Periclinal chimera technique: new plant breeding approach

Abstract

Plant interspecific periclinal chimeras are a mosaic formed by tissues from two species. They are manipulated here as an efficient plant breeding tool for cassava root yields. In this study, plants synthesized from two chimeras, designated as chimera 2 and chimera 4, were characterized morphologically and cytologically to unravel the origin of their tissue layers (L2 and L3). Root yield of the two chimeras was also evaluated. Chimera 2 that was developed from graft union between Manihot fortalezensis (F) as scion and M. esculenta (E) as rootstock and the same in chimera 4 was developed from grafting triploid cassava cultivar (2n = 54) (C) as scion and M. pohlii (P) (2n = 36) as rootstock. A new method of inducing interspecific chimeras without using hormones was also tested in this study. Five combinations between four cassava cultivars on one side and M. fortalezensis and an interspecific hybrid (M. glaziovii x M. esculenta) on the other side were experimented to determine compatibility between the parents. Wild species always gave L2 and L3, independent of being used as rootstock or scion. L3 is responsible for producing pericycle. Thus, its performance was different in each chimera due to specific epigenetic interaction. Of 48 grafts, it was obtained one chimera giving a percentage of 2.1% that is little lower than using hormones but much efficient to use. Chimera induction efficiency in this investigation was the same when using hormones. Thus, our new, less labor, and more cost-effective technique is as much efficient as hormones and is much potential to employ as an effective plant breeding method boosting cassava root yield.

Introduction

Cassava is the principal food for poor people in the tropics and subtropics. In 2014, it was grown on an area of 23 million hectares producing 268 million ton and feeding one billion individuals all over the world (Fao Ifad, 2014).

Cassava (Manihot esculenta Crantz) is a perennial dicotyledonous shrub that belongs to the family Euphorbiaceae (Rogers and Appan, 1973). There are 98 species of Manihot, all having 2n = 36 chromosomes.

The wide genetic diversity in the Manihot species has been utilized by introgressing important traits such as resistance to diseases (cassava mosaic virus disease and bacterial blight), high protein content, apomixis to cassava cultivars through interspecific hybridization (Nassar, 2007; Nassar et al., 2008; Nassar and Ortiz, 2009; Akinbo et al., 2015). However, there are still limitations for breeding cassava to produce high yielding edible roots. The world average yield is only 11.2 ton/ha (Fao Ifad, 2014) compared to the potential of 90 ton/ha (Akinnagbe, 2010; Akinbo et al., 2015).

Then, we need to look for new approaches to the improvement of this crop productivity. Grafting scion of the vigorous interspecific hybrid of the perennial wild tree Manihot glaziovii, onto cassava rootstock has improved root production by 30 to 100% (De Bruijn and Dharmaputra, 1974; Nassar and Ortiz, 2010). However, the instability of this increase due to the necessity of realizing grafting every year was an obstacle of adopting this technique.

Preclinical chimeras were noted by Nassar et al. (2012) to emerge from the graft junctions of such combinations, and it appeared to be a more promising method of developing highly productive cassava varieties and perpetuating this high productivity.

A chimera is a meristem with different genetic tissues in one or more of its layers. The components of plant grafts in a chimera do not lose their integrity but coexist harmoniously (Buder, 1911; Goffreda et al., 1990; Hirata et al., 2000).

Contrary to sectorial chimeras, periclinal ones are relatively stable and can be vegetatively propagated as new varieties. Based on this, the trials were started.

Two interspecific chimeras of cassava have recently been produced from grafts of the wild triploid cassava species, M. fortalezensis, on rootstocks of two cassava cultivars at Universidade de Brasília (UnB), Brazil. By using hormones, they gave 3- to 7-fold increase in tuber production relative to donor plants (Nassar and Bomfim, 2013; Bomfim and Nassar, 2014).

The inclusion of other wild species and cassava cultivars in grafting may provide new opportunity to improve productivity and other important traits of the crop. Inducing chimeras without the use of hormones may also serve as a more cost-effective method. This study was,therefore, designed to test the efficiency of developing cassava chimeras without the use of hormones in graft combinations involving various wild Manihot species and cassava cultivars under production and to study the histogenic differentiation of the various tissue layers of these chimeras and to assess their productivity.

Materials and Methods

The experiment was conducted at the experimental station of the UnB, and at Cáceres, Mato Grosso, Brazil, in Latossoil. Two wild Manihot species (M. pohlii and M. fortalezensis) and one interspecific hybrid (M. glaziovii x M. esculenta) and five M. esculenta cultivars (UnB 201, UnB 205, UnB 031, UnB 530 p, and UnB 530-19) were used in this study.

Synthesizing periclinal chimera

Scions and rootstocks with similar size and vigor were whips grafted in August 2016. Different graft combinations were made using the varieties as mentioned above (UnB 201, UnB 203, UnB 205, and UnB 530 p), one wild species M. fortalezensis, and an interspecific hybrid (M. glaziovii x M. esculenta). The scion was cut using a knife in a slanted position closer to a bud, and the rootstock was cut in the opposite direction. The scion and the rootstock were placed in close contact taking into consideration the juxtaposition of the scion and the rootstock buds. The buds were placed in close contact for interaction. A cello tape was used to fasten and hold them together (Figure 1). All auxiliary shoots and adventitious shoots arising from anywhere except near the graft union were removed as they appeared and this was done to prevent competition with chimera for water and nutrient. Shoot induction and chimera induction rates of the various graft combinations were determined, and compatible combinations of parents were identified.

Plant chimeras were identified based on alterations in stems and leaf morphology. Stem and leaf size and shape were used to differentiate between chimeras and their donor parents using previously established methods (Carlson and Chaleff, 1974; Hirata et al., 2000; Nielsen et al., 2003; Hashimoto-Freitas and Nassar, 2013).

Since the parental species used to synthesize chimeras were of different chromosome number, counting of the chromosomes was done at mitosis and meiosis to identify the origin of every layer.

Two-year-old plants vegetatively reproduced from cuttings of two chimeras designated as chimera 2 and chimera 4, induced in August 2013 and using hormones (Figure 1), were characterized morphologically and cytologically in August 2016 to determine the distinctiveness of the chimeras and to unravel their tissue origin and their productivity. The two chimeras already produced from 2014 grafting, namely chimera 2 and chimera 4, were analyzed morphologically and cytologically by counting chromosome numbers from root tips and anther meiosis. Chimera 2 was developed from graft union between M. fortalezensis (2n = 54) as the scion on M. esculenta UnB 031 (2n = 36) as the rootstock, and chimera 4 was developed from grafting M. pohlii (2n = 36) on M. esculenta UnB 530-19 (2n = 54).

Morphological characters of the chimeras were examined for growth habit and stem, leaf, inflorescence, fruit, and root characteristics and compared to those of their donor parents to determine the distinctiveness of the chimeras. Roots of two plants of each chimera were also weighed to determine their good root productivity as compared to the cassava parent.

Meiotic chromosome counts of anthers allow L2 characterization because gametes are usually derived from the L2 layer (Satina et al., 1940; Goffreda et al., 1990), while mitotic chromosome counts on adventitious root tips allow the determination of the L3, since roots originate from the pericycle, which is derived from L3 layer (Medina et al., 2007; Nassar and Bonfim, 2013). Therefore, meiotic chromosome counting of L2-derived anthers and mitotic chromosome counting of L3-derived adventitious root tip cells were used for cytological characterization because of the different chromosome number of the donor plants of both chimera shoots as given above. Twenty plates (cells) were observed for each of the chimeras, and chromosome configurations in metaphase were also observed to judge meiotic regularity.

Tetrads were examined too to judge regularity of meiosis. A total 10 male floral buds with a size between 2-3 mm were observed for this end. The presence of four tetrads of microspores without any micronuclei was considered normal tetrad and the presence of micronuclei with the tetrads, or formation of dyads and triads with or without micronuclei was considered a sign of irregular meiosis.

Results and Discussion

The experiment was conducted at the experimental station of the UnB, and at Cáceres, Mato Grosso, Brazil, in Latossoil. Two wild Manihot species (M. pohlii and M. fortalezensis) and one interspecific hybrid (M. glaziovii x M. esculenta) and five M. esculenta cultivars (UnB 201, UnB 205, UnB 031, UnB 530 p, and UnB 530-19) were used in this study.

Synthesizing periclinal chimera

Scions and rootstocks with similar size and vigor were whips grafted in August 2016. Different graft combinations were made using the varieties as mentioned above (UnB 201, UnB 203, UnB 205, and UnB 530 p), one wild species M. fortalezensis, and an interspecific hybrid (M. glaziovii x M. esculenta). The scion was cut using a knife in a slanted position closer to a bud, and the rootstock was cut in the opposite direction. The scion and the rootstock were placed in close contact taking into consideration the juxtaposition of the scion and the rootstock buds. The buds were placed in close contact for interaction. A cello tape was used to fasten and hold them together (Figure 1). All auxiliary shoots and adventitious shoots arising from anywhere except near the graft union were removed as they appeared and this was done to prevent competition with chimera for water and nutrient. Shoot induction and chimera induction rates of the various graft combinations were determined, and compatible combinations of parents were identified.

Plant chimeras were identified based on alterations in stems and leaf morphology. Stem and leaf size and shape were used to differentiate between chimeras and their donor parents using previously established methods (Carlson and Chaleff, 1974; Hirata et al., 2000; Nielsen et al., 2003; Hashimoto-Freitas and Nassar, 2013).

Since the parental species used to synthesize chimeras were of different chromosome number, counting of the chromosomes was done at mitosis and meiosis to identify the origin of every layer.

Two-year-old plants vegetatively reproduced from cuttings of two chimeras designated as chimera 2 and chimera 4, induced in August 2013 and using hormones (Figure 1), were characterized morphologically and cytologically in August 2016 to determine the distinctiveness of the chimeras and to unravel their tissue origin and their productivity. The two chimeras already produced from 2014 grafting, namely chimera 2 and chimera 4, were analyzed morphologically and cytologically by counting chromosome numbers from root tips and anther meiosis. Chimera 2 was developed from graft union between M. fortalezensis (2n = 54) as the scion on M. esculenta UnB 031 (2n = 36) as the rootstock, and chimera 4 was developed from grafting M. pohlii (2n = 36) on M. esculenta UnB 530-19 (2n = 54).

Morphological characters of the chimeras were examined for growth habit and stem, leaf, inflorescence, fruit, and root characteristics and compared to those of their donor parents to determine the distinctiveness of the chimeras. Roots of two plants of each chimera were also weighed to determine their good root productivity as compared to the cassava parent.

Meiotic chromosome counts of anthers allow L2 characterization because gametes are usually derived from the L2 layer (Satina et al., 1940; Goffreda et al., 1990), while mitotic chromosome counts on adventitious root tips allow the determination of the L3, since roots originate from the pericycle, which is derived from L3 layer (Medina et al., 2007; Nassar and Bonfim, 2013). Therefore, meiotic chromosome counting of L2-derived anthers and mitotic chromosome counting of L3-derived adventitious root tip cells were used for cytological characterization because of the different chromosome number of the donor plants of both chimera shoots as given above. Twenty plates (cells) were observed for each of the chimeras, and chromosome configurations in metaphase were also observed to judge meiotic regularity.

Tetrads were examined too to judge regularity of meiosis. A total 10 male floral buds with a size between 2-3 mm were observed for this end. The presence of four tetrads of microspores without any micronuclei was considered normal tetrad and the presence of micronuclei with the tetrads, or formation of dyads and triads with or without micronuclei was considered a sign of irregular meiosis.

Acknowledgments

P.M. Gakpetor is the recipient of a scholarship from Fundação Nagib Nassar para Desenvolvimento Científico e Sustentável (FUNAGIB) to whom she extends her gratefulness. Thanks are due to Nayra Bomfim, Welton Reis, and Helena Schuch for their help with laboratory preparations. The above wild cassava living collection was established at Universidade de Brasília with the help of the Canadian International Development Research (IDRC). N.M.A. Nassar is the recipient of a scholarship from Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq) to whom he is thankful.