Environmental conditions and host plant origin override endophyte effects on invertebrate communities

En i onmen al condi ions and hos plan o igin o e ide endophy e e ec s on in e eb a e communi ies Salla-Riikka Ves e lund &Ma jo Helande & S anley H. Fae h &Te ho Hy önen &Ka i Saikkonen Recei ed: 15 Decembe 2010 /Accep ed: 5 Janua y 2011 /Published online: 29 Janua y 2011 #The Au ho (s) 2011. This a icle is published wi h open access a Sp inge link.com Abs ac Sys emic ungal endophy es o g asses can p oduce high concen a ions o alkaloids ha a e known o de e in e eb a e he bi o es and educe hei abundance, especially in ag onomic g asses. G ass endophy es may also in luence in e eb a e communi y di e si y and composi ion. He e, we examined in a common ga den expe imen wi h wild all escue plan s and he ag onomic cul i a Ken ucky 31, whe he in ec ion by Neo yphodium coenophialum, he gene ic back- g ound (o igin) o he hos plan , abio ic ac o s, and hei in e ac ions a ec ed he in e eb a e communi ies li ing on all escue. We collec ed a o al o 18650 in e eb a es om he 480 all escue plan s, iden i ied hem o 97 mo phological axa (mainly o amily le el) ep esen ing i e eeding guilds (he bi o es, de i i o es, omni o es, p eda o s and pa asi- oids). In con as o pas li e a u e, endophy e in ec ion did no a ec abundances o any axon o eeding guild, o axonomic di e si y and he composi ion o he in e eb a e communi y. Ins ead, he in e eb a e communi y o na i e all escue appea s o be p ima ily d i en by en i onmen al condi ions and niche di e en ia ion among axonomical g oups o in e eb a es. We p opose ha communi y app oaches a e equi ed o unde s and o he ole o endophy es on a h opod abundances and di e si y in na u e. Keywo ds Endophy ic ungi .Tall escue .In e eb a e di e si y and communi y s uc u e .Func ional guild . He bi o y.Gene ic a ia ion In oduc ion Asexual Neo yphodium endophy es ( amily Cla icipi aceae) o m symbio ic ela ionships wi h many cool-season g asses belonging o he sub- amily Pooidae (Clay 1988,1990). In ec ions a e sys emic and he endophy e is ansmi ed e ically o he nex gene a ion h ough seeds (Scha dl e al. 2004;ClayandScha dl2002). Tall escue (Schedono us phoenix (Scop. Holub.) [ = Lolium a undinaceum (Sch eb.) Da bysh. = Schedono us a undinaceus (Sch eb.) Dumo ]) has been widely used as o age and u g ass in he Uni ed S a es o decades (Ball e al. 1993). Thus, one o he mos s udied g ass–endophy e associa ions is he N. coenophialum and all escue symbiosis (Saikkonen e al. 2006,2010). Tall escue cul i a s a e domina ed by a widely-adap ed cul i a named “Ken ucky 31”(he ea e e e ed o as K-31), which has a long g owing season and is esis an o pes s, d ough , poo soil condi ions, and a ia ions in soil pH (Ball e al. 1993). Based on he esea ch o his g ass–endophy e sys em, he ela ionship be ween he endophy ic ungus and i s hos has gene ally been hough o be mu ualis ic (Clay 1988; Clay e al. 1993; Saikkonen e al. 2006; Scha dl and Phillips 1997). Recen s udies ha e shown, howe e , ha his ela ionship can a y om mu ualism o an agonism, depending on he geno ype o he ungus and he hos as well as en i on- men al condi ions, especially in na i e g asses (Cheplick e al. 1989; Cheplick and Fae h 2009; Fae h 2002;Fae h and Saikkonen 2007; Fae h and Sulli an 2003). Saikkonen e al. (1998,2004,2006) he e o e p oposed ha he S.-R. Ves e lund (*):M. Helande Sec ion o Ecology, Depa men o Biology, Uni e si y o Tu ku, Tu ku FI-20014, Finland e-mail: [email p o ec ed] S. H. Fae h Depa men o Biology, Uni e si y o No h Ca olina a G eensbo o, G eensbo o, NC 27402, USA T. Hy önen :K. Saikkonen MTT Ag i ood Resea ch Finland, Plan P o ec ion, Jokioinen FI-31600 Finland Fungal Di e si y (2011) 47:109–118 DOI 10.1007/s13225-011-0089-x p e ailing concep o endophy es as mu ualis s is likely his o ical and sys em based a he han based on e idence om na u al popula ions. In he case o he all escue–N. coenophialum symbiosis, much o he esea ch has been done in he Uni ed S a es on ag onomic cul i a s such as K-31 (Saikkonen e al. 2006), al hough he o igins o his g ass a e in Eu asia. In hese ag onomic cul i a s plan ed ou side hei na i e dis ibu ional ange, Neo yphodium is widely known o cause de imen al e ec s (e.g., oxicosis) on e eb a e g aze s in high-nu ien ag onomic en i onmen s (Ball e al. 1993; Clay 1989,1990; Saikkonen e al. 2006,2010; Scha dl and Phillips 1997). These e ec s a e ela ed o high concen a ions o alkaloids (Clay 1990; Lyons e al. 1986), which a e known o de e bo h e eb a e and in e eb a e he bi o es (Bacon 1995; Bacon e al. 1977; Bazely e al. 1997; Siegel and Bush 1996, 1997; Vica i e al. 2002). Because alkaloids a e nu ien - ich compounds, hei syn hesis has cos o o he basic plan g ow h and ep oduc i e unc ions (Fae h 2002; Fae h and Bul man 2002; Fae h and Fagan 2002). These cos s may ou weigh he bene i s o he endophy e in ec ion in mos en i onmen s, bu pa icula ly so in nu ien -poo en i- onmen s in na u e (Ahlholm e al. 2002; Fae h 2002; Leh onen e al. 2005). Thus, in i s na i e habi a , in ec ed wild all escue may p oduce lowe le els and ewe ypes o alkaloids han i s cul i a ed and selec i e-b ed a ie ies in nu ien - ich en i onmen s in he in oduced ange (Saikkonen e al. 1998,2010;SiegelandBush1996;bu see Piano e al. 2005). Recen e idence suppo s his idea: (1) he le els and composi ion o alkaloids p oduced a ies among ungal species and geno ypes (e.g., Piano e al. 2005; Siegel and Bush 1997), also (2), alkaloid ypes and le els in na u al popula ions a y mo e han hose in ag onomic g asses (Bony e al. 2001; Fae h and Saikkonen 2007), (3) he numbe o non- oxic endophy e-in ec ed g asses exceed oxic ones (Fae h 2002), and (4) in some cases, in ec ion dec eased, a he han inc eased, he he bi o e esis ance o he hos plan (Fae h and Shocha 2010; Jani e al. 2010; Saikkonen e al. 1998; Schul hess and Fae h 1998). Al ough well-s udied in ag onomic cul i a s such as K-31 in in oduced a eas, he in e ac ions be ween all escue and Neo yphodium endophy es a e s ill la gely igno ed in hei na i e ange in Eu ope (Saa i e al. 2010; Zabalgogeazcoa and Bony 2005), p obably because all escue is no a p e e ed li es ock o age g ass (Niemeläinen e al. 2001) and li es ock oxicosis is a e (Zabalgogeazcoa and Bony 2005). The na u e and ecological impo ance o he all escue–N. coenophialum symbiosis may be di e en in i s na i e ange (Saikkonen 2000; Saikkonen e al. 1998; Siegel and Bush 1996). We examined whe he he N. coenophialum endophy e in ec ion and he o igin o he hos plan as well as abio ic ac o s and hei possible in e ac ions a ec he in e e- b a e communi y li ing on all escue. Besides he bi o es, ungal endophy es may also a ec de i i o es (e.g., Lemons e al. 2005) and he na u al enemies o he bi o es (Fae h and Shocha 2010; Ha ley and Gange 2009;Janie al. 2010; Omacini e al. 2001) o ende he bi o es mo e o less suscep ible o na u al enemies by a ec ing hei a ack a es (Ben ey and Denno 1997; Saa i e al. 2010) and delaying he bi o e de elopmen (e.g. B een 1994; Clay e al. 1985; Popay and Rowan 1994). Howe e , he e a e only a ew s udies ha ha e conside ed he impac o g ass endophy es on a h opod communi ies o unc ional g oups (e.g., A khami and Rudge s 2009; Fae h and Shocha 2010;Janie al.2010). In his s udy, we used a whole-in e eb a e communi y su ey o a con olled common ga den expe imen o es how in e eb a e di e si y and communi y s uc u e, and he numbe o indi iduals in unc ional in e eb a e axa and guilds di e s be ween (i) endophy e in ec ed (E+), endophy e ee (E-), and manipula i ely endophy e- ee (ME-) all escue, (ii) hos plan s o di e en o igin (wild popula ions om Åland, Go land, coas al Sweden and one ag onomical cul i a , K-31 om USA), and (iii) hos plan s g owing in di e en abio ic en i onmen s (nu ien and wa e ea men s). Based on he pas s udies on de ensi e endophy e-g ass mu ualism (Saikkonen e al. 2010)and he ew ecen s udies on how endophy es s uc u e a h opod communi ies in all escue (Rudge s and Clay 2008), we p edic ed ha : (1) endophy e in ec ion dec eases in e eb a e di e si y and communi y s uc u e, (2) he abundances o plan eeding in e eb a es (he bi o es, omni o es and de i i o es) a e highly a iable, bu in gene al, lowe in endophy e in ec ed (E+) hos plan s compa ed wi h endophy e- ee plan s, pa icula ly in he e ilized plan s, (3) he ag icul u al cul i a (K-31) shows s ong de e ence o plan eeding in e eb a es, and (4) endophy e in ec ion may bo h posi i ely and nega i ely a ec he na u al enemies o he bi o es (p eda o s and pa asi oids). Ma e ials and me hods Plan and seed ma e ial To es he e ec o in ec ion, hos plan o igin, and en i onmen al ac o s (wa e and nu ien ea men s), in Augus 2005, we collec ed seeds om mul iple na u al all escue popula ions by he Bal ic Sea in locali ies ha we e geog aphically sepa a ed om each o he by app oxima ely 500 km. These we e he island o Åland (8 popula ions), he island o Go land (9 popula ions), and he wes coas o Sweden (6 popula ions). 10 o 50 110 Fungal Di e si y (2011) 47:109–118 indi iduals we e collec ed om each popula ion, and h ee seeds om each plan indi idual we e s ained o mic oscopic examina ion o he endophy e in ec ion s a us (Saha e al. 1988). Neo yphodium coenophialum in ec i i y a ied be ween 85–100% in all all escue popula ions om he h ee loca ions. Unin ec ed (E-) and in ec ed (E+) seeds we e combined sepa a ely om popula ions wi hin each o he h ee s udy a eas (Åland, Go land, and coas al Sweden). In o he wo ds, we pooled all E- seeds and hen all E+ om he popula ions wi hin each loca ion o c ea e h ee ba ches o E- seeds and h ee ba ches o E+ seeds ha ep esen ed he h ee geog aphic o igins. In addi ion o plan s om na u al all escue popula ions, we used E+ and E- K-31 ( om T. Phillips, Uni e si y o Ken ucky) cul i a seeds in ou expe i- men . To es he ole o he endophy e on in e eb a e communi ies while con olling o plan geno ypic backg ound, we expe imen ally emo ed he endophy e om po ion o E+ seeds (manipula i ely endophy e- ee plan s = ME-). To kill he ungus while he seeds emained iable, he E+ seeds we e hea – ea ed by keeping he seeds in wa m wa e (56-57°C) o 10– 20 min. All all escue seeds om na u al popula ions, K-31 cul i a and endophy e- emo ed seeds we e ge mina ed on mois issue pape in Pe i-dishes in a g eenhouse and plan ed 7 days a e ge mina ion o indi idual po s wi h sand and pea mix u e. Table 2 The e ec s o endophy e s a us (E+ = endophy e in ec ed, E- = endophy e- ee, and manipula i ely endophy e ee = ME-), wa e and nu ien ea men s (C = con ol, N = nu ien , W = wa e , and WN = wa e + nu ien ), plan o igin (A = Åland, G = Go land, and S = coas al Sweden; K = cul i a “Ken ucky 31”) and plan biomass on he abundances o dip e ans, mi es (Aca i), Hymenop e ans, collembolas and Coleop e ans Taxon Feeding guild Endophy e s a us (E) T ea men (TRT) Plan o igin (PO) E*TRT E*PO TRT*PO Plan biomass d =2 d =3 d =3 d =6 d =6 d =9 d =1 F p F p F p Fp Fp Fp F p Dip e a he bi o ous 0.20 0.8202 2.34 0.0727 2.15 0.0931 2.30 0.0337 0.59 0.7402 2.57 0.0070 9.21 0.0026 de i i o ous 0.84 0.4317 11.62 <0.0001 3.04 0.0291 0.92 0.4807 1.06 0.3846 2.36 0.0133 5.47 0.0199 omni o ous 1.04 0.3540 0.97 0.4091 1.29 0.2791 3.04 0.0063 0.90 0.4948 1.18 0.3058 1.59 0.2077 pa asi ic 0.06 0.9398 0.97 0.4072 1.63 0.1820 1.40 0.2122 0.99 0.4289 0.77 0.6458 5.75 0.0169 p eda o y 1.52 0.2190 2.57 0.0537 1.07 0.3628 1.30 0.2541 0.45 0.8420 0.68 0.7289 0.31 0.5761 Aca i omni o ous & pa asi ic 1.16 0.3141 3.76 0.0110 0.07 0.9743 0.41 0.8735 1.69 0.1220 0.61 0.7885 4.66 0.0315 Hymenop e a pa asi ic 2.13 0.1204 0.68 0.5659 4.76 0.0028 0.51 0.7970 0.73 0.6279 1.48 0.1518 0.59 0.4446 A aneae p eda o y 0.47 0.6260 1.95 0.1213 1.16 0.3255 0.64 0.6975 1.05 0.3911 0.93 0.5025 4.13 0.0429 Collembola de i i o ous 0.97 0.3785 11.91 <0.0001 3.14 0.0253 2.68 0.0146 0.29 0.9404 0.75 0.6660 10.39 0.0014 Coleop e a de i i o ous 0.16 0.8514 23.63 <0.0001 3.10 0.0268 1.95 0.0716 0.31 0.9322 2.51 0.0084 0.07 0.7964 p eda o y 2.67 0.0708 18.81 <0.0001 1.28 0.2792 0.68 0.6669 1.60 0.1455 1.77 0.0730 2.85 0.0923 Table 1 In e eb a e axa collec ed om he expe imen al plan s Taxon Numbe o indi iduals Feeding guild Dip e a 1393 he bi o ous 704 de i i o ous 328 omni o ous 25 p eda o y 3 pa asi ic Hymenop e a 46 he bi o ous 606 pa asi ic Collembola 8360 de i i o ous Hemip e a 197 he bi o ous 51 p eda o y Homop e a 37 he bi o ous Coleop e a 28 he bi o ous 379 p eda o y 589 de i i o ous A aneae (A achnida) 281 p eda o y Aca i (A achnida) 4017 omni o ous / pa asi ic Thysanop e a 62 (guild no iden i ied) Fungal Di e si y (2011) 47:109–118 111 Field expe imen To es he ole o endophy e in ec ion, plan geog aphic o igin and en i onmen al ac o s, a common ga den ield expe imen was es ablished a Bo anical Ga den, Uni e - si y o Tu ku, Finland in 2004. The s udy si e is a he edge o he no he n dis ibu ion ange o na u al all escue popula ions and has been in cul i a ion in he pas . I was illed in he summe 2004 wi hou nu ien applica ion. The expe imen al a ea was enced o p e en la ge e eb a es (e.g., abbi s, dee ) om b owsing he plan s. Howe e , smalle e eb a es (e.g., oles) and in e eb a es we e allowed o eely access he a ea. The space be ween expe imen al plan s was ei he mowed, hand weeded o sp ayed wi h he bicide wo imes du ing he g owing season o p e en in e speci ic compe i ion in he ield. The expe imen al design was a andomized block consis ing o 10 blocks, each di ided o 4 plo s. The ou plo s in each block we e andomly designa ed o one o he ou ea men s: (i) con ol (C) ecei ing only ambien wa e and nu ien s, (ii) wa e ea men (W) wi h 3 li es o wa e applied o each plan sepa a ely h ee imes a week om June o Augus , (iii) nu ien ea men (N) whe e 1dl o N-P-K- e ilize (Nu men Y2, Kemi a KnowHow,[N-P-K/20-6-6])/plan was applied wo imes du ing he g owing season, and (i ) wa e – nu ien ea men (WN) combining bo h wa e and nu ien applica ions. The ea men s we e applied du ing he pe iod o 2005–2006. Tall escue plan s wi h 2-3 ille s we e plan ed in Augus 2004 abou 0.5 me e s apa om each o he and om he edge o he plo . Fo y plan s om each o igin (na u al popula ions A = Åland, G = Go land, and S = coas al Sweden; cul i a s K = “Ken ucky 31”) and in ec ion s a us (E+, E-, ME-) we e andomly chosen. Thus, he e we e 12 plan s in each o he 40 plo s o a o al o 480 plan s used in he p esen s udy. The in ec ion s a us o all indi idual plan s was con i med in 2006 ia seed s aining (Saha e al. 1988). The biomass o he abo e-g ound plan pa s was emo ed, d ied and weighed in au umn a he end o he g owing season 2006. Collec ion and iden i ica ion o in e eb a es In e eb a es we e collec ed om each plan indi idual wi h an Insec Vo is Vaccuum® sample (Bu ka d L d., UK) in July 2006. E e y plan was acuumed in he same way o 10 s om he middle o he plan . The samples we e placed in o eclosable plas ic bags and ozen immedia ely a e sampling. In e eb a es we e hen la e coun ed, iden i ied o amily le el unde a mic oscope, and assigned o he ollowing i e eeding guilds based on he key amily and species cha ac e is ics in li e a u e: he bi o es, omni o es, de i i o es, p eda o s and pa asi oids (Table 1). S a is ical analyses We used ANCOVA (wi h plan biomass as a co a ia e) in he Mixed model p ocedu e o SAS s a is ical so wa e (SAS U ili ies 9.1) o analyze he e ec s o endophy e s a us (E+, E-, and ME-), wa e and nu ien ea men s (W, N, WN, and C), plan o igin (A, G, K, S), and block (1–10) on he abundances o in e eb a es in he eeding guilds (he bi o es, omni o es, de i i- o es, p eda o s and pa asi oids) and axonomical g oups wi h su icien sample sizes o meaning ul s a is ical analyses (he bi o ous, de i i o ous, omni o- ous, pa asi ic, and p eda o y dip e ans; Aca i; Hyme- nop e ans; spide s; collembolas, and de i i o ous and p eda o y Coleop e ans) T ea men , o igin, endophy e, and hei possible in e ac ions we e conside ed o be ixed ac o s in all models, whe eas block was a andom ac o . Plan biomass was used as a co a ia e, because plan size may in luence in e eb a e abundan- Table 3 The e ec s o endophy e s a us (E+ = endophy e in ec ed, E- = endophy e- ee, and manipula i ely endophy e- ee = ME-), wa e and nu ien ea men s (C = con ol, N = nu ien , W = wa e , and WN = wa e + nu ien ), plan o igin (A = Åland, G = Go land, and S = coas al Sweden; K = cul i a “Ken ucky 31”) and plan biomass on abundances o he bi o es, de i i o es and p eda o s He bi o es De i i o es Omni o es Pa asi oids P eda o s d Fp F p Fp Fp F p Endophy e s a us (E) 2 0.35 0.7036 0.80 0.4484 0.29 0.8330 2.14 0.1192 2.31 0.1007 T ea men (TRT) 3 3.10 0.0268 15.05 <0.0001 0.71 0.5471 0.63 0.5987 15.38 <0.0001 Plan o igin (PO) 3 1.61 0.1870 3.99 0.0080 0.52 0.5932 4.59 0.0036 1.04 0.3730 E * TRT 6 2.62 0.0169 2.63 0.0165 0.50 0.8089 0.55 0.7674 0.68 0.6681 E * PO 6 0.74 0.6199 0.26 0.9565 0.87 0.5156 0.75 0.6119 1.04 0.3987 TRT * PO 9 1.94 0.0449 0.72 0.6885 0.44 0.9142 1.46 0.1591 1.45 0.1662 Plan biomass 1 9.67 0.0020 10.28 0.0015 0.04 0.8338 0.78 0.3781 3.22 0.0734 112 Fungal Di e si y (2011) 47:109–118 ces. Plan size was signi ican ly inc eased by wa e ing and e iliza ion (d =3, F=17.07, p<0.0001)(C: mean= 395g,SE=16.4;N:mean=414g,SE=22.1;W:mean= 422 g, SE=15.2; WN: mean=587 g, SE=24.2) excep in he case o he K- 31 cul i a . Resul s on plan g ow h and pe o mance will be epo ed and discussed in mo e de ail elsewhe e. The e ec s o endophy e s a us (E+, E-, and ME-), wa e and nu ien ea men s (W, N, WN, and C), plan o igin (A, G, K, S) and plan biomass on axonomic in e eb a e di e si y we e examined in wo ways. Fi s , we es ed he e ec s o he explana o y ac o s and hei in e ac ions on species numbe s and he Shannon di e si y index by a mixed model analysis o co a iance Table 4 Means and s anda d e o s (SE) o axonomic g oups o in e eb a es showing s a is ically signi ican (a) in e ac i e e ec s o wa e and nu ien ea men s (C = con ol, N = nu ien , W = wa e , and WN = wa e + nu ien ) and endophy e s a us (E+ = endophy e in ec ed, E- = endophy e- ee, and manipula i ely endophy e- ee = ME-), (b) e ec s o plan o igin (A = Åland, G = Go land, and S = coas al Sweden; K = cul i a “Ken ucky 31”) and (c) in e ac i e e ec s o wa e and endophy e s a us (see Table 2) Taxon a He bi o ous Dip e a Omni o ous Dip e a Collembola T ea men Endophy e s a us n mean SE mean SE mean SE C E+ 39 2.7 2.7 1.2 0.37 9.4 1.76 E- 39 3.4 3.4 0.5 0.14 10.2 2.03 ME- 40 3.7 3.7 0.6 0.12 11.7 2.54 W E+ 39 3.2 3.2 0.7 0.15 20.7 3.27 E- 40 2.6 2.6 0.6 0.13 14.3 2.31 ME- 39 2.1 2.1 0.8 0.25 11.4 1.81 N E+ 32 2.4 2.4 0.6 0.14 21.8 3.36 E- 37 2.4 2.4 0.5 0.13 28.7 5.10 ME- 34 3.6 3.6 0.6 0.13 25.9 3.66 WN E+ 38 3.9 3.9 0.7 0.18 33.7 6.22 E- 34 4.6 4.6 1.6 0.36 18.8 3.87 ME- 34 3.3 3.3 0.5 0.14 22.0 3.80 b De i i o ous Dip e a Hymenop e a Collembola Coleop e a Plan o igin n mean SE mean SE mean SE mean SE Åland 118 1.9 0.24 1.8 0.17 24.3 2.34 1.9 0.33 Go land 113 1.65 0.19 1.2 0.17 17.7 2.06 1.2 0.26 K-31 99 1.1 0.14 0.96 0.12 13.9 1.78 0.8 0.14 Sweden 115 1.6 0.17 1.4 0.13 18.4 2.10 1.3 0.18 c He bi o ous Dip e a De i i o ous Dip e a Coleop e a T ea men Plan o igin n mean SE mean SE mean SE C Åland 30 2.8 0.47 1.0 0.25 0.3 0.10 Go land 29 3.3 0.60 1.2 0.25 0.3 0.11 K-31 29 3.1 0.44 0.9 0.20 0.4 0.15 Sweden 30 3.6 0.32 1.0 0.26 0.4 0.12 W Åland 28 2.9 0.53 1.8 0.39 0.5 0.15 Go land 30 1.9 0.31 2.0 0.37 0.4 0.09 K-31 30 2.7 0.45 1.0 0.25 0.5 0.16 Sweden 30 3.1 0.64 1.6 0.35 0.7 0.22 N Åland 30 2.9 0.47 1.1 0.22 2.2 0.58 Go land 26 2.8 0.40 1.2 0.31 1.7 0.40 K-31 19 2.6 0.63 1.1 0.27 1.7 0.45 Sweden 28 2.8 0.44 1.3 0.27 1.7 0.33 WN Åland 30 6.1 0.76 3.9 0.72 4.5 1.00 Go land 28 3.6 0.65 2.2 0.52 2.7 0.89 K-31 21 2.2 0.71 1.4 0.38 1.0 0.33 Sweden 27 3.3 0.71 2.6 0.37 2.4 0.53 Fungal Di e si y (2011) 47:109–118 113 (ANCOVA) wi h plan biomass as a co a ia e, using he Mixed p ocedu e o SAS s a is ical so wa e (SAS U ili ies 9.1). The plan -speci ic Shannon index alue (H’) was calcula ed as ollows: H0¼PipilnðpiÞ whe e p i is he p opo ion o indi iduals in he i he axonomical g oups in he expe imen al plan s. Compa ed o species numbe o ichness, he ad an age o he Shannon index is ha i inco po a es he numbe o axonomical g oups and hei e enness. Second, o examine he amoun o a ia ion (%) ha endophy e s a us, wa e and nu ien ea men s and plan o igin explained in he in e eb a e communi y composi ion, we used a pa ial Canonical Co espondence Analysis CCA (Bo ca d e al. 1992) wi h CANOCO 4 so wa e (Te B aak and Šmilaue 1998). Only he a ia ion explained by s a is ically signi ican en i onmen al a iables was pa i- ioned (Økland 1999). The de aul op ions o CANOCO (excep log x + 1 da a ans o ma ion and downweighing o a e species) we e used. The signi icance o he i s CCA axis and he CCA model, as well as each en i onmen al a iable was e alua ed by Mon e Ca lo pe mu a ion es s (500 pe mu a ions) in all analyses. Nu ien and wa e ea men s along wi h plan biomass appea ed o be signi ican (p<0.01) in CCA. Resul s and discussion Recen li e a u e indica es ha ungal endophy es al e in e eb a e communi ies in bo h ag onomic and wild g ass popula ions (Rudge s and Clay 2007; Ben ey and Denno 1997; Fae h and Shocha 2010; Ha ley and Gange 2009; Jani e al. 2010; Lemons e al. 2005; Omacini e al. 2001; Saa i e al. 2010). Howe e , he Neo yphodium endophy e in ec ion ailed o in luence he abundances o he axa o eeding guild, o o e all axonomic di e si y and compo- si ion o he in e eb a e communi y in ou expe imen al common ga den s udy o wild all escue plan s collec ed ac oss he no he n dis ibu ion ange o he species and he well s udied all escue K-31 cul i a . Ins ead, he in e e- b a e communi y o na i e all escue in his expe imen appea s o be p ima ily d i en by en i onmen al condi ions in e ac ing wi h plan geog aphic o igin. In e eb a e abundance and communi y composi ion A o al o 18650 in e eb a es we e collec ed and iden i ied o amily le el om he expe imen al plan s. Sp ing ails (Collembola), mi es (Aca i), and lies and midges (Dip e a) comp ised 48%, 23% and 14% o he indi idual in e eb a es, espec i ely (a o al o 85%) (Table 1). The es 15% o he in e eb a es we e Coleop e ans (6%), Hymenop e ans (4%), spide s (2%), and Hemip e ans (2%). Only one pe cen age o species emained uniden i ied. 56% and 24% o he in e eb a e communi y consis ed o de i i o es and pa a- si oids, espec i ely, because o he high numbe o de i i o ous Collembola and Aca i mi es and pa asi ic Hymenop e ans in ou samples (Table 1). Only 10% o all in e eb a es we e he bi o es, bu his eeding guild was axonomically he mos di e se comp ising o 42 iden i ied axa. E+ plan s did no di e om E- and ME- plan s in he abundance o any axonomic in e eb a e g oup (Table 2) o eeding guild (Table 3). Howe e , endophy e in ec ion a ec ed he abundance o he bi o ous and omni o ous dip e ans, and collembolas in e ac i ely wi h wa e and nu ien ea men s. Fo example, he abundance o he bi o ous dip e ans was highe on wa e ed and e ilized E- and E+ plan s compa ed o he o he ea men and in ec ion combina ions, whe eas he abundance o omni - o ous dip e ans was highes on wa e ed and e ilized E- plan s, second highes on un ea ed E+ plan s, and lowes on e ilized E- plan s (Table 4a). In con as o dip e ans, de i i o ous Collembola (sp ing ails) we e much mo e ab CWN Numbe o he bi o es 0 1 2 3 4 5 6 7 E+ E- ME- CWNWN WN Numbe o de i i o es 0 10 20 30 40 50 E+ E- ME- Fig. 1 The e ec s o endophy e s a us (E+, E-, and ME-) and wa e and nu ien ea men s (W, N, WN, and C) on he o al numbe o he bi o es (a) and de i i o es (b) 114 Fungal Di e si y (2011) 47:109–118 abundan and appea ed o p e e wa e ed and e ilized E+ plan s (Table 4a; see also Fae h and Shocha 2010). Likewise, he o al numbe o he bi o es and de i i o es did no show a common end o p e e ence o a oidance o E+ o E- plan s in ei he low o high nu ien en i onmen s (Table 2,Fig.1). Plan o igin signi ican ly a ec ed he abundances o de i i o ous Dip e a, Hymenop e a, Collembola and Cole- op e a (Table 2), as hei mean abundances was highes on plan s collec ed om Åland and lowes on he cul i a “Ken ucky 31”in all g oups (Table 4b). In he cases o Coleop e a and bo h he bi o ous and de i i o ous Dip e a, abundances a ied among plan o igins in e ac i ely wi h wa e and nu ien ea men (Table 2), bu we e highes on plan s om Åland and lowes on he K-31 when he plan was wa e ed and e ilized (Table 4c). This indica es di e ences in esis ance among plan geno ypes in di e en en i onmen s. Plan size appea s o be posi i ely ela ed o in e eb a e abundance. Plan biomass explained signi ican ly he numbe s o he bi o ous, de i o ous and pa asi ic dip e ans, spide s (A aneae), and mi es (Aca i) (Table 2), and he abundances o hese axa we e posi i ely co ela ed wi h plan size excep in he case o pa asi ic dip e ans (he bi o ous Dip e a: n=445, =0.21, p=<0.0001; de i i o ous: n=445, =0.26, p=<0.0001; pa asi ic Dip e a: n=445, =0.06, p=<0.2035; Collembola: n=445, =0.24, p=<0.0001; A aneae: n=445, =0.13, p=0.0074). Likewise, he o al numbe o bo h he bi o es and de i i o es posi i ely co ela ed wi h plan biomass (He bi o es: n=445, =0.22, p=<0.0001; de i i o es: n=445, =0.26, p=<0.0001). In e eb a e ichness In e eb a e ichness ollowed he same ends as axonomic g oups. Nei he he numbe o axa no Shannon di e si y index a ied by he in ec ion s a us o he plan (Table 5). Ins ead, in e eb a e ichness was posi i ely co ela ed wi h plan size (plan biomass— he numbe o axa: n=444, = 0.27, p=<0.0001; plan biomass— he Shannon di e si y index: n=444, =0.15, p=0.0016) and i was signi ican ly highe on e ilized plan s (Table 5,Fig.2). In e eb a e communi y s uc u e Canonical Co espondence Analysis (CCA) sugges s ha in e eb a e communi y well mi o s abio ic en i onmen al condi ions and he size o he plan . Mos o he a ia ion in he axonomical composi ion was highly dependen on nu ien (Axis 1 in Fig. 3a) and wa e (Axis 2 in Fig. 3a) a ailabili y in he soil. The sum o all canonical eigen alues was 0.131. The i s axis explained 3.2% o axon a ia ion and 57.6% o he a ia ion o he axon-en i onmen ela ionship. In he Mon e Ca lo es , he signi icance o he i s axis was P=0.002 (F=14.2) and o all axes P= 0.002 (F=2.8). T ea men explained 73.3% o he a ia ion, ab CWNWNNW Numbe o axa 0 2 4 6 8 10 12 CNW Shannon H' 0,0 0,2 0,4 0,6 0,8 1,0 1,2 1,4 1,6 1,8 Fig. 2 Mean (±SE) numbe o axa (a) and he Shannon di e si y index in wa e and nu ien ea men s d Taxon numbe Shannon di e si y index Fp F p Endophy e s a us (E) 2 0.88 0.4148 0.37 0.6931 T ea men (TRT) 3 11.05 <0.0001 6.07 0.0005 Plan o igin (PO) 3 1.52 0.2086 0.80 0.4923 E * TRT 6 1.95 0.0714 0.60 0.7268 E * PO 6 1.25 0.2815 1.29 0.2605 TRT * PO 9 1.12 0.3456 1.03 0.4159 Plan biomass 1 12.23 0.0005 4.38 0.0369 Table 5 The e ec s o endophy e s a us (E+ = endophy e in ec ed, E- = endophy e- ee, and manipula i ely endophy e- ee = ME-), wa e and nu ien ea - men s (C = con ol, N = nu ien , W = wa e , and WN = wa e + nu ien ), plan o igin (A = Åland, G=Go land,andS=coas al Sweden; K = cul i a “Ken ucky 31”) and plan biomass on iden i ied axon numbe s and he Shannon di e si y index Fungal Di e si y (2011) 47:109–118 115 whe eas he p opo ion o he o he ac o s emained smalle (plan o igin 9.9%, endophy e s a us 7.6%, plan biomass 6.9%) and s a is ically insigni ican (C: F=7.0, P= 0.002; W: F=5.5, P=0.002; N: F=8.1, P=0.002; NW: F= 3.8, P=0.002; Biomass o he plan : F=1.986, P=0.002; E+: F=1.161, P=0.2196; E-: F=0.815, P=0.7884; ME-: F= 0.955, P=0.5250; A: F=1.083, P=0.3593; G: F=0.902, P= 0.6727; S: F=0.729, P=0.9022; K: F=0.884, P=0.6966). Howe e , he e was no common s uc u e in he in e e- b a e communi y ela ed o endophy e s a us, plan o igin o wa e and nu ien ea men s ac oss he axonomical g oups o eeding guilds (Fig. 3). In con as , CCA indica es ha each axonomical g oup in a eeding guild appea s o p e e di e en en i onmen s and he axonomical di e si y o a eeding guild is in ela ion o specializa ion in esou ce use acco ding o he idea o niche pa i ioning (El on 1927;Hu chinson1961;Richa dse al.2000;Diehl2003). Fo example, he he bi o e eeding guild was axonomically mos di e se (42 axa), bu he place o he bi o e axa in he expe imen al wa e and nu ien en i onmen s we e no iden ical (Fig. 3b) In o he wo ds, he species clea ly do no occupy exac ly he same hos ype. ab cd e Fig. 3 Canonical Co espon- dence Analysis (CCA) o he ela ionship be ween axonomical g oups and examined bio ic (endophy e s a us o he plan , plan o igin and plan biomass) and abio ic (wa e and nu ien ea men s) en i onmen al ac o s. Signi ican en i onmen al a iables (a) (W = wa e , N = ni ogen, WN = wa e and ni ogen, C = con ol) and plan biomass (BIOM) a e shown wi h i e axonomical in e eb a e g oups: he bi o es (b), de i i o es (c), omni o es (d), pa asi oids (e) and p eda o s ( ). Eigen alue o he i s axis was 0.171 and o he second axis 0.056 116 Fungal Di e si y (2011) 47:109–118 Conclusions Ou esul s demons a e ha (1) he axonomical di e si y and complexi y o an in e eb a e communi y can be e y high e en in ela i ely simple plan communi ies, and (2) he di e si y is commensu a e wi h p ima y p oduc ion and en i onmen al ac o s ha in e ac wi h plan o igin a he han endophy e in ec ions. Fu he mo e, in e eb a e commu- ni y, pa icula ly he mos di e se eeding guild, he bi o es, showed s ong di e en ia ion along he examined wa e and nu ien g adien s. This may d i e he communi y s uc u e o in e eb a e he bi o es in a pa chy en i onmen . The lack o inc eased o dec eased he bi o e esis ance migh be pa ly explained by he ac ha alkaloids in na i e Eu opean all escue a e no o he ype o le el ha educe (A khami and Rudge s 2009) o p omo e (Fae h and Shocha 2010;Janie al. 2010) plan eeding in e eb a es. Howe e , such di e - ences in alkaloid p o iles and o he plan cha ac e is ics due o di e ences among plan o endophy e geno ypes ails o explain he lack o axon, eeding guild and communi y le el esponses wi h he cul i a K-31. We p opose ha empi ical whole-communi y app oaches a e equi ed o unde s and he impo ance o endophy es and o he mechanisms d i ing plan popula ions and in e eb a e communi ies eeding on hem. Accumula ing e idence om endophy e media ed in e ac ions has e ealed ha endophy es can nega i ely a ec plan eeding he bi o es (Saikkonen e al. 2010). Howe e , he accumula ing e idence also indica es ha di e si y in esul s and in e p e a ions o he gene al impo ance o endophy es in g assland communi ies inc eases as new model sys ems appea . Cu en li e a u e appea s o be s ongly biased by wo model species, all escue and pe ennial yeg ass and hei ew cul i a s such as K-31, in in oduced and ag onomic en i onmen s, and his has dis ac ed he li e a u e (Saikkonen e al. 2006,2010). By using wild all escues in hei na i e con inen , we we e able o show ha en i onmen al condi ions and hos plan o igin o e ide endophy e e ec s on in e eb a e di e si y, com- muni y s uc u e, and eeding guilds. Acknowledgemen s This s udy was unded by he Academy o Finland (P ojec no. 110658). Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion Noncomme cial License which pe mi s any noncomme cial use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho (s) and sou ce a e c edi ed. Re e ences A khami ME, Rudge s JA (2009) Endophy e-media ed esis ance o he bi o es depends on he bi o e iden i y in he wild g ass Fes uca sub e icilla a. En i on En omol 38:1086–1095 Ahlholm JU, Helande M, Leh imäki S, Wäli P, Saikkonen K (2002) Ve ically ansmi ed ungal endophy es: di e en esponses o hos - pa asi e sys ems o en i onmen al condi ions. Oikos 99:173–183 Bacon CW (1995) Toxic endophy e-in ec ed all escue and ange g asses: his o ic pe spec i es. J Anim Sci 73:861–870 Bacon CW, Po e JK, Robbins JD, Lu ell ES (1977) Epichloë yphina om oxic all escue g asses. App En i on Mic ob 34:576–581 Ball DM, Pede sen JF, Lace ield GD (1993) The all- escue endophy e. E olu ion mee s economics in he ale o he na ion’s mos popula plan ed g ass, which owes many o i s quali ies o a ungus oxic o li es ock. Am Sci 81:370–379 Bazely DR, Vica i M, Emme ich S, Filip L, Lin D, Inman A (1997) In e ac ions be ween endophy e-in ec ed Fes uca ub a om he Sco ish islands o S Kilda, Benbecula and Rum. J Appl Ecol 34:847–860 Ben ey B, Denno RF (1997) The slow-g ow h-high-mo ali y hypo hesis: a es using he cabbage bu e ly. Ecology 78:987–999 Bony S, Pichon N, Ra el C, Du ix A, Bal ou ie C, Guillaumin J-J (2001) The ela ionship be ween myo oxin syn hesis in ungal endophy es o Lolium pe enne. New Phy ol 152:125–137 Bo ca d D, Legend e P, D apeau P (1992) Pa ialling ou he spa ial componen o ecological a ia ion. Ecology 73:1045–1055 B een JP (1994) Ac emonium endophy e in e ac ions wi h enhanced plan esis ance o insec s. Annu Re En omol 39:401–423 Cheplick GP, Fae h SH (2009) Ecology and e olu ion o he g ass- endophy e symbiosis. Ox o d Uni e si y P ess Cheplick GP, Clay K, Ma ks S (1989) In e ac ions be ween in ec ion by endophy ic ungi and nu ien limi a ion in he g asses Lolium pe enne and Fes uca a undinaceae. New Phy ol 111:89–97 Clay K (1988) Fungal endophy es o g asses: a de ensi e mu ualism be ween plan s and ungi. Ecology 69:10–16 Clay K (1989) Cla icipi aceous endophy es o g asses: hei po en ial as biocon ol agen s. Mycol Res 92:1–12 Clay K (1990) Fungal endophy es o g asses. Annu Re Ecol Sys 21:275–297 Clay K, Scha dl CL (2002) E olu iona y o igins and ecological consequences o endophy e symbiosis wi h g asses. Am Na 160:99–127 Clay K, Ha dy TN, Hammond AM J (1985) Fungal endophy es o g asses and hei e ec s on an insec he bi o e. Oecologia 66:1–6 Clay K, Ma ks S, Cheplick GP (1993) E ec s o insec he bi o y and ungal endophy e in ec ion on compe i i e in e ac ions among g asses. Ecology 74:1767–1777 Diehl S (2003) The e olu ion and main enance o omni o y: dynamic cons ain s and he ole o ood quali y. Ecology 84:2557–2567 El on CS (1927) Animal ecology. Sidgwick and Jackson, London Fae h SH (2002) A e endophy ic ungi de ensi e plan mu ualis s? Oikos 99:200–200 Fae h SH, Bul man TL (2002) Endophy ic ungi and in e ac ions among hos plan s, he bi o es and na u al enemies. In: Tscha n ke T, Hawkins BA (eds) Mul i ophic le el in e ac ions. Camb idge Uni e si y P ess, pp 89–123 Fae h SH, Fagan WF (2002) Fungal endophy es: common hos plan symbion sbu uncommon mu ualis s. In eg Comp Biol 42:360–368 Fae h SH, Saikkonen K (2007) Va iabili y is he na u e o he endophy e-g ass in e ac ion. P oceedings o he 6 h In e na ional Symposium on Fungal Endophy es o G asses “F om Lab o Fa m”. Ch is chu ch, New Zealand Ma ch 25–28, 2007, Popay AJ, Thom ER (eds) G assland Resea ch and P ac ice Se ies No. 13. New Zealand G assland Associa ion, Dunedin, NZ, ISSN 0110-8581 Fae h SH, Shocha E (2010) Inhe i ed mic obial symbion s inc ease he bi o e abundances and al e a h opod di e si y on a na i e g ass. Ecology (in p ess) Fungal Di e si y (2011) 47:109–118 117