PlantCell Reports
Plant Cell Reports (1986) 5:377-380
© Springer-Verlag1986
Stimulation of somatic embryogenesis and plant regeneration from anther culture of Vitis vinifera cv. Cabernet-Sauvignon M. CI. Mauro, C. Nef, and J. Fallot Ecole Nationale Supdrieure Agronomique de Toulouse, 145 Avenue de Muret, F-31076 Toulouse C6dex, France Received March 10, 1986 / Revised version received July 25, 1986 Communicated by A. M. Boudet
ABSTRACT Somatic embryogenesis and subsequent diploid plants have been obtained from anthers of Vitis vinifera Cabernet-Sauvignon, a cultivar so far considered as recalcitrant to in vitro regeneration. A n t h e ~ enclosing microsporcs near the first pollen mitosis were found to be the most responsive. However, from a practical point of view anther length proved to be an easier criterium for determining the optimal physiological anther stage. Calli derived from the anther sJ~m~tic tissues pr~duccd cmbryoids o~ly wh~Ll ~ultured on a medium supplemented with casein hydrolysate. Glutamine and adenine were found to stimulate this embryoid production. Evidence is presented that early removal of cotyledons increases the frequency of normal development of embryoids into plantlets. Abbreviations : MS : Murashige and Skoog medium (]962) 2,4-D : 2,4-dichlorophenoxyacetic acid NAA : ]-naphtaleneacetic acid BA : 6-benzylaminopurine. INTRODUCTION Some commercially important grapevine cultivars
(Yitis vinifera) present cultural defects leading to economic losses, in vitro techniques appear to be the only tool available for the genetic improvement of these varieties grown in vineyards where hybridization processes are not allowed. Regeneration of plantlets has been obtained by cultivation of nucellus in Yitis vinifera (MULLINS and SRINIVASAN 1976, SRINIVASAN and MULLINS ]980) and by cultivation of vegetative tissues essentially in interspecific hybrid grapevines (FAVRE 1977, KRUL and WORLEY 1977, HIRABAYASHI 1985). Embryo~ and plantlet production from cultured anthers of Yiti~ long reported to be unsuccessful (MULLINS 1971, GRESSHOFF and DOY ]974) has been achieved more recently (RAJASEKARAN and MULLINS 1979, BOUQUET et al. 1982, HIRABAYASHI 1982). However, the cultivar Cabernet-Sauvignon, used all over the world for the production of high quality red wines, appears to be recalcitrant to anther ~culture. Various physiological factors have been found to enhance the embryoid formation of numerous plants. For in vitro androgenesis through anther culture, the developmental stage of the anther seems to be important (SUNDERLAND et al. 1977, SIBI et al. 1979, DUNWELL 1986). Additionally, the presence in the culture medium of reduced nitrogen such as ammonium ion and certain amino acids was found to stimulate in vitro
Offprint requests to: M. C1. Mauro
embryogenesis of carrot in particular (HALPERIN and WETHERELL 1965, WETHERELL and DOUGALL 1976, KAMADA and HARADA ]979). The purpose of this paper is therfore to examine the influence of cultural conditions, such as anther lenght and the addition to the culture medium of various nitrogen compounds, in order to increase callus and embryoid yield of Cabernet-Sauvignon. MATERIALS and METHODS The methods for anther culture and induction of somatic embryos were adapted from BOUQUET et al. (1982). Cool-stored (4°C) cuttings of 3 clones (n ° ]5, 1565 and 337) of Cabernet-Sauvignon, provided by Mr Pouget (INRA Bordeaux) were grown in greenhouse according to the method of MULLINS (]966). Inflorescences were collected and chilled for 72 hours at 4°C. The flowers were then surface-sterilized (10 min) with a filtered 7 % (w/v) solution of calcium hypochlorite contained Tween 20 ( 0 . 1 % v/v) and rinsed several tim~ with sterile distilled water. Callus initiation. The anthers, 0.2 to I mm long, were excised from the flowers and cultured during 28 days 5 cm Petri dishes containing halfstrength MS (1962) medium solidified with 0.7 % agar and supplemented with sucrose (2 % w/v), 2,4-D (4.5 DM) and BA (I ~M). Callus proliferation. The anthers were transferred to medium supplemented with NAA (0~5 DM), BA (i DM) and casein hydrolysate (250 mg/l) and cultured during one month. The various nitrogen compounds examined were added before autoclaving to these two media. Embryogenetic callus proliferation. Calli were subcultured each month on a regeneration medium lacking casein hydrolysate and supplemented with NAA (0.5 ~M) and BA (] NM). Embryoids were individually transferred to test tubes containing BA (0.4 ~M) in order to develop plantlets. All the cultures were maintained at 24°C under a 16 h illumination regime (]00 pE ~2E~, except the callus initiation performed in the dark. The percentage of anthers producing callus was determined at the end of the first subculture upon embryogenetic medium. The percentage of calli producing embryoids was determined after three subcultures upon embryogenetic medium.
378 RESULTS and DISCUSSION I) EFFECT OF ANTHER LENGTH For three clones of Cabernet-Sauvignon (n ° 15, n ° 1565, n ° 337), the effect of anther length on callus and embryoid yield is presented in Table I. Table I : Effect of anther length on callus and embryoid initiation of 3 clones (n ° 15,1565,337) of Fitis vin~f~ra cv. CabernetSauvignon. = o =
Anther length (mm)
% Anthers producing callus
%a embryogenic calli
%a embryoids 0
0.3
850
7
O
0.3 to 0.5
750
8,7
3
3
0.5
to 0.7
450
0
0
O.8
500
0
0
11.6 ~m 3
0.3
600
10
0
0
0.3 to 0.5
1450
25 ~
0.8
0.8
0.5 to 0.7
1550
19.2 mm
1.7
76.2
0.8
1200
1.2
3.7
0.3
m
Number of anthers cultured
6.8
360
2O
0
0
O.3 to 0.5
820
25.8 m
0.4
0.4
0.5 to 0.7
980
43,6 ~
180
1.6
0.8
1.8
4.9
O
0
Fig. 2 : High-frequency somatic embryo formation from anthers of CabeTnet-Sauvignon clone n ° 15 (X5).
a : calculated compared to the number of anthers producing callus. Significantly different from the other classes (2x2 contengy test) at 0.05 level of probability ~
at O.Ol level of probability
For each clone the most suitable developmental stage of anther for producing callus was remarkably and significantly constant and corresponded to the two intermediate classes : 0.3 to 0.5 mm and 0.5 to 0.7 mm. These anther classes initiating most calli were also the most efficient at producing embryoids. A comparison of the three clones pointed out differences in their capacities to initiate calli and embryoids. Clone n ° 1565 was found to be significantly (P = 0.05) the least effective especially in producing calli. These calli had low embryogenic capacities (figure I). The two other clones n ° 15 and n ° 337 were more efficient in both callus and embryoid production. However, the percentages of embryogenic calli were in general rather weak since they never went beyond 1.8. An occasional high embryoid yield was observed (76.2 %) displaying the high embryogenic capacity of certain calli (figure 2).
These data confirm those of BOUQUET et al. (1982), showing that the ability of anthers to produce calli and embryoids depended on genotype. Untill now the most suitable anthers retained for grapevine somatic embryogenesis have been identified by their microspore state (RAJASEKARAN and MULLINS 1979, HIRABAYASHI 1982). According to the present results it seems that anther length is the best practical parameter. By the way, it was found that callus was produced by the anther wall and/or the connective and not by the microspores (RAJASEKARAN and MULLINS 1983). Thus, in the following experiments all the cultured anthers of Cabernet-Sauvignon were 0.5 mm long. This size was easily identified by a color shift of the anther wall, from green to yellow.
2) EFFECT OF CASEIN HYDROLYSATE, AMINO ACIDS AND ADENINE ON CALLUS AND EMBRYOID INITIATION
Fig.
I : Low-frequency somatic embryo formation from anthers of Cabernet-Sauvignon clone n ° 1565 (X5).
Anthers,of clone n ° 337 have been used in an experiment designed to define, primarily, the effect of casein hydrolysate added to the proliferation medium on callus and embryoid yield. The effect of casein hydrolysate on callus initiation was not appreciable ; with or without casein hydrolysate the percentages of calli were similar. However, the aspect of calli was greatly different, and without casein hydrolysate none of the calli, uniformly brown, were able to induce embryoids (figures 3 and 4).
379 Tablp 2 : A ~ n e § ~ t ~ o n of 311 the factor~ found tO sti~l~la~ ~mbryo~d formation of Cabernet-gauvignon clone n ~ 337 (anther length, casein hy~zolysmte, glu~amine, phenylalanine and adenine). The percentages were d e t e ~ i n e d at the same time as previously reported (v.i, materials and methods).
Nitrogen additives
%a Embryogenic callus
of
%a embryoids
% of embryoids forming plantlets
Phe+Ade
2].5 x
1.5
1.4
Glu+Ade
27.9 ~
2.3
8].8
19.4
26.3 ~
4.8
9.6
87.5
;1.5
O
0
Glu+Phe+Ade CONTROL
Fig. 3 : Callus production by anther cultured during 28 days upon a callus initiation medium and one month upon a proliferation callus medium supplemented with casein hydrolysate (X 2).
% Anther producing callus (~ >5 mm)
0
0
Results are for 315 anthers (0.5 mm long)/sample a : calculated compared to the number of anther producing Glu : glutamine added at 1OO mg/l ; Phe : Phenylalanlne Ade
callus.
added at 10 mg/l
;
: added at I mg/l.
~ significantly different from the control 0,05 level of probability.
(2x2 contingency
test) at
3) IMPROVEMENT OF EMBRYOID DEVELOPMENT The inaptitude of embryoids to develop into plantlets was attended by the occurrence of hypertrophic structures or by one or multiple cotyledons (figure 5).
Fig. 4 : Callus production by anther culture~ Ouring 28 days upon a callus initiation medium and one month upon a proliferation callus medium without casein hydrolysate (X 2).
Addi.tionally, L-glutamine and DL-phenylalanine previously found to enhance cell proliferation and adenine to increase embryogenic efficiency (MAURO et al. 1985) were also added to the culture medium. The effect of the association between these three compounds (glutamine, phenylalanine and adenine) was examined. The three media tested increased in a similar fashion and significantly the percentages of calli obtained from anther (table 2). When transferred to the regeneration medium, all the treated calli showed an enhanced capacity to induce embryoids. The most effective medium was that supplemented with the 3 organic compounds (4.8 % of embryogenic calli) and the weakest percentage (1.5 %) was observed in the absence of glutamine. These results pointed out that the association between glutamine and adenine was the most suitable. Therefore, associated to adenine, glutamine was effective not only on callus growth but also on Cabernet-Sauvignon embryogenesis. According to KAMADA and HARADA (~984) glutamine, known to be a key nitrogen donor, could support the active protein synthesis needed for rapid cell proliferation and em]~ryogenesis. Otherwise it has been shown by GAMBORG (1970) and BISTER-MIEL et al. (1985) that casein hydrolysate effectiveness could be due to its high concentration in glutamine. Occasionally, when a large percentage of embryoids were differentiated (81.8) (table 2), only some of them developped into normal plantlets (19.4 %). An inverse relationship was observed between the embryoid yield and the subsequent embryoid development, as if these two phenomena were antagonistic.
Fig. 5 : Cabernet-Sauvignon embryoid with primary and adventitious roots, elongated hypocotyl and abnormal cotyledons (X3).
Fig. 6 : Cabernet-Sauvignen plantlet after normal development, with primary root, hypocotyl, small cotyledons and leaves. Excision of cotyledons was not necessary (X 2).
380 Difficulties in producing grapevine plantlets from embryoids have been reported earlier by RAJASEKARAN and MULLINS (1979), BOUQUET et al. (1982) and HIRABAYASHI (1985), KRUL (1985). Attemps to improve the germination of somatic embryos, using differing basal media were unsuccessful (RAJASEKARAN and MULLINS 1979). Three to five weeks of chilling and also gibberellin treatments proved to be effective (RAJASEKAN and MULLINS 1979, TAKENO et al. 1983). In our attempts athree weeks chilling period at various stages of embryogenesis did not significantly increase the percentage of regeneration (data not shown). One hypothesis to explain the dormancy phenomenon in zygotic embryos suggests that cotyledons could exert an inhibitory effect on the germination of embryo axis (THEVENOT 1982). Consequently cotyledon removal could be a suitable technique to overcome the inaptitude of embryoids to develop into plantlets. Therefore, chilling and cotyledon removal were the two treatments performed on clone n ° 15 embryoids, obtained in large quantity from calli grown on medium supplemented only with casein hydrolysate. The data are presented in table 3. The most effective treatment was the removal of cotyledons associated with three weeks of chilling which caused about a 5-fold stimulation in plantlet regeneration (56 %) compared to the untreated embryoids (12 %). Early cotyledon removal alone proved to be nearly as effective (52 %). Performed later, when the abnormal cotyledons were thoroughly grown, cotyledon removal was far less effective (28 %) as was chilling when applied alone
(3t %) Table 3 : Effect of various treatments on the development of embryoids into normal plantlets (Cabernet-gauvignon cv n ° 15) during the third month upon the embryogenic medium.
Number of embryoids treated
% of embryoids producing normal plantlets
Embryoids not chilled
48
12
gmbryoids chilled 4°C for 3 weeks
52
31
60
52 ~
48
56 ~
46
28
Treatment
aEarly removal of embryoids cotyledons aEarly removal of embryoids cotyledons and chilling at 4°C for 3 weeks bLate removal of embryoids cotyledons
a : Removal was practised just when cotyledons appeared, before abnormality was noticed. b : Removal was practised when cotyledon abnormality was visible. significc~lydifferent from control (2x2 contingency test) at 0.05 level of probability.
In conclusion, it has been shown that the various cultural conditions studied could improve embryoid yield of Cabernet-Sauvignon. In particular, the visual selection of the most effective anthers is of practical
interest and the cotyledon removal a proper means to increase the number of regenerated plantlets. Regarding the embryoid abnormalities, further insight into the phenomenon and use of other factors will be necessary to further increase the percentage of regeneration. Nevertheless, the somatic embryogenesis efficiency displayed for Cabernet-Sauvignon should provide an interesting tool to get a new variability inside this variety. For the moment I00 somaclones obtained through this method have been transferred to greenhouse and their variability is now under study.
REFERENCES
Bister-Miel f, Guignard JL, Bury M, Agier C (1985) Plant Cell Rep. 4: 161-163. Bouquet A, Piganeau B, Lamaison AM (1982) C.R. Acad. Sc. Paris 295: 569-574. Dunwell JM (1986) In: Plant Tissue Culture and its Agricultural Application Ed. L.A. Withers and P.G. Anderson, Butterworths, London. 377-403. Favre JM (1977) Ann. Amelior. Plantes 8: 151-169. Gamborg OL (1970) Plant Physiol. 45: 372-375. Gresshoff PM, Doy CH (1974) Z. Pflphysiol. 73: 132-141. Halperin W, Wetherell DF (1965) Nature 205:408. Hirabayashi T (1982) Proc. 5th Intl. Cong. Plant Tissue and Cell Culture 547-548. Hirabayashi T (1985) Colloque Amelioration de la vigne et culture in vitro. Paris Avril 1985. In press. Kamada H, Harada H (1979) Z. Pflanzenphysiol. 91: 453463. Kamada H, Harada H (1984) Plant Sci. Letters 33: 7-13. Krul WR, Worley JF (1977) J. Amer. Soc. Hort. Sci. 102: 360-363. Krul WR (1985) Somatic embryo induction, expression and development. Ageneral model. Colloque Amelioration de la vigne et culture in vitro. Paris Avril 1985. In press. Mauro MC, Nef C, Ambid C, Fallot J (1985) 4~me Symp. Int. de la g~n~tique de la vigne. Verone Avril 1985~ In press. Mullins MG (1966) Nature Lond. i09: 419-420. Mullins MG (1971) Rep. CSIRO Division of horticultural Research 1969-1971. Adelaide Australia 26-27. Mullins MG, Srinivasan C (1976) J. Exp. Bot. 27: 10221030. Murashige T, Skoog F (1962) Physiol. Plant. 15: 473497. Rajasekaran K, Mullins MG (1979) J. Exp. Bot, 30: 399-407. Rajasekaran K, Mullins MG (1983) Am. J. Enol. Vitic. 34: 108-113. Sibi M, Duma s de Vaulx R, Chambonnet D (1979) Ann. Amglior. Plantes 29: 583-606. Srinivasan C, Mullins MG (1980) Scientia Hortic. 13: 245-252. Sunderland N, Dunwell JM (1977) In: Plant Tissue and Cell Culture, HE Street, ed. Blackwell Sci. Pub. 223-265. Takeno K, Koshioka M, Pharis RP, Rajasekaran K, Mullins MG (1983) Plant. Physiol. 73: 803-808. Thevenot C (1982) Z. Pflanzenphysiol. 106: 15-26. Wetherell DF, Dougall DK (]976) Physiol. Plant. 37: 97-103.
Plant Cell Reports (1986) 5:377-380
© Springer-Verlag1986
Stimulation of somatic embryogenesis and plant regeneration from anther culture of Vitis vinifera cv. Cabernet-Sauvignon M. CI. Mauro, C. Nef, and J. Fallot Ecole Nationale Supdrieure Agronomique de Toulouse, 145 Avenue de Muret, F-31076 Toulouse C6dex, France Received March 10, 1986 / Revised version received July 25, 1986 Communicated by A. M. Boudet
ABSTRACT Somatic embryogenesis and subsequent diploid plants have been obtained from anthers of Vitis vinifera Cabernet-Sauvignon, a cultivar so far considered as recalcitrant to in vitro regeneration. A n t h e ~ enclosing microsporcs near the first pollen mitosis were found to be the most responsive. However, from a practical point of view anther length proved to be an easier criterium for determining the optimal physiological anther stage. Calli derived from the anther sJ~m~tic tissues pr~duccd cmbryoids o~ly wh~Ll ~ultured on a medium supplemented with casein hydrolysate. Glutamine and adenine were found to stimulate this embryoid production. Evidence is presented that early removal of cotyledons increases the frequency of normal development of embryoids into plantlets. Abbreviations : MS : Murashige and Skoog medium (]962) 2,4-D : 2,4-dichlorophenoxyacetic acid NAA : ]-naphtaleneacetic acid BA : 6-benzylaminopurine. INTRODUCTION Some commercially important grapevine cultivars
(Yitis vinifera) present cultural defects leading to economic losses, in vitro techniques appear to be the only tool available for the genetic improvement of these varieties grown in vineyards where hybridization processes are not allowed. Regeneration of plantlets has been obtained by cultivation of nucellus in Yitis vinifera (MULLINS and SRINIVASAN 1976, SRINIVASAN and MULLINS ]980) and by cultivation of vegetative tissues essentially in interspecific hybrid grapevines (FAVRE 1977, KRUL and WORLEY 1977, HIRABAYASHI 1985). Embryo~ and plantlet production from cultured anthers of Yiti~ long reported to be unsuccessful (MULLINS 1971, GRESSHOFF and DOY ]974) has been achieved more recently (RAJASEKARAN and MULLINS 1979, BOUQUET et al. 1982, HIRABAYASHI 1982). However, the cultivar Cabernet-Sauvignon, used all over the world for the production of high quality red wines, appears to be recalcitrant to anther ~culture. Various physiological factors have been found to enhance the embryoid formation of numerous plants. For in vitro androgenesis through anther culture, the developmental stage of the anther seems to be important (SUNDERLAND et al. 1977, SIBI et al. 1979, DUNWELL 1986). Additionally, the presence in the culture medium of reduced nitrogen such as ammonium ion and certain amino acids was found to stimulate in vitro
Offprint requests to: M. C1. Mauro
embryogenesis of carrot in particular (HALPERIN and WETHERELL 1965, WETHERELL and DOUGALL 1976, KAMADA and HARADA ]979). The purpose of this paper is therfore to examine the influence of cultural conditions, such as anther lenght and the addition to the culture medium of various nitrogen compounds, in order to increase callus and embryoid yield of Cabernet-Sauvignon. MATERIALS and METHODS The methods for anther culture and induction of somatic embryos were adapted from BOUQUET et al. (1982). Cool-stored (4°C) cuttings of 3 clones (n ° ]5, 1565 and 337) of Cabernet-Sauvignon, provided by Mr Pouget (INRA Bordeaux) were grown in greenhouse according to the method of MULLINS (]966). Inflorescences were collected and chilled for 72 hours at 4°C. The flowers were then surface-sterilized (10 min) with a filtered 7 % (w/v) solution of calcium hypochlorite contained Tween 20 ( 0 . 1 % v/v) and rinsed several tim~ with sterile distilled water. Callus initiation. The anthers, 0.2 to I mm long, were excised from the flowers and cultured during 28 days 5 cm Petri dishes containing halfstrength MS (1962) medium solidified with 0.7 % agar and supplemented with sucrose (2 % w/v), 2,4-D (4.5 DM) and BA (I ~M). Callus proliferation. The anthers were transferred to medium supplemented with NAA (0~5 DM), BA (i DM) and casein hydrolysate (250 mg/l) and cultured during one month. The various nitrogen compounds examined were added before autoclaving to these two media. Embryogenetic callus proliferation. Calli were subcultured each month on a regeneration medium lacking casein hydrolysate and supplemented with NAA (0.5 ~M) and BA (] NM). Embryoids were individually transferred to test tubes containing BA (0.4 ~M) in order to develop plantlets. All the cultures were maintained at 24°C under a 16 h illumination regime (]00 pE ~2E~, except the callus initiation performed in the dark. The percentage of anthers producing callus was determined at the end of the first subculture upon embryogenetic medium. The percentage of calli producing embryoids was determined after three subcultures upon embryogenetic medium.
378 RESULTS and DISCUSSION I) EFFECT OF ANTHER LENGTH For three clones of Cabernet-Sauvignon (n ° 15, n ° 1565, n ° 337), the effect of anther length on callus and embryoid yield is presented in Table I. Table I : Effect of anther length on callus and embryoid initiation of 3 clones (n ° 15,1565,337) of Fitis vin~f~ra cv. CabernetSauvignon. = o =
Anther length (mm)
% Anthers producing callus
%a embryogenic calli
%a embryoids 0
0.3
850
7
O
0.3 to 0.5
750
8,7
3
3
0.5
to 0.7
450
0
0
O.8
500
0
0
11.6 ~m 3
0.3
600
10
0
0
0.3 to 0.5
1450
25 ~
0.8
0.8
0.5 to 0.7
1550
19.2 mm
1.7
76.2
0.8
1200
1.2
3.7
0.3
m
Number of anthers cultured
6.8
360
2O
0
0
O.3 to 0.5
820
25.8 m
0.4
0.4
0.5 to 0.7
980
43,6 ~
180
1.6
0.8
1.8
4.9
O
0
Fig. 2 : High-frequency somatic embryo formation from anthers of CabeTnet-Sauvignon clone n ° 15 (X5).
a : calculated compared to the number of anthers producing callus. Significantly different from the other classes (2x2 contengy test) at 0.05 level of probability ~
at O.Ol level of probability
For each clone the most suitable developmental stage of anther for producing callus was remarkably and significantly constant and corresponded to the two intermediate classes : 0.3 to 0.5 mm and 0.5 to 0.7 mm. These anther classes initiating most calli were also the most efficient at producing embryoids. A comparison of the three clones pointed out differences in their capacities to initiate calli and embryoids. Clone n ° 1565 was found to be significantly (P = 0.05) the least effective especially in producing calli. These calli had low embryogenic capacities (figure I). The two other clones n ° 15 and n ° 337 were more efficient in both callus and embryoid production. However, the percentages of embryogenic calli were in general rather weak since they never went beyond 1.8. An occasional high embryoid yield was observed (76.2 %) displaying the high embryogenic capacity of certain calli (figure 2).
These data confirm those of BOUQUET et al. (1982), showing that the ability of anthers to produce calli and embryoids depended on genotype. Untill now the most suitable anthers retained for grapevine somatic embryogenesis have been identified by their microspore state (RAJASEKARAN and MULLINS 1979, HIRABAYASHI 1982). According to the present results it seems that anther length is the best practical parameter. By the way, it was found that callus was produced by the anther wall and/or the connective and not by the microspores (RAJASEKARAN and MULLINS 1983). Thus, in the following experiments all the cultured anthers of Cabernet-Sauvignon were 0.5 mm long. This size was easily identified by a color shift of the anther wall, from green to yellow.
2) EFFECT OF CASEIN HYDROLYSATE, AMINO ACIDS AND ADENINE ON CALLUS AND EMBRYOID INITIATION
Fig.
I : Low-frequency somatic embryo formation from anthers of Cabernet-Sauvignon clone n ° 1565 (X5).
Anthers,of clone n ° 337 have been used in an experiment designed to define, primarily, the effect of casein hydrolysate added to the proliferation medium on callus and embryoid yield. The effect of casein hydrolysate on callus initiation was not appreciable ; with or without casein hydrolysate the percentages of calli were similar. However, the aspect of calli was greatly different, and without casein hydrolysate none of the calli, uniformly brown, were able to induce embryoids (figures 3 and 4).
379 Tablp 2 : A ~ n e § ~ t ~ o n of 311 the factor~ found tO sti~l~la~ ~mbryo~d formation of Cabernet-gauvignon clone n ~ 337 (anther length, casein hy~zolysmte, glu~amine, phenylalanine and adenine). The percentages were d e t e ~ i n e d at the same time as previously reported (v.i, materials and methods).
Nitrogen additives
%a Embryogenic callus
of
%a embryoids
% of embryoids forming plantlets
Phe+Ade
2].5 x
1.5
1.4
Glu+Ade
27.9 ~
2.3
8].8
19.4
26.3 ~
4.8
9.6
87.5
;1.5
O
0
Glu+Phe+Ade CONTROL
Fig. 3 : Callus production by anther cultured during 28 days upon a callus initiation medium and one month upon a proliferation callus medium supplemented with casein hydrolysate (X 2).
% Anther producing callus (~ >5 mm)
0
0
Results are for 315 anthers (0.5 mm long)/sample a : calculated compared to the number of anther producing Glu : glutamine added at 1OO mg/l ; Phe : Phenylalanlne Ade
callus.
added at 10 mg/l
;
: added at I mg/l.
~ significantly different from the control 0,05 level of probability.
(2x2 contingency
test) at
3) IMPROVEMENT OF EMBRYOID DEVELOPMENT The inaptitude of embryoids to develop into plantlets was attended by the occurrence of hypertrophic structures or by one or multiple cotyledons (figure 5).
Fig. 4 : Callus production by anther culture~ Ouring 28 days upon a callus initiation medium and one month upon a proliferation callus medium without casein hydrolysate (X 2).
Addi.tionally, L-glutamine and DL-phenylalanine previously found to enhance cell proliferation and adenine to increase embryogenic efficiency (MAURO et al. 1985) were also added to the culture medium. The effect of the association between these three compounds (glutamine, phenylalanine and adenine) was examined. The three media tested increased in a similar fashion and significantly the percentages of calli obtained from anther (table 2). When transferred to the regeneration medium, all the treated calli showed an enhanced capacity to induce embryoids. The most effective medium was that supplemented with the 3 organic compounds (4.8 % of embryogenic calli) and the weakest percentage (1.5 %) was observed in the absence of glutamine. These results pointed out that the association between glutamine and adenine was the most suitable. Therefore, associated to adenine, glutamine was effective not only on callus growth but also on Cabernet-Sauvignon embryogenesis. According to KAMADA and HARADA (~984) glutamine, known to be a key nitrogen donor, could support the active protein synthesis needed for rapid cell proliferation and em]~ryogenesis. Otherwise it has been shown by GAMBORG (1970) and BISTER-MIEL et al. (1985) that casein hydrolysate effectiveness could be due to its high concentration in glutamine. Occasionally, when a large percentage of embryoids were differentiated (81.8) (table 2), only some of them developped into normal plantlets (19.4 %). An inverse relationship was observed between the embryoid yield and the subsequent embryoid development, as if these two phenomena were antagonistic.
Fig. 5 : Cabernet-Sauvignon embryoid with primary and adventitious roots, elongated hypocotyl and abnormal cotyledons (X3).
Fig. 6 : Cabernet-Sauvignen plantlet after normal development, with primary root, hypocotyl, small cotyledons and leaves. Excision of cotyledons was not necessary (X 2).
380 Difficulties in producing grapevine plantlets from embryoids have been reported earlier by RAJASEKARAN and MULLINS (1979), BOUQUET et al. (1982) and HIRABAYASHI (1985), KRUL (1985). Attemps to improve the germination of somatic embryos, using differing basal media were unsuccessful (RAJASEKARAN and MULLINS 1979). Three to five weeks of chilling and also gibberellin treatments proved to be effective (RAJASEKAN and MULLINS 1979, TAKENO et al. 1983). In our attempts athree weeks chilling period at various stages of embryogenesis did not significantly increase the percentage of regeneration (data not shown). One hypothesis to explain the dormancy phenomenon in zygotic embryos suggests that cotyledons could exert an inhibitory effect on the germination of embryo axis (THEVENOT 1982). Consequently cotyledon removal could be a suitable technique to overcome the inaptitude of embryoids to develop into plantlets. Therefore, chilling and cotyledon removal were the two treatments performed on clone n ° 15 embryoids, obtained in large quantity from calli grown on medium supplemented only with casein hydrolysate. The data are presented in table 3. The most effective treatment was the removal of cotyledons associated with three weeks of chilling which caused about a 5-fold stimulation in plantlet regeneration (56 %) compared to the untreated embryoids (12 %). Early cotyledon removal alone proved to be nearly as effective (52 %). Performed later, when the abnormal cotyledons were thoroughly grown, cotyledon removal was far less effective (28 %) as was chilling when applied alone
(3t %) Table 3 : Effect of various treatments on the development of embryoids into normal plantlets (Cabernet-gauvignon cv n ° 15) during the third month upon the embryogenic medium.
Number of embryoids treated
% of embryoids producing normal plantlets
Embryoids not chilled
48
12
gmbryoids chilled 4°C for 3 weeks
52
31
60
52 ~
48
56 ~
46
28
Treatment
aEarly removal of embryoids cotyledons aEarly removal of embryoids cotyledons and chilling at 4°C for 3 weeks bLate removal of embryoids cotyledons
a : Removal was practised just when cotyledons appeared, before abnormality was noticed. b : Removal was practised when cotyledon abnormality was visible. significc~lydifferent from control (2x2 contingency test) at 0.05 level of probability.
In conclusion, it has been shown that the various cultural conditions studied could improve embryoid yield of Cabernet-Sauvignon. In particular, the visual selection of the most effective anthers is of practical
interest and the cotyledon removal a proper means to increase the number of regenerated plantlets. Regarding the embryoid abnormalities, further insight into the phenomenon and use of other factors will be necessary to further increase the percentage of regeneration. Nevertheless, the somatic embryogenesis efficiency displayed for Cabernet-Sauvignon should provide an interesting tool to get a new variability inside this variety. For the moment I00 somaclones obtained through this method have been transferred to greenhouse and their variability is now under study.
REFERENCES
Bister-Miel f, Guignard JL, Bury M, Agier C (1985) Plant Cell Rep. 4: 161-163. Bouquet A, Piganeau B, Lamaison AM (1982) C.R. Acad. Sc. Paris 295: 569-574. Dunwell JM (1986) In: Plant Tissue Culture and its Agricultural Application Ed. L.A. Withers and P.G. Anderson, Butterworths, London. 377-403. Favre JM (1977) Ann. Amelior. Plantes 8: 151-169. Gamborg OL (1970) Plant Physiol. 45: 372-375. Gresshoff PM, Doy CH (1974) Z. Pflphysiol. 73: 132-141. Halperin W, Wetherell DF (1965) Nature 205:408. Hirabayashi T (1982) Proc. 5th Intl. Cong. Plant Tissue and Cell Culture 547-548. Hirabayashi T (1985) Colloque Amelioration de la vigne et culture in vitro. Paris Avril 1985. In press. Kamada H, Harada H (1979) Z. Pflanzenphysiol. 91: 453463. Kamada H, Harada H (1984) Plant Sci. Letters 33: 7-13. Krul WR, Worley JF (1977) J. Amer. Soc. Hort. Sci. 102: 360-363. Krul WR (1985) Somatic embryo induction, expression and development. Ageneral model. Colloque Amelioration de la vigne et culture in vitro. Paris Avril 1985. In press. Mauro MC, Nef C, Ambid C, Fallot J (1985) 4~me Symp. Int. de la g~n~tique de la vigne. Verone Avril 1985~ In press. Mullins MG (1966) Nature Lond. i09: 419-420. Mullins MG (1971) Rep. CSIRO Division of horticultural Research 1969-1971. Adelaide Australia 26-27. Mullins MG, Srinivasan C (1976) J. Exp. Bot. 27: 10221030. Murashige T, Skoog F (1962) Physiol. Plant. 15: 473497. Rajasekaran K, Mullins MG (1979) J. Exp. Bot, 30: 399-407. Rajasekaran K, Mullins MG (1983) Am. J. Enol. Vitic. 34: 108-113. Sibi M, Duma s de Vaulx R, Chambonnet D (1979) Ann. Amglior. Plantes 29: 583-606. Srinivasan C, Mullins MG (1980) Scientia Hortic. 13: 245-252. Sunderland N, Dunwell JM (1977) In: Plant Tissue and Cell Culture, HE Street, ed. Blackwell Sci. Pub. 223-265. Takeno K, Koshioka M, Pharis RP, Rajasekaran K, Mullins MG (1983) Plant. Physiol. 73: 803-808. Thevenot C (1982) Z. Pflanzenphysiol. 106: 15-26. Wetherell DF, Dougall DK (]976) Physiol. Plant. 37: 97-103.
