Apomixis in different ploidy levels of cassava NAGIB M. A. NASSAR , ADALGISA CHAIB and AHMED Y. ELSAYED

Universidade de Brasilia, Brasilia, Brazil

Nassar, N. M. A., Chaib, A. and Elsayed, A. Y. 2011. Apomixis in different ploidy levels of cassava. – Hereditas 148: 125–128. Lund, Sweden. eISSN 1601-5223. Received 20 September 2011. Accepted 28 October 2011.

Two polyploid hybrids between cassava ( Manihot esculenta ) cultivar 307-2 and its wild relatives M. glaziovii and M. anomala , were studied to examine the relationship between ploidy level and the production of seeds without fertilization. A clearing method was applied to assess ovule sizes as an indication of multiembryony. The diploid cultivar 307-2 had regular 18 bivalents at meiotic metaphase 1 while the polyploid types showed chromosome confi gurations varying from 3 to 4 quadrivalents and 28 to 30 bivalents. A total of 14% of studied ovules of the polyploid hybrid involving M. glaziovii were multiebryonic, while the percentage of multi-embryony was as low as 2% in the polyploid hybrid M. anomala � M. esculenta . Diploid hybrid types did not show any multi-embryony. Adventitious embryos were found and documented for the fi rst time in polyploid hybrids M . esculenta � M. glaziovii . The association of multiple embryo formation with ovary size and pollination showed that apomictic embryos form independently from fertilization. Simple iodized carmine stain for measuring pollen viability proved as effi cient as the sophisticated Alexander method.

Nagib M. A. Nassar, Universidade de Brasilia, Brasilia, Brazil. E-mail: nagnassa@rudah.com.br

In apomixis seeds are produced without fertilization so that the embryo develops from an unreduced somatic cell. It fi xes heterozygosity and may perpetuate superior types (K OLTUNOW et al. 1995).

Polyploidy is often associated with apomixis ( ASKER and JERLING 1992; NASSAR 2006). Some wild Manihot species such as M. glaziovii and M. neuana reproduce through apo-mixis ( NASSAR 2001; NASSAR et al. 2007). We have tried to transfer this character further and to overcome any expected hybrid sterility through polyploidization. NASSAR (2006) and NASSAR et al. (2007, 2009), working with polyploidized between species hybrids of Manihot , recommend this as a possible solution to problems in cassava breeding.

Here we study the relationship between ploidy level and apomixis and whether pollination is necessary for triggering apomictic embryogenesis as seen in by poly-embryo formation.

MATERIAL AND METHODS

We use the diploid cultivar 307-2, selected for produc-ti vity from the progeny of the hybrid M. esculenta � M. glaziovii. This hybrid does not express apomixis while the M. glaziovii parent is apomictic. We attempt to see if any of the second generation offspring may segregate for apomixis, as seen through multiembryo formation ( OZIAS-AKINS 2006). In addition to this variety, we used two between species hybrids: M. esculenta � M. glaziovii and M. esulenta � M. anomala . These hybrids were poly-ploidized artifi cially so that colchicine was applied to lat-eral buds. The resulting polyploid shoots were multiplied vegetatively.

Multiembryo formation in ovules, the nucellus and teguments was taken as evidence for apomixis ( NASSAR et al. 2007, 2008). To study possible role of pollination and the stage of fl ower development in forming poly-embryonic ovules, fl owers of different sizes with ovaries measuring 3, 4, 5, 6, 7 and 8 mm in diameter were studied. The fl owers were protected with PVC plastic.

To study embryos the fl owers were fi xed for 24 h in Carnoy ’ s fl uid (alcohol and glacial acetic acid 3:1). They were thereafter examined using the clearing technique ( HERR 1982; NASSAR et al. 1998). Images were captured using software HonestechTVR 2.5. This technique allows the visualization of the embryo sac in toto and any abnor-malities in embryo development.

To study eventual meiotic behaviour, 15 metaphase cells from every plant were analyzed using the smear method. Images were digitized as mentioned above.

Hereditas 148: 125–128 (2011)

© 2011 The Authors. This is an Open Access article. DOI: 10.1111/j.1601-5223.2011.02243.x

Table 1. Apomixis in different Manihot type s .

Cultivar TreatmentTotal

ovulesTotal

apomicticTotal

atrophied% of

apomixis

307-2 FAA FP FPA

303 52

241

44 8

21

20 0 0

14% 15% 8.7%

M. esculenta � M. glaziovii

FAA FP FPA

206 43

236

5 11 26

21 −

16

2.4% 25% 11%

M. esculenta � M. anomala

FAA FP FPA

236 91

207

9 2

10

73 11 15

3.8% 2.1% 4.8%

FAA � fl ower before anthesis; FP � fl ower protected; FPA � fl ower after anthesis .

126 N. M. A. Nassar et al. Hereditas 148 (2011)

here showed chromosome confi gurations involving 3 through 4 quadrivalents and 28 through 30 bivalents (Fig. 1).

The clarifi ed ovules of cultivar 307-2 showed embryo development between the inner integument and the outer one, (Fig. 1-A). Diploid hybrid types do not show any multiembryony. Multiembryos were found in ovules mea-suring 3, 4 and 5 mm. Figure 2 shows that ovules measur-ing 8 mm across had both adventitious and multiembryos

Fig. 1. ( A ) Tegumental embryos of fl ower before anthesis. Mag. � 50. ( B ) Ovule with two embryo sacs. Mag. � 50. ( C ) Normal ovule. ( D ) Metaphase I of polyploid hybrid M. esculenta � M. glaziovii. Mag. � 400. ( E, F ) Pollen viability assessed using carmin stain and the Alexander method, respecively. Mag. � 100.

To assess pollen viability, two methods were applied: iodized carmine ( NASSAR et al. 2000 and Alexander’s stain ( ALEXANDER 1969)).

RESULTS AND DISCUSSION

Table 1 and Fig. 1 and 2 give the results in a concise form. The diploid cultivar 307-2 has a regular set of 18 bivalents at meiotic metaphase I, while the polyploid types studied

Hereditas 148 (2011) Apomixis in different ploidy levels of cassava 127

Fig. 2. Effect of ovule size on multiembryo formation in different manihot types: cultivar 307-2, hybrid M. esculenta � M. glaziovii and M. esculenta � M. anomala . FAA � fl ower before anthesis; FP � fl ower protected; FPA � fl ower after anthesis.

(Fig. 2). The association of multiple embryo formation with ovary size and pollination showed that apomictic embryos form independently from fertilization.

Adventitious embryos are here found and documented for the fi rst time in polyploid hybrids M . esculenta � M. glaziovii . Accordingly, in addition to apomictic apo-spory reported by NASSAR et al. (2000) cassava may have adventitious embryos. It shows that cassava has a plastic-ity of development involving different types of apomixis.

Multiembryos were found to form before anthesis. Their formation does not depend on fertilization. This is seen in cultivar 307-2 where the highest percentage of apomixis; 15% was found in protected fl owers. In fl owers examined before anthesis we observed 14% multi-embryos. The lower rate of apomixis detected was in the post-anthesis fl owers 8.7%. The higher level of apomixis in protected fl owers may be due to that unprotected ones were frequently attacked by insects and aborted.

A total of 14% of studied ovules of the polyploid hybrid involving M. glaziovii were multiebryonic, while the per-centage of multiembryony was as low as 2% in the poly-ploid hybrid M. anomala � M. esculenta .

The polyploidized M. esculenta � M. glaziovii had 25% multiple embryos in the examined 300 ovules while the diploid type showed 0.0%. Apparently apomixis is controlled by a model of additive gene action.

In contrast to studies suggesting monogenic inheritance ( LEBLANC et al. 1995), this fi nding may indicate polygenic interaction and additive action of genes for apomixis in Manihot . In some species of Hieracicum, apomixis is reported to be controlled by dominant one dominant allele ( BICKNELL et al. 2000), while in Paspalum more than one pair of genes is required for apomixis to manifest itself ( ASKER and JERLING 1992). Clearly, the kind of apomixis in Manihot is different from that in Hieracicum.

Polyploid plants of the hybrid M. anomala � M. esculenta showed a high proportion of hermaphrodite fl owers (80%). Pollen of the these fl owers showed

viability equal with the staminate fl owers. It is likely that hybridization followed by polyploidization seems to act in certain genotypes to canalize expression fl owering genes in this way ( SOLTIS and SOLTIS 2009).

The diploid cultivar 307-2 had a regular set of 18 biva-lents at meiotic metaphase I, with high pollen viability. The polyploid types studied here showed chromosome confi gurations involving 3 through 4 quadrivalents and 28 through 30 bivalents (Fig. 1) but they had a high propor-tion of viable pollen as well.

Our data indicate that the genetic nature of apomixis in Manihot could be recessive since it was absent in the F 1 and showed a segregation in F 2 . It can also be a poly-genic trait since it was manifested in polyploid type while it was absent in the diploid one. This is in agreement with the fi ndings reported earlier ( NASSAR et al. 2007, 2009, 2010).

Measuring pollen viability by both iodized carmine and the Alexander method showed the same result of 71 – 73% viability without any signifi cant difference. Accordingly, we recommended the use of carmin in such experiments because it is very much simpler to prepare and apply in comparison with the sophisticated Alexander method.

In conclusion we may state that there is no association between polyploidy and apomixis in hybrids of cassava with M. glaziovii or with M. anomala . It is apparent that the genetic behaviour of apomixis in Manihot does not follow the model found in some genera like Paspalum ( ASKER and JERLING 1992). Apomixis in Manihot is differ-ent in nature from other types since it combines both adventitious embryos and apospory. The formation of multiembryos in apomictic cassava ovule does not depend on pollination and fertilization of ovule. The use of simple iodinated carmine stain was as effi ecient as the sophisti-cated Alexander stain.

Acknowledgements – This work has been supported by the National council for Scientifi c Development (CNPq). We

128 N. M. A. Nassar et al. Hereditas 148 (2011)

Nassar, N. M. A. 2006. Chromosome doubling induces apomixis in a cassava � Manihot anomala hybrid. – Hereditas 143: 1 – 3.

Nassar, N. M. A., Vieira, M. A., Vieira, C. et al. 1998. Molecular and embryonic evidence of apomixis in cassava ( Manihot esculenta Crantz). – Euphytica 102: 9 – 13.

Nassar, N. M. A., Santos, E. and David, S. R. O. 2000. The transference of apomixis genes from Manihot Neusana Nassar to cassava, M. esculenta Crantz. – Hereditas 132: 167 – 170.

Nassar, N. M. A., Kalkmann, D. and Collevatti, R. G. 2007. A further study of microsatellite on apomixis in cassava. – Hereditas 144: 1 – 4.

Nassar, N. M. A., Hashimoto, D. Y. C. and Fernandes, S. D. C. 2008. Wild Manihot species: botanical aspects, geographic distribution and economic value. – Genet. Mol. Res. 7: 16 – 28.

Nassar, N. M. A., Gomes, P. T. C., Chaib, A. M. et al. 2009. Cytogenetic and molecular analysis of an apomictic cassava hybrid and its progeny. – Genet. Mol. Res. 8: 1323 – 1330.

Nassar, N. M. A., Graciano-Ribeiro, D., Gomes, P. F. et al. 2010. Alternation of reproduction system in a polyploidized cassava interspecifi c hybrid. – Hereditas 147: 58 – 61.

Ozias-Akins, P. 2006. Apomixis: developmental characteristics and genetics. – Critical Rev. Plant Sci. 25: 199 – 214.

Soltis, P. S. and Soltis, D. E. 2009. The role of hybridization in plant speciation. – Annu. Rev. Plant Biol. 60: 561 – 588.

express our gratitude towards the Canadian international Devel-opment Research center (IDRC) that helped to establish the living collection of Manihot .

REFERENCES

Alexander, M. P. 1969. Differential staining of aborted and nonaborted pollen. – Stain Technol. 44: 117 – 122.

Asker, S. E. and Jerling, L. 1992. Apomixis in plants. – CRC Press. Bicknell, R. A., Borst, N. K. and Koltunow, A. M. 2000. Monogenic

inheritance of apomixis in two Hieracium species with distinct developmental mechanisms. – Heredity 84: 228 – 237.

Herr, J. M. 1982. An analysis of methods for permanently mounting ovules cleared in four-and-a-half type clearing fl uids. – Stain Technol. 57 : 161 – 169.

Koltunow, A. M., Bicknell, R. A. and Chaudhury, A. M. 1995. Apomixis: molecular strategies for the generation of genetically identical seeds without fertilization. – Plant Physiol. 108: 1345 – 1352.

Leblanc, O., Grimanelli, D., Gonzalez-de-Leo, D. et al. 1995. Detection of the apomictic mode of reproduction in maize- Tripsacum hybrids using maize RFLP markers. – Theor. Appl. Genet. 90: 1198 – 1203.

Nassar, N. M. A. 2001. The nature of apomixis in cassava ( Manihot esculenta Crantz). – Hereditas 134: 185 – 187.