Subscriber access provided by University of Newcastle, Australia
Article
Non-invasive respiratory metabolite analysis associated with clinical disease in cetaceans: a Deepwater Horizon oil spill study Alberto Pasamontes, Alexander A Aksenov, Michael Schivo, Teri Rowles, Cynthia R. Smith, Lori H. Schwacke, Randall S. Wells, Laura Yeates, Stephanie Venn-Watson, and Cristina E. Davis Environ. Sci. Technol., Just Accepted Manuscript • DOI: 10.1021/acs.est.6b06482 • Publication Date (Web): 13 Apr 2017 Downloaded from http://pubs.acs.org on April 16, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 28
Environmental Science & Technology
1
Non-invasive respiratory metabolite analysis associated with clinical
2
disease in cetaceans: a Deepwater Horizon oil spill study
3 4
Alberto Pasamontes1, Alexander A. Aksenov1, Michael Schivo2,3, Teri Rowles4, Cynthia R. Smith5, Lori
5
H. Schwacke6, Randall S. Wells7, Laura Yeates5, Stephanie Venn-Watson5, Cristina E. Davis1,*
6 7
1 Mechanical and Aerospace Engineering, One Shields Avenue, University of California, Davis, CA 95616, USA
8
2 Department of Internal Medicine, 4150 V Street, Suite 3400, University of California, Sacramento, CA 95817,
9
USA
10
3 Center for Comparative Respiratory Biology and Medicine, University of California, Davis, CA 95616, USA
11
4 Office of Protected Resources, National Marine Fisheries Service, National Oceanic and Atmospheric
12
Administration, 1315 East West Highway, Silver Spring, MD 20910, USA
13
5 National Marine Mammal Foundation, 2240 Shelter Island Drive, Suite 200, San Diego, CA 92106, USA
14
6 National Centers for Coastal Ocean Science, National Oceanic and Atmospheric Administration, 331 Fort Johnson
15
Road, Charleston, SC 29412, USA
16
7 Chicago Zoological Society’s Sarasota Dolphin Research Program, c/o Mote Marine Laboratory, 1600 Ken
17
Thompson Parkway, Sarasota, Florida 34236, USA
18
*Correspondence: Mechanical and Aerospace Engineering, One Shields Avenue, University of California, Davis,
19
CA 95616, USA, +1 (530) 754-9004,
[email protected]
20
21
22
23 1 ACS Paragon Plus Environment
Environmental Science & Technology
24
ABSTRACT
25
Health assessments of wild cetaceans can be challenging due to the difficulty of gaining
26
access to conventional diagnostic matrices of blood, serum and others. While the non-invasive
27
detection of metabolites in exhaled breath could potentially help to address this problem, there
28
exists a knowledge gap regarding associations between known disease states and breath
29
metabolite profiles in cetaceans. This technology was applied to the largest marine oil spill in
30
U.S. history (The 2010 Deepwater Horizon oil spill in the Gulf of Mexico). An accurate analysis
31
was performed to test for associations between the exhaled breath metabolome and sonographic
32
lung abnormalities as well as hematological, serum biochemical, and endocrine hormone
33
parameters. Importantly, metabolites consistent with chronic inflammation, such as products of
34
lung epithelial cellular breakdown and arachidonic acid cascade metabolites were associated
35
with sonographic evidence of lung consolidation. Exhaled breath condensate (EBC) metabolite
36
profiles also correlated with serum hormone concentrations (cortisol and aldosterone),
37
hepatobiliary enzyme levels, white blood cell counts, and iron homeostasis. The correlations
38
among breath metabolites and conventional health measures suggest potential application of
39
breath sampling for remotely assessing health of wild cetaceans. This methodology may hold
40
promise for large cetaceans in the wild for which routine collection of blood and respiratory
41
anomalies are not currently feasible.
42
43
Keywords: Breath analysis, metabolomics, non-invasive method, cetaceans
44
45
46
47
2 ACS Paragon Plus Environment
Page 2 of 28
Page 3 of 28
Environmental Science & Technology
48
49
INTRODUCTION
50
Health assessments of common bottlenose dolphins (Tursiops truncatus) via temporary
51
capture and release have been valuable for understanding disease in wild populations, as well as
52
identifying underlying causes and risk factors for disease. Unfortunately, routine health
53
monitoring of cetaceans, especially large species, in their natural habitat can be challenging since
54
comprehensive health evaluations require samples, such as blood, urine, or ultrasound images,
55
that are difficult or impossible to obtain without physically handling the animal. One potential
56
way to address this issue is through the use of exhaled breath analysis for cetacean health
57
assessment. Breath analysis metabolomics is a very attractive tool due to the fact that the
58
necessary samples can be collected non-invasively and potentially remotely without having to
59
physically handle the animal1, 2.
60
In order to identify breath-based metabolic profiles that are associated with specific
61
disease states in cetaceans, there is a need to pair exhaled breath metabolite data with confirmed
62
clinical disease using standard diagnostic tools. Additionally, there is a need to use sophisticated
63
state-of-the-art technologies and analytical approaches, such as mass spectrometry3, 4. Deviations
64
in breath metabolic content due to any factors such as diseases and exposures can be potentially
65
evaluated by comparing against “normal” baseline metabolite profile from healthy animals5 or a
66
list of target biomarkers previously identified.
67
The Deepwater Horizon (DWH) oil spill began in April 2010 after the explosion of the
68
oil rig in the Macondo Prospect6, 7. As part of the DWH Natural Resource Damage Assessment
69
(NRDA) a number of studies were conducted to assess the impacts to populations of cetaceans8,
70
9
, including bay, sound and estuary (BSE) common bottlenose dolphins10-12. The studies
71
evaluated the health of bottlenose dolphins in two areas of the Northern Gulf of Mexico coast
72
that were heavily oiled (Barataria Bay and Mississippi Sound), as well as in a reference site
73
(Sarasota Bay, FL) that did not receive oiling from the DWH spill
74
dolphins that stranded and died within the oil spill footprint10. These studies have revealed that
75
dolphin populations within the oil spill footprint were severely affected by the spill, including
3 ACS Paragon Plus Environment
13
, as well as evaluation of
Environmental Science & Technology
76
increased presence of adrenal and lung disease, higher mortality and lower survival rates,
77
perinatal losses, and poor reproductive success11,
78
opportunity to evaluate affected animals and lay the ground work in determining specific breath
79
biomarkers associated with health conditions in dolphins affected by the oil spill. The goals of
80
the present study were to identify specific 1) breath-based metabolite profiles associated with a
81
dolphin population with adverse health effects following an oil spill compared to reference
82
populations, and 2) changes in breath metabolites associated with lung disease or clinical blood
83
values. By doing so, we aimed to demonstrate how the method of breath analysis could serve as
84
a valuable tool in wildlife conservation efforts and deepen our understanding of marine mammal
85
biology and physiology.
14-16
. The severe impacts have presented an
86 87
MATERIAL AND METHODS
88
Exhaled Breath Condensate (EBC) Sampling.
89
An exhaled breath condensate (EBC) collection device was specifically adapted for
90
marine mammal anatomy and physiology and was previously described4. Briefly, this device
91
consists of a glass tube with dry ice packed around it and a soft mask on the bottom which is
92
placed above the blowhole. Exhaled air is expelled through the chilled glass tube out of the
93
exhaust to the surrounding air.
94
EBC Sample Collection from Wild Dolphins.
95
Exhaled breath condensate from wild common bottlenose dolphins was collected during
96
capture and release exercises conducted in Sarasota Bay (SB), Florida USA in May 2011 and
97
May 2012 as a part of periodic health assessment studies, and in Barataria Bay (BB), Louisiana
98
USA during 3-16 August 2011 as part of a DWH NRDA study. The dolphin health assessment
99
study in SB was led by the Chicago Zoological Society and conducted under National Marine
100
Fisheries Service Scientific Research Permit No. 15543, and the study in BB was led by NOAA
101
and conducted under National Marine Fisheries Service Scientific Research Permit No. 932-
102
1905/MA-009526. Protocols were reviewed and approved by NOAA Institutional Animal Care
103
and Use Committees for Barataria Bay. In total, 21 breath samples were collected from dolphins
104
in SB and 19 breath samples were collected from dolphins in Barataria Bay. 4 ACS Paragon Plus Environment
Page 4 of 28
Page 5 of 28
Environmental Science & Technology
105
The catch and release procedure was as follows. The free swimming animals were
106
scouted and selected, and then restricted by deploying a seine net. Captured individuals were
107
restrained by handlers and, if necessary, disentangled from the net. Following the capture, the
108
animals were transferred onto a padded, shaded deck of a special research boat and a several
109
biological samples were collected within span of approximately one hour10, including exhaled
110
breath condensate. To procure a breath sample, the collection device was positioned above the
111
blowhole without immediate contact and 10 full breaths were captured over an approximately 5
112
min time frame. Approximately 200-500 µL of EBC were collected each time, depending on the
113
size and behavior of the animal during collection. The frozen EBC samples were extruded from
114
the sampling device using PTFE plunger, placed into a borosilicate vial, capped and immediately
115
placed on dry ice and then in a -80 °C freezer at until further metabolite biochemical analysis.
116
For each consecutive collection the device and plunger were thoroughly cleaned with 70%
117
ethanol and allowed to air dry.
118 119
EBC Sample Collection from Managed Animals under Human Care.
120
EBC sample collection was carried out from 10 managed animals in the U.S. Navy
121
Marine Mammal Program in San Diego, CA using previously published methods4. In total, 21
122
samples were utilized for analysis, as described below. The collection was conducted on animals
123
that breathed normally while positioned by trainers at the water surface. Approximately 10-20
124
breaths were collected over a 5 min time frame, depending on the animal size and behavior. The
125
samples were normalized by its weight. Typical EBC volumes were up to 500 µL of condensate.
126
The frozen EBC samples were extruded from the sampling device with PTFE plunger, placed
127
into a borosilicate vial, capped and immediately stored on dry ice and then in a -80 °C freezer
128
until further analysis. For each consecutive collection the device and plunger were thoroughly
129
cleaned with 70% ethanol and allowed to air dry.
130 131
Liquid Chromatography/Mass Spectrometry (LC/MS) Analysis of Non-volatile EBC
132
Metabolites.
133
Samples from both wild and managed animals were analyzed using the same analytical
134
protocols for metabolomic profiling that have been published in detail earlier4. Briefly, the 0.1 5 ACS Paragon Plus Environment
Environmental Science & Technology
135
mL aliquot of EBC was lyophilized and then redissolved in 0.1 mL of 90% acetonitrile in water.
136
Five µL of resuspended sample was then used for LC/MS analysis. The CUDA (12-
137
[[(cyclohexylamino)-carbonyl]amino]dodecanoic acid) in methanol/toluene, 9:1 v/v internal
138
standard, was added to each sample and used for quality control and to assess reproducibility.
139
Analysis was performed on an Agilent 1200 Series high-performance liquid chromatography
140
(HPLC)/6530 accurate-mass Q-TOF mass spectrometer system (Agilent Technologies, Santa
141
Clara, CA)4 on a Kinetex 2.6 µm (HILIC) 100 Å HPLC column (150 mm x 2.10 mm)
142
(Phenomenex, Torrance, CA), held at 40 ºC during analysis. The mass spectrometer was
143
equipped with an Agilent Jet Stream ESI source.
144
acetonitrile in water, 9:1 v/v (mobile phase B) to 45% over 15 min water (mobile phase A) at
145
0.35 mL/min. Ammonium acetate and acetic acid were added to pH 5 to each mobile phase. The
146
MS analysis was performed in both positive and negative ion modes; the mass range was set to
147
60-1700 Thomson (m/z); scan rate 4 spectra/sec, sample analysis time was 21 min per sample.
148
Fragmentor voltage was 120 V, sheath gas flow was 11 L/min, and sheath gas temp was 350 ºC.
149
For the reverse phase (RP) analysis, the 100 µL sample aliquot was dried down and re-suspended
150
in 100 µL 9:1 methanol: toluene, 3 µL of sample was then injected. The data-dependent MS/MS
151
analysis was performed on a pooled sample as described earlier17. For both analyses modes compound
152
annotations were performed by comparing sample MS/MS spectra to the METLIN library18.
153
Blood Diagnostics of Barataria Bay and Sarasota Bay Dolphins and Assessment of
154
Respiratory Anomalies.
The gradient was changed from 100%
155
The majority of the health diagnostic assays has been conducted routinely in previous
156
dolphin health assessment studies and were previously reported for the wild animal cohorts in
157
this study10, 19. Blood samples were sent to the Animal Health Diagnostic Center at Cornell
158
University College of Veterinary Medicine (Ithaca, NY) for hematology, serum chemistry, and
159
evaluation of endocrine function using standard methods19. The assessment of respiratory
160
anomalies was conducted as described previously using pulmonary ultrasound10.
161
Statistical Analysis.
162
The LC/MS raw data files were first processed with Mass Hunter Qualitative Analysis
163
B.05.00SP1 software in order to deconvolve each peak (Agilent; Santa Clara, CA). The peaks 6 ACS Paragon Plus Environment
Page 6 of 28
Page 7 of 28
Environmental Science & Technology
164
were aligned using Mass Profiler Professional 12.1 software. Data from both the managed
165
population (San Diego data set) and the wild capture/releases (Barataria and Sarasota data set)
166
were treated as a single data set for deconvolution and alignment to ensure the parameters used
167
were consistent across the study. Principal Component Analysis (PCA) was applied for
168
exploratory analysis and to study the internal variance of the data4. Metabolites present in less
169
than 10% of the samples were removed, assuming they were not sufficiently reproducible and
170
therefore of lesser diagnostic value.
171
Partial least-squares (PLS)20 regression is a multivariate latent-variable method that
172
relates one dependent variable (y) to a set of independent variables, X. Partial least-squares
173
discriminant analysis (PLSDA) is the application of PLS to the classification of problems in
174
which (y) is a vector that codifies the class of each sample. In our case, we were interested in
175
using metabolite abundances to predict the sampling location of each data set (Sarasota Bay =
176
SB; Barataria Bay = BB; San Diego = SD). Cross-validation where 66% of the samples were
177
used as a training group and 34% as a test group was then carried out. Quality assessment of a
178
model was performed by computing the sensitivity (samples of the class of interest correctly
179
assigned to its class) and the specificity (samples not belonging to the class of interest correctly
180
not assigned to that class) of the predicted class assignments4. The global accuracy of the model
181
is reported as the sum of the specificity and sensitivity divided by two21, 22. PLSDA also reports
182
information about which features makes the difference between groups; the chemicals were
183
ranked by the Variable Importance in Projection (VIP) score. VIP scores estimate the importance
184
of each variable in the PLSDA model. Variables with VIP scores higher than 0.8 were
185
considered important.
186
metabolites, a “70-20% filter23” was applied: peaks which were present between 70% in one
187
group and 20% of samples in another group were removed. Furthermore, peaks that were present
188
in less than 20% of samples in both groups were also removed.
Before applying PLSDA, in order to remove low reproducibility
189
Where available, the identities of metabolites were proposed based on MS/MS
190
fragmentation patterns. If MS/MS was not available, the compounds of interest have been
191
tentatively identified based on exact mass and were annotated by comparing either MS/MS
192
spectra or exact mass values to the METLIN library entries. The exact mass search was carried
193
out with 20 ppm window with protonation, ammoniation and sodiation as possible charge 7 ACS Paragon Plus Environment
Environmental Science & Technology
194
acquisition modes. The list of potential candidates was then manually reviewed and the most
195
likely candidate matches were selected. For example, protonated or ammoniated species were
196
given priority, while non-biogenic compounds or compounds that would not be expected to
197
occur in a marine environment have been excluded. In cases when no appropriate match was
198
found or a large number of plausible matches existed, i.e. more than 3 plausible disparate
199
structures or compound classes match the MS/MS and/or exact formula, the compound was
200
presumed unidentified. In many cases, where a number of possible isomers exists (e.g. for
201
lipids), although the exact structure could not be determined, the class of the molecular species
202
was postulated. All compound identifications are putative at the level 3-4 of the Metabolomics
203
Standards Initiative24.
204
Correlation of EBC metabolites with blood analysis parameters was carried out for BB
205
and SB dolphins. In order to fit a linear model between the predictors (EBC metabolites) and the
206
responses (blood analysis parameters), stepwise regression25 using forward selection was applied
207
to select the top 5 most important predictors for each response using the minimum BIC with K-
208
fold cross-validation (K=5). All calculations were performed using the PLS_Toolbox v. 7.9
209
(Eigenvector Research, Manson, WA, USA) in MATLAB v. 7.8 (The Mathworks AS,
210
Massachusetts, USA) and JMP10 (SAS Institute Inc., USA).
211
We next examined possible correlation between the EBC metabolite content and
212
pulmonary abnormalities identified by ultrasound examination including pulmonary nodules,
213
consolidation and alveolar-interstitial syndrome (AIS). In order to find potential clinically-
214
relevant biomarkers, a stepwise linear regression using forward selection was also applied to
215
each pulmonary abnormality.
216 217
RESULTS
218
Comprehensive Exhaled Breath Condensate (EBC) Metabolite Profiling.
219
We collected and analyzed exhaled breath condensate from a managed collection San
220
Diego (SD) and wild (Sarasota Bay (SB) and Barataria Bay (BB)) bottlenose dolphin
221
populations. Principal component analysis (PCA) was initially applied to the LC/MS data for SB
222
and BB animals (Figure S1). Partial least square discriminant analysis (PLSDA) was applied and 8 ACS Paragon Plus Environment
Page 8 of 28
Page 9 of 28
Environmental Science & Technology
223
validated using cross-validation, and the global accuracy of the PLSDA model was found to be
224
95%. Table S1 lists the most important metabolite which allow discrimination between these
225
groups (SB and BB dolphins) and corresponding tentative chemical metabolite identification.
226
This table includes the unique metabolites present in only one of the groups, as well as
227
statistically different common metabolites present in both groups, but at different abundance
228
levels.
229
Next, it was important to examine possible external confounding factors unrelated to
230
health conditions that could artificially influence any observed differences between groups,
231
including geographic location. To do this, we added a second healthy control group of managed
232
collection animals from San Diego Bay, CA in the analysis (Figure 1). The PCA-score plot
233
shows the SD dolphins cluster with the SB control population, indicating a similarity in the EBC
234
samples and total metabolite profile between these two healthy reference groups. Compared to
235
Figure 1S, the separation between SB and BB increased due to we are emphasizing the variance
236
related to the healthy condition and decreasing the impact of confounding factors. PLSDA was
237
also applied to the data comparing BB animals versus the combined SB and SD animals.
238
Compared to this combined group, BB dolphins had a number of unique metabolites, and the
239
accuracy of the model in this case was 85%. Table 1 lists the most important EBC metabolite
240
that discriminate BB dolphins from the two healthy reference groups and the corresponding
241
tentative chemical identifications.
242
EBC Metabolite Correlation with Pulmonary Radiographic Analysis.
243
We have examined possible correlation between the EBC metabolite content (BB
244
dolphins) and pulmonary abnormalities including pulmonary nodules, consolidation, and
245
alveolar-interstitial syndrome (AIS). No significant differences (p < 0.05) were observed for
246
pulmonary nodules or AIS. However, there was a distinct correlation between dolphins with
247
pulmonary consolidation (n=6) and specific breath-based metabolites. No pulmonary
248
consolidation was detected in Sarasota Bay dolphins. The list of breath chemicals that correlate
249
with pulmonary consolidation are considered putative biomarkers, and the corresponding
250
tentative identification of these potential biomarkers are provided (Table 2).
251
EBC Metabolite Correlation to Common Blood Measures. 9 ACS Paragon Plus Environment
Environmental Science & Technology
Page 10 of 28
252
In order to elucidate whether specific blood parameters correlate to the exhaled
253
metabolites, the first step was to apply PLSDA to obtain the most important blood parameter to
254
discriminate the SB and BB groups. The VIP versus coefficients of the PSLDA model plot is
255
shown (Figure S2). This plot identifies blood parameters that are influential relative to the fit for
256
the response. For example, we saw that cortisol and glucose have both VIP values that exceed
257
0.8 and relatively large coefficients. This finding is consistent with that reported in Schwacke et
258
al.10
259
The second step was to correlate the blood variables with EBC metabolites. To do so,
260
stepwise regression was applied to find the most important EBC features. As an example, results
261
for lymphocyte counts and cortisol are shown (Figure S3). To assess the quality of the results,
262
the coefficient of regression and RMSE (Root Mean Squared Error) were reported. As can be
263
seen, low values of cortisol and high values of lymphocytes are specific to BB dolphins. The
264
remaining blood variables along with top five EBC metabolites that correlate with each variable
265
are reported (Table S2).
266 267
DISCUSSION
268
Exhaled breath condensate (EBC) analysis.
269
We observed a distinctly different breath-based metabolome in the population within
270
Barataria Bay, Louisiana compared to reference populations in Sarasota Bay, Florida, and San
271
Diego, California. Importantly, we found putative breath-based biomarkers in Barataria Bay
272
dolphins that correlated with consolidations in lungs and multiple blood changes, including
273
lower blood glucose and cortisol concentrations consistent with the hypoadrenocorticism seen
274
previously observed10. Although the identification of a distinct breath-based metabolome in
275
Barataria Bay dolphins does not confirm that the origin of all metabolite differences were
276
directly related to petroleum product exposure, the findings of multiple unique compound classes
277
in dolphins exposed to petroleum products paired with breath-based metabolite profiles
278
associated with identified diseases caused by the oil spill provides a compelling case.
10 ACS Paragon Plus Environment
Page 11 of 28
Environmental Science & Technology
279
Extensive, multi-year and multidisciplinary investigation of live and dead bottlenose
280
dolphins in the northern Gulf of Mexico, especially those in heavily-oiled Barataria Bay,
281
Louisiana, demonstrated that exposure to the spill’s petroleum products caused chronic lung and
282
adrenal disease which contributed to increased mortality, lower survival, and poor reproductive
283
success14,
284
bacterial pneumonia and lung consolidation, possibly secondary to lung injury, and
285
hypoadrenocorticism evidenced by low blood cortisol, aldosterone, glucose, and thin adrenal
286
cortices. The results of this present study that are outlined in the previous sections corroborate
287
these findings. Data for the comprehensive health assessments then allowed us to document
288
specific breath-borne biomarkers that correlate with specific blood parameters as well as
289
specified health conditions - most importantly lung consolidations.
16
. Specifically, Barataria Bay post-spill dolphins had an increased prevalence of
290 291
EBC and Lung Consolidation.
292
Several interesting EBC metabolites were associated with lung consolidation identified
293
by pulmonary ultrasound of BB dolphins. Lung consolidation on ultrasound can be an indicator
294
of pneumonia in animals, including bottlenose dolphins and humans10, and can be used to help
295
identify bacterial, viral, or fungal infections in the lung. Tentative metabolites associated with
296
lung consolidation in BB dolphins are listed (Table 2). In several cases, postulated EBC
297
molecular structures associated with lung consolidation in dolphins contain a glycan moiety (e.g.
298
Ganglioside GT3 or Ganglioside GM2). These compounds tend to possess complex and difficult-
299
to-characterize isomers, which often defines their biological functions. Therefore, instead of
300
focusing on specific compounds, we considered general classes of identified compounds.
301
Encouragingly, a number of compounds found in dolphin EBC would be expected to be
302
observed given the presence of lung disease. In particular, we found evidence of cellular bilayer
303
degradation (phospholipid moieties) that indicates nonspecific lung epithelial breakdown in
304
lungs of animals with consolidation. Several of the tentatively proposed metabolites may be
305
related to DNA and cellular damage processes, but unambiguous chemical identification is
306
required to elucidate this possibility further. Some of the compounds were not expected in the
307
context of the study, and chemical identification of these compounds may be of lower certainty. 11 ACS Paragon Plus Environment
Environmental Science & Technology
Page 12 of 28
308
One such group of compounds includes glycosides, e.g., steroidal glycosides. Many of these
309
compounds are usually of plant origin, although they may also occur in the marine environment.
310
Also, we have tentatively identified several anti-infective agents that also often naturally occur in
311
plants (e.g. alkaloids), although these may originate from marine algae11,
312
bacterial metabolites. These compounds include, among other metabolites, leptomycin B,
313
norselic acid A, and yiamoloside. Leptomycin B, a secondary metabolites of the Streptomyces
314
spp actinobacteria, is an anti-fungal agent that also possesses anti-tumor activities10, 11, 14-16. The
315
Streptomyces can be found in soil and rotting vegetation that are resent in animals’ habitat in
316
coastal estuaries.
317
refractory to standard chemotherapy, and it appears to result in G1 cell cycle arrest. Norselic acid
318
A derives from the Crella sponge species, functions as an anti-infective steroid that displays
319
antibiotic activity against methicillin-resistant Staphylococcus aureus (MRSA), methicillin-
320
sensitive S. aureus (MSSA), vancomycin-resistant Enterococcus faecium (VRE), and Candida
321
albicans10. Yiamoloside B is a fungistatic chemical belonging to a class of saponin compounds
322
released from Phytolacca octandra10. At the present time, the origin and role of the detected
323
metabolites are putative. Although these compounds appear to be meaningful in the present
324
biological context, no conclusions should be drawn about biological role of these metabolites
325
until structural confirmation. However, when confirmed and validated, such compounds may
326
serve as non-invasive biomarkers of lung consolidation.
16
or as secondary
Leptomycin B has been studied in human lung cancer cells which are
327 328
EBC and Blood Analysis
329
The next phase of the analysis sought to correlate breath metabolites to blood parameters
330
in dolphins. Blood-based biomarkers are essential to understanding the health of dolphins in
331
managed populations, including detection of inflammatory processes and infectious diseases,
332
anemia and iron abnormalities, renal and liver diseases and more. Some of the metabolites in
333
breath are partitioned from blood at the alveoli and thus may directly correlate with blood
334
composition; this may be especially true of cetaceans, which have a double layer of capillaries in
335
the lung and exhale a large volume of deep lung air with each breath26. In humans and other
336
terrestrial animals, there is some evidence of breath-serum correlation in glucose metabolism27,
337
emphysema exacerbations28, and lung cancer29. These investigations are built on the premise that 12 ACS Paragon Plus Environment
Page 13 of 28
Environmental Science & Technology
338
systemic inflammation may be recognized in the lungs (and eventually breath) given the
339
extensive capillary-alveolar interface network.
340
correlated with breath biomarkers to varying degrees. Here, we restrict our discussion to those
341
correlations which are robust and show a clear separation between the SB and BB dolphins based
342
on the serum biomarkers.
In this study, several serum biomarkers
343 344
EBC and Hypoadrenocorticsm.
345
It has been previously demonstrated that cortisol levels were lower in Barataria Bay
346
dolphins compared to Sarasota Bay dolphins. Lower cortisol levels in BB dolphins, paired with
347
lower glucose, lower aldosterone, and thin adrenal gland cortices previously reported in this
348
population, support an impaired hypothalamus-pituitary-adrenal axis caused by oil spill
349
exposure. In the healthy SB dolphins, five breath metabolites (Table S2) correlated with higher
350
cortisol levels (R2=0.81) (Figure S3). Elevated cortisol levels in healthy dolphins during catch-
351
and-release protocols represent a normal response. Since we did not see consistent breath
352
metabolites in the BB dolphins that correlated with their serum markers of hypoadrenocorticism,
353
we speculate that the breath metabolites seen in the SB dolphins are markers of health rather than
354
disease. It is possible that the BB dolphins were not able to produce the breath metabolites that
355
indicate elevated, appropriate adrenal activity. This conjecture needs further evaluation. The
356
breath metabolites that correlated with the normally-elevated cortisol levels in the SB dolphins
357
include cellular membrane products and fatty acid hydrolyzing agents, although the exact
358
identification and possible explanations for roles of these compounds are limited at the present
359
time.
360
Next, five breath metabolites were found to correlate with glucose levels in the SB
361
dolphins (R2=0.79), and overall the SB dolphins had higher glucose levels compared to the BB
362
dolphins. The elevated glucose levels in SB dolphins appear to be normal in healthy animals30
363
and may represent a non-pathologic, potentially necessary response to feeding31. The tentatively
364
identified breath metabolites that correlate with elevated glucose levels include triglycerides and
365
Vitamin D2 3-glucuronide (Table S2). As with the cortisol levels described above, the lack of
366
correlation between breath metabolites and glucose levels in the BB dolphins may reflect a 13 ACS Paragon Plus Environment
Environmental Science & Technology
Page 14 of 28
367
disease state. Glucose levels in the BB dolphins were pathologically low compared to their SB
368
counterparts, possibly a marker of a poor glycemic response to feeding and/or stress. The
369
absence of breath metabolites likely indicates that the dolphins have abnormal glucose
370
metabolism which may be a part of a depressed hypothalamus-pituitary-adrenal axis. Overall,
371
glucose and cortisol levels seen in healthy dolphins correlate with several breath metabolites.
372
Further investigations are required to conclusively correlate the chemical identity of these
373
metabolites to specific physiological processes that cause such correlations. For example, for
374
blood glucose correlations, compounds that contain glycoside moiety were found (e.g. vitamin
375
D2 3-glucuronide), but the biological basis at this point is not elucidated. The fact that the BB
376
dolphins did not have correlative breath metabolites for glucose and cortisol levels also supports
377
impaired glucose and cortisol regulation. At the same time, though it is compelling to place
378
importance on one or two specific breath biomarkers, one must remember that the correlation
379
between any blood biomarker with breath metabolites must include the entire group of breath
380
metabolites.
381 382
EBC, inflammation, renal function, and iron metabolism.
383
Four additional serum biomarkers that differentiate BB from SB dolphins and correlate
384
with breath biomarkers in the BB group include: lymphocytes, blood urea nitrogen (BUN),
385
potassium levels, and transferrin saturation. Lymphocytes are white blood cells that generally
386
indicate a chronic infection or inflammation, as might be seen with a prolonged bacterial, viral,
387
or fungal pneumonia. Most acute infections engender an elevated peripheral segmented
388
neutrophil count, though this elevated neutrophil count may segue into a lymphocytosis given
389
enough time. Lymphocyte count is statistically higher overall in the BB group as shown Figure
390
S3b and this is likely a result of chronic inflammation and/or infection.
391
Last, levels of saturated transferrin correlated with several breath biomarkers in BB
392
dolphins (R2=0.81). Transferrin saturation is a biomarker of iron metabolism, and higher
393
saturations indicate iron overload. It has been described that bottlenose dolphins can develop iron
394
overload paired with insulin resistance, and that dolphins who eat fish with increased
395
heptadecanoic acid levels may have a reversal of the markers of iron overload and insulin 14 ACS Paragon Plus Environment
Page 15 of 28
Environmental Science & Technology
396
resistance32. The exact dietary differences between SB and BB dolphins are not known, though
397
one would expect potential changes in prey to be accentuated by a large oil spill. It is possible
398
that the BB group did not have access to enough fish with adequate heptadecanoic acid levels to
399
effectively reduce their transferrin saturation levels compared to the SB group. However, the
400
clinical consequence of this presently remains unclear.
401 402
EBC Metabolites in Barataria Bay versus Healthy Reference Dolphin Populations.
403
Some of the more striking findings in BB dolphin breath are the tentatively identified
404
phospholipid moieties. Phosphatidic acid, phosphatidylethanolamine, and steroids were all found
405
in either higher abundance or uniquely in BB dolphins (Table 1S). The specific isomers of these
406
molecular species were not distinguished. In general, these lipid products suggest cellular
407
destruction including remnants of lipid bilayers, presumably from disintegrated respiratory
408
epithelial cells. This is an expected finding as petroleum products are toxic to mammalian
409
lungs33. However, phospatidylglycerol was also identified which comes from the breakdown of
410
lung surfactant, a phospholipid-protein mixture that confers both mechanical and immunologic
411
protection in animals34 and humans35. The finding of surfactant breakdown products in exhaled
412
breath suggests that aspirated contents, such as contaminated water, or inhalation of airborne
413
polycyclic aromatic hydrocarbons (PAHs) at the water surface, may have some role in
414
disintegrating surfactant. Or, as with lung infections or aspiration in humans, a clinical syndrome
415
similar to the human acute respiratory distress syndrome (a hyper-inflammatory syndrome of the
416
lungs marked by a loss of surfactant, alveolar filling with exudative fluid, dyspnea, and
417
hypoxemia) may have developed36.
418
When comparing BB to SB dolphins, we tentatively identified compounds that are likely
419
to be naturally-occurring macrolide-type antibiotics, arachidonic acid metabolites, and products
420
of lipid breakdown (Table S1). We tentatively identified the following macrolide antibiotics in
421
the BB and not in the SB dolphins’ breath: chivosazole E, leukomycin A1 and A7, lansonolide
422
A, jadomycin B, and mycolactone. Structural isomers of these compounds are also possible.
423
The identification of leukotriene E3 in the breath of Barataria Bay dolphins is particularly
424
notable. This compound is a metabolite of arachidonic acid and is known to be a stable marker of 15 ACS Paragon Plus Environment
Environmental Science & Technology
Page 16 of 28
425
asthmatic inflammation found in biologic fluids including breath condensate37. Thus, leukotriene
426
E3 may be a breath-based indicator of lung disease in Barataria Bay dolphins.
427
In general, antibiotics identified in BB dolphins are most likely metabolic products of
428
bacteria and fungi, and their exclusive presence in the BB group (and not in the SB group)
429
suggests high microbial presence within animals’ respiratory system and/or in the Barataria Bay
430
waters. This hypothesis would be consistent with the high prevalence of Barataria Bay dolphins
431
that stranded and died with bacterial pneumonia38. Chivosazole E is produced by Sorangium
432
cellulosum, and possesses antifungal and cytotoxic properties10,
433
mammalian cells40. Leukomycin A1 and A7 belong to a group of macrolide antibiotics produced
434
from Streptomyces kitasatoensis41 with a wide range of antibiotic, and some anti-neoplastic,
435
properties. Lansonolide is produced from the Caribbean sponge, Forcepia sp.42, and possesses
436
defined anti-neoplastic activity against lung cancer cells43. Jadomycin B is produced by
437
Streptomyces venezuelae, and it has a wide range of antimicrobial activity including methicillin-
438
resistant Staphylococcus aureus11. One of the tentatively identified compounds, mycolactone is a
439
potent macrolide antibiotic which is produced by mycobacteria, including Mycobacterium
440
marinum and M. ulcerans44. M. marinum is a bacterium with a global presence in fresh and
441
saltwater 45. M. ulcerans is also associated with water, although sources of M. ulcerans appear to
442
be predominantly fresh water with a potential vector-borne (mosquito or other aquatic insect)
443
component
444
other animals, including nodules and ulcers
445
cutaneous tissues, by inhibiting local immune responses, and it is the main toxin associated with
446
human buruli ulcers linked to M. ulcerans outbreaks48. M. ulcerans outbreaks in humans
447
typically occur in subtropical and tropical wetlands during periods of environmental
448
disturbance49, though there are no published reports of the mycobacterium affecting the Gulf
449
region or dolphins before or after the BP oil spill. Further, there was no evidence of
450
Mycobacterium-associated skin lesions noted during live health assessments or in dead, stranded
451
dolphins in BB50. Oil-contaminated seawater could alter endemic seawater bacteria, though little
452
data exist that has identified the ramifications of such alterations51. It is described that bacterial
453
populations around off-shore oil platforms and oil-contaminated seawater are unique52, 53, though
454
to date there is little connection between oil-contamination and antibiotic production from
455
bacteria. Possible bacterial metabolites identified in the breath of Barataria Bay dolphins suggest
46
39
with specific toxicity to
. Both M. marinum and M. ulcerans can cause cutaneous lesions in humans and 47
. Mycolactone can damage tissues, primarily
16 ACS Paragon Plus Environment
Page 17 of 28
Environmental Science & Technology
456
that these metabolites reflected bacteria present in the lungs of dolphins with pneumonia,
457
although further studies are needed to confirm chemical identities of metabolites particular to the
458
BB animals, to demonstrate that bacterial populations change in response to oil contamination,
459
and that these changes result in increased antibiotic production. Therefore, at the present time the
460
findings of the antibiotic compounds should only be considered in the context of the present
461
study pending further confirmation.
462
Though the focus of this investigation is on the breath metabolites of the BB dolphin
463
population, there were some unique metabolites identified in the SB animals. These tentative
464
metabolites include L-gizzerosine and haemulcholic acid. L-gizzerosine may derive as a by-
465
product from mackerel ingestion, a common food source for dolphins, and haemulcholic acid is
466
found in bile. It is presumed that these metabolites may represent differences in prey availability
467
for SB and BB dolphins.
468
When the breath metabolites from the two unexposed groups (SB and SD dolphins) were
469
combined and compared to the BB group, many of the same compounds were identified (e.g.
470
DG, PC lipids, Leucomycin) while even more metabolites were found to be specific to BB
471
dolphins (Table 1). Many of these breath metabolites were again markers of lipid oxidation,
472
arachidonic acid metabolism and lung surfactant breakdown. These compounds were presumably
473
originating from epithelial injury seen at a higher proportion in the BB group than the SB or SD
474
dolphins. The increased number of BB-specific metabolites when compared to the combined
475
group of control healthy animals indicates that some of these compounds may be present in
476
control animals, albeit with lower occurrence. Increasing the number of healthy animals led to
477
some of these compounds to not pass the “70% filter” for peak selection (described in the
478
“Methods” supplemental material), as these metabolites were detectable in fewer healthy
479
animals.
480
It is important to point out that there is a large degree of overlap between exhaled
481
metabolites that differentiate dolphin populations (Table S2), and many of these biomarkers
482
statistically correlate with ultrasound findings and/or blood parameters. This strongly indicates
483
that these metabolites are most likely not statistical aberrations but rather meaningful chemicals
484
that are engendered by specific health conditions (e.g., pneumonia and inflammation). An
485
example compound is phosphatidylglycerol such as PG(12:0/15:1(9Z)) or isomer, which 17 ACS Paragon Plus Environment
Environmental Science & Technology
486
differentiates oiled from healthy dolphins (Table S1), as well as is one of the most important
487
breath metabolites that correlate with lung consolidation (Table 2). In another example, a
488
compound postulated as Leucomycin is one of the compounds that differentiates SB and BB
489
animals while also correlates with inflammation (lymphocytes, Table S2).
Page 18 of 28
490
This study supports that exhaled breath analysis may become a key tool to non-invasively
491
assess dolphin54. EBC biomarkers appear to change in response to dolphin health in general and
492
specific issues such as infection or lung consolidation, in particular. Despite potential
493
confounding factors that may contribute to breath metabolite differences between groups (e.g.
494
geography, food supply, or water quality), we found the BB group differed from the SB and SD
495
dolphins. The managed collection animals were healthy, while some of the wild animals may
496
have had sporadic normal baseline health variations such as minor infections. The documented
497
breath metabolites that distinguished the affected BB population mostly fell into two categories:
498
products of cellular breakdown and naturally-occurring bacterial and fungal metabolites. The
499
increased amount of cellular breakdown products fit the proposed mechanism of aspirating or
500
inhaling noxious agents, though it is not yet clear if the products we detected were specific to BP
501
oil spill-associated petroleum products, dispersants, or to pulmonary inflammation in general.
502
Although it is tempting to draw conclusions based on biomarker differences between
503
groups, we see the main impact of this study to be the correlation of EBC metabolites with
504
established radiographic and blood markers of cetacean health. If the EBC metabolites can serve
505
as a proxy for biomarkers in blood (or other matrices) and correlate with ultrasound or other
506
assessment methodologies data, then EBC analysis could be used in lieu of those methodologies
507
for field health assessments, both routine and during critical events, such as environmental
508
disasters or mortality events. The methodology will facilitate such assessments for some species
509
and will enable them for others, for which these assessments have never been conducted due to
510
logistical difficulties.
511 512
ACKNOWLEDGEMENTS
18 ACS Paragon Plus Environment
Page 19 of 28
Environmental Science & Technology
513
Support for these investigators was provided by: the Office of Naval Research grant #N-
514
00014-13-1-0580 [CED, SVW]; National Institutes of Health (NIH) grant number
515
#1U01EB022003-01 and #UL1RR024146-06 and #1P30ES023513-01A1 [CED]; NIH #T32-
516
HL007013
517
#8KL2TR000134-07 K12 mentored training award and NIH #1K23HL127185-01A1 [MS]; and
518
Dolphin Quest, Inc. [RSW]. Sample collections in Sarasota Bay and Barataria Bay were part of
519
the Deepwater Horizon NRDA being conducted cooperatively among NOAA, other Federal and
520
State Trustees, and BP. This work was performed under an animal care and use protocol
521
reviewed and approved by the Navy Marine Mammal Program Institutional Animal Care and
522
Use Committee and the Navy Bureau of Medicine. The authors would like to thank Daniel J.
523
Peirano for data analysis discussions.
524
responsibility of the authors and do not necessarily represent the official views of the funding
525
agencies.
526
Zoological Society and conducted under National Marine Fisheries Service Scientific Research
527
Permit No. 15543, and the study in Barataria Bay was led by NOAA and conducted under
528
National Marine Fisheries Service Scientific Research Permit No. 932-1905/MA-009526.
and #T32-ES007059 [MS]; UC Davis School of Medicine and NIH
The contents of this manuscript are solely the
The dolphin health assessment study in Sarasota Bay was led by the Chicago
529 530 531 532 533
The supporting information has 8 pages, containing 3 figures (Figure S1, Figure S2, and Figure S3) and 3 table (Table S0, Table S1, Table S2).
534 535 536 537 538 539 540 541 542 543 544 545
1. Wells, R.; Rhinehart, H.; Hansen, L.; Sweeney, J.; Townsend, F.; Stone, R.; Casper, D. R.; Scott, M.; Hohn, A.; Rowles, T., Bottlenose Dolphins as Marine Ecosystem Sentinels: Developing a Health Monitoring System. EcoHealth 2004, 1, (3), 246-254. 2. Amann, A.; Smith, D., Volatile Biomarkers Non-Invasive Diagnosis in Physiology and Medicine. Elsevier: 2013; p 563. 3. Aksenov, A. A.; Schivo, M.; Bardaweel, H.; Zrodnikov, Y.; Kwan, A. M.; Zamuruyev, K.; Cheung, W. H. K.; Peirano, D. J.; Davis, C. E., Volatile Organic Compounds in Human Breath: Biogenic Origin and Point-of-Care Analysis Approaches. Volatile Biomarkers: Non-Invasive Diagnosis in Physiology and Medicine 2013, 129-154. 4. Aksenov, A. A.; Yeates, L.; Pasamontes, A.; Siebe, C. A.; Zrodnikov, Y.; Simmons, J.; McCartney, M. M.; Delplanque, J. P.; Wells, R. S.; Davis, C. E., Metabolite content profiling of bottlenose dolphin exhaled breath. Anal Chem 2014, 86, (21), 10616-10624.
REFERENCES
19 ACS Paragon Plus Environment
Environmental Science & Technology
546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592
5. Szulejko, J. E.; McCulloch, M.; Jackson, J.; McKee, D. L.; Walker, J. C.; Solouki, T., Evidence for cancer biomarkers in exhaled breath. IEEE Sens. J. 2010, 10, (1), 185-210. 6. Guard, U. S. C.; Team, N. R., On Scene Coordinator Report: Deepwater Horizon Oil Spill. U.S. Department of Homeland Security, U.S. Coast Guard: 2011. 7. Biello, D., Is Using Dispersants on the BP Gulf Oil Spill Fighting Pollution with Pollution? Scientific American 2010. 8. Matkin, C. O.; Saulitis, E. L.; Ellis, G. M.; Olesiuk, P.; Rice, S. D., Ongoing population-level impacts on killer whales Orcinus orca following the ‘Exxon Valdez’ oil spill in Prince William Sound, Alaska. Marine Ecology Progress Series 2008, 356, 269-281. 9. Lipscomb, T. P.; Harris, R. K.; Moeller, R. B.; Pletcher, J. M.; Haebler, R. J.; Ballachey, B. E., Histopathologic Lesions in Sea Otters Exposed to Crude Oil. Veterinary Pathology 1993, 30, (1), 1-11. 10. Schwacke, L. H.; Smith, C. R.; Townsend, F. I.; Wells, R. S.; Hart, L. B.; Balmer, B. C.; Collier, T. K.; De Guise, S.; Fry, M. M.; Guillette, L. J.; Lamb, S. V.; Lane, S. M.; Mcfee, W. E.; Place, N. J.; Tumlin, M. C.; Ylitalo, G. M.; Zolman, E. S.; Rowles, T. K., Health of Common Bottlenose Dolphins (Tursiops truncatus) in Barataria Bay, Louisiana Following the Deepwater Horizon Oil Spill (vol 48, pg 93, 2014). Environmental Science & Technology 2014, 48, (17), 10528-10528. 11. Venn-Watson, S.; Colegrove, K. M.; Litz, J.; Kinsel, M.; Terio, K.; Saliki, J.; Fire, S.; Carmichael, R.; Chevis, C.; Hatchett, W.; Pitchford, J.; Tumlin, M.; Field, C.; Smith, S.; Ewing, R.; Fauquier, D.; Lovewell, G.; Whitehead, H.; Rotstein, D.; McFee, W.; Fougeres, E.; Rowles, T., Adrenal Gland and Lung Lesions in Gulf of Mexico Common Bottlenose Dolphins (Tursiops truncatus) Found Dead following the Deepwater Horizon Oil Spill. PLoS One 2015, 10, (5), e0126538. 12. Smith, C. R.; Rowles, T. K.; Hart, L. B.; Townsend, F. I.; Wells, R. S.; Zolman, E. S.; Balmer, B. C.; Quigley, B.; Ivancˇic΄, M.; McKercher, W.; Tumlin, M. C.; Mullin, K. D.; Adams, J. D.; Wu, Q.; McFee, W.; Collier, T. K.; Schwacke, L. H., Slow recovery of Barataria Bay dolphin health following the Deepwater Horizon oil spill (2013-2014), with evidence of persistent lung disease and impaired stress response. Endangered Species Research 2017, 33, 127-142. 13. Trustees, D. H. N. R. D. A. Deepwater Horizon oil spill: Final Programmatic Damage Assessment and Restoration Plan and Final Programmatic Environmental Impact Statement. http://www.gulfspillrestoration.noaa.gov/restoration-planning/gulf-plan 14. Lane, S. M.; Smith, C. R.; Mitchell, J.; Balmer, B. C.; Barry, K. P.; McDonald, T.; Mori, C. S.; Rosel, P. E.; Rowles, T. K.; Speakman, T. R.; Townsend, F. I.; Tumlin, M. C.; Wells, R. S.; Zolman, E. S.; Schwacke, L. H., Reproductive outcome and survival of common bottlenose dolphins sampled in Barataria Bay, Louisiana, USA, following the Deepwater Horizon oil spill. Proc Biol Sci 2015, 282, (1818), 20151944. 15. Colegrove, K. M.; Venn-Watson, S.; Litz, J.; Kinsel, M. J.; Terio, K. A.; Fougeres, E.; Ewing, R.; Pabst, D. A.; McLellan, W. A.; Raverty, S.; Saliki, J.; Fire, S.; Rappucci, G.; Bowen-Stevens, S.; Noble, L.; Costidis, A.; Barbieri, M.; Field, C.; Smith, S.; Carmichael, R. H.; Chevis, C.; Hatchett, W.; Shannon, D.; Tumlin, M.; Lovewell, G.; McFee, W.; Rowles, T. K., Fetal distress and in utero pneumonia in perinatal dolphins during the Northern Gulf of Mexico unusual mortality event. Diseases of aquatic organisms 2016, 119, (1), 1-16. 16. Venn-Watson, S.; Garrison, L.; Litz, J.; Fougeres, E.; Mase, B.; Rappucci, G.; Stratton, E.; Carmichael, R.; Odell, D.; Shannon, D.; Shippee, S.; Smith, S.; Staggs, L.; Tumlin, M.; Whitehead, H.; Rowles, T., Demographic clusters identified within the northern Gulf of Mexico common bottlenose dolphin (Tursiops truncates) unusual mortality event: January 2010-June 2013. PLoS One 2015, 10, (2), e0117248. 17. Konstantin, O. Z.; Alexander, A. A.; Mark, B.; Alberto, P.; Celeste, P.; Soraya, F.; Stephanie, V.-W.; Bart, C. W.; Jean-Pierre, D.; Cristina, E. D., Enhanced non-invasive respiratory sampling from bottlenose dolphins for breath metabolomics measurements. Journal of Breath Research 2016, 10, (4), 046005.
20 ACS Paragon Plus Environment
Page 20 of 28
Page 21 of 28
593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638
Environmental Science & Technology
18. Smith, C. A.; Maille, G. O.; Want, E. J.; Qin, C.; Trauger, S. A.; Brandon, T. R.; Custodio, D. E.; Abagyan, R.; Siuzdak, G., METLIN: A Metabolite Mass Spectral Database. Therapeutic Drug Monitoring 2005, 27, (6), 747-751. 19. NOAA Deepwater Horizon Natural Damage Assessment Data. https://dwhdiver.orr.noaa.gov/home 20. Geladi, P.; Kowalski, B. R., Partial least-squares regression: a tutorial. Anal. Chim. Acta 1986, 185, 1-17. 21. Westerhuis, J. A.; Hoefsloot, H. C. J.; Smit, S.; Vis, D. J.; Smilde, A. K.; van Velzen, E. J. J.; van Duijnhoven, J. P. M.; van Dorsten, F. A., Assessment of PLSDA cross validation. Metabolomics 2008, 4, (1), 81-89. 22. Wikoff, W. R.; Grapov, D.; Fahrmann, J. F.; DeFelice, B.; Rom, W. N.; Pass, H. I.; Kim, K.; Nguyen, U.; Taylor, S. L.; Gandara, D. R.; Kelly, K.; Fiehn, O.; Miyamoto, S., Metabolomic Markers of Altered Nucleotide Metabolism in Early Stage Adenocarcinoma. Cancer Prevention Research 2015, 8, (5), 410. 23. Pasamontes, A.; Cheung, W. H. K.; Simmons, J.; Aksenov, A. A.; Peirano, D. J.; Grafton-Cardwell, E. E.; Kapaun, T.; Dandekar, A. M.; Fiehn, O.; Davis, C. E., Citrus tristeza virus infection in sweet orange trees and a mandarin × tangor cross alters low molecular weight metabolites assessed using gas chromatography mass spectrometry (GC/MS). Metabolomics 2016, 12, (3), 41. 24. Sumner, L. W.; Amberg, A.; Barrett, D.; Beale, M. H.; Beger, R.; Daykin, C. A.; Fan, T. W. M.; Fiehn, O.; Goodacre, R.; Griffin, J. L.; Hankemeier, T.; Hardy, N.; Harnly, J.; Higashi, R.; Kopka, J.; Lane, A. N.; Lindon, J. C.; Marriott, P.; Nicholls, A. W.; Reily, M. D.; Thaden, J. J.; Viant, M. R., Proposed minimum reporting standards for chemical analysis Chemical Analysis Working Group (CAWG) Metabolomics Standards Initiative (MSI). Metabolomics : Official journal of the Metabolomic Society 2007, 3, (3), 211221. 25. Zhang, Z., Variable selection with stepwise and best subset approaches. Annals of Translational Medicine 2016, 4, (7), 136. 26. Smith, C. R.; Solano, M.; Lutmerding, B. A.; Johnson, S. P.; Meegan, J. M.; Le-Bert, C. R.; EmoryGomez, F.; Cassle, S.; Carlin, K.; Jensen, E. D., Pulmonary ultrasound findings in a bottlenose dolphin Tursiops truncatus population. Diseases of aquatic organisms 2012, 101, (3), 243-255. 27. Liu, X. L.; Lian, R.; Tao, Y. K.; Gu, C. D.; Zhang, G. Q., Lung ultrasonography: an effective way to diagnose community-acquired pneumonia. Emergency medicine journal : EMJ 2015, 32, (6), 433-8. 28. Guven, K. C.; Percot, A.; Sezik, E., Alkaloids in Marine Algae. Mar Drugs 2010, 8, (2), 269-284. 29. Lu, C. W.; Shao, C. X.; Cobos, E.; Singh, K. P.; Gao, W. M., Chemotherapeutic Sensitization of Leptomycin B Resistant Lung Cancer Cells by Pretreatment with Doxorubicin. Plos One 2012, 7, (3). 30. Ma, W. S.; Mutka, T.; Vesley, B.; Amsler, M. O.; McClintock, J. B.; Amsler, C. D.; Perman, J. A.; Singh, M. P.; Maiese, W. M.; Zaworotko, M. J.; Kyle, D. E.; Baker, B. J., Norselic Acids A-E, Highly Oxidized Anti-infective Steroids that Deter Mesograzer Predation, from the Antarctic Sponge Crella sp. J Nat Prod 2009, 72, (10), 1842-1846. 31. Moreno, M.; Rodriguez, V. M., Medicinal Pre-Hispanic Plants .9. Yiamoloside-B, a Fungistatic Saponin of Phytolacca-Octandra. Phytochemistry 1981, 20, (6), 1446-1447. 32. Ridgway, S. H.; Scronce, B. L.; Kanwisher, J., Respiration and deep diving in the bottlenose porpoise. Science 1969, 166, (Copyright (C) 2013 U.S. National Library of Medicine.), 1651-4. 33. Minh, T. D. C.; Blake, D. R.; Galassetti, P. R., The clinical potential of exhaled breath analysis for diabetes mellitus. Diabetes Res Clin Pr 2012, 97, (2), 195-205. 34. Koczulla, A. R.; Noeske, S.; Herr, C.; Koepke, J.; Jorres, R. A.; Nell, C.; Schmid, S.; Vogelmeier, C.; Bals, R., Alpha-1 antitrypsin is elevated in exhaled breath condensate and serum in exacerbated COPD patients. Resp Med 2012, 106, (1), 120-126.
21 ACS Paragon Plus Environment
Environmental Science & Technology
639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686
35. Carpagnano, G. E.; Palladino, G. P.; Lacedonia, D.; Koutelou, A.; Orlando, S.; Foschino-Barbaro, M. P., Neutrophilic airways inflammation in lung cancer: the role of exhaled LTB-4 and IL-8. Bmc Cancer 2011, 11. 36. Venn-Watson, S.; Carlin, K.; Ridgway, S., Dolphins as animal models for type 2 diabetes: Sustained, post-prandial hyperglycemia and hyperinsulinemia. Gen Comp Endocr 2011, 170, (1), 193199. 37. Schivo, M.; Aksenov, A. A.; Yeates, L. C.; Pasamontes, A.; Davis, C. E., Diabetes and the metabolic syndrome: possibilities of a new breath test in a dolphin model. Front Endocrinol (Lausanne) 2013, 4, 163. 38. Venn-Watson, S. K.; Parry, C.; Baird, M.; Stevenson, S.; Carlin, K.; Daniels, R.; Smith, C. R.; Jones, R.; Wells, R. S.; Ridgway, S.; Jensen, E. D., Increased Dietary Intake of Saturated Fatty Acid Heptadecanoic Acid (C17:0) Associated with Decreasing Ferritin and Alleviated Metabolic Syndrome in Dolphins. Plos One 2015, 10, (7), e0132117. 39. McKee, R. H.; White, R., The Mammalian Toxicological Hazards of Petroleum-Derived Substances: An Overview of the Petroleum Industry Response to the High Production Volume Challenge Program. Int J Toxicol 2014, 33, 4S-16S. 40. Foot, N. J.; Orgeig, S.; Donnellan, S.; Bertozzi, T.; Daniels, C. B., Positive selection in the Nterminal extramembrane domain of lung surfactant protein C (SP-C) in marine mammals. J Mol Evol 2007, 65, (1), 12-22. 41. Han, S.; Mallampalli, R. K., The Role of Surfactant in Lung Disease and Host Defense against Pulmonary Infections. Ann Am Thorac Soc 2015, 12, (5), 765-74. 42. Tibboel, J.; Reiss, I.; de Jongste, J. C.; Post, M., Sphingolipids in lung growth and repair. Chest 2014, 145, (1), 120-8. 43. Szefler, S. J.; Wenzel, S.; Brown, R.; Erzurum, S. C.; Fahy, J. V.; Hamilton, R. G.; Hunt, J. F.; Kita, H.; Liu, A. H.; Panettieri, R. A., Jr.; Schleimer, R. P.; Minnicozzi, M., Asthma outcomes: biomarkers. J Allergy Clin Immunol 2012, 129, (3 Suppl), S9-23. 44. Jansen, R.; Irschik, H.; Reichenbach, H.; Hofle, G., Antibiotics from gliding bacteria .80. Chivosazoles A-F: Novel antifungal and cytotoxic macrolides from Sorangium cellulosum (Myxobacteria). Liebigs Ann-Recl 1997, (8), 1725-1732. 45. Perlova, O.; Gerth, K.; Kaiser, O.; Hans, A.; Muller, R., Identification and analysis of the chivosazol biosynthetic gene cluster from the myxobacterial model strain Sorangium cellulosum So ce56. J Biotechnol 2006, 121, (2), 174-91. 46. Vezina, C.; Bolduc, C.; Kudelski, A.; Audet, P., Biosynthesis of kitasamycin (leucomycin) by leucine analog-resistant mutants of Streptomyces kitasatoensis. Antimicrob Agents Chemother 1979, 15, (5), 738-46. 47. Lee, E.; Song, H. Y.; Joo, J. M.; Kang, J. W.; Kim, D. S.; Jung, C. K.; Hong, C. Y.; Jeong, S.; Jeon, K., Synthesis of (+)-Lasonolide A: (−)-Lasonolide A is the biologically active enantiomer. Bioorganic & Medicinal Chemistry Letters 2002, 12, (24), 3519-3520. 48. Zhang, Y. W.; Ghosh, A. K.; Pommier, Y., Lasonolide A, a potent and reversible inducer of chromosome condensation. Cell Cycle 2012, 11, (23), 4424-35. 49. Jakeman, D. L.; Bandi, S.; Graham, C. L.; Reid, T. R.; Wentzell, J. R.; Douglas, S. E., Antimicrobial activities of jadomycin B and structurally related analogues. Antimicrob Agents Chemother 2009, 53, (3), 1245-7. 50. Yip, M. J.; Porter, J. L.; Fyfe, J. A.; Lavender, C. J.; Portaels, F.; Rhodes, M.; Kator, H.; Colorni, A.; Jenkin, G. A.; Stinear, T., Evolution of Mycobacterium ulcerans and other mycolactone-producing mycobacteria from a common Mycobacterium marinum progenitor. J Bacteriol 2007, 189, (5), 2021-9. 51. Collier, D. N., Cutaneous infections from coastal and marine bacteria. Dermatologic Therapy 2002, 15, (1), 1-9. 22 ACS Paragon Plus Environment
Page 22 of 28
Page 23 of 28
687 688 689 690 691 692 693 694 695 696 697
Environmental Science & Technology
52. Quek, T. Y.; Athan, E.; Henry, M. J.; Pasco, J. A.; Redden-Hoare, J.; Hughes, A.; Johnson, P. D., Risk factors for Mycobacterium ulcerans infection, southeastern Australia. Emerging infectious diseases 2007, 13, (11), 1661-6. 53. Marsollier, L.; Robert, R.; Aubry, J.; Saint Andre, J. P.; Kouakou, H.; Legras, P.; Manceau, A. L.; Mahaza, C.; Carbonnelle, B., Aquatic insects as a vector for Mycobacterium ulcerans. Applied and environmental microbiology 2002, 68, (9), 4623-8. 54. Jernigan, J. A.; Farr, B. M., Incubation period and sources of exposure for cutaneous Mycobacterium marinum infection: Case report and review of the literature. Clin Infect Dis 2000, 31, (2), 439-443. 55. Dunn, O. J., Multiple Comparisons among Means. Journal of the American Statistical Association 1961, 56, (293), 52-64.
698 699 700 701 702
23 ACS Paragon Plus Environment
Environmental Science & Technology
703
Page 24 of 28
TOC
704 705 706 707
Figure 1: PCA score plot of the EBC metabolome sampled from San Diego and Sarasota and Barataria. N is the number of samples.
708
709 710 711 712 713 714
24 ACS Paragon Plus Environment
Page 25 of 28
715 716 717
Environmental Science & Technology
Table 1. Significantly altered metabolites in exhaled breath condensate and key discriminants in oil spillaffected bottlenose dolphins living in Barataria Bay compared to healthy reference dolphins (San Diego and Sarasota Bay).
Retention Time
Tentative Chemical Identification
132.0788
4.392
Undetermined*
160.0873
1.754
4-Cyanoindole
246.0781
1.86
250.178
Mass
Class
Fold Change a
Directio nb Related to Healthy Referenc e Dolphins
pFDRc
4.65E09 2.36E12 5.67E08 3.53E03 5.92E07 8.79E03 1.06E10 5.82E07
78
UP
44
UP
Undetermined*
44
UP
2.172
Undetermined*
0.3
DOWN
300.1927
1.883
Undetermined*
133
UP
300.2773
9.072
Decanoylcholine
0.4
DOWN
310.1087
1.482
Undetermined*
32
UP
318.2025
1.884
Undetermined*, (Adrenosterone or other steroid)
Steroid/Li pid
96
UP
326.1946
4.464
Possibly 1-[(5-Amino-5carboxypentyl)amino]-1-deoxyfructose
Amino Acids Derivative
18
UP
2.72E21
365.2399
2.019
R/S-4-benzyl-3-((S)-3-hydroxy-2,2dimethyloctanoyl)oxazolidin-2-one
97
UP
3.29E11
1.762
lipid: Docosatetraenoyl Ethanolamide, Nethyl N-(2-hydroxy-ethyl) arachidonoyl amine etc.
0.5
DOWN
5.05E01
422.2343
1.467
Possibly 7α-(Thiomethyl)spironolactone sulfoxide or a 15beta-Hydroxy-7alphamercapto-pregn-4-ene-3,20-dione 7acetate
67
UP
1.39E48
429.3815
5.973
Possibly 25-Azacholesterol
0.5
DOWN
3.81E02
437.371
1.927
Steroid, e.g. 5beta-Cholestane3alpha,7alpha,24,26-tetrol, 5-bCholestane-3a-7-tetraol etc.
0.6
DOWN
1.05E01
458.2117
1.727
Undetermined*, possibly a metoxyphenol
67.3
UP
496.2308
6.459
Possibly WYE-354
62.6
UP
497.3201
2.311
Possibly a steroid, e.g. Pregnanediol-3glucuronide
Steroid/Li pid
138
UP
514.3156
1.956
Undetermined*, possibly G-418 or PG(17:1(9Z)/0:0)
Lipid
364
UP
393.3421
Amine
Lipid
Lipid
Lipid Lipid
25 ACS Paragon Plus Environment
4.93E07 3.43E12 3.11E37 5.79E07
Environmental Science & Technology
537.4229
1.938
Possibly N-Methylindolo[3,2-b]-5alphacholest-2-ene
549.2809
1.724
Undetermined* (Antimycin A1 or Myristinin A)
549.3429
1.821
PG(O-20:0/0:0)
566.2995
1.759
Imazamethabenz or Archangelolide or Antimycin A1
583.2537
1.704
Possibly Biliverdin IX
627.4904
1.861
A diglyceride, e.g. DG(17:1(9Z)/20:5(5Z,8Z,11Z,14Z,17Z)/0: 0)[iso2]
633.4085
1.967
Oleanolic acid 3-O-beta-D-glucosiduronic acid
681.4653
2.268
A Phosphatidylglycerol, e.g. PG(17:0/12:0)
693.4024
2.221
Phosphatidylglycerol
722.3751
1.684
25-Acetyl-6,7-didehydrofevicordin F 3glucoside
737.4277
2.328
Myxochromide S2
743.5735
2.446
A Phosphatidylethanolamine lipid
755.4858
2.35
A Phosphatidylglycerol,
759.569
2.27
A Phosphatidylethanolamine or Phosphatidylcholine
780.4212
1.711
Leucomycin A7
781.4519
2.441
Digoxin or Gitoxin
797.3611
1.649
Possibly Rifamycin B
805.5958
1.784
A Phosphatidylglycerol or Phosphatidylserine
809.5625
1.889
A Phosphatidylglycerol or Phosphatidylinositol
819.5708
1.953
Undetermined* (a Phosphatidylcholine)
825.479
2.568
Yiamoloside B
879.6141
1.93
A Phosphatidylglycerol
1.673
(3b,16a,20R)-3,16,20,22,25Pentahydroxy-5-cucurbiten-11-one 3[rhamnosyl-(1->4)-[glucosyl-(1->6)]glucoside]
994.5428
Steroid/Li pid Secondary bacterial metabolite Lipid Secondary bacterial metabolite
Page 26 of 28
80
UP
9.64E12
79
UP
1.26E05
374
UP
5.92E07
206
UP
6.05E07
139
UP
1.49E05
36.6
UP
3.13E11
72.4
UP
9.60E07
69
UP
7.34E08
40.9
UP
3.04E26
300.6
UP
1.45E05
69.2
UP
1.64E11
146
UP
46
UP
45
UP
381
UP
1.25E07
130
UP
1.27E44
272
UP
8.77E11
545
UP
1.86E11
53
UP
4.38E06
119
UP
188
UP
80
UP
301
UP
Lipid
Lipid Secondary bacterial metabolite
Secondary bacterial metabolite Lipid Lipid Lipid Secondary bacterial metabolite Secondary bacterial metabolite Lipid Lipid
Terpenoid Lipid
26 ACS Paragon Plus Environment
1.75E05 2.61E05 6.76E12
3.55E06 1.31E07 3.06E06 9.89E09
Page 27 of 28
Environmental Science & Technology
* no reasonable matches, too many possible alternatives or identification not possible a ratio of means relative to Sarasota and San Diego samples
718 719 720 721 722 723 724 725 726 727
b
direction of change in means relative to Sarasota samples
c
false discovery rate adjusted mixed effects model p-value
Table 2: List of the most important EBC metabolite that correlate with the pulmonary disease in Barataria Bay bottlenose dolphins (consolidation) as determined by ultrasound examinations conducted during 2011 following the 2010 Deepwater Horizon oil spill. Fold Compared to BB Dolphins without Lung Consolidationa
Mass
Retention Time
t.test adjusted pvalue55
246.1016
2.695
0.051
3.5
Debromohymenialdisine
363.2358
12.708
0.051
2.9
Dimethisterone or other steroid
310.1082
1.492
0.051
2.5
Undetermined*
416.1507
2.049
0.051
2.5
Undetermined*
422.2346
1.495
0.051
2.5
Possibly 7α-(Thiomethyl)spironolactone sulfoxide or a 15beta-Hydroxy-7alphamercapto-pregn-4-ene-3,20-dione 7-acetate
580.2943
8.601
0.051
2.5
Protoporphyrin IX
458.2125
1.709
0.051
2.4
Undetermined*, possibly Metoxyphenol
825.4798
2.568
0.051
2.3
Yiamoloside B
318.2064
1.902
0.051
2.1
Undetermined*, possibly Adrenosterone or other steroid
625.4772
2.087
0.051
2.1
Diacylglycerol
739.4724
1.971
0.051
2.1
Antanapeptin B or a steroid glucoronide
1004.7853
1.697
0.051
2.1
LacCer(d18:0/26:0) or NAPE(16:0/18:1(9Z)/20:4(5Z,8Z,11Z,14Z)
453.2938
2.193
0.051
2
Norselic acid A
472.2984
5.849
0.051
2
Deoxodeoxydihydrogedunin
737.4288
2.333
0.051
2
Myxochromide S2
781.4548
2.436
0.051
2
Digoxin or Gitoxin
999.7644
1.75
0.051
2
PI-Cer(t20:0/26:0(2OH))
Tentative chemical Identification
27 ACS Paragon Plus Environment
Environmental Science & Technology
742.4488
7.406
0.051
1.9
Erythromycin C or Phosphatidylinositol lipid, e.g. PI(10:0/16:1(9Z))
310.157
1.748
0.051
1.8
Undetermined*, possibly Butenachlor
541.3468
2.433
0.051
1.8
Leptomycin B or PG(20:0/0:0)
651.4197
1.864
0.051
1.8
Phosphatidylglycerol, e.g. PG(12:0/15:1(9Z))
695.5196
2.155
0.051
1.7
Undetermined*, likely a lipid e.g. PG(O16:0/15:0)
605.4459
1.974
0.051
1.5
Phospholipid, e.g. PA(P-18:0/12:0)
323.1987
3.877
0.001
