safranine fluorescent staining of wood cell walls

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Universidad Nacional de Colombia - Free Trial]On: 2 January 2009Access details: Access Details: [subscription number 905131491]Publisher Informa HealthcareInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Biotechnic and HistochemistryPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713692932

Safranine fluorescent staining of wood cell wallsJ. Bond a; L. Donaldson a; S. Hill b; K. Hitchcock b

a Cellwall Biotechnology Centre, b Biomaterials Engineering, Scion, Rotorua, New Zealand

First Published:June2008

To cite this Article Bond, J., Donaldson, L., Hill, S. and Hitchcock, K.(2008)'Safranine fluorescent staining of wood cell walls',Biotechnicand Histochemistry,83:3,161 — 171

To link to this Article: DOI: 10.1080/10520290802373354

URL: http://dx.doi.org/10.1080/10520290802373354

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://www.informaworld.com/smpp/title~content=t713692932

http://dx.doi.org/10.1080/10520290802373354

http://www.informaworld.com/terms-and-conditions-of-access.pdf

Safranine fluorescent staining of wood cell walls

J Bond1, L Donaldson1, S Hill2, K Hitchcock2

1Cellwall Biotechnology Centre, 2Biomaterials Engineering, Scion, Private Bag 3020, Rotorua, New Zealand

Submitted September 10, 2007; accepted January 28, 2008

Abstract

Safranine is an azo dye commonly used for plant microscopy, especially as a stain for lignifiedtissues such as xylem. Safranine fluorescently labels the wood cell wall, producing green/yellowfluorescence in the secondary cell wall and red/orange fluorescence in the middle lamella (ML)region. We examined the fluorescence behavior of safranine under blue light excitation using avariety of wood- and fiber-based samples of known composition to interpret the observed colordifferentiation of different cell wall types. We also examined the basis for the differences influorescence emission using spectral confocal microscopy to examine lignin-rich and cellulose-rich cell walls including reaction wood and decayed wood compared to normal wood. Our resultsindicate that lignin-rich cell walls, such as the ML of tracheids, the secondary wall of compressionwood tracheids, and wood decayed by brown rot, tend to fluoresce red or orange, while cellulose-rich cell walls such as resin canals, wood decayed by white rot, cotton fibers and the G-layer oftension wood fibers, tend to fluoresce green/yellow. This variation in fluorescence emissionseems to be due to factors including an emission shift toward red wavelengths combined with dyequenching at shorter wavelengths in regions with high lignin content. Safranine fluorescenceprovides a useful way to differentiate lignin-rich and cellulose-rich cell walls without counter-staining as required for bright field microscopy.

Key words: cell walls, cellulose, fluorescence, lignin, safranine, wood

Plant cell walls are composed of cellulose, non-cellulosic polysaccharides (hemicelluloses and pec-tins) and frequently lignin. The typical tracheid cellwall is made up of a primary cell wall and threesecondary cell wall layers (S1, S2, S3), which differ inthe orientation of their cellulose microfibrils(Donaldson and Xu 2005) and their degree oflignification (Donaldson 2001, Harris 2006). In allthese layers, the cellulose microfibrils are alignedwithin a matrix of lignin and hemicelluloses (Kerrand Goring 1975). Between each cell is the middlelamella (ML), which contains lignin and non-cellu-losic polysaccharides (Harris 2006). In leaning trees,a special type of wood called reaction wood forms;this is known as compression wood in conifers andtension wood in dicotyledons. In compression woodtracheids, lignin increases in the outer part of the

secondary cell wall (S2L), but is reduced in the ML(Donaldson et al. 2004). In tension wood fibers, agelatinous, or G-layer, forms as the innermost layerand consists mostly of cellulose (Bentum et al. 1969,Joseleau et al. 2004, Pilate et al. 2004, Daniel et al.2006, Gierlinger and Schwanninger 2006). Fungus-degraded wood also can contain cell walls ofdifferent composition. White rot fungi degradelignin and produce partially delignified cell walls,while brown rot fungi remove the cellulose and leavea lignin skeleton with little or no polysaccharideresidues (Eriksson et al. 1990). Similar degradationcan occur as a result of weathering mainly due toexposure to UV radiation in sunlight, which de-grades lignin in the cell walls near exposed woodsurfaces (George et al. 2005).

Despite the increasing number of methods forspecifically visualizing compounds within thewood cell wall, such as antibodies and carbohy-drate binding modules (Willats et al. 2000, Hosooet al. 2002, Joseleau et al. 2004, Daniel et al. 2006,McCartney et al. 2005, McCartney et al. 2006), dyesstill are commonly used to screen quickly and

Address for correspondence: J Bond, 6A Willow Ar, Hannahs

Bay, Rotorue, New Zealand. E-mail: [email protected].

– Biological Stain Commission

Biotechnic & Histochemistry 2008, 83(3�4): 161�171.

DOI:10.1080/10520290802373354 161

Downloaded By: [Universidad Nacional de Colombia - Free Trial] At: 14:02 2 January 2009

inexpensively for chemical differences among andwithin different cell wall types (Stockert et al. 1984,Hogetsu 1990, Knebel and Schnepf 1991, Ma et al.1993, Graham and Joshi 1995, Mori and Bellani1996, Vazquez-Cooz and Meyer 2002, Li and Reeve2004, Donaldson and Bond 2005). One of the mostcommon stains for botanical tissue is safranine(safranine O) (Ma et al. 1993, Vazquez-Cooz andMeyer 2002). This dye stains lignin, chromosomes,nucleoli, cutin, extracts (resins and gums) and cork(suberin) (Johansen 1940, Kasten 1989, Ruzin 1999,Horobin 2002). When used for plant material,safranine typically is used in conjunction with acounterstain such as fast green (Johansen 1940,Berlyn and Miksche 1976, O’Brien and McCully1981, Ma et al. 1993) or astra blue (Srebotnik andMessner 1994, Vazquez-Coos and Meyer 2002), sothat lignified tissue is stained red while unlignifiedtissues are stained in green or blue.

Traditionally, plant tissues stained with dyes areexamined using transmitted light; however, manydyes often are also fluorescent (Haseloff 2003).Owing to improvements in image resolution andcontrast provided by confocal fluorescence micro-scopy compared to wide field transmitted lightimaging, superior results can be achieved for arange of plant tissues, including wood, using thistechnique (Knebel and Schnepf 1991, Donaldsonand Lausberg 1998, Hepler and Gunning 1998,Haseloff 2003, Angeles et al. 2004, Li and Reeve2004, Donaldson and Bond 2005). Some cell wallcomponents, such as lignin, autofluoresce withmaximum absorbance in the UV range and requireno dye (Olmstead and Gray 1997, Albinsson et al.1999, Donaldson et al. 1999, 2004, Donaldson andBond 2005).

In general, dyes have specific excitation andemission wavelengths, but the chemical environ-ment, e.g., pH, may lead to changes in fluorescenceemission. Changes in fluorescence also may occurowing to increasing dye�dye or dye�substrateinteractions resulting in quenching and/or Forsterresonance energy transfer (FRET). Quenching(Lakowicz 2006) occurs when the fluorescenceintensity of the sample decreases with increasingdye concentration, as substrate concentration in-creases, or with the addition of a quenching agentthat binds or associates directly with the fluoro-phore. Because quenching usually is not advanta-geous, the optimal dye concentration and stainingtime must be found using a dilution series. FREToccurs when the emission spectrum of one dyemolecule (donor) overlaps with the absorbancespectrum of another dye molecule (acceptor)(Periasamy and Day 2005) resulting in loss of

fluorescence at shorter wavelengths and increasedfluorescence at longer wavelengths. In some cases,the donor and acceptor can be the same dye(Periasamy and Day 2005).

We examine here the fluorescence behavior ofwood cell walls stained with safranine to under-stand observed color differentiation among cellwall and tissue types.

Materials and methods

Wood samples and staining

Blocks of Pinus radiata D. Don (radiata pine) normaland compression wood, Pseudotsuga menziesii(Mirb.) Franco (Douglas fir), and Populus deltoidesBartr. ex Marsh. (poplar) tension wood were cutapproximately 1 cm2 blocks and fixed in formalinacetoalcohol (FAA; 5% formaldehyde, and 5% aceticacid in 95% ethanol) until they were used. Severaltypes of fungal-degraded wood also were examinedincluding brown rot-decayed radiata pine (cellulosespecific fungus species unknown) and two types ofwhite rot. White rot-degraded samples of radiatapine were produced by treating wood blocks with apure culture of Trametes versicolor (a non-selectivedegrader). Samples of Douglas fir degraded byPhellinus pini (lignin-specific degrader) were col-lected. Artificially weathered radiata pine wasproduced by treatment in a weathering chamber(Atlas Wi65 weatherometer (Atlas Electric DevicesCo., Chicago, IL)) using a cycle of 102 min ofultraviolet irradiation followed by 18 min of waterspray for a 3000 h exposure period.

Transverse sections of the wood samples wereprepared at 60 mm thickness using a sledge micro-tome; the block was kept wet with water duringsectioning. Wood sections and cotton fibers werestained in 0.1% aqueous safranine (safranin O, CI50240, #34067 B (BDH, Poole, England)) for5�10 min, and rinsed three times for 10 min eachin warm water at 30�40oC. Sections then were ovendried and mounted in immersion oil for fluores-cence microscopy. Cotton fibers were prepared as awhole mount. Sections were examined as quicklyas possible to avoid background staining caused bysafranine leaching from the sections.

A safranine dilution series was prepared usingwood sections stained for 10 min with 0.1, 0.02, 0.005or 0.002% safranine. Wood sections also werestained for 10 min with 0.1% safranine, then de-stained by washing in 50% ethanol for either 1 or 5 h.To test the effect of pH on safranine fluorescence,dye was dissolved in water, in water adjusted to

162 Biotechnic & Histochemistry 2008, 83(3�4): 161�171


pH 3 with HCl, or water adjusted to pH 11 withNaOH. Wood was stained as described above.

Microscopy

Conventional (filter based) confocal fluorescenceimaging was carried out using a Leica TCS NTconfocal microscope, while spectral confocal ima-ging was performed using a Leica SP2 spectralconfocal microscope (Leica Microsystems, Wetzlar,Germany). Sections stained with safranine wereexcited at 488 nm and imaged using a BP530 bandpass filter (30 nm band width) and an LP590 longpass filter (] 590 nm) for conventional confocalmicroscopy. Lignin autofluorescence was excitedwith 488/568 nm radiation and imaged using aBP530/LP590 filter set using slow scan speed andhigh laser power. To determine whether ligninautofluorescence contributed to the fluorescencesignal observed when wood was stained withsafranine, we examined a piece of unstained woodusing the same gain settings as used for safranineimaging. In mature radiata pine, autofluorescencewas nine times weaker than safranine stained tissueand therefore did not contribute to safranine fluo-rescence at our settings. All images were collected at1024� 1024 pixels using 4�32 frame averages.

For spectral scanning of normal, compressionand white rot-degraded wood, samples werestained with 0.1% safranine and excited at488 nm, and 20 fluorescence images were collectedat 10 nm intervals from 500�700 nm. For thesafranine dilution series, 30 fluorescence imageswere collected at 10 nm intervals from 500�700 nm.Images were 512� 512 pixels with four frameaverages. After collecting the images, regionswithin the cell wall were selected for spectralanalysis. The ML, S2L, and inner S2 region of whiterot-degraded wood (S2i) were selected to representhigh lignin and low cellulose content. The S2 regionof normal wood and the ML of compression woodwere used to represent low lignin levels. For thedilution series, spectral analysis was only per-formed on the S2L region of compression woodtracheids. For all confocal imaging, we tried tomatch the red/green overlay color to that seenwhen looking through the eye piece with a greenlong pass filter set. This usually was achieved bysaturating the gain on each detector channel forconventional confocal imaging.

Results

Safranine dye fluorescently labeled the wood cellwall producing green/yellow fluorescence in the

secondary cell wall and red/orange fluorescence inthe ML region in normal pine and the S2L region ofcompression wood (Fig. 1). Our filter sets assign515�545 nm to the green channel and wavelengthsover 590 nm to the red channel, giving an approx-imation of the true color of the sample seen whenusing a green long pass filter in wide fieldfluorescence. The lmax for safranine dye solutionwas 588 nm; therefore, the majority of fluorescencewas collected in the red channel. Separation of redand green emission channels (Fig. 1B,C,E,F)showed two different intensity profiles: the major-ity of the green fluorescence was in the secondarycell wall (Fig. 1B,E), and the red fluorescence wasin the more highly lignified ML (Fig. 1C, arrows),S3 regions (Fig. 1C) and in the S2L region ofcompression wood (Fig. 1F). Lignin autofluores-cence of unstained tissue (Fig. 2) showed highlignin content in the ML, which correlated withfluorescence seen in the red channel after safraninestaining (Fig. 1C). Lignin autofluorescence wasapproximately nine times weaker than safraninefluorescence, and at the settings used to imagesafranine fluorescence, there was no significantcontribution from lignin autofluorescence.

Chemical composition

To determine whether the safranine green/redfluorescence profiles shown in Fig. 1 are relatedto the composition of the cell wall, we examined arange of cell wall types of varying composition asshown earlier by chemical analysis (Fengel andWegener 1983). Cell wall types that predominantlyfluoresce red, such as poplar ray cells (Fig. 3A),radiata pine compression wood S2L (Fig. 3B),brown rot-degraded wood (Fig. 3C), white rot-degraded inner secondary cell wall (Fig. 3D) andthe ML region of tracheids (Fig. 3A,D,E,F and H)are all high in lignin and/or low in cellulose. Thesensitivity of safranine staining is shown in whiterot-degraded wood (Fig. 3D). This fungus isreported to be a non-selective cell wall degrader;however, safranine staining showed fluorescencedifferences in the secondary cell wall. Degradationstarted from the lumen side of the S2 layer andshowed a deep red band in the inner S2 region (Fig.3D). At the top of the image, most of the S2 layerwas degraded leaving a thin red residual cell wallregion. Extract found in the resin canals of radiatapine also showed red fluorescence (Fig. 3H); that isnot surprising given their polyphenolic composi-tion, which is generically similar to lignin (Fengeland Wegener 1983).

Safranine cell wall staining 163


Fig. 1 (Continued)



Cell wall types that predominantly fluorescegreen, such as the inner secondary wall layer ofpoplar tension wood fibers, known as the gelatinouslayer (Fig. 3A), white rot-degraded Douglas fir(Fig. 3E), resin canal cells (Bamber 1972)(Fig. 3H), artificially weathered radiata pine wood(Fig. 3F), and cotton fibers (Fig. 3G), are all low inlignin and/or high in cellulose. A gradient can beseen from green to red at the transition fromweathered to normal wood, showing a transitionfrom low to normal levels of lignin. The secondarycell walls of poplar (Fig. 3A), radiata pine compres-sion wood (Fig. 3B), non-decayed or partiallydecayed regions of brown rot- (Fig. 3C, betweenarrows) and white rot-decayed wood (Fig. 3D), non-weathered cell walls (Fig. 3F) and radiata pinenormal wood (Fig. 3H) all have a mixture of ligninand cellulose and these cell walls appeared varyingshades of yellow.

Quenching

To determine whether quenching affects the fluo-rescence properties of safranine, radiata pinecompression wood was stained with 0.1%, 0.02%,0.005% and 0.002% safranine (Fig. 4). This dye

concentration series showed that the red/greenfluorescence pattern that characterizes differentcell wall regions/types became less distinct withdecreasing safranine concentration. At 0.002%safranine, the color difference between high andlow lignin regions was lost and the cell wallfluoresced uniformly green/yellow with only in-tensity differences among cell wall layers (Fig.4D). Within the S2L region, green fluorescenceincreased relative to the rest of the cell wall assafranine concentration is reduced, suggesting thatquenching occurs within this region at greenwavelengths, but not at red wavelengths whereintensity declined equally in all cell wall regionswith reducing safranine concentration. This isshown graphically in Fig. 4 (left side) where greenand red lines represent intensity in the green andred channels. The white lines in Fig. 4 (right side)show where the intensity profiles for green andred channels were measured. The intensity profilefor the red and green channels with 0.1% safra-nine shows the high fluorescence of the S2L in thered channel compared to the green channel, whileat 0.002% both profiles are the same.

Compression wood also was stained with 0.1%safranine, then destained for 1 or 5 h and showed

Fig. 1. Confocal fluorescence images of radiata pine normal and compression wood stained with safranine. (A) Normalwood, red/green channel overlay; (B) Normal wood, green channel fluorescence, highlighting secondary cell walls;(C) Normal wood, red channel fluorescence, highlighting the ML and S3 layers. Arrows indicate ML and S3 layer;(D) Compression wood, red/green channel overlay; (E) Compression wood, green channel fluorescence, highlighting innersecondary walls and ML; (F) Compression wood, red channel fluorescence, highlighting S2L region of the secondary cellwall. Arrows indicate S2L region. Scale bar�40 mm.

Fig. 2. Autofluorescence image of radiata pine normal wood. High fluorescence intensity corresponds to high lignincontent. ML, middle lamella. Scale bar�40 mm.



Fig. 3. Safranine staining of wood types with varying levels of lignin and polysaccharides. (A) Poplar tension wood. Thegelatinous layer (G) is predominantly made up of cellulose. The ray cell (R) and ML are high in lignin; (B) Radiata pinecompression wood. The S2L region of the secondary cell wall has high lignin content; (C) Brown rot-degraded radiata pine.Arrows indicate a region of partially degraded wood containing crystalline cellulose, whereas the surrounding cells havelittle or no cellulose; (D) Trametes versicolor-degraded radiata pine; non-selective white rot. The inner cell wall (S2i) hasaltered chemistry as shown by the deep red fluorescence; the outer S2 (S2o) appears unaffected. The asterisk shows anundegraded cell; (E) Phellinus pini degraded Douglas fir; white rot, which selectively removes lignin. Arrows show lignin-degraded regions of the cell wall; (F) UV-degraded (artificially weathered) wood. Arrows show the exposed region at thewood surface where lignin has been removed by artificial weathering; (G) Cotton fibers; (H) A resin canal in radiata pinewood (RC). Resin canal cell walls do not contain lignin and fluoresce green when stained with safranine while extract (E)fluoresces red. Scale bars�40 mm.



Fig. 4. Radiata pine compression wood stained with (A) 0.1, (B) 0.02, (C) 0.005 and (D) 0.002% safranine. At 0.1%safranine, green fluorescence quenches in the S2L, as the safranine solution decreases in concentration the S2L becomesmore fluorescent in the green channel. At 0.002%, green and red fluorescence show the same fluorescence profile. Thegraphs to the left of each picture represent red (red line) and green (green line) intensity across the corresponding whiteline in the image. Scale bar�20 mm.



the same changes in fluorescence as describedabove (data not shown). After 1 h destaining,red/green color differentiation was still present,but after 5 h, both channels were similar andresulted in uniform green/yellow fluorescence.This confirms that quenching is at least partlyresponsible for the color differentiation observedat high safranine concentration.

Spectral imaging

Confocal fluorescence microscopy showed an ap-proximation of the true fluorescence color of thesample by assigning colors to the different detectorchannels (Fig. 1). We used spectral confocal ima-ging to determine whether the fluorescence colordifferences observed in the different regions of thecell wall are related to changes in the fluorescencespectrum of safranine bound to wood.

Wood was stained with 0.1%, 0.02%, 0.005% and0.002% safranine and the spectrum for each wasmeasured in the S2L region (Fig. 5). At 0.1%safranine, the S2L region had a lmax of 581 nm,which was close to the 588 nm lmax of the 0.1%safranine solution. The lmax of the S2L cell wallstained with the three lower safranine concentra-tions was 567 nm. Although the lmax for the lastthree concentrations was the same, the areasunder the curves of the last two dilutions (0.005and 0.002%) were shifted farther toward the greenend of the spectrum. As dye concentration wasreduced from 0.1 to 0.002%, fluorescence intensityincreased, with the maximum fluorescence inten-sity occurring at 0.005% safranine concentration(Fig. 5).

Comparison of fluorescence spectra from regionswithin the cell wall in compression wood, normal

wood and white rot-degraded wood stained with0.1% safranine (Fig. 6) showed a shift of lmax tolonger wavelengths (toward the red end of thespectrum) with increasing concentration of lignin.In normal radiata pine (Fig. 6A), the lmax of the MLwas 585 nm compared to 575 nm in the S2 region.In compression wood, the S2L region of thesecondary wall was shifted 10 nm from that inthe S2 region to 585 nm, and the ML was reduced10 nm to 575 nm, matching the S2 region of normalwood. The inner cell wall (S2i) of white rot-degraded wood was highly autofluorescent, sug-gesting a high lignin content, and fluoresced faintlyred with safranine staining (Fig. 3D, arrow) withthe lmax at 595 nm. The outer S2 region (S2o)appeared similar to undegraded wood.

Fig. 5. Fluorescence spectra taken from the S2L region ofcompression wood. Wood samples were stained with0.1%, 0.02%, 0.005%, and 0.002% safranine. The safra-nine staining solution was a 0.1% safranine solution inwater.

Fig. 6. Fluorescence spectra of wood stained with 0.1%safranine. (A) Safranine stained normal wood; (B) com-pression wood and (C) white rot-degraded radiata pine.Regions selected were high in lignin and low in poly-saccharides (A, ML; B, S2L; C, S2i) or low in lignin andhigh in polysaccharides (A, B, S2; C, ML compressionwood).



Both dye concentration and the amount of ligninaffected the fluorescence spectrum of safraninestained wood. The fluorescence spectrum in re-gions of high lignin was shifted toward redwavelengths and may also have quenched, thusfurther reducing the green fluorescence signal.Regions of low lignin fluoresced more toward thegreen end of the spectrum and showed lessquenching.

Discussion

Safranine staining reveals differences in chemicalcomposition and can therefore be used as an easymethod for screening samples for the presence ofreaction wood, fungal degradation or variations incell wall composition resulting from mutation orgenetic modification. The differential red/greenfluorescence of cell walls with high lignin or highcellulose eliminates the need to counterstain withother dyes (Ma et al. 1993, Srebotnik and Messner1994, Vazquez-Cooz and Meyer 2002), with theadded advantage of improved resolution of con-focal fluorescence microscopy over bright fieldlight microscopy. The counterstains commonlyused with safranine, such as fast green or astrablue, are not fluorescent, so cannot be used withsafranine for fluorescence microscopy.

Owing to the change in fluorescence emission,safranine can differentiate high and low levels oflignin. Regions of high lignin fluoresce red/orange,while regions with low lignin fluoresce green/yellow. Quenching, in combination with a shifttoward red wavelengths would explain why theS2L region of compression wood appears redcompared to the green/yellow appearance of theadjacent S2 region. This color effect occurs onlywhen the dye is quenching; therefore, both dyeconcentration and staining time must be tested inadvance to give the best effect. The spectral shift tolonger wavelengths when safranine binds ligninmay be the result of FRET (Periasamy and Day2005), because the prerequisite of overlappingabsorbance and emission spectra and close proxi-mity of dye molecules (Donaldson and Bond 2005)fits our observations. The nature of safraninebinding to wood has been reported by Stockertet al. (1984). These investigations report that safra-nine forms stacked p-p interactions with lignin byslotting between the aromatic residues. Less isknown about the interaction of safranine withpolysaccharides.

Safranine also stains polysaccharides, as shownby examination of artificially weathered wood inwhich lignin is removed from the surface tracheidsleaving a cellulose residue, and by the G-layer oftension wood, which contains mainly cellulose.Safranine interacts in some way with polysacchar-ides, because if it were unbound within the cellwall, we would expect it to show the maximumshift to the red end of the spectrum as shown byunbound safranine solution (Fig. 5, 0.1% safraninesolution), which is not the case. In general, safra-nine association with wood is weak and the dye isreadily leached from cell walls by alcohols (Vaz-quez-Cooz and Meyer 2002). This also precludesthe use of glycerol as a mounting medium. Safra-nine is a basic dye and thus it is attracted to acidicsites by polar bonding (Rost 1995).

Fungal-degraded wood is a useful way to char-acterize safranine fluorescence. The fungi we usedwere characterized as selective lignin, polysacchar-ide, or general cell wall degraders. Using thesesamples, we could show all stages of degradationwithin a single section. The high sensitivity ofconfocal imaging of safranine stained wood wasshown with the white rot fungus Trametes versicolor.This fungus has been reported to degrade all cellwall components equally; however, there was adark red layer around the inner S2 (S2i) aftersafranine staining. This zone was also highlyautofluorescent (data not shown) and thereforelikely to be high in lignin, reflecting an initial lossof cellulose followed by the collapse of the residuallignin-rich secondary cell wall.

Pear sclereids have been stained with safranineand also show some red/green fluorescence differ-entiation (Angeles et al. 2004). These investigatorsused 1% safranine and stained for at least 20 min.Based on our results, we would expect theseconditions to result in even greater quenching asshown by the concentration of fluorescence on cellwall surfaces including the lining of pit canals.Angeles et al. (2004) reported that the greenfluorescence was more specific to lignin than thered fluorescence, but this may be due to therelatively weak lignification of the sclereids. Otherstudies reported the lmax of safranine to beapproximately 530�540 nm (Durrenberger et al.2001, Horobin 2002) or 590 nm (Kasten 1989)compared to 588 nm as reported here. Durrenber-ger et al. (2001) found that safranine has anemission maximum of 540 nm when staining starchgrains and primary walls in foods such as yamsand bread. This value agrees with our observations



of the predominantly green fluorescence of un-lignified cell walls. The chemical environment ofsafranine may have a large influence on its fluores-cence emission characteristics. Haseloff (2003) alsoreported different results with safranine fluores-cence after embedding in paraffin, where primarycell walls fluoresced red rather than green. Thus, itis important to understand the effects of one’sexperimental procedures when using this dye.

The color differentiation among cell wall typesshown by safranine fluorescence is due to thecombined effects of quenching and a spectral shifttoward red emission when safranine is associa-ted with lignin-rich cell walls. Reduction of dyeconcentration or leaching of dye from the sampleremoves this color differentiation. The observedeffects are compatible with the theory that safra-nine molecules undergo FRET with each other inareas of heavy staining where the dye moleculesmay be in close proximity to each other. The colordifferentiation eliminates the need for counterstain-ing with fast green or astra blue as used for brightfield transmission light microscopy and thus pro-vides a useful method for determining the relativeamounts of lignin and cellulose in wood cell walls.

Acknowledgments

The authors thank Jacqui Ross, Department ofAnatomy with Radiology, Medical and HealthSciences, University of Auckland, New Zealandfor her help with spectral confocal microscopy; IanHood from Ensis Forest Biosecurity and Protection,SCION, New Zealand for providing fungal de-graded wood; and Tim Strabala from CellwallBiotechnology Centre, SCION, New Zealand forcritical review of the manuscript.

References

Albinsson B, Li SM, Lundquist K, Stomberg R (1999)The origin of lignin fluorescence. J. Mol. Struct. 508:19�27.Angeles G, Owens SA, Ewers FW (2004) Fluorescenceshell: a novel view of sclereid morphology with theconfocal laser scanning microscope. Microsc. Res. Techn.63: 282�288.Bamber RK (1972) Properties of the cell walls of the resincanal tissue of the sap wood and heart wood of Pinuslambertiana dougl. and P radiata d.don. J. Inst. Wood Sci. 6:32�35.Bentum ALK, Cote WA, Day AC, Timell TE (1969)Distribution of lignin in normal and tension wood. WoodSci. Technol. 3: 218�231.Berlyn PG, Miksche JP (1976) Botanical Microtechniqueand Cytochemistry. The Iowa State University Press, Ames,IA. p. 98.

Daniel G, Filonova L, Kallas AM, Teeri TT (2006)Chemical and morphological characterization of the G-layer in tension wood fibers of Populus tremula and Betulaverrucosa: labelling with cellulose-binding module CBM1HjCel7A and fluorescence and FESEM microscopy. Holz-forschung 60: 618�624.Donaldson LA (2001) Lignification and lignin topochem-istry � an ultrastructural view. Phytochemistry 57: 859�873.Donaldson LA, Bond J (2005) Fluorescence Microscopy ofWood. Scion, Rotorua, New Zealand. CD-ROM.Donaldson LA, Grace J, Downes GM (2004) Within-treevariation in anatomical properties of compression woodin radiata pine. IAWA J. 25: 253�271.Donaldson LA, Lausberg MJF (1998) Comparison ofconventional transmitted light and confocal microscopyfor measuring wood cell dimensions by image analysis.IAWA J. 19: 321�336.Donaldson LA, Singh AP, Yoshinaga A, Takabe K (1999)Lignin distribution in mild compression wood of Pinusradiata. Can. J. Bot. 77: 41�50.Donaldson LA, Xu P (2005) Microfibril orientation acrossthe secondary cell wall of radiata pine tracheids. Trees 19:644�653.Durrenberger MB, Handschin S, Conde-Petit S, EscherF (2001) Visualisation of food structure by confocal laserscanning microscopy (CLSM). Lebensm-Wiss. Technol. 34:11�17.Eriksson K-E, Blanchette RA, Ander P (1990) Microbialand enzymatic degradation of wood and wood components.Springer Verlag, Heidelberg. pp. 1�416.Fengel D, Wegener G (1983) Wood: Chemistry, Ultrastruc-ture, Reactions. De Gruyter, New York. pp. 1�613.George B, Suttie E, Merlin A, Deglise X (2005) Photo-degradation and photostabilisation of wood-the state ofthe art. Polym. Deg. Stab. 88: 268�274.Gierlinger N, Schwanninger M (2006) Chemical ima-ging of poplar wood cell walls by confocal Ramanmicroscopy. Plant Physiol. 140: 1246�1254.Graham ET, Joshi PA (1995) Novel fixation of planttissue, staining through paraffin with alcian blue andhematoxylin, and improved slide preparation. Biotech.Histochem. 70: 263�266.Harris PJ (2006) Primary and secondary plant cell walls:a comparative overview. NZ J. For. Sci. 36: 36�53.Haseloff J (2003) Old botanical techniques for newmicroscopes. BioTechniques 34: 1174�1182.Hepler PK, Gunning BES (1998) Confocal microscopy ofplant cells. Protoplasma 201: 121�157.Hogetsu T (1990) Detection of hemicelluloses specificto the cell wall of tracheary elements and phloemcells by fluorescein-conjugated lectins. Protoplasma 156:67�73.Horobin RW (2002) Conn’s Biological Stains: A Handbook ofDyes, Stains and Fluorochromes for Use in Biology andMedicine. 1st ed. BIOS Scientific Publ, Oxford. pp. 1�502.Hosoo Y, Yoshida M, Imai T, Okuyama T (2002) Diurnaldifference in the amount of immunogold-labelled gluco-mannans detected with field emission scanning electronmicroscopy at the innermost surface of developingsecondary walls of differentiating conifer tracheids.Planta 215: 1006�1012.



Johansen DA (1940) Plant Microtechnique. McGraw-Hill,New York. pp. 1�523.Joseleau JP, Imai T, Kuroda K, Ruel K (2004) Detectionin situ and characterisation of lignin in the G-layer oftension wood fibers of Populus deltoides. Planta 219: 338�345.Kasten FH (1989) The origins of modern fluorescentmicroscopy and fluorescent probes. In: Cell Structure andFunction by Microspectrofluorometry, Kohen E, HirschbergJG, Eds. Academic Press, San Diego, CA. pp. 3�5.Kerr AJ, Goring DAI (1975) The ultrastructural arrange-ment of the wood cell wall. Cellulose Chem. Technol. 9:563�573.Knebel W, Schnepf E (1991) Confocal laser scanningmicroscopy of fluorescently stained wood cells: a newmethod for three-dimensional imaging of xylem ele-ments. Trees 5: 1�4.Lakowicz JR, Ed. (2006) Principles of Fluorescence Spectro-scopy. Springer, New York. p. 17.Li K, Reeve DW (2004) Fluorescent labeling of lignin inthe wood pulp fiber wall. J. Wood Chem. Technol. 24: 169�181.Ma Y, Sawhney VK, Steeves TA (1993) Staining ofparaffin-embedded plant material in safranine and fastgreen without prior removal of the paraffin. Can. J Bot. 71:996�999.McCartney L, Blake WA, Flint J, Bolam DN, BorastonAB, Gilbert HJ, Knox JP (2006) Differential recognitionof plant cell walls by microbial xylan-specific carbohy-drate-binding modules. Proc. Nat. Acad. Sci. 103: 4765�4770.McCartney L, Marcus SE, Knox JP (2005) Monoclonalantibodies to plant cell wall xylans and arabinoxylans.J. Histochem Cytochem. 53: 543�546.

Mori B, Bellani LM (1996) Differential staining forcellulosic and modified plant cell walls. Biotech. &Histochem. 71: 71�72.O’Brien TP, McCully ME (1981) The study of PlantStructure. Principles and Selected Methods. ThermacarphiPty. Ltd, Melbourne, Australia. pp. 6.83�6.84.Olmstead JA, Gray DG (1997) Fluorescence spectro-scopy of cellulose, lignin and mechanical pulps: a review.J. PulpPaper Sci. 23 12: 571�581.Periasamy A, Day RN (2005) Molecular Imaging, FRETMicroscopy and Spectroscopy. Oxford University Press,New York. pp. 1�312.Pilate G, Chabbert B, Cathala B, Yoshinaga A, Leple J-C, Laurans F, Lapierre C, Ruel K (2004) Lignification andtension wood. CR Biol. 327: 889�901.Rost FWD (1995) Fluorescence Microscopy, Vol. II. Cam-bridge University Press, UK, Cambridge. pp. 1�266.Ruzin SE (1999) Plant Microtechnique and Microscopy.Oxford University Press, New York. pp. 1�322.Srebotnik E, Messner K (1994) A simple method thatuses differential staining and light microscopy to assessthe selectivity of wood delignification by white rot fungi.Appl. Env. Microbiol. 60: 1383�1386.Stockert JC, Canete M, Colman OD (1984) Histologicalmechanism for the orthochromatic staining and fluores-cence reaction of lignified tissues. Cell. Mol. Biol. 30: 503�508.Vazquez-Cooz I, Meyer RW (2002) A differential stain-ing method to identify lignified and unlignified tissues.Biotech. Histochem. 77: 277�282.Willats WGT, Steele-King CG, McCartney L, Orfila C,Marcus SE, Knox JP (2000) Making and using antibodyprobes to study plant cell walls. Plant Physiol. & Biochem.38: 27�36.



safranine fluorescent staining of wood cell walls

Documents

decayed wood

reaction wood

avariety of wood

normal wood

secondary cell wall

celluloserich cell walls

ligninrich cell walls

glayer oftension wood