HORTSCIENCE 35(1):125–127. 2000. Materials and Methods

The allotetraploid somatic hybrid ‘Valencia Response of Cybrids and a Somatic + Femminello’ and the two ‘Femminello’ cybrid plants used for this study, one Hybrid of Lemon to tracheiphila diploid (2n = 2x = 18) and one tetraploid (2n = 4x = 36), were produced by polyethylene Infections glycol (PEG)–induced somatic fusion of nu- cellus-derived embryogenic protoplasts with Nicasio Tusa1 and Sergio Fatta Del Bosco protoplasts derived from nucellar seedlings, as previously described (Grosser et al., 1996; Centro di Studio per il Miglioramento Genetico degli Agrumi, C.N.R., Viale Tusa et al., 1990). Buds were grafted onto 2- delle Scienze, 90128 Palermo, Italy year-old sour (C. aurantium L.) seed- lings in 1989. Franco Nigro and Antonio Ippolito Mal secco resistance was assayed by spe- Dipartimento Protezione delle Piante dalle Malattie. Universita’ degli Studi cific tests performed in a growth chamber di Bari, Via Amendola, 165/a 70126, Bari, Italy maintained at 20 ± 1 °C and artificially illumi- nated for 12 h (Gro-lux tubes, type F4T12/ Additional index words. symmetrical and asymmetrical protoplast fusion, mal secco, Gro; Sylvania, Danvers, Mass.) with a light limon, Citrus improvement intensity of 100 µE. The somatic hybrid and cybrid plants were grown in a mixture of soil Abstract. The reaction of lemon hybrids obtained by symmetrical and asymmetrical (70%) and peat (30%) in 3-L plastic contain- protoplast fusion, toward “mal secco” infection caused by Phoma tracheiphila (Petri) ers. ‘Femminello’ and ‘Monachello’ , Kanc. et Ghik. was examined. Resistance was tested in ‘Valencia’ sweet orange [Citrus highly susceptible and resistant to the disease, sinensis (L.) Osbeck] and ‘Femminello’ lemon [C. limon (L.) Burm. f.] somatic hybrid and respectively, were used as controls. The cybrids, by stem and leaf inoculation tests and by analysis of propagule number of P. budwood of ‘Monachello’ and ‘Femminello’ tracheiphila in the xylem of stem-inoculated plants. In general, the somatic hybrid and the lemon was collected from 7-year-old plants. cybrids showed an intermediate degree of resistance, with slight differences in disease Stem and leaf inoculation were performed symptoms, in comparison with resistant ‘Monachello’ lemon and susceptible ‘Femminello’ as described by Luisi et al. (1978), since these lemon, used as controls. The lower mortality in the asymmetrical lemon cybrids suggests techniques were effective for evaluating host that specific mechanisms of resistance to the disease could be activated in these genotypes. resistance against the disease, and the results are closely related to the susceptibility to mal The development of improved Citrus cul- genomes, one can regenerate allotetraploid secco in mature plants in the groves (De Cicco tivars is an important breeding objective, since somatic hybrid plants, which can be used as et al., 1988; Ippolito et al., 1990). Stem inocu- many factors limit the market potential for pollen parents in backcrosses, in efforts to lation was performed on sour orange root- Italian citriculture (e.g., seediness, small fruit produce triploid with desirable horti- stock, 4 to 5 cm above the crown. The inocu- size, unfruitfulness, early maturity). For lemon, cultural traits and disease tolerance. However, lum (a drop of conidia adjusted to 106 conidia/ a high priority is the development of disease- in several somatic hybridization experiments, mL) was introduced into the transpiration resistant plants, since most commercial lemon spontaneous elimination of the nucleus of one stream of 10 replicate plants per genotype by cultivars are seriously damaged by a systemic fusion partner sometimes occurs; as a result, notching the xylem vessels with the blade of a fungal disease, called “mal secco” and caused “cybrids,” individuals that contain the nucleus knife on opposite sides of the stem. Leaf by Phoma tracheiphila. The mainly of only one partner in combination with some inoculation was performed on 6-month-old penetrates through wounds and easily colo- organellar components of the other partner, leaves by pricking the petiole with three ento- nizes the xylem, inducing withering and die- may be recovered. Direct utilization of such mological pins mounted on a cork and placing back of twigs and branches. A sudden wilt of asymmetrical somatic hybrids (cybrids) in a 10-µL drop of conidial suspension (107 whole branches or trees may also occur when commercial orchards may be possible if ad- conidia/mL) at each inoculation site (two per infections take place on the main roots or equate horticultural characteristics are main- leaf). Each treatment was comprised of five trunk. The few resistant lemon cultivars tained and disease tolerance is achieved plants of each genotype and five leaves were (‘Monachello’ and ‘Interdonato’) lack the pro- (Grosser et al., 1996; Vardi et al., 1987). The inoculated on each plant. ductivity and fruit quality of the susceptible practical value of citrus cybrids is, however, When the symptoms appeared (20 and 15 d ‘Femminello’ (Salerno and Cutuli, 1977). currently unknown since the nuclear/cytoplas- after inoculation for stem and leaf test, respec- Somatic hybridization via symmetrical and mic interactions are still obscure in Citrus. tively), the development of the disease was asymmetrical protoplast fusion could repre- The objective of this research was to evalu- assessed approximately weekly, using empiri- sent a primary strategy in obtaining improved ate tolerance toward mal secco infections in cal scales. Severity of symptoms following disease-tolerant scions and rootstocks (Gmitter the interspecific allotetraploid somatic hybrid stem inoculations was rated on a scale of 0 to et al., 1992; Grosser et al., 1990a, 1990b, plant ‘Valencia’ sweet orange + ‘Femminello’ 5: 0 = no disease symptoms; 1 = one or two 1990c). By symmetrically combining nuclear lemon (Tusa et al., 1990) and in two apical leaves showing vein chlorosis; 2 = <50% ‘Femminello’ lemon cybrids (Grosser et al., leaves with vein or extended chlorosis, wilted Received for publication 9 Mar. 1998. Accepted for 1996) that were regenerated following fusion or fallen; 3 = >50% leaves with symptoms or publication 18 May 1999. We thank Dr. L.W. Timmer of embryogenic with mesophyll protoplasts of fallen; 4 = all or almost all leaves wilted or (Univ. of Florida, IFAS, Lake Alfred) for useful the parents. Evaluation of the mal secco toler- fallen, necrosis of the plant starting from the comments and for a critical reviewing of the manu- ance of the allotetraploid hybrid ‘Valencia + top. The test was concluded when all plants of script. The cost of publishing this paper was defrayed Femminello’ is extremely important, since it the same genotype had reached a score of 4. in part by the payment of page charges. Under postal is fertile and could be used in interploid sexual Moreover, at the end of the test the number of regulations, this paper therefore must be hereby hybridization with diploid ‘Femminello’ lemon dead plants was also recorded. Disease sever- marked advertisement solely to indicate this fact. to produce improved seedless triploid lemon- ity following leaf inoculations was also rated 1To whom reprint requests should be addressed: Dr. N. Tusa, Centro di Studio per il Miglioramento types tolerant to mal secco. Understanding the on a scale of 0 to 5: 0 = no sign of infection; 1 Genetico degli Agrumi. C.N.R., Viale delle Scienze status of ‘Femminello’ lemon cybrids would = chlorotic halo around the inoculation point; c/o Facolta’ di Agraria, 90128 Palermo, Italy (phone: provide new information on the nuclear/cyto- 2 = chlorosis of the vein close to the inocula- 39 91 423398–484482–6521100; fax: 39 91 plasmic interactions involved in important hor- tion point; 3 = vein chlorosis extending as far 6521098; e-mail: [email protected]). ticultural traits. as the leaf edge; 4 = generalized chlorosis and

HORTSCIENCE, VOL. 35(1), FEBRUARY 2000 125 BREEDING, CULTIVARS, ROOTSTOCKS, & GERMPLASM RESOURCES

/or browning of the veins close to the inocula- Table1. Parameters of mal secco infections on stem and leaf of somatic hybrid and cybrids of lemon. tion point (Luisi et al., 1978). The test was ‘Femminello’ and ‘Monachello’ lemon used as control. concluded when 90% of the leaves of a geno- Genotype type scored 4. The values of the empirical Days after Diploid Valencia + Tetraploid scale were utilized to compute the McKinney Observation inoculation Femminello cybrid Femminello cybrid Monachello index (McKinney, 1923), by means of the Severity of disease 45 2.07 az 1.37 c 1.34 c 1.69 b 1.77 b formula: 90 3.42 ab 3.68 a 3.21 b 3.17 b 2.33 c Mi = [∑(d·f)/(Tn·D)] · 100 No. propagules of Phoma tracheiphila/g of wood 10.4 a 7.6 ab 3.3 b 3.8 b 1.7 b where d is the degree of disease intensity assessed on the leaf and f its frequency; Tn is McKinney index values 20 16.9 a 12.4 b 9.5 c 15.8 a 2.3 d 65 91.4 a 89.4 a 80.4 a 80.5 a 37.1 b the total number of the leaves examined z ≤ (healthy and diseased); and D the highest Mean separation within rows by Duncan’s multiple range test, P 0.05. degree of disease intensity occurring on the empirical scale. This index represents the termediate, were not significantly different by organelle DNA (Grosser et al., 1996). weighted average of the disease severity, ex- from that of the susceptible control; Considering the results obtained in the pressed as actual percentage with respect to ‘Monachello’ lemon plants showed a signifi- stem inoculation test, a slower development of the possible maximum level of disease (100%). cantly lower disease intensity than did the the disease in the initial stages seems to be a The number of propagules of the pathogen other genotypes tested. peculiar trait of the somatic hybrid ‘Valencia in the xylem of stem-inoculated lemon plants However, differences in the percentages of + Femminello’ and both the lemon cybrids, as was evaluated after the last assessment of dead plants between the cybrids and the resis- compared with the susceptible control, while, disease intensity, using the technique of tant genotype ‘Monachello’ were nonsignifi- in the leaf inoculation test, only the somatic Buchenauer and Erwin (1972) as modified by cant (χ2 = 1.44; P = 0.21). Only 10% of the hybrid ‘Valencia + Femminello’ and the dip- Lima et al. (1994) for P. tracheiphila. For each diploid and tetraploid cybrids were dead loid lemon cybrid show a slower initial symp- genotype, the stems of three plants showing a compared with 40% of the susceptible tom progression. However, compared with the disease intensity between 3 and 4 were chosen ‘Femminello’ and 20% of resistant ‘Monachello’. behavior of the resistant ‘Monachello’ lemon, in order to evaluate the rate of xylem coloniza- The somatic hybrid ‘Valencia + Femminello’, the somatic hybrid and both cybrids showed a tion at similar degrees of disease symptom at 30% mortality, was intermediate. On the higher intensity of disease symptoms at the intensity. Each plant was separately processed. surviving plants of this somatic hybrid, abun- end of the various tests, except for the propagule After aseptically removing the bark, each stem dant gummy exudates were observed near the numbers of P. tracheiphila in the xylem of was cut into smaller pieces, and the pieces leaf scars. stem-inoculated plants. In this case, a low were weighed and separately homogenized The number of propagules of P. tracheiphila number of propagules was found in the xylem for 3 min in 100 mL of cold, sterile water. The per gram of wood was significantly lower (P ≤ of the somatic hybrid ‘Valencia + Femminello’, homogenate was serially diluted (1/10, 1/100, 0.05) on the stems of the somatic hybrid the tetraploid cybrid, and the resistant control. 1/1000) and 250 µL were mixed in petri dishes ‘Valencia + Femminello’ and tetraploid cybrids An intermediate reaction was observed in the with 12 mL of potato sucrose agar at 50 °C, than on the susceptible ‘Femminello’ (Table diploid cybrid with respect to the susceptible amended with streptomycin sulfate (250 1). The value for the diploid cybrid, while and resistant controls. This result may indicate mg·L–1). For each stem and dilution, five petri numerically different, was not statistically dif- that this character is connected to the observed plates were utilized. The number of P. ferent from that for the susceptible control. slower disease development. In ‘Monachello’ tracheiphila colonies were counted after 4 to 6 However, ‘Monachello’ lemon showed the lemon the resistance to mal secco has been d of incubation at 21 ± 1 °C. lowest value. characterized and its expression is related to Stem and leaf inoculation tests were ar- Twenty days after inoculation, McKinney gum reactions, which delay the progress of the ranged in a randomized complete-block de- index values for leaf-inoculated plants of the pathogen in vascular tissue and allow the plants sign. For stem inoculations, five replicates, allotetraploid somatic hybrid ‘Valencia + to compartmentalize the infection (Perrotta et each comprising two plants per genotype, were Femminello’ and the diploid ‘Femminello’ al., 1980; Somma et al., 1979). A similar used; for leaf inoculation, five replicates, each lemon cybrid were significantly lower than mechanism could be hypothesized for the so- comprising one plant with five inoculated that for the susceptible control (Table 1). The matic hybrid ‘Valencia + Femminello’, since leaves, were used. The data were subjected to resistant ‘Monachello’ genotype had the low- abundant gummy exudates were observed near analysis of variance and the mean values com- est value. However, at the end of the test (65 the leaf scars of the surviving plants. Our data pared using Duncan’s multiple range test d after inoculation) differences among the on propagules of the pathogen in the xylem are (Snedecor and Cochran, 1980). The percent- somatic hybrid, both cybrids, and the in agreement with others (Magnano di San Lio age data (i.e., McKinney index) showed high ‘Femminello’ susceptible control were non- et al., 1992), indicating that extensive vascular heterogeneity of variance when checked by significant. colonization is a prerequisite for symptom Bartlett’s test; therefore, before analysis, the expression in mal secco–infected sour orange data were converted into Bliss angular values Discussion seedlings. In fact, the difference in disease (arcsin √%), to reduce heterogeneity. The data intensity observed on ‘Femminello’ and were then retested and subjected to chi-square In the present work, the resistance to P. ‘Monachello’ seems to be positively related to normality (Gomez and Gomez, 1984; Little tracheiphila infections was tested in three the propagule numbers of the pathogen in the and Hills, 1972). different genotypes, the amphidiploid somatic xylem. The low percentages of dead plants hybrid ‘Valencia + Femminello’ and two recorded for diploid and tetraploid lemon Results ‘Femminello’ lemon cybrids, obtained by sym- cybrids could confirm this hypothesis. More- Forty-five days after stem inoculation, metrical and asymmetrical protoplast fusion, over, the lower percentage of dead plants in the allotetraploid somatic ‘Valencia + respectively. The value of fertile somatic hy- both ‘Femminello’ lemon cybrids than in Femminello’ and the diploid and tetraploid brids in Citrus scion improvement is known, ‘Monachello’ lemon suggests that mechanisms ‘Femminello’ cybrids showed a significantly since they can be used in interploid crosses of resistance to P. tracheiphila invasion into (P ≤ 0.05) lower intensity of mal secco disease with diploid cultivars to obtain triploid seed- the xylem could be activated as an effect of than did the susceptible ‘Femminello’ control less plants for selection (Grosser and Gmitter, nuclear/cytoplasmic DNA interactions; in (Table1). Three months after inoculation, rat- 1990a, 1990b; Soost and Cameron, 1975). these genotypes, the mitochondrial genome is ings for the allotetraploid somatic hybrid Yet, the practical value of citrus cybrids is derived from the resistant ‘Valencia’ sweet ‘Valencia + Femminello’ and the tetraploid currently unknown, because no horticultur- orange and the nuclear genome from the ‘Femminello’ cybrid, while numerically in- ally important traits are known to be encoded susceptible ‘Femminello’ lemon.

126 HORTSCIENCE, VOL. 35(1), FEBRUARY 2000 The allotetraploid somatic hybrid ‘Valencia acters of the xylem, gummy reactions, etc.) il mal secco degli agrumi mediante isolati + Femminello’ was developed in an effort to deserves further investigation. ipovirulenti del patogeno. La Difesa delle piante combine the good quality and performance of 17:135–144. Little, T.M. and F.J. Hills. 1972. Statistical meth- ‘Femminello’ lemon with the cold hardiness Literature Cited and mal secco tolerance of ‘Valencia’ (Tusa et ods in agricultural research. Univ. of California, Buchenauer, H. and D.C. Erwin. 1972. Control of Davis. al., 1990). In our experiment, disease devel- Luisi, N., V. De Cicco, G. Cutuli, and M. Salerno. oped more slowly in ‘Valencia + Femminello’ Verticillium wilt of cotton by spraying with acidic solutions of benomyl, methyl 2-benzimi- 1978. Factors in early testing for citrus mal in the initial stages; moreover, the lower rate of dazole carbamate, and thiabendazole. secco resistance. Proc. Intl. Soc. Citricult. 1:197– propagules in the xylem and the lower percent- Phytopathol. Z. 75:124–129. 200. age of dead plants as compared with the sus- De Cicco, V., A. Ippolito, and M. Salerno. 1988. Magnano di San Lio, G., S.O. Cacciola, A. Pane, ceptible ‘Femminello’ would indicate an in- Response to mal secco infections of two puta- and S. Grasso. 1992. Relationship between xy- termediate degree of resistance to mal secco tive resistant lemon lines under controlled envi- lem colonization and symptom expression in infection. The performance is not surprising, ronment. p. 99–103. In: Proc. 2nd Intl. Mtg. mal secco infected sour orange seedlings. Proc. since this genotype is an amphidiploid somatic Mediterranean Tree Crops. Chania, Crete, Intl. Soc. Citricult. 2:873–876. Greece, 2–4 Nov. 1988. McKinney, H.H. 1923. Influence of soil tempera- hybrid obtained from symmetrical protoplast ture and moisture on infection of wheat seed- fusion, in which the complete nuclear ge- Gmitter, F.G., Jr., J.W. Grosser, and G.A. Moore. 1992. Citrus, p. 335–370. In: R.E. Litz and F. lings by Helminthosporium sativum. J. Agr. nomes of both parents were combined. The Hammerschlag (eds.). Biotechnology of peren- Res. 26:195–218. additive process offers the advantage of trans- nial fruit crops. CAB Intl., Oxon, U.K. Perrotta, G., G. Magnano di San Lio, and M. Bassi. ferring important traits for which expression is Gomez, K.A. and A.A. Gomez. 1984. Statistical 1980. Some anatomical and morpho-functional controlled by complex gene systems; as a procedures for agricultural research. 2nd ed. aspect of resistance to Phoma tracheiphila in result, somatic hybrids may retain many of the Wiley, New York. citrus plants. Phytopathol. Z. 98:346– 358. attributes of both parents (Grosser et al., 1995). Grosser, J.W. and F.G. Gmitter, Jr. 1990a. Proto- Salerno, M. and G. Cutuli. 1977. Control of citrus The behavior to P. tracheiphila infections plast fusion and citrus improvement. Plant Breed. mal secco in Italy today. Proc. Intl. Soc. Citricult. Rev. 8:339–374. 3:1001–1003. seems to be confirmed by preliminary field Snedecor, G.W. and W.G. Cochran, 1980. Statisti- observations; the somatic hybrid ‘Valencia + Grosser, J.W. and F.G. Gmitter, Jr. 1990b. Wide hybridization of citrus via protoplast fusion: cal methods. The Iowa State Univ. Press, Ames. Femminello’ and both of the lemon cybrids Progress, strategies and limitations, p. 31–41. Somma, V., G. Cutuli, and M. Salerno. 1979. have been in experimental field trials for lemon In: A.B. Bennet and S.D. O’Neil (eds.). Horti- Ricerche sul saggio della resistenza al mal secco since 1990 to evaluate their horticultural per- cultural biotechnology, plant biology. vol. 11. di giovani limoni innestati, p. 71–80. In: Proc. formance and their resistance or tolerance to Wiley-Liss, New York. secondo seminario di studio sul miglioramento natural infection of mal secco. To date, all Grosser, J.W. and F.G. Gmitter, Jr. 1990c. Somatic genetico del limone. Giovinazzo, Bari, Italy, 5– three exhibit adequate horticultural character- hybridization of citrus with wild relatives for 6 Apr. 1979. istics, being vigorous and having a higher germplasm enhancement and develop- Soost, R.K. and J.W. Cameron. 1975. Citrus, p. 507–540. In: J. Janick and J.N. Moore (eds.). degree of tolerance to mal secco infection than ment. HortScience 25:147–151. Advanced fruit breeding. Purdue Univ. Press, the original ‘Femminello’ lemon. Complete Grosser, J.W., F.G. Gmitter, Jr., W.S. Castle, and J.L. Chandler. 1995. Production and evaluation West Lafayette, Ind. evaluation of these genotypes is, however, of citrus somatic hybrid rootstocks: Progress Tusa, N., S. Fatta Del Bosco, L. Nardi, and S. still in progress and is a very long-term and report. Proc. Fla. Hort. Soc. 108:140–143. Lucretti. 1996. Obtaining triploid plants by cross- complex effort. Moreover, ‘Valencia + Grosser, J.W., F.G. Gmitter, Jr., N. Tusa, G. ing citrus lemon cv. ‘Femminello’ 2n x 4n al- Femminello’ has confirmed its great potential Reforgiato Recupero, and P. Cucinotta. 1996. lotetraploid somatic hybrids. Proc. Intl. Soc. in citrus cultivar development; in fact, it has Further evidence of a cybridization requirement Citricult. 1:133–136. been used as a male parent in interploid back- for plant regeneration from citrus leaf proto- Tusa, N., J.W. Grosser, and F.G. Gmitter, Jr. 1990. crosses with diploid ‘Femminello’ lemon, and plasts following somatic fusion. Plant Cell Rpt. Plant regeneration of ‘Valencia’ sweet orange, a large triploid progeny has been rescued (Tusa 15:672–676. ‘Femminello’ lemon, and the interspecific so- et al., 1996). Ippolito, A., V. De Cicco, and E. Traversa. 1990. matic hybrid following protoplast fusion. J. Behaviour of some grafted lemon rootstocks Amer. Soc. Hort. Sci. 115:1043–1046. This is the first report in which somatic towards mal secco infections in controlled Vardi, A., A. Breiman, and E. Galun. 1987. Citrus hybrids and cybrids of lemon have been tested enviroment. XXIII Intl. Hort. Congr., Florence, cybrids: Production by donor-recipient proto- for their resistance to mal secco infections. Abstr. p. 512. plast fusion and verification by mitochondrial However, the characterization of the observed Lima, G., F. Nigro, A. Santomauro, and A. Ippolito. restriction profiles. Theor. Appl. Genet. 75:51– resistance (i.e., anatomical and functional char- 1994. Ulteriori tentativi di lotta biologica contro 58.

HORTSCIENCE, VOL. 35(1), FEBRUARY 2000 127