Mycotrophy in Gilliesieae , a threatened and poorly known tribe of Alliaceae from central Chile Micotrofía en Gilliesieae , una tribu amenazada y poco conocida de Alliaceae de Chile central

The fi ve known genera of Gilliesieae have their diversity center in the Mediterranean zone of central Chile, where many of their habitats are threatened by urban expansion, industrial and agroforestry activities, as well as other anthropogenic impacts. Very little is known about the biology of these particular geophytes, the majority of which currently have either vulnerable or endangered status, mainly due to their dispersed and small populations generally associated to remnants of native vegetation. As mycorrhizal associations are essential for soil resource acquisition and stress mitigation in most plants, our objective was to assess the hitherto unknown mycotrophic status of ten species of Gilliesieae from central Chile by qualitative and quantitative assessment of intraradical fungal structures. All sampled genera (Gethyum, Gilliesia, Miersia, Solaria, Speea) showed regular presence of arbuscular mycorrhiza, while other mycorrhizal or putatively mutualistic associations, like dark septate endophytes, were practically absent. Mycorrhizal colonization of fi ne roots reached a mean of ca. 45 % across all examined taxa, with highly variable values ranging from 9 % to 82 % in Miersia tenuiseta and Gilliesia curicana, respectively. The high level of mycorrhization indicates that arbuscular mycorrhiza should be considered for conservation strategies of threatened species or biotechnological use in plant propagation. The main future task is the identifi cation of the associated fungal taxa.


INTRODUCTION
The Gilliesieae Lindl.are a poorly known South American tribe of the Alliaceae whose most studied feature so far has been floral morphology, which mimics insects in several species (Ravenna 2000, Rudall et al. 2002).The highest diversity of the group is encountered in central Chile, with about 14 species belonging to the fi ve genera Gilliesia Lindl.(5 species), Miersia Lindl.(5), Solaria Phil.(1), Speea Loes.(1) and Gethyum Phil.(2) (Muñoz et al. 2000, Ravenna 2005, Escobar 2012) (Fig. 1).Some taxa are also known from Argentina, Bolivia and Per u.Most species typically grow in the Mediterranean climate zone of central Chile between 33º and 38º S, which is considered a hotspot of biodiversity (Arroyo et al. 1999).Populations are found in the understor y of, or close to, native forest dominated by sclerophyllous and deciduous trees.Distribution and population size is ver y heterogeneous across species (Escobar et al. 2010) and the majority is classifi ed as endangered (Ravenna et al. 1998), due to overlapping of their main distribution area with the most densely populated urban and rural zones as well as the centers of agriculture and forestr y in the countr y.The recent, major revision of Chilean Gilliesieae by the second author (the first since Reiche 1893 and Hutchinson 1939) was based on freshly collected material of most recorded species, some of them ver y rare.This gave us the opportunity to scrutinize for the fi rst time the presence of mycorrhizal associations in species of all known Chilean genera.
Generally, ecology of Gilliesiae, especially association with other organisms like soil microorganisms, pathogenic (Jackson 1926) and mutualistic fungi, pollinators, etc., to date remains largely unknown.Knowledge of mycotrophism is especially impor tant f o r u n d e r s t a n d i n g a c q u i s i t i o n o f s o i l resources by Gilliesieae and their possible integration into interspecific mycor rhizal networks.In other Alliaceae and related monocot families, arbuscular mycor rhizal associations are commonly found (Wang & Qiu 2006).The arbuscular mycorrhizal (AM) symbiosis is one of the most important and widespread mutualistic associations between absorbing roots and soil fungi of the phylum Glomeromycota (Schüssler et al. 2001, Smith andRead 2008).The primar y benefi t of the mycorrhizal association for the plant host is the extension of its bioactive zone in the soil via fungal mycelia and mycorrhizal networks.This allows a more ef ficient access to soil water and nutrients (Smith & Read 2008) as well as a wider range of allelopathic protection (Sikes 2010, Bar to et al. 2011), therefore having a direct effect on plant performance and community structure.The relative contribution of the mycobionts to the mentioned benefi ts has been shown to be species-specific (Lendenmann et al. 2011).This implies that investigation of the general mycotrophic status of endangered plant species should be followed by an assessment of the mycorrhizal fungal species involved.Both natural mycorrhization and application of mycorrhizal fungal inocula have been shown to be effi cient in restoration processes (Herrera et al. 1993, White et al. 2008).As a fi rst step, we analyzed presence and abundance of arbuscular mycorrhizal fungal structures in roots of ten species belonging to fi ve genera of Gilliesieae (Fig. 1), providing the fi rst dataset about mycotrophism in this particular and interesting plant group.

Study site
Chilean Gilliesieae were sampled from central Chile between 33º11' S (Región Metropolitana, Santiago) and 37º28' S (VIII Región, Biobío).The sampling zone is generally characterized by a temperate climate but is heterogeneous in terms of precipitation regime and temperatures along latitudinal and altitudinal gradients, which results in a mosaic of different ecosystems and habitats.The northern part of the zone is characterized by dry summers lasting seven to eight months, less than 40 mm of rain per month and almost 80 % of rainfall in winter, and low relative humidity (annual mean 70 %); also there is a high thermal amplitude with almost 13 ºC difference between the warmer and colder months; daily maximum and minimum difference ranges between 14 and 16 ºC (Dirección Meteorológica de Chile, http:// www.meteochile.gob.cl/).
The southern range is characterized by a shorter summer (less than four months), precipitations up to 1300 mm per year and winter rainfall making up 65-70 % of annual total precipitation.The proximity to the ocean produces a narrower annual thermal amplitude with 7.5 ºC difference between the warmer and colder months; similarly, daily thermal amplitude is only reaching up to 10 ºC (Dirección Meteorológica de Chile, http://www.meteochile.gob.cl/)

Sampling, processing and staining of arbuscular mycorrhizal roots
Between early and mid spring (August-October) of 2007, the roots of 10 previously identifi ed species of Gilliesieae were collected (Table 1).Notably, populations of some species were very small so that sampling had to be restricted to the minimum in order to avoid major damage.Roots were dug out, cut at the hypocotyl, cleaned with distilled water and stored in Falcon tubes in a cooler for transport.In the laboratory, the material was carefully washed in tap water to remove soil particles and debris, and then fi xed in 70 % ethanol.Before staining, the roots were cleared in 10 % KOH solution at 121 ºC for 20 min, subsequently washed with distilled water and stored in water/lactic acid/glycerol, according to Brundrett et al. (1996).
The staining of the AM fungal structures was performed in 10 ml Eppendorf tubes with Trypan Blue 0.01 % at 121 ºC for 20 min.Before microscopy, roots were rinsed with distilled water and deposited overnight in Falcon tubes with lactoglycerol to remove excessive stain.Selected root segments were then mounted on glass slides in lactoglycerol and observed under a light microscope (Leitz Dialux, Leitz, Wetzlar, Germany) at distinct magnifi cations (100x to 1000x).Details of root and mycorrhizal structures were documented by digital photomicrographs using a Nikon Coolpix 950 camera (Nikon, Tokyo, Japan).

Relative mycorrhization
Percentage of mycorrhization was calculated using the Trouvelot method (Trouvelot et al. 1986, also cited in Covacevich et al. 2001).For every specimen examined, twenty 1 cm segments of stained fi ne roots were mounted in parallel on a slide and observed at 40x magnifi cation.Every root segment was assigned to one of six categories (0-5) of relative mycorrhization (intensity of colonization by mycorrhizal fungal structures) ranging from 0 % to > 95 % mycorrhization.The mycorrhization percentage was then calculated as follows: M Trouvelot (%) = (n1 + 5n2 + 30n3 + 70n4 + 95n5) / N N is the number of observed segments, n1 to n5 represent the number of segments categorized as one to fi ve respectively.The mycorrhization percentage was obtained for each plant (Covacevich et al. 2001).When diagnostic mycorrhizal structures were observed in more than three samples (n > 3) of each species, we calculated the mycorrhization percentage and the frequency of each element (hyphal coils, intraradical hyphae, vesicles, arbuscules).

RESULTS
Roots of all examined genera showed one or more diagnostic structures of arbuscular m y c o r r h i z a e ( T a b l e 1 , F i g . 2 ) , v i z .: appressoria, aseptate hyphae, hyphal coils, arbuscules, and vesicles.Differences between colonization patter ns and morphology of individual structures (e.g., Fig. 2C, 2D) suggest the presence of more than one species of arbuscular mycorrhizal fungi (AMF).Other fungus-root associations that could be expected, such as dark septate endophytes (DSE), were practically absent, with septate hyphae only found in G. curicana.

DISCUSSION
A r b u s c u l a r m y c o r r h i z a s a r e r e g u l a r l y present in Chilean Gilliesieae and, based on our data, seem to be the exclusive form of mycorrhizal symbiosis in this plant group.The range of relative mycorrhization found across all examined species of Gilliesieae is in concordance with available data of AM colonization of plants in natural ecosystems (Read et al. 1976, Smith & Read 2008, Treseder & Cross 2006).Morphological dif ferences between diagnostic intraradical fungal str uctures such as vesicle shape, hyphal patterns and staining intensity across or within plant host species suggest the presence of more than one species of AMF, a common situation in herbaceous plants (Smith & Read 2008).Other typical fungus roots associations like DSE were practically absent, with some endophytic, dark septate hyphae found only in roots of G. curicana.
Differences in the phenology of diagnostic fungal structures appear across samples with some roots only showing either vesicles or arbuscules.Apar t from taxon-specific traits of involved glomeromycotan fungi, this could be an expression of seasonal dynamics of mycorrhiza formation which is not backed by our data obtained from a one-off sampling campaign.Dif ferent AM str uctures were present in root samples from all visited habitats (plantation, native forest and wayside), without a characteristic trend.Never theless, the highest value of AM percentage was found in the wayside samples.It has been shown that soil disturbance by human activities can lead to changes in AMF communities, especially by reducing their diversity, given that different genera respond differently to such disturbance (Dodd 2000).This has been extensively proven in agricultural soils where AMF species vanish over longer periods of regular tillage (e.g., Jansa et al. 2002, Oehl et al. 2005).Jha et al. (1992) found that in degraded forest habitats, AMF populations were smaller than in less disturbed sites.
Quality and quantity of mycor rhization in Chilean Gilliesieae is probably relevant for its conser vation status as well: many of their habitats which are mostly located in the Mediter ranean zone of central Chile, are patchy, having suf fered from land use changes, especially during the last decades, and are degraded or surrounded by agricultural land or exotic forestr y plantations (Arroyo et al. 1999).It has been shown that the stability and diversity of plant communities and soil microbial communities (including AMF) are mutually cor related and that common mycorrhizal networks can structurally and functionally connect not only roots of conspecifi c host individuals but also of different plant species in a defi ned area (Haystead et al. 1988;McNaughton & Oesterheld 1990, van der Heijden & Horton 2009).Whereas Gilliesieae grow preferentially on spots where other herbaceous plants are, at least seasonally, rare or absent, they are often found close to native tree species like Cryptocarya alba (Lauraceae), Quillaja saponaria (Rosaceae) and Peumus boldus (Monimiaceae), among others (personal observation), which have been shown to form AM as well (Garrido 1988, Godoy et al. 1991, Carrillo et al. 1992).This could be a hint for the existence of shared mycorrhizal mycelia which so far in Chilean monocots has only been shown for the achlorophyllous and obligate Mycorrhizal percentage (M %) and frequency (%) of arbuscular mycorrhizal structures in species of Gilliesieae from central Chile.Habitats: NF = native forest, WS = wayside, PL = plantation.Arbuscular mycorrhizae structures: HC = hyphal coils, IH= intraradical hypha, Arb = arbuscules, Ves = vesicles.Mycorrhization values are means ± SD. * n was too low for mean calculation, +: present, but frequency was not determined.
In this context, future work should be focused on the dependency of Gilliesieae on common mycor r hizal networks and the identification of the associated fungal species.Apart from the potential signifi cance of mycor r hiza for restoration practices (Haselwandter 1997), it has been shown that despite of the worldwide distribution of many AMF taxa, there seems to be a more specifi c relationship between certain, possibly locally adapted species and their plant hosts (van der Heijden 1998, Klironomos 2003, Vogelsang et al. 2006).