85": v2 éliasfiufinurwwWHL. 2.2.. ....2. . .. . _ .. .. 2 2 . ... , . . hie-«22 ...-..1. 3.2.6225. .2 4422?»): 2.1-2. 2. 2 25... «...-2h... . .s.2 . .. .. .2322. 5 V2. . ,...£.p2}$wmn§5w.wV ..me omn..fi§...2.wwwmwnu ....nmnwmwbr..,é2 r. . . . 2&5. a. .m . V flaw“? ...Vcwm. «.322... .....JWM2wa-m2gfi-..& 2-... M222? H . «...?é. .. «..w WE! .u-Mnmafiwwwz . V .V. . “Md-Winn" 2a.? immei-uusmfl. . ......2..,..v.-2 . Es. «nfififimqudr .m 2. a .2...» .52 . egg-2.22.2: ...-.5. . V .2222.wa 35.0w. . 2 .. 25.72 2...“. .2l. r ”H2 .2.” I. 3WD .9er » r.-.n.V 7.42.“:2 . .2 «we ..3 wafimflfi H4522 22... 2.2m .... . 2.5.2“. .V. .....d. 2 ”2.... ,. «fun-4.2. horcmvir "...-V22. .3 . 2... V-. . . 1.1%.). u. 2 . .2282... 2 'INJslvxiiv2rl‘V22A3rs 2. 1:. «2s... a. 1- VV 8 294.222 21 \ V 2..- \c 2.210. «Khan ‘5! ravuViJhw-w. 2. LhA-vh. 22.2]... 2 But-.... 4.2 .. .fixstrtwmuhilfll 2.92m o 222.252....“ 22.. ..r...«.........2...h fin»2km}....w-fl,rmu._2 «aw. . V2..- 2m$.%m.n2..v2fi 12, Army»? «kg-...: :2 end-m 522.2% hwflhtmrm2st2qlm2. a........-....mtsmmnflfi2V2.w-flz-&u.¥s.2¥«a22m6 . 2. - 2 I22 .. ...“.«Jrfififizm kin-r «Mal-u; .. s 2‘ h b I ’2P‘ I ... . I 2 22' I 2 . - 42.2%.... ibusYhquJuorfis 22-pin . «22,2 2- . 3.22.2 VH- 222: hrwflslc. .....22222 Afgkfliris 2.222221222222222 JJ.2...,2222- . 2352222 “UK-2.2“ 2.2.1.2224; . . -2 . «clsb . Vo '20.!!!0'222 . v . . . . a 2 2 2 s - 2 - 2.. - 2. . 2 u . V b 2 2 Y I: 2 l . I ‘ 2 . 2.... - 2 "a V .721: . lor- . 2. .n... Jria 12....2ts. as l.V-.27'Hl\.liv (A 80... .2 42 . V V 2. 3... .2 , 2. .- 2 .... .12 ..2. .2 2 .I 2 It. : 2.2%.“? v «1.2? .2 I .2 Wflfirfiu .4. a . . 2... 22 V .2. 3.! . . ... . 25.2m2hn22u22 2%. .Ps: «my .2 2.223 2.272 «.1. .. .. 2?. .V 2 .22. "22422.2 .. 2 22... ,. ...... V 2 2 2.2.2.2522 . 2. V. use-um. Hawaii-2., Miss-{mum.VM2WWL2fl-m12224v2n222mék A «We 2.5nsfirquflwkmfiflmmhwvfflr-n $2.. . .. .. V. nun-”2.2 ....2u.«u222«s2..2.&..u2. . .. 2222.. Vania-“nuanfimnv52nz. ....m... annumfi ... ii???“2.222...2.s-.—2....H...«2r-.2. «wfi: 2. u...u2.2wfiu2fln.-n.«22- 221-22 2.2ng ...fi-Mfim max.» .22222 22.22.22... -....véfV nun-22.22.27. :2 . .V . 2.13.22: 522.322....» 2":- 2...» .0... .nn2u822221mmfizsv «any-222., ,2 2.2km ..aul I 722...}.2931. .- Null-“‘14.“ VII. 211%.}? . u . .‘rrIL .21 h. . 2.2.2! 9 .92: ...“.I-Do Isl-‘42! . .4 3A2‘v. ... givi- «smw.?..u .24.": 2 22222... V.......22.+ xagyfiuiLbX-L V V §2¢$ ..2 2.1- .1335, «222%.. ...2.2.UM2...~2§.J..2.§«. ....2. 22-2.? 7.2.2.? 2%.. ma. .. 2. 222 2.- ..N.V22.. .. 22.2.2.1): .... Ln . 22 .r.--.«..- . 2 ..fi2?r2§k.v. .qust.x-..2r.§§unn2.2fiw.hfir ammu. 2«Ms.. zsiumm-J Laos-«... 5.99%?23 2.0!... .- 42..»2 EV- {£29. 2: «n2,_-22.2§22..22:.22 «"2222. 22:12-21:32 -- 1. .2 2 i!- .n Hfith: Evita-war... .2 v 51:22.92: v .. u’Vltfiiv 2;.C AB. 222%». no 22102: 2|. nut-222 T'ftinlo Zak)“ « . .. _ ....22. . . 2...... ..2- . rrwmkdxr) V...........-._2...,2..2V-.9..2.fi. ..L.-2. .«2...«H-.2..22.:.V 2.22.24.21.22 $0222.52? w. . . _ .... _ .«.sI« . . 2. 2 .V . « .... 2.2. nun. Aggy...» 2422?»,2ffip22q22r22mr.22fl=223. ffLWV-i‘ . 2 .993: 212.“...22322... 2., V. ..2 fluff ...; ”9.2.. an-uiafistnu..s.2.-.2..2-2Wm2-\24-2!22s2.§ 2 VV . mummpfiz . 2 .. cur-1.22.2222? nah-Wur- iss. 2a.. 2§n$2fi2hhr2n~v¢hfihhMmMuUW2£fiuk 2122.2 . . .. .. . . . 2 2 -2.. 2 .«2......4..2..¢.2m22l- -2.-«..2-.%. . 2.2.3... 2. .3222”... 224- LEI-£22222 22..~2«n..«.. .22 .. - 2pm.... .2 . . ...-15:22:22-? 21.222222}. 228-. 2 .- . 2 «... s- .. s22. 2 ... dust-s2 21.2- .h n «2 i -.u-2.-2.2.-.«sa.-. ....m.-22.2.:-2m«sl #352 .2. 2 u..- u. . .2 .2.- 3flh$uhummhwfl2 «hwwuflWFs! _ .2 trauma-22222.. 2.2.1.2.“...9222P-«s222us23w22 2. 2... .V 22 .2 2.1-2.5 m: if . . .2 12.72.222.32, ssss 292 ...: 5.2.222- 2. ...s 222202-2213. .2. -s2. .22.. .. run-«s «2 I. . .2 ..H..J. radar-.mu-mni-wp J§2mnm2 . ”.22 ... 2.2.2.2135 .-..2 V... . ..... . 2.2”»? ..h 222., . . . 2.. «kW-2.5%.... 22s»... 2.2-flfimwmnymwfiwsflwmw2nknmfi2 22mmmm NW2.»- . 2 . . V . 2 . . . 2.22222; {ammo-...... ......m... .- unamwné mnadfwflfi ......Mn Vuw..ulm_w.w2vm2vmm.wv.n222%2.uss2¥1-fi . .. . . . . ...2 2 -.n . . . . .2 . .V .2 b. 2-22..- .s-.2V. .u... . “ha-22.2.5222- ...2 2.. . .n».....Q.-«?2... a progeny ratio of 1:2:1 homogametic supermale (MM), heterogametic male (Mm), and homogametic female (mm), respectively. Supermales are imam; important 1J1 asparagus breeding programs because the result of crossing a supermale and £1 female is EN] all male population. Generally, male plants give rise to more spears and are consequently higher yielding than female plants (Franken 1970). It should be noted that supermales can also be produced via anther culture (Falavigna et a1. 1996). Breeding Goals Asparagus has an average harvest period of 10 years and also demands a great amount of financial and labor input before any economic return is realized. Therefore, improvement of cultivars by breeding has high priority. Important breeding aims are all-male hybrids, high yield and quality characterized by en1 increased number (N? spears of large diameter per plant, uniformity among the spears, low level of fiber in the spears, high disease resistance, salt tolerance and climatic adaptability (Reuther 1984). In the Netherlands, breeding for earliness is important because of the cool climate. Earlier cultivars would allow the harvest to occur along with Southern Europe which is typically a month sooner (Scholten and Boonen 1996). 6 Fusarium If; considered the rmxfi: limiting factor ijl asparagus production (Mace et a1. 1981; Peirce 1987). Fusarium oxysporium (Schlecht) f. Sp. asparagi Cohen (FOA) is the causal agent of the wilt and root rot disease and Fusarium moniliforme Sheld. emend. Snyd. & Hans (FM) is the causal agent of stem and crown rot (Cohen and Heald 1941; Johnston et al. 1979). Because of fusarium, asparagus plantings decline to uneconomical levels as soon as 5—6 years due to the reduction in plant vigor and loss of crowns. Replanting in fields where asparagus was grown previously results in losses of up to 50% of the new plants within the first year. This problem is described as the “asparagus decline and replant syndrome” (Hanna 1947; Lake et al. 1993). Because FOA and PM are transmitted through soil, the Luxe of fungicides and fumigation for long-term control is limited. So, breeding for disease resistant varieties is time most effective yum; for control of this disease. Takatori and Southern (1978) in California and Ellison (1986) in New Jersey made attempts to select for asparagus resistance to fusarium, but to date no such cultivar has been developed. However, new lines introduced from Rutgers such as Jersey Giant have considerable tolerance to this disease. Ellison (1986) attempted to use exotic germplasm 1J1 breeding tin: fusarium resistance. He collected numerous seed samples of A. acutifolius in the wild from Greece, Italy and Spain and seed of wild A. maritimus from Yugoslavia. Unfortunately, all of those accessions were 7 found to be susceptible to fusarium. Only ornamental species such as A. densiflorus cv. Sprengerii and A. plumosus have been found t1) be highly resistant to fusarium (Lewis and Shoemaker 1964; Stephens et al. 1989), but neither of them hybridize sexually with A. officinalis (Ellison 1986). 131 addition IX) fusariunn rust (Puccinia .asparagi), phytophthora spear rot (Phytophthora spp.), botrytis blight (Botrytis cinerea), stemphylium leaf spot (Stemphylium vesicarium) and asparagus virus 1 (AVl) and asparagus virus 2 (AV2) effect asparagus production. These diseases are not as ubiquitous as fusarium but can severely affect production if conditions permit (Peirce 1987). With the exceptions of fusarium and viruses, the fungal diseases can be controlled by using resistant varieties, fungicides and cultural practices (Broadhurst 1996; Scholten and Boonen 1996). There are several differences in the growth of female and male plants that make the males more desirable for field production. Male plants give rise to more spears and are consequently higher yielding than the females. Female spears generally are heavier individually but fewer in number than in male plants (Ellison and Scheer 1959). Also, male plants do not produce seedlings which compete with the established crowns for nutrition and may favor disease epidemics. For these reasons, breeding efforts are concentrated on developing all-male cultivars that will produce higher, more stable yields of spears (Ellison 1986). 8 W-Movm Somatic embryogenesis is the process by which haploid or diploid somatic cells develop into differentiated. plants through characteristic embryological stages without fusion of gametes (Williams and Maheswaran 1986). This process occurs in Vivo but is confined to intra-ovular tissues including the nucellus, inner integument, synergids, antipodals and endosperm (Tisserat et a1. 1979). Currently however, somatic embryogenesis :hs best IOKNHI as an .h? vitro developmental pathway. In vitro somatic embryogenesis was first observed in carrot cell cultures where structures resembling zygotic embryos, “embryoids”, developed anxi gave rise tx> plants (Reinert 1958; Steward et a1. 1958). These structures were eventually termed somatic embryos because they developed from somatic cells without gametic fusion (Ammirato 1987). Somatic embryos are characterized as structures that progress through morphologically defined stages of development (globular, heart-shaped, torpedo and cotyledonary) and culminate in the formation of a bipolar structure possessing both a shoot and root apex at maturity (Ammirato 1987). Somatic embryos ck) not have endospenn due to tflua lack of gametic fusion and are generally genotypically identical to the donor parent if derived from somatic cells. Also, seed coats are not present on somatic embryos since they are the result of the integuments hardening around the zygotic embryo. Somatic embryos can be distinguished from shoots in vitro by 9 the presence of a “closed” vascular system connecting both apices. The vascular system of a shoot or a root extends into the parental tissue. In addition, embryos give rise to cotyledons prior to true leaves while shoots produce only leaves (Haccius 1978). In vitro somatic embryos develop via either direct or indirect embryogenesis (Sharp et a1. 1980; Evans et al. 1981). Direct embryogenesis involves tflua development (n3 embryos directly from tissues without an intervening callus. The embryos arise from “pre-embryogenic determined cells” (PEDC’s) which are embryogenic at the time of explanting and undergo a consecutive mitotic development sequence characteristic of zygotic embryogenesis when environmental conditions are favorable. PEDC’s are most prominent in embryonic tissues such as cotyledons, hypocotyls and the embryonic axis, but can also occur in leaves, stems and other non-embryo associated explants. In indirect embryogenesis, cells must dedifferenti- ate before they can express embryogenesis. Dedifferentiation of non-PEDC's converts them to induced embryogenic determined cells (IEDC’s) which are embryogenically competent under the proper conditions. IEDC’s are primarily present in embryogenic callus and suspension cells. Auxin is the most important factor for the regulation of induction and development of embryogenesis and has different effects on different phases of embryogenesis (Komamine et al. 1992). Auxin is required for the formation of embryogenic 10 cell clusters from non-induced competent cells. Once IEDC's are formed, the same level of auxin required for induction of embryogenesis is inhibitory for expression of embryogenesis. IEDC’s will express embryogenesis upon auxin removal. io n lo Somatic embryogenesis The first report of somatic embryogenesis in a monocot was with Asparagus officinalis by Wilmar and Hellendoorn (1968). Totipotency of cultured plant cells had only recently been demonstrated in dicots (Steward et al., 1958) and so asparagus was chosen to determine if monocots also had the ability to develop whole plants from undifferentiated cells. A green callus was induced on hypocotyls from sterile seedlings incubated on LS medium with 1 ng/l 2,4— dichlorophenoxyacetic acid (2,4—D) and 0.315 mg/l kinetin. Upon omission of the growth regulators, shoots and some roots developed from the calluses similar to previous reports with carrot (Steward et al. 1958). To investigate embryo development in more detail, callus cells were placed in LS liquid medium containing the same levels of 2,4-D and kinetin and agitated on a shaker. Small cytoplasmically dense cells developed along with globular embryos in suspension and these embryos developed further as the 2,4—D concentration was reduced. Steward et al. (1958) observed that the embryos developed according to the pattern of zygotic embryos in vivo 11 and many eventually elongated to form banana—shaped structures up to 2 mm in length. They also described polarity, a criteria tin: embryogenesis, as EH1 increase 111 chlorophyll content on one side of the embryos and elongation to form the radicle on the other. Some embryos germinated into plants when placed on medium lacking PGR’s although transferring the embryos at the right stage from suspension to this medium was critical. Normal diploid plants were obtained as well as some tetraploids. During the same period, Steward and Mapes (1971) were also .investigating totipotency 111 asparagus euui the possibility of developing a mass propagation system for the crop. Asparagus was considered to be recalcitrant since it is a monocot, and a challenge to establish in tissue culture. The authors proposed that if a propagation system from free cells could be cmmeloped, elite genotypes could be indefinitely nmltiplied. Stem segments (n5 asparagus were placed on White’s medium with 10% coconut water and NAA to establish callus cultures. Subsequently primary calluses were placed in liquid MS medium with coconut water and 2,4-D to produce morphogenic suspension cultures. NAA satisfactorily promoted callus growth, but produced little evidence of organized growth, whereas 2,4—D resulted in more regular and organized cell clusters. It was believed that 2,4-D provided a morphogenic stimulus necessary for embryogenesis and that NAA did not. When the cells were transferred to liquid medium 12 containing NAA, “embryoids” and cell clusters with roots developed. If the 2,4-D cultured cells were transferred to MS medium containing only coconut water, vigorous root growth occurred. However, when the cells were transferred to liquid MS medium alone, the cell clusters gave rise to numerous feathery shoots. These feathery shoot cultures could be placed on semi—solid MS medium and rooted. This study had shown that cultured. asparagus cells luui all the genetic information and all the cytoplasmic machinery to support their totipotent development given the proper stimuli. This report demonstrated that ea clonal propagation system \dji somatic embryogenesis was possible in asparagus. Levi and Sink (1990; 1991; 1992) investigated several factors influencing the induction, development and subsequent conversion of somatic embryos in Asparagus officinalis L. The effects of carbohydrates on embryogenesis were tested (Levi and Sink 1990). Calluses were incubated on induction medium containing LS salts, 1.5 mg/l 2,4-D and 0.3 mg/l 2iP or 1.5 mg/l NAA and 0.3 ng/l bfi-(2-isopenntenyl)adenine (2iP), and 2, IL. 4, 5, or 6% (w/v) of either sucrose, glucose or fructose for four weeks. Callus tissues were subsequently placed on subculture media containing LS salts, 0.08 mg/l NAA, 0.2 mg/l ZiP and the same sugars as above. Upon subculture, calluses were placed on all combinations of the three sugars (nine treatments per level) at the various levels. Glucose induced the greatest number of globular embryos on both 2,4—D 13 and NAA although the frequency was not significantly different from sucrose on NAA containing medium. Embryo induction generally increased at carbohydrate levels up to 3% and then significantly decreased at concentrations of 5% or higher. Regardless of auxin type, fructose induced the fewest embryos and those embryos tended to be large and vitrified. In general, upon subculture as the carbohydrate concentration increased, so did the development of bipolar embryos. It was finally determined that when 2,4—D was used that sucrose in the induction medium and fructose in the subculture medium gave rise to the most plantlets. When NAA was used, glucose in the induction medium and fructose in the subculture medium was best. The results also showed that more embryos converted into plants when NAA was used in the induction medium compared to 2,4—D. Levi and Sink (1991) found that the choice of explant and growth regulator can significantly effect embryogenesis and plant regeneration. Embryogenic callus was derived from spear cross sections (SS), in vitro crowns (IVC) and lateral buds (LB) on either 0, 0.01, 0.1, 1.0 or 10 mg/l 2,4-D or NAA and 0, 0.1, 1.0 or 10 mg/l kinetin. 2,4-D at 1-10 mg/l in combination with 0-1 mg/l kinetin induced more globular embryos from all three explant sources than NAA at the same concentrations. NAA promoted a higher frequency of bipolar embryo development compared to 2,4-D after the cultures were placed on embryo development medium (EDM). In addition, 14 bipolar embryos initiated on NAA media converted into significantly more plants than from 2,4—D derived embryos. 2,4-D promoted the development of abnormal embryos especially at 10 mg/l which was noted to be the most likely cause for low embryo conversion. In regards to explant source, SS cultured on 2,4-D yielded the most globular embryos, bipolar embryos and plantlets per gram of callus than IVC and LB explants. When NAA was used to initiate embryogenesis, LB explants produced the greatest number of globular embryos, bipolar embryos and plantlets. The authors concluded that induction of embryogenic callus was best carried out from LB explants on 10 mg/l (50 uM) NAA. This scheme enhanced embryogenesis over other sources and promoted conversion of normal plantlets on embryo development medium. Histology was performed on some of these embryos and their ontogeny was compared to that of asparagus zygotic embryos (Levi and Sink 1991). The authors identified embryogenic cells within asparagus callus that were cytoplasmically dense in contrast to the surrounding cells and tracked their development up to the nature bipolar stage. Mature somatic embryos either closely resembled their zygotic counterparts having a banana or crescent shape with an elongated cotyledon and a lateral shoot apex, or they had a short and wide cotyledon that did not resemble the zygotic embryos. Both types were reported to convert into plantlets with equal frequencies. 15 In a final study, Levi and Sink (1992) re-examined the production of asparagus somatic embryos from suspension culture and investigated the effect of NAA on embryo induction along with the effect of carbohydrate concentration and type on embrye> maturatitwi and. conversion. Their attempts to utilize previously published protocols (Wilmar and Hellendoorn 1968; Steward and Mapes 1971) resulted in mostly abnormal structures and few plantlets. These earlier reports had also failed tx> quantify' embryogenic responses run: mention the germination frequency of embryos to plantlets. Callus cells were placed into liquid MS medium with NAA and 2iP at non- embryogenic levels to establish suspension cultures. Cells from these suspensions were placed in liquid MS medium containing 0.54, 5.4, 16, 54, 81 or 107 (JM NAA. The suspensions were evaluated for embryogenic potential by assessing (1) the frequency of single elongated vacuolated cells, (2) the number of organized clusters of cells with enriched cytoplasm, (3) the number of proembryogenic masses consisting of elongated vacuolated cells and globular embryos at early developmental stages (8 to 64 cells), and (4) the quantity of translucent globular embryos with a smooth spherical profile. NAA at 16 uM or below produced suspensions that contained mostly single, elongated vacuolated cells which is characteristic of a lack of embryogenic potential. These suspensions also contained some clusters of cytoplasmically dense cells and many abnormal embryos that failed to convert 16 into plantlets on embryo development medium. In contrast, suspensions containing NAA at 54 uM or higher were quite embryogenic. They also possessed some elongated cells along with many clusters of dense cells with organized divisions and globular embryos. Although less tJMNI 50% (Hf the initial suspensions at these levels were embryogenic, those that were gave a high frequency of mature bipolar embryos and vigorous plantlets. Based on these observations, the authors state that levels of NAA at 54 uM or higher are required to induce a high frequency of embryogenesis in asparagus cells and that stringent selection of cell characteristics must be used to obtain and maintain embryogenic suspensions. To evaluate the effect of carbohydrate type and concentration on embryo development, suspension cells were placed on LS medium containing either sucrose at 0.06, 0.12, 0.18, 0.24 or 0.3 M, or glucose or fructose at 0.11, 0.22, 0.33, 0.44 or 0.55 M for two weeks and then subcultured to the same sugar at 0.06 or 0.11 M for an additional four weeks. Low sugar concentrations supported. secondary' embryogenesis, whereas, higher levels promoted the development of bipolar embryos with an average size that correlated to the increase in carbohydrate molarity. Fructose or sucrose concentration sequences of 0.33 to 0.11 M and 0.24 to 0.06 M, respectively, were optimal for the production of vigorous plantlets after two to four weeks on embryo development medium. l7 Somatic embryogenesis of asparagus has been Viewed as a suitable alternative to clonal propagation via adventitious shoots. Kohmura et al. (1994) established a micropropagation system in which bud clusters derived from shoot apices gave rise to somatic embryos. The bud clusters were established by incubating shoot apices on MS medium supplemented with 10 mg/l ancymidol for several months. The compact structures or bud clusters were placed on 10 uM 2,4—D and formed embryogenic calli after several months with bimonthly subcultures. The callus was placed in liquid LS medium without growth regulators on a rotary shaker to allow for somatic embryo development. Mature embryos germinated into plantlets when placed on solid medium without growth regulators. The bud clusters could be maintained indefinitely on ancymidol which allowed for a constant source of tissue for induction of embryogenic callus. The authors established over 2000 somatic embryo derived plants in the field and did not observe the occurrence of any abnormal plants. Cytology was performed on a random sample of plants and all had the expected 2n = 2x = 20 chromosomes. Odake et al. (1993) investigated the use of suspension cultures to ndcropropagate asparagus via somatic embryogenesis. NAA at 3 mg/l in combination with kinetin at 1 mg/l was used exclusively to initiate and maintain embryogenic cultures. Full strength MS medium was most effective for embryogenesis when compared to half-strength MS 18 or B5 medium. Embryos were placed onto half strength MS medium where most were reported to germinate normally with shoots and roots. Most plantlets were successfully transferred to the greenhouse and acclimatized. Cytology on 33 plants revealed that over 90% were tetraploid while the original stock plants were diploid. The authors speculated that the growth regulators used may have been responsible for chromosome doubling. Saito et al.(l991) investigated several factors to increase the quality of asparagus somatic embryos and improve germination frequency. Both NAA and 2,4-D were used separately to induce embryogenic callus from asparagus epicotyls. 2,4-D was used in most of the study because the frequency of embryogenic callus initiation from NAA was very low. The granular yellowish white callus that developed was placed in liquid MS with 5 uM 2,4—D where a fine embryogenic suspension rapidly developed. The suspensions were plated onto solid or liquid MS medium without plant growth regulators (PGR’s) to determine if medium condition influenced embryo development. Three times as many elongated bipolar embryos developed on solid medium versus liquid medium. The effect of the vessel capping material and Gelrite concentration on embryo maturation and conversion was investigated. Vessels containing embryogenic cultures on 0.2, 0.5 or 1.0% Gelrite were sealed with aluminum foil or a ventilative membrane. The highest Gelrite concentration was best for either closure, 19 however the use of a ventilative filter drastically increased the number of elongated embryos and plantlets that formed. The relative humidity inside of the vessels with the ventilative filters was lower that those sealed with foil and Parafilm®. In addition, the moisture content of the embryos from the ventilated vessels was 6% lower than the foil closed containers. The use of Gelrite and the ventilative filter had a desiccating effect that prevented vitrification and proved beneficial for embryo maturation and conversion. Fourteen genotypes of A. officinalis were tested for their ability to exhibit somatic embryogenesis from several explant sources (Delbreil et al. 1994). Shoot apices, cladophylls and isolated mesophyll cells were incubated for one month on 10, 1 or 0.1 mg/l NAA, respectively. Following the induction period, the tissues were subcultured to PGR-free medium for embryo development. All tissue sources from each genotype were able to produce embryogenic callus at a frequency of 1-20%. The embryogenic callus which was composed of mostly globular embryos could be subcultured monthly onto medium lacking PGR’s and maintained as long-term embryogenic lines. These lines were considered to be habituated and continuously produced embryos although only some were able to convert into plants from ten of the genotypes. Histological studies showed that the long-term cultures grew by recurrent embryogenesis with each new embryo arising from a single epidermal cell of a pre-existing embryo. 20 In a related study, the same authors found that some plants regenerated from the long—term cultures exhibited a high embryogenic response when recultured (Delbreil and Jullien 1994). When shoot apices from somatic embryo-derived plants (R0) were placed back on embryo induction medium, the frequency of embryogenic callus formation ranged from 16.5 to 91.3%. In addition, these same R0 lines produced embryogenic callus at.ea frequency of 26 tx>.57% when the explants were placed on PGR—free medium. None of the control plants that the R0 plants were derived from produced embryogenic callus on PGR-free medium. To determine if the high embryogenic trait was a stable mutation and sexually transmitted, a high embryogenic R0 line was crossed to a low embryogenic line and the ngprogeny backcrossed to the low embryogenic line. They found that a stable mutation had occurred at a single locus and observed segregation ratios of 1:1 in the F1 as well as the BC1 populations. The authors state that id? this high embryogenic trait could be transmitted to desirable genotypes without disrupting existing asparagus agronomic character- istics that a very efficient micropropagation system could be developed. Somatic embryogenesis Imus been achieved 1J1 Asparagus cooperi Baker which has medicinal as well as horticultural value (Ghosh and Sen 1991). Callus was initiated from spear sections on MS medium with 1 mg/l NAA and kinetin. After three monthly subcultures, the callus could be induced to form 21 embryos when placed on the same medium with 2.9 g/l potassium nitrate. Subsequently, the influence of organic and inorganic nitrogen sources on multiplication of this embryogenic callus was investigated. The addition of 300 mg/l glutamine, 1000 mg/l casein hydrolysate and 1200 mg/l ammonium nitrate to the basal medium produced the greatest enhancement of the callus multiplication rate. Mature embryos failed to germinate when they were transferred to MS medium lacking PGR’s. However, transferring embryos to 1 mg/l zeatin promoted embryo conversion with a maximum conversion frequency of 38%. Cytology was performed on 80 plants and all were karyotypically normal and free of any noticeable phenotypic variability. The authors further investigated encapsulating A. cooperi somatic embryos in alginate to produce synthetic seeds (Ghosh and Sen, 1994). A conversion frequency of over 30% was achieved from encapsulated embryos by the combination of 3.5% sodium alginate (Sigma) and 50 mM calcium chloride whereas naked embryos converted at 45% under similar conditions. When embryos were stored at 4°C for up to 30 days, the germination frequency was higher for the naked embryos. However, encapsulated embryos stored for 60 and 90 days had a higher conversion frequency. In addition to studies that attempt to optimize the production and development. of asparagus somatic embryos, somatic embryos or embryogenic cultures have themselves been used to optimize other technologies or facilitate the recovery 22 of genetically unique plants. Cryopreservation systems were developed to store asparagus somatic embryos and embryogenic cultures over long periods of time (Uragami et a1. 1989; Nishizawa et al. 1992; 1993). Somatic embryogenesis has been used successfully to recover plants from protoplasts (Kunitake and Mii 1990; Mukhopadhyay and Desjardins 1994a; 1994b; May and Sink 1995), haploid plants from anther and ndcrospore culture (Feng and Wolyn 1991; 1993; 1994), and transgenic plants from. Agrobacterium .Lnoculated. embryogenic calluses (Delbreil et al. 1993). Protoplast culture Bui-Dang-Ha and Mackenzie (1973) developed the first asparagus protoplast culture scheme in hopes of using it for mass producing valuable cultivars and for accepting foreign genetic: material for (xxx) improvement. Protoplasts were isolated from cladodes (fern tissue) from a single cultivar. In many cases, at least 30% of the initial protoplast population died vdthin time first three days. Those that survived achieved first division by eight days although this was quite variable 1J1 that some isolations did run: divide until 17 days. They reported that 4% of the populations would at least divide once but actual plating efficiency (PE) was not presented. In an attempt to enhance protoplast division, PGR's were added at various concentrations without success. Sustained divisions were only achieved after glutamine was 23 increased from 200 mg/l to 1000 mg/l. The authors mentioned that a reduced nitrogen source such as glutamine may be beneficial to growth whereas ammonium as the sole nitrogen source may be inhibitory. Selected colonies were placed on solid. MS medium. and adventitious shoot regeneration was achieved. A subsequent paper concentrated on plant regeneration from protoplast-derived calluses (Bui-Dang-Ha et al. 1975). Calluses incubated solely on 2,4-D or ‘NAA at (equimolar concentrations exhibited callus proliferation and prolific rooting. The addition of BA to either NAA or IAA initiated shoots while using zeatin or 2,4-D in any combination was ineffective. The combination of IAA and BA produced up to 90% shoot initiation in protoplast-derived calluses whereas up to only 20% shoot initiation occurred on NAA and BA. The authors also observed somatic embryogenesis from the calluses. If the calluses were grown on equimolar BA and NAA with 40 mg/l adenine sulfate for 6 - 8 weeks and then subcultured to medium lacking PGR’s, a friable callus formed at the margins of about 15% of the calluses. This new callus was composed of various stages of somatic embryos. The embryos required IAA and zeatin for germination into plants because embryos left on medium lacking PGR's failed to develop further. Histological evidence showed the similarity of the somatic embryos to zygotic asparagus embryos. 24 Elmer et al.(1989) utilized callus cultures from four asparagus genotypes to investigate protoplast isolation, culture and shoot regeneration parameters. Callus was derived from spears of genotypes Jersey Giant #8, Jersey Giant #14, E2 and A19 on six different media containing various combinations of NAA, 2,4-D, BA and/or kinetin. The greatest protoplast yield came from callus grown on 2.5 mg/l 2,4-D and 1.0 mg/l kinetin, euui 20 days following subculture versus 10 (n: 30 days. Jersey Giant #8 (JG8) was the only genotype whose protoplasts were capable of sustained divisions and colony formation. Protoplasts from the other three genotypes died within 3-4 weeks after plating. Protoplasts were cultured in liquid Kao and Michayluk (1981) medium (KM medium) containing several combinations and concentrations of NAA, 2,4-D and BA. JG8 protoplasts divided only in KM32 medium containing 1 mg/l 2,4-D aumi 0.5 mg/l BA” In addition, plating density was important because at densities below 5 x 104 protoplasts/ml, no divisions occurred whereas at 5 x 10‘1 to 105' plating efficiencies in) to 7.3% were achieved. For shoot regeneration, the protoplast colonies were moved to semi-solid media when they had grown to 0.5 mm in diameter. The regeneration medium contained 2iP, kinetin, BA, or zeatin at 0.1 or 1.0 mg/l in combination with 0.1 mg/l NAA. All of the cytokinins induced shoots in about 30% of the calluses depending (N1 the cytokinin concentration. However, the authors did observe differences in shoot growth depending on 25 the cytokinind BA produced slow growing dark green ferns whereas zeatin and 2iP promoted thin elongated ferns. Eight protoplast—derived plants were found to be aneuploid with 2n = 2x = 22 — 38 chromosomes. Dan and Stephens (1991) developed a protoplast culture system specifically for A. officinalis L. cv. Lucullus 234 that has resistance to fusarium (Stephens et al. 1989). They tested the effects of three culture methods (liquid, agarose layer, and bead cultures), growth regulators and osmoticum on plating efficiency and colony formation. In the first experiment, they found that more divisions occurred when protoplasts were embedded in agarose versus liquid medium, and that KM medium containing 0.5 mg/l NAA, 0.5 mg/l 2,4-D and 0.5 mg/l kinetin was superior to other PGR combinations. Reducing the osmoticum from 800 mOsmol/kg to 360 increased the plating efficiency from 1.08 to 19.12%. Colony formation was also well supported at the lower osmoticum. They suggested that maintaining a high osmoticum may impair protoplast growth and metabolism and that the agarose protects the fragile cells and permits a lower osmoticum to be used. Following this, they re-evaluated the effect of PGR’s on plating efficiency using protoplasts embedded in agarose beads in the lower osmotic medium. NAA, 2,4-D and kinetin were used in concert at 0.25, 0.5 or 1.0 mg/l, and the plating efficiency reached 19% with all PGR’s at 0.5 mg/l. The protoplast-derived calluses were transferred to shoot regeneration medium containing various 26 combinations of NAA, 2,4-D, kinetin, zeatin and BA for four weeks and then to medium lacking hormones until shoots developed. The most effective medium contained 0.25 mg/l BA, 0.125 mg/l NAA and 0.125 mg/l 2,4—D and induced shoots in 92% of the calluses tested. Twenty eight plants were transferred to the greenhouse and 20 survived. The studies by Elmer et al.(1989) and Dan and Stephens(1991) regenerated plants from. protoplast—derived calluses \fija adventitious shoot formation. The following reports contrast those where plants were regenerated via somatb: embryogenesis. Kunitake and bMJ. (1990) isolated protoplasts from embryogenic callus that had been initiated from spear sections of A. officinalis L. cv. Mary Washington (n1 1 mg/l 2,4-D. They found that they could dramatically increase the viability and protoplast yield by pretreating the callus for 4-7 days in MS medium lacking PGR’s. The viability of the pretreated protoplasts was greater than 95% as determined by staining with fluorescein diacetate (FDA). The protoplasts were adjusted In) 1. x 105 /nd in MS medium containing several combinations of NAA, 2,4-D, BA and zeatin, different osmotic agents, with or without 1000 mg/l glutamine and solidified vnlfli 0.1 Gellan gum“ They found tflua most beneficial combination of factors for cell divisions of the embryogenic protoplasts was 1 mg/l NAA, 0.5 mg/l zeatin, 1000 mg/l glutamine and 0.6 M glucose as the sole osmoticum. This medium produced a plating efficiency of over 7% whereas any 27 other combination did not yield PE’s above 0.5%. After 40-50 days of culture, the protoplast-derived colonies were placed on % MS medium with 1% sucrose and lacking PGR’s. After about two weeks friable callus composed of various stages of embryos developed. Pretreating the embryos in distilled water for one week was critical for germination which occurred on 1’2 MS medium with 1 mg/l IBA, 1 mg/l GAy % sucrose and 0.2% Gellan gum” Embryos that were not pmetreated rarely germinated. Twelve plants were moved to the greenhouse and all had 2n = 2x = 20 chromosomes. Mukhopadhyay and Desjardins (1994a; 1994b) wanted to develop a simplified and reliable protocol for culturing embryogenic protoplasts from two genotypes of asparagus and regenerate plants. via jprotoplast-derived. somatic embryos. Similar to other studies, they investigated the effects of culture media, culture modes, plating densities and carbon sources on plating efficiency. Embryogenic callus, used as the source of protoplasts, was initiated from young spears of A. officinalis genotypes 6203 and 6171 on MS medium with 1 mg/l 2,4-D. The callus was subcultured every four weeks until a friable callus containing various stages of embryos developed. Their first experiment investigated the plating efficiency of 6203 and G171 protoplasts cultured in liquid or semisolid (0.1% Gelrite) KM medium or % strength MS medium with 1 mg/l 2,4-D and 0.5 mg/l BA or 1 mg/l NAA and 0.5 mg/l zeatin“ KM medium failed to pmoduce plating efficiencies 28 above 0.25% for either genotype whereas MS medium produced PE’s from 2.3 to 12.5%, and 1 mg/l NAA and 0.5 mg/l zeatin was more beneficial than 1 mg/l 2,4—D and 0.5 mg/l BA. As observed in other studies, the immobilized protoplasts (0.1% Gelrite) had a significantly higher PE than those in liquid culture only. They also observed consistently higher PE's with genotype G203 than 6171. In a second experiment, they compared the effect of inoculum density and osmoticum on PE with both genotypes. The protoplast densities for this study, 104, 5 x 10% JIP, and 2 x 105 and the osmoticum carbohydrates and levels, 0.1 M sucrose, 0.6 M glucose, and 0.7 M mannitol were used because they were optimal in other asparagus protoplast studies (Kong and Chin 1988; Elmer et al. 1989; Kunitake and DMJ. 1990; Mukhopadhyay and Eesjardins 1994). Similar to the other reports, they found the optimal protoplast culture density was 1 x 105 (Elmer et al. 1989) and the highest PE was in protoplasts cultured with 0.6 M glucose as the osmoticum (Kunitake and Mii 1990). Genotype G203 again produced higher PE’s than. G171 in each treatment. 'The protoplast-derived calluses were placed. on semi-solid. MS medium with 1 mg/l 2,4-D and rapidly grew into friable and nodular calluses of an embryogenic nature. After four weeks, the embryogenic callus of G203 was subcultured to E MS medium lacking PGR's with 1% sucrose for embryo development. Of the embryos that developed, approx. 40% showed normal development and. when those embryos were transferred 133 La MS medium 29 containing 1 mg/l GA3, 1% sucrose and 0.2 % Gelrite, normal roots and shoots developed. Cytology was performed on 10 plants and each was found to have 2n = 2x = 20 chromosomes. Plant regeneration for genotype G171 was not mentioned. Hsu et al. (1990) investigated several parameters for protoplast isolation and culture from asparagus suspension cultures since suspensions can provide large uniform populations of cells. Suspension cultures were initiated from anther—derived callus of A. officinalis cv. UC New Dwarf 5. One month-old callus was macerated and placed in 25 nfl.<1f liquid MS medium with 2 mg/l 2,4-D and agitated at 150 rpm under dim light. The suspensions were subcultured weekly by placing 5 ml of culture into 20 ml of fresh medium. Protoplasts were isolated 2, 4, 6 and 8 days after the 12th subculture to determine the effect of culture age on protoplast yield. Yield was greatest at 2 and 4 days with 42% and 50% of the cells yielding protoplasts, respectively. However, yield dropped off dramatically at 6 and 8 days with 15% and 5.8% yields, respectively. The reduction in yield may have been due to the presence of secondary cell walls that form as cell growth slows and are consequently more difficult to digest. Optimal protoplast yield was obtained at 4 days post subculture for up to 24 generations, although PE decreased with the age of the culture. They also found that protoplast yield. was more than double when substituting glucose for mannitol as the osmoticum in the digestion 30 solution. Protoplasts were cultured at 1 x 105hnl in liquid KM medium. When colonies reached 0.1 to 0.3 mm in dia. they were placed on semi-solid MS medium with 2 mg/l 2,4—D until a friable embryogenic callus was produced. Embryos were then transferred to MS medium with 0.3 mg/l NAA and 0.1 mg/l kinetin for germination to plants. Two similar studies examined the use of a polypropylene membrane as a support system to enhance asparagus protoplast culture (Kong and Chin 1988; Chin et al. 1988). Protoplasts were isolated from. 5-6 day-old cell suspension cultures initiated from asparagus seedling-derived callus. The choice of osmoticum in the digestion solution affected speed in which first divisions occurred] Protoplasts digested iJI glucose divided faster than when digested in mannitol or sorbitol. They also found that the method of culture significantly affected divisions. Protoplasts failed to divide in liquid droplets alone although when the droplets were placed on the membrane floating on liquid medium, a plating efficiency of 1- 5% was observed. Protoplasts embedded in agarose droplets produced around 1% divisions and the plating efficiency rose to at least 10% when the agarose was placed on the membrane system. The highest plating efficiency was achieved at 1 x IIP protoplasts/ml. The authors stated an advantage of using a porous membrane, besides enhancing cell division from protoplasts, is that the membrane can be easily handled and 31 placed on fresh medium without disturbing the protoplasts or developing colonies. A distinctive disadvantage is that the protoplasts can not be monitored with a microscope due to the membrane. Transformation Asparagus sp. are susceptible to infection by Agrobacterium tumefaciens. The lack of a wound response and the failure of the bacteria to attach to the plant cell wall are believed to be part of the reason why monocot species are recalcitrant to Agrobacterium transformation (Smith and Hood 1995). Asparagus, however, exhibits both of these requisites to transformation. The wound response in A. officinalis was studied using mechanically isolated mesophyll cells which are viable, damaged in a uniform manner and available in large numbers (Harikrishna et al. 1991). Two days after isolation, the cells underwent dramatic changes. fmua cells began to expand and divide by day five, respiration increased along with RNA and DNA synthesis, and several novel peptides had appeared which were not present in the unwounded tissue. These cytological, physiological and molecular changes are all characteristic of a wound response. Two studies documented the ability of A. tumefaciens to attaCh to asparagus mesophyll cells. Draper et al.(1983) examined several factors that effected attachment of Agrobacterium t1) isolated 14. officinalis mesophyll. cells. 32 Cell attachment was measured as the amount of plant cell aggregation or clumping that resulted from bacterial attachment. Increasing time, plant cell density and bacteria density all increased the rate of aggregatbmn. There was virtually no effect on aggregation when the mesophyll cells were fixed with glutaraldehyde. However, when the cells were heat killed, the rate of aggregation decreased. This indicated that viability was not required for attachment and that the heat treatment probably altered or cfisrupted the surface receptor molecules. Several strains of A. tumefaciens were studied and all wild-type strains could aggregate the cells. Moreover, when the same strains were cured of their Ti-plasmids, aggregation still occurred. This confirmed that the genes responsible for cell attachment lie on the bacterial chromosome. Terouchi et al. (1990) compared the ability of A. tumefaciens to attach to dicot and monocot cells by SEM. The .Agrobacterjtmi was observed. to .attach to ‘vinca, rice and asparagus cells at the same rate. Bacteria cured of their virulence plasmids attached at the same rate as wild-type strains confirming the results of the above study that the genes for cell attachment are chromosomal. Both studies concluded that bacterial attachment is not the limiting factor in Agrobacterium transformation of monocots or asparagus. Asparagus officinalis was the first monocot plant in which hormone-independent and opine—producing crown gall tissue could be isolated. Hernalsteens et a1. (1984) 33 inoculated young spear sections from seedling derived plants of A. officinalis cv. ‘Roem van Brunswijk' with the oncogenic A. tumefaciens strain C58. After incubating the explants on 1 mg/l BA and NAA, and 500 mg/l cefotaxime for four weeks, proliferating calluses arising from the sites of inoculation were placed on PGR—free medium. Some of the isolated calluses continued to proliferate on this medium. The authors reported that obtaining hormone-independent callus was reproducible but the frequency of such occurrences was about 10% based on total explants inoculated. Also, that the youngest or most apical portion of time spears typically gave ea positive response. Several of the hormone-autotrophic calluses were tested for the presence of opines. Each of the transformed cultures contained nopaline whereas the control cultures did not. This confirmed the first successful isolation of transgenic tissue in a monocot or asparagus. Conner et al. (1988) investigated the ability of four oncogenic strains of IQ. tumefaciens tx> produce tumors cm seventeen genotypes of A. officinalis. Of the four strains, C58, A722, A281 and A4T, only C58 was capable of causing small raised tumors to form at the wound sites. Of the seventeen genotypes tested, only three produced nopaline positive tumors when strain C58 was used exclusively. Similar to the report by Hernalsteens et al.(1984), tumors were only observed on the upper portions of young spears. 34 Opine positive tissues have been isolated from two additional species of asparagus following inoculation with A. tumefaciens. Stems of A. sprengeri Regal and A. tetragonous Bresler were wounded and inoculated with wild-type strain A208. Both species produced nopaline positive tumors (Suseelan et al. 1987). Prinsen EH1 al. (1990) investigated time levels of endogenous PGR’s in hormone-autotrophic crown gall tissues of asparagus that had been transformed with wildtype A. tumefaciens strain C58. Genes 1 (iaaM) enui 2 (iaaH) from oncogenic A. tumefaciens code for enzymes in the IAA biosynthesis pathway and gene 4 (ipt) catalyzes the first step in cytokinin biosynthesis. The presence of these genes in hormone-autotrophic callus was confirmed. by Southern hybridization. In regard to auxin biosynthesis, there was virtually no enhanced IAA levels in the tumor cell lines compared to the untransformed callus. However, there was a significant increase in endogenous cytokinins detected. This indicated that hormone-autotrophic growth of asparagus crown gall cells is only dependent upon an active gene 4 for increased cytokinin levels. The first molecular evidence of T-DNA integration into a monocot plant was in A. officinalis (Bytebier et al. 1987). Hormone autotrophic tissues were produced as described by Hernalsteens et al. (1984) via transformation with oncogenic A. tumefaciens strain C58. Southern analysis of one of the 35 tumor tissues using several restriction enzymes and four probes spanning the T-DNA region, showed that the T-DNA was integrated into the asparagus genome and that no detectable deletions resulted from transfer or integration. Hence, the authors concluded that T-DNA could be stably integrated into monocots (asparagus) and tflem: the T-DNA 1J1 asparagus was identical to integration observed in dicots. In the second part of this study, the first reported transgenic asparagus plants transformed via Agrobacterium were produced and analyzed. Young spear sections were inoculated with the non- oncogenic A. tumefaciens strain C58C1 containing pGV3850::1103neo. This cointegrated vector possessed the NPTII gene for kanamycin resistance and the nopaline synthase gene. After one month, 2 mm slices of infected tissue were cultured on LS medium with 1 mg/l BA and 1 mg/l NAA and 200 mg/l glutamine. The explants were subcultured to the same medium after a month for one additional month of culture. The tissue was then transferred to the same medium containing 50 mg/l kanamycin. Three of 25 calluses isolated continued to grow on kanamycin containing medium. Nopaline was detected in each of the kanamycin resistant tissues. Two of the three resistant cell lines gave rise to transgenic plants after being subcultured to medium containing 40 mg/l adenine, 4 mg/l BA and 1 mg/l NAA. Southern analysis of three plants from the same callus line showed integration of the T-DNA into high- molecular weight DNA and that no major rearrangements occurred 36 during regeneration. The authors concluded that T-DNA integration in asparagus is comparable to that in dicots. Conner et al. (1988) tested the ability of four strains of wild-type A. tumefaciens to produce tumors on 17 genotypes of A. officinalis. From this study they determined that Agrobacterium strain C58 produced the most tumors on asparagus genotype CRD 157. ‘Based on these data, they attempted to produce transgenic plants of genotype CRD157 using the non- oncogenic version of time same strain, C58C1, containing pGV3850::1103neo. The cointegrate vector possessed both the NPTII anmi nopaline synthase genes. .D? vitro explants of CRD157 were inoculated with the bacteria and cocultivated on callusing medium (MS medium containing 200 mg/l glutamine, 1 mg/l NAA and 1 mg/l kinetin) for 2-3 days. Subsequently, the cultures were placed on the same medium containing 250 mg/l cefotaxime to prevent bacteria growth. Kanamycin selection (100 mg/l) was imposed after 10 days of culture. After seven months of culture on kanamycin, 3 of 125 explants produced resistant calluses. .All three of time cell lines produced nopaline. Two of the cell lines failed to continue to grow beyond five subcultures and the third one maintained stable kanamycin resistance and nopaline synthesis for over 18 months. Transgenic plants were regenerated from this last cell line and Southern hybridization confirmed the integration of the NPTII gene. 37 Long-term embryogenic callus of A. officinalis was used as a target for Agrobacterium-mediated transformation to determine ii? these cultures could kme efficiently used ti) produce transgenic material (Delbreil et al. 1993). Three long-term embryogenic cultures, L1, L2 and L3, were established from three male genotypes as described by Delbreil et al. (1994). The cultures were inoculated with A. tumefaciens strain C58 cemtaining p3SSGUSINT. The binary vector possesses the NPTII gene for kanamycin resistance and an intron containing GUS gene that can only be expressed in the plant cell. One gram of somatic embryos were inoculated in 5 ml of OD600 0.6 - 0.8 bacteria suspension for 15 minutes. The embryos were then blotted dry, placed.cn113 medium for cocultivation and incubated in darkness for 48 hr. Following cocultivation, the embryos were subcultured.ti> B selection medium containing 400 mg/l cefotaxime and 100 mg/l kanamycin. After two to three months on selection medium, resistant embryogenic lines began to develop from necrosed primary embryos. A total of 23 resistant cultures were isolated from the three original cell lines. Lines L1, L2, and L3 produced 14” 8 and 1. resistant line(s), respectively. The trans- formation frequencies achieved for each long-term embryogenic line based on resistant cell lines isolated per gram of embryos inoculated was 0.6, 4 and 1 for L1, L2 and L3, respectively. Only three of the kanamycin resistant lines from the L1 culture gave rise to plantlets out of all the 38 resistant lines isolated. All resistant cell lines and the resulting plantlets were positive for GUS expression. Embryo stage, cocultivation medium and timing of selection were compared in an effort to increase the transformation efficiency. The results showed that globular embryos were resistant to transformation compared to elongated embryos, solid medium was superior to liquid during cocultivation and that. applying' selection. pressure ‘within seven.1days after cocultivation was beneficial. Southern analysis confirmed the stable integration of the NPTII and GUS genes in four resistant cell lines and the integration of the NPTII gene in one transgenic asparagus plant. One and two sites of T-DNA integration were identified in the material tested similar to integration patterns found in dicot plants transformed with Agrobacterium. The overall transformation frequency was very low although time authors did rmfl: consider this a: limiting factor. Since each gram of material contained about 300 embryos, the low transformation frequency could be compensated for by increasing the amount of material inoculated. There has been one report of direct-gene-mediated transformation in asparagus (Mukhopadhyay and Desjardins 1994). Factors influencing transient and stable GUS expression following electroporation. of plasmid DNA into callus-derived protoplasts were studied. Field strengths at 500 and 750 V/cm were optimal for transient expression but reduced protoplast viability. Both heat shock and the 39 addition of polyethylene glycol to the electroporation solution enhanced transient and stable GUS expression. Two genotypes were tested and 6203 outperformed G171 under each experimental parameter. After the electroporated protoplasts had been cultured for 28 days, the colonies were subcultured to 50 mg/l kanamycin. Kanamycin resistant calluses developed within 2 — 3 weeks and were tested for stable GUS expression. anh one of the resistant colonies was also GUS positive. Although GUS positive/kanamycin resistant colonies were obtained, regeneration of transgenic plants was not reported. LIST OF REFERENCES 40 LIST OF REFERENCES Ammirato, P.V. 1987. Organization events during somatic embryogenesis. In: Plant tissue and cell culture. C.E. Green, D.A. Somers, W.P. Hackett and D.D. Biesboer (Eds.) Alan R. Liss, Inc. New York. pp. 57-81. Bailey, L.H., and E.Z. Bailey. 1976. Hortus 3rd. Macmillan, New York, 118-119. Broadhurst, P.G. 1996. Stemphylium disease tolerance in Asparagus officinalis L. Acta Hort. 415:387-391. Bui-Dang-Ha, D. , I.AH Mackenzie. 1973. Time division. of protoplasts from Asparagus officinalis L. and their growth and differentiation. Protoplasma 78:215-221. Bui-Dang-Ha, D., B. Norreel and A. Masset. 1975. Regeneration of Asparagus officinalis L. through callus cultures derived from protoplasts. J. Exp. Botany. 26:263—270. Bytebier, B., F. Deboeck, H. De Greve, M. Van Montagu and J.- P. Hernalsteens. 1987. T-DNA organization in tumor cultures and transgenic plants of the monocotyledon Asparagus officinalis. Proc. Natl. Acad. Sci. USA. 84:5345-5349. Chin, C.-K., Y. Kong and H. Pedersen. 1988. Culture of droplets containing asparagus cells and protoplasts on polypropylene membane. 1988. Plant Cell Tis. Org. Cult. 15:59- 65. Cohen, S.I., and F.D. Heald. 1941. A wilt and root rot of asparagus caused by Fusarium oxysporum (Schlecht.) Plant Dis. Rep. 25:503—509. Conner, A.J., M.K. Williams, S.C. Deroles and R.C. Gardner. Agrobacterium-mediated transformation of asparagus. In: K.S. McWhirter, R.W. Downes, B.J. Read (eds.) Proceedings of the 9th Australian Plant Breeding Conference, Wagga Wagga, N.S.W. Australia. 1988. pp. 131-132. 41 Dan, ‘Y., and.(ZJF. Stephens. 1991. Studies (n3 protoplast culture types and. plant regeneration from callus-derived protoplasts cxf Asparagus officinalis 1“. cv. Lucullus 234. Plant Cell Tis. Org. Cult. 27:321-331. Delbreil, B., P. Guerche and M. Jullien. 1993. Agrobacterium- mediated transformation of Asparagus officinalis L. long-term embryogenic callus and regeneration of transgenic plants. Plant Cell Rep. 12:129-132. Delbreil, B., I. Goebel-Tourand, C. Lefrancois and M. Jullien. 1994. Isolation and characterization of long-term embryogenic lines in Asparagus officinalis L. J. Plant. Physiol. 144:194- 200. Delbreil, B., and.bt. Jullien. 1994. Evidence for ;U7 vitro induced.:mutation. which improves somatic embryogenesis in Asparagus officinalis L. Plant Cell Rep. 13:372—376. Desjardins, Y. Micropropagation (n3 asparagus (Asparagus officinalis I“), In:Y.P.S. Bajaj (Ed.), Biotechnology in agriculture and forestry, Vol. 19. High-tech and micropropagation III. Springer-Verlag Berlin, 1992, pp. 26-41. Draper, J., I.A. Mackenzie, M.R. Davey and J.P. Freeman. 1983. Attachment of Agrobacterium tumefaciens to mechannically isolated asparagus cells. Plant Sci. Let. 29:227—236. Ellison, J.H., and D.F. Scheer. 1959. Yield related to brush vigor in asparagus. Proc. Am. Soc. Hort. Sci. 73:339-344. Evans, D.A., W.R. Sharp and C.E. Flick. 1981. In: T.A. Thorpe (ed.). Plant tissue culture: methods and applications in agriculture. Proceedings of UNESCO symposium, Sao Paulo, Academic Press, New York. pp. 45-113. Falavigna, A., P.E. Casali, and M.G. Tacconi. 1996. Advances in asparagus breeding following in vitro anther culture. Acta Hort. 415:137—142. Feng, X.R., and D.J. Wolyn. 1991. High frequency production of haploid embryos in asparagus anther culture. Plant Cell Rep. 10:574-578. Feng, X.R., and D.J. Wolyn. 1993. Development of haploid asparagus embryos from liquid cultures of anther-derived calli is enhanced by ancymidol. Plant Cell Rep. 12:281-285. 42 Feng, X.R., and D.J. Wolyn. 1994. Recovery of haploid plants from asparagus microspore culture. Can. J. Bot. 72:296-300. Franken, A.A. 1970. Sex characteristics and inheritance of sex in asparagus. Euphytica 19:277—287. Galli, M.G., M. Bracale, A. Falavigna, F. Raffaldi, C. Savini and A. Vigo. 1993. Edfferent kinds of male flowers in the dioecious plant Asparagus officinalis L. Sex. Plant. Reprod. 6:16—21. Ghosh, B., and S. Sen. 1991. Plant regeneration through somatic embryogenesis from spear callus culture of Asparagus cooperi Baker. Plant Cell Rep. 9:667-670. Ghosh, B., and S. Sen. 1994. Plant regeneration from alginate encapsulated somatic embryos of Asparagus cooperi Baker. Plant Cell Rep. 13:381-385. Haccius, B. 1978. Question of unicellular origin of non- zygotic embryos in callus cultures. Phytomorph. 28:74-81. Harikrishna, K., E. Paul, R. Darby and J. Draper. 1991. Wound response in mechanically isolated asparagus mesophyll cells: A model monocotyledon system. J. Exp. Bot. 42:791-799. Henderson, P. 1890. Handbook of plants. Peter Henderson and Co. New York. Hernalsteens, J.—P., 1h Thia-Toong, (I. Schell. and FL Van Montagu. 1984. An Agrobacterium-transformed cell culture from the monocot Asparagus officinalis. EMBO J. 3:3039-3041. Hus, J.Y., C.C. Yeh, T.P. Yang, W.C. Lin and H.S. Tsay. 1990. Initiation of cell suspension cultures and plant regeneration from protoplasts of asparagus. Acta Hort. 271:135-143. Johnston, S.A., J.K. Springer and G.D. Lewis. 1979. Fusarium moniliforme as a cause of stem and crown rot of asparagus and its association with asparagus decline. Phytopath. 69:778-780. Kohmura, PL, S. Chokyu and 11. Harada. 1994. [U1 effective micropropagation system using embryogenic calli induced from bud clusters in Asparagus officinalis L. J. Japan. Soc. Hort. Sci. 63:51-59. 43 Komamine, A., R. Kawahara, M. Matsumoto, S. Sunabori, T. Toya, A. Fujiwara, M. Tsukahara, J. Smith, M. Ito, H. Fukuda, K. Nomura, and T. Fujimura. 1992. Mechanisms of somatic embryogenesis in cell cultures: Physiology, biochemistry, and molecular biology. In Vitro Cell. Dev. Biol. 28:11—14. Kong, Y., and C.—K. Chin. 1988. Culture of asparagus protoplasts on porous polypropylene membrane. Plant Cell Rep. 7:67-69. Kunitake, 1L, enui M. Mii. 1990. Somatic embryogenesis and plant regeneration from protoplasts of asparagus (Asparagus officinalis L.). Plant Cell Rep. 8:706-710. Lake, R.J., P.G. Falloon. and D.W.M. Cook. 1993. Replant problem. and chemical components of asparagus roots. New Zealand J. Crop Hort. Sci. 21:53—58. Levi, A., and K.C. Sink. 1990. Differntial effects of sucrose, glucose and fructose during somatic embryogenesis in asparagus. J. Plant Physiol. 137:184-189. Levi, A., and K.C. Sink. 1991. Histology and morphology of asparagus somatic embryos. Hortsci. 26:1322-1324. Levi, A., and. K.C. Sink. 1991. Somatic embryogenesis in asparagus: the role of explants and growth regulators. Plant Cell Rep. 10:71-75. Levi, It”. and K.C. Sink. 1992. Asparagus somatic embryos: Production in suspension culture and conversion to plantlets on solidified medium as influenced by carbohydrate regime. Plant Cell, Tissue and Organ Cult. 31:115-122. Lewis, G.D., and P.B. Shoemaker. 1964. Resistance of asparagus species to Evsarium oxysporum f. asparagi. Plant Dis. Rep. 46:364-365. Loptien, H., 1979. Identification of the sex chromosome pair in asparagus (Asparagus officinalis L.). Z. Pflanzenzuchtg, 82:162-173. Luzny, J. 1979. The history of asparagus as a vegetable, the tradition of its growing in Czechoslovakia (CSSR) and prospect of its further propagation and breeding. In: Proc. 5th International Asparagus Symposium. G. Reuther, (ed.) Eucarpia Section Vegetables. Geisenheim Forschungsanstalt, Germany. pp. 17-21. 44 Mace, M.E., A.A. Belland, and C.H. Beckman. 1981. Fungal wilt diseases of plants. Academic press. New York. pp. 639. May, R.A., and K.C. Sink. 1995. Genotype and auxin influence direct somatic embryogenesis from protoplasts derived from embryogenic cell suspensions of Asparagus officinalis L. Plant Sci. 108:71-84. Mukhopadhyay, S., anmi Y. Desjardins. 1994a. IX comparative study on mode of culture and plant regeneration from protoplast-derived somatic embryos of two genotypes of Asparagus officinalis L. Plant Sci. 100:97-104. Mukhopadhyay, S., and Y. Desjardins. 1994b. Plant regeneration from. jprotoplast-derived somatic embryos of .Asparagus officinalis L. J. Plant Physiol. 144:94-99. Mukhopadhyay, S., and Y. Desjardins. 1994. Direct gene transfer to protoplasts of two genotypes of Asparagus officinalis L. by electroporation. Plant Cell Rep. 13:421- 424. Nichols, M.A. 1990. Asparagus: The world scene. Acta Hort. 271:25-31. Nishizawa, S., A. Sakai, Y. Amano and T. Netsuzawa. 1992. Cryopreservation of asparagus (Asparagus officinalis L.) embryogenic cells and subsequent plant regeneration by a simple freezing method. Cryo-Letters 13:379-388. Nishizawa, S., A. Sakai, Y. Amano and T. Netsuzawa. 1993. Cryopreservation of asparagus (Asparagus officinalis L.) embryogenic suspension cells and subsequent plant regeneration by vitrification. Plant Sci. 91:67—73. Nonnecke, I.L. Vegetable Production. Van Nostrand Reinhold, New York. 1989. pp. 570—580. Odake, Y., A. Udagawa, H. Saga and M. Mii. 1993. Somatic embryogenesis of tetraploid plants from internodal segments of a diploid cultivar of Asparagus officinalis L. grown in liquid culture. Plant Sci. 94:173—177. Peirce, L.C. Vegetables, characteristics, production and marketing. John Wiley and Sons, New York. 1987. pp. 183-189. 45 Prinsen, E., B. Bytebier, J.-P. Hernalsteens, J. De Greef and H. Van Onckelen. 1990. Functional expression of Agrobacterium tumefaciens T-DNA onc-genes in asparagus crown gall tissues. Plant Cell Physiol. 31:69—75. Reinert, J. 1958. Morphogenese und ihre kontrolle an gewebekulturen aus carotten. Naturwissen. 45:344-345. Reuther, G. 1977. Adventitious formation and somatic embryogenesis in callus of asparagus and iris and its possible applications. Acta Hort. 78:217-224. Reuther, G. Asparagus. Handbook of Plant Cell Culture. Vol. 2. pp. 211—242. R.S. Sharp, D.A., Evans, P.V. Ammirato, and Y. Yamada (eds.) Macmillan Publishing Co. New York, 1984, pp. 211-242. Saito, T., S. Nishizawa and S. Nishimura. 1991. Improved culture conditions for somatic embryogenesis from Asparagus officinalis L. using an aseptic ventilative filter. Plant Cell Rep. 10:230-234. Scholten, C.T.J., and P.H.G. Boonen. 1996. Asparagus breeding in the Netherlands. Acta Hort. 415:67-70. Sharp, W.R., M.R. Sohndahl, A.E. Evans, L.A. Caldas and S.B. Maraffa. 1980. The physiology cflf in vitro .asexual embryogenesis. Hort. Rev. 2:268-310. Smith, R.H., and E.E. Hood. 1995. Agrobacterium tumefaciens transformation of monocots. Crop Sci. 35:301-309. Stephens, C.T., R.M. De Vries and K.C. Sink. 1989. Evaluation of Asparagus species for resistance to Fusarium oxysporum f. sp. asparagi and F. moniliforme. Hortsci. 24:365-368. Sturtevant, A.H. 1919. Sturtevant's notes on the edible plants. U.P. Hedrick (ed.) J.B. Lyon Company, Albany, New York. Steward, F.C., M.O. Mapes and K. Mears. Growth and organized development of cultured cells. II. Organization in cultures grown from freely suspended cells. Am. J. Bot. 445:705-708. Steward, F.C., and M.O. Mapes. 1971. Morphogenesis and plant propagation in aseptic cultures of asparagus. Bot. Gaz. 132:70-79. 46 Suseelan, K.N., A. Bhagwat, H. Mathews and C.R. Bhatia. 1987. Agrobacterium tumefaciens-induced tumour formation (N1 some tropical dicot and monocot plants. Cur. Sci. 56:888-889. Takatori, F., and F.D. Souther. (1978). Asparagus Workshop Proceedings. Dept. of Plant Sciences, Univ. of Calif., Riverside. pp. 100. Terouchi, N., S. Hasezawa, H. Matsushima, Y. Kaneko and K. Syono. 1990. Observation by SEM of the attachment of Agrobacterium tumefaciens to the surface of vinca, asparagus and rice cells. Bot. Mag. Tokyo 103:11-23. Tisserat, B., E.B. Esan and T. Murashige. 1979. Somatic embrogenesis in angiosperms. Hort. Reviews. 1:1-77. Uragami, EL, .A. Sakai, 1L Nagai and if. Takahashi. 1989. Survival of cultured cells and somatic embryos of Asparagus officinalis cryopreserved by vitrification. Plant Cell Rep. 8:418-421. Williams, E.G., and G. Maheswaran. 1986. Somatic embryogenesis: Factors influencing coordinated behavior of cells as an embryogenic group. Ann. Bot. 57:443-462. Wilmar, C., and M. Hellendoorn. 1968. Growth and morphogenesis of asparagus cells cultured in vitro. Nature 217:369-370. Chapter Two Genotype and auxin influence direct somatic embryogenesis from protoplasts derived from embryogenic cell suspensions of Asparagus officinalis L. ABSTRACT Embryogenic callus from four asparagus genotypes, Jersey Giant No. 8, MD10, Rutgers 22, and 86SOM1 was simultaneously initiated from spear explants on semisolid LS medium containing 5 uM 2,4-D or 50 uM NAA. Calluses were used to initiate cell suspensions in liqubd LS medimn of the same composition. The eight sets of cell suspensions were used as protoplast donors at both two and five months of age. Protoplasts were immobilized at lOSAnl density in MS medium with 0.6% agarose and overlaid with liquid KM medium; both containing the same type and concentration of auxin used for the corresponding donor cells or with plant growth regulator- free (PGR-free) medium. There was a significant interaction between genotype, suspension auxin, and inclusion or exclusion of PGRs in the protoplast culture media on plating efficiency, and colony and somatic embryo formation. Plating efficiencies at 14 days ranged from.C)- 40%. Globular somatic embryos developed directly from protoplasts in 10 - 14 days and bipolar embryos could be transferred in 3 - 4 weeks to embryo maturation medium (EM medium) composed of LS medium with 2% sucrose and 1% Phytagel. Conversion to plants occurred as rapidly as 1 - 2 weeks after transfer to EM medium or 5 - 6 weeks after initial protoplast culture. Although all four genotypes regenerated plants, Rutgers 22 had the highest conversion frequency at 42%. Most plants recovered from the 2,4-D-derived protoplasts were karyotypically aberrant while a higher frequency of normal plants were obtained from the NAA-derived protoplast cultures. 50 Introduction Asparagus, (Asparagus officinalis L.), is an economically important crop cultivated throughout the world for its edible spears (Nichols 1990). Genetic improvement of asparagus by somatic Ihybridization. and 'transformation. via Hon oaumEOm Ho moflcoHoo mo monaszm .Houuo cumccmum .H IddNMOHMIMHMdadWI poHMCHEHoO om H o H N «<2 aaz o mcoz «<2 G H GN mH mH o on QIH.N ouv.N o mcoz ona.N OHoz 0 HH mH om «Hz «Hz 0 m NN OH mcoz «Hz 0 G o m QIH.N QIH.N o mcoz o-s.N HZOme 0 «Hz «Hz 0 o om om 6:02 «Hz N H N H 0 me om one.N ouv.N o o om om mcoz QIH.N N on s H mm HH mH o OH aaz «AZ A H NH HN ON 0 om mcoz «<2 N H N H a me on 9-4.N Q-H.N o o N N 6:02 onv.N NN mumsusm H V Hooam cho HHHmo AH mom mom masuocmo coemmm>coo cam uoom Doom ummamououm coemcommcm Eouw mo>HQEo oaumEOm co>fluoclummaaououm mo mucmam on .moHSDHSU coamcmmmsm pHounucoEIN coeumuo>coo cam coeumCHEHow .m canoe 67 cultures such as JG 8 and 86SOM1 had a high degree of callus formation and low embryo conversion (Table 3). Plant regeneration occurred for emMiI genotype from. protoplasts derived from 5-month-old suspensions, however, conversion was lower overall compared to those of the 2-month-old cultures (Table 4). Conversion of Rutgers 22 NAA (-)PGR embryos decreased from 42% to 19%, and from 26% to 19% for MD10 2,4-D (+)PGR. Although JG 8 NAA (-)PGR protoplasts had the highest PE for both 2- and 5-month-old suspensions, mature embryos were not produced. Some protoplast—derived somatic embryos from Rutgers 22 and MD10 converted into plantlets as rapidLy as two weeks after subculture to EH4 mediuml (Fig. 1F). However, the majority of embryos required a longer time to mature (4 - 5 wks) due primarily to the presence of less developed stages at the end of protoplast culture. These embryos subsequently germinated upon subculture to PDM medium and could be transferred to the greenhouse as rapidly as three months after initial protoplast isolation. Comparison of embryo conversion from donor cell suspensions and their derived protoplasts Conversion frequency was greater for each genotype for the suspension cell-derived embryos; however, Rutgers 22 NAA and 2,4-D, aumi MD10 2,4-D cultures were rmfl: significantly different from the protoplast-derived cultures (Table 5). 68 .coHumoHHdoH Hog cmumHQ ma Ho ESEmeE m rues mo>HQEo UHHMEOm Ho mchoHoo wo HoQEszm .HOHHo Unaccepm 1H o 0 HH NH «<2 «<2 H H N N NH HG GN 6202 «<2 H H NH N NN o NH NIHIN ouH.N NN H N o H 6:02 NIH.N oHaz 0 «<2 «<2 0 o Nm Nm mcoz «<2 N H H H mm mN GN NIH.N NIH.N o N mN GN mcoz NIH.N Hzommm 0 «<2 «<2 0 0 SN GN 6202 «<2 m H N N NN Hm ON NIH.N ouH.N o 0 GH NH 6:02 ouH.N N on o ««2 ««2 H H NH NH NH G GN 6:02 «<2 H H N N Nm NH SN N-H.N NIH.N G H OH O GH 6202 NIH.N NN mHmNHsm ANV Hoonm NHHco HHHmo A2 «om mom wasnocmo coemno>coo pcm uoom uoom unmadououm conconsm Imdxmmauldwmmadm poumcHEHoo .moucuaso QOHmconSm odoueucoEim Eouw mo>HbEw 0HumEOm oo>HHoclummamououd mo mucmad ou coamuo>coo cam coflumcHEHoo .H canoe 69 Both 86SOM1 2,4-D and NAA suspension cells had higher embryo conversion than from protoplasts. JG 8 2,4-D and MD10 NAA suspensions had significantly more embryos that converted into plantlets than.:fimmn their protoplasts. The rmxfi: notable difference was observed between MD10 NAA cultures with suspension-derived embryos converting at 58% compared to 8% for protoplast-derived embryos. Protoplasts from JG 8 failed to regenerate plants and the donor suspensions also proved to be non-embryogenic. Table 5. Comparison of conversion frequency of somatic embryos from 5- month-old suspension culture cells with their derived protoplasts. Protoplast Suspension derived derived Suspension Protoplast Conversion Conversion Genotype PGR PGR N (%) N (%) t-test Rutgers 22 2,4-D 2,4—D 96 9 i l 80 13 i 4 ns NAA None 96 19 i 4 80 21 i 3 ns JG 8 2,4-D 2,4—D 80 9 i 3 80 33 i 4 ** NAA None 96 0 0 O 86SOM1 2,4-D 2,4-D 96 4 i 2 80 13 i 3 * NAA None 32 0 80 10 i 3 MD10 2,4-D 2,4-D 48 19 i 4 80 20 i 4 ns NAA None 96 8 i 4 80 58 i 7 ** NS, *, ** Significant at P > 0.05, and P < 0.05 and 0.01, respectively. i, standard error. Cytology Cytology was performed on a maximum of five protoplast derived-plants from each culture-regeneration treatment (Table 6). In general, plants regenerated from NAA cultures had a higher frequency of normal karyotypes (2n=20) than those from 70 .ON u Cm mH od>po>umx HmEHoza ON N-OHO2 ON N-OHO2 NN mcoz «<2 HIOHoz OH NIOHOE OH HIOHOE HNINN N-OHO2 OO NnoHoz ON O-H.N OIH.N HuoHoz ON NquomON ON HIHEONON OO NIHZONON ON NIHzoNON NN OIH.N OIH.N HquoNON ON HIN ON ON NIN ON ON «<2 «<2 H-OHO2 ON NIN OO O0-0N N-OHOz NN O-H.N O-H.N Hum on ON-NH HIOHoz ON NINN mHmNHsm ONIHN NIOHOz ON HINN mHmOHsm NNINO NnoHoz ON NuNN mHmOHsm OOIHN OIH.N O-H.N HIOHOE ON NINN mHmNHsm NN OIH.N O-H.N HIN ON ON 6202 «<2 H-NN mHmOHsm ON «<2 «<2 NuNN mHmOHem NN NINN mHmNHsm ON «<2 «<2 HlNN mHmOHsm NNION HINN mHmOHsm ON HuNN Humousm NN NINN mHmOHsm ON NINN mHmOHDN HH-NN NINN mHmNHsm ON NINN mHOOpdN ON-OH OIH.N O-H.N HINN mHmNHsN HN 6202 «<2 H-NN mHmOHsm ON-OO mcoz OIH.N HuNN mHmOHsm ON-NH mcoz OuH.N HINN mHmOHsm .oc mom mum ucmuocooom .oc mwm mom ucmumcomom oEomoEOHLU ummamouonm conconsm mEomoEouno ummaaouonm conchmsm mconcoomSWIoHoucucoelm WCOchonsw UH0|CucoEIN a.moH5uHso conconcm paOunucoelm cam IN Eoum cm>HHoo mmcoHoouonm wo mHmQESC oEomoEoubo .m canoe 71 2,4-D cultures. All plants except one (86SOM1-5) derived from protoplasts of 2,4-D suspensions were aneuploid, polyploid or mixoploid. Culture age (Uri not appear ti) be a”) influence since cytogenetically normal and aberrant plants were regenerated from the same donor suspensions at two or five months. Phenotypic differences between regenerated plants from 2,4-D or NAA derived protoplasts were observed in vitro primarily in regenerants of MD10. MD10 NAA-derived plants were vigorous in both shoot and root production compared to 2,4-D-derived plants that produced short thickened shoots that elongated poorly and had little or no root formation (Fig. 2). Discussion Factors such as protoplast plating density (1 x 105/m1), osmoticum (glucose) and. plating technique (agarose) have already been evaluated for asparagus protoplasts by others (Kong and Chin 1988; Elmer et al. 1989; Kunitake and Mii 1990; Dan and Stephens 1991) and these optimal parameters were incorporated into our study. However, the effect of genotype and auxin had not been examined and showed in our study to strongly influence protoplast culture and plant regeneration. Protoplast yields from the cell suspensions with the exception of JG 8 NAA were fairly low, particularly those from the 2,4-D cultures. The isolation procedure was the probable cause of the low yields because most of the starch filled 72 Fig. 2. Phenotypic differences in 2,4-D- and NAA-derived protoclones of MD10. The rooting response and shoot development from the 2,4—D-derived plants (left) was poor compared to the vigorous root and shoot growth from the NAA— derived regenerants (right). 73 protoplasts pelleted instead of floating. Poor recovery of protoplasts from embryogenic cells due to sucrose floatation was similarly observed in Brassica and MUsa cultures (Millam et al. 1991; Panis et al. 1993). Both studies were able to increase yield by sieving only and eliminating floating, and in subsequent experiments, we found that sieving isolations alone through 61 um and 35 um mesh routinely yielded 4 - 6 x 106 protoplasts Emu: gram (fl? cells compared ti) about 1. x 106/gram after floatation. Protoplasts of all fOur genotypes exhibited somatic embryogenesis although the degree of response was dependent on the genotype, donor cell PGR, and the presence or absence of the same PGR in the protoplast culture medium. In the study by Elmer 6H: al.(1989) only protoplasts of (K3 8 sustained divisions whereas protoplasts from three other genotypes, different from those used herein, failed to divide. In that study, 2,4-D was optimal for both donor callus and protoplast culture. J.G. 8 protoplasts cultured herein in 2,4-D achieved comparable PEs although frequency of plant regeneration via somatic embryogenesis was lower than previously reported via organogenesis (Elmer et al. 1989). The two PGR regimes tested in this study were based on previous asparagus protoplast and somatic embryogenesis studies. 2,4-D at 5 (m4 has been commonly used ti) induce somatic embryogenesis (Levi and Sink 1991; Saito et al. 1991) 74 and was optimal for protoplast culture in a previous report from our laboratory (Elmer et al. 1989). NAA also induced embryogenesis and 54 uM NAA produced a higher frequency of normal somatic embryos from callus cultures that subsequently converted into plants compared to embryos induced with 5 uM 2,4-D (Levi and Sink 1991). In addition, NAA was more effective than 2,4-D in inducing divisions in protoplasts of cv. Mary Washington (Kunitake and Mii 1990). Five uM 2,4-D enii 50 uM NAA were selected because both are effective in establishing embryogenic cell suspensions of asparagus (Saito et al. 1991; Levi and Sink 1992). Embryo conversion was highest for Rutgers 22 from NAA cultures, from both 2,4-D and NAA for MD10, and exclusively from 2,4-D derived protoplasts from JG 8 and 86SOM1. JG 8 and 86SOM1 are both selections from the seed variety Jersey Giant, and although regeneration frequencies were similar for both, PE and colony formation differed significantly illustrating the heterogeneity of even closely related genotypes. Treatments in which media lacking PGRs were included to potentially expedite somatic embryogenic development. Our hypothesis being that since the release of embryogenic potential typically occurs after auxin is removed from embryogenically determined cells, protoplasts derived from an embryogenic cell culture may respond similarly. This was the case since embryos developed directly from protoplasts in a rapid manner and without an intermediate callus stage. 75 Somatic embryogenesis from asparagus protoplasts has been previously reported to occurred after an initial callus stage and plantlet formation required 3 - 4 months (Hsu et al. 1990; Kunitake and Mii 1990). Direct somatic embryo formation from protoplasts has been observed in a number of species using embryogenic calli or suspension cultures as the protoplast donor sources (Roest and Gilissen 1989; 1993). For these species, isolated protoplasts are able to maintain their cell polarity and restore it rapidly upon culture (Roest and Gilissen 1993). This response was observed previously in asparagus protoplasts although embryogenesis through a callus phase was emphasized since only a few direct embryos formed in comparison to cell colonies (Kunitake and Mii 1990). NAA appeared to potentiate cells more than 2,4-D to give rise ti) embryogenically determined pmotoplasts capable of developing directly into somatic embryos particularly in PGR- free medium. Conversely, 2,4-D-derived protoplasts responded poorly in (-)PGR medium and the inclusion of 2,4-D was required for satisfactory divisions and for enhanced embryo formation. Similarly, protoplasts from 2,4-D derived embryogenic suspensions of Panicum maximum also produced relatively few divisions and exhibited no embryo formation in medium. 00 PGR compared ti) those cultured 1J1 2,4-D which promoted vigorous cell divisions and globular somatic embryos (Lu et al. 1981). 76 We compared embryo conversion from the 5-month-old suspension cells and from their derived protoplasts to determine if the protoplast culture scheme had an effect on plant regeneration. As well, to determine the regeneration ability of the donor suspensions which had been advised prior to protoplast isolation (Ahmed and Sagi 1993). Seven of the eight cell suspensions were capable of plant regeneration and their derived protoplasts either maintained totipotency or exhibited a significant reduction. The JG 8 NAA protoplasts rapidly divided into colonies similar to those from which they were isolated but plant regeneration did not occur. Unlike other cultures, the JG 8 NAA suspension cells lacked starch which has been related to low regeneration potential (Hirosawa, 1992). The majority of cytogenetic variability was observed in 2,4-D derived protoclones; although, a direct comparison of the effects of 2,4-D and NAA on such variability cannot be made since equal PGR concentrations were not used. However, our observations on somaclonal variation should be taken into consideration when using the same levels of PGRs and culture protocol as listed in this study. There are reports of plants with normal karyotypes regenerated from asparagus protoplasts when 2,4-D was either in the donor culture medium (Kunitake and Mii 1990) or in the protoplast culture medium (Kong and Chin 1988). However, Elmer et al. (1989) regenerated only aneuploid JG 8 plants when donor callus and protoplasts were 77 cultured in 2,4-D. Our results corroborate with Elmer et al. (1989) since only aneuploid J.G. 8 plants were recovered when 2,4-D was used exclusively at the same level. Two observations indicated that cytogenetic variability probably existed iii the donor cells. Firstly, two pdants regenerated from 2,4-D derived Rutgers 22 protoplasts cultured in the absence of 2,4-D had the same amount of variation as from the same protoplasts cultured in the presence of 2,4-D. Plants regenerated from MD10 suspensions were phenotypically identical to those derived from. protoplasts; NAA-derived plants were vigorous and rooted well and those from the 2,4-D suspension were stunted and rooted poorly similar to their protoplast counterparts. Fitter and Krikorian (1988) found considerable somaclonal variation among Hemerocallis plants regenerated thiin embryogenic suspensions emui their derived protoplasts compared with organogenic cultures. They concluded that the majority of the variation arising in the suspension and protoplast-derived plants originated from the cytogenetically heterogeneous donor cells. All four asparagus genotypes were capable (n5 plant regeneration from protoplasts through our system. Potentially, other genotypes could also be regenerated if a source of embryogenic donor cells can be established. Developing a protoplast culture system for a given genotype that favors direct somatic embryogenesis may aid direct-gene- 78 transfer and somatic hybridization studies by hastening plant regeneration and decreasing the time in culture. LIST OF REFERENCES 79 LITERATURE CITED Ahmed, K.Z., and F. Sagi. 1993. Culture of and fertile plant regeneration from regenerable embryogenic suspension cell- derived protoplasts of wheat (Triticum aestivum L.). Plant Cell Rep. 12:175-179. Chin, C.-KJ, Y. Kong' and. H. Pedersen. 1988. Culture of droplets containing asparagus cells and protoplasts on polypropylene membrane. Plant Cell Tiss. Org. Cult. 5:59-65. Dan, Y., anui C.T. Stephens. 1991. Studies (n5 protoplast culture types and. plant regeneration from. callus-derived protoplasts (If Asparagus officinalis IL. cv. Lucullus 234. Plant Cell Tiss. Org. Cult. 27:321-331. Elmer, W.H., T. Ball, M. VOlokita, C.T. Stephens and K.C. Sink. 1989. Plant regeneration from callus-derived protoplasts of asparagus. J. Amer. Soc. Hort. Sci. 114:1019-1024. Frearson, E.M., J.B. Power and E.C. Cocking. 1973. The isolation, culture enui regeneration (If petunia leaf protoplasts. Dev. Biol. 33:130-137. Fitter, M.S., and A.D. Krikorian. 1988. In: F.A. Valentine (Ed.), Forest and Crop Biotechnology: Progress and Prospects. Springer—Verlag, Berlin. pp. 242-256. Guri, A., M. Volokita and K.C. Sink. 1987. Plant regeneration from leaf protoplasts of Solanum torvum. Plant Cell Rep. 6:302-304. Bui Dang Ha, D., and I.A. Mackenzie. 1973. The division of protoplasts from Asparagus officinalis L. and their growth and differentiation. Protoplasma 78:215-221. Bui Dang Ha, D., B. Norreel, and A. Masset. 1975. Regeneration of Asparagus officinalis L. through callus cultures derived from protoplasts. J. Expt. Bot. 26:263-270. 80 Hsu, J.Y., C.C. Yeh, T.P. Yang, W.C. Lin and H.S. Tsay. 1990. Initiation of cell suspension cultures and plant regeneration from protoplasts of asparagus. Acta Hort. 271:135-143. Jamieson, J.L., T.Y. Slimmon. and IL. Tiessen. 1985. Time required to establish tissue culture clones. In: E.C. Lougheed and H. Tiessen (Eds.), Proceedings of the Sixth International Asparagus Symposium, University of Guelph. pp. 89-96. Hirosawa, T. In vitro mass propagation of rice. 1992. In: K. Kurata anmi'T. Kozai (Eds.), Transplant Production Systems. Kluwer Academic Publishers, Netherlands. pp. 195-212. Kao, K.N., and M.R. Michayluk. 1981. Embryoid formation in alfalfa cell suspension cultures from different plants. In Vitro. 17:645-648. Kong, Y., and C.-K. Chin. 1988. Culture of asparagus protoplasts on porous polypropylene membrane. Plant Cell Rep. 7:67-69. Kunitake, EL, evil M. Mii. 1990. Somatic embryogenesis and plant regeneration from protoplasts of asparagus (Asparagus officinalis L.). Plant Cell Rep. 8:706-710. Lai, F.-M., and B.D. McKersie. 1993. Effect of nutrition on maturation of alfalfa (Medicago sativa L.) somatic embryos. Plant Sci. 91:87—95. Levi, A., and K.C. Sink. 1991. Histology and morphology of asparagus somatic embryos. Hortsci. 26:1322-1324. Levi, .A., eumi K.C. Sink. 1991. Somatic embryogenesis in asparagus: the role of explants and growth regulators. Plant Cell Rep. 10:71-75. Levi, AH, enmi K.C. Sink. 1992. Asparagus somatic embryos: Production in suspension culture and conversion to plantlets on solidified medium as influenced by carbohydrate regime. Plant Cell Tiss. Org. Cult. 31:115-122. Linsmaier, E.U., and F. Skoog. 1965. Organic growth factors requirements of tobacco tissue cultures. Physiol. Plant. 18:100-127. 81 Lu, C.-Y., V. Vasil and I.K. Vasil. 1981. Isolation and culture of protoplasts of Panicum maximum Jacq. (Guinea Grass): Somatic embryogenesis and plantlet formation. Z. Pflanzenphysiol. 104:311-318. Millam, S., A.T.H. Burns and T.J. Hocking. 1991. A comparative assessment of purification techniques for mesophyll protoplasts of Brassica napus L. Plant Cell Tiss. Org. Cult. 24:43-48. Murashige, T., and F. Skoog. 1962. A revised medium for rapid growth and bioassays with tobacco tissue culture. Physiol. Plant. 15:473-497. Nichols, M.A. 1990. Asparagus - The world scene. Acta Hort. 271:25-31. Panis, B., A. Van Wauwe and R. Swennen. 1993. Plant regeneration through direct somatic embryogenesis from protoplasts of banana (MUsa spp.). Plant Cell Rep. 12:403-407. Roest, S., and L.J.W. Gilissen. 1989. Plant regeneration from protoplasts: a literature review. Acta Bot. Neerl. 38:1-23. Roest, S., and L.J.W. Gilissen. 1993. Regeneration from protoplasts - a supplementary literature review. Acta Bot. Neerl. 42:1-23. Saito, T., S. Nishizawa and S. Nishimura. 1991. Improved culture conditions for somatic embryogenesis from Asparagus officinalis L. using an aseptic ventilative filter. Plant Cell Rep. 10:230-234. Slimmon, T.Y., J.L. Jamieson and it. Tiessen. 1985. Multiplication potential of tissue cultured asparagus. In: E.C. Lougheed and H. Tiessen (Eds.), Proceedings of the Sixth International Asparagus Symposium, University of Guelph. pp. 97-104. Tsay, H.S., P.C. Lai, L.J. Chen and N.C. Chi. 1981. The development of haploid plants of Asparagus officinalis L. through anther culture, Proceedings of the Fourth International Symposium of SABRAO Kuala Lumpur, Malaysia. pp. 313-324. Chapter Three Agrobacterium-mediated transformation of embryogenic suspension cells of Asparagus oflicinalis L. ABSTRACT Asparagus officinalis IL. has been transformed with Agrobacterium tumefaciens in previous studies although the frequency at which transgenic tissues and plants have been produced was very low. In an effort to develop an efficient Agrobacterium transformation protocol, highly regenerable embryogenic suspension cultures were used as the target cells and important transformation parameters were evaluated via transient GUS expression. The binary vector, pCNL56, possesses EH1 intron containing (HHS gene that (iNI only be expressed 1J1 plant cells vflflmii was used 1J1 all transient expression experiments. Cocultivation duration was optimal at 4 days and GUS expression was greatest at an inoculum density of 5 )< 107 cfu/ml. Iknn: A. tumefaciens strains, EHA105, GV3101(pMP90), GV3101(pGV2260) and LBA4404, were tested and EHA105 and GV3101(pMP90)produced the greatest number of GUS foci. GV3101(pGV2260) gave significantly few GUS foci but LBA4404 was unresponsive. In addition, GUS expression with EHA105 and. GV3101(pMP90) induced. with acetosyringone was significantly greater than with uninduced bacteria. Kanamycin, G418 and glufosinate-ammonium were evaluated as potential selective agents for stable transformation studies. Glufosinate was effective from 2 - 5 mg/l, G418 inhibited growth at 100 mg/l and kanamycin was not inhibitory at 100 mg/l. The optimal parameters from the transient expression experiments were combined with EHA105 containing pGPTV-BAR and transgenic tissues were selected on 2 mg/l glufosinate. A single transgenic plant was produced in which integration of the BAR gene was confirmed Via Southern hybridization. Transient and stable GUS expression were compared in the same experiment vniii both EHA105 anui GV3101(pMP90). Transient expression was ten fold greater than stable expression as measured by solid blue colonies or somatic embryos after 8 weeks of culture on 100 mg/l G418. This indicates that the T- DNA is entering the cells but integration into genomic DNA is low which is the most probable reason for the low frequency of stably transformation tissues. 85 Introduction Genetic transformation has the potential to increase the productivity of horticultural crops as well as to improve individual desired traits. Asparagus officinalis L. is one such crop that would benefit from this technology. For example, herbicide resistance would be useful for asparagus seed production in addition to enhancing cultural practices. To produce a transgenic hybrid asparagus cultivar, at least one of the parents (male or female) must be transformed. Since inbreeding would destroy the genetic integrity of the transgenic parent, it must remain hemizygous for the transgene. Assuming that the introduced gene was present at a single locus in the transgenic parent, the resulting hybrid population would segregate 1:1 for the new trait. However, a seedling population could be treated with the herbicide for which the transgene confers resistance and thus eliminate the non-transgenic plants (Conner and Abernethy 1996). The BAR or PAT genes, or modified EPSPS gene that confers resistance to the non-selective herbicides L-phosphinothricin (Liberty), and glyphosate (Roundup), respectively, would work very well for this purpose. Asparagus officinalis In is susceptible ti) certain diseases and. pests that could kme overcome through transformation. Asparagus can harbor several viruses (Mink and Uyeda 1977; Falloon et al. 1986). Among these, asparagus virus I (AV-I), a potyvirus, and asparagus virus II (AV-II), 86 an ilarvirus, have been reported in the United States, Europe and the Orient. The presence of either or both of these viruses is associated with a decline in field vigor and productivity, a decrease in rooting capacity and survival in tissue culture, and an increase in susceptibility to Fusarium crown and root rot (Yang 1979; Evans and Stephens 1989a; 1989b). AV-II is time most jprevalent virus 1J1 Michigan asparagus fields (Hartung et al. 1985; Evans et al. 1990), although AV-I also occurs at a lower frequency (Evans and Stephens 1989a; Evans et al. 1990). Transforming the coat protein cDNA gene of the Virus of interest into a susceptible plant has conferred coat protein-mediated protection of that virus in a number of crops (Gonsalves and Slightom 1993). The AV-II coat protein gene has been cloned and could be used to produce transgenic asparagus resistant to the AV-II virus (D. Plunkett 1995, pers. comm.). The common and spotted asparagus beetles are serious insect pests of asparagus (Putnam et al. 1983). The common asparagus beetle overwinters as an adult, emerges in April and begins feeding and laying eggs on the emerging spears. Consequently, spears that are covered with eggs or have feeding damage may be rejected by the broker or processor. The spotted asparagus beetle emerges later and primarily feeds on the fern reducing the vigor of the plant. The larvae feed on the fern as well as ripening berries causing additional damage. Bacillus thuringiensis var. tenebrionis produces the 87 CryIII 6-endotoxin that is insecticidal to coleoptera (beetles) and harmless to humans (Estruch et al. 1997). The cryIIIA gene has been cloned and transformed into potato plants where it provided protection against the Colorado potato beetle (Perlak et al. 1993). If the CryIII é-endotoxin is insecticidal to asparagus beetles, transformed asparagus containing the cryIIIA gene would provide control for these pests and reduce insecticide input to the crop. “Tip breakdown” is a serious postharvest disorder of asparagus in which freshly harvested asparagus spears deteriorate rapidly and become unsaleable within 2-6 days at ambient temperatures (Lipton 1990). The spear tips show a rapid decline in respiration rate, lose protein and soluble carbohydrates, and accumulate ammonium ions (King et al. 1990; Lill et al. 1990). Regulating the gene(s) that are responsible for ammonium accumulation or assimilation (glutamine dehydrogenase, glutamine synthase) might not be a worthwhile approach since the spear tip already has an efficient system that reassimilates large amounts of potentially toxic ammonium (Hurst and Clark 1993; Downs et al. 1996). A better approach to overcoming tip breakdown would be to prevent proteolysis following harvest either by providing more carbohydrate (starch or sugars) for use as respiratory substrate or by slowing its present rate of usage (Hurst and Clark 1993). Altering gene expression in the sucrose synthase 88 pathway may be the most likely approach to delay tip breakdown (Irving and Hurst 1993). To be able to produce transgenic asparagus with these improved traits, an efficient transformation system must be developed. Asparagus is susceptible to infection by Agrobacterium tumefaciens and transgenic asparagus plants have been produced via Agrobacterium-mediated transformation (Bytebier et al. 1987; Conner et al. 1988; Delbreil et al. 1993). However, only one to few transgenic plants were produced in each report indicating the low efficiency of the transformation systems. There has been one report of direct- gene-mediated transformation of asparagus protoplasts (Mukhopadhyay enii Desjardins, 1994). Kanamycin resistant calluses were produced that were also positive for GUS expression, although no transgenic plants were regenerated. The approach used herein to increase the efficiency of asparagus transformation was to target somatic embryogenic cells from suspension cultures with A. tumefaciens and monitor transient tranformation events with an intron containing GUS gene. IIt is generally considered that somatic embryos are derived from single cells (Haccius 1978). Therefore, targeting such cells from. highly regenerable embryogenic cultures is an efficient pathway for producing non-chimeric, genetically transformed plants (Litz aumi Gray 1995). As previously stated, Agrobacterium-mediated transformation of asparagus has yielded limited amounts of transgenic plants. 89 However, Agrobacterium does have certain advantages over direct-gene-mediated transformation, such as higher rates of transformation, and more efficient and predictable patterns of gene integration, that warrant further development of an efficient Agrobacterium system (Smith and ikiii 1995). The effectiveness of various transformation parameters was monitored kn! transient (HHS expression. It) prevent. false positives from GUS expression in Agrobacterium, an intron- containing GUS gene was used that is only expressed in plant cells due to the presence of a modified plant intron in the GUS gene coding region (Li et al. 1992; Ritchie et al. 1993). To date, kanamycin is the only selective agent that has been used to select for transgenic asparagus tissues (Bytebier et al. 1987; Conner et al. 1988; Delbreil et al. 1993). However, monocots are relatively insensitive to kanamycin and using a more effective selective agent may benefit the production and selection of transgenic tissues (Wilmink and Dons 1993). In this study, I attempted to increase the Agrobacterium transformation frequency of asparagus by employing highly regenerable embryogenic suspensions as the target cells, evaluating several selective agents for efficient selection, and using an intron-containing GUS gene to monitor transient transformation expression of each parameter tested. 90 Materials and.Methods Plant material Asparagus officinalis L. cv. Rutgers 22 was micro- propagated according to Slimmon et al.(1985). The genotype Rutgers 22 was used because it is embryogenic as determined from the previous protoplast study. Establishment of embryogenic cell suspensions Elongated shoots taken from micropropagated crowns of Rutgers 22 were cut transversely into 1 - 2 cm explants, each possessing lateral buds, and placed on Linsmaier and Skoog (LS) medium (1965), containing 2% sucrose, 0.8% Bacto agar (pH 5.8) and 50 uM NAA. Cultures were incubated at 26W3, and 25 umolm‘zs‘1 light supplied by Philips F96T12/CW cool white bulbs on a 16-h photoperiod. Callus originating from the lateral buds was subcultured monthly on the same medium. After three subcultures, yellowish-white embryogenic callus possessing immature somatic embryos that formed at the surface of mucilaginous callus were used to initiate cell suspensions. Approximately 250 mg of callus was inoculated into each of four 250 ml Erlenmeyer flasks each containimg 40 ml of liquid medium of the same composition as the callus medium. The cultures were incubated at ZONE under 4 umolm'zs‘1 illumination for 16-h (H) a gyratory shaker“ at 110 rpm. Subcultures were performed weekly by sieving cells through 1 mm nylon mesh and transferring approx. 0.5 ml of cells, 91 settled cell volume (SCV), to 40 ml fresh medium. The suspensions were maintained on a pedigree basis for the first month and then one of the four cultures was visually selected for further subculture based on the presence of pro- embryogenic masses (PEMs) and clusters of globular somatic embryos. New cultures were initiated every three months. Agrobacterium strains The non-oncogenic A. tumefaciens strains used in this study are listed in Table l. EHA105 was kindly provided by S. Table l. Disarmed Agrobacterium tumefaciens strains and their relevant characteristics. Antibiotic Strain Vir plasmid resistance Characteristics Reference EHA105 pEHA105 Rif C58 chromosome; Hood et a1. 1993 agropine type GV3101 pMP90 Rif, Gm C58 chromosome; Koncz and Schell nopaline type 1986 GV3101 pBV2260 Rif, Cb C58 chromosome; Deblaere et al. octopine type 1985 LBA4404 pAL4404 Rif Ach5 chromosome; Hoekema et al. octopine type 1983 Rif, rifampicin; Gm, gentamycin; Cb, carbenicillin. Gelvin (Purdue Univ., USA), GV3101(pMP90) from C. Koncz (Max- Planck Institute, Germany), and GV3101(pGV2260) from J. Cardoen (Plant Genetic Systems, Gent, Belgium). 92 Bacteria culture The following protocol was used for those experiments in which the Agrobacterium virulence genes were induced prior to transformation (induction. protocol). Bacteria from e-8OWZ freezer stocks were streaked onto 100 x 15 mm Petri dishes containing semi-solid AB medium(Chilton et al. 1974) with the appropriate antibiotics and incubated for three days at 280C in darkness. A colony was placed in a 16 x 125 mm tube containing 2.5 ml of YEP medium (Chilton et al. 1974) with antibiotics and was grown overnight at 280C and 250 rpm. The next morning 1.0 ml of the overnight culture was added to 25 ml of AB medium with antibiotics in a 125 ml flask and incubated at 28%: and 200 rpm (7 - 9 hr) until an optical density at 600 nm (OD600) of 0.8 - 1.0 was reached. The cultures were diluted to OD600 = 0.3 in 25 m1 induction medium (IM) that consisted of AB salts, 2 mM sodium phosphate buffer, pH 5.6, 20 mM MES, 0.5% glucose and 100 uM acetosyringone (AS) (Li et al. 1992). The bacteria culture was incubated overnight at 25%: and 200 rpm until the OD600 reached 0.8 - 1.2. The induced bacteria were then centrifuged for 10 min at 20%: and 4000 x g, and resuspended at a particular density in 20 ml of IM lacking antibiotics. Non-induced bacteria was pmepared tn! the following protocol (non-induced protocol). A fresh bacteria colony was inoculated into 2.5 ml of YEP medium with antibiotics in a 150 x 16 mm tube and incubated overnight at 250 rpm and 28W3. The 93 next morning, 2.0 ml of bacteria was placed into 25 ml of AB medium with antibiotics in a 125 m1 flask. The culture was incubated at 200 rpm and 28%: until the OD600 reached 0.8 - 1.0. After reaching time proper density, time culture was centrifuged for 10 min at 4000 x g and resuspended to the proper density AB medium without antibiotics. For growth and maintenance cflf.A. tumefaciens, the antibiotics rifampicin, carbenicillin, gentamycin, and kanamycin were used 1J1 the media at concentrations of 10, 25, 50, and 100 ug/ml, respectively. Binary vectors The binary vector used for all GUS expression studies was pCNL56 (Li et al. 1992), kindly provided by S. Gelvin (Purdue Univ., USA). pCNL56 is a 15.4 kb pBIN19-derived vector that possesses the NPT—II gene and a GUS reporter gene between its T-DNA borders. The expression of the GUS gene is driven by the mas/358 promoter and the coding region contains a modified plant intron for plant cell specific expression (Raineri et al. 1990). All of the strains listed in Table 1 containing pCNL56 were tested for transient GUS expression in tobacco prior to asparagus transformation. The binary vector pGPTV-BAR was used for the BAR gene transformation studies (Becker et al. 1992), and was kindly provided by D. Becker (Max-Planck Institute, Germany). pGPTV- BAR is a 13.4 kb pBIN19-derived vector containing the BAR gene 94 driven by the nos promoter. The BAR gene in this vector is derived from Streptomyces hygroscopicus, codes for phosphinothricin acetyltransferase (PAT), and confers resistance ti) the non-selective herbicides glufosinate and bialaphos (White et al., 1990). The integrity of the BAR gene was confirmed by transforming tobacco leaf explants with the vector aumi regenerating enmi rooting' transgenic shoots on lethal levels of glufosinate ammonium (5 mg/l). The binary vectors were transformed into A. tumefaciens via a modified freeze-thaw method (Chen et al. 1994). Histochemical GUS assay Transient and stable GUS expression was determined histochemically according to the procedure of Jefferson (1987). Unless stated otherwise, cells from a single dish (replication) were placed into a well of a Corning 6-well dish and stained with 1.5 ml of GUS solution containing 1 mM X-gluc in 100 mM phosphate buffer, 10 mM EDTA and 0.1% Triton X-100 overnight at 37%:. Tissues were not fixed before staining. GUS expression (GUS foci) was scored as isolated blue spots on cell colonies or somatic embryos, a tight group of spots, or solid blue cell colonies or embryos. Data were analyzed by analysis of variance using MSTAT-C (version 1.2). Least significant differences (LSD) were calculated at the 5% level of probability. 95 Experiments Evaluation of selective agents Kanamycin (Sigma), G418 (Geneticin; Boehringer Mannheim) and glufosinate ammonium (Hoechst-Roussel) were added to embryo maturation (EM) medium composed of MS salts, 0.4 mg/l thiamine, 100 mg/l inositol, 20 g/l sucrose and 1% phytagel (Sigma) at 0, 12.5, 25, 50 and 100 mg/l. The selective agents were filter sterilized (0.22 pm) and added to warm (60°C) media. after .autoclaving. Five 100 >< 15 rmn dishes each containing 30 ml of medium were prepared for each treatment. Embryogenic suspensions cultures were harvested 5 days after the previous subculture, sieved through 1 mm nylon mesh and washed three times with medium composed of MS salts, 0.4 mg/l thiamine, 100 me/l inositol, and 2%) g/l sucrose (pH 5.7)(EMW). 0.5 ml of washed cells (SCV) was resuspended in 50 ml of EAL. One ml of the washed suspension (10 - 15 mg of cells) was placed on each dish. The treatments were arranged in a randomized complete block (RCB) design and incubated for 1illumination for 16-h. four weeks at 26%: and 25 umolmfls' Data were taken on fresh weight at the end of four weeks and presented as percent of the control. Based on the results of the above experiment, glufosinate was further tested to determine optimal levels for transformation selection. Glufosinate ammonium was added to EM medium at 0, 1, 2, 3, 4 and 5 mg/l. Five 100 x 15 mm dishes were prepared for each treatment. Cells were added to 96 each dish as described above. The treatments were arranged in a RCB design and incubated for four weeks at ZOTIand 25 umolm' 7-1 “s illumination for 16-h. Data were taken on fresh weight at the end of four weeks and presented as percent of the control. Cocultivation duration for optimal transient GUS expression EHA105:pCNL56 was prepared by the non-induced protocol and adjusted to 5 x 108 colony forming units per milliliter (cfu/ml) in EMW. Embryogenic suspensions were sieved through 1 mm nylon mesh, washed 3X in EM and 1.5 g of washed cells was placed in each of four 100 x 15 mm dishes. The cells were inoculated with 20 m1 of bacteria for 15 min on a gyratory shaker at 40 rpm and 25T2in darkness. After inoculation, the bacteria was pipetted off and 250 mg of cells was placed on each of 20 100 x 15 mm EM dishes. One and a half grams of uninoculated control cells was handled identical to the inoculated cells and 250 mg of those cells were placed on five EM dishes. The cultures were cocultivated in darkness at 25%:. Five random inoculated dishes were subcultured to EM medium with 300 mg/l timentin (EMT) at 2, 3, and 4 days and incubated under the same conditions. Timentin is a bacteriostatic antibiotic used to prevent growth of Gram negative bacteria such as Agrobacterium. The fourth inoculated treatment remained on EM medium for the entire 6 days. At 6 days, the cells of each inoculated dish and the 97 control was stained for transient GUS expression. Data were taken on the number of GUS foci per sample. Effect of Agrobacterium strain and acetosyringone on transient GUS expression The four strains listed in Table 1 containing pCNL56 were prepared for transformation by the induction protocol and by the non-induced protocol to compare the two methods. Each strain lacking pCNL56 was prepared by the non-induced protocol as the negative controls. All bacteria were adjusted to a density of 5 x 108 cfu/ml prior to transformation. Embryogenic suspensions were sieved through 1 mm mesh, washed three times with BMW and 250 mg of cells were placed in each of five 60 x 15 mm dishes per treatment. The cells in each dish were inoculated with 6 ml of bacteria for 15 min at 25%: and 30 rpm on a gyratory shaker. Following inoculation, the agrobacterium was pipetted off and the inoculated cells were plated onto 100 x 15 mm dishes containing solidified EM medium with 100 (m4.AS (EMAS). Those treatments with non-induced bacteria were placed on EM medium without AS. The inoculated negative controls for each strain were plated onto two dishes containing EM medium. All of the cultures were cocultivated for 4 days at 25%: in darkness. Subsequently, the cultures were placed on EMT medium for two additional days under the same environment. The cultures were assayed for transient GUS 98 expression after six days of culture and the number of blue foci was scored in each treatment. Effect of inoculum density on transient GUS expression Embryogenic suspensions were prepared for transformation by sieving through lime mesh and washed three times with BMW. Cells, 250 mg, were placed in each of six 60 x 15 mm dishes for each of the five treatments. Agrobacterium strain EHA105:pCNL56 was prepared for transformation by the induction protocol. Five 40 ml aliquots of the bacteria were adjusted to 107, 5 x 107, 10% ES><1IW and 109 cfu/ml in IM medium prior to transformation. Six milliliters of a bacteria suspension were placed in each of the six dishes containing embryogenic material for' each. density ‘treatment. The cultures were inoculated for 15 min at ZSTIand 30 rpm on a gyratory shaker. Following inoculation, the bacteria was removed and the cells were placed on 100 x 15 mm dishes containing EMAS medium. The cultures were cocultivated for four days in darkness at 25%:. After four days the cells were transferred to EMT dishes and incubated under time same conditions for EU] additional two days. Subsequently, the cells were assayed for GUS expression and scored for the number of blue foci. 99 Effect of inoculation duration with EHA105:pGPTV-BAR on the recovery of putative transgenic plants from. embryogenic suspension cells A. tumefaciens strain EHA105 containing pGPTV-BAR or lacking pGPTV-BAR (control bacteria) were pmepared tn! the induction protocol and adjusted to 5 x 107 cfu/ml prior to transformation. Embryogenic suspension cultures were sieved through 1 mmP mesh and washed three times in EMW prior to inoculation. mee embryogenic cells were inoculated for 15 min, 1 hr and 8 hr at 25°C in darkness and 30 rpm on a gyratory shaker. Following inoculation, the bacteria was removed and 250 mg of cells were placed on each 100 x 15 mm dish containing EMAS medium and cocultivated for 4 days in darkness at 25%L Subsequently, the three treatments and the negative control were transferred onto 100 x 20 mm dishes containing EMT medium with 2 mg/l glufosinate (EMTI medium). The positive control was placed on EMT medium. The dishes were arranged in a RCB design and incubated at 26%: and 25 1 illumination on a 16—h photoperiod. innolmfzs‘ In the first experiment, the positive and negative controls were inoculated for 15 min only and two dishes of each were prepared. Three replications of each of the three treatments were also prepared. In the second experiment, two negative and two positive control dishes were prepared for each inoculation duration, and six replications were made for each of the three treatments. The cultures were subcultured 100 to fresh EMTI and EMT dishes every two weeks for a total of six weeks. Following culture on EM media, all bipolar embryos from each treatment were subcultured to 100 )< 20 mm dishes containing embryo germination medium (EGTI) composed of MS salts and vitamins, 100 mg/l inositol, 0.65 mg/l ancymidol, 40 g/l glucose, 300 mg/l timentin, 4 mg/l glufosinate, and 8 g/l Bacto agar (pH 5.7). Twenty five bipolar embryos from the EMT positive control dishes were placed on emmii of four EGTI dishes ems negative germination controls and (iii) four EGT dishes lacking glufosinate as positive germination controls for both experiments. The cultures were subcultured to fresh medium of the same composition at three weeks and cultured for an additional three weeks. The cultures were incubated under illuminated conditions as pueviously described. Putative transgenics that had a visible root and shoot in comparison to the negative control embryos were grown further by being placed individually into 150 x 25 mm culture tubes containing 20 ml of EGTI medium (5 mg/l glufosinate). Germinated embryos from the positive control treatment were placed individually into 150 x 25 mm tubes containing EGT medium and EGTI medium for positive and negative controls, respectively. 101 Effect of Agrobacterium strain and inoculation duration on transient and stable GUS expression Agrobacterium strains EHA105 and GV3101(pMP90), both possessing pCNL56, were prepared for transformation by the induction protocol. The bacteria were adjusted to 5 x 107 in IM medium prior to transformation. Embryogenic suspensions were sieved through 1 mm? mesh, washed three times with EM medium, and 1.5 g of cells were placed in each of six 100 x 15 mm dishes. Each culture was inoculated with either EHA105 or GV3101 for 15 min, 1 hr or 8 hr (six treatments) and incubated at 25%L in darkness anmi 30 rpm (N1 a gyratory shaker. Following inoculation, the bacteria was removed and 250 mg of cells were placed on each of six 100 x 15 mm dishes containing EMAS medium for each treatment. Six control dishes containing 250 mg of uninoculated suspension cells were prepared and included in the experiment. The cultures were cocultivated for 4 days in darkness at 25%: in a RCB design. Following cocultivation, three replications were subcultured to 100 x 20 mm dishes containing EMT medium for transient expression and three replications were subcultured to 11H) x 20 rmn dishes containing EMT medium with 100 mg/l G418 (EMTG medium). Two control samples were subcultured to EMT for negative transient expression controls, and for stable expression two samples were subcultured to EMT medium (positive control) and two samples were placed on EMTG medium (negative control). The samples for transient expression were incubated for two 102 additional days in darkness and then evaluated for GUS expression. The cultures for the stable expression experiment 1 illumination on a 16-h were incubated at 26T2and 25 umolmfis' photoperiod, and subcultured to fresh EMTG medium every two weeks for 8 weeks. Stable GUS expression was evaluated by incubating time entire contents (fl? each. dish LU] the GUS solution and recording the number of colonies and somatic embryos expressing GUS. DNA isolation and Southern analysis Total genomic DNA was isolated from fern tissue of in vitro control and putative transgenic asparagus plants via the CTAB. (cetyltrimethylammoniuml bromide) extractibml procedure (Doyle and Doyle 1990). The procedure was modified by using cold (-20%3) ethanol (2.5 volumes of time DNA solution) in place of isopropanol to precipitate the DNA (Lewis and Sink 1996). The DNA was dried, dissolved in 400 pl of TE buffer (10 mM Tris-HCl, 1 mM EDTA, pH 7.4) and incubated with RNAse for 1 hr at 3TTL The samples were quantified by fluorometer and adjusted to a final concentration of 300 ng/ul. The RFLP probe for the BAR gene was generated by the polymerase chain reaction (PCR). A left primer 5'-CAT-GAG- CCC-AGA-ACG-ACG-CC-3' and a right primer 5'-GCA-GGC-TGA-AGT- CCA-GCT-GC-3' were used to amplify a 512 bp fragment from the BAR gene coding sequence (White et al. 1990) in pGPTV-BAR. The amplifications were carried out in 50 ul reaction mixtures 103 containing 15 ng of pGPTV-BAR template DNA, 64 ng (200 uM) of each primer, 1 mM MgCly 200 uM each of dNTPs, and 1.25 U Taq DNA polymerase in 1X PCR buffer. The amplification profile consisted of 4 min at 94°C, and 35 cycles of l min at 94°C, 1 min at 60%:, and 2 min at 72%:. The reaction was run in a Perkin-Elmer Cetus 9600 PCR system. TNme PCR product was checked for amplification and size by electrophoresis on a 1% agarose gel for 1.5 hr at 80 V in 1X TAE buffer followed by staining in ethidium bromide. Following product confirmation, the probe DNA was diluted to 20 ng/ul in TE buffer (pH 8.0). Probe DNA (25 ng) was labeled with radioactive (”P) dCTP by the random primer method using the Gibco BRL RadPrime kit. DNA (6 (mm of an untransformed control aumi the two putative transgenic asparagus plants were digested with HindIII (8 U enzyme per ug DNA) and separated on a 0.9% TAE agarose gel for 7 hr at 150 V. A lane containing 2 ug of Lambda DNA digested with HindIII was included as a DNA size marker. Blotting and hybridization were performed according to Wang et al. (1995). Banding patterns were visualized by exposure (Amersham Hyperfilm-MP film) to membranes in a cassette with an intensifying screen at -80%:. Results Evaluation of selective agents Kanamycin, G418 and glufosinate were compared to establish which selective agent would be most effective for 104 selecting transgenic tissues based on growth inhibition of untransformed embryogenic asparagus cells. The cultures had the greatest sensitivity to glufosinate (Figure 1). At the lowest level tested, 12.5 mg/l, growth of the cultures was only 10% of that of the untreated control after four weeks. A small amount of a nonembryogenic white granular callus developed in this treatment although most of the tissue was yellowish brown (necrotic), and no bipolar somatic embryos were observed. Higher levels (ME glufosinate completely inhibited cell growth. G418 allowed for growth up to 50% of the control at 50 mg/l. The treatments containing 12.5 and 25 mg/l G418 were visually similar to the control although some growth inhibition was measured. The highest level of G418, 100 mg/l, did suitably inhibit cell growth and expression of embryogenesis. Kanamycin was the least effective of the three agents tested. Growth was still 80% of the control up to 50 mg/l. However, at 25 mg/l somatic embryos began to turn white or “bleach” from the kanamycin and at 50 mg/l only bleached embryos were present. At 100 mg/l, growth was still approx. 60% of the control which is not suitable for efficient cell selection. This highest level of kanamycin did not inhibit embryo development, however the embryos present were white versus green in the control. The effect of the selective agents on cell viability was tested using fluorescein diacetate (FDA). Cells showed no loss of viability at the highest level of kanamycin selection. 105 100 — + Kanamycin —l— G418 + Glufosinate 75 I- 80 H H a CD C) e... ¢> 60 H H A: DD 0- “E .E 40 _ m d) J: \° ° 20 H 0 fit A I I l I I 0 25 50 75 100 Figure 1. Comparison of the ability of three selective agents to inhibit the growth of embryogenic asparagus cells. 106 At 50 mg/l of G418, cell viability was approx. 50% and was less than 5% at 100 mg/l. Cells on glufosinate showed little (s 1%) or no viability on any of the treatments. The selective agent experiment was repeated using only glufosinate at low levels ti) determine the lowest concentration to obtain suitable growth inhibition (Figure 2). At 1 mg/l, growth was significantly inhibited to 22% of the control and embryogenesis did not proceed past the globular stage although some green tissues were present. Two milligrame per liter' of glufosinate inhibited growth to approx. 13% of the control. Embryogenesis was arrested and the tissues were mostly necrotic. Cultures on 3-5 mg/l glufosinate were similar with slightly less growth than occurred at 2 mg/l. Based on these data, selection for transgenic cells was performed using 2 mg/l glufosinate. Glufosinate was also tested for its ability to inhibit the growth of germinated embryos and root and shoot growth of in vitro asparagus crowns. After 4 weeks on embryo germination medium with 5 mg/l glufosinate, germinated embryos (Figure 3a) and in vitro crowns (Figure 3b) failed to develop and quickly became necrotic. Thus, glufosinate at 5 mg/l would successfully inhibit root and shoot formation of non- transgenic escapes. 107 100 — _ O «‘3 75 :1 _ O t) “-1 O ea .: .2.” 50 _ Q) 3 .: m 2 {—4 o 25 — e\ T A 0 I l l l I l 0 1 2 3 4 5 mg/l glufosinate Figure 2. The effect of low levels of glufosinate on the growth of embryogenic asparagus cells. 108 Figure 3. Effect of 5 mg/l glufosinate on untransformed asparagus germinated embryos and in vitro crowns after four weeks. Glufosinate treatment on left and control on right. (A) Germinated embryos. (B) In vitro crowns. 109 Evaluation of parameters for transient GUS expression Transient GUS expression in embryogenic suspension cells manifested itself as individual blue foci, tight groups of foci (n: solid kflime regions 22 250 Inn (Figure 4). These patterns of expression were consistant throughout the studies. GUS expression was never observed in the negative controls or in the Agrobacterium itself. Cocultivation duration The optimal length of cocultivation was established to maximize transient GUS expression in all future experiments. GUS expression was observed in all of the treatments and was optimal after 4 days of cocultivation (Figure 5). Four days of cocultivation produced significantly more GUS foci than 2 or 3 days. Six days produced the second greatest response and was not significantly different from 2, 3, or 4 days although the number of GUS foci was 38% less than for 4 days. Four days of cocultivation was used for all further transformation studies. Effects of Agrobacterium strain and acetosyringone Inducing the Agrobacterium prior to transformation was beneficial for transformation but only in the most responsive strains. Only the strains with the C58 chromosomal background (see Table 1) produced GUS expression in the cell cultures (Figure 6). Strain GV3101(pGV2260) gave significantly less 110 Figure 4. Typical transient GUS expression in embryogenic asparagus suspension cells. 111 50 I» 40— 30— 20— No. of GUS foci 10— 2 Days 3 Days 4 Days 6 Days Cocultivation duration Figure 5. Effect of cocultivation duration with EHA105:pCNL56 on transient GUS expression in embryogenic asparagus cells. LSDmm5= 19.0. 112 40 D Non-induced - Induced | 30 H o- C) CD In E 20 — CD e... CD 5 Z 10 — c c 0 _ It] EHA105 GV3101 GV3101 LBA4404 (pMP90) (va2260) Agrobacterium strain Figure EL 'The effect of Agrobacterium strain and vir-gene induction on transient GUS expression in embryogenic asparagus cells. LSD0_05 = 7.1. 113 expression than EHA105 and GV3101(pMP90), and induction had no effect. No expression was observed in the LBA4404 treatment. GUS expression was highest in GV3101(pMP90) treatments, although mean GUS foci were not significantly different when comparing induced cultures and uninduced cultures of EHA105 to GV3101(pMP90). GUS expression was significantly greater within those two strains when the bacteria were induced with AS. Inoculum density Transient GUS expression was enhanced with lower inoculum densities of EHA105:pCNL56 (Figure 7). The mean number of GUS foci was greatest at 107 and 5 x.JIW cfu/ml and decreased signicantly at higher bacteria densities. GUS expression at 108, 5 x 108 and 109 was not significantly different and the number' of foci was lowest at 109 cfu/md. for the entire experiment. The cells that were inoculated with bacteria at 5 x 108 and 109 took on a brownish appearance at the end of 6 days, whereas the cultures from the lower densities remained yellowish-white anmi healthy 1J1 appearance. Although. the number of GUS foci was not significantly different between the two lowest treatments, 5 x 10’cfu/ml was used for all further experiments. Stable transformation studies with EHA105:pGPTV-BAR The optimal transformation parameters that were determined in the transient expression and the selective 114 40 35— a 30— 25— No. of GUS foci l 15 — 10 — 5 _ 0 7 7 8 s 9 10 5 x 10 10 5 x 10 10 Agrobacterium density (cfu/ml) Figure 7. Effect of EHA105:pCNL56 inoculum density on transient GUS expression in embryogenic asparagus cells. LSD0.05 = 6I6. 115 agents studies were applied to stable expression experiments using EHA105:pGPTV-BAR. After six weeks culture on EMTI medium considerably more green, bipolar embryos were present in the EHA105:pGPTV-BAR treatments than. in the negative controls (Table 2). This was an initial indication that transgenic tissues may have been produced. Table 2. Comparison of the number of green, bipolar embryos present in EHA105 control and EHA105:pGPTV-BAR treatments after 6 weeks of culture on 2 mg/l glufosinate. Inoculation Mea um f i l r m r duration Experiment 1 Experiment 2 (+) control > 300 > 300 15 min (-) control 1.0 i 1 9.5 i 0.5 1 hr (-) control 14.4 i 0.5 8 hr (-) control 18.0 i 1.0 15 min BAR 13.7 i 0.9 32.3 i 3.1 1 hr BAR 11.0 i 3.2 31.2 i 3.4 8 hr BAR 12.7 i 0.9 51.7 i 5.0 i, standard error of the mean. Each positive control dish contained at least 300 mature embryos, whereas the negative controls contained 1 - 18 embryos on average depending on the experiment. This indicated that selection was working, although not completely, but enough to suppress the development of most non-transgenic tissues. After six weeks of culture on EGT germination medium, approx. 95% of the positive control embryos had germinated 116 with elongated roots, and 26% and 53% had shoots that were 2 1 cm in length in the first and second experiments, respectively. In contrast, development was almost completely arrested 1J1 the negative controls. Glufosinate at.14 mg/l (EGTI medium) inhibited radical elongation in embryos past 0.5 cm, completely prevented shoot formation and caused 80 — 90% of the embryos to become necrotic. In the same time, one embryo from the 1 hr inoculation duration treatment from both experiments (2 embryos total) germinated with both roots and shoots. These two embryos did not root as vigorously as the positive controls, but they did germinate and survive selection on 4 mg/l glufosinate. These two putative transgenic plantlets were placed in culture tubes containing EGTI H3 mg/l glufosinate) for fUrther selection aumi crown formation. After four weeks culture, the two putative transgenic plants survived and elongated roots CH1 the EGTI medium (Figure 8). The untranformed control plants were completely necrotic and did not develop past the initial explant stage on the same medium. During the first four weeks of culture, the putative transgenics rooted similar to the positive controls although shoot formation was depressed. Upon subculture ti) the same medium and.en1 additional four weeks of incubation for crown development, the putative transgenic from the first experiment produced a single new shoot of 3 cm and a few roots that elongated less than 2 cm. The putative transgenic from the second experiment was more 117 Figure 8. Appearance of putative transgenic somatic embryo— derived plants after four weeks of culture on 5 mg/l glufosinate. Left two tubes, untransformed controls without glufosinate. Middle two tubes, untransformed controls on glufosinate. Right two tubes, putative transgenics from Exp. 1 and Exp. 2, respectively, on glufosinate. 118 vigorous and had a root system similar to some of the positive controls and produced two new shoots up to 8 cm in length. In Figure 9, a single ~5 kb band representing the BAR transgene was resolved by Southern hybridization for the putative transgenic genomic DNA from Exp. 2 (lane 4). Bands were absent in the negative control (lane 2) and in the putative transgenic DNA from Exp. 1 (lane 3). This confirmed that only the plant from the second experiment was transgenic. Further studies could not be performed because the transformed plant became contaminated and did not survive transfer to the greenhouse. Relating transient to stable GUS expression Since stable transformation was very low in the previous study, transient and stable expression were compared within the same experiment to see how well the two related with the Agrobacterium strains used and the asparagus cell cultures. Transient GUS expression was greatest with both EHA105 and GV3101(pMP90) with either a 15 min or 1 hr inoculation time (Figure 10). Both strains were very similar in response. The mean number of GUS fOci pmoduced for those treatments was notably high for this study. 'Transient expression was significantly less at 8 hr for both of the strains. The 8 hr treatments did have a brownish appearance which may have been due to the detrimental effects of the bacteria and lead to reduced GUS expression. 119 6.6kb—> w <— 5 kb Figure 19. Southern blot for the BAR gene in putative transgenic asparagus plants transformed with EHA105:pGPTV-BAR. Lane 1, lambda DNA digested with HindIII. Lane 2, untransformed asparagus control. Lane 3, putative transgenic from Exp. 1. Lane 4, transgenic asparagus from Exp. 2 with a single T-DNA insert. 120 60 E GV3101(pMP90) - EHA105 504 40— 20— No. of GUS foci l 10 — 0 15 min 1 hr 8 hr Inoculation duration Figure 10. The effect of inoculation duration and Agrobacterium strain on transient GUS expression in embryogenic asparagus cells. LSme = 11.2. 121 After 8 weeks of culture on 100 mg/l G418, the cultures were evaluated for stable GUS expression. Growth was suitably inhibited with 100 mg/l G418 in the negative control and no GUS expression was observed in this treatment. GUS expressing colonies and embryos (Figure 11) were produced in all of the treatments, but stable expression was ten times lower than transient expression (Figure 12). The data were not significant as determined by ANOVA. 122 Figure 11. Somatic embryos exhibiting stable GUS expression after 8 weeks of culture on 100 mg/l G418. Far left, uninoculated control embryo. 123 m 5 O 3‘ III! GV3101 colonies Q GV3101 embryos E - EHA105 colonies ; 4 g m EHA105 embryos : d! m d) 'E c 3 H _ O C) O!) a 0— % 2 _ d) ‘5. I 23 m / :3 1‘4 0 a... ° z 15 min 1 hr 8 hr Inoculation duration Figure 12. The effect of inoculation duration and Agrobacterium strain on stable GUS expression in cell colonies and somatic embryos after 8 weeks of culture on 100 mg/l G418. Treatment effects were not significant. 124 Discussion An efficient Agrobacterium-mediated transformation system for asparagus was sought by combining highly regenerable embryogenic asparagus cell cultures with an appropriate A. tumefaciens strain eumi optimal transformation. parameters. Transient GUS expression assays are ideal for optimizing transformation parameters because they provide data quickly and easily without having to select for transgenic tissues. Such assays have been found useful for optimizing parameters that subsequently are applied to the production of transgenic plants (Janssen and Gardner 1989; Van Wordragen et al. 1992; Janssen and Gardner 1993). However, in this study transient expression assays were not found to be good predictors for optimizing stable transformation. It should be noted that transient expression is not an indication of integration but rather an assessment of T-DNA transfer and transcription within the plant cells. A. tumefaciens strain EHA105 was used in several preliminary studies and was continued to be used because it produced consistant transient GUS expression. In contrast, when the optimumi parameters derived. from. the ‘transient expression studies were applied to stable expression using EHA105:pGPTV- BAR, only one transgenic plant was recovered. This finding prompted further investigations into the relationship between transient and stable expression relative to asparagus. The wildtype strain A281 is the progenitor to EHA105 and failed to 125 produce swellings or galls (N1 spears of 17 .4. officinalis genotypes (Conner et al. 1988). Although lack of tumorigenesis does not necessarily mean the absence of stable transformation especially in monocots (Godwin et al. 1992), there is a strong possibility that the frequency of T-DNA integration is quite low using EHA105 in asparagus. A. tumefaciens strain GV3101(pMP90) was included along with EHA105 because ii: had been reported ti) produce transgenic asparagus tissues in other studies (Hernalsteens et al. 1984; Conner et al. 1988; Prinsen et al. 1990; Delbreil et al. 1993). Transient GUS expression in the final study had occurred at levels exceeding previous experiments for both strains, but after 8 weeks on selection medium(lOO mg/l G418), the number of colonies and somatic embryos expressing GUS was ten-fold lower than for the initial transient expression assay at 6 days. Apparently, the T-DNA was entering the cells and transcription of the GUS gene occurred but the frequency of integration into genomic DNA was very low. Narasimhulu et al.(1996) studied the early events involved with T-DNA transfer, transcription and integration into cells of tobacco and maize. They observed that the initial kinetics for the appearance of GUS transcripts was approximately equal for both species and that there was no major difference in T-DNA transfer, nuclear targeting or conversion to a double-stranded DNA femmu However, while GUS transcripts were present in tobacco cells for at least 7 days, they dissappeared in maize 126 within 36 hr after inoculation. This result was an indication that the T-DNA was being lost or degraded in maize; whereas, in tobacco it was being integrated into the tobacco genome. Since transient expression in maize was equivalent to that of tobacco, they suggested the inability to efficiently transform maize with Agrobacterium does not involve the infection process but rather there is a block at the level of T-DNA integration iimi) the genome. This block may kme the same reason for the low frequency of stable transformation observed in asparagus using Agrobacterium. Another potential reason for the low frequency of stable transformation could be genotype. A. officinalis cv. Rutgers 22 was used in this study because it is amenable to forming embryogenic cultures as determined in previous work in our laboratory and for its potential breeding value. Studies have shown that genotypes of the same species can vary widely in their response to transient and stable Agrobacterium transformation (Van Wordragen et al. 1991; 1992). Conner et al. (1988) investigated the Agrobacteriwn x genotype interaction in asparagus using four wild-type strains of A. tumefaciens and 17 asparagus genotypes. Only A. tumefaciens wild-type strain. C58 produced tumors on five of the 17 genotypes. Moreover, A. officinalis cv. CRD 157 produced the majority of tumors. There is the possibility that Rutgers 22 is recalcitrant to T-DNA integration and that other genotypes may have yielded a higher frequency of transgenic tissues. 127 The parameters evaluated in this study comprised those known to Ime most influencial for successful Agrobacterium transformation. Among those, the most important being Agrobacteruim strain. The Agrobacterium strains that were used corresponded either to those that have successfully produced transgenic asparagus tissues or have potential merit in asparagus transformation. In previous studies in which transgenLc plants were pmoduced, strains corresponding to GV3101(pMP90) (C58C1:pGV3850::1103neo) (Bytebier et al. 1987; Conner et al. 1988) and GV3101(pGV2260) (Delbreil et al. 1993) were used. Strain EHA105, derived from the wild-type A281, carries virulence genes from pTiB0542 and is considered to be hypervirulent (Hood et al. 1986). A281 was reported to be ineffective in forming tumors on asparagus spears(Conner et al. 1988) but has been effective in transforming other monocots (Raineri et al. 1990; Dong et al. 1996). Finally, strain Ach5 is the wild-type progenitor to LBA4404. Ach5 has the ability to aggregate asparagus cells which was used as an indication of plant cell attachment (Draper et al. 1983). Cell attachment is one of the first steps in the Agrobacterium transformation process. Aggregation by Ach5 actually exceeded that caused by C58 strains in that study. The strain of Agrobacterium used did prove to be an important factor in this study. The strains that produced transgenic plants in studies by Bytebier et al. (1987) and Conner et al.(1988) were equivalent to GV3101(pMP90). 128 GV3101(pMP90) produced the greatest amount of transient GUS expression along with EHA105. Delbreil et al. (1993) successfully produced transformed embryogenic cell cultures of asparagus with GV3101(pGV2260). However, this C58-derived strain was not found to be effective in this study. Since transient GUS expression was not significantly different between EHA105 and GV3101(pMP90) and there was no significant amount of transgenic tissues produced, ii: is difficult to assertain which strain is best. At the time of this study, kanamycin was the only selective agent that had been used to select for transgenic asparagus tissues(Bytebier et al. 1987; Conner et al. 1988; Delbreil et al. 1993; Mukhopadhyay and Desjardins 1994). Few transgenic plants were produced in those studies. Since the efficiency of selection is an important factor in the suscessful recovery of transgenic plants (Lindsey and Jones 1990), evaluating selective agents other than kanamycin was warranted. IN; a potential alternative ti) kanamycin, G418 (geneticin) was tested because it is also an aminoglycoside antibiotic that is detoxified by the NPTII gene and has been shown to be more effective than kanamycin in some dicots and most monocots(Norelli and Aldwinckle 1993; Wilmink and Dons 1993; Laparra et al. 1995). Previous asparagus transformation studies selected tissues using up to 100 mg/l kanamycin. This level was found to be insufficient for suitably inhibiting growth of embryogenic cells and escapes could be anticipated. 129 The asparagus cells were approximately twice as sensitive to G418 ems they were to kanamycin. IN: is believed that the increased sensitivity to G418 is caused by more effective binding to the ribosomes compared to kanamycin (Wilmink and Dons 1993). The asparagus cells exhibited the greatest amount of sensitivity to glufosinate. Unlike the aminoglycosides that suppress protein synthesis, glufosinate is an herbicide that inhibits glutamine synthase and causes rapid cell death from the accumulation of ammonia in the cells (Tachibana et al. 1986). Although G418 and glufosinate were determined to be more effective than kanamycin in inhibiting growth, the actual effect of the selective agents on transformation and regeneration of transgenic plants could rim: be determined since a population of transgenic plants was not produced. The other parameters that were tested in this study aided in T-DNA transfer and transient GUS expression. Determining the length of cocultivation provided information as to when the greatest amount of GUS expression could be detected to aid in scoring the experiments. This information was applied directly to timing of selection of transformed cells. The proper inoculum density is a balance between the size of the inoculum required for high levels of DNA transfer weighed against deleterious effects fo bacterial growth on the inoculated tissues (Lindsey and Jones 1990). Transient GUS expression was the greatest at the lowest densities tested which may be an indication that inoculum densities above 108 130 might tie deleterious. .At densities of 5 )< 109 cfu/ml and above, the cells were observed to be brownish in appearance indicating that the lower expression may have been due to deleterious conditions. Part of the transformation process involves the attraction of Agrobacterium to the cell surface or wound site and the activation of the vir-genes by phenolic signal molecules. The most commonly used compound for inducing the Vir-genes prior to transformation to aid T-DNA transfer is acetosyringone (Stachel et al. 1985). Asparagus does produce a moderate amount of phenolic signal molecules compared to other monocots which may be an explanation why it can be naturally infected (Messens et al. 1990). However, while acetosyringone did significantly enhance T-DNA transfer and transient GUS expression, it did not have an effect on T- DNA integration. In this study, Agrobacterium tumefaciens had the ability to transform asparagus as exhibited by transient GUS expression assays. However, an efficient transformation system that routinely yields a high frequency of transgenic plants will require a greater understanding of the requirements for T-DNA integration into the asparagus genome. LIST OF REFERENCES 131 LITERATURE CITED Becker, D., E. Kemper, J. Schell and R. Masterson. 1992. New plant binary vectors with selectable markers located proximal to the left T-DNA border. Plant Mol. Biol. 20:1195-1197. Bytebier, B., F. Deboeck, H. De Greve, M. Van Montagu and J. -P. Hernalsteens. 1987. T-DNA organization in tumor cultures and transgenic plants of the monocotyledon Asparagus officinalis. Proc. Natl. Acad. Sci. USA. 84:5345-5349. Chen, H., R.S. Nelson and J.L. Sherwood. 1994. Enhanced recovery of transformants of Agrobacterium tumefaciens after freeze-thaw transformation and drug selection. Biotech. 16:664-669. Chilton, M.-D., T.C. Currier, S.K. Farrand, A.J. Bendich, M.P. Gordon and E.W. Nester. 1974. Agrobacterium tumefaciens DNA and P88 bacteriophage DNA not detected in crown gall tumors. Proc. Nat. Acad. Sci. USA 71:3672-3676. Conner, A.J., M.K. Williams, S.C. Deroles and R.C. Gardner. Agrobacterium-mediated transformation of asparagus. In: K.S. WcWhirter, R.W. Downes, B.J. Read (eds.) Proceedings of the 9th Australian Plant Breeding Conference, Wagga Wagga, N.S.W. Australia. 1988. pp. 131-132 Conner, A.J., and D.J. Abernethy. 1996. Genetic engineering of asparagus: Assessment of methods, field testing and safety considerations. Acta Hort. 415:51-58. Delblaere, R., B. Bytebier, H. De Greve, F. Deboeck, J. Schell, M. Van Montagu and J. Leemans. 1985. Efficient octopine Ti-plasmid-derived.vectors for Agrobacterium-mediated gene transfer to plants. 13:4777-4788. Delbreil, B., P. Guerche and M. Jullien. 1993. Agrobacterium- mediated transformation of Asparagus officinalis L. long-term embryogenic callus and regeneration of transgenic plants. Plant Cell Rep. 12:129-132. 132 Dong, J., W. Teng, W.G. Buchholz and T.C. Hall. 1996. Agrobacterium-mediated transformation of Javanica rice. Mol. Breed. 2:267-276. Downs, C.G., W.M. Borst, P.L. Hurst, G.A. King and D.G. Stevenson. 1996. Changes in glutamine synthase in harvested asparagus spears: An opportunity for genetic modification? Acta Hort. 415:315-321. Doyle, J.J., and J.L. Doyle. 1990. Isolation of plant DNA from fresh tissue. Focus 12:13-15. Draper, J., I.A. Mackenzie, M.R. Davey and J.P. Freeman. 1983. Attachment of Agrobacterium tumefaciens to mechanically isolated asparagus cells. Plant Sci. Let. 29:227-236. Estruch, J.J., N.B. Carozzi, N. Desai, N.B. Duck, G.W. Warren and M.G. Koziel. 1997. Transgenic plants: An emerging approach to pest control. Nat. Biotech. 15:137-141. Evans, T.A., and C.T. Stephens. 1989a. First report of asparagus virus I in asparagus in Michigan. Plant Dis. 73:273. Evans, T.A., and C.T. Stephens. 1989b. Increased susceptibility to Fusarium crown and root rot in virus- infected asparagus. Phytopath. 79:253-256. Evans, T.A., R.M. DeVries, T.L. Wacker and C.T. Stephens. 1990. Epidemiology of asparagus viruses in Michigan asparagus. Acta Hort. 271:285-290. Falloon, P.G., L.M. Falloon and R.C. Grogan. 1986. A survey of California asparagus for asparagus virus I, asparagus virus II and tobacco streak virus. Plant Dis. 70:103-105. Godwin, I.D., B.V. Ford-Lloyd and H.J. Newbury. 1992. In vitro approaches to extending the host-range of Agrobacterium for plant transformation. Aust. J. Bot. 40:751-763. Gonsalves, D., and J.L. Slightom. 1993. Coat protein-mediated protection: analysis of transgenic plants for resistance in a variety of crops. Sem. Virol. 4:397-405. Haccius, B. 1978. Question of unicellular origin of non- zygotic embryos in callus cultures. Phytomorph. 28:74-81. Hartung, A.C., T.A. Evans and C.T. Stephens. 1985. Occurrence of asparagus virus II iii commercial asparagus fields in Michigan. Plant Dis. 69:501-504. 133 Hernalsteens, J.-P., I“ Thia-Toong, II. Schell. and FL Van Montagu. 1984. An Agrobacterium-transformed cell culture from the monocot Asparagus officinalis. EMBO J. 3:3039-3041. Hoekema, A., P.R. Hirsch, P.J.J. Hooykaas and R.A. Schilperoort. 1983. A binary plant vector strategy based on separation of Vir- and T-region of the Agrobacterium tumefaciens Ti-plasmid. Nature. 303:179-180. Hood, E.E., G.L. Helmer, R.T. Fraley and M.-D. Chilton. 1986. The hypervirulence of Agrobacterium tumefaciens A281 is encoded in a region of pTiBo542 outside of T-DNA. J. Bact. 168:1291-1301. Hood, E.E., S.B. Gelvin, L.S. Melchers and A. Hoekema. 1993. New Agrobacterium helper plasmids for gene transfer to plants. Trans. Res. 2:208-218. Hurst, P.L., and C.J. Clark. 1993. Postharvest changes in ammonium, amino acids and enzymes of amino acid metabolism in asparagus spear tips. J. Sci. Food. Agric. 63:465-471. Irving, D.E., and. P.L. jHurst. 1993. Respiration, soluble carbohydrates and enzymes of carbohydrate metabolism in tips of harvested asparagus spears. Plant Sci. 94:89-97. Janssen, B.-J., and R.C. Gardner. 1989. Localized transient expression of GUS in leaf discs following cocultivation with Agrobacterium. Plant Mol. Biol. 14:61-72. Janssen, B.-J., and R.C. Gardner. 1993. The use of transient GUS expression ti) develop an .Agrobacterium-mediated gene transfer system for kiwifruit. Plant Cell Rep. 13:28-31. Jefferson, R.A. 1987. Assaying chimeric genes in plants: The GUS gene fusion system. Plant Mol. Biol. Rep. 5:387-405. King, G.A., D.C. Woollard, D.E. Irving and W.M. Borst. 1990. Physiological changes in asparagus spear tips after harvest. Physiol. Plant. 80:393-400. Koncz, C. And J. Schell. 1986. The promoter of Tg-DNA gene 5 controls the tissue-specific expression an chimaeric genes carried by a novel type of Agrobacterium binary vector. Mol. Gen. Genet. 204:383-396. 134 Laparra, H., M. Burrus, R. Hunold, B. Damm, A.-M. Bravo-Angel, R. Bronner and G. Hahne. 1995. Expression of foreign genes in sunflower (Helianthus annuus L.)- evaluation of three gene transfer methods. Euphyt. 85:63-74. Lewis, M.E., and K.C. Sink. 1996. RFLP linkage map of asparagus. Genome 39:622-627. Li, X.-Q., C.-N. Liu, S.W. Ritchie, J.-Y. Peng, S.B. Gelvin and T.K. Hodges. 1992. Factors influencing Agrobacterium- mediated transient expression of gusA in rice. Plant Mol. Biol. 20:1037-1048. Lill, R.E., G.A. King and E.M. O’Donoghue. 1990. Physiological changes in asparagus spears immediately after harvest. Sci. Hort. 44:191-199. Lindsey, K., and M.G.K. Jones. 1990. Selection of transformed cells. In: Plant Cell Line Selection. P.J. Dix (Ed.),. VCH Publishers Press, New York. Pp. 317-339. Linsmaier, E.U., and F. Skoog. 1965. Organic growth factors requirements of tobacco tissue cultures. Physiol. Plant. 18:100-127. Lipton, W.J. 1990. Postharvest biology of fresh asparagus. Hort. Rev. 12:69-155. Litz, R.E., and D.J. Gray. 1995. Somatic embryogenesis for agricultural improvement. World J. Microbiol. Biotech. 11:416- 425. Messens, E., R. Dekeyser and S.E. Stachel. 1990. A nontransformable Triticum monococcum monocotyledonous culture produces the potent Agrobacterium vir—inducing compound ethyl ferulate. Proc. Natl. Acad. Sci. USA. 87:4368-4372. Mink, G.I., and I. Uyeda. 1977. Three mechanically transmissible viruses isolated from asparagus in Michigan. Plant Dis. Rep. 61:398-401. Mukhopadhyay, S., and Y. Desjardins. 1994. Direct gene transfer to protoplasts of two genotypes of Asparagus officinalis L. by electroporation. Plant Cell Rep. 13:421- 424. Narashimhulu, S.B., X.-B. Deng, R. Sarria and S.B. Gelvin. 1996. Early transcriptflmu of Agrobacteriwn T-DNA genes in tobacco and maize. Plant Cell 8:873-886. 135 Norelli, J.L., and H.S. Aldwinckle. 1993. The role of aminoglycoside antibiotics in the regeneration and selection of neomycin phosphotransferase-transgenic apple tissue. J) Amer. Soc. Hort. Sci. 118:311-316. Perlak, F.J., T.B. Stone, Y.M. Muskopf, L.J. Peterson, G.B. Parker, S.A. McPherson, J. Wyman, S. Love, G. Reed, D. Biever and D.A. Fischoff. 1993. Genetically improved potatoes: Protection from damage by Colorado potato beetles. Plant Mol. Biol. 22:313-321. Prinsen, E., B. Bytebier, J.-P. Hernalsteens, J. De Greef and H. Van Onckelen. 1990. Functional expression of Agrobacterium tumefaciens T-DNA onc-genes in asparagus crown gall tissues. Plant Cell Physiol. 31:69-75. Putnam, A.R., R. Stuckey, M.L. Lacy, E. Grafius and G.W. Bird. 1983. Common asparagus pests. Michigan State Univ. Ext. Bul. E-959. Raineri, D.M., P. Bottino, M.P. Gordon and E.W. Nester. 1990. Agrobacterium-mediated transformation of rice (Oryza sativa L.). Bio/Technol. 8:33-38. Ritchie, S.W., C.-N. Lui, J.C. Sellmer, H. Kononowicz, T.K. Hodges and. S.B. Gelvin. 1993. Agrobacterium tumefaciens- mediated expression of gusA in maize tissues. Trans. Res. 2:252-265. Slimmon, T.Y., J.L. Jamieson and H. Tiessen. 1985. Multiplication potential of tissue cultured asparagus. In: Lougheed, E.C., and H. Tiessen (Eds.), Proceedings of the Sixth International Asparagus Symposium, University of Guelph. pp. 97-104 Smith, R.H., and E.E. Hood. 1995. Agrobacterium tumefaciens transformation of monocotyledons. Crop Sci. 35:301-309. Stachel, S.E., E. Messens, M. Van Montagu and P. Zambryski. 1985. Identification of the signal molecules produced by wounded plant cells that activate T-DNA transfer in Agrobacterium tumefaciens. Nature 318:624-629. Tachibana, K., T. watanabe, Y. Sekizuwa and T. Takematsu. 1986. Action. mechanisml of bialaphos. 2. Accumulation of ammonia iii plants treated vdiit bialaphos. II. Pestic. Sci. 11:33-37. 136 Van Wordragen, M.F., J. de Jong, H. B.M. Huitema and H.J.M. Dons. 1991. Genetic transformation of Chrysanthemum using wild type Agrobacterium strains; strain and cultivar specificity. Plant Cell Rep. 9:505-508. Van Wordragen, M.F., J. De Jong, M.J. Schornagel and H.J.M. Dons. 1992. Rapid screening for host-bacterium interactions in Agrobacterium-mediated gene transfer to Chrysanthemum, by using the GUS-intron gene. Plant Sci. 81:207-214. Wang, J., M.E. Lewis, J.H. Whallon and K.C. Sink. 1995. Chromosome mapping of T-DNA inserts in transgenic petunia by in situ hybridization. Trans. Res. 4:241-246. White, J., S.-Y.P. Chang, M.J. Bibb and M.J. Bibb. 1990. A cassette containing the bar gene of Streptomyces hygroscopicus: a selectable marker for plant transformation. Nucl. Acid. Res. 18:1062. Wilmink, A., enmi J.J.M. Dons. 1993. Selective agents and marker genes for use in transformation of monocotyledonous plants. Plant Mol. Biol. Rep. 11:165-185. Yang, H.J. 1979. Early effects of viruses on the growth and productivity of asparagus plants. Hortsci. 14:734-735. SUMMARY AND RECOMMENDATIONS 138 Protoplast Culture I found that genotype and auxin source influences asparagus protoplast culture. In addition, protoplasts derived from a rughly embryogenic cell source can undergo direct somatic embryogenesis under time proper conditions. Upon initiating protoplast culture with an untested genotype, embryogenesis should be attempted with both 2,4-D and NAA. If both auxins produce embryogenic cultures then preference should be given to the NAA cultures due to the lower frequency of somaclonal variation. It 1A5 very important ti) select embryogenic material, such as clusters of somatic embryos, for initiating suspension cultures. This will ensure that the population of donor cells are as embryogenic as possible prior to protoplast isolation. If non-embryogenic callus is used for this purpose, the suspension cell population will be more heterogenous and contain fewer embryogenic cells. For protoplast culture, the X—plate system described in chapter two is very effect. For NAA derived protoplasts, it would be useful to examine using levels of NAA far below 50 uM for protoplast culture. The lack of NAA in the protoplast culture medium allowed for the development of direct somatic embryos from NAA-derived protoplasts, however, a low level of NAA may be beneficial for protoplast culture with other genotypes. Using 50 uM was apparently toxic in our study but may expedite cell divisions at lower levels. 139 Transformation The production of transgenic asparagus tissues and plants via Agrobacterium-mediated transformation has occurred at a very low frequency in previous reports. This study attempted to develop an efficient Agrobacterium-mediated transformation protocol that was based on using embryogenic cells to expedite regeneration of transgenic plants via somatic embryogenesis. Unfortunately, the frequency of transformation was also very low 1J1 our study. Transformation cflf asparagus should kme pursued until an efficient and genotype independent system is developedr The following are recommendations for fUrther studies. The embryogenic colonies were placed on EM medium for embryo development. and.:maturation directly following inoculation. An alternative to this is to place the inoculated tissues onto a medium containing a selective agent (glufosinate) that promotes the development of embryogenic callus. This would allow the transformed regions of the cell masses to proliferate instead of depending on single transformed cells ti) give rise immediately ti) transformed somatic embryos. A second recommendation is to inoculated smaller cell colonies. Smaller or less developed colonies are rapidly growing, potentially more transformation competent and transgenic regions on such colonies are easily selected for. A final recommendation is to use direct-gene—mediated transformation in the form of the gun. T-DNA is associated with proteins from the Agrobacterium virulence genes that may ..rr 9...: 112' .lm 140 be recognized by the cell causing degradation or bdocking integration into the asparagus genome. Direct-gene-mediated transformation is not associated with any proteins that could be recognized tax the cell. Although transformation via biolistics is less efficient than with Agrobacterium, use of the gene gnu) is warranted for asparagus and mey'rmfl: be as genotype dependent.