Silica Stain -AMJB

7/28/2019 Silica Stain -AMJB

1/7

'HWHFWLRQRI6LOLFDLQ3ODQWV$XWKRUV3'D\DQDQGDQ3%.DXIPDQ&,)UDQNOLQ6RXUFH$PHULFDQ-RXUQDORI%RWDQ\9RO1R$XJSS3XEOLVKHGE\Botanical Society of America6WDEOH85/http://www.jstor.org/stable/2442818 .

$FFHVVHG

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless

you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you

may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at .http://www.jstor.org/action/showPublisher?publisherCode=botsam. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed

page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms

of scholarship. For more information about JSTOR, please contact [email protected].

Botanical Society of America is collaborating with JSTOR to digitize, preserve and extend access toAmerican

Journal of Botany.

http://www.jstor.org
http://www.jstor.org/action/showPublisher?publisherCode=botsamhttp://www.jstor.org/stable/2442818?origin=JSTOR-pdfhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/action/showPublisher?publisherCode=botsamhttp://www.jstor.org/action/showPublisher?publisherCode=botsamhttp://www.jstor.org/page/info/about/policies/terms.jsphttp://www.jstor.org/stable/2442818?origin=JSTOR-pdfhttp://www.jstor.org/action/showPublisher?publisherCode=botsam


2/7

Amer. J. Bot. 70(7): 1079-1084. 1983.

DETECTION OF SILICA IN PLANTS1P. DAYANANDAN, P. B. KAUFMAN, AND C. I. FRANKLIN

Division of Biological Sciences, University of Michigan, Ann Arbor, Michigan 48109

ABSTRACTSilica in plants can be stained by silver chromate, methyl red, and a colorless crystal violetlactone which are adsorbed by the silanol groups resulting in red-brown, red, and blue colors,respectively. Specialized silica cells in grasses can also be detected through polarization colorsdue to form birefringence. Silica in the bulliform and silica cells of rice leaves is amorphousand is made up of 1-2-nm particles aggregating into 2.5 X 0.4-,um rods with oblique ends.

SILICON IS DEPOSITED in plants as hydratedamorphous silica (SiO2 nH2O) through the po-lymerization of monosilicic acid (Si(OH)4) ab-sorbed by roots from soil solutions (Jones andHandreck, 1967). Molisch (1923) summarizedfive methods for the microscopic detection ofsilica in plants: 1) formation of sodium sili-cofluoride crystals by the action of NaCl andHF; 2) dry ashing; 3) wet ashing; 4) stainingwith basic fuchsin and malachite green; and 5)mounting in phenol. Netolitzky (1929) men-tioned gentian violet, methylene blue, and saf-ranin as additional stains for silica, but for lackof specificity, staining reactions were aban-doned early (Viehoever and Prusky, 1938), andno color reaction is in current use. Wet ashingwith acids (Parry and Smithson, 1958), mi-croincineration to produce a spodogram of ashresidues (Lanning, Hopkins and Loera, 1980)or clearing with phenol or other reagents haveremained the only methods for the detectionof silica by light microscopy. All these methodshave tended to treat silica as an inert material,and investigators have resorted to differencesin refractive indices (RI) of silica (RI 1.43) andthe mounting media to render contrast. Whenmounted in phenol (RI 1.54) or Canada balsam(RI 1.54), cell walls and associated matter (RI1.56) are obscured and silica is seen with higherrelief.We report here three distinct histochemicalreactions based on the reactivity of the silanol(SiOH) groups of silica. We also report the useof polarizing microscopy to impart diffractioncolors for the detection of a special type of silicabody in grasses. Ultrastructural information isalso provided for the silica we have investi-gated.I Received for publication 20 September 1982; revisionaccepted 14 February 1983.We thank Dr. R. K. lier for valuable suggestions, Dr.T. T. Chang for rice seeds, Mr. David Bay for photographicassistance and Nyacol Products Inc. for the gift of crystalviolet lactone. This research was supported by the NSFGrant PCM 80-23923.

MATERIALS AND MICROSCOPY-The follow-ing species of rice (Oryza) obtained from theInternational Rice Research Institute, LosBafios, Philippines, and grown in clay in plasticpots in a growth chamber were the primarymaterials used in this investigation: 0. altaSwallen, 0. australiensis Domin, 0. brach-yantha A. Chev. et Roehr, 0. eichingeri A.Peter, 0. minuta J.S. Presl ex C.B. Presl, 0.nivara Sharma et Shastry, 0. officinalis Wall.ex Watt, 0. perrieri A. Camus, 0. punctataKotschy ex Steud., 0. rufipogon Griff., and 0.sativa L. Other silica accumulating monocotsand dicots were used as supplementary ma-terial to confirm the applicability of the stain-ing techniques to these other plants. Epidermalpeels, free-hand sections, isolated cuticles andsilica bodies were further treated as describedunder individual staining procedures. Sinceepidermal peels are not easy to isolate fromrice leaves, about 1-cm-long pieces of freshleaves were scraped on abaxial or adaxial sidesto remove most of the cells above the under-lying epidermis. Cuticle and silica bodies wereisolated by immersing pieces of leaves in conc.H2SO4 for 1-4 days, and diluting with waterto float the cuticles, and sediment the silicabodies. Cuticles were rinsed in water and trans-ferred to slides for air drying. After severalwashings, the silica bodies were isolated bycentrifugation or a long period of undisturbedsedimentation.An Olympus VANOX research microscopewith polarizing and epifluorescence optics wasused. Color photographs were recorded eitheron Kodacolor II negative or Kodachrome 40color transparency films. For transmissionelectron microscopy (TEM), rice leaves werefixed in 4%glutaraldehyde in 0.05 Mcacodylatebuffer at pH 6.8 overnight at 4 C. Either withor without post-fixation in 2% OS04 for 2 hr,they were dehydrated in an ethanol series andembedded in Spurr's medium. Isolated silicabodies were similarly embedded but were not1079


3/7

1080 AMERICAN JOURNAL OF BOTANY [Vol. 70fixed. Thin sections were collected on uncoatedor holey-carbon coated 200-mesh copper grids.Sections were examined without staining orafter staining with uranyl acetate and lead ci-trate. They were examined with a JEM 100 CXelectron microscope operated at 100 Kev.METHODSND RESULTS-Birefringentsilicacells-Isolated cuticles of rice leaves often havesilica cells attached to them. Such silica cellsoccur in 1-3 rows over the sclerenchyma as-sociated with vascular bundles on both theabaxial and adaxial epidermises. When ex-amined with polarized light with the polarizerand the analyzer in the conventional crossedpositions (Bennett, 1950), the silica cells ap-pear birefringent. When a 1st order plate of5 30 nm is introduced with its slow axis runningfrom SW to NE positions in the microscopefield, the field appears uniformly red, and thesilica cells show addition (blue) and subtraction(yellow) colors (Fig. 1-3). The resulting colorscan be interpreted on the basis of the slow axisof the material responsible for birefringencerunning parallel to the outlines of the silicacells. Ultrastructuralanalysis of silica cells show(Fig. 12) numerous 2.5 X 0.4-,Amrods of silicatightly packed inside each cell. These rodsthemselves are not responsible for the bire-fringence since their orientations do not par-allel the expected arrangement. Also, the silicain these, as in all other parts of the rice plant,does not reveal any crystallinity when exam-ined for electron diffraction. Figure 13 showsvery fine lines about 10-20 nm thick, runningparallel to the cell outlines. We interpret theseto be impressions left by cellulose microfibrilssince cellulose microfibrils are of these dimen-sions. Our unpublished observations suggestthat during the ontogeny of a silica cell, the cellwall swells to fill a considerable portion of the

cell lumen and that silica is polymerized aroundwall cellulose fibrils. Thus, the birefringenceand the color of these silica cells are due toform birefringence caused by the impressionsor concavities left by the microfibrils withinthe amorphous silica matrix. The birefringenceis not due to intrinsic birefringence since silicais amorphous and any cellulose would havebeen digested by the conc. H2SO4 used in theisolation procedure. Silica bodies isolated fromlong epidermal cells and bulliform cells do notshow the birefringence typical of the silica cells.This may indicate a basic difference in the on-togeny of different types of silica accumulatingstructures.

Silver-ammine chromate (SAC) reaction-Amorphous silica reacts with the light-yellow-colored SAC producing red-brown-stained sil-ica. The basis of this reaction is the adsorptiveremoval of ammonia which demasks the red-brown silver chromate from the masked yel-low-colored solution. The silver chromate pre-cipitates as reddish brown particles on the silicasurface (Feigl, 1943). SAC is prepared by mix-ing solutions of AgNO3 and K2CrO4, washingthe precipitate in hot water to remove solubleKNO3, and dissolving the Ag2CrO4precipitatein less than the required amount of NH40Hto completely dissolve the precipitate. SAC so-lution exists in the following equilibrium:(AgNH3)2CrO4 = 2 NH3 + Ag2CrO4. Excessammonia interferes with efficient demaskingof silver chromate by producing CrO4- ionswhich make the solution yellow in color. Weprepare SAC by separately dissolving 34 gAgNO3 and 20 g K2CrO4in 100 ml of watereach and mixing these solutions to obtain theprecipitate. The precipitate is then washed inhot water (50-60 C) and dissolved in 200 mlof 3% NH3 (20 ml of strong ammonium hy-

Fig. 1-11. Silica in Oryza species. 1-3. Silica cells attached to isolated cuticles of O. sativa cv. 'Basmati' photographedwith polarizing microscope with 1st order red plate. The cuticles were isolated by immersing pieces of leaves in conc.H2SO4 for 3 days. All X400. 1. In conventional position with the slow axis of red plate oriented SW-NE. Length ofleaf blade parallels the silica cell rows from left to right. 2. Portion of the same cells oriented at 450 to the polarizerbut silica cell rows parallel to the slow axis of red plate. 3. Same field oriented in the opposite direction as in Fig. 2with rows at 900 to the slow axis of red plate. 4. A row of silica cells of 0. sativa cv. 'Homarenichiki' stained with SACafter 30 sec etching. X200. 5. Silica body isolated from a bulliform cell of 0. sativa cv. 'FR13A' stained with SACwithout etching. Xl,OOO.6. MR-stained silica in a bulliform cell of 0. alta viewed from the underside of adaxialepidermis. Xl,000. 7. Silica from a bulliform cell of 0. sativa cv. 'Nipponbare' lying on its side in an epidermalpreparation stained with CVL. X1,000. 8. Adaxial epidermis of 0. officinalis stained with MR followed by CVL. Largebulliform cell silica bodies lie between rows of silica cells. X100. 9. Abaxial cuticle with rows of silica cells still attached,from 0. glaberrima stained with MR. Large papillae between the two vertical rows of silica cells lie over a majorvascular bundle. These and other smaller papillae and protruberances around the stomata show silicification. X200.10. C. S. of leaf of 0. sativa cv. 'Basmati' showing upper epidermis with bulliform cell silica body stained with MRand adjacent tissues counterstained with fast green. Xl,000. 11. Similar to Fig. 10 but silica stained with CVL andadjacent tissues counterstained with safranin. X1,000.


4/7

August, 1983] DAYANANDAN ET AL. -SILICA DETECTION 1081

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .~~~~~. ....

k _ ' _ MR _ w

.5. 7 i _

6 _Al _~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~..".~I..".,..,,'

ioi!"c?0 k j d $i,; ,, b ,tt


5/7

1082 AMERICAN JOURNAL OF BOTANY [Vol. 70

12

Fig. 12-15. Ultrastructure f silica in Oryzasativa. 12. Paradermal ectionof a silica cell from a leaf of cv. 'L-III-125'showingrods of silica.Dark bandsarefoldings hatdevelopduringsectioning.Arrow ndicates he orientationof the longitudinalaxis of leaf blade. X7,200. 13. Portion of the silica cell in Fig. 12 (circle)at highermagnificationshowing mpressionsof cellulose microfibrils. 29,760. 14.C. S. of outerepidermalwallshowinganinnerand anouterlayerof silica. The outer layerof silica is intimatelyassociated with the cuticle and the papilla.Whenin associationwith the cell wall of outerepidermis,silicaappearsaggregatednto sphericalmasses,unlike the typicalrodsseen whensilicapolymerizesn cell lumen.x7,470. 15. C. S. of portionof anisolated silicabodyfrom 'Basmati' ice leafbulliformcell. Individualrodsof silica are about 2.5 X 0.4 gm in dimensions and each bulliformcell may contain as many as100,000 such rods. Each rod is in turn made up of ultimateparticlesof 1-2 nm diam. X5,435.

droxide of 30% NH3 in 190 ml of H20). Fil-teredSACcan be stored at room temperaturesin tightlystopperedbottlesfor a long time andcan be refilteredwhen turbid.Pieces of leaves with one surface scrapedwith a blade are dipped in 50%H2SO4 for 2min. Afterthoroughwashing,they aremount-ed in a few drops of SAC and immediatelycoveredwith a coverslip.Leavescan befurtherscraped o isolatethe silica from bulliformcells(Fig. 5) before treatingwith SAC. The acidtreatmentenhances the intensity of the stainandresults nthestainingof moresilica bodies.This may be due to partial digestion of thesurrounding issues to provide easy access ofthe reagent.However, any trace of excess acidleads to nonspecific precipitationsince SACcan react with acids, acid salts, and acid an-hydrides(Feigl, 1943). The above treatmentdoes not always stain all the silica bodies be-cause of the presenceof tightly packed silicaat theouter boundariesof the silica bodies that

preventsentrance of SAC. Mild etching afteracid treatmentovercomes this problem (Fig.4). Etching can be done with HF and/or am-monium fluoride NH4F).Sincethe silicabod-ies areverysmall, etchingshould bedonecare-fully lest the entiresilica body is dissolved. A30-60 sec etching in room temperaturewith0.1-0.2% HF with or without the addition ofNH4F to give a 1%concentration has givensuccessful results with rice leaves. It is alsopossible to add theetchingcompoundsdirectlyin H2SO4. In all events, thoroughwashing isessential.SAC-stainedmaterialcan be washedin water and dehydratedthrough an alcoholseries and mounted in permanentmountingmedia. Dehydrationchanges color from red-brownto black offeringexcellentcontrastforphotomicrography.Methyl red (MR) and crystal violet lactone(CVL)-Non-ionic MR and cationic CVLarereadilyadsorbed from non-polarsolvents by


6/7

August, 1983] DAYANANDAN ET AL. -SILICA DETECTION 1083amorphous silica gel. When dissolved in ben-zene, MR appears light yellow-orange in color,and CVL is colorless. When adsorbed by silica,MR imparts a bright red color, while CVLstains deep blue to violet. The commercial uses,chemical structure and basis of reaction of CVLhave been described (Iler, 1979). Basically, inthe absence of water, alcohol or other agentswhich tend to hydrogen bond with the SiOHgroups on silica, the leucobasic CVL is ad-sorbed on the SiOH groups and undergoes astructural shift resulting in a change from col-orless to blue. The adsorption of MR by thesilanol groups on silica is the basis for the useof this dye in surface area measurements (Sha-piro and Kolthoff, 1950).Crystal violet lactone obtained from NyacolProducts Inc. (Megunco Road, Ashland, Mas-sachusetts 01721, USA) is used as a 0.1% so-lution in benzene. The acid form of MR (Ko-dak Co.) is used as a benzene-saturated andfiltered solution. The acid form of MR is pref-erable to the sodium salt or the hydrochlorideof MR. The latter two compounds do not stainsilica as intensely as the acid form. If the hy-drochloride of MR is used, it should be firstdissolved in alcohol before the addition of ben-zene. Sections, peels, or small pieces of entireleaves are immersed in 50% H2S04 for 2-5min and rinsed in tap water. After dehydrationthrough an ethanol or acetone series, they aretransferredgraduallyto pure benzene, and frombenzene, to MR or CVL. Color develops onthe silica bodies almost immediately, but long-er treatments ensure maximum intensity (Fig.6, 7). Isolated cuticles can be similarly pro-cessed (Fig. 9). Where handling the delicatecuticles is a problem, they can be allowed todry on the slide, and CVL or MR applied di-rectly. Excess MR or CVL should be removedby rinsing sections in pure benzene. Stainedmaterial can be made permanent by mountingin Permount (Fig. 6-11).By mixing varying proportions of MR andCVL together, or by staining in CVL followedby MR or vice versa, intermediate shades ofcolor can be imparted to silica bodies (Fig. 8).It is also possible to counterstain the non-silic-ified regions with a number of conventionalstains. Figures 10 and 11 show counter-stain-ing with fast green and safranin, respectively.These stains should be applied before trans-ferring samples to ethanol-benzene mixture.Fast green is best applied in 100% ethanol,while safranin can be applied while in wateror 70% ethanol. Methyl red and CVL do notstain cellulose when applied in benzene. How-ever, lignin, pectin-rich material, and calciumcarbonate do stain with different intensities and

shades of red or blue. When care is taken tocompletely dehydrate the material and removeall excess MR and CVL, these nonspecificstaining reactions are minimized. Interferingcomponents can also be selectively removedthrough treatments that do not dissolve silica.With SAC reagent, the rods or aggregates ofrods that make up silica in bulliform cells andsilica cells, as well as in trichomes and lemmaand palea of rice can be visualized. MR andCVL also stain silica deposited in long epi-dermal cells, trichomes, sclerenchyma, sto-matal ledges, and epidermal papillae (Fig. 9).We have confirmed the presence of silica inthese regions by x-ray microanalysis. In theouter wall of rice epidermis, silica is depositedin two distinct layers (Fig. 14). The outer layerof silica is in intimate contact with the cuticularlayer and the numerous papillae characteristicof the rice epidermis (Fig. 9, 14). MR and CVLare excellent stains for isolated cuticle and as-sociated siliceous matter in grasses. Thesestaining reactions are applicable to all thespecies of rice mentioned in Materials and Mi-croscopy. They readily reveal differences in size,distribution, degree of staining and number ofsilica bodies in these different species. For ex-ample, while most species of rices and cultivarsof 0. sativa possess silica cells only over thevascular bundles (Fig. 4), intercostal epidermalsilica cells occur in 0. minuta, 0. perrieri and0. punctata.A preliminary survey indicates that SAC,MR, and CVL also stain silica in other gra-minaceous and non-graminaceous plants. Alarge number of silica bodies in bulliform cellsare revealed in Coix lacryma-jobi L. and inbamboos such as Phyllostachys sp., and Shi-bataea kumasasa Makino. Silica in the outerepidermal wall and in specialized thickeningsof stomata in Equisetum are also stained. Thesestains also localize silica in the rings of cells atthe bases of trichomes in the leaf epidermis ofHumulus lupulus L., and in the epidermal tri-chomes of CommelinabenghalensisL., Gali-umpalustreL. and Urticadioica L.DISCUSSION-The reactivity of silica is dueto the presence of the silanol (SiOH) groups,typically one silanol group per atom of silicon.Since biogenic silica is polymerized from silicicacid solutions under moderate temperatures,one could expect about 4-6 OH nm-2 of sur-face. The polymerization of biogenic silica intorodlike subunits (Fig. 15) offers large surfacearea for extensive chemical reactions. SinceMR and CVL possess molecular areas of about1.16 and 1.64 nm2, respectively, the surfaces ofsilica should have these minimum dimensions


7/7

1084 AMERICAN JOURNAL OF BOTANY [Vol. 70for the entry of these molecules. Most silicacells and some bulliform silica are not readilystained by MR and CVL unless etched. Thismay indicate that in some populations of silicabodies the outer surfaces are tightly packedwith particles of silica that do not possess largeenough pores for the molecules to enter. It mayalso mean that the silanol groups have beenmodified by other reactions such as esterifi-cation.Positive results with SAC, MR, and CVLpoint to the similarity between biogenic silicaand the numerous commercial silica gels thatreact similarly. Commercial silica is known tobe highly reactive, capable of forming numer-ous ionic, nonionic and covalent bonds (Iler,1979). It should be possible to develop manystaining reactions that can be applied to bio-genic silica. These could include direct visu-alization with stains such as Janus red andmalachite green, deposits of metals such ascobalt, copper, nickel, silver, and zinc (Smithand Jacobson, 1956), and indirect staining ofsubstances that are first adsorbed on the silicasurface. Silica can also be visualized with flu-orescence microscopy through the adsorptionof a metal such as uranium or a stain such asrhodamine 6GB. Molisch (1923) indicated thatsilica mounted in phenol can show a reddishhue. The basis of this reaction can be attributedto the ability of silica to adsorb phenols (Davis,Deuchar and Jbbitson, 1973). Subsequent ox-idation of phenol can lead to color develop-ment. It should thus be possible to adsorbphenols onto silica and to react with dizoniumsalts to produce the desired colors.The physical and chemical similarities be-tween commercial silica and biogenic silica al-low one to take advantage of the extensiveknowledge regardingthe former to characterizethe physical and chemical nature of biogenic

silica, especially in connection with the levelof hydration, pore size, pore volume, surfacearea, silanol number, and the presence of ad-sorbed metals and organic molecules. Such in-formation will give us a better understandingof the functional and structural roles of silicain plants.

LITERATURE CITEDBENNETT,H. S. 1950. The microscopical investigationof biological materials with polarized light. In R. M.Jones [ed.], McClung's handbook of microscopicaltechnique, pp. 591-677. Hoeber Inc., New York.DAVIS, K. M. C., J. A. DEUCHAR, AND D. A. IBBITSON.1973. Adsorption of phenols from non-polar solventson to silica gel. J. Chem. Soc. Faraday Trans. I. 69:1117-1126.FEIGL, . 1943. Laboratory manual of spot tests, pp. 155-156. Academic Press, New York.ILER,R. K. 1979. The chemistry of silica, pp. 622-729.John Wiley & Sons, New York.JONES,L. H. P., ANDK. A. HANDRECK. 1967. Silica insoils, plants and animals. Adv. Agron. 19: 107-149.LANNING, F. C., T. L. HOPKINS,ANDJ. C. LOERA. 1980.Silica and ash content and depositional patterns intissues of mature Zea mays L. plants. Ann. Bot. 45:549-554.MOLISCH,H. 1923. Mikrochemie der Pflanze, pp. 74-76. Gustav Fischer, Jena.NETOLITZKY,. 1929. Die Kieselkorper die Kalksalzeals Zellinhaltskorper. In K. Linsbauer [ed.], Hand-buch der Pflanzenanatomie, pp. 1-1 9. GebriuderBorn-traeger, Berlin.PARRY, D. W., AND F. SMITHSON. 1958. Techniques forstudying opaline silica in grass leaves. Ann. Bot. 22:543-551.SHAPIRO, I., AND I. M. KOLTHOFF. 1950. Studies on agingof precipitates and co-precipitation. XLIII. Thermalaging of precipitated silica (silica gel). J. Amer. Chem.Soc. 72: 776-782.SMITH, G. W., AND H. W. JACOBSON. 1956. Character-istics of adsorption complex metal-ammines and oth-er complex ions of zinc, copper, cobalt, nickel, andsilver on silica gel. J. Phys. Chem. 60: 1008-1012.VIEHOEVER, A., AND S. C. PRUSKY. 1938. Biochemistryof silica. Amer. J. Pharm. 110: 99-120.

Documents

Silica Stain -AMJB