Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Storage protein gene expression in zygotic and somatic embryos of interior spruce Flinn, Barry Stanley 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1992_fall_flinn_barry_stanley.pdf [ 2.78MB ]
Metadata
JSON: 831-1.0086520.json
JSON-LD: 831-1.0086520-ld.json
RDF/XML (Pretty): 831-1.0086520-rdf.xml
RDF/JSON: 831-1.0086520-rdf.json
Turtle: 831-1.0086520-turtle.txt
N-Triples: 831-1.0086520-rdf-ntriples.txt
Original Record: 831-1.0086520-source.json
Full Text
831-1.0086520-fulltext.txt
Citation
831-1.0086520.ris

Full Text

STORAGE PROTEIN GENE EXPRESSION IN ZYGOTIC AND SOMATICEMBRYOS OF INTERIOR SPRUCEbyBARRY STANLEY FLINNB.Sc., Queen’s University, 1982M.Sc., Queen’s University, 1987A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFDOCTOR OF PHILOSOPHYinTHE FACULTY OF GRADUATE STUDIESDepartment of BotanyWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAMay 1992© Barry Stanley FlinnSignature(s) removed to protect privacyIn presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my wiittenpermission.(Signature)________________Department of______________The University of British ColumbiaVancouver, CanadaDate Pla..y ) 192_-DE-6 (2/88)Signature(s) removed to protect privacy11ABSTRACTStorage proteins from interior spruce (Picea glauca/engelmanhicomplex) were identified, partially characterized and used asmarkers to compare the developmental fidelity between zygotic andsomatic embryos. The major storage proteins expressed in bothembryo types had molecular weights of approximately 41, 35, 33,24 and 22 kD. The 41 kD protein was buffer and low salt-soluble,whereas the 35—33 kD and 24-22 kD proteins were high salt-solubleand disulfide linked. All of the proteins possessed severalisoelectric variants. Based on solubility and disulfide linkagecharacteristics, as well as cDNA sequences, these storageproteins were homologous to angiosperm vicilin-type (41 kD) andlegumin-type (35-33 kD, 24-22 kD) storage proteins.Somatic embryos of different genotypes matured on 40 jM ABAaccumulated significant levels of storage protein, similar to orhigher than levels found in zygotic embryos. Somatic embryos on10 M ABA displayed initial storage protein accumulation, but thelevels did not reach those found in zygotic embryos or somaticembryos matured on 40 jM ABA.Zygotic embryos and somatic embryos differentiated on 40 ,M or10 M ABA displayed differential storage protein accumulation,with the legumin-type proteins apparent before the vicilin-type,although all showed major accumulations during cotyledondevelopment. Zygotic embryos displayed a rapid, transient periodof storage protein accumulation, with maximum storage proteinlevels attained at least 1 month prior to mature seed shed. Incontrast, somatic embryos differentiated on 40 M ABA displayed aiiimore prolonged, gradual accumulation of storage proteins, whichwere still on the increase after 9 weeks of maturation on ABA.Somatic embryos on 10 jM ABA initally accumulated storageproteins, but these were rapidly degraded as the embryosgerminated precociously.Analysis of storage protein mRNA5 indicated they were presentby torpedo stage in zygotic embryos and somatic embryos maturedon 40 .LM and 10 I.LM ABA. In all cases, the transcripts increasedduring development, with those of legumin reaching high levelsprior to those of vicilin. Transcript levels in zygotic embryosincreased during cotyledon development and then declined rapidlyto very low levels at least 1 month prior to mature seed shed.Somatic embryos on 40 ,.LM ABA displayed high transcript levels fora prolonged period, and these were still present after 9 weeks,although they had declined to 50% of maximum levels. Low levelsof storage protein transcripts also appeared in somatic embryoson 10 jM ABA, but declined during precocious germination,although they were still detectable after several weeks ofprecocious germination.Osmotic stress, caused by the culture of somatic embryos onmedium containing 15% mannitol, induced storage protein andstorage protein transcript accumulation. This could be inhibitedby inclusion of the ABA-biosynthetic inhibitor, fluridone,suggesting that the increase was due to osmotic stress—inducedABA biosynthesis.ivTABLE OF CONTENTSPageABSTRACT iiTABLE OF CONTENTS ivLIST OF TABLES viLIST OF FIGURES viiACKNOWLEDGEMENTS x1. INTRODUCTION 12 . LITEP.ATtJIE REVIEW 52.1. Somatic embryogenesis in spruce 52.1.1 General introduction 52.1.2. Somatic embryo induction 52.1.3. Somatic embryo maturation 92.2. Protein reserves during embryogenesis 132.2.1. Regulation of angiosperm storageprotein accumulation 132.2.2. Spatial patterns of storage proteinaccumulation 172.2.3. Storage protein accumulation duringconifer embryogenesis 182.2.4. Comparison of storage protein accumulationin zygotic and non—zygotic embryos 192.3. Abscisic acid, osmotic stress and embryodevelopment 212.4. Summary 303. MATERIALS AND METHODS 323.l.Zygoticembryomaterial 323.2.Somaticembryomaterial 333.3. Protein extraction and electrophoresis 383.4. Protein body isolation and analysis 403 . 5 . Microscopy 413.6. Antibody production 423.7. Immunoblotting 423.8. Chlorophyll analysis 433.9. In vivo protein labelling andimmunoprecipitation 443.10. 41 kD storage protein cDNA isolation 463.11. Generation of deletion constructs forsequencing 493.12. DNA sequencing 533.13. Genomic DNA extraction, electrophoresisand blotting 553.14. RNA extraction, electrophoresis andblotting 57VPage3.15. cDNA probe production and hybridizationto blots 594 . RESULTS 614.1. Identification and characterization of zygoticembryo storage proteins 614.2. storage protein accumulation during zygoticembryo development 714.3. Identification and characterization of somaticembryo storage proteins and comparison withzygotic embryo storage proteins 784.4. storage protein accumulation in somatic embryosand comparison with zygotic embryos 934.5. Identification and characterization of a cDNAencoding the 41 kD storage protein 1014.6. RNA gel blot analysis of storage proteingene expression in zygotic embryos and somaticembryos differentiated on 40 M and 10 /iM ABA 1164.7. Analysis of storage protein expression insomatic embryos in response to osmotic stress 1265. DISCUSSION 1365.1. Interior spruce storage proteins 1365.2. Zygotic embryo storage protein expression 1385.3. Somatic embryo storage proteins 1435.4. Developmental expression of storage proteinsin somatic embryos on 40 .LM and 10 M ABA 1465.5. Osmotic stress and storage protein geneexpression 1535.6. Concluding statement 1576. LITERATURE CITED 159viLIST OF TABLESTable Page1. Somatic embryo developmental stages on 40 M ABA 362. The presence (+) or absence (—) of various storageproteins during zygotic embryo development 793. Major storage protein distribution in EK1O zygoticand W29 somatic embryos 944. Developmental changes in storage protein mRNAsin zygotic embryos as determined by scanningdensitometry of RNA gel blots 1205. Chlorophyll content (mg/g FW) of Stage 4—6 and 9week ABA somatic embryos of genotypes W29 and W70 .... 1216. Developmental changes in storage protein mRNAsin somatic embryos differentiated on 40 jM ABA asdetermined by scanning densitometry of RNA gelblots 1257. Developmental changes in storage protein mRNAsin somatic embryos differentiated on 10 ,AM ABA asdetermined by scanning densitometry of RNA gelblots 129viiLIST OF FIGURESFigure Page1. Coomassie—stained SDS—PAGE of embryo proteins 632. Coomassie-stained SDS-PAGE of zygotic embryoprotein body extracts under reduced (A), non—reduced (B) and two-dimensional SDS-PAGE ofnon—reduced extract under non—reducing conditionsfollowed by electrophoresis under reducingconditions 663. Coomassie-stained SDS-PAGE of protein bodysamples extracted under different conditions 684. Light micrograph of a longitudinal cotyledonsection from a late maturation stage zygoticembryo, stained by the periodic acid—Schiff’stechnique and counter—stained with aniline blueb lac]c 705. Silver—stained two—dimensional electrophoretogramsof zygotic embryo protein body extract examinedusing pH 5-8 ampholytes (A) or pH 3-10ampholytes (B) 736. Silver—stained two—dimensional electrophoretogramsof 3 representative zygotic embryo stages collectedduring the summer of 1988 757. Coomassie-stained SDS-PAGE of developmental stagechanges in total protein during EK1O zygoticembryogenesis 778. Abscisic acid-dependent developmental profile ofgenotype W29 819. Examples of somatic embryo developmental stages 8310. Coomassie-stained SDS-PAGE of zygotic embryo,zygotic protein body, 9 week ABA somatic embryoand somatic embryo protein body extracts 8611. Coomassie-stained SDS-PAGE of somatic embryoextracts under reduced (A), non—reduced (B) andtwo—dimensional SDS—PAGE of non—reduced extractunder non—reducing conditions followed byelectrophoresis under reducing conditions (C) 88viiiFigure Page12. Coomassie-stained SDS-PAGE of somatic embryoprotein body samples extracted under differentconditions 9013. Silver—stained two—dimensional electrophoretogramsof somatic embryo protein body extract examinedusing pH 5-8 ampholytes (A) or pH 3-10ampholytes (B) 9214. Coomassie-stained SDS-PAGE of total proteins fortwo zygotic embryo genotypes and different somaticembryo genotypes derived from them 9615. Silver—stained two—dimensional electrophoretogramsof total proteins from stage 4-9 (A) and stage 3-2(B) somatic embryos differentiated on 40 jM ABA 9816. Coomassie-stained SDS-PAGE of developmental stagechanges in total somatic embryo protein ofgenotype W29 matured on 40 jM ABA 10017. Relative quantification of 41 kD protein immunoblotsduring zygotic and somatic embryo development 10318. Relative quantification of 24 + 22 kD proteinimmunoblots during zygotic and somatic embryodevelopment 10519. Sequence analysis of the spruce 115A.0 cDNA clone 10820. Amino acid sequence comparison of 115A.0 with otherangiosperm vicilin—type storage protein sequences 11021. Pulse:chase labelling of late cotyledonary somaticembryos differentiated on 40 j.LM ABA 11322. DNA gel blot analysis of spruce DNA probed with 115A.0cDNA 11523. changes in total proteins and storage protein mRNAduring zygotic embryo development 11824. Changes in total proteins and storage protein mRNAduring somatic embryo development on 40 M ABA 12325. Changes in total proteins and storage protein mRNAduring somatic embryo development on 10 jM ABA 12826. Effects of culture on media containing no growthregulators, 40 M ABA, 15% mannitol or fluridoneon somatic embryo development 131ixFigure Page27. changes in total proteins and storage protein mRNAin somatic embryos exposed to no growth regulators,40 M ABA, 15% mannitol or fluridone 134xACKNOWLEDGEMENTSI would like to acknowledge the financial support provided bythe B.C. Science Council GREAT Award program and the NationalResearch Council of Canada, as well as funds provided byForestry Canada and the B.C. Ministry of Forests through theForest Resource Development Agreement.Of the many people I would like to acknowledge, I wouldinitially like to thank my parents, for their love and supportthroughout the years, and I would like to extend my heartfeltaffection and gratitude to Karen Jackson for her friendship andcompanionship over the past 1½ years. Many thanks are extendedto the members of the Forest Biotechnology Centre at B.C.Research, especially Wayne Lazaroff, Stuart Pritchard, FionaWebster and Dr. Craig Newton, for their encouragement, help anddiscussions over the past few years. I am truly indebted toSusanna Grimes, Stephanie Mclnnis and Jocelyn Steedman for theirhelp with media preparation during this project.I would like to thank the members of my supervisory committee(Drs. Carl Douglas, Anthony Glass and Ben Sutton) for their inputand suggestions during this work. I would like to thank my cosupervisor, Dr. lain E.P. Taylor, for discussions and numerousglasses of wine. Last, but certainly not least, I’d like toextend my appreciation to my lab mate, Dr. Dane Roberts, forbeing a great friend and supervisor, and for his discussions andsuggestions during the course of this project. FINISHED ATLAST!!!11. INTRODUCTIONForestry is the primary natural resource industry in BritishColumbia and throughout Canada. Spruce species (Picea spp.) formthe second largest standing timber volume, after pines, andprovide the second largest log source in B.C. (Council of ForestIndustries of B.C. 1987). The propagation of these economicallyimportant conifers is thus an essential part of any nationaleffort to ensure the long term replacement of harvested trees.Tissue culture methods are widely believed to hold the bestprospect of achieving the required mass propagation of superiorgenotypes (Cheliak and Rogers 1990, Karnosky 1981) which willprovide the opportunity to maintain and even enhance the qualityof the forest stock. Tissue culture propagation by somaticembryogenesis (embryo differentiation from somatic cells) allowsrapid production of large numbers of plants and the fastexploitation of genetic gains that have been achieved byconventional forest tree breeding programs (Cheliak and Rogers1990). Use of somatic embryogenesis improves the economicfeasibility of propagation via tissue culture, because itprovides large scale production and “bulk up” of the selectedstock. It also offers the potential for the genetic engineeringof single cells, for such traits as herbicide resistance andproduction of the anti—spruce budworm Bacillus thuringiensistoxin, and their regeneration to plants.Now that the technologies for somatic embryogenesis have beendeveloped for spruce (Attree et al. 1990ab, Becwar et al. 1989,Hakman and von Arnold 1988, Krogstrup 1990, Lelu and Bornman21990, Roberts et al. 1990ab, Tremblay 1990, von Arnold and Hakman1988, Webster et al. 1990) and other conifer species (Becwar etal. 1990, Bourgkard and Favre 1988, Durzan and Gupta 1987, Fineret al. 1989, Gupta and Durzan 1987, Nørgaard and Krogstrup 1991,von Aderkas and Bonga 1988), emphasis has moved to ensure thatthe derived plants are both healthy and vigorous. Evaluation ofalfalfa artificial seed (Redenbaugh et al. 1986) showed thatsomatic embryos that produced vigorous germinants more closelyresembled their zygotic counterparts than did the embryos thatproduced less vigorous plants. This has led to the currentwisdom that the potential of somatic embryos to produce vigorousplants can be predicted by determination of their biochemical,physiological and morphological similarity to normal zygoticembryos.The periods of synthesis and deposition of storage materialare critical for zygotic embryogenesis, seed development,germination and development of robust seedlings. In matureconifer embryos, proteins and lipids (triglycerides) are the mostprominent reserves (Bewley and Black 1985, Cyr et al. 1991, Flinnet al. 1989), and are excellent biochemical markers of zygoticand somatic embryogenesis. However, there are few reportsconcerning biochemical development in conifer zygotic embryos(see for example Janick et al. 1991, Johnson et al. 1987) andonly recently, during the course of this present study have otherlabs reported identification and characterization of coniferstorage proteins (Gifford 1988, Gifford and Tolley 1989, Green etal. 1991, Misra and Green 1990, Misra and Green 1991). Clearly,3we need to understand the biochemical processes that occur duringconifer zygotic embryogenesis before we can evaluate conifersomatic embryos.The involvement of plant growth regulators in the control ofstorage reserve deposition during angiosperm seed development hasbeen well documented (Bewley and Black 1985). Endogenousabscisic acid (ABA) levels increase during the period of reservesynthesis and deposition in developing seeds (Bewley and Black1985, Finkelstein et al. 1985) and exogenous ABA can inducestorage protein accumulation (Barratt 1986, Crouch and Sussex1981, Goffner et al. 1990). Furthermore, the addition of ABA tocultures during somatic embryo differentiation preventsprecocious germination, increases the number of embryos thatmature, and reduces the occurrence of morphologically—abnormalembryos (Ammirato 1974, Roberts et al. l990a, von Arnold andHakman 1988, Webster et al. 1990).While most studies have been carried out with angiosperms,little is known about the role of ABA in conifer embryogenesis.Most of the angiosperm studies have shown that patterns ofreserve substance deposition in somatic embryos differstemporally and/or quantitatively from those in zygotic embryos(Avjioglu and Knox 1989, Crouch 1982, Shoemaker et al. 1987,Krochko et al. 1989, Stuart et al. 1988). No comparableinformation is available from conifers, although Feirer et al.(1989) and Cyr et al. (1991) found that spruce somatic embryoscontained less triglyceride reserves than zygotic embryos andHakman and von Arnold (1988) and von Arnold and Hakman (1988)4showed that spruce somatic embryos contained storage proteinbodies and protein profiles similar to those of zygotic embryos(Hakman et al. 1990). These studies did not providedevelopmental comparisons of storage protein gene expressionbetween zygotic and somatic embryos, nor did they report the roleof ABA on these reserves.The hypothesis of this present thesis is that somatic embryosmatured in the presence of high, adequate levels of ABA willresemble their zygotic counterparts with respect to patterns ofstorage protein gene expression and overall embryo developmentthan somatic embryos matured using low levels of ABA. To testthis hypothesis, the goals of this study will be to: 1) Identifythe major storage proteins of spruce zygotic and somatic embryos;2) compare the developmental expression of these proteins at theprotein and gene transcript level during zygotic and somaticembryogenesis and; 3) Examine the effect of high and low levelsof ABA on the formation of these reserves during spruce somaticembryo development.52. LITERATURE REVIEW2.1. SOMATIC EMBRYOGENESIS IN SPRUCE2.1.1. General introductionThe advances in conifer tissue culture during the past 6—7years have led to the production of somatic embryos in species ofLarix (Klimaszewska 1989, Nagmani and Bonga 1985, von Aderkas andBonga 1988, von Aderkas et al. 1990), Abies (Nørgaard andKrogstrup 1991, Schuller et al. 1989), Picea (Attree et al.1990a, Becwar et al. 1989, Gupta and Durzan 1986a, Hakman and vonArnold 1985, Krogstrup et al. 1988, Lelu and Bornman 1990,Roberts et al. 1990ab, Roberts et al. 1991, Tautorus et al. 1990,Tremblay 1990, von Arnold and Hakman 1988, von Arnold andWoodward 1988, Webb et al. 1989, Webster et al. 1990), Pinus(Becwar et al. 1990, Becwar et al. 1991, Finer et al. 1989, Guptaand Durzan 1986b, Gupta and Durzan 1987, Jam et al. 1989, Laméand David 1990), Pseudotsuga (Durzan and Gupta 1987) and Sequoia(Bourgkard and Favre 1988). Since most of the work has beencarried out with Picea species and interior spruce (Piceaglauca/engelmanhi complex) was the object of this study, thefollowing literature review deals only with spruce.2.1.2. Somatic embryo inductionSomatic embryo induction depends on a variety of factors.Explant age plays a major role in the ability to respond tomorphogenic stimuli in culture and somatic embryogenesis has beenobtained primarily from immature embryos (for review see Attreeand Fowke 1991). There is commonly a window of competence, inwhich specific developmental stages are more amenable to6induction, and a declining capacity for embryogenic inductionwith increasing embryo age (Hakman et al. 1985, Lu and Thorpe1987, Webb et al. 1989). Generally,in Picea, early cotyledonaryembryos are the most responsive. The loss of competence isattributed to morphological, physiological and biochemicalchanges that occur during embryo maturation (Roberts et al.1989).Culture media commonly used for induction are modifications ofthe LP medium (von Arnold and Eriksson 1981) used at either full-or half-strength (Hakman and von Arnold 1988, von Arnold andHakman 1988, Webb et al. 1989), although optimal media strengthappears to be species dependent (Tautorus et al. 1990). Somaticembryogenesis has also been obtained using modified forms ofMurashige and Skoog’s (1962) medium (Gupta and Durzan l986a,Krogstrup et al. 1988) and Litvay’s (1985) medium (Tremblay 1990,Tremblay and Tremblay l99lab). The latter medium is aformulation based on the chemical composition of conifer seedmegagametophyte. Since total N, reduced N, and N03:NH4 arecritical for conifer organogenesis (David et al. 1982, Flinn etal. 1986) and angiosperm eiubryogenesis (Gleddie et al. 1983,Walker and Sato 1981, Wetherell and Dougall 1976), the differentamounts and ratios of nitrogen in the different mediaformulations are probably an important factor.Early studies with mature embryos revealed that inductionfrequencies were lower than from immature embryo explants.However, more recently, induction frequencies similar to thosefrom immature embryos have been obtained (Tremblay 1990, Verhagen7and Wann 1989, von Arnold 1987). The expansion of the window ofcompetence to include mature embryos has been obtained by variousmedia manipulations. Half—strength media formulations are oftenused (Tautorus et al. 1990, Tremblay 1990, Verhagen and Wann1989, von Arnold 1987, von Arnold and Woodward 1989). Themanipulation of NH4O3 levels, pH and amino acid composition hassubstantially enhanced induction from mature embryos (Tautorus etal. 1990, Verhagen and Wann 1989, von Arnold 1987).Somatic embryogenesis has also been achieved from germinatedseedlings ranging from several days (Lelu and Bornman 1990, Leluet al. 1990) to 3-5 weeks old (Attree et al. l990a, Mo and vonArnold 1991). Induction was achieved by manipulating media andcultural environment conditions. Some studies have suggestedthat explant pretreatment with cytokinin enhanced the embryogenicresponse (Lelu and Bornman 1990, Lelu et al. 1990), althoughothers found this not to be beneficial (Attree et al. 1990, Moand von Arnold 1991).While explant developmental stage and media composition mayaffect the capacity for embryogenic induction, other factors arealso involved. Cold storage of immature cones enhancedsubsequent induction from the excised embryos (Hakman and vonArnold 1985, Hakman and Fowke 1987), although Tremblay (1990)reported that low temperature exposure of mature excised embryosstrongly inhibited embryogenesis. The duration of mature seedstorage significantly decreased embryogenic potential (Tremblay1990). Furthermore, the vigour of the explant (Tremblay 1990),8as well as its genetic background (Hakman and von Arnold 1988,Tremblay 1990, Webb et al. 1989) are important.Auxin—cytokinin combinations are used to initiateembryogenesis. While 2,4-dichiorophenoxyacetic acid (2,4-D) andN6—benzyladenine (BA) (see Attree and Fowke 1990) are mostcommonly used, other workers have found combinations of othercytokinins with 2,4-D to be effective (Gupta and Durzan l986a,Krogstrup et al. 1988). As well, somatic embryos have beeninduced using other auxins in combination with cytokinins (Lu andThorpe 1987, Verhagen and Wann 1989, von Arnold and Hakman 1988).Another important component included in the tissue culturemedium is the carbohydrate source, with sucrose the most commonlyused. The optimum sucrose concentration varies between 1% and3%, depending on the basal medium (von Arnold and Hakman 1986,von Arnold 1987) and may also vary between different seed sourcesused for explants (Webb et al. 1989). The sucrose level used canalso influence the morphogenic response. Webb et al. (1989)found that low sucrose (1%) favoured embryogenic induction fromimmature embryos, while higher levels (3-4%) favoured theinduction of adventitious shoots.The culture environment during induction is also important.Kvaalen and von Arnold (1991) reported that the gaseousenvironment of the culture vessel affected induction. Low 02 incombination with full strength medium stimulated induction, whileon half-strength medium, high 2 was better. High CO2 levelsalso promoted embryo induction. This effect was attributed to astimulation of embryogenic tissue growth or to an inhibitory9effect by CO2 on ethylene synthesis, since ethylene is known toinhibit the growth of embryogenic tissue (Kumar et al. 1989).It has been suggested that light inhibits embryogenic tissueinduction. Von Arnold (1987) reported that culture in the darkwas better for induction and most workers follow this practice(Hakman and Fowke 1987, Gupta and Durzan l986a, Webb et al.1989). However, Verhagen and Wann (1987) reported no significantdifferences between light and dark treatments on induction.2.1.3. Somatic embryo maturationThe embryogenic cultures described in the above studiesconsist of small, single cells and cell aggregates, as well asnumerous proembryo—like structures with small, opaque, denselycytoplasmic embryos (Hakman et al. 1987, von Arnold and Woodward1988, Webb et al. 1989). Culture of embryogenic tissues underconditions used for induction results in the continued growth andproliferation of these proembryonal structures and the preventionof subsequent development. Further embryo development requiresthe transfer to conditions allowing maturation. Embryomaturation involves cell division and the expansion of theembryonal cells to form globular embryos, which then elongate anddevelop cotyledon primordia, followed by cotyledon and hypocotylelongation (Dunstan et al. 1988, Hakman and von Arnold 1988).Developmental progress to maturation or to precocious germinationdepends on the manipulation of media and other cultureconditions.In all cases, the continued maturation of spruce somaticembryos past the proembryo stage requires, at the least, the10reduction or removal of embryo—inducing growth regulator levels.Some studies have used reduced 2,4—D levels for furtherdevelopment (Gupta and Durzan 1986a, Lu and Thorpe 1987, Tremblay1990), although embryo quality and yield were low. Better yieldand maturation have been obtained in the absence of 2,4—D andwith the use of ABA alone or in combination with an auxin.However, Lu and Thorpe (1987) reported no improvement of somaticembryo development by ABA. Low levels of ABA and an auxin (1 J.LMeach) increased the number of embryos produced and the numberreaching maturity (Becwar et al. 1989, Jam et al. 1988).Furthermore, buthionine sulfoximine (an inhibitor of reducingagents) tripled somatic embryo maturation in combination with theabove growth regulators (Jam et al. 1988). Higher ABAconcentrations (5—16 M) have been used by others to promoteembryo maturation (Attree et al. 1990b, Krogstrup et al. 1988,von Arnold and Hakman 1988), but levels in excess of this rangereduce regeneration (von Arnold and Hakman 1988) and lowermaturation and germination values (Attree et al. 1990b).Comprehensive studies on the effects of ABA on quality andmaturation of interior spruce somatic embryos revealed that theoptimum ABA level for most genotypes was between 40 and 60 M,although there were genotype—related differences in maturation(Webster et al. 1990). The incorporation of low levels (0.1-10jiM) of indole-3-butyric acid (IBA) with ABA enhanced embryoproduction, cotyledon development and morphology (Roberts et al.1990a). ABA levels in the range of 10-20 jiM resulted in fewerembryos, most of which germinated precociously (Roberts et al.111990a). It is evident that a wide range of ABA levels have beenused for maturation. Boulay et al. (1988) reported that thenumber of subcultures on ABA and the concentration required tomaximize somatic embryo recovery was dependent on the number ofprevious subcultures under proliferation conditions.In addition to ABA, ABA analogues have been tested for theireffects on spruce somatic embryo maturation (Dunstan et al. 1988,1991). Of those tested, abscisyl alcohol produced resultssimilar to ABA, while the others were inhibitory.Most studies have utilized ABA exposures of 4—5 weeks.Prolonged exposures have been reported to cause swelling (vonArnold and Hakman 1988), inhibit hypocotyl and cotyledonelongation (Boulay et al. 1988) and result in poor rootdevelopment (Dunstan et al. 1988). However, Dunstan et al.(1991) reported that 9 weeks of ABA exposure stimulated epicotyldevelopment in the resulting plantlets.While ABA influences embryo maturation, other factors, such ascarbohydrates, are also important. Von Arnold and Hakman (1988)found that 90 mM (3%) sucrose was optimal for development. Highsucrose levels (120-150 mN; 4-5%) stimulated early embryodevelopment but repressed further maturation. Other workers havereported that 6% sucrose was optimal for somatic embryodevelopment (Lu and Thorpe 1987, Tremblay 1990, Tremblay andTremblay l991a). Individual carbohydrates may act differently onembryo maturation in different species. Tremblay and Tremblay(199la) reported that 6% fructose was more effective than glucose12or sucrose for maturation in red spruce, whereas in black spruceall three carbohydrates were equally effective at the 6% level.Some of the carbohydrate supplied during maturation may bereplaced by an osmoticum (Lu and Thorpe 1987, Tremblay andTremblay l991a), indicating that the carbohydrate effect onmaturation is partly osmotic. Roberts (1991) reported that lowlevels of osmoticum (2-6% mannitol) promoted globular embryoformation, and that ABA was required for development to thecotyledonary stage. After cotyledon development, high levels ofosmoticum (13-20% mannitol) could replace ABA as an inhibitor ofprecocious germination. Furthermore, a 1 week mannitol pulse incombination with ABA, followed by maturation on ABA alone,doubled the production of late cotyledonary stage embryos(Roberts 1991).The culture environment during maturation also affects theextent of embryo development. Embryo maturation was stimulatedby combinations of low 02 and high CO2 on medium containing 7.6tM ABA (Kvaalen and von Arnold 1991). This effect was not asevident at the higher (60 M) ABA levels tested. Light can alsoinfluence embryo development, with maturation in the lightfavouring the production of a greater number of mature embryoscompared to darkness (Tremblay and Tremblay l99lb).Thus, while several factors must be considered during somaticembryo induction and maturation, it is possible to obtain welldeveloped spruce somatic embryos from a number of species. Thiscapability provides the potential for biotechnologicalapplications to spruce species, as well as a system to study13factors that affect conifer embryo development withoutcomplications that arise from the effects of embryo excision fromthe seed, and the presence of the surrounding maternal tissues.2.2. PROTEIN RESERVES DURING EMBRYOGENESIS2.2.1. Regulation of angiosperm storage protein accumulationMost mature seeds contain protein, lipid and carbohydratereserves which are used by the embryo during germination and postgermination processes. Most studies have concentrated on cerealand legume reserves, due to their economic importance. However,little emphasis has been placed on the study of the major coniferreserve materials. Since proteins are prominent conifer seedreserves, they represent useful biochemical markers to study andcompare development in zygotic and somatic embryogenesis.Storage proteins are synthesized during embryo development andare degraded during germination to supply amino acids, nitrogenand carbon skeletons to the developing seedling. These depositsoccur in distinct protein bodies and are confined largely to theembryo and surrounding storage tissue (megagametophyte ingymnosperms). Many storage proteins undergo post—translationalmodifications during deposition to convert them to their correctsize for deposition (MUntz 1989, Shotwell and Larkins 1989).Storage protein synthesis tends to be limited to the cellexpansion phase, following cell division and prior to maturationand desiccation (Bewley and. Black 1985). Some pea leguminstorage protein accumulation has been noted during the celldivision phase (Domoney et al. 1980), but it was not known ifthis occurred in dividing cells or a relatively few non-dividing14cells. Recent work with pea, using in situ hybridization, showedthat storage protein mRNAs only accumulated in cells lackingmitotic activity (Hauxwell et al. 1990).Most seeds contain more than one class of storage protein,each of which has a distinct temporal accumulation pattern. Inrapeseed, the 2S protein, napin, and the uS protein, cruciferin,start to accumulate during early embryo development, with napindetectable slightly earlier than cruciferin (Crouch and Sussex1981, Murphy et al. 1989). These proteins accumulate rapidlybetween 5 and 7 weeks post—anthesis, after which accumulationslows down and levels off (Crouch and Sussex 1981, Murphy et al.1989). Napin accumulation ends when the embryo water contentbegins to decline, while cruciferin accumulation continues untilseed maturity (Crouch and Sussex 1981).In pea, accumulation of the 7S protein, vicilin, begins oneday earlier than that of the llS protein, legumin. During thefirst few days, vicilin synthesis predominates (Boulter et al.1987), reaches a maximum by 14/15 days after flowering (DAF) andthen declines. During the period from 18/19 to 20 DAF, leguminsynthesis and accumulation peaks and remains constant.In soybean, the 7S 13—conglycinin — and cL’—subunit proteinsare detectable 18 to 20 days after anthesis (DAA), while the liSglycinin subunit proteins start to accumulate between 19 and 21DAA. However, the 13—subunit of 13—conglycinin does not accumulateuntil after the onset of glycinin accumulation (Meinke et al.1981).15The temporally distinct accumulation patterns described abovefor storage proteins also occur for their mRNA levels. Inrapeseed, napin mRNA is detectable by 18 DAA, achieves itsmaximum at 27 DAA and remains high until 40 DAA. In contrast,cruciferin mRNA is detected 3 days later at 21 DAA and peaks at40 DAA (Finkeistein et al. 1985). In pea, vicilin mRNA isdetectable before legumin mRNA, peaks at 14 DAF, then declines.Legumin mRNA is detectable about 1 day after vicilin, peaks at 18DAF, then declines (Boulter et al. 1987, Yang et al. 1990).Similarily, in soybean, 13-conglycinin mRNA is detectable severaldays prior to glycinin mRNA (Walling et al. 1986).Members of the same protein family may or may not exhibit thesame temporal mRNA accumulation patterns. In soybean, eachglycinin gene is expressed in the same temporal framework duringembryo development (Nielsen et al. 1989). In contrast, thesoybean 13—conglycinin gene family shows differences. The mRNAsfor the o.— and a’—subunits accumulate prior to i3—subunit mRNA(Harada et al. 1989). In rapeseed, one subfamily of napin (gNa)has mRNA levels that peak and decline earlier than those of theother members of the napin family (Blundy et al. 1991).The level of storage protein mRNA5 appears to be controlledprimarily at the transcriptional level, Storage protein genesare transcriptionally activated early in embryogenesis, attainrelatively high transcription rates by mid-maturation and arerepressed prior to desiccation and dormancy (Delisle and Crouch1989, Evans et al. 1984, Harada et al. 1989, Nielsen et al. 1989,16Walling et al. 1986). These changes roughly parallel those inendogenous mRNA levels.Apart from transcriptional regulation, post—transcriptionalregulation also plays a role in controlling storage protein mRNAlevels (Delisle and Crouch 1989, Harada et al. 1989, Nielsen etal. 1989, Walling et al. 1986). In soybean, Gy2, Gy5 and G*members of the glycinin gene family are transcriptionallyactivated and repressed at the same developmental stages. By 35DAF, Gy2 and Gy5 show similar transcription rates, which arelower than that of G*. However, steady state Gy2 and Gy5 mRNAlevels are higher than those of G*, suggesting that the G* mRNAis more unstable (Nielsen et al. 1989). With soybean 13-conglycinin, the cL/0.’_ and B—subunit genes, which are alsotranscriptionally activated at the same developmental stage, showsimilar transcription rates by 25 DAF, although the c’/o.’-subunitmRNA levels are higher than 13-subunit mRNA (Harada et al. 1989).These results suggest that at this developmental stage, the B-subunit mRNA is less stable. In rapeseed both napin andcruciferin transcription rates are on the decline by 38 DAA.However, cruciferin mRNA levels remain elevated while napin mRNAlevels drop, suggesting that cruciferin mRNA is more stableduring the later stages of embryo maturation (Delisle and Crouch1989). All of these above studies suggest that developmentallyspecific changes in mRI[A stability may affect steady state mRNAlevels.While storage protein accumulation can be regulated by mRNAavailability, it can also be regulated at the level of protein17synthesis. In developing oat seeds (Chesnut et al. 1989), theamounts of avenin storage protein mRNA are equal to or greaterthan globulin storage protein mRNA during most of development,although globulin is the predominant storage protein in the seed.The high proportion of globulin suggests some type of post-transcriptional regulation, which is believed to betranslational, because there is a greater proportion of globulinmessage incorporated into polysomes (Fabijanski and Altosaar1985). Post—translational regulation also regulates storageprotein accumulation (Shuttuck-Eidens and Beachy 1985). Theseworkers used pulse:chase—labelling to detect the synthesis of 13—conglycinin 13—subunit protein well before its accumulation duringembryo development. However, this protein was rapidly turnedover by proteolysis during the early stages of cotyledonmaturation. As maturation proceeded, the stability of theprotein increased, allowing its accumulation.2.2.2. Spatial patterns of storage protein accumulationSeed storage proteins are differentially expressed, bothquantitatively and qualitatively in various tissues, and indifferent cells within the same tissue during development. Insoybean, the relative proportions of 7S and llS proteins differbetween embryonic axes and cotyledons. Axes contain very littleliS storage protein, as well as reduced levels of 13—conglycininoc.—subunit and no 13—subunit. In addition, axes contain a 13—conglycinin subunit that is not found in cotyledons (Meinke etal. 1981). In maize endosperm, storage protein compositionvaries with increasing distance from the aleurone layer. Cells18adjacent to the aleurone accumulate 13— and —zeins, but little orno °‘—zein, whereas cells further away from the aleurone containmore oc-zein (Lending and Larkins 1989). The 2S storage proteinsof Arabidopsis thaliana also show differential expressionpatterns (Guerche et al. 1990). Of the 4 members of the at2Sgene family, at2S2, at2S3 and at2S4 are expressed throughout theembryo, while at2Sl is strongly expressed only in the axis. Arecent study using in situ hybridization pointed out the changingcellular patterns of storage protein expression during rapeseedembryo development (Fernandez et al. 1991). Napin and cruciferinmRNAs accumulated initially in the cortex of the axis during lateheart stage, then in the outer face of the cotyledons duringtorpedo stage, followed by a “wave—like” spread to the inner faceof the cotyledons. No expression occurred in the root or shootmeristem during the early stages of embryogenesis, but duringmaturation drying, both mRNAs were detected in the shootmeristem.The results of the above studies indicate that theaccumulation of storage proteins during embryogenesis and seeddevelopment is a highly regulated event, in which gene expressionis controlled at a variety of levels both temporally andspatially.2.2.3. storage protein accumulation during conifer embryogenesisIt is well known that conifer seed embryos contain prominentprotein bodies (Flinn et al. 1989, Green et al. 1991, Mia andDurzan 1974), but qualitative descriptions of conifer storageproteins have only been reported over the past few years,19(Gifford 1988, Gif ford and Tolley 1989, Green et al. 1991, Misraand Green 1990, Misra and Green 1991, Stabel et al. 1990) duringthe course of this present work. Few biochemical studies havebeen carried out during conifer embryogenesis. Johnson et al.(1987) followed changes in buffer-soluble protein in developingred and white pine embryos. However, this study did not dealspecifically with storage proteins, and since most conifer seedproteins are buffer—insoluble and require SDS or urea forsolubilization (Gifford 1988, Gifford and Tolley 1989, Green etal. 1991), this study did not examine the major proteinaccumulation patterns. Recently, Janick et al. (1991) followedchanges in embryo length, dry weight and fatty acid contentduring loblolly pine seed development, but storage proteinaccumulation was not addressed specifically. Therefore, thebiochemistry, molecular biology and developmental regulation ofconifer storage proteins requires further study.2.2.4. Comparison of storage protein accumulation in zygotic andnon—zygotic embryosThe studies described in the previous sections dealt withzygotic embryogenesis. Only a few studies have concentrated onstorage protein gene expression in non—zygotic (somatic andmicrospore—derived) embryos and have shown that non—zygoticembryo storage protein expression differs temporally and/orquantitatively from the patterns observed in zygotic embryos.Shoemaker et al. (1987) noted that the pattern of storage proteinsynthesis, processing and accumulation in cotton somatic embryosparallelled that reported in zygotic embryos, although somatic20embryos accumulated their proteins at earlier stages and insmaller amounts. Similar observations were reported formicrospore—derived rapeseed embryos (Crouch 1982). In alfalfasomatic embryos, Stuart et al. (1988) found that both 7S and uSstorage proteins accumulated to only 10% of the level found inzygotic embryos, while Krochko et al. (1989) found that alfalfasomatic embryos contained altered proportions of the 7S and llSstorage proteins.However, none of the above studies included ABA during theirdifferentiation protocols. Since ABA is known to influenceembryo maturation and storage protein gene expression (seesection 2.3 of Literature Review), the absence of exogenous ABAcould explain the altered storage protein expression found inthese embryos. Recent evidence may support this view. Rapeseedmicrospore-derived embryos treated with ABA displayed similarstorage protein mRNA induction to that observed in equivalentstage zygotic embryos, however, storage protein accumulation wasnot examined. The developmental timing of storage protein geneexpression was also similar to that of zygotic embryos (Wilen etal. 1990).The few investigations carried out with conifer somaticembryos showed that those of Norway spruce and white sprucecontained distinct protein bodies (Von Arnold and Hakman 1988,Hakman and von Arnold 1988). In addition, Hakman et al. (1990)reported that Norway spruce somatic embryos contained similarproteins to those found in zygotic embryos, and Joy IV et al.(1991) found that white spruce somatic embryos that21differentiated on low ABA levels contained less lipid and totalprotein, but more starch than their zygotic counterparts.The work to date with conifer somatic embryos has not includeddevelopmental comparisons of storage protein accumulation andgene expression. These studies are required if we are to makecomparisons between zygotic and somatic embryos and increase ourunderstanding of conifer embryonic gene expression.2.3. ABSCISIC ACID, OSMOTIC STRESS AND EMBRYO DEVELOPMENTThe plant growth regulator ABA is believed to modulatenumerous aspects of plant growth and development (for reviews seeCreelman 1989, Zeevaart and Creelman 1988). It has beenimplicated in such diverse processes as stomatal closure, seedand bud dormancy, stress adaptation, gravitropism and growthinhibition. For the purpose of this review, discussion will belimited to detailed studies of ABA effects on seed/embryodevelopment.Abscisic acid has been found to increase during angiospermseed development, showing one or two peaks of accumulation,followed by a decline during late embryo development anddesiccation (Ackerson 1984a, Finkelstein et al. 1985, Galau etal. 1987, Pence 1991). It is commonly considered to function asa growth inhibitor. Immature zygotic embryos will germinateprecociously when cultured in vitro on growth regulator—freemedium, conditions which allow depletion of the endogenous ABApool (Ackerson 1984b, Finkelstein et al. 1985, Rivin and Grudt1991). Furthermore, the continued culture of these embryos onABA-containing medium prevents this germination (Eisenberg and22Mascarenhas 1985, Finkelstein et al. 1985, Rivin and Grudt 1991).Application of exogenous ABA also prevents the precociousgermination of somatic embryos (Amrnirato 1974, Boulay et al.1988, Roberts et al. 1990a). Further support for the role ofendogenous ABA in the control of embryo germination comes fromthe study of ABA-deficient and ABA-insensitive mutants, such asthe maize viviparous (vp) mutants, which are characterized by anuninterrupted progression from embryogenesis to germination (Krizet al. 1990, Rivin and Grudt 1991).While these studies suggest that ABA is inhibitory during thelater stages of development, it also has a promotory role inembryo maturation. Crouch and Sussex (1981) found that ABApromoted embryo growth in rapeseed. Similarily, soybean embryoscultured in the presence of ABA showed a close correlationbetween ABA levels and growth, with a stimulation of growth anddry weight accumulation during early phases of embryogenesis, andgrowth suppression by mid—stage embryo development (Ackersonl984b). A further promotory effect has been noted in somaticembryo differentiation systems, where ABA enhances both thenumber and morphological normality of the embryos (Ammirato 1974,Kamada and Harada 1981, Roberts et al. l990a).Abscisic acid has also been implicated in the development ofdesiccation tolerance during embryogenesis. Late EmbryogenesisAbundant (Lea) proteins accumulate during the mid to later stagesof embryo development, just prior to maturation drying (Galau etal. 1987). These proteins, because of their structure and timingof accumulation, are believed to serve a protective, role during23desiccation (Dure et al. 1989). The expression of some of theLea mRNA5 mirror changes in endogenous ABA levels, while othersdo not (Galau et al. 1987), and the precocious appearance ofthese proteins can be induced by exogenous ABA (Galau et al.1986). Furthermore, excised zygotic or somatic embryos treatedwith ABA show enhanced tolerance to partial or completedesiccation (Bochicchio et al. 1991, Kim and Janick 1991, Robertset al. l990b, Senaratna et al. 1990). Koornneef et al. (1989)found that ABA was required for the development of desiccationtolerance in Arabidopsis thaliana mutants. Using recombinantsfrom crosses containing mutations for ABA deficiency (aba) andreduced ABA sensitivity (abi3), these workers found thatdesiccation intolerance in seeds only occurred when both maternaland embryonic genotypes were double recessive; ie. seedshomozygous recessive for both ABA deficiency and reduced ABAsensitivity (aba/abi3) that arose from the self-fertilization ofan individual heterozygous for ABA deficiency and homozygousrecessive for reduced ABA sensitivity (aba/ABA, abi3/abi3) werestill capable of desiccation and were viable.In addition to the above effects on seed/embryo development,ABA has also been implicated in dormancy induction. The mostcompelling evidence for this again comes from the study ofArabidopsis mutants. Mutants deficient in ABA or with reducedABA sensitivity display reduced dormancy (Karssen et al. 1983,Koornneef et al. 1984). Furthermore, the use of double mutantshas shown that dormancy induction in the developing seed requiresembryonic ABA. Maternal ABA or exogenously applied ABA did not24induce dormancy in seeds that were homozygous for the abamutation (Koornneef et al. 1989).Different stages of seed growth and development arecharacterized by patterns of differential gene expression. Ithas been suggested that approximately 20,000 diverse genes areexpressed at the mRNA level at any particular stage of seeddevelopment (Goldberg et al. 1989). Several proteins that areexpressed during embryo development have been found to beaffected by ABA (Bartels et al. 1991, Bochicchio et al. 1991,Galau et al. 1986, Hatzopoulis et al. 1990, Williamson andQuatrano 1988). The most highly expressed genes during seeddevelopment are those of the storage proteins, whose mRNAs canrepresent at least 50% of total mRNA at mid-maturation (Goldberget al. 1989). The potential role of ABA in the regulation ofexpression of these genes has been studied intensively.Evidence for the in situ regulation of storage proteinaccumulation by ABA in zygotic embryos is based primarily on thefact that endogenous ABA levels increase during embryodevelopment and are correlated with the synthesis of storagereserves (Bewley and Black 1985, Finkelstein et al. 1985). Mostwork with zygotic embryos has utilized excised embryos culturedon ABA-containing media. This type of experiment has shown thatABA induces storage protein accumulation in immature embryos ofrapeseed (Crouch and Sussex 1981), mustard (Croissant—Sych andBopp 1988), soybean (Ackerson 1984a, Eisenberg and Mascarenhas1985) and wheat (Williamson et al. 1985), as well as in culturedbroad bean cotyledons (Barratt 1986) and endosperm of Solanum25species (Smith and Desborough 1987). However, thischaracteristic is not true for all plants, as ABA does not appearto be associated with storage protein expression in cotton,either in vivo or in vitro (Dure and Galau 1981, Galau et al.1987)In embryos where ABA does affect storage protein accumulation,the response to ABA varies with developmental stage. Youngerembryos are more responsive to ABA, while older embryos are lessso (Bray and Beachy 1985, Croissant-Sych and Bopp 1988, Eisenbergand Mascarenhas 1985, Finkelstein et al. 1985), suggesting adecline in ABA sensitivity during maturation. In addition, whileABA is required for us storage protein and mRNA accumulation inearly and mid—maturation cultured soybean embryos, it causes adecline in uS protein mRNA (Eisenberg and Mascarenhas 1985) atyounger stages.In embryos that do show storage protein response to ABA, notall storage proteins are equally affected. In cultured soybeancotyledons, the accumulation of the 13-subunit protein and mRNA of13-conglycinin is stimulated by ABA, while the oL/o.’-subunits arenot (Bray and Beachy 1985). Also, legumin storage proteinaccumulation by cultured pea embryos is not stimulated by ABA(Davies and Bedford 1982), whereas pea vicilin synthesis isenhanced by exogenous ABA application to seeds (Schroeder 1984).The enhancement of storage protein accumulation by ABA appearsto be due to increased storage protein mRNAs (Finkeistein et al.1985, Goffner et al. 1990, Kriz et al. 1990, Taylor et al. 1990).These increases are dose—responsive, with higher ABA causing26increased mRNA levels (Finkeistein et al. 1985). Differentstorage protein mRNA5 are increased to different degrees by ABA,as seen in cultured 27 DAA rapeseed embryos, where cruciferinmRNA levels were increased 2—fold and napin mRNA levels wereincreased 4-fold by 1 IM ABA (Delisle and Crouch 1989). Inaddition, Eisenberg and Mascarenhas (1985) noted that the mRNA5of different members of the soybean glycinin gene familyaccumulated to different degrees in response to ABA.The mechanism by which ABA maintains or stimulates storageprotein synthesis and accumulation is complex. The response toABA appears primarily transcriptional (Mansfield and Raikhel1990, Williamson and Quatrano 1988), with the higher mRNA levelsparallelled by higher transcription rates (Delisle and Crouch1989). However, post—transcriptional, translational and post—translational regulation have also been suggested. Williamsonand Quatrano (1988) noted that wheat Em protein (an albuminstorage protein) mRNA showed ABA-mediated stabilization,suggesting post—transcriptional regulation. However, Em proteinsynthesis rates closely parallelled Em mRNA levels, suggesting nocontrol by ABA at the translational level. Some post-translational regulation is implicated, because, in the absenceof ABA, Em synthesis continued but no appreciable accumulationoccurred. This suggests that ABA may help to stabilize theprotein and allow its accumulation (Williamson et al. 1985).Finkelstein et al. (1985) found that excised rapeseed embryoscultured on 1 M ABA accumulated cruciferin at the same rate asin seed embryos, although the mRNA levels were lower in the27cultured embryos. These workers suggested that ABA may haveenhanced cruciferin translation, since there was less cruciferinmRNA but more cruciferin protein accumulated per embryo atincreased ABA concentrations.The previously described studies used exogenous ABA to studyeffects on storage protein accumulation. Further insight intoABA effects on storage proteins have been gained from ABAmutants. Groot et al. (1991) noted that tomato mutant sitiens,which has strongly reduced ABA levels, did not differ from wildtype in seed fresh weight, dry weight or storage protein content.While this suggested no role for ABA in seed development, therewere indications that this mutant was “leaky”, with enoughendogenous ABA to allow normal development. Koornneef et al.(1989) had examined this poposal earlier by constructingArabidopsis double mutants using recombinants of ABA-deficientaba with ABA-insensitive abil or abi3. They found that aba, abi3double mutants did not display storage protein accumulation andthat aberrant seed development occurred when the reduced ABAlevel of the aba mutant was combined with the reduced sensitivityof the abi3 mutant. Finkelstein and Somerville (1990) found thatArabidopsis wild type and abil and abi2 mutants had similarlevels of storage protein, but abi3 mutants containedsignificantly less storage protein and storage protein mRNA.These results indicated that the abi3 locus is important for seedresponses to ABA and that ABA levels and sensitivity areimportant for storage protein gene expression during embryodevelopment.28The normal process of seed development includes a loss ofwater content. It has been suggested that changes in theinternal osmotic environment of the seed play a role in embryomaturation. In support of this, numerous studies have shown thatosmotic stress influences embryo development. In somatic embryodifferentiation systems, osmotic stress during early developmentcan enhance the number of globular embryos produced (Litz 1986,Roberts 1991). However, development past this point may requiresubculture to medium of lower osmolarity (Litz 1986). Inaddition, later stages of embryo development can be enhanced andaccelerated (Nadel et al. 1989) and precocious germinationinhibited by osmotic stress (Finkelstein and Crouch 1986, Roberts1991, Xu et al. 1990).Apart from morphological development, osmotic stress has beenshown to stimulate storage protein accumulation in excisedzygotic (Finkelstein and Crouch 1986, Goffner et al. 1990, Xu etal. 1990) and microspore-derived embryos (Wilen et al. 1990).Since osmotic stress can cause elevated ABA levels in vegetativetissues (Skriver and Mundy 1990), it has been suggested thatosmotic effects on embryo development are mediated via increasedABA levels. Some studies support this. Wilen et al. (1990)noted that rapeseed microspore—derived embryos cultured on 12.5%sorbitol showed induction of storage protein mRNAs, albeit atslower rates than after exogenous ABA application. Measurementsof endogenous ABA changes during sorbitol treatment revealed anincrease in ABA that preceded storage protein mRNA induction,suggesting that osmotically-induced ABA stimulated storage29protein gene expression. Similarily, broad bean cotyledonscultured on high osmoticum (18% sucrose) accumulated vicilin andlegumin storage proteins (Barratt 1986). The combined effect offluridone (an inhibitor of endogenous ABA biosynthesis) with thehigh osmoticum treatment inhibited the storage protein increase.Exogenous ABA counteracted this inhibition. Additionally, maizeABA-deficient vp5 mutants accumulated globulin storage proteinsafter exposure to ABA, whereas those cultured on high osmoticumalone did not (Rivin and Grudt 1991). This was attributed to thefact that these embryos could not synthesize ABA in response toosmotic stress.All of these studies suggest that osmotic stress-inducedalterations in ABA levels are responsible for the changes in geneexpression, and that ABA is directly involved in this response.In contrast, other workers have found that ABA changes do notoccur in response to osmotic stress. Finkelstein and Crouch(1986) noted that rapeseed embryos cultured on sorbitolaccumulated storage protein and storage protein mRNA, but notransient increase in ABA content was observed. In sunflower,llS storage protein and mRNA increased during culture on sorbitol(Goffner et al. 1990). The increase in mRNA level was morepronounced for culture on sorbitol than on ABA, and nostimulation of ABA levels or ABA biosynthesis was detected inresponse to the osmotic stress. Barratt et al. (1989) found thatpea pods cultured for 14 days on medium containing ABA or ABAplus fluridone contained no differences in legumin or vicilintranslatable mRNA, suggesting that ABA did not play a role in pea30storage protein gene expression. Finally, Xu et al. (1990) notedthat alfalfa embryos cultured on ABA or osmoticum remainedembryonic and did not germinate, but only embryos cultured onosmoticum synthesized embryo developmental proteins (includingstorage proteins). This suggested that the osmotic prevention ofgermination and maintenance of developmental gene expression wasnot mediated through ABA.Since ABA and osmoticum do inhibit water uptake (Finkeisteinand Crouch 1986, Schopfer and Plachy 1984), it may be that theinvolvement of ABA in the above responses is indirect, and thatthe major effector causing changes in developmental geneexpression is the water status of the embryo, which is altered byosmotic stress or by ABA.It is evident that the role of ABA in the regulation of geneexpression is still unclear. Further work is required toidentify putative ABA receptors and elucidate the signaltransduction pathways. The identification of upstream regulatoryregions of ABA—responsive and osmotically—induced genes shouldhelp to clarify the role of ABA and osmotic stress in generegulation.2.4. SUMMARYMajor advances in tissue culture have made the attainment ofconifer somatic embryos and plants a reality. In order to gaugethe quality of these embryos and plants, biochemical andphysiological comparisons between zygotic embryo— and somaticembryo—derived material is required. Comparisons of storageproteins in angiosperm zygotic and somatic embryos revealed31several differences between the two systems. These differencesmay be attributed to the absence of ABA during differentiation invitro. Work to define storage protein accumulation patterns andthe effects of ABA on storage proteins at both the protein andmolecular level during conifer embryogenesis has still to beaddressed.323. MATERIALS AND METHODS3.1. ZYGOTIC EMBRYO MATERIALInterior spruce from the interior of British Columbia is amixture of two closely related species Picea glauca (Moench) Vossand Picea engelrnanhi Parry (Owens and Molder 1984). Seed coneswere collected during 1988 and 1989 at the British ColumbiaMinistry of Forests Kalamalka Research Station, Vernon, B.C. fromsource tree EK1O, which was open pollinated and possessed a Piceaengelmanii maternal background. Due to poor EK1O cone yieldduring 1991, seed cones were collected from the adjacent treeEK11. Cones were received the day after collection and stored at4°C until embryo excision, which took place within a few days ofreceipt. Embryos were classified according to theirdevelopmental stage based on morphological criteria establishedby Buchholz and Stiemert (1945) and stored at -80°C untilrequired. Embryos with suspensors and small embryonal heads(proembryos) were obtained in very small numbers. Stage 2embryos were torpedo—shaped, possessed a visible apical dome butlacked cotyledon primordia. Stage 3 embryos possessed cotyledonsthat had not developed above the apical dome. Embryos collectedduring 1989 were abundant enough to allow further sub—classification into Stage 3A (cotyledons visible as small stubs,but not very large) and Stage 3B (cotyledons larger with distinctclefts between them, but not overgrowing the shoot apex).Collections during 1988 and 1991 did not contain sufficientembryos to allow this separation, so all Stage 3 embryos weregrouped together. Stage 3-4 embryos had cotyledons that had33overgrown the shoot apex, but had not yet completely covered theapical dome. Stage 4 embryos had cotyledons that completelyenclosed the apical dome. Collections were made weekly during1989 and 1991, with the last dated collection being mature seedembryos. Collections made during 1988 were not as frequent, andStage 4 embryos (excluding mature desiccated embryos) were onlyobtained at 3 collection dates, which represented increasingstages of embryo age (stage 4-1: collected July 13; Stage 4—2:collected July 27; Stage 4-3: collected August 24).To study the effects of germination on embryo proteins, seedswere surface sterilized in 10% (v/v) commercial bleach containing0.1% (v/v) Tween 80 for 15 mm, washed three times with steriledistilled water and imbibed overnight at 4°C in the dark.Imbibed seeds were placed on a water—saturated Kimpak (SeedburoEquipment Co., Chicago, IL) inside a Magenta GA7 vessel (MagentaCorp., Chicago, IL) and germinated at 27°C in a 16-h photoperiodwith a light intensity of 30 ILE/m2/sec. seeds were examined atthree—day intervals and classified according to extent of radicalemergence. Germinants of similar development were removed andstored at -80°C.3.2. SOMATIC EMBRYO MATERIALAll culture media used were autoclaved, all growth regulatorswere added prior to sterilization and all media were solidifiedwith 0.6% Noble agar (Difco Laboratories, Detroit, MI) unlessotherwise stated. The ABA used in experiments was (+,—)—cis,trans-Abscisic acid (Sigma Chemical Co., St. Louis, MO). Allcultural manipulations were carried out under sterile conditions34and all cultures were grown at 27°C in either the dark or light(25—40 bLE/m2/sec).Somatic embryo cultures were available which had been inducedfrom individual zygotic embryos of trees EK1O and PG118 (Webb etal. 1989), and identified as distinct lines (W29, W74, W76, W77from EK1O; W70 and W46 from PGll8). Embryogenic cultures weremaintained in the dark on Litvay’s medium (Litvay et al. 1985)using half—strength macronutrients containing 1% sucrose (1% ½LMmedium) and 10 tM 2,4-D and 5 M BA. Cultures were subculturedto fresh media every 2 weeks.Somatic embryo differentiation was carried out in the light,and initially used the protocol of Roberts et al. (1990a).Embryogenic tissues containing proembryos were subcultured to VEmedium (von Arnold and Eriksson 1981), containing 3.4% sucrose(3.4% VE medium), no growth regulators and 1% charcoal, for oneweek. The ABA-dependent developmental profile was obtained fromembryogenic tissues (approximately 150 mg pieces) that wereplaced onto 3.4% VE medium containing 1 jM indole-3-butyricacid (IBA) and various levels (0, 20, 40 or 60 M) of ABA for 5weeks, with subculture to fresh medium every 2 weeks. Followingthis maturation, embryos were classified according tomorphological appearance (Roberts et al. 1990a) as mature(remained opaque with well—developed cotyledons) or precociouslygerminating (exhibited extensive greening and elongation of thecotyledon—hypocotyl axis). Aberrant structures that appeared asan abnormal elongation of cotyledons from a basal callus wereclassifed as shooty structures. The number of visible structures35were counted after 5 weeks. All subsequent somatic embryodifferentiation was performed using 40 M ABA, except for thedevelopmental study to compare effects of high and low ABAlevels, in which case 10 M ABA was also used.Somatic embryos on 40 ,M ABA were classified according todevelopmental stage (Table 1), using a modification of the schemedescribed by von Arnold and Hakman (1988). Early cotyledonarysomatic embryos were designated Stage 3 embryos. Once thecotyledons had overgrown the shoot apex (Stage 3—4), the somaticembryos resembled Stage 4 zygotic embryos, and were classified asStage 4 somatic embryos. Somatic embryos reached maximum lengthafter approximately 6 weeks on ABA and did not lengthen much moreduring the time of these experiments, and were thereafterclassified as 7-week (Stage 4-7), 8-week (Stage 4-8) or 9—week(Stage 4—9) embryos. In order to compare the pattern of storageprotein accumulation during the later stages of maturation,zygotic and somatic embryos were identified on the basis of timesince onset of cotyledon development.Somatic embryos on 10 M ABA developed in a similar manner tothose on 40 ABA, but once the cotyledons had overgrown theshoot apex (Stage 4-2), the embryos showed the extensivechlorophyll accumulation and hypocotyl/cotyledon elongation thatwas characteristic of precocious germination. These embryoscould not be classified using the length/morphology criteriaapplied to non-germinating embryos on 40 jM ABA and hence werereferred to as germinants (Germ 1—6). For the developmentalcomparisons, embryos cultured on both 10 and 40 jM ABA were36TABLE 1. Somatic embryo developmental stages on 40 jM ABAStage Weeks after onset Descriptionof cotyledondevelopmentProembryos on maintenancecalliEmbryos with globularheadsTorpedo embryos withround headsTorpedo embryos with flatheadsSmall cotyledon primordiabarely visiblePrimordia enlarged withclefts visible betweenthem, but not overgrownthe shoot apexCotyledons enlarged andslightly overgrown theshoot apex, but embryoless than 1 mm longEmbryos 1.0-1.25 mm longEmbryos 1.3-1.5 mm longEmbryos 1.55—1.75 mm longEmbryos 1.8-2.0 nun longEmbryos 2.05-2.25 mm longEmbryos 2.3-2.5 mm longEmbryos 2.55-2.75 imu longEmbryos 2.55—2.75 mm longand on ABA for 7 weeksEmbryos 2.55—2.75 mm longand on ABA for 8 weeksEmbryos 2.55—2.75 nun longand on ABA for 9 weeksPeGlobRoundF 1atEarly 3-1 0Cotyledon3—2 0.23—3 1.03—4 1.44—1 1.74—2 2.14—3 2.44—4 2.94—5 3.34—6 4.14—7 4.64—8 5.6Late 4-9 6.6Cotyledon37collected for analysis after the same exposure time to ABA. Germ1 (2-2.25 nun in length) precociously germinating embryos werecollected at the same time as Stage 4-3 embryos, and Germ 6 (8+mm in length) advanced precocious germinants were collected atthe same time as Stage 4—8 embryos. The intermediate stagesformed a developmental continuum of increasing size.Embryos differentiated during the 40 M/lO jM ABAdevelopmental comparison used a modified differentiationprotocol. Embryogenic tissues were grown for 1 week insuspension culture (10% w/v) in liquid 1% ½LM medium containing2,4-D and BA rotating at 120 rpm. The suspensions were thenfiltered through 2 layers of Miracloth (Calbiochem Corp., LaJolla, CA) and washed 3 times with 1% ½LN medium containing nogrowth regulators. The resulting tissue mass was resuspended(20% w/v) in growth regulator-free 1% ½LM medium. Theresuspended tissue (750 jl) was plated onto black 0.8 j.un filterdiscs (Millipore Corp., Bedford, MA), with a filter paperabsorbant backing to remove excess medium. The filter disc wasseparated from the absorbant pad and placed onto solid 3.4% VEmedium containing 1 M IBA and 40 M ABA as described previously.For the osmotic stress experiments, embryogenic culturesobtained via the suspension culture method were differentiated onsolid 3.4% VE medium containing 1 ,M IBA and 40 ,M ABA.Embryogenic tissues containing Stage 3—4 embryos were removed andplaced onto solid 3.4% VE medium containing no growth regulators,40 M ABA, fluridone, 15% mannitol or combinations of the above.Fluridone (kindly supplied by Dr. Franklin Fong, Texas A & N38University) was made up as a 2.5 mg/mi stock solution in acetoneand added as required. All of the above compounds were added tothe media prior to autoclaving. Cultures were grown for afurther 2 weeks and then embryos were removed and stored at —8 0°Cuntil required. Four cultures were used per treatment.3.3. PROTEIN EXTRACTION AND ELECTROPHORESISAll protein and immunoblot analyses were performed using lineW29 except where indicated. For analysis of proteins by sodiumdodecylsulfate-polyacrylamide electrophoresis (SDS-PAGE),embryonic tissues were homogenized in Eppendorf tubes with 30-40j.l 4x SDS sample buffer [562.5 l 0.5 N Tris—HC1 pH 6.8, 225 j.Ll2—mercaptoethanol, 225 Ll glycerol, 9% (w/v) SDS, 0.5 mgbromophenol blue] per mg tissue. Samples were boiled for 7 mm,centrifuged (10 mm; 14,000 rpm) in a bench top microfuge andstored at -80°C until needed. Protein was determined by themodified method of Ghosh et al. (1988), in which 2 jl of samplewas dot-blotted, incubated in 1 ml 1% (w/v) SDS overnight andmeasured spectrophotometrically at 595 nm. Bovine serum albumin(BSA) was used as a standard.For two—dimensional electrophoresis, extracts in 4x SDS samplebuffer were dialyzed overnight at 4°C against 4 L deionizedwater, frozen at —80°C and lyophilized. The samples wereresuspended in 13 1 modified 2D-NH (Mayer et al. 1987) buffer[9.6 ml of buffer contained 100 jl Pharmalyte 3-10 (PharmaciaInc., Montreal, QU), 175 mg NaC1, 3.7 mg Na2EDTA, 3.8 mg EGTA,200 1 Triton X—lOO, 8.8 ing ascorbic acid, 154 mg dithiothreitol(DTT), 100 g leupeptin, 100 go-macroglobulin, 2% (w/v) SDS and39deionized water] containing 9.5 M urea. Just before sampleapplication, 0.5 p1 Pharmalyte (5-8 or 3-10) was added.For one-dimensional SDS-PAGE analysis, gels consisted of a 12%separating gel and a 5% stacking gel, with the Laenunli buffersystem (Laemmli 1970). For two—dimensional analysis, sampleswere run into a 10 cm x 2 nun tube gel. The gel solutionconsisted 4.86 g urea, 2.88 ml deionized water, 1.18 mlacrylamide-bis-acrylamide (31% T, 4.3% C), 2.03 ml 10% (v/v)Triton X-100 and 500 p1 Pharmalytes (Pharmalyte 3-10 or 450 p1Pharmalyte 5—8 + 50 p1 Pharinalyte 3-10), and was polymerized with2 p1 20% aimnonium persulfate and 3p1 TEMED per ml of gelsolution. The sample was applied at the cathode end and coveredwith overlay buffer [2 ml 10% (v/v) Triton X-lOO, Pharmalytes(500 p1 Pharmalyte 3-10 or 450 p1 Pharmalyte 5—8 + 50 p1Pharmalyte 3—10), 7.5 ml deionized water]. The cathode solutionwas 0.1 N NaOH and the anode solution was 0.06% phosphoric acid.Gels were run at 400 V constant voltage to a total of 7100 Vh.After electrophoresis, tubes were placed on ice for 20 mm, thegels extruded, incubated in 4x SDS sample buffer for 20 mm andapplied to a stacking gel. The tube gel was overlaid with 1%(w/v) agarose in 12.5 mlvi Tris-HC1 pH 6.8. After electrophoresis,gels were fixed with aqueous 40% methanol + 10% acetic acid andstained with Coomassie R-250 or silver reagent (Wray et al.1981)The gels were photographed using Ektachrome 160 (Kodak,Toronto, ON) colour slide film and quantification of proteinbands was carried out by scanning densitometry of each colour40slide at 560 nm using a DU—64 Spectrophotometer (BeckmanInstruments Inc., Mississauga, ON). Results from scans of 3individual total protein profiles were used to calculate thepercentage of the major storage proteins relative to totalprotein.3.4. PROTEIN BODY ISOLATION AND ANALYSISProtein bodies were isolated from whole embryos based on thetechnique of Chrispeels et al. (1982). Approximately 80-100mature zygotic embryos or 50 stage 4—9 somatic embryos wereplaced on ice in a petri dish containing 2 ml Medium A (100 mMTris-HCL pH 7.8, 1 mM EDTA), 12 % (w/w) sucrose and 2 mM MgC12.Embryos were macerated and the resulting slurry filtered throughone layer of 53 ,.t.m nylon mesh. The petri dish was rinsed with 1ml of the same medium. This 3 ml filtrate was layered over 8 mlMedium A containing 16% (w/w) sucrose and 2 mM MgC12 in acentrifuge tube and centrifuged (40 mm: 2,800 rpm) at 4°C, usinga SW41 swinging bucket rotor in an L8-M Ultracentrifuge (BeckmanInstruments Inc., Mississauga, ON). The protein body pellet wasresuspended in 80-100 4x SDS sample buffer, and the sampletreated as described above for SDS—PAGE. Extractions under non—reducing conditions were performed in 4x SDS sample buffer minus2—mercaptoethanol. The non—reduced samples were analyzed underreducing conditions by cutting slices from non—reducing gelsfollowing SDS-PAGE and incubating them in 4x SDS sample buffer.They were then placed on top of the stacking gel andelectrophoresed under standard conditions.41Buffer—soluble and buffer—insoluble proteins were separatedfrom the protein body pellet by extraction with 35 l of 0.05 Msodium phosphate buffer (pH 7.5) and centrifugation (10 mm;14,000 rpm). The supernatant was removed, the pellet re—extracted and centrifuged, and the supernatant fractions werepooled to give the buffer-soluble fraction. This was mixed withan equal volume of 4X SDS sample buffer and treated as describedpreviously for SDS-PAGE. The insoluble pellet was extracted with100 1 of a 1:1 mixture of sodium phosphate buffer:4x SDS samplebuffer and treated for SDS-PAGE as described previously. Thesolubility characteristics of the storage proteins were examinedusing Osborne’s (1924) criteria. The protein body pellet wasextracted twice with 25 jl of either deionized water, sodiumphosphate buffer, buffer containing 0.2 M NaC1 or buffercontaining 1 M NaC1. The extracts were mixed with 50 ,l of 4xSDS sample buffer and treated for SDS-PAGE as described above.3.5. MICROSCOPYEmbryos from late August seed collections were excised, fixedovernight in 25 mM potassium phosphate buffer (pH 6.8) containing2.5% (v/v) glutaraldehyde and post-fixed for 2 hr in buffered 1%(w/v) osmium tetroxide. Specimens were dehydrated through agraduated acetone series, transferred to propylene oxide, thenpropylene oxide:Spurr’s resin (1:1) (Spurr 1969) and embedded inSpurr’s resin. After polymerization of the resin for 24 hr at60°C, sections (1 p.m thick) were cut, stained using the periodicacid—Schiff’s (PAS) reagent (Jensen 1962) for polysaccharides,42and counter—stained with aniline blue—black (Jensen and Fisher1968) for proteins.3.6. ANTIBODY PRODUCTIONThe 41 kD or 24 + 22 kD bands were excised from SDS-PAGE gelsof protein body extracts, dried, ground to a fine powder inliquid nitrogen and added to a 1:1 mixture of phosphate-bufferedsaline:Freund’s complete adjuvant. A total of 100—200 j.g proteinwas injected into New Zealand white rabbits. Booster injectionsof 100—200 jg protein in a mixture of phosphate—bufferedsaline:Freund’s incomplete adjuvant (1:1) were given 4 weekslater. Rabbits were bled 1 week later, the blood allowed to clotovernight at 4°C then the supernatant was removed and centrifuged(15 mm; 4,000 rpm). The straw—coloured supernatant anti—serumwas removed and used for subsequent immunoblots. The 41 kD antiserum, which displayed slight cross-reactivity with a highmolecular weight embryo protein, was affinity-purified by themethod of Smith and Fisher (1984).3.7. IMMUNOBLOTTINGProteins separated by SDS-PAGE were blotted ontonitrocellulose overnight at 30 V using the Transblot apparatus(Bio-Rad, Mississauga, ON) and transfer buffer consisting ofstock buffer [glycine 100 g/L, Tris base 80 g/L pH 8.7, 2 g/LSDS]:methanol:deionized water (1:2:7). The filters were treatedwith Tris-buffered saline pH 7.4 (TBS) containing 3% (w/v)Carnation powdered skim milk (CM) for 1 hr, washed 3 times for 5mm each with TBS + 0.5% (v/v) Tween 80, then incubated overnightin TBS containing 1% (w/v) CM and a 1:400 dilution of primary43antibody solution. Filters were rinsed 3 times for 5 mm eachwith TBS + Tween, incubated for 1 hr in TBS + 1% (w/v) CMcontaining a 1:3000 dilution of alkaline phosphatase—conjugatedgoat anti-rabbit IgG (Bio-Rad, Mississauga, ON). Filters werethen rinsed 3 times for 5 mm each with TBS + Tween and colourdeveloped following the Bio-Rad protocol. The blots werephotographed and scanned by densitometry of photographic slidesas described previously.3.8. CHLOROPHYLL ANALYSISChlorophyll analysis of somatic embryos was based on theprotocol of Hiscox and Israelstam (1979). Somatic embryos oflines W29 and W70 were collected at Stages 4-6 and 4-9. Embryoswere placed into screw-capped vials containing 20 l dimethylsulfoxide (DMSO)/mg fresh weight, that had been heated at 65°Cfor 10 mm. Samples were shaken to submerge the embryos insolvent, and incubated at 65°C for 45 mm. The vials werecentrifuged (14,000 rpm) for a few seconds to recover solventdroplets condensed on the side, and the absorbence of each samplewas measured at 645 and 663 nm against a DMSO blank. Totalchlorophyll was calculated using the equation:Total [Chlorophyll) in mg/g = (20.2 0645 + 8.02 D663) x V (ml)1000 x W (g)Where D = absorbence at the wavelength statedV = total volume of the chlorophyll solutionW = weight of the fresh tissue extractedAnalyses were carried out twice with samples from separateexperiments.443.9. IN VIVO PROTEIN LABELLING AND INMUNOPRECIPITATIONStage 4—7 somatic embryos of line W29 were placed into asterile Eppendorf tube containing 30 pCi of Tran 355-Labei (amixture of 70% L-methionine, 15% L-cysteine; ICN Biomedicals,Mississauga, ON; loll Ci/mmole) in 300 p1 of liquid 3.4% VEmedium containing 1 pM IBA and 40 pM ABA. Embryos were allowedto label for 4 hr at room temperature, the solution was removed,the embryos were rinsed with medium containing 1 mM unlabelledmethionine and cultured in 500 p1 of medium containing 10 pMunlabelled methionine. Embryo samples were collected after 0(immediately after the rinse), 2, 4, 8, 16 and 24 hr of chase.Proteins were extracted from embryos using 4x SDS sample bufferas described previously and the trichioroacetic acid (TCA)precipitable radioactivity determined. Briefly, five p1 of eachsample was spotted onto a filter paper disc, allowed to dry, andincubated in cold 10% (w/v) TCA containing 1 mg/mi methionine for10 mm. The disc was boiled in hot 10% (w/v) TCA for 10 mm,washed by vacuum suction with cold 10% (w/v) TCA followed by 95%ethanol, and allowed to dry. The sample spots were cut out,placed into a scintillation vial containing 1 ml of 0.1% (w/v)SDS and allowed to stand for 2 hr. Following this, 10 ml ofAquasol (Du Pont Corp., Mississauga, ON) was added and theradioactivity was determined using an LS 1801 LiquidScintillation Counter (Beckman Instruments Inc., Mississauga,ON).Total labelled protein profiles were obtained by applicationof samples containing equal TCA-precipitable radioactivity to45each gel, followed by SDS—PAGE as described previously.Following fixation, gels were rinsed twice (30 mm each) indeionized water to remove excess fixative, placed in fluor(Enlightning; Du Pont Corp., Mississauga, ON) containing 10%(v/v) glycerol for 1 hr, dried at 60°C for 2 hr, and allowed toexpose at -80°C using Kodak X-Omat AR film.For iinmunoprecipitation, equal amounts of TCA-precipitableradioactivity were used for each reaction. For each individualsample, 11-15 l of labelled protein extract was added to 1200 ,lof BBS [10 mM sodium borate, 160 mM NaCl pH 7.9]. To this wereadded 36 j.Ll of 16.5% (w/v) BSA, 60 jl of 10% (v/v) Triton X-l00,60 l of 250 inN methionine and 5 jl (to give a final 1:275dilution) of 41 kD protein antiserum. The reaction mixture wasincubated for 1 hr with mixing at 37°C, and kept overnight at4°C. The whole solution was added to a 20 il bed volume(approximately 11.5 mg) of Protein A-Sepharose CL4B (PharmaciaInc., Montreal, QU), which had been washed 3 times with 100 leach of BBS. This mixture was shaken gently for 1 hr at roomtemperature and centrifuged (2 mm; 14,000 rpm). The supernatantwas discarded and the Protein A-Sepharose washed 3 times with BBScontaining 10% (v/v) Triton X-lOO and 250 mM methionine, followedby a single wash with 0.3x BBS. Fifteen jl of 4x SDS samplebuffer was added to the Protein A—Sepharose, the sample boiledfor 7 mm, centrifuged (5 mm; 14,000 rpm) and the supernatant,containing the immunoprecipitated protein, removed. Ten j.Ll ofeach immunoprecipitated sample was analyzed by SDS-PAGE and thegel processed for fluorography as described above.463.10. 41 kD STORAGE PROTEIN CDNA ISOLATIONA cDNA library constructed using poly (A) RNA from latecotyledonary somatic embryos cultured on 40 M ABA was obtainedfrom Dr. Craig Newton at B.C. Research. Inserts had been clonedinto the Sma I site of pUC 18 and plasmids transformed intocompetent SURE cells (Stratagene Cloning Systems, La Jolla, CA).The cDNA library was partially screened using 2 plates, eachcontaining approximately 7500 colony forming units. Colonieswere grown overnight at 37°C on YT medium [8 g/L bacto-tryptone,5 g/L bacto-yeast extract, 5 g/L NaC1] containing 100 tg/mlampicillin (YT + amp medium) and solidified with 15 g/L agar.Replica filters of the plates were made using nitrocellulose(Bio-Rad, Mississauga, ON) and placed colony side up on YT + ampplates covered with 50 ,l of 100 inN Isopropyl .B-D—Thiogalactopyranoside (IPTG). After overnight growth at 37°C,colonies on the filters were lysed for 30 mm in chloroformvapour by suspension in a TLC tank equilibrated 30 mm withchloroform. Filters were placed in a solution of 50 mM Tris-HC1pH 7.5, 150 inN NaC1 and 5 mM MgC12, containing 1 g/ml DNase, 40g/ml lysozyme and 3% (w/v) CM, and incubated overnight withagitation. They were rinsed 3 times with TBS-Tween, thenincubated in TBS-Tween containing a 1:400 dilution of affinitypurified 41 kD antibody. Washing, secondary antibody incubationand colour development of filter immunoblots was as described forSDS-PAGE immunoblots.An antibody-positive colony was identified on the replicafilter, the corresponding colony sampled and inoculated into 2 ml47of YT + amp broth for overnight growth. Following this, adilution series was made and plated onto YT + amp plates. Onehundred jAl of a dilution containing io2 colonies/mi was plated,grown overnight, a replica filter made and the immunoscreeningprocedure repeated as described previously. Several antibody—positive colonies were isolated, grown overnight in YT + ampbroth and their plasmid DNA isolated by the alkaline lysis methoddescribed by Sambrook et al. (1989). Briefly, for each sample,1.5 ml of bacterial culture was centrifuged (30 sec; 14,000 rpm),the medium removed and the pellet resuspended with 100 jLl ofSolution I [50 mM glucose, 25 mM Tris-HC1, 10 mM EDTA pH 8].Following this, 200 il of freshly prepared Solution II [0.2 NNaOH, 1% (w/v) SDS] were added and the contents mixed byinverting the tube rapidly 5 times, and then 150 jl of ice coldSolution III [60 ml 5 M potassium acetate, 11.5 ml glacial aceticacid, 28.5 ml deionized water] were added, mixed, the tube kepton ice for 5 mm and centrifuged (5 mm; 14,000 rpm) at 4°C. Thesupernatant was transferred to a fresh tube, extracted with anequal volume of phenol:chloroform (1:1), mixed, centrifuged (5mm; 14,000 rpm), and the aqueous phase removed and mixed with 2vol of 95% ethanol. The DNA was allowed to precipitate for 2 mmat room temperature, the sample centrifuged (15 mm; 14,000 rpm)at 4°C, the pellet washed with 95% ethanol, then 70% ethanol,vacuum dried and resuspended with 50 ,.L1 TE buffer [10 inN TrisHC1, 1 inN EDTA pH 8].The plasmid cDNA inserts were excised using restriction enzymedigests. All restriction enzymes were used with the appropriate48buffers and conditions described by the suppliers, and allplasmid DNA samples were treated with 5% (v/v) of 10 mg/mi RNaseduring digestion to destroy contaminating RNAs. Two p1 aliquotsof the plasmid DNA samples isolated above were digested with EcoRI—Pst I and analyzed by agarose gel electrophoresis. Ten p1 ofeach digest mix was added to 1 p1 Ficoll dye mix [0.25% (w/v)bromophenol blue, 0.25% (w/v) xylene cyanol FF, 15% (w/v) Ficoil(Pharmacia) in water], and separated on a 1.2% (w/v) agarose gelin lx TAE [50x TAE = 2 N Tris-acetate, 0.05 N EDTA pH 8],containing 0.5 pg/mi ethidium bromide, with the electrophoresisbuffer consisting of ix TAE. Examination of samples under UVlight revealed that they all contained the same sized insertfragment (1.7 kb). One colony was chosen for further use anddesignated as 115A.0. This plasmid DNA was digested with Eco RIPst I and the insert fragment separated on an agarose gel. Theinsert band was visualized under UV light, cut out and gelpurified using the Prep-A-Gene method (Bio-Rad, Mississauga, ON).Briefly, the gel fragment was placed into an Eppendorf tube, thevolume estimated and 3 vol binding buffer [60 mM sodiumperchlorate, 1 mM EDTA, 50 mM Tris-HC1 pH 7.8] was added. Thesample was heated at 37—55°C for several minutes to dissolve theagarose, after which 10 p1 of Prep-A-Gene binding matrix wasadded. The sample was mixed by vortexing, then shaken for 10 mmat room temperature. The sample was centrifuged (30 sec; 14,000rpm), the supernatant removed and the pellet rinsed 3 times with1 ml of a 1:1 mixture of 95% ethanol:wash buffer [400 mM NaCl, 2mM EDTA, 20 mM Tris-HC1 pH 7.4], followed by suspension with TE49buffer. This was mixed by vortexing, centrifuged as describedabove, and the supernatant TE transferred to a fresh tube. Onetl of the TE fraction was diluted to 100 jl with fresh TE bufferand the DNA quantified spectrophotometrically at 260 nm. Thegel-purified DNA was used for 115A.0 probe. Other gel purifiedprobes (XI5H and lBS rRNA) were obtained from Dr. Craig Newton.3.11. GENERATION OF DELETION CONSTRUCTS FOR SEQUENCINGTwo hundred ml of YT + amp broth was inoculated with bacteriacontaining the 115A.0 insert. Following overnight incubation at37°C, nucleic acids were isolated using the alkaline lysisprocedure. The resulting pellet was resuspended in 2.4 ml of TEbuffer and the plasmid DNA isolated using cesium chlorideethidium bromide (CsC1-EtBr) gradient centrifugation. Briefly,CsC1 (4.2 g) and 400 l of 10 mg/ml EtBr were added to the 2.4 mlplasmid solution, and centrifuged (5 mm; 6,000 rpm) at roomtemperature in a J2—2l Centrifuge (Beckman Instruments Inc.,Mississauga, ON). A Ti70..l centrifuge tube was partially filledwith 8 ml of light CsCl solution (63 g/lOO ml), and the plasmidsolution placed at the bottom of this tube. The tube wasbalanced, filled with light CsC1 solution, sealed and centrifugedovernight (18 hr; 40,000 rpm) in an L8—M Ultracentrifuge.Following this, the lower band was removed from the tube using a21—gauge needle and extracted 4 times with an equal volume ofwater—saturated isobutanol. The lower, aqueous phase wastransferred to a clean tube, 3 vol of deionized water and 2 volof 95% ethanol were added, and the DNA allowed to precipitate for30 mm at -800C. The sample was centrifuged (15 mm; 14,00050rpm), the pellet washed with ethanol, dried under vacuum andresuspended with TE buffer.The resulting DNA preparation was used for deletion constructformation. In order to sequence both strands, the cDNA insertwas required in both orientations. The 115A.0 insert was excisedby Eco RI-Sal I double restriction digestion and the resultingfragment gel purified and blunt-ended. To blunt end thefragment, 500 ng of 115A.0 DNA in 3 of TE buffer was added toa solution containing 6 ul of reaction buffer [1.5 inN DTT, 30 mMmagnesium acetate, 200 InN potassium acetate, 100 inN Tris—acetatepH 7.9], 2 j.tl dNTP mix [125 .LM each of dATP, dCTP, dGTP, dTTP]and 1 unit of Klenow DNA polymerase, and incubated for 15 mm at37°C. This was extracted with phenol:chloroform, the aqueousphase was mixed with 0.5 vol 7.5 M ainmonium acetate andprecipitated with 2 vol of 95% ethanol at -80°C. The pelletobtained following centrifugation (15 mm; 14,000 rpm) was washedwith ethanol, dried, resuspended with TE buffer and the DNAcontent was determined.Ten jg of pEMBL 8 vector (without insert) was digested withSma I, and the resulting linearized plasmid treated with calfintestinal alkaline phosphatase (CIP) to prevent plasmidrecircularization. Briefly, following restriction digestion, thesample was extracted with phenol:chloroform, the aqueous phasemixed with ammonium acetate and precipitated with 95% ethanol asdescribed previously. The resulting pellet was resuspended in 90j1 of 10 mM Tris-HC1 (pH 8.3). To this was added 10 jil of lOxCIP buffer [10 inN ZnC12, 10 mM MgC12, 100 inN Tris-HC1 pH 8.3] and511 ul (22 units) CIP. The sample was incubated at 37°C for 15mm, followed by the addition of 1 l of CIP and incubation at550C for 45 mm. CIP was inactivated by heating at 65°C for 1 hrin the presence of 5 mM EDTA (pH 8), after which the sample wasextracted with phenol:chloroform, the aqueous phase mixed withammonium acetate and precipitated with 95% ethanol as describedabove. The pellet was resuspended with TE buffer and the DNAcontent determined.The vector and insert were ligated using a 3:1 (300 ng:100 ng)ratio of insert:vector. The insert + vector were contained in atotal of 3 tl of TE buffer, to which was added 1 l of lOx ligasebuffer [500 mM Tris—HC1 pH 7.6, 100 mM MgC12, 10 mM ATP], 5 l of10 mM DTT and 1 l of T4 DNA ligase (5 units). Ligation wasallowed to proceed overnight at room temperature, then the 10 illigation mixture was added to 100 l of competent SURE cells(Stratagene Cloning Systems, La Jolla, CA), mixed gently,incubated for 30 mm on ice, heated at 42°C for 2 minutes andthen placed on ice for 2 mm. Two hundred ,l of SOC broth [20g/L bacto-tryptone, 5 g/L bacto-yeast extract, 0.5 g/L NaC1, 3.6g/L glucose] was added, the sample warmed to 37°C for 5 mm andshaken at 37°C for 1 hr. Cells were plated onto YT + amp mediumand grown overnight at 37°C. Ten individual colonies wereremoved, the plasmid DNA isolated by alkaline lysis, the DNAdigested with Barn HI, and separated on an agarose gel. Sampleswere compared to Barn HI-digested plasmid containing II5A.0 in theforward orientation. A colony containing altered fragment sizes52expected from the reverse orientation was identified anddesignated as RII5A.0.Plasmids containing 115A.0 and RII5A.0 were used forunidirectional deletion construct formation following the PromegaErase-A-Base system (Promega Corp., Madison, WI). Plasmid DNA(5-10 g) was digested with Sal I overnight, and a small amountwas analyzed by agarose gel electrophoresis to assay for completedigestion. After completion, the plasmid DNA was digested withPst I for 6 hr, and extracted with an equal volume ofphenol:chloroform (1:1). The aqueous phase was mixed withammonium acetate and precipitated with 95% ethanol as describedpreviously. The pellet was resuspended in 30 jl of ix Exo IIIbuffer [lOx Exo III buffer = 660 mM Tris-HC1 pH 8, 6.6 mM MgC12],heated at 35°C for 10 mm, and then 250 units of Exo III added.Digestion was carried out at 35°C. After a 25 sec lag period,2.5 ,1 aliquots were collected every 30 sec for up to 5 mm.These were placed into tubes containing 7.5 l of Si mix [172 jldeionized water, 27 1 7.4x Si buffer (0.3 M potassium acetate pH4.6, 2.5 M NaC1, 10 mM ZnSO4, 50% glycerol), 60 units Sinuclease) and incubated at room temperature for 30 mm, then 1 jlSi stop buffer [0.3 M Tris base, 0.05 M EDTA] was added and thesamples heated at 70°C for 10 mm. The extent of digestion wasdetermined by analysis of 2 1 of each sample by agarose gelelectrophoresis. Samples were heated to 37°C and 1 l of Klenowmix [30 l ix Klenow buffer (20 mM Tris-HC1 pH 8, 100 mM MgC12),5 units Kienow DNA polymerase] was added, the samples incubatedfor 3 mm, followed by the addition of 1 j1 dNTP mix [125 ,M each53of dATP, dCTP, dGTP, dTTP] and incubation for 5 mm. Sampleswere moved to room temperature, 40 ,l of ligase mix [890 jldeionized water, 100 ,l lOx ligase buffer (500 mM Tris-HC1 pH7.6, 100 mM MgC12, 10 mM ATP), 10 .tl DTT, 5 units T4 DNA ligase]added, and ligation allowed to proceed overnight. Ligation mix(10 l) was then used to transform 100 l of competent cells.These were plated onto YT + amp medium and grown overnight. Tenindividual colonies from each deletion time point were used forplasmid DNA isolation by alkaline lysis. The resulting DNA wasdouble digested with Eco RI-Hind III to excise the insert, andanalysed by agarose gel electrophoresis. Colonies containingdeletions were identified in this manner. The deletions obtainedspanned the entire length of the 115A.0 and RII5A.0 inserts.3.12. DNA SEQUENCINGDNA sequences were obtained using the dideoxy method. Foreach colony representing a deletion time point to be sequenced,plasmid DNA was isolated from a 2-4 ml YT + amp broth overnightculture using alkaline lysis. The DNA was resuspended in 100 jllof TE buffer and then treated with 1 l RNase (10 mg/nil) for 1hr at 37°C to digest contaminating RNA. Following this, 60 l of2.5 M NaCl/20% PEG 8000 was added, the contents mixed, left atroom temperature for 15 mm, centrifuged (10 mm; 14,000 rpm) at4°C and the supernatant (containing the RNA) removed. The pelletwas resuspended in 100 ul TE buffer, extracted with an equalvolume of phenol:chloroform, the aqueous phase removed and mixedwith ammonium acetate and precipitated with 95* ethanol as54described previously. The pellet was resuspended in TE bufferand the DNA quantified.A portion of the template DNA isolated above (2-4 g) in 16 ,.tlTE buffer was added to 4 1 of 1 M NaOH, incubated at roomtemperature for 10 mm, and precipitated with 0.5 vol of 7.5 Mammonium acetate and 2 vol of 95% ethanol at -80°C. The pelletobtained following centrifugation was washed with ethanol, driedand resuspended in a solution of 5 jl deionized water, 1 jl lOxsequencing sample buffer [500 mM Tris-HC1 pH 7.5, 100 mM NaC1,100 mM MgC12] and 1 1 of 2 ng/tl forward primer(5’ GTAAAACGACGGCCAGT3). This was mixed by vortexing,centrifuged briefly to collect the sample and incubated at 37-45°C for 15 mm. The sample was placed on ice and 1 1 of 15 MdATP, 1 il 100 mM DTT, 1.5 l o-32P-dATP (NEN-Du Pont Corp.,Mississauga, ON; 3000 Ci/mmole) and 4.5 l of a 0.5 unit solutionof Klenow DNA polymerase added. This solution was thetemplate/primer/label mix.In a petri dish on ice, 1.5 l of each dideoxynucleotide/deoxynucleotide (ddN/dN) mix was spotted in the dish[(G mix: 18 l 1 mM ddG, 16 l 0.1 mM dGTP, 32 1 each of 1 mMdTTP and 1 mM dCTP, 102 1 deionized water), (A mix: 23 jl 1 mMddA, 22 l each of 1 mM dGTP, 1 m dTTP and 1 mM dCTP, 111 ldeionized water), (T mix: 12.5 ,l 10 mM ddT, 32 jl 1 mM dGTP, 261 0.1 mM dTTP, 32 1 1 mM dCTP, 97 ,il deionized water), (C mix:75 ,1 1 mM ddC, 32 1 each of 1 mM dGTP and 1 mM dTTP, 41 jll 0.1mM dCTP, 20 jl deionized water). The template/primer/label mix(3.5 l) was added to each of the ddN/dN spots, and the petri55dish placed onto a water bath at 37-42°C for 5 mm. Followingthis, the petri dish was placed back on ice, 1 l of chasesolution [1 mM each of dGTP, dATP, dTTP, dCTP] added to each spotand, the dish placed onto the water bath for a further 5 mm.Reactions were stopped by placeing the petri dish on ice andadding 5 jl formamide dye mix [98% (v/v) deionized formamide, 10inN EDTA pH 8, 0.025% (w/v) xylene cyanol FF, 0.025% (w/v)bromophenol blue] to each spot. The petri dish was placed on aboiling water bath and the samples heated for 2 mm. Samples (1-2 l) were then loaded onto sequencing gels.Sequencing gels were made using 37.5 g urea, 7.5 ml lOxmodified TBE [162 g/L Tris base, 27 g/L boric acid, 9.4 g/LNa2EDTA.2H0], 11.25 ml 40% (w/v) acrylamide and 32 ml deionizedwater, to give a final urea concentration of 8 M and acrylamideconcentration of 6%. The gel mix was filtered, degassed for 30iuin and polymerized with 300 jl 10% ammonium persulfate and 15 jlTEMED. Gels were run at 2000 V using lx TBE as theelectrophoresis buffer. Following electrophoresis, gels weredried at 80°C for 1 hr and allowed to expose at -80°C using KodakX-Omat RP film. Sequences for both strands were obtained and thecoding strand sequence compared to sequences in the EMBL databank using PC/GENE (IntelliGenetics Inc., Mountain View, CA).3.13. GENOMIC DNA EXTRACTION, ELECTROPHORESIS AND BLOTTINGSpruce needles (5 g) were ground in liquid nitrogen, suspendedin 90 ml of extraction buffer [50 ml/L 1 M Tris-HC1 pH 8, 10 ml/L0.5 M EDTA, 63.78 g/L sorbitol, 100 g/L PEG 8000, 5 g/Lspermidine, 5 g/L spermine, 1 ml/L 2-mercaptoethanol] and56filtered through 12 layers of cheesecloth. The solution wasfiltered through 1 layer of Miracioth and 1 layer of 53 m nylonmesh, and centrifuged (20 mm; 10,000 rpm) at 4°C in a J2-2lCentrifuge. The supernatant was removed, the pellet resuspendedin 3.6 ml CsC1—lauryl sarcosine solution [17.5 g CsC1 in 10 mlsarcosine solution (50 inN Tris-HC1 pH 7.8, 10 mM EDTA, 0.5% (w/v)sodium lauryl sarcosine)], and heated for 10 mm at 60°C.Ethidium bromide (400 l of a 10 mg/mi solution) was added andthe sample centrifuged (10 mm; 10,000 rpm) at room temperature.The clear solution was removed and placed at the bottom of aTi70.1 centrifuge tube containing 8 ml of light CsC1 solution,and the sample centrifuged as described previously for CsC1-EtBrgradient centrifugation. The DNA band was removed, extractedwith isobutanol, precipitated as described previously andresuspended in TE buffer.Five of DNA were used for each restriction digest, and,following digestion, the DNA was precipitated with ammoniumacetate and ethanol as described previously. The resultingpellet was resuspended in 20 .tl TE buffer containing 5 l Ficoildye mix, and the DNA separated on a 1% agarose gel. The gel wasthen rinsed 3 times (10 mm each) with 0.5 M NaOH + 1.5 M NaC1,followed by 3 rinses(10 mm each) with 1 M ammonium acetate, andthen blotted overnight onto Hybond-N nylon membrane (AmershamCorp., Arlington Heights, IL) using 1 N ammonium acetate. Theblot was air dried and the DNA cross-linked to the membrane by UVexposure for 10 mm. Gene copy reconstructions to approximatethe number of vicilin gene copies were constructed using 115A.057insert and assuming a haploid DNA content of 8.5 pg, which hasbeen described for white spruce (Dhillon 1987).3.14. RNA EXTRACTION, ELECTROPHORESIS AND BLOTTINGSolutions for RNA work were made using diethylpyrocarbonate—treated [0.1% (v/v)] deionized water. Total RNA was extractedusing the protocol of Verwoerd et al. (1989). Tissues wereisolated, frozen immediately in liquid nitrogen and stored at—80°C until needed. Frozen tissues were ground in liquidnitrogen to a fine powder in Eppendorf tubes using a glass rod(precooled in liquid nitrogen) that fitted precisely into thetubes. After grinding, 500 l of hot (80°C) extraction buffer[phenol: 0.1 M LiC1, 100 inN Tris-HC1 pH 8, 10 mM EDTA, 1% (w/v)SDS (1:1)] was added. The samples were mixed by vortexing for 30sec, then chloroform (250 l) was added, followed by another 30sec vortex. After centrifugation (5 mm; 14,000 rpm), the waterphase was removed, an equal volume of extraction buffer (minusphenol) was added to each tube and the sample back—extracted.After centrifugation, the two water phases from each sampleextraction were pooled and mixed with 1 vol of 4 M Lid. RNA5were allowed to precipitate overnight at 4°C, then collected bycentrifugation (15 mm; 14,000 rpm). The pellets were dissolvedin 250 .Ll water, 0.1 vol of 3 N sodium acetate pH 5.2 was addedand the RNAs precipitated with 2 vol of 95% ethanol at -80°C for30 mm. After centrifugation, the pellets were washed withethanol, dried and resuspended in TE buffer containing 0.1% (w/v)SDS.58The purity and integrity of samples was checked by agarose gelelectrophoresis. Two ,l of RITA sample was added to 8 l of TEbuffer containing 0.1% (w/v) SDS and 1 l Ficoll dye mix. Thesamples were heated at 65°C for 5 mm and then applied to a gelconsisting of 1% (w/v) agarose in lx TAE containing 0.1% (w/v)SDS. Electrophoresis buffer consisted of lx TAE with 0.1% (w/v)SDS. For RNA quantification, 1 l of sample was diluted to 100il with TE buffer and quantified spectrophotometrically at 260nm.For RNA electrophoresis and blotting, the protocol of Fourneyet al. (1988) was used. RITA samples were adjusted to 5 jl withwater, 25 l of electrophoresis sample buffer [750 ,l deionizedformamide, 150 l lOx MOPS/EDTA buffer (0.2 M 3-(N—morpholino)propanesulfonic acid, 50 mM sodium acetate, 10 mM EDTA, pH 7 andautoclaved), 240 jl 37% formaldehyde, 100 1 water, 100 lglycerol, 80 ,.Ll 10% (w/v) bromophenol blue] added, and heated at65°C for 15 mm. Samples were fractionated on denaturingformaldehyde gels [225 ml water, 2.5 g agarose, 25 ml lOxMOPS/EDTA buffer, 4.1 ml 37% formaldehyde, 0.46 g iodoacetamide,5 l of 10 mg/nil EtBr] run overnight at 25 V, using lx MOPS/EDTAas the electrophoresis buffer.Following electrophoresis, gels were rinsed 20 mm in 0.05 MNaOH in lx SSC [20x SSC = 3 N NaC1, 0.3 M Na3 citrate], followedby 2 rinses (20 mm each) in lOx SSC. They were then blottedovernight onto Hybond-N nylon membrane using lOx SSC. Blots wereallowed to air dry, and the RNA crosslinked to the membrane byexposure to UV light for 5 mm.593.15. cDNA PROBE PRODUCTION ND HYBRIDIZATION TO BLOTSAll probes were generated using the random oligonucleotideprimer method. In a 1.5 p1 Eppendorf tube, 1 p1 of template cDNA(25—50 ng) was mixed with 9.5 p1 of deionized water, heated at98°C for 10 mm, then snap cooled on ice. To this was added 2 p1lOx buffer [900 mM HEPES pH 6.6, 100 mM MgC12], 1 p1 1 mg/mi BSA,1 p1 100 mM DTT, 2 p1 dNTP mix [10 mM each of dGTP and dTTP], 1p1 random hexamer primer, 2 p1 each of oc.-(32P-dATP) anddCTP) (NEN-Du Pont Corp., Mississauga, ON; each with 3000Ci/mmol) and 1 p1 Klenow DNA polymerase (5 units). This wasincubated overnight at room temperature. The probe wasprecipitated by addition of 1 p1 of 0.5 N EDTA, 80 p1 deionizedwater, 3 p1 E. coil carrier tRNA (2 mg/mi), 50 p1 7.5 M ammoniumacetate and 375 p1 95% ethanol to the tube and incubation at -80°C for 30 mm, followed by centrifugation (15 mm; 14,000 rpm)at 4°C. The pellet was washed with ethanol, vacuum dried andresuspended in 100 p1 TE buffer. One p1 of labelled probe wasplaced in a scintillation vial and the radioactivity determined.Probes were heated at 98°C for 5 mm and then snap cooled on iceimmediately prior to their use.DNA blots were pre-hybridized overnight at 68°C inapproximately 30 ml of pre-hyb/hyb solution [5x SSPE (20x SSPE =3.6 M NaC1, 0.2 M Na2HPO4.7H0, 0.02 N EDTA), 100—200 pg/mi calfthymus DNA, 1% (w/v) SDS, 5x Denhardt’s solution (bOx Denhardt’s= 2% (w/v) BSA, 2% (w/v) polyvinylpyrrolidone, 2% (w/v) Ficoll),0.05% (w/v) sodium pyrophosphate, 10% (w/v) dextran sulfate].Following pre-hybridization, blots were hybridized at 68°C for 260days with labelled probe (1-2 x 106 cpm/ml of solution) inapproximately 30 ml of hyb solution. Blots were washed 3 times(30 mm each) at 68°C with 2x SSC + 1% (w/v) SDS and allowed toexpose at -8 0°C with an intensifying screen using Kodak X-Omat ARfilm.RNA blots were pre—hybridized overnight at 680C in pre-hyb/hybsolution [0.5 M sodium phosphate pH 7.2, 7% (w/v) SDS, 1 itiNEDTA]. Following this, blots were hybridized for 22-24 hr at68°C in approximately 30 ml of hyb solution containing 1-2 x 106cpm of labelled probe/mi of solution. After hybridization, blotswere washed twice (20 mm each) with a solution of lx SSC + 0.1%(w/v) SDS at room temperature, followed by a 20 mm wash at 68°C.Blots were then allowed to expose at -800C with an intensifyingscreen using Kodak X-Omat AR film. For reprobing of blots, eachblot was stripped with several 250 ml washes of boiling 0.1%(w/v) SDS and the membrane was checked by overnight exposure toensure complete removal of probes. Autoradiographs of RNA blotswere scanned at 595 nm in a DU—64 Spectrophotometer.614. RESULTS4.1. IDENTIFICATION AND CHARACTERIZATION OF ZYGOTIC EMBRYOSTORAGE PROTEINSThe storage proteins of interior spruce were identified bySDS-PAGE analysis of total proteins from mature seed embryos(Fig. 1). Prominent proteins with apparent molecular weights of41, 35, 33, 24, 22, 17 and 16 kD were identified. Similarproteins, with the exception of the 17 and 16 kD proteins, weredetected in megagametophyte tissue extracts (data not shown). Todetermine if any of the proteins were storage proteins, embryoprotein profiles were examined during germination (Fig. 1A). Byday 3 after sowing, at which time radicle emergence had occurred,the 41, 35, 33, 24 and 22 kD proteins were almost undetectable,indicating rapid degradation during germination, characteristicof storage protein mobilization. However, the 17 and 16 kDproteins were still detected on day 9 of germination (Fig. 1A),suggesting that they were not storage proteins. To test theassumption that the 5 proteins missing by day 3 were storageproteins, protein bodies were isolated from mature embryonictissues, solubilized and analyzed electrophoretically (Fig. 1B).The 41, 35, 33, 24 and 22 kD bands were prominent in thisprofile. Minor bands of 30 and 27.5 kD were also observed (Fig.1B). These proteins were also present in megagametophyte tissue(data not shown).To determine if individual proteins were joined by disulfidelinkages, extracts of isolated protein bodies made under non—reducing conditions were examined by SDS—PAGE under non—reducing,62FIGURE 1. Coomassie-stained SDS-PAGE of embryo proteins.A. Total embryo protein changes during germination. Lane 1,mature embryos; Lane 2, 3 days after sowing; Lane 3, 6 days aftersowing; Lane 4, 9 days after sowing; Lane 5, 12 days aftersowing; MW, molecular weight standards. Fifteen jg protein wasapplied to Lane 1 and 10 ,g protein to all others. The 41, 35,33, 24 and 22 kD proteins (solid arrows), as well as the 17 and16 kD proteins (open arrows) are indicated.B. Proteins from mature embryos (Lane 1) and isolated proteinbodies (Lane 2). NW, molecular weight standards. Fifteen ,gprotein was applied to each lane. Arrows denote the 41, 35, 33,30, 27.5, 24 and 22 kD proteins.kD97.466.242.731.0b4121.514.4B45MWkDI—97.4123,,A64and then reducing conditions (Fig. 2). The patterns observedunder reducing conditions differed from those observed under non—reducing conditions. The 55-57 kD protein doublet present undernon—reducing conditions (Fig. 2B) was absent afterelectrophoresis under reducing conditions and replaced by fourbands of 35, 33, 24 and 22 kD (Figs. 2A and 2C), which werepreviously linked by disulfide bonds. It was difficult todetermine the specific association of the proteins in the 55-57kD doublet (Fig. 2C). The minor 34 kD protein observed undernon—reducing conditions was also resolved into two componentproteins of 14 and 22 kD when analysed under reducing conditions.The 14 kD protein was often not readily observed in the proteinbody fraction and was less prominent than the minor 30 and 27.5kD proteins. Apparently, the disulfide linked 22 and 14 kDcomplex is a very minor component of the protein body fraction.The storage proteins identified previously were classifiedaccording to their solubility characteristics. Isolated proteinbodies that were extracted with phosphate buffer or water yieldedonly the major 41 kD and minor 30 and 27.5 kD bands (Figs. 3A and3B). The remaining proteins were only soluble in SDS-containingbuffer (Fig. 3A) or buffer containing 1 M NaCl (Fig. 3B). Therewere thus two major solubility classes of proteins in isolatedprotein bodies. Microscopic analysis of protein bodies from latecotyledonary zygotic embryos revealed two distinct protein—staining regions (Fig. 4), indicating the heterogenous nature ofthe proteins within these organelles.65FIGURE 2. Coomassie-stained SDS-PAGE of zygotic embryo proteinbody extracts under reduced (A), non—reduced (B), and two—dimensional SDS—PAGE of non—reduced extract under non—reducingconditions followed by electrophoresis under reducing conditions(C). The 41, 35, 33, 30, 27.5, 24 and 22 kD proteins areindicated by open arrows (A). The 55-57 kD doublet present undernon-reduced conditions is indicated by solid arrows (B). Themolecular weights of non—reduced proteins and the correspondingproteins obtained by second dimension SDS-PAGE are indicated (C).Each sample contains 4 jtg protein.m 0.C C)e—410..4Non—Reduced27.530354155-57I’ll——35—331-30—27.5F,-24-220’C67FIGURE 3. Coomassie-stained SDS-PAGE of protein body samplesextracted under different conditions.A. Buffer-soluble (Lane 1) and SDS-soluble (Lane 2) proteinsextracted from isolated protein bodies. A 10 l sample wasapplied to each lane. MW, molecular weight standards.B. Water-soluble (Lane 1), buffer—soluble (Lane 2), low saltsoluble (Lane 3) and high salt-soluble (Lane 4) proteinsextracted from isolated protein bodies. A 10 jl sample wasapplied to each lane. MW, molecular weight standards.31.O—34I.—Lr.iii-i-42.7121kDMW97.466.2,.‘J42.7MWkO97.466.22.jzJ21.5hfluia’31.0,-21.5A-14.4B69FIGURE 4. Light micrograph of a longitudinal cotyledon sectionfrom a late maturation stage zygotic embryo, stained by theperiodic acid-Schiff’s technique and counter-stained with anilineblue black. Protein bodies (PB) containing light and darkstaining zones are visible. N, nucleus. x 1100.L*4.a/.40a‘aS10SISzaSrAj71Protein body samples were subjected to two—dimensionalelectrophoresis using both narrow pH (5-8 ampholytes) and broadpH (3-10 ampholytes) ranges (Fig. 5). The 41 kD proteincontained the largest number of isoelectric variants, withapproximately 10 or more proteins found across the range of thenarrow pH gradient gel. It was often difficult to resolvedistinct spots for the 35, 33, 24 and 22 kD proteins and althoughdifferent gradients and running conditions were tried, theseproteins were usually seen as smears. The 35 and 33 kD proteinvariants were located primarily at the basic end of the narrowgradient gel. However, the 24 and 22 kD proteins which were alsobasic variants were more readily observed in the broader pHgradient gel (Fig. SB). The minor 30 and 27.5 kD proteins werelocated towards the acidic region of the gels and were morereadily visualised in the narrow pH gradient gel (Fig. 5A).4.2. STORAGE PROTEIN ACCUMULATION DURING ZYGOTIC EMBRYODEVELOPMENTTo follow the appearance of the major storage proteins overtime, embryos from Stage 2 to maturity were analyzed by one— andtwo—dimensional electrophoresis. Embryos collected during thesummer of 1988 were used for two—dimensional analysis (Fig. 6),and a more complete developmental sequence was obtained from theone-dimensional analysis of embryos collected during 1989 (Fig.7). Major accumulations of storage proteins, especially the 41kD protein, occurred in the Stage 4 embryo, when cotyledons werealready well developed (Figs. 6 and 7). The other storageproteins were detectable, albeit at low levels, at much earlier72FIGURE 5. Silver-stained two-dimensional electrophoretograms ofzygotic embryo protein body extract examined using pH 5-8ampholytes (A) or pH 3-10 ampholytes (B). The acidic (+) andbasic (-) ends of the gel are indicated. A total of 12 ugprotein was analyzed for each pH gradient.r)3+IEF— kD I;— 4— 1 — 427—31.0—21.5A—42.7—l74FIGURE 6. Silver-stained two-dimensional electrophoretograms of3 representative zygotic embryo stages collected during thesummer of 1988. Proteins were separated using pH 5—8 ampholytes,and the representative portions of the gels containing the majorstorage proteins (boxed regions) are shown. Acidic (+) and basic(—) ends of the gels are indicated.Stage4-1rg+IEFSDS—- ,PAGEStage2ii-1MatureS76FIGURE 7. Coomassie-stained SDS-PAGE of developmental stagechanges in total protein during EK1O zygotic embryogenesis.Zygotic embryos collected during the summer of 1989, ranging indevelopment from Stage 2 to mature seed embryos (Stage 4, August28 collection) are shown. Molecular weight standard locationsare shown on the left. Twelve pg total protein was loaded ineach lane.StageStageStageStageStage423A383—4JNkO2027411182518152328.4——-.————a—.-———.——,————..—.—•*•••.—..JLAU97.4—66.2—.42.7—31.0—21.5—1...4b-.—-,.__—I--I—..—.-.—-.-*.--.qb-4I---..—.I*1__iPrr-’ZE.W*-78stages of development, suggesting the differential regulation ofstorage proteins. Although the 35, 33, 24 and 22 kD proteins didnot show major accumulations until cotyledonary development, theywere detected at low levels in stage 2 embryos (Fig. 6). Asummary of the developmental appearance of the major storageproteins in 1988 collection embryos is presented in Table 2.4.3. IDENTIFICATION AND CHARACTERIZATION OF SOMATIC EMBRYOSTORAGE PROTEINS AND COMPARISON WITH ZYGOTIC EMBRYO STORAGEPROTEINSSomatic embryos were matured in the presence of various levelsof ABA (0-60 jIM) and an ABA-dependent developmental profile wasobtained (Fig. 8). The level of ABA affected the number and typeof structures that developed. Low ABA levels favouredproduction of shooty structures or precocious germinants withextensive chlorophyll development. Higher ABA concentrationsstimulated the production of late cotyledonary somatic embryosthat did not germinate precociously during the experimentalperiod. The optimal level of ABA for line W29 was between 40-60M. Based on morphological characteristics described by Buchholzand Stiemert (1945), somatic embryos differentiated on 40 M ABAproceeded through embryogenesis in a manner similar to that ofconifer zygotic embryos (Fig. 9). The late cotyledonary somaticembryos were larger in girth but similar in length and overallappearance to zygotic embryos after 9 weeks of maturation on ABA(Fig. 9). These embryos were used for the identification andcharacterization of somatic embryo storage proteins.79TABLE 2. The presence (+) or absence (—) of various storageproteins during zygotic embryo development, 1988 collection.Storage Embryo developmental stageprotein( kD)2 3 3—4 4—1 4—2 4—3 Mature(JL13) (JL27) (AU24)41 - — - - + + +35 + + + + + + +33 + + + + + + +24 + + + + + + +22 + + + + + + +80FIGURE 8. Abscisic acid-dependent developmental profile ofgenotype W29, showing the types and numbers of structuresobtained with maturation using different ABA concentrations. ME,late cotyledonary somatic embryo; PE, precociously germinatingsomatic embryo; SH, shooty structure. The bar graphs representquantitative data for the cultures depicted below.500 -ME450400PESH(J] 350 -LJ100 -500-0 20 40 60ABA (tM)82FIGURE 9. Examples of somatic embryo developmental stages. PE,proembryo; G, globular embryo; R, round head torpedo embryo; F,flat head torpedo embryo; EC, early cotyledonary embryo; LC, latecotyledonary embryo; ZE, zygotic embryo.1mm84SDS-PAGE analysis of total embryo proteins revealed thatproteins were similar in zygotic and somatic embryos (Fig. 10),as well as in the isolated protein bodies of both embryo types.Total protein gels suggested that there were differences betweenstorage protein levels of the two embryo types (Fig. 10).However, the relative levels of 35, 33, 24 and 22 kD proteins inisolated protein bodies were similar from both sources, althoughthe level of 41 kD protein was greater in somatic embryo proteinbodies (Fig. 10).An analysis of the somatic embryo protein body fraction underreducing and non—reducing conditions revealed distinctdifferences between the two treatments (Fig. 11), similar tothose observed for zygotic embryo storage proteins (Fig. 2).The somatic embryo storage proteins were characterized furtherusing solubility criteria. Isolated somatic embryo protein bodypreparations contained phosphate buffer-soluble 41, 30 and 27.5kD proteins, with the remaining proteins requiring SDS forsolubilization (Fig. 12). These results were similar to thoseobtained for zygotic embryo storage proteins (Fig. 3).Somatic embryo protein body samples which were analysed bytwo—dimensional electrophoresis using narrow and broad pHgradients (Fig. 13) were similar to those observed for zygoticembryo storage proteins.A comparison of the steady state storage protein levelsbetween zygotic and W29 somatic embryos by scanning densitometryof total protein profiles revealed some differences between theproportion of major storage proteins in the two embryo types85FIGURE 10. Coomassie-stained SDS-PAGE of zygotic embryo (Lane1), zygotic protein body (Lane 2), 9-week ABA somatic embryo(Lane 3) and somatic embryo protein body (Lane 4) extracts.Arrows indicate the 41, 35, 33, 30, 27.5, 24 and 22 kD proteins.Lanes 1 and 3 contain 15 protein and lanes 2 and 4 contain 5protein. MW, molecular weight standards.kD MW 1 2 3 4-- —.-97.4 I66.2 442.7______ ___________I__I___ 1l4 j —— rIi21.5 LZem14.4 — - -1_._____— —87FIGURE 11. Coomassie-stained SDS-PAGE of somatic embryo proteinbody extracts under reduced (A), non—reduced (B), and two—dimensional SDS—PAGE of non—reduced extract under non—reducingconditions followed by electrophoresis under reducing conditions(C). The 41, 35, 33, 30, 27.5, 24 and 22 kD proteins areindicated by open arrows (A). The 55-57 kD doublet present undernon-reduced conditions is indicated by the solid arrow (B). Themolecular weights of non—reduced proteins and the correspondingproteins obtained by second dimension SDS-PAGE are indicated (C).Each sample contains 5 jg protein.IwLI.4_z00o.C,UI •bJ IlIIg)U1UI89FIGURE 12. Coomassie-stained SDS-PAGE of somatic embryo proteinbody samples extracted under different conditions. Buffer—soluble (Lane 1) and SDS-soluble (Lane 2) proteins extracted fromisolated protein body samples are shown. A 10 ,l sample wasapplied to each lane. NW, molecular weight standards.cjo21.5144I1 2kD MW97.466.242.731.091FIGURE 13. Silver-stained two-dimensional electrophoretogram ofsomatic embryo protein body extract examined using pH 5-8ampholytes (A) or pH 3-10 ampholytes (B). The acidic (+) andbasic (-) ends of the gel are indicated. A total of 20 jgprotein was analyzed for each pH gradient. MW, molecular weightstandards.+cia- .rwI:H* II jI 111AIEF— MW kD 1SDSIPAGE42.7______411’!1 I. 21.5B93(Table 3). Somatic embryos of genotype W29, after 9 weeks onABA, contained more of the major storage proteins as a percentageof total protein than did mature zygotic embryos (Table 3).To determine the extent to which storage proteins accumulatedin other somatic embryo genotypes, the protein profiles of latecotyledonary (7 weeks on ABA) somatic embryos of uniform lengthwere compared with those from genetically related seed (PG118 orEK1O) (Fig. 14). Somatic embryos of several genotypes derivedfrom EK1O seed showed variable levels of storage proteins, withW29 embryos containing the highest levels (Fig. 14A). Embryosfrom two different genotypes derived from PG11B seed had storageprotein levels similar to each other, which were similar to orslightly higher than the levels in PG118 zygotic embryos (Fig.l4B). However, storage proteins accumulated to a significantdegree in late cotyledonary embryos of each genotype.4.4. STORAGE PROTEIN ACCUMULATION IN SOMATIC EMBRYOS ANDCOMPARISON WITH ZYGOTIC EMBRYOSA comparison of early cotyledonary (Stage 3—2) and latecotyledonary (Stage 4-9) W29 somatic embryos by two-dimensionalelectrophoresis revealed a differential appearance of the majorstorage proteins during development. In late cotyledonarysomatic embryos, the 41 kD protein was abundant (Fig. 15).However, during earlier embryogenic stages, the 35, 33, 24 and 22kD proteins were detected at low levels, but the 41 kD proteinwas not. In addition, the major accumulation of all storageproteins occurred during the later stages of embryo maturation(Fig. 16). The differential appearance of the 41 kD versus the94TABLE 3. Major storage protein distribution in EK1O zygotic andW29 somatic embryos as determined by densitometry. Mean ± SE.Storage Zygotic embryo Somatic embryoprotein( kD)% of total protein % of total protein41 24±3 32±135 + 33 5 ± 3 13 ± 124+22 12±2 13±2Totalprotein 41 ± 4 58 ± 195FIGURE 14. Coomassie—stained SDS-PAGE of total proteins for twozygotic embryo genotypes and different somatic embryo genotypesderived from them. Protein profiles of the different somaticembryo genotypes derived from (A) EK1O (W29, W74, W76, W77) and(B)PG118 (W46, W70) seed are indicated. Each lane contains 7.5g protein. MW, molecular weight standards.21.5-14.4BW70W46EKkDMWW76W77W74W291097.466.2—42.7_______—.-31.021.5PG 118MWkD 97.466.242.731.0--:--14.4I0•A97FIGURE 15. Silver-stained two dimensional electrophoretograms oftotal proteins from stage 4-9 (A) and stage 3-2 (B) somaticembryos differentiated on 40 ,.M ABA. Proteins were separatedusing pH 3-10 ampholytes, and the representative portions of thegels containing the major storage proteins (boxed regions) areshown. Acidic (+) and basic (-) regions of the gels areindicated.C0I1-4+99FIGURE 16. Coomassie-stained SDS-PAGE of developmental stagechanges in total somatic embryo protein of genotype W29 maturedon 40 ,M ABA. Storage proteins are indicated by arrows. G,globular embryos; R, round head torpedo embryos; F, flat headtorpedo embryos; EC, early cotyledonary embryos; LC, latecotyledonary embryos. The range from early cotyledonary to latecotyledonary includes the range of embryos from the onset ofStage 3 development to 9 weeks on ABA as described in Table 1.Each lane contains 15 jg protein. MW, molecular weight standards.I00G R FIEC +LCj MW--- ._.._ p1,4 e .197.466.242.7— — .31.021.5— 14.410135, 33, 24 and 22 kD proteins during the early stages of embryomaturation was similar in zygotic and somatic embryos.Polyclonal antibodies were made against the 41 and 24 + 22 lcDstorage proteins, and analysis by immunoblotting (Figs. 17 and18) showed that these proteins accumulated in both somatic andzygotic embryos, although there were differences between theaccumulation patterns of the two embryo types. Quantification ofimmunoblots by densitometry indicated that the 41 kD protein insomatic embryos showed an initial rapid accumulation whichcontinued during the 6—week period of cotyledon/embryo maturationin our differentiation protocol (Fig. 17). In zygotic embryos,there was also an initial, rapid accumulation of the 41 kDprotein over a 3—week period, but thereafter the protein levelsremained relatively constant or increased only slightly (Fig.17). The accumulation of the 24 + 22 kD proteins displayedsimilar differences between somatic and zygotic embryos. Therewas a prolonged period of accumulation in the somatic embryos,and a more rapid, shorter accumulation period in the zygoticembryos (Fig. 18). The relative abundance of the 24 + 22 kDproteins actually declined during the later stages of embryomaturation in the zygotic samples.4.5. IDENTIFICATION AND CHARACTERIZATION OF A CDNA ENCODING THE41. ]cD STORAGE PROTEINComparison of storage protein gene expression at the molecularlevel in zygotic and somatic embryos required the use of storageprotein cDNA probes. A cDNA library was constructed with theexpression vector pUC 18, using mRNA extracted from late102FIGURE 17. Relative quantification of 41 kD protein immunoblotsduring zygotic and somatic embryo development.UPPER PANEL: Immunoblot of zygotic embryo cotyledonary stagescollected during 1989. ND, not detected on scan.LOWER PANEL: Immunoblot of somatic embryo developmental stageson 40 jM ABA. The developmental stages from early to latecotyledonary are described in Table 1. ND, not detected on scan.o3006 ZE-4lkD0.050.040.03->0.020.010—— ———— a. —— a — — —0.5 I 2 3 4 5 6 7 8 9 10 10.7WEEKS OF COTY1EDON DEVELOPMENTEARLY LATECOTYLtDONARY COWLEDONARY0.060.050.040.030.020.010— —. — —— :1.4 1.7 2.1 2.4 2.9 3.3 4.1 4.6 5.6 6.6WEEKS OF COTYLEDON DEVELOPIIENTEARLY LATECOTYLEDONARY COrILEDONARYI I IHI I104FIGURE 18. Relative quantification of 24 + 22 kD proteinimmunoblots during zygotic and somatic embryo development.UPPER PANEL: Immunoblot of zygotic embryo cotyledonary stagescollected during 1989. ND, not detected on scan.LOWER PANEL: Immunoblot of somatic embryo developmental stageson 40 jM ABA. The developmental stages from early to latecotyledonary are described in Table 1. ND, not detected on scan.‘Os0.0240.0220.0200.0180.0160.0140.0120.0100.0080.0060.0040.0020.220.200.180.160.140.12a.0100.080.060.040.02SE-24+22kD1 1.4 1.7 2.1 2.4 2.9 3.3 4.1 4.6 5.6 6.6WEEKS OF COTYLEDON DEVELOPIIENTEARLY____________________LATECOTYLEDONARY COTYLEDONARY3 4 5 6 7 8WEEKS OF COTYLEDON DEVELOPMENTEARLY LATECOTYLEDONARY COTYLEDONARYND. rnrnFiFiITTi106cotyledonary somatic embryos. A portion of the library wasscreened using polyclonal antibodies against purified 41 kDstorage protein. A positive clone, designated 115A.0 wasidentifed and characterized further. This clone contained aninsert of approximately 1.7 kb, making it large enough to codefor the 41 kD protein class. Other clones have been identifiedby cross hybridization with 115A..0 and are of similar size (Dr.Craig Newton, personal communication). Sequencing of the 115A.0clone revealed a 1633 bp insert (Fig. 19), with a reading frameof 1406 bp, encoding a putative precursor protein of 466 aminoacids which ran from the ATG start codon at base 9 to the stopcodon at base 1406. This protein had a predicted molecular massof 51,835. The putative polypeptide was compared with knownprotein sequences in the EMBL data bank and showed a high degreeof similarity to several angiosperm vicilin—type storage proteinsequences (Fig. 20). It was concluded that the 115A.0 cDNA was aspruce storage protein cDNA.The predicted molecular mass of the putative storage proteinwas slightly higher than the apparent molecular mass of themature protein, as determined by SDS-PAGE. Amino-terminalsequencing of the protein suggested that the mature proteinstarted at the glycine residue which corresponded to amino acid48 (Dr. Craig Newton, personal communication), with thepreceeding amino acid region serving as a signal sequence whichwas co—translationally cleaved. Computer analysis to identifythe potential hydrophobic signal sequence indicated that thefirst 18 amino acid residues could serve as the signal sequence107FIGURE 19. Sequence analysis of the spruce 115A.0 cDNA clone.The nucleotide sequence of the spruce 115A.0 cDNA described inthe Results is shown. The deduced amino acid sequence is givenin single—letter code. The N—terminal amino acid sequencedisplaying almost complete duplication is underlined, and thepotential N—terminal amino acid residue of the mature protein isindicated by the solid triangle.108GCATCATCATGGTTTTCGCTTCTTTACTTATGATTCTTCTTGCAATCTCCTCCTCCTCGGCTGCCCTCACCGAG 74N V F A S L L MILL A IS S S S A ALT ECCACTAGCCAGCACGGCCAATCCAACCTCCTCCTCCTCGGCTGCCCTCACTGAGCCACTATCCAGCACGGCCAATP LAST A N PT S S SS A ALT E P L SS TANCCAGGAGTTTTTCCTGAATATCTCGGCCGAGGCCGAGGGAGACGAGAAGAAGAGCGAGAGGAGAATCCATACGTA 224PG V F P E Y L G R G R G R RE E ER E EN P Y V— ATTCCACAGTGACAGCTTCAGGACCAGAGCATCATCTGAAGCTGGTGAAATCAGAGCTCTGCCGAACTTTGGGGAGF H SD SF R T R A S SEA GE IRA L P N F GEGTCTCTGAACTTCTTGAAGGGATTAGAAAATTCAGAGTTACCTGCATTGAAATGAAACCCAATACAGTGATGCTC 374VS EL LEG IRK FR VT CI EM K P N TV MLCCTCACTATATTGATGCGACATGGATCTTATATGTTACTAGAGGAAGAGGCTACATAGCCTATGTGCACCAGAATPH Y IDA T WILY VT R G R G Y IA Y V H Q NGAGCTGGTTAMAGAAAGTTGGAGGAAGGAGATGTATTCGGTGTTCCAAGTGGTCATACATTTTATCTCGTTAAC 524EL V KR K LEE GD V F G VP S G H T F Y LV NAACGATGACCATAGCACCCTTCGCATTGCCAGTCTCCTGCGTCCCGTGTCTACGATCCCAGGAGAATATCAGCCCND D H ST L RI A S L L R P V ST I PG E Y Q PTTCTACGTTGCGGGAGGTCGGAATCCTCAGAGTGTTTACTCTGCCTTTAGCGATGATGTTCTCGAGGCTGCATTC 674F Y VA G G RN P Q S V Y S A F SD DV LEA A FAATACGAACGTACAGCAGCTTGAACGTATTTTCGGTGGACACAAAAGCGGAGTCATAATCCACGCAAATGAAGAANT N V Q Q L ER IF G G H KS G VII H A NE ECAGATTAGAGAAATGATGAGGAAACGGGGATTATCAGCAGGATCCATGTCTGCACCTGAGCACCCCAAGCCTTTC 824Q IRE M MR KR G L SAGS N SAP E H P K P FAACCTTCGGAACCAGAAGCCAGATTTCGAGAACGAAATGGCAGGTTTACTATTGCTGGTCCCAAAAATTATCCTN L RN Q K PD F EN ENGR F TI AG P K NY PTTTCTAGACGCGCTCGACGTTTCTGTTGGGCTTGCCGATTTGAATCCTGGATCCATGACAGCCCCATCTCTCAAC 974FL D AL DV S V GLAD L N PG SM TAPS L NTCGAAATCAACGTCAATCGGCATTGTTACGAATGGGGAAGGAAGGATTGAGATGGCATGCCCGCACCTTGGTCAAS K ST SI G IV TN GE G RI EM AC PH L G QCATAGCTGGTCTAGTCCGCGTGAGAGAGGCGACCAAGATATTACTTACCAGAGAGTCTGGGCAAAGCTGAGGACC 1124H SW S S PR ERG D Q DI T Y Q R V WA K L RTGGCAGCGTTTATATTGTTCCTGCTGGTCATCCAATCACGGAGATAGCTTCAACAAACAGCCGCCTGCAAATCTTGG S V Y IV PA G H PIT El A S TN SR L Q ILTGGTTTGATCTTAATACCCGCGGCAATGAGAGACAATTCCTGGCAGGAAAGAACAATGTGCTTAACACGTTGGAG 1274W F DL NT R G N ER Q FLAG K N N V L NT L EAGGGAGATCAGGCAGATATCCTTCAACGTACCACGTGGGGAAGAGATTGAAGAAGTGTTGCAGGCACAAAAGGATRE I R Q 1SF N VP R GEE I E E V L Q A Q K DCAAGTGATCCTCAGAGGCCCCCAACGACGAAGCCGGGACGAGGCGAGGAGCTCTTCTTAGATCCATGTCATCATT 1424Q VI L R G P Q R R SR DEARS S S *GCAGATCGCATTATGGACGACATGACAAGAGTTTCTCCACGTTCACTCTTAATATGTAGTTAAGAATAAGCTATCCATAAATGTGTTCGAAGATGAACTCTTTCTGTTTAAATGAATTATGTATGAGTCTAACAAAGCTATCGTTGGGCT 1574CCTCTTTCTACTTCAATGCAATGAAACGCAGGTCTTCTCTTAAAAAAAAAAAAAAAAAA109FIGURE 20. Amino acid sequence comparison of 115A.0 with otherangiosperm vicilin—type storage protein sequences. The figurecompares the amino acid sequence of spruce 115A.0 with those ofthe cotton vicilin (alpha-globulin A) precursor, the broad beanvicilin precursor (BRDBN) and the soybean 13-conglycinin alphachain precursor (SOYBN). Sequences were aligned to maximizeidentities. Positions in the alignment that are perfectlyconserved are boxed. Positions that are well conserved areindicated by asterisks.110-FA4IL- - IS-SSS-----AALTEP- 115A. 0I N----VRNKSVFVjLj4FsLFL FGLLCSAKDFPGR--—-RSEDDPQQRY COTTONCLSS BRDBN-- -MRARFPLL-VVFtJ.SVSVSFG 1AYWEKQNPSHNKCLRSCNSEKDSYR SOYBN* * * *LASTANPT S SSSAALTE 115A. 0- EDCRKRCQLETPGQTEQDKCED RSETQLKEEQQRDGEDPQR- COTTONBRDBNNQACHARCNLLKVEE-—EEECEEGQIPRPRPQHPERERQQHGEKEEDEGEQPRPF SOYBNPLSSTANPGVFPEYL 115A. 0RYQDCRQHCQQEE RRLRPHCEQSCREQYEKQQQQQPDKQFKECQQ COTTONBRDBNPFPRPRQPHQEEEHEQKEEHEWHRKEEKHGGKGSEEEQDEREHPRPHQPHQKEEE SOYBNGRGRGRREEEREE 115A, 0RCQWQEQRPERKQQCVKECREQYQEDPWKGERENKWREEEEEESD-EGEQQQR*I COTTONRSDQLjNI BRDBNKHEWQ-HKQEKHQ- -GKESEEEEEDQDEDEEQDXESQESEGSESQREPRRMKNIJ SOYBNEM 1l4EP II 5A. oCOTTONFLPjQ BRDBN• FI14SKJF TLFKNQ\1$ J* SOYBNYITWILYVT1RGYIAYVHQNELVKRKLEDVFG’HTFY II 5A. 0HcDAEKIYVVT4RGTVTFVTHENKESYNVVrVVVR*STVY COTTONQDA)FILVVLcAILTVLLPNDRNSFSLEIDTIK$jr BRDBNSOYBN* * ** *** * ** *** * * *I I 5A. 0L NREIIA 1N-QFQ4FPAGQENPQfLRIjFR COTTONBRDJ3NSOYBND\l4F14VQQLERIF —HKS GVIIUANEiRE I 15A. 0EIjLE?IVFl4ISEQLDELP -44RQSHRRQQGQGMFRKASQIjRA COTTONN E FZ’41)JYKEI EKVLLEEHGKEKYHR1KDRRQRGQEENVIVKISRKIIEE BRDBNN E YEtCFEEINKVLF---GREEGQQcj-EERLQ----ESVIVEISKKbRE SOYBN** ** **** * ** - *** * * * *MNRKRGLSAGSMSAPEHPK N Q1j)FFTIAGPKNYPF-)ALDVS 115A.OLSQG---ATSPRGKGSEGY N COTTONL-NKNAKSSSKKSTSSESE N SRjPtY F(FFEITPKRNP—I4DLNIF BRDBNL-SKHAKSSSRETI SSEDK N SRIjJt Y(IJCLFEIT-QRNP—JDLDVF SOYBN* *** ** ** * .*** ** * ** ** * ***VGLADIMTLcSTSIGIVTEGRIACPHLGQHS--WSSPRE--R II 5A. 0VVAFEIN4 I COTTONVNYVEILLISRAIVIVTVN$(GDFjEVGQRNENQQGLREEYDEEKE BRDENSOYBN* ** ** *** * *** ** ** * **GDQDIT 115A. oCOTTONQGEEEIRK__QVQNYAPGDVLVIPAGfrT__AIKASSt4t44LVGI*EN 8RDBNSOYBN** ** * * ** ** * * * ** * ** **EI-KVLNTLEREIRQINVPRGEEIEEVLQAQKDQV—ILRGPQRRS 115A. 0I-KNVRQWDRQAKELNF3VE-SRLVDEVFNNNPQESYFVSGRORRG COTTONSQIHKPVKELFtPGS-AQEVDTLLENQKQ-SHFANAQPRER BRDBNNLJSKUJVISQIPSQVQEL1PRS-AKDI ENLIKSQSE—SYFVDAQPQQK SOYBN* * ** ** * **** * *****•*** * * ** *1DEARSSS 115A.0FDERRGSNNPLSPFLDFARLF COTTONEIGSQEIKDHLYSILG--SF— BRDI3NEEGNKGRKGPLSSILR--AFY SOYBN* ** *111(MVFASLLMILLAISSSSA/ALT). Interestingly, part of the potentialsignal sequence was contained in an almost complete sequenceduplication at the amino terminus (Fig. 19) The predictedmolecular weight of the protein encoded by amino acids 48 through466 was approximately 47 kD, still slightly larger than themature protein, suggesting that further post—translationalprocessing might occur. To test this assumption, latecotyledonary somatic embryos were labelled for 4 hours with 35_methionine, and then chased for up to 24 hours. Total proteinprofiles (Fig. 21, upper panel) and immunoprecipitation oflabelled proteins with 41 kD polyclonal antibodies (Fig. 21,lower panel) revealed a precursor protein that was approximately3-4 kD larger than the mature protein. Processing of theprecursor was not evident until at least 4 hours of chase, andwas almost complete by 8 hours of chase (Fig. 21). Anotherprotein was detected by immunoprecipitation that did not changeduring the chase period (Fig. 21). Since no major proteins ofthis size were observed in total labelled protein profiles, itwas concluded that this is a cross—reacting protein.Two-dimensional protein analysis suggested that the 41 kDprotein class consisted of several isoelectric variants, possiblyencoded by a multi-gene family (Figs. 5 and 13). Spruce genomicDNA was digested with several different restriction enzymes,blotted and probed with 115A.O cDNA under moderately-stringentconditions (Fig. 22). Numerous bands were observed in eachdigest and copy reconstruction standards suggested the presenceof approximately 10-15 copies per genome (Fig. 22).112FIGURE 21. Pulse:chase-labelling of late cotyledonary somaticembryos differentiated on 40 M ABA. Embryos were labelled for 4hours with 35S-methionine (0 time), then chased with coldmethionine for up to 24 hours.UPPER PANEL: Total labelled protein profile. Open arrow denotesthe potential precursor to the mature 41 kD protein (closedarrow).LOWER PANEL: Labelled proteins from the above treatmentsinuuunoprecipitated with 41 kD antibody. The disappearance of theprecursor protein (open arrow) and the appearance of the matureprotein (closed arrow) is visible.“3Chase IhrIkD 0 2 4 8 16 2499.4—68.0—43.0— 129.0—18.4— iiii— ——14.3—E. —43.o_vV.A114FIGURE 22. DNA gel blot analysis of spruce DNA probed with115A.0 cDNA. Lane 1 contains lambda Hind III molecular weightstandards. Lanes 2 through 5 contain 5 g of spruce DNA digestedwith Hind III (Lane 2), Barn HI (Lane 3), Eco RI (Lane 4) or Xba I(Lane 5). Lanes 6 through 10 contain gene copy numberreconstructions at 0.8 copy per haploid genome (Lane 6), 1.6copies (Lane 7), 2.4 copies (Lane 8), 4 copies (Lane 9) or 8copies (Lane 10). Gene copy reconstructions were made using115A..0 cDNA, and a haploid DNA content of 8.5 pg.p!43M40)0•II- F) c)I01 0) -.4 CD 0III116During the course of the above work, other cDNAs were alsoisolated and sequenced. The results revealed similarity toangiosperm legumin—type storage proteins (Dr. Craig Newton,personal communication). The spruce legumin—type sequences,designated XI5H, encoded precursor proteins with conservedproteolytic processing sites which would yield disulf ide-linked35-33 and 24—22 kD proteins (Dr. Craig Newton, personalcommunication), similar to the storage proteins identified inthis study. Thus, probes were available to study the expressionof the spruce vicilin-type (41 kD) and legumin-type (35-33, 24-22kD) storage proteins.4.6. RNA GEL BLOT ANALYSIS OF LEGUMIN AND VICILIN STORAGEPROTEIN TRANSCRIPTS IN ZYGOTIC EMBRYOS AND SOMATIC EMBRYOSDIFFERENTIATED ON 40 LM AND 10 jtM ABAAnalysis of zygotic embryos collected during 1991 revealed, asdescribed previously, that the storage proteins displayeddifferential accumulation patterns, accumulated to high levelsduring cotyledon development, and reached steady state levels 1—1½ months prior to mature seed shed (Fig. 23). Analysis of themRNA levels for the spruce vicilin and legumin proteins revealedthat, although the different protein classes displayeddifferential accumulation patterns, their messages weredetectable at the same time, torpedo stage (stage 2), ofdevelopment (Fig. 23). The messages accumulated rapidly,reaching high levels during cotyledon development, and thendeclined rapidly to low levels, Scanning densitometry of RNA gelblots (using values normalized to constant rRNA) was used to117FIGURE 23. Changes in total proteins and storage protein mRNAduring zygotic embryo development.UPPER PANEL: Coomassie-stained SDS-PAGE of total proteins fromzygotic embryos collected during 1991, ranging in developmentfrom proembryos (PE) to mature seed embryos (Stage 4, September4). The spruce vicilin (V) and legumin (L) proteins areindicated. Each lane contains 12 g protein. MW, molecularweight standards.LOWER PANEL: RNA gel blot analysis of total RNA from the samezygotic embryo developmental stages described above, probed withspruce vicilin (115A.O), legumin (XI5H) or yeast 18S rRNA cDNA.kD97.466.245.031.0• I,StagePE2 33-4___94Stage Stage 4PE2 3..4 JL AU SEMW 9 16 23 30 6 13 22 28 4---:21.514.4Stage 4IvJLJLViciliriLeguminrRNA119compare changes in levels of the mRNAs during development. Thelow loading in the proembryo and Stage 2 lanes, caused byinadequate supply of zygotic material from these early stages,required use of correction values (on the order of 68x) tonormalize these values. This increased the error probability forthe normalized values, so only samples from Stage 3 onwards werenormalized. Densitometry revealed that the leguinin mRNA reachedhigh levels earlier in cotyledon development than the vicilinmRNA (Table 4). The vicilin mRNA remained slightly higher inStage 4 embryos during the period between July 16 to July 30, butthen both mRNA classes declined to around 1% or less of maximallevels (Table 4).A different genotype (W70) was used during this phase of theresearch because of a loss in vigour of genotype W29. Somaticembryos of W70 were similar in general appearance to those of W29although, by 9 weeks on 40 M ABA, those of genotype W70 had ahigher chlorophyll content (Table 5), but had not precociouslygerminated.The W70 embryos which were matured on 40 jM ABA displayed asimilar trend in storage protein accumulation to that observedfor genotype W29. Legumin protein accumulations were detectableby SDS-PAGE in early cotyledonary stages, prior to the vicilin(Fig. 24). All proteins accumulated during cotyledonarydevelopment, and accumulation was still evident after 9 weeks onABA (Fig. 24). RNA gel blots revealed that both vicilin andlegumin mRNAs were detectable at the same time in round torpedostage embryos (Fig. 24), similar to the observations with zygotic120TABLE 4. Developmental changes in storage protein mRNAs inzygotic embryos as determined by scanning densitometry of RNA gelblots.Developmental Stage Storage protein mRNA level(% of maximal hybridization)Vicilin LeguminStage 3 72 % 100 %Stage 3—4 96 % 50 %Stage 4 (JL 9) 100 % 59 %Stage 4 (JL 16) 93 % 39 %Stage 4 (JL 23) 32 % 11 %Stage 4 (JL 30) 7 % 1 %Stage 4 (AU 6) 1 % < 1 %Stage 4 (AU 13) 1 % < 1 %Stage 4 (AU 28) 1 % < 1 %Mature 2% <1%121TABLE 5. Chlorophyll content (mg/g FW) of Stage 4-6 and 9 weekABA somatic embryos of genotypes W29 and W70. Mean ± SE.Genotype Stage of embryo development4-6 9 weeks ABAW29 0.014 ± 0.006 0.011 ± 0.001W70 0.019 ± 0.001 0.098 ± 0.030122FIGURE 24. Changes in total proteins and storage protein mRNAduring somatic embryo development on 40 M ABA.UPPER PANEL: Coomassie-stained SDS-PAGE of total proteins fromsomatic embryos matured up to 9 weeks on ABA. Developmentalstages are as described for Fig. 16. The vicilin (V) and legumin(L) proteins are indicated. Each lane contains 12 g protein.NW, molecular weight standards.LOWER PANEL: RNA gel blot analysis of total RNA from the same•somatic embryo developmental stages described above, probed withspruce vicilin (115A.0), legumin (XI5H) or yeast lBS rRNA cDNA.31.021.514.4PE G R F(ECkD MWPE G RFii97.4 —66.245.0I+LC)124embryos. Both nRNA classes increased during cotyledondevelopment, but declined during the later stages. RNA gel blotswere analyzed by densitometry as described previously. Thelegumin mRNA reached high levels earlier in cotyledon developmentthan the vicilin mRNA (Table 6), similar to observations madewith zygotic embryos. While both mRNA classes declined aftertheir peak levels (Fig. 24, Table 6), their decline was not aslarge as observed for zygotic embryos (Fig. 23, Table 4) and,after 9 weeks on ABA, both mRNA classes were present atapproximately 50% of their maximal levels (Table 6).Somatic embryos which matured on 10 M ABA developed in asimilar manner to those on 40 j.M ABA up to the mid-cotyledonstage. However, once the cotyledons had overgrown the shoot apex(after Stage 4-2), the embryos accumulated chlorophyll andexhibited hypocotyl/cotyledon elongation, characteristic ofprecocious germination. These differences prohibited theclassification of later stage embryos on 10 ).LM ABA based on thelength/morphological standards used for non—germinating embryoson 40 j.M ABA, and they were thus classified as germinants (Germ1-6). Embryos from both 10 M and 40 iM ABA were collected foranalysis after the same duration of ABA exposure. Germ 1 sampleswere precociously germinating embryos collected at the same timeas Stage 4—3 embryos on 40 M ABA and Germ 6 were more advancedprecocious germinants collected at the same time as Stage 4—8embryos (8 weeks). The intermediate stages formed adevelopmental size continuum of precocious embryos.125TABLE 6. Developmental changes in storage protein mRNAs insomatic embryos differentiated on 40 M ABA as determined byscanning densitometry of RNA gel blots.Developmental Stage Storage protein mRNA level(% of maximal hybridization)Vicilin LeguminProembryo 0 % 0 %Globular 0 % 0 %Round head 8 % 7 %Flat head 36 % 29 %Stage 3—1 62 % 66 %Stage 3—2 84 % 100 %Stage 3—3 76 % 63 %Stage 3—4 80 % 55 %Stage 4—1 50 * 50 %Stage 4—2 86 % 64 %Stage 4—3 100 % 67 %Stage 4—4 69 % 40 %Stage 4—5 73 % 46 *Stage 4—6 60 % 41 %Stage 4—7 61 % 47 %Stage 4—8 46 * 41 *Stage 4—9 51 * 50 *126Somatic embryos on 10 M ABA also displayed differentialstorage protein accumulation during the early stages of cotyledondevelopment. Legumin accumulation was detected prior to vicilin(Fig. 25), although these proteins occurred at a slightly laterstage in cotyledon development than in 40 ,M ABA embryos.However, the storage protein levels declined as precociousgermination commenced (Fig. 25) and did not reach the levelsfound in 40 jM ABA somatic embryos. RNA gel blots revealed thatboth vicilin and legumin mRNA5 were detectable in flat torpedostage embryos (Fig. 25). Both mRNA classes increased duringcotyledon development and declined as precocious germinationcommenced. However, very low levels were still detectable at theGerm 6 stage, several weeks after precocious germination hadstarted (Fig. 25). Densitometric analysis of RNA gel blotsrevealed that the legumin mRNA increased earlier than the vicilinmRNA (Table 7), similar to the pattern observed for zygoticembryos and somatic embryos on 40 uM ABA. Comparison of 10 MABA samples to a control sample from 40 M ABA somatic embryos onthe same blots revealed that the maximum vicilin and legumin mRNAlevels attained in the 10 ,LM ABA samples were approximately 70%of the maximum levels found in 40 jM ABA somatic embryos.4.7. ANALYSIS OF STORAGE PROTEIN EXPRESSION IN SOMATIC EMBRYOSIN RESPONSE TO OSMOTIC STRESSEarly cotyledonary (Stage 3-4) embryos differentiated on 40 pMABA were cultured for 2 weeks on medium containing no growthregulators, 40 pM ABA, 15% mannitol or mannitol plus the ABAbiosynthetic inhibitor fluridone, to determine if osmotic stress-127FIGURE 25. changes in total proteins and storage protein mRNAduring somatic embryo development on 10 jM ABA.UPPER PANEL: Coomassie-stained SDS-PAGE of total proteins fromsomatic embryos matured up to 8 weeks on ABA. Somatic embryoswere collected after the same duration of total ABA exposure asthose on 40 ).LM ABA. Embryos were classified by the samedevelopmental stages as those on 40 jtM ABA until after mid—cotyledon development (Stage 4-2), after which they began toprecociously germinate (Germ 1). Precocious germinants werecollected at increasing stages of development (Germ 1—Germ 6; seetext). The vicilin (V) and legumin (L) proteins are indicated.Each lane contains 12 jg protein. MW, molecular weightstandards.LOWER PANEL: RNA gel blot analysis of total RNA from the samesomatic embryo developmental stages described above, probed withspruce vicilin (115A.0), legumin (XI5H) or yeast 18S rRNA cDNA.GERM GERMkD MWPE G R FIEC 1 697.4 •. ,... . •—- - . •-•.4- -.h -66.245.031.0 ...21.514.4--- ——..rPEGR FIECGERM GERM1AILLegumin129TABLE 7. Developmental changes in storage protein mRNAs insomatic embryos differentiated on 10 jM ABA as determined byscanning densitometry of RNA gel blots.Developmental Stage Storage protein mRNA level(% of maximal hybridization)Vicilin LeguminProembryo 0 % 0 %Globular 0 % 0 %Round head 0 % 0 %Flat head 5 % 12 %Stage 3—1 15 % 47 %Stage 3—2 23 % 60 %Stage 3—3 82 % 81 %Stage 3—4 100 % 95 %Stage 4—1 88 % 100 %Stage 4—2 75 % 68 %Germi 41% 67%Germ2 9% 14%Germ3 1% 2%Germ4 1% 2%Germ5 1% 2%Germ6 3% 2%130FIGURE 26. Effects of culture on media containing no growthregulators, 40 M ABA, 15% inannitol or fluridone on somaticembryo development. Somatic embryos were matured on 40 M ABAuntil early cotyledonary stage (EC; Stage 3-4), after which theywere cultured for a further 2 weeks on media containing no growthregulators (GRF), 40 jM ABA (ABA), 50 mg/L fluridone (50 FL), 15%mannitol (MAN), mannitol + 50 mg/L fluridone (MAN + 50 FL),mannitol + 10 mg/L fluridone (MAN + 10 FL) or mannitol + 50 mg/Lfluridone + 40 ,M ABA (MAN + 50 FL + ABA).%31++-Jzzu<<0z—2mm-JLi.z -JLi.0It)LL0<132induced storage protein accumulation was mediated via ABA. After2 weeks on growth regulator—free medium (with or withoutfluridone), embryos had precociously germinated (Fig. 26),although those cultured with fluridone were bleached inappearance. These results indicated that fluridone was notinhibitory to embryo survival or growth. Embryos cultured on 40M ABA or 15% mannitol did not germinate precociously, andembryos cultured on mannitol were smaller in size than those on40 M ABA (Fig. 26). Embryos cultured on mannitol, mannitol plusfluridone or mannitol plus fluridone and ABA were similar insize, although those exposed to fluridone were bleached inappearance (Fig. 26).Protein profile differences between the various treatmentsafter the 2 week culture period were analyzed by SDS-PAGE (Fig.27). Early cotyledonary embryos collected prior to the start ofthe treatments contained low levels of storage proteins (Fig.27). After 2 weeks on growth regulator-free medium (with orwithout fluridone), no storage protein accumulation was observed,and the low levels initially present in the embryos had declined(Fig. 27). Embryos on 40 ABA accumulated storage protein, asdid those exposed to 15% mannitol, although accumulation wasslightly less in the mannitol-treated embryos (Fig. 27). Embryoscultured on mannitol plus 50 mg/L fluridone contained lowerlevels of storage proteins compared to the mannitol treatment,while embryos on mannitol plus 10 mg/L fluridone or mannitol plus50 mg/L fluridone and 40 J.LM ABA displayed higher storage protein133FIGURE 27. Changes in total proteins and storage protein mRNA insomatic embryos exposed to no growth regulators, 40 j&M ABA, 15%mannitol or fluridone.UPPER PANEL: Coomassie-stained SDS-PAGE of total proteins fromsomatic embryos. Embryos were treated as described for Fig. 26.The vicilin (V) and legumin (L) proteins are indicated. Eachlane contains 12 g protein.LOWER PANEL: RNA gel blot analysis of total RNA from the samesomatic embryo treatments as above, probed with spruce vicilin(115A. 0), legumin (XI5H) or yeast 18S rRNA cDNA.-C.)4ø)(Ø_-(71O1-1Ii7flUt’autirrn in1MAN+5OFLMAN+1OFLI‘ ktiEC GRFABA5OFLMANMAN+5OFLMAN+1OFLMAN+5OFL+ABAAN+5OFL+ABA135levels than those in the mannitol plus 50 mg/L fluridonetreatment (Fig. 27).Analysis of storage protein mRNAs were carried out for theabove treatments. Vicilin and legumin mRNA5 were present ininitial explants, declined to very low levels in the growthregulator—free treatment, and were undetectable in embryosexposed to growth regulator-free conditions with 50 mg/Lfluridone. Storage protein mRNAs were present at high levels inembryos cultured on 40 j.M ABA or 15% mannitol (Fig. 27). Embryosexposed to high levels of fluridone (50 mg/L) while on mannitolhad reduced storage protein mRNA levels, but embryos exposed tolow fluridone levels (10 mg/L) contained higher storage proteinmRNA levels, similar to the mannitol—only and ABA treatments(Fig. 27). However, embryos cultured on medium containingmannitol plus 50 mg/L fluridone and ABA contained vicilin andlegumin mRNA levels similar to the mannitol plus 50 Lug/Lfluridone treatment. The addition of ABA to this treatment didnot enhance storage protein transcript levels.1365. DISCUSSION5.1. INTERIOR SPRUCE STORAGE PROTEINSThis study showed that the major storage proteins of interiorspruce were a buffer-soluble 41 kD matrix protein and high salt-soluble disulf ide-linked crystalloid proteins of 35-33 kD and 24-22 kD. Minor proteins of 30 and 27.5 kD were also identified.Storage proteins of similar molecular weight, solubility anddisulf ide—linkage characteristics have been described from otherconifers such as white spruce (Gifford and Tolley 1989, Misra andGreen 1990), Pinus species (Gifford 1988) and Douglas fir (Greenet al. 1991), and similar disuif ide-linked proteins were found inthe non—coniferous gymnosperm Ginkgo biloba (Jensen and Berthold1989). In addition to these proteins, there were 2 minordisuif ide-linked proteins of 22 kD and 14 kD, similar to the“small-dimer” protein found in Ginkgo biloba (Jensen and Berthold1989)The results of this present study differ somewhat from thoseof Misra and Green (1990, 1991), who described a 42 kDcrystalloid storage protein. However, this may have been anartifact arising from incomplete extraction procedures. The factthat antibodies raised against the 42 kD crystalloid proteincross—reacted with the matrix 42 kD protein (Misra and Green1991) supports this conclusion.The solubility and disulf ide-linkage characteristics of thespruce 35-33 kD and 24-22 kD proteins were similar to those ofangiosperm ll-12S globulin legumin-type storage proteins, whereasthe 41 kD protein was similar to angiosperm albumin or 7S137globulin vicilin-type proteins (Higgins 1984, Shotwell andLarkins 1989). The identification and characterization of 2 cDNAclasses that are highly expressed during spruce embryodevelopment revealed further similarity to angiosperm vicilin—type and legumin-type storage proteins (this study and Dr. CraigNewton, personal communication), suggesting that the majorspruce storage proteins are homologous to these two angiospermstorage protein classes.Further similarities between interior spruce storage proteinsand angiosperm storage proteins included isoelectricheterogeneity for each protein class, revealed by two—dimensionalelectrophoresis, suggesting that both protein groups were membersof multi-gene families. This was substantiated for the sprucevicilin using DNA gel blots which indicated the presence ofapproximately 10-15 gene copies, a similar result to reports thatangiosperm storage proteins are encoded by multi—gene families(Harada et al. 1989, Higgins 1984, Nielsen et al. 1989, Shotwelland Larkins 1989).The relatively simple mixture of storage proteins in spruceprotein body samples and embryos suggested that complexproteolytic processing patterns did not occur during theirsynthesis. In angiosperms, both 7S (vicilin-type) and liS(legumin—type) storage proteins can show post—translationalprocessing, apart from co-translational signal sequence cleavage,to yield mature proteins (Müntz 1989, Shotwell and Larkins 1989).Angiosperm uS proteins are synthesized as large precursorproteins. The precursor protein forms an intrachain disulfide138linkage and is proteolytically processed at either a singlepoint to yield two disulf ide-linked proteins, or the precursor isproteolytically processed at two sites to yield two disulfidelinked proteins and a short, free linker polypeptide (Shotwelland Larkins 1989). Further processing of the disuif ide-linkedproteins may also occur (Dure 1989, Raynal et al. 1987). Thespruce legumin cDNA sequence revealed a similar single conservedproteolytic cleavage site (Dr. Craig Newton, personalcommunication).Pulse:chase labelling revealed that the mature interior sprucevicilin protein arose from the processing of a precursor protein3-4 kD larger, suggesting that extensive proteolytic processingof the spruce vicilin did not occur. Some angiosperm vicilin—type proteins, such as 13-conglycinin (Meinke et al. 1981) andphaseolin (Shotwell and Larkins 1989) also do not undergoextensive proteolytic processing apart from signal sequenceremoval. In contrast, some angiosperm vicilins, such as thosefrom cotton (Dure 1989) and pea (Spencer et al. 1983) undergoextensive post—translational processing to form complex proteinpatterns.5.2. ZYGOTIC EMBRYO STORAGE PROTEIN EXPRESSIONThe results of this present study indicate that the sprucelegumin-type proteins were detectable before the vicilin-type,thus indicating differential accumulation patterns. Low levelsof legumin proteins were detected in pre—cotyledonary torpedo—stage embryos, whereas the vicilin protein was only detected incotyledonary embryos, although both classes displayed major139accumulations during cotyledon development and reached peaklevels well before mature seed shed, as water loss associatedwith maturation drying began (data not shown). This pattern wassimilar to that commonly described for angiosperms, wheredifferent storage protein classes within the same embryodisplayed different accumulation patterns. In pea, vicilinsynthesis and accumulation precedes that of legumin (Boulter etal. 1987), while in rapeseed napin accumulates earlier thancruciferin (Crouch and Sussex 1981, Murphy et al. 1989) and insoybean, 13—conglycinin proteins are detectable prior to glycinin(Meinke et al. 1981).Storage protein accumulation occurs during the cell expansionphase, after cell division and prior to drying and desiccation(Bewley and Black 1985). The major accumulation of sprucestorage proteins during cotyledon development and prior tomaturation drying is similar to these observations withangiosperms. The presence of low levels of spruce leguminstorage proteins in torpedo—stage embryos indicated that somestorage protein expression occurred prior to cotyledondevelopment/expansion. Domoney et al. (1980) also reported thatsome pea legumin accumulated prior to the expansion phase.Recent work has shown that storage protein mRNAs only accumulatein cells lacking mitotic activity (Hauxwell et al. 1990), sospruce storage protein expression in pre—cotyledonary embryos mayreflect expression in non—dividing, expanding cells of theseearly stage embryos. It is also possible that the differentialappearance of spruce legumins in pre—cotyledonary embryos140reflected differences in storage protein expression patternsbetween the embryonic axis and cotyledons, with the vicilinsprimarily expressed in cotyledonary cells. Organ—specificstorage protein expression patterns in soybean showed littleglycinin accumulation compared to 13—conglycinin in embryonicaxes, but prominent levels of both storage proteins in cotyledons(Meinke et al. 1981).The pattern of storage protein accumulation described in thisstudy does not agree with most of the results reported recentlyfor white spruce (Misra and Green 1991). While these workersalso reported accumulations of some crystalloid proteins prior tothe accumulation of a 42 kD protein, they found that the 35-34 kDcrystalloids accumulated before 24-23 kD crystalloids. Incontrast, the present results indicated that the 35-33 kD and 24-22 kD crystalloids displayed the same temporal pattern. Based onsolubility, disulf ide-linkage and cDNA sequence characteristics,the spruce 35-33 kD and 24-22 kD proteins appear homologous tothe angiosperm legumins and, since the disulf ide-linkedangiosperm legumin proteins show concurrent accumulation patterns(Higgins 1984, Meinke et al. 1981, MUntz 1989, Shotwell andLarkins 1989), the results of this present study agree with theangiosperm data. Misra and Green (1991) also reported that themajor accumulation of the 42 kD protein occurred between the lasttwo collection dates (Aug. 29 and mature seed), well after theattainment of maximum dry weight and during the period of seeddevelopment in which maturation drying occurs. These resultscontrasted with those of this present study, where greatest141protein levels were attained at least 1 month prior to matureseed shed. This maximum coincided with steady state dry weightand the onset of maturation drying (data not shown), similar toresults with angiosperms. Some of the discrepancies betweenMisra and Green (1991) and this work arise from the methods usedby these workers to carry out their protein extractions andantibody production. It was also difficult to compare thepresent results with those of Misra and Green (1991) becausethey did not include data on morphological development, so thatthe protein samples analyzed were related only to collectiondate.Analysis of interior spruce zygotic embryos revealed that bothvicilin and legumin messages were detectable in torpedo stageembryos, even though the storage proteins displayed differentaccumulation patterns. Both the mRNA5 increased to peak levelsduring cotyledon development, with legumin mRNA peaking anddeclining prior to that of vicilin. However, both dropped to lowlevels during the last month or so of zygotic seed development,coincident with the drying phase. The simultaneous appearance ofboth storage protein mRNAs may have been indicative of coordinate regulation. Most angiosperm embryos which showdifferential storage protein accumulation also show differentialappearance of storage protein mRNA5, with the different mRNAclasses displaying temporal differences in appearance by a fewdays (Boulter et al. 1987, Finkelstein et al. 1985, walling etal. 1986, Yang et al. 1990), although Meinke et al. (1981) notedthat soybean 7S and uS storage protein mRNAs were detectable at142the same time during development. It is possible that the sprucevicilin and legumin mRNA5 do display different appearancepatterns, but these were not detected due to the time frame ofthe zygotic embryo collections. More frequent, meticulouscollections during the early stages of zygotic embryo developmentare required to determine if different appearance patterns existfor the various spruce storage protein mRNAs.Angiosperm storage protein mRNAs appear well afterfertilization and during early embryo development (Boulter et al.1987, Finkeistein et al. 1985, Harada et al. 1989, Nielsen et al.1989), as did those in spruce (this study). However, recent workby Kamalay et al. (1991) described the appearance of legumin—likemRNAs in megagametophyte tissue of eastern white pine justfollowing fertilization, and Misra and Green (1991) reported lowlevels of crystalloid protein in white spruce megagametophytesjust following fertilization, well before their visibleaccumulation in this study. These results suggest that lowlevels of transient storage protein expression inmegagametophytes may be associated with fertilization or theinduction of embryo development. Recent work with rapeseedmicrospore—derived embryos revealed that the heat shock treatmentused to induce embryogenesis also caused the transient appearanceof storage protein (napin) mRNA (Boutilier et al. 1991),providing further support for this hypothesis.The different accumulation patterns for the spruce storageprotein mRNAs were similar to results described for angiosperms.In pea, vicilin mRNA and protein showed peak accumulations prior143to those for legumin (Boulter et al. 1987, Yang et al. 1990) and,in rapeseed, napin mRNA and protein showed peak accumulationsprior to those of cruciferin (Crouch and Sussex 1981, Finkelsteinet al. 1985). The accumulation of angiosperm storage proteintranscripts during development is primarily regulated at thetranscriptional level, although differences in mRNA stabilityalso play a role (Delisle and Crouch 1989, Evans et al. 1984,Harada et al. 1989, Nielsen et al. 1989). While analyses oftranscriptional rates were not performed, the different storageprotein accumulation patterns, and the accumulation patterns oftheir mRNA5 suggest, by analogy with other systems, thatregulation in spruce may also be primarily at the transcriptionallevel, although other factors may also be involved. Both storageprotein mRNAs are present at the same early stage, but thevicilin protein accumulates later than the legumin proteins,suggesting that translational or post—translational regulationmay also be involved. In soybean, vicilin mRNA is detectablewell before protein accumulation and the proteins aresynthesized, but then degraded rapidly (Shuttuck-Eidens andBeachy 1985). As embryo development proceeds, the proteinbecomes more stable and accumulates, indicating thatdevelopmental changes in post—translational regulation forsoybean vicilin exist. A similar type of mechanism may beassociated with spruce vicilin accumulation.5.3. SOMATIC EMBRYO STORAGE PROTEINSThis study revealed that interior spruce somatic embryosaccumulated the same major storage proteins as zygotic embryos,144based on molecular weight, solubility and disuif ide-linkagecharacteristics, and migration in two—dimensional gels. Hakmanet al. (1990) also reported that Norway spruce somatic andzygotic embryos contained similar proteins, while Joy IV et al.(1991) reported that white spruce somatic and zygotic embryototal protein profiles were similar, although some of the majorproteins were absent or were not highly expressed. Work withangiosperm systems has also shown that non—zygotic (somatic andmicrospore—derived) embryos contain the same proteins as theirzygotic counterparts (Crouch 1982, Krochko et al. 1989, Shoemakeret al. 1987, Stuart et al. 1988, Tewes et al. 1991). Theseresults suggest that the tissue culture process does not alterthe regions of storage protein genes encoding structuralinformation, leading to the expression of the same storageproteins as found in zygotic embryos. This is important ifsomatic embryo—derived material is to be used forbiotechnological applications.The levels of storage proteins that accumulated in sprucesomatic embryos in the presence of 40 M ABA were similar to orhigher than those found in mature zygotic embryos. Recent workin our lab (Cyr et al. 1991) has confirmed that somatic embryostorage protein levels may differ from those of zygotic embryos.These results are in contrast to the angiosperm data, wherestorage proteins were significantly lower in non—zygotic embryos(Crouch 1982, Krochko et al. 1989, Shoemaker et al. 1987, Stuartet al. 1988, Taylor et al. 1990). Also, storage proteinproportions are altered in alfalfa somatic embryos (Krochko et145al. 1989, Stuart et al. 1988). The alteration of storage proteinlevels in alfalfa somatic embryos is believed to be due to lowtranslation efficiency of storage protein mRNAs during earlydevelopment, a reflection of their inability to be incorporatedinto polysomes (Pramanik et al. 1991). However, none of theseangiosperm studies utilized ABA during somatic embryodifferentiation, with the exception of Taylor et al. (1990), whoused a short pulse of low ABA level during embryo maturation.Since ABA induces storage protein accumulation in culturedzygotic embryo tissues (Barratt 1986, Bray and Beachy 1985,Croissant—Sych and Bopp 1988, Eisenberg and Mascarenhas 1985,Finkelstein et al. 1985) and may enhance storage proteintranslation (Finkelstein et al. 1985), the fact that sprucesomatic embryos were differentiated continuously on 40 M ABAduring maturation could explain the high levels of storageproteins observed in somatic embryos in this study. Hakman etal. (1990) also noted that Norway spruce somatic embryosdifferentiated on ABA contained abundant levels of storageproteins, although quantitative comparisons were not made.Recently, Joy IV et al. (1991) reported that white spruce somaticembryos contained significantly lower levels of total proteinsand some major proteins as identified by SDS-PAGE. However,these workers used low levels of ABA (10 suM) compared to the 40M level used here.The analysis of different somatic embryo genotypes indicatedthat differences in storage protein levels occurred amongcotyledonary somatic embryos after 7 weeks of maturation on 40 ,.LM146ABA, although all genotypes accumulated significant levels. Thevariations observed between the different genotypes could havereflected effects of the differentiation protocol, which may haveto be optimized for each embryogenic line in order to obtainsimilar storage protein levels in each line. On the other hand,the variations may have reflected genetic differences arisingfrom the individual seed embryos used for embryogenic tissueinduction, since it is known that genotype—dependent differencesin storage protein accumulation can occur (Higgins 1984).5.4. DEVELOPMENTAL EXPRESSION OF STORAGE PROTEINS IN SOMATICEMBRYOS ON 40 JLM AND 10 jLM ABASpruce somatic embryos cultured on 40 M ABA displayeddifferential storage protein accumulation patterns similar tothose described for zygotic embryos. The legumin-type proteinsappeared earlier in development, but both types accumulatedprimarily during cotyledon development. The appearance ofvicilin and legumin mRNAs in torpedo stage somatic embryos on 40M ABA was also similar to their developmental appearance inzygotic embryos. Also, the developmental accumulation pattern ofstorage protein mRNA5 in somatic embryos on 40 ).LM ABA was similarto that in zygotic embryos, with legumin mRNA peaking andstarting to decline prior to vicilin mRNA. All of these resultssuggest that the patterns of storage protein gene induction andaccumulation were similar between the two embryo types, althoughtotal storage protein mRNA levels were not compared between them.These results are different from those commonly reported forangiosperm non—zygotic embryos, which show temporally—altered147storage protein expression, with accumulations earlier indevelopment than in their zygotic counterparts (Crouch 1982,Shoemaker et al. 1987). However, these studies did not use ABAduring differentiation. Since ABA promotes more normal embryomaturation in excised zygotic embryos (Ackerson l984ab) anddeveloping somatic embryos (this study, Ammirato 1974, Kamada andHarada 1981, Roberts et al. 1990a), as well as induces storageprotein gene expression in angiosperms, these altereddevelopmental expression patterns may be due to the lack ofexogenous ABA, which, if it were supplied would help to promotemore normal patterns of gene expression in these embryos.Support for this proposal comes from a recent study by Wilen etal. (1990) using microspore—derived rapeseed embryos. Theseworkers found that exposure of embryos to a 48 hour pulse of ABAat different stages of development promoted the correctdevelopmental induction of napin and cruciferin transcripts.They also obtained levels of storage protein mRNA similar to thatobserved in equivalent stage zygotic embryos, although thedevelopmental accumulation patterns of the storage proteinsthemselves were not characterized.Apart from the similarities described above for spruce zygoticand somatic embryos, there were also differences in storageprotein gene expression. Storage proteins accumulated graduallyand continuously for a prolonged period in somatic embryos, andwere still increasing after 9 weeks of maturation on 40 ).LM ABA.High levels of storage protein mRNA5 were also observed duringthis period, such that after 9 weeks on ABA, they were still148present at 50% of their maximal levels. This pattern was inmarked contrast to that observed for zygotic embryos, whichdisplayed a more rapid and transient period of storage proteinand storage protein mRNA accumulation, after which protein levelsdid not increase and mRNA levels declined rapidly, toapproximately 1% or less of maximal levels. The continuousaccumulation of storage proteins and the high levels of theirtranscripts may reflect the constant exposure of somatic embryosto high levels of ABA throughout maturation. Also, somaticembryos remain fully hydrated during maturation and, based onstorage protein accumulation, metabolically active. In contrast,zygotic embryos desiccate and enter a period of metabolicquiescence during later stages of seed development (Bewley andBlack 1985). Storage protein synthesis and message levelsgenerally decline during the maturation drying and desiccationstage of embryo development (Dure and Galau 1981, Finkelstein etal. 1985, Galau et al. 1987, Kermode et al. 1989). Recentresults in our lab have indicated that storage protein transcriptlevels also decline rapidly in spruce somatic embryos exposed toan artificial drying regime (Dr. David Cyr, personalcommunication) that is used to mimic normal seed/embryo drying.Spruce somatic embryos differentiated on low levels (10 jIM) ofABA also accumulated some storage protein. The differentialstorage protein accumulation patterns observed in zygotic embryosand somatic embryos exposed to 40 M ABA were also observed on 10M ABA. Furthermore, both vicilin and legumin mRNA5 weredetected by torpedo stage in these embryos, as were the149differential accumulation patterns, with legumin mRNA increasingprior to vicilin. This was similar to the pattern observed withzygotic embryos and somatic embryos on 40 ABA, although themaximal storage protein mRNA levels attained on 10 M ABA wereless than those for 40 J.LM ABA somatic embryos. This isattributed to the lower ABA level used. Similar ABA dose-dependence has been shown for storage protein gene expression incultured zygotic (Finkelstein et al. 1985, Delisle and Crouch1989) and non-zygotic (Wilen et al. 1990) embryos, with higherABA treatments resulting in higher storage protein transcriptlevels. Interestingly, somatic embryos on high or low ABAdisplayed similar developmental induction and differentialregulation of storage proteins, suggesting that ABA couldmaintain the normal developmental pattern of gene expressionregardless of the level used, although ABA did affect thequantitative expression of storage protein genes in similar waysto its action in microspore-derived embryos (Wilen et al. 1990).The fact that both high and low ABA-treated somatic embryos didnot exhibit storage protein transcripts until torpedo stage ofdevelopment, although they had been exposed to ABA since theproembryo stage, suggested that the competence to respond to ABA,as manifested by storage protein mRNA induction, did not occuruntil torpedo stage. The differences between the relative levelsof vicilin and legumin transcripts during their accumulation inearly cotyledonary embryos were not as great in somatic embryoson high levels of ABA. This may be attributed to a morestimulatory effect of high levels of ABA on vicilin transcript150levels. The preferential enhancement by ABA of vicilin proteinsand transcripts over other storage proteins has been documented(Bray and Beachy 1985, Schroeder 1984).In contrast to the pattern observed in somatic embryos on 40jM ABA, embryos on 10 uM ABA began to germinate precociously oncethey had reached mid-cotyledon stage of development. This wasparallelled by a decline in storage protein mRNAs and storageprotein levels, so that storage proteins did not accumulate tothe high levels observed in 40 ,M ABA somatic embryos. Recently,Joy IV et al. (1991) also reported that white spruce somaticembryos that were differentiated on low (10 M) ABA did notcontain storage proteins at the level of their zygoticcounterparts. While low ABA could maintain embryo developmentand prevent precocious germination initially, it was no longereffective by the mid-cotyledonary stage, suggesting that a lossof sensitivity to ABA occurred as the embryo matured. Loss ofABA sensitivity has been reported to occur during zygotic embryomaturation (Eisenberg and Mascarenhas 1985, Finkelstein et al.1985, Kermode et al. 1989, Rivin and Grudt 1991), and this changein sensitivity has been attributed to changes caused bymaturation drying and desiccation (Kermode et al. 1989). Thedecline in storage protein transcripts commonly observed duringlate zygotic embryogeny (this study, Finkelstein et al. 1985,Harada et al. 1989, Nielsen et al. 1989, Walling et al. 1986) maypartially reflect changes due to lowered ABA levels and reducedABA sensitivity.151While changes in ABA sensitivity occur in response todesiccation, work with maize viviparous mutants, which do notundergo desiccation, has shown that embryos also undergo a lossof ABA sensitivity during maturation without desiccation (Rivinand Grudt 1991). These results and the present results withspruce somatic embryos on 10 M ABA indicate that there may bedevelopmental changes in ABA sensitivity that occur duringmaturation, regardless of drying, allowing precociousgermination.Spruce somatic embryos that germinated precociously on 10 jMABA still contained low levels of storage protein transcripts,even after several weeks of germination. Both zygotic embryosand partially—dried somatic embryos have undetectable levels ofstorage protein transcripts within days of germination (Dr. DavidCyr, personal communication). It has been suggested that dryingor desiccation is required to switch from an embryo maturation toa germinative program (Kermode and Bewley 1989, Kermode et al.1989). Furthermore, spruce somatic embryos given a partialdrying treatment prior to germination exhibit much more normalgerminative and post-germinative growth (Roberts et al. l990b,1991). The role of embryo developmental stage on gene expressionduring precocious germination was explored by Finkeistein andCrouch (1984). Mid-cotyledon stage rapeseed zygotic embryosplaced on germination media germinated abnormally, with rootdevelopment but little hypocotyl elongation and secondarycotyledon development formation as opposed to leaf formation.These embryos continued to express storage protein genes and152accumulate storage proteins during the 4 weeks of precociousgermination. Older, maturation stage embryos germinated intonormal—looking seedlings, but storage protein transcripts werestill detectable and storage protein degradation took severalweeks. This was attributed to the continued synthesis andturnover of storage proteins, indicating concurrent expression ofembryo developmental and germinative programs. Only mature, dry,seed embryos responded normally during germination, with nostorage protein transcript retention and rapid degradation ofstorage proteins. These results suggested that drying wasrequired to switch developmental programs, and that embryos werecapable of expressing both embryo maturation and germinationprograms simultaneously. Kriz et al. (1990) also reported thatprecociously germinating embryos of viviparous mutants, which didnot undergo desiccation, exhibited prolonged expression ofstorage protein transcripts and an incomplete switch from embryodevelopment to germination, further indicating the importance ofdrying for the change from embryo developmental to germinativegene expression. These results are similar to the observationswith spruce somatic embryos on 10 iM ABA, which appear havegerminated, but still contain low levels of storage proteintranscripts, suggesting simultaneous expression of embryo andgermination programs, in contrast to germinants of normal embryosor partially-dried somatic embryos (Dr. David Cyr, personalcommunication).It was suggested that somatic embryos on 10 M ABA lost theirsensitivity to ABA and germinated precociously. However, somatic153embryos on 40 M ABA did not display such an evident change.After 9 weeks on ABA, these embryos had not yet precociouslygerminated, although prolonged culture on these high levels ofABA will eventually lead to their germination (data not shown,Dunstan et al. 1991), suggesting that they also exhibit adevelopmental change in ABA sensitivity. Somatic embryos on 40btM ABA displayed a decline in storage protein transcript levelsby 9 weeks of maturation. This decline may be indicative of achange in ABA sensitivity, suggesting that the embryos are, atleast at the molecular level, starting to switch into aprecocious germination mode even though protein levels are stillhigh. Further studies using germination-specific probes arerequired to determine if this is the case. The use of molecularprobes to determine the state of embryo development/germinationhas important implications in determining the appropriatedevelopmental stage at which maturing somatic embryos should beremoved for partial drying and artificial seed production.5.5. OSMOTIC STRESS AND STORAGE PROTEIN GENE EXPRESSIONEarly cotyledonary spruce somatic embryos cultured on 40 LMABA or 15% mannitol for 2 weeks matured into well developedcotyledonary stage embryos which did not precociously germinate,although embryos on 15% mannitol were smaller in size. Thispromotion and/or enhancement of somatic embryo differentiationhas been reported for angiosperm (Litz 1986, Nadel et al. 1989)and conifer (Lu and Thorpe 1987, Roberts 1991, Tremblay andTremblay 1991a) systems. In this study, somatic embryos maturedon ABA or mannitol contained high levels of storage protein154transcripts and storage proteins, although storage protein levelsin the mannitol-treated embryos were not as high as in the ABA-treated embryos. This confirms previous results that indicatedan osmotic stress—induced accumulation of storage proteins inspruce somatic embryos (Roberts 1991) and agrees with angiospermdata showing the same pattern in cultured embryo tissues (Barratt1986, Finkelstein and Crouch 1986, Goffner et al. 1990, Xu et al.1990) and developing microspore-derived embryos (Wilen et al.1990)Osmotic stress—induced storage protein accumulations have beenattributed to a stimulation of endogenous ABA levels by thestress (Barratt 1986, Rivin and Grudt 1991, Wilen et al. 1990),although others have suggested that endogenous ABA is notdirectly associated with the response (Barratt et al. 1989,Finkeistein and Crouch 1986, Goffner et al. 1990, Xu et al.1990). Since both ABA and osmoticum inhibit water uptake(Finkelstein and Crouch 1986, Schopfer and Plachy 1984), it hasbeen suggested that alteration in embryo cell osmotic potential,and not ABA, is the primary effector of the storage proteinresponse. To study the potential role of ABA in osmotic stress—induced storage protein gene expression in spruce, developingsomatic embryos were exposed to the ABA biosynthetic inhibitor,fluridone, during osmotic stress. This study indicated thatsomatic embryos exposed to high levels of fluridone contained lowlevels of storage protein transcripts and did not accumulate highlevels of storage proteins. Furthermore, lower fluridone levelsin combination with osmotic stress, which should have allowed155more endogenous ABA biosynthesis, displayed higher levels ofstorage protein transcripts and storage protein accumulation.These results suggested that osmotic stress—induced storageprotein gene expression in spruce somatic embryos was mediatedvia ABA. However, while embryos matured on mannitol—containirigmedium plus high levels of fluridone and ABA contained higherstorage protein levels compared to those on mannitol plusfluridone, the storage protein transcript levels did not increasesubstantially. This suggested that exogenous ABA couldnot stimulate the fluridone-induced inhibition of storage proteintranscript levels under the conditions used here. The inabilityof exogenous ABA to stimulate storage protein transcript levelsmay have been due to little uptake of applied ABA, due to thehighly negative osmotic potential of the mannitol-containingmedium. The possible inhibition of ABA uptake by osmotic stresswas suggested by the results of Bray and Beachy (1985). Theseworkers noted that low levels of osmoticum (0.5—3% sucrose)stimulated endogenous ABA levels and exogenous ABA applicationsenhanced these levels. However, at the highest sucroseconcentration tested (10%), endogenous ABA levels were lower thanthose from the 0.5-3% sucrose treatments,and exogenous ABAapplication did not enhance these levels.If ABA uptake was low due to the high osmoticum used in thisstudy, this could potentially account for the discrepancyobserved between storage protein and transcript levels.Finkelstein et al. (1985) reported that excised zygotic embryoscultured on low ABA did not accumulate storage protein156transcripts to the levels found in seed embryos, although storageproteins did reach their normal levels, suggesting that ABAenhanced translation. It is possible that the ABA level to whichthe spruce somatic embryos were exposed during the inannitol plusfluridone and ABA culture was not sufficient to induce a majorincrease in steady state storage protein transcript levels, butenhanced translation of the message to allow storage proteinaccumulation.It is also possible that the inability of ABA to completelyrestore storage protein transcript levels in the presence of highlevels of fluridone may have been due to some other indirecteffect of fluridone, although fluridone exposure did not inhibitembryo survival or growth. Interestingly, Fong et al. (1983)noted that exogenous ABA could only partially reverse fluridone—induced vivipary in maize embryos.While this present study did not measure endogenous ABA levelsduring development, other studies have shown that fluridone doesinhibit ABA biosynthesis and reduces endogenous ABA levels(Barratt et al. 1989, Bray and Beachy 1985). ABA is synthesizedvia the carotenoid pathway (Parry and Horgan 1991, Rock andZeevaart 1991), and fluridone inhibits phytoene desaturation inthis pathway, thereby inhibiting ABA biosynthesis (Zeevaart andCreelman 1988). The bleached appearance of the somatic embryosexposed to fluridone was indicative of carotenoid biosyntheticsuppression (Fong et al. 1983), and therefore, ABA biosynthesis.It is possible that alterations in endogenous ABA pools inresponse to the osmotic stress could have occurred, but other157studies using fluridone have shown that endogenous ABA levels dodecline, and the two week period of culture during thisexperiment should have allowed depletion of endogenous ABA poolswithout replenishment via biosynthesis.5.6. CONCLUDING STATEMENTThis study showed that both zygotic and somatic embryos ofspruce expressed the same storage proteins. These proteinsappeared homologous to known angiosperm storage protein classes,based on solubility, disulfide linkage and cDNA sequencecharacteristics. The two different storage protein classesdisplayed differential accumulation patterns. This was similarto the differential accumulation patterns often reported forangiosperm storage proteins. Spruce storage protein geneexpression was regulated by ABA and while somatic embryos maturedon high or low levels of ABA displayed a similar developmentalinduction of storage protein transcripts and their proteins tozygotic embryos, only those embryos cultured on high levels ofABA were capable of prolonged development without precociousgermination. This allowed the accumulation of storage proteinsto levels similar to or higher than levels found in zygoticembryos, although the period of storage protein transcript andprotein accumulation was more prolonged than that observed inzygotic embryos. These results suggested that, using storageprotein gene expression as a marker, somatic embryos that maturedon high levels of ABA were developmentally similar to zygoticembryos. This finding bodes well for their utilization inbiotechnological applications.158Somatic embryos must exhibit vigorous germination and post—germinative growth if they are to be useful. It is known thatsomatic embryos subject to an artificial drying regime, whichmimics normal seed/embryo desiccation and switches geneexpression from an embryo developmental to germinative mode,perform much better than if not dried (Parrott et al. 1988,Roberts et al. 1990b). Therefore, the removal and drying ofsomatic embryos at the appropriate time, prior to precociousgermination, is critical to optimize plantlet recovery. Whilesomatic embryos on high ABA levels did not germinateprecociously during the culture period used here, they eventuallydid, indicating an apparent loss of ABA sensitivity during thelater stages of somatic embryo maturation. Although the embryoshad a prolonged period of storage protein accumulation thatindicated a delay in precocious germination, the decline ofstorage protein mRNAs may have indicated the onset of loss of ABAsensitivity and precocious germination. Further studies, usingembryo developmental and germination—specific probes are requiredto confirm this, but molecular markers clearly have the potentialto gauge the maturity of somatic embryos and could indicate theonset of precocious germination well before protein changesbecome apparent. This would facilitate the removal of somaticembryos for partial drying and artificial seed production.1596. LITERATURE CITEDAckerson, R.C. 1984a. Regulation of soybean embryogenesis byabscisic acid. J. Exp. Bot. 35(152): 403—413.Ackerson, R.C. 1984b. Abscisic acid and precocious germination insoybeans. J. Exp. Bot. 35(152): 414—421.Anuuirato, P.V. 1974. The effects of abscisic acid on thedevelopment of somatic embryos from cells of caraway (Carum carviL.) Bot. Gaz. 135: 328—337.Attree, S.M., Budimir, S. and Fowke, L.C. 1990a. Somaticembryogenesis and plantlet regeneration from cultured shoots andcotyledons from stored seeds of black and white spruce (Piceamariana and Picea glauca). Can. J. Bot. 68: 30-34.Attree, S .M., Tautorus, T. E., Dunstan, D. I. and Fowke, L. C.1990b. Somatic embryo maturation, germination, and soilestablishment of plants of black and white spruce (Picea marianaand Picea glauca). Can. J. Bot. 68: 2583—2589.Attree, S.M. and Fowke, L.C. 1991. Micropropagation throughsomatic embryogenesis in conifers. In Biotechnology inAgriculture and Forestry, Vol. 17, High-Tech and MicropropagationI. Edited by Y.P.S. Bajaj. Springer-Verlag, Berlin, pp 53-70.Avjioglu, A. and Knox, R.B. 1989. Storage lipid accumulation byzygotic and somatic embryos in culture. Ann. Bot. 63: 409-420.Barratt, D.H.P. 1986. Modulation by abscisic acid of storageprotein accumulation in Vicia faba L. cotyledons cultured invitro. Plant Sci. 46: 159—167.Barratt, D.H.P., Whitford, P.N., Cook, S.K., Butcher, G. andWang, T.L. 1989. An analysis of seed development in Pisumsativum. VIII. Does abscisic acid prevent precocious germinationand control storage protein synthesis? J. Exp. Bot. 40(218):1009—1014.Bartels, D., Engelhardt, K., Roncarati, R., Schneider, K.,Rotter, M. and Salamini, F. 1991. An ABA and GA modulated geneexpressed in the barley embryo encodes an aldose reductaserelated protein. EMBO J. 10(5): 1037-1043.Becwar, M.R., Noland, T.L. and Wyckoff, J.L. 1989. Maturation,germination, and conversion of Norway spruce (Picea abies L.)somatic embryos to plants. In Vitro Cell. Dev. Biol. 25(6): 575-580.Becwar, M.R., Nagmani, R. and Wann, S.R. 1990. Initiation ofembryogenic cultures and somatic embryo development in loblollypine (Pinus taeda). Can. J. For. Res. 20: 810-817.160Becwar, M.R., Blush, T.D., Brown, D.W. and Chesick, E.E. 1991.Multiple paternal genotypes in embryogenic tissue derived fromindividual immature loblolly pine seeds. Plant Cell Tiss. Org.Cult. 26: 37—44.Bewley, J.D. and Black, M. 1985. Seeds: germination, structureand composition. In Seeds: Physiology of Development andGermination. Edited by J.D. Bewley and M. Black. Plenum Press,New York, pp 1-28.Blundy, K.S., Blundy, M.A.C. and Crouch, M.L. 1991. Differentialexpression of members of the napin storage protein gene familyduring exnbryogenesis in Brassica napus. Plant. Mol. Biol. 17:1099—1104.Bochicchio, A., Vazzana, C., Velasco, R., Singh, M. and Bartels,D. 1991. Exogenous ABA induces desiccation tolerance and leads tothe synthesis of specific gene transcripts in immature embryos ofmaize. Maydica 36: 11-16.Boulay, M.P., Gupta, P.K. Krogstrup, P. and Durzan, D.J. 1988.Development of somatic embryos from cell suspension cultures ofNorway spruce (Picea abies Karst.) Plant Cell Rep. 7: 134—137.Boulter, D., Evans, I.M., Ellis, J.R., Shirsat, A., Gatehouse,J.A. and Croy, R.R.D. 1987. Differential gene expression in thedevelopment of Pisum sativum. Plant Physiol. Biochem. 25(3): 283-289.Bourgkard, F. and Favre, J.M. 1988. Somatic embryos from callusof Sequoia sempervirens. Plant Cell Rep. 7: 445-448.Boutilier, K., Nozzolillo, C. and Miki, B.L. 1991. Expression ofnapin in Brassica napus microspores induced for embryogenesis.ISPMB 3rd International Congress, Tucson, AZ, Abstract no. 750.Bray, E. and Beachy, R.N. 1985. Regulation by ABA of 13-conglycinin expression in cultured developing soybean cotyledons.Plant Physiol. 79: 746—750.Buchholz, J.T. and Stiemert, M.L. 1945. Development of seed andembryos in Pinus ponderosa with special reference to seed size.Trans. Ill. Acad. Sci. 38: 27—50.Cheliak, W.M. and Rogers, D.L. 1990. Integrating biotechnologyinto tree improvement programs. Can. J. For. Res. 20: 452-463.Chesnut, R.S., Shotwell, M.A., Boyer, S.K. and Larkins, B.A.1989. Analysis of avenin proteins and the expression of theirmRNA5 in developing oat seeds. Plant Cell 1: 913-924.Chrispeels, M.J., Higgins, T.J.V., Craig, S. and Spencer, D.1982. Role of the endoplasmic reticulum in the synthesis of161reserve proteins and the kinetics of their transport to proteinbodies in developing pea cotyledons. J. Cell Biol. 93: 5-14.Council of Forest Industries of British Columbia. 1987. COFIAnnual Report.Creelman, R.A. 1989. Abscisic acid physiology and biosynthesis inhigher plants, Physiol. Plant. 75: 131-136.Croissant-Sych, Y. and Bopp, M. 1988. Formation and degradationof storage proteins in the embryo of Sinapis alba. J. PlantPhysiol. 132: 520—528.Crouch, M.L. 1982. Non—zygotic embryos of Brassica napus L.contain embryo—specific storage proteins. Planta 156: 520-524.Crouch, M.L. and Sussex, I.M. 1981. Development and storageprotein synthesis in Brassica napus L. embryos in vivo and invitro. Planta 153: 64-74.Cyr, D.R., Webster, F.B. and Roberts, D.R. 1991. Biochemicalevents during germination and early growth of somatic embryos andseed of interior spruce (Picea glauca engelmanii complex). SeedSci. Res. 1: 91—97.David, A., David, H. and Mateille, T. 1982. In vitro adventitiousbudding on Pinus pinaster cotyledons and needles. Physiol. Plant.56: 102—107.Davies, D.R. and Bedford, I.D. 1982. Abscisic acid and storageprotein accumulation in Pisum sativum embryos grown in vitro.Plant Sci. Lett. 27: 337—343.Delisle, A.J. and Crouch, M.L. 1989. Seed storage proteintranscription and mRNA levels in Brassica napus duringdevelopment and in response to exogenous abscisic acid. PlantPhysiol. 91: 617—623.Dhillon, S.S. 1987. DNA in tree species. In Cell and TissueCulture in Forestry, Vol. 1: General Principles andBiotechnology. Edited by J.M Bonga and D.J. Durzan. MartinusNijhoff, Boston, pp 298—313.Domoney, C., Davies, D.R. and Casey, R. 1980. The initiation oflegumin synthesis in immature embryos of Pisum sativum L. grownin vivo and in vitro. Planta 149: 454-460.Dunstan, D.I., Bekkaoui, F., Pilon, M., Fowke, L.C. and Abrams,S.R. 1988. Effects of abscisic acid and analogues on thematuration of white spruce (Picea glauca) somatic embryos. PlantSci. 58: 77—84.162Dunstan, D.I., Bethune, T.D. and Abrams, S.R. 1991. Racemicabscisic acid and abscisyl alcohol promote maturation of whitespruce (Picea glauca) somatic embryos. Plant Sci. 76: 219-228.Dure III, L. 1989. Characteristics of the storage proteins ofcotton. JAOCS 66(3): 356-359.Dure III, L. and Galau, G.A. 1981. Developmental biochemistry ofcottondeed embryogenesis and germination. XIII. Regulation ofbiosynthesis of principal storage proteins. Plant Physiol. 68:187—194.Dure III, L., Crouch, M., Harada, J., Ho, T. H. D., Mundy, J.,Quatrano, R., Thomas, T. and Sung, Z.R. 1989. Common amino acidsequence domains among the LEA proteins of higher plants. PlantMol. Biol. 12: 475—486.Durzan, D.J. and Gupta, P.K. 1987. Somatic eiubryogenesis andpolyembryogenesis in Douglas fir cell suspension cultures. PlantSci. 52: 229—235.Eisenberg, A.J. and Mascarenhas, J.P. 1985. Abscisic acid and theregulation of the synthesis of specific seed proteins and theirmessenger PNAs during culture of soybean embryos. Planta 166:505—514.Evans, I.M., Gatehouse, J.A., Croy, R.R.D. and Boulter, D. 1984.Regulation of the transcription of storage protein mRNA in nucleiisolated from developing pea (Pisum sativum L.) cotyledons.Planta 160: 559—568.Fabijanski, S. and Altosaar, I. 1985. Evidence for translationalcontrol of storage protein biosynthesis during embryogenesis ofAvena sativa L. (oat endosperm). Plant Mol. Biol. 4: 211—218.Feirer, R.P., Conkey, J.H. and Verhagen, S.A. 1989. Triglyceridesin embryogenic conifer calli: a comparison with zygotic embryos.Plant Cell Rep. 8: 207-209.Fernandez, D.E. , Turner, F.R. and Crouch, M.L. 1991. In situlocalizatiomn of storage protein mRNA5 in developing meristems ofBrassica napus embryos. Development 111: 299—313.Finer, J.J., Kriebel, H.B. and Becwar, M.R. 1989. Initiation ofembryogenic callus and suspension cultures of eastern white pine(Pinus strobus L.). Plant Cell Rep. 8: 203-206.Finkelstein, R.R. and Crouch, M.L. 1984. Precociously germinatingrapeseed embryos retain characteristics of embryogeny. Planta162: 125—131.Finkelstein, R.R., Tenbarge, K.M., Shumway, J.E. and Crouch, M.L.1985. Role of ABA in maturation of rapeseed embryos. Plant.Physiol. 78: 630—636.163Finkelstein, R.R. and Crouch, M.L. 1986. Rapeseed embryodevelopment in culture on high osmoticum is similar to that inseeds. Plant Physiol. 81: 907—912.Finkelstein, R.R. and Somerville, C.R. 1990. Three classes ofabscisic acid (ABA)-insensitive mutations of Arabidopsis definegenes that control overlapping subsets of ABA responses. PlantPhysiol. 94: 1172—1179.Flinn, B.S., Webb, D.T. and Georgis, W. 1986. In vitro control ofcaulogenesis by growth regulators and media components inembryonic explants of eastern white pine (Pinus strobus). Can. J.Bot. 64: 1948—1956.Flinn, B.S., Webb, D.T. and Newcomb, W. 1989. Morphometricanalysis of reserve substances and ultrastructural changes duringcaulogenic determination and loss of competence of Eastern whitepine (Pinus strobus) cotyledons in vitro. Can. J. Bot. 67: 779-789.Fong, F., Smith, J.D. and Koehier, D.E. 1983. Early events inmaize seed development. l-Methyl-3-phenyl-5-(3-[trifluromethyl]phenyl) -4-(l H) -pyridinone induction of vivipary.Plant Physiol. 73: 899—901.Fourney, R.M, Miyakoshi, J., Day III, R.S. and Paterson, M.C.1988. Northern blotting: efficient RNA staining and transfer. BRLFocus 10: 5—7.Galau, G.A., Hughes, D.W. and Dure III, L. 1986. Abscisic acidinduction of cloned cotton late embryogenesis—abundant (Lea)mRNAs. Plant Mol. Biol. 7: 155—170.Galau, G.A., Bijaisoradat, N. and Hughes, D.W. 1987. Accumulationkinetics of cotton late embryogenesis—abundant mRNAs and storageprotein mRNAs: Coordinate regulation during embryogenesis and therole of abscisic acid. Dev. Biol. 123: 198—212.Ghosh, S., Gepstein, S., Heikkila, J. and Dumbroff, E.B. 1988.Use of a scanning densitometer or an ELISA plate reader formeasurement of nanogram amounts of protein in crude extracts frombiological tissues. Anal. Biochem. 169: 227—233.Gifford, D.J. 1988. An electrophoretic analysis of the seedproteins from Pinus monticola and eight other species of pine.Can. J. Bot. 66: 1808—1812.Gif ford, D.J. and Tolley, M.C. 1989. The seed proteins of whitespruce and their mobilization following germination. Physiol.Plant. 77: 254—261.Gleddie, S., Keller, W. and Setterfield, G. 1983. Somaticembryogenesis and plant regeneration from leaf explants and cell164suspensions of Solanum melongena (eggplant). Can. J. Bot. 61:656—666.Goffner, D., This, P. and Delseny, N. 1990. Effects of abscisicacid and osmotica on helianthinin gene expression in sunflowercotyledons in vitro. Plant Sci. 66: 211-219.Goldberg, R.B., Barker, S.J. and Perez-Grau, L. 1989. Regulationof gene expression during plant embryogenesis. Cell 56: 149-160.Green, N.J., McLeod, J.K. and Misra, S. 1991. Characterization ofDouglas fir protein body composition by SDS-PAGE and electronmicroscopy. Plant Physiol. Biochem. 29(1): 49—55.Groot, S.P.C., van Yperen, 1.1. and Karssen, C.M. 1991. Stronglyreduced levels of endogenous abscisic acid in developing seeds oftomato mutant sitiens do not influence in vivo accumulation ofdry matter and storage proteins. Physiol. Plant. 81: 73-78.Guerche, P., Tire, C., de Sa, F.G., De Clercq, A.D., Van Montagu,N. and Krebbers, E. 1990. Differential expression of theArabidopsis 2S albumin genes and the effect of increasing genefamily size. Plant Cell 2: 469—478.Gupta, P.K. and Durzan, D.J.. 1986a. Plantlet regeneration viasomatic embryogenesis from subcultured callus of mature embryosof Picea abies (Norway spruce). In Vitro Cell. Dev. Biol. 22(11):685—688.Gupta, P.K. and Durzan, D.J. 1986b. Somatic polyembryogenesisfrom callus of mature sugar pine embryos. BioTechnology 4: 643—645.Gupta, P.K. and Durzan, D.J. 1987. Biotechnology of somaticpolyembryogenesis and plantlet regeneration in loblolly pine.BioTechnology 5: 147-151.Hakman, I., Fowke, L. C., von Arnold, S. and Eriksson, T. 1985.The development of somatic embryos in tissue cultures initiatedfrom immature embryos of Picea abies (Norway spruce). Plant Sd.38: 53—59.Hakman, I. and von Arnold, S. 1985. Plantlet regeneration throughsomatic embryogenesis in Picea abies (Norway spruce). J. PlantPhysiol. 121: 149—158Hakman, I. and Fowke, L.C. 1987. Somatic embryogenesis in Piceaglauca (white spruce) and Picea mariana (black spruce). Can. J.Bot. 65: 656—659.Hakman, I. and von Arnold, S. 1988. Somatic embryogenesis andplant regeneration from suspension cultures of Picea glauca(white spruce). Physiol. Plant. 72: 579-587.165Hakman, I., Stabel, P., Engström, P. and Eriksson, T. 1990.Storage protein accumulation during zygotic and somatic embryodevelopment in Picea abies (Norway spruce). Physiol. Plant. 80:441—445.Harada, J.J., Barker, S.J. and Goldberg, R.B. 1989. Soybean 13-conglycinin genes are clustered in several DNA regions and areregulated by transcriptional and post—transcriptional processes.Plant Cell 1: 415—425.Hatzopoulis, P., Fong, F. and Sung, Z.R. 1990. Abscisic acidregulation of DC8, a carrot embryonic gene. Plant Physiol. 94:690—695.Hauxwell, A.J., Corke, F.M.K., Hedley, C.L. and Wang. T.L. 1990.Storage protein gene expression is localised to regions lackingmitotic activity in developing pea embryos. An analysis of seeddevelopment in Pisum sativum XIV. Development 110: 283-289.Higgins, T.J.V. 1984. Synthesis and regulation of major proteinsin seeds. Ann. Rev. Plant Physiol. 35: 191-221.Hiscox, J.D. and Israelstam, G.F. 1979. A method for theextraction of chlorophyll from leaf tissue without maceration.Can. J. Bot. 57: 1332—1334.Jam, S.M., Newton, R.J. and Soltes, E.J. 1988. Enhancement ofsomatic embryogenesis in Norway spruce (Picea abies L.) Theor.Appl. Genet. 76: 501-506.Jam, S.M., Dong, N. and Newton, R.J. 1989. Somatic embryogenesisin slash pine (Pinus elliottii) from immature embryos cultured invitro. Plant Sci. 65: 233—241.Janick, J., Velho, C.C. and Whipkey, A. 1991. Developmentalchanges in seeds of loblolly pine. J. Amer. Soc. Hortic. Sci.116(2): 297—301.Jensen, U. and Berthold, H. 1989. Legumin-like proteins ingymnosperms. Phytochemistry 28(5): 1389-1394.Jensen, W.A. 1962. Botanical histochemistry. W.H. Freeman andCompany, San Francisco.Jensen, W.A. and Fisher, D.B. 1968. Cotton embryogenesis: theentrance and discharge of the pollen tube in the embryo sac.Planta 78: 158—183.Johnson, M.A., Carlson, J.A., Conkey, J.H. and Noland, T.L. 1987.Biochemical changes associated with zygotic pine embryodevelopment. J. Exp. Bot. 38(188): 518-524.166Joy IV, R.W., Yeung, E.C., Kong, L. and Thorpe, T.A. 1991.Development of white spruce somatic embryos. I. Storage productdeposition. In Vitro Cell. Dev. Biol. 27: 32-41.Kamada, H. and Harada, H. 1981. Changes in the endogenous leveland effects of abscisic acid during somatic embryogenesis ofDaucus carota L. Plant Cell Physiol. 22: 1423-1429.Kamalay, J.C., Baker, S., Rugh, C. and Whitmore, B. 1991. Geneproducts activated at fertilization in white pinemegagametophytes. ISPMB 3rd International Congress, Tucson, AZ,Abstract no. 545.Karnosky, D.L. 1981. Potential for forest tree improvement viatissue culture. BioSci. 31: 114—120.Karssen, C.M., Brinkhorst—van der Swan, D.L.C., Breekiand, A.E.and Koornneef, M. 1983. Induction of dormancy during seeddevelopment by endogenous abscisic acid: studies on abscisic aciddeficient genotypes of Arabidopdis thaliana (L.) Heynh. Planta157: 158—165.Kermode, A.R. and Bewley, J.D. 1989. Developing seeds of Ricinuscommunis L., when detached and maintained in an atmosphere ofhigh relative humidity, switch to a germinative mode without therequirement for complete desiccation. Plant Physiol. 90: 702—707.Kermode, A.R., Dumbroff, E.B. and Bewley, J.D. 1989. The role ofmaturation drying in the transition from seed development togermination. VII. Effects of partial and complete desiccation onabscisic acid levels and sensitivity in Ricinus communis L.seeds. J. Exp. Bot. 40: 303-313.Kim, Y.-H. and Janick, J. 1991. Abscisic acid and proline improvedesiccation tolerance and increase fatty acid content of celerysomatic embryos. Plant Cell Tiss. Org. Cult. 24: 83-89.Klimaszewska, K. 1989. Plantlet development from immature zygoticembryos of hybrid larch through somatic embryogenesis. Plant Sci.63: 95—103.Koornneef, 14., Reuling, G. and Karssen, C.M. 1984. The isolationand characterization of abscisic acid insensitive mutants ofArabidopsis thaliana. Physiol. Plant. 61: 377-383.Koornneef, N., Hanhart, C.J., Hilhorst, H.W.M. and Karssen, C.M.1989. In vivo inhibition of seed development and reserve proteinaccumulation in recombinants of abscisic acid biosynthesis andresponsiveness mutants in Arabidopsis thaliana. Plant Physiol.90: 463—469.Kriz, A.R., Wallace, M.S. and Paiva, R. 1990. Globulin geneexpression in embryos of maize viviparous mutants. Evidence forregulation of the Gibi gene by ABA. Plant Physiol. 92: 538-542.167Krochko, J.E., Coulter, K.M., Greenwood, J.S. and Bewley, J.D.1989. Comparison of storage proteins in zygotic and somaticembryos of alfalfa. Plant Physiol. 89: S—172.Krogstrup, P., Eriksen, E.N., Møller, J.D. and Roulund, H. 1988.Somatic embryogenesis in Sitka spruce (Picea sitchensis (Bong.)Carr.). Plant Cell Rep. 7: 594-597.Kumar, P.P., Joy IV, R.W. and Thorpe, TA. 1989. Ethylene andcarbon dioxide accumulation, and growth of cell suspensioncultures of Picea glauca (white spruce). J. Plant Physiol. 135:592—596.Kvaalen, H. and von Arnold, S. 1991. Effects of various partialpressures of oxygen and carbon dioxide on different stages ofsomatic embryogenesis in Picea abies. Plant Cell Tiss. Org. Cult.27: 49—57.Laemmli, U.K. 1970. Cleavage of structural proteins during theassembly of the head of bacteriophage T4. Nature 227: 680-685.Lamé, E. and David, A. 1990. Somatic embryogenesis in immatureembryos and protoplasts of Pinus caribaea. Plant Sci. 69: 215-224.Lelu, M.-A.P. and Bornman, C.H. 1990. Induction of somaticembryogenesis in excised cotyledons of Picea glauca and Piceamariana. Plant Physiol. Biochem. 28(6): 785-791.Lelu, M.—A.P., Boulay, M.P. and Borninan, C.H. 1990. Somaticembryogenesis in cotyledons of Picea abies is enhanced by anadventitious bud-inducing treatment. New Forests 4: 125-135.Lending, C.R. and Larkins, B.A. 1989. Changes in the zeincomposition of protein bodies during maize endosperm development.Plant Cell 1: 1011—1023.Litvay, J.D., Verma, D.C. and Johnson, M.A. 1985. Influence ofloblolly pine (Pinus taeda L.) culture medium and its componentson growth and somatic embryogenesis of the wild carrot (Daucuscarota L.). Plant Cell Rep. 4: 325—328.Litz, R.E. 1986. Effect of osmotic stress on somaticembryogenesis in Carica suspension cultures. J. Amer. Soc.Hortic. Sci. 111(6): 969—972.Lu, C.-Y. and Thorpe, T.A. 1987. Somatic embryogenesis andplantlet regeneration in cultured immature embryos of Piceaglauca. J. Plant Physiol. 128: 297-302.Mansfield, M.A. and Raikhel, N.y. 1990. Abscisic acid enhancesthe transcription of wheat-germ agglutinin mRNA without alteringits tissue-specific expression. Planta 180: 548—554.168Mayer, J.E., Hahne, G., Palme, K. and Schell, J. 1987. A simpleand general plant tissue extraction procedure for two—dimensionalgel electrophoresis. Plant Cell Rep. 6: 77-81.Meinke, D.W., Chen, J. and Beachy, R.N. 1981. Expression ofstorage protein genes during soybean seed development. Planta153: 130—139.Mia, A.J. and Durzan, D.J. 1974. Cytochemical and subcellularorganization od the shoot apical meristem of dry and germinatingjack pine embryos. Can. J. For. Res. 4: 39—54.Misra, S. and Green, M.J. 1990. Developmental gene expression inconifer embryogenesis and germination. I. Seed proteins andprotein body composition of mature embryo and the megagametophyteof white spruce (Picea glauca [Moench] Voss.) Plant Sci. 68: 163-173.Misra, S. and Green, M.J. 1991. Developmental gene expression inconifer embryogenesis and germination. II. Crystalloid proteinsynthesis in the developing embryo and megagametophyte of whitespruce (Picea glauca [Moench] Voss.) Plant Sci. 78: 61—71.Mo, L.H. and von Arnold, 5. 1991. Origin and development ofembryogenic cultures from seedlings of Norway spruce (Piceaabies). J. Plant Physiol. 138: 223—230.MUntz, K. 1989. Intracellular protein sorting and the formationof protein reserves in storage tissue cells of plant seeds.Biochem. Physiol. Pflanzen 185: 315-335.Murashige, T. and Skoog, F. 1962. A revised medium for rapidgrowth and bioassays with tobacco tissue cultures. Physiol.Plant. 15: 473—497.Murphy, D. J., Cummins, I. and Ryan, A. J. 1989. Immunocytochemicaland biochemical study of the biosynthesis and mobilisation of themajor seed proteins of Brassica napus. Plant Physiol. Biochem.27(5): 647—657.Nadel, B.L., Altman, A. and Ziv, M. 1989. Regulation of somaticembryogenesis in celery cell suspensions. 1. Promoting effects ofmannitol on somatic embryo development. Plant Cell Tiss. Org.Cult. 18: 181—189.Nagmani, R. and Bonga, J.M. 1985. Embryogenesis in subculturedcallus of Larix decidua. Can. J. For. Res. 15: 1088-1091.Nielsen, N.C., Dickinson, C.D., Cho, T.J., Thanh, V.J., Scallon,B.J., Fischer, R.L., Sims, T.L., Drews, G.N. and Goldberg, R.B.1989. Characterization of the glycinin gene family in soybean.Plant Cell 1: 313—328.169Nørgaard, J.V. and Krogstrup, P. 1991. Cytokinin induced somaticembryogenesis from immature embryos of Abies nordmanniana Lk.Plant Cell Rep. 9: 509-513.Osborne, T.B. 1924. The vegetable proteins. In Monographs inBiochemistry. Edited by R.H. Plummer and F.G. Hopkins.Longmans, Green and Company, London, 154 p.Owens, J.N. and Molder, M. 1984. The reproductive cycle ofinterior spruce. Province of British Columbia, InformationServices Branch, Ministry of Forests, Victoria, 31 p.Parry, A.D. and Horgan, R. 1991. Carotenoids and abscisic acid(ABA) biosynthesis in higher plants. Physiol. Plant. 82: 320-326.Parrott, W.A., Dryden, G., Vogt, S., Hidebrand, D.F., Collins,G.B. and Williams, E.G. 1988. Optimization of somaticembryogenesis and embryo germination in soybean. In Vitro Cell.Dev. Biol. 24(8): 817—820.Pence, V.C. 1991. Abscisic acid in developing zygotic embryos ofTheobroma cacao. Plant Physiol. 95: 1291-1293.Pramanik, S.J., Krochko, J.E. and Bewley, J.D. 1991. Storageprotein mRNAs are repressed in early stages of alfalfa somaticembryo development. ISPMB 3rd International Congress, Tucson,AZ, Abstract no. 452.Raynal, M., Aspart, L., This, P. and Delseny, N. 1987.Biosynthesis of cruciferin polypeptides in inuuature radish seeds.Plant Physiol. Biochem. 25(4): 439-444.Redenbaugh, K., Paasch, B.D., Nichol, J.W., Kossler, M.E., Viss,P.R. and Walker, K.A. 1986. Somatic seeds: Encapsulation ofasexual plant embryos. BioTechnology 4: 797—801.Rivin, C.J. and Grudt, T. 1991. Abscisic acid and thedevelopmental regulation of embryo storage proteins in maize.Plant Physiol. 95: 358—365.Roberts, D.R. 1991. Abscisic acid and mannitol promote earlydevelopment, maturation and storage protein accumulation insomatic embryos of interior spruce. Physiol. Plant. 83: 247-254.Roberts, D.R., Flinn, B.S., Webb, D.T., Webster, F.B. and Sutton,B.C.S. 1989. Characterization of immature embryos of interiorspruce by SDS-PAGE and microscopy in relation to their competencefor somatic embryogenesis. Plant Cell Rep. 8: 285-288.Roberts, D.R., Flinn, B.S., Webb, D.T., Webster, F.W. and Sutton,B.C.S. l990a. Abscisic acid and indole-3-butyric acid regulationof maturation and accumulation of storage proteins in somaticembryos of interior spruce. Physiol. Plant. 78: 355-360.170Roberts, D.R., Sutton. B.C.S. and Flinn, B.S. 1990b. Synchronousand high frequency germination of interior spruce somatic embryosfollowing partial drying at high relative humidity. Can. J. Bot.68: 1086—1090.Roberts, D.R., Lazaroff, W.R. and Webster, F.B. 1991. Interactionbetween maturation and high relative humidity treatments andtheir effects on germination of Sitka spruce somatic embryos. J.Plant Physiol. 138: 1-6.Rock, C.D. and Zeevaart, J.A.D. 1991. The aba mutant ofArabidopsis thaliana is impaired in epoxy-carotenoidbiosynthesis. Proc. Natl. Acad. Sci. USA 88: 7496-7499.Sambrook, J., Fritsch, E.F. and Maniatis, T. 1989. Molecularcloning, 2nd edition. Cold Spring Harbor Laboratory Press, ColdSpring Harbor.Schopfer, P. and Plachy, C. 1984. Control of seed germination byabscisic acid. II. Effect on embryo water uptake in Brassicanapus L. Plant Physiol. 76: 155-160.Schroeder, H.E. 1984. Effects of applied growth regulators on podgrowth and seed protein composition in Pisum sativum L. J. Exp.Bot. 35: 813—821.Schuller, A., Reuther, G. and Geier, T. 1989. Somaticembryogenesis from seed explants of Abies alba. Plant Cell Tiss.Org. Cult. 17: 53-58.Senaratna, T., McKersie, B.D. and Bowley, S.R. 1990. Artificialseeds of alfalfa (Medicago sativa L.). Induction of desiccationtolerance in somatic embryos. In Vitro Cell. Dev. Biol. 26: 85-90.Shoemaker, R.C., Christofferson, S.E. and Galbraith, D.W. 1987.Storage protein accumulation patterns in somatic embryos ofcotton (Gossypium hirsutum L.) Plant Cell Rep. 6: 12-15.Shotwell, M.A. and Larkins, B.A. 1989. The biochemistry andmolecular biology of seed storage proteins. In The Biochemistryof Plants, Vol. 15, Molecular Biology. Edited by A. Marcus.Academic Press, New York, pp 297-345.Shuttuck-Eidens, D.M. and Beachy, R.N. 1985. Degradation of 13-conglycinin in early stages of soybean embryogenesis. PlantPhysiol. 78: 895—898.Skriver, K. and Mundy, J. 1990. Gene expression in response toabscisic acid and osmotic stress. Plant Cell 2: 503—512.Smith, D.E. and Fisher, P.A. 1984. Identification, developmentalregulation and response to heat shock of two antigenicallyrelated forms of a major nuclear envelope protein in Drosophila171embryos: Application of an improved method for affinitypurification of antibodies using polypeptides immobilized onnitrocellulose blots. J. Cell Biol. 99: 20-28.Smith, J.A. and Desborough, S.L. 1987. The endosperm seed proteinsolin: biochemical characterization, induction by ABA andspecies—specific subunits. Theor. Appl. Genet. 74: 739—745.Spencer, D., Chandler, P.M., Higgins, T.J.V., Inglis, A. andRubira, M. 1983. Sequence interrelationships between the subunitsof vicilin from pea seeds. Plant Mol. Biol. 2: 259—267.Spurr, A. 1969. A low-viscosity epoxy resin embedding medium forelectron microscopy. J. Ultrastruct. Res. 26: 31—43.Stabel, P., Eriksson, T. and Engström, P. 1990. Changes inprotein synthesis upon cytokinin-mediated adventitious budinduction and during seedling development in Norway spruce, Piceaabies. Plant Physiol. 92: 1174—1183.Stuart, D.A., Nelsen, J. and Nichol, J.W. 1988. Expression of 7Sand uS alfalfa seed storage proteins in somatic embryos. J.Plant Physiol. 132: 134—139.Tautorus, T.E., Attree, S.M., Fowke, L.C. and Dunstan, D.I. 1990.Somatic embryogenesis from immature and mature zygotic embryos,and embryo regeneration from protoplasts in black spruce (Piceamariana Mill.) Plant Sci. 67: 115—124.Taylor, D.C., Weber, N., Underhill, E.W., Pomeroy, M.K., Keller,W.A., Scowcroft, W.R., Wilen, R.W., Moloney, M.M. and Holbrook,L.A. 1990. Storage protein regulation and lipid accumulation inmicrospore embryos of Brassica napus L. Planta 181: 18—26.Tewes, A., Manteuffel, R., Adler, K., Weber, E. and Wobus, U.1991. Long—term cultures of barley synthesize and correctlydeposit seed storage proteins. Plant Cell Rep. 10: 467-470.Tremblay, F.M. 1990. Somatic embryogenesis and plantletregeneration from embryos isolated from stored seeds of Piceaglauca. Can. J. Bot. 68: 236-242.Tremblay, L. and Tremblay, F.M. 1991a. Carbohydrate requirementsfor the development of black spruce (Picea mariana (Mill.)B.S.P.) and red spruce (P. rubens Sarg.) somatic embryos. PlantCell Tiss. Org. Cult. 27: 95—103.Tremblay, L. and Tremblay, F.M. 1991b. Effects of gelling agents,ainmonium nitrate and light on the development of Picea mariana(Mill.) B.S.P. (black spruce) and Picea rubens Sarg. (red spruce)somatic embryos. Plant Sci. 77: 233—242.172Verhagen, S.A. and Wann, S.R. 1989. Norway spruce somaticembryogenesis: high-frequency initiation from light-culturedmature embryos. Plant Cell Tiss. Org. Cult. 16: 103-111.Verwoerd, T. C., Dekker, B.M.M. and HoekeiTta, A. 1989. A smallscale procedure for the rapid isolation of plant RNAs. NucleicAcids Res. 17: 2362.von Aderkas, P. and Bonga, J.M. 1988. Embryoid formation in Larixdecidua. Amer. J. Bot. 75: 690-700.von Aderkas, P., Klimaszewska, K. and Bonga, J.M. 1990. Diploidand haploid embryogenesis in Larix leptolepsis, L. decidua, andtheir reciprocal hybrids. Can. J. For. Res. 20: 9—14.von Arnold, S. 1987. Improved efficiency of somatic embryogenesisin mature embryos of Picea abies (L.) Karst. J. Plant Physiol.128: 233—244.von Arnold, S. and Eriksson, T. 1981. In vitro studies ofadventitious shoot formation in Pinus contorta. Can. J. Bot. 59:870—874.von Arnold, S. and Hakman, I. 1986. Effect of sucrose oninitiation of embryogenic callus cultures from mature zygoticembryos of Picea abies (L.) Karst. (Norway spruce). J. PlantPhysiol. 122: 261—265.von Arnold, S. and Hakman, I. 1988. Regulation of somatic embryodevelopment in Picea abies by abscisic acid (ABA). J. PlantPhysiol. 132: 164—169.von Arnold, S. and Woodward, S. 1988. Organogenesis andembryogenesis in mature zygotic embryos of Picea sitchensis. TreePhysiol. 4: 291—300.Walker, K. A. and Sato, S.J. 1981. Morphogenesis in callus tissueof Medi cago sativa: the role of ammonium ion in somaticembryogenesis. Plant Cell Tiss. Org. Cult. 1: 109—121.Walling, L., Drews, G.N. and Goldberg, R.B. 1986. Transcriptionaland post—transcriptional regulation of soybean seed protein mRNAlevels. Proc. Natl. Acad. Sci. USA 83: 2123—2127.Webb, D.T., Webster, F., Flinn, B.S., Roberts, D.R. and Ellis,D.D. 1989. Factors influencing the induction of embryogenic andcaulogenic callus from embryos of Picea glauca and P. engelmanhi.Can. J. For. Res. 19: 1301—1308.Webster, F.B., Roberts, D.R., Mclnnis, S.M. and Sutton, B.C.S.1990. Propagation of interior spruce by somatic embryogenesis.Can. J. For. Res. 20: 1759—1765.173Wetherell, D.F. and Dougall, D.K. 1976. Sources of nitrogensupporting growth and embryogenesis in cultured wild carrottissue. Physiol. Plant. 37: 97-103.Wilen, R.W., Mandel, R.M., Pharis, R.P., Holbrook, L.A. andMoloney, M.M. 1990. Effects of abscisic acid and high osmoticumon storage protein gene expression in microspore embryos ofBrassica napus. Plant Physiol. 94: 875-881.Williamson, J.D., Quatrano, R.S. and Cuming, A.C. 1985. Empolypeptide and its messenger RNA levels are modulated by ABAduring embryogenesis in wheat. Eur. J. Biochem. 152: 501-507.Williamson, J.D. and Quatrano, R.S. 1988. ABA-regulation of twoclasses of embryo—specific sequences in mature wheat embryos.Plant Physiol. 86: 208—215.Wray, W., Boulikas, T., Wray, V.P. and Hancock, R. 1981. Silverstaining of proteins in polyacrylamide gels. Anal. Biochem. 118:197—203.Xu, N., Coulter, K.M. and Bewley, J.D. 1990. Abscisic acid andosmoticum prevent germination of developing alfalfa embryos, butonly osmoticum maintains the synthesis of developmental proteins.Planta 182: 382—390.Yang, L.J., Barratt, D.H.P., Domoney, C., Hedley, C.L. and Wang,T.L. 1990. An analysis of seed development in Pisum sativum. X.Expression of storage protein genes in cultured embryos. J. Exp.Bot. 41(224): 283—288.Zeevaart, J.A.D. and Creelman, R.A. 1988. Metabolism andphysiology of abscisic acid. Ann. Rev. Plant Physiol. Plant Mol.Biol. 39: 439—473.

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086520/manifest

Comment

Related Items