Studies On Pollen Morphology And Molecular Systematics Of The Genus Berteroella And Its Related Genera

Brassicaceae is always recognized a rather homogeneous group based on the charactersof six stamens in a teradynamous pattern (two short and four long), cruciform corolla, distinctcapsular fruit,and pungency. Although a lot work has been doing about the classification ofinfrafamily, none of systerms is agreed by almost researchers. The main problems lie on tribalclassification, delimation of genera and placement of some species.Pollen morphology of 112 species belonging to 59 genera of 10 tribes were studied byusing scanning electron microscope (SEM)and light microscope(LM). The results indicatethat pollen morphology of the family is very similar. Their shape varies from spheroidal,prolate to perprolate with reticulate ornamentation. The pollen grains are most tricolpate (rarelyteracolpate), 6-colpate or weakly aperturate in a few species. There are exine and endexine in wallcomposition. The endexine is thicker than the exine, or the thickness of them is the same, or the boundrybetween them is unclear. Nine pollen types can be distinguished based on the shape of pollengrans, the size of the lumina and the number of the aperture: (1) Pollen grains 3-colpate, perprolate,with coarsely reticulate ornamentation;(2) Pollen grains 3-colpate, perprolate, with reticulateornamentation;(3) Pollen grains 3-colpate, perprolate, with microreticulate ornamentation;(4) Pollengrains 3-colpate, prolate, with reticulate ornamentation;(5) Pollen grains 3-colpate, prolate, with reticulateornamentation;(6) Pollen grains 3-colpate, prolate, with microreticulate ornamentation;(7) Pollen grains3-colpate,spherodial, with reticulate ornamentation;(8) Pollen grains 3-colpate, spherodial, withmicroreticulate ornamentation;(9) Pollen grains 6-colpate, prolate, with microreticulate ornamentation.About the species studied in the thesis, there is more than one type of pollen grains in the same tribe. Forexample, six types in Sisymbrieae, five types in Arabideae. And different types pollen grains are obserbedwithin the same genus. Such as in Cardamine and Lepidium. There is not distinct rule about the pollengrains morphology on the level of tribe and genus. But the characters of pollen grains were found useful todistinguish among species in the same genus. In addition, the 6-colpate pollen grains of Cruciferae isreported for the first time in the present paper. It indicates the genus Berteroella with the 6-colpate pollengrains is rather unique in Cruciferae. Therefore the present authors raise the genus Berteroella to the rankof tribe, and treat it as a new tribe, Berteroelleae F. Z. Li, G. Y. Tang & Z. Y. Sun, trib. nov. on the basis ofthe unique 6-colpate pollen grains in Cruciferae.We also studied molecular phylogenetics of Berteroella and the allied genera based onnuclear ribosomal ITS DNA sequence analysis. Using parsimony, we reconstructed thephylogeny of the gene. Our phylogenetic results indicate that 34 ingroup accessions of 8 tribesfall into 9 clades. As one clade alone, Berteroella isolates from the allied gerera, paralleling toother tribe clades. Which indicated that Berteroella evolved independly to the other tribes inBrassicaceae. The results support the viewpoint that the genus should be raised to the rank oftribe based on the pollen grains morphology. In the phylogenetic tree, the accessions ofSisymbrieae fall into 5 clades, the accessions of Arabideae fall into Cardamine Clade andArabis Clade, the accessions of Lepidieae fall into Lepidium Clade and Arabis Clade, theresults support koch(1999),Warwich and Black (1993,1994 and 1997),Al-Ahehbaz'(2006)viewpoint that the three tribes are polyphyletic. Camelina of Sisymbrieae fall into Clade Arabis,which is consistent with Mark A. Beilstein (2006), dissupport that Z. Y. Wu et al (2003) putthe genus in Lepidieae. Descurainia of Sisymbrieae falling into Clade Lepidieae is consistentwith Mark A. Beilstein (2006). Coronopus of Lepidieae fall into Lepidium, which agrees onthe studies based on chloroplast DNA restriction site variation (Karsten Zunk et al 1999) andNumerical taxonomy (M.El Naggar 1993). Thlaspi also fall into Lepidium, which agree withKlaus-Peter Kolbe's (1982) view. Capsella falling into Arabis Clade is consistent with theresults of Koch.et al (2001),Abdel Khalik K. (2002) and Mark A. Beilstein (2006). Inaddition, the placement of Brassiceae is different from contradional classificationary systerm.In our clade tree, Brassiceae falls into Sisymbrieae, not parallel to other tribes. So furtherstudies need to do to test if Brassiceae should be united into sisymbrieae or be still keptisolated tribe.