192247

8/13/2019 192247

1/92

Coffee(Coffeaarabicacv.Rubi)seedgermination: mechanism and regulation

Edvaldo AparecidoAmaraldaSilva

CENTRALE LANDBOUWCATALOGUS

8/13/2019 192247

2/92

Promoter: Prof.Dr.L.H.W.vanderPlasHoogleraarindeP lantenfysiologieW ageningen Universiteit

Co-promotoren: Dr.H.W .M. HilhorstUniversitair docentLaboratorium voor PlantenfysiologieW ageningen UniversiteitDr. A.A.M.vanLammerenUniversitair hoofddocentLaboratorium voor PlantencelbiologieW ageningen Universiteit

Oppositie: Prof.J.D.Bew ley, UniversityofGuelph, CanadaProf.M.L.MoreiradeCarvalho, Universidade FederaldeLavras,BrazilieProf.Dr.A.M.C.Emon s, W ageningen U niversiteitDr.S.P.C.Groot, Plant Research International

8/13/2019 192247

3/92

A / A / o ^ l , 3 3 fPropo sitions (Stellingen)

1- Con trary to seed germination in tom ato, the comp letion of coffee seed germination is thenet result of embryo grow th and endosperm weakening(this thesis).2- Ab scisic acid inhibits coffee seed germination by suppressing the increase in embry ogrowth potential and he second stepof endosperm cap w eakening(this thesis).3- Coffee seed germination is both stimulated and inhibited by gibberellins at phy siologica l'concentrations,(this thesis).

4- Acc elerating coffee seed germination and seedling establishment by one month gre atlyreduces he abour andcost ofestablishingnew coffee shrubs.5- Th e study of the behav ior of tree seeds during developmen t, germination and storag e ismand atory in he preservation of biodiversity.6- Fac ts are he air of scientists. W ithout hem you can never fly.LinusPauling7- Collaboration amongdeveloped and developing countries san outstanding w ay ofpromo ting scientific and echnologic development.8-The B razilians should be more proud of heir country.

These propositions belong to the P hD thesis entitled: Coffee(Coffea arabicacv Rubi) seedgermination: mechanism and regulation.

AmaraldaSilvaWageningen,26June 2002

8/13/2019 192247

4/92

j ; ; ; ^

Coffee(Coffeaarabicacv.Ru bi) seedgerm ination: m echanism and reg ulation

Edvaldo AparecidoAmaraldaSilva

Proefschriftter verkrijging van degraadvan doctorop gezag van de rector magnificusvan Wageningen Universiteitprof.dr. ir. L.Speelman,in het openb aar te verdedigen

op woensdag 26Juni 2002desnamiddagse half twee in de aula.

J >

8/13/2019 192247

5/92

Cov er illustration: Coffee tree bearingunripe (green) and ripe (red) fruits that each con taintwo seeds.

ISBN 90-5808-650-x

8/13/2019 192247

6/92

PrefaceThis workisndebted to awide range of people.First Iwould liketo saythanks toWageningen University and theFederal University ofLavras for giving me theunforgettable opportunity topursue this PhDprogram in TheNetherlands.I want to say thanks to my Brazilian sponsor (CAP ES) and the Laboratory of PlantPhysiology (WU )for the inancial support of my studies.Prof.Linus van derPlasisvery much acknowledged for acce pting me in theLaboratory ofPlant Physiology as a PhDstudent underhissupervision and or all thesupport during thelast4 years.My special thanks goes to meneer Dr.Henk Hilhorst, for thebright supervision, forproviding mewithagood education duringmystayin TheNetherlands, for the opportunitiesgiven that allowme toexpose myself to other scientists and toknow other cu ltures. Whatyoutaughtme is priceless.I also wantto saythanks to myco-promoter Andre vanLammeren and to hisgroup at theLaboratory ofPlant Celt Biologyfor thesupervision andenthusiastic discussions during ourweekly meetings.I sincerely thankDr.Peter Tooropfor the riendship, discussions andwise advice during myPhD program and o hiswife RoseTooropfor the riendship andortheEnglish lessons.Ireally appreciated thehelpofAdriaan vanAelstandJaap Nijsse during the microscopywork, also their valuable discussions, their help withthepictures and for teaching me how tomakethe layoutof this thesis.ToProf.Derek Bewleyfor hissupervision andhospitality duringmystayinCanada. Ihavelearneda lot inyour lab, thanks also orthe opportunity.My gratitude goesto the to thepeople of theSeed Laboratory at theFederal University ofLavras for thecontinuous motivation, for helping me toachieve myobjectives and forshipping thecoffee seedsto TheNetherlands every year.Toallof you Muito obrigado .Iwould like to express myappreciation to thestaff m embers and PhDstudents at theLaboratory of Plant Physiology and the Laboratory of Plant Cell Biology, and to myroommate JoseMarciofor theagreeable coexistence, supportandhelp duringmy studies.TotheBrazilian comm unity inWageningen thankyouvery muchfor thesupport, friendship,helpand patient.FinallyIcouldnothave don e this work w ithout thesupportandunderstanding of my family,tomywife Claudia and my sonSamuelfor their comprehension andorgive me or myalienation duringthelast4 years.Amaral

8/13/2019 192247

7/92

/dedicate his thesisTomy wife Claudiaand my son Samuel.

8/13/2019 192247

8/92

ContentsChapter1 General introduction 1Chap ter 2 Anatom y and morphology of thecoffee(Coffeaarabicacv.Rubi) seed 7

and fruit during g ermination

Chapter 3 ABA regulates embryo growth potential and endosperm cap weakening 19during coffee(Coffeaarabicacv. Rubi) seed germination

Cha pter 4 Supra-optima l GA concentrations inhibit coffee(Coffeaarabicacv. 41Rubi) seed germination and lead to death of the embryo

Chapter 5 ABAreduces the abundance of microtubules and inhibits transversal 59organization of the microtubules, em bryo cell elongation and celldivision du ring coffee(Coffeaarabicacv.Rubi) seed germination

Chapter 6 Molecular cloning of cDNA s encoding an endo-|3-mannanase and(3- 75mann osidase from theendosperm caps of germinating coffee(Coffeaarabicacv. Rubi) seeds

Chap ter 7 General discussion 87Summary 93Samenvatting 97Sumario 101Curriculum vitae 105

8/13/2019 192247

9/92

CHAPTER 1

General Introduction

8/13/2019 192247

10/92

8/13/2019 192247

11/92

General Introduction

bytheinternal cell turgor pressure \|/p)(Bewley,1997).Brassica napu sis anexampleof aseed whereanncreasein heurgorandcell w all extensibilityofhe embryoisaprerequisitefor radicle protrusion (SchopferandPlachy,1985).

The third possibilityisthatthetissues env elopingtheembryo w eaken, allowingradicle grow th.Inseeds that showasevere constraintonradicle cell growth impo sedbysurrounding structures,thepressure potential \|/p) in theembryoaswellas theturgorareinsufficient todrive cell wall expa nsion.Inthis case weakeningof thecell wallsof theconstraining tissuesbyactionof hydrolases is required for declineof themechanicalresistance (Bewley, 1997).Thecurrent modelfor tomato seed germination from Toorop(1998)ispresentedinFigure1.

Objectives

Coffee seed germination is slow and shows wide variation in the timingofemergence.Theoverall objectiveofhis thesiswas ounravelthemechanismofcoffee seedgermination aswellas tsegulationbyabscisic acidandgibberellic acid.

Mo re specifically,theobjectivesofhis hesisare:

1. Structural analysistonvestigate endosperm cell wall morphologyanddegradation duringcoffee seed germinationand tssignificancetoadicle protrusion;

2. Studyofhe involvementofenzyme s requiredforendosperm cell wall degradation duringgermination,aswellascontrolbyabscisic acidofhe germination process;

3. Understandingtheoleofendogenousandexogenous gibberellinsinembryo growthandendosperm degradation during germination;

4. Studyof theeffectofexogenous abscisic acidon hecell cycle m achineryin thecoffeeembryo during germ ination;5. Cloningofendo-P-mannanaseandP-mannosidase gene s during coffee seed germination

as wellas heimingandocationofendo-p-mannanaseandP-mannosidasein thedifferent seed parts during germ ination.

8/13/2019 192247

12/92

Chapter 1

lateralendosperm

endospermcap

I A B A e n d o -* OAendo

I

embryo

inductioninhibitionnoeffect

testa

ABAe

OAe,\

JW >0 55MPaW < 0.55 MPa

Figure. 1.Model for the mechanism and regulation of tomato seed germination as proposed by Tooro p(1998).The model is based on the concept that only endosperm weakening is required for radicleprotrusion as the embryo has a water potential of approximately - 2 MPa. En dogenou s GA s, synthesizedin the emb ryo and secreted into the endosperm, induce the enzymatic degradation of cell walls in theendospe rm cap (1) and lateral endosperm (2) and endogenous A BA inhibits these processes (6). Of thetwo-step degradation of the endosperm ca p the first step, mediated by endo -p-m anna nase, is induced byendog enous and exogenou s G As (1). The first step is not affected by ABA (3) but is inhibited by anexternal osm otic potential < - 0.55 M Pa (9). Degradation of the lateral endosperm through endo-p -mannanase is inhibited by ABA (7). The second step of endosperm softening is promoted byendogenousandexogenous GAs (5) and inhibited by ABA (4) and an external osm otic potential > -0.55MPa(10), which did not affect the first step (8). Th e enzym e(s) involve d in the secon d s tep ofendosp erm weakening are not known ( ? )

8/13/2019 192247

13/92

General Introduction

Scope of the thesis

Chapter-General IntroductionTh e origin and the importance of coffee is presented, as well as the problems occurring ingerm inating coffee see d, theobjectives of the hesis as well as he hesis structure;

Chapter2-Morphology and anatomy of the coffee Coffea arabicacv. Rubi) fruit andseed during germination.Light microscopy and low temperature scanning electron m icroscopy (Cryo-SEM ) w ere usedto describe the different fruit and seed tissues and cells. Morphology and anatomy ofendosperm, endosperm cap and embryo during coffee seed germination are presented. C ertainaspects of cell and cell wall morphology were investigated in more detail through lightmicroscopy and Cryo-SEM.

Chapter 3 -ABA regulates embryo growth potential and endosperm cap weakeningduring coffee Coffeaarabicacv. Rubi) seed germin ation.The involvement of hydrolytic enzymes in endosperm cap degradation w as investigatedduring coffee seed germination as well as the time and duration of these events in relationwith the completion of germ ination. Additional investigations of the location of these eve ntswithin the seed were done by tissue printing. Puncture force w as measured to determin eendosperm weakening. D ifferentisoformsofendo-(i-mannanasefrom the endosperm cap andrest of the endosperm were identified by using isoelectric focussing. Endogeno us AB A leve lsof coffee e mbryos w eredetermined and the possible role of this hormone is addressed .

Chapter 4 -Supra -optima l GA concen trations inhibit germination in coffee seed Coffeaarabicacv. Rubi) and eadsto deathof theembryo.Th e role of endogenous G A and the inhibitory effect of exogeno us GA s were studied duringcoffee seed germ ination.

Chapter5 -ABA reduces the abundance of the microtubules and inhibits transversalorgan isation of the m icrotubu les, emb ryo cell elongation and cell division during coffeeCoffeaarabicacv.Rubi) seed germination.

The effect of exogenous ABA on DNA synthesis,P-tubulinaccumulation and assembly of themicrotub ules w ere studied in coffee seed embryos during germination.

8/13/2019 192247

14/92

8/13/2019 192247

15/92

CHAPTER 2

Anatomy a nd morphology of he coffee Coffea arabicacv.Rubi) seedand fruit du ring g ermination

E.A.Amaralda Silva,Peter E.Toorop,Adriaan C.va nAelstandHenk W.M. Hilhorst

8/13/2019 192247

16/92

Chapter 2

Abstract

The coffee Coffeaarabicacv.Rubi) fruit is adrupe containing two seeds. The coffeeseed is comprised of an endosperm, embryo andspermodermor silver skin . The thickenedcell walls of the endosperm are composed m ainly of mann ans with 2% of galactose. Theendosperm contains both polygonal and rectangular cell types. The rectangular cell type waslocated adjacent to the embryo in the so-called internal endosperm whereas the p olygonalcells were located in the external endosperm. The endosperm cap cells have sm aller andthinner cell walls than the rest of the endosperm, which indicates that the region w here theradicle will protrude is predestined in coffee seeds. Radicle protrusion in the dark at 30 Cwas initiated around day 5 of imbibition and at day 10, 50% of the seed population showedradicle protrusion. The endosperm cap of the coffee seed changed during g ermination. Cellcompression was followed by loss of cell integrity, appearance of a protuberance andoccurrenc e of cell wall porosity. The observations indicated that embryo grow th and chan gesin he endosperm cap region control radicle protrusion in coffee seed.

Keywords:Coffee seed, light microscopy, cryo-scanning electron m icroscopy, morpho logy,anatomy, germination.

8/13/2019 192247

17/92

8/13/2019 192247

18/92

Chapter2

seeds w ere kept at 30 1C in the dark (Huxley, 1965; Valio, 1976). At least 3 seeds w eretaken randomly every day during germination for light microscopy and low temperaturescanning electron microscopy studies.

Light microscopy. The entire imbibed seeds were sectioned using a microtome(Reich ert, Austria). The sections of 20-30 urn thickness from the endosperm cap and the restof the endosperm were first transferred to demineralized water and then fixed in liquidK aise r's glycerol gelatin (Merck, German y) for observations. Observations w ere made in aNikon Optiphot microscope in bright field mode. Photographs were taken with a digitalPanasonic Colour Video Camera or a Sony CCD CameraDKR 700.Images of the coffee fruit,seeds and embryo were aken by using aLeica binocular.

Cryo-scanningelectron microscopy.Coffee seeds were prepared for C ryo-Sca nningElectron M icrocopy (C ryo-SEM ). The seeds were longitudinally sectioned w ith a razor bladeand mounted on a cup shaped holder with tissue freezing medium. After mounting, thesamp les were plunge-frozen and stored in liquid nitrogen for subsequentcryo-planingandobservations.Cryo-Planing,which attempts to produce flat surfaces for observations in Cryo-SEM , was performed using acryo-ultramicrotomewith a diamond knife, acco rding to N ijsseetal.,(1999). For observa tions the specimens were heated up to -90 C, sputter-coated withplatinum and placed in the cryostat of the scanning electron m icroscope (JEOL 630 0 Fieldemission SEM ). Observations w ere made at -180 C using a 2.5-5kV accelerating voltage.Digital images w ere taken and printed. Alternatively, the seeds were freeze-fractured with acold scalpel knife, heated up to -90 C,partially freeze-dried and sputter-coated w ith 5nmofPt.

Resu lts and discussionCoffee ruit andseedmorphology

^^E^ufl

A

1 Empty1 j3 area K a |

1 BH fiCd^ E x t e m a T ^ ^ ^Endosperm

PPExocarp^Bj-MesocarpiL^.Endocarp^H>-Embryo

InternalEndosperm

Figure1A:Green coffee Coffea arabicacv.Rubi)fruit; B:Transversalsection ofcoffee fruit showing he nternalstructures.

The coffeefruit (Coffea arabicaL.) is a drupecontaining two seeds(Fig. 1A and B ). Thethick exocarp is easilyremoved, revealingthe. oft mes ocarp . Theouter cover of the seedis formed by a hard

10

8/13/2019 192247

19/92

8/13/2019 192247

20/92

Chapter 2

(Giorgini and Ca mp os, 1992) . I t i s 3 to 4 mm long and is com posed of an axis and twocoty ledo ns (F ig. 3); i t is localised clos e to the conve x surface of the seed (Ren aetal., 1986).Du r ing embry o developm ent hypocotyl format ion is preceded by the format ion of thecotyledons, but embryo development takes p lace af ter endosperm development (Arci la-Pulgarfn and O r o z c o - C a s ta n a , 1987) . Polyem bryon y, more than one embry o per seed, andem pty seed s hav e been obs erved in coffee seed at a frequency of 1,2% (M en des , 1944 ).

Germination characteristics

Coffee seed s germin ate s lowly (Renaetal, 1986) . Seedl ing eme rgenc e f rom the soi lstarts 50 to 60 days after so win g in the warm er period s of the year (M aestr i and V ieira, 1961).W hen tempe ratures are lower the em ergen ce per iod may increase to 90 days (W ent , 1957) .Fol lowing germinat ion, the cof fee cotyledons grow by absorbing the endosperm and turngreen (W ellm an, 1961 ). Th e first seed parts to em erge from the soil are the cot yled ons ,character iz ing epigeal germinat ion, and 3 to 4 weeks are required for the cotyledons tocom plete ly deple te the endo sperm and be free from any residual endospe rm (Hu xley, 1964) .

3 days 6 daysDuring g ermination

9 days

9 days 25 daysAfter germination

Fig ure 4 Germinationsensu strictoof the coffee seed(Coffeaarabicacv. Rubi). A fully imbibed seed isshown at day 3 of imbibition with no visible protuberance; a protuberance is visible from day 6 ofimbibition onw ards and radicle protrusion starts at day 9. Following germination, the radicle grow s andthe endosperm remains attached to the cotyledons. The cotyledons will completely dissolve theendos perm before they become green and autotrophic.

Ra dic le protrusion in coffee seed s und er optimal cond itions (30 C, in the dark) started aroundday 5 or 6 and at day 10 of imbibit ion 5 0% of the seed populatio n disp laye d radicleprotrusion. At day 15 of imbibit ion most of the seeds had shown radicle protrusion.Obv iously , germ inat ion is faster unde r opt imal condi t ions when env ironm ental ef fects such asvariation in day-night temperatures and soil water potential are absent. In addition,

12

8/13/2019 192247

21/92

Anatomy andmorphology ofhecoffee Coffea arabicacv.Rubi)seed

germina tion unde r field conditions is defined as seedling emergence from the soil; radicleprotrusion has already been completed som e time before em ergence . Th e developme nt of thegermina tion processsensustricton coffee seed ispresented in figure 4.

Structural description of heendospermand embryoduringgermination

The mainhemicellulosein the cell walls of coffee seeds is an insolubleP- l4)D-mannanwith 2% of galactose present in the side chains that may serve as a c arbohy dratereserve (W olfrometal.,1961;Bewley and Black, 1994).The galactose units are also found inarabinogalac tans in he coffee seed (Wolfrom and Patin, 1965).The coffee seed belongs to hegroup of seeds hat have a elatively high amount ofmannans(Wolfrometal.,1961).

Protein s, lipids and minerals are also present in the cytoplasm of the endos perm cellsand could be another source of reserves (Dentan, 1985). Endosperm c ells have rectan gularand polygonal cell types (Fig. 5 A and B ). The rectang ular cells are located adjacent to theembryo and are observed in the region of the internal endosperm w hereas the polygonal cellswere located in the external endosperm (Fig. 5 A and B). The cells of the rest of theendospe rm displayed thick cell walls, indicating the source of reserves. These cell walls areprobably degraded following germination to provide a source of energy to the growingseedlings.A different morphology was observed in the cells of the endosperm cap. These cellsare smaller and the cell w alls are thinner than the cells of the rest of the endosperm that hasthicker cell walls with thin-walled regions (Fig. 5C, 6 C and D). Plasmod esma ta have b eenobserved in the primary pit-field of the coffee endosperm walls (Dentan, 1985). Thedifference in cell size and in cell wall morphology betw een en dosperm cap and rest of theendos perm indicates that the region wh ere the radicle will protrude is predestined in coffeeseeds o allow embryo grow th, although it may not exclude the requiremen t of endosperm capdegradation prior toradicle protrusion in order ofacilitate ra dicle p rotrusion.

During the first 3 days of germination the endosperm cells expand probably as aresult of the water uptake. A t day 3 of imbibition thecells are apparently turgid indicating thatphase 2 of the germination proce ss has been attained. H owever, at day 6 of imb ibition, stillcells that were not fully imbibed, often surrounded by fully imbibed cells, are visible in theendospe rm ca p and in the rest of the endosperm (F ig. 6 C and D ). There were mo re fullyimbibed cells in the endosperm cap than in the rest of the endosperm. Apparently, thethickness of the cell wa lls and probably also the accum ulation of solutes are very im portant incontrolling wa ter uptake during germ ination.

13

8/13/2019 192247

22/92

Chapter 2

Figure5 Light m icroscopy images of the endosperm cap and rest of the endosperm of coffee seedduring germ ination. A: Cells of the rest of the endosperm adjacent to the embryo during imbibition(bar indicates 100u,m).Note the uniformity of the cell size in this region.B:Cells of the rest of theendosperm during imbibition showing the border between internal endosperm (ie) and externalendosperm (bar indicates 100\im).Note that the cells of the internal endosperm adjacent to theemb ryo are rectangular and will be the first cells to be consumed, following germination. Th e cells ofthe external endosperm (ee) have a polygonal shap e; these cells will be consumed later. C: H ighermagnification of the external endosperm region of the rest of the endosperm show ing thin-wa lledareas (bar indicates 10|lm).D: Endosperm cap of a 9 day-imbibed seed showing remn ants of asuspe nsor at the radic le tip (dark spot) (bar indicates 100um).E: Endosperm cap(ec)region andembryo (em) at 6 days of imbibition show ing compressed cells in the endosperm (bar indicates 10(im).F: Endosperm cap (ec) region and em bryo (em) in a 10-day imbibed seed sh owing loss of cellintegrity ust before radicle protrusion (bar indicates100u,m).

14

8/13/2019 192247

23/92

Anatomy andmorphology ofhecoffee Coffea arabicacv.Rubi)seed

In the endosperm cap region, various compressed cells walls were evident prior toradicle protrusion at day 6 of the imbibition p rocess (F ig. 5 E).The remna nts of the suspensorwere observed at the endosperm c ap ust prior to radicle protrusion (F ig. 5 D) and were alsoobserved in the endosperm cap region outside of the seed surface when the protub eranceappeared. As germination proceeded the compressed cells lost their integrity just beforeradicle protrusion (F ig. 5 F). Compressed cells and loss of cell integrity coincided with theappearan ce of the protuberanc e observed in the endosperm c ap preceding radicle protrusion(Fig 4,6B).Endosperm ce ll walls surrounding the embryo nic axis, ust below the radicle tip,also showed c ompress ed c ells, possibly caused by the lateral expansion of the embryo insidethe endosp erm during germination (not shown).

Obv iously, compressed c ells, oss of cell integrity and appearance of the protuberancewere he result of embryo growth inside he endosperm prior to radicle protrusion that may bedriven by embryo cell expansion, elongation or division. In the embryo cells manyintercellular spaces were observed (Fig. 6 E). At day 9 of imbibition p resence of nuc lei,nucleo li, protein bodies and large central vacuoles were observed (Fig. 6 E). Appa rently, thevacuoles are fused to form a large central vacuole prior to radicle protrusion in the embryocells,since in earlier stages of imbibition more vacuoles w ere observed in individual em bryocells.Plastids that may contain starch as a source of reserves for the growing coffee emb ryowere observed at day 9of imbibition (Fig. 6E).

Concomitantly with the morphological changes in the endosperm cap region weobserved th e developm ent of porosity in the cell walls at day 9 of imbibition (Fig. 6 F). Fortomato seed s it has been shown that hese pores are caused by he evaporation of water duringthe freeze-drying proce ss, indicating the absence of cell wall components and coinciding w itha decrea se in the force required to punc ture the endosperm, as w ell as an increase in endo-(i-man nanase activity (EC 3.2.1.78) (Tooropet al, 2000). Therefore, this porosity of the cellwalls in the endospe rm ca p of coffee seed indicated a lso that cell w all degradation took placeduring coffee seed germ ination, possibly to weaken the endosperm cap in order to facilitateradicle protrusion. Furtherm ore, degradation of the rest of the endosperm (lateral en dosperm )during germ ination m ay not be ruled out. Thu s, our observations indicate that endo spermdegradation in the coffee seed is important during germination, not only to weaken theendosp erm ca p but also as a source of reserve materials during seedling establishment w henendosperm degradation also takes place in rest of the endosperm. Prev ious w ork in coffeeseed has suggested that mobilization of mannan-rich cell walls is a post-germinativephenom enon sinc e endo-P-m annanase activity responsible for endosperm mobilization of themannanpolyme rs was d etected only after radicle protrusion (G iorgini and Co moli, 1996;Marraccinietal.,2001).

15

8/13/2019 192247

24/92

Chapter 2

Light microscopy observations did not show the presence of an aleurone layersurrounding the endosperm that could be the source of hydrolytic enzym es. In addition,incubation of endospe rm slices in etrazolium(2,3,5riphenytetrazolium chloride) solutions at30C for 16 hours, show ed a positive reaction (data not show n). This indicates that coffeeendosperm cells are alive and may be themselves the source of the hydrolytic enzymespresent w ithin the endosperm rather than being depen dent on aspecialized aleuron e cell layeras source of enzy me s. Finally, the results indicate that both embryo growth and change s inthe endospe rm tissue control germination in coffee seed.

16

8/13/2019 192247

25/92

Anatom y and morph ology of the coffee(Coffea arabicacv. Rubi) seed

Figure6Low tem perature scanning electron m icroscopy of a coffee seed. A: Transversal section of theendosperm cap from a 9 day-imbibed seed. Observe the embryo (em), endosperm cap (ec) andremnants of the sperm oderm (sd) or silver skin at the convex seed surface. B: Longitudinal section of a6 day-imbibed seed, showing the endosperm cap (ec) and rest of the endosperm (roe). Ma rk thelocalization of the embryo (em ) and the radicle tip (rt) within the endosperm prior to radicle protrusion,and the lateral expansion of the embryo cau sing a protuberance. C: Endosperm cap of a 6-day imbibedseed show ing the thinner cell walls. Observe that some cells are not completely hydrated (arrows),surrounded by fully hydrated cells of endosperm cap (ec) and em bryo (em ). D: Cells of the rest of theendosperm (roe) of 6-day imbibed seeds. Note that these cell walls are thicker than the cell walls of theendosperm cap region (Fig. 6 C) and are apparently not hydrated. Again thin-walled regions can b eobserved locally (arrows) E : Embryo cells in a 9-day imbibed seed showing intercellular space (is) anda large vacuole (V). At the cell periphery nuclei (n), nucleoli (nu) and plastids (p) are visible. F:Endos perm cap cell wall of a 9-day imbibed seed (prior to radicle protrusion), showing porosity,indicating cell wall degradation.

A c k n o w l e d g e m e n t

W e t h a n k C A P E S (Coordenacao de Aperfeicoamento de Pessoal deNfvel Superior) forfinancial support of the studies of E.A.Amaral da Silva. The seed lab at Lavras FederalUniversi ty , M G, Brazi l (UFL A) is acknow ledged for handl ing and shipping the seeds to Th eNeth er land s. We are grateful toWimvan Veenend aal a t the Laboratory of P lant Cel l Biologyfor his help with the l ight micro scop y studie s.

References

A r c i l a - P u l g a r i n ,MIa n d O r o z c o - C a s t a n o , FJ(1987) Est udio mo rfolog ico del desarro llo delembrionde cafe.Cenicafe 3 8 ,62-63 .

B e w l e y , JD and Black , M (1994) Seeds. Physiology of development and germinat ion.P lenum P ress , New York , London , .

Chin, H F and Rober ts , EH (1980) Recalci trant crop seeds. Tropical Press , Kuala L um pur .D e d e c c a , D M (1957 )Anatomiaedesenvolvimentoontogenet ico de Coffea arabica L. var.

Typ ica Cramer .Bragantia 16 ,3 1 5 - 3 5 5 .D e n t a n , E (1985) The microscopic structure of the coffee bean.In MN Cl i f ford and KC

W ilson, eds, Coffee bo tany, b iochemist ry and product ion of beans and beve rage. The AviPubl ish ing Co mp any , W estpor t , Conne ct icut , pp 284-3 04.

Giorg in i , JF and C omol i , E(1996)Effect of emb ryo and exogenous G A3o nendospermicendo- |3-m annanas e act iv i ty of Coffea arabica L. dur ing germinat ion and ear ly seedl inggrowth .Revista B rasileira de Fisiologia Vegetal 8, 43 -49.

17

8/13/2019 192247

26/92

Chapter2

Giorgini, JF and C ampos, CASP(1992)Cha nges in the content of soluble sugars and starchsynthesis and degradation du ring germination and seedling growth ofCoffeaarabicaL.RevistaBrasileirade Fisiologia Vegetal4 , 11-15.

Huxley, PA(1964) Some factors w hich can regulate germination and influence viability ofcoffee seeds .Proceedingsof he nternationalSeedTestingAssociation 2 9,33-60.

Huxley, PA (1965) Coffee germination test recommendations and defective seed types.Proceedings of he nternational SeedTestingAssociation3 0,705-715.

Krug, CA and C arvalho, A(1939) Genetical proof of the existence of coffee endosp erm.Nature144,515.

Ma estri, M and V ieira,C(1961)Nota sobre areducaodaporcentagemdegerminacaodesementes decafe por efeito doacido giberelico.Revista Ceres11 ,247-249.

Marraccini, P, Rogers, WJ and Allard, C (2001) Molecular and biochemicalcharac terization ofendo-(3-mannanasefrom germina ting coffee(Coffea arabica)grains.Planta213,296-308.

Mendes,AJT (1941) Cytological observations in Coffea. VI. Embryo and endospermdevelopment inCoffeaarabicaL.AmericanJournal ofBotany28,784-789.

Mendes, AJT(1944)Observacoescitologicas em coffea VHI-poliembrionia.Bragantia4,693-708.

Nijsse, J and V anAelst,AC(1999)Cryo-Planingfor cryo-scanning electron microsc opy.Scanning21 ,372-378.

Nijsse, J, Erbe, E, Brantjes, NBM, Schel, JHN and W ergin, WP(1998)Low-temperaturescanning electron microsco pic observations on endosperm in imbibed and germ inatedlettuce seeds.CanadianJournal of Botany76,509-516.

Rena, AB , Malavolta, E, Rocha, M andYamada,T(1986) Cultura d o cafeeiro-fatores queafetama produtividade. 447.Piracicaba; Potafos.

Toorop , PE , Van Aelst, AC and H ilhorst, HWM (2000) The second step of the biphasicendosperm cap weakening that mediates tomato (Lycopersicon esculentum) seedgermination isunder con trol ofABA.Journal ofExperimentalBotany51,1371-1379.

Valio IFM(1996)Germination of coffee seeds(CoffeaarabicaL. cv. Mundo N ovo).Journalof Experimental Botany27,983-991.

Went, FW (1957)The expe rimental control of plant growth. Ronald Press, New York.Wellman,PL(1961) Coffee: botan y, cultivation, and utilization. London, Leonard Hill.Wolfrom, M L, Laver, ML and Patin, DL(1961)Carbohydrates of coffee bean. II. Isolation

and characterization of amannan.JournalofOrganicChemistry26,4533-4536.Wolfrom ML ,Patin DL(1964) Isolation and characterization of cellulose in the coffee bean.

Agricultural and Food Chem istry12,376-377.

18

8/13/2019 192247

27/92

CHAPTER 3

ABA regulates em bryo growth potential and endosperm cap weakeningdu rin g coffee Coffeaarabicacv.Rubi) seed germ ination

E.A.Amaralda Silva,Peter E. Toorop,Adriaan C.vanAelstandHenkW.M.Hilhorst

8/13/2019 192247

28/92

Chapter3

Abstract

The m echanism and regulation of coffee seed germination w ere studied inCoffeaarabicacv. Rubi. The coffee em bryo grew inside the endosperm prior to radicle protrusion and AB Ainhibited the increase in pressure potential. There were two steps of endosperm cap wea kening.An increase in cellulase activity coincided with the first step and an increase in endo-P-manna nase activity w ith the second step. ABA inhibited the second step of endosperm capwe akening presum ably by inhibiting the activities of least two endo-P-m annana se isoforms. T heincrease in the activity of endo-P-m annanase and cellulase coincided with the decrease in theforce required to punc ture the endosperm and with the appearance of porosity in the cell walls asobserved by low tem perature sca nning electronic microscopy. Tissue printing showed that endo -P-man nanase activity w as spatially regulated in the endosperm. A ctivity w as initiated in theendosp erm cap wh ereas later during germination it could also be detected in the rest of theendosperm. Tissue printing revealed that ABA inhibited endo-P-mannanase activity in theendospe rm cap, but not in the rest of the endosperm.ABAdid not inhibit cellulase activity. The rewas a transient rise in ABA content in the embryo during imbibition, suggesting that alsoendog enous A BA m ay control emb ryo growth potential and the second step of endosperm c apwea kening during coffee seed germination.

Keywords:coffee seed, endosperm w eakening, abscisic acid, endo-P -man nanase , cellulase, cryo-scanning electron microsco py, puncture force.

20

8/13/2019 192247

29/92

8/13/2019 192247

30/92

Chapter 3

and Bradford, 1990). InDatura eroxthe increase in embryo grow th potential was insufficient toallow germination (de Miguel and Sanches, 1992). In tomato seed endo-P-mannanase(E.C.3.2.1.78)activity correlated with weakening of the endosperm cap (Groot et al.,1988;Tooropetal.,2000 ). Endo-P-m annanase activity also correlated with porosity in the end ospermcap cell walls, as observed by cryo-scanning electronic microscopy (cryo-SEM) and with adecrease in the required puncture force (Toorop et al., 2000). Other enzymes such aspolygalacturonase (Sitrit etal., 1999), cellulase (Leviatov et al., 1995) and arabinosidase(Bradford etal., 2000) have also been shown to increase in activity during tomato seedgermination. Also in muskmelon seed cellular degradation and weakening occurredconcomitantly with the decrease in puncture force (Welbaum etal., 1995). InDatura sppscanning electron micrograph s and analyses of endosperm cell wall polysacc haride c omp ositionshowed m orphological change s in themicropylarendosperm before radicle protrusion (S anchesetal.,1990). In peppe r seeds the endosperm cap displayed c ompressed cells and loss of integritybefore radicle protrusion (Watkinsetal.,1985)as well as a decrease in the required punctureforce (Watkinsetal.,1983).How ever, endo-P-mannana se activity was only detected after radicleprotrusion (Watkinsetal.,1985).

Cell wall hydrolytic enzyme s have previously been studied in coffee seed. These includ e,oc-galactosidase(EC 3.2.1.22) (Petek and Dong, 1961; Shadaksharaswamy and Ram achandra,1967), cellulase (EC 3.2.1.4), (Takaki and Dietrich, 1980 and Giorgini, 1992) and endo-P-ma nnan ase, (Giorgini and Com oli, 1996 and M arracinie tal, 2001 ). How ever, there is littleinformation about enzym e activity in relation to he germination mecha nism and its regulation.

Abscisic acid (AB A) is known to induce dormancy and inhibit seed germination (Bewleyand Black, 1994). In lettuce seed endogenous A BA inhibits endo-P-m annana se activity (Dulsonetal, 1988) and cellulase activity (Bewley, 1997).In fenugreek and carob seeds ABA suppressesthe activity ofendo-B-mannanasein the endosperm (Kontoset al., 1996). In tobacco P-1,3-glucanase (EC 3.2.1.39) correlates with endosperm rupture and ABA delays this rupture(Leubner-Metzgeret al., 1995). In the endosperm cap of tomato seed AB A does no t inhibitcellulase (T oorop, 1998 and Bradfordetal.,2000) and endo-P-m annana se activity (Tooropetal.,1996; Still and Bradford, 1997) but radicle protrusion is prevented. In the em bryo ofBrassicanapus,Schopfer and Plachy (198 5) have shown that ABA inhibited cell wall loosening. In coffeeseedValio(1976) found that endogenousABA-likesubstances and exogenous ABA causedinhibition of germination through inhibition of embryo g rowth. How ever, the role ofABAduringcoffee seed germ ination has not been described in clear detail.

22

8/13/2019 192247

31/92

ABA egulatesgrowthpotential andendospermcapweakening

The aim of the present work is to determine the targets and mechanism of the ABAcontrolled inhibition of coffee seed germination.Materials and methods

Seed source. Coffee seeds fromCoffeaarabicaL. cultivar Rubi w ere harvested in 1998inLavras-MG -Brazil,depulped me chanically, fermented and dried to 12% of moisture conten tand shipped to The Nethe rlands where hey were stored at10C.

Germination con ditions.The seed coat was removed by hand and the surface sterilised in1 %of sodium hypoc hlorite for 2 minutes . Subsequently, seeds were rinsed in water and imbibedin 10mldemineralizedwater or ABA solution in a concentration of 1000 (iM,lOOuMor 10uM(racemicm ixture; Sigma, St. Louis, Mo., USA ). Seeds w ere placed in 94-mm Petri dishes onfilter paper (no. 860, Schleiche r & Schuell, Dass el, Ge rman y). During imb ibition see ds were keptat 30 1C in the dark (Hu xley, 1965;Valio,1976).ABA solutions were prepared by dissolvingthe comp ound in1N of KOH followed by neutralisation with1N ofHC1.Fluridone solution wasprepared by dissolving the com pound in 0.1 % of acetone until com plete dissolution. Co ntrolexperiments showed that the acetone concentration used did not affect germination. Thegermination pe rcentage w as recorded daily.

Imbibitioncurve.Intact seeds were imbibed as described abov e and the fresh weight w asmeasured daily.Embryo growth.Tw enty embryos from wa ter-imbibed seeds we re isolated by cutting the

endosperm with a razor blade. Embryo length was measured by using callipers . After lengthmeasurement the embryos were separated in embryonic axes and cotyledons and these w eremeasured again.

Waterpotential y)measurements.The w ater potential(v|/)and osmotic potential \|/H)ofcoffee emb ryos from seeds imbibed in water or in AB A solutions were measured by using acalibrated thermocouple psychrometer (Model HR-33T , W escor, USA) C-52 sample cham ber(We scor, USA ). Sam ples were equilibrated for 40 minutes and 2 readings were taken beforestarting the experiments to ensure that equilibrium had been attained. Cooling time was 45seconds.The C -52 chambe r was placed in an airtight glove box kept at 100% elative humidity bya stream of water-saturated air at a constant temperature of 25 1C. Embryos w ere isolated asdescribed above and placed in the C-52 chamber for measurements. Three replications of 5embryo s were used for the measu reme nts. After mea surement of the water potential the em bryoswere put in liquid nitrogen for determ ination of the osmotic potential(\|/).After 2 hours in liquid

23

8/13/2019 192247

32/92

Chapter 3

Resul ts

Germination. Rad ic le prot rus ion s ta r ted a t 5 daysof imbibi t ion in water . ABA a t 1000 |JM comple te lyinhibi ted germinat ion. Lower concent ra t ions of 100|0.M fand 10 |xM a l l owed ge rmi na t i on fo r 3 6% and 49%, I

orespec t ive ly (Fig. 1) . Flur id one , an inhibi tor of caro tenoidbiosy nthes i s tha t a l so inhibi t s A BA a ccum ula t ion (Li andW al ton , 19 90) , acce lera ted radic le prot rus ion s igni f icant lyat a optimal concentrat ion of 50 |J .M. In the presence off lur idone the seeds requi red 8.9 days to reach 50 % ofgerm inat ion , where as in water the seeds requi red 9.9 days(F i g . 1 ) . AB A (10 00nM) in the presence of 50iiM off lur idone did not a l low radic le prot rus ion (Fig. 1) .

- O - Control0 1000uM-ABA

- - IOOUM-ABA- A - I O U M - A B A

50iMFluridone- O -Fluridone+ABA

Fig ure 1 Germination of coffeeseeds in water, 1000xM,100uMor 10 uM of AB A, 50uM ofFluridone or 50 |iM of Fluridone+ 1000 uM AB A. Data points areaverage of 4 replications of 25seeds; error bars indicate standarddeviation.Imbibition curve. T he fresh weig ht of intact seedsdu ring imb ibi t ion increa sed (p hase I) to reach a pla teau (ph ase II ) a t day 3 of imbibi t ion, an drem ained con stant unt i l day15of imbib it ion (F ig. 2).

0 2 4 6 8 10 12 14 16 18 20

Fi gu re 2 Imbibition curve during coffee seed germ ination. Datapoints are average of 100 seeds; error bars ind icate standarddeviations. Arrow indicates radicle protrusion of 50% of theseeds.

Embryo growth. Th e em bry o grew ins ide the end osp erm before radic le prot ru s ion. Inwater- imb ibed seeds there was a s igni f icant increase in both the length of the embry onic axes an dof the cotyledon s (Fig. 3 ) wh en 5 0% of the seed popu la t ion had germ inated. Th e increase inem bry o length was 1 .06 mm (3 5% ) unti l 10 days of imbibi t ion (P

8/13/2019 192247

33/92

AB A regulates growth potential and endosperm cap w eakening

Fi gu re 3 Length of emb ryo, axis and cotyledons from coffeeseeds imbibed in water. The embryos were isolatedimm ediately before measu rement. D ata points are average of20 embryos;error bars indicate standard dev iation.

Figure 4 Water potential 4* ( ) ,osmotic potential4*,,(O) and pressurepotential H^ (T) of coffee embryosisolated from water-imbibed (A) seedsand from seeds imbibed in1000 jMofABA (B); error bars indicate standarddeviation.

- # - Embryo- O - Axis- ^ - Cotyledons I , x i _A

1 2 3 4 5 6 7 8 9 10Time(d)

Water potential measurements.Psych rom et r ic measu rem ents were s tar ted a t 2 daysof imbibi t ion. The embryo water potent ia l was -4.4 0 M Pa and increased to - 0 .96 M Pa a t 5 day s ofimbibi t ion. Th e osmot ic potent ia l increased from -4.5 0 MP a a t 2 days of imbibi t ion to -2.59 MP a a t 5days of imbibi t ion. Con seque nt ly, the pressu repotential increased from0.11MP a to 1 .62 M Pa a t5 days . At 6 days of imbibi t ion there was adecrease in both the water and o smo t ic poten t ia l .Th e water potent ia l decrea sed from -0 .96 M Pa to -3 .64 MPa and the osmot ic potent ia l f rom -2.59MPa to -3 .55 MPa. The pressure potent ia l a l sodecreased from1.62M Pa to a roun d 0 M Pa . After 6days of imbibi t ion water potent ia l and osmot icpotential increased ag ain (Fig. 4a ).

In ABA the embryo water potent ia lincreased from -4.31 M Pa to -1 .53 M Pa a t 5 daysof imbibi t ion and the osmot ic potent ia l f rom -4.50M Pa to -1 .85 M Pa a t 5 day s of imbibi t ion. At 6

days of imbibi t ion there was a decrea se in water potent ia l f rom -1.53 M Pa to -3 .63 M Pa . Th eosm ot ic potent ia l a l so decreas ed from -1.85 MP a to -3 .84 M Pa. No cha nge in pressure po tent ia lin AB A- im bibe d seeds wa s observed (Fig. 4b ) ; va lues were a lways s l ight ly abov e zero.

27

8/13/2019 192247

34/92

Chapter3

of endo-pVmannanase(pi4.5,pi6.5 andpi7.0). In AB A-imbibed seeds twoisoforms werecom pletely inhibited in the endosperm cap(pi4.5 andpi6.5). There was an extra isoform thatseemed to be specific for the rest of he endosperm (pi5.5) since this isoform w as not observed inthe endosperm cap (Fig. 8).ABAnhibited only one soform in the rest of theendosperm(pi4.5).

Endosperm structure during germination. The endosperm cap expanded duringimbibition prior to radicle protrusion. Endosperm expansion prior to radicle protrusion has bee ndescribed in other species as a protuberance (Werker, 1997). The protuberance observed in theendosp erm cap was detected after five days of imbibition (Fig. 9A); it was inhibited by A BA andincreased in size until radicle protrusion. Three to four cell layers wereobserved in theendospe rm cap in front of the radicle tip (Fig. 9B ).Cryo-SEM revealed compressed cells and lossof integrity of endospe rm c ap cells before radicle protrusion, coinciding w ith the protuberance(Fig. 9G). The endo sperm c ap showed thinner-walled cells than the rest of the endosperm (Fig.9B and C ). Concomitantly with the occurrence of the protuberance porosity in the walls of theendosp erm cap w as observed (Fig. 9E) as well as in the rest of the endosperm (Fig. 9F), but noporosity w as observed ea rlier during imbibition (Fig. 9 D ). From day 3 to 9 of imbibition thenumb er of cell layers in the endosperm cap showing porosity increased. The re was a gradient inporosity from higher porosity in the cell walls close to the embryo to lower porosity in cell wallsclose to the epidermal c ells. In ABA -imbibed seeds the endosperm cap also showed the sam egradient in porosity as observed in water-imbibed seeds. In the rest of the endospe rm porositywas also observed in ABA -imbibed seeds from day 6 onwards (results not shown). Initially, thepores appea red in the cell walls that were close to the embryo and at day 9 of imbibition the firstcell wall layer adjacent to he embryo was completely eroded .

Figure 9A.Coffee seedafter5daysof mbibition in water with ndication ofendosperm cap(ec) and herest of the endosperm (roe). Note the occurrence of a protuberance. B. Scanning electron micrograph oftheendosperm cap ec)at3daysof mbibition inwater;embryo(em). C. Scanning electron micrographofthe estofheendosperm (roe) at3daysof mbibition inwater.Note hat hecell wallsare hicker than ntheendospermcap.D.Scanningelectron micrographs of heendospermcap ec)at2daysof mbibition nwater.Noporosity wasdetected. E. Scanning electron micrograph of heendosperm cap (ec) at 6days ofimbibition in water.Highly porouscell wallscanbeobserved. F.Scanning electron micrograph ofheestof the endosperm (roe) at 6 days of imbibition in water. Porosity can be observed throughout the cellwalls. G. Scanning electron micrograph of theendosperm cap (ec) at9 days of imbibition in water.Cellsappearcompressed andshow ossof ntegrity.

30

8/13/2019 192247

35/92

ABA regulates growth potential and endosperm cap weaken ing

31

8/13/2019 192247

36/92

Chapter3

activities(r2=0.86 for endo-P-man nanase andr2=0.83 for cellulase). Endospe rm wea kening priorto radicle protrusion has also been demonstrated to occur in muskmelon (Welbaum et al.,1995),Capsicum annuum(Wa tkins and Cantiliffe, 1983) andDatura erox (de Miguel and Sanches,1992)which also coincided with the occurrence of enzyme activity in the endosperm . H owev er,wecan not exclude hat the embryo growth during coffee seed germination may also contribute tothe second phase of the endosperm cap we akening.

ABA only inhibited the second step of endosperm cap w eakening as w ell as endo -P-man nanase activity. Thus, the first phase of the decrease in the required puncture force cannot beattributed to endo-6 -mann anase activity w hereas the second phase may be under control of thisenzyme. Isoelectric focussing showed that there were three different isoforms of endo-P-man nanase in the endospe rm cap of coffee seed and that ABA inhibited at least two of them(pi4.5 andpi6.5). This suggests that there are endo-P-m annanase isoforms that may have a decisiverole during the second step of endosperm cap wea kening. In the rest of the endos perm A BAinhibited only oneisoform (pi4.5). In total we observed four different isoforms of endo-P-man nanase in the endosperm of coffee seed whereas Marraciniet al.,(2001) observed moreisoforms. T his difference may be due to the fact that we used seeds prior to radicle protrusion inisoelectric focusing studies w hereas these authors used 28-days imbibed seedsi.e.after radicleprotrusion. Different isoforms ofendo-B-mannanaseare also present in tomato seeds duringgermination (Tooropetal ,1996;Nonogakietal.,1998).

The first step of endosperm c ap weakening w as not inhibited by ABA and AB A did notinhibit cellulase activity. Indeed, the ncrease in cellulase activity coincided w ith the first phase ofdecrea se in puncture force in AB A-imbibed seeds. The presence of cellulase has previously beendemonstrated in coffee seed (Takaki and Dietrich, 1980; Giorgini, 1992). Tissue printingdem onstrated that cellulase activity was present throughout the endosperm du ring imbibition andno differences were observed w ith and w ithout AB A. Also in tomato seeds A BA did not inhibitcellulase activity (Tooropet al., 1998). A cDN A having high homology with known p-1-4glucanases was isolated from radicle and endosperm cap of tomato seeds prior to radicleprotrusion and A BA had no effect on its expression (Bradford, 200 0). Tom ato seeds also show abiphasic endosperm cap w eakening (Tooropet al.,2000). During the first p hase the decrea se inrequired puncture force correlated with an increase of endo-B-mannanase activity and theoccurre nce of ice crystal-induced po rosity in the cell wall as observed by scanning electronmicroscopy. During the second phase endo-P-mannanase activity and required puncture were

34

8/13/2019 192247

37/92

ABA egulatesgrowth potential andendosperm capweakening

uncoupled in AB A-imbibed seeds. Thus, tomato seeds show a similar behaviour in endospe rmweaken ing but a dissimilar pattern of endo-P-mannana se activity ascompared with coffee seeds

Cryo-SEM studies showed that the endosperm cap cells were compressed and lostintegrity before radicle protrusion. Evidently, growth of the embryo inside the endosperm cause dthe occurrence of the protuberance, as well as the compression of cells in the endosperm cap andloss of cell integrity. Cryo-SEM also showed porosity in the endosperm ca p and in the rest of theendosperm before radicle protrusion. There was a progressive increase in porosity before radicleprotrusion in the endosperm cap and in he rest of theendosperm. The same rend, however, albeitat lower levels, was observed in ABA-imbibed seeds. The porosity in the endosperm capcoincided with the decrease in required puncture force, increase in cellulase andendo-B-mannanaseactivity, and with the occurrence of specific endo-p-mannanase isoforms in theendosperm cap and in the rest of endosperm. In tomato seeds the development of porosity in theendosperm cap coincided w ith the increase in endo-P-mannanase activity and the overall decreasein required puncture force (Too ropet ai, 2000). Also inDaturaspp. eroded cell walls werepresent in the micropylar endosperm before radicle protrusion (Sanchezet ai, 1990). Moreov er,the coffee e ndosperm cap cell walls are thinner compared with the cell walls in the rest of theendosperm . The same structural difference has been described in the endosperm cap of tomato(Hilhorstetai , 1998), muskmelon (Welbaumet ai, 1990) and inDaturaspecies (Sanchezetal,1990).

There was a transient rise in ABA content in coffee embryos around day two ofimbibition and a second peak around day 5 (Fig.11).The low ering in ABA content to near ze rovalues at day 8 of imbibition coincided with radicle protrusion (50% of the seed population). Itshows that AB A is synthesizedde novoin the embryo during coffee seed imbibition and isdegraded or leached out thereafter. Fluridone, an inhibitor of ABA biosyn thesis (Li and W alton,1990)significantly advanced radicle protrusion. Therefore, ABA biosynthesis during coffee seedimbibition may contribute to the slow radicle protrusion observed in coffee seeds. Wehypothesise that ABA controls the embryo growth potential during germination, and the secondstepof endosperm cap weakening by nhibiting wo isoforms of endo -p-ma nnana se.

In conclusion, embryo growth and weakening of the endosperm cap control coffee seedgermination and ABA inhibits seed germination by controlling the second step of endosperm c apweakening and the increase in pressure potential in heembryo.

35

8/13/2019 192247

38/92

Chapter3

Leubner-Metzer, G, Friindt, C,Vogeli-Lange,R and Meins Jr, F (1995) Class I 6-1,3glucanases in endosperm of tobacco during germination.PlantPhysiology109,751-759.

Leviatov, S, Shoseyov,Oan dWolf S (1995) Involvement of endom annanase in the control oftomato seed germination at ow temperature conditions.Annuals ofBotany76,1-6.

Li, Y and W alton, DC (1990)Violaxanthin is an abscisic acid precursor in water-stressed dark-grown bean leaves.Plant Physiology92,551-559.

Ma rraccini, P, Rogers, WJ and Allard,C (2001) Molecular and biochemical characterizationofendo-P-mannanasefrom germinating coffee(Coffeaarabica)grains.Planta213,296-308.

Maestri, M a nd Vieira,C(1961)Nota sobre areducaoda porcentagem degerminacaodesementes decafe por efeito doacido giberelico.Revista Ceres11,247-249.

Mendes, AJT (1941) Cytological observations in Coffea. VI. Embryo and endospermdevelopm ent inCoffeaarabicaL.American Journal ofBotany28,784-789.

Nijsse, J , and vanAelst,AC (1999)Cryo-Planing for cryo-scanning electron microscopy.Scanning21,372-378.

Nonogaki, H, Nomaguchi, M and Okumoto,N (1998) Temporal and spatial pattern of thebiochemical activation of the endosperm during and following germination of tomato seed.Physiologia Plantarum 102,236-242.

Petek, F and Dong , T(1961) Separation et etude de deuxoi-galactosidasesdes graines du cafe.Enzimologia23,133 -142.

Raykhel, NV, Hunghes, DW and Galau, GA(1987) Anenzyme-immunoassayfor quantitativeanalysis of abscisic acid in wheat.InMolecular Biology of Plant Growth Con trol. Alan R .Liss,New York, pp. 197-207.

Rena, A and M aestri, M (1986) Fisio logia do cafeeiro.InAB Rena, E M alavolta, M R ocha, TYa mad a, eds, Cultura do cafeeiro- fatores queafetama produtividade.AssociasaoBrasileirapara aPesquisa da P otassa e do Fosfato.PojosdeCaldas-MG, pp13-85.

Sanchez, R A, Sunell, L, Labav itch, JM and Bonner, BA(1990)Changes in the endosperm cellwalls of twoDaturaspecies before radicle protrusion.Plant Physiology93,89-97.

Shadaksharaswamy,M andRamachandra,G(1968) Changes in the oligosa ccharides and theoc-galactosidasecontent of coffee seeds during soaking and germination.Phytochemistry7,715-719.

Schopfer, P and PlachyC (1985) Control of seed germination by abscisic ac id. III. Effect onembryo growth potential (minimum turgor pressure) and growth coefficient (cell wallextensibility) inBrassica napusL.Plant Physiology77,676-686.

38

8/13/2019 192247

39/92

ABA egulatesgrowthpotential andendospermcapweakening

Sitrit, Y, Hadfield, KA, Bennett, AB , Bradford, KJ and D ownie, B.(1999) Expression ofpolygalacturo nase assoc iated with tomato seed germination.Plant Physiology121,419-428.

Steinbauer, GP(1937)Dormancy and germination ofFraxinusSeeds.PlantPhysiology12,813-824.

Still, DW and Bradford, KJ (1997) Endo-6-mannanase activity from individual tomatoendosp erm caps and radicle tips in relation togermination rates.PlantPhysiology113,21-29.

Takaki, M and D ietrich, SMC (1980) Effect ofGA3and light on polysaccharide levels andmetabolism in germinating coffee seeds.Journal ofExperimentalBotany31,1643 -1649.

Takeba,G (1980) Accumula tion of free am ino acids in the tips ofnon-thermodormantembryonicaxes accounts for the increase in he growth potential of New York lettuce seeds.Plant &CellPhysiology21,1639-1644.

Toorop,PE,Bew ley, JD and Hilhorst, HWM (1996)Endo-B-mannanase soformsare present inendos perm c ap and embryo of tomato seeds, but arenot essentially linked to he completion ofgermination.Planta200,153-158.

Takeba, G(1980) Accu mulation of free am ino acids in the ips of non-thermo dorma nt emb ryonicaxes accou nts for the increase in he growth potential of New York lettuce seeds.Plant &CellPhysiology21,1639-1644.

Toorop,PE,Bewley, JD and H ilhorst, HW M (1996) Endo-6-m annanase isoforms are present inendospe rm ca p and embryo of tomato seed s, but are not essentially linked to thecomp letion ofgermination.Planta200,153 -158.

Toorop, PE(1998) The role ofendo-p-mannanaseactivity in tomato seed germ ination. PhDthesis, W ageningen A gricultural University, W ageningen, The Netherlands

Toorop, PE , van Aelst, AC and H ilhorst, HWM (2000) The second step of the biphasicendosperm cap w eakening that mediates tomato Lycopersiconesculentum)seed germ inationisunder control of ABA .Journal Experimental Botany51,1371-1379.

Valio,IFM(1976)Germination of coffee seeds(CoffeaarabicaL.cv. Mundo N ovo).JournalofExperimentalBotany27,983-991.

Van der Toorn , P and K arssen, CM(1992)Analysis of embryo growth in m ature fruits ofcelery Apiumgraveolens) Plant Physiology84,593-599.

W atkins, JT and C antliffe, DJ(1983) Mechanical resistance of the seed coat and endo spermduring germination ofCapsicumannuumat low emperature.PlantPhysiology72, 146-150.

Wa tkins, JT, Cantliffe, J D, Huber, DJ and Nell, TA (1985) Gibberellic acid stimulateddegrada tion of endosperm in pepper.Journal oftheAmerican Society of Horticulture Science110,61-65.

39

8/13/2019 192247

40/92

Chapter 4

Abstract

The m echanism of inhibition of coffee Coffeaarabicacv. Rubi) seed germination byexoge nous gibbere llins (GA s) and the requirement of germination for endogenous G A werestudied. Exogenous GA 4+ 7inhibited coffee seed g ermination. The respon se to GA4+7 showedtwo sensitivity thresh olds: a lower one between 0 and1lMand a higher one between10and100 fiM. H owe ver, radicle protrusion in coffee seed depended on thede novosynthesis ofGAs. Endogenous GAs were required for embryo cell elongation and endosperm capwe aken ing. Incubation of coffee seed in exogenousGAi+7ead to loss of embryo viability anddead cells were observed by low temperature scanning m icroscopy only when the endosp ermwas surrounding the embryo. The results described here indicate that the inhibition ofgermina tion by exogeno us GA s seems to be caused by factors that are released from theendosperm during or after its weakening, causing cell death in the embryo leading toinhibition of radicle protrusion.

Keywords: Gibberellins, endo-)3-mannanase, fS-mannosidase, cell death, puncture force,coffee seed, germination.

42

8/13/2019 192247

41/92

Supra-optimal GAconcentrations nhibitcoffee Coffea arabicacv.Rubi) seed germination

Introduction

Gibbere llins (GA s) are an absolute prerequisite for seed germination of many spec ies(Bewley, 1997). For example, GA-deficient mutants of Arabidopsis and tomato do notgerm inate in the absence of exogenous GA (K oornneef and van der Ve en, 1980; Groot andKarssen,1987).Tetcyclac is and paclobutrazol are inhibitors of GA -biosynthesis and may thusprevent seed germination (Karssen,et al. 1989; Radem acher, 2000) as at their effectiveconcentrations these inhibitors do not exert side effects. Addition of exogenous GAscom pletely reve rts the inhibitory effect of tetcyclacis and paclobutrazol, e.g. in Arabidopsis(Debeaujon and Koornneef, 2 000).

There is ample evidence that GAs induce endosperm de gradation by stimulating cellwall hydrolase activity. In pepper seeds GA 4+7 induced the decrease in the required force topuncture the endosperm cap(Watkins and Cantliffe, 1983) and stimulated endosp ermdegradation (Watkinsetal.,1985).Incelery seeds gibberellic acid induced chan ges in the cellwall structure in the endosperm (Jacobsenetal.,1976). In tobacco seeds GA4induced (3-1-3glucana se activity in the micropylar endosperm , which corresponded with endospe rm rupture(Leubner-Metzgeret al.1996). Endo-(3-mannanase and p-mannosidase, both involved inhydrolysis of galacto-ma nnans, were nduced in the micropylar endosperm ofDatura eroxbyGA (S anche z and de Miguel, 1997). In the tomato seed G A, possibly from the em bryo,triggered we akening of the endosperm ca p, induced degradation of the endosperm ce ll wallsand allowed radicle protrusion(Grpotand Karssen, 1987; Grootet al.,1988). Grootet al.(1988) have shown that the activity of endo-(3-mannanase, P-mannosidase and oc-galactosidase was enhanced in GA deficient m utant seeds gib-I) treated with exogenousGA4+7. In the absence of GA only a-galactosidase could be detected but no endo-P-mannanase and P-mannosidase.

Also in tomato GA stimulated em bryo growth (Karssenet al.1 989; Karssen andLacka,1986), possibly by enhancing the embryo growth potential. GA has been shown tostimulate elongation in hypocotyls of dark-grown lettuce seedlings (Katsu and K amisaka,1981) and in Arabidop sis GA co ntrols cell elongation in light- and dark-grow n hypoco tyls(Cow ling and Harberd, 1999).

Contrary to many reports on the stimulatory effect of GA during seed germinationand tissue elongation,GAiinhibited radicle protrusion (Valio, 1976; Takaki and D rietrich,1979a,1979b; Takaki and D ietrich, 1980) and radicle emerge nce in coffee seed (Maestri andVieira, 1961). This inhibition was proposed to be caused by ma nnose, a degradation productof the hydrolysis of mannans (Takaki and Dietrich,1980).Coffee end osperm cell walls arecompo sed m ainly of mannans (W olfrometal.,1961).Mannose has been show to inhibit A TP

43

8/13/2019 192247

42/92

Chapter4

Enzy me activity in the supernatant was assayed using75u.l Macllvainebuffer pH 5 .0, 15xl10mMp-nitrophenyl-(3-D-mannopyranno side dissolved in Ma cllvaine buffer, p H 5.0 and 60|llenzym e extract. After incubation for 2 h at 37 C he reaction w as stopped by adding 75ill0.2 MNa2CO.vThe yellow color produced w as measured atOD405n amicrotiter plate reader.The enzym e activity was expressed asp-nitrophenolreleased(nmolsec 1) per endospermcap.

Tetrazoliwnstain.Embryos were isolated and incubated in 0.1% (w/v) of 2,3,5triphenyltetrazolium c hloride (Sigma) at30Cin the dark for 16 hours according to Dias andSilva, (1986). The tetrazolium salts are used to measure the activity of dehydrogenaseenzym es as an index of the respiration rate and seed viability, distinguishing between viableand dead tissues (Copeland and McD onald, 1 995).

Cryo-scanning electron microscopy. Seeds from water-, G/Wr and mannose-imbibed seeds were prepared for cryo-scanning electron microcopy (cryo-SEM). Theembryo s were mo unted on aluminum rivets with a drop of tissue freezing m edium (TissueTek,Sakura, The Ne therlands ). After mo unting, the samples w ere plunge-frozen in liquidpropane and stored in liquid nitrogen for subsequentcryo-planingand observations. Theembryos werecryo-planedat - 90 C in acryo-ultramicrotome (Reichert-Jung UltracutE/FC4D)with a diamond knife (8 mm w ide; 45, Drukker Internationa l, The N etherlands)according toNijsse and vanAelst(1999). The planed surfaces were freeze dried for 3minutesat -89 Cand10 4Pa and sputter-coated with platinum in an OxfordCT1500HFcryotransferunit. The surfaces were photographed in a cryo-SEM (JEOL 63 00 F) at -180Cand 2.5-5.0

kV using adigital imaging system.Statistical analysis. Statistical analyses were

performed by using the general linear model (SPSS10.0.5).

Results

Radicle protrusion of coffee seeds started atday five of imbibition in water (da Silvaetal.2002)and light partially inhibited coffee seed germination(Fig. 1). GA4+7 inhibited radicle protrusion in aconcentration dependent manner. However, the dose

100

80

60

40

20

0

- O -Water,dark- O - GA4+7 KXJOiiM" - GA4+7IOOJIM

V ^4+7fM- O 2*4,7

8/13/2019 192247

43/92

Supra-optimal GA concentrations inhibit coffee(Co ffea arabicacv. Rubi) seed germ ination

0 100 200 300 400PaclobutrazolandTecyclacis iM

Figure 2 Germination of coffeeseeds in paclobutrazol and intetcyclacis. Data points areaverages of 4 replications of 25seeds each; error bars indicatestandard deviation.

.0 2.0 5.0 10.0

Figure 3 Germination of coffeeseeds in various GA 4+7concentrations in the presence of3 00U.Mpaclobu trazol. Data po intsare averages of 4 replications of 25seeds each; error bars indicatestandard deviation.

G A 4 + 7 resulted in a reduction of the maximalge rmi na t i on by 3 5 % whe rea s 100 and 1000 |iMGA4+7l ead to a fur ther reduct ion by 30%. Apparent ly, theconcent ra t ion thresho lds for inhibi t ion were be tween 0and 10.Mand10and100u,M, resp ectively .G A 4 + 7was mo re effect ive thanGAjin inhib i t ing rad icleprot rus ion and w as used in al l expe rimen ts ( resul t s notshown). The inhibi t ion of radic le prot rus ion byexog enou s GA s was only observed in coffee seed. Th esame GA 4 + 7solut ion w as used to germinate tom ato seedand no inhibit ion of germ inat ion w as observed but thereG A ^ increased the speed of radic le prot rus ion (resul t snot shown). Te tcyc lac is and pac lobut razol , inhibi torsof GA b i osynt he s is (R ademach e r , 2000) , com pl e t e l yinhibi ted germ inat ion a t concen t ra t ions of 400(iMand300 nM, respec t ive ly (Fig. 2) . Appl ica t ion ofe x o g e n o u s G A 4 + 7overcam e the inhibit ion and a l low edgerminat ion, which exc luded the poss ibi l i ty of s ideeffects during imbibit ion, and confirmed therequi rement for GA-biosynthes i s of coffee seedgerminat ion (Fig. 3 ) . The dose-response curvedisplayed a narrow opt im um of approxim ate ly 2 Mo fG A 4 + 7 a t the pac lobut razol conc ent ra t ion used .Ge rmi na t i on i n AT P or KH 2 P 0 4 + K2 H P 0 4 did notovercome the inhibi tory e ffec t of exogenousgibbere l l ins during coffee seed germinat ion (Fig. 4) .

Water GA 0.1ATP 1ATP 10ATPPhosphate source [mM]

Fig ur e 4 Germination of coffee seeds in0.1,land 10mMATP and 20mMof P(KH2P04+K2H P 0 4) in the p resenceof 100uMof GA4+ 7. Com pare to germination in water andin 100 uM of GA 4+ 7( GA ). Data points are averages of 4replications of 25 seeds each; error bars indicate standarddeviation.

47

8/13/2019 192247

44/92

Chapter4

day 5 (Fig. 9A ). InG/Wrimbibedseeds endo-P-manna nase a ctivity in the endosperm c apfrom seeds wa s substantially higher(3-10fold) than in water-imbibed seeds until 8 days ofimbibition. After that, the activity decreased to he level of water-imbibed seeds at day 10andwas ma intained at that level in the non-germinating se eds. Endo-P -mann anase activity wasalmost com pletely inhibited in tetcyclacis andpaclobutrazol-imbibedseeds at all im bibitionintervals.

P-mannosidaseactivity. The activity of P-mannosidase in the endospe rm capfollowed the same trend as endo-P-mannana se activity. In water-imbibed seeds the activityalso increased before radicle protrusion. In GA 4+7 the activity w as higher (2-6 fold) until 8days of imbibition and decreased thereafter. Slightly lower levels of p-manno sidase activitywere detected in etcyclacis and in paclobutrazol (Fig.9B) as compared to he water control.

1.29| 1.0

\V

- # - Water- O - 100J1MGA4+7H f - 400xMTetcyclacis- - 300nMPaclobutrazol

Tetrazoliumstain.Embryos lost their viabilitywhen the seeds were imbibed in GA solutions. Thiswas particularly pronounced in the axes. The axesshowed a brown c olor, confirming that the tissue haddied. In the control the hypocotyls showed an intensered color, indicating that the embryos w ere re spiringand alive (Fig.10A).

Cryo-scanning electron microscopy. Sincetetrazolium staining demo nstrated that the embryo nicaxes ofGA4+7-imbibedseeds were dead, we visualisedthat particular region with cryo-SEM . The hypocotylregions contained patches of deteriorated cells (Fig.10B).Higher magnification revealed that the cells in thisregion had collapsed and/or had lost cellular

compartmentalization(Fig 10 C and D ). The dead or dying cells were surround ed by intactcells. The cell contents of the collapsed cells could be observed in the intercellular spaces ofthe hypocotyl cells (Fig. 10D).The groups of dead cells were located in he epiderm is, cortexand in the vascular region. W ater- andmannose-imbibedseeds only showed normal turgidcells (results not show n).

Figure 8The equired puncture forceof water-, tetcyclacis-, G A ^ - andpaclobutrazol-imbibed seeds beforeradicle protrusion. Data points areaverages of 30 measurements; errorbars ndicatestandard errorofmean.

50

8/13/2019 192247

45/92

Supra-op timal GA con centrations inhibit coffee Coffea arabicacv. Rubi) seed germ ination

Discuss ion

Action ofGA

GA4+7 substant ia lly inhibi tedgermination of coffee seeds at aconcen t ra t ion as low as 1 ( iM. Th erespo nse to GA4+7exhibi ted two sensi t ivi tythreshold s : a lowe r one be tw een 0 and 1|xM and a high er betw een 10 and 100 |i .M.This may be caused by a he terogeneouspopula t ion, consis t ing of sub-popula t ionsof seeds, displaying different sensi t ivi t iest o t he horm one . Howev e r , p rev i ous dose -re sponse expe r i ment s wi t h AB A neve rindica ted any he terogenei ty of the seedbatch. The two steps of inhibi t ion ofgerminat ion may a l so be caused by twosi tes or mechanisms of inhibi t ion wi thdi ffe rent sens i t ivi t i es . When GA-biosyn thes i s was blocked by te tcyc lac is orpac lobut razol , appl ied GA 4 + 7 s t imula tedgerminat ion up to the opt imum of 2 |J.Mafter which i t became inhibi tory. Clearly,the amoun t of appl ied GA adds up to the

endogenously synthes ized hormone. From these da ta we es t imate tha t the amount of GAssynthes ized in the seed i s in the order of a few ( iM of exogen ous G A equ iva lents . W e do no tkno w to wha t percentage ap pl ied GA s are taken up by the coffee seeds . Th e opt imal range ofGA co ncen t ra t ions for germinat ion appeared to be very narrow (Fig. 3 ) .

Ou r resul t s indica te tha t germinat ion of coffee seeds depend s onde novo synthes i s ofGA s. This has been shown for a la rge num ber of spec ies , inc luding Arab idop sis and to mato(Kars senet al., 1989; Nam baraet al., 1991) . Ho wev er , to our know ledge coffee i s the onlyspec ies tha t di splays inhibi t ion of germinat ion by GA s a t physiolog ica l co ncen t ra t ions .

Th e s i te of GA ac t ion has been propose d to be both in the end osp erm and in theemb ryo (Ka rs senet al., 1989) . In tomato seeds GA induces both em bry o growth (Ka rssenetal.,(1989) and endosp e rm cap weakeni ng (Grootetal.,19 97 ;G r o o tetal.,1998).

6 8 10Time (d)

Figure 9 Endo-fJ-mannanase (A) and |5-mannosidase (B) activities in water, 100uMofGA 4+7, 300 uM of paclobutrazol and 400uMoftetcyclacis. Data represent 3 replications of extractsfrom10endo sperm caps isolated from seeds priorto radicle protrusion. After radicle protrusion theseeds do not have endosperm cap. Error barsindicate standard d eviation.

51

8/13/2019 192247

46/92

Chapter4

Chapter4

by reactive oxygen spe cies. Our results showed that he hypocotyl turned brown in 100(iMofGA4+7,an indication of oxidation stress and/or the absence of sufficient 'reducing po wer'.Thus , what triggers cell death in the endosperm cap during or after its degradation may affectthe embryo as well, eading tocell death and,consequently, inhibition of radicle protrusion.

In the coffee seed exogenous GA m ay speed up germination related process es, e.g.endosp erm ca p w eakening. It is possible that under these conditions normal cell death of theendosperm occurs too early, with respect to embryo growth. The embryo would then beaffected by the dam aging comp onents from the endosperm c ells. In other words, too muchGA disregulates the synchronization of germination processes occ urring in the embryo andendosperm.

Ahypoth esis of he ecologicalrelevance of G A-inhibitedgermination

Argum ents in favor of a possible ecological significance of this phenom enon are 1)The narrow optimum forGA-stimulated germination. This enables fine tuning of theresponse w ithin anarrow range of GAconcentrations and enables the seed orespond to smallchange s in its environmen t; 2) Light inhibits germination of coffee seeds (V alio, 1976 andFig. 1).This makes sense in an ecological context sinceCoffeaarabicais originally classifiedas a shadow plant (Rena and Maestri, 1986). Light induces GA-biosynthesis in seeds(Hilhorst and Karssen, 1 992; Toyomasuetal.,1993). To avoid germination under full sunlight coffee seeds may have dev eloped this inhibiting mechanism . It is likely that un dernatural conditions embryonic cell death does not occur. At lower levels factors that con tributeto cell death m ay inhibit cellular processes rather than kill thecells.

Acknowledgement

We thank CAPES(CoordenacaodeAperfeicoamentode Pessoal deNfvel Superior) forfinancial support of the studies of E.A.Amaralda Silva. The seed lab at Lavras FederalUniversity-M G - Brazil (UFL A) is acknowledged for handling and shipping the seeds to TheNetherlands. We also thank M sKatjaGrolle of the department of Food Science for hertechnical advice with the material tester. Plant Research International is acknow ledged for theuse of the psychrom eter.Prof.W. Rademacher of BAS F (Germany) is acknowledged for hisgift of tetcyclacis and SYNGENTA (Enkhuizen, The Netherlands) for the gift ofpaclobutrazol.

8/13/2019 192247

47/92

Supra-optimal GAconcentrations nhibitcoffee Coffea arabicacv.Rubi)seedgermination

References

Bethke, PC, Fath, A and Jones, RL(2001) Regulation of viability and cell death horm onesin cereal aleurone.Journal of PlantPhysiology158,429-438.

Bewley,JD(1997) Seed germination and dormancy.ThePlant Cell9, 1055-1066.Copeland, LO and McDonald, MB(1995) Seed Science and Technology, Ed .3 .Kluwer

Academic Publishers, M assachusetts.Cowling RJ and H arberd, PH (1999)Gibberellins controlArabidopsishypocotyl growth via

regulation of cellular elongation.Journal ofExperimentalBotany50,1351-1357.Da Silva, EAA , Toorop, PE , van Aelst, AC and Hilhorst, HW M (2002) ABA regulates

embryo growth potential and endosperm c ap weakening du ring coffee(C offea arabicacv.Rubi) seed germination.This hesis. Chapter3.

Debeaujon, I andKoornneef M (2000) Gibberellin requirement for Arabidopsis seedgermination is determined both by testa characteristics and embryonic absc isic acid.PlantPhysiology122,415-424.

Dias, MCL a nd Silva, WR(1986)Determinacaoda viabilidade de sem entes de cafe atravesdo este de etrazolio.PesquisaAgropecudria Brasileira.21,1139-1145.

Downie, B, Hilhorst, HW M and Bewley, JD(1994)A new assay for quantifying endo-fi-mann anase activity using Congo red dye.Phytochemistry36,829-835.

Fath, A, Bethke, PC and Jones RL(2001) Enzymes that scavenge reactive oxygen speciesare down-regulated prior to gibberellic acid-induced programm ed ce ll death in barleyaleurone.Plant P hysiology126,156-166.

Groot, SPC and Kar ssen, CM(1987) Gibberellins regulate seed germination in toma to byendosperm wea kening: a study with gibberellin-deficient mutants.PlantaYl\, 525-531.

Groot, SPC , Kieliszewska-Rokicka, B, Vermeer, E and Karssen, CM(1988) Gibberellin-induced hydrolysis of endosperm cell w alls in gibberellin-deficient tomato seeds prior toradicleprotrusion.PlantaV74 500-504.

Herold, A and Lewis, DH(1977) Mannose and green plants: occurren ce, physiology andmetabolism, and use as ool to study the role of orthophosphate.New Phytologist79,1-40.

Hilhorst, HWM and Karssen, CM(1992) Seed dormancy and germ ination: The role ofabscisic acid and gibberellins and the importance of hormone m utants.Plant GrowthRegulation 11,225-238.

Huxley, PA (1965) Coffee germination test recommendations and defective seed types.Proceedings of he nternational SeedTestingAssociation 30,705-715.

55

8/13/2019 192247

48/92

Chapter4

Jacobsen, JV, Pressm an, E and Pyiiotis, NA(1976) Gibberellin-induced separation of ce llsin isolated endos perm of celery seed.Planta129,113-122.

Jang,JC and Sheen,J(1994) Sugar sensing n higher plants.ThePlant Cell6, 1665-1679.Karssen, CM and Lacka,E (1986) A revision of the hormone b alance theory of seed

dorma ncy: studies on gibberellinand/orabscisic acid-deficient mutants ofArabidopsisthaliana. InM Bopp,ed,Plant Growth Sub stances. Springer-V erlag, Berlin, pp315-323.

Karssen, CM, Zagorski, S, Kepczynski, J and Groot, SPC (1989) A key role forendoge nous gibberellins in the control of seed germination.Annuals ofBotany63 , 71-80.

Katsu, N and Kam isaka, S(1981) Effect of gibberellic acid and m etabolic inhibitors ofDN A and RNA synthesis on hypocotyl elongation and cell wall loosening in dark-grownlettuce seedlings.Plantand CellPhysiology22,327-331.

Koornneef M and van der Veen, JH(1980)Induction and analysis of gibberellin sensitivemutants inArabidopsis thaliana(L.) Heynh.Theoretical and Applied G enetics5 8, 257-263.

Leubner-Metzger, G, Frundt, C and M eins Jr, F(1996) Effects of gibbere llins, darknessand osm otica on endosperm rupture and class131-3 glucanase induction in tobacco seedgermination.Planta199,282-288.

Maestri, M and V ieira,C . (1961) Nota sobre areducaoda porcentagem degerminacaodeseme ntes de cafe CoffeaarabicaL.var. Bourbon), por efeito do acido g iberelico.RevistaCeres11 ,247-249.

Nambara,E, Akazawa, T and M cCourt, P(1991)Effects of the gibberellin biosynthesisinhibitor uniconazol on mutantsofArabidopsis.Plant Physiology97,736-738.

Nijsse, Jandvan Aelst, AC(1999)Cryo-Planingfor cryo-scanning electron microsc opy.Scanning21,372-378.

Rademacher, W (2000) Growth retardants: Effects on gibberellin biosynthe sis and othermetabolic pathwa ys.Annual R eview of Plant Physiologyand Plant Molecular Biology5 1 ,501-531.Rena, AB and M aestri, M (1986) Fisiologia do cafeeiro.InRena, AB , Malavolta, E, R ocha,M .,Yamada,T, eds, Cultura do cafeeiro - fatores queafetama produtividade . Piracicaba,Potafos, Brasil.

Sanchez, RA and de Miguel, L (1997) Phytochrome promotion of mannan-degradingenzyme activities n themicropylarendosperm ofDatura eroxseeds requires the presenceof the embry o and gibberellin synthesis.SeedScienceResearch7,27-33.

Stein, JC and Hansen, G (1999) M annose induces an endo nuclease res ponsible for D NAladdering in plant cells.Plant Physiology121,71-79.

Takaki, M and D ietrich, SMC(1980)Effect of G A3and light on polysaccharide levels andmetabo lism in germina ting cofee seeds.Journal ofExperimentalBotany3 1,1643 -1649.

56

8/13/2019 192247

49/92

Supra-optimal G A concentrations inh ibit coffee(Coffeaarabicacv.Rubi) seed germination

Takaki , M , D ie tr ich , SM C and F urtado , JS (1979a) Anatomical changes in the hardend osp erm of gibbere l l ic ac id- trea ted coffee seeds during germinat ion.Revista Brasileirade Botanica 2, 103-10 6.

T a k a k i , M a n d D i e t r i c h , S M C(1979 b) Effect of gibbere l l ic acid on perox idas e from coffeeseeds during germina t ion.Hoennea 8,29- 3 3 .

Too rop , P E, van Aels t , AC and H i lhors t , H W M (2000) The secon d s tep of the biphas icendospe rm cap weakeni ng t ha t medi a t e s t oma t o (Lycopersicon esculentum) seedgerm inat ion i s und er cont rol of AB A.Journal of Experimental Botany 5 1 , 13 71-13 79 .

Toyomasu , T , Tsuj i , H , Yamane , H , Nakayama, M , Yamaguch i , I , M urofush i , N ,T a k a h a s h i , N a n dInoueY(1993)Ligh t effects on endo gen ou s levels of gibberel l in s inphotob las t ic le t tuce seeds .Journal of Plant G rowth Regulation 12, 85-9 0.

V a l i o , IF M (1976) Germinat ion of coffee seeds (Coffea arabica L . cv . M undo No vo) .Journal o f Experimental Botany 2 7 ,9 8 3 - 9 9 1 .

W atkins , JT and Cant l i f f e , DJ(1983)Mechanica l res i s tance of the seed coa t and end osp ermduring germinat ion ofCapsicum annuum a t low tempera ture . Plant Physiology 7 2 , 146-150.

W atkins , JT, Cant l if f e , JD , H uber , DJ and N el l , TA (1985) Gibberel l ic acid st imulateddegra dat ion of end ospe rm in pepper .Journal of the American Society of Horticulture 110,61 -65 .

W o l f r o m , M L , L a v e r , M L a n d P a t i n , D L(1961)Carb ohy drate s of coffee bea n. II. Isolat io nand charac ter iza t ion of a man nan.Journal of Organic Chem istry 2 6 , 4 5 3 3 - 4 5 3 6 .

57

8/13/2019 192247

50/92

CHAPTER 5ABA reduces the abunda nce of microtubules and inhibits the irtransv ersal organisation, emb ryo cellelongation and cell division

du rin g coffee Coffeaarabicacv.Rubi) seed g ermination

E.A. A. da Silva,P. van E keren, P. E.Toorop, A.A.M. vanLammerenand Henk W. M.Hilhorst

8/13/2019 192247

51/92

Chapter5

Abstract

W e studied the growth of the coffee CoffeaarabicaL. cv. Rubi) embryo during andfollowing germ ination of water- andABA-imbibedseed s. The coffee e mbryo g rows first byincrease in cell width followed by longitudinal growth. These events coincided withreorientation and incre ase in abundance of the microtubules and w ith accumulation of (3-tubulin in he embryonic axis during and following g ermination. DNA synthesis was activatedduring im bibition and cell division started prior to radicle protrusion and its rate increasedfollowing ge rmina tion. AB A decreased the number of microtubu les, inhibited the longitudinalgrowth of the emb ryo cells, the reorganization of the microtubules and cell division in theem bryonic ax is. How ever, ABA neither inhibited sw elling of the embryo ce lls nor organellarand nuclear DNA replication.

Keywords:coffee, c ell expansion, cell elongation, cell division, microtubu les, (3-tubulin,abscisic acid.

60

8/13/2019 192247

52/92

ABA educes heabundanceofmicrotubules and nhibits ransversal organization

IntroductionSeed germination can be defined as heevent that starts with water uptak e by he seed

and is terminated with the radicle protrusion (Bew ley and Black, 1994). Ma ny b iochem icaland molecular eve nts take place during imbibition. Among these events D NA replication isnecessary at the beginning of the imbibition process for DNA repair (Osborn e, 1983 ). DN Asynthesis associated with cell division is a late event during imbibition (Bew ley, 1997).How ever, P-tubulin ac cumu lation, assembly of microtubules, nuclear DNA synthesis and celldivision have been shown to occur during imbibition of tomato seeds prior to radicleprotrusion (de Castroet al., 1995, 2000). Coffee Coffea arabicaL.) seeds contain anendosperm tissue that envelops the embryo (Krug and Carvalho, 1939; Mendes, 1941 ;Chapter 2).The differentiated embryo lies inside an embryo cavity (Rena and M aestri, 1986;Chapter 2 ), has a length of 3 to 4 mm and consists of a radicle, an axis and two smallcotyledons (M endes, 1941). During coffee seed germination the embryo grows inside theendosperm before radicle protrusion by increasing the pressure potential and cell wallextensibility in the embryo cells (Chapter 2 and Chapter 3 ).Abscisic acid (ABA) is known toinhibit seed germination in many species (Bewley and Black, 1994). During germinationAB A acts both in the embryo and in the endosperm (Hilhorst, 1995). AB A inhibits cell wallloosening in the coffee embryo (Chapter 3) as shown to occur inBrassicanapus(Schopferand Plachy, 1985).

In the embryo of apple A BA inhibited the transition of nuclei to the G 2 phase of thecell cycle and, consequ ently, cell division (Bouvier-Durand et al.,1989). How ever, ABA didnot inhibit DNA repair in embryos ofAvena atua (Elder and Osbo rne, 1 993). The coffeeembryo disp lays growth during germination but it is unknown whether this growth is causedby cell elongation, cell division or both. In coffee seeds ABA inhibited the increase in turgorand, thus, inhibited coffee embryo growth during germination (Valio, 1976; Chapter 3).Micro tubules play a crucial role in both cell elongation and cell division (God dardet al.,1994).Here we address the question w hether inhibition of germination by ABA is targeted atthe assembly and organisation of microtubules. For this we have used light m icroscopicanalysis, immuno-histochemical detection of DNA synthesis, visualization of themicrotubular cytoskeleton, flow cytometry, and detection of P-tubulin accumulation byWe stern blotting in water- and ABA -imbibed seeds.

61

8/13/2019 192247

53/92

Chapter5

- # - Water- T - 1000uMABA

pH , 6.9), for 4 hours and subsequently rinsed in 1:5 diluted M SB for 4 x 15 min utes. Duringfixation th e embryos were subm itted to vacuum.

The embryos were dehydrated in an ethanol series and embedded inbutylmethylmetacrylateaccording to Baskine t al.(199 2). To avoid oxidation during sam plepreparation ascorbic acid (0 .1 % w/v) was added. For the analysis of themicrotubularcytoskeleton the embryos w ere solated and plunge-frozen in liquid propane and transferred tofreeze substitution medium containing water free methanol and 0.1% glutaraldehyde. Thecryo-tubes con taining the embryos and freeze-fixation medium w ere incubated in a freeze

substitution unit (Cryotech-Benelux). After freezesubstitution the freeze fixation m edium w as replaced byethanol followed by embedding inbutylmethylmetacrylate andUVpolymerisation at -20Caccording to Baskine t al.(1992). Sections w ere madeeither for the detection of DNA synthesis (incorporatedBrdU) or for the visualization of the microtubularcytoskeleton (p-tubulin). Mouseanti-P-tubulin (Sigma)dilution 1:200 v/v and anti-BrdU (Amersham ) dilution1:1 v/v w ere applied. In both cases, the second antibodyused was goat anti-mouse IgG conjugated withFITC(Molecular Probes) diluted 1:100 v/v. Nuc lear DNA wascounterstained with 1mg ml 1 propidium iodide (PI)(Molecular Probes,Eugene, OR, USA).

Statistical analysis. Statistical analyses wereperformed by using the general linear model (SPSS10.0.5). ;

Results

Germination.Ra dicle protrusion of the first seed startedat day 5 of imbibition in water. ABA at 1000xMcompletely inhibited germination (Fig. 1).Hydroxyurea, a nuclear DNA synthesis inhibitor(Gornik,et al.,1997) did not inhibit germination up toa concentration of 10mM(Fig.2).

Figure 1 Germination of coffeeseeds in water and in 1000 uM ofABA. Data points are average of100 seeds; error bars indicatestandard deviation.

- - WaterV10mMhydroxyurea1mMhydroxyurea- O - 0.1mMhydroxyurea

10 12 14

Figure2Germination ofcoffee seedsin water and in differentconcentrations of hydroxyurea. Datapoints areaverage of 100seeds;errorbars ndicatestandarddeviation.

64

8/13/2019 192247

54/92

ABA educes heabundanceofmicrotubulesand nhibits ransversal organization

S 255

WaterO 1000 nMABA

0 6 d A B A 3d Water3dABA

20 25 30 35 40Length (fim)

Figure 3 Changes in dimensions of thecells of the embryonic axis uponimbibition in water during and followinggermination (GERM) (), and in 1000u.MABA (O). The embryonic axis wasdivided in 10equal parts and the cells nthe regions3 ,7 and 9 were measuredfrom days 3, 6 and 9 of imbibition. Datapoints represent the average cell lengthplotted against average cell width from150 cells in urn. Bars indicate maximalstandard deviation.Note he difference inscaleat heX-axisandY-axis.

Light microscopy.Coffee embryo axes w eretaken for light microscopy studies since this part ofthe coffee embryo ap peared to contribute m ost tocoffee em bryo growth during germination (Chapter3 ,4). The embryonic axis was divided into 10 equalparts; microscopic analysis showed that cell growthoccurred evenly in these 10 egions (data no show n).Therefore, three regions in the axis were taken forfurther analysis viz. regions 3, 7, and 9 from thecortex region. In water-imbibed seeds the cellsinitially increased both in length and w idth, and hadthe tendency to increase in length later duringgermination (Fig. 3). This indicates that the cellsfirst grew about isodiametrically and thenlongitudinally (Fig. 3). Embryo growth at themom ent of radicle protrusion w as mainly controlledby cell elongation, since the increase in width fromday 9 of imbibition to the moment of radicleprotrusion was only 3% (Fig. 3). In AB A-inhibited

seeds he points representing 3,6 and 9 days of imbibition showed littledifferences. (Fig. 3 ).He re, the cells kept their isodiametric shape. This implies that ABA effectively reduce