ORIGINAL RESEARCH ARTICLE
published: 17 Sep embe 2014
doi: 10.3389/ pls.2014.00478
Silicon, endophy es and seconda y me aboli es as g ass
de enses agains mammalian he bi o es
O so Hui u1*, K is ian M. Fo bes1, Ma jo Helande 2, Rii a Julkunen-Tii o3, Xa ie Lambin4,
Ka i Saikkonen 5, Pe e S ua 1†, Sini Sulkama1† and Sue Ha ley6
1Suonenjoki Resea ch Uni , Finnish Fo es Resea ch Ins i u e, Suonenjoki, Finland
2Sec ion o Ecology, Uni e si y o Tu ku, Tu ku, Finland
3Depa men o Biology, Uni e si y o Eas e n Finland, Joensuu, Finland
4School o Biological Sciences, Uni e si y o Abe deen, Abe deen, UK
5Plan P o ec ion, Ag i ood Resea ch Finland, Jokioinen, Finland
6Depa men o Biology, Uni e si y o Yo k,Yo k, UK
Edi ed by:
Jane DeGab iel, Uni e si y o Wes e n
Sydney, Aus alia
Re iewed by:
Daniel J. Ballho n, Po land S a e
Uni e si y, USA
Ca oline S ol e , Uni e si y o
Hambu g, Ge many
*Co espondence:
O so Hui u, Suonenjoki Resea ch
Uni , Finnish Fo es Resea ch
Ins i u e, Jun in ie 154, FI-77600
Suonenjoki, Finland
e-mail: o so.hui u@me la.fi
†P esen add ess:
Pe e S ua , Depa men o Bo any
and Zoology, Masa yk Uni e si y,
B no, Czech Republic;
Sini Sulkama, Depa men o
Molecula Neu ology, Uni e si y o
Helsinki, Finland
G asses ha e been conside ed o p ima ily employ ole ance in lieu o de ense in mi iga ing
damage caused by he bi o y.Ye a numbe o mechanisms ha e been iden ified in g asses,
which may de e eeding by g aze s. These include enhanced silicon up ake, hos ing o
oxin-p oducing endophy ic ungi and induc ion o seconda y me aboli es. While hese
mechanisms ha e been indi idually s udied, hei syne gis ic esponses o g azing, as
well as hei e ec s on g aze s, a e poo ly known. A field expe imen was ca ied ou
in5×5 m ou doo enclosu es o quan i y phy ochemical changes o ei he endophy e-
in ec ed (E+) o endophy e- ee (E−) meadow escue (Schedono us p a ensis) in esponse
o medium in ensi y (co esponding wi h densi ies o ca. 1200 oles/ha o 5 weeks du ing
3 mon hs) o hea y in ensi y (ca. 1200 oles/ha o 8 weeks du ing 3 mon hs) g azing by
a mammalian he bi o e, he field ole (Mic o us ag es is). A labo a o y expe imen was
hen conduc ed o e alua e he e ec s o endophy e in ec ion s a us and g azing his o y o
he g ass die on ole pe o mance. As p edic ed, g azing inc eased olia silicon con en ,
by up o 13%. G azing also inc eased olia le els o phospho ous and se e al phenolic
compounds, mos no ably hose o he fla onols iso hamne in-diglycoside and hamne in
de i a i e. Silicon concen a ions we e consis en ly ci ca 16% highe in E+g asses han
in E-g asses, a all le els o g azing. Simila ly, concen a ions o chlo ogenic acid de i a i e
we e ound o be consis en ly highe in E+ han in E−g asses. Female oles main ained on
hea ily g azed g asses su e ed highe mo ali y a es in he labo a o y han emale oles
ed ung azed g ass, ega dless o endophy e in ec ion s a us. Ou esul s conclusi ely
demons a e ha , in addi ion o ole ance, g asses employ mul i- ie ed, e ec i e de enses
agains mammalian g aze s.
Keywo ds: de ense, endophy es, g asses, g azing, phenolics, seconda y me aboli es, silicon, oles
INTRODUCTION
G asses a e highly ole an o g azing by means o hei apid
eg ow h capaci y, basal me is ems, unde g ound s o age o gans
and ille ing capaci y (Dye e al., 1991;Ka ban and Baldwin,
2007). Indeed, g asses ha e been conside ed o p ima ily ely
on hese ole ance ai s in lieu o physical o chemical de enses
o mi iga e damage caused by he bi o y. Howe e , a numbe o
mechanisms ha e been iden ified in g asses ha may de e eeding
by g aze s, and hence ac as de enses (Vica i and Bazely, 1993).
The mos p ominen o hese is enhanced silicon (Si) up ake in
esponse o damage (McNaugh on and Ta an s, 1983;Massey
e al., 2007a).
Following he bi o e damage, Si is d awn om he soil and
deposi ed sys emically in he cell walls o lumina o new shoo s
as silica bodies o phy oli hs (Ma and Yamaji, 2006;Cu ie and
Pe y, 2007). These o ms o Si a e ex emely ha d, and he eby
inc ease he ab asi eness o lea ma e ial (Massey and Ha ley,
2006), po en ially leading o accele a ed oo h wea in g aze s.
High Si le els also inhibi he abso p ion o ni ogen om diges ed
plan ma e ial by mammalian he bi o es (Massey and Ha ley,
2006).
G aze s ha e o en been obse ed o a oid plan ma e ial wi h
high Si con en (Gali-Muh asib e al., 1992;Co e ill e al., 2007;
Reynolds e al., 2012). Howe e , he e ec s o Si on mammalian
he bi o e eeding beha io appea in many cases o be mode a ed
o e en o e - idden by a ange o o he ac o s including species
iden i y (o bo h g ass and he bi o e), geno ype o g owing en i-
onmen and o he aspec s o o age quali y (Shewmake e al.,
1989;Massey e al., 2009;Soininen e al., 2013a).
Many g ass species ha bo sys emic, e ically ansmi ed
endophy ic ungi, which may also inc ease he esis ance o he
hos o he bi o y (Clay, 1988;Cheplick and Clay, 1988;Saikko-
nen e al., 1998,2006). Endophy ic ungi a e o en p oduce s o
myco oxins, which ei he educe he pala abili y o he g ass o
ende i oxic o he he bi o e (Clay, 1988;Powell and Pe oski,
1993;Siegel and Bush, 1996;Saikkonen e al., 1998,2006). Bo h
www. on ie sin.o g Sep embe 2014 |Volume 5 |A icle 478 |1
Hui u e al. G ass de enses agains mammalian he bi o es
he ungal load o g asses and he amoun o oxins hey p oduce
may inc ease in esponse o he bi o y, and hence ac as induced
de enses (Bazely e al., 1997).
While ad e se e ec s o endophy e consump ion a e well doc-
umen ed o domes ic li es ock ( e iewed by Clay, 1988), esul s
oms udiesonwildmammaliang aze sa esca ceandambiguous
a bes . Some s udies epo nega i e e ec s, such as dec eases in
popula ion densi y (Coley e al., 1995), inc eased oxici y-induced
mo ali y (Cono e , 1998) and supp ession o ep oduc ion and
g ow h (Jackson e al., 1996;Du ham and Tannenbaum, 1998;
Fo ie e al., 2000;Cono e , 2003), while o he s ha e ound no
e ec sa all(Ba ge and Tannenbaum, 1998;Saa i e al., 2010).
Thus, he e ec s o endophy ic ungi on he bi o es may be dose-
dependen and may a y be ween g ass, ungi and he bi o e
species.
G asses a e known o p oduce an a ay o seconda y me abo-
li es, o example hyd oxamic acids (Niemeye , 1988), condensed
annins (Be nays e al., 1989), cyanogenic glycosides (Jones, 1998)
and alkaloids, albei a le els much lowe han dico yledons
(Coughenou , 1985;Vica i and Bazely, 1993). Seconda y me abo-
li es ound in g asses ha e been shown o ha e ad e se e ec s
on he pe o mance o oden s when consumed in a ificial die s
(Lind o h and Ba zli, 1984). Howe e , only a ely ha e hese com-
pounds had measu able nega i e impac s on ee- anging mam-
malian he bi o es a he concen a ions in which hey na u ally
occu in g asses. Excep ions again in ol e li es ock: cyanogenic
glycosides poisoning ca le (Geo giadis and McNaugh on, 1988)
and indole alkaloids educing he pala abili y o g ass o sheep
(Simons and Ma en, 1971;Ma en e al., 1973).
Mos s udies ha e ailed o iden i y significan e ec s o g azing
on le els o de ensi e seconda y me aboli es in g asses (Lind o h
and Ba zli, 1986;Klemola e al., 2000). None heless, g amini -
o ous he bi o es a e highly selec i e eede s bo h wi hin and
be ween plan species (F eeland, 1974;Hjäl én e al., 1996). As
his beha io is o a g ea ex en go e ned also by he seconda y
chemis y o hei oodplan s(Jung andBa zli,1981;Be ge onand
Joudoin, 1989;Massey e al., 2007b), a ia ion in g ass seconda y
me aboli e concen a ions holds he po en ial o influence g az-
ing mammals and hei popula ion demog aphy (F eeland and
Janzen, 1974).
Al hough hese h ee po en ial de ensi e mechanisms o
g asses – Si induc ion, endophy es, and seconda y me aboli es
– ha e been indi idually s udied, hei syne gis ic esponses o
g azing and in e ac i e e ec s on he bi o es a e poo ly known.
The aim o his s udy was o simul aneously quan i y how le els
o Si, o he nu ien s and seconda y me aboli e con en in e ac in
g asses in esponse o mammalian g azing and endophy e in ec-
ion s a us. Fu he mo e, he aim was o e alua e he syne gis ic
e ec o hese po en ial de ense mechanisms on he physiologi-
cal pe o mance o a g amini o ous he bi o e. Al hough all o
he g ass de enses ocussed on in his s udy a e mechanisms
h ough which he bi o e popula ions may po en ially be a ec ed,
he e idence o impac s o mechanisms in na u al popula ions is
missing.
Meadow escue (Sche odonus p a ensis ex. Lolium p a ense and
Fes uca p a ensis), which is commonly in ec ed by he endophy ic
ungus Epichloë uncina a (Saa i e al., 2009) was used as a model
g ass species. As an indica o o seconda y me aboli e p oduc ion,
we analyzed le els o phenolics, a widesp ead g oup o impo -
an de ensi e compounds (Rhoades,1979) known o ha e ad e se
e ec s on g amini o ous he bi o es (Lind o h and Ba zli, 1984).
A common Eu opean g assland oden , he field ole Mic o us
ag es is se ed as ou model g azing he bi o e.
The ollowing specific hypo heses we e es ed: (1) ole g azing
ele a es le els o bo h Si and phenolics in g asses, (2) he mag-
ni ude o esponse o Si and phenolics o g azing di e s be ween
endophy e-in ec ed and non-in ec ed g asses, and (3) consump-
ion o hea ily g azed g ass, as compa ed o consump ion o less
hea ily g azed g asses, has nega i e impac s on ole physiological
condi ion, and mo e so i he g ass is endophy e-in ec ed.
MATERIALS AND METHODS
STUDY SPECIES
Meadow escue is one o he mos impo an na i e o age g asses
in No dic coun ies. I is also common ou side o ag onomic use
in meadows, oadsides, and was elands (Häme -Ah i e al., 1998).
Meadow escue cul i a s a e o en colonized by he sys emic endo-
phy e E. uncina a, which p oduces loline alkaloids, which appea
o be non- oxic o la ge mammal he bi o es (Clay and Scha dl,
2002) bu can be noxious o in e eb a es and small e eb a es
(Cono e , 2003;Saikkonen e al., 2006;Hui u e al., 2008). A
common meadow escue cul i a “Kaspe ,” egis e ed and com-
me cialized o use in No dic coun ies in 1989 (Saa i e al., 2009)
was used o pu poses o his s udy.
The field ole is a common and widesp ead small oden species
in Eu ope, including Fennoscandia (Myllymäki, 1977). I exhibi s
high-ampli ude 3–4 yea popula ion cycles which a e synch onous
ac oss la ge geog aphic a eas (Sundell e al., 2004;Ko pela e al.,
2013). Field oles p e e en ially inhabi g assland habi a s (Hans-
son, 1971;Myllymäki, 1977). Thei die consis s p edomina ely o
g asses and o bs (Hansson, 1971).
EXPERIMENTAL DESIGN AND PROTOCOL
The s udy was conduc ed in enced enclosu es in Jokioinen,
Sou h–Wes Finland (60◦49N, 23◦30E) du ing summe
2011. The enclosu es (20 in o al in 5 ×4 configu a ion, each
25 m ×39 m; Figu e 1) we e es ablished in May 2006 in an
old ag icul u al field by illing, e ilizing wi h cow manu e (30
000 kg/ha) and sowing as monocul u es ei he endophy e- ee
seed lo s (E−; 0% endophy e equency) o endophy e-in ec ed
seed lo s (E+; 79% in ec ion equency) o he meadow es-
cue cul i a “Kaspe ” a a a e o 20 kg ha−1(Saikkonen e al.,
2013). In June 2007, plo s we e e ilized again wi h a com-
me cial e ilize [16:9:22 (N:P:K) wi h mic onu ien s, Kemi a,
p oduc numbe : 0647334]. E+and E− ea men s we e an-
domly assigned o 10 enclosu e pai s. Seed lo s we e ob ained
om seed p oduc ion a ms ia he Finnish Food Sa e y Au ho -
i y (EVIRA), Seed Ce ifica ion Uni , Loimaa, Finland. Fou
yea s a e es ablishmen , he co e pe cen ages o meadow
escue had dec eased om 100 o 75% and 98% in E– and
E+plo s, espec i ely, due o weed in asion (Saikkonen e al.,
2013).
Each enclosu e was cons uc ed o a shee me al ence (embed-
ded 60 cm below g ound while 60 cm emained abo e g ound) in
F on ie s in Plan Science |Func ional Plan Ecology Sep embe 2014 |Volume 5 |A icle 478 |2
Hui u e al. G ass de enses agains mammalian he bi o es
FIGURE 1 |Ae ial pho o o expe imen al enclosu e complex.The
deno a ion E+indica es hose 25 ×39 m enclosu es which we e
ege a ed by endophy e-in ec ed meadow escue, and E− hose wi h
endophy e- ee meadow escue. The deno a ion G+in he smalle plo s
pic u ed in he inse indica e hose plo s in which field oles g azed h ough
he summe , and G− hose, which we e no g azed. Pho o cou esy o he
Na ional Land Su ey o Finland.
o de o e ain he expe imen al oles and exclude oles o na u al
popula ions and small mammal p eda o s.
E e y second enclosu e pai o he 10 a ailable pai s was
selec ed o he expe imen (Figu e 1). Fou 5m ×5 m shee me al
plo s a anged in 2 ×2 squa es we e cons uc ed (Figu e 1)inone
andomly assigned co ne (co ne s bo de ing he ou side o he
enclosu e complex we e excluded o educe p eda o a ac ion)
o each o he 10 enclosu es (5 ×E+and 5 ×E−). Two plo s pe
enclosu e we e andomly designa ed as ole g azing plo s (G+)
and wo as non-g azed con ol plo s (G−).
In o al, he expe imen consis ed o 10 eplica es o each o he
ollowing ac o ial ea men s: ole g azing in endophy e-in ec ed
plo s (G+E+), ole g azing in endophy e- ee plo s (G+E−), no
g azing in endophy e-in ec ed plo s (G−E+), and no g azing in
endophy e- ee plo s (G−E−). The g azing ea men was begun
in la e May 2011 by in oducing wo adul field oles, apped in
he icini y o he enclosu es and in Suonenjoki, Eas e n Finland,
in oeachG+plo . P io o in oduc ion, all oles we e housed
in s anda d labo a o y oden cages (43cm ×26cm ×15 cm;
Tecniplas , I aly) wi h ad libi um wa e , eed pelle s (Al omin
1314F; Al omin Spezial u e GmbH and Co. Ge many), and
u nip.
Two le els o g azing in ensi y we applied o ensu e adequa e
esponses o g asses o he g azing ea men . The medium g azing
in ensi y ea men in ol ed wo oles g azing in each plo con in-
uously o 2 weeks, and he ea e ou oles g azing o 3 weeks,
ei he di ec ly a e he fi s 2 weeks, o a e a3weekpe iodo no
g azing. The hea y g azing in ensi y ea men in ol ed wo oles
g azing in each plo con inuously o 2 weeks, and he ea e ou
oles g azing fi s o 3 weeks, ollowed by 3 weeks o no g azing
and finally ano he 3 weeks o g azing by ou oles.
In summa y, he medium g azing in ensi y plo s ecei ed
14 days o g azing a a ole densi y co esponding o 800 oles
/ ha (2 oles ×0.0025 ha =800 oles/ha) plus 21 days o
g azing a a ole densi y co esponding wi h 1600 oles/ha (4
oles ×0.0025 ha =1600 oles/ha). The high in ensi y g az-
ing ea men s ecei ed 14 days o g azing a 800 oles / ha plus
42 days o g azing a 1600 oles/ha. The collec i e g azing in en-
si y o e he expe imen app oxima ed 1200 oles/ha o 5 weeks
du ing 3 mon hs and ca. 1200 oles / ha o 8 weeks du ing
3 mon hs o he medium and hea y g azing in ensi y ea -
men s, espec i ely. The le el o g azing on indi idual g asses was
no measu ed, bu isual inspec ion indica ed ha while hea y
g azing did ha e a subs an ial e ec on ege a ion, g asses we e
a no poin du ing he expe imen en i ely deple e wi hin he
plo s.
Li e apping was conduc ed in mid-June in all enclosu es o
measu e he body mass o oles in he G+enclosu es and o e i y
ha no oles had en e ed he G−enclosu es. As expec ed, no
oles we e encoun e ed in he G−enclosu es. Howe e , only ci ca
25% o he o iginally in oduced oles we e apped in he G+
enclosu es o e h ee nigh s. This s ongly indica ed p eda ion
wi hin he expe imen al enclosu es, and a ca was occasionally
seen a ound he enclosu e complex.
The e o e, he expe imen al g azing p o ocol was al e ed om
i s o iginal o m. In mid-June, all oles we e emo ed om he
plo s. Fou field oles we e hen in oduced in o hal o he G+
plo s o a pe iod o 3 weeks. The ea e , he oles we e apped
and ansloca ed o he emaining plo s. This was again epea ed
a e 3 weeks, wi h ou oles e u ned o he plo s ha we e fi s
g azed. Du ing his 9 week pe iod, only se en oles disappea ed,
and hose ha did we e eplaced upon ansloca ion. All oles we e
li e apped ou o he plo s a he end o Augus .
SAMPLE COLLECTION AND PHYTOCHEMICAL ANALYSES
To e alua e he e ec s o g azing on phy ochemis y, meadow es-
cue samples we e collec ed in each o he plo s h ee imes du ing
he summe : immedia ely p io o in oduc ion o oles, once a
he beginning o Augus and once a he beginning o Sep embe
(1 week a e oles had been emo ed). Each ime, one ille was
emo ed a he base om h ee andomly selec ed meadow escue
plan s pe plo and pooled in o one sample. Du ing he fi s sam-
pling occasion, samples om he wo G+and wo G−plo s pe
enclosu e we e also pooled.
The Si, phospho ous (P), ni ogen (N), and ca bon (C) con-
en s o he samples we e analyzed o each o he h ee sampling
occasions. Fo hese analyses, g ass samples we e washed unde
wa e , o en d ied a 80◦C and g ound in o a fine, non-fib ous
powde . Th ee hund ed millig am o his powde was p essed in a
hyd aulic p ess in oa3mm hick, 13 mm diame e pelle wi h a
p essu e o 13 ba . Si and P con en (exp essed as pe cen age d y
weigh ) was analyzed om he pelle s wi h a po able X- ay fluo-
escencespec ome e (Ni onXL3 900GOLDDAnalyze ; The mo
Scien ific,Winches e , UK),as desc ibed by Reidinge e al. (2012).
www. on ie sin.o g Sep embe 2014 |Volume 5 |A icle 478 |3
Hui u e al. G ass de enses agains mammalian he bi o es
As his echnique is non-des uc ible, he same pelle s we e used
o he analysis o N and C. This was done by ISO 13878 and ISO
10694 based me hods, espec i ely, using a LECO/CHN analyse
(Leco Co., USA).
Phenolic compounds we e analyzed only om samples col-
lec ed on he las occasion, a e he g azing ea men . These
samples we e ai d ied a oom empe a u e o 1 week p io
o analyses. All soluble non- annin phenolic compounds we e
ex ac ed and HPLC-analyzed om he g ass lea samples acco d-
ing o he p o ocol desc ibed in Nybakken e al. (2012).
VOLE LABORATORY EXPERIMENT
A he beginning o Sep embe , a labo a o y expe imen was
ca ied ou o e alua e how ea lie g azing and endophy e in ec-
ion s a us a ec ed ole physiological condi ion and pe o mance.
96 wild field oles, cap u ed nea he owns o Suonenjoki,
cen al Finland (62◦N, 27◦E) and Jokioinen, we e housed
in same sex pai s in a decommissioned g eenhouse, in s an-
da d labo a o y cages con aining pea and s aw as bedding and
nes ma e ial. The oles we e assigned o one o ou ood
ea men g oups (n=24 oles pe ea men g oup), which
ecei ed ad libi um g ass collec ed daily om he expe imen-
al plo s ( ea men s: E+G+,E+G−,E−G+, and E−G−).
Fo his segmen o he s udy, all g ass collec ed om g azed
enclosu es was pooled and no dis inguished be ween hea y o
medium g azing. Voles also ecei ed pieces o po a o o ex a
hyd a ion.
Expe imen al eeding began on Sep embe 10 h and con in-
ued un il he conclusion o he expe imen on Sep embe 26 h.
Th oughou his pe iod, oles we e checked daily o su i al,
and each animal was weighed using a sp ing balance (Pesola
AG, Swi ze land; accu acy ±1 g) e e y 3–4 days (six mea-
su emen occasions in o al). In he beginning and a he end
o he eeding pe iod, a blood sample was collec ed om he
e o-o bi al sinus o each ole using a capilla y ube. Tubes
we e hen cen i uged a 12000 g o fi e minu es, and haema-
oc i exp essed as he pe cen age o packed ed blood cells in
o al olume. The eeding expe imen was e mina ed and all
oles eu hanized due o an inc eased a e o mo ali y in he
end o he second expe imen al week, o easons ha emain
uniden ified.
The expe imen was ca ied ou unde pe mi om he Animal
E hics Council o he S a e P o incial O fice o Sou he n Finland.
STATISTICAL ANALYSES
Va ia ion in g ass Si, P, N, and C con en was analyzed wi h linea
mixed models using he maximum likelihood me hod, whe e each
o he elemen s wasin u n he esponse a iable. G azingin ensi y
(hea y, medium o con ol), endophy e in ec ion s a us (E+o
E−), ime (June, Augus , Sep embe ) and all in e ac ions we e
ini ially en e ed in he models as class explana o y a iables. Time
was designa ed a epea ed ac o , wi h plo as he subjec . In all
cases, AIC-based model selec ion (Bu nham and Ande son, 2000)
indica ed uns uc u ed as he mos pa simonious co a iance ype.
Enclosu e was en e ed as a andom ac o .
Va ia ion in phenolics concen a ions a he end o summe
was also analyzed wi h sepa a e linea mixed models o each
compound using he maximum likelihood me hod. Ini ially, all
peak in ensi y alues o he di e en compounds we e loge-
ans o med. G azing in ensi y, endophy e in ec ion s a us and
hei in e ac ion we e en e ed as fixed explana o y a iables, while
enclosu e and plo (nes ed wi hin enclosu e) we e designa ed
andom ac o s.
The o e all esponse o meadow escue o he ea men s in
e mso hei phenolic con en wasalsoanalyzedwi h mul i a ia e
analysis o a iance using all 16 iden ified phenolic compounds
as esponse a iables, and g azing in ensi y, endophy e in ec ion
s a us and hei in e ac ion as explana o y a iables.
Changes in he body mass and haema oc i alue o oles ha
su i ed o he end o he labo a o y expe imen we e analyzed
wi h linea mixed models,wi h he pe cen age o change ela i e o
he beginning o he expe imen as he esponse a iable. G azing
his o y (g azed o no ) and endophy e in ec ion s a us o he g ass
hey we e ed, ole sex and body mass a he beginning o he
expe imen , oge he wi h all possible h ee-way in e ac ions, we e
en e ed as fixed explana o y a iables. Iden i y o he cage in which
he oles we e housed was en e ed as a andom ac o . Full models
we e educed by emo ing explana o y e ms on he basis o AIC.
Model fi was assessed by isual obse a ion o he final model
esiduals.
Vole su i al du ing he labo a o y expe imen was analyzed
wi h gene alized linea mixed models, using ole su i al as a
bina y esponse a iable wi h a logi link. G azing his o y and
endophy e in ec ion s a us o he g ass, ole sex, and all in e ac-
ions, we e en e ed as fixed explana o y a iables and ole cage as
a andom ac o .
Full models we e educed in all analyses by emo ing explana-
o y e ms on he basis o AIC – e ms we e emo ed i his educed
he AIC alue by >2 uni s (Bu nham and Ande son, 2000). Model
fi was assessed in he linea mixed models by isual inspec ion
o he final model esiduals, and by he a io o he gene alized
chi-squa e s a is ic and i s deg ees o eedom in he gene al-
ized linea mixed model. All analyses we e ca ied ou wi h SAS
s a is ical so wa e 9.3 (SAS® Ins i u e Inc., Ca y, NC, USA).
RESULTS
G azing ele a ed he Si con en o g asses (F2,24.4 =3.50,
p=0.046). Highes le els we e eco ded in he hea y g az-
ing ea men (mean ±SE. Si con en as pe cen age d y
weigh =1.56 ±0.07%), ollowed by medium g azing
(1.54 ±0.08%), while con ol g asses had he lowes alues
(1.36 ±0.05%; Figu e 2). Si le els we e consis en ly highe in
E+g asses han in E−g asses, wi h no significan in e ac i e
e ec o g azing (E+: mean ±SE. Si con en as pe cen age d y
weigh =1.60 ±0.06%; E−=1.38 ±0.06%; Figu e 2).
Likewise, g azing ele a ed he P con en o g asses
(F2,28.4 =3.78, p=0.035). Highes le els we e eco ded in he
medium g azing ea men (mean ±SE. P con en as pe cen -
age d y weigh =0.42 ±0.01%), ollowed by hea y g azing
(0.42 ±0.01%), while con ol g asses had he lowes alues
(0.40 ±0.01%; con ol s. hea y24.8 =−2.03, p=0.054; con ol
s. medium31.2 =−2.43, p=0.021; hea y s. medium30.5 =−0.35,
p=0.73). The le el o P declined in E+g asses o e he sum-
me , whe eas ha in E−g asses emained s able (F2,27.6 =3.79,
F on ie s in Plan Science |Func ional Plan Ecology Sep embe 2014 |Volume 5 |A icle 478 |4
Hui u e al. G ass de enses agains mammalian he bi o es
FIGURE 2 |Silicon con en , exp essed as pe cen d y weigh in
endophy e- ee (E−; g ay ba s) and endophy e-in ec ed (E+; black
ba s) g asses om con ol, medium g azing and hea y g azing
in ensi y enclosu es a end o summe . Di e ences be ween g azing
in ensi y ea men s: con ol s. hea y18.1 =−2.38, p=0.028; con ol s.
medium28.5 =−1.83, p=0.079; hea y s. medium30.3 =0.24, p=0.81.
Di e ences in Si con en be ween endophy e-in ec ed and non-in ec ed
g asses: Fendophy e1,9.33 =6.97, p=0.026).
p=0.035). No di e ences we e ound in concen a ions o N, C,
o in C/N- a io ela i e o g azing in ensi y, endophy e in ec ion
s a us, ime o hei in e ac ions (all p- alues >0.08).
A wo-way mul i a ia e analysis o a iance e ealed a sig-
nifican in e ac ion be ween g azing in ensi y and endophy e
in ec ion s a us on he phenolics p ofile o he g asses (Wilks’
Lambda =0.12, F32,38 =2.27, p=0.008). One-way mul i a ia e
es swi hei he E+o E−samplesdidno clea lyindica ewhicho
he main e ec s had a s onge e ec on g ass phy ochemis y. The
a ying deg ees o g azing in ensi y ended o influence he phe-
nolics p ofile o endophy e- ee samples (Wilks’ Lambda =0.001,
F32,4 =4.58, p=0.074), bu hese di e ences we e less p o-
nounced in endophy e-in ec ed samples (Wilks’ Lambda =0.02,
F32,4 =2.40, p=0.21).
Fi e o he 16 analyzed phenolic compounds exhibi ed an
ele a ed esponse o g azing – chlo ogenic acid, iso hamne in-
diglycoside, my ice in-glycoside, iso hamne in 3-glucoside, and
hamne in de i a i e (Table 1). Concen a ions o chlo ogenic
acid de i a i e we e highe in E+ han in E−g asses. Two com-
pounds we e a ec ed by he in e ac ion be ween g azing in ensi y
and endophy e s a us. Le els o que ci in, in pa icula , we e pos-
i i ely associa ed wi h g azing in ensi y, bu only in E+g asses
(Table 1).
The body mass o oles main ained in he labo a o y on
endophy e-in ec ed and endophy e- ee g asses esponded di -
e en ly o he g azing his o y o hei die . Voles eeding on
endophy e- ee g ass ha had been g azed h oughou he sum-
me los ela i ely mo e mass han oles in he o he h ee
g oups (mean ±SE. mass change du ing expe imen as pe -
cen age: E−G−=+3.90 ±3.12%, E−G+=−8.92 ±3.08%,
E+G−=+0.74 ±2.88%, E+G+=+0.62 ±3.18%). Feed-
ing on p e iously g azed g ass also had a significan nega i e
main e ec on ole body mass change (mean ±SE. mass change:
G−=+2.32 ±2.15%, G+=−4.15 ±2.18%). Females los mo e
body mass han males, i espec i e o die a y g oup (mean ±SE.
mass change: males =+3.21 ±2.26, emales =−5.04 ±2.06%;
Figu e 3).
Females ha had ed on endophy e-in ec ed g ass had ele-
a ed haema oc i le els compa ed o hose ed on endophy e- ee
g asses (mean ±SE.: E+=51.3 ±0.8, E−=48.9 ±0.9;
F1,19.7 =4.55, p=0.046). Con e sely, emales main ained on
hea ily g azed g asses su e ed ca. 45% highe mo ali y a es
han emales ed ung azed g ass, ega dless o endophy e in ec ion
s a us; hese di e ences did no mani es in males (Figu e 4).
DISCUSSION
In suppo o ou hypo heses, ole g azing ele a ed le els o bo h
Si and phenolics in g asses. Endophy e in ec ion alone also ele-
a ed g ass Si con en , o an ex en compa able o ha o g azing.
Howe e , changes in he le els o Si in esponse o g azing did
no di e be ween E+and E−g asses. G azing and endophy e
in ec ion s a us did in e ac in al e ing he phenolic compound
p ofile o g asses. Also in suppo o ou hypo hesis, consump ion
o hea ily g azed g ass had nega i e impac s on ole physiological
condi ion, bu hese e ec s we e condi ional on endophy e in ec-
ion s a us. Al hough endophy e in ec ion ele a es Si con en as
much as g azing alone does, ou esul s o e all sugges ha g az-
ing has a mo e p ominen e ec on g ass de enses han endophy e
in ec ion s a us.
GRASS RESPONSES TO GRAZING
As p edic ed, g azing-induced he up ake o Si om he soil in o
he g ass. This esponse has been documen ed se e al imes in
associa ion wi h a ious g aze axa (McNaugh on and Ta an s,
1983;Massey and Ha ley,2006;Massey e al., 2007a), and he bio-
chemical p ocesses unde lying i a e well known (Ma and Yamaji,
2006;Cu ie and Pe y,2007). Si accumula ion has been iden ified
as he mos plausible, and po en de ense mechanism o g asses
agains g amini o ous oles (Massey and Ha ley, 2006). I has
e en been sugges ed as a candida e ac o o gene a ing mul i-
annual popula ion cycles in oden s, as i s induc ion in esponse
o g azing may occu wi h a delay ela i e o oden densi y, and
he induced le els o Si pe sis in he g asses o su ficien ly long
(Massey e al., 2008;Reynolds e al., 2012).
Al hough he accumula ion o Si has in many s udy sys ems
ecei ed suppo as a de ense mechanism agains g aze s, many
o he s ha e ailed o do so. Fo example, Shewmake e al. (1989)
couldno demons a eanye ec so Sicon en on he g azing p e -
e ence o sheep, which appea o be less esponsi e o a ia ions in
g ass Si con en han small oden s (Massey e al., 2009). Densi ies
o A ican g azing ungula es ha e been shown o be ei he nega-
i ely (McNaugh on and Ta an s, 1983) o posi i ely (Geo giadis
and McNaugh on, 1990) associa ed wi h he Si con en o hei
ood plan s. This con ingency in esponses is likely o eflec di e -
ences in g azing le els and hence in he magni ude o Si induc ion,
gi en he known impac s o bo h equency and in ensi y o dam-
age on he Si esponse (Massey e al., 2007a;Reynolds e al., 2012),
as well as a ia ions in he ange o plan , he bi o e and en i-
onmen al ac o s influencing plan -he bi o e in e ac ions (e.g.,
Soininen e al., 2013a).
In suba c ic Fennoscandia, whe e ole densi ies a e 2–3 o de s
o magni ude lowe han simula ed in his expe imen , g ass Si
www. on ie sin.o g Sep embe 2014 |Volume 5 |A icle 478 |5

Hui u e al. G ass de enses agains mammalian he bi o es
Table 1 |Mean peak in ensi y alues o he phenolic compounds analyzed, g ouped by g azing in ensi y, and endophy e s a us o he g asses.
G azing in ensi y
Phenolic compound Con ol Medium Hea y Signi ican
e ec
Di ec ion
E−E+E−E+E−E+
Gen isic acid 0.12 0.10 0.08 0.09 0.09 0.10 n.s.
Neochlo ogenic acid 0.80 0.91 0.97 0.88 0.65 0.87 n.s.
Chlo ogenic acid 11.35 10.80 12.10 13.01 11.12 12.11 g azing C <M(p=0.057)
Chlo ogenic acid de i a i e 0.12 0.17 0.13 0.16 0.11 0.15 endophy e E−<E+,p=0.051
Que ce in-diglycoside 1.36 1.12 1.37 1.42 1.23 1.34 n.s.
Iso hamne in-diglycoside 0.70 0.56 0.77 0.87 0.74 0.88 g azing C <H and M
My ice in-glycoside 0.01 0.01 0.01 0.02 0.03 0.04 g azing H >C and M
Iso hamne in-glycoside 0.02 0.03 0.03 0.04 0.06 0.05 n.s.
Que ce in-glucoside 1.29 1.59 1.66 1.85 1.36 2.11 n.s.
Kaemp e ol-glycoside 0.02 0.01 0.02 0.04 0.02 0.03 n.s.
Que ci in 0.03 0.01 0.03 0.05 0.03 0.06 in e ac ion C E−>CE+;HE+>E−;
CE+<ME+andHE+
Kaemp e ol 3-glucoside 0.07 0.10 0.11 0.11 0.09 0.11 in e ac ion C E−<ME−;CE−<M
E+(p=0.054)
Iso hamne in 3-glucoside 0.27 0.35 0.44 0.51 0.36 0.62 g azing C <M and H
Rhamne in de i a i e 1 0.20 0.20 0.21 0.23 0.20 0.23 n.s.
Que ce in de i a i e 0.01 0.01 0.02 0.04 0.02 0.03 n.s.
Rhamne in de i a i e 2 0.11 0.12 0.13 0.15 0.12 0.14 g azing C <M(p=0.008) and H
(p=0.059)
Significan e ec deno es which o hese e ms was significan a a isk le el o α=0.05 (bo de line alues indica ed in column Di ec ion). Di ec ion deno es pai wise
di e ences be ween le els o he significan e ec and hei di ec ion; C, M, and H deno e con ol, medium and hea y g azing in ensi ies, E+and E−indica e
endophy e-in ec ed and non-in ec ed g asses, espec i ely.
le els appea o be influenced mo e by g ass species, hei geno ype
o hei g owing en i onmen han by he mos ly low g azing p es-
su e (Soininen e al., 2013a). Indeed, he majo i y o expe imen al
s udies in which Si has become induced in esponse o oden g az-
ing ha e been conduc ed in labo a o y o g eenhouse condi ions
(Massey and Ha ley, 2006;Massey e al., 2007a;Reynolds e al.,
2012, bu see Massey e al., 2008), o as is he case he e, in small
ou doo enclosu es. In hese expe imen al condi ionsi is ela i ely
easy o impose high damage le els in e ms o biomass emo ed
(Reynolds e al., 2012) o high ole densi ies ( his s udy). I seems
likely ha bo h ole densi ies and he equency and in ensi y o
plan damage a e likely o be lowe in na u al sys ems, which a e
also subjec o a g ea e ange o en i onmen al a iabili y (includ-
ing clima ic ac o s, soil ype, plan geno ypes, and g ow h s age),
making he de ec ion o Si induc ion in he field a he landscape
scale mo e challenging.
While no di e ences we e ound in concen a ions o N, C, o
in C/N- a io ela i e o g azing in ensi y o endophy e in ec ion
s a us, g azing did ele a e he P con en o g asses. Le els o phos-
pho ous ha e been ound o inc ease in esponse o g azing in
g asses, bu his p ima ily occu s o acili a e shoo eg ow h a e
de olia ion, a he han o ac as an induced de ense agains g aze s
(Chapin and Slack, 1979;Donaldson e al., 1984;McNaugh on
and Chapin, 1985). I is e iden ha he p ima y chemis y o
he g asses did no exhibi p onounced changes in esponse o
de olia ion by g azing.
G azing inc eased he con en o fi e o he 16 analyzed
phenolic compounds – chlo ogenic acid and fi e fla onols,
iso hamne in-diglycoside, my ice in-glycoside, iso hamne in 3-
glucoside, and hamne in de i a i e (Table 1). Many phenolics
a e inducible de ensi e compounds, which ei he inhibi diges ion
by he bi o es, a e de e en o e en di ec ly oxic (Lind o h and
Ba zli, 1984;Ka ban and Baldwin, 2007). Thei concen a ions in
g asses a e gene ally low ela i e o o he plan axa (Be nays e al.,
1989) and hei collec i e le els appea no o espond g ea ly o
g azing, a leas by oden s (Lind o h and Ba zli, 1986). Howe e ,
in esponse o he bi o y, he ela i e concen a ions o di e en
indi idual phenolic compounds may change d ama ically wi hou
changing o al concen a ions, and i has been demons a ed ha
indi idual compounds can impac ad e sely on oles (Lind o h
and Ba zli, 1984).
A clea ca ea o ou s udy on g ass phy ochemical esponses
o g azing is he ac ha concen a ions o all possible de ensi e
compounds, such as cyanogenic glycosides, hyd oxamic acids, o
alkaloids we e no comp ehensi ely analyzed. A numbe o hese
ha e been ound o espond o g azing and o ha e ad e se e ec s
F on ie s in Plan Science |Func ional Plan Ecology Sep embe 2014 |Volume 5 |A icle 478 |6
Hui u e al. G ass de enses agains mammalian he bi o es
FIGURE 3 |Changes in body mass o oles main ained in he
labo a o y on p e iously g azed o ung azed endophy e-in ec ed (E+)
o endophy e- ee (E−) g ass; (A) males, (B) emales. Di e ences o
changes in body mass o oles ela i e o he ea men s and ole sex:
Fendophy e ×g azing1,40.9 =4.14, p=0.049; Fg azing1,38.8 =4.38,
p=0.043; Fsex1,40.3 =7.20, p=0.01.
FIGURE 4 |Su i al o male ( illed symbols) and emale (open
symbols) oles main ained in he labo a o y on p e iously ung azed
o g azed g ass. Le e s abo e he symbols deno e significan di e ences
be ween he g oups a α<0.05. Fsex ×g azing1,45.41 =6.83, p=0.013.
on he bi o es ( e iewed byVica i and Bazely,1993). The na u e o
hese associa ions in ou s udy sys em emains o be in es iga ed
in u u e esea ch.
EFFECTS OF ENDOPHYTE INFECTION ON GRASS PHYTOCHEMISTRY
Si le els we e ound o be o be sligh ly highe in E+g asses han
in E−g asses. Si has been shown o be posi i ely associa ed wi h
d ough esis ance in g asses, as i o ms physical s uc u es on
lea es which educe anspi a ion h ough s oma a (Lux e al.,
2002). Si is also known o i s capaci y o inhibi he g ow h
o pa hogenic ungi (Fau eux e al., 2005) and al hough o en
desc ibed as ha ing mu ualis ic a he han an agonis ic in e ac-
ions wi h i s hos plan s (Saikkonen e al., 2006), i is plausible
ha endophy ic ungi a e in some cases iden ified by hei hos s as
ha m ul pa hogens (Saikkonen e al., 1998,2004). An al e na i e
explana ion o he ele a ed Si le els obse ed in E+g asses in
his s udy may he e o e be an e olu iona ily conse ed de ensi e
esponse agains he in usion o a o eign ungal o ganism, e en
po en ially mu ualis ic ones.
Simila ly, chlo ogenic acid de i a i e le els we e highe in E+
han in E−g asses. Chlo ogenic acid has been shown o ac as
a chemical de ense agains in e eb a e he bi o es (Be nays e al.,
2000;Ikonen e al., 2001), hough e y li le is known o how, o
i , chlo ogenic acid ac s as a de ense agains ungal in ec ions in
plan s, as Si does.
INTERACTIVE EFFECTS OF GRAZING AND ENDOPHYTE INFECTION ON
GRASS PHYTOCHEMISTRY
Twophenolic compoundswe ea ec edby hein e ac ionbe ween
g azing in ensi y and endophy e s a us. Le els o que ci in (also
known as que ce in 3- hamnoside), in pa icula , we e posi-
i ely associa ed wi h g azing in ensi y, bu only in E+g asses
(Table 1). Que ci in consump ion has been shown o educe
he g ow h o weanling oles, and mos p o oundly so when
die a y p o ein le els a e low (Lind o h and Ba zli, 1984). The
e ec s o que ci in on oles appea o be oxic, as die a y
p o ein le els ha e been shown no o a ec p o ein diges ibil-
i y – a he , low die a y p o ein le els appea o educe he
capaci y o oles o de oxi y que ci in (Lind o h and Ba zli,
1984).
The g azing-induced ele a ion o Si may ha e mo e a -
eaching e ec s han he up- egula ion o he physical de ense
o plan s. Fo example, Si appea s also o ha e a unc ional ole
in he sys emic esponse o plan s o pa hogens, h ough he ac i-
a ion o gene al de ensi e pa hways, e.g., in ol ing salicylic, o
jasmonic acids (Fawe e al., 1998;Rod igues e al., 2004;Fau eux
e al., 2005). Ele a ed Si le els ha e also been shown o inc ease
he accumula ion o lignin, phenolic compounds, chi inases, and
pe oxidases (Fawe e al., 1998). As such, Si is likely o be in eg al in
he egula ion o e pene o phenolic –based de enses in s essed
plan s (Fau eux e al., 2005).
A ca ea o ou s udy is ha we we e unable o de e mine he
e ec s o ole g azing on ei he he amoun o endophy e hyphae
pe g ass lea o he concen a ions o myco oxin alkaloids ha he
ungus p oduces. Bazely e al. (1997) demons a ed ha in ensi e
sheep g azing inc eased hyphae loads and/o alkaloid p oduc ion.
A simila mechanism may occu wi h ole g azing, hough his oo
emains a opic o u he in es iga ion.
EFFECTS OF DIET QUALITY VARIATION ON VOLE CONDITION
Voles ha consumed p e iously g azed g ass los body mass in
he labo a o y. Massey and Ha ley (2006) es ablished ha an
ele a ed concen a ion o Si in he die is de imen al o g amini -
o ous oles, in ha i educes hei capaci y o abso b ni ogen
om hei die , leading o educed g ow h a es in ju eniles and
www. on ie sin.o g Sep embe 2014 |Volume 5 |A icle 478 |7
Hui u e al. G ass de enses agains mammalian he bi o es
adul emales, and ou s udy confi ms his. Con e sely, he e a e
a ewe cases in which seconda y me aboli es ha e been shown
o ha e simila nega i e e ec s on g amini o ous mammals (see
Simons and Ma en, 1971;Ma en e al., 1973;Geo giadis and
McNaugh on, 1988), le alone oles. Al hough g asses p oduce
a wide a ay o seconda y me aboli es ( e iewed by Vica i and
Bazely, 1993), hei e ec s on oden s appea o mani es only
when animals a e ed on expe imen al die s in labo a o y condi-
ions (e.g., Lind o h and Ba zli, 1984). In ligh o his, i seems
mo e pa simonious o a ibu e he obse ed body mass loss o
oles o ele a ed Si con en in he g asses han phenolics con en .
Howe e , ou expe imen al design does no enable us o fi mly
disc imina e be ween hese ac o s.
Females, in pa icula , su e ed mass loss while main ained in
he labo a o y, ega dless o ea men . The oles we e o e ed only
one species o g ass as die . Consump ion o such a mono onous
die is highly unlikely o oles in na u al habi a s (Myllymäki,
1977;Soininen e al., 2013b). The ac ha emales su e ed g ea e
mass loss han males may be due o in e sexual di e ences in
die a y equi emen s ela ed o, e.g., ep oduc ion (Ba boza and
Bowye , 2000; o no e, none o he emales in ou expe imen
we e g a id). Un o una ely, i ually no in o ma ion exis s on
sex di e ences in die a y in ake in oden s.
Females ha had ed on endophy e-in ec ed g ass had ele-
a ed haema oc i le els compa ed o hose ed on endophy e- ee
g asses. Haema oc i measu es he ela i e olume o ed blood
cells in whole blood, and i s alues may inc ease due o ei he
p oli e a ion o ed blood cells, o , mo e commonly, due o a
educ ion in blood plasma con en due o dehyd a ion (Fai e al.,
2007). A dehyd a ion-induced educ ion in haema oc i o en
occu s in associa ion wi h p olonged as ing in animals (Vleck
e al., 2000). I is likely ha he ul ima e ac o s which lead o
emale mass loss and high dea h a e also con ibu ed p oxi-
ma ely o hei haema oc i le els. The eby, i is plausible ha
emales we e mo e sensi i e ei he o he inc eased Si con en
o he E+g asses, o o he endophy e i sel , han we e male
oles.
Highes le els o mo ali y we e eco ded o emales main-
ained on g azed g asses. Con e sely, endophy e in ec ion s a us
did no a ec mo ali y a es. Al hough endophy e in ec ion s a-
us appea ed o be associa ed wi h emale ood a oidance in he
labo a o y, as judged by hema oc i le els, hese e ec s we e no
p onounced enough o impac su i al. Collec i ely, ou esul s
indica e ha g azing-induced changes in he quali y o die , i.e.,
inc eased Si and phenolics con en , ou weigh he po en ial nega-
i e e ec s o endophy e in ec ion on he physiological condi ion
o oles (see Lind o h and Ba zli,1984;Massey and Ha ley, 2006).
Fu he mo e, we demons a e he e ha hea y and p olonged g az-
ing may educe he quali y o g amini o ous ole die o he ex en
ha i has nega i e e ec s on ole su i al. By ex ension, as he
mos p o ound nega i e impac s we e obse ed in emales, hese
e ec s hold he po en ial o ca y o e o popula ion g ow h. How-
e e , we acknowledge ha he a ificial na u e o ou labo a o y
may ha e gene a ed spu ious associa ions be ween ood quali y
and ole pe o mance. Mo e de ailed expe imen al esea ch is
he e o e s ill called o o elucida e hese causal ela ionships in
en i ely na u al su oundings.
CONCLUSION
Ou esul s ende suppo o he hypo hesis ha ood quali y
may indeed ha e a limi ing e ec on ole popula ion g ow h, in
cases when g azing is se e e and long enough (see Reynolds e al.,
2012). We conclusi ely demons a e he e ha g asses a e capable
o employing mul i- ie ed, e ec i e de enses agains mammalian
g aze s. We do no imply ha hese findings a e uni e sally appli-
cable o all g azing ecosys ems – a he , we highligh he need o
mo e de ailed in es iga ions on he en i e me abolome and he el-
a i e e ec s o de ense mechanisms hus a iden ified in g asses
o elucida e he ole o ood quali y as a de e minan o he bi o e
popula ion dynamics.
ACKNOWLEDGMENTS
This s udy was suppo ed by he Academy o Finland (g an no.
133495 o O so Hui u; g an s no. 137909 and 110658 o Ka i
Saikkonen) and by he NERC (g an no. NE/F003994/1 o Xa ie
Lambin). We hank D . S e an Reidinge and D . James S ockdale
o help wi h he silicon analyses. Technician Sinikka So sa con-
duc ed he phenolic ex ac ions. S ephen Ryan and Anaïs Zimme
assis ed wi h field wo k.
REFERENCES
Ba boza, P. S., and Bowye , R. T. (2000). Sexual seg ega ion in dimo phic dee :
a new gas ocen ic hypo hesis. J. Mammal. 81, 473–489. doi: 10.1644/1545-
1542(2000)081<0473:SSIDDA>2.0.CO;2
Ba ge , J. L., and Tannenbaum, M. G. (1998). Consump ion o endophy e-in ec ed
escue seeds and osmo egula ion in whi e- oo ed mice. J. Mammal. 79, 464–474.
doi: 10.2307/1382977
Bazely, D. R., Vica i, M., Emme ich, S., Filip, L., Lin, D., and Inman, A. (1997).
In e ac ions be ween he bi o es and 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. doi:
10.2307/2405276
Be ge on, J. M., and Joudoin, L. (1989). Pa e ns o esou ce use, ood quali y,
and heal h s a us o oles (Mic o us pennsyl anicus) apped om fluc ua ing
popula ions. Oecologia 79, 306–314. doi: 10.1007/BF00384309
Be nays, E. A., Coope -D i e , G. A., and Bilgene , M. (1989). He bi o es
and plan annins. Ad . Ecol. Res. 19, 263–303. doi: 10.1016/S0065-2504(08)
60160-9
Be nays, E. A., Oppenheim, S., Chapman, R. F., Kwon, H., and Gould, F. (2000).
Tas e sensi i i y o insec he bi o es o de e en s is g ea e in specialis s han in
gene alis s: a beha io al es o he hypo hesis wi h wo closely ela ed ca e pilla s.
J. Chem. Ecol. 26, 547–563. doi: 10.1023/A:1005430010314
Bu nham, K. P., and Ande son, D. R. (2000). Model Selec ion and In e ence: A
P ac ical In o ma ion-Theo e ic App oach. New Yo k, NY: Sp inge .
Chapin, F. S. III, and Slack, M. (1979). E ec o de olia ion upon oo g ow h,
phospha e abso p ion and espi a ion in nu ien -limi ed und a g aminoids.
Oecologia 42, 67–79.
Cheplick, G. P., and Clay, K. (1988). Acqui ed chemical de ences in g asses: he ole
o ungal endophy es. Oikos 52, 309–318. doi: 10.2307/3565204
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. doi: 10.2307/1943155
Clay, K., and Scha dl, C. (2002). E olu iona y o igins and ecological consequences
o endophy e symbiosis wi h g asses. Am. Na . 160, S99–S127. doi: 10.1086/
342161
Coley, A. B., F ibou g, H. A., Pel on, M. R., and Gwinn, K. D. (1995). E ec s o all
escue endophy e in es a ion on ela i e abundance o small mammals. J. En i on.
Qual. 24, 472–475. doi: 10.2134/jeq1995.00472425002400030012x
Cono e , M. R. (1998). Impac o consuming all escue lea es wi h he endophy ic
ungus, Ac emonium coenophialum, on meadow oles. J. Mammal. 79, 457–463.
doi: 10.2307/1382976
Cono e , M. R. (2003). Impac o he consump ion o endophy e-in ec ed pe en-
nial yeg ass by meadow oles. Ag ic. Ecosys . En i on. 97, 199–203. doi:
10.1016/S0167-8809(03)00037-9
F on ie s in Plan Science |Func ional Plan Ecology Sep embe 2014 |Volume 5 |A icle 478 |8
Hui u e al. G ass de enses agains mammalian he bi o es
Co e ill, J. V., Wa kins, R. W., B ennon, C. B., and Cowan, D. P. (2007). Boos ing
silica le els in whea lea es educes g azing by abbi s. Pes Manag. Sci. 63, 247–
253. doi: 10.1002/ps.1302
Coughenou , M. B. (1985). G aminoid esponses o g azing by la ge he bi o es:
adap a ions, exap a ions, and in e ac ing p ocesses. Ann. Mo. Bo . Ga d. 72,
852–863. doi: 10.2307/2399227
Cu ie, H. A., and Pe y, C. C. (2007). Silica in plan s: biological, biochemical and
chemical s udies. Ann. Bo . 100, 1383–1389. doi: 10.1093/aob/mcm247
Donaldson, C. H., Roo man, G., and G ossman, D. (1984). Long e m ni o-
gen and phospho us applica ion o eld. J. G assl. Soc. So. A . 1, 27–32. doi:
10.1080/02566702.1984.9647972
Du ham, W. F., and Tannenbaum, M. G. (1998). E ec s o endophy e consump ion
on ood in ake, g ow h, and ep oduc ion in p ai ie oles. Can. J. Zool. 76,
960–969. doi: 10.1139/z98-021
Dye , M. I., Ac a, M. A., Wang, G. M., Coleman, D. C., F eckman, D. W.,
McNaugh on, S. J., e al. (1991). Sou ce-sink ca bon ela ions in wo Pan-
icum colo a um eco ypes in esponse o he bi o y. Ecology 72, 1472–1483. doi:
10.2307/1941120
Fai , J., Whi ake , S., and Pea son, B. (2007). Sou ces o a ia ion in
haema oc i in bi ds. Ibis 149, 535–552. doi: 10.1111/j.1474-919X.2007.0
0680.x
Fau eux, F., Rémus-Bo el, W., Menzies, J. G., and Bélange , R. R. (2005). Silicon and
plan disease esis ance agains pa hogenic ungi. FEMS Mic obiol. Le . 249, 1–6.
doi: 10.1016/j. emsle.2005.06.034
Fawe,A.,Abou-Zaid,M., Menzies, J. G.,and Bélange , R.R. (1998). Silicon-media ed
accumula ion o fla onoid phy oalexins in cucumbe . Phy opa hology 88, 396–
401. doi: 10.1094/PHYTO.1998.88.5.396
Fo ie , G. M., Ba d, N., Jansen, M., and Clay, K. (2000). E ec s o all escue endo-
phy e in ec ion and popula ion densi y on g ow h and ep oduc ion in p ai ie
oles. J. Wildl. Manag. 64, 122–128. doi: 10.2307/3802981
F eeland, W. J. (1974). Vole cycles: ano he hypo hesis. Am. Na . 108, 238–245. doi:
10.1086/282905
F eeland, W. J., and Janzen, D. H. (1974). S a egies in he bi o y by mammals:
he ole o plan seconda y compounds. Am. Na . 108, 269–289. doi: 10.1086/
282907
Gali-Muh asib, H. U., Smi h, C. C., and Higgins, J. J. (1992). The e ec o silica in
g asses on he eeding beha io o he p ai ie ole, Mic o us och ogas e .Ecology
73, 1724–1729. doi: 10.2307/1940024
Geo giadis, N. J., and McNaugh on, S. J. (1988). In e ac ions be ween g az-
e s and a cyanogenic g ass, Cynodon plec os achyus.Oikos 51, 343–350. doi:
10.2307/3565316
Geo giadis, N. J., and McNaugh on, S. J. (1990). Elemen al and fib e con en s o
sa anna g asses: a ia ion wi h g azing, soil ype, season and species. J. Appl. Ecol.
27, 623–634. doi: 10.2307/2404307
Häme -Ah i,L.,Suominen, J., Ul inen, T.,andUo ila,P. (eds). (1998). Re keilykas io
(Field flo a o Finland). Helsinki: Finnish Museum o Na u al His o y, Bo anical
Museum.
Hansson, L. (1971). Small oden ood, eeding and popula ion dynamics: a com-
pa ison be ween g ani o ous and he bi o ous species in Scandina ia. Oikos 22,
183–198. doi: 10.2307/3543724
Hjäl én, J., Danell, K., and E icson, L. (1996). Food selec ion by wo ole species in
ela ion o plan g ow h s a egies and plan chemis y. Oikos 76, 181–190. doi:
10.2307/3545760
Hui u, O., Helande , M., Leh onen, P., and Saikkonen, K. (2008). Consump ion
o g ass endophy es al e s he ul a iole spec um o ole u ine. Oecologia 156,
333–340. doi: 10.1007/s00442-008-0984-3
Ikonen, A., Tah anainen, J., and Roininen, H. (2001). Chlo ogenic acid as an an i-
he bi o e de ence o willows agains lea bee les. En omol. Exp. Appl. 99, 47–54.
doi: 10.1046/j.1570-7458.2001.00800.x
Jackson, J. A., Va ney, D. R., Pe oski, R. J., Powell, R. G., Bush, L. P., Siegel,
M. R. e al. (1996). Physiological esponses o a s ed loline and e go alka-
loids om endophy e-in ec ed all escue. D ug Chem. Toxicol. 19, 85–96. doi:
10.3109/01480549609002198
Jones, D. A. (1998). Why a e so many ood plan s cyanogenic? Phy ochemis y 47,
155–162. doi: 10.1016/S0031-9422(97)00425-1
Jung, H. J. G., and Ba zli, G. O. (1981). Nu i ional ecology o mic o ine oden s:
e ec s o plan ex ac s on he g ow h o a c ic mic o ines. J. Mammal. 62, 286–
292. doi: 10.2307/1380705
Ka ban, R., and Baldwin, I. T. (2007). Induced Responses o He bi o y. Chicago, IL:
Uni e si y o Chicago P ess.
Klemola, T., No dahl, K., and Ko pimäki, E. (2000). Do delayed e ec s
o o e g azing explain popula ion cycles in oles? Oikos 90, 509–516. doi:
10.1034/j.1600-0706.2000.900309.x
Ko pela, K., Delgado, M., Hen onen, H., Ko pimäki, E., Koskela, E., O askainen,
O., e al. (2013). Nonlinea e ec s o clima e on bo eal oden dynamics: mild
win e s do no nega e high-ampli ude cycles. Glob. Chang. Biol. 19, 697–710. doi:
10.1111/gcb.12099
Lind o h, R. L., and Ba zli, G. O. (1984). Plan phenolics as chemical
de enses: e ec s o na u al phenolics on su i al and g ow h o p ai ie oles
(Mic o us och ogas e ). J. Chem. Ecol. 10, 229–244. doi: 10.1007/BF009
87851
Lind o h, R. L., and Ba zli, G. O. (1986). Inducible plan chemical de ences: a cause
o ole popula ion cycles? J. Anim. Ecol. 55, 431–449. doi: 10.2307/4729
Lux, A., Luxo a, M., Ha o i, T., Inanaga, S., and Sugimo o, Y. (2002). Silicifica ion
in so ghum (So ghum bicolo ) cul i a s wi h di e en d ough ole ance. Physiol.
Plan . 115, 87–92. doi: 10.1034/j.1399-3054.2002.1150110.x
Ma, J. F., and Yamaji, N. (2006). Silicon up ake and accumula ion in highe plan s.
T ends Plan Sci. 11, 392–397. doi: 10.1016/j. plan s.2006.06.007
Ma en, G. C., Ba nes, R. F., Simons, A. B., and Wooding, F. J. (1973). Alkaloids and
pala abili y o Phala is a undinacea L. g own in di e se en i onmen s. Ag on. J.
65, 199–201. doi: 10.2134/ag onj1973.00021962006500020004x
Massey, F. P., Ennos, A. R., and Ha ley, S. E. (2007a). He bi o e specific induc ion
o silica-based plan de ences. Oecologia 152, 677–683. doi: 10.1007/s00442-007-
0703-5
Massey, F. P., Ennos, R., and Ha ley, S. E. (2007b). G asses and he esou ce a ail-
abili y hypo hesis: he impo ance o silica-based de ences. J. Ecol. 95, 414–442.
doi: 10.1111/j.1365-2745.2007.01223.x
Massey, F. P., and Ha ley, S. E. (2006). Expe imen al demons a ion o he an ihe -
bi o e e ec s o silica in g asses: impac s on oliage diges ibili y and ole g ow h
a es. P oc. R. Soc. B 273, 2299–2304. doi: 10.1098/ spb.2006.3586
Massey, F. P., Massey, K., Roland Ennos, A., and Ha ley, S. E. (2009). Impac s o
silica-based de ences in g asses on he eeding p e e ences o sheep. Basic Appl.
Ecol. 10, 622–630. doi: 10.1016/j.baae.2009.04.004
Massey, F. P., Smi h, M. J., Lambin, X., and Ha ley, S. E. (2008). A e silica de ences
d i ing ole popula ion cycles? Biol. Le . 4, 419–422. doi: 10.1098/ sbl.2008.0106
McNaugh on, S. J., and Chapin, F. S. (1985). E ec s o phospho us nu i ion and
de olia ion on C4 g aminoids om he Se enge i plains. Ecology 66, 1617–1629.
doi: 10.2307/1938024
McNaugh on, S. J., and Ta an s, J. L. (1983). G ass lea silicifica ion: na u al selec-
ion o an inducible de ense agains he bi o es. P oc. Na . Acad. Sci. U.S.A. 80,
790–791. doi: 10.1073/pnas.80.3.790
Myllymäki, A. (1977). In aspecific compe i ion and home ange dynamics in he
field ole Mic o us ag es is.Oikos 29, 553–569. doi: 10.2307/3543594
Niemeye , H. M. (1988). Hyd oxamic acids (4-hyd oxy-1, 4-benzoxazin-3-ones),
de ence chemicals in he G amineae. Phy ochemis y 27, 3349–3358. doi:
10.1016/0031-9422(88)80731-3
Nybakken, L., Hö kkä, R., and Julkunen-Tii o, R. (2012). Combined enhancemen s
o empe a u e and UVB influence g ow h and phenolics in clones o he sexually
dimo phic Salix my sini olia.Physiol. Plan . 145, 551–564. doi: 10.1111/j.1399-
3054.2011.01565.x
Powell, R. G., and Pe oski, R. J. (1993). Alkaloid oxins in endophy e-in ec ed
g asses. Na . Toxins 1, 163–170. doi: 10.1002/n .2620010304
Reidinge , S., Ramsey, M. H., and Ha ley, S. E. (2012). Rapid and accu a e anal-
yses o silicon and phospho us in plan s using a po able X- ay fluo escence
spec ome e . New Phy ol. 195, 699–706. doi: 10.1111/j.1469-8137.2012.04179.x
Reynolds, J. J., Lambin, X., Massey, F. P., Reidinge , S., She a , J. A., Smi h, M.
J., e al. (2012). Delayed induced silica de ences in g asses and hei po en ial
o des abilising he bi o e popula ion dynamics. Oecologia 170, 445–456. doi:
10.1007/s00442-012-2326-8
Rhoades, D. F. (1979). “E olu ion o plan chemical de ense agains he bi o es,” in
He bi o es: Thei In e ac ion Wi h Seconda y Plan Me aboli es,eds G.A. Rosen hal
and D. H. Janzen (New Yo k, NY: Academic P ess), 1–55.
Rod igues, F. Á., McNally, D. J., Da no , L. E., Jones, J. B., Labbé, C., Benhamou,
N., e al. (2004). Silicon enhances he accumula ion o di e penoid phy oalexins
in ice: a po en ial mechanism o blas esis ance. Phy opa hology 94, 177–183.
doi: 10.1094/PHYTO.2004.94.2.177
www. on ie sin.o g Sep embe 2014 |Volume 5 |A icle 478 |9