In Vitro Cell. Dev. Biol.—Plant 42:114–118, March– April 2006 q 2006 Society for In Vitro Biology 1054-5476/06 $18.00+0.00

DOI: 10.1079/IVP2005733

HIGHLY EFFICIENT EMBRYOGENESIS AND PLANT REGENERATION OF TALL FESCUE (FESTUCA ARUNDINACEA SCHREB.) FROM MATURE SEED-DERIVED CALLI WAN-JUN ZHANG, JIANG-LI DONG, BEN-GUO LIANG, YONG-SHENG JIN,

AND

TAO WANG*

State Key Laboratory for Agro-biotechnology, China Agricultural University, Beijing 100094, China (Received 14 March 2005; accepted 28 July 2005; editor F. Engelmann)

Summary We report a protocol for efficient plant regeneration of four tall fescue (Festuca arundinacea Schreb.) cultivars (‘Surpro’, ‘Coronado’, ‘Summer Lawn,’ and ‘Fawn’) via somatic embryogenesis. Calli were initiated from mature seeds grown on modified Murashige and Skoog (MMS) medium supplemented with 7.0 mg l21 (31.7 mM) 2,4-dichlorophenoxyacetic acid (2,4-D) and 0.05 mg l21 (0.23 mM) kinetin (Kin). Calli were maintained and proliferated by subculture at monthly intervals on MMS medium containing 4.5 mg l21 (20.4 mM) 2,4-D and 0.2 mg l21 (0.9 mM) Kin. Somatic embryos (SE) were induced from seed-derived calli on SE-induction medium (MMS supplemented with 2.0 mg l21 2,4-D and 0.2 mg l21 Kin). Plantlets were regenerated from somatic embryogenic calli grown on modified SH medium supplemented with 2 mg l21 Kin. Using this optimized protocol, 78.6– 82.3% of mature seeds of all four cultivars produced SE clusters, of which 93.5– 95.3% regenerated into plants within 10 wk. The regenerants showed no phenotypic abnormalities. Key words: culture medium; Festuca arundinacea Schreb.; plant regeneration; somatic embryogenesis.

system for plant regeneration of four tall fescue cultivars. via somatic embryogenesis.

Introduction Tall fescue (Festuca arundinacea Schreb.) is an open-pollinated, cool-season, perennial pasture and turf grass (Buckner et al., 1979; Eigenza and Dahleen, 1990; Bai and Qu, 2001). It has great potential use as a turf grass in transition zones where heat- and drought-tolerant grass cultivars are desirable (Bai and Qu, 2000). Despite the progress in fescue breeding (NTEP, 1996; Wang et al., 2001), genetic engineering has proved to be an efficient approach for improvement of tall fescue, owing to its genetic complexity and the associated difficulties encountered with conventional breeding methods (Wang et al., 1992; Garcia et al., 1994; Kuai et al., 1999; Bai and Qu, 2001; Wang et al., 2001; Bettany et al., 2003). An efficient plant regeneration system is a prerequisite for plant improvement through genetic transformation (Dalton, 1988; Fennell et al., 1996). Results of previous studies showed that tall fescue plants could be regenerated from various explants, including mature seed embryos (Lowe and Conger, 1979; Takamizo et al., 1994), shoot tips (Dale and Dalton, 1983), matured seeds (Wang et al., 1992; Bai and Qu, 2000), inflorescences (Eigenza and Dahleen, 1990), anther (Kasperbauer and Springer, 1980.) and protoplasts isolated from cell suspension culture (Dalton, 1988; Sam et al., 1992). However, these protocols require 15 – 45 wk for regeneration and the regeneration frequencies are also low. More effort is therefore required to establish a routine and efficient regeneration system. In this study, we investigated the factors influencing plant regeneration, with an aim to develop an efficient tissue culture

Materials and Methods Plant material and culture conditions. We used four tall fescue cultivars namely, three elite turf-type (Surpro, Coronado, and Summer Lawn) and one forage-type (Fawn) in this study. Mature seeds were husked, rinsed in 70% (v/v) alcohol for 2 min, sterilized in 0.1% (w/v) mercuric chloride for 10 min, washed four times in sterile distilled water, and then placed on the callus-induction media. All media used were solidified with 8 g l21 agar (Sigma) at pH 5.8 and autoclaved at 1218C for 20 min. All plant materials were cultured at 25 ^ 28C under a 14 h photoperiod at a light intensity of 30–35 mmol m22 s21 unless otherwise mentioned. Experiment 1: callus induction. Modified MS (Murashige and Skoog, 1962) medium (MMS) was used as the basic callus-induction medium. This medium contained MS basic salt and 9.9 mg l21 thiamine, 4.5 mg l21 nicotinic acid, 9.5 mg l21 pyridoxine, 1.0 g l21 casein enzymatic hydrolysate (Sigma), 8 g l21 agar (Sigma), and 30 g l21 sucrose (Zhang and Wang, 2003). To optimize the concentration of 2,4-dichlorophenoxyacetic acid (2,4-D) for callus induction, culture response of cv. ‘Surpro’ in MMS containing 0.05 mg l21 kinetin (Kin) and different 2,4-D concentrations (3.0, 5.0, 7.0, 9.0, and 11.0 mg l21) was evaluated. MS containing 9.0 mg l21 2,4-D was used as a control. Each treatment consisted of three replicates, each containing 100 seeds with 10 seeds plated in one 50 ml culture bottle. Observations were made every 3 d during the course of callus induction. The time of callus initiation (the first visible callus development) and callus size were recorded. The callus– induction ratio and size of callus clusters were scored after 30 d. The callus– induction ratio was defined as the number of calli formed per seed divided by the total number of viable seeds. A viable seed is referred to as germinating seed or seed that formed calli on callus induction medium (Wang et al., 2002b). Experiment 2: embryogenic callus promotion. The effects of different 2,4-D and Kin (2,4-D 2.0– 5.0 mg l21, kin 0.05–0.2 mg l21) combinations on embryogenesis promotion and somatic embryo (SE) induction were also investigated. This was conducted by using the 30-d-old callus clusters

*Author to whom correspondence should be addressed: Email wangt@ cau.edu.cn

114

115

PLANT REGENERATION FROM TALL FESCUE (3 £ 3 mm2) of ‘Surpro’ initiated on MMS medium containing 7.0 mg l21 2,4D. Each treatment consisted of three replicates, each containing 50 callus clusters, with five callus clusters placed in one bottle. The somatic embryogenesis frequency and the sizes of embryogenic callus clusters were scored after 30 d. Experiment 3: embryogenic callus differentiation and conversion of SE to plants. To promote SE formation and plant regeneration, embryogenic calli were cultured on three basic media, namely, MMS, SH (Schenk and Hildebrandt, 1972), and modified SH medium (MSH, Zhang and Wang, 2002) containing 2 or 3 mg l21 Kin. MSH contained SH macronutrients and MMS micronutrients and organic components. In these media, three levels of sucrose (10, 20, and 30 g l21) were used. Callus clusters (3 £ 3 mm2) were placed on the medium in a 50 ml culture bottle in their original polarity. Each treatment consisted of three replicates, each containing 50 callus clusters with five calli per culture bottle. Embryogenic calli were cultured under the same conditions as described for callus induction. The callus– differentiation ratio, mean number of shoots per callus cluster, and number of roots per shoot were scored after 30 d. The callus differentiation ratio was defined as the number of embryogenic calli forming at least one shoot divided by the total number of calli. Experiment 4: regeneration of other tall fescue cultivars. After the protocol for high-frequency plant regeneration of cv. ‘Surpro’ was optimized, similar culture optimization for plant regeneration of other tall fescue cultivars were also conducted. Embryogenic calli were induced by culturing husked seeds of cvs. ‘Coronado’, ‘Summer Lawn,’ and ‘Fawn’ on MMS medium supplemented with 7.0 mg l21 2,4-D and 0.05 mg l21 Kin for 2 wk, followed by transferring to MMS medium containing 4.5 mg l21 2,4-D and 0.2 mg l21 Kin. For SE induction, embryogenic calli were transferred to MMS medium containing 2.0 mg l21 2,4-D and 0.2 mg l21 Kin for 3 wk. The resulting SEs were then transferred to MSH medium containing 2.0 mg l21 Kin and 20 g l21 sucrose for plant conversion. For each step, each treatment consisted of three replicates, each containing 50 explants, with five explants per culture bottle. The SE induction ratio, the callus–differentiation ratio, and number of plantlets regenerated from callus clusters were scored after 30 d. The SE induction ratio was defined as the number of calli forming somatic embryos divided by the total number of calli. Statistical analysis. All experiments were conducted using the randomized designs. The mean of the callus–induction ratio, SE–induction ratio, and differentiation ratio (the total number of embryogenic calli with at least one shoot divided by the number of calli used for differentiation) and the mean number of plantlets per callus (the total number of plantlets divided the number of calli used for differentiation) were calculated. To fulfill the normal distribution assumption, data were arcsine transformed. The differences between the means were analyzed statistically using Duncan’s multiple range test (Duncan, 1955).

Results and Discussion Results from previous studies showed that the presence of 2,4-D at 5 – 9 mg l21 promoted callus induction from mature seeds of tall fescue (Bai and Qu, 2001; Wang et al., 2003). The concentration of

2,4-D required for callus initiation differs among seeds and tall fescue cultivars. Our results indicated that the frequency of callus formation varied significantly among cultivars and media (data not shown). MMS medium was shown to be more suitable for tall fescue callus induction than MS medium (Table 1). In general, calli formed on the germination side of seeds. Calli appeared to be sticky and watery initially but, after several days of growth, some calli became compact, yellowish in color, and had a variable proliferation rate. The callus size and callus initiation time were significantly affected by the concentration of 2,4-D. We observed a correlation between the callus initiation time and 2,4-D concentration, which has also been reported in ryegrass (Zhang and Wang, 2003). Callus initiation was accelerated with an increase in 2,4-D concentration. In contrast, callus size was negatively correlated with the 2,4-D concentration (Table 1). On the MMS5 medium, we obtained the highest callus-induction rate (90.3%) from the cv. ‘Surpro’. On the other hand, the time required for callus initiation was somewhat longer on MMS5 than MMS7, MMS9 or MMS11. In the first experiment, with the exception of MMS11, the callusinduction rate was generally higher than that on MS9. The difference between MS and MMS medium lies in the organic components; vitamin B (thiamine, nicotinic acid, and pyridoxine) and casein enzymatic hydrolysate replaced the organic components of MS medium in the MMS medium. These results indicate that the use of vitamin B in conjunction with casein enzymatic hydrolysate is suitable for callus formation from tall fescue seeds. The capacity of callus formation is greatly influenced by the medium during subculture, especially when nutrients or hormone in the medium are present at the sub-optimal level (Wang et al., 2002a). In general, hormones are the most likely candidates for regulation of callus developmental switches (Wang et al., 2001). Results in this study showed that callus proliferation and somatic embryogenesis induction ratios of cv. ‘Surpro’ were greatly influenced by the concentration of 2,4-D and Kin (Table 2). For the same level of 2,4-D with increasing concentration of Kin (0.1–0.2 mg l21), the somatic embryogenesis induction ratios of the calli tended to increase, while the callus size decreased. Furthermore, for the same level of Kin, an increase in the concentration of 2,4-D (2–5 mg l21) increased the size of calli, whereas the somatic embryogenesis induction ratios decreased. The balance of 2,4-D and Kin concentrations is important to callus differentiation or proliferation, whereas the presence of 2,4-D at supra-optimal levels promoted callus induction and root formation. Thus after callus formation, it is necessary to reduce

TABLE 1 EFFECT OF MEDIUM AND 2,4-D ON CALLUS FORMATION FROM CULTURED MATURE SEEDS OF TALL FESCUE CV. ‘SURPRO’ Hormone (mg l21) Medium MMS3 MMS5 MMS7 MMS9 MMS11 MS9 (MS)

2,4-D 3 5 7 9 11 9

Kin 0.05 0.05 0.05 0.05 0.05 0.05

Callus formation frequency (%) z

76.2 ^ 0.2 c 90.3 ^ 0.3 a 85.4 ^ 0.1 b 79.6 ^ 0.8 c 48.5 ^ 0.6 e 61.4 ^ 0.6 d

Callus size (mm) z

7.3 ^ 0.5 a 7.6a ^ 0.4 b 5.5 ^ 0.6 b 4.2 ^ 0.3 c 2.7 ^ 0.2 d 4.3 ^ 0.2 c

Initiation of callus (d) 10 ^ 0.5 az 9 ^ 0.2 a 7 ^ 0.3 b 6 ^ 0.3 b 6 ^ 0.4 b 7 ^ 0.4 b

z Values are the mean ^ SE of three replicates, each with 100 seeds cultured in 10 bottles. Means within columns followed by the same letter are not significantly different at the 5% level of significance.

116

ZHANG ET AL. TABLE 2

EFFECT OF 2,4-D AND KINETIN ON PROLIFERATION AND SOMATIC EMBRYOGENESIS OF CV. ‘SURPRO’ CALLI GROWN ON MMS7 MEDIUM AFTER SUBCULTURE FOR 30 D Hormone (mg 121) Medium MMS5 MMS MMS MMS MMS MMS MMS MMS

2,4-D 2 2 2 4 4.5 5 5

Kin 0.1 0.2 0.5 0.2 0.2 0.1 0.2

SE induction ratio (%)

Embryogenic callus (mm)

10.2 ^ 0.4 gz 85.3 ^ 0.3 b 91.4 ^ 0.4 a 53.3 ^ 0.6 e 75.6 ^ 0.2 c 86.5 ^ 0.1 b 40.5 ^ 0.8 f 63.6 ^ 0.6 d

10.2 ^ 0.5 bcz 12.3 ^ 0.3 ab 11.4 ^ 0.4 b 6.1 ^ 0.5 c 10.2 ^ 0.2 bc 13.6 ^ 0.6 a 9.5 ^ 0.4 c 8.3 ^ 0.4 c

z

Values are the mean ^ SE of three replicates, each with 50 calli cultured in 10 bottles. Means within columns followed by the same letter are not significantly different at the 5% level of significance.

the 2,4-D concentration to promote somatic embryogenesis and callus proliferation. In this study, formation of SEs from calli was observed after calli were transferred to the medium with decreased concentration of 2,4-D and increased Kin for 2 – 3 wk. On this medium, some parts of the plump and sticky callus clusters became yellowish, grainy, and compact initially but after 2 – 5 d, small visible green global embryos appeared on the surface of the compact yellowish calli. Shortly after, small shoots with roots began to develop. In contrast, calli maintained on MMS5 produced only hairy roots on the callus surface. MMS medium containing 4.5 mg l21 2,4-D and 0.2 mg l21 Kin showed a higher SE induction ratio, larger size of embryogenic calli, and better callus appearance compared to those grown on other media. This medium was, therefore, selected for embryogenic callus maintenance and proliferation. On the other hand, MMS medium containing 2 mg l21 2,4-D and 0.2 mg l21 Kin showed the highest SE induction ratio and was chosen for SE induction during culture optimization. Acquisition of regenerated plants largely relies on embryogenic callus differentiation or conversion of SEs into plants. Our results

indicated that the concentrations of macronutrients, organic nutrients, sucrose, and Kin in the medium had significant effects on cell differentiation leading to plant regeneration of tall fescue. MSH containing 2 mg l21 Kin and 20 g l21 sucrose showed a particularly high plant regeneration rate (96.3%), mean shoot number per callus (28.2), and mean root number per plant (7.2). It was, therefore, selected as the plant-regeneration medium. In contrast, MMS medium containing the same level of Kin and sucrose showed lower callus-differentiation rate (37.2%) and mean shoot number per callus (5.4). On the other hand, the basic SH medium containing the same level of Kin and sucrose yielded a callus-differentiation rate of 71.3% and 12.4 shoots per callus. The difference between the MMS and MSH media is the concentration of macronutrients, while organic components in MSH and SH differ. Results of this study clearly show that the macronutrient content of SH in combination with vitamin B and casein enzymatic hydrolysate is most suitable for the conversion of embryogenic calli to plants in tall fescue (Tables 3 and 4). The role of sucrose in regulating plant development and embryogenic callus differentiation has been reported previously (Ohto et al., 2001; Zhang and Wang, 2002, 2003). In this study, the callus-differentiation rate of cv. ‘Surpro’ grown on MMS medium increased with an increase in sucrose concentrations, whereas the differentiation rate for the MSH medium decreased when the sucrose concentration increased from 20 to 30 g l21. With respect to rooting, exogenous cytokinin is usually not required because Kin restrains root formation to some degree. In this study, somatic embryogenic calli were converted to plantlets on the differentiation medium containing 2 mg l21 Kin. The regenerated plantlets produced roots at the same time as they developed shoots. After 30 d, the regenerated plantlets could be transplanted to pots (Fig. 1). Beside cv. ‘Surpro,’ we found a wide variation in the callus induction frequency among cvs. ‘Coronado,’ ‘Summer Lawn’ and ‘Fawn.’ Although the callus-induction frequency has been used as an index for evaluating plant regeneration (Dale and Dalton, 1983; Bai and Qu, 2001), it may not be a reliable parameter as calli can be proliferated in a large scale by subculturing. On the other hand, the callus induction ratio may not reflect the frequency of SE formation or plant differentiation from callus. This is supported by the results

TABLE 3 EFFECTS OF MEDIUM COMPONENTS ON PLANT REGENERATION FROM EMBRYOGENIC CALLI OF CV. ‘SURPRO’ AFTER 1 MO. OF CULTURE Medium MMS MMS MMS MSH MSH MSH MSH SH SH SH

Kinetin (mg l21)

Sucrose (g l21)

Differentiation rate of calli

Mean number of green shoots per callus

Mean number of roots per shoot

2 2 2 2 2 2 3 2 2 2

10 20 30 10 20 30 20 10 20 30

30.2 ^ 0.4 h 37.2 ^ 0.4 g 54.4 ^ 0.3 f 91.2 ^ 0.9 b 96.3 ^ 0.4 a 80.6 ^ 0.4 d 85.4 ^ 0.5 c 56.3 ^ 0.8 f 71.3 ^ 0.4 e 67.6 ^ 0.6 e

3.6 ^ 0.4 e (albino) 5.4 ^ 0.8 d (albino) 7 ^ 0.2 d 24.5 ^ 0.1 b 28.2 ^ 0.6 a 22.6 ^ 0.3 b 22.1 ^ 0.2 b 6.2 ^ 0.4 e 12.4 ^ 0.3 c 7.6 ^ 0.4 e

0g 1.1 ^ 0.1 f 2.3 ^ 0.5 e 6.5 ^ 0.3 b 7.2 ^ 0.1 a 5.3 ^ 0.7 c 3.8 ^ 0.4 d 0g 2.8 ^ 0.2 e 1.4 ^ 0.4 f

Values are the mean ^ SE of three replicates, each with 50 calli cultured in 10 bottles. Means within columns followed by the same letter are not significantly different at the 5% level of significance.

117

PLANT REGENERATION FROM TALL FESCUE TABLE 4 DIFFERENTIAL CULTURE RESPONSES OF FOUR TALL FESCUE CULTIVARS Cultivars

SE induction ratio (%)

Differentiation ratio (%)

Number of plantlets

Surpro Coronado Summer Lawn Fawn

81.4 ^ 0.5 a 79.8 ^ 0.2 a 78.6 ^ 0.1 a 82.3 ^ 0.6 a

95.3 ^ 0.1 a 94.3 ^ 0.8 a 93.5 ^ 0.6 a 94.2 ^ 0.5 a

18.9 ^ 0.2 a 16.4 ^ 0.3 b 15.3 ^ 0.1 b 12.6 ^ 0.2 c

Values are the mean ^ SE of three replicates, each with 50 calli cultured in 10 bottles. Means within columns followed by the same letter are not significantly different at the 5% level of significance.

FIG . 1. The regeneration of tall fescue cv. ‘Surpro’. A–C, Different stages of somatic embryogenesis of tall fescue calli; D, plants obtained from tall fescue somatic embryos; E, regenerated plants transplant and survive.

of this study, in which the callus induction frequency of different tall fescue cultivars varied significantly (data not shown) but, after subculture and SE induction, embryogenic calli of all cultivars showed high rate of plant regeneration. The protocol described in this study is suitable for high frequency plant regeneration of tall fescue cvs. ‘Surpro’, ‘Coronado’, ‘Summer Lawn,’ and ‘Fawn’ via callus proliferation and induction of SEs. Acknowledgments This work was supported by Hi-Tech & Program of China (J2002-B-0003). We are grateful to Dr Fei-shui Zhang and Dr Lin-xia Sun for their constructive suggestions.

References Bai, Y.; Qu, R. An evaluation on callus induction and plant regeneration of 25 turf-type tall fescue (Festuca arundinacea Schreb.) cultivars. Grass Forage Sci. 55:326–330; 2000. Bai, Y.; Qu, R. Factors influencing tissue culture responses of mature seeds and immature embryos in turf-type tall fescue. Plant Breed. 120:239– 242; 2001.

Bettany, A. J. E.; Dalton, S. J.; Timms, E.; Manderyck, B.; Dhanoa, M. S.; Morris, P. Agrobacterium tumefaciens-mediated transformation of Festuca arundinacea (Schreb.) and Lolium multiflorum. Plant Cell Rep. 21:437–444; 2003. Buckner, R. C.; Powell, J. B.; Frakes, R. V. Historical development. In: Buckner, R. C.; Bush, L. P., eds. Tall fescue. Agronomy 20: 1 –8; 1979. Dale, P. J.; Dalton, S. J. Immature inflorescence culture in Lolium, Festuca, Phleum, and Dactylis. Z. Pflanzenphysiol. 111:39–45; 1983. Dalton, S. J. Plant regeneration from cell suspension protoplasts of Festuca arundinacea Schreb. (tall fescue), and Lolium perenne L. (perennial ryegrass). J. Plant Physiol. 132:170–175; 1988. Duncan, D. B. Multiple range and multiple F test. Biometrics 11:1– 42; 1955. Eigenza, G. C.; Dahleen, L. S. Callus production, regeneration and evaluation of plants from cultured inflorescences of tall fescue (Festuca arundinacea Schreb.). Plant Cell Tiss. Organ Cult. 22:7– 15; 1990. Fennell, S.; Bohorova, N.; Ginkel, M. V.; Grossa, J.; Hoisington, D. Plant regeneration from immature embryos of 48 elite CIMMYT bread wheat. Theor. Appl. Genet. 92:163–169; 1996. Garcia, A.; Dalton, S. J.; Humphreys, M. O. Reproductive disturbances and phosphoglucoisomerase instability in Festuca arundinacea (tall fescue) plants regenerated from callus and cell suspension cultures. Heredity 73:355–362; 1994.

118

ZHANG ET AL.

Kasperbauer, M. J.; Springer, W. D. Haploid plants by anther-panicle culture of tall fescue. Crop Sci. 19:457–460; 1980. Kuai, B.; Dalton, S. J.; Bettany, A. J.; Morris, P. E. Regeneration of fertile transgenic tall fescue plants with a stable highly expressed foreign gene. Plant Cell Tiss. Organ Cult. 58:149– 154; 1999. Lowe, K. W.; Conger, B. V. Root and shoot formation from callus culture of tall fescue. Crop Sci. 19:397– 400; 1979. Murashige, T.; Skoog, F. A revised medium for rapid growth and bioassay with tobacco tissue cultures. Physiol. Plant. 15:473–497; 1962. National Turfgrass Evaluation Program (NTEP). National tall fescue test– 1992. Final report 1993–95. NTEP No. 96-13. Beltsville, MD: USDA-ARS; 1996. Ohto, M.; Onai, K.; Furukawa, Y. Effects of sugar on vegetative development and floral transition in Arabidopsis. Plant Physiol. 127:252–261; 2001. Sam, B. H.; Fang, S. W.; Tracy, K. T. Transgenic turf-type tall fescue (Festuca arundinacea Schreb.) plants regenerated from protoplasts. Plant Cell Rep. 11:601–604; 1992. Schenk, B. V.; Hildebrandt, A. C. Medium and techniques for induction and growth of monocotyledonous and dicotyledonous plant cell cultures. Can. J. Bot. 50:199–204; 1972. Takamizo, T.; Fukase, N.; Saruwatari, Y.; Suginobu, K. Genetic variation in plant regeneration from callus culture of tall fescue (Festuca

arundinacea Schreb.) cultivars. Jpn. Agric. Res. Q. 28:200–205; 1994. Wang, Y. X.; Ruemmele, B. A.; Chandlee, J. M.; Sullivan, W. M.; Knapp, J. E.; Kausch, A. P. Embryogenic callus induction and plant regeneration media for bentgrasses and annual bluegrass. In Vitro Cell. Dev. Biol. Plant 38:460– 467; 2002b. Wang, Z. Y.; Bell, J.; Ge, Y. X.; Lehmann, D. Inheritance of transgenes in transgenic tall fescue (Festuca arundinacea Schreb.) 39: 277 –282; 2003. Wang, Z. Y.; Hopkins, A.; Main, R. Forage and turf grass biotechnology. Crit. Rev. Plant Sci. 20:573–619; 2001. Wang, Z. Y.; Hopkins, A.; Scott, M. Plant regeneration from embryogenic cell suspension cultures of Lolium temulentum. In Vitro Cell. Dev. Biol. Plant 38:446–450; 2002a. Wang, Z. Y.; Takamizo, T.; Iglesias, U. S. Transgenic plants of tall fescue (Festuca arundinacea Schreb.) obtained by direct gene transfer to protoplasts. Bio/Technology 10:691–696; 1992. Zhang, W. J.; Wang, T. Construction of a high frequency plant regeneration system from callus cultures of alfalfa and a study of the effects. Sci. Agric. Sin. 35:1579–1583; 2002 (in Chinese). Zhang, W. J.; Wang, T. A study on the tissue culture and plant regeneration of perennial ryegrass. Acta Agrestia Sin. 11:219–222; 2003 (in Chinese).