Root fungal associations in some non-orchidaceous vascular lithophytes

Plant roots in natural ecosystems are colonized by a diverse group of fungi among which the most common and widespread are arbuscular mycorrhizal (AM) and dark septate endophyte (DSE) fungi. Th ough AM and DSE fungal associations are well reported for terricolous plant species, they are rather poorly known for lithophytic plant species. In this study, we examined AM and DSE fungal association in 72 non-orchidaceous vascular plant species growing as lithophytes in Siruvani Hills, Western Ghats of Tamilnadu, India. Sixty-nine plant species had AM and 58 species had DSE fungal associations. To our knowledge, we report AM fungal association in 42 and DSE fungal association in 53 plant species for the fi rst time. Th ere were signifi cant diff erences in total root length colonization and root length colonized by diff erent AM and DSE fungal structures among plant species. In contrast, the diff erences in AM and DSE fungal colonization among plants in various life-forms and lifecycles were not signifi cant. AM morphology reported for the fi rst time in 56 plant species was dominated by intermediate type AM morphology. A signifi cant negative relationship existed between total root length colonized by AM and DSE fungi. Th ese results clearly suggest that AM and DSE fungal associations are widespread in lithophytes.


Introduction
Plants grow in a variety of habitats, including those which are not normally suitable for plant growth.One of the harshest habitats is the surfaces and crevices of rock where low moisture, nutrients and high temperatures limit plant's existence.Plants growing on rocks are termed lithophytes and these types of plants occur in diff erent arid (Zwieniecki & Newton 1995;Bashan et al. 2002;Lopez et al. 2009) and non arid (Tozer et al. 2005;Grӧger & Huber 2007;Muthuraja et al. 2014) ecosystems.Nevertheless, lithophytes also colonize rocks that are constantly misted by waterfalls or rocks dripping with water.Lithophytes have certain biological peculiarities, including capturing moisture from rain, dew, melting snow or water trickling down the rocks, and acquire nutrients from rain water, litter, and feeding on insects or their own dead tissues (Alves & Kolbeck 1993;Gold & Bliss 1995;Tozer et al. 2005).Th ese features enable lithophytes to occupy and thrive on the rock surfaces.Based on their existence and dependence on rock, lithophytes are classifi ed as epilithic or epipetric and endolithic forms.Epilithic forms grow superfi cially on rock surfaces (e.g., orchids, ferns, algae, bryophytes), whereas endolithic forms grow embedded in rocks (e.g., algae, fungi).Further, plants growing exclusively on rocks are known as obligate lithophytes and those that growing on rocky as well as on other substrates are known as the facultative lithophytes.
Generally, success of plants growing on harsh environment depends on their ability to associate with microorganisms (Puente et al. 2004a;b;Bashan et al. 2007).Plants growing in different environments are colonized by arbuscular mycorrhizal (AM) fungi belonging to Glomeromycota, and it is one of the most widespread associations among land plants (Smith & Read 2008).The AM fungi aids plants in their uptake of water and nutrients, and impart tolerance against various stresses.This association enables plant's to survive and thrive under harsh environments that are not conducive for plant growth.Although the roots of several terricolous plant taxa have been examined for the presence of AM fungal association, their occurrence in lithophytic or saxicolous plant species is limited to certain plant groups.For example, the occurrence of AM fungal association has been examined in lithophytic mosses (Zhang & Guo 2007) and ferns (Lehnert et al. 2009;Kessler et al. 2010;Muthukumar & Prabha 2013;Muthuraja et al. 2014).The results of some of these studies suggest that non-mycorrhizal condition appears to be more prevalent in lithophytic than in terricolous taxa (Kessler et al. 2010).
A great diversity exists in inter-and intracellular distribution of AM fungal structures within plant roots.Based on the distribution of the fungal structures within plant roots, the AM colonization patterns have been termed as Arum-, Paris-and intermediate types.The intraradical fungal hyphae are linear and intercellular in Arum-type, and intracellular and coiled in Paris-type.In between these two extremes, there exists a diverse colonization patterns termed as intermediate types.Though AM colonization patterns in terricolous plant species are widely reported, it is rather scanty for lithophytic species (e.g., Muthukumar & Prabha 2013;Muthuraja et al. 2014).
A group of predominantly ascomycetous fungi belonging to the order Helotiales also colonizes plant roots either individually or along with AM fungi.These fungi termed as dark septate endophyte (DSE) fungi form dematiaceous, regularly septate hyphae with microsclerotia and/or moniliform cells in plant roots.Although these fungi have been recorded in roots of more than 1000 plant species growing in a range of habitats (Jumpponen & Trappe 1998;Mandyam & Jumpponen 2005;Muthukumar & Prabha 2013;Muthuraja et al. 2014;Kumar & Muthukumar 2014), their occurrences in many plant taxa and habitats are yet to be examined.Like mycorrhizal fungi, DSE fungi also associate with plants under extreme environmental conditions (Read & Haselwandter 1981;Newsham et al. 2009).The widespread occurrence of DSE fungi in stressed environments suggests that DSE fungi might act as surrogate mutualists in these environments (Bledsoe et al. 1990;Newsham 2011).Inspite of their frequent occurrence in plant roots, the role of DSE fungi on plant growth, development and ecology is largely unknown.Limited studies exploring the role of DSE fungi on plant growth show that these fungi could enhance plant performance when nutrients are available predominantly in organic forms (Newsham 2011).Like the incidence of AM, DSE fungal associations in lithophytes are also limited to certain plant groups.The occurrence of DSE fungal association in lithophytic fern and fern-allies has been reported recently from the Eastern and Western Ghats of southern India (Muthukumar et al. 2014;Muthuraja et al. 2014).
As lithophytic habitats are characterized by extreme environmental conditions and plants in stressed habitats depend more on association with endophytic microorganisms for their growth and survival, we hypothesized that the incidence of endophytic root fungal associations in lithophytes are high.To test this hypothesis, we examined the roots of non-orchidaceous lithophytes to assess the incidence and extent of AM and DSE fungal colonization.In addition, we assessed the influence of plant habit and life cycle on the occurrence of AM and DSE fungal associations.Finally, we also determined the existence of any possible relationship between the AM and DSE fungal variables as these fungal types occupy the same niche within plant roots.

Sampling site and investigated plant species
Kovai Kutralam (10°56'12.9"N76°41'14.3"E),located 22 miles West of Coimbatore City in Siruvani Hills is a part of the Nilgiri Biosphere Reserve of Western Ghats, which is one of the biodiversity hotspots of the world (Balaji & Hariharan 2013).In addition to providing drinking water to the Coimbatore city, Siruvani Hills are also home to certain tribes like the Mudugars and Irulars.The vegetation at lower to higher altitudes ranges from southern dry mixed deciduous forest, southern moist mixed deciduous forest to tropical evergreen forest in hill tops (Balaji & Hariharan 2013).This region receives a rainfall varying from 750 to 2000 mm, mostly from the northeast monsoon during September to November, with a dry period from December to May.The minimum and maximum temperature range from 10 to 22 °C (average 16 °C) and 35 to 43 °C (average 39 °C) respectively (Balaji & Hariharan 2013).The 72 nonorchidaceous plant species in this study belong to 65 genera and 38 families (Tab.1).Twelve of these were pteridophytes belonging to seven families.Fifty-five plant species were herbs, 14 were shrubs/under shrubs, and three were trees.Seventeen species were annuals, and 55 were perennials (Tab.1).Plant nomenclature follows The Plant list (www.ipni.org).
The roots were scooped off the rock surfaces, and the debris and mosses attached to roots were removed.The roots were then gently washed in water and fixed in FAA (formalinacetic acid-alcohol, 5ml:5ml:90ml) and transported to the laboratory for processing.As most of the roots were not associated with soil, or identifiable spores were absent in the substrates attached to the roots, we did not enumerate AM fungal spore numbers or diversity.

Preparation of roots for AM and DSE fungal assessment
Fixed roots were washed free of FAA, cut into 1-cm fragments, cleared in 10 % KOH (Phillips & Hayman 1970), acidified with 5 N HCl and stained with trypan blue (0.5% in lacto glycerol) overnight.Roots that remained dark after clearing were bleached in alkaline H 2 O 2 prior to acidification.The percentage of root length colonized by AM and DSE fungal structures and total root length colonized were estimated according to a magnified intersection method (McGonigle et al. 1990).Briefly, five intersections between the microscopic eyepiece crosshair and roots were recorded in 10 1-cm long root bits for each plant at ×400 in an Olympus BX5 trinolular microscope.The presence of AM and DSE fungal structures was noted at each intersections were recorded.The percentage of root length colonized by AM and DSE fungal structures and total root length colonized were estimated as the proportion of root length containing different fungal structures (McGonigle et al. 1990).
The AM-morphology was classified as Arum-, Parisand intermediate types based on whether the fungal hyphae were present, mainly as hyphae running through intercellular spaces or within cells as coils, respectively following descriptions of Dickson (2004).The intermediate subtypes categorized by Dickson (2004)

Statistical analysis
Analysis of variance (ANOVA) was used to assess the significance of variation for fungal variables among plant species.Similarly, ANOVA and t-test was used to test the influence of plant life-forms and lifecycle patterns on the fungal variables.Regression analysis was used to assess the relationship between AM and DSE fungal variables.Percentage data on AM and DSE fungal root colonization was arcsine transformed prior to analysis.All statistical analysis was performed using SPSS for Windows (Version 9).

Extent of AM fungal colonization
The extent of AM fungal colonization and AM fungal structures varied significantly within plant species.The percentage root length with total colonization (%RLTC) varied significantly among plant species (F The average percentage of root length with different AM fungal structures (except %RLA) and %RLTC was maximum in trees and minimum in herbs (Fig. 2A).Among plants of different lifecycles, average %RLH, %RLA, %RLV was higher in annuals than in perennials (Fig. 2B).Contrarily, perennials had a higher %RLH, %RLAC and %RLTC than annuals.However, the differences between plant life-forms and life cycles were not significant.

AM morphological types
Intermediate type AM morphology was dominant among the various life-forms and lifecycles.Typical Paris-type AM was present only in the fern Nephrolepis cordifolia.Typical Arum-type AM was observed in 14 species and among the intermediate types; I1 was present in 10 species, I2 in 11 species and I4 in 33 species.
Among the various plant life-forms examined, 79% of the herbs, 71% of the under shrubs/shrubs and 100% of the trees had intermediate type AM (Fig. 3A).Arum-type AM was present in <30% of the herbs and shrubs.Like plants in different life-forms, plants with different lifecycles were also dominated by intermediate type AM with 71% of the annuals and 76% of the perennials possessing it (Fig. 3B).

Incidence of AM fungal colonization in lithophytes
In this study, the lithophytic plant community consisted of species belonging to different plant groups and families.These results agree with the studies where lithophytic plant communities have been reported to contain a diverse plant assemblage (Gröger & Huber 2007;Jayanthi et al. 2011).Ninety-six percent of the non-orchidaceous lithophytes examined in the present study was colonized by AM fungi and to our knowledge, AM symbiosis is reported for the first time in 42 plant species.High prevalence of AM fungal association in lithophytes in this study is in accordance with other studies where such an observation has been reported (Muthukumar & Udaiyan 2000b;Zhang & Guo 2007;Muthukumar et al. 2014;Muthuraja et al. 2014).Plants growing on rocks or shallow substrates employ several adaptations, including reduction or absence of root hairs and increasing the root diameter to maximize returns from limited carbon investments (Zwieniecki & Newton 1995;Poot et al. 2012).This might have contributed to the high incidence of AM in lithophytes examined in the present study as coarse roots tend to increase the mycorrhizal dependency of plants (Fitter & Moyersoen 1996).
The mycorrhizal status of some plant species appears to be significantly influenced by the nature of the substrate they grow.For example, Drynaria quercifolia reported mycorrhizal when growing on soil was nonmycorrhizal in the present study (Muthukumar et al. 2014;Muthuraja et al. 2014).This fern has been reported to be both mycorrhizal and nonmycorrhizal when growing as an epiphyte (Muthukumar & Udaiyan 2000b;Muthukumar & Prabha 2013).Like D. quercifolia, E. robustum examined was nonmycorrhizal.Species of Eriocaulon were reported to be both nonmycorrhizal (Ragupathy & Mahadevan 1993;Muthukumar & Udaiyan 2000a;Kumar & Muthukumar 2014) and mycorrhizal (Khan & Belik 1995;Brundrett et al. 1995;Rodrigues & Bukhari 1995;Ragupathy et al. 1990;Harikumar 2001).Species in Eriocaulon have adaptations to cope with an extreme shortage of soil nutrients like N and P. In addition, plants of Eriocaulon are evergreen, and  any nutrient investment in leaves is amortized over an extended period, which considerably reduces its reliance on soil resources and subsequently on AM association (Keddy 2010).Like other species of the genus, M. nudiflora was nonmycorrhizal in the present study (Logan et al. 1989;Brundrett et al. 1995;Tester et al. 1987;Silva et al. 2001;Muthukumar & Udaiyan 2000a).However, non-functional mycorrhiza (lacking arbuscules) has been reported in Murdannia semiteres by Radhika & Rodrigues (2007).
In contrast to earlier observations, Striga asiatica, the obligate root hemiparasite associated with grasses was colonized by both AM and DSE fungi (Ragupathy & Mahadevan 1993;Muthukumar & Udaiyan 2000a;Gworgwor & Weber 2003).Nevertheless, atypical AM characterized by the presence of only hyphae and vesicles were reported in Striga gesnerioides parasitizing Lepidagathis hamiltoniana from the high-altitude region of Maharashtra, India (Kamble & Agre 2013).Similarly, species like Chamaecrista pumila, Canscora heteroclita, Utricularia uliginosa, Lygodium flexuosum, Ficus microcarpa, Biophytum sensitivum, and Pouzolzia zeylanica previously reported to be nonmycorrhizal had AM fungal structures in the present study (Ragupathy & Mahadevan 1993;Muthukumar & Udaiyan 2000a;Khade & Rodrigues 2002).This clearly emphasizes the need for more rigorous examination of plants from different habitats to resolve the true mycorrhizal status of a plant species (Wang & Qiu 2006;Brundrett 2009).Habitat conditions like the availability of adequate AM fungal propagules or conditions necessary for the AM formation along with host factors contribute to the varied mycorrhizal status of a plant species (Smith & Read 2008).Variations are likely to occur for facultatively mycorrhizal plant species that occupy diverse habitats (Brundrett 2009).Though the mycorrhizal status of sedges could be influenced by habitat conditions (Muthukumar et al. 2004),Cyperus cyperinus examined in the present study was mycorrhizal.This corresponds with the findings of Muthukumar et al. (1996) who also found AM in this sedge growing on soil.Results of recent experimental studies do indicate that sedges could benefit from AM association like other plant species (Lagrange et al. 2011;2013).
Carnivorous plants inhabit nutrient-poor habitats and have highly specialized nutrient-capture strategies.These plants acquire substantial proportion of their nutrients by digesting their prey, and some have long and dense root hairs (Brundrett 2009;Ellison & Adamec 2011).Subsequently, carnivorous plants have become partially or fully independent of mycorrhizal associations (Brundrett 2009).However, the roots of Utricularia uliginosa examined in the present study had both AM and DSE fungal structures.This is in accordance with studies where mycorrhizal structures have been reported in the roots of carnivorous plants (Fuchs & Haselwandter 2004;Harikumar 2013).Although, the endophytic fungal associations have been reported in carnivorous plants, the nature and function of the associations have yet to be ascertained.
It is interesting to note that in spite of absence of spores in their substrate, more than 48% of lithophytes had >50% of their roots colonized by AM fungi.This emphasizes the importance of propagules other than spores in initiating colonization under lithophytic conditions.The relevance of the extra radical mycelia and mycorrhizal roots as a propagule source in mycorrhization has already been well recognized (Schalamuk & Cabello 2010).

AM morphology
We found the intermediate type AM to be common among lithophytes and even in different life-forms and lifecycles than typical Arum-and Paris-types.This is similar to some recent studies where intermediate type AM was found to be common among plant species examined.To our knowledge, we report AM morphology for the first time in 56 plant species.This complements Dickson et al. (2007) observation that even in plant species for which mycorrhizal status is available, AM morphologies is not known.Further, we also found AM morphologies different from those already reported in 12 plant species.This corroborates studies where different AM morphologies have been reported for the same plant species (see Dickson et al. 2007 and references therein).For example, Ageratum conyzoides possessing intermediate AM morphology in the present study is also capable of both Arum-and intermediate type morphologies (Muthukumar et al. 2003;Kumar & Muthukumar 2014).Research had established that AM colonization patterns in plant roots are an interaction between plant root structure and the fungal partners.Although not examined in the present study, root structure, especially those of the cortical region tend to exhibit modifications in plants growing on rocks (Zwieniecki & Newton 1995).This could have contributed to the variation in AM morphology in certain plant species from those already reported.Brundrett & Kendrick (1990) indicated that the presence or absence of intercellular air spaces in the root cortical region to be an important determinant of AM morphology.Later studies elaborated this view and speculated the influence of factors like light, moisture and temperature of the substrates on AM patterns in roots as these factors can influence root growth (Yamato & Iwasaki 2002;Yamato 2004;Becerra et al. 2007).However, detailed studies exploring root anatomy of lithophytic plant species, especially the intercellular spaces in roots could reveal their contribution to AM morphology.In addition to these, Cavagnaro et al. (2001) and Smith et al. (2004) have shown that AM fungal species colonizing roots could also influence AM fungal colonization patterns.

Incidence of DSE fungal colonization in lithophytes
Occurrence of DSE fungal association has been reported in a wide range of terricolous plant species belonging to

Influence of plant habit and lifecycle on the occurrence of AM and DSE fungal associations
In the present study, we did not find any significant difference in the extent of colonization for plants in different life-forms and lifecycles.This contrasts the observations of Muthukumar & Udaiyan (2000a) where terricolous shrubs and trees growing in the Western Ghats region of southern India had higher colonization levels than herbs.Nevertheless, the average %RLTC of <50% in annuals and perennials accords with the observations of Muthukumar & Udaiyan (2000a) where a higher proportion of annuals and perennials examined had colonization levels between 25% and 50%.These clearly indicate that AM fungi are equally important for plant species with different life-forms and lifecycle strategies even under lithophytic conditions.Like AM fungal association, the extent of DSE fungal colonization also failed to exhibit significant variations among plants of different life-forms and lifecycle strategies.Nevertheless, the results of the present study suggest that lithophytes are more dependent on AM than DSE fungi.This is evidenced by the existence of a significantly higher average %RLTC than %RLDTC (t 71 = 13.095;P<0.001) in general as well as among different life-forms and lifecycles.For example, the %RLTC in herbs, shrubs and trees was respectively 3.21, 2.82 and 7.30 folds higher than %RLDTC.Similarly, the %RLTC of annuals and perennials was 3.04 and 3.80 folds respectively higher than %RLDTC.This may be due to more conducive conditions for AM formation and functioning and is consistent with the trend observed for terricolous species (Lugo et al. 2009;Muthukumar & Prabha 2013).

Relationship between AM and DSE fungi
Co-occurrence of AM and DSE fungi were found in roots of 79% plant species examined.This is similar to studies where a high proportion of dual colonization of AM and DSE fungi in roots of plant species examined have been reported (Massenssini et al. 2014;Majewska et al. 2015).Our observation of a negative correlation between %RLTC and %RLDTC is in agreement with studies where an inverse relationship between these fungal variables has been reported (Muthukumar et al. 2006;Mandyam & Jumpponen 2008;Muthukumar & Tamilselvi 2010).While an interaction between AM and DSE fungal type is expected as they share the same niche within the root system, the nature of the relation between these two fungal types has received far less attention.It is generally believed that DSE fungi could aid plant growth and survival under circumstances where conditions are not conducive for AM functioning.For example, Barrow (2003) suggested that DSE fungi could enhance plant's performance in very dry soils of arid ecosystems.Similarly, DSE fungal colonization of the dominant grass Bouteloua gracilis from semi-arid grasslands of Mexico was four times higher than AM colonization (Medina-Roldán et al. 2008).The extent of root colonization by fungal endophytes depends on the carbon allocation by the host plant as well as the host, fungal and soil factors (Medina-Roldán et al. 2008;Smith & Read 2008).Recent research on the control of plant-fungal mutualism indicates that the symbionts offering the best benefits in terms of nutrients or carbon are rewarded, and the control is bidirectional (Kiers et al. 2011;Hart et al. 2013).Therefore, differential allocation of carbon to these fungal types by host plants may reason for the inverse relation between these fungal variables in the present study.

Conclusion
Overall, the results of the present study indicate that AM and DSE fungal associations are widespread among plants growing on rocks like those of saxicolous plant species.This is the first report of AM and DSE fungal associations for several plant species.We speculate that the dual association of AM and DSE fungi in lithophytes could improve plant adaptability and survival under rapidly changing conditions of rock surfaces as shown for some harsh habitats.Despite their common occurrence, the roles of these fungal associations on plant adaptations to grow and survive on rock surfaces are not known unlike the role of bacteria (Puente et al. 2004a;b).Experimental studies are, therefore, necessary to ascertain the role of AM and DSE fungal associations on growth and nutrition of plants growing on rock surfaces.
Root fungal associations in some non-orchidaceous vascular lithophytes are: linear hyphae are inter-and intracellular with the former possessing Arum-type arbuscules (intermediate type-1, I1); intracellular linear hyphae bearing Arum-type arbuscules (intermediate type-2, I2); intracellular linear hyphae bearing arbuscules and intercellular linear hyphae (intermediate type-3, I3); hyphal/ arbusculate coils along with intercellular linear hyphae (intermediate type-4, I4).We checked the previous reports on AM and DSE fungal associations and AM colonization patterns, and the results are presented in Tab. 1.

Table 1 .
Life-form, life cycle, arbuscular mycorrhizal (AM) and dark septate endophyte (DSE) fungal status, AM morphology (in parenthesis) of non-orchidaceous lithophytes in Siruvani Hills of Western Ghats.In the last column, previous reports on the occurrence of AM and DSE fungal associations and AM morphological types (in parenthesis) are listed.

Thangavelu Muthukumar, Marimuthu Chinnathambi and Perumalsamy Priyadharsini diverse
Lugo et al. (2009)e 1998)ar & Prabha 2013;Muthuraja et al. 2014on the occurrence of DSE fungal association in non-terricolous plant species are limited.We found DSE fungal association in 82% of the plant species examined, which included Eriocaulon robustum and M. nudiflora the two of the three nonmycorrhizal species.The presence of DSE fungal association in lithophytes that lacked AM association resembles the observations ofLugo et al. (2009)in South American bromeliads where all the epiphytic nonmycorrhizal plants were invariably colonized by DSE fungi.Roots of Costus speciosus (Costaceae), Parahemionitis cordata and Drynaria quercifolia reported possessing DSE fungal association in earlier studies lacked DSE fungal structures in the present study(Uma et al. 2010;Muthukumar & Prabha 2013;Muthuraja et al. 2014).To our knowledge, we report DSE fungal presence in 53 plant species for the first time.This clearly suggests the lack of information on the occurrence of DSE fungal occurrence in many habitats, especially in tropical regions(Jumpponen & Trappe 1998).Based on their observations,Lugo et al. (2009)hypothesized that DSE fungi could enable plants to survive extreme arid conditions prevailing in nonterricolous habitats by enhancing the drought tolerance mechanisms and nutrient capture.