Oenothera, a unique model to study the role of plastids in speciation

The subject of this thesis was to develop molecular approaches
appropriate to investigate speciation processes. The genus
Oenothera was chosen for study, since it offers the unique
possibility to exchange plastids, individual or more chromosomes
and/or even entire haploid genomes (so-called Renner complexes)
between species. In addition, a rich stock of information in
taxonomy, cytogenetics and formal genetics is available, collected
for more than a century of research. Interspecific exchange of
plastids, nuclear genomes or chromosomes often leads to
mis-development of the resulting hybrids. These inviable hybrids
form hybridization barriers responsible for speciation. In the case
of plastid and nuclear genome exchange, hybrid bleaching is
frequently observed, which results from plastome-genome
incompatibility (PGI) due to compartmental co-evolution.
Traditional work on Oenothera was almost exclusively restricted to
classical genetic and cytogenetic approaches. Subsection Oenothera,
the best studied of the five subsections in the section Oenothera,
was used in this work. It is comprised of three basic nuclear
genomes, A, B and C, which occur in homozygous (AA, BB, CC) or
stable heterozygous (AB, AC, BC) combination. In nature, the
nuclear genomes are associated with five basic, genetically
discernible plastid types (I - V) in distinct combinations. The
following results were obtained: (i) Biochemistry with Oenothera is
not trivial due to exceedingly high amounts of mucilage and tannins
which adversely interfere with the isolation of macromolecules and
enzymatic reactions. A basic biochemistry for the material was
therefore developed initially, notably to obtain appropriate
subcellular fractions, restricable, amplifyable and clonable DNA,
RNA, supramolecular protein assemblies and proteins of appropriate
purity. (ii) Evaluation of the PGI literature clearly indicates
that PGI can form hybridization barriers according to the
Dobzhansky-Muller gene pair model of speciation, even if the genes
reside in different cellular compartments. (iii) Oenothera PGIs
could be classified into four genetically distinct categories,
which influence hybridization barriers in different ways. (iv)
Co-dominant marker systems (SSLP and CAPS) were generated for both,
nuclear genome and plastome. Their potential was successfully
evaluated with crossing programs designed to exchange plastomes,
genomes, or individual chromosomes between species. (v) The
plastome markers allowed to genotype 41 subplastomes to judge
inter- and intraplastome diversity and displayed molecular loci
linked to the genetic behaviour of basic plastome types I - V. (vi)
A single, highly polymorphic marker (M40) was sufficient to
genotype 29 different Renner complexes of the basic genome types A,
B and C. (vii) Markers specific for all seven Oenothera chromosomes
were selected. Combined with the genetics of a partial permanent
translocation heterozygote (ring of 12 chromosomes plus 1 bivalent,
which behave as two distinct linkage groups) they allowed the
assignment of molecular linkage group 7 to chromosome 9•8 of the
classical Oenothera map. Material for the assignment of the
remaining chromosomes and their arms was produced or selected so
that both map types can now be fully integrated. (viii) In parallel
to work on the nuclear genome, the sequences of the five basic
Oenothera plastomes were completed (in cooperation). Elaborated in
this thesis, due to its limited coding potential, conserved nature,
and substantial knowledge about photosynthesis, plastid chromosomes
provide relatively easy access to “speciation genes” and selection
pressures causing speciation. (ix) Phylogenetic analysis of the
sequences provided a plastome pedigree, and also an idea about the
age of the subsection, i.e. back to the middle of Pleistocene,
approximately 1 mya ago. This contributed to solve a long lasting
question in the Oenothera literature. (x) Application of
appropriate algorithms uncovered for the first time that plastomes
are subject to natural selection and hence contribute to
speciation. This was questioned repeatedly. (xi) A novel weighting
strategy, combining classical genetic data on plastome-genome
compatibility/incompatibility with molecular data and bioinformatic
approaches, was applied to deduce potential plastid determinants
for PGI. (xii) In a case study it could be shown that a single
plastid locus contributes substantially to PGI in the interspecific
hybrid AB-I, which was found to be defective in photosystem II. A
plastome I-specific deletion in the bidirectional promoter region
between psbB and clpP was found to be responsible for the phenotype
observed. The finding is consistent with reduced levels of psbB
mRNA and its product CP47 chlorophyll a apoprotein of photosystem
II, with spectroscopic data and phenotype. (xiii) Available data
indicate that interspecific plastome-genome hybrids represent some
sort of “network mutants”. This would imply that speciation is
predominantly a regulatory phenomenon. In the studied cases PGIs
are is involved in the fine-tuning of regulation of photosynthesis,
rather than in an adaptation of its structural components. This is
considered as a major finding of this thesis.