Observations on the mycorrhizal status of Polygonum viviparum in the Polish Tatra Mts. (Western Carpathians)

Polygonum viviparum is one of very few herbaceous plants known to form ectomycorrhiza; in the Tatra Mts. it is one of dominants in the alpine zone, but also descends down to the feet of the massif. Specimens of this plant were collected from 5 sites at the altitude range 900– 2150 m, from granite and limestone. It allowed an estimation of the ectomycorrhizal diversity as well as preliminary ecological observations. Roots were also stained in order to check potential presence of arbuscular mycorrhizal colonization. Ectomycorrhizae were present in all specimens (with 2–5 morphotypes observed on single plants). In total, 17 morphotypes were observed and briefly described. The most widespread were the mycorrhiza of Cenococcum geophilum and a brightly coloured morphotype resembling the ectomycorrhizae of Russula sp. No important differences in ectomycorrhizal colonization between low and high localities were found. Observed general differences in abundance and diversity of mycorrhiza in P. viviparum between sites could most probably be connected with plant community composition (presence/absence of ectomycorrhizal shrubs maintaining ectomycorrhizal fungi), although mycorrhizae were present also in sites devoid of other ectomycorrhizal plants. Structures associated to arbuscular colonization (vesicles, hyphal coils) were occassionally observed, but without formation of arbuscules.


INTRODUCTION
Polygonum viviparum L. (Polygonaceae) is a widely distributed arctic-alpine species (Pa w ł o w s k a 1972), commonly found in the arctic tundra and high mountain regions of the northern hemisphere. In the Tatra Mts. (the highest massif of the Carpathians, with maximum altitude of 2663 m) it is one of dominant species in the alpine and subnival zones (up to 2480 m a.s.l. in the Lodowy massif; Pa w ł o w s k i 1956), but also descends down to the meadows at feet of the massif (900 m), which makes a very large altitudinal range. It is one of very few herbaceous plants known to form ectomycorrhiza (reviewed by G a r d e s and D a h l b e r g 1996). H e s s e l m a n (1900) first reported the presence of ectomycorrhizal tips with a Hartig net on the root system of P. viviparum. This observation was then confirmed for plants from different sites as Central Alps (C o n s t a n t i n and M a g r o u 1926; Pe y r o n e l 1930, 1937; Fo n t a n a 1977), Eastern Alps (H a s e l w a n d t e r and Re a d 1980; Re a d and H a s e l w a n d t e r 1981; B l a s c h k e 1991a, 1991b), Tatra  Some data suggest presence of two types of mycorrhizal colonization in P. viviparum. The paralell colonization of root system by arbuscular mycorrhizal fungi and ectomycorrhizal fungi was reported by S t a h l (1900) from arctic site and by B l a sc h k e (1991a, 1991b) from the Bavarian Alps. N e s p i a k (1953) also mentioned the AMF colonization of roots of P. viviparum in the Tatra Mts.
The aim of the present work was to describe the mycorrhizal status of P. viviparum in the Polish Tatra Mts. (Western Carpathians) and to contribute to the study of the diversity of mycorrhiza of this interesting arctic-alpine species. It is thought as a pilot study initiating more complex analyses of ectomycorrhizae in alpine habitats and their role in building the arctic-alpine macromycete diversity in the Tatra Mts.

MATERIALS AND METHODS
Description of the study sites and sampling of plant specimens. Five sampling sites were chosen on the stations of Polygonum viviparum growing on granitic and calcareous bedrock in Polish Tatra Mts., in different parts of the altitudinal range of the species (Tab. 1). Samples of plant root systems were collected twice during the vegetation period on three sites and once on two others (Tab. 1); 4-5 plants were collected from each site. Plants were taken together with the embedding soil (about 20×20×15 cm), transported in plastic bags and stored in a fridge (when quickly analyzed) or frozen at -20° C. Samples were soaked in water, then roots of P. viviparum were washed and carefully separated from roots of other plants. Only roots connected to rhizome were considered. Ectomycorrhizal colonization was analyzed and afterwards the roots were stained for checking the endophytic colonization.
Additionally, ectomycorrhizal structure on longitudinal and cross sections were observed. Ectomycorrhizal tips for sections were embedded in the synthetic resin (Historesin embedding kit -Leica, Germany; prepared according to manufacturer's instructions) and cut on the microtome (Leica RM 2135, Germany; 6−7 μm). Slides were observed using microscope with differential interference contrast (DIC). Ectomycorrhizal tips were stored in FAA solution (G e r l a c h 1972).
For observations in scanning electron microscope, fresh ectomycorrhizal tips were fixed in 2% glutaraldehyde solution in cacodylate buffer and dehydrated in the increasing acetone and ethanol concentration series (M a s s i c o t t e et al. 1987).
EM quantitative comparison. Quantitative comparison of ectomycorrhizal colonization was very difficult to estimate due to the tight mixture of roots in case of alpine grassland samples and thus only approximate data could be obtained. Ectomycorrhizal tips were counted and related to the length of the roots, as follows:

M=
number of mycorrhizal tips length of roots [cm] Each site was characterized by a mean M value from all specimens (4-5). Additionally, the ratio of alive and dead mycorrhizal tips (M l/d ) was calculated.
AM colonization. For estimation of endophytic colonization, roots of P. viviparum were stained according to the modified method of P h i l i p s and H a y m a n (1970). Roots were softened using 7% KOH solution, washed with water and bleached with H 2 O 2 containing NH 3 (10:1 v/v) for a few minutes. The material was then acidified in 5% lactic acid solution and stained with 0,01% cotton blue (anilin blue, methyl blue) solution in lactic acid. All steps were conducted in room temperature and lasted (apart from bleaching) 24 h each. Stained material was stored in pure lactic acid. Prior to staining the roots were kept in 50% ethanol solution.
Ectomycorrhizae were formed mainly on delicate, secondary roots, the side branches of dark roots growing from the rhizome. All ectomycorrhizae were simple and did not form any ramifications. In some tips traces of renewed growth were visible in the form of slight segmentation (beaded mycorrhiza; A g e r e r 1991).
Hartig net was of paraepidermal type in all examined ectomycorrhizae (A g er e r 1991), fungal colonization was limited to anticlinal walls of rhizodermal cells (Fig. 1F). The hyphae of Hartig net were ramified and closely cohering together forming "palmetti" structures. Root cells of Hartig net zone were strongly elongated transversally (CCq=0.32, comp. A g e r e r 1987-2002).
Key for the determination of ectomycorrhizae. The dichotomous key is presented below in order to facilitate the identification of morphotypes described in this paper. Both macroscopical features (colour, surface structure) and microscopical characteristics of fungal mantle and extramatrical structures in "plan view" were included. Average density of mycorrhizal tips for all analysed plants was almost equal in spring and autumn: 0.084 and 0.082 respectively. The ratio: alive vs dead mycorrhizae, was estimated for all sites. In all spring samples, number of living mycorrhizae was considerably lower than dead (M l/d < 1). This quotient was the lowest on the site 3 (M l/d < 0.16).
The share of morphotypes in the total number of mycorrhizal tips was very differentiated. Cenococcum geophilum and "Polygonirhiza lacteocinerea" were clearly dominant, constituting appr. 23 % of all mycorrhizal tips each. "Polygonirhiza vulpina", "P. aspera", "P. arenaria" and "P. fusca" represented 8-9 % of total number of ectomycorrhizae each. Other morphotypes were less numerous, and sometimes limited only to few tips ("P. granulosa", "P. aurata", "P. terrea", "P. maculata"). Domination of Cenococcum geophilum, "Polygonirhiza lacteocinerea" and "P. vulpina" was correlated with their presence in all (in the case of two first) or most (3 -in case of the third) sites; to the contrary, "P. tuberoidea" and "P. arenaria" were limited to single sites only. Some differences were observed in presence of morphotypes in samples collected in spring and autumn (sites 1, 2 and 3). Five morphotypes on the site 1 ("P. aspera", "P. rufocystidiata", "P. radiata", "P. salebrosa", "P. tenua") and two on the site 2 ("P. tuberoidea" and "P. aspera") were present only in spring (in the case of "P. tuberoidea" it was connected with a clear domination of this morphotype in the samples). On the other hand, Cenococcum geophilum was almost absent in spring, while in autumn it was very frequent in all sites. This was also true for "P. arenaria" on the site 1 (only 4 tips observed in spring and very abundant occurrence in autumn). These single observations are too scarce, however, to formulate any general conclusions.
There were no clear differences between sites in the average number of mycorrhizae in relation to altitude. Although the M value for the high mountain site 3 was low, the data for two other high-altitude stands (4 and 5) were higher and comparable with data for the lower located sites (1 and 2).
Endomycorrhiza in Polygonum viviparum. Prevailing part of roots did not manifest any traces of AMF colonization, however, several roots taken from site 2 in spring contained intraradical structures resembling those formed by endomycorrhizal fungi (members of Glomeromycota) -massive hyphae, coils and vesicles. No arbuscules, however, were observed.

DISCUSSION
Presence and diversity of ectomycorrhizae on Polygonum viviparum in the Tatra Mts. Seventeen ectomycorrhizal morphotypes were described in the samples from the Tatra Mts. Comparable number was reported in the observations from Italian Alps (Fo n t a n a 1977). The diversity of morphotypes on single specimens of P. viviparum reported from the Alps also corresponds well with the situation in the Tatra Mts. Fo n t a n a (1977) observed 2-3 morphotypes on average on a single plant, with maximum of 5 different mycorrhizae. The plants from Denali National Park (Alaska) had at least 3 morphotypes each (Tr e u et al. 1996). Also in the material from Rocky Mts. "several morphotypes" were mentioned (M a s s i c o t t e et al. 1998). The universal presence of ectomycorrhizal colonization in P. viviparum on sites in whole altitudinal range of Tatra Mts. is in agreement with majority of observations. However, it does not correspond with some data, especially that of N e s p i a k (1953), who found specimens without any ectomycorrhizal colonization in two alpine sites in High Tatra. However, comparison of data (from literature and present observations) for plants growing on different sites reveal rather small direct role of the position above sea level in the detected number of ectomycorrhizae, even though mycorrhizal colonization generally decreases with altitude (rewieved in K ö r n e r 1999). More probably, it could be suspected that the composition of the plant communities might have a strong influence on the ectomycorrhizal population of P. viviparum. A distinct, positive relationship was observed between the diversity of ectomycorrhizae of P. viviparum, and the presence of other ectomycorrhizal plants in its vicinity (cf. Tab Importance of this factor increases with the environmental stress at high altitude locations, diminishing the capability of fungi to grow and form fruitbodies. Nevertheless, a quite high level of mycorrhizal colonization on the site 4 (although with only 3 morphotypes, two of them common for all the investigated sites), also devoid of ectomycorrhizal shrubs, suggests a notable autonomy of P. viviparum in the formation and maintenance of ectomycorrhiza.
The presence of alive ectomycorrhizae in samples collected in spring and in autumn, together with a similar average number of mycorrhizal tips, suggest that the colonization is generally stable throughout the year. The absence of several morphotypes in spring could be the result of weaker ability to recolonize the roots after winter dormancy, however other reasons (eg. patchy distribution of mycelium) cannot be excluded.

Morphotypes recorded on Polygonum viviparum.
A comparison of ectomycorrhizae found in the Tatra Mts. with those from other sites is difficult, as very few descriptions are available. D o m i n i k et al. (1954) described an "ectotrophic mycorrhiza of the A type"; this type comprises a simple mycorrhiza without important ramifications nor growth modifications, with a primitive (plectenchymatous), loose mantle and Hartig net of different depth (D o m i n i k 1961). The paper includes some general remarks on the mycorrhiza of P. viviparum, but does not allow comparison of morphotypes. One of dominant morphotypes in Tatra Mts. was formed by an ascomycete Cenococcum geophilum. As a result of its commonness and a very characteristic appearance, the presence of this mycorrhiza is reported in most investigations from the whole distribution area of P. viviparum (e.g. Fo n t a n a 1977; Re a d and H a s e l w a n d t e r 1981; Tr e u et al. 1996; M a s s i c o t t e et al. 1998). It is not surprising, as the mycorrhiza of this fungus was described from many plant hosts (M a i a et al. 1996), and it was found on several other arctic-alpine species, as Dryas octopetala, D. integrifolia, Salix spp., Kobresia belliardi (Fo n t a n a 1963; Tr a p p e 1964; Re a d , H a s e l w a n d t e r 1981; M a s s i c o t t e 1998). In lowlands Cenococcum often dominates in dry environments (Tr a p p e 1964). It is not the case in the mountains, characterized by relatively high falls and long snow depositions. As suggested by Re a d & H a s e l w a n d t e r (1981), the commonness of this fungus in such areas could be connected with an efficient spread strategy, that is the production of very resistant and abundant sclerotia, rather than any special symbiotic features. The high resistance of Cenococcum to frost, experimentally demonstrated by C o r b e r y and L e Ta c o n (1997) can also be important factor. The species from the genera Amanita, Inocybe and Russula were also reported to form mycorrhiza with P. viviparum (rewieved in G a r d e s and D a h l b e r g 1996); an ectomycorrhiza of Alnicola cholea and P. viviparum was also recently described (M o r e a u et al. 2006). A morphotype called "Polygonirhiza lacteocinerea" strongly resembles an alpine mycorrhiza formed by Russula nana (=Russula emetica Fr. var. alpestris Boud.), described by Fo n t a n a (1977). The presence of this mycorrhiza in Tatra Mts. is probable since this fungus is very common in the alpine belt there (N e s p i a k 1960; M . Ro n i k i e r , pers. obs.). In the case of some morphotypes, many characteristics link them to the mycorrhizae described on trees. "Polygonirhiza epidermoidea" share mantle features with e.g. the mycorrhizae of some Russula spp. (epidermoid mantle structure with inner plectenchymatous layer, scarce extramatrical hyphae). A similar mycorrhiza is formed by Russula firmula on Pinus mugho (Tr e u 1990). The morphotype "Polygonirhiza aurata" seems to be close to the ectomycorrhiza of Tomentella galzini described on Quercus (J a k u c s et al. 1997 as "Quercirhiza fibulocystidiata"; K õ l j a l g et al. 2001). Both have an olive-green colour, pseudoparenchymatous, angular mantle structure and very characteristic cystidia with single clamps and thick walls below them. The only difference between these two morphotypes seems to be the lack of ramified emanating hyphae in the mycorrhiza of P. viviparum. Also the features of the group of "black" mycorrhizae found on P. viviparum lead to suppose their possible relationships with several tree mycorrhizae, e.g. "Piceirhiza nigra" (G r o n b a c h 1988), identified as formed by a member of Thelephoraceae (A g e r e r et al. 1995). These morphotypes, having a pseudoparenchymatous mantle structure ("Polygonirhiza aspera", "P. rufocystidiata", "P. salebrosa", "P. fusca"), could be formed by this group of fungi. Representatives of Thelephoraceae forming dark mycorrhizae are mostly species producing resupinate fruit-bodies on wood; interestingly, such morphotypes clearly dominated in sites in the vicinity of Picea abies or Pinus mugo (cf. Tab. 1, 2), so they could be formed not by alpine fungi but species related with trees -Tomentella spp. It is not possible, however, to refer such an assumption to available mycological data from the Tatra Mts. as this group of fungi has not been studied in this area so far. A very characteristic star-like hyphae arrangment and the presence of dark, incrusted emanating hyphae in "Polygonirhiza aspera" resembles strongly "Fagirhiza setifera" (B r a n d 1991), however the Fagus mycorrhiza has abundant cystidia, lacking in the P. viviparum mycorrhiza.
The general morphological aspects of ectomycorrhiza formed by P. viviparum in the Tatra  Presence of arbuscular mycorrhiza in Polygonum viviparum. Traces of probable arbuscular mycorrhizal colonization were sporadically observed. The roots contained some characteristic structures typically associated with the arbuscular mycorrhiza, as hyphae, coils and vesicles, but they were not accompanied by arbuscules. Most observations of P. viviparum reported lack of endomycorrhizal colonization, although it was incidentally found in arctic sites (S t a h l 1900) as well as in the mountains (N e s p i a k 1953; B l a s c h k e 1991a, 1991b). N e s p i a k (1953) noticed exclusively endomycorrhizal colonization without ectomycorrhizae in the same root system, while B l a s c h k e (1991a, 1991b) found both kinds of symbiosis occurring together. None of these authors, however, mentioned the formation of arbuscules in the roots. The infection of non-host roots by AM fungi, including formation of vesicles, was reported in some cases; the main signal which controls the development of functional symbiosis probably acts by trigerring fungal genes responsible for change of hyphal growth and physiology during arbuscule formation (review in G i o v a n n e t t i and S b r a n a 1998). Considering the presence of some AM structures in P. viviparum roots, the capacity of this plant to form this kind of symbiosis seems to be probable even if not important ecologically. Lack of arbuscules could possibly be also due to their short-lived appearance during the vegetation period, as it was reported in the study of the high-mountain Ranunculus adoneus colonization by Glomus tenuis (M u l l e n , S c h m i d t 1993). Regular phenological study or controlled cultures would be necessary to verify potential factors responsible for establishment of arbuscular mycorrhiza in P. viviparum.
The present study is the first contribution focused on the mycorrhiza of Polygonum viviparum in the Tatra Mts. and the Carpathians. The results showing that ectomycorrhizal colonization is a regular situation in this species, but affected by several factors, should be the starting point for future studies employing rigorous morphological/anatomical descriptions of morphotypes, regular survey of carpophores on permanent plots and employing DNA comparisons of mycorrhizae and carpophores. Including comparative analysis of diversity of mycorrhizae in neighbouring ectomycorrhizal plants in plant communities with P. viviparum would al-low a direct estimation of share/independence of the ectomycorrhizal diversity of P. viviparum. of Prof. Katarzyna Turnau (Jagiellonian University, Kraków). The authors thank Prof. K Turnau for helpful discussions and comments, and Dr. Anna Ronikier for help in gathering plant samples and comments on the manuscript. Polygonum viviparum jest jednym z nielicznych gatunków roślin zielnych, które tworzą ektomikoryzę. W Tatrach rdest żyworodny należy do gatunków dominujących w piętrze alpejskim, występuje również niżej sięgając do podnóży masywu. Celem badań była wstępna analiza różnorodności ektomikoryz tworzonych przez ten gatunek w Tatrach oraz ogólna analiza jej zależności od warunków ekologicznych takich jak wysokość nad poziom morza oraz skład zbiorowisk roślinnych. Korzenie P. viviparum były również dodatkowo badane pod kątem obecności kolonizacji endomikoryzowej.