a molecular approach to the study of birds’ dna in bats...
TRANSCRIPT
1
A molecular approach to the study of birds’ DNA in bats faeces 1
2
David Pastor-Beviá*, Carlos Ibáñez, Juan Luís García-Mudarra, Javier Juste 3
Estación Biológica de Doñana, Consejo Superior de Investigaciones Científicas (CSIC), 4
Sevilla, Spain 5
6
*Corresponding author: David Pastor-Beviá 7
Address: Estación Biológica de Doñana, Consejo Superior de Investigaciones Científicas, c/ 8
Américo Vespucio, s/n, 41092, Seville, Spain 9
Phone: +34669026480 10
E-mail: [email protected] 11
12
Running title: Birds DNA study in bat faeces. 13
14
Keywords: giant noctule bat, faeces, DNA extraction, conservation, primer design, bird 15
DNA, amplification success. 16
17
18
Abstract 19
20
The molecular identification of prey in faeces is an efficient non-invasive technique to 21
study diet which requires both a satisfactory method of DNA extraction and the design of 22
specific primers to selectively amplify prey’s DNA. In this study we evaluated and compared 23
the efficiency of two total DNA extraction methods and five primer pairs for the molecular 24
2
identification of birds from scats, in particular from the giant noctule bat (Nyctalus 25
lasiopterus). A modified DNA stool Mini Kit of Qiagen was tested against a modified silica 26
method with a guanidinium thiocianate (GuSCN) applied after freezing and pulverizing the 27
samples. We also checked two published vertebrate- and bird-generalist primer pairs and three 28
bird-specific primer pairs designed by us (two pairs targeting the cytochrome b and one the 29
cytochrome oxidase subunit I genes) that amplified shorter DNA fragments. The results show 30
that pulverizing the scat remains before extraction was a very important step, presumably 31
facilitating access to the well preserved DNA located inside the rachis of the feathers. The 32
combination of our bird-specific designed primers showed a higher amplification rate than the 33
generalist primers and allowed successful bird identification from the feathers excreted by the 34
giant noctule bat in all the scat samples analyzed, independent of the conservation method 35
used (dried and frozen). These methodological improvements will allow not only the study of 36
the avian diet composition of the enigmatic giant noctule, but the extension of this 37
methodology to other bird predators such as raptors. 38
39
Introduction 40
41
Understanding food webs is a central question in ecology (Montoya & Solé, 2002). 42
Predator-prey interactions have long captured researchers’ attention because of the important 43
impacts that these interactions have both on prey and predator population dynamics and on 44
the functioning of the complete ecosystem (Lima, 1998). The selection of a particular prey is 45
a complex decision process that results from integrating factors such as prey size, density, and 46
availability (Griffiths, 1975; Symondson, 2002). Direct observations of feeding events are 47
often difficult in the field and diets are studied conventionally by the morphological 48
3
identification of remains in faeces which many times is a challenging task due the effect of 49
mastication and digestion processes on these remains. 50
In recent times molecular and particularly, PCR-based techniques have proved to be 51
highly effective and versatile for studying diets and are likely to rapidly displace other 52
approaches (Symondson, 2002). PCR-based techniques target preys DNA in predators´ 53
faeces. Nevertheless, detecting and amplifying degraded or semi-digested DNA is not always 54
an easy task mainly due to the presence in faeces of both reaction inhibitors and multiple 55
DNA templates. Besides, due to its higher proportion, the predator’s DNA can swamp the 56
result of the PCR amplification, although this problem can be normally avoided by using 57
species- or group-specific primers that exclusively target prey DNA sequences (King et al., 58
2008). As a result, prey identification through the amplification of the prey´s DNA is now 59
possible even from highly degraded samples such as faeces, guts content or regurgitates 60
(Zeale et al., 2011). This possibility is changing the perspective of diet studies, and in fact the 61
approach has allowed identifying prey species in a variety of biological remains from spiders 62
(Agustí et al., 2003) to pinnipeds (Parsons et al., 2005), Macarony penguins (Deagle et al., 63
2007) or bats (Clare et al., 2009; Zeale et al., 2011; Alberdi et al., 2012).Mitochondrial DNA 64
(mtDNA) has been widely employed in species identification due to its relative abundance 65
and high mutation rate, which facilitates the distinction between species, even between close 66
related ones (Brown et al., 1979). For this reason, mtDNA markers have been widely used in 67
diet studies, which usually rely on the partial or complete sequencing of genes such as 68
cytochrome b (cyt b) or cytochrome oxidase subunit I (COI). Particularly, the use of COI is 69
increasing owing primarily to its adoption by the Barcode for Life Consortium (Tobe et al., 70
2010). DNA barcoding employs sequence diversity in short, standardized gene regions to aid 71
species identification and discovery in large assemblages of species. The Barcode of Life 72
4
Data System (BOLD) — www. barcodinglife.org — provides an integrated bioinformatics 73
platform that supports all phases of the analytical pathway from specimen collection to a 74
tightly validated barcode library (Ratnasingham & Hebert, 2007). 75
Bats typically show a wide variety of diets that range from arthropods, vertebrates, 76
fruits, nectar and pollen, to leaves and even blood. This wide diversification of diet is striking 77
in comparison to other mammal groups or even vertebrates (Altringham, 2011). In temperate 78
regions, bats are mainly insectivorous and they are major consumers of nocturnal insects 79
many of which are economically important pests (Agosta, 2002). 80
Diet studies of insectivorous bats traditionally relied on the morphological identification 81
in faeces of microscopic remains –primarily fragments of arthropod cuticles- resulting from 82
chewing prey into tiny pieces. These remains do not allow taxonomic identification normally 83
further than to Order, which usually provides only a gross estimate of the number and 84
diversity of insects eaten, hence providing only a reduced perspective on the real diet 85
(Whitaker et al., 2009; Zeale et al., 2011). On the other hand, the use of molecular techniques 86
has increased the accuracy of prey identification in bat diet studies even to the species level 87
(Zeale et al., 2011). 88
The giant noctule bat (Nyctalus lasiopterus) was found to be the first Palaearctic bat to 89
feed on flying passerines after the discovery of feathers in their faecal pellets (Dondini & 90
Vergari, 2000; Ibáñez et al., 2001). Small birds are preyed upon during their fall and spring 91
migrations (Ibáñez et al., 2001). This seasonal feeding switch from insects to birds was later 92
confirmed by stable isotopes analysis. The signature of the bat´s blood isotopic composition 93
peaked during Spring and Autumn, coinciding with the periods in which feathers appear in 94
the droppings and matching the birds migration patterns (Popa-Lisseanu et al., 2007). 95
Nevertheless, species composition of the avian diet of this bat remains still unknown since 96
5
almost all the remains consist on little fragments of feathers (Dondini & Vergari, 2000) that 97
do not allow further morphological identification 98
Our objectives in this research are to 1) test if different conservation methods of faecal 99
samples and/or storage time affect DNA preservation, 2) develop an efficient protocol for 100
DNA extraction from bat faeces and 3) search for diagnostic mtDNA fragments that 101
amplifying only from birds’ DNA against other DNAs present in the bat scats, allow birds’ 102
identification at species level. 103
104
Material and Methods 105
106
Collection, storage and preservation of faeces 107
For this study, we selected faecal pellets that showed presence of feathers from a total 108
of 47 giant noctule bats netted either over water courses in three points across Spain (La 109
Rioja, Malaga and Jaen), or captured as they returned to their tree roosts in a city-park in 110
Seville. Bats were individually kept in clean cloth bags and thereafter sexed, weighed, 111
measured, ringed and released at the site of capture. A total of 23 faecal pellets were 112
preserved by drying in paper bags at room temperature (11 collected in 2000 and 12 in 2005), 113
whereas another batch of 24 faecal pellets was preserved by freezing at -20ºC (13 collected in 114
2006 and 11 in 2009). All samples were processed in 2012. 115
116
DNA extraction 117
Total DNA extraction was performed in a sterile DNA laboratory. DNA was extracted 118
from about 0.01 – 0.05 g of faecal pellet unit using two different protocols: a) the Zeale et al. 119
(2011)’s method (based on a modified DNA stool Mini Kit of Qiagen) and starting by 120
6
washing the pellet, and b) the Rohland & Hofreiter (2007)’s silica method with a guanidinium 121
thiocianate (GuSCN), modified by first freezing the samples with liquid nitrogen and then 122
pulverizing them using 2 2.5 mm Ø steel balls. Both extraction methods were used in most of 123
scats (Table 2). DNA extracts were recovered with milli-Q water to a total volume of 60 µl. 124
Blanks without samples (one per extraction) were systematically included to detect possible 125
contaminations. 126
127
Primer design and testing 128
The resolution in bird identification of previously published general primers was tested 129
against newly designed ones both for fragments of cyt b and COI. We used the vertebrate 130
generalist primer pairs M13BC-FW/BCV-RV1 for a first PCR amplification, and M13/BCV-131
RV2 for the products re-amplification (Alcaide et al., 2009) of a COI’s final fragment of 758 132
bp. In addition, the bird-specific primer pair Bird-F1/COIbirdR2, used previously for bar-133
coding a bird community (Kerr et al., 2009), was used to amplify a COI 648 bp fragment. The 134
diagnostic resolution of these generalist and relatively long fragments were tested against 135
specific and short fragments of cyt b and COI obtained using newly designed primers. These 136
primers were designed from a selection of 81 homologous cyt b and 77 COI sequences from 137
86 birds downloaded from GenBank (http://www.ncbi.nlm.nih.gov/genbank) and BOLD 138
Systems (www.barcodinglife.org) and aligned with SEQUENCHER v. 4.9 (Gene Codes 139
Corporation Ann Arbor, MI USA). The selection was based on a list of potential bird preys of 140
the giant noctule in the Iberian Peninsula according to their weight (< 80 g) and wingspan (< 141
33 cm) and following Del Hoyo et al. (2004-2011). The final list of potential prey as 142
follows:7 Alaudidae, 3 Hirundinidae, 7 Motacillidae, 1 Troglodytidae, 9 Turdidae, 2 143
Muscicapidae, 2 Regulidae, 1 Cisticolidae, 23 Silviidae, 1 Paradoxornithidae, 5 Paridae, 1 144
7
Remizidae, 1 Aegithalidae, 1 Sittidae, 1 Certhiidae, 1 Laniidae, 4 Passeridae, 10 Fringillidae 145
and 6 Emberizidae). Numbers of GenBank accession of all the sequences used are available in 146
Supplementary Material 1. 147
Primers were designed using the software Primer3 v. 2.3.2. (Koressaar & Remm, 2007) in 148
conserved regions and selected avoiding primer-dimer formation, self-complementarity, too 149
low melting temperature (Tm) and incorrect internal stability profile (Rychlik, 1995). 150
The sequences obtained with the new primers were first aligned with the programme 151
SEQUENCHER v. 4.9 and then inspected using PAUP 4.0.b10 (Swofford, 2001) to assure 152
that they represented unique haplotypes and that we were able to correctly distinguish each of 153
the species within the set of birds. 154
155
PCR amplification and sequencing 156
The 47 DNA extracts were amplified in a 28 µl total volume and a mixed composition 157
of: 3 µl 10 X NH4 buffer, 0.8 µl 50 mM MgCl2, 5 µg bovine serum albumin (BSA), 1.2 µg 158
dimethyl sulfoxide (DMSO), 0.2 mM dNTP mix, 0.75 µl 10 mM forward and reverse primer 159
and 0.16 µl 5 u/µl BIOTAQ DNA polymerase. Purified water was added to complete the 28 160
µl volume and 3 µl of DNA extract were added up to a final volume of 31 µl. 161
The PCR thermocycling included a 3 min initial denaturation step at 94 ºC followed by 35 162
cycles of 1 min at 94 ºC, 1 min at annealing temperatures between 45-55 ºC (depending on 163
primers), 1 min at 72 ºC followed by a final extension of 5 min. at 72 ºC. PCR products were 164
run and visualized on a 1% agarose ⁄ TBE gels stained with SYBR Safe, and purified by Big 165
Dye. Products were sequenced on an ABI 377 automated sequencer (PE Biosystems, 166
Warrington, UK) following the manufacturer´s protocols. 167
168
8
Data analysis 169
In this study we investigated the effects on the amplification success of prey DNA in bat 170
scats of the time elapsed between faeces collection and DNA extraction (3 to 12 years), 171
conservation method (dried versus frozen) and DNA extraction method (two protocols). 172
Additionally, we checked the efficiency in the specific amplification of bird preys DNA of the 173
different pairs of primers targeting two mtDNA genes (cyt b and COI). To test the 174
significance of all these effects on DNA amplification, we performed Generalized Linear 175
Models (GLM) using the package “MASS” of the program R (Venables & Ripley, 2002). 176
Model construction followed a forward stepwise procedure and the selection was based on the 177
percentage of deviance explained and the Akaike’s Information Criterion (AIC), where 178
smaller AIC values suggesting a better fit to data. The analysis was performed in two steps: we 179
first tested for a determining influence on amplification by any of the selected factors using a 180
binomial GLM with all treatments and ‘DNA amplification’ as dependent variable. Then, we 181
tested for factors that influence the success in birds’ DNA amplification considering only 182
those treatments that yielded positive amplifications and using this time a negative binomial 183
GLM according to the distribution of the data. Amplifications were considered as ‘positive’ 184
only when the resulting sequences were recognized as being from a bird. For this second 185
design, amplifications of bats’ DNA, of DNA from other organisms and ambiguous sequences 186
were considered ‘negative’ as the null amplifications since the possible prey –in case of being 187
amplified- couldn’t be identified unambiguously. 188
189
Results 190
191
9
We selected 2 primer pairs for the cyt b and one primer pair for COI that amplified 192
fragments of 160 or 380 bp (Table 1). In the first analysis of DNA amplification success, most 193
of the variation (66.56%) was explained by a model that included all the factors (storing time, 194
conservation method, extraction protocol and pair of primers), although any of them had 195
significant influence by itself. For the second analysis, based on positive amplifications only, 196
the model with lower AIC and higher explanation of the variation (45.02%) included as 197
significant factors ‘conservation method’, ‘extraction protocol’ and their interaction, and 198
‘primers selection’. The factor ‘storing time’ was the one explaining less percentage of 199
variation (less than 1%) and it was excluded of the model. The results showed statistically 200
significant differences between extraction methods depending of the conservation method 201
utilized. Thus, the modified silica extraction with pulverization was statistically significant 202
when the faeces had been dried, but not when they had been frozen (Z = -1.303, p = 0.192). 203
On the other hand, the amplification rate of our designed primers was always higher than the 204
amplification rate of vertebrate and bird generalist primers independent of the extraction 205
method whereas there were no statistically significant differences in amplification success 206
between the newly designed primers (Table 3, Figure 2). 207
All bird DNA fragments amplified were reliably identified to species level, indicating 208
that the sequences had a minimum of 4 bp differences between closely related species in the 209
case of the cyt b fragments (2.5% in CytbSPreyFW/RW and1.1% in CytbLPreyFW/RW, n=81 210
sequences), and 9 bp in the case of the COI fragments (1.2% in BC-FW-13/BCV-RV1, 1.4% 211
in Bird-F1/COIbirdR2 and 2.4% in COIPrey-FW/RW, n=77 sequences) according to the 212
absolute distance matrix obtained with PAUP for each primer set. These results indicated that 213
there is no risk of misidentifying species when we compare the sequences in GenBank. 214
215
10
Discussion 216
217
Non-invasive techniques based on the amplification of genetic markers have proven to 218
be very effective in diet studies, offering significantly greater resolution in prey identification 219
than most of the traditional techniques (Symondson, 2002; King et al., 2008; Parsons et al., 220
2005; Deagle et al., 2007; Dunshea, 2009; Zeale et al., 2011). Together with the DNA of the 221
consumed prey, faeces contain generally a mixture of DNAs originated from epithelial cells 222
that cover the intestinal walls of the predator, and from bacteria, fungi and parasites among 223
other sources. Nevertheless, the existence of multiple templates doesn’t affect to the sequence 224
determination since chimeras and contaminants were easily identified in the eye-inspection or 225
in the BLAST. In this study, bird DNA found in the excrements of noctule bats seems to be 226
better preserved within the feather’s rachis, which serves as a protective structure. This DNA 227
is less susceptible to the effects of gastric acids, degradation by external agents and the effect 228
of organisms (e.g. Haag et al., 2009; Farrell et al., 2000; Murphy et al., 2007; Brinkman et 229
al., 2010). Therefore, the amplification of relatively larger fragments (160-380 bp) than those 230
usually analyzed (100-250 bp) in diet studies (Pompanon et al., 2011) is more likely. In our 231
analyses, the GuSCN-based extraction method with previous pulverization has proven to be 232
more effective than the washing Qiagen method on dry-preserved scats. This result is 233
probably due to the fact that when pulverizing the faecal pellet, the within-rachis DNA is also 234
extracted and made available. Besides, it is expected that this DNA will be in better 235
conditions than other DNAs present in the faeces, and which amplification is highly 236
dependent on preservation and extraction methods (Frantzen et al., 1998; Piggott & Taylor, 237
2003; Prugh et al., 2005). In the case of the modified Mini Kit of Qiagen without 238
pulverization, only the DNA that remains more exposed is extracted and made available, 239
11
whereas the DNA inside the rachis is not reached. This fact would explain why the 240
conservation method has a significant effect on the amplification success only in combination 241
with this extraction method. Thus, DNA amplification from material preserved frozen (and 242
not pulverized) yields better results -most probably due to less DNA degradation - than the 243
amplification from dried-preserved material. Finally, storing time does not significantly 244
influence DNA amplification success, probably because the period of time elapsed since the 245
collection of the samples and their study was not long enough for significant differences in 246
DNA’s breakdown and despite possible limitations in the sampling. 247
The mtDNA markers cyt b and COI have been used extensively by mammalogists and 248
ornithologists for species identification. At the beginning of this study, databases for cyt b 249
were more extensive for birds than those for COI although the later has been favored by 250
ornithologists due to the development of the DNA barcode and standardized reference 251
libraries. In our study both cyt b and COI showed similar resolution. Vertebrate and avian 252
generalist primers were only relatively efficient (25-38%) in the detection of avian DNA 253
within the bat’s faecal pellets. This was most probably due to their lower specificity and the 254
relatively large size (for degraded DNA) of the fragments that they amplified. It is well 255
known that smaller fragments survive better to digestion and are more detectable than larger 256
ones (eg. Sheppard & Harwood, 2005). Using the modified silica method extraction with 257
pulverization and the designed primers, the success percentage was between 81-89% and 258
none of the primers amplified bat’s DNA. These amplification success values are quite 259
satisfactory in relation to other studies. In fact, the percentage of predator DNA amplified 260
from fresh scat samples of captive jaguars was 87% (Haag et al., 2009), 76.8% from dried 261
scats of Andean cats (Napolitano et al., 2008), and 88% from scats of pine marten and stone 262
marten stored in ethanol or frozen (Ruiz-González et al., 2008). 263
12
The three fragments obtained with our designed primers showed similar efficiency and 264
they can be used together for prey identification since the combination of two primer pairs 265
increases the amplification success rate up to 100%, although the pair CytbSPreyFW/RW 266
showed a wider range of amplification success percentages. We recommend combining the 267
COI primer pair (COIPrey-FW/RW) with CytbLPreyFW/RW to ensure that at least one 268
sequence can be found in the DNA databases. The bird species selection used to study 269
specificity is highly representative of the potential preys availability to the bats and since 270
there were no ambiguous haplotypes in the chosen fragments for any of the potential preys, 271
species identification will be unequivocal at least within the Iberian Peninsula. 272
Due to their specificity and high amplification rate, it will be highly useful to check 273
these new primers on a wider variety of test cases to establish their possible broader 274
applicability for studying the small bird component in the diet of other predators in the 275
Palaeactic, for example, of the other known bird-eating bats such as the great evening bat Ia io 276
(Thabah et al., 2007), or the bird-like noctule Nyctalus aviator (Fukui et al., 2013). This 277
primer combination should work even for raptors such as the Eleonora´s falcon (Falco 278
eleonorae) and the sooty falcon (Falco concolor), which are the only two diurnal predators 279
that show a similar predatory strategy than the giant noctule bat in the Mediterranean region 280
(Ibáñez et al., 2001). So far, bird identification in the diet studies of these raptors is based on 281
the study of remains found near the nests, and this is not always possible (e.g. De León et al., 282
2007). Our primers would help identify unambiguously the species to which all the remains 283
belong to. 284
285
In summary, the modified silica method with GuSCN binding solution (Rohland & 286
Hofreiter, 2007), in which samples are frozen by liquid nitrogen and then pulverized, turned 287
13
out to be more effective than DNA stool Mini Kit (Qiagen) when the faeces are not fresh, due 288
to the fact that the DNA is extracted from feather remains protected of degradation inside the 289
rachis. When using the modified silica with GuSCN extraction protocol, the conservation 290
method used to store faecal samples as well as storing time do not affect amplification success 291
due to the pulverization of the sample. Our designed primers allow amplify satisfactorily 292
mtDNA fragments of birds consumed by the giant noctule bat and the identification of the 293
bird species for all the samples. These primers will allow us to study in detail the avian diet 294
composition of the poorly known giant noctule bat, unique in its partial carnivore diet in the 295
entire European continent, and could be used in the diet analysis of other bird predators. 296
297
Acknowledgements 298
299
Logistical support was provided by the ‘Laboratorio de Ecología Molecular’, at the 300
Estación Biológica de Doñana, CSIC (LEM-EBD). The EBD receives financial support from 301
the Spanish Ministry of Economy and Competitiveness, through the Severo Ochoa Program 302
for Centres of Excellence in R+D+I (SEV-2012-0262). This study was part of the MICINN 303
project CGL2009-12393 coordinated by J. Aihartza, UPV/EHU. We are grateful to Kent 304
Rylander for his help in the English usage. DPB acknowledges the financial support of the 305
Consejo Superior de Investigaciones Científicas (CSIC) through a JAE-PreDoc fellowship. 306
307
308
309
14
References 310
311
AGOSTA, S.J. 2002. Habitat use, diet and roost selection by the big brown bat (Eptesicus 312
fuscus) in North America: a case for conserving an abundant species. Mammal Review, 313
32: 179–198. 314
AGUSTÍ, N., S.P. SHAYLER, J.D. HARWOOD, I.P. VAUGHAN, K.D. SUNDERLAND, andW.O.C. 315
SYMONDSON. 2003. Collembola as alternative prey sustaining spiders in arable 316
ecosystems: prey detection within predators using molecular markers. Molecular 317
Ecology, 12: 3467-3475. 318
ALBERDI, A., I. GARIN, O. AIZPURUA, and J. AIHARTZA. 2012. The foraging ecology of the 319
mountain long-eared bat (Plecotus macrobullaris) revealed with DNA mini-barcodes. 320
PLoS ONE, 7: e35692. 321
ALCAIDE, M., C. RICO, S. RUIZ, R. SORIGUER, J. MUÑOZ, and J. FIGUEROLA. 2009. 322
Disentangling vector-borne transmission networks: a universal DNA barcoding method 323
to identify vertebrate hosts from arthropod bloodmeals. PLoS ONE, 4: e7092. 324
ALTRINGHAM, J.D. 2011. Bats. From evolution to conservation. (Second ed.) Oxford 325
University Press. 326
BRINKMAN, T.J., M.K. SCHWARTZ, D.K. PERSON, K.L. PILGRIM, and K.J. HUNDERTMARK. 327
2010. Effects of time and rainfall on PCR success using DNA extracted from deer fecal 328
pellets. Conservation Genetics, 11: 1547-1552. 329
BROWN, W.M., M. GEORGE, and A.C. WILSON. 1979. Rapid evolution of animal 330
mitochondrial DNA. PNAS, 76: 1967-1971. 331
15
CLARE, E.L., E.E. FRASER, H.E. BRAID, M.B. FENTON, and P.D. HEBERT. 2009. Species on the 332
menu of a generalist predator, the eastern red bat (Lasiurus borealis): using a molecular 333
approach to detect arthropod prey. Molecular Ecology, 18: 2532-2542. 334
DEAGLE, B.E., N.J. GALES, K. EVANS, S.N. JARMAN, S. ROBINSON, R. TREBILCO, and M.A. 335
HINDELL. 2007. Studying seabird diet through genetic analysis of faeces: a case study on 336
Macaroni penguins (Eudyptes chrysolophus). PLoS ONE, 2: e831. 337
DE LEÓN, L., B. RODRÍGUEZ, and A. MARTÍN. 2007. Status, distribution and diet of Eleonora´s 338
falcon (Falco eleonorae) in the Canary Islands. Journal of Raptor Research, 41: 331-336. 339
DEL HOYO, J., A. ELLIOT, and D.A. CHRISTIE. 2004-2011. Handbook of the Birds of the 340
World, vols. 9-16, Lynx Edicions, Barcelona. 341
DONDINI, G., and S. VERGARI. 2000. Carnivory in the greater noctule bat (Nyctalus 342
lasiopterus) in Italy. Journal of Zoology, 251: 233-236. 343
FARRELL, L.E., J. ROMAN, and M.E. SUNQUIST. 2000. Dietary separation of sympatric 344
carnivores identified by molecular analysis of scats. Molecular Ecology, 9: 1583-1590. 345
FRANTZEN, M.A.J., J.B. SILK, J.W.H. FERGUSON, R.K. WAYNE, and M.H. KOHN. 1998. 346
Empirical evaluation of preservation methods for faecal DNA. Molecular Ecology, 7: 347
1423-1428. 348
FREEMAN, P.W. 2000. Macroevolution in Microchiroptera: recoupling morphology and 349
ecology with phylogeny. Evolutionary Ecology Research, 2: 317–335. 350
FUKUI, D., H. DEWA, S. KATSUTA, and A. SATO. 2013. Bird predation by the birdlike noctule 351
in Japan. Journal of Mammalogy, 94: 657-661. 352
GRIFFITHS, D. 1975. Prey availability and the food of predators. Ecology, 56: 1209-1214. 353
HAAG, T., A.S. SANTOS, C. DE ANGELO, A.C. SRBEK-ARAUJO, D.A. SANA, R.G. MORATO, 354
F.M. SALZANO, and E. EIZIRIK. 2009. Development and testing of an optimized method 355
16
for DNA-based identification of jaguar (Panthera onca) and puma (Puma concolor) fecal 356
samples for use in ecological and genetics studies. Genetica, 136: 505-512. 357
IBÁÑEZ, C., J. JUSTE, J.L. GARCÍA-MUDARRA, and P.T. AGIRRE-MENDI. 2001. Bat predation on 358
nocturnally migrating birds. PNAS, 98: 9700-9702. 359
KERR, K.C.R., D.A. LIJTMAER, A.S. BARREIRA, P.D. HEBERT, and P.L. TUBARO. 2009. Probing 360
evolutionary patterns in Neotropical birds through DNA barcodes. PLoS ONE, 4: e4379. 361
KING, R., D. READ, M. TRAUGOTT, and W. SYMONDSON. 2008. Molecular analysis of 362
predation: a review of best practice for DNA-based approaches. Molecular Ecology, 17: 363
947–963. 364
LIMA, S.L. 1998. Nonlethal effects in the ecology of predator-prey interactions. What are the 365
ecological effects of anti-predator decision-making? BioScience, 48: 25-34. 366
MONTOYA, J.M., and R.V. SOLÉ. 2002. Small world patterns in food webs. Journal of 367
Theoretical Biology, 2147: 405-412. 368
MURPHY, M.A., K.C. KENDALL, A. ROBINSON, and L.P. WAITS. 2007. The impact of time and 369
field conditions of brown bear (Ursus arctos) fecal DNA amplification. Conservation 370
Genetics, 8: 1219-1224. 371
NAPOLITANO, C., M. BENNETT, W.E. JOHNSON, S.J. O´BRIEN, P.A. MARQUET, I. BARRÍA, E. 372
POULIN, and A. IRIARTE. 2008. Ecological and biogeographical inferences on two 373
sympatric and enigmatic Andean cat species using genetic identification of faecal 374
samples. Molecular Ecology, 17: 678-690. 375
PARSONS, K.M., S.B. PIERTNEY, S.J. MIDDLEMAS, P.S. HAMMOND, and J.D. ARMSTRONG. 376
2005. DNA-based identification of salmonid prey species in seal faeces. Journal of 377
Zoology, 266: 275-281. 378
17
PIGGOTT, M.P., and A.C. TAYLOR. 2003. Extensive evaluation of fecal preservation and DNA 379
extraction methods in Australian native and introduced species. Australian Journal of 380
Zoology, 51: 341-355. 381
POPA-LISSEANU, A.G., A. DELGADO-HUERTAS, M.G. FORERO, A. RODRÍGUEZ, R. ARLETTAZ, 382
and C. IBÁÑEZ. 2007. Bats´conquest of a formidable foraging niche: the myriads of 383
nocturnally migrating songbirds. PLoS ONE, 2: e205. 384
PRUGH, L.R., C.E. RITLAND, S.M. ARTHUR, and C.J. KREBS. 2005. Monitoring coyote 385
population dynamics by genotyping faeces. Molecular Ecology, 14: 1585-1596. 386
RATNASINGHAM, S., and P.D.N HEBERT. 2007. BARCODING. BOLD: The Barcode of Life 387
Data System (www.barcodinglife.org). Molecular Ecology Notes, 7: 355-364. 388
ROHLAND, N., and M. HOFREITER. 2007. Comparison and optimization of ancient DNA 389
extraction. BioTechniques, 42: 343-352. 390
RUIZ-GONZÁLEZ, A., J. RUBINES, O. BERDIÓN, and B.J. GÓMEZ-MOLINER. 2008. A non-391
invasive genetic method to identify the sympatric mustelids pine marten (Martes martes) 392
and stone marten (Martes foina): preliminary distribution survey on the northern Iberian 393
Peninsula. European Journal of Wildlife Research, 54: 253-261. 394
RYCHLIK, W. 1995. Selection of primers for Polymerase Chain Reaction. Molecular 395
Biotechnology, 3: 129-134. 396
SHEPPARD, S.K., and J.D. HARWOOD. 2005. Advances in molecular ecology: tracking trophic 397
links through predator-prey food webs. Functional Ecology, 19: 751-762. 398
SWOFFORD, D.L. 2001. PAUP* Phylogenetic Analysis Using Parsimony (and other methods) 399
Version 4.0b10. Sinauer Associates, Sunderland, MA, USA. 400
SYMONDSON, W.O.C. 2002. Molecular identification of prey in predator diets. Molecular 401
Ecology, 11: 627–641. 402
18
THABAH, A., G. LI, Y. WANG, B. LIANG, K. HU, S. ZHANG, and G. JONES. 2007. Diet, 403
echolocation, calls and phylogenetic affinities of the great evening bat (Ia io; 404
Vespertilionidae): another carnivorous bat. Journal of Mammalogy, 88: 728-735. 405
TOBE, S.S., A.C. KITCHENER, and A.M.T. LINACRE. 2010. Reconstructing mammalian 406
phylogenies: a detailed comparison of the cytochrome b and cytochrome oxidase subunit 407
I mitochondrial genes. PLoS ONE, 5: e14156. 408
WHITAKER, J.O., G.F. MCCRAKEN, and B.M SIEMERS. 2009. Food habits analysis of 409
insectivorous bats. In: Ecological and behavioral methods for the study of bats (eds. Kunz 410
TH, Parsons S), Second ed., The Johns Hopkins University Press, Baltimore, pp. 567-411
592. 412
ZEALE, M.R.K., R.K BUTLIN, G.L. BARKER, and D.C. LEES. 2011. Taxon-specific PCR for 413
DNA barcoding arthropod prey in bat faeces. Molecular Ecology Resources, 11: 236-244. 414
415
416
19
Figure Legends 417
418
Figure 1. Range, box-plots with 95% confidence intervals and mean values of the rate of 419
positive amplification success for the two extraction methods used. 420
421
Figure 2. Range, box-plots with 95% confidence intervals and mean values of the rate of 422
positive amplification success for the primers pairs used. 423
424
425
426
Supporting Information 427
428
Supplementary Material 1. List of species, numbers of GenBank accession of the sequences 429
used and mitochondrial region to which they belong. 430
431
432
433
434
435
436
437
438
439
20
Table 1. Description and amplification conditions of the primer pairs designed. Forward 440
(FW) and Reverse (RW) primers are signalized following primer names. 441
442
443
Target gene Name Sequence Fragment size (bp)
Annealing temperature (ºC)
Cyt b CytbSPrey (FW) AAT GGG ATT TTG TCG CAG TC 160 52 CytbSPrey (RW) TCT CAG CCA TCC CCT ACA TC
Cyt b CytbLPrey (FW) GAY AAA ATY CCM TTY CAC C 380 47 CytbLPrey (RW) TGT TCD ACD GGY TGG CT
COI COIPrey (FW) CGA GCA GAR CTA GGC CAA CC 380 55 COIprey (RW) GCA GGC GGT TTT ATG TTG ATT GCT G
444
445
446
447
448
Table 2. Amplification success for each primer pair broken-down into extraction method 449
(with and without pulverization), conservation, and collection time. Positives: number of 450
faeces in which a bird sequence is amplified. Negatives: number of faeces in which there is no 451
bird DNA amplification (divided by samples without amplification and samples with 452
ambiguous identification). 453
454
455
456
457
458
21
Without pulverization Whit pulverization
CytbSPreyFW/RW
CytbSPreyFW/RW
Method of conservation
Positives
Negatives Positives Negatives
n Amplifications No amplify Amplify others
n Amplifications No amplify Amplify others
Dry 2000 13 0 (0.0%) 13 0 11 9 (81.8%) 2 0
Dry 2005 12 4 (33.3%) 8 0 12 11 (91.7%) 1 0
Frozen 2006 8 0 (0.0%) 8 0 13 10 (76.9%) 3 0
Frozen 2009 9 1 (11.1%) 8 0 11 8 /72.7%) 3 0
TOTAL 42 5 (11.9%) 37 0 47 38 (80.6%) 9 0
CytBbLPreyFW/RW CytBbLPreyFW/RW
Positives Negatives
Positives Negatives
n Amplifications
No amplify Amplify others
n Amplifications
No amplify Amplify others
Dry 2000 13 0 (0.0%) 13 0 11 11 (100%) 0 0
Dry 2005 12 1 (8.3%) 11 0 12 9 (75%) 3 0
Frozen 2006 8 5 (62.5%) 3 0 13 12 (92.3%) 1 0
Frozen 2009 9 5 (55.6%) 4 0 11 10 (90.9%) 1 0
TOTAL 42 11 (26.2%) 31 0 47 42 (89.4%) 5 0
COIPreyFW/RW COIPreyFW/RW
Positives Negatives Positives Negatives
n Amplifications
No amplify Amplify others
n Amplifications
No amplify Amplify others
Dry 2000 13 6 (46.2%) 7 0 11 11 (100%) 0 0
Dry 2005 12 10 (83.3%) 2 0 12 10 (83.3%) 2 0
Frozen 2006 8 5 (62.5%) 3 0 13 11 (84.6%) 2 0
Frozen 2009 9 7 (77.8%) 2 0 11 8 (72.7%) 3 0
TOTAL 42 28 (66.7%) 14 0 47 40 (85.1%) 7 0
Bird-F1/COIbirdR2 Bird-F1/COIbirdR2
Positives Negatives Positives Negatives
n Amplifications
No amplify Amplify others
n Amplifications
No amplify Amplify others
Dry 2000 13 0 (0.0%) 13 0 11 3 (27.3%) 5 3
Dry 2005 12 0 (0.0%) 11 1 12 7 (58.3%) 4 1
Frozen 2006 8 5 (62.5%) 3 0 13 6 (46.2%) 6 1
Frozen 2009 9 3 (33.3%) 6 0 11 2 (18.2%) 5 4
TOTAL 42 8 (19.0%) 33 1 47 18 (38.3%) 20 9
BCFW-M13/BCRV1
BCFW-M13/BCRV1
Positives Negatives
Positives
Negatives
n Amplifications
No amplify Amplify others
n Amplifications
No amplify Amplify others
Dry 2000 13 0 (0.0%) 7 6 11 5 (45.5%) 4 2
Dry 2005 12 0 (0.0%) 6 6 12 4 (33.3%) 2 6
Frozen 2006 8 2 (25.0%) 1 5 13 2 (15.4%) 7 4
Frozen 2009 9 0 (0.0%) 3 6 11 1 (9.1%) 6 4
TOTAL 42 2 (4.8%) 17 23 47 12 (25.5%) 19 16
459
22
Table 3. Estimate, standard error (SE), Z-value and P-value of the factors that can influence 460
in the amplification success. Negative results were excluded. Significance codes: 0 ‘***’ 461
0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1. 462
463
Coefficients Estimate SE Z-value P-value
Intercept 4.4512 0.2006 22.192 < 2e-16 ***
CytbSPreyFW/RW -0.1117 0.2407 -0.464 0.64274
BirdF1/COIBirdR2 -0.7458 0.2462 -3.029 0.00245 **
COIPreyFW/RW 0.2488 0.2232 1.114 0.26514
BCFW-M13/BCRV1 -0.7439 0.2601 -2.86 0.00423 **
No pulverized:Frozen -0.3397 0.2073 -1.639 0.10126
Pulverized:Frozen -0.2522 0.1936 -1.303 0.19265
No pulverized:Dry -0.8443 0.2686 -3.143 0.00167 **
464