E L S E V I E R Field Crops Research 37 (1994) 113-119

F i e l d C r o p s R e s e a r c h

Effect of in vitro culture on rust resistance and yield in sugarcane

J.P. P6rosa'*,E. Bonnel a'l, D. Roques a, F. Paulet b

aD~partement des Cultures Annuelles, Centre de Cooperation lnternationale en Recherche Agronomique pour le D~veloppement, 97487 Saint Denis, lie de la R~union France bBP5045, 34032 MontpeUier Cedex, France

(Received 6 September 1993; accepted 3 May 1994)

Abstract

The purpose of this study was to compare the susceptibility to rust (Puccinia melanocephala) and the yield characteristics of sugarcane plants derived either from callus or bud cultures with those of plants obtained by the normal method of vegetative propagation. Plants regenerated from callus cultures (somaclones) generally had higher rust severity, thinner stalks and smaller leaves than the control. The magnitude of these effects depended on the method used to obtain the somaclones. Plants derived from bud cultures also showed an altered rust resistance and had improved yield components in two trials. The effect on rust resistance persisted for in vitro-derived plants even after two vegetative propagations. It is thus necessary to take account of these effects when using in vitro culture techniques for sugarcane improvement or mass propagation.

Keywords: Juvenile traits; Micropropagation; Puccinia melanocephala; Rust resistance; Somaclonal variation; Sugarcane; Yield

1. Introduction

Plant tissue culture and the regeneration of whole plants from cultured cells or tissues can result in exten- sive genetic changes (Larkin, 1987; Brown, 1991). Such somaclonal variation was proposed as a new method for improvement of sugarcane (complex hybrids of Saccharum) in the early 1970s (Heinz, 1973). Variants of sugarcane were successfully iso- lated which expressed improved resistance to diseases (Krishnamurthi and Tlaskal, 1974; Heinz et al., 1977; Liu, 1981; Larkin and Scowcroft, 1983). Other para-

*Corresponding author. Present address: Institut National de la Recherche Agronomique, Station de Reeherches Viticoles, BP 13, 34751 Villeneuve les Maguelonne Cedex, France. ~Present address: Germicopa S.A., 8 rue Olivier de Serres, 29103

Quimper, France.

0378-4290/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved S S D 1 0 3 7 8 - 4 2 9 0 ( 9 4 ) 0 0 0 3 1 - 7

meters like sucrose content (Liu et al., 1984), flower- ing and isozyme patterns (P6ros et al., 1989) may be affected. It was also reported that the extent of soma- clonal variation in sugarcane may be more limited than expected (Irvine, 1984; Kresovitch et al., 1986; Lour- ens and Martin, 1987; Bonnel et al., 1988). Anderlini and Kostka (1986) indeed suggested using regenera- tion from callus culture as an alternative method for clonal propagation. Nevertheless, when somaclones were compared to material obtained by the normal method of vegetative propagation, whole population shifts were observed. Examples include changes in resistance to eyespot (Larkin and Scrowcroft, 1983), to rust due to Puccinia melanocephala H. and P. Sydow (P6ros et al., 1989) and to smut (Bayley and Bechet, 1989), and changes in yield components like tillering and stalk diameter (Kresovitch et al., 1986; Bonnel et al., 1988; Bayley and Bechet, 1989), and in flowering

114 J.P. Pdros et al./ Field Crops Research 37 (1994) 113-119

(Hogarth, pers. commun., 1987; Bonnel et al., 1988). However, the shifts were generally observed when

attempts were made to obtain variants. Thus, the exper- imental designs used were not suitable to study those effects. Moreover, the effects due to in vitro culture without regeneration were often not evaluated since plants derived from bud culture were not included in trials. Finally, the stability of the effects after propa- gation of the material in the field was not assessed.

It thus appeared necessary to conduct further exper- iments in order to measure the extent and stability of the effects of in vitro culture of sugarcane. We compare here the rust susceptibility and agronomic perform- ances of somaclones and plants derived from bud cul- ture with those obtained from the normal method of vegetative propagation in repeated field trials.

2. Materials and methods

Two consecutive experiments were performed in Reunion Island with the cultivar B43-62. The material from Experiment 1 provided part of the material tested in Experiment 2. Both experiments are presented in Fig. 1. Methods for transplantation in pots and culture in the greenhouse were those of Bonnel et al. (1988). Plants in field trials were fertilized once with 105 kg- ha- 1 of N, 49 kg. ha- ~ of P and 168 kg- ha- 1 of K.

2.1. Experiment I

Somaclones of cv. B43-62 were obtained in Mont- peUier as described earlier (P6ros et al., 1989) and sent to Reunion Island in October 1985. Somaclones (S) and bud cultures (B) initiated in April 1984 as described by P6ros et al. (1989) were simultaneously subcultured once. The control plants (C) were included when plantlets were transplanted in pots. In the field (P0 trial), plants of the three treatments (S, B and C) alternated along the row (0.5-m spacing between plants, 1-m interrow spacing). Data were collected in December 1986 on number of tillers per plant, stalk length (SL) from the stalk base to the first visible dewlap, length (LL) and width (LW) of the blade of the youngest leaf with a visible dewlap (TVD leaf). The number of uredinia of Puccinia melanocephala per cm 2 leaf area was determinated as described by

P6ros (1989). Two main stalks and corresponding TVD leaves were observed for each plant.

2.2. Experiment2

The P0 trial was ratooned in May 1987 and ten culms of S, B and C plants were cut in July 1987. Cylinders of juvenile leaf tissue were placed on a modified Mura- shige and Skoog medium containing 3 mg. 1- ~ of 2,4 dichlorophenoxyacetic acid (Bonnel et al., 1983). Six weeks later, the callussing explants were placed on the same medium without 2,4-D (propagation medium) to induce regeneration. To study the effect of the length of callus culture, half of the calli obtained from B plants were placed on the propagation medium after 12 weeks of culture. Only one plantlet was selected from each regenerative explant. The somaclones were identified as follows: SS originated from S plants, SB1 and SB2 originated from 6-week-old calli and 12-week-old calli of B plants, respectively, and, SC originated from C plants. All somaclones were subcultured twice on a rooting medium (propagation medium with 60 g. 1-1 sucrose) simultaneously with plantlets from bud cul- tures (B).

Control (C) plants were included in the experiment when SB1, SB2, SC and B plantlets were transferred to pots. The TVD leaf was inoculated with uredospores of P. melanocephala in a settling tower as described by P6ros (1989). The greenhouse trial was arranged in a randomized complete block (RCB) design with 6 replicates and 2 plants of each treatment per replicate. The number of uredinia per cm 2 leaf area was deter- minated 15 days after inoculation.

The P'0 trial was arranged in a RCB design with six replicates and a row with 6 plants (0.5-m interplant spacing, 1.5-m interrow spacing) of each treatment per replicate. Number of uredinia per cm 2 leaf area was recorded monthly from April to July 1989 for the TVD leaves of the two main stalks of each plant. We calcu- lated the area under the disease progress curve (AUDPC) using the formula:

AUDPC = Ei( ni + ni + 1 ) / 2,

with n~---number of uredinia per cm 2 at month i and ni+ ~ = number of uredinia per cm 2 at month i + 1. Data were collected in August 1989 on SN per plant, SL, SD, LL and LW of TVD. The two main stalks and corresponding TVD leaves were observed for each

J.P. P dros et al. /Fie ld Crops Research 37 (1994) 113-119 115

I Somacloncs

S

Y I

C

v Feb. 1986: Transplantalion in pots

Y Y Y May 19S6:P0 trial (4o laamSlz~r t~mnmt)

July 1987: ~ a n e production

SB1 SB2 SC

Y Y Nov. 1988: Greenhouse real, RCB

2 plants per rqaic~c I I

r ! Dec. 19~: P'0 lzial, RCB

6 ~ 6 n a o m i . 6 ~anu wr na~ica~

V' Sept. 1989: I~l trial, RCB 6 t a~ t me t , 8 na~kat~

3-m mw t~r aqflima:

Y March 1991:I~2 ~k~l, RCB 3 m = m a t ~ 6 m a i t r e . 3 x 5-m rowper ~

Fig. 1. Description and calendar of experiments conducted to study the effects of in vitro culture on rust susceptibility and yield components of sugarcane; S: somaclones, B: bud cultures, C: control plants, S S: somaclones from calli of somaclones S, SB 1: somaclones from 6-week-old calli of B plants, SB2: somaclones from 12-week-old calli of B plants, and RCB: random complete block designs. P0 and P'0 trials included plants transplanted from greenhouse without propagation, P' 1 and P'2 trials included plants propagated from P'0 and P' 1 trials, respectively.

! 16 J.P. Pdros et al./Field Crops Research 37 (1994) 113-119

plant. In addition, leaf tissues were analyzed for major nutrient contents (P~ros, 1990).

larger number of tillers and longer main stalks whereas S plants had smaller leaves.

The P' 1 (propagation 1 ) trial was arranged in a RCB design with eight replicates. Each plot consisted of a 3- m-long row planted with seven 3-bud cuttings per meter. Natural rust infection was recorded monthly from December 1989 to May 1990 on five TVD leaves per plot and we calculated AUDPC as above. Data were collected on SN per plot, SL, and SD in July 1990. Three stalks were observed per plot. Plots were har- vested in August 1990 and weighted. Data on LL and LW of TVD leaf were collected in first ratoon in Feb- ruary 1991.

The P'2 trial (propagation 2) included only the SC, B and C treatments. It was arranged in a RCB design with six replicates. Each plot consisted of three rows, 5 m in length, planted with seven 3-bud cuttings per meter. Number of uredinia per cm 2 leaf area was recorded monthly from June 1991 to November 1991 on 10 TVD leaves per plot (central row) and AUDPC was calculated. In November 1991, SN was recorded in the central row and SL and SD were determinated for 10 stalks in the central row. Stalks from entire plots were weighed. In order to compare the results of P' 1 and P'2 trials, SN and plot weight were calculated per row of I m. Technological parameters were determined in a routine laboratory. A stalk sample was ground and the resulting pulp was pressured. Fiber % was calcu- lated as 0.55 × weight of pressured pulp divided by weight of pulp. Brix was measured with a refractometer and Pol with a polarimeter after clarification of the juice. Purity was calculated as (Pol/Brix) × 100.

3. Results

3.1. Experiment I

Survival of transplanted plants was on an average 81% with no large differences among the three treat- ments (Table 1 ). The S plants seemed more susceptible to rust infection but this difference was not significant at the 5% level (P=0.058) (Table 1). Transplanting at the beginning of the austral winter of potted plants was not favourable for growth which appeared very weak. Significant differences were demonstrated among the treatments (Table 1): B plants exhibited a

3.2. Experiment 2

The number of uredinia per c m 2 leaf area obtained after inoculation of potted plants was significantly higher for in vitro-derived than for C plants (Table 2). No mortality was observed after moving the potted plants to the field. The lower rust susceptibility of C plants was revealed by AUDPCs and SS plants were significantly more affected than other in vitro-derived plants (Table'2). Number of stalks and stalk length were similar for all treatments after 8 months of growth (Table 2). However, significant differences were found for stalk diameter and leaf size. These differences varied according to the method which was used to obtain the in vitro-derived plants. Clearly, SS plants had thinner stalks and smaller leaves than those from all other treatments, SC, SB 1 and SB2 plants had sig- nificantly decreased stalk diameter, and SC and SB 1 had decreased leaf width compared with the control (Table 2). Compared with in vitro-derived plants, C plants had a non-significant reduced N content (0.94 vs. 1.03% of dry weight) and significant higher K con- tent ( 1.39 vs. 1.25% of dry weight) (data not shown).

AUDPCs for the P' 1 trial showed that only SS plants exhibited significantly higher rust susceptibility than the C plants (Table 3). In addition, SS plants had a significant decrease in stalk number, stalk diameter, leaf width, and a dramatic decrease in yield (Table 3). Differences among other groups of plants seemed attenuated. However, yield was significantly lower for C than B plants (Table 3).

In the P'2 trial, only the treaments C, B and SC were compared. All previous experiments clearly demon- strated that SB 1 and SB2 somaclones showed similar behaviour to SC somaclones and that the SS group exhibited dramatic changes. The AUDPCs indicated a significantly higher susceptibility for in vitro-derived plants than C plants (Table 4). Among the agronomic parameters, only stalk diameter was affected by the treatment. The SC plants had thinner stalks but this had no significant consequence for the stalk yield (Table 4). No significant difference was noted for Brix, purity, and fiber.

J.P. P~ros et al. / Field Crops Research 37 (1994) 113-119 117

Table 1 Susceptibility to Puccinia melanocephala, stalk characteristics and leaf size of sugarcane derived from cuttings (C), bud culture (B), and callus culture (S) after transfer of potted plants to the field (Exp. 1, Trial P0)

Source a Total Uredinia Tillers Main stalk Leaf blade Leaf blade plants per cm 2 per plant length length width (no.) leaf area (no.) (cm) (cm) (cm)

C 32 9.0 19.1 54.1 130.6 3.4 B 34 9.2 23.1" 61.9" 126.7 3.3 S 31 11.5 18.9 53.5 112.5" 2.4**

aRefer to Materials and Methods and Fig. 1. */**Significant difference (P < 0.05/P < 0.01 ) according to the Student test.

Table 2 Susceptibility to Puccinia melanocephala, stalk characteristics and leaf size of sugarcane derived from cuttings (C), bud culture (B), and callus culture (SC, SB1, SB2, SS) (Exp. 2, greenhouse test and P'0 trial)

Source x Rust severity Stalk characteristics Leaf blade size

U cm - 2 y AUDPC z No plant- l Length Diameter Length Width (em) (era) (era) (cm)

C 19.5a 9.6a 6.1 97.2 2.5a 124. la 3.9a B 25.8b 13.4b 7.1 109.3 2.4a 122.4a 3.7ab SC 28.0b 13.0b 7.3 111.1 2.3b 119.9a 3.5bc SB I 30.8b 13.5b 6.5 96.2 2.2b 114.6a 3.3c SB2 30.3b 13.0b 7.7 100.1 2.2b 116.4a 3.6abe SS 30.5b 20.2c 6. I 87.1 1.8c 106.8b 3.0d

XRefer to Materials and Methods and Fig. I. YNumber of uredinia per cm 2 after inoculation in the greenhouse. ZArea under disease progress curve in the field. Means followed by different letters are significantly different at P < 0.05 (Newman-Keuls test).

Table 3 Susceptibility to Puccinia melanocephala, stalk characteristics and leaf size of sugarcane derived from cuttings (C), bud culture ( B ), and callus culture (SC, SB 1, SB2, SS) after one vegetative propagation in the field ( Exp. 2, Trial P' 1 )

Source x AUDPC y Stalk characteristics z Leaf blade size

No m - ~ Length Diameter kg m - 1 Length Width (cm) (em) (cm) (cm)

C 31.1a 16.2a 214.6 2.5a 19.0b 171.6 5.1a B 30.0a 19.1a 221.8 2.5a 25.3a 168.2 5.1a SC 31.4a 21.8a 219.8 2.4a 23.7ab 173.5 4.8a SB 1 35.5a 21.4a 224,4 2.4a 22.4ab 171.2 5.0a SB2 35.4a 21.5a 217.5 2.3a 22.3ab 170.7 4.9a SS 57.0b 10.7b 208.5 1.9c 8.5c 173.3 3.6b

XRefer to Materials and Methods and Fig. 1. YArea under disease progress curve in the field. ZNumber and fresh mass are expressed per meter of sugarcane row. Means followed by different letters are significantly different at P < 0.05 (Newman-Keuls test).

118 J.P. Pdros et al. / Field Crops Research 37 (1994) 113-119

Table 4 Susceptibility to Puccinia melanocephala and yield components of sugarcane derived from cuttings (C), bud culture (B), and callus culture (SC) after two vegetative propagations in the field (Exp. 2, trial P'2)

Source x AUDPC y Stalk characteristics z Technological traits

No m- ~ Length Diameter kg m- t Brix Purity Fiber (cm) (cm)

C 12.6a 14.8 162.0 3.0a 13.7 18.2 89.0 14.8 B 18.3b 15.1 160.2 3.0a 13.0 18.9 91.3 15.6 SC 18.9b 15. l 163.1 2.9b 12.9 17.7 87.7 15.0

XRefer to Materials and Methods and Fig. 1. YAUDPC: Area under disease progress curve in the field. ZNumber and fresh mass are expressed per meter of sugarcane row. Means followed by different letters are significantly different at P < 0.05 (Newman-Keuls test).

4. Discussion and conclusion

The somaclonal population obtained in Montpellier showed marked differences for rust susceptibility and stalk yield compared with the donor cultivar B43-62. The SS somaclones, obtained from this material after another cycle of culture, had the same defects. No dif- ferences appeared among somaclones obtained from C plants (SC) and from B plants (SB1 and SB2). The SC, SB 1 and SB2 somaclones were less affected than the SS somaclones but also had higher rust suscepti- bility and thinner stalks than the control plants. Thus, whole population shifts may vary according to the method used to obtain the somaclones or to the number of in vitro culture cycles.

The plants derived from bud cultures conformed more closely to those of C plants. However, B plants also exhibited an increased susceptibility to rust and had a higher yield than C plants in P0 and P' 1 trials. This can be explained only by the occurrence of epi- genetic effects, i.e. physiological effects due to in vitro culture. Such effects were responsible for the shifts observed in populations of plants derived from either callus or bud.

As quoted by Nozeran (1978), in vitro-cultured plants have meristems with a reduced size. This would explain the expression of juvenile characteristics already observed in several plant species (Nozeran, 1980). Grenan (1982) specifically studied such mod- ified expression of unaltered genome in grapevine, another vegetatively propagated crop. Sugarcane in vitro-derived plants do in fact reveal some similarities to seedlings such as high tillering and smaller leaves.

In addition, breeding for rust was facilitated by the clear indication of the disease at the seedling stage.

Regarding the stability of the effects, evidence is presented here that in vitro-derived plants have an increased rust susceptibility but no difference for yield after two vegetative propagations. Interactions between epigenetic changes and environmental conditions may be suspected, since differences between SC, B and C plants were not similar in P' 1 and P'2 trials.

In conclusion, epigenetic changes due to the use of in vitro culture methods in sugarcane may be con- founded with true genetic variation. Their persistance after normal vegetative propagation indicated that more time is necessary for breeders to be confident of true genetic variation. Moreover, any commercial applica- tion of tissue culture needs foregoing experiments to evaluate, on a cultivar-by-cultivar basis, the magnitude and direction of the induced shifts. In particular, the advantage of tissue culture to provide pathogen-free seed material (Anderlini and Kostka, 1986) would be nullified in the ~ ,-.se of stable altered disease resistance of in vitro-derived plants.

Acknowledgements

The authors thank H. Lombard and H. T61ismard for their technical assistance. They also thank the Centre Technique Interprofessionnel de la Canne et du Sucre, Saint Denis, Ile de la R6union, for the technological analyses.

J.P. Pdros et al. / FieM Crops Research 37 (1994) 113-119 119

References

Anderlini, T.A. and Kostka, S.J., 1986. Initial yield responses of Kleentek tissue culture produced seed cane in Louisiana. Proc. Int. Soc. Sugar Cane Technol., 19: 391-401.

Bayley, R.A. and Bechet, G.R., 1989. A comparison of seedcane derived from tissue culture with conventional seedcane. Proc. S. Afr. Sugar Technol. Assoc., 63: 125-129.

Bonnel, E., Demady, Y. and Essad, S., 1983. Evolution anatomique des tissus folialres de eanne h sucre (Saccharum sp.) cultivts in vitro. Can. J. Bot., 61: 830-836.

Bonnel, E., P~ros, J.P. and Girard, J.C., 1988. Evaluation of sugar- cane somaclones for gumming disease resistance and agronomic performances. Sugar Cane, 4: 15-17.

Brown, P.T.H., 1991. The spectrum of molecular changes associated with somaclonal variation. Int. Assoc. Plant Tissue Cult. News- lea., 66: 14-25.

Grenan, S., 1982. Quelques r~flexions ~t propos de modifications morphogtn~tiques cons~cutives h la culture in vitro chez la vigne. Ann. Sci. Nat. Bot., 4: 135-146.

Heinz, D.J., 1973. Sugarcane improvement through induced muta- tions using vegetative propagules and cell culture techniques. In: Induced Mutations in Vegetatively Propagated Plants. Int. At. Energy Agency, Vienna, 1972, pp. 53-59.

Heinz, D.J., Knsnamurthi, M., Nickell, L.G. and Maretzki, A., 1977. Cell, tissue and organ culture in sugarcane improvement. In: J. Reinert and Y.P.S. Bajaj (Editors), Plant Cell, Tissue and Organ Culture. Springer-Verlag, Berlin, pp. 3-17.

lrvine, J.E., 1984. The frequency of marker changes in sugarcane plants regenerated from callus culture. Plant Cell Tissue Organ Cult., 3: 201-209.

Kresovitch, S., McGee, R.E., Dr'awe, H.J. and Rivera, J.L., 1986. Variability of agronomic characters in populations of tissue cul-

ture-derived and vegetatively-propagated sugarcane. Proc. Int. Soc. Sugar Cane Teclmol., 19: 528-532.

Krisnamurthi, M. and Tlaskal, J., 1974. Fiji disease resistant Sac- charum oflficinarum vat'. Pindar sub-clones from tissue cultures. Proc. Int. Soc. Sugar Cane Technol., 15: 130-137.

Larkin, P.J., 1987. Somaclonal variation, history, method and mean- ing. Iowa State J. Res., 61: 393-434.

Larkin, P.J. and Scowcroft, W.R., 1983. Somaclonal variation and eyespot toxin tolerance in sugarcane. Plant Cell Tissue Organ Cult., 2: 111-121.

Liu, M.C., 1981. In vitro methods applied to sugarcane improvement. In: T.H. Thorpe (Editor), Plant Tissue Culture. Methods and Applications in Agriculture. Academic Press, London, pp. 299- 323.

Liu, M.C., Yeh, H.S. and Chen, W.H., 1984. Tissue and cell culture as aids to sugarcane breeding a high-sucrose and vigorously growing callicione 71-4829. Taiwan Sugar, 31: 77-83.

Lourens, A.G. and Martin, F.A., 1987. Evaluation of in vitro prop- agated hybrids for somaclonal variation. Crop Sci., 27: 793-796.

Nozeran, R., 1978. Polymorphisme des individus issus de la multi- plication vtgttative des vtgttaux suptrieurs, avec conservation du potentiel g~nttique. Physiol. Vtgtt., 16: 177-194.

Nozeran, R., 1980. La multiplication v~gttative chez les v~gttaux sup&ieurs. In: La Multiplication Vtg~tafive des Plantes Suptr- ieures. Gauthiers-Villars, Paris, pp. 1-29.

Ptros, J.P., 1989. Etude quantitative de la rtsistance de la canne ~t sucre h Puccinia melanocephala. Agron. Trop., 44: 123-127.

Ptros, J.P., 1990. Relations entre les teneurs en min~raux des limbes de canne ~t sucre et l'infection par Puccinia melanocephala. I. Essais en pots. Agron. Trop., 45: 205-212.

Ptros, J.P., Bonnel, E., Ftrtol, L. and Mauboussin, J.C., 1989. Altered rust resistance and yield components arising from leaf tissue and bud culture in sugarcane. Proc. Int. Soc. Sugar Cane Technol., 20: 732-739.