A diploid in the Arctic – genetic and morphological variation of Cochlearia groenlandica L.

Abstract

The genus Cochlearia, together with its sister genus Ionopsidium Rchb., forms the tribe Cochlearieae, a rather small tribe which comprises about 30 taxa, which either x=6 or x=7 basic chromosome number. The x=6 series forms a polyploid complex, which is distributed throughout Europe, while the x=7 basic chromosome number is only found only in diploid plants (2n=14), which are widespread in the arctic. Plants with both basic chromosome number occur as diploids only in Iceland (2n=12, 14). As the Icelandic plants show morphologically similarity to the arctic plants, it has been debated whether they both should be included in the widespread arctic species C. groenlandica. The aim of this thesis was to obtain more knowledge about the relationships and taxonomic status of arctic C. groenlandica. For this purpose, populations of C. groenlandica have been sampled from Alaska, Canada, West Greenland, East Greenland and Svalbard in addition to both coastal and alpine Cochlearia populations from Iceland. The collected populations represented various habitats such as beach cliffs, tundra, and urban areas. Single nucleotide polymorphisms (SNPs), derived from RAD-sequencing (RADseq), were applied to study the genetic variation of these populations, morphological variation were analyzed in plants cultivated under controlled conditions, and finally ploidy level and chromosome number were assessed by flow cytometry and chromosome counting. The RADseq data showed a clear geographical structure among the C. groenlandica populations, which is in line with previous studies using RADseq to analyze Cochlearia populations. The populations from West Greenland constituted one genetic group, which was separated from a group consisting of populations from East Greenland, Svalbard and Alaska. The Canadian population had an intermediate position between these two groups. When the Icelandic populations were included, the Icelandic alpine population clustered closer to East Greenland and Svalbard than to the Icelandic coastal populations, in line with previous reports suggesting a genetic link between the Icelandic alpine populations and C. groenlandica populations in Svalbard. For the morphological analyses, the same leaf and flower traits as have been used to study other Cochlearia populations were used to evaluate leaf shape, leaf base shape and leaf size, as well as petal length. There was a large variation in leaf shape and leaf base shape within and between the different populations, but no clear pattern which could be related to the geographically structure genetic variation. However, morphological differences between populations were found also after the plants had been grown in controlled conditions, which indicates that genetic factors or epigenetics underlie the observed phenotypic variation. Even though not consistently, there seemed to be a correspondence between leaf size and leaf shape and the habitat that the plants grew in, as all the tundra populations had the small lanceolate leaves with cuneate/trullate base, while costal and urban populations had significantly larger reniform leaves with subcordate/cordate base. The flow cytometry analysis and chromosome counting confirmed that all the C. groenlandica populations were diploid with x=7, that there was no overlap with plants with x=6, and that C. groenlandica indeed seems to be a diploid in the Arctic. The results in this thesis show that there are both morphological and genetic differences between the analyzed populations of C. groenlandica, but there are no indications that these constitute more than one species. Previous suggestions to refer the Icelandic alpine populations to C. groenlandica are supported in this thesis, but it is still uncertain whether they might be referred to as ecotypes or subspecies. However, the Icelandic coastal populations do not show similar close associations to the Arctic.

The genus Cochlearia, together with its sister genus Ionopsidium Rchb., forms the tribe Cochlearieae, a rather small tribe which comprises about 30 taxa, which either x=6 or x=7 basic chromosome number. The x=6 series forms a polyploid complex, which is distributed throughout Europe, while the x=7 basic chromosome number is only found only in diploid plants (2n=14), which are widespread in the arctic. Plants with both basic chromosome number occur as diploids only in Iceland (2n=12, 14). As the Icelandic plants show morphologically similarity to the arctic plants, it has been debated whether they both should be included in the widespread arctic species C. groenlandica. The aim of this thesis was to obtain more knowledge about the relationships and taxonomic status of arctic C. groenlandica. For this purpose, populations of C. groenlandica have been sampled from Alaska, Canada, West Greenland, East Greenland and Svalbard in addition to both coastal and alpine Cochlearia populations from Iceland. The collected populations represented various habitats such as beach cliffs, tundra, and urban areas. Single nucleotide polymorphisms (SNPs), derived from RAD-sequencing (RADseq), were applied to study the genetic variation of these populations, morphological variation were analyzed in plants cultivated under controlled conditions, and finally ploidy level and chromosome number were assessed by flow cytometry and chromosome counting. The RADseq data showed a clear geographical structure among the C. groenlandica populations, which is in line with previous studies using RADseq to analyze Cochlearia populations. The populations from West Greenland constituted one genetic group, which was separated from a group consisting of populations from East Greenland, Svalbard and Alaska. The Canadian population had an intermediate position between these two groups. When the Icelandic populations were included, the Icelandic alpine population clustered closer to East Greenland and Svalbard than to the Icelandic coastal populations, in line with previous reports suggesting a genetic link between the Icelandic alpine populations and C. groenlandica populations in Svalbard. For the morphological analyses, the same leaf and flower traits as have been used to study other Cochlearia populations were used to evaluate leaf shape, leaf base shape and leaf size, as well as petal length. There was a large variation in leaf shape and leaf base shape within and between the different populations, but no clear pattern which could be related to the geographically structure genetic variation. However, morphological differences between populations were found also after the plants had been grown in controlled conditions, which indicates that genetic factors or epigenetics underlie the observed phenotypic variation. Even though not consistently, there seemed to be a correspondence between leaf size and leaf shape and the habitat that the plants grew in, as all the tundra populations had the small lanceolate leaves with cuneate/trullate base, while costal and urban populations had significantly larger reniform leaves with subcordate/cordate base. The flow cytometry analysis and chromosome counting confirmed that all the C. groenlandica populations were diploid with x=7, that there was no overlap with plants with x=6, and that C. groenlandica indeed seems to be a diploid in the Arctic. The results in this thesis show that there are both morphological and genetic differences between the analyzed populations of C. groenlandica, but there are no indications that these constitute more than one species. Previous suggestions to refer the Icelandic alpine populations to C. groenlandica are supported in this thesis, but it is still uncertain whether they might be referred to as ecotypes or subspecies. However, the Icelandic coastal populations do not show similar close associations to the Arctic.