- Email: [email protected]
The microbiome of a striped dolphin (Stenella coeruleoalba) stranded in Portugal
Accepted Manuscript The microbiome of a striped dolphin (Stenella coeruleoalba) stranded in Portugal Filipa Godoy-Vitorino, Arnold Rodriguez-Hilario, Ana Luísa Alves, Filipa Gonçalves, Beatriz Cabrera-Colon, Cristina Sousa Mesquita, Pedro Soares-Castro, Marisa Ferreira, Ana Marçalo, José Vingada, Catarina Eira, Pedro Miguel Santos PII:
S0923-2508(16)30096-1
DOI:
10.1016/j.resmic.2016.08.004
Reference:
RESMIC 3534
To appear in:
Research in Microbiology
Received Date: 5 May 2016 Revised Date:
20 July 2016
Accepted Date: 23 August 2016
Please cite this article as: F. Godoy-Vitorino, A. Rodriguez-Hilario, A.L. Alves, F. Gonçalves, B. CabreraColon, C.S. Mesquita, P. Soares-Castro, M. Ferreira, A. Marçalo, J. Vingada, C. Eira, P.M. Santos, The microbiome of a striped dolphin (Stenella coeruleoalba) stranded in Portugal, Research in Microbiologoy (2016), doi: 10.1016/j.resmic.2016.08.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
For publication
2
The microbiome of a striped dolphin (Stenella coeruleoalba) stranded in
3
Portugal
4
Filipa Godoy-Vitorinoa, Arnold Rodriguez-Hilarioa, Ana Luísa Alvesb, Filipa Gonçalvesb,
5
Beatriz Cabrera-Colona, Cristina Sousa Mesquitab, Pedro Soares-Castrob, Marisa Ferreirab,c, Ana
6
Marçalod, José Vingadac,e, Catarina Eirac,d, Pedro Miguel Santosb*
SC
RI PT
1
a
Microbial Ecology and Genomics Lab, Department of Natural Sciences, Inter American University of
7
b
Department of Biology and Centre for Molecular and Environmental Biology (CBMA), University of
9
Minho, Campus de Gualtar 4710-087 Braga, Portugal
10 11
M AN U
Puerto Rico, Metropolitan Campus, P.O. Box 191293 San Juan, Puerto Rico 00919-1293
8
c
Portuguese Wildlife Society (SPVS), Quiaios Field Station, Apartado 16 EC Quiaios, 3081-101 Figueira da Foz, Portugal
12
22 23 24 25 26 27
Department of Biology and CESAM, University of Minho, Campus de Gualtar 4710-087 Braga, Portugal
E-mail addresses: [email protected] (F. Godoy-Vitorino), [email protected] (A. Rodriguez-Hilario), [email protected] (B. Cabrera-Colon), [email protected] (A.L. Alves), [email protected] (F. Gonçalves), [email protected] (C.S. Mesquita), [email protected] (P. Soares-Castro), [email protected] (M. Ferreira), [email protected] (A. Marçalo), [email protected] (J. Vingada), [email protected] (C. Eira), [email protected]* (P.M. Santos) *Corresponding author
EP
15 16 17 18 19 20 21
e
AC C
14
Department of Biology and CESAM, University of Aveiro, 3810-193 Aveiro, Portugal
TE D
d
13
ACCEPTED MANUSCRIPT 2 28
Abstract Infectious diseases with epizootic consequences have not been fully studied in marine
30
mammals. Presently, the unprecedented depth of sequencing, made available by high-throughput
31
approaches, allows detailed comparisons of the microbiome in health and disease. This is the
32
first report of the striped dolphin microbiome in different body sites. Samples from one striped
33
female edematous dolphin were acquired from a variety of body niches, including the blowhole,
34
oral cavity, oral mucosa, tongue, stomach, intestines and genital mucosa. Detailed 16S rRNA
35
analysis of over half a million sequences identified 235 OTUs. Beta diversity analyses indicated
36
that microbial communities vary in structure and cluster by sample origin. Pathogenic, Gram-
37
negative, facultative and obligate anaerobic taxa were significantly detected, including
38
Cetobacterium, Fusobacterium and Ureaplasma. Phocoenobacter and Arcobacter dominated the
39
oral-type samples, while Cardiobacteriaceae and Vibrio were associated with the blowhole and
40
Photobacterium were abundant in the gut. We report for the first time the association of
41
Epulopiscium with a marine mammal gut.
TE D
M AN U
SC
RI PT
29
The striped dolphin microbiota shows variation in structure and diversity according to the
43
organ type. The high dominance of Gram-negative anaerobic pathogens evidences a cetacean
44
microbiome affected by human-related bacteria.
45
Keywords: Cetacean; Striped dolphin; 16S rRNA; Microbiota; Metagenomics
46
1. Introduction
AC C
47
EP
42
The integrity and ecological sustainability of marine ecosystems are under increasing
48
threats mainly due to long-term consequences of climate change, habitat degradation and human
49
impact [1,2]. Marine mammal species are top predators and have long life spans. Often, they
50
need to cope with exposure to a variable combination of environmental stressors (e.g. xenobiotic
ACCEPTED MANUSCRIPT 3 pollutants) and pathogens [3]. Therefore, marine mammals can be considered prime sentinel
52
species of aquatic ecosystem health. In fact, the low resilience of marine mammal species results
53
from their low fecundity rate and their position at the top of food webs, which makes them more
54
susceptible to many human-induced pressures [4]. In Europe, there is growing concern about the
55
current health status of marine mammal populations. Significantly, a decline in some coastal
56
species such as the harbor porpoise (Phocoena phocoena) was registered [5–7], including the
57
population of the Iberian Atlantic coast, where a new porpoise ecotype was recently proposed [8].
58
Likewise, the health status of striped dolphin (Stenella coeruleoalba) populations is of great
59
concern, particularly the Mediterranean population, considering their very high mortality rates
60
during Morbillivirus (MV) outbreaks [9–11].
M AN U
SC
RI PT
51
Information on microbiological infections in cetaceans worldwide is critical for
62
understanding epidemiology in their populations, as well as warning about ocean health changes,
63
even more so in light of additional non-infectious aggressions mainly due to human activities,
64
that already render these populations susceptible to disease. In some reports, the death of
65
stranded dolphins has been associated with viruses (in addition to MV outbreaks), bacteria, fungi
66
and parasites. In striped dolphins, these included bacteria such as Brucella affecting the central
67
nervous system [12], fungal yeast such as Cryptococcus gattii associated with pulmonary disease
68
[13] and parasites such as Toxoplasma gondii in cerebral toxoplasmosis [14]. In bottlenose
69
dolphins (Tursiops truncatus), Papillomaviruses (PVs) have been associated with malignant
70
genital tumors [15]. However, information on striped dolphin health remains scarce, and most
71
available studies are based on bottlenose dolphins, the most common cetacean species in
72
captivity around the world. For example, a study performed in a bottlenose dolphin in captivity
73
over a 7-year period documented potentially zoonotic bacterial pathogens through isolation of
AC C
EP
TE D
61
ACCEPTED MANUSCRIPT 4 74
bacteria such as Proteus mirabilis, Staphylococcus aureus and Pseudomonas aeruginosa [16].
75
Microbial isolates from free-ranging bottlenose dolphins in Florida, Texas and North Carolina
76
evidenced abundant Staphylococcus spp. Contents, suggesting direct human impact [17]. In order to treat live-stranded or captive dolphins, we need to be able to assess the
78
vulnerability of free-ranging dolphin populations and to evaluate the risk of disease transmission
79
between humans and dolphins. Also, to further assess the proposed role of dolphins as
80
environmental sentinels, we need detailed characterization of the dolphin microbiota in health
81
and disease contexts. Since ~99% of all microorganisms are recalcitrant to isolation under
82
conventional laboratory conditions [18], next-generation sequencing technologies applied to 16S
83
rRNA genes have been playing a pivotal role in
84
natural ecosystems [19]. The unprecedented depth of sequencing made available by high-
85
throughput approaches allows for finely detailed descriptions of the microbiome of marine
86
mammals, and thus provides a relevant approach to environmental surveillance of the marine
87
environment. Very recently, the microbial taxonomic composition of 48 healthy bottlenose
88
dolphins (38 in captivity and 10 wild animals) and 18 California sea lions (Zalophus
89
californianus) was reported [20]. The authors found as many as 30 phyla in the dolphin
90
specimens and
91
overall bacterial taxonomic compositions of the marine mammals were distinct from dietary fish
92
and seawater. Similar findings have been reported for terrestrial vertebrate gut microbiota [21].
SC
M AN U
microbial diversity in
EP
TE D
characterization of
significant differences between wild and captive dolphins, concluding that the
AC C
93
RI PT
77
A multidisciplinary team from Portuguese and Spanish research centers was established
94
some years ago to evaluate the health status of marine mammal populations along the Iberian
95
Atlantic coast,
96
species by collecting data regarding trait history, contaminants, bacteriological and fungal
systematically
assessing population status and health of coastal cetacean
ACCEPTED MANUSCRIPT 5 97
imbalances, prevalence of viral diseases, parasite load and tissue damage [22,23]. In the context
98
of this integrative approach, standardized sample collection procedures for evaluating the
99
taxonomic composition of the microbiome of cetaceans were implemented. Here,
first results
emerging from this collaborative effort are reported. Specifically, a 16S rRNA gene-based
101
Illumina MiSeq sequencing approach was used to investigate
102
composition and structure in an edematous female striped dolphin stranded along the Algarve
103
coast in Portugal.
104
2. Materials and methods
105
2.1. Sample collection
RI PT
100
M AN U
SC
microbial community
The striped dolphin sampled in the present study was stranded alive in the Algarve at
107
the Meia Praia Beach (37.115885ºN, -8,648677°E), and died shortly after the arrival of a rescue
108
team from the Marine Animal Stranding Network (coordinated by the Instituto para a
109
Conservação da Natureza e Florestas, http://www.icnf.pt/portal/icnf, Portugal, in cooperation
110
with the Portuguese Wildlife Society (SPVS)). The dead animal was taken to the Quiaios Field
111
Station in Figueira da Foz (Portugal), where the respective necropsy took place. The dolphin was
112
an immature female with a total length of 191 cm and 63.6 kg of weight. Post-mortem analysis
113
was performed within 12 h after the animal’s death according to standard protocols [24]. The
114
occurrence of gross pathologies was recorded if any evidence of pathology (lesions or abnormal
115
appearance) was found in any organ system [25]. The animal was emaciated and presented low
116
consistency of brain, liver and pancreas tissues, lung lesions compatible with pneumonia and
117
ulcerous formations in the stomach chambers.
AC C
EP
TE D
106
ACCEPTED MANUSCRIPT 6 During the necropsy, tissue and swab samples from different body sites (blowhole, colon,
119
intestine, stomach, tongue, oral cavity (mouth), duodenum, genitals) were collected in triplicate
120
and stored in RNA-Later and frozen within an hour after the necropsy at -80 °C.
121
2.2. Genomic DNA extraction, PCR Amplifications and Sequencing
RI PT
118
Total DNA extractions were done in triplicate (for each sample) using the PowerSoil DNA
123
isolation kit (MoBio). Briefly, 250 mg of tissue or 250 ml of total pellet recovered from swabs
124
(following extensive washing with saline solution) were added to powerbead tubes (MoBio). All
125
subsequent steps were performed according to the manufacturer’s instructions. The concentration
126
and quality of DNA was evaluated using a NanoDrop™ 1000 spectrophotometer.
M AN U
SC
122
To determine whether the extracted DNA samples included intact DNA of bacterial
128
origin, PCR amplification of the whole 16S rRNA gene was carried out using universal primers
129
27F and 1492R [26]. Due to varying PCR yields across samples, PCR reactions were pooled
130
from triplicate samples from each origin prior to sequencing.
TE D
127
The composition of microbial communities from the different striped dolphin samples
132
was assessed using primers for paired-end V3-V4 16S rRNA sequencing on the Illumina MiSeq
133
platform [27].
134
2.3. Data deposition
136 137
The sequences used in this study have been deposited with BioProject database ID
AC C
135
EP
131
PRJNA316486 in SRA with accession number SRP072736. 2.4. Sequence processing and data analyses
138
Sequencing reads were processed using QIIME [28] with strict quality and size filtering,
139
de-replication and clustering (Usearch [29]). The resulting high-quality dataset was then
140
screened for chimeric sequences and contaminant chloroplast DNA (Uchime [30]). Filtered
ACCEPTED MANUSCRIPT 7 141
sequences were used to evaluate the diversity and taxonomic composition of target samples
142
using
143
performed. Sequence binning was done using the 97% identity threshold with the Silva database
144
[31]. Data were subsampled,
145
Richness curves were built using the The chao 1 index that uses low frequency counts for
146
richness (singletons and doubletons) to estimate the number of missing species [32].
beta and alpha diversity measures and significance tests were
thus mitigating biases due to differences in sampling depth.
RI PT
QIIME 1.9.1, where all
Principal component analyses of bacterial communities from the eight body site samples
148
were compared phylogenetically using UniFrac distances [33], a β-diversity measure that uses
149
phylogenetic information to compare samples in which the distances between them represent
150
their ecological distances. The “R” Vegan package [34] was used to draw heat maps, and ggplot2
151
[35] and reshape2 [36] were used to draw dendrograms. PICRUSt (phylogenetic investigation of
152
communities by reconstruction of unobserved states) was used for functional metagenome
153
prediction, estimating gene families (KEGG KO functions) contributing to a bacterial
154
metagenome identified by 16S rRNA sequencing [37].
155
3. Results
TE D
M AN U
SC
147
In order to monitor the microbiome composition of stranded cetaceans, a standardized
157
procedure pipeline was implemented, as depicted in Fig. 1. Briefly, following cetacean stranding
158
registration, the animal is evaluated and, whenever it evidences a fresh status (time of death
159
below 24 h), samples for microbiome content evaluation are collected during necropsy.
160
Following total DNA extraction, all samples evidencing the presence of bacteria (assessed by
161
PCR amplification of the full 16S rRNA region) are then sequenced using V3-V4 16S rRNA
162
protocol on the Illumina MiSeq platform. Here, the first results of this procedure are presented.
163
AC C
EP
156
ACCEPTED MANUSCRIPT 8 164
3.1. Community diversity and structure An initial set of 923,066 raw reads underwent strict filtering, including dereplication,
166
quality and size filtering, removing reads with poor quality scores, chimeras and chloroplasts. A
167
total of 639,116 sequences with an average read length of about 460 bp passed preprocessing.
168
The number of sequences per sample varied from a minimum of 24,083 to a maximum of
169
289,253 sequences, with an average of 79,904 sequences per sample (Table 1). Differences
170
between the total number of retrieved sequences may suggest significant variations in bacterial
171
richness and abundance according to the sampled body site. Binning of the 639,116 sequences
172
resulted in 235 operational taxonomic units (OTUs) (Table 1). All samples were rarefied to an
173
even sampling depth of 24,000 sequences prior to statistical analysis. The samples with the
174
highest number of OTUs were the oral cavity, the tongue and the blowhole (Table 1). Oral cavity
175
bacteria shared more OTUs with the tongue compared to all other body sites, followed by the
176
genital sample, which also shared OTUs with both the tongue and oral cavity (Supplementary
177
Table 1S). PCoA of UniFrac distances revealed that bacterial communities from each body site
178
segregated the duodenum and stomach from all other samples (Fig. 2). The genital sample shared
179
fewer OTUs with the other samples, and there existed community similarities between tongue
180
and oral samples as well as between colon and intestine, with the blowhole having a community
181
similar to the oral and intestinal samples (Fig. 2). The tongue, oral, blowhole and genitals
182
samples showed the highest Chao1 richness values and were separate from gut samples (Fig. 2,
183
Supplementary Fig. 1S, Supplementary Fig. 2S).
AC C
EP
TE D
M AN U
SC
RI PT
165
184 185 186
3.2. Microbial taxonomic profiles A total of 15 phyla, with the 5 most abundant (>5%) being Proteobacteria, Fusobacteria,
ACCEPTED MANUSCRIPT 9 Firmicutes, Bacteroidetes and Tenericutes, were found. The remaining phyla had less than 5%
188
abundance. Proteobacteria, Fusobacteria Firmicutes and Bacteroidetes were present across all
189
samples (Fig. 3). Genitals and blowholes were the external body orifices with a dominance of
190
Fusobacteria, Bacteroidetes and Proteobacteria, with half of its community dominated by
191
Cardiobacteriales and Pasteurellales (~43%) (Fig. 3, Supplementary Fig. 3S). Oral and tongue
192
samples were dominated by Proteobacterial orders Pasteurellales and Campylobacterales (~85%).
193
The stomach sample had a dominance of Tenericutes order Mycoplasmatales (~43%) followed
194
by Firmicutes, and the duodenum had a dominance of Firmicutes order Clostridiales (~98%),
195
Intestines and colon evidenced a similar community structure dominated by Proteobacteria,
196
Fusobacteria and Firmicutes. Interestingly, the amount of Bacteroidetes decreased from open
197
orifices, through the oral cavity and gut, with virtually no such taxa in the intestine or colon; the
198
opposite occurred with the Firmicutes.
M AN U
SC
RI PT
187
A total of 108 genera were detected, most of which had residual abundance (<0.5%), and
200
each sample had a dominance of different taxa; the blowhole had ~27% of uncultured
201
Cardiobacteraceae,
202
Tenacibaculum
203
Porphyromonas (~20%), Actinobacillus (15%), Parvimonas (~7%), and Ureaplasma (5%). The
204
oral cavity had dominant Phocoenobacter (~30%) and Arcobacter (24%), but also
205
Porphyromonas (~14%) very similar to the tongue’s genus profile, with a slightly higher amount
206
of
207
Peptostreptococaceae-IncertaeSedis (~98%), representing a dominance of ~40% in the stomach,
208
followed by Ureaplasma (~36%), Mycoplasma (~7%) and Helicobacter (4%). The intestine had
209
a dominance of Photobacterium (64%), Cetobacterium (20%) and Peptostreptococaceae-
TE D
199
~18%
Helcococcus.
Pasteurellaceae,
The
genital
~9%
sample
had
Actinobacillus
and
Fusobacterium
5%
(25%),
AC C
EP
and
uncultured
Phocoenobacter
(~52%).
The
duodenum
samples
evidenced
a
dominance
of
ACCEPTED MANUSCRIPT 10 210
IncertaeSedis (14%). The colon had a dominance of Actinobacillus (~33%), Photobacterium
211
(30%), Cetobacterium (10%) and Epulopiscium (8%) (Fig. 3). Tongue and oral cavity samples
212
apparently shared Phocoenobacter,
213
(Supplementary Fig. 4S). Shared taxa between colon and intestine included Photobacterium,
214
Actinobacillus and Cetobacterium. The blowhole shared Actinobacillus with the colon and
215
intestine (Supplementary Fig. 4S).
RI PT
Arcobacter and Porphyromonas with a genital sample
Significantly different taxa (p<0.05) included Streptobacillus, Arcobacter, Vibrio and
217
Tenacibacter in the blowhole (taxa absent in all other samples). In the colon, Actinobacillus and
218
Epulopiscium were significantly more abundant while, in the intestine,
219
damselae and Cetobacterium predominated; the stomach had Helicobacter heilmanii,
220
Mycoplasma, Ureaplasma and Paeniclostridium sordellii; in the tongue, Phocoenobacter was
221
significantly more abundant (Fig. 4). Oral samples had diverse and highly abundant taxa
222
including Campylobacter, Parvimonas, Tenacibaculum, Porphyromonas, Psychrobacter and
223
Arcobacter sp., while in the duodenum, Paeniclostridium sordellii dominated and in the genitals
224
Actinobacillus scotiae and Fusobacterium were significantly more abundant (Fig. 4).
227
M AN U
TE D
EP
226
Photobacterium
3.3. Phenotypic and metabolic inference Based on inference of taxonomic-to-phenotypic mapping of metabolism using
AC C
225
SC
216
228
METAGENassist [38], ammonia oxidizing pathways in the blowhole, sulfate reducers and
229
oxidizers in the colon, chitin degradation and denitrification in the intestine, nitrogen fixation in
230
the stomach, ammonia oxidizers and dehalogenation in the oral cavity and ammonia oxidizers in
231
the genitals were found (Supplementary Fig. 5S). As a whole, we found mostly anaerobic and
232
facultative anaerobic bacteria, and a dominance of Gram-negative pathogenic bacteria in most
ACCEPTED MANUSCRIPT 11 organs except for the duodenum and intestines (Supplementary Fig. 5S). Picrust analyses for oral
234
and gut samples revealed that taxa in these samples included metabolic pathways previously
235
associated with human diseases such as cell signaling in H. pylori infections, Tuberculosis, the
236
Vibrio chloreae cycle and immunodeficiency were significantly associated with gut samples.
237
Pathways associated with pertussis and Huntington’s disease were significantly associated with
238
taxa found in the oral samples (Supplementary Fig. 6S).
239
4.
SC
Discussion
RI PT
233
Studies on the impact of microbial pathogens in dolphin strandings have relied on
241
cultivation techniques, PCR and clonal sequencing; however, very few studies have focused on
242
the inventory of bacteria in different body sites using next-generation sequencing, a thorough and
243
sophisticated approach to understanding the overall health status of the animal. Here, the first
244
microbiome inventory of a striped dolphin, a small oceanic delphinid species, is presented. One
245
of the potential bottlenecks of this approach was the high risk of sample contamination with
246
external sources during stranded animal handling up until sample collection. Nonetheless,
247
bacterial inventory of the striped dolphin specimen under analysis included a variety of bacteria
248
in a range of 15 phyla,
249
conducted on bottlenose dolphins along the US Coast [20], although with a different relative
250
abundance of the main phyla. This corroborates the validity of
251
sample collection from stranded specimens. Compared to a large marine mammal study by Bik
252
and co-workers [20], we confirm that oral samples were richer than gut samples, as were those in
253
the blowhole and genitals.
TE D
M AN U
240
AC C
EP
also reported in a more extensive microbiome composition survey
implemented procedures for
254
Most of the bacteria within the 15 phyla changed their relative abundance in each organ
255
type, but overall, we found a dominance of mostly facultative, anaerobic, Gram-negative taxa,
ACCEPTED MANUSCRIPT 12 some of which are host-adapted and
could be traced to human infections. Some of these
257
dominant bacteria included common residents of the mucous membranes of animals, such as
258
Pasteurellaceae OTUs, which can be commensals, as well as Actinobacillus and
259
Phocoenobacter, which were recently recovered from marine mammal carcasses and were
260
related to septicemia [39]. Actinobacillus were also found in high abundance in gastric rectal
261
samples from healthy dolphins, but Phocoenobacter is apparently absent in the healthy animals
262
[20]. These taxa were found in the striped dolphin in very high abundance, corresponding to
263
nearly half of the bacterial community in the blowhole and colon as well as in the tongue and
264
oral samples. We found several obligatory anaerobic bacteria such as Arcobacter, a genus
265
associated with human and animal illness [40] as well as
266
hemolytic bacteria [42,43] and Tenacibacullum (bacteria common in the GI tract and fish
267
pathogen [44]) are all obligatory anaerobic bacteria. Campylobacter of human and livestock
268
origin present in animal oral samples, are regarded as zoonotic bacteria, causing gastroenteritis
269
and septicemia and occasionally other conditions in marine mammals [45]. Cardiobacteriaceae
270
OTUs were also found in the blowhole, tongue and oral cavity of the dolphin, and these
271
anaerobic taxa are associated with bacteremia and wound infections in humans [46]. The fact
272
that these taxa are anaerobic and have entry portals mostly in the gut, and are found at a highly
273
oxygenized area of the animal such as the blowhole, indicates
274
stranded striped dolphins from Italy were diagnosed with cardiovascular pathologies, including
275
ventricular dilation associated with hypoplasia of the tricuspid chordae and valvular fibrosis,
276
ventricular hypertrophy and myocarditis; however, no pathogens were found [48]. Additionally,
277
another study reporting the first case of sepsis in a live-stranded sperm whale
278
pathogen Edwardsiella tarda, relating these bacteria to stranding and death of the animal [49].
M AN U
SC
RI PT
256
AC C
EP
TE D
marine sediments [41]; Helcococcus -
sepsis [47]. Recently, nine
found the
ACCEPTED MANUSCRIPT 13 The abundance of Cardiobacteriaceae OTUs in the dolphin may suggest a microbial role for
280
similar cardiopathologies. Other obligate Gram-negatives such as Cetobacterium were abundant
281
in the colon and intestines and had already been isolated from other marine mammals [50].
282
Pathogenic Vibrionaceae were also abundant in the colon and intestines (Photobacterium), as
283
well as Vibrio, a known cytotoxic bacteria, in the blowhole [51]. Gastric inflammation has been
284
considered a cytologic marker of systemic illness in dolphins; however, no clinically significant
285
findings of cultured bacteria and fungi have been reported, except for a case of biliary cirrhosis
286
produced by Campula spp [52]. In agreement with this, the presence of Ureaplasma in the
287
stomach of healthy asymptomatic bottlenose dolphins was previously reported [20]; thus, the
288
impact of these organisms in the health of dolphins remains unknown. Another taxon that may be
289
a biomarker for disease is, for instance, genital Fusobacterium, a Gram-negative, obligatory
290
anaerobic species that is a common resident of the GI tract, but that possesses a number of
291
pathogenic traits with relevance to gut diseases such as inflammatory bowel disease [53].
292
Interestingly, we found Epulopiscium, a large symbiotic bacteria found in the tropical surgeon
293
fish [54], probably helping fish to break down algal nutrients in the diet. This is the first report
294
of Epulopiscium in the gut of a marine mammal.
SC
M AN U
TE D
EP
295
RI PT
279
In conclusion, the main goal of this work was to reveal, with high resolution, the microbiome of a critically ill female edematous striped dolphin.
297
diverse community, mainly dominated by Gram-negative anaerobic bacteria of human origin.
298
The veterinary bibliography on marine mammal research regarding infectious diseases in
299
strandings reveals a remarkable scarcity of comprehensive literature about diseases/disorders
300
affecting different body systems of dolphins, and these data are quite limited in marine mammals
301
inhabiting Portuguese and northern Spanish waters. Data exist on the effects of accidental
AC C
296
As a result, we
report a
ACCEPTED MANUSCRIPT 14 capture [55], the prevalence of some parasite species [56,57] and pollutants [5,22,23] in marine
303
mammal populations. However, studies on infectious disease epidemiology in cetacean species
304
are lacking despite recent reports of re-emerging or new infectious diseases in marine mammal
305
populations, with epizootic and potential zoonotic consequences.
RI PT
302
Our in-depth study using next-generation sequencing technologies is the first bacterial
307
metagenomic survey of an edematous striped dolphin to show that the analyzed animal
308
harbored multiple bacterial pathogens resembling those present in humans. Whether human
309
activities
310
light on the cetacean pathogenic microbiome that critically impacted a striped dolphin on the
311
Iberian Peninsula. Detailed epidemiological models using next-generation sequencing and
312
metagenomics as tools to predict new outbreaks and facilitate their control are warranted.
SC
306
M AN U
actually contribute to such a burden remains to be clarified. This study sheds new
313
315
Acknowledgements
TE D
314
The authors would like to acknowledge Jörg Becker and João Sobral for
sequencing
services provided at the Instituto Gulbenkian de Ciência and to Ricardo Araújo for preliminary
317
DNA extraction tests.
EP
316
This work was supported by project CetSenti RECI/AAG-GLO/0470/2012 (FCOMP-01-
319
0124-FEDER-027472), FCT/MCTES (PIDDAC) and FEDER-COMPETE (POFC), the strategic
320
program
321
through the FCT I.P., by the ERDF through the COMPETE2020 - Programa Operacional
322
Competitividade e Internacionalização (POCI) and through PhD grants (SFRH/BD/76894/2011,
323
SFRH/BD/98558/2013) attributed to P.S-C and C.S.M, respectively. The work was also
324
supported by the FCT through CESAM UID/AMB/50017/2013 co-funded by the FCT/MEC and
AC C
318
UID/BIA/04050/2013 (POCI-01-0145-FEDER-007569) funded by national funds
ACCEPTED MANUSCRIPT 15 325
FEDER, within PT2020 and Compete 2020 (C.E.) and grants SFRH/BD/30240/2006 (M.F.) and
326
SFRH/BPD/64889/2009 (A.M.).
327
The authors declare no conflict of interest.
329 330
References
331
[1]
SC
Dayton PK. The importance of the natural sciences to conservation: (an American Society of Naturalists symposium paper). Am Nat 2003;162:1–13.
332 333
RI PT
Conflicts of interest
[2]
Marques M, da Costa MF, Mayorga MI de O, Pinheiro PR. Water environments:
M AN U
328
334
anthropogenic pressures and ecosystem changes in the Atlantic drainage basins of Brazil.
335
Ambio 2004;33:68–77. [3]
2011;48:676–690.
337 338
Bossart GD. Marine mammals as sentinel species for oceans and human health. Vet Pathol
[4]
TE D
336
Peltier H, Baagøe HJ, Camphuysen KCJ, Czeck R, Dabin W, Daniel P, et al. The stranding anomaly as population indicator: the case of harbour porpoise Phocoena
340
phocoena in North-Western Europe. PLoS One 2013;8:e62180. [5]
and ecological factors related to trace element levels in harbour porpoises (Phocoena
342
phocoena) from European waters. Mar Environ Res 2007;64:247–266.
343 344
Lahaye V, Bustamante P, Law RJ, Learmonth JA, Santos MB, Boon JP, et al. Biological
AC C
341
EP
339
[6]
Mahfouz C, Henry F, Courcot L, Pezeril S, Bouveroux T, Dabin W, et al. Harbour
345
porpoises (Phocoena phocoena) stranded along the southern North Sea: an assessment
346
through metallic contamination. Environ Res 2014;133:266–273.
347
[7]
Siebert U, Wünschmann A, Weiss R, Frank H, Benke H, Frese K. Post-mortem findings in
ACCEPTED MANUSCRIPT 16 348
harbour porpoises (Phocoena phocoena) from the German North and Baltic Seas. J Comp
349
Pathol 2001;124:102–114.
350
[8]
Fontaine MC, Roland K, Calves I, Austerlitz F, Palstra FP, Tolley KA, et al. Postglacial climate changes and rise of three ecotypes of harbour porpoises, Phocoena phocoena, in
352
western Palearctic waters. Mol Ecol 2014;23:3306–3321.
353
[9]
RI PT
351
Van Bressem M-F, Duignan PJ, Banyard A, Barbieri M, Colegrove KM, De Guise S, et al. Cetacean morbillivirus: current knowledge and future directions. Viruses 2014;6:5145–
355
181. [10]
1993;22:524–528.
357 358
Aguilar A, Raga JA. The Striped Dolphin Epizootic in the Mediterranean Sea. Ambio
M AN U
356
SC
354
[11]
Raga J-A, Banyard A, Domingo M, Corteyn M, Van Bressem M-F, Fernández M, et al. Dolphin morbillivirus epizootic resurgence, Mediterranean Sea. Emerg Infect Dis
360
2008;14:471–473.
361
[12]
TE D
359
González-Barrientos R, Morales J-A, Hernández-Mora G, Barquero-Calvo E, GuzmánVerri C, Chaves-Olarte E, et al. Pathology of striped dolphins (Stenella coeruleoalba)
363
infected with Brucella ceti. J Comp Pathol 2010;142:347–52. [13]
coeruleoalba). J Wildl Dis 1985;21:443–446.
365 366 367 368
Gales N, Wallace G, Dickson J. Pulmonary cryptococcosis in a striped dolphin (Stenella
[14]
AC C
364
EP
362
Di Guardo G, Proietto U, Di Francesco CE, Marsilio F, Zaccaroni A, Scaravelli D, et al. Cerebral toxoplasmosis in striped dolphins (Stenella coeruleoalba) stranded along the Ligurian Sea coast of Italy. Vet Pathol 2010;47:245–253.
369
[15] Rehtanz M, Bossart GD, Doescher B, Rector A, Van Ranst M, Fair PA, et al. Bottlenose
370
dolphin (Tursiops truncatus) papillomaviruses: vaccine antigen candidates and screening
ACCEPTED MANUSCRIPT 17 test development. Vet Microbiol 2009;133:43–53.
371 372
[16] Chan OS, Mukherjee J, Yuen CS. Microbial flora of blowhole samples in captive bottlenose dolphins (Tursiops truncatus aduncas-type) in Hong Kong, 1993– 2000. Proc.
374
32nd Annu. Conf. Int. Assoc. Aquat. Anim. Med., Tampa, FL: 2001, p. 24–25.
375
[17]
RI PT
373
Buck JD, Wells RS, Rhinehart HL, Hansen LJ. Aerobic micro-organisms associated with freeranging dolphins in coastal Gulf of Mexico and Atlantic Ocean waters. J Wildl Dis
377
2006;42:536–544. [18]
culture in a simulated natural environment. Science 2002;296:1127–1129.
379 380
Kaeberlein T, Lewis K, Epstein SS. Isolating “uncultivable” microorganisms in pure
[19]
M AN U
378
SC
376
Brooks JP, Edwards DJ, Harwich Jr. MD, Rivera MC, Fettweis JM, Serrano MG, et al.
381
The truth about metagenomics: quantifying and counteracting bias in 16S rRNA studies.
382
BMC Microbiol 2015;15:66. [20]
Bik EM, Costello EK, Switzer AD, Callahan BJ, Holmes SP, Wells RS, et al. Marine
TE D
383 384
mammals harbor unique microbiotas shaped by and yet distinct from the sea. Nat
385
Commun 2016;7:10516.
evolution of the vertebrate gut microbiota. Nat Rev Microbiol 2008;6:776–788.
387 388
[22]
(Phocoena phocoena) from Portugal (western Iberian Peninsula). Environ Pollut
390
2016;210:293–302.
391
393
Ferreira M, Monteiro SS, Torres J, Oliveira I, Sequeira M, López A, et al. Biological variables and health status affecting inorganic element concentrations in harbour porpoises
389
392
Ley RE, Lozupone CA, Hamady M, Knight R, Gordon JI. Worlds within worlds:
EP
[21]
AC C
386
[23]
Monteiro SS, Torres J, Ferreira M, Marçalo A, Nicolau L, Vingada J V, et al. Ecological variables influencing trace element concentrations in bottlenose dolphins (Tursiops
ACCEPTED MANUSCRIPT 18 394
truncatus, Montagu 1821) stranded in continental Portugal. Sci Total Environ
395
2016;544:837–844.
396
[24]
Kuiken T, Hartmann MG. Proceedings of the First ECS Workshop on Cetacean Pathology : dissection techniques and tissue sampling, Leiden, Netherlands: [European
398
Cetacean Society]; 1991.
399
[25]
RI PT
397
Pugliares KR, Bogomolni A, Touhey KM, Herzig SM, Harry CT, Moore MJ. Marine mammal necropsy : an introductory guide for stranding responders and field biologists.
401
Woods Hole, MA: Woods Hole Oceanographic Institution; 2007. [26]
1991.
403 404
Lane DJ. 16S/23S rRNA sequencing. Chichester, United Kingdom: John Wiley and Sons;
M AN U
402
SC
400
[27]
Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, et al. Ultrahigh-throughput microbial community analysis on the Illumina HiSeq and MiSeq
406
platforms. ISME J 2012;6:1621–1624.
407
[28]
TE D
405
Kuczynski J, Stombaugh J, Walters WA, González A, Caporaso JG, Knight R. Using QIIME to analyze 16S rRNA gene sequences from microbial communities. Curr Protoc
409
Bioinformatics 2011;Chapter 10:Unit 10.7. [29]
2010;26:2460–2461.
411 412
[30]
Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 2011;27:2194–2200.
413 414
Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics
AC C
410
EP
408
[31]
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal
415
RNA gene database project: improved data processing and web-based tools. Nucleic
416
Acids Res 2013;41:D590–6.
ACCEPTED MANUSCRIPT 19 [32]
completeness rather than size. Ecology 2012;93:2533–2547.
418 419
[33]
Lozupone C, Knight R. UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol 2005;71:8228–8235.
420 421
Chao A, Jost L. Coverage-based rarefaction and extrapolation: standardizing samples by
[34]
RI PT
417
Oksanen J, Blanchet F, Kindt R, Legendre P, Minchin P, O’Hara R, et al. vegan: Community Ecology Package. R package version 2.3-4 2016. https://cran.r-
423
project.org/package=vegan. [35]
Wickham H. ggplot2 - Elegant Graphics for Data Analysis, Hadley Wickham, Springer. 1st ed. Springer-Verlag New York; 2009.
425
M AN U
424
SC
422
426
[36]
Wickham H. Reshaping Data with the reshape Package. J Stat Softw 2007;21:1–20.
427
[37]
Langille MGI, Zaneveld J, Caporaso JG, McDonald D, Knights D, Reyes JA, et al. Predictive functional profiling of microbial communities using 16S rRNA marker gene
429
sequences. Nat Biotechnol 2013;31:814–821.
430
[38]
TE D
428
Arndt D, Xia J, Liu Y, Zhou Y, Guo AC, Cruz JA, et al. METAGENassist: a comprehensive web server for comparative metagenomics. Nucleic Acids Res
432
2012;40:W88–95. [39]
uteri gen. nov., sp. nov., a new member of the family Pasteurellaceae Pohl (1979) 1981
434
isolated from a harbour porpoise (Phocoena phocoena). Int J Syst Evol Microbiol 2000;50
435
Pt 1:135–139.
436 437
Foster G, Ross HM, Malnick H, Willems A, Hutson RA, Reid RJ, et al. Phocoenobacter
AC C
433
EP
431
[40]
Lee C, Agidi S, Marion JW, Lee J. Arcobacter in Lake Erie beach waters: an emerging
438
gastrointestinal pathogen linked with human-associated fecal contamination. Appl
439
Environ Microbiol 2012;78:5511–5519.
ACCEPTED MANUSCRIPT 20 440
[41]
players in sulphidic habitats. Nat Rev Microbiol 2006;4:458–468.
441 442
Campbell BJ, Engel AS, Porter ML, Takai K. The versatile epsilon-proteobacteria: key
[42] Chow SK, Clarridge III JE. Identification and clinical significance of Helcococcus species, with description of Helcococcus seattlensis sp. nov. from a patient with urosepsis. J Clin
444
Microbiol 2014;52:854–858. [43]
from a human wound. Int J Syst Evol Microbiol 2004;54:1557–1560.
446 447
Collins MD, Falsen E, Brownlee K, Lawson PA. Helcococcus sueciensis sp. nov., isolated
[44]
SC
445
RI PT
443
Småge SB, Brevik ØJ, Duesund H, Ottem KF, Watanabe K, Nylund A. Tenacibaculum finnmarkense sp. nov., a fish pathogenic bacterium of the family Flavobacteriaceae
449
isolated from Atlantic salmon. Antonie Van Leeuwenhoek 2016;109:273–285.
450
[45]
M AN U
448
Stoddard RA, Atwill ER, Gulland FMD, Miller MA, Dabritz HA, Paradies DM, et al. Risk factors for infection with pathogenic and antimicrobial-resistant fecal bacteria in northern
452
elephant seals in California. Public Health Rep 2008;123:360–370. [46]
caused by HACEK microorganisms. Annu Rev Med 1997;48:25–33. [47]
1982;31:95–104.
456 457
[48]
Scaglione FE, Bollo E, Pregel P, Chiappino L, Sereno A, Mignone W, et al. Heart pathologies in dolphins stranded along the northwestern Italian coast. Dis Aquat Organ
458
2013;107:31–36.
459 460
Nord CE. Anaerobic bacteria in septicaemia and endocarditis. Scand J Infect Dis Suppl
EP
454 455
Das M, Badley AD, Cockerill FR, Steckelberg JM, Wilson WR. Infective endocarditis
AC C
453
TE D
451
[49]
Cools P, Haelters J, Lopes dos Santos Santiago G, Claeys G, Boelens J, Leroux-Roels I, et
461
al. Edwardsiella tarda sepsis in a live-stranded sperm whale (Physeter macrocephalus).
462
Vet Microbiol 2013;166:311-315.
ACCEPTED MANUSCRIPT 21 463
[50]
Foster G, Ross HM, Naylor RD, Collins MD, Ramos CP, Fernandez Garayzabal F, et al.
464
Cetobacterium ceti gen. nov., sp. nov., a new Gram-negative obligate anaerobe from sea
465
mammals. Lett Appl Microbiol 1995;21:202–206. [51]
Naim R, Yanagihara I, Iida T, Honda T. Vibrio parahaemolyticus thermostable direct
RI PT
466 467
hemolysin can induce an apoptotic cell death in Rat-1 cells from inside and outside of the
468
cells. FEMS Microbiol Lett 2001;195:237–244. [52]
Jaber JR, Pérez J, Rotstein D, Zafra R, Herráez P, Carrascosa C, et al. Biliary cirrhosis
SC
469
caused by Campula spp. in a dolphin and four porpoises. Dis Aquat Organ 2013;106:79–
471
84.
472
[53]
Allen-Vercoe E, Strauss J, Chadee K. Fusobacterium nucleatum: an emerging gut pathogen? Gut Microbes 2011;2:294–298.
473 474
M AN U
470
[54] Fishelson L, Montgomery L, Myrberg W, Arthur A. A unique symbiosis in the gut of tropical herbivorous surgeonfish (Acanthuridae: teleostei) from the red sea. Science
476
1985;229:49–51.
477
[55]
TE D
475
López A, Santos MB, Pierce GJ, González AF, Valeiras X, Guerra A. Trends in strandings and by-catch of marine mammals in north-west Spain during the 1990s. J Mar Biol Assoc
479
UK 2002;82:513–521. [56]
lesions in cetacean stranded on the northwestern Spanish Atlantic coast. Dis Aquat Organ
481
1998;32:227–231.
482 483 484 485
Abollo E, López A, Benavente P, Cabaleiro M, Pascual S. Macroparasites andgross
AC C
480
EP
478
[57]
Pascual S, Abollo E, López A. Elemental analysis of cetacean skull lesions associated with nematode infections. Dis Aquat Organ 2000;42:71–75.
ACCEPTED MANUSCRIPT 22 486 487
Table and Figure Legends
488
Table 1. Number of sequences and OTU estimates across body sites.
490
Fig. 1. Schematic representation of the procedure outline for assessing the microbiome
491
composition of stranded cetaceans.
492
Fig. 2. Measures of microbial diversity of the eight body sites. Panel A depicts a beta
493
diversity plot based on UniFrac distances calculated on the OTU table of the eight body sites.
494
Panel B shows alpha diversity rarefaction curves for Chao1 richness estimates for each body site.
495
Fig. 3. Taxonomic distribution of the OTUs at the phylum level (panel A) and genus level
496
(panel B) for each body site. Panel A: A total of 15 phyla were detected; the five most abundant
497
(>5%) were Proteobacteria, Fusobacteria, Firmicutes, Bacteroidetes and Tenericutes. The
498
remaining phyla had a relative abundance below 5%. Panel B: A total of 108 genera were
499
detected, but some had only residual abundance (<0.5%).
500
Fig. 4. Genus level heat map depicting significantly different OTUs according to body site.
501
The 51 OTUs were selected after a statistical G-test comparing OTU frequencies in the sample
502
groups. Only those taxa that showed p<0.05 were selected. The dendrograms were made through
503
hierarchical clustering using the ggdendro and ggplot2 packages in R. Percentage of identity (ID)
504
of the best-hit matches of the respective representative sequences, using BLASTN analysis, are
505
indicated next to the assigned taxonomy.
506
AC C
EP
TE D
M AN U
SC
RI PT
489
ACCEPTED MANUSCRIPT Table 1. Number of sequences and OTU estimates across body sites
Number
of Total number Number
sequences
of OTUs per unique OTUs sample
per sample
43,896
122
32
Colon
39,579
75
1
Intestine
74,465
54
0
Stomach
62,886
85
3
Duodenum
57,705
70
0
Tongue
47,249
149
1
Oral
289,253
156
14
Genitals
24,083
95
0
Total
639,116
235(unique)
51
M AN U
TE D EP
SC
Blowhole
AC C
of
RI PT
Sample Type
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0923-2508(16)30096-1
DOI:
10.1016/j.resmic.2016.08.004
Reference:
RESMIC 3534
To appear in:
Research in Microbiology
Received Date: 5 May 2016 Revised Date:
20 July 2016
Accepted Date: 23 August 2016
Please cite this article as: F. Godoy-Vitorino, A. Rodriguez-Hilario, A.L. Alves, F. Gonçalves, B. CabreraColon, C.S. Mesquita, P. Soares-Castro, M. Ferreira, A. Marçalo, J. Vingada, C. Eira, P.M. Santos, The microbiome of a striped dolphin (Stenella coeruleoalba) stranded in Portugal, Research in Microbiologoy (2016), doi: 10.1016/j.resmic.2016.08.004. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
For publication
2
The microbiome of a striped dolphin (Stenella coeruleoalba) stranded in
3
Portugal
4
Filipa Godoy-Vitorinoa, Arnold Rodriguez-Hilarioa, Ana Luísa Alvesb, Filipa Gonçalvesb,
5
Beatriz Cabrera-Colona, Cristina Sousa Mesquitab, Pedro Soares-Castrob, Marisa Ferreirab,c, Ana
6
Marçalod, José Vingadac,e, Catarina Eirac,d, Pedro Miguel Santosb*
SC
RI PT
1
a
Microbial Ecology and Genomics Lab, Department of Natural Sciences, Inter American University of
7
b
Department of Biology and Centre for Molecular and Environmental Biology (CBMA), University of
9
Minho, Campus de Gualtar 4710-087 Braga, Portugal
10 11
M AN U
Puerto Rico, Metropolitan Campus, P.O. Box 191293 San Juan, Puerto Rico 00919-1293
8
c
Portuguese Wildlife Society (SPVS), Quiaios Field Station, Apartado 16 EC Quiaios, 3081-101 Figueira da Foz, Portugal
12
22 23 24 25 26 27
Department of Biology and CESAM, University of Minho, Campus de Gualtar 4710-087 Braga, Portugal
E-mail addresses: [email protected] (F. Godoy-Vitorino), [email protected] (A. Rodriguez-Hilario), [email protected] (B. Cabrera-Colon), [email protected] (A.L. Alves), [email protected] (F. Gonçalves), [email protected] (C.S. Mesquita), [email protected] (P. Soares-Castro), [email protected] (M. Ferreira), [email protected] (A. Marçalo), [email protected] (J. Vingada), [email protected] (C. Eira), [email protected]* (P.M. Santos) *Corresponding author
EP
15 16 17 18 19 20 21
e
AC C
14
Department of Biology and CESAM, University of Aveiro, 3810-193 Aveiro, Portugal
TE D
d
13
ACCEPTED MANUSCRIPT 2 28
Abstract Infectious diseases with epizootic consequences have not been fully studied in marine
30
mammals. Presently, the unprecedented depth of sequencing, made available by high-throughput
31
approaches, allows detailed comparisons of the microbiome in health and disease. This is the
32
first report of the striped dolphin microbiome in different body sites. Samples from one striped
33
female edematous dolphin were acquired from a variety of body niches, including the blowhole,
34
oral cavity, oral mucosa, tongue, stomach, intestines and genital mucosa. Detailed 16S rRNA
35
analysis of over half a million sequences identified 235 OTUs. Beta diversity analyses indicated
36
that microbial communities vary in structure and cluster by sample origin. Pathogenic, Gram-
37
negative, facultative and obligate anaerobic taxa were significantly detected, including
38
Cetobacterium, Fusobacterium and Ureaplasma. Phocoenobacter and Arcobacter dominated the
39
oral-type samples, while Cardiobacteriaceae and Vibrio were associated with the blowhole and
40
Photobacterium were abundant in the gut. We report for the first time the association of
41
Epulopiscium with a marine mammal gut.
TE D
M AN U
SC
RI PT
29
The striped dolphin microbiota shows variation in structure and diversity according to the
43
organ type. The high dominance of Gram-negative anaerobic pathogens evidences a cetacean
44
microbiome affected by human-related bacteria.
45
Keywords: Cetacean; Striped dolphin; 16S rRNA; Microbiota; Metagenomics
46
1. Introduction
AC C
47
EP
42
The integrity and ecological sustainability of marine ecosystems are under increasing
48
threats mainly due to long-term consequences of climate change, habitat degradation and human
49
impact [1,2]. Marine mammal species are top predators and have long life spans. Often, they
50
need to cope with exposure to a variable combination of environmental stressors (e.g. xenobiotic
ACCEPTED MANUSCRIPT 3 pollutants) and pathogens [3]. Therefore, marine mammals can be considered prime sentinel
52
species of aquatic ecosystem health. In fact, the low resilience of marine mammal species results
53
from their low fecundity rate and their position at the top of food webs, which makes them more
54
susceptible to many human-induced pressures [4]. In Europe, there is growing concern about the
55
current health status of marine mammal populations. Significantly, a decline in some coastal
56
species such as the harbor porpoise (Phocoena phocoena) was registered [5–7], including the
57
population of the Iberian Atlantic coast, where a new porpoise ecotype was recently proposed [8].
58
Likewise, the health status of striped dolphin (Stenella coeruleoalba) populations is of great
59
concern, particularly the Mediterranean population, considering their very high mortality rates
60
during Morbillivirus (MV) outbreaks [9–11].
M AN U
SC
RI PT
51
Information on microbiological infections in cetaceans worldwide is critical for
62
understanding epidemiology in their populations, as well as warning about ocean health changes,
63
even more so in light of additional non-infectious aggressions mainly due to human activities,
64
that already render these populations susceptible to disease. In some reports, the death of
65
stranded dolphins has been associated with viruses (in addition to MV outbreaks), bacteria, fungi
66
and parasites. In striped dolphins, these included bacteria such as Brucella affecting the central
67
nervous system [12], fungal yeast such as Cryptococcus gattii associated with pulmonary disease
68
[13] and parasites such as Toxoplasma gondii in cerebral toxoplasmosis [14]. In bottlenose
69
dolphins (Tursiops truncatus), Papillomaviruses (PVs) have been associated with malignant
70
genital tumors [15]. However, information on striped dolphin health remains scarce, and most
71
available studies are based on bottlenose dolphins, the most common cetacean species in
72
captivity around the world. For example, a study performed in a bottlenose dolphin in captivity
73
over a 7-year period documented potentially zoonotic bacterial pathogens through isolation of
AC C
EP
TE D
61
ACCEPTED MANUSCRIPT 4 74
bacteria such as Proteus mirabilis, Staphylococcus aureus and Pseudomonas aeruginosa [16].
75
Microbial isolates from free-ranging bottlenose dolphins in Florida, Texas and North Carolina
76
evidenced abundant Staphylococcus spp. Contents, suggesting direct human impact [17]. In order to treat live-stranded or captive dolphins, we need to be able to assess the
78
vulnerability of free-ranging dolphin populations and to evaluate the risk of disease transmission
79
between humans and dolphins. Also, to further assess the proposed role of dolphins as
80
environmental sentinels, we need detailed characterization of the dolphin microbiota in health
81
and disease contexts. Since ~99% of all microorganisms are recalcitrant to isolation under
82
conventional laboratory conditions [18], next-generation sequencing technologies applied to 16S
83
rRNA genes have been playing a pivotal role in
84
natural ecosystems [19]. The unprecedented depth of sequencing made available by high-
85
throughput approaches allows for finely detailed descriptions of the microbiome of marine
86
mammals, and thus provides a relevant approach to environmental surveillance of the marine
87
environment. Very recently, the microbial taxonomic composition of 48 healthy bottlenose
88
dolphins (38 in captivity and 10 wild animals) and 18 California sea lions (Zalophus
89
californianus) was reported [20]. The authors found as many as 30 phyla in the dolphin
90
specimens and
91
overall bacterial taxonomic compositions of the marine mammals were distinct from dietary fish
92
and seawater. Similar findings have been reported for terrestrial vertebrate gut microbiota [21].
SC
M AN U
microbial diversity in
EP
TE D
characterization of
significant differences between wild and captive dolphins, concluding that the
AC C
93
RI PT
77
A multidisciplinary team from Portuguese and Spanish research centers was established
94
some years ago to evaluate the health status of marine mammal populations along the Iberian
95
Atlantic coast,
96
species by collecting data regarding trait history, contaminants, bacteriological and fungal
systematically
assessing population status and health of coastal cetacean
ACCEPTED MANUSCRIPT 5 97
imbalances, prevalence of viral diseases, parasite load and tissue damage [22,23]. In the context
98
of this integrative approach, standardized sample collection procedures for evaluating the
99
taxonomic composition of the microbiome of cetaceans were implemented. Here,
first results
emerging from this collaborative effort are reported. Specifically, a 16S rRNA gene-based
101
Illumina MiSeq sequencing approach was used to investigate
102
composition and structure in an edematous female striped dolphin stranded along the Algarve
103
coast in Portugal.
104
2. Materials and methods
105
2.1. Sample collection
RI PT
100
M AN U
SC
microbial community
The striped dolphin sampled in the present study was stranded alive in the Algarve at
107
the Meia Praia Beach (37.115885ºN, -8,648677°E), and died shortly after the arrival of a rescue
108
team from the Marine Animal Stranding Network (coordinated by the Instituto para a
109
Conservação da Natureza e Florestas, http://www.icnf.pt/portal/icnf, Portugal, in cooperation
110
with the Portuguese Wildlife Society (SPVS)). The dead animal was taken to the Quiaios Field
111
Station in Figueira da Foz (Portugal), where the respective necropsy took place. The dolphin was
112
an immature female with a total length of 191 cm and 63.6 kg of weight. Post-mortem analysis
113
was performed within 12 h after the animal’s death according to standard protocols [24]. The
114
occurrence of gross pathologies was recorded if any evidence of pathology (lesions or abnormal
115
appearance) was found in any organ system [25]. The animal was emaciated and presented low
116
consistency of brain, liver and pancreas tissues, lung lesions compatible with pneumonia and
117
ulcerous formations in the stomach chambers.
AC C
EP
TE D
106
ACCEPTED MANUSCRIPT 6 During the necropsy, tissue and swab samples from different body sites (blowhole, colon,
119
intestine, stomach, tongue, oral cavity (mouth), duodenum, genitals) were collected in triplicate
120
and stored in RNA-Later and frozen within an hour after the necropsy at -80 °C.
121
2.2. Genomic DNA extraction, PCR Amplifications and Sequencing
RI PT
118
Total DNA extractions were done in triplicate (for each sample) using the PowerSoil DNA
123
isolation kit (MoBio). Briefly, 250 mg of tissue or 250 ml of total pellet recovered from swabs
124
(following extensive washing with saline solution) were added to powerbead tubes (MoBio). All
125
subsequent steps were performed according to the manufacturer’s instructions. The concentration
126
and quality of DNA was evaluated using a NanoDrop™ 1000 spectrophotometer.
M AN U
SC
122
To determine whether the extracted DNA samples included intact DNA of bacterial
128
origin, PCR amplification of the whole 16S rRNA gene was carried out using universal primers
129
27F and 1492R [26]. Due to varying PCR yields across samples, PCR reactions were pooled
130
from triplicate samples from each origin prior to sequencing.
TE D
127
The composition of microbial communities from the different striped dolphin samples
132
was assessed using primers for paired-end V3-V4 16S rRNA sequencing on the Illumina MiSeq
133
platform [27].
134
2.3. Data deposition
136 137
The sequences used in this study have been deposited with BioProject database ID
AC C
135
EP
131
PRJNA316486 in SRA with accession number SRP072736. 2.4. Sequence processing and data analyses
138
Sequencing reads were processed using QIIME [28] with strict quality and size filtering,
139
de-replication and clustering (Usearch [29]). The resulting high-quality dataset was then
140
screened for chimeric sequences and contaminant chloroplast DNA (Uchime [30]). Filtered
ACCEPTED MANUSCRIPT 7 141
sequences were used to evaluate the diversity and taxonomic composition of target samples
142
using
143
performed. Sequence binning was done using the 97% identity threshold with the Silva database
144
[31]. Data were subsampled,
145
Richness curves were built using the The chao 1 index that uses low frequency counts for
146
richness (singletons and doubletons) to estimate the number of missing species [32].
beta and alpha diversity measures and significance tests were
thus mitigating biases due to differences in sampling depth.
RI PT
QIIME 1.9.1, where all
Principal component analyses of bacterial communities from the eight body site samples
148
were compared phylogenetically using UniFrac distances [33], a β-diversity measure that uses
149
phylogenetic information to compare samples in which the distances between them represent
150
their ecological distances. The “R” Vegan package [34] was used to draw heat maps, and ggplot2
151
[35] and reshape2 [36] were used to draw dendrograms. PICRUSt (phylogenetic investigation of
152
communities by reconstruction of unobserved states) was used for functional metagenome
153
prediction, estimating gene families (KEGG KO functions) contributing to a bacterial
154
metagenome identified by 16S rRNA sequencing [37].
155
3. Results
TE D
M AN U
SC
147
In order to monitor the microbiome composition of stranded cetaceans, a standardized
157
procedure pipeline was implemented, as depicted in Fig. 1. Briefly, following cetacean stranding
158
registration, the animal is evaluated and, whenever it evidences a fresh status (time of death
159
below 24 h), samples for microbiome content evaluation are collected during necropsy.
160
Following total DNA extraction, all samples evidencing the presence of bacteria (assessed by
161
PCR amplification of the full 16S rRNA region) are then sequenced using V3-V4 16S rRNA
162
protocol on the Illumina MiSeq platform. Here, the first results of this procedure are presented.
163
AC C
EP
156
ACCEPTED MANUSCRIPT 8 164
3.1. Community diversity and structure An initial set of 923,066 raw reads underwent strict filtering, including dereplication,
166
quality and size filtering, removing reads with poor quality scores, chimeras and chloroplasts. A
167
total of 639,116 sequences with an average read length of about 460 bp passed preprocessing.
168
The number of sequences per sample varied from a minimum of 24,083 to a maximum of
169
289,253 sequences, with an average of 79,904 sequences per sample (Table 1). Differences
170
between the total number of retrieved sequences may suggest significant variations in bacterial
171
richness and abundance according to the sampled body site. Binning of the 639,116 sequences
172
resulted in 235 operational taxonomic units (OTUs) (Table 1). All samples were rarefied to an
173
even sampling depth of 24,000 sequences prior to statistical analysis. The samples with the
174
highest number of OTUs were the oral cavity, the tongue and the blowhole (Table 1). Oral cavity
175
bacteria shared more OTUs with the tongue compared to all other body sites, followed by the
176
genital sample, which also shared OTUs with both the tongue and oral cavity (Supplementary
177
Table 1S). PCoA of UniFrac distances revealed that bacterial communities from each body site
178
segregated the duodenum and stomach from all other samples (Fig. 2). The genital sample shared
179
fewer OTUs with the other samples, and there existed community similarities between tongue
180
and oral samples as well as between colon and intestine, with the blowhole having a community
181
similar to the oral and intestinal samples (Fig. 2). The tongue, oral, blowhole and genitals
182
samples showed the highest Chao1 richness values and were separate from gut samples (Fig. 2,
183
Supplementary Fig. 1S, Supplementary Fig. 2S).
AC C
EP
TE D
M AN U
SC
RI PT
165
184 185 186
3.2. Microbial taxonomic profiles A total of 15 phyla, with the 5 most abundant (>5%) being Proteobacteria, Fusobacteria,
ACCEPTED MANUSCRIPT 9 Firmicutes, Bacteroidetes and Tenericutes, were found. The remaining phyla had less than 5%
188
abundance. Proteobacteria, Fusobacteria Firmicutes and Bacteroidetes were present across all
189
samples (Fig. 3). Genitals and blowholes were the external body orifices with a dominance of
190
Fusobacteria, Bacteroidetes and Proteobacteria, with half of its community dominated by
191
Cardiobacteriales and Pasteurellales (~43%) (Fig. 3, Supplementary Fig. 3S). Oral and tongue
192
samples were dominated by Proteobacterial orders Pasteurellales and Campylobacterales (~85%).
193
The stomach sample had a dominance of Tenericutes order Mycoplasmatales (~43%) followed
194
by Firmicutes, and the duodenum had a dominance of Firmicutes order Clostridiales (~98%),
195
Intestines and colon evidenced a similar community structure dominated by Proteobacteria,
196
Fusobacteria and Firmicutes. Interestingly, the amount of Bacteroidetes decreased from open
197
orifices, through the oral cavity and gut, with virtually no such taxa in the intestine or colon; the
198
opposite occurred with the Firmicutes.
M AN U
SC
RI PT
187
A total of 108 genera were detected, most of which had residual abundance (<0.5%), and
200
each sample had a dominance of different taxa; the blowhole had ~27% of uncultured
201
Cardiobacteraceae,
202
Tenacibaculum
203
Porphyromonas (~20%), Actinobacillus (15%), Parvimonas (~7%), and Ureaplasma (5%). The
204
oral cavity had dominant Phocoenobacter (~30%) and Arcobacter (24%), but also
205
Porphyromonas (~14%) very similar to the tongue’s genus profile, with a slightly higher amount
206
of
207
Peptostreptococaceae-IncertaeSedis (~98%), representing a dominance of ~40% in the stomach,
208
followed by Ureaplasma (~36%), Mycoplasma (~7%) and Helicobacter (4%). The intestine had
209
a dominance of Photobacterium (64%), Cetobacterium (20%) and Peptostreptococaceae-
TE D
199
~18%
Helcococcus.
Pasteurellaceae,
The
genital
~9%
sample
had
Actinobacillus
and
Fusobacterium
5%
(25%),
AC C
EP
and
uncultured
Phocoenobacter
(~52%).
The
duodenum
samples
evidenced
a
dominance
of
ACCEPTED MANUSCRIPT 10 210
IncertaeSedis (14%). The colon had a dominance of Actinobacillus (~33%), Photobacterium
211
(30%), Cetobacterium (10%) and Epulopiscium (8%) (Fig. 3). Tongue and oral cavity samples
212
apparently shared Phocoenobacter,
213
(Supplementary Fig. 4S). Shared taxa between colon and intestine included Photobacterium,
214
Actinobacillus and Cetobacterium. The blowhole shared Actinobacillus with the colon and
215
intestine (Supplementary Fig. 4S).
RI PT
Arcobacter and Porphyromonas with a genital sample
Significantly different taxa (p<0.05) included Streptobacillus, Arcobacter, Vibrio and
217
Tenacibacter in the blowhole (taxa absent in all other samples). In the colon, Actinobacillus and
218
Epulopiscium were significantly more abundant while, in the intestine,
219
damselae and Cetobacterium predominated; the stomach had Helicobacter heilmanii,
220
Mycoplasma, Ureaplasma and Paeniclostridium sordellii; in the tongue, Phocoenobacter was
221
significantly more abundant (Fig. 4). Oral samples had diverse and highly abundant taxa
222
including Campylobacter, Parvimonas, Tenacibaculum, Porphyromonas, Psychrobacter and
223
Arcobacter sp., while in the duodenum, Paeniclostridium sordellii dominated and in the genitals
224
Actinobacillus scotiae and Fusobacterium were significantly more abundant (Fig. 4).
227
M AN U
TE D
EP
226
Photobacterium
3.3. Phenotypic and metabolic inference Based on inference of taxonomic-to-phenotypic mapping of metabolism using
AC C
225
SC
216
228
METAGENassist [38], ammonia oxidizing pathways in the blowhole, sulfate reducers and
229
oxidizers in the colon, chitin degradation and denitrification in the intestine, nitrogen fixation in
230
the stomach, ammonia oxidizers and dehalogenation in the oral cavity and ammonia oxidizers in
231
the genitals were found (Supplementary Fig. 5S). As a whole, we found mostly anaerobic and
232
facultative anaerobic bacteria, and a dominance of Gram-negative pathogenic bacteria in most
ACCEPTED MANUSCRIPT 11 organs except for the duodenum and intestines (Supplementary Fig. 5S). Picrust analyses for oral
234
and gut samples revealed that taxa in these samples included metabolic pathways previously
235
associated with human diseases such as cell signaling in H. pylori infections, Tuberculosis, the
236
Vibrio chloreae cycle and immunodeficiency were significantly associated with gut samples.
237
Pathways associated with pertussis and Huntington’s disease were significantly associated with
238
taxa found in the oral samples (Supplementary Fig. 6S).
239
4.
SC
Discussion
RI PT
233
Studies on the impact of microbial pathogens in dolphin strandings have relied on
241
cultivation techniques, PCR and clonal sequencing; however, very few studies have focused on
242
the inventory of bacteria in different body sites using next-generation sequencing, a thorough and
243
sophisticated approach to understanding the overall health status of the animal. Here, the first
244
microbiome inventory of a striped dolphin, a small oceanic delphinid species, is presented. One
245
of the potential bottlenecks of this approach was the high risk of sample contamination with
246
external sources during stranded animal handling up until sample collection. Nonetheless,
247
bacterial inventory of the striped dolphin specimen under analysis included a variety of bacteria
248
in a range of 15 phyla,
249
conducted on bottlenose dolphins along the US Coast [20], although with a different relative
250
abundance of the main phyla. This corroborates the validity of
251
sample collection from stranded specimens. Compared to a large marine mammal study by Bik
252
and co-workers [20], we confirm that oral samples were richer than gut samples, as were those in
253
the blowhole and genitals.
TE D
M AN U
240
AC C
EP
also reported in a more extensive microbiome composition survey
implemented procedures for
254
Most of the bacteria within the 15 phyla changed their relative abundance in each organ
255
type, but overall, we found a dominance of mostly facultative, anaerobic, Gram-negative taxa,
ACCEPTED MANUSCRIPT 12 some of which are host-adapted and
could be traced to human infections. Some of these
257
dominant bacteria included common residents of the mucous membranes of animals, such as
258
Pasteurellaceae OTUs, which can be commensals, as well as Actinobacillus and
259
Phocoenobacter, which were recently recovered from marine mammal carcasses and were
260
related to septicemia [39]. Actinobacillus were also found in high abundance in gastric rectal
261
samples from healthy dolphins, but Phocoenobacter is apparently absent in the healthy animals
262
[20]. These taxa were found in the striped dolphin in very high abundance, corresponding to
263
nearly half of the bacterial community in the blowhole and colon as well as in the tongue and
264
oral samples. We found several obligatory anaerobic bacteria such as Arcobacter, a genus
265
associated with human and animal illness [40] as well as
266
hemolytic bacteria [42,43] and Tenacibacullum (bacteria common in the GI tract and fish
267
pathogen [44]) are all obligatory anaerobic bacteria. Campylobacter of human and livestock
268
origin present in animal oral samples, are regarded as zoonotic bacteria, causing gastroenteritis
269
and septicemia and occasionally other conditions in marine mammals [45]. Cardiobacteriaceae
270
OTUs were also found in the blowhole, tongue and oral cavity of the dolphin, and these
271
anaerobic taxa are associated with bacteremia and wound infections in humans [46]. The fact
272
that these taxa are anaerobic and have entry portals mostly in the gut, and are found at a highly
273
oxygenized area of the animal such as the blowhole, indicates
274
stranded striped dolphins from Italy were diagnosed with cardiovascular pathologies, including
275
ventricular dilation associated with hypoplasia of the tricuspid chordae and valvular fibrosis,
276
ventricular hypertrophy and myocarditis; however, no pathogens were found [48]. Additionally,
277
another study reporting the first case of sepsis in a live-stranded sperm whale
278
pathogen Edwardsiella tarda, relating these bacteria to stranding and death of the animal [49].
M AN U
SC
RI PT
256
AC C
EP
TE D
marine sediments [41]; Helcococcus -
sepsis [47]. Recently, nine
found the
ACCEPTED MANUSCRIPT 13 The abundance of Cardiobacteriaceae OTUs in the dolphin may suggest a microbial role for
280
similar cardiopathologies. Other obligate Gram-negatives such as Cetobacterium were abundant
281
in the colon and intestines and had already been isolated from other marine mammals [50].
282
Pathogenic Vibrionaceae were also abundant in the colon and intestines (Photobacterium), as
283
well as Vibrio, a known cytotoxic bacteria, in the blowhole [51]. Gastric inflammation has been
284
considered a cytologic marker of systemic illness in dolphins; however, no clinically significant
285
findings of cultured bacteria and fungi have been reported, except for a case of biliary cirrhosis
286
produced by Campula spp [52]. In agreement with this, the presence of Ureaplasma in the
287
stomach of healthy asymptomatic bottlenose dolphins was previously reported [20]; thus, the
288
impact of these organisms in the health of dolphins remains unknown. Another taxon that may be
289
a biomarker for disease is, for instance, genital Fusobacterium, a Gram-negative, obligatory
290
anaerobic species that is a common resident of the GI tract, but that possesses a number of
291
pathogenic traits with relevance to gut diseases such as inflammatory bowel disease [53].
292
Interestingly, we found Epulopiscium, a large symbiotic bacteria found in the tropical surgeon
293
fish [54], probably helping fish to break down algal nutrients in the diet. This is the first report
294
of Epulopiscium in the gut of a marine mammal.
SC
M AN U
TE D
EP
295
RI PT
279
In conclusion, the main goal of this work was to reveal, with high resolution, the microbiome of a critically ill female edematous striped dolphin.
297
diverse community, mainly dominated by Gram-negative anaerobic bacteria of human origin.
298
The veterinary bibliography on marine mammal research regarding infectious diseases in
299
strandings reveals a remarkable scarcity of comprehensive literature about diseases/disorders
300
affecting different body systems of dolphins, and these data are quite limited in marine mammals
301
inhabiting Portuguese and northern Spanish waters. Data exist on the effects of accidental
AC C
296
As a result, we
report a
ACCEPTED MANUSCRIPT 14 capture [55], the prevalence of some parasite species [56,57] and pollutants [5,22,23] in marine
303
mammal populations. However, studies on infectious disease epidemiology in cetacean species
304
are lacking despite recent reports of re-emerging or new infectious diseases in marine mammal
305
populations, with epizootic and potential zoonotic consequences.
RI PT
302
Our in-depth study using next-generation sequencing technologies is the first bacterial
307
metagenomic survey of an edematous striped dolphin to show that the analyzed animal
308
harbored multiple bacterial pathogens resembling those present in humans. Whether human
309
activities
310
light on the cetacean pathogenic microbiome that critically impacted a striped dolphin on the
311
Iberian Peninsula. Detailed epidemiological models using next-generation sequencing and
312
metagenomics as tools to predict new outbreaks and facilitate their control are warranted.
SC
306
M AN U
actually contribute to such a burden remains to be clarified. This study sheds new
313
315
Acknowledgements
TE D
314
The authors would like to acknowledge Jörg Becker and João Sobral for
sequencing
services provided at the Instituto Gulbenkian de Ciência and to Ricardo Araújo for preliminary
317
DNA extraction tests.
EP
316
This work was supported by project CetSenti RECI/AAG-GLO/0470/2012 (FCOMP-01-
319
0124-FEDER-027472), FCT/MCTES (PIDDAC) and FEDER-COMPETE (POFC), the strategic
320
program
321
through the FCT I.P., by the ERDF through the COMPETE2020 - Programa Operacional
322
Competitividade e Internacionalização (POCI) and through PhD grants (SFRH/BD/76894/2011,
323
SFRH/BD/98558/2013) attributed to P.S-C and C.S.M, respectively. The work was also
324
supported by the FCT through CESAM UID/AMB/50017/2013 co-funded by the FCT/MEC and
AC C
318
UID/BIA/04050/2013 (POCI-01-0145-FEDER-007569) funded by national funds
ACCEPTED MANUSCRIPT 15 325
FEDER, within PT2020 and Compete 2020 (C.E.) and grants SFRH/BD/30240/2006 (M.F.) and
326
SFRH/BPD/64889/2009 (A.M.).
327
The authors declare no conflict of interest.
329 330
References
331
[1]
SC
Dayton PK. The importance of the natural sciences to conservation: (an American Society of Naturalists symposium paper). Am Nat 2003;162:1–13.
332 333
RI PT
Conflicts of interest
[2]
Marques M, da Costa MF, Mayorga MI de O, Pinheiro PR. Water environments:
M AN U
328
334
anthropogenic pressures and ecosystem changes in the Atlantic drainage basins of Brazil.
335
Ambio 2004;33:68–77. [3]
2011;48:676–690.
337 338
Bossart GD. Marine mammals as sentinel species for oceans and human health. Vet Pathol
[4]
TE D
336
Peltier H, Baagøe HJ, Camphuysen KCJ, Czeck R, Dabin W, Daniel P, et al. The stranding anomaly as population indicator: the case of harbour porpoise Phocoena
340
phocoena in North-Western Europe. PLoS One 2013;8:e62180. [5]
and ecological factors related to trace element levels in harbour porpoises (Phocoena
342
phocoena) from European waters. Mar Environ Res 2007;64:247–266.
343 344
Lahaye V, Bustamante P, Law RJ, Learmonth JA, Santos MB, Boon JP, et al. Biological
AC C
341
EP
339
[6]
Mahfouz C, Henry F, Courcot L, Pezeril S, Bouveroux T, Dabin W, et al. Harbour
345
porpoises (Phocoena phocoena) stranded along the southern North Sea: an assessment
346
through metallic contamination. Environ Res 2014;133:266–273.
347
[7]
Siebert U, Wünschmann A, Weiss R, Frank H, Benke H, Frese K. Post-mortem findings in
ACCEPTED MANUSCRIPT 16 348
harbour porpoises (Phocoena phocoena) from the German North and Baltic Seas. J Comp
349
Pathol 2001;124:102–114.
350
[8]
Fontaine MC, Roland K, Calves I, Austerlitz F, Palstra FP, Tolley KA, et al. Postglacial climate changes and rise of three ecotypes of harbour porpoises, Phocoena phocoena, in
352
western Palearctic waters. Mol Ecol 2014;23:3306–3321.
353
[9]
RI PT
351
Van Bressem M-F, Duignan PJ, Banyard A, Barbieri M, Colegrove KM, De Guise S, et al. Cetacean morbillivirus: current knowledge and future directions. Viruses 2014;6:5145–
355
181. [10]
1993;22:524–528.
357 358
Aguilar A, Raga JA. The Striped Dolphin Epizootic in the Mediterranean Sea. Ambio
M AN U
356
SC
354
[11]
Raga J-A, Banyard A, Domingo M, Corteyn M, Van Bressem M-F, Fernández M, et al. Dolphin morbillivirus epizootic resurgence, Mediterranean Sea. Emerg Infect Dis
360
2008;14:471–473.
361
[12]
TE D
359
González-Barrientos R, Morales J-A, Hernández-Mora G, Barquero-Calvo E, GuzmánVerri C, Chaves-Olarte E, et al. Pathology of striped dolphins (Stenella coeruleoalba)
363
infected with Brucella ceti. J Comp Pathol 2010;142:347–52. [13]
coeruleoalba). J Wildl Dis 1985;21:443–446.
365 366 367 368
Gales N, Wallace G, Dickson J. Pulmonary cryptococcosis in a striped dolphin (Stenella
[14]
AC C
364
EP
362
Di Guardo G, Proietto U, Di Francesco CE, Marsilio F, Zaccaroni A, Scaravelli D, et al. Cerebral toxoplasmosis in striped dolphins (Stenella coeruleoalba) stranded along the Ligurian Sea coast of Italy. Vet Pathol 2010;47:245–253.
369
[15] Rehtanz M, Bossart GD, Doescher B, Rector A, Van Ranst M, Fair PA, et al. Bottlenose
370
dolphin (Tursiops truncatus) papillomaviruses: vaccine antigen candidates and screening
ACCEPTED MANUSCRIPT 17 test development. Vet Microbiol 2009;133:43–53.
371 372
[16] Chan OS, Mukherjee J, Yuen CS. Microbial flora of blowhole samples in captive bottlenose dolphins (Tursiops truncatus aduncas-type) in Hong Kong, 1993– 2000. Proc.
374
32nd Annu. Conf. Int. Assoc. Aquat. Anim. Med., Tampa, FL: 2001, p. 24–25.
375
[17]
RI PT
373
Buck JD, Wells RS, Rhinehart HL, Hansen LJ. Aerobic micro-organisms associated with freeranging dolphins in coastal Gulf of Mexico and Atlantic Ocean waters. J Wildl Dis
377
2006;42:536–544. [18]
culture in a simulated natural environment. Science 2002;296:1127–1129.
379 380
Kaeberlein T, Lewis K, Epstein SS. Isolating “uncultivable” microorganisms in pure
[19]
M AN U
378
SC
376
Brooks JP, Edwards DJ, Harwich Jr. MD, Rivera MC, Fettweis JM, Serrano MG, et al.
381
The truth about metagenomics: quantifying and counteracting bias in 16S rRNA studies.
382
BMC Microbiol 2015;15:66. [20]
Bik EM, Costello EK, Switzer AD, Callahan BJ, Holmes SP, Wells RS, et al. Marine
TE D
383 384
mammals harbor unique microbiotas shaped by and yet distinct from the sea. Nat
385
Commun 2016;7:10516.
evolution of the vertebrate gut microbiota. Nat Rev Microbiol 2008;6:776–788.
387 388
[22]
(Phocoena phocoena) from Portugal (western Iberian Peninsula). Environ Pollut
390
2016;210:293–302.
391
393
Ferreira M, Monteiro SS, Torres J, Oliveira I, Sequeira M, López A, et al. Biological variables and health status affecting inorganic element concentrations in harbour porpoises
389
392
Ley RE, Lozupone CA, Hamady M, Knight R, Gordon JI. Worlds within worlds:
EP
[21]
AC C
386
[23]
Monteiro SS, Torres J, Ferreira M, Marçalo A, Nicolau L, Vingada J V, et al. Ecological variables influencing trace element concentrations in bottlenose dolphins (Tursiops
ACCEPTED MANUSCRIPT 18 394
truncatus, Montagu 1821) stranded in continental Portugal. Sci Total Environ
395
2016;544:837–844.
396
[24]
Kuiken T, Hartmann MG. Proceedings of the First ECS Workshop on Cetacean Pathology : dissection techniques and tissue sampling, Leiden, Netherlands: [European
398
Cetacean Society]; 1991.
399
[25]
RI PT
397
Pugliares KR, Bogomolni A, Touhey KM, Herzig SM, Harry CT, Moore MJ. Marine mammal necropsy : an introductory guide for stranding responders and field biologists.
401
Woods Hole, MA: Woods Hole Oceanographic Institution; 2007. [26]
1991.
403 404
Lane DJ. 16S/23S rRNA sequencing. Chichester, United Kingdom: John Wiley and Sons;
M AN U
402
SC
400
[27]
Caporaso JG, Lauber CL, Walters WA, Berg-Lyons D, Huntley J, Fierer N, et al. Ultrahigh-throughput microbial community analysis on the Illumina HiSeq and MiSeq
406
platforms. ISME J 2012;6:1621–1624.
407
[28]
TE D
405
Kuczynski J, Stombaugh J, Walters WA, González A, Caporaso JG, Knight R. Using QIIME to analyze 16S rRNA gene sequences from microbial communities. Curr Protoc
409
Bioinformatics 2011;Chapter 10:Unit 10.7. [29]
2010;26:2460–2461.
411 412
[30]
Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 2011;27:2194–2200.
413 414
Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics
AC C
410
EP
408
[31]
Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, et al. The SILVA ribosomal
415
RNA gene database project: improved data processing and web-based tools. Nucleic
416
Acids Res 2013;41:D590–6.
ACCEPTED MANUSCRIPT 19 [32]
completeness rather than size. Ecology 2012;93:2533–2547.
418 419
[33]
Lozupone C, Knight R. UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol 2005;71:8228–8235.
420 421
Chao A, Jost L. Coverage-based rarefaction and extrapolation: standardizing samples by
[34]
RI PT
417
Oksanen J, Blanchet F, Kindt R, Legendre P, Minchin P, O’Hara R, et al. vegan: Community Ecology Package. R package version 2.3-4 2016. https://cran.r-
423
project.org/package=vegan. [35]
Wickham H. ggplot2 - Elegant Graphics for Data Analysis, Hadley Wickham, Springer. 1st ed. Springer-Verlag New York; 2009.
425
M AN U
424
SC
422
426
[36]
Wickham H. Reshaping Data with the reshape Package. J Stat Softw 2007;21:1–20.
427
[37]
Langille MGI, Zaneveld J, Caporaso JG, McDonald D, Knights D, Reyes JA, et al. Predictive functional profiling of microbial communities using 16S rRNA marker gene
429
sequences. Nat Biotechnol 2013;31:814–821.
430
[38]
TE D
428
Arndt D, Xia J, Liu Y, Zhou Y, Guo AC, Cruz JA, et al. METAGENassist: a comprehensive web server for comparative metagenomics. Nucleic Acids Res
432
2012;40:W88–95. [39]
uteri gen. nov., sp. nov., a new member of the family Pasteurellaceae Pohl (1979) 1981
434
isolated from a harbour porpoise (Phocoena phocoena). Int J Syst Evol Microbiol 2000;50
435
Pt 1:135–139.
436 437
Foster G, Ross HM, Malnick H, Willems A, Hutson RA, Reid RJ, et al. Phocoenobacter
AC C
433
EP
431
[40]
Lee C, Agidi S, Marion JW, Lee J. Arcobacter in Lake Erie beach waters: an emerging
438
gastrointestinal pathogen linked with human-associated fecal contamination. Appl
439
Environ Microbiol 2012;78:5511–5519.
ACCEPTED MANUSCRIPT 20 440
[41]
players in sulphidic habitats. Nat Rev Microbiol 2006;4:458–468.
441 442
Campbell BJ, Engel AS, Porter ML, Takai K. The versatile epsilon-proteobacteria: key
[42] Chow SK, Clarridge III JE. Identification and clinical significance of Helcococcus species, with description of Helcococcus seattlensis sp. nov. from a patient with urosepsis. J Clin
444
Microbiol 2014;52:854–858. [43]
from a human wound. Int J Syst Evol Microbiol 2004;54:1557–1560.
446 447
Collins MD, Falsen E, Brownlee K, Lawson PA. Helcococcus sueciensis sp. nov., isolated
[44]
SC
445
RI PT
443
Småge SB, Brevik ØJ, Duesund H, Ottem KF, Watanabe K, Nylund A. Tenacibaculum finnmarkense sp. nov., a fish pathogenic bacterium of the family Flavobacteriaceae
449
isolated from Atlantic salmon. Antonie Van Leeuwenhoek 2016;109:273–285.
450
[45]
M AN U
448
Stoddard RA, Atwill ER, Gulland FMD, Miller MA, Dabritz HA, Paradies DM, et al. Risk factors for infection with pathogenic and antimicrobial-resistant fecal bacteria in northern
452
elephant seals in California. Public Health Rep 2008;123:360–370. [46]
caused by HACEK microorganisms. Annu Rev Med 1997;48:25–33. [47]
1982;31:95–104.
456 457
[48]
Scaglione FE, Bollo E, Pregel P, Chiappino L, Sereno A, Mignone W, et al. Heart pathologies in dolphins stranded along the northwestern Italian coast. Dis Aquat Organ
458
2013;107:31–36.
459 460
Nord CE. Anaerobic bacteria in septicaemia and endocarditis. Scand J Infect Dis Suppl
EP
454 455
Das M, Badley AD, Cockerill FR, Steckelberg JM, Wilson WR. Infective endocarditis
AC C
453
TE D
451
[49]
Cools P, Haelters J, Lopes dos Santos Santiago G, Claeys G, Boelens J, Leroux-Roels I, et
461
al. Edwardsiella tarda sepsis in a live-stranded sperm whale (Physeter macrocephalus).
462
Vet Microbiol 2013;166:311-315.
ACCEPTED MANUSCRIPT 21 463
[50]
Foster G, Ross HM, Naylor RD, Collins MD, Ramos CP, Fernandez Garayzabal F, et al.
464
Cetobacterium ceti gen. nov., sp. nov., a new Gram-negative obligate anaerobe from sea
465
mammals. Lett Appl Microbiol 1995;21:202–206. [51]
Naim R, Yanagihara I, Iida T, Honda T. Vibrio parahaemolyticus thermostable direct
RI PT
466 467
hemolysin can induce an apoptotic cell death in Rat-1 cells from inside and outside of the
468
cells. FEMS Microbiol Lett 2001;195:237–244. [52]
Jaber JR, Pérez J, Rotstein D, Zafra R, Herráez P, Carrascosa C, et al. Biliary cirrhosis
SC
469
caused by Campula spp. in a dolphin and four porpoises. Dis Aquat Organ 2013;106:79–
471
84.
472
[53]
Allen-Vercoe E, Strauss J, Chadee K. Fusobacterium nucleatum: an emerging gut pathogen? Gut Microbes 2011;2:294–298.
473 474
M AN U
470
[54] Fishelson L, Montgomery L, Myrberg W, Arthur A. A unique symbiosis in the gut of tropical herbivorous surgeonfish (Acanthuridae: teleostei) from the red sea. Science
476
1985;229:49–51.
477
[55]
TE D
475
López A, Santos MB, Pierce GJ, González AF, Valeiras X, Guerra A. Trends in strandings and by-catch of marine mammals in north-west Spain during the 1990s. J Mar Biol Assoc
479
UK 2002;82:513–521. [56]
lesions in cetacean stranded on the northwestern Spanish Atlantic coast. Dis Aquat Organ
481
1998;32:227–231.
482 483 484 485
Abollo E, López A, Benavente P, Cabaleiro M, Pascual S. Macroparasites andgross
AC C
480
EP
478
[57]
Pascual S, Abollo E, López A. Elemental analysis of cetacean skull lesions associated with nematode infections. Dis Aquat Organ 2000;42:71–75.
ACCEPTED MANUSCRIPT 22 486 487
Table and Figure Legends
488
Table 1. Number of sequences and OTU estimates across body sites.
490
Fig. 1. Schematic representation of the procedure outline for assessing the microbiome
491
composition of stranded cetaceans.
492
Fig. 2. Measures of microbial diversity of the eight body sites. Panel A depicts a beta
493
diversity plot based on UniFrac distances calculated on the OTU table of the eight body sites.
494
Panel B shows alpha diversity rarefaction curves for Chao1 richness estimates for each body site.
495
Fig. 3. Taxonomic distribution of the OTUs at the phylum level (panel A) and genus level
496
(panel B) for each body site. Panel A: A total of 15 phyla were detected; the five most abundant
497
(>5%) were Proteobacteria, Fusobacteria, Firmicutes, Bacteroidetes and Tenericutes. The
498
remaining phyla had a relative abundance below 5%. Panel B: A total of 108 genera were
499
detected, but some had only residual abundance (<0.5%).
500
Fig. 4. Genus level heat map depicting significantly different OTUs according to body site.
501
The 51 OTUs were selected after a statistical G-test comparing OTU frequencies in the sample
502
groups. Only those taxa that showed p<0.05 were selected. The dendrograms were made through
503
hierarchical clustering using the ggdendro and ggplot2 packages in R. Percentage of identity (ID)
504
of the best-hit matches of the respective representative sequences, using BLASTN analysis, are
505
indicated next to the assigned taxonomy.
506
AC C
EP
TE D
M AN U
SC
RI PT
489
ACCEPTED MANUSCRIPT Table 1. Number of sequences and OTU estimates across body sites
Number
of Total number Number
sequences
of OTUs per unique OTUs sample
per sample
43,896
122
32
Colon
39,579
75
1
Intestine
74,465
54
0
Stomach
62,886
85
3
Duodenum
57,705
70
0
Tongue
47,249
149
1
Oral
289,253
156
14
Genitals
24,083
95
0
Total
639,116
235(unique)
51
M AN U
TE D EP
SC
Blowhole
AC C
of
RI PT
Sample Type
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT