Planta (1996)199:459 466

P l a n t a (c] Springer-Verlag 1996

Expression of abundant mRNAs during somatic embryogenesis of white spruce [Picea glauca (Moench) Voss-I Jin-Zhuo Dong, David I. Dunstan Plant BiotechnologyInstitute,National Research Council of Canada, 110 GymnasiumPlace, Saskatoon, SaskatchewanS7N 0W9, Canada Received: 16 October 1995/Accepted: 8 December 1995 Abstract. Embryogenic tissues of white spruce [Picea glauca (Moench) Voss] remain in an early developmental

Introduction

stage while cultured on 2,4-dichlorophenoxyacetic acid and N6-benzyladenine, but develop to cotyledonary embryos when these phytohormones are replaced by abscisic acid. Twenty-eight cDNAs were isolated from cotyledonary embryos by differential screening against immature embryo and non-embryonic tissues. Temporal expression patterns of these cDNAs during ABA-stimulated somatic embryo development were observed. This showed that clones could be allocated to various groups, including embryo-abundant, embryo-maturation-induced, and those whose expression was modulated during embryo development, germination or in non-embryogenic tissues. Expression corresponding to these cDNA clones showed that there were various responses to exogenous ABA or polyethylene glycol during a period of 48 h. Analyses of DNA and predicted amino acid sequence revealed that 12 of 28 cDNA clones had no known homologues, while others were predicted to encode different late-embryogenesis-abundant proteins, early methionine-labelled proteins, storage proteins, heat-shock proteins, glycine-rich cell wall protein, metallothionein-like protein and some other metabolic enzymes.

Plant embryogenesis has been divided into three phases based on morphological changes and physiological data, these are embryo differentiation, embryo maturation and seed dehydration (Goldberg et al. 1989; Thomas 1993). Recently, increased attention has been given to isolation of plant embryo-specific or embryogenesis-related genes, identification of their regulatory elements and observation of their responses to various environmental stresses and chemicals during embryo development. To investigate gene expression during cotton zygotic embryogenesis, Hughes and Galau (1989) randomly selected 47 cDNA clones in a cDNA library made from cotyledonary mRNAs. These cDNA clones could be grouped into different temporally abundant components, i.e. associated with the cotyledon stage, endogenous abscisic acid (ABA), maturation, ovule abscission, or germination. Among them eight late-embryogenesis-abundant (LEA) proteins and their genes were characterized (Baker et al. 1988; Galau et al. 1993). Recently, the accumulation kinetics of 18 cDNAs were observed during Arabidopsis seed development. These included four storage protein genes, five L E A genes and three ABAinducible genes (Parcy et al. 1994). The mRNAs belonged to distinct groups classified according to their temporal expression. An important class of genes, the L E A genes, has been extensively investigated in the last few years to elucidate plant embryogenesis. The L E A genes are identified by a pronounced increase in their expression in embryos during late embryogenesis and their quiescence during subsequent germination stages. Products of the L E A gene are believed to prevent embryos from damage from desiccation and from precocious germination during seed development. Since the early methionine-labelled protein gene (Era) (Litts et al. 1987) and cotton L E A genes (Baker et al, 1988) were cloned, a number of their homologues have been identified in various angiosperm plants using heterologous DNA probes (Dure 1993). The identification of embryo-specific and embryo-stage-specific genes, especially those involved in early events of embryo

Key words: Abscisic acid - Embryo abundant gene Gene expression - Polyethylene glycol - Picea - Somatic embryogenesis

Abbreviations: 2,4-D = 2,4-dichlorophenoxyacetic acid; ABA = abscisic acid; BA = N6-benzyladenine; cDNA = complementary deoxyribonucleic acid; Em = early methionine-labelled; HSP = heat-shock protein; LEA = late embryogenesisabundant; PEG = polyethyleneglycol Correspondence to: D.I. Dunstan; Tel: (1) 306 975 5283; FAX: (1) 306 975 4839; E-mail: [email protected]

460

J.-Z. Dong and D.I. Dunstan: Gene expression during somatic embryogenesis

development, remains a challenge for understanding the molecular mechanisms of plant e m b r y o development. Somatic or asexual embryogenesis is the process by which somatic cells develop into embryos t h r o u g h m o r p h o l o g i c a l stages similar to those of their zygotic counterparts. Because of experimental accessibility, somatic embryogenesis and in-vitro culture of zygotic e m b r y o s are being used to investigate factors which regulate and affect e m b r y o development. Embryogenic suspension cultures can be used to provide large a m o u n t s of somatic e m b r y o s at particular developmental stages, and the environment of the somatic e m b r y o can be altered to study effects on gene expression. F o r example, carrot somatic embryogenesis is a well characterized system and considered to be representative of angiosperms in the molecular analysis of somatic embryogenesis (Zimmerman, 1993). C a r r o t somatic e m b r y o development occurs u p o n removal of 2,4-dichlorophenoxyacetic acid (2,4-D) from suspension cultures, which results in the sequential formation of structures morphologically analogous to globular, heart and t o r p e d o stages in zygotic e m b r y o development. The genes DC3 (Seffens et al. 1990), DC8 (Franz et al. 1989) and EMB1 (Wurtele et al. 1993) have been cloned from e m b r y o n i c tissues and have been shown to be somatic embryogenesis-associated. The temporal expression p r o g r a m of e m b r y o - a b u n d a n t genes during somatic embryogenesis has not been extensively described to date. In conifers, several storage-protein genes have been reported in white spruce (Newton et al. 1992; Leal and Misra 1993) and their expression during zygotic e m b r y o development has been observed (Newton et al. 1992; Leal et al. 1995). Genes typical in angiosperm e m b r y o development, such as L E A and Em, have not been reported for conifers. Somatic embryogenesis has attracted attention in conifer biotechnology because it provides a useful system for p r o d u c i n g tree clones and an avenue for genetic m a n i p u l a t i o n (Bommineni et al. 1993). Conifers, one of the m a j o r g y m n o s p e r m groups, show differences from angiosperm plants in m o r p h o l o g y during seed development [-for a description of conifer seed development consult Singh and Owens (1981)]. Also, unlike most angiosperm somatic e m b r y o cultures, conifer somatic e m b r y o develo p m e n t usually has to be stimulated by exogenous ABA. Developmental stages that are observed during conifer zygotic embryogenesis can also be observed during somatic embryogenesis (Dunstan et al. 1995). Unlike angiosperm embryogenesis, there is little that is k n o w n about gene regulation and expression patterns during conifer e m b r y o development. To contribute to an understanding of the molecular events of conifer e m b r y o development, c D N A probes from embryogenic masses (Stage 1) and n o n - e m b r y o g e n i c tissues were used to differentially screen a c D N A library m a d e from c o t y l e d o n a r y (Stage 3) embryos. Twentyeight c D N A s have been analyzed for insert D N A sequences and predicted amino-acid sequences, developmental expression, and responses to ABA and the osmotic agent polyethylene glycol (PEG). With the exception of several h o m o l o g u e s of conifer storage-protein genes, m a n y of the c D N A s have not been previously reported.

Materials

and methods

Embryogenic tissues and somatic embryo maturation. Description of the embryogenic culture of white spruce [Picea glauca (Moench)

Voss] containing immature somatic embryos (Stage 1) and of its maintenance in liquid suspension, have been previously reported (Bommineni et al. 1993). Induction of somatic embryo maturation was conducted as described by Bommineni et al. (1993). Somatic embryos at different morphological stages were harvested relative to time elapsed following culture under maturation condition, i.e. at 0 (inoculation time), 1, 2, 3, 5 and 7 d (Stage 1), at 10 and 14 d (Stage 2), at 28 d (Stage 3a), and at 42 d (Stage 3b). Mature embryos (Stage 3b) were transferred to phytohormone-free 0.5 x GMD (Mohammed et al. 1986) medium for germination, and cultured for another six weeks to obtain plantlets (Stage 4). Rooted plantlets were transplanted to soil in pots and grown for two years in a greenhouse. Needle tissues were harvested from these plants. Treatment with ABA and PEG. To determine the influence of ABA and PEG on expression of genes associated with somatic embryo development, embryogenic tissues were resuspended in half-strength LP medium (von Arnold and Eriksson 1981) containing either 48 ~tM ( _+)-ABA or 5% (w/v) PEG and the cultures were agitated at 150 rpm on a gyratory shaker at 25 ~ in the dark. The tissues were sampled at 0, 1, 3, 6, 12, 24 and 48 h after culture. Racemic ABA [( _+)-ABA] was dissolved in ethanol, and then diluted by halfstrefigth LP medium with a final concentration of 10% (v/v) ethanol. The ABA stock solution w a s 1 0 - 3 M. Extraction of RNA. Total RNA was extracted following the method of Dong and Dunstan (1996) and was used in Northern hybridization. Polyadenylated mRNA was obtained from total RNA using biotinylated oligo (dT) primers bound to paramagnetic particles (Promega, Ottawa, Ontario, Canada) and used for cDNA synthesis as follows. Construction ofa cDNA library. Mature somatic embryos (Stage 3b) were used as a tissue source of the mRNA for construction of the cDNA library. First-strand cDNAs were synthesized from 5 gg poly(A)+RNA by Moloney-Murine leukemia virus reverse transcriptase (M-MuLVRT) in the presence of a primer consisting of an oligo (dT) stretch and an Xho I enzyme restriction sequence. The nucleotide 5-methyl dCTP was used instead of dCTP in the reactions. After the doubled-strand cDNA was synthesized, EcoR I adapters were added. EcoR I-Xho I cDNA fragments were inserted in Uni-ZAP XR vectors (Stratagene, La Jolla, Cal., USA). The cDNA library was packaged and plated using Escheriehia coil cell line XL1-Blue MRF' with a titre of 9.31 x 1 0 6 plaque-forming units per gg cDNA. DifJerential screening of the cDNA library. Polyadenylated RNA from suspension-cultured Stage 1 embryo tissues or Stage 4 plantlets as well as from Stage 3b embryo tissues were used as the templates for synthesizing first-strand cDNA probes to screen the cDNA library just described. Differentially hybridized plaques were picked off and the inserts were excised in vitro into plasmid pBluescript SK( - ) (Stratagene, La Jolla, Cal., USA) for further analysis. Sequencing and analysis of DNA. Plasmid DNA from pBluescript

SK( - ) containing an insert of interest was prepared by an alkali lysis method. Nucleotide sequences were determined by the dideoxy nucleotide chain-termination method (Sanger et al. 1977). Sequencing started at both ends of the insert using T3 and T7 primers in pBluescript SK( - ). Complete cDNA insert sequences were obtained by continuous sequencing with new synthesized primers. Sequence analyses used the EUGENE program (Lark Sequencing Technologies Inc., Houston, Tex., USA) and similarity analysis to known sequences in GenBank used the BLAST E-mail server program (National Centre for Biotechnology Information, Bethesda, Md., USA).

J.-Z. Dong and D.I. Dunstan: Gene expression during somatic embryogenesis

Northern blot analysis. Total RNA was denatured at 70~ for 10 min before fractionation in a 1.2% (w/v)agarose gel following the method of Pell6 and Murphy (1993) and blotted on charged nylon § membrane (Amersham, Oakville, Ontario, Canada). Hybridization was at 68 ~ in hybridization solution (Stratagene) in the presence of a DNA probe containing a [32p]dCTP-labelled insert (specific activities were approximately 108 cpm/lag DNA) using the random-primed DNA labelling method (Boehringer Mannheim, Laval, Quebec, Canada). After washing the membrane once with 2 x SSC (1 x SSC = 0.15 M NaC1, 0.015 M Na3-citrate, pH 7) and 0.1% (w/v) SDS for 15 min at 23 ~ and once with 0.1 x SSC and 0.1% (w/v) SDS for 30 min at 60~ the membrane was exposed to X-ray film at - 70 ~ for 2 or 3 d (Kodak Diagnostic Film; Eastman Kodak, Rochester, N.Y., USA).

Results

Somatic embryo development of white spruce. Stage 1 embryos in suspension culture consisted of an embryonic region with densely cytoplasmic cells subtended by a translucent suspensor with long, highly vacuolate cells, equivalent to the embryonal mass in zygotic embryogenesis (Singh and Owens 1981). For further somatic embryo development, embryogenic tissues were pretreated for one week in medium without 2,4-D before placement onto ABA- and PEG-containing maturation medium. After 10-14 d on maturation medium, the distal regions of the embryos became prominent and opaque with a smooth and glossy surface subtended by a suspensor, equivalent to the globular embryo stages in zygotic embryogenesis (Singh and Owens 1981). Such somatic embryos were cream in color and are termed Stage 2 (10-14 d). Stage 3 embryos were cotyledonary, and were visible after another one to two weeks in culture. In the somatic system these are termed Stage 3a embryos (21 28 d), defined as cream-coloured with primordial cotyledons located below the prominent central shoot apex, and Stage 3b embryos ( 3 5 4 2 d) which were cream to pale green in color, and had cotyledons overtopping the shoot apex. These stages have the equivalent counterparts in zygotic embryogenesis (Singh and Owens 1981). Green plantlets developed from mature embryos when placed on germination medium, and were characterized by possession of elongating epicotyl, cotyledons, hypocotyl, and radicle.

Identification of cDNAs associated with somatic embryo development. Approximately 7.5 x l04 plaques were screened and 146 individual plaques which showed obvious differences in expression were analysed further. The cDNA inserts of these clones were partially sequenced at the insert 5' end using T3 primer. Based on homology analysis among clones, identical or highly similar clones were grouped. Fifty-seven cDNA clones were used as probes to determine the developmental expression patterns of corresponding genes during somatic embryo development and subsequent germination, resulting in 45 individual cDNAs being selected for further comparison. Full-length insert DNA sequences were obtained for 28 of these cDNAs (Table 1). Among them, 15 represent fulllength mRNAs while 11 represent partial mRNA sequences. In addition, clones PgEMB4 and 28, respectively

461

1151 and 1254 nucleotides in length, had no recognizable open-reading frames. The cDNA sequences were compared with those in GenBank with the result that 12 of the 28 sequences failed to find any homologues (Table 1), while others were predicted to encode different LEA, Em, storage and heat-shock proteins (HSPs), glycine-rich cell wall protein, metallothionein-like protein, and some other metabolic enzymes (Table 1).

Developmental

expression. Expression profiles of 27 cDNA clones during somatic embryogenesis and subsequent germination are shown in Fig. 1. In terms of mRNA accumulation patterns, the corresponding cDNAs could be categorized into three major groups as follows. 1. Embryo-abundant. Gene expression of these cDNA clones occurred in Stage 1 embryo masses during suspension culture (0 d) before exposure to maturation conditions. The mRNAs occurred in somatic embryo developmental Stages 1 through to 3b, but were not detectable in germinants or potted plants. Four clones belonged to this group (Table 1, Fig. 1). The PgEMB1 gene showed expression through Stages 1 to 3b embryos. For another two clones (PgEMB3 and 4), mRNA accumulation rose once suspension-cultured Stage 1 embryos were placed onto maturation medium with higher abundance detected at Stages 2 to 3b. Transcripts related to PgEMB5 were at very low abundance in suspension-cultured tissues (0 d), but increased in subsequent embryo development. 2. Maturation-induced. Accumulation of transcripts corresponding to these cDNAs was not detectable in suspension-cultured embryonic tissues before exposure to maturation induction (0 d). Expression appeared to be induced following exposure to maturation conditions, and no gene expression was detected in needle tissues of transplanted plants. Two subgroups (2a and 2b) were derived based on whether or not expression was detectable in Stage 4 plantlets. 2a: mRNAs were not detectable in Stage 4. There were 13 cDNAs (PgEMB9-20 and 44) showing this pattern (Table 1). Transcripts related to seven clones (PgEMB9-15) began to accumulate once embryogenic tissues were placed under maturation conditions, and subsequently increased their abundance during development, with the maximum abundance during Stages 2 to 3b (Fig. 1). Gene expression patterns for clones PgEMB16-20 showed a trend toward expression only during Stages 2 to 3b (Fig. 1). PgEMB44, encoding an Em-like protein (Table 1) had 97% identity to its homologue PgEMB19 over their cDNA coding regions and 92% identity between the two predicted proteins. As detected by using insert cDNA probes, PgEMB44 showed a similar developmental expression pattern to PgEMB19 during spruce somatic embryogenesis (data not shown). 2b: This subgroup of cDNAs represented expression detectable in developing embryos as well as in Stage 4. Three clones (PgEMB21, 23 and 25) belonged to this subgroup (Table 1, Fig. 1). PgEMB21 showed low level expression through each stage. PgEMB23 showed uniform expression during development, with decreased expression in Stage 4. Transcripts related to PgEMB25 were detectable only during Stages 2 3b and at a very low level in Stage 4.

462

J.-Z. Dong and D.I. Dunstan: Gene expression during somatic embryogenesis

Table 1. cDNA clones associated with spruce somatic embryogenesis Clone name

Insert DNA length (bp)

1. Embryo-abundant PgEMB1 758 PgEMB3 741 PgEMB4 1151b PgEMB5 338"

Predicted amino acid length (aa)

Homologues if anyd

Homology score, %~

Reference for homologues

cDNA accession number in GenBank

158 152

HSP (18.2)

71 (67)

Oh et al. 1993

L47609 L47601 L47610 L47604

2. Maturation-induced 2a: Expressed only in embryonic tissues PgEMB9 568" PgEMB10 399a PgEMBll 588 92 PgEMB12 788 190 PgEMB13 508a PgEMB14 756 82 PgEMB15 551~ 100 PgEMB16 634" 101 PgEMB17 471a 98 PgEMB18 1609 451 PgEMB19 636 91 PgEMB20 1495 294 PgE M B44f 601 91 2b: Expressed in embryonic tissues and 9erminants PgEMB21 1535a 463 PgEMB23 898 175 PgEMB25 728 165

IAA-induced protein Group 3 LEA

36 ( 4 7 ) 47 (135)

Yamamoto et al. 1992 Baker et al. 1988

Tomato LE25 Seed protein

54 (61) 80 (96)

Cohen and Bray 1992 Jarvis et al. 1995

Legumin-like Vicilin-like Era-like

95 (63) 97 (444) 77 (85)

Leal and Misra 1 9 9 3 Newton et al. 1992 Litts et al. 1987

Em-like

86 (86)

Litts et al. 1987

HSP70-1ike

78 (402)

Winter et al. 1988

Albumin-like

82 (159)

Rice and Kamalay 1991

3. Modulated abundance 3a: Expressed in in-vitro-cultured embryos, germinants and potted plants PgEMB26 431a PgEMB28 1245b PgEMB30 413 60 MT-like

74 (46)

PgEMB31 PgEMB43 c

60 (45) 88 (53)

Ledger and Gardner 1994 Keller et al. 1988 van de Loo et al. 1995

62 (32) 91 ( 1 4 2 )

Galau et al. 1993 Karpinski et al. 1 9 9 2

454 405"

76 66

Cell wall protein Fatty acid desaturase

3b: Expressed only in in vitro-cultured embryos and ,qerminants PgEMB32 600 95 Cotton LEA5 PgEMB33 615a Dismutase PgEMB35 742 133

L47611 L47612 L47628 L42465 L47613 L47606 L47607 L47629 L47630 L47744 L42464 L47749 L47750 L47751 L47603 L47745

L47747 L47743 L47746 L47748

L47602 L47742 L47605

a Partial mRNA sequences b NO recognizable open-reading frames were found c This cDNA clone is undergoing further characterization, the DNA sequence has not yet been submitted to GenBank d Abbreviations in this column: HSP, heat-shock protein; LEA, late-embryogenesis-abundantprotein; Em, early-methionine-labeledprotein; MT, metallothionein-likeprotein e Homology score (%) means the similarity at the protein level, for the amino acid length indicated in parentheses, except for PgEMB33 in which the data are at the DNA level since no proper protein sequence could be predicted from it f This clone had 97% identity to its homologue PgEMB19 over their cDNA coding regions, and 92% identity between the two predicted proteins

3. M o d u l a t e d a b u n d a n c e . The last m a j o r g r o u p of c D N A s were those c o u n t e r p a r t s of m R N A s detectable in all d e v e l o p m e n t a l stages, i n c l u d i n g plants. T w o s u b g r o u p s were derived (3a a n d 3b) based on whether expression occurred in vitro a n d ex vitro, or only in vitro. 3a: G e n e expression c o r r e s p o n d i n g to five c D N A s (PgEMB26, 28, 30, 31 a n d 43) was detectable in all d e v e l o p m e n t a l stages assessed (Fig. 1). The p a t t e r n for P g E M B 2 6 showed high a b u n d a n c e d u r i n g Stage 3. T r a n scripts c o r r e s p o n d i n g to P g E M B 2 8 a n d 43 gradually increased after tissues were placed u n d e r m a t u r a t i o n con-

ditions, reached a m a x i m u m in Stage 3, a n d then decreased. Transcripts c o r r e s p o n d i n g to P g E M B 3 0 a n d 31 genes were a b u n d a n t from Stage 2 t h r o u g h Stage 4 a n d in potted plants. 3b: G e n e expression c o r r e s p o n d i n g to the three c D N A s in this s u b g r o u p (PgEMB32, 33 a n d 35) was of low a b u n d a n c e in Stage 4 plantlets cultured in vitro, b u t was n o t detectable in t r a n s p l a n t s g r o w n in pots. Transcripts c o r r e s p o n d i n g to P g E M B 3 2 increased d u r i n g e m b r y o development, reaching the greatest level at Stage 3 (Fig. 1). I n contrast, clone P g E M B 3 3 c o r r e s p o n d e d to

J.-Z. Dong and DT Dunstan: Gene expression during somatic embryogenesis

463

Embryo stage Days after culture

Pg m3 iw dn4

PgEMB10 PgEMBll

PgEMBI2 Ib.,EMB13 PgEMB14

PgEMBI5 PgEMB16 PgEMBI7

PgEMB18 Fig. 1. Northern blot analysis of gene expression programs of 27 somatic embryogenesis-associated cDNA clones during somatic embryo development, and subsequent embryo-derived plantlets (Pl) and mature needle tissues (N) in white spruce. Three micrograms total RNA was loaded in each lane. Days after culture: days after transfer to maturation conditions

a decreased expression during embryo development. Expression related to PgEMB35 was much lower in suspension-cultured embryogenic tissues and embryo-derived germinants, than in embryo tissues cultured under maturation conditions. Induction by A B A and PEG. Twelve of the 28 cDNA clones were found to be ABA-responsive (Fig. 2). Gene expression of 11 among them was induced or enhanced by exogenous ABA. In the clones showing ABA-enhanced expression, transcripts corresponding to five (PgEMB1, 3, 28, 32 and 43) gradually increased during 48 h exposure to ABA. PgEMB35 showed a transient increase in transcript levels after tissues were exposed to ABA. Five ABAresponsive cDNAs (PgEMB5, 12-14 and 23) could be

categorized as ABA-inducible, that is, their mRNAs were not detectable in embryogenic tissues in the absence of exogenous ABA. Gene expression corresponding to four cDNAs (PgEMB31, 32, 35 and 43) indicated PEG-responsiveness. Both ABA and P E G affected gene expression related to PgEMB32, 35 and 43. The PgEMB31 gene was only PEG-inducible after 48 h.

Discussion Somatic embryogenesis in vitro is a developmental pathway by which plant sporophytic and gametophytic cells undergo a sequence of morphological changes

464

J.-Z. Dong and D.I. Dunstan: Gene expression during somatic embryogenesis

Fig. 2. Northern blot analysis of temporal gene expression of ABA- and PEGresponsive cDNAs associated with somatic embryogenesisin embryogenic tissues of white spruce. Three micrograms total RNA was loaded in each lane

culminating in an asexual embryo with root and shoot poles, capable of further development into a plant. Morphological changes are similar to those observed in zygotic embryogenesis (Thomas 1993; Dunstan et al. 1995). During zygotic and somatic embryo maturation storage products such as proteins and lipids are deposited, and desiccation often occurs. Using zygotic embryogenesis in dicotyledonary plants and in spruce species as guides, the equivalent maturation phase of conifer somatic embryogenesis would be between Stages 2 and 3b. Conifer somatic embryo development between Stages 1 and 2 equates to the cell division and differentiation phase of zygotic embryogenesis (Goldberg et al. 1989). At the molecular level, zygotic embryo developmental stages are characterized by the accumulation of distinct sets of mRNAs and corresponding proteins in the embryos (Goldberg et al. 1989; Hughes and Galau 1989; Parcy et al. 1994). In dicotyledonary plants, storage proteins are characteristic markers of maturation stages, followed by the accumulation of various LEA proteins. Based on temporal gene expression profiles of cDNAs of white spruce during somatic embryo development and subsequent germination, three major groups can be differentiated as described in the results. The embryo-abundant cDNAs of Group 1 represent genes which express in embryo development from Stages 1 to 3b. Group 2

cDNAs represent genes associated with conifer somatic embryogenesis which were maturation-medium-inducible (Group 2a) and cotyledonary abundant (Group 2b). Group 3 cDNAs represent genes which were detected in all developmental embryo stages and also in plants. Gene expression programs for embryo germination might be established in the mature embryos, as observed in embryos of rapeseed (Finkelstein and Crouch 1986), sunflower (Goffner et al. 1990), wheat (Morris et al. 1990) or cotton (Galau et al. 1991). Genes showing differential expression, temporally or developmentally, could serve as useful molecular markers for monitoring somatic embryo development. In conifer somatic embryogenesis, ABA is critical for stimulation of embryo development (Dunstan et al. 1995) and PEG has been reported to substantively alter Stage 3 embryo quality (Attree et al. 1991). We found that ABA could influence expression of approximately 30% of the embryo-abundant cDNAs described in this report, while PEG affected 15%. Only three cDNAs (10%) were found to be affected by both ABA and PEG. From these results there appear to be different molecular pathways related to ABA and PEG during somatic embryo development, each of which is likely regulated under distinct expression progams. For some genes there is also a probable common regulatory pathway involving ABA and PEG. In addition, approximately 60% of embryo-abundant

J.-Z. Dong and D.I. Dunstan: Gene expression during somatic embryogenesis m R N A s were not obviously affected by A B A and P E G during the first 48 h of induction, suggesting involvement of other regulatory factors. A L E A - l i k e c D N A (PgEMB21) predicted to encode a G r o u p 3 L E A protein, was cloned from spruce somatic e m b r y o tissues, but its expression pattern was quite different from patterns observed in angiosperm zygotic embryogenesis. Transcripts started to accumulate immediately after Stage 1 e m b r y o s were exposed to ABA. Transcripts of similar L E A genes were accumulated only after m a t u r a t i o n during zygotic embryogenesis of cotton, wheat and Brassica (Baker et al. 1988; D u r e et al. 1989; C u r r y et al. 1991). A n o t h e r g r o u p of L E A proteins are the Em-like proteins (Dure et al. 1989). W h e a t E m transcription regulation elements have been analyzed, indicating that Em was ABA responsive (Marcotte et al. 1988). H o w ever, the spruce c D N A P g E M B 1 9 corresponding to the E m gene was not apparently ABA inducible or P E G inducible by 48 h. A n u m b e r of c D N A s described in this study also showed similar gene expression patterns to L E A genes, but did not share significant h o m o l o g y with any k n o w n L E A gene sequences. Heat-shock-protein genes have been shown to be developmentally regulated during embryogenesis of animals, insects and mammals, especially at early embryogenesis stages, as well as being responsive to heat-shock induction (Heikkila 1993). T w o c D N A s were identified in conifer somatic e m b r y o tissues which were predicted to encode HSP-like proteins, and which showed developmental regulation under n o r m a l culture conditions. Differences in their m R N A accumulation patterns indicate that they might have different roles. The functions of H S P s in plant embryogenesis have not been resolved to date. It is believed that H S P s have specific roles in plant developmental pathways as well as protection of embryos from heat-shock stress. The c D N A s described in this study were identified from cotyledonary embryos, and represent the most abundantly expressed. Their regulation is likely to occur downstream of genes more directly involved in initiating e m b r y o development programs. The expression patterns reported in this w o r k suggest that spruce somatic e m b r y o development involves a temporal sequence of expression of different sets of genes. A l t h o u g h there are obvious differences in e m b r y o m o r p h o l o g y and development between angiosperms and gymnosperms, this observation is similar to those m a d e with angiosperm zygotic embryogenesis (Goldberg et al. 1989; Hughes and G a l a u 1989; Parcy et al: 1994) and somatic embryogenesis ( T h o m a s 1993). However, there were also features of gene expression and A B A responsiveness of spruce e m b r y o a b u n d a n t c D N A s which differed from those observed in angiosperms. M o r e than half of the e m b r y o - a b u n d a n t spruce c D N A s have not been previously described. Expression patterns and sequence data related to the 28 clones will be useful in the study of plant embryogenesis, especially for g y m n o s p e r m plants. The authors thank Mr. Terry Bethune for his assistance, and Dr. Larry Pelcher, Mr. Barry Panchuk and Mr. Don Schwab for DNA sequencing and primer synthesis. This is National Research Council of Canada publication number 38929.

465

References Attree SM, Moore D, Sawhney VK, Fowke LC (1991) Enhanced maturation and desiccation tolerance of white spruce [Picea glauca (Moench) Voss] somatic embryos: Effects of a non-plasmolysing water stress and abscisic acid. Ann Bot 68:519 525 Baker J, Steele C, Dure III L (1988) Sequence and characterization of 6 lea proteins and their genes from cotton. Plant Mol Biol 11: 277-291 Bommineni VR, Chibbar RN, Dalta RSS, Tsang EWT (1993) Transformation of white spruce (Picea glauca) somatic embryos by microprojectile bombardment. Plant Cell Rep 13:17-23 Cohen A, Bray EA (1992) Nucleotide sequence of an ABA-induced tomato gene that is expressed in wilted vegetative organs and developing seeds. Plant Mol Biol 18:411413 Curry J, Morris CF, Walker-Simmons MK (1991) Sequence analysis ofa cDNA encoding a group 3 LEA mRNA inducible by ABA or dehydration stress in wheat. Plant Mol Biol 16:1073 1076 Dong J-Z, Dunstan DI (1996) A reliable method for extraction of RNA from various conifer tissues. Plant Cell Rep 15:516-521 Dunstan DI, Tautorus TE, Thorpe TA (1995) Somatic embryogenesis in woody plants. In: Thorpe TA (ed) In vitro embryogenesis in plants. Kluwer Academic Publishers, Dordrecht, Netherlands, pp 473-539 Dure L (1993) The lea proteins of higher plants. In: Verma DPS (ed) Controls of plant gene expression. CRC Press, pp 325-335 Dure L, Crouch M, Harada J, Ho T-H D, Mundy J, Quatrano R, Thomas T, Sung ZR (1989) Common amino acid sequence domains among the LEA proteins of higher plants. Plant Mol Biol 12:475~486 Finkelstein RR, Crouch ML (1986) Rapeseed embryo development in culture on high osmoticum is similar to that in seeds. Plant Physiol 81:907-912 Franz G, Hatzopoulos P, Jones TJ, Krauss M, Sung ZR (1989) Molecular and genetic analysis of an embryonic gene, DC8, from Daucus carota L. Mol Gen Genet 218:143-151 Galau GA, Jacobsen KS, Hughes DW (1991) The controls of late dicot embryogenesis and early germination. Physiol Plant 81: 280-288 Galau GA, Wang H T-C, Hughes DW (1993) Cotton lea5 and leal4 encode atypical late embryogenesis-abundant proteins. Plant Physiol 101:695 696 Goffner D, This P, Delseny M (1990) Effects of abscisic acid and osmotica on helianthinin gene expression in sunflower cotyledons in vitro. Plant Sci 66:211 219 Goldberg RB, Barker SJ, Perez-Grau L (1989) Regulation of gene expression during plant embryogenesis. Cell 56:149-160 Heikkila JJ (1993) Heat shock gene expression and development. I. An overview of fungal, plant, and poikilothermic animal developmental systems. Dev Genet 14: ! 5 Hughes DW, Galau GA (1989) Temporally modular gene expression during cotyledon development. Gene Dev 3:358 369 Jarvis S, Taylor MA, MacLeod MR, Davies HV (1995) Cloning and characterization of 3 cDNA clones of genes that are differentially expressed during dormancy-breakage .in seeds of Douglas fir (Pseudotsuga menziesii). Plant Mol Biol, in press Karpinski S, Wingsle G, Olsson O, Haellgren JE (1992) Characterization of cDNAs encoding CuZn-superoxide dismutases in Scots pine. Plant Mol Biol 18:545-555 Keller B, Sauer N, Lamb CJ (1988) Glycine-rich cell wall proteins in bean: gene structure and association of the proteins with the cascular system. EMBO J 7:3625 3633 Leal I, Misra S (1993) Molecular cloning and characterization of a legumin-like storage protein cDNA of Douglas fir seeds. Plant Mol Biol 21:709-715 Leal I, Misra S, Attree SM, Fowke LC (1995) Effect of abscisic acid, osmoticum and desiccation on 11S storage protein gene expression in somatic embryos of white spruce. Plant Sci 106: 121 128 Ledger SE, Gardner RC (1994) Cloning and characterization of five cDNAs for genes differentially expressed during fruit

466

J.-Z. Dong and D.I. Dunstan: Gene expression during somatic embryogenesis

development of kiwifruit (Actinidia deIiciosa vat. deliciosa). Plant Mol Biol 25:877 886 Litts JC, Colwell GW, Chakerian RL, Quatrano RS (1987) The nucleotide sequence of a cDNA clone encoding the wheat Em protein. Nucleic Acids Res 15:3607-3618 Marcotte WR, Bayley CC, Quatrano RS (1988) Regulation of wheat promoter by abscisic acid in rice protoplasts. Nature 335: 454~457 Mohammed GH, Dunstan DI, Thorpe TA (1986) Influence of nutrient medium upon shoot initiation on vegetative explants from 15- to 18-year-old Picea glauca. NZ J For Sci 16:297 305 Morris PC, Kumar A, Bowles D J, Cuming AC (1990) Osmotic stress and abscisic acid induce expression of wheat Em genes. Eur J Biochem 190:625-630 Newton CH, Flinn BS, Sutton BC (1992) Vicilin-like seed storage proteins in the gymnosperm interior spruce (Picea glauca/engelmannii). Plant Mol Biol 20:315-322 Oh S, Iwahori A, Kato S (1993) Human cDNA encoding DnaJ protein homologue. Biochim Biophys Acta 1174:114-116 Parcy F, Valon C, Raynal M, Gaubier-Comella P, Belseny M, Giraudat J (1994) Regulation of gene expression programs during Arabidopsis seed development: roles of the ABI3 locus and of endogenous abscisic acid. Plant Cell 6:1567 1582 Pell6 R, Murphy NB (1993) Northern hybridization: rapid and simple electrophoretic condition. Nucleic Acids Res 21:2783-2784 Rice J, Kamalay JC (1991) A family of pine seed proteins related to 2S albumins. GenBank accession number X62435

Sanger F, Nicklen S, Coulson AR (1977) DNA sequencing with chain terminating inhibitors. Proc Natl Acad Sci USA 74: 5463 5467 Seffens WS, Almoguera C, Wilde HD, Vonder RA, Thomas TL (1990) Molecular analysis of a phylogenetically conserved carrot gene: developmental and environmental regulation. Dev Genet 11:65-76 Singh H, Owens JH (1981) Sexual reproduction of Engelmann spruce (Picea engelmannii). Can J Bot 59:793-810 Thomas TL (1993) Gene expression during plant embryogenesis and germination: an overview. Plant Cell 5:1401 1410 van de Loo FJ, Broun P, Turner S, Somerville CR (1995) An oleate 12-hydroxylase from Ricinus communis L. is a fatty acyl desaturase homolog. Proc Natl Acad Sci USA 92:6743-6747 von Arnold S, Eriksson T (1981) In vitro studies of adventitious shoot formation in Pinus contorta. Can J Bot 59:870-874 Winter J, Wright R, Duck N, Gasser C, Fraley R, Shah D (1988) The inhibition of petunia hsp 70 mRNA processing during CdCI2 stress. Mol Gen Genet 211:315-319 Wurtele ES, Wang H, Durgerian A, Nikolau BJ, Ulrich TH (1993) Characterization of a gene that is expressed early in somatic embryogenesis of Daucus carota. Plant Physiol 102:303-312 Yamamoto KT, Mori H, Imaseki H (1992) Novel mRNA sequences induced by indole-3-acetic acid in sections of elongating hypocotyls ofmung bean (Vigna radiata). Plant Cell Physio133:13 20 Zimmerman JL (1993) Somatic embryogenesis: A model for early development in higher plants. Plant Cell 5:1411-1423