Foraging ecology of horseshoe bats and pest consumption through molecular techniques.

FORAGING ECOLOGY OF HORSESHOE BATS AND PEST CONSUMPTION THROUGH MOLECULAR TECHNIQUES PhD Thesis by Miren Aldasoro Lezea Leioa, 2025 Foraging Ecology of Horseshoe Bats and Pest Consumption through Molecular Techniques PhD Thesis by Miren Aldasoro Lezea Leioa, 2025 Under de supervision of: Dr. Joxerra Aihartza Azurtza Dr. Oihane Diaz de Cerio Arruabarrena (cc) 2025 Miren Aldasoro Lecea (cc by-nc-sa 4.0) Etxekoei. "In nature, nothing exists alone." Rachel Carson FUNDING This research was funded by: ● Grant PID2019-108123GB-I00, funded by MICIU/AEI/10.13039/501100011033 and the European Union. ● Basque Government (Research grant IT1571-22). ● Basque Government Biodiversity Directorate (Expedient No 049‐2022‐42). ● Genomic Resources Research group of the University of the Basque Country (UPV/EHU) TABLE OF CONTENTS ESKER ONAK - ACKNOWLEDGMENTS..........................................................................9 I. LABURPENA....................................................................................................................... 1 I. SUMMARY........................................................................................................................... 5 II. SARRERA OROKORRA.................................................................................................. 9 II. GENERAL INTRODUCTION........................................................................................21 III. STATE OF THE ART..................................................................................................... 43 III.1. HYPOTHESIS.......................................................................................................... 43 III.2. OBJECTIVES........................................................................................................... 44 IV. RESULTS.......................................................................................................................... 45 CHAPTER 1: Optimise primers combination for analysing the diet of horseshoe bats by metabarcoding............................................................................................................... 47 LABURPENA................................................................................................................ 49 ABSTRACT....................................................................................................................49 1.1. INTRODUCTION................................................................................................... 50 1.2. MATERIALS AND METHODS.............................................................................52 1.3. RESULTS................................................................................................................ 54 1.4. DISCUSSION..........................................................................................................59 1.5. REFERENCES........................................................................................................ 62 1.6. SUPPORTING INFORMATION............................................................................ 65 1.I. Addendum to CHAPTER 1:....................................................................................67 1.I.1. REASON FOR THE ADDENDUM..................................................................... 67 1.I.2. MATERIALS AND METHODS...........................................................................68 1.I.3. RESULTS..............................................................................................................69 1.I.4. DISCUSSION AND CONCLUSIONS.................................................................73 1.I.5. REFERENCES......................................................................................................75 1.I.6. SUPPORTING INFORMATION..........................................................................77 CHAPTER 2: Quantitative pest detection in bat faeces: first attempt with multiplex ligation-dependent probe amplification (MLPA)............................................................... 79 LABURPENA................................................................................................................ 81 ABSTRACT....................................................................................................................81 2.1. INTRODUCTION................................................................................................... 82 2.2. MATERIALS AND METHODS.............................................................................85 2.3. RESULTS................................................................................................................ 91 2.4. DISCUSSION..........................................................................................................96 2.5. REFERENCES........................................................................................................ 99 CHAPTER 3: Seasonal and geographic variation in the trophic ecology and habitat dependence of Rhinolophus hipposideros........................................................................103 LABURPENA.............................................................................................................. 104 ABSTRACT..................................................................................................................105 3.1. INTRODUCTION................................................................................................. 106 Summary I. SUMMARY This dissertation presents a significant and timely contribution to the understanding of horseshoe bats' foraging ecology. It adopts a comprehensive approach that considers seasonal and regional dietary variations, environmental interactions, and the influence of prey availability and habitat structure. Intraand interspecific competition also play a crucial role in shaping foraging strategies and resource partitioning among coexisting species. The use of DNA metabarcoding, a powerful tool for investigating trophic ecology, provides high-resolution taxonomic identification of prey items. The specific focus is on the trophic ecology of three horseshoe bat species, with a particular emphasis on the lesser (Rhinolophus hipposideros) and the greater (Rhinolophus ferrumequinum) horseshoe bats. The study has been performed in the Basque Country (Northern Iberian Peninsula), a core area of the distribution of these two species. In the first part of the thesis, we focused on the methodological part of the study, aiming to optimise the metabarcoding approach and test a new semi-quantitative assay for the multiplex identification of specific prey species. On the one hand, since using a single primer set can introduce biases in the dietary composition analyses by metabarcoding, we first evaluated whether combining two different primer sets improved the coverage of identified prey items. Each primer set revealed a distinct subset of prey, with only a small proportion detected by both. Consequently, each primer provided a different perspective on the trophic niche of the studied species. Our finding confirmed that using complementary primers significantly enhances the accuracy of diet characterisation in generalist insectivorous bats. This evaluation demonstrated that the combination of ANML and FWH1 was the most effective, maximising prey diversity coverage, prey quantity, and predator identification. This primer combination was, therefore, selected for the subsequent studies. On the other hand, we sought to develop a reliable molecular assay for the simultaneous detection and semi-quantification of multiple pest species in bat faeces using a multiplex ligation-dependent probe amplification (MLPA) assay. This study represents the first attempt to adapt the MLPA approach for detecting prey DNA in faecal samples. We designed specific probes targeting four key moth pest species—Thaumetopoea pityocampa, Cydia pomonella, Autographa gamma, and Lobesia botrana. While we successfully tested three probes on pure moth DNA and detected T. pityocampa in bat faeces, further optimisation is needed to determine detection limits and validate multiplexing capabilities. Therefore, we left this technique aside and focused on metabarcoding. 5 Summary Once the methodological part was finished, we focused on the main objectives of the thesis. While significant advances have been made in understanding bat foraging habitats, previous studies based on radio telemetry provide only short-term data, as they monitor individuals for limited periods. In contrast, metabarcoding enables species-level identification of consumed prey, allowing for an indirect assessment of source habitats. To explore seasonal and spatial dietary variations, we analysed faecal samples from three R. hipposideros colonies across different climatic zones from spring to late summer. Using metabarcoding, we examined the variation of the lesser horseshoe bat diet, checking how their most-consumed prey varied seasonally and across landscapes. Our results revealed seasonal diet composition shifts, which were dissimilar in each studied landscape, likely echoing respective prey availability. While woodlands and shrublands were primary prey source habitats, R. hipposideros also foraged in open habitats more frequently than expected, demonstrating high trophic plasticity. A similar study was conducted on R. ferrumequinum across colonies in diverse landscapes, urbanisation levels, and climatic conditions. We observed significant spatial and temporal dietary differences, with a stronger reliance on riparian habitats in Mediterranean areas. The species exhibited remarkable ecological adaptability, adjusting its foraging preferences among forests, riparian habitats, shrubs, and grasslands. Our results emphasise the importance of preserving these habitats for conservation management purposes. These findings underscore the importance of preserving a mosaic of interconnected habitats to support horseshoe bats and their prey. Building on these findings, we investigated the role of horseshoe bats in agricultural pest suppression. While insectivorous bats are increasingly recognised for their role in controlling pest populations, there remains a need to quantify their impact. We reanalysed the diets of the six colonies—three R. hipposideros and three R. ferrumequinum—focusing on pest consumption. Both species preyed on a substantial variety of pest insects, with some species preyed upon consistently and massively, including previously unreported species, with predation patterns aligning with pest outbreaks. We estimated the total biomass of agricultural pests consumed per colony during the breeding season and the overall pest consumption by these bat species in the studied area, revealing a significant contribution to natural pest control. However, bats' body size is key in pest suppression, directly influencing nightly food intake. While R. hipposideros exhibited a higher relative incidence of pest consumption, the total biomass of pests consumed was greater in R. ferrumequinum. Furthermore, both species preyed on more pest species in highly anthropised areas, emphasising their potential role in integrated pest management. Following these investigations, we examined the impact of developmental stage on diet composition by comparing juvenile and adult diets of R. hipposideros, R. ferrumequinum, and the Mediterranean horseshoe bat (Rhinolophus euryale). We analysed dietary differences both taxonomically and based on prey traits (e.g., size, flight speed, hardness). While juveniles of R. hipposideros and R. euryale consumed significantly different prey than adults, R. ferrumequinum showed less pronounced 6 Summary taxonomic differences. We could only observe discernible diet patterns through the trait analysis for the latter. A shared pattern across species was that juveniles preferentially consumed smaller, softer, and slower prey, likely reflecting developmental constraints on prey capture. These findings highlight the importance of considering age-related dietary needs in conservation strategies. Collectively, the findings of this dissertation demonstrate that the foraging ecology of R. hipposideros and R. ferrumequinum is far more complex than previously assumed. These species exhibit high dietary plasticity, adjusting their prey selection seasonally and responding to prey outbreaks. Their foraging strategies extend beyond expected habitat dependencies, reflecting a dynamic interaction with their landscapes. Moreover, juvenile dietary preferences differ from those of adults, highlighting developmental constraints on foraging efficiency. Finally, our research underscores the crucial role of horseshoe bats in agricultural pest suppression, reinforcing the importance of maintaining healthy bat populations in anthropised landscapes. These findings have practical implications for future conservation efforts, which should account for this ecological complexity, ensuring the protection of diverse foraging habitats and fostering coexistence between bats and human-altered environments. 7 II. SARRERA OROKORRA SARRERA OROKORRA II.1 SAGUZAR INTSEKTUJALEEN BAZKA EKOLOGIA Saguzarrek (Chiroptera ordena) 1.400 espezie inguru biltzen dituzte, eta, beraz, ugaztunen bigarren ordenik anitzena dira, deskribatutako ugaztun espezie guztien laurdena direlarik (Wilson & Mittermeier, 2019). Lurreko ekosistema gehienetan paper garrantzitsua betetzen dute kontinente guztietan zehar, Antartikan izan ezik (Kunz & Pierson, 1994), eta ugaztunen artean bakarrak dira hegaldi propultsatuaren garapenean. Hegaldiaren eta ekokokapenaren eboluzioa funtsezkoak izan dira saguzarrek intsektu gautar mota ezberdinak ustiatzeko (Jones & Rydell, 2003), saguzarren % 70 inguru intsektujaleak baitira (Simmons, 2005). Saguzar intsektujaleen artean dagoen aniztasun ekologiko handia hegakerarekin eta ekokapenarekin lotura estua duten hainbat egokitzapenekin lotuta dago. Esaterako, espezie handiago eta azkarrago batzuek maiztasun baxuko ekokokapen-deiak erabiltzen dituzte espazio zabaletako intsektuak ehizatzeko (adb., Nyctalus lasiopterus; Ibañez & Juste, 2023), eta saguzar txikiago eta motelagoak, berriz, maniobragarriagoak dira eta beren harrapakinak dentsitate altuagoko eremuetan harrapatzen dituzte (adb., Myotis bechsteinii; Kerth & van Schaik, 2023). Gainera, saguzar intsektujaleek burezur-egitura, baraila eta hortz morfologia ezberdinak dituzte, beren behar dietetikoekin bat datozenak, eta horrek haien hozkada-indarrari eta harrapakin motei eragiten die. Gorputz gogorreko harrapakinak kontsumitzen dituzten saguzarrek burezur eta baraila sendoagoak dituzte, eta letagin luzeagoak, harrapakinen hautespen biguna jaten dutenekin alderatuta (Freeman, 1979, 1998; Ramírez-Fráncel et al., 2021; Santana et al., 2011, 2022). Saguzar espezie bakoitzaren egokitzapen espezifiko hauek eragina dute haien bazka-ekologian, non eta zer ehizatu dezaketen baldintzatuz (adb., Swartz et al., 2003). Bazka funtsezko jarduera da animalien biziraupenerako. Saguzar intsektujaleen kasuan, harrapakin potentzialak bilatzea, detektatzea eta ezagutzean datza. Era berean, harrapakinaren atzetik ibili, harrapatu eta azkenean kontsumitu edo ez erabaki behar da (Stephens & Krebs, 1986). Horrela, harrapari-harrapakin elkarrekintza harrapakinen errentagarritasunaren eta defentsa-estrategien eraginkortasunaren, hala nola kamuflajea, mimetizatze gaitasuna edo ihes-portaeraren mende daude, (Stephens & Krebs, 1986; Spitz et al., 2014). Harrapakinaren tamaina, gogortasuna edo iheskortasuna bezalako faktoreek errentagarritasunari eragiten diote, eta harrapakin eskuragarritasunak eta harraparien nitxo malgutasunak, berriz, dieta baldintzatzen dute (Araújo et al., 2011). Saguzar intsektujaleen bazka-ekologiaren ikerketa alor konplexua eta aldeaniztuna da, eta askotariko ikerketa-metodoetan oinarritu da. Metodo horiek, analisi dietetikotik hasi eta soinu-analisiaren eta irrati-telemetriaren bidez bazkatzeko portaerak monitorizatzeraino, saguzarren ekologia ulertzeko balio izan dute. Saguzarren morfologiari buruzko ikerketek haien elikatze-ereduei eta baliabideen banaketari buruzko hainbat iragarpen ekarri dituzte, inferentzia ekomorfologiko hauek indartzen dituzten dieta azterketek babestuta (adb., Fenton, 1982). Ikerketa-metodoen aniztasun horrek 10 SARRERA OROKORRA saguzarrak talde desberdinetan sailkatzea ekarri du, bazkatzeko estrategietan, habitatean edo dietan oinarrituta (Allen, 1939; Denzinger et al., 2018; Fenton, 1990, hurrenez hurren). II.2 EKOMORFOLOGIA ETA BAZKA-TALDEAK Saguzarren hegalen forma eta ekokokapen deiak erabiltzen dituzten bazka-nitxo espezifikoetara egokituta daude. Korrelazio honek saguzarrak taldeetan sailkatzeko aukera ematen digu. Adibidez, espazio irekien ehizatzen dutenek maiztasun baxuko ekokokapen dei luze eta frekuentzia banda estua erabiltzen dituzte, distantzia luzeetara iritsi daitezkeenak (Norberg & Rayner, 1987). Hegal zorrotzak, hego-zama eta itxura-ratio handiekin batera, azkar hegan egiteko aukera ematen diete, nahiz eta honek maniobrabilitate txikiagoa dakarren. Aldiz, landarediatik gertu hegan egiten duten saguzarrek, normalean, hego-zama txikiagoa eta hegal borobilduagoak izaten dituzte, beren maniobrabilitate handitzen dutenak (Vaughan, 1959). Saguzar horiek, gainera, ekokokapen deietan aniztasun ikaragarria erakusten dute, ehiza teknikaren arabera. Intsektu hegalariak distantzia txikietan atzematen dituzten saguzarrek, frekuentzia konstantea erabiltzen dute (CF), maiztasun altuagoko maiztasun modulatuko (FM) luzapen laburrekin batera (FM) (Schnitzler, 1984). Dei luze bat edo dei laburren sorta bat igorri dezakete Doppler efektua konpentsatzeko, beren helburuak detektatzeko eta identifikatzeko gai izanez (Schnitzler, 1968). Patroi hau hegan harrapatzen duten (flycatching) saguzarren tipikoa da (Schnitzler et al., 1985). Landarediaren artean ehizatzen diren beste saguzar batzuek, aldiz, maiztasun altu edo baxuko konbinazio motz edo luzeak izan ditzaketen FM deiak erabiltzen dituzte inguruan ezkutatutako harrapakinak aurkitzeko (Simmons & Stein, 1980). Azaleretatik ehizatzen duten saguzarrek (gleaning), bestetik, FM dei laburrak eta intentsitate txikikoak erabiltzen dituzte beren harrapakinen testura eta egiturak bereizteko (Habersetzer & Vogler, 1983). Eredu orokor horietan oinarrituta, Fentonek (1990) bazka-habitataren araberako hiru talde handi egin zituen : espazio zabaleko ehiztariak, ertz eta hutsuneetako ehiztariak eta espazio estuko ehiztariak. Schnitzlerrek eta Kalkok (2001) antzeko sailkapena egin zuten habitataren arabera (zabalak, erdi-itxia, oso itxiak (uncluttered, background-cluttered, highly cluttered), eta habitat bakoitzerako azpikategoriak bereizi zituzten animalien ehiza tekniketan oinarrituta, hala nola saguzar airekoak edo azalerakoak. Oraintsuago, Denzinger eta kideek (2018) bazka-moduetan oinarritutako sailkapena egin zuten, eta horiek erabiltzen duten habitatarekin ere lotuta daude, hala nola, aireko harrapari ibiltariak (hawkers), arraste harrapariak (trawlers), azaleretako harrapari aktiboak, pasiboak eta landaredia trinkoan (clutter) detektatzen duten harrapariak (flutter-detectors). Tradizionalki, talde horiek habitat espezifikoekin lotu izan dira. Horrela, itxurako-ratio baxuak dituzten espezieak baso trinkoekin eta zuhaixka-eremuekin lotu dira, eta itxura-ratio handiak dituztenak, berriz, ingurune irekiekin lotu dira, hala nola belardiekin, landazabalekin edo altitude handiekin (Findley et al., 1972). 11 SARRERA OROKORRA II.3. BAZKALEKUAK Azken hamarkadetan, saguzar intsektujaleen bazka-ekologiari buruz dugun jakintza nabarmen aurreratu da, batez ere habitatari eta eskakizun trofikoei dagokienez. Aurrerapen hori, ikerketa-metodologien eta berrikuntza teknologikoen hobekuntzei dagokie, batez ere, hala nola irrati-telemetriari (adb., Fenton, 1997; Bontadina et al., 2002; Butschkoski, 2005; Dressler et al., 2016; Lagerveld et al., 2017) eta jarraipen akustikoaren hobekuntzari (Vaughan et al., 1997; Russo & Jones, 2003; Frick, 2013; Ancilloto et al., 2023). Ikerketa hauek saguzarren jokabideari eta habitataren erabilerari buruzko ezagutza sakonagoak eman dituzte. Habitat anitz eta kalitate handikoen garrantzia azpimarratzen dute, urtean zehar saguzar intsektujaleentzat intsektuen ugaritasun eta aniztasun aberatsa bermatzen dutenak (Nicholls & Racey, 2006; Jung & Kalko, 2010; Hagen & Sabo, 2016; Straka et al., 2016). Kalitate handiko bazkalekuak identifikatzea saguzarren kontserbazioaren funtsezkoa da (Frick et al., 2024); habitataren galera –basoen ustiaketa eta nekazaritza barne– mundu osoko biodibertsitatearentzat mehatxu nagusia baita (Frick et al., 2020). Hala ere, eginkizun hau erronka bat da. Izan ere, habitatak babesteko eskala eta konfigurazio egokia zehaztea, batez ere saguzarren gaueko mugimendu zabalak kontuan hartuta (McCracken et al., 2016; Goldshtein et al., 2020; O 'Mara et al., 2021), lan konplexua eta zaila da. Bestalde, irrati eta ultrasoinu bidezko ikerketek ere mugak dituzte. Ikerketa hauek neketsuak dira eta langile eta ekipamendu garrantzitsuak behar dituzte, nahiz eta askotan lagin-tamaina txikiak eta jarraipen-iraupen mugatuak lortzen diren (Bontadina et al., 2002; Murray & Kurta, 2004; Smith & Racey, 2005). Horregatik, harrapakinen eskuragarritasunean eta baldintza klimatikoetan izandako aldaketek eragindako saguzarren artean ohikoak diren habitat-erabileren garaikako aldaketak askotan ez dituzte erakusten. Habitatari eta dietari buruzko ikerketa integratua funtsezkoa da, saguzarren habitat-eskakizunei eta haien elikadurari buruzko ikerketei zuzendutako irrati-jarraipeneko ikerketa gehienak ez baitira askotan aldi berean egiten, eta ez baitira eskuragarritasuna eta aukeratzen dutena erakustera iristen (baina ikus Goiti et al., 2008; Flanders & Jones, 2009; Napal et al., 2013). Gainera, artropodoek bizi-ziklo konplexuak dituztenez, habitatak babesteko ahaleginek organismo horien bizi-ziklo osoari eusten dioten habitaten osotasuna bermatu behar dute (Arrizabalaga-Escudero et al., 2015). Horregatik, saguzarren bazka-habitat espezifikoen dependentzia eta haien harrapakin-elementu espezifikoak zehatz-mehatz identifikatzea funtsezkoa da saguzarren populazioak eraginkortasunez babesteko. 12 SARRERA OROKORRA II.4 DIETA Harraparien bazkaren ikerketak, historikoki, dietaren osaera eta aberastasuna aztertu ditu. Ikuspegi horren ondorioz, harrapariak monotipikoak espezializatu edo selektibo gisa sailkatu dira, eta harrapari mota anitzez elikatzen direnak, berriz, harrapari jeneralista edo oportunista gisa (Spitz et al., 2014). Saguzar intsektujaleek jokabide dietetiko ugari izaten dituzte, jeneralistetik oso espezializatuetara. Espezie gehienek dieta malguak dituzte (adb., Jones & Rydell, 2003; Clare et al., 2009; Aizpurua et al., 2013; Aihartza et al., 2023). Hala ere, espezie batzuk intsektu taxoi espezifikoen espezialistatzat sailkatu dira, haien portaerari konplexutasun geruza bat gehituz. Adibidez, Pipistrellus nathusii eltxoez elikatzen da nagusiki (Krüger et al., 2014), eta Barbastella sp. (Sierro & Arlettaz, 1997) eta Rhinolophus euryale (Goiti et al., 2008; Salsamendi et al., 2012; Andreas et al., 2013; Arrizabalaga-Escudero et al., 2015) ezagunak dira sitsak hobesten dituztelako. Gainera, saguzar batzuk, hala nola Scotophilus viridis eta Myotis myotis, kakalardoez elikatzen espezializatuta daudela ikusi da (Fenton et al., 1977; Arlettaz, 1999, hurrenez hurren). Hala eta guztiz ere, espezializazio-maila sasoi edo faktore geografikoen arabera alda daiteke. Adibidez, Scotophilus viridis espeziean sasoiko elikadura aldaketak hauteman dira, sasoi lehorrean dieta zabalagoa izan erakutsiz (Fenton & Thomas, 1980), elikagaia urria denean eskura dituen intsektuak kontsumitzen dituela adieraziz. Myotis emarginatus eskualdearen aldakortasunaren adibide garbia da, nagusiki Alemaniako eta Belgikako behi kortetan ehizatzen baitu (Beck, 1995; Steck & Brinkmann, 2006; Kervyn et al., 2012), baina hegoaldeko eskualdeetan armiarmek dietaren zati handi bat osatzen dute (Goiti et al., 2011; Galan et al., 2017). Interesgarria da, baita kolonia mailan ere Vallejo et al. (2019, 2023) eulien eta armiarmen arteko sasoikako elikadura aldaketa behatu dituztela, harrapakinen eskuragarritasunaren eta errentagarritasunaren arabera. Gainera, ikerketek lehentasun dietetikoak aldatu egin daitezkeela sexuaren (Mata et al., 2016) edo adinaren (Hamilton & Barclay, 1998; Salsamendi et al., 2008; Arrizabalaga-Escudero et al., 2019) arabera. Dietaren aldaketa horiek estrategia aldaketa bat isla dezakete saguzarrek harrapakinen ugaritasunari (Fenton, 1982), energia-eskakizunei (Mata et al., 2016) edo bazka-trebetasunei (Hamilton & Barclay, 1998) erantzuteko. Horregatik, saguzar intsektujaleak espezialista gisa sailkatzea ez da hain xamurra (Fenton, 1982). II.5 METABARCODING BIDEZ IKERKETA TROFIKOETAN IZANDAKO AURRERAPENAK Hamarkada bat baino gehiago igaro da DNA metabarcoding-a gorozkien ohiko analisi morfologikoaren alternatiba gisa sortu zenetik, dieta azterketak irauliz (adb., Clare et al., 2009; Deagle et al., 2009; Pegard et al., 2009). Teknika horri esker, ingurumen-lagin konplexuetan, hala nola lurzoruan, uretan edo erraietan, espezie ugari zehatz identifika daitezke, ohiko metodo morfologikoek baino bereizmen taxonomiko hobea eskainiz (Bohmann et al., 2014). Gainera, azkarragoa da, ez da 13 SARRERA OROKORRA irrati-telemetria eta dietaren analisia, gorozkien azterketa morfologikoaren eta intsektuen eskuragarritasunaren azterketaren bidez. Oraintsuago, Arrizabalaga-Escuderok teknika molekularrak erabili zituen bere doktore-tesian (Arrizabalaga-Escudero, 2016) espezie horren dieta eta ekologia sakonago aztertzeko. Haren aurkikuntzek agerian utzi zuten espeziea paisaia-elementu espezifikoen menpekotasunak eta harrapakinen jatorrizko habitatak babestu beharrak ondorio nabarmenak dituztela saguzarraren kontserbazioan (Arrizabalaga-Escudero et al., 2015). Gainera, saguzar baten dieta ezaugarri funtzionalen ikuspegitik aztertzen lehena izan zen, aldaketa trofiko intraspekifikoa harrapakinen ezaugarrien bidez (kokapena, urtaroa, tamaina, sexua eta adina) ebaluatuz (Arrizabalaga-Escudero et al., 2019). Ferra-saguzar txikia eta handia, azterketa-eremu osoan ohikoak, sakonago ikertu beharrean daude. Lurralde osoan banatuta egon eta batzuetan kolonia handiak sor ditzaketen arren, haien bazka-ekologia ez da ferra-saguzar mediterranearrarena bezain sakon aztertu. Ferra-saguzar handiaren dietari buruzko ikerketa molekular bakarra Frantzian egin da (Tournayre et al., 2021). Aurreko ikerketek gorotzen analisi morfologikoan oinarritu ziren (adb., Jones, 1990; Flanders & Jones, 2009; Koselj, Schnitzler & Siemers, 2011), zehaztasun gutxiagoko emaitzak emanez. Horrela, argi dago espezie horren bazka-ekologian sakontzeko beharra. Aitzitik, azterketa-eremuan ferra-saguzar txikiaren dieta aztertu izan da, baina mahastiekin eta haien izurrite-kontsumoarekin lotura estua izan (Baroja et al., 2019, 2021). Espezie honi buruzko aurreko ikerketak morfologikoak izan ziren ere (adb., McAney & Fairley, 1989; Lino et al., 2014; Mitschunas & Wagner, 2015). Gainera, bi espezieei buruzko ikerketa gehienak beren banaketen iparraldeko ertzean egin dira, eta, beraz, baliteke ez izatea espeziearen funtsezko bazka-ekologiaren guztiz adierazgarri (adb., R. ferrumequinum: Jones, 1990; Jones & Morton, 1992; Duvergé, 1996; Ransome, 1996; Pir, 1994; Dietz, Pir & Hillen, 2013; R. hipposideros: McAney & Fairley, 1988; Arlettaz, 2000; Gody & Boyer, 2002; R. hipposideros: McAney & Fairley, 2002; Goyet, 2002). Beraz, ferra-saguzar txiki eta handien ekologia trofikoa oraindik ikertzeke dago. 20 II. GENERAL INTRODUCTION General Introduction II.1. FORAGING ECOLOGY OF INSECTIVOROUS BATS Bats (order Chiroptera) comprise approximately 1,400 species, making them the second most diverse order of mammals, accounting for nearly a quarter of all described mammal species (Wilson & Mittermeier, 2019). They play a significant role in most terrestrial ecosystems across all continents except Antarctica (Kunz & Pierson, 1994) and are unique among mammals in their evolution of powered flight. The evolution of flight and echolocation has been essential for bats to exploit the diverse range of nocturnal insects (Jones & Rydell, 2003) since about 70% of bats are insectivorous (Simmons, 2005). The high ecological diversity among insectivorous bats is linked to various adaptations closely related to flight and echolocation. For instance, some larger and faster species use low-frequency echolocation calls to hunt airborne insects in open spaces (e.g., Nyctalus lasiopterus; Ibañez & Juste, 2023), while smaller, slower bats are more manoeuvrable and capture their food in cluttered environments (e.g., Myotis bechsteinii; Kerth & van Schaik, 2023). Additionally, insectivorous bats exhibit diverse skull structures, mandibles, and teeth shapes that align with their dietary needs, which affects their bite force and prey. Bats that consume hard-bodied prey have more robust skulls and mandibles, and longer canines, compared to those that eat soft-bodied prey (Freeman, 1979, 1998; Ramírez-Fráncel et al., 2021; Santana et al., 2011, 2022). These species-specific adaptations influence their foraging ecology, determining where and what they can hunt (e.g., Swartz et al., 2003). Foraging is a fundamental activity necessary for the survival of animals. In the case of insectivorous bats, it involves searching for, detecting, and recognising potential prey. It also requires deciding whether to pursue the prey, actually pursue it, catch it, and finally consume it (Stephens & Krebs, 1986). Thus, predator-prey interactions depend on prey profitability and the effectiveness of defensive strategies, such as camouflage, mimicry, or escape behaviour (Stephens & Krebs, 1986; Spitz et al., 2014). Factors like prey size, hardness or evasiveness affect profitability, while prey availability and predators’ niche flexibility shape diet (Araújo et al., 2011). The study of the foraging ecology of insectivorous bats is a complex and multi-faceted field that has relied on a diverse range of research methods. These methods, ranging from dietary analysis to monitoring feeding behaviour through sound analysis and radio-telemetry, have provided a comprehensive understanding of bat ecology. Research on the morphology of bats has led to various predictions about their feeding patterns and resource partitioning, often supported by dietary studies reinforcing these inferences drawn from the ecomorphological analyses (e.g., Fenton, 1982). This diversity of research methods has led to the classification of bats into different guilds based on their foraging strategies, habitat or diet (Allen, 1939; Denzinger et al., 2018; Fenton, 1990, respectively). 23 General Introduction II.2. ECOMORPHOLOGY AND FORAGING GUILDS Bats' wing shape and echolocation calls are adapted to the specific foraging niches they exploit. This correlation allows us to classify bats into groups or guilds. For instance, open-space foragers primarily use long, narrowband echolocation calls at low frequencies that can travel over long distances (reviewed in Norberg & Rayner, 1987). Their pointed wingtips, combined with high wing loading and aspect ratios, enable them to fly quickly, although this comes at the cost of lower manoeuvrability. In contrast, bats that fly near or among clutter typically exhibit lower wing loading and rounder wingtips, which enhance their manoeuvrability (Vaughan, 1959). These bats also show an impressive diversity of echolocation calls depending on their hunting behaviours. Some bats that detect fluttering insects at short range use constant frequency (CF) calls along with brief frequency-modulated (FM) sweeps at higher frequencies (Schnitzler, 1984). They may emit a long call or a series of shorter calls to compensate for the Doppler effect, allowing them to detect and identify their targets (Schnitzler, 1968). This calling pattern is typical of flycatching bats (Schnitzler et al., 1985). Additionally, other bats that fly among clutter utilise FM calls that can include a variety of short or long combinations, with steep or shallow frequencies, to locate prey hidden in their surroundings (Simmons & Stein, 1980). Gleaning bats, which hunt on surfaces, use short, low-intensity FM calls to discriminate their targets' fine textures (Habersetzer & Vogler, 1983). Based on these general patterns, Fenton (1990) made three big groups depending on their foraging habitat: open-space foragers, edge and gap foragers, and narrow-space foragers. Schnitzler and Kalko (2001) made a similar classification based on the habitat (uncluttered, background-cluttered, and highly cluttered), and they differentiated subcategories for each habitat based on the animals' hunting techniques, such as aerial or gleaning bats. More recently, Denzinger et al. (2018) made a classification based on the foraging modes, which are also linked to the habitat they use, as aerial hawkers, trawlers, active gleaners, passive gleaners, and flutter detectors. Traditionally, these guilds have been associated with specific habitats. In this way, species with low wing aspect ratios have been linked to dense forests and shrubby areas, whereas those with high aspect ratios have been linked to open environments such as meadows, moorlands or high altitudes (Findley et al., 1972). II.3. FORAGING AREAS Over the last decades, our understanding of the foraging ecology of insectivorous bats has significantly advanced, especially regarding their habitat and trophic requirements. This progress is primarily attributed to improvements in research methodologies and technological innovations, such as radio telemetry (e.g., Fenton, 1997; Bontadina et al., 2002; Butschkoski, 2005; Dressler et al., 2016; Lagerveld et al., 2017) and improved acoustic monitoring (Vaughan et al., 1997; Russo & Jones, 2003; Frick, 2013; Ancilloto et al., 2023; Kotila et al., 2023). These studies have provided more profound insights into bat behaviour and habitat use. Research highlights the importance of 24 General Introduction diverse and high-quality habitats for insectivorous bats, which support a rich abundance and diversity of insects throughout the year (Nicholls & Racey, 2006; Jung & Kalko, 2010; Hagen & Sabo, 2016; Straka et al., 2016). Identifying high-quality foraging areas is a crucial aspect of bats' conservation (Frick et al., 2024), given that habitat loss—including logging and agriculture—, is the primary threat to bat biodiversity worldwide (Frick et al., 2020). However, this task is not without challenges. Determining the appropriate scale and configuration for habitat protection, especially considering the extensive nightly movements of bats (McCracken et al., 2016; Goldshtein et al., 2020; O’Mara et al., 2021), is a complex and difficult task. On the other hand, radio-tracking and ultrasound-based studies also faced limitations. These studies are labour and resource-intensive, requiring significant field personnel and equipment, often resulting in small sample sizes and limited tracking durations (Bontadina et al., 2002; Murray & Kurta, 2004; Smith & Racey, 2005). Therefore, they barely reveal the seasonal shifts in habitat use, common among bats, driven by changes in prey availability and climatic conditions. The need for integrated research on habitat and diet is essential, as most radio-tracking studies targeting the habitat requirements of bats and their diet studies are not often simultaneously carried out and fail to address what the animals have available in their foraging grounds and what they select there (but see Goiti et al., 2008; Flanders & Jones, 2009; Napal et al., 2013). Furthermore, because arthropods have complex life cycles, habitat protection efforts must ensure the integrity of various habitats that support these organisms' entire life cycle (Arrizabalaga-Escudero et al., 2015). Therefore, accurately identifying the specific foraging habitats that bats depend on and their specific prey items is crucial for effectively protecting bat populations. II.4. DIET The study of predators' foraging has historically focused on analysing the species composition and richness of prey in their diets. This approach has led to the classification of monotypic predators as specialised or selective, while those that feed on a wide variety of prey species are categorised as generalist or opportunistic predators (Spitz et al., 2014). Insectivorous bats exhibit a broad array of dietary behaviours, ranging from generalist to highly specialised. Most species display considerable diet flexibility (e.g., Jones & Rydell, 2003; Clare et al., 2009; Aizpurua et al., 2013; Aihartza et al., 2023). However, some species have been addressed as specialised in consuming specific insect taxa, adding a layer of complexity to their behaviour. For instance, Pipistrellus nathusii primarily feeds on mosquitoes (Krüger et al., 2014), while Barbastella sp. (Sierro & Arlettaz, 1997) and Rhinolophus euryale (Goiti et al., 2008; Salsamendi et al., 2012; Andreas et al., 2013; Arrizabalaga-Escudero et al., 2015) are known for their preference for moths. 25 General Introduction Additionally, some bats, such as Scotophilus viridis and Myotis myotis, have been observed to specialise in feeding on beetles (Fenton et al., 1977; Arlettaz, 1999, respectively). Nonetheless, the specialisation level can change based on seasonal or geographical factors. For instance, seasonal dietary shifts have also been noted in Scotophilus viridis, which tends to have a broader diet during the dry season (Fenton & Thomas, 1980), indicating that it consumes whatever available insects it can find when food is scarce. Myotis emarginatus is a clear example of regional variability since it primarily hunts flies in cowsheds in Germany and Belgium (Beck, 1995; Steck & Brinkmann, 2006; Kervyn et al., 2012), but in southern regions, spiders make up a significant portion of their diet (Goiti et al., 2011; Galan et al., 2017). Interestingly, even within the same colonies, Vallejo et al. (2019, 2023) have observed seasonal shifts between flies and spiders depending on prey availability and profitability. Additionally, studies have shown that dietary preferences can vary between individuals based on factors such as sex (Mata et al., 2016) or age (Hamilton & Barclay, 1998; Salsamendi et al., 2008; Arrizabalaga-Escudero et al., 2019). These changes in diet may reflect a change in strategy by the bats responding to variations in prey abundance (Fenton, 1982), differing energy demands (Mata et al., 2016), or differences in foraging skills (Hamilton & Barclay, 1998). Therefore, it is challenging to classify insectivorous bats strictly as dietary specialists (Fenton, 1982). II.5. ADVANCES IN TROPHIC STUDIES THROUGH METABARCODING More than a decade has passed since DNA metabarcoding emerged as a ground-breaking alternative to traditional morphological analysis of faeces, revolutionising diet studies (e.g., Clare et al., 2009; Deagle et al., 2009; Pegard et al., 2009). This technique enables the precise identification of multiple species in complex environmental samples, such as soil, water, or gut contents, offering finer taxonomic resolution than traditional morphological methods (reviewed in Bohmann et al., 2014). Additionally, it is faster, does not rely on expert taxonomic skills, and can identify fragments that are difficult to classify morphologically (reviewed in Pompanon et al., 2012). Since the pioneering work by Clare et al. (2009), the development of arthropod-specific short primers for diet analyses (e.g., Zeale et al., 2011) and advancements in Next Generation Sequencing Technologies (reviewed in Pompanon et al., 2012) have enhanced the resolution and cost-effectiveness of diet analyses, despite some limitations (Clare, 2014; Alberdi et al., 2018). Therefore, bat dietary studies have also been enriched from this methodology (e.g., Whitaker & Karataş, 2009; Clare et al., 2011; Bohman et al., 2011; Razgour et al., 2011; Zeale et al., 2011). The rapid advancement of metabarcoding has led to the development of several primer sets targeting arthropods (e.g., Zeale et al., 2011; Elbrecht & Leese, 2017; Galan et al., 2017; Vamos et al., 2017; Wangensteen et al., 2018; Jusino et al., 2019). However, researchers have shown that primer sets often exhibit biases toward certain arthropod orders (Clarke et al., 2014; Piñol et al., 2015; Jusino et al., 26 General Introduction 2019; Krehenwinkel et al., 2017; Elbrecht et al., 2019). As a result, there has been a trend toward designing more degenerated primers (Krehenwinkel et al., 2017; Vamos et al., 2017; Jusino et al., 2019) to enable the amplification of DNA from a broader range of arthropod groups. Additionally, combining different primer sets may be a more effective approach to minimise biases associated with individual primer sets. This strategy improves dietary coverage, providing a better understanding of predators' ecology (Esnaola et al., 2018; Jusino et al., 2019; Penner et al., 2024). Moreover, sequencing depth is also crucial to this technique (Alberdi et al., 2018). A few years ago, various sequencing platforms were utilised; however, following the results of several studies comparing the performance of these platforms (Loman et al., 2012; Mak et al., 2017; Taberlet et al., 2018), the use of Illumina became predominant. Recent advancements in sequencing technology have enabled us to achieve up to 400 million reads per run with the NovaSeq system (Illumina), in contrast to a maximum of 25 million reads per run with the older MiSeq system (Illumina). These improvements facilitate more comprehensive metabarcoding studies, offering more profound insights into ecology by allowing for increased sampling sizes and identifying more prey items (Caporaso et al., 2012; Han et al., 2024). II.6. BEYOND TAXONOMIC ANALYSIS OF DIET Prey-predator relationships are often examined primarily through a taxonomic lens, focusing on which predator feeds on which taxa without considering the characteristics of the prey (Spitz et al., 2014). Many molecular studies on bat diets have primarily been limited to providing extensive lists of prey (e.g., Bohman et al., 2011; Clare et al., 2013; Rolfe et al., 2014; Van den Bussche, 2016; Gordon et al., 2019; Wray et al., 2020; O’Rourke et al., 2021). In contrast, analysing diets with a broader, more ambitious perspective—incorporating a focused sampling design while also considering prey characteristics, predators, environment, and seasonal variations—can greatly enhance our understanding of bat ecology. Molecular studies have shown that bats consume hundreds of species, which vary over time (Razgour et al., 2011; Arrizabalaga‐Escudero et al., 2015; Aihartza et al., 2023); space (Clare et al., 2014); and among individuals (Mata et al., 2016), allowing them to adjust their diet based on availability (e.g., McCracken et al., 2012; Almenar et al., 2013; Napal et al., 2013; Aihartza et al., 2023; Vallejo et al., 2023). Metabarcoding studies have also shed light on the role of niche partitioning among co-occurring sibling species (e.g., Arrizabalaga-Escudero et al., 2018; Novella-Fernandez et al., 2020; Andriollo et al., 2021; Blanch et al., 2023), and have even identified many agroforestry pest species, unveiling that bats may act as pest suppressors (e.g., Aizpurua et al., 2017; Baroja et al., 2019; Andriollo et al., 2021; Liu et al., 2023). However, accurate assessments of foraging ecology in predators like insectivorous bats require more than a taxonomic listing. It is essential to move beyond and address the question: Why a prey is a prey? This approach involves understanding the functional relationships between predators and prey. 27 General Introduction Insectivorous bats feed on a wide variety of insects, which differ in size, flight patterns, hardness and evasive behaviours, traits that likely influence their profitability to bats (Schnitzler, 1987). This variability implies that predators must be able to distinguish between prey types of varying profitability (i.e. energy gained per time unit of handling: detecting, pursuing, capturing and consuming) and adjust prey targets when prey availability changes to ensure an average positive energy balance. Prey profitability may vary with environmental conditions (e.g., prey abundance) or predator-specific requirements (e.g., breeding season, sex‐specific energetic requirements or individual differences in prey-capturing skills) (Stephens & Krebs, 1986). For instance, male and female Tadarida teniotis bats differ in the type of moths they consume (Mata et al., 2016). Similarly, differences between adult and juveniles have been reported in Eptesicus fuscus (Hamilton & Barclay, 1998) or Rhinolophus euryale (Arrizabalaga-Escudero et al., 2019) and suggested that flight and echolocation experience may be causing such differences. Thus, what factors influence their dietary adjustments? The characteristics of their prey are more significant than taxonomic classification. Trait-based approaches, which connect predator and prey morphology and ecology, offer a deeper insight into the factors that shape prey selection (Spitz et al., 2014). Pioneering this field on bats, Arrizabalaga-Escudero et al. (2019) applied trait-based methods to explore the trophic ecology of horseshoe bats. The study examined how the energetic needs of males, females in different breeding stages, and juveniles versus adults influenced their choices of prey size or flight characteristics. Other research has also linked prey size, habitat, seasonality, and the bats' body condition to their foraging behaviour (Ancilloto et al., 2023). This expanding field of study highlights the complexity of bat foraging strategies and the crucial role of prey traits in shaping their dietary choices. Moreover, the precise taxonomic resolution provided by metabarcoding enables linking prey species to their habitats (Clare et al., 2011; Razgour et al., 2011; Alberdi et al., 2012). While it cannot determine where the consumed prey were captured (sink habitat), it can indicate the origin of those prey (source habitat) (Arrizabalaga-Escudero et al., 2015). Consequently, molecular diet analysis has become paramount for determining the foraging dependencies of bats by identifying the habitats where they hunt or the sources of their prey throughout the year. Notably, studies on molecular diets suggest that foraging guilds are more flexible than previously thought. For instance, open-forager bat species also hunt in —or above, or around— various habitats, including forests, in response to prey availability peaks (Garin et al., 2019; Aihartza et al., 2023). Similarly, clutter-specialist bats will likely take advantage of prey outbreaks in different habitats, using their echolocation and flight skills at the microhabitat level by navigating nearby vegetation. 28 General Introduction II.7. ECOSYSTEM SERVICES DNA metabarcoding has revealed that bats, with their diverse diet of pest arthropods, play a crucial role in responding to outbreaks (McCracken et al., 2012; Charbonnier et al., 2014; Puig-Montserrat et al., 2015; Aizpurua et al., 2018; Krauel et al., 2018; Baroja et al., 2019, 2021; Garin et al., 2019; Korine et al., 2020; Andriollo et al., 2021; Liu et al., 2023). In large-scale monocultures, insect pests are a significant part of their diet (Symondson et al., 2002; Segoli & Rosenheim, 2012), and exclosure experiments have confirmed their role in reducing agricultural pests (Maas et al., 2013, 2016; Linden et al., 2019; Ancilloto et al., 2024; Tuneu-Corral et al., 2024). This underscores the potential of bats in sustainable pest management, offering a ray of hope for the future of agriculture. However, most studies rely on presence/absence data (Aizpurua et al., 2018; Baroja et al., 2019; Weier et al., 2019). It is important to note that metabarcoding biases can lead to false negatives due to differential primer affinity and competition among DNA sequences (Pompanon et al., 2012; Evans et al., 2016). To address this, targeted approaches, such as species-specific primers with PCR-based methods, offer greater precision. Conventional PCR (cPCR) and quantitative PCR (qPCR) can detect minimum DNA quantities (Jarman et al., 2004) but are limited to single target sequences. Multiplex PCR allows amplification of multiple sequences (Chamberlain et al., 1988), but the use of different primer pairs may lead to amplification bias for certain DNA templates (Mustorp et al., 2011). The PCR-based multiplex ligation-dependent probe amplification (MLPA) technique enhances multiplexing by using a single primer pair to amplify specific probes, improving quantification accuracy (Schouten et al., 2002). Despite the challenges in faecal DNA analysis (e.g., Piñol et al., 2015; Alberdi et al., 2018; Deagle et al., 2019), studies have estimated bat consumption rates (McCracken et al., 2012; Aihartza et al., 2023) and their economic value in pest suppression (reviewed by Tuneu-Corral et al., 2023). Although seasonal and landscape variations affect estimates, understanding bat diets in agroecosystems is crucial for assessing their ecosystem services (Russo et al., 2018). II.8. FORAGING ECOLOGY OF HORSESHOE BATS Horseshoe bats (Rhinolophidae Gray, 1825) are a diverse family of insectivorous bats known for their distinctive horseshoe-shaped noseleaves, which play a crucial role in echolocation. These nocturnal mammals rely on high-frequency calls to navigate and hunt in complex environments. Distributed across the southern Paleartic region of the Old World, they are cave-dwelling species particularly abundant in karst areas, where underground roosts are abundant (Csorba et al., 2003). They mainly use subterranean roosts, either natural (caves) or artificial (mines, tunnels, etc.), but also inhabit buildings, especially for maternity colonies (Russo et al., 2023). In Europe, they are pretty common in areas such as the Northern Iberian Peninsula (Palomo et al., 2007), France (Arthur & Lemaire, 2021), Italy (Agnelli et al., 2004; GIRC, 2004), and the Balkans (e.g., Tvrtkovic, 2006; Kryštufek & Donev, 29 General Introduction Goiti, U., Aihartza, J., & Garin, I. (2004). Diet and prey selection in the Mediterranean horseshoe bat Rhinolophus euryale (Chiroptera, Rhinolophidae) during the pre-breeding season. Mammalia, 68(4), 397-402. https://doi.org/10.1515/mamm.2004.039 Goiti, U., Aihartza, J. R., Garin, I., & Zabala, J. (2003). Influence of habitat on the foraging behaviour of the Mediterranean horseshoe bat, Rhinolophus euryale. Acta Chiropterologica, 5(1), 75-84. Goiti, U., Aihartza, J. R., Almenar, D., Salsamendi, E., & Garin, I. (2006). Seasonal foraging by Rhinolophus euryale (Rhinolophidae) in an Atlantic rural landscape in northern Iberian Peninsula. Acta Chiropterologica, 8(1), 141-155. Goiti, U., Aihartza, J., Guiu, M., Salsamendi, E., Almenar, D., Napal, M., & Garin, I. (2011). Geoffroy's bat, Myotis emarginatus, preys preferentially on spiders in multistratified dense habitats: a study of foraging bats in the Mediterranean. Folia Zoologica, 60(1), 17-24. Goiti, U., Garin, I., Almenar, D., Salsamendi, E., & Aihartza, J. (2008). Foraging by Mediterranean horseshoe bats (Rhinolophus euryale) in relation to prey distribution and edge habitat. Journal of Mammalogy, 89(2), 493-502. https://doi.org/10.1644/07-MAMM-A-054R2.1 Goldshtein, A., Handel, M., Eitan, O., Bonstein, A., Shaler, T., Collet, S., ... & Yovel, Y. (2020). Reinforcement learning enables resource partitioning in foraging bats. Current Biology, 30(20), 4096-4102. Gordon, R., Ivens, S., Ammerman, L. K., Fenton, M. B., Littlefair, J. E., Ratcliffe, J. M., & Clare, E. L. (2019). Molecular diet analysis finds an insectivorous desert bat community dominated by resource sharing despite diverse echolocation and foraging strategies. Ecology and Evolution, 9(6), 3117-3129. Gruppo Italiano Ricerca Chirotteri (GIRC) (2004). The Italian bat roost project: a preliminary inventory of sites and conservation perspectives. Hystrix, 15(2), 55-68. Habersetzer, J., & Vogler, B. (1983). Discrimination of surface-structured targets by the echolocating bat Myotis myotis during flight. Journal of comparative physiology, 152, 275-282. Hagen, E. M., & Sabo, J. L. (2011). A landscape perspective on bat foraging ecology along rivers: does channel confinement and insect availability influence the response of bats to aquatic resources in riverine landscapes?. Oecologia, 166(3), 751-760. Hamilton, I. M., & Barclay, R. M. (1998). Diets of juvenile, yearling, and adult big brown bats (Eptesicus fuscus) in southeastern Alberta. Journal of Mammalogy, 79(3), 764-771. Han, H., Lee, H. J., Kim, K. S., Chung, J., & Na, H. S. (2024). Comparison of the performance of MiSeq and NovaSeq in oral microbiome study. Journal of Oral Microbiology, 16(1), 2344293. Ibañez, C., & Juste, J. (2023). Greater Noctule Bat Nyctalus lasiopterus (Schreber, 1780). In D. Russo (Ed.), Chiroptera, Handbook of the Mammals of Europe (pp. 439-462). Springer Nature Switzerland. Ikeda, Y., Jiang, T., Oh, H., Csorba, G., & Motokawa, M. (2020). Geographic variations of skull morphology in the Rhinolophus ferrumequinum species complex (Mammalia: Chiroptera). Zoologischer Anzeiger, 288, 125-138. Jarman, S. N., Deagle, B. E., & Gales, N. J. (2004). Group‐specific polymerase chain reaction for DNA‐based analysis of species diversity and identity in dietary samples. Molecular Ecology, 13(5), 1313-1322. Jones, G. (1990). Prey selection by the greater horseshoe bat (Rhinolophus ferrumequinum): optimal foraging by echolocation?. The Journal of Animal Ecology, 587-602. Jones, G., Gordon, T., & Nightingale, J. (1992). Sex and age differences in the echolocation calls of the lesser horseshoe bat, Rhinolophus hipposideros. Mammalia, 56(2), 189-194. Jones, G., & Morton, M. (1992). Radio-tracking studies on habitat use by greater horseshoe bats (Rhinolophus ferrumequinum). In Wildlife Telemetry. Remote Monitoring and Tracking of Animals (pp. 521-537). Jones, G., & Rayner, J. M. (1989). Foraging behavior and echolocation of wild horseshoe bats Rhinolophus ferrumequinum and R. hipposideros (Chiroptera, Rhinolophidae). Behavioral Ecology and Sociobiology, 25, 183-191. Jones, G., & Rydell, J. (2003). Attack and Defense: Interactions between Echolocating Bats and Their Insect Prey. In T. H. Kunz, & M. B. Fenton (Eds), Bat Ecology (pp. 301-345). The University Chicago Press, Ltd., London. Jung, K., & Kalko, E. K. V. (2010). Where forest meets urbanization: foraging plasticity of aerial insectivorous bats in an anthropogenically altered environment. Journal of Mammalogy. Jusino, M. A., Banik, M. T., Palmer, J. M., Wray, A. K., Xiao, L., Pelton, E., Barber, J. R., Kawahara, A. Y., Gratton, C., Peery, M. Z., & Lindner, D. L. (2019). An improved method for utilizing high‐throughput amplicon sequencing to determine the diets of insectivorous animals. Molecular ecology resources, 19(1), 176-190. Kay, L., & Pickvance, T. J. (1963). Ultrasonic emissions of the lesser horseshoe bat Rhinolophus hipposideros (Bech.). In Proceedings of the Zoological Society of London (Vol. 141, No. 1, pp. 163-171). Oxford, UK: Blackwell Publishing Ltd. 36 General Introduction Kerth, G., & van Schaik, J. (2023). Bechstein’s Bat Myotis bechsteinii (Kuhl, 1817). In D. Russo (Ed.), Chiroptera, Handbook of the Mammals of Europe (pp. 113-135). Springer Nature Switzerland. Kervyn, T., Godin, M. C., Jocqué, R., Grootaert, P., & Libois, R. (2012). Web-building spiders and blood-feeding flies as prey of the notch-eared bat (Myotis emarginatus). Belgian Journal of Zoology, 142(1). Kober, R., & Schnitzler, H. U. (1990). Information in sonar echoes of fluttering insects available for echolocating bats. The Journal of the Acoustical Society of America, 87(2), 882-896. Koh, H. S., Jo, J. E., Oh, J. G., Kweon, G. H., Ahn, N. H., Sin, W. H., & Sin, D. S. (2014). Little genetic divergence of the greater horseshoe bat Rhinolophus ferrumequinum from far-eastern Asia, with a preliminary report on genetic differentiation of R. ferrumequinum from Eurasia and northern Africa examined from cytochrome b sequences. Russian Journal of Theriology, 13(2), 97-103. Korine, C., Niv, A., Axelrod, M., & Dahan, T. (2020). Species richness and activity of insectivorous bats in cotton fields in semi-arid and mesic Mediterranean agroecosystems. Mammalian Biology, 100(1), 73-80. Koselj, K. (2002). Diet and ecology of the Mediterranean horseshoe bat (Rhinolophus euryale Blasius, 1853; Mammalia: Chiroptera) in south-eastern Slovenia. Graduation thesis, University of Ljubljana, Biothechnical Faculty, Department of Biology, Ljubljana: IX. Koselj, K., Schnitzler, H. U., & Siemers, B. M. (2011). Horseshoe bats make adaptive prey-selection decisions, informed by echo cues. Proceedings of the Royal Society B: Biological Sciences, 278(1721), 3034-3041. Kotila, M., Suominen, K. M., Vasko, V. V., Blomberg, A. S., Lehikoinen, A., Andersson, T., ... & Lilley, T. M. (2023). Large‐scale long‐term passive‐acoustic monitoring reveals spatio‐temporal activity patterns of boreal bats. Ecography, 2023(6), e06617. Krauel, J. J., Brown, V. A., Westbrook, J. K., & McCracken, G. F. (2018). Predator–prey interaction reveals local effects of high-altitude insect migration. Oecologia, 186, 49-58. Krüger, F., Clare, E. L., Greif, S., Siemers, B. M., Symondson, W. O. C., & Sommer, R. S. (2014). An integrative approach to detect subtle trophic niche differentiation in the sympatric trawling bat species Myotis dasycneme and Myotis daubentonii. Molecular Ecology, 23(15), 3657-3671. Kunz, T.H., & Pierson, E.D. (1994). Bats of the World: An Introduction. In R. M. Nowak (Ed.), Walker’s Bats of the World (pp. 1-46). The Johns Hopkins University Press, Baltimore, MD, USA. Kryštufek, B., & Donev, N. R. (2005). The atlas of Slovenian bats (Chiroptera). Atlas netopirjev Slovenije (Chiroptera). Scopolia, 55, 1-92. Lagerveld, S., Janssen, R., Manshanden, J., Haarsma, A. J., de Vries, S., Brabant, R., & Scholl, M. (2017). Telemetry for migratory bats: a feasibility study (No. C011/17). Wageningen Marine Research. Le Moal, T. (2007). Le rhinolophe euryale (Rhinolophus euryale) en vallee des Aldudes. SFPEM, Toulouse. Lino, A., Fonseca, C., Goiti, U., & Pereira, M. J. R. (2014). Prey selection by Rhinolophus hipposideros (Chiroptera, Rhinolophidae) in a modified forest in Southwest Europe. Acta Chiropterologica, 16(1), 75-83. Linden, V. M., Grass, I., Joubert, E., Tscharntke, T., Weier, S. M., & Taylor, P. J. (2019). Ecosystem services and disservices by birds, bats and monkeys change with macadamia landscape heterogeneity. Journal of Applied Ecology, 56(8), 2069-2078. Liu, L., Li, Y., Li, S., Hu, N., He, Y., Pong, R., ... & Law, M. (2012). Comparison of next‐generation sequencing systems. BioMed research international, 2012(1), 251364. Liu, Y., Si, M., Huang, Z., Feng, J., & Jiang, T. (2023). Bats are sentinels for invasive pest surveillance based on DNA metabarcoding. Ecological Indicators, 152, 110354. Loman, N. J., Misra, R. V., Dallman, T. J., Constantinidou, C., Gharbia, S. E., Wain, J., & Pallen, M. J. (2012). Performance comparison of benchtop high-throughput sequencing platforms. Nature biotechnology, 30(5), 434-439. Maas, B., Clough, Y., & Tscharntke, T. (2013). Bats and birds increase crop yield in tropical agroforestry landscapes. Ecology letters, 16(12), 1480-1487. Maas, B., Karp, D. S., Bumrungsri, S., Darras, K., Gonthier, D., Huang, J. C. C., ... & Williams‐Guillén, K. (2016). Bird and bat predation services in tropical forests and agroforestry landscapes. Biological Reviews, 91(4), 1081-1101. Mak, S. S. T., Gopalakrishnan, S., Carøe, C., & Geng, C. (2017). Comparativeperformance of the BGISEQ-500 vs Illumina HiSeq2500 sequencingplatforms for palaeogenomic sequencing. GigaScience, 6(8), 1–13. Mata, V. A., Amorim, F., Corley, M. F., McCracken, G. F., Rebelo, H., & Beja, P. (2016). Female dietary bias towards large migratory moths in the European free-tailed bat (Tadarida teniotis). Biology letters, 12(3), 20150988. McAney, C. M., & Fairley, J. S. (1988). Habitat preference and overnight and seasonal variation in the foraging activity of lesser horseshoe bats. Acta theriologica, 33(28), 393-402. 37 General Introduction McAney, C. M., & Fairley, J. S. (1989). Analysis of the diet of the lesser horseshoe bat Rhinolophus hipposideros in the West of Ireland. Journal of Zoology, 217(3), 491-498. McCracken, G. F., Safi, K., Kunz, T. H., Dechmann, D. K., Swartz, S. M., & Wikelski, M. (2016). Airplane tracking documents the fastest flight speeds recorded for bats. Royal Society Open Science, 3(11), 160398. McCracken, G. F., Westbrook, J. K., Brown, V. A., Eldridge, M., Federico, P., & Kunz, T. H. (2012). Bats track and exploit changes in insect pest populations. PLoS ONE, 7(8): e43839. Mitschunas, N., & Wagner, M. (2015). Diet of the lesser horseshoe bat (Rhinolophus hipposideros) in Central Germany and its seasonal and site-specific variation. Acta Chiropterologica, 17(2), 379-392. Motte, G., & Libois, R. (2002). Conservation of the lesser horseshoe bat (Rhinolophus hipposideros Bechstein, 1800) (Mammalia: Chiroptera) in Belgium. A case study of feeding habitat requirements. Belgian Journal of Zoology, 132(1). Murray, S. W. and Kurta, A. (2004). Nocturnal activity of the endangered indiana bat (Myotis sodalis). Journal of Zoology, 262(2), 197-206. Mustorp, S. L., Drømtorp, S. M., & Holck, A. L. (2011). Multiplex, quantitative, ligation-dependent probe amplification for determination of allergens in food. Journal of agricultural and food chemistry, 59(10), 5231-5239. Napal, M., Garin, I., Goiti, U., Salsamendi, E., & Aihartza, J. (2013). Past deforestation of Mediterranean Europe explains the present distribution of the strict forest dweller Myotis bechsteinii. Forest Ecology and Management, 293, 161-170. Nemoz, M. (2007) Etude de l’activité et des habitats de chasse des rhinolophes euryales de la colonie de Magnagues (Lot, France). SFPEM, Toulouse. Nicholls, B., & Racey, P. A. (2006). Habitat selection as a mechanism of resource partitioning in two cryptic bat species Pipistrellus pipistrellus and Pipistrellus pygmaeus. Ecography, 29(5), 697-708. https://doi.org/10.1111/j.2006.0906-7590.04575.x Norberg, U. M., & Rayner, J. M. (1987). Ecological morphology and flight in bats (Mammalia; Chiroptera): wing adaptations, flight performance, foraging strategy and echolocation. Philosophical Transactions of the Royal Society of London. B, Biological Sciences, 316(1179), 335-427. Novella‐Fernandez, R., Ibañez, C., Juste, J., Clare, E. L., Doncaster, C. P., & Razgour, O. (2020). Trophic resource partitioning drives fine‐scale coexistence in cryptic bat species. Ecology and evolution, 10(24), 14122-14136. O’Mara, M. T., Amorim, F., Scacco, M., McCracken, G. F., Safi, K., Mata, V., Tomé, R., Swartz, S., Wikelski, M., Beja, P., Rebelo, H., & Dechmann, D. K. (2021). Bats use topography and nocturnal updrafts to fly high and fast. Current Biology, 31(6), 1311-1316. O'Rourke, D.R., Mangan, M.T., Mangan, K.E., Bokulich, N.A., MacManes, M.D. & Foster, J. T. (2021). Lord of the Diptera (and Moths and a Spider): Molecular Diet Analyses and Foraging Ecology of Indiana Bats in Illinois. Frontier in Ecology and Evolution. 9:623655. doi: 10.3389/fevo.2021.623655 Palomo, L. J., Gisbert, J., & Blanco, J. C. (2007). Atlas y Libro Rojo de los Mamíferos Terrestres de España. Dirección General para la Biodiversidad -SECEM-SECEMU, Madrid. Pegard, A., Miquel, C., Valentini, A., Coissac, E., Bouvier, F., François, D., Taberlet, P., Engel, E., & Pompanon, F. (2009). Universal DNA-based methods for assessing the diet of grazing livestock and wildlife from feces. Journal of agricultural and food chemistry, 57(13), 5700-5706. Penner, F.V., Lynggaard, C., Argolo, L.A., da Silva, Y.K.R., de Oliveira, G.C., Bohmann, K., & Gastauer, M. (2024). Navigating diversity: primer impact on arthropod recovery in molecular environmental monitoring in tropical Amazon. Front. Ecol. Evol. 12:1411388. Piñol, J., Mir, G., Gomez‐Polo, P., & Agustí, N. (2015). Universal and blocking primer mismatches limit the use of high‐throughput DNA sequencing for the quantitative metabarcoding of arthropods. Molecular ecology resources, 15(4), 819-830. Pir, J. (1994). Etho-oÈkologische Untersuchung einer Wochenstubenkolonie der Grossen Hufeisennase (Rhinolophus ferrumequinum, Schreber 1774) in Luxemburg. Unpubl. Diplom arbeit. Justus-Liebig-Univ. Giessen. Piraccini, R. 2016. Rhinolophus ferrumequinum. The IUCN Red List of Threatened Species 2016. https://www.iucnredlist.org/species/19517/21973253. Accessed on 12 December 2024. Pompanon, F., Deagle, B. E., Symondson, W. O., Brown, D. S., Jarman, S. N., & Taberlet, P. (2012). Who is eating what: diet assessment using next generation sequencing. Molecular ecology, 21(8), 1931-1950. Presetnik, P., Paunović, M., Karapandža, B., Đurović, M., Ivanović, Č., Ždralević, M., Benda, P., & Budinski, I. (2014). Distribution of bats (Chiroptera) in Montenegro. Vespertilio, 17, 129-156. 38 General Introduction Puig-Montserrat, X., Torre, I., López-Baucells, A., Guerrieri, E., Monti, M. M., Ràfols-García, R., ... & Flaquer, C. (2015). Pest control service provided by bats in Mediterranean rice paddies: linking agroecosystems structure to ecological functions. Mammalian Biology, 80(3), 237-245. Ramírez-Fráncel, L. A., García-Herrera, L. V., Losada-Prado, S., Reinoso-Flórez, G., Lim, B. K., Sánchez, F., ... & Guevara, G. (2021). Skull morphology, bite force, and diet in insectivorous bats from tropical dry forests in Colombia. Biology, 10(10), 1012. Ransome, R.D. (1991). Lesser horseshoe bat. In G. Corbet & S. Harris (Eds.). The handbook of British mammals (3rd edition, pp. 95-97). Blackwell Scientific Publications, Oxford. Ransome, R. D. (1996). The management of feeding areas for greater horseshoe bats. English Nature Research Reports (174). Peterborough, UK: English Nature. Ransome, R. D. (2023). Greater Horseshoe Bat Rhinolophus ferrumequinum (Schreber, 1774). In D. Russo (Ed.), Chiroptera, Handbook of the Mammals of Europe (pp. 839-874). Springer Nature Switzerland. Razgour, O., Clare, E. L., Zeale, M. R., Hanmer, J., Schnell, I. B., Rasmussen, M., Gilbert, T., & Jones, G. (2011). High‐throughput sequencing offers insight into mechanisms of resource partitioning in cryptic bat species. Ecology and evolution, 1(4), 556-570. Rolfe, A. K., Kurta, A., & Clemans, D. L. (2014). Species-level analysis of diets of two mormoopid bats from Puerto Rico. Journal of Mammalogy, 95(3), 587-596. Rossiter, S. J., Benda, P., Dietz, C., Zhang, S., & Jones, G. (2007). Rangewide phylogeography in the greater horseshoe bat inferred from microsatellites: implications for population history, taxonomy and conservation. Molecular Ecology, 16(22), 4699-4714. Russ, J. (Ed.). (2021). Bat calls of Britain and Europe: A guide to species identification. Pelagic Publishing Ltd. Russo, D. (2023). Chiroptera, Handbook of the Mammals of Europe. Springer Nature Switzerland. Russo, D., Almenar, D., Aihartza, J., Goiti, U., Salsamendi, E., & Garin, I. (2005). Habitat selection in sympatric Rhinolophus mehelyi and R. euryale (Mammalia: Chiroptera). What mechanism of niche segregation allows the coexistence of sympatric sibling rhinolophid bats? Journal of Zoology, 266, 327–332. Russo, D., Bosso, L., & Ancillotto, L. (2018). Novel perspectives on bat insectivory highlight the value of this ecosystem service in farmland: Research frontiers and management implications. Agriculture, Ecosystems & Environment, 266, 31-38. Russo, D., & Jones, G. (2002). Identification of twenty‐two bat species (Mammalia: Chiroptera) from Italy by analysis of time‐expanded recordings of echolocation calls. Journal of Zoology, 258(1), 91-103. Russo, D., & Jones, G. (2003). Use of foraging habitats by bats in a Mediterranean area determined by acoustic surveys: conservation implications. Ecography, 26(2), 197-209. Russo, D., Jones, G., & Migliozzi, A. (2002). Habitat selection by the Mediterranean horseshoe bat, Rhinolophus euryale (Chiroptera: Rhinolophidae) in a rural area of southern Italy and implications for conservation. Biological Conservation, 107(1), 71-81. Salsamendi, E., Garin, I., Almenar, D., Goiti, U., Napal, M., & Aihartza, J. (2008). Diet and prey selection in Mehelyi's horseshoe bat Rhinolophus mehelyi (Chiroptera, Rhinolophidae) in the south-western Iberian Peninsula. Acta chiropterologica, 10(2), 279-286. Salsamendi, E., Garin, I., Arostegui, I., Goiti, U., & Aihartza, J. (2012). What mechanism of niche segregation allows the coexistence of sympatric sibling rhinolophid bats?. Frontiers in Zoology, 9, 1-12. Sano, A. (2015). Rhinolophus ferrumequinum (Schreber, 1774). In Ohdachi, S.D., Ishibashi, Y., Iwasa, M.A., Fukui, D., Saitoh, T. (Eds.), The Wild Mammals of Japan, second ed. (pp. 58-59) Shoukadoh Book Sellers, Kyoto. Santana, S. E., Grossnickle, D. M., Sadier, A., Patterson, E., & Sears, K. E. (2022). Bat dentitions: a model system for studies at the interface of development, biomechanics, and evolution. Integrative and Comparative Biology, 62(3), 762-773. Santana, S. E., Strait, S., & Dumont, E. R. (2011). The better to eat you with: functional correlates of tooth structure in bats. Functional Ecology, 25(4), 839-847. Schnitzler, H. U. (1968). The ultrasonic sounds of horseshoe bats (Chiroptera-Rhinolophidae) in different orientation situations. Zeitschrift für vergleichende Physiologie, 57, 376-408. Schnitzler, H. U. (1984). The performance of bat sonar systems. In Localization and orientation in biology and engineering (pp. 211-224). Berlin, Heidelberg: Springer Berlin Heidelberg. Schnitzler, H. U., Hackbarth, H., Heilmann, U., & Herbert, H. (1985). Echolocation behavior of rufous horseshoe bats hunting for insects in the flycatcher-style. Journal of Comparative Physiology A, 157, 39-46. 39 General Introduction Schnitzler, H. U., & Kalko, E. K. (2001). Echolocation by insect-eating bats: we define four distinct functional groups of bats and find differences in signal structure that correlate with the typical echolocation tasks faced by each group. Bioscience, 51(7), 557-569. Schofield, H., Messenger, J., Birks, J., & Jermyn, D. (2002). Foraging and roosting behaviour of lesser horseshoe bats at the Ciliau, Radnor. Vincent Wildlife Trust, Ledbury, United Kingdom. Schouten, J. P., McElgunn, C. J., Waaijer, R., Zwijnenburg, D., Diepvens, F., & Pals, G. (2002). Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic acids research, 30(12), e57-e57. Segoli, M., & Rosenheim, J. A. (2012). Should increasing the field size of monocultural crops be expected to exacerbate pest damage?. Agriculture, ecosystems & environment, 150, 38-44. Siemers, B. M., & Ivanova, T. (2004). Ground gleaning in horseshoe bats: comparative evidence from Rhinolophus blasii, R. euryale and R. mehelyi. Behavioral Ecology and Sociobiology, 56, 464-471. Sierro, A., & Arlettaz, R. (1997). Barbastelle bats (Barbastella spp.) specialize in the predation of moths: implications for foraging tactics and conservation. Acta Oecologica, 18(2), 91-106. Simmons, N.B. (2005). Order chiroptera. In D. E. Wilson & D. M. Reeder (Eds.), Mammal Species of the World: A Taxonomic and Geographic Reference (3rd edn, pp. 312–529). Smithsonian Institution Press, Washington, District of Columbia. Simmons, J. A., & Stein, R. A. (1980). Acoustic imaging in bat sonar: echolocation signals and the evolution of echolocation. Journal of comparative physiology, 135, 61-84. Smith, P., & Racey, P. (2005). Optimum effort to estimate habitat use when the individual animal is the sampling unit. Mammal Review, 35(3-4), 295-301. Spitz, J., Ridoux, V., & Brind'Amour, A. (2014). Let's go beyond taxonomy in diet description: testing a trait‐based approach to prey–predator relationships. Journal of Animal Ecology, 83(5), 1137-1148. Steck, C. E., & Brinkmann, R. (2006). The trophic niche of the Geoffroy's bat (Myotis emarginatus) in south-western Germany. Acta chiropterologica, 8(2), 445-450. Stephens, D. W., & Krebs, J. R. (1986). Foraging theory (Vol. 6). Princeton university press. Princeton, NJ, USA. Straka, T. M., Lentini, P. E., Lumsden, L. F., Wintle, B. A., & Ree, R. v. d. (2016). Urban bat communities are affected by wetland size, quality, and pollution levels. Ecology and Evolution, 6(14), 4761-4774. Swartz, S.M., Freeman, P.W., & Stockwell, E.F. (2003). Ecomorphology of Bats: Comparative and Experimental Approaches Relating Structural Design to Ecology. In T. H. Kunz & M. B. Fenton (Eds), Bat Ecology (pp. 257-300). The University Chicago Press, Ltd., London Symondson, W. O. C., Sunderland, K. D., & Greenstone, M. H. (2002). Can generalist predators be effective biocontrol agents?. Annual review of entomology, 47(1), 561-594. Taberlet, P., Bonin, A., Zinger, L., & Coissac, E. (2018). Environmental DNA: For biodiversity research and monitoring. Oxford University Press. Taylor, P. 2016. Rhinolophus hipposideros. The IUCN Red List of Threatened Species 2016. https://www.iucnredlist.org/species/19518/21972794. Accessed on 12 December 2024. Tournayre, O., Leuchtmann, M., Galan, M., Trillat, M., Piry, S., Pinaud, D., Filippi-Codaccioni, O., Pontier, D., & Charbonnel, N. (2021). eDNA metabarcoding reveals a core and secondary diets of the greater horseshoe bat with strong spatio‐temporal plasticity. Environmental DNA, 3(1), 277-296. Tuneu-Corral, C., Puig-Montserrat, X., Flaquer, C., Mata, V. A., Rebelo, H., Cabeza, M., & López-Baucells, A. (2024). Bats and rice: Quantifying the role of insectivorous bats as agricultural pest suppressors in rice fields. Ecosystem Services, 66, 101603. Tuneu‐Corral, C., Puig‐Montserrat, X., Riba‐Bertolín, D., Russo, D., Rebelo, H., Cabeza, M., & López‐Baucells, A. (2023). Pest suppression by bats and management strategies to favour it: a global review. Biological Reviews, 98(5), 1564-1582. Tvrtkovic, N. (2006). Crvena knjiga sisavaca Hrvatske. Ministarstvo kulture, Dr`avni zavod za zastitu prirode, Zagreb. Vamos, E. E., Elbrecht, V., & Leese, F. (2017). Short COI markers for freshwater macroinvertebrate metabarcoding. PeerJ Preprints. No. e3037v2. Van Den Bussche, R. A., Lee, D. N., Judkins, M. E., Dyer, J. E., Thompson, D. M., Stark, R. C., ... & Fuller, B. (2016). Molecular dietary analysis of the endangered Ozark big-eared bat (Corynorhinus townsendii ingens). Acta Chiropterologica, 18(1), 181-191. 40 General Introduction Vallejo, N., Aihartza, J., Goiti, U., Arrizabalaga-Escudero, A., Flaquer, C., Puig, X., Aldasoro, M., Baroja, U., & Garin, I. (2019). The diet of the notch-eared bat (Myotis emarginatus) across the Iberian Peninsula analysed by amplicon metabarcoding. Hystrix, 30(1), 59. Vallejo, N., Aihartza, J., Olasagasti, L., Aldasoro, M., Goiti, U., & Garin, I. (2023). Seasonal shift in the diet of the notched-eared bat (Myotis emarginatus) in the Basque Country: from flies to spiders. Mammalian Biology, 103(4), 419-431. Wilson, D.E., Mittermeier, R.A. (2019). Handbook of the Mammals of the World (Volume 9: Bats, Linx, Ed. Assoc. Conserv. Int. IUCN: Barcelona, Spain. 41 State of the Art III. STATE OF THE ART Understanding the foraging ecology of bats requires a comprehensive approach that accounts for seasonal and regional variations in diet, and the species' interactions with its environment, including the surrounding habitats and prey availability. Additionally, intraand interspecific competition is crucial in shaping foraging strategies and resource partitioning among coexisting species. High-throughput sequencing techniques, such as DNA metabarcoding, provide high-resolution taxonomic identification of prey items, making them invaluable tools for investigating trophic ecology. This dissertation aims to enhance the understanding of the trophic ecology of the three horseshoe bat species present in the study area (Basque Country, Northern Iberian Peninsula), with a particular emphasis on the lesser (Rhinolophus hipposideros) and greater (Rhinolophus ferrumequinum) horseshoe bats. By integrating DNA metabarcoding with ecological trait analyses, this research seeks to provide novel insights into dietary variation, habitat dependences, and the ecological factors influencing prey selection. The findings will contribute to a broader understanding of horseshoe bat foraging strategies and their role within the ecosystem. III.1. HYPOTHESIS Recent advances in molecular techniques enable a comprehensive analysis of the trophic ecology of both horseshoe bat species (Rhinolophus hipposideros and R. ferrumequinum), including temporal and spatial variation in diet, habitat dependency, and their functional role as natural pest suppressors. Furthermore, these methods will facilitate the identification of dietary differences between juvenile and adult individuals, providing insight into their distinct trophic requirements and ecological niches. 43 State of the Art III.2. OBJECTIVES The specific objectives outlined in the following chapters underscore the significance of this research; the first two deal with methodological improvements, while the subsequent ones were designed to prove the hypothesis and contribute to a broader understanding of bat foraging ecology: OBJECTIVE 1: To identify and select the most suitable primers for analysing the diet of horseshoe bats, enabling accurate and reliable insights into their feeding habits and ecological interactions. This objective is developed in Chapter 1. The rapid advancements in DNA metabarcoding during the development of the study necessitated a review and update of the previously used primers, leading to an additional analysis for ongoing research. This section is included in an addendum to Chapter 1. OBJECTIVE 2: To develop a reliable assay for detecting and semi-quantifying multiple pest species in bat faeces using a multiplex ligation-dependent probe amplification (MLPA) assay. This objective is developed in Chapter 2. OBJECTIVE 3: To analyse the fundamental trophic ecology of both horseshoe bats by examining seasonal and geographical variations in their diet throughout their active season. The study also aims to understand the habitat and landscape dependence of the species by examining the relations of preferred prey and their habitats. This objective is developed in Chapter 3 for R. hipposideros and Chapter 4 for R. ferrumequinum. OBJECTIVE 4: To study the primary pest species consumed by R. hipposideros and R. ferrumequinum across various agroforestry landscapes throughout the season and estimate the total mass of pests consumed, contributing to understanding their role in pest control. This objective is developed in Chapter 5 OBJECTIVE 5: To compare the diets of juveniles and adults across the three horseshoe bat species of the study area –R. hipposideros, R. ferrumequinum and R. euryale– by analysing their diet at the taxonomic and prey trait levels (size, flying speed, hardness). This research aims to test age-related dietary differences and the impacts of development and trophic specialisation on these differences. This objective is developed in Chapter 5. 44 IV. RESULTS CHAPTER 1 1.2. MATERIALS AND METHODS 1.2.1. Study Area The study was carried out in Karrantza and Lea-Artibai Valleys (Basque Country, Northern Iberian Peninsula). Karrantza is a hilly valley with elevations of 200–855 m a.s.l. (30T 46968E, 478950N) where the prevailing landscape consists of a mosaic of small meadows and pastures, dedicated to dairy cattle breeding, surrounded by an important hedgerow network consisting mainly of shrubs and deciduous trees. Lea-Artibai Valley is also a hilly and steep valley with elevations ranging ca 40–700 ma.s.l. (30T 53647E, 479442N), where prevailing plantations of Pinus radiata–and less frequently Eucalyptus globulus–are interspersed with small farming patches and small deciduous and holm oak woodland patches. Limestone massifs that provide abundant natural cavities surround both valleys, characterised by Atlantic temperate oceanic climate, where rainfall occurs throughout the year (annual mean 1400mm) (Arrizabalaga-Escudero et al., 2015) 1.2.2. Sample Collection Sampling was carried out during the breeding season, in July 2012. Within each sampling area (Karrantza and Lea-Artibai), each bat species was sampled in a different capture site. There were three roosts in Karrantza–one for each species–and two roosts in Aulesti–one used by R. euryale and R. ferrumequinum, and another one by R. hipposideros. Bats were captured with a 2 × 2 m harp trap (Tuttle, 1974) located in the entrance of the colony roosts from 00:30 a.m. onwards, as bats returned to the caves after foraging. Each captured bat was held individually in a clean cloth bag until it defecated (a maximum of 40–90 min). Each bag was used only once to avoid cross-contamination of faecal samples. Faecal material collected from each individual bat was frozen within 6 h since collection time. Bats were immediately released into the cave after handling. Considering both capture sites altogether, 24 R. ferrumequinum, 31 R. hipposideros and 18 R. euryale individuals were sampled. Individual bats were considered as sample units (Whitaker et al., 1996). Capture and handling protocols followed published guidelines for treatment of animals in research and teaching (Sherwin, 2006) and were approved by the Ethics Committee at the University of the Basque Country (Ref. CEBA/219/2012/GARIN ATORRASAGASTI). Captures were performed under license from the Department of the Environment of the Regional Council of Biscay (Permit numbers G13 1061; G13 1064 and G13 1066). 1.2.3. DNA Extraction, PCR Amplification, Library Preparation and Sequencing Individual faecal samples of 10–40 mg were used for DNA extraction with the DNeasy Power-Soil Kit (Qiagen, Valencia, CA), following the manufacturer steps. Extracted DNA was PCRamplified twice using to different primer sets, targeting different mini-COI segments of the mitochondrial DNA cytochrome c oxidase subunit I barcode region (COI): Zeale primers (ZBJ-ArtF1c and ZBJ-ArtR2c) 52 CHAPTER 1 (Zeale et al., 2011) were used to amplify a 157 bp section, and Gillet primers (modified LepF1 and EPT-long-univR, following (Gillet et al., 2015) to amplify another 133bp section. Both amplifications were performed using QIAGEN Multiplex PCR Kit (Qiagen Iberia, S.L. Madrid) in 25 μl PCR reactions. Each reaction contained 2.5 μl Buffer 10X, 1.5 μl MgCl2 50mM, 0.5 μl nNTPs 25mM and 0.125 μl of taq polymerase. In the case of Zeale primers, 0.6 μl of each primer (10μM) (forward and reverse), 17.175 μl deionised water and 2 μl sample DNA were added. With Gillet primers, 0.75 μl of each primer (10μM), 14.875 μl deionised water and 4 μl sample DNA were added. Each primer set had its own PCR program, modified from the reference to the used reactive. Thermocycler conditions for Zeale primers were: 95˚C– 15 min; 50 cycles of 94˚C– 30 sec, 52˚C– 30 sec, 72˚C– 30 sec; 72˚C– 6 min (modified from Razgour et al., 2011). For Gillet primers we used: 95˚C– 15 min; 40 cycles of 94˚C– 30 sec, 45˚C– 45 sec, 72˚C– 30 sec; 72˚C– 10 min (Gillet et al., 2015). For the library preparation, each sample was tagged with a unique combination of Multiplex Identifier primers (MID) (Binladen et al., 2007). PCR outputs were sequenced by Ion Torrent sequencing platform, one run making above one million reads. 1.2.4. Bioinformatic Analyses Quality control, sequence pre-processing and collapsing of identical sequences into a single sequence were performed using CUTADAPT (Martin, 2011) and USEARCH (Edgar, 2010). Clustering of sequences into Operational Taxonomic Units (OTU) was carried out with the VSEARCH (Rognes et al., 2016), at a 97% similarity threshold. OTUs were normalised in order to avoid disparities in sample reads and the ones with less than 1% frequency were filtered. The taxonomic assignment of each OTU was performed by comparing the representative sequence of each OTU against reference sequences in the Barcode of Life Database (BOLD; www.boldsystems.org/) using BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and GenBank (http://www.ncbi.nlm.nih.gov), following the identification criteria of Clare et al. (2014). The distribution range of each species was checked in order to verify that it encompasses our study area. Species level assignments were performed when query sequences matched reference sequences above 98% similarity and 75% overlap (Clare et al., 2014). When query sequences matched more than one species in the database, the hit with the longest alignment length was selected. Besides, as a rule, only hits with e-value below 1e-20 were accepted (Vesterinen et al., 2013) to make sure that the match did not occur by chance. Primer outputs were also tested to see whether any of the OTUs built from them could also identify the predators themselves (See Appendix A for script details). 1.2.5. Data Analysis To study the effect of primers on the species composition observed in the diet, we performed a permutational multivariate analysis of variance (PERMANOVA) using adonis with 999 random permutations in vegan 2.5–1 package (Oksanen et al., 2018) for R version 3.3.2 (2016). First, we 53 CHAPTER 1 measured the difference among colonies/sampling sites and, as it was not significant, it was no longer considered. Then we used primer set and bat species as predictor variables and the number of occurrences of prey as response variable. Jaccard’s distance measure was used to calculate dissimilarities between samples. We performed NMDS in vegan 2.5–1 package for R to visualise dissimilarities in species composition among samples. The percentage of occurrence (POO) of a given prey taxon refers to the percentage obtained with the number of occurrences of each taxon when compared with the total number of occurrences of all taxa and the frequency of occurrence (FOO) to the number of bat individuals where each taxon was found compared with the sample size (Deagle et al., 2019). Pianka’s (Pianka, 1973) measure of niche overlap has been carried out to compare the interspecific resource partitioning of the species. For the comparison of the diet of the three bats, PERMANOVA analyses were performed using adonis and for the pairwise analysis we used pairwise.perm.manova from RVAideMemoire 0.9–72 package (Hervé, 2014). See Appendix C for R script details. 1.3. RESULTS 1.3.1. Diet of Horseshoe Bats We successfully extracted and amplified DNA from faeces of 24 Rhinolophus ferrumequinum individuals, 18 R. euryale and 31 R. hipposideros, obtaining above one million sequence reads (Table 1.1). 309 OTUs were then built and 135 of them were assigned to potential prey species consumed by bats. We identified 62 prey species of R. ferrumequinum: 34 lepidopterans, 17 dipterans, 7 coleopterans, 2 neuropterans and 1 trichopteran. Lepidoptera and Diptera were the most frequently consumed, followed by Coleoptera (Tables A, B and C in S1.1 Supporting Information File). Among the most frequently consumed species Pharmacis fusconebulosa (Hepialidae; FOO = 46%) prevailed among Lepidoptera and Rhipidia maculata (Limoniidae; FOO = 71%) and Tipula maxima (Tipulidae; FOO = 29%) among Diptera. Within Coleoptera, the most consumed were the elaterid Stenagostus rhombeus (FOO = 46%) and the cerambycids Arhopalus rusticus and Prionus coriarius (FOO = 42% and 17% respectively), completed with scarabeids Aphodius sp. and Serica brunnea (FOO = 29% and 17% respectively). Out of the 81 prey species identified in faeces of R. euryale, 61 were lepidopterans, 10 dipterans, 5 neuropterans, 2 ephemeropterans, 1 trichopteran, 1 hemipteran and 1 hymenopteran. Lepidoptera was the most frequently consumed order followed by Diptera (Tables A, B and C in S1.1 Supporting Information File), although the FOO of most of them was less than 20%. The exceptions were the noctuids Capsula sparganii, Cosmia trapezina and Lycophotia porphyria (FOO > 28%), the geometrid Idaea biselata (FOO = 56%) and the limonid Austrolimnophila ochracea (FOO = 94%). 54 CHAPTER 1 Finally, 73 prey species were identified for R. hipposideros, including 33 lepidopterans, 28 dipterans, 3 hemipterans, 3 neuropterans, 2 coleopterans, 1 hymenopteran, 1 trichopteran, 1 spider and 1 psocopteran. Among them, Diptera were the most frequently consumed, followed in descending order by Lepidoptera, Ephemeroptera, Trichoptera and Neuroptera (Tables A, B and C in S1.1 Supporting Information File). Among Diptera, lesser horseshoe bats mostly preyed upon limonids Austrolimnophila ochracea (FOO = 100%), Rhipidia maculate (FOO = 58%), Neolimonia dumetorum (FOO = 52%), Limonia nubeculosa (FOO = 29%) and Dicranomyia modesta (FOO = 26%), and the tipulid Tipula helvola (FOO = 55%). They also consistently preyed upon the neuropterans Hemerobious humulinus and Wesmaelius nervosus (FOO > 19%). Noteworthy, the occurrence of most moth species was below 3 with a maximum of 8 occurrences of the autostichid Anania hortulata. Table 1.1. Results obtained in the different steps of bioinformatic analyses with each of the primer sets. “Taxa” are the sum of OTUs identified up to species and genus level. (Sequence reads: Total of reads generated from the sequencing; Primary OTUs: Total of built OTUs; Identified OTUs: Number of OTUs which have been identified in the databases with the established similarity and overlap levels; Potential prey taxa: Total number of taxa identified up to genus or species level; Potential prey species: Total number of identified species; Occurrences of potential prey: Total number of occurrences of the identified OTUs.). ZEALE GILLET TOTAL Sequence reads 112191 1003689 1115880 Primary OTUs 179 130 309 Identified OTUs 122 (68%) 69 (53%) 191 Potential taxa 112 58 (61)* 147 (150)* Identified species 101 54 (57)* 135(138)* Occurrences of identified sp. 350 278 (294)* 628 (644)* *: The number in brackets belongs to the total species number identified in Gillet’s samples, and the previous one to the potential prey species (i.e., excluding those considered environmental pollution). As a whole, 12 prey species have been identified in the faeces of the three predators: 6 were lepidopterans (Acronicta rumicis, Cyclophora punctaria, Idaea degeneraria, Anaplectoides prasina, Noctua sp. and Udea ferrugalis), 5 dipterans (Rhipidia maculata, Austrolimnophila ochracea, Limonia nubeculosa, Neolimonia dumetorum and Tipula helvola) and one neuropteran (Hemerobius humulinus). 1.3.2. Performance of Primers Gillet primers yielded the highest numbers of reads, whereas Zeale ones got the highest numbers of either primary OTUs, positively identified OTUs, occurrences of prey and prey species identified (Table 1.1). Moreover, some of the OTUs built from Gillet primers were identified as belonging to algae and mammal species (4.41% of the total taxa), and so they must be considered as environmental pollution instead of "potential prey" consumed by bats. 55 CHAPTER 1 Figure 1.1. NMDS ordination of samples. Stress = 0.1997; k = 2; non-metric fit R2 = 0.96. Dots represent prey species and colours different primer sets (Red: Zeale; Green: Gillet). More distant dots indicate more different prey composition of samples. Individual bat samples are represented as grey triangles. We first tested that there was not significant geographical effect of the two sampling sites in the diet (F(1,68) = 1.031; R2 = 0.132; p = 0.37). Therefore, the location variable was not considered in further analyses. The difference between species diets is significant for the whole data set (F(2,135) = 8.277; R2 = 0.092; p = 0.001), but also for the results obtained with each of the primer sets by their own (Gillet: F(2,70) = 10.466; R2 = 0.230; p = 0.001; Zeale: F(2,65) = 4,772; R2 = 0.128; p = 0.001). The primer choice significantly affected the resulting diet composition (F(1,135) = 14.438; R2 = 0.082; p = 0.001; Figure 1.1). Consequently, the sum of the partial results enlarged the entire prey species list. Of the total species identified as potential prey 40% have been identified with Gillet’s set and 74.8% with Zeale’s, i.e., only 21 out of the 135 (15.5%) potential prey species have been amplified by both primer sets. Anyway, we can see that both primer and bat species affect the list of consumed prey, with a slightly higher explanation of the variation in the case of the bat species. The interaction of primer and species also shows a significant difference among the results, even if it explains less variation than primers and species on their own (F(2,135) = 4.841; R2 = 0.055; p = 0.001). The complete lists of potential prey species identified with each primer set is are included in Tables A, B and C; sequences of all the OTUs built are available in Table D, all of them in S1.1 Supporting Information File. There were big qualitative and quantitative differences among primers at a broader taxonomic level as well (F(1,135) = 61.157; R2 = 0.239; p = 0.001) (Figure 1.2, Tables 1.2 and 1.3, and Table A in Supporting Information File S1.1). Thus, Zeale primers were able to identify five orders of potential prey–namely Lepidoptera, Diptera, Neuroptera, Hemiptera and Psocoptera–whereas OTUs yielded 56 CHAPTER 1 from Gillet primers were assigned to species of fifteen orders: namely Lepidoptera, Diptera, Coleoptera, Trichoptera, Neuroptera, Ephemeroptera, Hemiptera, Hymenoptera and Araneae for prey species, as well as Mucorales, Artiodactyla, Primates and Chiroptera for environmental DNA. Besides, Zeale primers yielded more occurrences than Gillet ones (Table 1.1). The three predator species were identified with Gillet primers in all samples but in two R. ferrumequinum. The interspecific overlap of the diet obtained with Zeale primer sets is not significantly different than the expected by chance (Ojk = 0.18, P = 0.076). R. ferrumequinum and R. hipposideros show the highest overlap (Ojk = 0.34), followed by R. euryale and R. ferrumequinum (Ojk = 0.12), and R. euryale and R. hipposideros (Ojk = 0.08). On the contrary, the overlap based on Gillet primers was significantly higher than expected by chance (Ojk = 0.50, P = 0.002) with the least overlap between R. euryale and R. ferrumequinum (Ojk = 0.25) and the highest overlap between the other two species pairs (R. euryale—R. hipposideros: Ojk = 0.74; R. ferrumequinum—R. hipposideros: Ojk = 0.52). In any case, the effect size is still generally large and the p-value generally small. Similarly, when results of both primer sets are combined the overlap was higher than expected by chance (Ojk = 0.34, P = 0.001). Again, the least overlap was showed by R. euryale and R. ferrumequinum (Ojk = 0.15), while the other two couples show a higher overlap (R. euryale—R. hipposideros: Ojk = 0.38; R. ferrumequinum—R. hipposideros: Ojk = 0.50). Table 1.2. Species identified in faeces and their occurrences with each primer set (Zeale’s and Gillet’s), and combining results, arranged by prey orders. ORDER ZEALE GILLET COMB Occurrences Species Occurrences Species Occurrences Species Araneae 0 0 2 1 2 1 Coleoptera 0 0 44 10 44 10 Diptera 68 19 165 37 269 55 Ephemeroptera 0 0 3 2 3 2 Hemiptera 1 1 5 3 6 4 Hymenoptera 0 0 3 2 3 2 Lepidoptera 256 119 43 17 281 128 Neuroptera 26 6 10 5 33 10 Psocoptera 1 1 0 0 1 1 Trichoptera 0 0 5 2 6 3 57 CHAPTER 1 Figure 1.2. Results of the three bats’ diet obtained with each primer and combining both primers. Results are represented as percentages of occurrences (POO) 2a: R. ferrumequinum, 2b: R. euryale, 2c: R. hipposideros). “Others” comprise the orders with lesser frequencies: Araneae, hemiptera, hymenoptera, psocoptera and trichoptera. GIL: Gillet; ZEA: Zeale; COMB: Combination of both primer sets. Table 1.3. Main orders of prey consumed identified in faeces of the three species of horseshoe bats. ORDER R. ferrumequinum R. euryale R. hipposideros FOO Species FOO Species FOO Species Araneae 0.00 0 0.00 0 6.45 1 Coleoptera 70.83 8 0.00 0 9.68 2 Diptera 95.83 17 94.44 10 100.00 28 Ephemeroptera 0.00 0 16.67 2 0.00 0 Hemiptera 4.17 0 11.11 1 12.90 3 Hymenoptera 0.00 0 5.56 1 6.45 1 Lepidoptera 79.17 34 100.00 61 70.97 33 Neuroptera 20.83 2 22.22 5 48.39 3 Psocoptera 0.00 0 0.00 0 3.23 1 Trichoptera 4.17 1 5.56 1 12.90 1 58 CHAPTER 1 1.4. DISCUSSION Our results confirm the relevance of combining complementary primers to describe the diet of generalist insectivorous bats with amplicon metabarcoding techniques. In general, each pair of primers revealed a subset of the prey composition, with a small fraction of the species being detected by both of them. As a result, the interplay between the primer taxonomic affinity and dietary composition of the bat affected the niche overlap among the three horseshoe bats pictured by each primer. Due to their more generalist character, Gillet primers (Gillet et al., 2015) amplify and identify a higher number of different orders, showing a more diverse diet composition. This generalist character, though, does not cover a full representation or important prey orders such as Lepidoptera. Moreover, their high amplification success comes with the impossibility of identifying a substantial fraction of the amplified DNA (Table 1.1). On the contrary, the higher selectivity of the Zeale primer set for Lepidoptera and some Diptera (Clarke et al., 2014; Aizpurua et al., 2018) might elicit the underestimation of other groups of consumed prey, such as Coleoptera, Ephemeroptera, Hymenoptera or Orthoptera (Aizpurua et al., 2018). On the positive side, maybe due to their lesser degeneration level, a higher proportion of the OTUs got with Zeale primers were assigned to know taxa (68%, Table 1.1), providing a deep coverage of lepidopteran prey species. As both primer sets used in our study amplify regions of the same well-represented COI marker region, the final prey list did not depend of the availability of model species’ sequences in the databases. Moreover, both primers target mini-COI segments of similar size, short enough to be present in faecal samples after digestion (Meusnier et al., 2008; Zeale et al., 2011; Brandon-Mong, 2015; Gillet et al., 2015). In fact, the slightly higher amount of primary OTUs yielded by Gillet primers is consistent with the fact that this primer set amplifies moderately shorter fragments than Zeale ones (133 vs 157 bp, respectively). Nevertheless, the more degenerated Gillet primers could have amplified more DNA fragments, generating more OTUs but with a lower assignment to prey taxa whereas Zeale primers produced less OTUs but with a higher assignment to taxa, consistently with their lower degree of degeneration. The latest molecular study carried out by Galan et al. (2018) described the diet of Rhinolophus ferrumequinum as mainly consisting in Lepidoptera and Diptera, whereas morphological studies have described Coleoptera and Lepidoptera as the most important prey orders (Schober & Grimmberger, 1987; Duvergé & Jones, 1994; Beck, 1997; Andreas et al., 2013). In our study, Diptera occurs in 96% of the samples, closely followed by Lepidoptera (79%) and Coleoptera (71%). Nevertheless, lepidopterans were the main prey order detected with Zeale primers, followed by dipterans, whereas with Gillet ones coleopterans and dipterans prevailed, lepidopterans falling down to a modest third place. In fact, coleopterans were only amplified by Gillet primers and some of the most important lepidopteran families (namely Geometridae, Noctuidae and Totricidae) were disclosed by Zeale. We 59 CHAPTER 1 report the family Geometridae and frequently occurring species such as Rhipidia maculata (Diptera, Limoniidae), Serica brunnea (Scarabaeidae) and Pharmacis fusconebulosa (Lepidoptera, Hepialidae), for the first time among prey of R. ferrumequinum. R. euryale has been widely considered a moth specialist (Schober & Grimmberger, 1987) and, according to Koselj (2002) and Dietz et al. (2009), lepidopterans make up 90% of its diet. In our study, separate molecular studies performed with the two primers showed a narrow specialization level of R. euryale for lepidopterans, but seasonally complemented by ephemeropterans, hemipterans, hymenopterans and trichopterans (Arrizabalaga et al., 2015; Galan et al., 2018). In fact, ephemeropterans had been previously reported as prey of R. euryale in North Africa (Ahmim & Moali, 2013). Noteworthy, lepidopterans are almost the only preyed order if using Zeale, whereas Gillet gives similar importance to lepidopterans and dipterans. Four out of the five most frequently occurring lepidopterans—namely Capsula sparganii (Noctuidae), Udea ferrugalis Crambidae), Lycopohotia porphyrea (Noctuidae), Scoparia sp. (Crambidae) where solely amplified by Zeale. The two most preyed species Capsula sparganii and Austrolimnophila ochracea have not been described before in the diet of R. euryale. R. hipposideros is known to prey mostly upon Diptera Nematocera, followed by Lepidoptera and Neuroptera (McAney & Fairley, 1989; Lino et al., 2014). In our study, the combined use of both primer sets overall confirms the diet composition depicted in previous studies (McAney & Fairley, 1989; Beck, 1997; Feldman et al., 2000; Ahmim & Moali, 2013; Lino et al., 2014; Galan et al., 2018), even if the family choices within dipterans and neuropterans differ. When only Gillet primers were used, though, prevalence of dipterans (mainly limonids) inflated, while that of lepidopterans and neuropterans (hemerobids) deflated. For Zeale, instead, dipterans and lepidopterans appeared almost in the same frequencies, closely followed by neuropterans. This results agree with Andreas et al. (Andreas et al., 2013) who reported a highly prevalence of Lepidoptera in the pellets. Some of the most important families within Diptera reported by morphological studies (Vaughan, 1997), namely Tipulidae, Empididae, Muscidae and Culicidae are also represented within the most frequent prey species in the current study, Empididae only amplified by Gillet and Muscidae only by Zeale. Two of the most frequent limonid prey species—namely Neolimonia dumetorum and Limonia nubeculosa—had been previously reported by Galan et al. (2018). Conversely, some other frequent limonids such as Austrolimnophila ochracea, Rhipidia maculata, Dicranomyia modesta or along with other frequent prey species—Hemerobius humulinus (Neuroptera, Hemerobiidae), Wesmaelius nervosus (Neuroptera, Hemerobiidae), or Pseudatemelia josephinae (Lepidoptera)—had not been reported before. Noteworthy, most of the molecular diet studies carried out on bats exclusively with Zeale primers not surprisingly have concluded that moths or/and Diptera were their main prey: e. g. Barbastella barbastellus, Plecotus macrobullaris, Chalinolobus gouldii, Vespadelus regulus, Nyctophilus gouldi, 60 CHAPTER 1 Eptesicus nilssonii, Myotis brandtii, M. daubentonii, M. mystacinus and Plecotus auritus (Zeale et al., 2011; Alberdi et al., 2012; Burgar et al., 2014; Vesterinen et al., 2016). In some of the studies (Burgar et al., 2014) previously known prey species–such as Coleoptera, Hymenoptera, Isoptera and Trichoptera–were lacking. Notwithstanding the wellsettled importance of moths and dipterans as prey of insectivorous bats (e.g., Beck, 1997; Whitaker et al., 2009), the reliability of trophic scenarios depicted so far only with Zeale primers is still to be ascertained. Therefore, new studies using combination of primers are highly advisable in order to acquire the fullest dietary view whether to confirm the results obtained with Zeale. Furthermore, the strong primer bias reported herein cast doubts on the results of previous studies comparing the trophic niche overlap between sibling bat species, carried out exclusively with a single primer set (Zeale). For example, a study comparing the niches of R. euryale and R. mehelyi (Arrizabalaga-Escudero et al., 218) showed a high degree of diet overlap. Razgour et al. (2011) obtained similar results for Plecotus austriacus and P. auritus. Some other studies have also analysed the diets of sympatric bat species (Vesterinen et al., 2018) based on Zeale primers. Even though the diet overlaps these studies reported cannot be denied, other primers may well unveil additional consumed prey and higher levels of resource partitioning among the species pairs. Last but not least, previous studies have shown that Gillet primers are useful to identify predators’ DNA (Galan et al., 2018; Esnaola et al., 2018). In this study we identified almost all the faecal samples for their predator, in except from two R. ferrumequinum samples. Galan et al. (2018) argued that a mismatch (T/C) at the 30-end of the reverse primer could be at the origin of their higher rates of amplification failure for some bat species, including R. ferrumequinum. We also identified DNA remains indicating unexpected interactions, including secondary predation events. Thus, we found one R. euryale faecal sample containing Bos taurus sequences, likely traces of bovine animal excrements coming from the common housefly (Musca domestica). Lichtheimia ramose was identified in R. euryale and R. hipposideros. This is a fungus living in soil and vegetable wastes that infects both animals and humans. These results must be considered with caution, though, because field contamination cannot be fully discarded. 61 CHAPTER 1 1.I.2. MATERIALS AND METHODS We conducted sequencing and assessed the performance and efficiency of selected primers (see below) on 18 samples from each rhinolophid species—Rhinolophus euryale, R. hipposideros, and R. ferrumequinum—totalling 54 samples. These samples were previously amplified for Chapter 1. 1.I.2.1. Selection of Universal Primers To address the biases of the primer sets used in Chapter 1, we selected novel primer sets targeting mini-COI regions. Our chosen primers have a higher degree of degeneracy than those of Zeale et al. (2011) while amplifying longer sequences than those of Gillet (2015). Specifically, we employed the ANML primer pair (Jusino et al., 2019), which is derived from Zeale et al. (2010) but exhibits greater degeneracy and was designed to amplify prey DNA from bat faecal samples. Additionally, we used FWH1 and FWH2 (Vamos et al., 2017), which are based on multiple arthropod-targeting primer sets (Folmer et al., 1994; Zeale et al., 2011; Leray et al., 2013; Gibson et al., 2015) and have an increased level of degeneracy to enhance amplification success. These primers were previously tested on R. ferrumequinum faecal samples, demonstrating good performance (Tournayre et al., 2020). Table 1.I.1. Primer sets used in this analysis, including ID names of their Forward (For) and Reverse (Rev) primers, their length (bp), and their reference. ID Forward Reverse Length Reference ANML LCOI1490 COI-CFMRa 180 bp Jusino et al. 2019 FWH1 LCOI1490 ZBJ-ArtR2c 178 bp Vamos et al. 2017 FWH2 mlCOIintF ArR5 205 bp Vamos et al. 2017 1.I.2.2. PCR Amplification, Library Preparation And Sequencing Three mini-COI segments of mitochondrial DNA were amplified using the FWH1, FWH2, and ANML primers (178, 205, and 180 bp, respectively). PCR amplification was performed following Jusino et al. (2019) and Tournayre et al. (2020), with modifications. Libraries were constructed using Illumina’s Nextera XT kit, and samples were sequenced on an Illumina MiSeq. PCR amplification, DNA library construction, and sequencing were carried out at the Genomics and Proteomics General Service (SGIker) of the University of the Basque Country (UPV/EHU). 1.I.2.3 Bioinformatic Analyses Separation by primers, quality control, sequence pre-processing, collapsing of identical sequences, and OTU clustering were performed following the steps outlined in Chapter 1. Similarly, the taxonomic assignment of each OTU was carried out by comparing the representative sequence of each OTU against reference sequences in the Barcode of Life Database (BOLD; www.boldsystems.org) and GenBank (www.ncbi.nlm.nih.gov), accessed on 16th July 2022, using the same criteria as in Chapter 1 (S1.2 and S1.3 Supporting Information Files). 68 CHAPTER 1 1.I.2.4. Data Analysis We evaluated the performance of the three primer sets based on read counts (previously normalised) and estimated OTU numbers across predator, prey, environmental, contamination, and unassigned categories. OTUs identified at the genus (e.g., Rhinolophus sp.) or species level in their respective samples (R. hipposideros, R. euryale, or R. ferrumequinum) were assigned as predator. Arthropods from the study area were classified as potential prey. Sequences from fungi, lichens, plants, and the microbiome, along with those unlikely to be naturally found in faeces—such as those from humans, other bats, or mammals—were classified as environmental contamination. Additionally, we accounted for identified prey species across different orders. We analysed all three horseshoe bat species both collectively and individually. Finally, results were combined by performing pairwise primer combinations to identify the optimal set. 1.I.3. RESULTS The sequencing output varied significantly among the different primer sets, both in terms of the quality of sequence reads and the number of rough operational taxonomic units (OTUs) identified (Figure 1.1a). The FWH2 primers produced the highest number of rough OTUs, totalling 2,904, followed by the ANML primers with 2,313 OTUs, and the FWH1 primers with 2,178 OTUs. However, with respect to the number of prey reads, the ANML primers yielded the highest proportion, accounting for 92.98% of the reads, followed by FWH2 (68.84%) and FWH1 (63.25%) (Figure 1.1a). In terms of prey OTUs identified, the ANML primers led with 775 OTUs (33.51%), followed by FWH1 with 659 OTUs (30.26%) and FWH2 with 579 OTUs (19.94%) (Figure 1.1a). This pattern was consistent across the performance of each bat species (Figure 1.1b). FWH1 and FWH2 generated the greatest number of predator reads (6.98% and 6.30%, respectively), with their predator OTUs comprising the largest numbers of sequences, reaching up to 250,418 sequences for FWH1 and 186,275 sequences for FWH2. In comparison, the largest predator OTU in the ANML dataset consisted of only 2,475 sequences (Figure 1.1a). FWH1 also led to the highest number of predator OTUs, with 11, while both ANML and FWH2 each identified only 3 predator OTUs (Figure 1.1a). Additionally, R. euryale had the most OTUs (8 with FWH1, and 2 with FWH2 and ANML) and the highest number of reads (18.96% with FWH2, 18.17% with FWH1, and 0.52% with ANML), while the read counts for R. hipposideros and R. ferrumequinum were significantly lower (2.% with FWH1, and below 0.01% with FWH2 and ANML; 0.94% with FWH1, 0.28% with FWH2, and 0.02% with ANML, respectively) (Figure 1.1b). 69 CHAPTER 1 Figure 1.I.1. Total number of OTUs and Read Counts obtained by the three primer sets (ANML, FWH1 and FWH2) for the categories contamination (red), environmental (blue), predator (green), prey (purple) and unassigned (yellow). a) Total and b) divided by bat species. REU: R. euryale, RHI: R. hipposideros, RFE: R. ferrumequinum. Regarding environmental contamination, FWH1 showed the highest values (13.49% of the reads; 6.89% of the OTUs, respectively), followed by FWH2 (6.63% of the reads; 6.06% of the OTUs, respectively), while ANML showed the lowest (1.47% of the reads; 1.95% of the OTUs, respectively) (Figure 1.1a). Although the percentages of unassigned OTUs were high for all three primers (ANML 64.42%; FWH1 62.44%; FWH2 73.86%), the percentages of unassigned reads were much smaller (ANML 5.38%; FWH1 16.28%; FWH2 18.22%), showing that the unassigned OTUs consisted of small OTUs composed of only a few sequences, which may be artefacts or sequencing errors. However, ANML achieved the best results in this regard (Figure 1.I.1a). FWH1 was the primer set that identified the highest number of prey species, with 534 species, followed by ANML with 522 species, and FWH2 with 483 species. However, ANML consistently identified more prey species for each bat species (R. hipposideros: 379, R. euryale: 334, R. ferrumequinum: 239), while FWH2 identified the fewest (R. hipposideros: 314, R. euryale: 253, R. ferrumequinum: 191). The highest total number of prey species was identified in R. hipposideros (529), followed by R. euryale (454), and the fewest in R. ferrumequinum (368) (Table 1.I.2). At the order level, all primers together identified 13 prey orders. In all cases, the most abundant prey orders were Lepidoptera and Diptera, although the relative importance of the smaller groups varied (Figure 1.I.2). ANML led to a higher number of lepidopteran reads and occurrences, while FWH1 identified more dipterans and coleopterans (Table 1.I.3). FWH1 and FWH2 identified more hemipterans (Figure 1.I.2). 70 CHAPTER 1 Table 1.I.2. Table showing the total prey species of different prey orders identified in the diet of three Rhinolophus bat species (R. ferrumequinum, R. hipposideros, and R. euryale) using different primer combinations (ANML+FWH1; ANML+FWH2; FWH1+FWH2). The data include absolute counts and the total percentages of identified species of each combination across samples. 71 CHAPTER 1 Table 1.I.3. Percentage of occurrences and read counts of secondary prey orders detected by each primer set (ANML, FWH1, and FWH2). Read Counts Percentage of Occurrences Order ANML FWH1 FWH2 ANML FWH1 FWH2 Araneae 0.21 0.48 0.00 0.47 1.00 0.12 Blattodea 0.00 0.39 0.06 0.04 0.43 0.49 Coleoptera 0.86 4.00 2.38 2.35 5.29 4.52 Diptera 41.35 43.61 45.50 27.82 30.96 32.77 Ephemeroptera 0.03 0.62 0.19 0.25 0.49 0.25 Hemiptera 0.21 0.89 2.36 1.38 5.10 7.47 Hymenoptera 0.36 0.31 0.20 0.91 1.73 2.38 Lepidoptera 53.87 46.73 46.80 62.37 47.58 44.35 Mantodea 0.01 0.00 0.00 0.04 0.00 0.00 Neuroptera 2.14 2.38 1.61 2.95 6.20 6.28 Odonata 0.00 0.10 0.00 0.00 0.40 0.00 Orthoptera 0.04 0.00 0.01 0.19 0.12 0.21 Trichoptera 0.92 0.49 0.87 1.22 0.70 1.15 Regarding primer combinations, the pairing that identified the most prey species was ANML+FWH1, detecting 675 out of 752 species, yielding an identification rate of 89.76%. This was closely followed by the ANML+FWH2 combination, which identified 673 species, or 89.49%. The combination with the lowest number of determined species was FWH1+FWH2, which identified 636 species, corresponding to 84.57% (Table 1.I.2). The coverage pattern for each bat species was similar (Table 1.I.2). Figure 1.I.2. Percentage of occurrences and read counts of secondary prey orders detected by each primer set (ANML, FWH1, and FWH2), excluding Lepidoptera and Diptera. 72 CHAPTER 1 When examining prey spectrum coverage, ANML+FWH1 exhibited the best coverage across the main prey orders: Lepidoptera (94.12%), Diptera (89.16%), Hemiptera (83.87%), and Coleoptera (88.46%), both overall and for each bat species (Table 1.I.2). In R. euryale, a moth specialist, 94.89% of the species were identified by ANML+FWH1, followed by ANML+FWH2 (92.70%) and FWH1+FWH2 (76.64%). The pattern for Lepidoptera remained similar across the three bat species. In R. ferrumequinum, the identification of Diptera and Coleoptera was also significant. For Diptera, the best combination was still ANML+FWH1 (84.76%), followed by FWH1+FWH2 and ANML+FWH2 (81.90%). For Coleoptera, the optimal combination was FWH1+FWH2 (87.5%), followed by ANML+FWH1 and ANML+FWH2 identifying the fewest (81.25%). Finally, for R. hipposideros, the identification of Diptera and Hemiptera was also important. For Diptera, the best combination remained ANML+FWH1 (89.61%), followed by ANML+FWH2 (88.96%), and FWH1+FWH2 (86.36%). For Hemiptera ANML+FWH2 achieved the highest coverage (90.74%), followed by the other two combinations (87.04%). 1.I.4. DISCUSSION AND CONCLUSIONS To overcome the biases observed in previously used primers—where the ZBJ primers (Zeale et al., 2011) exhibited a narrow taxonomic range (Clarke et al., 2014; Brandon‐Mong et al., 2015; Mallott et al., 2015), and the Gillet primers (2015) amplified too many unidentifiable sequences (Esnaola et al., 2018)—we aimed to identify more effective primers that would broaden taxonomic coverage while maintaining specificity. The three selected primer sets outperformed the previous ones, identifying a greater number of prey species and orders. Among these, ANML successfully amplified the highest number of prey species. Notably, even the FWH2 primer set, which detected the fewest species among the three, identified 483 species, significantly surpassing the Zeale primers (101 species) and the Gillet primers (54 species) (Chapter 1; Aldasoro et al., 2019). Additionally, when analysing bulk faecal samples collected from beneath the animals in the roost, identifying the predator is crucial, as faeces from other species may be mixed with those of the target species. Among the tested primers, FWH1 yielded the best results for predator identification. The ANML primers provided the highest coverage of prey species, while the FWH1 primers were the most effective for detecting predator DNA. Although all three primers primarily amplified lepidopterans and dipterans, they varied in their ability to identify smaller groups, which could be seasonally significant for each predator species. For instance, FWH1 identified the most coleopteran species, which are an important part of Rhinolophus ferrumequinum’s diet (e.g., Ransome, 1996; Jones, 1990; Flanders & Jones, 2009; Tournayre, 2021). Similarly, larger, harder prey such as blattodeans or orthopterans were also amplified by FWH1 primer set. In contrast, small prey are crucial for R. hipposideros (e.g., McAney & Fairley, 1989; Bontadina et al., 2008; Lino et al., 2014). In this case, the identification of hemipterans and neuropterans is important, and FWH1 and FWH2 were the primers that amplified the highest number of species from this order. 73 CHAPTER 1 We confirmed that using multiple primers increased the number of identified prey taxa and provided more comprehensive diversity coverage, consistent with previous studies (Esnaola et al., 2018; Aldasoro et al., 2019; Jusino et al., 2019). The combination of ANML and FWH1 identified 89.8% of the total species, followed by ANML and FWH2 with 89.5%. For the most consumed prey orders, Lepidoptera and Diptera, the ANML and FWH1 combination demonstrated the highest species coverage, while allowing the identification of the predator itself. While identifying a higher number of prey species can be beneficial, more is not always better, and further analysis is required to assess the ecological relevance of these results. For instance, it remains uncertain whether R. hipposideros could realistically consume 19 species of Coleoptera, highlighting the need for species-level investigations into prey size and ecological plausibility. Some detections may result from cross-contamination between samples, potentially from R. ferrumequinum. Additionally, many identified species belong to low-read, low-occurrence OTUs, which could represent rare species or secondary prey (e.g., Tournayre et al., 2021; Aihartza et al., 2023). Therefore, when investigating foraging ecology beyond simply listing prey, it is crucial to determine whether the ecological significance lies in the most frequently consumed species or whether more detailed taxonomic analyses are necessary. Nonetheless, selecting appropriate primers remains fundamental to obtaining a reliable species list, which serves as the foundation for any subsequent ecological interpretation. In this context, we concluded that the combination of ANML and FWH1 was the most effective for the three horseshoe bat species studied, maximising prey diversity coverage, prey quantity, and predator identification. We will therefore choose this combination for future studies. 74 CHAPTER 1 1.I.5. REFERENCES Aihartza, J., Vallejo, N., Aldasoro, M., García-Mudarra, J. L., Goiti, U., Nogueras, J., & Ibáñez, C. (2023). Aerospace-foraging bats eat seasonably across varying habitats. Scientific Reports, 13(1), 19576. Alberdi, A., Garin, I., Aizpurua, O., & Aihartza, J. (2012). The foraging ecology of the mountain long-eared bat Plecotus macrobullaris revealed with DNA mini-barcodes. PLoS ONE, 7(4), e35692. Aldasoro, M., Garin, I., Vallejo, N., Baroja, U., Arrizabalaga-Escudero, A., Goiti, U., & Aihartza, J. (2019). Gaining ecological insight on dietary allocation among horseshoe bats through molecular primer combination. PLoS ONE, 14(7), e0220081. Arrizabalaga-Escudero, A., Garin, I., García-Mudarra, J. L., Alberdi, A., Aihartza, J., & Goiti, U. (2015). Trophic requirements beyond foraging habitats: The importance of prey source habitats in bat conservation. Biological Conservation, 191, 512-519. Bakker, J., Wangensteen, O. S., Chapman, D. D., Boussarie, G., Buddo, D., Guttridge, T. L., ... & Mariani, S. (2017). Environmental DNA reveals tropical shark diversity in contrasting levels of anthropogenic impact. Scientific reports, 7(1), 16886. Bontadina, F., Schmied, S. F., Beck, A., & Arlettaz, R. (2008). Changes in prey abundance unlikely to explain the demography of a critically endangered Central European bat. Journal of Applied Ecology, 45(2), 641-648. Brandon-Mong, G. J., Gan, H. M., Sing, K. W., Lee, P. S., Lim, P. E., & Wilson, J. J. (2015). DNA metabarcoding of insects and allies: an evaluation of primers and pipelines. Bulletin of entomological research, 105(6), 717-727. Burgar, J. M., Murray, D. C., Craig, M. D., Haile, J., Houston, J., Stokes, V., & Bunce, M. (2014). Who's for dinner? High‐throughput sequencing reveals bat dietary differentiation in a biodiversity hotspot where prey taxonomy is largely undescribed. Molecular Ecology, 23(15), 3605-3617. Clarke, L. J., Soubrier, J., Weyrich, L. S., & Cooper, A. (2014). Environmental metabarcodes for insects: in silico PCR reveals potential for taxonomic bias. Molecular ecology resources, 14(6), 1160-1170. Elbrecht, V., & Leese, F. (2017). Validation and development of COI metabarcoding primers for freshwater macroinvertebrate bioassessment. Frontiers in Environmental Science, 5, 11. Esnaola, A., Arrizabalaga-Escudero, A., González-Esteban, J., Elosegi, A., & Aihartza, J. (2018). Determining diet from faeces: Selection of metabarcoding primers for the insectivore Pyrenean desman (Galemys pyrenaicus). PLoS ONE, 13(12), e0208986. Jeunen, G. J., Knapp, M., Spencer, H. G., Lamare, M. D., Taylor, H. R., Stat, M., ... & Gemmell, N. J. (2019). Environmental DNA (eDNA) metabarcoding reveals strong discrimination among diverse marine habitats connected by water movement. Molecular Ecology Resources, 19(2), 426-438. Flanders, J., & Jones, G. (2009). Roost use, ranging behavior, and diet of greater horseshoe bats (Rhinolophus ferrumequinum) using a transitional roost. Journal of Mammalogy, 90(4), 888-896. Folmer, O., Black, M., Hoeh, W., Lutz, R., Vrijenhoek, R. (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Molecular Marine Biology and Biotechnology, 3, 294–299. Gibson, J. F., Shokralla, S., Curry, C., Baird, D. J., Monk, W. A., King, I., & Hajibabaei, M. (2015). Large-scale biomonitoring of remote and threatened ecosystems via high-throughput sequencing. PLoS ONE, 10(10), e0138432. Gillet, F., Tiouchichine, M. L., Galan, M., Blan, F., Némoz, M., Aulagnier, S., & Michaux, J. R. (2015). A new method to identify the endangered Pyrenean desman (Galemys pyrenaicus) and to study its diet, using next generation sequencing from faeces. Mammalian Biology, 80(6), 505-509. Hebert, P. D., Cywinska, A., Ball, S. L., & DeWaard, J. R. (2003a). Biological identifications through DNA barcodes. Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(1512), 313-321. Hebert, P. D., Ratnasingham, S., & De Waard, J. R. (2003b). Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(suppl_1), S96-S99. Jones, G. (1990). Prey selection by the greater horseshoe bat (Rhinolophus ferrumequinum): optimal foraging by echolocation?. The Journal of Animal Ecology, 587-602. Jusino, M. A., Banik, M. T., Palmer, J. M., Wray, A. K., Xiao, L., Pelton, E., ... & Lindner, D. L. (2019). An improved method for utilizing high‐throughput amplicon sequencing to determine the diets of insectivorous animals. Molecular ecology resources, 19(1), 176-190. Kirse, A., Bourlat, S. J., Langen, K., & Fonseca, V. G. (2021). Metabarcoding Malaise traps and soil eDNA reveals seasonal and local arthropod diversity shifts. Scientific Reports, 11(1), 10498. 75 CHAPTER 1 Krehenwinkel, H., Wolf, M., Lim, J. Y., Rominger, A. J., Simison, W. B., & Gillespie, R. G. (2017). Estimating and mitigating amplification bias in qualitative and quantitative arthropod metabarcoding. ScientificRreports, 7(1), 17668. Krüger, F., Clare, E. L., Greif, S., Siemers, B. M., Symondson, W. O. C., & Sommer, R. S. (2014). An integrative approach to detect subtle trophic niche differentiation in the sympatric trawling bat species Myotis dasycneme and Myotis daubentonii. Molecular Ecology, 23(15), 3657-3671. Leray, M., Yang, J. Y., Meyer, C. P., Mills, S. C., Agudelo, N., Ranwez, V., ... & Machida, R. J. (2013). A new versatile primer set targeting a short fragment of the mitochondrial COI region for metabarcoding metazoan diversity: application for characterizing coral reef fish gut contents. Frontiers in Zoology, 10, 1-14. Lino, A., Fonseca, C., Goiti, U., & Pereira, M. J. R. (2014). Prey selection by Rhinolophus hipposideros (Chiroptera, Rhinolophidae) in a modified forest in Southwest Europe. Acta Chiropterologica, 16(1), 75-83. Mallott, E. K., Malhi, R. S., & Garber, P. A. (2015). High‐throughput sequencing of fecal DNA to identify insects consumed by wild W eddell's saddleback tamarins (Saguinus weddelli, Cebidae, Primates) in B olivia. American Journal of Physical Anthropology, 156(3), 474-481. Marquina, D., Andersson, A. F., & Ronquist, F. (2019). New mitochondrial primers for metabarcoding of insects, designed and evaluated using in silico methods. Molecular Ecology Resources, 19(1), 90-104. McAney, C. M., & Fairley, J. S. (1989). Analysis of the diet of the lesser horseshoe bat Rhinolophus hipposideros in the West of Ireland. Journal of Zoology, 217(3), 491-498. Ransome, R. D. (1996). The management of feeding areas for greater horseshoe bats. English Nature Research Reports (174). Peterborough, UK: English Nature. Ratnasingham, S., & Hebert, P. D. (2007). BOLD: The Barcode of Life Data System (http://www. barcodinglife. org). Molecular ecology notes, 7(3), 355-364. Roswag, A., Becker, N. I., & Encarnação, J. A. (2015). Importance of multi-dimensional analyses of resource partitioning in highly mobile species assemblages. Population Ecology, 57, 601-611. Taylor, P. J., Matamba, E., Steyn, J. N. K., Nangammbi, T., Zepeda-Mendoza, M. L., & Bohmann, K. (2017). Diet determined by next generation sequencing reveals pest consumption and opportunistic foraging by bats in macadamia orchards in South Africa. Acta Chiropterologica, 19(2), 239-254. Tournayre, O., Leuchtmann, M., Filippi‐Codaccioni, O., Trillat, M., Piry, S., Pontier, D., ... & Galan, M. (2020). In silico and empirical evaluation of twelve metabarcoding primer sets for insectivorous diet analyses. Ecology and Evolution, 10(13), 6310-6332. Tournayre, O., Leuchtmann, M., Galan, M., Trillat, M., Piry, S., Pinaud, D., Filippi-Codaccioni, O., Pontier, D., & Charbonnel, N. (2021). eDNA metabarcoding reveals a core and secondary diets of the greater horseshoe bat with strong spatio‐temporal plasticity. Environmental DNA, 3(1), 277-296. Vamos, E. E., Elbrecht, V., & Leese, F. (2017). Short COI markers for freshwater macroinvertebrate metabarcoding. PeerJ Preprints, No. e3037v2. Vesterinen, E. J., Ruokolainen, L., Wahlberg, N., Peña, C., Roslin, T., Laine, V. N., ... & Lilley, T. M. (2016). What you need is what you eat? Prey selection by the bat Myotis daubentonii. Molecular ecology, 25(7), 1581-1594. Wangensteen, O. S., Palacín, C., Guardiola, M., & Turon, X. (2018). DNA metabarcoding of littoral hard-bottom communities: high diversity and database gaps revealed by two molecular markers. PeerJ, 6, e4705. Zeale, M. R., Butlin, R. K., Barker, G. L., Lees, D. C., & Jones, G. (2011). Taxon‐specific PCR for DNA barcoding arthropod prey in bat faeces. Molecular ecology resources, 11(2), 236-244. 76 CHAPTER 1 1.I.6. SUPPORTING INFORMATION S1.2 Supporting Information File. Table A. Identifications of OTUs. Table C. OTU Table. (xlsx) https://tinyurl.com/aldasoroS12 S1.3 Supporting Information File. Table B. Sequences of all the OTUs built with the three primer sets. (txt) https://tinyurl.com/aldasoroS13 77 CHAPTER 2 Figure 2.1. Schematic representation of the Multiplex Ligation-dependent Probe Amplification (MLPA) process (adapted from Schouten et al., 2002). (1) Denaturation and hybridisation: Target DNA is denatured, and MLPA probes hybridise to complementary sequences. (2) Probe ligation: If both probe halves hybridise correctly, a DNA ligase joins them into a single strand. (3) Amplification: Ligated probes are amplified using a universal primer pair with fluorescent labelling for detection. (4) Fragment separation and data analysis: PCR products are separated via capillary electrophoresis, producing an electropherogram where peak heights correspond to the relative quantity of target DNA. However, some COI primers may exhibit species-specific mismatches, leading to biased amplification and reduced detection efficiency (Deagle et al., 2014), leading to the exploration of alternative DNA metabarcoding markers (Clarke et al., 2014; Deagle et al., 2014). The internal transcribed spacer (ITS) region has been recognised as a viable alternative, demonstrating successful application in various taxa (Cruickshank, 2002; Wiemers et al., 2009). In most eukaryotes, ribosomal DNA (rDNA) is organised as tandemly repeated multicopy gene clusters (Arnheim et al., 1980; Coen et al., 1982), undergoing concerted evolution in animals to 84 CHAPTER 2 maintain sequence homogeneity (Hillis et al., 1991; Schlötterer & Tautz, 1994). Copy numbers of the animal rRNA genes range from 39 to 19,300 (Prokopowich et al., 2003), making these genes valuable for molecular species identification. Each transcribed unit contains 18S, 5.8S, and 28S rRNA genes separated by ITS1 and ITS2 (Hernandez et al., 1993). Remarkably, the ITS region of rDNA is worthwhile due to its conserved flanking regions and species-specific size variability. ITS1, a nuclear noncoding fragment with high interspecific variability, has been widely used to develop species-specific primers for single-step PCR in pest identification (Liu, 2004; Asokan et al., 2007; Yeh et al., 2015; Farris et al., 2020). It has facilitated species identification in mosquitoes (Collins & Paskewitz, 1996; Perera et al., 1998) and moths (Perera et al., 2015; Tsai et al., 2020; Park et al., 2022), showing consistency across moth populations (Lewter et al., 2006). However, no attempts have been made to amplify arthropod ITS from faecal samples. Considering all the aspects mentioned, this work aims to develop an MLPA assay to be used in bat faeces for the simultaneous screening and quantification of DNA from various agricultural pests using the ITS1 region as species-specific. 2.2. MATERIALS AND METHODS 2.2.1. Sample Selection Target species were chosen based on 1) their prevalences in previous samplings and metabarcoding analysis and their consumption by diverse bat species (Aizpurua et al., 2018; Baroja et al., 2019; Garin et al., 2019; Aihartza et al., 2024); 2) Our specimen availability in the collection or the possibility of obtaining specimens for DNA extraction; 3) whether genome information was available in the NCBI database; and 4) their impact and interest as agricultural pests (Gil et al., 2014; European and Mediterranean Plant Protection Organization (EPPO) Global Database (https://gd.eppo.int/)). Considering all this, we selected four species to design the MLPA probes: the pine processionary (Thaumetopoea pityocampa, TPI), the codling moth (Cydia pomonella, CPO), the silver Y (Autographa gamma, AGA) and the European vine moth (Lobesia botrana, LBO). 2.2.2. Design of MLPA Oligonucleotide Probes for 18S rDNA -ITS1 Region We extracted five specimens of each species (single specimen in AGA due to availability) using the DNeasy Blood & Tissue DNA extraction kit (Qiagen) and following the manufacturer´s instructions. DNA concentration was measured with a NanoDrop ND-1000 spectrophotometer (NanoDrop Technologies Inc., Montchanin, DE). To sequence the ITS1 region of the targeted species, we compared the genome database of different moth species in NCBI (National Center for Biotechnology Information; http://www.ncbi.nlm.nih.gov; accessed on 9 June 2021) to identify the preserved sequences at the ribosomal DNA fragments 18S to 5.8S. Then, we designed a degenerated primer set for Lepidoptera Internal Transcribed Spacer 1 85 CHAPTER 2 (ITS1) which partially amplified 18S and 5.8S rDNA, and the whole ITS1 sequence, which will vary from species to species: Fw: 5´-AGTCGTAACAAGGTTTCCGTAGG-3´; Rv: 5´-CKATGACGCRCAGTTTRCTGC-3´. Briefly, we performed a PCR on the extracted samples. First, the DNA was denatured at 95ºC for 15 minutes, followed by 40 cycles of amplification (30 seconds at 94ºC for denaturation, 30 seconds at 57ºC for annealing, and 30 seconds at 72ºC for extension), and a final extension at 72ºC for 6 minutes. The PCR was performed with the HotStarTaq DNA Polymerase (Qiagen) in a Mastercycler pro S Thermocycler (Eppendorf). Amplicons were observed in 2% agarose gel stained with Ethidium Bromide. Then, the positive amplicons were Sanger sequenced by the team of the Genomics Service of SGIker (UPV/EHU/ERDF, EU). Once we had our target sequences, they were aligned (ClustalW) and analysed (Figure 2.2) to find suitable regions for the design of the MLPA species-specific probe set (Table 2.1). The specificity of the probes was evaluated in silico: 1) by BLAST search within NCBI GenBank (accessed on 5 October 2021), and 2) looking at the potential secondary structure with OligoAnalyzer Tool (IDT, https://www.idtdna.com/pages/tools/oligoanalyzer; accessed on 5 October 2021), in this way we avoided interfering cross-reactivities with the target moths and other related species. As far as possible, the oligonucleotide probe attributes were designed following the general suggestions of MRC-Holland (Amsterdam, The Netherlands). The 5´ end has to be phosphorylated since it allows the ligase to join probe halves, ensuring selective amplification of correctly hybridised targets, which improves assay specificity and reliability (Schouten et al., 2002). Probes were purchased from IDT (Integrated DNA Technologies, Iowa, USA) (https://eu.idtdna.com/site). 86 CHAPTER 2 Figure 2.2. Alignment of the consensus sequences belonging to the four target species (AGA, TPI, CPO, and LBO) along with the probe binding sites. The Left Probe Oligo binding site is highlighted in blue, while the Right Probe Oligo binding site is shown in yellow. Areas belonging to 18S rDNA and 5.8S rDNA are inside black squares. 87 CHAPTER 2 Table 2.1. Probes and primers of the 18SrDNA-ITS1-5.8SrDNA MLPA systems. The primer binding sites are highlighted in bold capitals, and the 5' end of RPOs were phosphorylated (*). 88 CHAPTER 2 2.2.3. MLPA Assay Validation Workflow We conducted the MLPA assay following the steps below (Figure 2.3): 1. Testing the MLPA design in pure samples We first conducted an in vitro analysis to detect a specific target DNA, ensuring the proper functionality of the probes with individual DNA extractions. DNA templates extracted from the target moth species were individually analysed in separate reactions using their corresponding MLPA probe sets. This analysis was performed using TPI, CPO, AGA, and LBO. We studied three specimens per species, except for AGA, for which only a single specimen was available. Each sample was tested in duplicate, and a no-template control (NTC) was included for each probe combination. The DNA concentration ranges (ng/µL) of the moth extractions for each species were as follows: TPI (11.6–19.7), CPO (6.2–20.6), LBO (15.4–19.7), and AGA (6.0). 2. Determining MLPA detection limit in pure moth DNA samples After verification, we performed serial dilutions of the DNA extractions to determine the detection limit for TPI, CPO, and AGA at 1:2.5, 1:5, and 1:10. Since we failed to detect LBO in the first step, we set it aside for the subsequent steps. Due to the low initial DNA concentration, sample concentration was not feasible, and further dilution of the initial samples would be nonsense; therefore, the samples were not equimolarised. We analysed three TPI and CPO samples and one AGA sample, with two replicates per dilution and an NTC per probe combination. 3. Multiplexing the probes in pure moth DNA samples After confirming the effectiveness of individual probes, we conducted a multiplex assay using a single target DNA and multiple MLPA probes simultaneously. This assay included a triplex combination (CPO + TPI + AGA) and three duplex combinations (CPO + TPI, CPO + AGA, and TPI + AGA). We included an NTC sample for each combination. Analyses were performed in duplicates. 4. Moth MLPA assay in bat faeces Finally, we conducted a preliminary test to identify pest species in bat faeces. Samples were selected based on previous studies where the target species had been detected in DNA extractions through metabarcoding (positive samples from Chapter 6 and Aihartza et al., 2023) (Table 2.2). Analyses were performed in triplicates, and an NTC sample was included. The samples in which the target DNA was detected were diluted to their original 1:10 concentrations through serial dilutions to establish the detection limit. 2.2.4. MLPA Assay and Analysis All steps of the MLPA assay were performed using a SimpliAmp Thermal Cycler (Thermo Fisher Scientific, USA) and the SALSA MLPA EK1 reagent kit (MRC-Holland) according to the manufacturer's instructions. A negative sample (NTC) was included in each test to detect possible contamination or other potential problems. The reaction components are included in the SALSA 89 CHAPTER 2 MLPA EK1 (FAM-6) reagent kit, excluding the DNA sample, probe mix, water, and TE buffer. Briefly, 4 µL of sample DNA were denatured with 1 µL of TE at 95ºC for 5 minutes using 0.2 µL tubes. A non-template control containing 5 µL of TE was included in each reaction to detect any potential reagent contamination. After denaturation, the samples were cooled to 25ºC and removed from the thermocycler to add the hybridisation mix, which consisted of 1.5 µL of SALSA MLPA buffer and 1.5 µL of the synthesised probe mixture in TE. The samples were then heated to 95ºC for 1 minute and incubated at 70ºC for 16 hours to facilitate hybridisation. For the ligation reaction of the hybridised probes, 32 µL of the Ligase master mix (comprising 3 µL of ligase buffer A, 3 µL of ligase buffer B, 25 µL of H2O, and 1 µL of Ligase-65) were added and incubated for 15 minutes at 54ºC. Afterwards, the samples were heated to 98ºC for 5 minutes to inactivate the enzyme. Simultaneous amplification of the ligation products was performed using universal SALSA PCR primers. To initiate the amplification, we added 10 µL of the polymerase master mix (containing 7.5 µL of H2O, 2 µL of PCR Primer Mix, and 0.5 µL of SALSA polymerase). We executed the following program: 35 cycles of amplification (30 seconds at 95ºC for denaturation, 30 seconds at 60ºC for annealing, and 1 minute at 72ºC for extension), followed by a final extension at 72ºC for 20 minutes and a cool-down step to 15ºC. The PCR products were stored in the dark until further analysis. In each experiment, DNA samples were analysed in duplicate to evaluate the reproducibility of the developed assay. Finally, the amplicons were analysed by electrophoresis using the ABI PRISM 3130XL (Thermo Fisher Scientific, USA) by the Genomics Service of the SGIker team (UPV/EHU/ERDF, EU). Table 2.2. The faecal samples used for the MLPA test in bat faeces, including their original DNA concentration (ng/µL), relative read abundance (RRA), and weighted percentage of occurrence (wPOO) from the metabarcoding analysis. Amplification indicates whether the MLPA assay successfully detected the expected peak. For FaTPI1 and FaTPI2, DNA concentrations of the serial dilutions are also provided. SAMPLE [DNA] (ng/µL) [1:2.5] (ng/µL) [1:5] (ng/µL) [1:10] (ng/µL) RRA (%) wPOO (%) FaCPO1 37.97 63.4 12.60 FaTPI1 5.1 2.04 1.02 0.51 62.1 9.09 FaTPI2 7.6 3.04 1.52 0.76 69.8 9.09 FaTPI3 3.9 94.2 33.33 FaTPI4 8.0 66.1 14.28 FaTPI5 7.7 61.4 10 90 CHAPTER 2 Figure 2.3. Schematic representation of the MLPA assay workflow for detecting moth pest species in bat faeces. The top section illustrates the target moth species and their corresponding probes, in vitro moth DNA extractions, and DNA extractions from bat faeces. The bottom section presents the experimental steps: (1) testing the MLPA design with pure samples, (2) determining the detection limit for MLPA, (3) multiplexing the probes, and (4) applying MLPA to bat faeces samples to assess detection efficiency. Green checkmarks indicate successful detection, while red crosses signify failed detection or detection limits. TPI: Thaumetopoea pityocampa; CPO: Cydia pomonella; AGA: Autographa gamma; LBO: Lobesia botrana. 2.3. RESULTS 2.3.1. Evaluation of the ITS1 MLPA Probe Specificity The designed primers successfully amplified the 18S-ITS1-5.8S region. After analysing the forward and reverse Sanger sequencing results, we obtained the consensus sequence for each species by comparing our sequenced fragments with sequences available in GenBank (The consensus sequences have been deposited in GenBank under the following accession numbers: PQ605054, PQ605056, PQ605057, and PV156760). We observed homology in the 3´region of the 18S rDNA, while the 5´region of the 5.8S region exhibited nucleotide variations between species (Figure 2.2). In length, the 91 CHAPTER 2 ITS1 region varied significantly between species, ranging from the shortest in LBO (146 bp) to the longest in CPO (468 bp), with AGA measuring 331 bp and TPI 432 bp. The ITS1 MLPA design was first in silico tested for possible cross-homologies between DNA from the target species (TPI, CPO, AGA, LBO) and non-target arthropod species. To do so, we aligned the complete hybridising sequences of each probe (left and right hybridisation sequences) against sequences available within NCBI GenBank using BLASTn. Our analysis showed 100% identity for TPI, 100% for CPO, 99.01% for LBO, and 95.92% for AGA compared to the expected species in a blast search against the entire nucleotide database. The analysis of potential heterodimer formation of the probes indicated a generally low risk, with only one high-risk heterodimer (ΔG < -9 kcal/mol) observed for each species probe combination. In any case, these did not align with the expected length. 2.3.2. Testing the MLPA Design in Pure Samples DNA templates extracted from the target moth species were individually analysed in separate reactions using their corresponding MLPA probe sets. Three out of the four species (TPI, CPO, and AGA) were successfully detected, producing reproducible signals of the expected sizes (TPI: 155 nt, CPO: 180 nt, AGA: 168 nt), while LBO was not detected (Figure 2.4). Due to the absence of detectable peaks in LBO, we excluded it from further analyses. Figure 2.4. Electropherograms of the MLPA assay for independent detection in pure samples. (a) Cydia pomonella (CPO): 180 nt; (b) Thaumetopoea pityocampa (TPI): 155 nt; (c) Autographa gamma (AGA): 168 nt; and (d) Lobesia botrana (LBO), which was not detected. The red peak at 50 bp represents the lower marker of the electropherogram. The X-axis represents sequence length (nt), while the Y-axis shows fluorescence units. A slight 1–2 nt shift from the expected fragment lengths was consistently observed (Table 2.1). This shift was likely due to variations in DNA fragment mobility and/or differences in labelling between 92 CHAPTER 2 the standard and amplified fragments, as reported in previous studies (Ehlert et al., 2009; Unterberger et al., 2014; Garcia-Garcia et al., 2017). 2.3.3. Determining MLPA Detection Limit Even though we could detect specific peaks in the dilutions (ranging from 1:2.5 to 1:10) for TPI and AGA, the peak heights remained nearly constant despite the variations in DNA concentrations (Figure 2.5). Only one of the TPI samples (dark blue in Figure 2.5b) showed a decline in fluorescence with decreasing DNA concentrations. For CPO, however, the MLPA assay did not work as expected, and we identified nonspecific peaks (at 113 bp, 126 bp and 146 bp, which likely corresponded to different dimers) in addition to the expected 180 bp peak. Figure 2.5. DNA concentration (ng/µL) and specific peak heights from electrophoresis results of serial dilutions (1:1, 1:2.5, 1:5, and 1:10) in in vitro samples of (a) Autographa gamma (AGA), showing dilutions from a single sample and its duplicates, and (b) Thaumetopoea pityocampa (TPI), showing dilutions from three different samples and their duplicates. Different shades of blue in (b) represent the three original TPI samples. The red dotted line shows the mean trend line, and the blue lines show the trend line of each sample. 2.3.4. Multiplexing the Probes No specific peaks corresponding to the target species were detected when we conducted the in vitro multiplex assay using a single target DNA and multiple MLPA probes simultaneously (including all three probes). Instead, a peak of approximately 115 bp was observed in all samples (Figure 2.6a). When the assay was performed using probe pairs (TPI+CPO, TPI+AGA, and CPO+AGA), the 115 bp 93 CHAPTER 2 Ehlert, A., Demmel, A., Hupfer, C., Busch, U., & Engel, K. H. (2009). Simultaneous detection of DNA from 10 food allergens by ligation-dependent probe amplification. Food Additives and Contaminants, 26(4), 409-418. Esnaola, A., Arrizabalaga-Escudero, A., González-Esteban, J., Elosegi, A., & Aihartza, J. (2018). Determining diet from faeces: Selection of metabarcoding primers for the insectivore Pyrenean desman (Galemys pyrenaicus). PLoS ONE, 13(12), e0208986. Evans, D. M., Kitson, J. J., Lunt, D. H., Straw, N. A., & Pocock, M. J. (2016). Merging DNA metabarcoding and ecological network analysis to understand and build resilient terrestrial ecosystems. Functional Ecology, 30(12), 1904-1916. Farris, R. E., Ruiz-Arce, R., Ciomperlik, M., Vasquez, J. D., & DeLeón, R. (2010). Development of a ribosomal DNA ITS2 marker for the identification of the thrips, Scirtothrips dorsalis. Journal of Insect Science, 10(1), 26. García-García, A., Madrid, R., García, T., Martín, R., & González, I. (2018). Use of multiplex ligation-dependent probe amplification (MLPA) for screening of wheat, barley, rye and oats in foods. Food Control, 84, 268-277. Garin, I., Aihartza, J., Goiti, U., Arrizabalaga-Escudero, A., Nogueras, J., & Ibáñez, C. (2019). Bats from different foraging guilds prey upon the pine processionary moth. PeerJ, 7, e7169. Gil, Á. M., de Ojer, J. L. R. S., Pérez, M. R., & Ministerio de Agricultura. (2014). Guía de gestión integrada de plagas. Ministerio de Agricultura, Alimentación y Medio Ambiente, Centro de Publicaciones. Gillet, F., Tiouchichine, M. L., Galan, M., Blan, F., Némoz, M., Aulagnier, S., & Michaux, J. R. (2015). A new method to identify the endangered Pyrenean desman (Galemys pyrenaicus) and to study its diet, using next generation sequencing from faeces. Mammalian Biology, 80(6), 505-509. Gonzalez, J. M., Portillo, M. C., & Saiz‐Jimenez, C. (2005). Multiple displacement amplification as a pre‐polymerase chain reaction (pre‐PCR) to process difficult to amplify samples and low copy number sequences from natural environments. Environmental Microbiology, 7(7), 1024-1028. Grohmann, L., Barbante, A., Eriksson, R., Gatto, F., Georgieva, T., Huber, I., …, Zdeňková, K., (2021). Guidance document on multiplex real-time PCR methods. EUR 30708 EN, Publications Office of the European Union, Luxembourg. Guertler, P., Goerlich, O., & Busch, U. (2014). Development of a multiplex ligation dependent probe amplification (MLPA) module for simultaneous detection of five genetically modified rapeseed events. Agricultural Sciences, 2014. Hajibabaei, M., Smith, M. A., Janzen, D. H., Rodriguez, J. J., Whitfield, J. B., & Hebert, P. D. (2006). A minimalist barcode can identify a specimen whose DNA is degraded. Molecular ecology notes, 6(4), 959-964. Hebert, P. D., Cywinska, A., Ball, S. L., & DeWaard, J. R. (2003a). Biological identifications through DNA barcodes. Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(1512), 313-321. Hebert, P. D., Ratnasingham, S., & De Waard, J. R. (2003b). Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(suppl_1), S96-S99. Hernandez, P., Martín‐Parras, L., Martínez‐Robles, M. L., & Schvartzman, J. B. (1993). Conserved features in the mode of replication of eukaryotic ribosomal RNA genes. The EMBO journal, 12(4), 1475-1485. Hillis, D. M., Moritz, C., Porter, C. A., & Baker, R. J. (1991). Evidence for biased gene conversion in concerted evolution of ribosomal DNA. Science, 251(4991), 308-310. Janssen, D., González-Miras, E., & Rodríguez, E. (2023). Tuta absoluta-Specific DNA in Domestic and Synanthropic Vertebrate Insectivore Feces. Insects, 14(8), 673. Jarman, S. N., Deagle, B. E., & Gales, N. J. (2004). Group‐specific polymerase chain reaction for DNA‐based analysis of species diversity and identity in dietary samples. Molecular Ecology, 13(5), 1313-1322. Jusino, M. A., Banik, M. T., Palmer, J. M., Wray, A. K., Xiao, L., Pelton, E., ... & Lindner, D. L. (2019). An improved method for utilizing high‐throughput amplicon sequencing to determine the diets of insectivorous animals. Molecular ecology resources, 19(1), 176-190. Korine, C., Niv, A., Axelrod, M., & Dahan, T. (2020). Species richness and activity of insectivorous bats in cotton fields in semi-arid and mesic Mediterranean agroecosystems. Mammalian Biology, 100(1), 73-80. Krauel, J. J., Brown, V. A., Westbrook, J. K., & McCracken, G. F. (2018). Predator–prey interaction reveals local effects of high-altitude insect migration. Oecologia, 186, 49-58. Lewter, J. A., Szalanski, A. L., Nagoshi, R. N., Meagher Jr, R. L., Owens, C. B., & Luttrell, R. G. (2006). Genetic variation within and between strains of the fall armyworm, Spodoptera frugiperda (Lepidoptera: Noctuidae). Florida Entomologist, 89(1), 63-68. Liu, Y. C. (2004). Molecular identification of a plant quarantine pest (Frankliniella occidentalis) by one-tube nested PCR targeting ribosomal DNA internal transcribed spacer regions. Plant Protection Bulletin, 46(1), 27-46. 100 CHAPTER 2 López-Calleja, I. M., García, A., Madrid, R., García, T., Martín, R., & González, I. (2017). Multiplex ligation-dependent probe amplification (MLPA) for simultaneous detection of DNA from sunflower, poppy, flaxseed, sesame and soy allergenic ingredients in commercial food products. Food Control, 71, 301-310. Lorenz, T. C. (2012). Polymerase chain reaction: basic protocol plus troubleshooting and optimization strategies. Journal of visualized experiments: Journal of visualized experiments, 63, 3998. Mata, V. A., Rebelo, H., Amorim, F., McCracken, G. F., Jarman, S., & Beja, P. (2019). How much is enough? Effects of technical and biological replication on metabarcoding dietary analysis. Molecular Ecology, 28(2), 165-175. McCracken, G. F., Westbrook, J. K., Brown, V. A., Eldridge, M., Federico, P., & Kunz, T. H. (2012). Bats track and exploit changes in insect pest populations. PLoS ONE, 7(8), e43839. Meusnier, I., Singer, G. A., Landry, J. F., Hickey, D. A., Hebert, P. D., & Hajibabaei, M. (2008). A universal DNA mini-barcode for biodiversity analysis. BMC Genomics, 9, 1-4. Monteiro, L., Bonnemaison, D., Vekris, A., Petry, K. G., Bonnet, J., Vidal, R., ... & Mégraud, F. (1997). Complex polysaccharides as PCR inhibitors in feces: Helicobacter pylori model. Journal of Clinical Microbiology, 35(4), 995-998. Mustorp, S. L., Drømtorp, S. M., & Holck, A. L. (2011). Multiplex, quantitative, ligation-dependent probe amplification for determination of allergens in food. Journal of agricultural and food chemistry, 59(10), 5231-5239. Oerke, E. C. (2006). Crop losses to pests. The Journal of Agricultural Science, 144(1), 31-43. Oikarinen, S., Tauriainen, S., Viskari, H., Simell, O., Knip, M., Virtanen, S., & Hyöty, H. (2009). PCR inhibition in stool samples in relation to age of infants. Journal of Clinical Virology, 44(3), 211-214. Park, S. R., Lee, D. E., Nam, H. Y., Kim, J., Lee, S. H., & Kim, J. H. (2022). Development of Multiplex PCR-based Protocols for simultaneous caterpillar diagnosis of three Spodoptera and One Mamestra Species (Lepidoptera: Noctuidae). Journal of Economic Entomology, 115(5), 1703-1711. Perera, O. P., Allen, K. C., Jain, D., Purcell, M., Little, N. S., & Luttrell, R. G. (2015). Rapid identification of Helicoverpa armigera and Helicoverpa zea (Lepidoptera: Noctuidae) using ribosomal RNA internal transcribed spacer 1. Journal of Insect Science, 15(1), 155. Perera, O. P., Cockburn, A. F., Mitchell, S. E., Conn, J., & Seawright, J. A. (1998). Species-specific repeat units in the intergenic spacer of the ribosomal RNA cistron of Anopheles aquasalis Curry. The American journal of tropical medicine and hygiene, 59(5), 673-678. Piñol, J., Mir, G., Gomez‐Polo, P. R. I. S. C. I. L. A., & Agustí, N. (2015). Universal and blocking primer mismatches limit the use of high‐throughput DNA sequencing for the quantitative metabarcoding of arthropods. Molecular ecology resources, 15(4), 819-830. Puig-Montserrat, X., Torre, I., López-Baucells, A., Guerrieri, E., Monti, M. M., Ràfols-García, R., ... & Flaquer, C. (2015). Pest control service provided by bats in Mediterranean rice paddies: linking agroecosystems structure to ecological functions. Mammalian Biology, 80(3), 237-245. Pompanon, F., Deagle, B. E., Symondson, W. O., Brown, D. S., Jarman, S. N., & Taberlet, P. (2012). Who is eating what: diet assessment using next generation sequencing. Molecular Ecology, 21(8), 1931-1950. Prokopowich, C. D., Gregory, T. R., & Crease, T. J. (2003). The correlation between rDNA copy number and genome size in eukaryotes. Genome, 46(1), 48-50. Schierwater, B., Metzler, D., Krüger, K., & Streit, B. (1996). The effects of nested primer binding sites on the reproducibility of PCR: mathematical modeling and computer simulation studies. Journal of Computational Biology, 3(2), 235-251. Schlötterer, C., & Tautz, D. (1994). Chromosomal homogeneity of Drosophila ribosomal DNA arrays suggests intrachromosomal exchanges drive concerted evolution. Current Biology, 4(9), 777-783. Schouten, J. P., McElgunn, C. J., Waaijer, R., Zwijnenburg, D., Diepvens, F., & Pals, G. (2002). Relative quantification of 40 nucleic acid sequences by multiplex ligation-dependent probe amplification. Nucleic acids research, 30(12), e57-e57. Segoli, M., & Rosenheim, J. A. (2012). Should increasing the field size of monocultural crops be expected to exacerbate pest damage?. Agriculture, ecosystems & environment, 150, 38-44. Sun, J., Xu, J., Liang, P., Mao, Q., Huang, Y., Lv, X., ... & Yu, X. (2011). Molecular identification of Clonorchis sinensis and discrimination with other opisthorchid liver fluke species using multiple Ligation-depended Probe Amplification (MLPA). Parasites & vectors, 4, 1-12. Symondson, W. O. C., Sunderland, K. D., & Greenstone, M. H. (2002). Can generalist predators be effective biocontrol agents?. Annual review of entomology, 47(1), 561-594. 101 CHAPTER 2 Taberlet, P., Bonin, A., Zinger, L., & Coissac, E. (2018). Environmental DNA: For biodiversity research and monitoring. Oxford University Press. Terefework, Z., Pham, C. L., Prosperi, A. C., Entius, M. M., Errami, A., van Spanning, R. J., ... & Crielaard, W. (2008). MLPA diagnostics of complex microbial communities: relative quantification of bacterial species in oral biofilms. Journal of microbiological methods, 75(3), 558-565. Thornton, C. G., & Passen, S. (2004). Inhibition of PCR amplification by phytic acid, and treatment of bovine fecal specimens with phytase to reduce inhibition. Journal of microbiological methods, 59(1), 43-52. Tsai, C. L., Chu, I. H., Chou, M. H., Chareonviriyaphap, T., Chiang, M. Y., Lin, P. A., ... & Yeh, W. B. (2020). Rapid identification of the invasive fall armyworm Spodoptera frugiperda (Lepidoptera, Noctuidae) using species-specific primers in multiplex PCR. Scientific reports, 10(1), 16508. Unterberger, C., Luber, F., Demmel, A., Grünwald, K., Huber, I., Engel, K. H., & Busch, U. (2014). Simultaneous detection of allergenic fish, cephalopods and shellfish in food by multiplex ligation-dependent probe amplification. European Food Research and Technology, 239, 559-566. Vamos, E. E., Elbrecht, V., & Leese, F. (2017). Short COI markers for freshwater macroinvertebrate metabarcoding. PeerJ Preprints, e3037v2. Verginelli, D., Paternò, A., De Marchis, M. L., Quarchioni, C., Vinciguerra, D., Bonini, P., ... & Marchesi, U. (2020). Development and comparative study of a pat/bar real‐time PCR assay for integrating the screening strategy of a GMO testing laboratory. Journal of the Science of Food and Agriculture, 100(5), 2121-2129. Wang, X. C., Liu, C., Huang, L., Bengtsson‐Palme, J., Chen, H., Zhang, J. H., ... & Li, J. Q. (2015). ITS 1: a DNA barcode better than ITS 2 in eukaryotes?. Molecular ecology resources, 15(3), 573-586. Wiemers, M., Keller, A., & Wolf, M. (2009). ITS2 secondary structure improves phylogeny estimation in a radiation of blue butterflies of the subgenus Agrodiaetus (Lepidoptera: Lycaenidae: Polyommatus). BMC Evolutionary Biology, 9, 1-27. Yeh, W. B., Tseng, M. J., Chang, N. T., Wu, S. Y., & Tsai, Y. S. (2015). Agronomically important thrips: development of species-specific primers in multiplex PCR and microarray assay using internal transcribed spacer 1 (ITS1) sequences for identification. Bulletin of Entomological Research, 105(1), 52-59. Zeale, M. R., Butlin, R. K., Barker, G. L., Lees, D. C., & Jones, G. (2011). Taxon‐specific PCR for DNA barcoding arthropod prey in bat faeces. Molecular ecology resources, 11(2), 236-244. Zhou, Y., Nie, J., Yu, S., Hu, Z., & Wang, B. (2020). Multiplex ligation-dependent probe amplification for simultaneous identification of Bungarus multicinctus and its common adulterants in a single assay. Frontiers in pharmacology, 11, 501. 102 CHAPTER 3 Seasonal and geographic variation in the trophic ecology and habitat dependence of Rhinolophus hipposideros ARTICLE Aldasoro, M., de Cerio, O. D., Russo, D., Vallejo, N., Olasagasti, L., Goiti, U., & Aihartza, J. (2025). Molecular Dietary Analysis Reveals Plasticity in Habitat Requirements of a Clutter Specialist Bat. Basic and Applied Ecology, 84, 101-109. CONGRESS XVIIth European Bat Research Symposium (EBRS). Poster communication. “The Lesser Horseshoe bat, a flexible clutter foraging specialist”. Girona, Catalonia. CHAPTER 3 LABURPENA Azken hamarkadetan aurrerapen handiak izan dira saguzarren bazka-habitaten azterketan. Hala ere, ikerketa hauek bere beharren ulermen mugatua besterik ez dute eskaintzen. Metabarcoding-ak aukera ematen du kontsumitutako harrapakinak espezie mailan identifikatzeko, haien jatorrizko habitatak zehazteko aukera emanez. Azterketa honetan, Rhinolophus hipposideros-en hiru kolonietako gorotz laginak hartu ziren, eremu klimatiko ezberdinetan, udaberritik uda bukaera arte. Metabarcoding-a baliatuz, ferra-saguzar txikiaren dieta denboran zeharreko dieta aldakortasuna aztertu ziren, eta haien harrapakin nagusiak sasoi eta eremu ezberdinen arabera aldatzen diren. Emaitzek erakusten dutenez, saguzarren dietak aldatu egiten dira urtaroaren arabera, eta kolonien arteko dieta ezberdina da, askotan harrapakinen agerraldiei erantzunez. Saguzarren jandako harrapakinetatik habitat eskakizun desberdinak dituztela ondorioztatu genuen. Basoak eta zuhaixkak habitat nagusiak dira, nahiz eta beste ingurune batzuekiko ere menpekotasuna erakusten duten. Habitat irekiak bereziki uste baino maizago ustiatzen direla ondorioztatu zen, bazkatze habitaten beharretan plastikotasun-maila handia erakutsiz. Hortaz, saguzarren kontserbaziorako askotariko harrapakinak eskaintzen dituzten eta elkar konektatuta dauden paisai anitzak babestea funtsezkoa da. ABSTRACT In recent decades, there has been significant progress in studying the foraging habitats of bats. However, these studies provide only a limited understanding of their requirements. Metabarcoding allows species-level identification of consumed prey, allowing us to determine their source habitats. In this study, we sampled faeces from three Rhinolophus hipposideros colonies in different climatic zones from spring to late summer. Using metabarcoding, we examined how the lesser horseshoe bat diet changes over time and whether their most-consumed prey varies seasonally across landscapes. Our results show that bat diets change seasonally and differ between colonies, often presumably in response to new prey outbreaks. We deduced from the prey eaten by bats that they have varied habitat requirements. While woodland and shrubs are primary prey source habitats, bats also rely on other environments. We inferred that, in particular, open habitats are exploited more frequently than expected, indicating a high degree of plasticity in their trophic habitat needs. Therefore, protecting diverse, interconnected landscapes with varied prey is crucial for their conservation. 105 CHAPTER 3 3.1. INTRODUCTION Understanding habitat preferences of elusive, rare, or small foraging animals under 20g is challenging, even with GPS advancements. VHF radio tags are the primary tracking method for small passerines (Bauer et al., 2005) and bats (Dietz et al., 2016). However, analysing their diet can indirectly reveal the importance of some habitats by identifying food sources. This approach provides crucial habitat information for species conservation (e.g., Collins et al., 2005; Mitschunas & Wagner, 2015). Morphological diet studies were labour-intensive and often inaccurate, offering limited ecological and behavioural insights and capturing only static snapshots of bat diets (Hope et al., 2014; Alberdi et al., 2020). These technical limitations constrained accurate habitat identification. However, DNA metabarcoding now allows detailed species-level identification of prey remains (Krüger et al., 2014; Galan et al., 2018; Baroja et al., 2019), paving the way for more accurate foraging habitat identification (Clare et al., 2011; Stockdale, 2018). Such studies have provided insights into the habitats associated with bat prey (Alberdi et al., 2012). While they cannot pinpoint where prey was captured (sink habitat), they indicate prey origins (source habitat) (Arrizabalaga-Escudero et al., 2015). Molecular studies show bats consume hundreds of species that vary over time (Razgour et al., 2011; Arrizabalaga‐Escudero et al., 2015; Aihartza et al., 2023), space (Clare et al., 2014), and among individuals (Mata et al., 2016), adjusting their diet to prey availability (e.g., Almenar et al., 2013; Napal et al., 2013; Aihartza et al., 2023; Vallejo et al., 2023). Consequently, molecular diet analysis has become paramount for determining bats' foraging dependencies by identifying prey-origin habitats throughout the year. Different bat guilds, defined by echolocation and flight abilities, have varied foraging strategies (Fenton, 1990; Schnitzler & Kalko, 2001). Clutter specialist use short broadband calls to navigate dense vegetation, while aerial hawkers use high-frequency calls to catch insects in open spaces. Surprisingly, molecular diet studies show open-space bats also forage in diverse habitats, including forests, relying on prey availability peaks (Garin et al., 2019; Aihartza et al., 2023). Similarly, clutter specialist bats likely exploit prey outbreaks in diverse habitats, using their echolocation and flight skills at the microhabitat level by flying close to vegetation. The lesser horseshoe bat, Rhinolophus hipposideros (Bechstein, 1800), is the smallest European rhinolophid and uses high CF calls (about 112 kHz) to hunt near vegetation, making it a specialized clutter hunter (Jones & Rayner, 1989; Schofield, 1996). It is widely distributed in the western and central Palaearctic, found in almost all European countries (Taylor, 2016), partly due to its adaptability to anthropic areas. Despite this, it is classified as of community importance in the EU, requiring Special Areas of Conservation (Council Directive 92/43/EEC). The species is at risk due to the destruction of native woodlands, (Bontadina et al., 2002; Motte & Libois, 2002; Reiter, 2004; 106 CHAPTER 3 Zahn et al., 2008), roost loss or disturbance, artificial illumination and use of pesticides (Schofield et al., 2023). The lesser horseshoe bat prefers broadleaf woodland habitats, but also uses other habitats for foraging and commuting, such as grazed grasslands and areas with high habitat diversity (Bontadina, 2002). In agroforestry areas, it favours woodland patches and hedgerows near cultivated fields (Motte & Libois, 2002). The presence of maternity roosts, crucial for the species' persistence, depends on the availability of wooded elements near small built areas and the integration of roosts into a connected network (Motte & Libois, 2002; Tournant, 2013). R. hipposideros has a foraging area with an average radius of 600 m from home, with a maximum-recorded distance of 4.2 km from the roost (Bontadina et al., 2002). Therefore, its conservation relies on the availability of suitable roosting sites and productive foraging areas nearby. As a result, understanding their habitat dependence is essential. Their similar diet across pristine and modified habitats in Europe (e.g., McAney & Fairley, 1989; Russo & Jones, 2003; Lino et al., 2014; Baroja et al., 2019), suggests that maintaining high insect diversity is crucial for their foraging. Studying this species' dietary patterns in human-modified habitats may help enhance these areas for foraging (Lino et al., 2014). In this study, we analysed the diet changes of three maternity colonies of R. hipposideros in the Basque Country (Northern Iberian Peninsula) occurring in areas varying in landscape, urbanisation level, and climate. We examined the species’ diet across their active season, paying attention to the seasonal and latitudinal variations. We also analysed the habitats in which the most consumed prey depends. Our hypotheses were: The lesser horseshoe bat diet will change (1) with time, showing a seasonally shifting of most-consumed prey, and (2) with the site, depending on the prey availability of each colony. (3) Most consumed prey will depend on different landscapes or habitats, causing seasonal changes in the required environmental resources for the bats, and thus, (4) such prey-habitat dependence may also vary among colonies. 3.2. MATERIALS AND METHODS 3.2.1. Study Area We sampled three R. hipposideros maternity roosts with varying land use and climate (Figure 3.1). Using QGIS v3.16 and Spanish Government (SIGPAC) maps, we mapped habitat and land cover units within 1, 2, and 3 km buffers around the roosts. These buffers cover all potential foraging areas, as bats typically move up to 4.2 km from their roosts (Bontadina et al., 2002). Matxinbenta (MA): Over 100 R. hipposideros roosted in a rural church dome in a village surrounded by deciduous and conifer forests, meadows, and pasture. The area has an Atlantic temperate oceanic climate (293 MASL) with year-round rainfall (annual mean 1400 mm). 107 CHAPTER 3 Zarautz (ZA): About 100 R. hipposideros roosted in a church dome in an urban area on the Cantabrian coast. The surroundings include vineyards, fruit trees, pastures, and deciduous and coniferous forests. The area has an Atlantic temperate oceanic climate (4 MASL). Caseda (CA): Up to 30 R. hipposideros roosted in an empty building that was once part of a hydroelectric power plant. Located near a large river, the area is dominated by dry cereal farmland to the north and Mediterranean bush and forest to the south, with some vineyards and fruit orchards nearby. The weather is Mediterranean temperate, with hot, dry summers (367 MASL). Figure 3.1. Study Area: Primary habitat and land cover units within 3 km around Rhinolophus hipposideros roosts where dietary analysis was conducted. ZA: Zarautz colony; MA: Matxinbenta colony; CA: Caseda colony. Charts show habitat availability at 1, 2, and 3 km buffers. Sea in ZA is excluded as it’s not a prey source. Map created with QGIS v3.16, based on Spanish Government (SIGPAC) public maps. 3.2.2. Sampling Bat Droppings We sampled the three colonies in 2022, throughout the maternity season (mid-spring and summer). Samplings began in Julian Week 17 (late April) when bats arrived. MA bats were the last to leave (September 30th, Week 39), while bats from ZA and CA left two weeks earlier (Week 37). Sampling occurred every fortnight in even weeks using non-invasive methods to avoid disturbing the animals. We placed paper collectors under the animals to keep the droppings as dry as possible, and we examined the faeces visually to only collect the most recent bat faeces to avoid degradation of the sample's DNA. Paper was changed after each sampling. We collected up to 24 small “community samples” of four to six pellets each for each sampling event (adapted from Andriollo et al. 2019, 2021). These samples aim to reflect the diet of the entire maternity colony throughout the sampling 108 CHAPTER 3 period, rather than the contributions from individual animals. In total, 413 faecal samples for sequencing procedures and further analysis. 3.2.3. DNA Extraction, PCR Amplification, Library Preparation and Sequencing We extracted DNA from 413 frozen faecal samples weighing up to 40 mg (median of 19.4 mg) with the Kingfisher extraction tool (Thermo Fisher Scientific Inc.), previous homogenisation with Precellys 24 Touch Homogenizer (Bertin Technologies SAS, France). From the extracts, two 178 and 180 bp different mini-COI segments of the mitochondrial DNA cytochrome c oxidase subunit I (COI) barcode region were amplified with FWH1 (fwhF1/fwhR1) (Vamos et al., 2017) and ANML (CO1490/CO1‐CFMRa) (Jusino et al., 2019) primers, respectively. Amplifications were performed using the QIAGEN Multiplex PCR Kit (Qiagen Iberia, S.L. Madrid) in 25 μl PCR reactions, following the original protocols. PCR products were migrated in agarose gel electrophoresis to test the success of the amplification process. Subsequently, a second PCR reaction was performed to attach a unique combination of tags and Illumina sequencing adapters to each amplicon using the Nextera XT Index Kit (Illumina, 2013). Finally, 413 samples were pooled and sequenced using Illumina NovaSeq technology (SP Flow Cell kit v2 PE: 2 x 250bp (500 cycles) / 800 M reads). DNA library construction and sequencing processes were carried out at the Genomics and Proteomics General Service (SGIker) of the University of the Basque Country. 3.2.4. Bioinformatic analyses We performed separation by primers, quality control, sequence pre-processing, and collapsing identical sequences into single ones using CUTADAPT (Martin, 2011) and VSEARCH (Rognes et al., 2016). We clustered sequences into Operational Taxonomic Units (OTU) by VSEARCH at a 97% similarity threshold (Hebert et al., 2003) using the “–cluster_size” command. Subsequently, we cleaned up chimaera OTUs with VSEARCH’s “-uchime3_denovo” command. Only OTUs with an abundance higher than 0.5% in any sample unit were selected for the analysis (See Appendix B for script details). The taxonomic assignment of each OTU was performed by comparing the representative sequence against reference sequences in GenBank (www.ncbi.nlm.nih.gov; accesed on 2023-08-28) using the “-blastn” command (Chen et al., 2015). Only hits with pairwise identity above 98% and e-values below 1e-20 were accepted (Vesterinen et al., 2013; Clare et al., 2014) to ensure that the match did not occur by chance. The species databases of arthropods for Spain, France and Portugal were downloaded from GBIF (www.gbif.org; accessed on 2022-11-03) to verify that the identified species encompassed our study area. The Iberian Fauna Database (Iberfauna) was also checked for assuring the species distribution (iberfauna.mncn.csic.es; accessed on 2022-11-03). OTUs showing multiple potential assignments were carefully reviewed and refined manually to assign the most accurate taxa 109 CHAPTER 3 Figure 3.5. (a-b) RLQ analysis results for Rhinolophus hipposideros diet: (a) Eigenvalues and weekly colony sample scores (MA Green: Matxinbenta; ZA Blue: Zarautz; CA Orange: Caseda). (b) Prey trait coefficients. Panels display the first two axes with d-values indicating grid size. Monte-Carlo test: observed statistic = 0.419, p < 0.001. (c) 4th corner analysis results. Red indicates positive correlation, and blue indicates negative correlation (p < 0.001). Forests and shrub pastures were crucial for all colonies year-round. In temperate colonies (ZA, MA), they provided over 50% wPOO of diets, while, in the Mediterranean colony (CA), these habitats were less significant but still important (Figure 3.6). In MA, the diet consisted of 32.1% forest, 23.5% shrub pastures, and 13.5% grassland. Despite being the main prey-source habitat, forests were used less than available (Figure 3.6a). Key woodland species included dipterans Dicranomyia sp., Nephrotoma flavipalpis, Helina impucta, and Neolimonia dumetorus, along with lepidopterans Teleiodes wagae and Stathmopoda pedella. Shrub pasture species, such as the moth Argyresthia goedartella and hemipteran Piezodorus lituratus, were positively selected. Grasslands, with species like Limonia nubeculosa, Lasius sp., and Nabis pseudoferus, were also favoured, especially in summer. Riparian forests, with trichopterans 116 CHAPTER 3 Limnephilus rhombus and Stenophylax vibex, and fruit trees, including Cydia fagiglandana and Zeuzera pyrina, were important for the colony’s diet, especially later in the season (Figure 3.6a). In the ZA colony, forest and shrub pastures were the main prey-source habitats (32.9% and 21.6%, respectively), with "orchards" in third place at 12.9%. According to the forest availability, we can infer that they flew farther than 1km to hunt in these habitats (Figure 3.7b), the same as in shrub pastures. Orchards became significant from week 27, peaking in August (weeks 31-33) with intense predation of Opogona sacchari (Figure 3.6b). Although orchards were positively selected, urban species were rarely consumed, indicating this habitat was underused (Figure 3.7b). Figure 3.6. Percentage habitat contributions as prey sources (in wPOO in diet) to weekly diets of lesser horseshoe bats: colonies in CA: Caseda, MA: Matxinbenta and ZA: Zarautz. Numbers on the x-axis refer to Julian Weeks. Figure 3.7. Mosaic plots of available versus important habitats for core prey in Rhinolophus hipposideros diets: (a) MA: Matxinbenta; (b) ZA: Zarautz; (c) CA: Caseda. Square sizes show category percentages; colours indicate standard residuals—blue for positive selection, red for negative. Asterisks (*) mark habitats with significant positive selection (p < 0.05). 117 CHAPTER 3 In CA, forests remained a key prey-source-habitat but decreased to 21.9%, followed closely by riparian forest at 21.7% and grasslands at 20.8%. Shrub pastures contributed 17.8%, but habitat importance varied by date (Figure 3.6c). Woodland species' high consumption was linked to forest availability within 3 km, peaking in May-June (weeks 19-25) and midSeptember (week 37) (Figure 3.7c). However, their diet contribution declined in summer as other habitats, especially riparian forest, became more significant. The latter was heavily used from August, particularly in week 35, contributing 40% of the diet due to dipterans like Gonomyia tenella and Rheopelopia ornata, and the ephemeropteran Caenis luctuosa. In contrast, grasslands were a major prey source in spring, contributing up to 49% of the diet, while shrub pastures gained importance later. Croplands were underused except at the end of April (week 17). 3.4. DISCUSSION While bat diets are widely studied, this research, alongside Aihartza et al. (2023), is among the first to use detailed molecular data to infer habitat dependencies. The lesser horseshoe bats' diet has been addressed in various regions of Europe (McAney & Fairlay, 1989; Lino et al., 2014; Mitschunas & Wagner, 2015; Baroja et al., 2019), but our study is the most comprehensive to date, identifying prey at the species level using molecular methodologies, through extensive faecal sample analysis. Using NovaSeq "ultra-high-throughput sequencing" provided deeper sequencing than previous studies, with a consistent species-level prey identification. Additionally, by examining colonies in a climatic transition zone, we compared temperate and Mediterranean environments. Lastly, conducting the study throughout the maternity season offered a thorough understanding of lesser horseshoe bats' diet and ecology. The findings on the lesser horseshoe bat diet align with previous morphological studies (e.g., McAney & Fairley, 1989; Lino et al., 2014; Mitschunas & Wagner, 2015). Lepidoptera was the most consumed order, followed by Diptera and Hemiptera. Notably, caddisflies play a significant role in the Mediterranean colony (Caseda: CA), suggesting bats rely more on rivers and streams in Mediterranean regions, consistent with Lino et al. (2014). Previous molecular diet studies on this species are limited. The 197 taxa identified by Baroja et al. (2019) from May to September were all found in our research, with 63 part of our core diets. Aldasoro et al. (2019) (Chapter 1) assessed the diet in July in a similar area to the ZA and MA colonies, identifying key dipteran species such as Rhipidia maculata, Neolimonia dumetorum, Limonia nubeculosa, Dicranomyia modesta and Tipula helvola. Limonia nubeculosa and Neolimonia dumetorum were notably the most consumed species in the MA colony in late July. The species-level diet analysis supports our hypothesis of dietary differences between colonies. Our findings align with previous R. hipposideros (Mitschunas & Wagner, 2015) and other bat studies (Aizpurua et al., 2018; Vallejo et al., 2019), indicating that bat diets vary by location. Our results have 118 CHAPTER 3 also validated a seasonal diet shift within each colony, as bats take advantage of the alternation of occasional occurrence of many different prey peaks. This opportunistic behaviour aligns with patterns in insectivorous bats, which profit from hot spots and insect outbreaks (Aizpurua et al., 2013; Lino et al., 2014; Aihartza et al., 2023; Vallejo et al., 2023). In this context, we would like to emphasise the substantial consumption of the crop pest Opogona sacchari by the ZA colony. This observation underscores the species' opportunistic nature and its potential role as a consumer and controller of agricultural pests (Baroja et al., 2019; Chapter 5). Insectivorous bats use surrounding habitats in diverse ways. Fenton (1990) categorised them as "open-space foragers," "edge and gap foragers," and "narrow-space foragers." Schnitzler and Kalko (2001) categorised habitats as "uncluttered," "background-cluttered," and "highly cluttered," with subcategories based on hunting techniques: aerial and gleaning bats. R. hipposideros fits the "narrow-space forager" or "aerial in highly cluttered space" guilds. Radio-tracking studies have confirmed so, showing these bats mainly forage in woodlands and hedgerows (e.g., Motte & Libois, 2002; Russo & Jones, 2003; Reiter et al., 2013). Though R. hipposideros is thought to be restricted to specific habitats, it also uses grasslands, orchards (Zahn et al., 2008; Mitschunas & Wagner, 2015), and urban areas near building walls for hunting (Jones & Rayner, 1989). Dietz found that southern regions show a divergence from the prevalent use of forests in northern Europe, with gardens, orchards, hedgerows, and grazed pastures being most used in Bulgaria. This study confirms that the lesser horseshoe bat relies on diverse prey from various habitats. Our data suggests that although classified as "clutter-feeders", they forage for species originating in several vegetation types besides woodlands. Like for the "open-space forager" Miniopterus schreibersii (Aihartza et al., 2023), clutter specialist bats also shift their habitat reliance seasonally and across locations, tracking emerging prey outbreak peaks (Baroja et al., 2021; Aihartza et al., 2023; Vallejo et al., 2023). This adaptability underlines their generalist and opportunistic feeding behaviour, enabling them to exploit hot spot and abundant resources throughout the year. As previously pointed out (e.g., Lino et al., 2014; Arrizabalaga-Escudero et al., 2015; Ancillotto et al., 2023), we confirm the importance of protecting well-connected heterogeneous habitats, mosaic landscapes and ecotones with high prey diversity around their roosts to ensure the conservation of the lesser horseshoe bat. 119 CHAPTER 3 3.5. REFERENCES Aihartza, J., Vallejo, N., Aldasoro, M., García-Mudarra, J. L., Goiti, U., Nogueras, J., & Ibáñez, C. (2023). Aerospace-foraging bats eat seasonably across varying habitats. Scientific Reports, 13(1), 19576. Aizpurua, O., Budinski, I., Georgiakakis, P., Gopalakrishnan, S., Ibañez, C., Mata, V., ... & Alberdi, A. (2018). Agriculture shapes the trophic niche of a bat preying on multiple pest arthropods across Europe: Evidence from DNA metabarcoding. Molecular ecology, 27(3), 815-825. Aizpurua, O., Garin, I., Alberdi, A., Salsamendi, E., Baagøe, H., & Aihartza, J. (2013). Fishing long-fingered bats (Myotis capaccinii) prey regularly upon exotic fish. PLoS ONE, 8(11), e80163. Alberdi, A., & Gilbert, M. T. P. (2019). hilldiv: an R package for the integral analysis of diversity based on Hill numbers. Biorxiv, 545665. Alberdi, A., Garin, I., Aizpurua, O., & Aihartza, J. (2012). The foraging ecology of the mountain long-eared bat Plecotus macrobullaris revealed with DNA mini-barcodes. PLoS ONE, 7(4), e35692. Alberdi, A., Razgour, O., Aizpurua, O., Novella‐Fernandez, R., Aihartza, J., Budinski, İ., … & Gilbert, M. T. P. (2020). DNA metabarcoding and spatial modelling link diet diversification with distribution homogeneity in European bats. Nature Communications, 11(1). Aldasoro, M., Garin, I., Vallejo, N., Baroja, U., Arrizabalaga-Escudero, A., Goiti, U., & Aihartza, J. (2019). Gaining ecological insight on dietary allocation among horseshoe bats through molecular primer combination. PLoS ONE, 14(7), e0220081. Almenar, D., Aihartza, J., Goiti, U., Salsamendi, E. & Garin, I. (2013). Hierarchical patch choice by an insectivorous bat through prey availability components. Behavioral Ecology and Sociobiology, 67(2), 311–320. Ancillotto, L., Labadessa, R., Roscioni, F., Montioni, F., Fulco, E., Zollo, L., & Spilinga, C. (2023). Protected habitats support bats in Mediterranean dry grasslands. Science of The Total Environment, 882, 163415. Anderson, M. J. (2001). A new method for non‐parametric multivariate analysis of variance. Austral ecology, 26(1), 32-46. Anderson, M. J., & Walsh, D. C. (2013). PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: what null hypothesis are you testing?. Ecological monographs, 83(4), 557-574. Anderson, M. J., Ellingsen, K. E., & McArdle, B. H. (2006). Multivariate dispersion as a measure of beta diversity. Ecology letters, 9(6), 683-693. Andriollo T, Gillet F, Michaux JR, Ruedi M (2019). The menu varies with metabarcoding practices: A case study with the bat Plecotus auritus. PloS ONE, 14(7), e0219135. Andriollo T, Michaux JR, Ruedi M (2021). Food for everyone: Differential feeding habits of cryptic bat species inferred from DNA metabarcoding. Molecular Ecology, 30(18), 4584-4600. Arrizabalaga-Escudero, A., Garin, I., García-Mudarra, J. L., Alberdi, A., Aihartza, J., & Goiti, U. (2015). Trophic requirements beyond foraging habitats: The importance of prey source habitats in bat conservation. Biological Conservation, 191, 512-519. Baroja, U., Garin, I., Aihartza, J., Arrizabalaga-Escudero, A., Vallejo, N., Aldasoro, M., & Goiti, U. (2019). Pest consumption in a vineyard system by the lesser horseshoe bat (Rhinolophus hipposideros). PLoS ONE, 14(7), e0219265. Baroja, U., Garin, I., Vallejo, N., Aihartza, J., Rebelo, H., & Goiti, U. (2021). Bats actively track and prey on grape pest populations. Ecological Indicators, 126, 107718. Bauer, H. G., Bezzel, E., & Fiedler, W. (2005). Das Kompendium der Vögel Mitteleuropas. Alles über Biologie, Gefährdung und Schutz, 1, 3. Bontadina, F., Schofield, H., & Naef-Daenzer, B. (2002). Radio-tracking reveals that lesser horseshoe bats (Rhinolophus hipposideros) forage in woodland. Journal of Zoology, 258(3), 281-290. Chao, A., Gotelli, N. J., Hsieh, T. C., Sander, E. L., Ma, K. H., Colwell, R. K., & Ellison, A. M. (2014). Rarefaction and extrapolation with Hill numbers: a framework for sampling and estimation in species diversity studies. Ecological monographs, 84(1), 45-67. Chao, A., & Jost, L. (2012). Coverage‐based rarefaction and extrapolation: standardizing samples by completeness rather than size. Ecology, 93(12), 2533-2547. Chen, Y., Ye, W., Zhang, Y., & Xu, Y. (2015). High speed BLASTN: an accelerated MegaBLAST search tool. Nucleic acids research, 43(16), 7762-7768. Clare, E. L., Barber, B. R., Sweeney, B. W., Hebert, P. D. N., & Fenton, M. B. (2011). Eating local: influences of habitat on the diet of little brown bats (Myotis lucifugus). Molecular Ecology, 20(8), 1772-1780. 120 CHAPTER 3 Clare, E. L., Symondson, W. O., Broders, H., Fabianek, F., Fraser, E. E., MacKenzie, A., ... & Reimer, J. P. (2014). The diet of Myotis lucifugus across Canada: assessing foraging quality and diet variability. Molecular Ecology, 23(15), 3618-3632. Collins, P. W., Guthrie, D. A., Rick, T. C., & Erlandson, J. M. (2005). Analysis of prey remains excavated from an historic Bald Eagle nest site on San Miguel Island, California. In Proceedings of the Sixth California Islands Symposium. Institute for Wildlife Studies and National Park Service (Vol. 103120). Cuff, J. P., Windsor, F. M., Tercel, M. P., Kitson, J. J., & Evans, D. M. (2022). Overcoming the pitfalls of merging dietary metabarcoding into ecological networks. Methods in Ecology and Evolution, 13(3), 545-559. Deagle, B. E., Thomas, A. C., McInnes, J. C., Clarke, L. J., Vesterinen, E. J., Clare, E. L., ... & Eveson, J. P. (2019). Counting with DNA in metabarcoding studies: How should we convert sequence reads to dietary data?. Molecular ecology, 28(2), 391-406. Dietz, C., Nill, D., & von Helversen, O. (2016). Handbuch der Fledermäuse – Europa und Nordwestafrika (2nd ed.). Stuttgart: Kosmos. Directive, H. (1992). Council Directive 92/43/EEC of 21 May 1992 on the conservation of natural habitats and of wild fauna and flora. Official Journal of the European Union, 206(7), 50. Dray, S., & Siberchicot, M. A. (2017). Package ‘ade4’. France: Université de Lyon. Fenton, M. B., Swanepoel, C. M., Brigham, R. M., Cebek, J., & Hickey, M. B. C. (1990). Foraging behavior and prey selection by large slit-faced bats (Nycteris grandis; Chiroptera: Nycteridae). Biotropica, 22(1), 2-8. Galan, M., Pons, J. B., Tournayre, O., Pierre, E., Leuchtmann, M., Pontier, D., & Charbonnel, N. (2018). Metabarcoding for the parallel identification of several hundred predators and their prey: Application to bat species diet analysis. Molecular ecology resources, 18(3), 474-489. Garin, I., Aihartza, J., Goiti, U., Arrizabalaga-Escudero, A., Nogueras, J., & Ibáñez, C. (2019). Bats from different foraging guilds prey upon the pine processionary moth. PeerJ, 7, e7169. Hebert, P. D., Ratnasingham, S., & De Waard, J. R. (2003). Barcoding animal life: cytochrome c oxidase subunit 1 divergences among closely related species. Proceedings of the Royal Society of London. Series B: Biological Sciences, 270(suppl_1), S96-S99. Hope, P. R., Bohmann, K., Gilbert, M. T. P., Mendoza, M. L. Z., Razgour, O., & Jones, G. (2014). Second generation sequencing and morphological faecal analysis reveal unexpected foraging behaviour by Myotis nattereri (chiroptera, vespertilionidae) in winter. Frontiers in Zoology, 11(1), 39. Hsieh, T. C., Ma, K., & Chao, A. (2016). iNEXT: an R package for rarefaction and extrapolation of species diversity (Hill numbers). Methods in ecology and evolution, 7(12), 14511456. Jones, G., & Rayner, J. M. (1989). Foraging behaviour and echolocation of wild horseshoe bats Rhinolophus ferrumequinum and R. hipposideros (Chiroptera, Rhinolophidae). Behavioral Ecology and Sociobiology, 25, 183-191. Jusino, M. A., Banik, M. T., Palmer, J. M., Wray, A. K., Xiao, L., Pelton, E., ... & Lindner, D. L. (2019). An improved method for utilizing high‐throughput amplicon sequencing to determine the diets of insectivorous animals. Molecular ecology resources, 19(1), 176190. Krüger, F., Clare, E. L., Symondson, W. O., Keišs, O., & Pētersons, G. (2014). Diet of the insectivorous bat Pipistrellus nathusii during autumn migration and summer residence. Molecular ecology, 23(15), 3672-3683. Lino, A., Fonseca, C., Goiti, U., & Pereira, M. J. R. (2014). Prey selection by Rhinolophus hipposideros (Chiroptera, Rhinolophidae) in a modified forest in Southwest Europe. Acta Chiropterologica, 16(1), 75-83. Martin, M. (2011). Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet. journal, 17(1), 10-12. Mata, V. A., Amorim, F., Corley, M. F., McCracken, G. F., Rebelo, H., & Beja, P. (2016). Female dietary bias towards large migratory moths in the European free-tailed bat (Tadarida teniotis). Biology letters, 12(3), 20150988. McAney, C. M., & Fairley, J. S. (1989). Analysis of the diet of the lesser horseshoe bat Rhinolophus hipposideros in the West of Ireland. Journal of Zoology, 217(3), 491-498. Mitschunas, N., & Wagner, M. (2015). Diet of the lesser horseshoe bat (Rhinolophus hipposideros) in Central Germany and its seasonal and site-specific variation. Acta Chiropterologica, 17(2), 379-392. Motte, G., & Libois, R. (2002). Conservation of the lesser horseshoe bat (Rhinolophus hipposideros Bechstein, 1800) (Mammalia: Chiroptera) in Belgium. A case study of feeding habitat requirements. Belgian Journal of Zoology, 132(1). Napal, M., Garin, I., Goiti, U., Salsamendi, E., & Aihartza, J. (2013). Past deforestation of Mediterranean Europe explains the present distribution of the strict forest dweller Myotis bechsteinii. Forest Ecology and Management, 293, 161-170. 121 CHAPTER 3 Oksanen J., Blanchet F. G., Friendly M., Kindt R., Legendre P., McGlinn D., …, & Wagner H. (2022). Vegan: Community Ecology Package. R package version 2.6–4. Pianka E. R. (1973). The structure of lizard communities. Annual review of ecology and systematics, 4(1), 53–74. Razgour, O., Clare, E. L., Zeale, M. R., Hanmer, J., Schnell, I. B., Rasmussen, M., ... & Jones, G. (2011). High‐throughput sequencing offers insight into mechanisms of resource partitioning in cryptic bat species. Ecology and evolution, 1(4), 556-570. Reiter, G. (2004). Postnatal growth and reproductive biology of Rhinolophus hipposideros (Chiroptera: Rhinolophidae). Journal of Zoology, 262(3), 231-241. Reiter, G., Pölzer, E., Mixanig, H., Bontadina, F., & Hüttmeir, U. (2013). Impact of landscape fragmentation on a specialised woodland bat, Rhinolophus hipposideros. Mammalian Biology, 78(4), 283-289. Rognes, T., Flouri, T., Nichols, B., Quince, C., & Mahé, F. (2016). VSEARCH: a versatile open source tool for metagenomics. PeerJ, 4, e2584. Russo, D., & Jones, G. (2003). Use of foraging habitats by bats in a Mediterranean area determined by acoustic surveys: conservation implications. Ecography, 26(2), 197209. Schnitzler, H. U., & Kalko, E. K. (2001). Echolocation by insect-eating bats: we define four distinct functional groups of bats and find differences in signal structure that correlate with the typical echolocation tasks faced by each group. Bioscience, 51(7), 557-569. Schofield, H. W. (1996). The ecology and conservation biology of Rhinolophus hipposideros, the lesser horseshoe bat (Doctoral dissertation, University of Aberdeen). Schofield, H., Reiter, G. & Dool, S. E. (2023). Lesser Horseshoe Bat Rhinolophus hipposideros. In Russo D. (Ed.), Chiroptera (pp. 875-908). Springer Nature. Stockdale, J. (2018). Using high-throughput sequencing to track habitat use by thrushes exploiting heterogeneous farmland landscapes (Doctoral dissertation, Cardiff University). Taylor, P. (2016). Rhinolophus hipposideros. The IUCN Red List of Threatened Species 2016: e.T19518A21972794. https://www.iucnredlist.org/species/19518/21972794. Accessed on 20 June 2024. Tournant, P., Afonso, E., Roué, S., Giraudoux, P., & Foltête, J. C. (2013). Evaluating the effect of habitat connectivity on the distribution of lesser horseshoe bat maternity roosts using landscape graphs. Biological Conservation, 164, 39-49. Tournayre, O., Leuchtmann, M., Galan, M., Trillat, M., Piry, S., Pinaud, D., ... & Charbonnel, N. (2021). eDNA metabarcoding reveals a core and secondary diets of the greater horseshoe bat with strong spatio‐temporal plasticity. Environmental DNA, 3(1), 277296. Vallejo, N., Aihartza, J., Goiti, U., Arrizabalaga-Escudero, A., Flaquer, C., Puig, X., ... & Garin, I. (2019). The diet of the notch-eared bat (Myotis emarginatus) across the Iberian Peninsula analysed by amplicon metabarcoding. Hystrix, 30(1), 59. Vallejo, N., Aihartza, J., Olasagasti, L., Aldasoro, M., Goiti, U., & Garin, I. (2023). Seasonal shift in the diet of the notched-eared bat (Myotis emarginatus) in the Basque Country: from flies to spiders. Mammalian Biology, 103(4), 419-431. Vamos, E. E., Elbrecht, V., & Leese, F. (2017). Short COI markers for freshwater macroinvertebrate metabarcoding. PeerJ Preprints, No. e3037v2. Vesterinen, E. J., Lilley, T., Laine, V. N., & Wahlberg, N. (2013). Next generation sequencing of fecal DNA reveals the dietary diversity of the widespread insectivorous predator Daubenton’s bat (Myotis daubentonii) in Southwestern Finland. PLoS ONE, 8(11), e82168. Zahn, A., Holzhaider, J., Kriner, E., Maier, A., & Kayikcioglu, A. (2008). Foraging activity of Rhinolophus hipposideros on the island of Herrenchiemsee, Upper Bavaria. Mammalian Biology, 73(3), 222-229. 122 CHAPTER 3 3.6. SUPPORTING INFORMATION Supporting Information 3.1. Table A. Species of the core diets used for the Landscape analysis and their habitat descriptions. Table B. Bibliography for assessing their source habitats (xlsx) https://tinyurl.com/aldosoroS31 Supporting Information 3.2. Table A. Identified prey taxa and their wPOO (weighted percentage of occurrence) and FOO (frequency of occurrence) for each colony; Table B. Presence/absence values for individual bats. Table C. wPOO values for individual bats. (xlsx) https://tinyurl.com/aldasoroS32 123 CHAPTER 4 Seasonal and geographic variation in the trophic ecology and habitat dependence of Rhinolophus ferrumequinum CHAPTER 4 rarefaction curves with iNEXT to determine whether the colonies had reached a species diversity asymptote, indicating that the sample sizes were sufficient for meaningful comparisons. We compared the differences in the diet composition between colonies using Pianka’s (1973) measure of niche overlap with the pianka function in the package EcoSimR (Gotelli et al. 2015) at the order and species level. The following steps were carried out using the R package vegan (Oksanen et al. 2020). We calculated Bray-Curtis dissimilarities between all samples using the function vegdist (Oksanen et al., 2020). Using the adonis function, we explored the relationship of diet composition with Month, Colony and their interaction through PERMANOVA (Anderson, 2001) at the order and species level. We performed an NMDS (Non-metric Multi-Dimensional Scaling) with the metaMDS function to visualise differences in species composition between samples. See Appendix C for R script details. 4.2.5. Characterization of Prey’s Source Habitats We collected information on the habitat preferences of prey from various bibliographic sources (Supporting Information S4.1B). Next, we categorised these habitats based on land use, aligning them as closely as possible with the SIGPAC habitats. This process enabled us to create groupings of habitats that have functional value for bats, considering the level of clutter and the potential productivity of prey (Table 4.1). This classification will be referred to as “prey-source-habitat” from now on. Table 4.1. Habitat categories for the second RLQ analysis and their description. Habitat category Description Cropland Agricultural fields, cropland Forest Conifer or broadleaved woodland Riparian forest Rivers, streams, and riparian vegetation Other water bodies Ponds, small streams without riparian vegetation Grassland Grassland, meadows pastures, grazing pastures (cattle), wetlands Shrub grassland Scrubs, shrubland, heathland, woodland edges, hedgerows Vineyards Vineyards Fruit trees Fruit-trees, nut-trees, olive grooves Orchards Orchards Greenhouse Greenhouses, crops under plastic Urban Urban, artificial habitats… 4.2.6. Landscape inference analysis We performed RLQ and fourth-corner analyses using the ade4 package for R (Thioulouse et al., 2018) to explore how relationships with prey source habitats vary across different locations and seasons. We aimed to emphasize the significance of the source-habitats of prey and to determine whether particular taxa were consumed in specific habitat types, thereby identifying habitat-specific prey groups. In this way, we analysed how the consumption of habitat-specific prey groups varies monthly in the three 132 CHAPTER 4 colonies. This approach is based on constructing three matrices according to the method outlined by Arrizabalaga-Escudero et al., (2019). In our dataset, the sampling event (colony and sampling month) represents the “environmental variables”, the habitat-specific prey group (e.g., Forest Lepidoptera, Grassland Coleoptera…) represents the “trait” matrix, and the interaction matrix consists of the wPOO abundance of each prey species within each bat’s diet. We restricted the analysis to the core diets of each bat colony (Supporting Information S4.1A) to obtain a better picture of the source habitats of the most consumed prey, which in turn were possibly positively selected. Before the RLQ analysis, multivariate ordination is needed on all three matrices, namely a Correspondence Analysis on the L matrix since the latter includes quantitative data, and Multiple Correspondence Analysis (MCA) on matrices R and Q, which include categorical variables. We performed a randomization test using the function randtest to test the relationship between the environmental matrix (R) and the traits matrix (Q) through the species' occurrence or abundance (L) (Model 2). The number of permutations was elevated to 9,999. This model examines how traits respond to environmental traits indirectly via species distributions. In our case, we investigatedhow habitat-specific prey groups respond to seasonality (sampling month) and space (colony). After finding that sampling events were grouped by colony (see results, Figure 4.4), we performed an RLQ analysis for each colony to see the seasonality of habitat-specific prey groups without the interference of the site. For the three buffers in each colony —see above— we compared habitat availability with their trophic contribution (used habitat from now on), calculated by multiplying each prey-source-habitat availability by their wPOO values in the diet. This way, we obtained the “use” values for each of the prey-source-habitats. After that, we performed a Pearson's Chi-Square Test to find correlations between habitat availability and use, and we added the Monte Carlo test option to obtain a statistical significance p-value (simulate.p.value = T). See Appendix C for R script details. 4.3. RESULTS 4.3.1. Diet composition After sequencing, we obtained valuable sequences from 377 out of 379 samples. The NovaSeq sequencing led to 66,726,490 reads, which generated up to 150,000 OTUs, of which only 5,349 had an abundance higher than 0.5% in any of the samples. These results provided a comprehensive view of the bat's diet composition. We identified 2320 OTUs (43.37% of the total): five OTUs belonged to the predator, Rhinolophus ferrumequinum (1.85% of the reads), 200 (4.43% of reads) were classified as environmental contamination, and 2,115 (51.22% of reads) were identified as potential prey items. The remaining 3,029 (42.5% of reads) did not match with any sequence in the databases, at least at the 98% identity level. 133 CHAPTER 4 CA and UX colonies were mixed along with Myotis emarginatus, and after sequencing, we detected DNA contamination in some of the samples. Therefore, we decided to withdraw the contaminated samples, merge the clean samples, and analyse them based on the sampling month instead of the week. This decision was made to ensure a statistically valuable sample size and to account for potential variations in the bat's diet over longer periods. A possible correlation between Myotis emarginatus OTUs and prey species was analysed after filtering samples. No clear correlations were found, leading us to assume that the species examined in the diet were potential prey of Rhinolophus ferrumequinum. After discarding the contaminated samples, we worked with 231 small community samples. We identified up to 924 prey species belonging to 624 genera, and another 44 prey taxa were recognised at the genus level. In addition, 34 genera had OTUs identified at both the genus and species levels. In these cases, a genus-level occurrence was added to the dietary data. Overall, we analysed 1,002 prey taxa, identified at the genus or species levels, belonging to 14 orders (Supporting Information S4.2). Our data analysis was thorough, ensuring a comprehensive understanding of the bat species' diet. The most consumed order was Lepidoptera (wPOO 46.99%), followed by Diptera (26.67%) and Coleoptera (6.79%) (Figure 4.2). The most frequent prey species was the moth Coleophora sp. —Fam. Coleophoridae—, identified in 66 samples (Frequency of occurrence (FOO) 27,85%), accounted for 2.07% wPOO. The next two most frequent prey items were the caddisfly Hydropsyche exocellata —Fam. Hydropsychidae—, and the moth Opogona sacchari —Fam. Tineiidae—, identified in 63 and 52 samples, with 1.58% and 0.85% wPOO, respectively. Most prey items appeared occasionally in faeces: 449 out of 1,002 prey taxa were only identified in a single sample each, 166 in two samples, 302 in a range between three and ten samples, and 87 species occurred in more than ten samples. Finally, we added the species with a frequency of occurrence (FOO) greater than 5% (Tournayre et al., 2021) of each colony to obtain the core diets. These core diets, a significant part of our findings, gave a total of 173 taxa (species and genera) for the analysis. We identified 695 prey taxa for the diet of the Balmaseda (BA) colony. No prey type predominated with relevance in the diet, as the banana moth Opogona sacchari was the most consumed species in the whole season (accounting for 1.96% of the wPOO) and from May to July. The second most consumed species was the trichopteran Hydropsyche exocellata (1.34%), peaking in July and August, and thirdly, we found the limonid Dicranomyia sp. (1.27%) with a consumption maintained over time. In the diet of the Caseda (CA) colony, we identified 528 prey taxa. The most consumed species was the moth Eurodachtha canigella (2.83%), followed by the caddisfly Hydropsyche exocellata (2.25%), both especially consumed in spring, and the moth Coleophora sp. (1.95%) showing a sustained consumption until July. Finally, we assigned 289 prey taxa for the Uxue (UX) colony's diet. The most consumed species were moths. Coleophora sp. (4.09%) was widely consumed during the season, Phalonidia contractana 134 CHAPTER 4 (3.98%) was especially consumed in March, while Rhodometra sacraria (2.91%) was mainly consumed in March and June. Figure 4.2. (a) Total diet in the three colonies of Rhinolophus ferrumequinum, showing the consumed prey at the order level, measured as the weighted percentage of total occurrences (wPOO) (Deagle et al., 2019) and monthly diet of (b) BA, Balmaseda (c) CA, Caseda and (d) UX, Uxue. Axis y represents the wPOO values, while Axis x indicates the colony in plot (a) and the month in plots (b), (c), and (d). 4.3.2. Multivariate analysis of diet variability and homogeneity The colonies reached sample coverage of over 75% (81% BA, 80% CA and 77% UX). Our Pianka's niche overlap analysis at the prey order level revealed that all three colonies share a similar general diet. However, their diets varied significantly at the prey species level, with a similarity index of 46.52%. Notably, the geographic distance between the colonies played a significant role in these dietary differences. For instance, BA and UX, being the most distant, exhibited the most distinct diets, with a similarity index of only 33.90%, while UX and CA, being closer and having a more similar climate, had the most similar diets at 59.43%. BA and CA, falling in between, showing a 46.25% 135 CHAPTER 4 similarity. Regarding the core diet, the differences between colonies increased, with a total similarity index of 20.55%. Figure 4.3. NMDS multidimensional ordination of the samples based on their diets measured as wPOO, considering the whole diet. BA: Balmaseda, CA: Caseda, UX: Uxue. Points belong to bat faecal samples, and colours convey the different sampling dates. Grey dots show samples from the other colonies. The PERMANOVA analysis confirmed what the NMDS suggested (Figure 4.3), indicating that diet composition at the species and order level differed significantly based on the variables "colony" and "sampling month", as well as their interaction at the species level (p < 0.001; R2 = 0.044; R2 = 0.054; R2 = 0.055, respectively), and at the order level (p < 0.001: R2 = 0.046; R2 = 0.075; and p = 0.002; R2 = 0.055, respectively). Additionally, significant differences (p < 0.001) were observed between the sampling months when analysing the diets of each colony separately at the species level (R2BA = 0.122; R2CA = 0.097; R2UX = 0.112, respectively). 4.3.3. Landscape inference analysis The RLQ randomization tests were statistically significant for the whole analysis and for the three colonies (Model 2: p < 0.0001), testing the relationship between sampling event/seasons (R) and the habitat-prey group association (Q) through the diet matrix (L). In the complete RLQ analysis, the first two axes accounted for 24.97% and 39.24% of the total variance. The ordination showed a clear grouping of sampling events by colonies, suggesting that the site plays a crucial role in the differing consumption of habitat-specific prey groups (Figure 4.4a). BA sampling events were grouped on the right side of the plot and were associated with varying groups, such as orchardand 136 CHAPTER 4 urban-lepidopterans or diverse forest groups. Besides, CA events were mainly grouped on the down-left, associated with riparian species and some cropland groups. Finally, UX events were grouped on the top left and linked to grassland and shrubland groups. Nonetheless, no clear association can be made from this analysis. After performing the RLQ for each colony, we confirmed that the consumption of some habitat-specific prey groups was associated with specific months (Figure 4.3a,b,c). In contrast, the consumption of other groups showed a more homogeneous distribution over time. However, these associations differ between colonies. In the RLQ analysis for BA, the first two axes accounted for 44.8% and 22.6% of the total variance, respectively (see Figure 4.4b). We observed that during spring (April-May), forest dipterans were the most commonly consumed prey group in BA and remained significant throughout the entire season. Diptera associated with different habitats were consumed homogeneously, although especially prevalent during spring. In April, grassland coleopterans, dipterans, and hymenopterans were highly consumed. Starting in May, the orchard lepidopteran Opogona sacchari became one of the most consumed species and continued to be significant until the end of the season. Forest coleopterans were primarily consumed from May to July, with Stenagostus rhombeus being the dominant species in May and June, while Arhopalus rusticus became more prevalent in July and August. During summer, trichopterans, linked to riparian habitats, emerged as a key prey with a significant intake of Hydropsyche exocellata. Forest lepidopterans were mainly associated with July, highlighting the consumption of Thaumetopoea pityocampa or Dendrolimus pini, species linked to pine plantations. In contrast, grassland and shrubland lepidopterans showed a more general intake and were not explicitly linked to any particular month (see Figure 4.4b). For CA, the first two axes of the RLQ accounted for 43.4% and 24.9% of the total variance, respectively (see Figure 4.4c). Riparian prey groups were primarily associated with the early season. In April, the most consumed species was the trichopteran Hydropsyche exocellata. Forest and grassland coleopterans were linked to the summer months (June to August), including the dung geotrupids Typhaeus typhoeus and Geotrupes sp. and forest Trichoferus fasciculatus. Prey groups related to croplands were also prevalent in summer (July and August), with notable consumption of species like Tipula repanda (Diptera), and Agrotis segetum (Lepidoptera). In September, the primary prey species shifted back to lepidopterans mainly associated with grasslands and shrublands. Lepidoptera associated with fruit trees (Cydia fagiglandana) were strongly linked to September. However, these groups did not show a clear association because the consumption of lepidopterans from shrublands, grasslands, and forests was distributed throughout the season. Therefore, they were positioned in the middle of the plot (see Figure 4.4c). 137 CHAPTER 4 Figure 4.4. Results of RLQ analyses for the Rhinolophus ferrumequinum diet, all together (a) and by colony: (b) B: Balmaseda; (c) C: Caseda; (d) U: Uxue. The plots on the left show the eigenvalues and scores of colony monthly samples. The plots on the right side show coefficients for prey traits (habitat-specific prey groups). Panels display the first two axes only, with d‐values referring to grid size. 138 CHAPTER 4 Figure 4.5. Percentage contribution of habitats as core prey source (in wPOO in diet) to the monthly diets of the greater horseshoe bats colonies in CA: Caseda, BA: Balmaseda, and UX: Uxue. Finally, in the RLQ analysis for UX, the first two axes explained 46.7% and 32.7% of the variation, respectively (see Figure 4.4d). The consumption of dung coleopterans (Geotrupes sp.) distinguished May from the other months. In contrast, forest coleopterans (Stenagostus rhombeus) were consumed during March, April, and June. Grassland dipterans like Symplecta stictica were predominantly consumed in March. Forest and shrubland lepidopterans, and shrubland dipterans and hemipterans, were especially associated with April and July. Ephemeropterans (Habrophlebia lauta) were linked to both April and May. Forest species gained prominence in June, significantly consuming lepidopterans (Coleophora sp.) and dipterans (Zaira cinerea). Finally, in July, the primary prey group consisted of moths from shrubland (Gymnoscelis rufifasciata), although moths from both forest and grassland were also consumed in large quantities. Grassland lepidopterans were located in the right part of the plot because they were consumed throughout the season, except in May, when Geotrupes were the most consumed prey (see Figure 4.4d). Forests were an important prey-source-habitat throughout the year, but did not have the same significance in the different colonies and throughout the entire season (Figure 4.5). They were the most used prey-source-habitat in the temperate climate colony (BA), with a mean value of 45% of the diet. In the Mediterranean colonies, (CA and UX) forests were relatively less significant, and grasslands and shrub grasslands gained importance. 139 CHAPTER 4 Figure 4.6. Mosaic plots of available habitat types versus their importance as prey source habitats of the most consumed prey species in Rhinolophus ferrumequinum diet: colonies in (a) BA, Balmaseda (b) CA, Caseda (c) UX, Uxue. The habitat categories are arranged in descending order according to their use. The size of the squares represents the combined percentage of that category in the dataset, and colours show the values of the standard residuals. Blueish colours represent a positive selection and reddish negative selection. The asterisks (*) show the habitats that have been positively selected with significant Pearson’s correlation coefficient (p > 0.05). Boxes with an “X” represent the absence of such habitat. In BA, 45.2% of the core diet came from woodlands, followed by grasslands (18.5%) and shrub grasslands (16.4%) (Figures 4.5a, 4.6a). As prey-source-habitat, woodlands were underused relative to availability, while grasslands and shrub grasslands were used proportionally (Figure 4.6a). Riparian forests, instead, were important in July and August. In the CA colony, shrub grasslands were the most used core prey-source-habitat (26.4%), followed by forests (23.7%) and grasslands (22.1%) (Figures 4.5b, 4.6b). Forests over 1 km away were favoured, and grasslands were also positively selected (Figure 4.6b). Riparian forests were positively selected and were crucial in April-May (Figure 4.5b). Finally, croplands gained importance in July-August. In contrast, fruit trees were more used in September, even if they were overall underused (Figure 4.5b). Finally, in UX, grasslands (27.8.5%) and forests (25.9%) were the main prey source habitats, with their importance varying by month. Both shrub grasslands and riparian forests accounted for 19.9% of usage (Figures 4.5c and 4.6c). Shrub grasslands peaked in July with 43.3% of consumption, while riparian forests were crucial in spring. Orchards were used according to availability, particularly in March, May, and June (Figures 4.5c, 4.6c). Grasslands and forests were positively selected, while riparian forests were favoured within 1-3 km (Figure 4.6c). Prey species linked to shrub grasslands were used less than available, and it is noteworthy to mention that taxa from croplands were strongly underused. 140 CHAPTER 4 4.4. DISCUSSION Our study, which delves into two of the three critical aspects of bat conservation identified by Fenton (1997) and Pierson (1998) —the prey base and the inference of their foraging requirements— has significantly advanced our understanding of the greater horseshoe bats' fundamental trophic niche. On the first subject, we have unveiled substantial differences in the species’ diet and main prey-source-habitats in temperate and Mediterranean regions and seasonally, introducing a wealth of new knowledge to the field of bat' ecology. Our comprehensive study, in conjunction with Tournayre’s research (2021), indicates that Rhinolophus ferrumequinum has a remarkably diverse diet. Our sampling period is the most extensive to date, encompassing three distinct colonies in vastly different climates and with varied landscapes. Furthermore, the NovaSeq sequencing method we employed has a higher sequencing depth than any previously used, enabling us to identify approximately one thousand prey taxa down to the species or genus levels. Our initial hypothesis was supported by the results, indicating that prey availability is a crucial factor influencing the composition and spatio-temporal variations of the R. ferrumequinum diet (Tournayre et al., 2021), as well as the relative importance of different prey groups. Overall, the consumption of Lepidoptera, the primary order of prey, remained stable. However, it exhibited variations linked to the consumption of other prey groups, such as Diptera, Coleoptera, and Hemiptera. Instead, the source habitat of these prey groups was found to vary throughout the season. Even if the core diets primarily consisted of Lepidoptera, Diptera and Coleoptera (Jones, 1990; Flanders & Jones, 2009; Tournayre et al., 2021), other orders, such as Trichoptera, were significant seasonally, particularly in spring. The greater horseshoe bat is known for hunting beetles (Jones, 1990; Ransome, 1996, 2002), but their consumption varies by colony in species, timing, and habitat. In UX, Geotrupes sp. was the most consumed species in May, with beetles being less relevant in other months. In CA, while beetles were consumed throughout the season, they were not the main prey. In BA, beetles constituted a significant part of the diet from May to August, with prey species changing seasonally and being linked to forest habitats rather than grazed pastures. Our results suggest that the species' ability to select prey, a significant trait in bat ecology (Koselj et al., 2011), is flexible, depending on the habitat and season. These findings are consistent with a general trend observed in insectivorous bats, highlighting their adaptable behaviour in exploiting hotspots and insect outbreaks (Aizpurua et al., 2013; Lino et al., 2014; O’Rourke et al., 2022; Aihartza et al., 2023; Andreas et al., 2023; Vallejo et al., 2023). On the second matter, regarding the foraging habitat requirements of the greater horseshoe bats, it is worth mentioning that our landscape inference analysis underlines the relative importance of their main prey-source-habitats. This evaluation, while not determining where such main prey have been 141