Morphological and genetic diversity of European cranberry (Vaccinium oxycoccos L., Ericaceae) clones in Lithuanian reserves

Of the 50000 or more plant species used by man for food and medicine, only a small fraction is cultivated, the remainder being harvested directly from their natural habitats [1]. However, due to human pressure, such as habitat loss, land conversion or over-harvesting, many of these plants, especially those of medicinal value, are considered to be at risk. The cultivation of such species could improve the status of natural populations by alleviating the pressure of over-harvesting, assure unrestricted access to that particular crop and allow the selection of more resistant and productive varieties [1]. One European example of such a plant species is Vaccinium oxycoccos L. (syn. Oxycoccus quadripetalus Gilib., Oxycoccus palustris Pers.; Ericaceae Juss.), which is widely wild-harvested throughout much of its natural range and cultivated only in Russia and Estonia [2]. This species, the European cranberry, is a dwarf, woody, evergreen clonal shrub with slender, rooting stems, occasionally up to 0.8–1.0 m tall, with short, usually erect flowering shoots. The leaves are leathery, dark, glossy green dorsally, glaucous ventrally and frequently revolute with an entire blade margin. Racemes of 1–5, white, pink or red, protandrous flowers are pollinated mostly by solitary or social bees [3] and high fruit production frequently occurs following autogamy [4]. The fruit is an over-wintering, edible berry (the cranberry). Although fruit-set in natural populations may be high, the plant mostly reproduces vegetatively, forming large clones some hundreds of years old [3]. This plant has three (or four, depending on taxonomic treatment) ploidy levels: mainly tetra-and hexaploid populations are found, but pentaploids are also reported from the Czech Republic and Sweden. Diploids are usually treated as a separate species, namely V. microcarpum (Turcz. ex Rupr.) Schmalh. [5,6]. Vaccinium oxycoccos has a circumboreal distribution. In Europe it usually grows on Sphagnum peat bogs and is present in the north-western part of the continent, from Ireland, the British Isles and Scandinavia, throughout Central and Eastern Europe, the Balkan countries, Bulgaria, and even extending as far east as Siberia (N. Asia) and Japan. It also occurs in Greenland and the northern part of North America [3]. The wild-harvested fruit of V. oxycoccos is considered a substitute for that of V. macrocarpon Aiton, widely cultivated in the US and GB [7], and is commercially used in Scandinavia, the Baltic States, Poland, Belarus, Ukraine, Russia, Alpine zone of Switzerland, France, and Italy [8,9], as well as the USA and Canada. The fruit Abstract

is a source of phenolic compounds and athocyanins that have antibacterial, anticarcinogenic and antioxidant properties [10].
In nature, the plant grows on peat in poorly drained sites with a very high water level, and on very acidic soils (pH ranging from 3.0 to 4.5).In recent years, however, the peat bog vegetation has been seriously threatened by the effects of land reclamation.Large areas of raised bogs have also suffered from eutrophication, which has had an adverse effect on species composition.For example, in Lithuania, a drop in water level has promoted the growth of associated shrubs [Ledum palustre L., Calluna vulgaris (L.) Hull, Vaccinium uliginosum L.] and has reduced the vitality of V. oxycoccos, by causing a critical reduction in the natural resources of the species [11].Despite this considerable habitat loss, cranberries are still harvested in the wild, which further erodes natural populations of the species.This, in turn, has stimulated interest in the further cultivation and breeding of this crop [12].However, in order to promote future cranberry breeding and production, especially in countries having no such tradition, the morphological and horticulturally important characters of the plant require investigation [13,14].Above all, there is a need for information about the genetic variability of the species, so as to optimize the sampling strategy, especially as the plant displays great intraspecific morphological diversity.For example, individual cranberry clones from the same population may differ considerably in terms of berry size, color and shape, as well as shoot length.This high degree of morphological variability has been reported, for instance, from Poland [15] and the Czech Republic [5].Particular clones can also differ in their production of medicinally useful phytochemicals [10].Investigations carried out in Lithuania during 1965-1970 confirmed that berry shape is a very variable character [11].Since then, breeding and diversity of Lithuanian genotypes has become the subject of broad research and has provided a basis for V. oxycoccos studies.The application of molecular markers proved to be especially useful in assessing the diversity of the collected plant material [16].
It is important to select natural forms that display highest productivity, resistance to adverse environmental factors (diseases) and good fruit taste and size [17].Owing to the complicated system of morphological descriptions used in separating individual cranberry clones, clone identification is prone to errors.Molecular markers, however, allow the direct assessment of genetic diversity as a means of determining objectively differences in genetic material.
The aim of this investigation, which is based on the results of earlier morphological studies [11,12], is to investigate phenological, morphological and genetic diversity, and horticultural value of Lithuanian V. oxycoccos clones collected from the wild at two nature reserves, namely, Žuvintas and Čepkeliai, where we have previously observed a high degree of morphological variation between populations.

Plant material and evaluation of phenological, morphological, and horticultural characters
Plant material for the study was collected from two strictly protected reserves, Žuvintas and Čepkeliai, during 1998-1999 (Fig. 1).The Žuvintas reserve is situated in the southern part of the central Lithuanian lowlands, (N54°29' E23°40').It comprises of a complex of Žuvintas and Amalvas wetlands, covering 6847 ha.The Žuvintas reserve is notable for its diverse plant communities.The Čepkeliai reserve is situated in Southern Lithuania, close to its border with Belarus (N54°00' E24°30').Raised bogs cover about 80% of the Čepkeliai wetlands.
During field work at both sites, we collected 29 distinctive clones differing clearly in vegetative characters, including berry size, shape, and color.Cuttings of selected clones (size 10-15 cm) were transferred into the field collection at Kaunas University Botanic Garden, Kaunas, Lithuania.The annual precipitation for Kaunas district is 500-750 mm, and the average temperature exceeds 6.7°C.The cuttings were planted in acid peat beds (pH 4.0-5.0)and cultivated under ex situ conditions for further investigations.
Phenological observations were conducted throughout the entire vegetative growth period during 2000-2010.On the same days, twice a week, we checked the collection, and the following phenological phases were recorded for each of the clones: commencement of shoot growth, commencement of flower bud development, commencement of flowering, end of flowering, commencement of fruit ripening, end of fruit ripening, and the end of the vegetative growth period.
Detailed evaluation of morphological diversity for these clones was carried out during the years 2004-2010.We measured or assessed the following characters for each clone: leaf size and shape, shape of leaf apex and base, recurving of leaf margin, color of fully opened flower, length of peduncle, berry size, berry shape and color, shape of berry in cross-section, extent of waxy layer of fruit, and color of flesh (mesocarp) of berry.Berries were weighed using an analytical balance (Ishida Co., Japan, model DJ-150E; sensitivity of 0.01 g) and the average weight calculated.For each clone, three replicates of 50 fruit were weighed.The yield production of each clone was calculated, again for triplicate samples, by weighing the total berries per 1 m 2 .The average generative shoot length for each clone was calculated based on the measurement of 50 randomly selected shoots.The mean area of a leaf was determined by scanning triplicate samples of 30 randomly selected leaves from each clone with a CI-202 (CID Bio-Science, USA) portable laser leaf area meter.

Genetic analysis
For DNA extraction, we used 100-130 mg of fresh, new cranberry shoots collected in spring.For RADP analysis, we used nine 10 nt-long primers of random sequence (Fermentas, Lithuania; Roth, Germany; Tab. 1).DNA was PCR-amplified using an automatic thermocycler (Mastercycler, Eppendorf, Germany) under the following conditions: initial denaturation for 4 min at 94°C, 44 cycles of denaturation for 1 min at 94°C, primers annealing for 1 min at 35°C, extension for 2 min at 72°C followed by a final extension for 5 min at 72°C.PCR reaction per primer was done not less than twice.The reaction products were fractionated by electrophoresis in 1.5% agarose gel and visualized using ethidium bromide stain and UV light.The length of DNA bands was estimated according to the gene ruler Gene RulerTM 100 bp DNA Ladder (Fermentas, Lithuania).
For SSR DNA amplification we used five 20-23 nt-long primers (Biomers, Germany; Tab. 2).SSR primers were chosen and the DNA amplification reaction performed according to Boches et al. [18].
PCR products obtained by ABI 3130 xl Genetic analyzer (Applied Biosystems), length of fragments were set using standard of ROX-500 (Applied Biosystems) as an internal size standard.Allele sizes were visualized using GeneMapper v. 3.5 software (Applied Biosystems).

Data analysis
Data was analyzed using the statistical package STATISTICA 6 (Stat Soft., Inc.).Statistical differences were identified with ANOVA, followed by Fisher's LSD test at P ≤ 0.05 and 0.01.Population genetic analysis, such as principal coordinates analysis (PCA) and analysis of molecular variance (AMOVA), analysis were performed using the GenAlEx 6 [19].Calculation of the observed number of alleles, Nei's gene diversity H [20], Shannon's information index I, total gene diversity Ht, gene diversity within populations Hs, gene diversity between populations Gst = (Ht − Hs /Ht), gene flow Nm = 0.5 (1 − Gst)/Gst and the generation of a Nei's genetic distance based dendrogram were achieved using POPGENE V 1.31 software.

Phenological and morphological diversity
Vaccinium oxycoccos clones showed significant phenological plasticity.Although the duration of the vegetative growth period for the years 2004-2010 did not differ statistically between clones, in other years we observed great variation in the commencement of certain phenological phases (ranging from 10-19 days).For example, there were 3-9 day-long shifts between clones in the commencement of flower bud formation and 7-20 days-long shifts between clones in the commencement of berry ripening.The flowering phase, depending on the clone, started from mid May to the first ten days of June.Clones 99-Ž-02, 99-Ž-07, 98-Č-01, and 98-Č-09, in particular, would commence anthesis as late as the first ten days of June.Of the numerous molecular markers available, the random amplified polymorphic DNA (RAPD) technique has become the subject of much debate.Nevertheless, limitations to the reproducibility of RAPD markers have been largely overcome by improvements in laboratory techniques and band scoring procedures [26] and this method has been successfully used to investigate the degree of cloning in many plant species [27,28].Indeed, comparison of RAPD and AFLP molecular markers has confirmed the reproducibility of RAPD markers [29].
In our study, RAPD markers proved to be a powerful method for the detection of spatial genetic variation, allowing the selection of particularly valuable genotypes.For example, it has been demonstrated for V. stamineum L., that plants with the greatest genetic diversity within and between populations are better adapted to cope with different environmental conditions [27].
Analyses of the genetic structure of V. macrocarpon and certain other plants showed that, for many species, the greatest genetic variation may be detected within populations [30,31], in contrast to the results obtained for Oryza rufipogon Griff.[32].Average molecular variance between populations of Vaccinium species was 87.7%, whereas the value obtained from within populations of the same species was 27.7% [30].With regard to the genus Vaccinium, greatest intrapopulation variation was detected for American cranberry, V. macrocarpon: more than 91% [30], followed by V. uliginosum L. -90.3% [27] and V. myrtillus L. -86.19% [29,33].By contrast, our study of V. oxycoccos populations exhibited relatively low (71%) intrapopulation genetic variation, based on RAPD.This also contrasts with our SSR results, which showed 97% genetic variability between Žuvintas and Čepkeliai populations, whereas in other vegetatively propagated clonal species it ranged from 71-86% [34][35][36].
Greater genetic variation was found in the Čepkeliai population than in the Žuvintas population.This may be due to greater penetration of the latter site and the intensive picking of berries resulting in a reduction in propagation by seed and the promotion of clonal growth.
In conclusion, V. oxycoccos seems a promising crop for cultivation under Central European conditions.This study shows that some useful morphological characters such as leaf size, berry size and berry shape can be used to assess potentially interesting genotypes.The considerable genetic diversity found within the studied populations indicates that the selected clones from Čepkeliai and Žuvintas reserves are well suited to the prevailing environmental conditions and may prove a useful source of plant material for future study.

Tab. 1 Tab. 2
Primers and their sequences used for RAPD analysis of two populations of Vaccinium oxycoccos in Lithuania.Primers and their sequences used for microsatellite analysis of two populations of Vaccinium oxycoccos in Lithuania.