0 75 80Cbehavior in the chlorophyll fluorescence (i.e., was not

400 450 500 5 5 0 600 6 5 0 700 750 800 actinic) at our detection limits.WAVELENGTH (Urn) Simultaneous excitation by one actinic source and

41F spectra on intact leaves: (a) olive (Olea europea cv. the UV laser probe [traces (b) and (d)] required; (b) medica (Medicago arborea); (c) dehydrated olive leaf; optimization of the probe intensity in order to maxi-infiltrated olive leaf. Laser intensity on the leaf target . f5 photons/(cm 2 x pulse). Each spectrum is an average mize the ratio between fluorescence variation that isasurements; each measurement is a (fluorescence/laser due to the actinic light and fluorescence induced bytio, corrected for the spectral response of our detection the probe itself [which is the first step fluorescenceid normalized to its maximum value. level on traces (b) and (d)].

336 APPLIED OPTICS / Vol. 32, No. 3 / 20 January 1993

I-I

II

I

time FLUORESCENCE WAVELENGTHS (mm)

30s 740 680 530 440

O o 0ff on off

O. off on ff o off ff

(b)

off

H ni o~~~~~~~~~ff o off

ff)of

Fig. 3. Actinic effect measurements on olive leaves (see text).The UV pulsed laser (308 nm) has been used as a probe. Laserintensity on the leaf target is 5 x 1014 photons/(cm2 x pulse): (a),(c) actinic-induced fluorescence signal (lock-in output); (b), (d) LIFsignal (boxcar output). In the blue-green spectral region theprobe intensity has been decreased in order to have a LIF signalcomparable with the signal at 680-740 nm. On and off indicatethe switching modes of the actinic light illumination; the first stepfluorescence level on traces (b) and (d) is the probe-inducedfluorescence.

If blue-green fluorescence is emitted by some ma-jor links of the photosynthetic chain, we shouldobserve the fluorescence intensity at these wave-lengths to increase or decrease after the addition ofactinic light. Of course, at these wavelengths nosignal can be present on trace (a) and possible fluores-cence signals could be masked by the strong emissionof the actinic source on trace (c). Nevertheless, ifpresent, it should be apparent and time discriminatedby the LIF [traces (b) and (d)]. Several checks havebeen made at different intensities of the laser probe:in our experimental conditions neither increase norquenching was detectable outside of chlorophyll fluo-rescence. We, therefore, presumed that most of theblue-green fluorescence emission from intact leaveswas not directly connected to the photosyntheticapparatus and masked eventual small actinic varia-tions.

Following the hypothesis that a strong contribu-tion to the blue-green fluorescence could be ascribedto extrachloroplastic substances, we collected LIFspectra of cell, protoplast, and chloroplast suspen-sions (Fig. 4).

According to our hypothesis, only a weak blue-green fluorescence signal was detectable in chloro-plasts. Moreover, the blue fluorescence was domi-nant in protoplasts with respect to green fluorescence,suggesting that a green contribution to the LIFspectrum can be carried by the cell wall.17

In order to discriminate between membrane andthe vacuolar blue contribution, one can modulate thevacuolar volume by gently shrinking protoplasts withKCl. Modulation of green/red fluorescence emission

1. 0

0.8

0.6

0.4

0 .2

0I I I I

400 450 500 550I I I I 1

600 650 700 750 800

WAVELENGTH (nH)

Fig. 4. LIF spectra of (a) cell, (b) protoplast, and (c) chloroplastsuspensions of olive leaves. Each spectrum is an average of threemeasurements, each measurement is a (fluorescence/laser power)ratio, corrected for the fluorescence spectrum of the suspensionmedium and the spectral response of our detection system andfinally normalized to its maximum value.

in protoplasts should appear by selective photo-oxidation of chlorophyll. However, it seems evidentto us that, with our apparatus, identification andcharacterization of the blue-green fluorescence thatis connected to photosynthesis should be done onactive chloroplast suspensions first.

Following the pioneering works of Duysens andAmesz,21 we searched for increasing fluorescence,excited at 340 nm and detected at 460 nm, on photo-synthesizing chloroplast suspensions. In Figs. 5 and6 we show the best results we obtained on spinachchloroplast suspensions: of course, the increase influorescence is a function of the quality of our suspen-sion (intactness and activity of the chloroplasts).

This first detection of photosynthetically reducedNADPH by LIF connects to the possibilities of time-resolved fluorescence.23 Excitation of chloroplast (oralgae) suspensions by a suitable picosecond systemand comparison of time-resolved fluorescence emis-sion in dark and light conditions should enable us toidentify the NADPH signal, in order to single it out inintact systems.

3

-DARK---LIGHT - DARK -

2

I .

I

Id

i , ,

0 1 2 3

TIME (in)

Fig. 5. Actinic effect on the LIF of spinach chloroplasts:wavelengths, 340-nm excitation, 460-nm detection; suspensionmedium according to Ref. 13; 0.015-mg/mL chlorophyll concentra-tion.

20 January 1993 / Vol. 32, No. 3 / APPLIED OPTICS 337

'A , *'/" W\ I(o)~ ~~~

'11 :"~ l.Ni/ I;~ '"'

IaIdIzI

I3

IP.

0o

400 500 600

WAVELENGTH (nm)

Fig. 6. LIF spectra of the same spinach chloroplast suspension asin Fig. 5 in dark and light conditions.

Conclusions

Most of the blue-green fluorescence that is inducedfrom intact leaves by laser excitation at 308 nm can beascribed to extracellular structures, cell envelopes,and vacuolar solutes. A large number of importantstress factors (drought, salt, photo-oxidation, etc.)can change the leaf structure, the vacuolar volume,and its content: in these cases blue-green fluores-cence variations could be a powerful diagnostic tool.

In vivo discrimination of the weak chloroplasticblue fluorescence has not been possible at 308-nmexcitation wavelength. Nevertheless, NADPH pro-duced in photosynthesis can be detected by laser-induced fluorescence exciting at 340 nm, in chlo-roplast suspensions. Time resolution of thisfluorescence in chloroplast, protoplast, and cell sus-pensions could provide information when we searchfor NADPH fluorescence on intact leaves.

Most of the reported results were presented at theInternational Workshop on the use of chlorophyllfluorescence and other noninvasive spectroscopic tech-niques in plant stress physiology that was held inWageningen, The Netherlands, 14-16 August 1989.

Cell, protoplast, and chloroplast suspensions fromolive leaves were prepared by G. Bongi and M.Mencuccini at Consiglio Nazionale delle Ricerche,Centro di Studi per l'Olivicoltura, Madonna Alta,Perugia, Italy, to whom I am thankful.

References1. W. L. Butler and M. Kitajima, "Fluorescence quenching in

photosystem II of chloroplasts," Biochim. Biophys. Acta 376,116-125 (1975).

2. M. Bradbury and N. R. Baker, "Analysis of the induction ofchlorophyll fluorescence in leaves and isolated thylakoids:contribution of photochemical and nonphotochemicalquenching," Proc. R. Soc. London Ser. B 220, 251-264 (1983).

3. U. Schreiber and U. Schliwa, "A solid-state, portable instru-ment for measurement of chlorophyll luminescence inductionin plants," Photosynth. Res. 11, 173-182 (1987).

4. H. K. Lichtenthaler, "In vivo chlorophyll fluorescence as a toolfor stress detection in plants," in Applications of ChlorophyllFluorescence, H. K. Lichtenthaler, ed. (Kluwer, Dordrecht,The Netherlands, 1988), pp. 129-142.

5. J. Barber, S. Malkin, and A. Telfer, "The origin of chlorophyll

fluorescence in vivo and its quenching by the photosystem tworeaction centre," Philos. Trans. R. Soc. London Ser. B 323,1-13 (1989).

6. E. W. Chappelle, F. M. Wood, Jr., J. E. McMurtrey, III, andW. W. Newcomb, "Laser-induced fluorescence of green plants.1: A technique for remote detection of plant stress andspecies differentiation," Appl. Opt. 23, 134-138 (1984).

7. E. W. Chappelle, D. L. Williams, R. F. Nelson, and J. E.McMurtrey, III, "Laser may help in remote assessment ofvegetation," Laser Focus 25(6), 123-132 (1989).

8. E. W. Chappelle, J. E. McMurtrey, III, F. M. Wood, Jr., andW. W. Newcomb, "Laser induced fluorescence of green plants.2: LIF caused by nutrient deficiencies in corn," Appl. Opt.23, 139-142 (1984).

9. E. W. Chappelle, F. M. Wood, Jr., W. W. Newcomb, and J. E.McMurtrey, III, "Laser induced fluorescence of green plants.3: LIF spectral signature of five major plant types," Appl.Opt. 24, 74-80 (1985).

10. M. Senorer, "The nanosecond decay of variable chlorophyllfluorescence in leaves of higher plants," Biochim. Biophys.Acta 849, 374-380 (1986).

11. L. A. Cafoas, A. M. Wyssmann, and M. C. Benbadis, "Isolation,culture and division of olive (Olea europea L.) protoplasts,"Plant Cell Rep. 6, 369-371 (1987).

12. G. Mourioux and R. Douce, "Slow passive diffusion andorthophosphate between intact isolated chloroplasts and sus-pending medium," Plant Physiol. 67, 470-473 (1981).

13. D. A. Walker, "Preparation of higher plant chloroplasts,"Methods Enzymol. 69, 94-104 (1980).

14. H. Y. Nakatani and J. Barber, "An improved method forisolating chloroplasts retaining their outer membranes," Bio-chim. Biophys. Acta 461, 510-512 (1977).

15. A. F. Theisen, "Fluorescence changes in a drying maple leafobserved in the visible and near-infrared," in Applications ofChlorophyll Fluorescence, H. K. Lichtenthaler, ed. (Kluwer,Dordrecht, The Netherlands, 1988), pp. 197-201.

16. R. B. Knox and M. B. Singh, "Immunofluorescence applica-tions in plant cells," in Botanical Microscopy, A. W. Robards,ed. (Oxford U. Press, London, 1985), pp. 205-232.

17. P. J. Harris and R. D. Hartley, "Detection of bound ferulicacidin cell walls of the Graminae by ultraviolet fluorescencemicroscopy," Nature (London) 259, 508-510 (1976).

18. B. A. Palevitz, D. J. O'Kane, R. E. Kobres, and N. V. Raikhec,"The vacuole system in stomatal cells of Allium. Vacuolemovements and changes in morphology and differentiatingcells as revealed by epifluorescence video and electronmicroscopy," Protoplasma 109,23-55 (1981).

19. P. Matile, S. Ginsburg, M. Shellenberg, and H. Thomas,"Cathabolites of chlorophyll in senescing barley leaves arelocalized in the vacuoles of mesophyll cells," Proc. Natl. Acad.Sci. USA 85,9529-9532 (1988).

20. H. K. Lichtenthaler, F. Stober, C. Buschmann, U. Rinderle,and R. Hak, "Laser-induced chlorophyll fluorescence and bluefluorescence of plants," in Proceedings of the InternationalGeoscience Remote Sensing Symposium (U. Maryland Press,Washington, D.C., 1990), Vol. 3, pp. 1913-1918.

21. L. N. M. Duysens and J. Amesz, "Fluorescence spectrophotom-etry of reduced phosphopyridine nucleotide in intact cells inthe near ultraviolet and visible region," Biochim. Biophys.Acta 24, 19-26 (1957).

22. E. W. Chappelle, J. E. McMurtrey, and M. S. Kim, "Laserinduced blue fluorescence in vegetation," in Proceedings of theInternational Geoscience Remote Sensing Symposium (U.Maryland Press, Washington, D.C., 1990), Vol. 3, pp. 1919-1922.

23. Y. Goulas, I. Moya, and G. Schmuck, "Time-resolved spectros-copy of the blue fluorescence of spinach leaves," Photosynth.Res. 25, 299-307 (1990).

338 APPLIED OPTICS / Vol. 32, No. 3 / 20 January 1993

>I

;1

~4

400 500 600

2