Tracking year-to-year changes in intestinal nematodecommunities of rufous mouse lemurs (Microcebus rufus)

TUOMAS AIVELO1,2*, ALAN MEDLAR1, ARI LÖYTYNOJA1, JUHA LAAKKONEN3 andJUKKA JERNVALL1

1 Institute of Biotechnology, University of Helsinki, P.O. Box 56, FI-00014 Helsinki, Finland2Department of Biosciences, University of Helsinki, Helsinki, Finland3Department of Veterinary Biosciences, University of Helsinki, P.O. Box 66, FI-00014 Helsinki, Finland

(Received 18 December 2014; revised 13 March 2015; accepted 13 March 2015; !rst published online 20 April 2015)

SUMMARY

While it is known that intestinal parasite communities vary in their composition over time, there is a lack of studies addres-sing how variation in component communities (between-hosts) manifests in infracommunities (within-host) during thehost lifespan. In this study, we investigate the changes in the intestinal parasite infracommunities in wild-living rufousmouse lemurs (Microcebus rufus) from Ranomafana National Park in southeastern Madagascar from 2010 to 2012. Weused high-throughput barcoding of the 18S rRNA gene to interrogate parasite community structure. Our results showthat in these nematode communities, there were two frequently occurring putative species and four rarer putativespecies. All putative species were randomly distributed over host individuals and they did not occur in clear temporal pat-terns. For the individuals caught in at least two di!erent years, there was high turnover of putative species and high vari-ation in fecal egg counts. Our study shows that while there was remarkable variation in infracommunities over time, thecomponent community was relatively stable. Nevertheless, the patterns of prevalence varied substantially between years ineach component community.

Key words: nematodes, mouse lemurs, longitudinal study, infracommunity analysis, metabarcoding.

INTRODUCTION

Many aspects of parasite dynamics remain unknown.How parasite communities develop and what factorsa!ect the turnover of parasite individuals and speciesduring the lifespan of individual hosts is poorlyunderstood. The most common method in long-term parasite studies is cross-sectional killing ofhosts and examination of their gastrointestinaltract. As a result, the parasite studies on temporalchanges tend to focus on the component populationlevel (i.e. all the parasite populations in a host popu-lation; between-host populations (Bush et al. 1997)),whereas the longitudinal studies on infrapopulations(i.e. parasite populations in individual hosts; within-host populations (Bush et al. 1997)) are limited.The general expectation is that the distribution of

helminth parasites within the host population isoverdispersed: many individuals are uninfected orlightly infected while minority of hosts are heavilyinfected and therefore responsible for a higher pro-portion of parasite transmission compared withothers (Poulin, 1997; Hayward, 2013). Some indi-viduals are simultaneously infected with multipleparasite species and there is also evidence of mixedinfections from multiple strains of the same species(Dubey et al. 2009; Balmer and Tanner, 2011;

Gri"ths et al. 2011). Multiple-strain infectionswith negative e!ects seem to be common (Balmerand Caccone, 2008) for two possible reasons: theso-called ‘vicious cycle’ of individuals in poorhealth getting infected and interactions betweenpathogens promoting simultaneous co-infection(Stephenson et al. 2000). Co-infections, however,can also lower the total virulence of parasites(Alizon, 2013; Viney and Graham, 2013).The dynamics of component communities are well

studied, but there are con#icting results on the stab-ility of these communities: common helminthspecies appear to have long-term stability(Haukisalmi et al. 1988) and predictability (Behnkeet al. 2008a), even over large spatial scales (De Roijand MacColl, 2012), but there is also evidence forhigh turnover in parasite communities (Behnkeet al. 2008b; Ebbert et al. 2013). Infracommunitystudies are uncommon as following the same hostindividuals over their lifespan is di"cult. Despitethese di"culties, there have been recent advances:Telfer et al. (2010) have shown that there arestrong interactions between microparasite specieswithin the infracommunity. Most of the macropara-site studies have concentrated on fecal egg counts(FECs) and hosts’ resistance to helminths. Forexample, ponies in seminatural settings have strongseasonal variation in FEC and complex within-indi-vidual dynamics in shedding of eggs (Wood et al.2013). One of the most detailed studies with wild

* Corresponding author. Institute of Biotechnology,University of Helsinki, P.O. Box 56, FI-00014 Helsinki,Finland. E-mail: tuomas.aivelo@helsinki.$

1095

Parasitology (2015), 142, 1095–1107. © Cambridge University Press 2015doi:10.1017/S0031182015000438

animals was conducted with Soay sheep fromSt. Kilda (Clutton-Brock and Pemberton, 2004),mostly on di!erences in FEC over individual lifehistories (e.g. Jones et al. 2005; Hayward et al.2009), but also in succession of nematode infracom-munity after anthelminthic treatment (Craig et al.2009).The e!ect of host hibernation on parasite infra-

communities is poorly understood. A number of lab-oratory studies were performed in the early 20thcentury on helminths (reviewed by Barnes, 1970)with the general result of parasite eliminationduring host hibernation; however, there is also evi-dence to the contrary (Chute, 1960; Coggings et al.1982). Simitch and Petrovitch (1954) induced hiber-nation in European ground squirrel (Citellus citellus)and noted that di!erent helminth species survive fordi!erent periods of time. The mucosal lining ofintestine atrophies during hibernation (Carey,1990) and the expectation is that available nutrientsfor helminths are reduced. The immune systemseems to be strongly downregulated during hosthibernation (Bouma et al. 2010), though some intes-tinal immune functions are also upregulated (Kurtzand Carey, 2007). There seems to be signi$cantdi!erences in the microbiota between active andhibernation seasons, but clear trends have yet toemerge (Sonoyama et al. 2009; Carey et al. 2012).A $eld study of alpine marmots (Marmotamarmota) by Callait and Gauthier (2000) showedthat helminths can overwinter in an intermediatehost or the soil, but some manage to survive in theatrophied intestines of the infected marmots aswell. Even bears, which have a higher body tempera-ture than marmots during hibernation, seem to clearparasites prior to hibernation (Gau et al. 1999).Hibernation in tropical regions is common, but ithas not been studied in relation to the parasitism(Dausmann et al. 2005).This study describes the temporal changes in the

parasite communities of microchip-marked free-ranging rufous mouse lemurs (Microcebus rufus)both at the population-level and by tracking individ-uals longitudinally. The long lifespan of mouselemurs in the wild (up to 9 years, Zohdy et al.2014.) makes the mouse lemur an eminently suitablesystem to examine longitudinal patterns in parasiteinfections. Furthermore, rufous mouse lemurs arefacultatively hibernating primates, as individuals ingood body condition often hibernate during thedry season (April–September) (Atsalis, 2008). Westudied nematode communities, as the previousstudies with di!erent mouse lemur species haveshown that most of their gastrointestinal parasitesare nematodes (Raharivololona and Ganzhorn,2009, 2010). Due to the ongoing longitudinal studyon mouse lemur senescence and for ethical reasons,we were not able to dissect mouse lemurs in orderto examine adult nematodes. Instead, we developed

an approach to survey fecal samples from recapturedmouse lemurs and identi$ed operational taxonomicunits (OTUs) and putative species (PS) of parasitesvia barcode sequencing of the small ribosomalsubunit gene (Blaxter et al. 2005). This non-invasivemethod should provide a robust estimate of theactual nematode community in the mouse lemurs’intestines (Gillespie, 2006; Wilson et al. 2014).Our aims were to investigate nematode infection

persistence in individual lemur hosts. Speci$callywe had the following research questions. (i) Isthere a successional turnover of nematodes throughtime in within-host or between-host communities?(ii) Can we characterize the variability of nematodeprevalence between or within years? (iii) Do nema-tode community go extinct during mouse lemurhibernation? We hypothesized that after possiblehibernation, the nematode community would startsuccession from a blank state. This extinctionwould manifest as substantial di!erences in parasitecommunities from year to year. We predicted vari-ation in nematode prevalence between years due todi!ering biotic and abiotic factors: namely warmertemperatures and higher rainfall should lead tohigher prey densities and therefore to higher nema-tode prevalence. We also predicted an increasingprevalence during trapping seasons due to the suc-cession of infracommunities.

MATERIALS AND METHODS

Fecal samples were collected from October 2010 toDecember 2012 in Ranomafana National Park inSoutheastern Madagascar (21°16!S latitude and 47°20!E longitude). Ranomafana National park covers43 500 ha of lowland to mountainous moist ever-green forest from 500 to 1500 m elevation. Therainy season is from December to March and thedry season from April to August but there is exten-sive variation in rainfall patterns between years(Table 1). Traps were set on alternating nights attwo sites separated by the river Namorona. The$rst trap site located within the National Parkalong the Talatakely trail system within an area ofsecondary forest. The second trap site was locatedin the peripheral zone of the park in the campsiteof Centre Valbio. The second site was cleared andregenerated over 10 years with both endemic andnon-endemic trees (Wright and Andriamihaja,2002). Trapping sites were separated by a riverthat is considered to be a dispersal barrier.There were 19 trapping nights (October–

November) in 2010, 75 nights in 2011(September–December) and 85 nights in 2012(August–December). Mouse lemurs are most activeduring the mating season and our samplingcoincided with this season. Trapping success ratesare known to be low in Ranomafana during the dryseason when the mouse lemurs are assumed to be

1096Tuomas Aivelo and others

hibernating (Atsalis, 2008). At each site, 50 live traps(22·2 ! 6·6! 6·6 cm; XKL, Sherman Traps Inc.,FL, USA) were set along trails in pairs at 50 m inter-vals. Each trap was placed approximately 1·5 mabove the ground on tree branches and baited withfresh banana 1 h before the sunset. The traps werechecked 2 h after sunset, the mouse lemursmeasured, samples collected and any previouslyunseen brown mouse lemurs were tagged withmicrochips (FECAVA Eurochips, Vetcare,Finland) to allow for later identi$cation. Themouse lemurs were released on the same branchthey were caught on as soon as possible. Themedian time from setting the traps to release forany individual was 4 h, while the maximum durationwas 8 h.Animal handling procedures were approved by

the trilateral commission (CAFF/CORE) inMadagascar (permits: 115/10/MEF/SG/DGF/DCB.SAP/SCBSE, 203/11/MEF/SG/DGF/DCB.SAP/SCBSE and 203/12/MEF/SG/DGF/DCB.SAP/SCBSE) and the research ethics committee atViikki campus, University of Helsinki.

DNA extraction

Nematodes were isolated from fecal matter usingBaermann’s method (Baermann, 1917). We followedthe procedure by Gibbons et al. (2014) without$xing the samples. Baermannisation has beenshown to be less sensitive than other methods, butwe found it to be more practical under $eld con-ditions (Rinaldi et al. 2010). Fecal matter wasweighed and placed on a tissue (one half of 1-plyKimwipe, Kimberly-Clark Europe Ltd., Surrey,UK). The tissue was folded, tied with string andplaced on a sterile glass funnel with rubber tubing.The tubing was clamped and the funnel $lled with37 °C distilled water. The nematode larvae in thesample hatch and swim out from the sample to thewater. After 2 days the clamp was released and theliquid collected in a microcentrifuge tube (2 mL,

Sarstedt AG, Nümbrech, Germany). The samplewas centrifuged for 5 min at 2800 rcf to collect allthe larvae at the bottom of the tube and the super-natant discarded. The resultant pellet was spreadover one or two microscope slides, depending onthe size of the pellet, and examined under anoptical microscope with 100! magni$cation. Wequanti$ed all nematode larvae in the samples twice.If the results di!ered we used the mean countrounded to nearest whole number. The number oflarvae was assumed to be equal to the FEC. Thelarvae were then stored in 70% ethanol in an 18 °Cfreezer. We also surveyed a subset of fecal samples(n = 19) after nematode isolation in #otation liquid(concentrated MgSO4 solution) to validate that allnematodes were collected. We did not $nd anyremaining nematode eggs or larvae in the subset.We performed the following steps on samples con-

taining nematodes in addition to $ve samples thathad no observed nematodes for control purposes.For DNA extraction, the sample was washed withwater, incubated for 2 h at room temperature inmilliQ water and washed again with water. Celllysis was performed by incubating the sample over-night in proteinase K and TE bu!er at 56 °C. DNAwas collected with isopropanol precipitation(Sambrook et al. 1989) and washed with ethanoltwice.

Amplicon amplifying and sequencing

The small ribosomal subunit gene (18S) wasampli$ed using the primers from Bhadury andAusten (2010): 5!-AGRGGTGAAATYCGTGGAC-3! and 5!-TCTCGCTCGTTATCGGAAT-3!. These primers amplify approximately a regionof 450 bp within the 18S gene and are thought tobe nematode-speci$c and universal for all nema-todes. Phusion high-$delity polymerase (ThermoFisher Scienti$c, Inc, Waltham, MA, USA) wasused with 1–5 !L of template. The PCR programwas the following: initial denaturation at 98 °C for2 min, then 30 cycles of denaturing at 98 °C for 15s, annealing at 53 °C for 30 s and extension at 72 °C for 30 s and $nishing with 10 min of $nalelongation in 72 °C. The success of PCR was visual-ized using a 1% agarose gel. The $ve control sampleswith no nematodes failed to amplify and were there-fore not sequenced.Amplicons were sequenced at the sequencing

facility of Institute of Biotechnology, University ofHelsinki using Roche 454 Genome SequencerFLX. We aimed to produce at least 1500 sequencesper sample.

Sequence analysis

Sequence analysis was performed with the Séancebioinformatics pipeline (ver. 0.1) for reference-based

Table 1. Rainfall (mm) in Ranomafana NationalPark 2010–2012

2010 2011 2012

Jan 676 532 697Feb 262 1168 890Mar 821 441 672Apr 200 242 558May 254 73 223Jun 188 60 244Jul 434 84 79Aug 188 217 19Sep 41 147 243Oct 79 125 127Nov 96 320 384Dec 168 440 244

1097Intestinal nematode communities in rufous mouse lemur

phylogenetic analysis (Medlar et al. 2014). In brief,amplicons with an average quality of less than 25,or length less than 300 bp were discarded. Anyreads that contained ambiguous bases, errors in theMID (barcode) or homopolymers longer than 8 bpwere similarly discarded. Denoising was performedon each sample using Ampliconnoise (ver. 1.29)(Quince et al. 2011) and putative chimeric readswere identi$ed using UCHIME (ver. 4.2.40)(Edgar et al. 2011). All amplicons were then trun-cated to 250 bp.After preprocessing, reads with a copy number

greater than 5 were clustered into OTUs at 99%similarity. OTUs were labelled by performing aMegaBLAST search of the NCBI NR (non-redun-dant) database using the centroid sequence of eachcluster as the query. Taxonomic labels werederived from the lowest common ancestor from theNCBI taxonomy of the top scoring BLAST hits.Clusters containing sequences from less than twosamples were discarded.Phylogenetic analysis was performed using

Séance’s phylogenetic placement approach. A refer-ence phylogeny was inferred using RAxML (ver.7.2.8) (Stamatakis, 2006) on the full-length rRNA18S gene sequences from all 1320 nematode speciesfound in the SILVA database (Small subunit refer-ence version 115) (Quast et al. 2013). Séance usesthe Pagan multiple sequence aligner (ver. 0.55)(Löytynoja et al. 2012) to extend the reference phy-logenetic tree with the cluster centroid sequences.We used both the OTU labels and phylogenetic

placement-derived phylogeny to identify putativespecies. OTUs with non-nematode labels wereremoved.All nematodeOTUsco-occurringwith dip-teran contamination were deemed to be a contami-nation due to parasites of dipterans laying eggs insamples during the nematode isolation byBaermann’s method. OTUs with the same labelswere merged into putative species: these often had‘head-tail’ structure with one highly prevalent OTUco-occurring with several lower abundance OTUs.

Statistical testing

Parasite prevalence was tested with a logisticregression model where year, site and sex were inde-pendent variables. As we have repeated capturesfrom the same individuals, we needed to use gener-alized estimating equations (Liang and Zeger,1986). The FEC data and the number of putativespecies were transformed to van der Waerdennormal scores to allow for using ANOVA for com-parisons of year, sex and site e!ects (Conover, 1999).The mean $rst date of capture was tested with

repeated measures ANOVA as the data was normallydistributed and there were no signs of heteroscedas-ticity. The infection probability on the $rst catchwas tested with logistic generalized estimating

equations with year, site, sex and date used as inde-pendent variables.For individual case histories, all the individuals

caught at least twice in consecutive years with suc-cessful sequencing were considered. The randomdistribution of infection among the individuals wastested by calculating the number of times each puta-tive species was found in each individual and rando-mizing this to all years. The individuals which hadthe same status, either presence or absence of agiven putative species, was calculated for both, thiswas performed for 1000 iterations. The proportionof iterations with greater than or equal to thenumber of uniform pairs compared with the actualdata was calculated. This proportion shows thepossibility of having the current or stronger aggrega-tion by chance.We used two di!erent approaches to test the puta-

tive species co-occurrence following the methodsused by Fenton et al. (2014). First, we did a !2-testto see if the patterns of occurrence in putativespecies di!er from being randomly distributed. Totake into account pseudoreplication due to therepeated samples from the same individuals, wechose from every individual one sample at random,tested the distribution and iterated it this procedure100 times. To identify any confounding factors, weused longitudinal GLMs with site, sex, year andinfection status from the previous month as indepen-dent variables and mouse lemur individuals asrandom e!ect.The overdispersion of parasite aggregation was

tested with the variance-to-mean ratio and standar-dized residuals were compared with the !2-distri-bution (Shaw and Dobson, 1995). For therepeatability tests, the FEC were log-transformedand 10 added to all values. The repeatability ofFEC is the ratio of the between-host variancecomponent to the sum of the between-host andwithin-host components, also known as the intra-class correlation coe"cient (Sokal and Roh#, 1995).Statistical tests were run using R together with the

stats (R Core Team, 2013), psych (Revelle, 2015),lme4 (Bates et al. 2014) and geepack (Højsgaardet al. 2006) packages. Graphs were generated usingggplot2 (Wickham and Chang, 2014). Rarefactioncurves were generated using the vegan R package(Oksanen et al. 2013) using subsampling withoutreplacement with 100 permutations (Gotelli andColwell, 2001). The repeatability ratios were calcu-lated using theRpackage rptRwith an analysis of var-iance-based test (Nakagawa and Schielzeth, 2010).

RESULTS

Operational taxonomic units

We caught a total of 136 unique individuals, ofwhich 45% were female (Table 2). A total of 636

1098Tuomas Aivelo and others

mouse lemurs’ fecal samples were collected, of which38% were from females. Sixty per cent of the catchesand 65% of unique individuals were from theTalatakely site. A total of 471 samples (74·1% of allsamples) were positive for nematodes (Table 3).Ampli$cation and sequencing succeeded in 243

samples resulting in 828 333 raw reads. After pre-processing, 511 687 sequences were retained foranalysis. These sequences clustered into 29 OTUs.We removed 16 OTUs identi$ed as non-nematodes,10 of which were dipterans. Two OTUs had theirclosest match to nematode species known to be para-sitic on dipteran parasites and, as they consistentlyco-occurred with the dipteran contamination, wereremoved. One OTUwas identi$ed as a soil-dwellingnematode species (Bursaphelenchus sp.) and wastherefore considered to be contamination. BothCaenorhabditis (2 OTUs) and Panagrellus (1 OTU)clusters were retained despite being soil-dwellingnematodes, as the sequences were only 99% identicalto these soil-dwelling taxa. Furthermore, the pres-ence-patterns of Caenorhabditis were consistentwith intestinal parasitism: after positive samplingthe OTUs were present with probability 0·45 inthe successive samples in comparison with 0·35overall. 12 out of 102 samples with Caenorhabditisand all samples with Panagrellus were collecteddirectly from mouse lemurs without contact tosoil-contaminated surfaces. Ultimately, there were10 OTUs determined to be mouse lemurs’ intestinalparasites, which were merged based on identicallabels and phylogenetic placement forming 6 puta-tive species in 216 individual mouse lemurs. Theaccumulation curve for these putative species isshown in Fig. 1, which shows the number of puta-tive species plateaus in both sexes.The raw sequence data are archived in the Short

Read Archive (SRA) under project numberSRP042187. The sample metadata is published inFigshare (doi: 10.6084/m9.$gshare.1108080) andincludes the correspondence between individualsequencing runs in SRA and mouse lemur captures.

Di"erences in intestinal component community betweenyears, sex and site

At the host population-level, there were 2 highlyprevalent putative species, numbered 1 and 2,which occurred every year in both sexes, while theother putative species were much rarer (Table 3).Parasite prevalence di!erences were statistically sig-ni$cantly across years: the prevalence was highest in2010, lowest in 2011 (generalized estimatingequations; odds ratio to 2010: 0·04, 95% con$denceinterval 0·003–0·39, Table S1) and relatively low in2012 (OR: 0·08, CI: 0·007–0·84, Table S1). Maleshad higher nematode prevalence than females (OR:2·1, CI: 1·1–3·9, Table S1), while the trapping sitedid not have a signi$cant e!ect. The interactionT

able

2.The

numbe

rof

samples

inthestud

y.Total

catche

ssign

i$es

thetotalsam

plenu

mbe

rforallthree

trap

ping

season

s(201

0–20

12),in

twosites(C

ampsite,

Talatakely)

divide

dby

sexes.Uniqu

eindividu

alsmeans

how

man

ydi!eren

tindividu

alsha

vebe

encaug

htdu

ring

that

1year

andindividu

alscaug

htprevious

years

sign

ifyho

wman

yof

theun

ique

individu

alsha

dbe

encaug

htdu

ring

previous

years.Firstcatchof

theseason

isthemeanda

teforthe$rstcatch

ofindividu

als:thiswas

sign

i$cantly

laterin

females

(fem

ales:O

ct13

vsmales

Sep

24)a

ndTalatakely(T

alatakelyOct

5vs

Cam

psiteSep

28)site,bu

tthe

rewas

nodi!eren

cebe

tweenyears.

Thiswas

notcalculated

for20

10be

causesamplingstartedlaterthan

inothe

ryears

2010

2011

2012

Cam

psite

Talatakely

Cam

psite

Talatakely

Cam

psite

Talatakely

Male

Fem

ale

Male

Fem

ale

Male

Fem

ale

Male

Fem

ale

Male

Fem

ale

Male

Fem

ale

Total

catche

s17

430

3239

5975

5383

5615

335

Uniqu

eindividu

als

53

1513

1614

2021

1419

2613

Med

iannu

mbe

rof

catche

spe

rindividu

al1

21

12

11

12

33

2In

dividu

alscaug

htprevious

years

61

57

46

53

Mean$rstcatchof

theseason

Sep

24Sep

27Oct

5Oct

18Sep

14Oct

17Sep

21Oct

10S.D

.(days)

11·6

12·2

13·6

16·7

12·3

20·2

16·5

14·7

1099Intestinal nematode communities in rufous mouse lemur

terms were dropped during model selection and wecould not test for individual putative species due tolow goodness of $t (Tables S2–S8).There was no signi$cant e!ects of sex nor site to

the number of putative nematode species or toFECs (ANOVA; PS, sex F(1,462) = 0·001, P=0·972, site F(1,462) = 0·633, P= 0·633, interactionF(1,462) = 0·080, P= 0·777; FEC sex F(1,462) =0·001, P= 0·972; site F(1,462) = 0·393, P= 0·531,interaction F(1,208) = 2·815, P= 0·0949). Themean number of putative species in all individualswas 1·52 ± 0·77 and the mean FEC 10·44 ± 21·01.

Temporal succession in intestinal nematode componentcommunity

In 2011, there were two distinct peaks in nematodeprevalence at the end of September and duringNovember (Fig. 2a and b). This contrasts with 2012when the peaks were less conspicuous: the parasiteprevalence inmales peak inmid-September but other-wise stays over 75%. Parasite prevalence in females hadan upward trend peaking at 100% in mid-November2012. In 2010, the trapping season was from mid-October to mid-November, but the prevalence was100% throughout this period. The seasonal patternsof putative species community di!er between the 2years: the highest diversity of nematodes occurs in2011 between the two prevalence peaks in earlyOctober (Fig. 3b), whereas in 2012 the diversity ismore evenly spread over the trapping season(Fig. 3c). The diversity is lower in 2010 with onlythe most prevalent putative species 1 and 2 present(Fig. 3a). The variation in species number is depen-dent on the presence of the rare putative species(PS3–6) because PS1 is constantly present, whereasPS2 is absent only in November 2011 (Fig. 3b and c).There was a statistically signi$cant e!ect on sex (F

(1,117) = 52·305, P< 0·001) and site (F(1,117) =9·817, P = 0·002) on the $rst capture of the yearT

able

3.The

totalp

revalenc

eof

parasiticne

matod

esan

dtheprevalen

cesof

putative

speciesdivide

dby

years(201

0–20

12)an

dby

sexeswiththebino

miale

rrors.

Prevalenc

epe

rcen

tagesreferto

allsam

pled

mou

selemurs.Poten

tialtaxo

nisthelowesttaxo

nomicalleveltha

tthe

centroid

sequ

ence

uneq

uivo

cally

belong

sto

based

onBLAST

search

results

2010

2011

2012

Poten

tial

taxo

nAllsamples

Both

FM

Both

FM

Both

FM

Putativespecies1

Strongyloides

72%

(3%)

100%

(7%)

100%

(6%)

100%

(5%)

49%

(13%

)41

%(11%

)56

%(10%

)81

%(8%)

62%

(7%)

89%

(8%)

Putativespecies2

Caenorhabditis

36%

(10%

)7%

(1%)

13%

(3%)

5%(1%)

23%

(10%

)17

%(7%)

29%

(7%)

48%

(8%)

43%

(6%)

50%

(7%)

Putativespecies3

Strongylid

a7%

(6%)

0%(1%)

0%(1%)

0%(1%)

11%

(7%)

13%

(6%)

7%(4%)

6%(4%)

0%(1%)

9%(3%)

Putativespecies4

Chrom

adorea

7%(6%)

0%(1%)

0%(1%)

0%(1%)

4%(5%)

4%(4%)

5%(3%)

9%(4%)

8%(3%)

10%

(4%)

Putativespecies5

Enterobius

2%(4%)

0%(1%)

0%(1%)

0%(1%)

0%(2%)

0%(2%)

0%(1%)

4%(3%)

4%(2%)

4%(2%)

Putativespecies6

Pan

agrellu

s1%

(3%)

0%(1%)

0%(1%)

0%(1%)

2%(3%)

4%(4%)

0%(1%)

1%(2%)

2%(2%)

1%(1%)

Any

nematod

e74

%(2%)

100%

(5%)

100%

(7%)

100%

(7%)

51%

(5%)

41%

(7%)

61%

(6%)

83%

(3%)

66%

(6%)

90%

(3%)

Suc

cessfully

iden

ti$ed

216

289

1947

2225

141

3410

7Sam

ples

withne

matod

es47

183

3647

116

4670

272

6021

2Total

sampled

636

8336

4722

611

211

432

791

236

Fig. 1. The accumulation curve for the putative parasitespecies for all samples shows the sampling has beensu"cient. The shaded area represents the standarddeviation in the random permutations of the data.

1100Tuomas Aivelo and others

(Table 2, Table S9). The lemurs were more likely tobe infected on $rst capture in 2012 (logisticregression model; OR 4·6, CI: 2·0–10·5,Table S10) than in 2011 and the individuals caughtprevious years were less likely to be infected (OR:0·33, CI: 0·12–0·97), though the signi$cant inter-action between age and sex means the old maleswere more likely to be infected than young females(OR: 9·3, CI: 2·1–41·0). There were no signi$cante!ects of the trapping site or the date of trapping.We did not include 2010 in this analysis due to thetrapping starting later in the year.The variation in FEC during the season was high

and the temporal trends di!ered between years: 2011and 2012 had contrasting trends as FECs peaked

early in the season in 2011 (Fig. 4b) and late in 2012(Fig. 4c). In 2010, FECs were high overall (Fig. 4a).

Temporal succession of parasite infracommunity inindividual mouse lemurs

We mapped individual case histories for all lemurindividuals caught in two or more years and wherewe were able to sequence the nematode community(n = 16, Figs 5 and 6). The putative species wererandomly distributed (ratio of iterations withgreater than or equal pairing 0·498). For singleoccurrences all of the putative species were uni-formly distributed (PS1: 0·118, PS2: 0·500, PS3:1·000, PS4: 0·338, PS5: 1·000 and PS6: 1·000)

Fig. 2. Prevalences of nematodes in a) females and b) males for 10-day periods shows di!ering trends between years. Theprevalence is highest in the late season but early season has low prevalence in 2011 and high prevalence in 2012. We hadsamples for 2010 in only the late season.

Fig. 3. Prevalence of di!erent putative species in mouse lemurs for (a) 2010, (b) 2011 and (c) 2012 for the time periods inwhich we succeeded to sequence more than one fecal sample. The upper limit shows the total prevalence of nematodes andthe colored areas show the proportional occurrence of di!erent putative species. PS1 and PS2 are prevalent in all years,whereas the other putative species vary.

1101Intestinal nematode communities in rufous mouse lemur

(Fig. 6). The infection data were highly redundant:all individuals for whom there were more than 3successful ampli$cations, both putative species 1and 2 were present every year, except for a singleindividual (Turandot) in 2012.Putative species 1 and 2 were both present in the

same samples signi$cantly more often thanrandom (Table 4), while other putative speciespairs had much lower proportions of signi$cantresults. Previous infection did not seem to facilitateinfection though, as there were no signi$cante!ects of having previous infection in predictingoccurrences of either PS1 or PS2 (Tables S11and S12).The number of parasite species per host was not

aggregated (the variance-to-mean ratio was 0·60,di!erence from Poisson distribution: !2 = 211,P267 = 0·99) but FECs were strongly aggregated(VMR: 43. !2 = 10999, P524 < 0·001). The within-year

repeatability of FECs within individuals was 0·332(intraclass correlation coe"cient, P< 0·001) in2011, while it was not statistically signi$cantlydi!erent from zero in 2012 (P = 0·051) (Fig. 5).We could not perform the analysis for 2010 due tolow number of samples.

DISCUSSION

Our data show that the mouse lemur population wasparasitized by a relatively small number of intestinalnematode taxa (Table 3). The intestinal nematodesseemed to follow the expected pattern where a fewof the parasite species are more abundant and moreor less stable, while most of the species are rare(Stear et al. 1998). The saturated species accumu-lation curve implies that we have adequatelysampled the nematode parasites (Fig. 1; Dove andCribb, 2006). Our results are not directly

Fig. 4. The fecal egg counts for (a) 2010, (b) 2011 and (c) 2012 vary both in level and in overall trend between years. Thedaily mean fecal egg counts are plotted and the continuous line signi$es the rolling 10-day mean.

Fig. 5. The temporal variation of fecal egg counts in individuals caught in two or more years is pervasive. Size of circlerepresents fecal egg counts on logarithmic scale. The crosses represent samplings that had no nematodes present.

1102Tuomas Aivelo and others

comparable with the two previous surveys of mouselemur parasites conducted by Raharivololona andGanzhorn (2009, 2010), as the 18S sequences formorphologically identi$ed Malagasy parasites are

not available. Nevertheless, the total number oftaxa and patterns of infection are similar betweenstudies: both previous surveys found six di!erentspecies, as we did, and the dominant species

Fig. 6. The temporal variation in nematode community composition in the individuals caught in two or more years. Theuppermost row represents samples which did not have nematodes present, the second uppermost row represents sampleswhich had nematodes but the ampli$cation was not successful and the consequent rows show presence of putative speciesat time points when the individuals were caught.

1103Intestinal nematode communities in rufous mouse lemur

remained abundant throughout the survey period.The trapping e!ort was similar every night so wedo not expect sampling bias, but the data from2010 span a much shorter period than 2011 and2012. Thus it is more di"cult to make inferencesregarding the situation in 2010.Apart from the stable occurrence of putative

species 1 and 2 in the component community,there were no sustained trends in the occurrence ofputative species. In 2010, the parasite diversity waslow throughout the trapping period, in 2011 theparasite diversity was the highest during themating season in October, but in 2012 the rare puta-tive species were present throughout the trappingperiod with no overall change in parasite diversity(Fig. 3c). The most abundant parasite, putativespecies 1, identi$ed as belonging to the genusStrongyloides, appeared with high prevalencethroughout the study. One potential reason for thehigh prevalence of Strongyloides is their free-livingstage in the soil (Sandground, 1926), thus having areservoir stage capable of infecting active mouselemurs. Strongyloides is also capable of autoinfectionto sustain their population (Nishigori, 1928). Whilstnot a perfect match, putative species 2 was highlysimilar to the soil-living Caenorhabditis elegans.The infections of putative species 2 were sustainedand also occurred in cases where the feces were col-lected directly from mouse lemurs without contactwith the soil. We therefore hypothesize that putativespecies 2 is a parasitic species from the familyRhabditidae (Anderson et al. 2009).Of the rare nematodes, putative species 3 was ident-

i$ed as belonging to the suborder Strongylida andputative species 4 to the species-rich classChromadorea. Both of these taxons contain majorgroups of diverse parasitic nematodes (Andersonet al. 2009). It is therefore di"cult to speculateabout their transmission mode. Putative species 5,identi$ed as Enterobius is a genus known to infect pri-mates. The typical transmission happens via thefecal–oral route but the species is also autoinfecting(Anderson et al. 2009). Putative species 6 was asimilar case to putative species 2 with closest sequence

similarity toPanagrellus, a free-living nematode. Thisputative species could represent a parasitic speciesfrom the family Panagrolaimidae. The rare occur-rence of these putative species could be caused by sea-sonal elimination of nematodes by mouse lemurs’hibernation and weak infection in relation to theStrongyloides that has a free-living stage. Mouselemurs rarely eat animals other than insects (Atsalis,2008), so parasites with indirect life cycles should belimited to those having insects as intermediate hosts.The co-occurrence of putative species 1 and 2 was

positively correlated (Table 4), but it is di"cult tosay whether this due to a species interaction as obser-vational correlations are highly unreliable (Fentonet al. 2014). The current infection status was not sig-ni$cantly related to the infection status during theprevious month, so there does not seem to befacilitation of infection between these species. Co-occurrence can be also due to factors other than para-site interaction, like heterogeneities among thesampled host population (Fenton et al. 2010).Rainfall could explain the lower parasite preva-

lence in 2011 compared with 2012 (Fig. 2a and b).The rainfall from March to September was muchlower in 2011 than in 2012 (1263·5 mm vs 2038·1mm, Table 1). High rainfall has been shown toresult in high insect abundance and high presenceof insects in mouse lemur diets (Atsalis, 2008). Asjuvenile lemurs are unlikely to go into prolongedhibernation because of their low-fat reserves(Atsalis, 2008), their parasite levels might havestayed high all through the dry season, ensuringhigh parasite load at the start of the trappingseason (Fig. 4c). The prevalence data for 2010 arenot as reliable because they do not cover as long atime period, but rainfall during March–Septemberwas 2126 mm which supports our explanation andsuggests that rainfall patterns during the dryseason can lead to higher parasite prevalence.There were both similarities and di!erences

between sexes. Both FEC and parasite diversitywere highly similar between males and females.This could be partly due to mouse lemur femalesand males having similar testosterone levels (Zohdyet al. 2014). The e!ect of testosterone has been impli-cated as one of the main reasons males generally havelower immunocompetence (Klein, 2004) and thussimilar levels of testosterone can reduce di!erencesin immunocompetence. The main di!erencebetween sexes was that nematode prevalence washigher in males in the early season (Fig. 2b). This islikely due to the sex-dependent activity patterns:the mean $rst catch of males was 3 weeks earlierthan females (Table 2). Our $rst catch dates offemales are similar than in previous studies (Atsalis,2008). The males are expected to wake up alreadyinAugust as the earlymale activity is due to thempre-paring for the mating season (Atsalis, 2008). Eachindividual tended to get nematodes quite quickly

Table 4. Pairwise correlation matrix for the puta-tive species. The percentage value is the proportionof iterations found to be signi$cant with a !2 test.The only clearly signi$cant correlation in occurrencewas between putative species 1 and 2. All corre-lations were positive

PS2 PS3 PS4 PS5 PS6 (%) PS3–6

PS1 100% 0% 0% 0% 0 2%PS2 0% 29% 0% 1 4%PS3 0% 0% 25PS4 0% 23PS5 0

1104Tuomas Aivelo and others

(Fig. 6), but the overall prevalence did not rise uni-formly during the season as comparedwith beginningof the season. This is probably due to the staggeredactivity start times as well, and is suggestive that thedrop in prevalence seen during October 2011 couldbe due to older individuals waking up from hiber-nation later than younger individuals (Fig. 2b).While we expected that the mouse lemurs caught

early in the season would not have nematode infec-tion due to the nematode extinction during hiber-nation, the patterns were inconclusive (Fig. 5). Sixout of 32 (19%) yearly capture series with nematodes,lacked nematodes in their initial capture (Fig. 5). Itis perhaps noteworthy that four cases were oldfemales that generally hibernate longer than males(Atsalis, 2008). A previous study on grey mouselemurs by Raharivololona and Ganzhorn (2010) inSouthern Madagascar has shown signi$cantlylower parasite loads during the cold season, butthis cannot be currently determined for our studyanimals due to them not going to traps during thehibernation season. Moreover, because our datawere based on fecal samples, there remains a possi-bility that nematodes themselves, while survivingin the host, were dormant during hibernation.There was remarkable variation in parasite pres-

ence in individual mouse lemurs through time: onsome recaptures some lemur individuals did notshed any eggs though they had previously beeninfected (Fig. 6: e.g. Chewbacca in 2011) and someinfections were present only ephemerally (Fig. 6:e.g. Rachootin in 2011). We could not verify theground truth of whether these were actual changesin species composition or if this was only due to irre-gular egg shedding. Regardless, our $ndings high-light the fact that a single fecal sampling cannot beused as a representative of the total parasite diversityof a host species. Also, the distribution of putativespecies between hosts was not signi$cantly di!erentfrom a random distribution. Due to small samplesize, it is di"cult to draw conclusions, but thismight be indicative of quite small di!erences inmouse lemurs’ susceptibility to the parasites studied.We could not survey the adult parasites (see

above) but had to use their eggs and larvae as aproxy. There is always a time lag between the infec-tion and the start of egg shedding (Craig et al. 2006),which is not known for these parasites. There is alsoheterogeneity in egg shedding depending on the hostand parasite individuals (Grenfell et al. 1995; Dreyeret al. 1996; Uparanukraw et al. 1999) that may havean e!ect on the presumed correlation betweenappearance of eggs and larvae and the presence ofadult nematodes. Thus, the egg counts do notnecessarily represent the ratios of di!erent species(Dobson, 1986) or infection intensity (Gillespie,2006). Nevertheless, as we only used the eggs as anindicator of presence, our results should be relativelyaccurate, only slightly underestimating the parasite

prevalence. Also, our low measure of repeatabilitywas comparable with that of a previous study(Fig. 5, Stear et al. 1995).Our studydemonstrates thedi"culty of collecting a

sample size large enough for understanding parasitedynamics. The success rate for our method was lowand highlights the problems experienced workingwith the minute amounts of DNA obtainable fromnematodes in fecal matter. These issues will be dis-cussed in detail elsewhere (high-throughput barcod-ing of fecal samples identi$es the composition ofparasitic nematode communities,Aivelo et al. inprep-aration). There is a clear need for further studies onmultiple-parasite systems as the e!ects of parasitedynamics on their hosts can be as signi$cant as themore commonly considered factors such as season orhost age (Telfer et al. 2010). The intestinal parasitecommunity dynamics of mouse lemurs might bedriven by the stability of Strongyloides infections asthe temporally persistent parasites are expected tohave the strongest in#uence on temporal dynamicsof host–parasite interactions (Pilosof et al. 2013).Stear et al. (2011) have attributed similar $ndings ofa robust nematode community in lambs as evidencefor a long co-evolution between host and parasite.In summary, we uncovered pervasive turnover

and variation in egg shedding in infracommunitiesof mouse lemurs’ intestinal nematode parasitesusing barcoding of nematode larvae collected non-invasively from fecal samples. The component com-munity remained remarkably stable both within andbetween years due to two abundant putative species,but there were di!erences in the patterns of preva-lence between years.

SUPPLEMENTARY MATERIAL

To view supplementary material for this article, pleasevisit http://dx.doi.org/10.1017/S0031182015000438.

ACKNOWLEDGEMENTS

The authors thank the two anonymous reviewers for theircomments. The authors also wish to thank the MadagascarNational Parks, Madagascar Institute pour laConservation des Ecosystèmes Tropicaux and theInstitute for the Conservation of Tropical Environmentsand research station Centre Valbio for the logisticalsupport, Andry Herman Rafalinirina, Victor Rasendryand Hannah Price for the help in the $eld work, RitvaRice, Agnes Viherä and Raija Savolainen for assistancein the laboratory work, Lars Paulin in the BI sequencinglaboratory and Jani Anttila for useful discussions.

FINANCIAL SUPPORT

This work was supported by grants from SuomaLoimaranta-Airila Fund, Research Foundation ofHelsinki University, Otto A. Malm Foundationand Oscar Ö#und Foundation to T.A. andAcademy of Finland to J.J.

1105Intestinal nematode communities in rufous mouse lemur

REFERENCES

Alizon, S. (2013). Parasite co-transmission and the evolutionary epide-miology of virulence. Evolution 67, 921–933.Anderson, R. C., Chabaud, A. G. and Willmott, S. (2009). Keys to theNematode Parasites of Vertebrates, 1st Edn. CAB International,Wallingford, UK.Atsalis, S. (2008). A Natural History of the Brown Mouse Lemur. PearsonEducation, Upper Saddle River, NJ, USA.Baermann, G. (1917). Eine einfache methode zur au"ndung vonAnkylostomum (Nematoden) larven in erdproben. Tijdschrid Ergeneeskd57, 131–137.Balmer, O. and Caccone, A. (2008). Multiple-strain infections ofTrypanosoma brucei across Africa. Acta Tropica 107, 275–279.Balmer, O. and Tanner, M. (2011). Prevalence and implications of mul-tiple-strain infections. Lancet Infectious Diseases 11, 868–878.Barnes, E.M. (1970). E!ect of hibernation intestinal #ora. The AmericanJournal of Clinical Nutrition 23, 1519–1524.Bates, D., Maechler, M., Bolker, B. and Walker, S. (2014). lme4: linearmixed-e"ects models using Eigen and S4. R package version 1.1–7. http://CRAN.R-project.org/package=lme4Behnke, J.M., Bajer, A., Harris, P. D., Newington, L., Pidgeon, E.,Rowlands, G., Sheri!, C., Kulis !-Malkowska, K., Sin !ski, E.,Gilbert, F. S. and Barnard, C. J. (2008a). Temporal and between-sitevariation in helminth communities of bank voles (Myodes glareolus) fromN.E. Poland. 2. The infracommunity level. Parasitology 135, 999–1018.Behnke, J.M., Bajer, A., Harris, P. D., Newington, L., Pidgeon, E.,Rowlands, G., Sheri!, C., Kulis !-Malkowska, K., Sin !ski, E.,Gilbert, F. S. and Barnard, C. J. (2008b). Temporal and between-sitevariation in helminth communities of bank voles (Myodes glareolus) fromN.E. Poland. 1. Regional fauna and component community levels.Parasitology 135, 985–997.Bhadury, P. and Austen, M. C. (2010). Barcoding marine nematodes : animproved set of nematode 18S rRNA primers to overcome eukaryotic co-interference. Hydrobiologia 641, 245–251.Blaxter, M., Mann, J., Chapman, T., Thomas, F., Whitton, C.,Floyd, R. and Abebe, E. (2005). De$ning operational taxonomic unitsusing DNA barcode data. Philosophical Transactions of the Royal SocietyB: Biological Sciences 360, 1935–1943.Bouma, H. R., Carey, H. V. and Kroese, F. G.M. (2010). Hibernation:the immune system at rest? Journal of Leukocyte Biology 88, 619–624.Bush, A. O., La!erty, K. D., Lotz, J.M. and Shostak, A.W. (1997).Parasitology meets ecology on its own terms: Margolis et al. revisited.The Journal of Parasitology 83, 575–583.Callait, M. and Gauthier, D. (2000). Parasite adaptations to hibernationin Alpine marmots (Marmota marmota). In Life in the Cold (ed.Heldmaier, G. and Klingenspor, M.), pp. 139–146. Springer-Verlag,Berlin.Carey, H. V. (1990). Seasonal changes in mucosal structure and function inground squirrel intestine. American Journal of Physiology – Regulatory,Integrative and Comparative Physiology 259, R385–R392.Carey, H. V., Pike, A. C., Weber, J. L., Turner, A. K., Visser, S. C.,Beijer-Liefers, H. R., Bouma, H. R. and Kroese, F. G.M. (2012).Impact of hibernation on gut microbiota and intestinal barrier functionin ground squirrels. In Living in a Seasonal World: Thermoregulatory andMetabolic Adaptations (ed. Ruf, T., Bieber, C., Arnold, W. andMillesi, E.), pp. 281–291. Springer, Heidelberg, Germany.Chute, R.M. (1960). Overwintering of helminths in hibernating animals.Journal of Parasitology, 46, 539.Clutton-Brock, T. H. and Pemberton, J.M. (2004). Individuals andpopulations. In Soay Sheep: Dynamics and Selection in an IslandPopulation (ed. Clutton-Brock, T. H. and Pemberton, J.M.), pp. 1–16.CUP, Cambridge, UK.Coggings, J. R., Tedesco, J. L. and Ruppercht, C. E. (1982). Seasonalchanges and overwintering of parasites in the bat, Myotis lucifugus (LeConte), in a Wisconsin hibernaculum. American Midland Naturalist 107,305–315.Conover, W. J. (1999). Practical Nonparametric Statistics, 3rd Edn. JohnWiley & Sons, Hoboken, NJ, USA.Craig, B. H., Pilkington, J. G. and Pemberton, J.M. (2006).Gastrointestinal nematode species burdens and host mortality in a feralsheep population. Parasitology 133, 485–496.Craig, B. H., Jones, O. R., Pilkington, J. G. and Pemberton, J.M.(2009). Re-establishment of nematode infra-community and host survivor-ship in wild Soay sheep following anthelmintic treatment. VeterinaryParasitology 161, 47–52.Dausmann, K. H., Glos, J., Ganzhorn, J. U. and Heldmaier, G.(2005). Hibernation in the tropics: lessons from a primate. Journal of

Comparative Physiology B: Biochemical, Systems, and EnvironmentalPhysiology 175, 147–155.De Roij, J. and MacColl, A. D. C. (2012). Consistent di!erencesin macroparasite community composition among populations ofthree-spined sticklebacks, Gasterosteus aculeatus L. Parasitology 139,1478–1491.Dobson, A. (1986). Inequalities in the individual reproductive success ofparasites. Parasitology 92, 675–682.Dove, A. D.M. and Cribb, T. H. (2006). Species accumulation curves andtheir applications in parasite ecology. Trends in Parasitology 22, 568–574.Dreyer, G., Fernandes-Silva, E., Alves, S., Rocha, A.,Albuquerque, R. and Addiss, D. (1996). Patterns of detection ofStrongyloides stercoralis in stool specimens: implications for diagnosis andclinical trials. Journal of Clinical Microbiology 34, 2569–2571.Dubey, J. P., Moura, L., Majumdar, D., Sundar, N., Velmurugan, G.V., Kwok, O. C. H., Kelly, P., Krecek, R. C. and Su, C. (2009). Isolationand characterization of viable Toxoplasma gondii isolates revealed possiblehigh frequency of mixed infection in feral cats (Felis domesticus) from StKitts, West Indies. Parasitology 136, 589–594.Ebbert, M. A., McGrew, W. C. and Marchant, L. F. (2013).Community composition, correlations among taxa, prevalence, and rich-ness in gastrointestinal parasites of baboons in Senegal, West Africa.Primates 54, 183–189.Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C. and Knight, R.(2011). UCHIME improves sensitivity and speed of chimera detection.Bioinformatics 27, 2194–2200.Fenton, A., Viney, M. E. and Lello, J. (2010). Detecting interspeci$cmacroparasite interactions from ecological data: patterns and process.Ecology Letters 13, 606–615.Fenton, A., Knowles, S. C. L., Petchey, O. L. and Pedersen, A. (2014).The reliability of observational approaches for detecting interspeci$c para-site interactions: comparison with experimental results. InternationalJournal of Parasitology 44, 437–445.Gau, R. J., Kutz, S. and Elkin, B. T. (1999). Parasites in grizzly bears fromthe central Canadian Arctic. Journal of Wildlife Diseases 35, 618–621.Gibbons, L.M., Jacobs, D. E., Fox, M. T. and Hansen, J. (2014). TheRVC/FAO Guide to Veterinary Diagnostic Parasitology. http://www.rvc.ac.uk/review/Parasitology/Index/Index.htm.Gillespie, T. R. (2006). Noninvasive assessment of gastrointestinal para-site infections in free-ranging primates. International Journal ofPrimatology 27, 1129–1143.Gotelli, N. J. and Colwell, R. K. (2001). Quantifying biodiversity: pro-cedures and pitfalls in the measurement and comparison of species rich-ness. Ecology Letters 4, 379–391.Grenfell, B. T., Wilson, K., Isham, V. S., Boyd, H. E. and Dietz, K.(1995). Modelling patterns of parasite aggregation in natural populations:trichostrongylid nematode-ruminant interactions as a case study.Parasitology 111 (Suppl.), S135–S151.Gri"ths, E. C., Pedersen, A. B., Fenton, A. and Petchey, O. L. (2011).The nature and consequences of coinfection in humans. The Journal ofInfection 63, 200–206.Haukisalmi, V., Henttonen, H. and Tenora, F. (1988). Populationdynamics of common and rare helminths in cyclic vole populations.Journal of Animal Ecology 57, 807–825.Hayward, A. D. (2013). Causes and consequences of intra- and inter-host heterogeneity in defense against nematodes. Parasite Immunology 35,362–373.Hayward, A. D., Wilson, A. J., Pilkington, J. G., Pemberton, J.M. andKruuk, L. E. B. (2009). Ageing in a variable habitat: environmental stressa!ects senescence in parasite resistance in St Kilda Soay sheep.Proceedings of the Royal Society B: Biological Sciences 276, 3477–3485.Højsgaard, S., Halekoh, U. and Yan, J. (2006). The R package geepackfor generalized estimating equations. Journal of Statistical Software 15,1–11.Jones, O. R., Crawley, M. J., Pilkington, J. G. and Pemberton, J.M.(2005). Predictors of early survival in Soay sheep: cohort-, maternal- andindividual-level variation. Proceedings of the Royal Society B: Biologicalsciences 272, 2619–2625.Klein, S. (2004). Hormonal and immunological mechanisms mediating sexdi!erences in parasite infection. Parasite Immunology 26, 247–264.Kurtz, C. C. and Carey, H. V. (2007). Seasonal changes in the intestinalimmune system of hibernating ground squirrels. Developmental andComparative Immunology 31, 415–428.Liang, K.-Y. and Zeger, S. L. (1986). Longitudinal data analysis usinggeneralized linear models. Biometrika 73, 13–22.Löytynoja, A., Vilella, A. J. and Goldman, N. (2012). Accurate exten-sion of multiple sequence alignments using a phylogeny-aware graph algor-ithm. Bioinformatics 28, 1684–1691.

1106Tuomas Aivelo and others

Medlar, A., Aivelo, T. and Löytynoja, A. (2014). Séance: reference-based phylogenetic analysis for 18S rRNA studies. BMC EvolutionaryBiology 14, 235.Nakagawa, S. and Schielzeth, H. (2010). Repeatability for Gaussianand non-Gaussian data: a practical guide for biologists. BiologicalReviews of the Cambridge Philosophical Society 85, 935–956.Nishigori, M. (1928). The factors which in#uence the external develop-ment of Strongyloides stercoralis and on auto-infection with this parasite.Journal of Formosan Medical Association 247, 1–56.Oksanen, J., Blanchet, F. G., Kindt, R., Legendre, P., Minchin, P. R.,O’Hara, R. B., Simpson, G. L., Solymos, P., Stevens, M.H. H. andWagner, H. (2013). Vegan: Community Ecology Package. R packageversion 2.0–1o0. http://CRAN.R-project.org/package=veganPilosof, S., Fortuna,M. A., Vinarski, M. V., Korallo-Vinarskaya, N. P.and Krasnov, B. R. (2013). Temporal dynamics of direct reciprocal andindirect e!ects in a host-parasite network. The Journal of Animal Ecology82, 987–996.Poulin, R. (1997). Population abundance and sex ratio in dioecious hel-minth parasites. Oecologia 111, 375–380.Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P.,Peplies, J. and Glöckner, F. O. (2013). The SILVA ribosomal RNA genedatabase project: improved data processing and web-based tools. NucleicAcids Research 41, D590–D596.Quince, C., Lanzen, A., Davenport, R. J. and Turnbaugh, P. J. (2011).Removing noise frompyrosequenced amplicons.BMCBioinformatics 12, 38.R Core Team (2013). R: a Language and Environment for StatisticalComputing. R Foundation for Statistical Computing, Vienna, Austria.http://www.R-project.org/Raharivololona, B.M. and Ganzhorn, J. U. (2009). Gastrointestinalparasite infection of the gray mouse lemur (Microcebus murinus) in the lit-toral forest of Mandena, Madagascar: e!ects of forest fragmentation anddegradation. Madagascar Conservation and Development 4, 103–112.Raharivololona, B. and Ganzhorn, J. (2010). Seasonal variations in gas-trointestinal parasites excreted by the gray mouse lemurMicrocebus murinusin Madagascar. Endangered Species Research 11, 113–122.Revelle, W. (2015). Psych: Procedures for Psychological, Psychometric andPersonality Research. R package version 2.0–10. http://personality-project.org/r/psych/Rinaldi, L., Maurelli, M. P., Musella, V., Santaniello, A., Coles, G. C.and Cringoli, G. (2010). FLOTAC: an improved method for diagnosis oflungworm infection in sheep. Veterinary Parasitology 11, 395–398.Sambrook, J., Fritsch, E. F. andManiatis, T. (1989).Molecular Cloning:a Laboratory Manual, 2nd Edn. Cold Spring Harbor Laboratory Press,Long Island, NY, USA.Sandground, B. J.H. (1926). Biological studies on the life-cycle in thegenusStrongyloides grassi, 1879.American Journal ofEpidemiology 6, 337–388.Shaw, D. J. and Dobson, A. P. (1995). Patterns of macroparasite abun-dance and aggregation in wildlife populations: a quantitative review.Parasitology 111, S111–S133.Simitch, T. and Petrovitch, Z. (1954). Ce qu’il advient avec les hel-minthes du Citellus citellus au cours du sommeil hibernal de ce rongeur.Rivista di Parassitologia 15, 655–662.

Sokal, R. R. andRoh#, F. J. (1995).The Principles and Practice of Statisticsin Biological Research, 3rd Edn. W.H. Freeman, New York, USA.Sonoyama, K., Fujiwara, R., Takemura, N., Ogasawara, T.,Watanabe, J., Ito, H. and Morita, T. (2009). Response of gut microbiotato fasting and hibernation in Syrian hamsters. Applied and EnvironmentalMicrobiology 75, 6451–6456.Stamatakis, A. (2006). RAxML-VI-HPC: maximum likelihood-basedphylogenetic analyses with thousands of taxa and mixed models.Bioinformatics 22, 2688–2690.Stear, M. J., Bishop, S. C., Duncan, J. L., Mckellar, Q. A. andMurray, M. (1995). The repeatability of faecal egg counts, peripheral eosi-nophil counts, and plasma pepsinogen concentrations during deliberateinfection with Ostertagia circumcincta. International Journal forParasitology 25, 375–380.Stear, M. J., Bairden, K., Bishop, S. C., Gettinby, G., Mckellar, Q. A.,Park, M., Strain, S. and Wallace, D. S. (1998). The processes in#uen-cing the distribution of parasitic nematodes among naturally infectedlambs. Parasitology 117, 165–171.Stear, M. J., Singleton, D. and Matthews, L. (2011). An evolutionaryperspective on gastrointestinal nematodes of sheep. Journal ofHelminthology 85, 113–120.Stephenson, L. S., Latham, M. C. and Ottesen, E. A. (2000).Malnutrition and parasitic helminth infections. Parasitology 121,S23–S38.Telfer, S., Lambin, X., Birtles, R., Beldomenico, P., Burthe, S.,Paterson, S. and Begon, M. (2010). Species interactions in a parasitecommunity drive infection risk in a wildlife population. Science 330,243–246.Uparanukraw, P., Phongsri, S. and Morakote, N. (1999). Fluctuationsof larval excretion in Strongyloides stercoralis infection. The AmericanJournal of Tropical Medicine and Hygiene 60, 967–973.Viney, M. E. and Graham, A. L. (2013). Patterns and processes in para-site co-infection. Advances in Parasitology 82, 321–369.Wickham, H. and Chang, W. (2014). ggplot2: an Implementation of theGrammar of Graphics. R package version 1.0.0. http://ggplot2.org/Wilson, K., Grenfell, B. T., Pilkington, J. G., Boyd, H. E. andGulland, F.M. (2014). Parasites and their impact. In Soay Sheep:Dynamics and Selection in an Island Population (ed. Clutton-Brock, T. H. and Pemberton, J.M.), pp. 113–165. CambridgeUniversity Press, Cambridge, UK.Wood, E. L. D., Matthews, J. B., Stephenson, S., Slote, M. andNussey, D. H. (2013). Variation in fecal egg counts in horses managedfor conservation purposes: individual egg shedding consistency, agee!ects and seasonal variation. Parasitology 140, 115–128.Wright, P. C. and Andriamihaja, B. (2002). Making a rain forest nationalpark work in Madagascar: Ranomafana National Park and its long-termresearch commitment. In Making Parks Work: Strategies for PreservingTropical Nature (ed. Terborgh, J., van Schaik, C., Davenport, L. andRao, M.), pp. 112–136. Island Press, Washington, DC, USA.Zohdy, S., Gerber, B. D., Tecot, S., Blanco, M. B., Winchester, J. M.,Wright, P. C. and Jernvall, J. (2014). Teeth, sex, and testosterone: agingin the world’s smallest primate. PLoS ONE 9, e109528.

1107Intestinal nematode communities in rufous mouse lemur