- Email: [email protected]
The Equine Volatilome: Volatile Organic Compounds as Discriminatory Markers
Accepted Manuscript The Equine Volatilome: Volatile Organic Compounds as Discriminatory Markers Ketaki Deshpande, Kenneth g. Furton, De etta k. Mills
PII:
S0737-0806(17)30431-8
DOI:
10.1016/j.jevs.2017.05.013
Reference:
YJEVS 2331
To appear in:
Journal of Equine Veterinary Science
Received Date: 4 May 2017 Revised Date:
26 May 2017
Accepted Date: 29 May 2017
Please cite this article as: Deshpande K, Furton Kg, Mills Dek, The Equine Volatilome: Volatile Organic Compounds as Discriminatory Markers, Journal of Equine Veterinary Science (2017), doi: 10.1016/ j.jevs.2017.05.013. 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
SPME-GC-MS analysis of hair samples of domestic horses determines kinship as well as
AC C
EP
TE D
M AN U
SC
RI PT
provides evidence of breed specific signals based on volatile organic compound profiles.
ACCEPTED MANUSCRIPT
Title - THE EQUINE VOLATILOME: VOLATILE ORGANIC COMPOUNDS AS DISCRIMINATORY MARKERS
RI PT
Authors - KETAKI DESHPANDE 1,2, KENNETH G. FURTON 1,3 & DE ETTA K. MILLS* 1,2
International Forensic Research Institute; 2 Department of Biological Sciences;
3
Department of Chemistry and Biochemistry, Florida International University, Miami,
SC
1
M AN U
FL 33199 USA
Corresponding Author - DeEtta K. Mills, Florida International University. 11200 SW 8th Street, OE 167, Biological Sciences, Miami, FL 33199, United States.
ABSTRACT
TE D
Phone: 305-348-7410, Fax Number: 305-348-1986, E-mail: [email protected]
Background: Animals rely on subtle signals perceived between individuals that convey
EP
information such as sex, reproductive status, individual identity, ownership, competitive ability as well as health status. These cues have important influences on behaviors that
AC C
are vital for reproductive success, such as parent–offspring attachment, recognition of relatedness, mate choice and territorial marking.
Objective: This study investigates
individual odor profiles as discriminatory markers possibility affecting equine behavior. Methodology: This study investigated the volatile compounds in horses using solid phase microextraction gas chromatography-mass spectrometry and hair samples. Study design: Two horses breeds, Appaloosa (n=6) and Quarter Horses (n=17), were used as a model to
ACCEPTED MANUSCRIPT
assess the volatilome that may be used to identify individuals and kin recognition. Results: A total of 187 volatiles were identified and demonstrated that the volatilome carries information on genetic relatedness. Interestingly, apart from individualization and
RI PT
kinship, the volatiles detected were also able to discriminate between breeds. Main limitations: The study is limited exclusively two horse breeds and should be expanded to include other domestic and wild horses. Additionally, compounds were identified based
SC
on NIST 98 mass spectral library. Conclusions: The results of this study support the individual odor hypothesis suggesting that each individual possesses a unique scent that
M AN U
acts as a characteristic or odor fingerprint. Such qualitative and quantitative patterns have not been explored in domestic or wild animals and there is a need to understand more about the influence of odor on behavior and interactions.
AC C
EP
hair
TE D
Key Words – Kin; breed-specific; domestic; volatilome; Volatile organic compounds;
ACCEPTED MANUSCRIPT
ABSTRACT
2
Background: Animals rely on subtle signals perceived between individuals that convey
3
information such as sex, reproductive status, individual identity, ownership, competitive
4
ability as well as health status. These cues have important influences on behaviors that
5
are vital for reproductive success, such as parent–offspring attachment, recognition of
6
relatedness, mate choice and territorial marking.
7
individual odor profiles as discriminatory markers possibility affecting equine behavior.
8
Methodology: This study investigated the volatile compounds in horses using solid phase
9
microextraction gas chromatography-mass spectrometry and hair samples. Study design:
10
Two horse breeds, Appaloosa (n=6) and Quarter Horses (n=17), were used as a model to
11
assess the volatilome that may be used to identify individuals and kin recognition.
12
Results: A total of 187 volatiles were identified and demonstrated that the volatilome
13
carries information on genetic relatedness. Interestingly, apart from individualization and
14
kinship, the volatiles detected were also able to discriminate between breeds. Main
15
limitations: The study is limited exclusively two horse breeds and should be expanded to
16
include other domestic and wild horses. Additionally, compounds were identified based
17
on NIST 98 mass spectral library. Conclusions: The results of this study support the
18
individual odor hypothesis suggesting that each individual possesses a unique scent that
19
acts as a characteristic or odor fingerprint. Such qualitative and quantitative patterns
20
have not been explored in domestic or wild animals and there is a need to understand
21
more about the influence of odor on behavior and interactions.
RI PT
1
AC C
EP
TE D
M AN U
SC
Objective: This study investigates
1
ACCEPTED MANUSCRIPT
Key Words – Kin; breed-specific; domestic; volatilome; Volatile organic compounds;
23
hair
24
INTRODUCTION
RI PT
22
Olfactory cues function in the animal kingdom to distinguish between kin and
26
predators, detect food sources and even environmental toxins. The core body odors
27
(volatilome) are emitted volatile organic compounds (VOCs) that are the end products of
28
metabolism [1]. Factors such as genetics, diet, environment, the microbiome present on
29
the body, and even exogenous materials modulate the VOCs and can contribute to an
30
individual’s odor profile. Moreover, disease processes, such as infection, parasite load, or
31
endogenous metabolic disorders, can influence an individual’s odor profile by producing
32
different volatiles or by changing the proportions of VOCs that are normally produced
33
[2]; thus, they convey information about an individual’s metabolic or physiological
34
status. The olfactory differentiation for mating preferences and conspecific identification
35
or kin recognition has been studied across various taxonomic groups including rodents [3,
36
4,5] fish [6], Swedish sand lizards [7], birds [8], Tuco-tucos [9], otters [10] and lemurs
37
[11,12]. Although most of these studies did not focus on the individual components of the
38
odor, they led to the conclusion that individuals have a distinct body-odor type, which is
39
influenced by their inherited MHC alleles, and plays a pivotal role in kin recognition,
40
mate selection and identification of dissimilar or similar individuals [13].
M AN U
TE D
EP
AC C
41
SC
25
Horses, like all mammals, can recognize and discriminate chemical signals, which
42
provide essential information for individual and herd survival and greatly influence their
43
social behavior [14]. The large olfactory bulbs in a horse’s brain exhibit a convoluted
2
ACCEPTED MANUSCRIPT
surface, while the large size of the olfactory epithelium suggest that olfactory information
45
is vital to horses. Additionally, horses exhibit a well-developed vomeronasal organ that
46
is receptive to nonvolatile, large, species-specific molecules found in body secretions
47
[14]; hence, they have a highly developed olfactory capacity. Observational studies of
48
domestic and wild/feral horses have described how horses recognize each other on the
49
basis of body odors, urine, and feces [15,16,17]. Recognition at the individual level
50
guides the horse’s response based on previous experiences and determines the outcome of
51
the interactions. Close proximity mutual sniffing has been seen during horse greetings
52
and sexual advances that include blowing air on the face, standing parallel and sniffing
53
the neck and under another’s belly [18]. Horses often sniff excrement (dung piles) [19]
54
that allows horses to recognize other individuals [20,21] and to differentiate the sex of the
55
horse by its feces [18]. In fact, stallions create fecal piles known as stud piles and
56
repeatedly return to them to defecate as a territorial marking behavior [22,23]. Similar
57
practices are seen when the harem stallion covers an area with his urine where the harem
58
female horses have previously urinated or defecated [24,25]. Such greetings and scent
59
marking displayed by feral and domestic horses, suggest that odor is crucial in social
60
encounters and it is used to gather useful information from the chemical cues [20].
SC
M AN U
TE D
EP
AC C
61
RI PT
44
Curran et.al have proposed that individuals emanate “primary odor” VOCs that
62
are unique and stable over time regardless of diet or environmental factors and has
63
demonstrated using SPME/GC-MS that comparison of the presence and ratios of these
64
primary odor VOCs, individual humans can be differentiated [26,27,28]. Penn et al. have
65
proposed an “individual odor hypothesis” also suggesting that each individual possesses a
3
ACCEPTED MANUSCRIPT
unique VOC fingerprint [29]. The objective of the present study was to investigate odor
67
profiles from domestic horses, Equus caballus. Evidence for equine body volatiles that
68
are presumably used for individual discrimination is lacking; therefore, this study
69
examined the VOC profiles of domestic horses to determine the components of an
70
individual’s odor and whether these profiles can reliably indicate a degree kinship.
RI PT
66
METHODS AND MATERIALS
73
Sample Collection
M AN U
72
SC
71
Triplicate hair samples were donated by the horse owners and plucked from the
75
manes of 23 domestic horses (Table 1). Mane hair samples (each ≈10 mg) were collected
76
from each horse using sterile disposable gloves. The sample were immediately placed
77
into 10-mL glass clear, screw top vials (Supelco, Bellefonte, PA), sealed and allowed to
78
equilibrate for 24 h prior to SPME extraction. The usages of specific horse fly sprays,
79
feed, bathing routines were noted for all samples.
80
sampling and use of fly spray was avoided with the exception of Horse H5, who suffered
81
from insect bite hypersensitivity and was under treatment. Empty 10-mL vials (Supelco)
82
were exposed to the stable’s atmosphere for environmental VOC collection at each
83
sampling time. All the horses included in this study were fed the same feed and managed
84
in a similar fashion.
85
Solid-Phase Microextraction (SPME)- Gas Chromatography-Mass Spectrometry (GC-
86
MS) Procedures
Horses were not bathed before
AC C
EP
TE D
74
4
ACCEPTED MANUSCRIPT
Solid-Phase Microextraction (SPME) coupled with GC-MS is a sampling
88
technique that involves the use of a fiber coated with an extracting phase, that can capture
89
different analytes from samples. The GC works on the principle that a mixture will
90
separate into individual substances when heated, while mass spectrometry identifies
91
compounds by the mass of the analyte molecule. This enables the identification of VOCs,
92
which are secondary volatile metabolites and contribute to an individual’s unique odor
93
profile. Divinylbenzene/Carboxen on Polydimethylsiloxane (CAR/DVB on PDMS) 50/30
94
µm fibers (Supelco) were used to extract the VOCs from the headspace of the vials
95
containing the hair samples and environmental blanks. Fiber exposure was conducted at
96
room temperature for 12 h. The samples were separated and analyzed by GC-MS using
97
an Agilent 6890 GC with a 5973 MS. A HP5-MS column, with helium as the carrier gas
98
at flow rate of 1.0 mL/min for the separation of the analytes was used. The extracted
99
VOCs were desorbed in the injection port at 250°C for 10 min in splitless mode. The
100
temperature program was: initial oven temperature of 40°C for 5 min, 10°C per minute
101
ramp to a final temperature of 250°C, and a final hold for 2 min for a total run time of 32
102
min. The mass spectrometer used was an HP 5973 MSD with a quadrupole analyzer in
103
full scan mode. The compounds were identified using the NIST 98 mass spectral library.
104
The criterion for the identification of compounds was based on the quality of the detected
105
peak, set at greater than or equal to 40%. Additionally, compounds detected in all three
106
replicates at the same retention time were considered as part of an individual’s
107
volatilome.
108
Statistical Analysis
AC C
EP
TE D
M AN U
SC
RI PT
87
5
ACCEPTED MANUSCRIPT
Prior to statistical analyses, data were reduced by removing compounds present in
110
the environmental blanks and in Pyranha fly spray based on the MSDS description.
111
Further, compound names were searched on ChemSpider [30] to assess if they were
112
related to equine husbandry practices or presence in the environment.
113
commonly associated with ointments and plant/feed related compounds were removed
114
from the analysis. The relative ratio of each VOC’s abundance was calculated for all
115
VOCs and, subsequently transformed using square-root transformation.
116
Spearman Rank Correlation
RI PT
109
M AN U
SC
Compounds
117
Each horse’s VOC profile was statistically evaluated to determine the similarity
118
between each replicate. The replicate profiles for each horse were then averaged to
119
produce a single representative VOC profile.
120
correlated, in a pair wise manner, to the rest of the horses using Spearman rank
121
correlations.
122
Multivariate Analyses of Volatiles
TE D
Each individual VOC profile was
A non-metric multidimensional scaling (nMDS) plot is a visualization of the
124
pattern of proximities based on similarities or distances among the objects. The analysis
125
was performed in order to identify similarity/differences of (i) replicates from each horse
126
(ii) horse breeds and (iii) kinship. The analyses were performed using Primer-E ver 7
127
software [31] to create a similarity matrix among variables using the Bray-Curtis
128
similarity coefficient. The relative ratios were used for hierarchical cluster analysis
129
(HCA) using the complete linkage method. In order to observe percent similarity, the
AC C
EP
123
6
ACCEPTED MANUSCRIPT
cluster analysis output was overlaid on the nMDS ordination plots, indicated by circles
131
grouping the data.
132
Differences in composition of compounds within groups were tested using an Analysis of
133
Similarities (ANOSIM) [32]. Similarity Percentage analysis (SIMPER) was performed to
134
determine which compounds influenced the discrimination observed between the
135
individuals, breeds and kin. Heatmaps were generated to visualize the contribution of top
136
47 compounds obtained in SIMPER analyses. The heatmap is based on relative ratio of
137
compounds where the regions of white represent absence of a compound, blue represents
138
low relative abundance while dark red indicates higher relative abundance with highest
139
value of 0.52.
M AN U
SC
RI PT
130
140 RESULTS
TE D
141
The hair samples revealed compounds from various functional groups: alkanes,
143
alcohols, hydrocarbons, aldehydes, alkenes, amines/amides, ketones, esters, and indols
144
(n=187).
Three VOC with varying abundances were detected in 100% of horses
145
sampled;
Nonanal,
146
Dibromo-2-Methyl- Undecane. Ninety-eight compounds were specific to Quarter Horses;
147
however, no one compound was common across all 17 Quarter Horses.
148
Spearman Rank Correlations
Fluoren-9-Ol,
3,6-Dimethoxy-9-(2-Phenylethynyl)-,
and
1,2-
AC C
149
EP
142
Among horses of known relationships (parent-offspring, full sibling sharing dam
150
and sire, half sibling sharing dam or sire), correlation coefficient of ≥ 0.7 (S1) was
151
observed. These values demonstrate that, though there are qualitative similarities with
7
ACCEPTED MANUSCRIPT
presence/absence of VOCs between horses, the difference in relative abundance ratios
153
allowed for a higher percentage of discrimination between related and unrelated horses.
154
Multivariate Analysis
RI PT
152
The nMDS plot (Figure 1) with a stress value of 0.15 and 80% similarity shows
156
the VOC replicates for each horse and indicated that VOC from hair samples were stable
157
over time and reproducible. In support of the ‘individual odor hypothesis’, there was a
158
significant difference among each horse’s profile (R = 0.90; P < 0.001) in comparison to
159
other horses. The nMDS plot with a stress value of 0.15 (Figure 2) using the averages
160
showed a clear separation of Appaloosa and Quarter Horse breeds (R=0.814; P < 0.001).
161
The 40% similarity grouping was observed for related horses while a 20% similarity was
162
seen between horse breeds.
M AN U
SC
155
The nMDS plot (Figure 3) with a stress value of 0.15 indicated that the sampling
164
site (stables) had minimal effect on grouping of the horses when related individuals are
165
included in the study. This suggests that animal husbandry practices did not have an
166
influence on the VOC profiles. Overall, plots showed that VOC profiles are different for
167
each horse and consequently, one can individualize them. The HCA clearly grouped the
168
Appaloosas separate from the Quarter Horses demonstrating breed specific patterns.
169
Although clustering of related horses was quite evident, clustering of distant relationships
170
was also observed for horses H15 – H16 – H17, were H15 is the uncle to H16 and H17
171
from the dam’s lineage.
AC C
EP
TE D
163
172
SIMPER analyses of the dominant 47 compounds were able to discriminate the
173
Appaloosa and Quarter Horse breeds (S2). The average dissimilarity between the breeds
8
ACCEPTED MANUSCRIPT
was 77.03%. Using these 47 compounds, the heatmap (Figure 4) indicated the relative
175
ratio differences that provide individual uniqueness and the clustering of the two breeds.
176
The heatmap revealed that, rather than just a simple presence or absence of compounds,
177
the relative ratio of compounds differed across the horses sampled.
RI PT
174
178 DISCUSSION
SC
179
The potential role of olfaction in horses is known by a relatively small number of
181
published studies concentrated primarily on the role of odor in horse mating behavior and
182
sex identification. Such studies have focused on volatile chemical markers from feces
183
and urine samples that fluctuate with ovulation or the estrus cycle in mares [33,34] and
184
can elicit a response from stallions. The current study targeted odor profiles from hair
185
samples in order to learn whether particular individuals have a specific odor profile.
186
Cumulative VOC profiles considering presence/absence of compounds along with
187
relative ratios can successfully individualize horses supporting the “individual odor
188
hypothesis” [29]. As has been seen in vertebrate species like lemurs [35], mandrills [36],
189
deer [37], and now the odor profiles in horses, a high percentage of chemical compounds
190
were shared among profiles of all horses. However, the significant differences between
191
each individual suggested that the uniqueness in the chemical profile depends more on
192
the concentration of compounds and on complex interactions between compounds than
193
on the simple presence or absence of specific chemicals compounds [28,38]. Using
194
headspace SPME method Curran et al., further showed that a combination of the relative
195
ratios of compounds and the presence of differing compounds allows for the
AC C
EP
TE D
M AN U
180
9
ACCEPTED MANUSCRIPT
chromatographic distinction among individuals [27]. Curran et al. found that compounds
197
were present in differing ratio patterns between the males and females, thus indicating
198
qualitative similarities among individuals with quantitative differences [28].
RI PT
196
In a study conducted by Celerier et al., it was shown that mice were able to detect
200
baboons based on individual odor differences and were able to perceive a higher odor
201
similarity between related baboons than between unrelated baboons. The ability of mice
202
to discriminate other mammal species based on their odor suggests that odors may play a
203
role in both the signaling of individual characteristics and of relatedness among
204
individuals [39]. This supports the observed variation in volatiles between horses and
205
was indicative of an individual-specific odor that could aid in individual identification,
206
which may influence the behavioral responses. In this study, it is interesting to observe
207
that volatiles patterns differentiated between breeds even when stabled in different
208
locations. Horses that were housed at the same location but not related demonstrated that
209
housing and management appeared to have little influence on their volatile profiles. This
210
is contrary to the findings in otters housed in the same location where the similarity was
211
attributed to common diet as a potential explanation [40]. Grouping of horses with
212
unknown relationships may be due to variables that were not investigated in this study,
213
for example, age, current reproductive or health status of the individuals.
M AN U
TE D
EP
AC C
214
SC
199
From an evolutionary standpoint, kin recognition aids in parental care, kin
215
altruism, inbreeding avoidance, and maintenance of optimal outbreeding.
216
choice, most animals mate with unrelated or distant relatives. This innate selection is
217
thought to be evolutionarily favored, as it should improve their inclusive fitness and
Given a
10
ACCEPTED MANUSCRIPT
decrease the effects of inbreeding on the population [12]. In lemurs it was observed that
219
chemical cues coded genetic relatedness within and between sexes. Additionally, these
220
chemicals showed a seasonal pattern demonstrating variation in accordance to the
221
competitive breeding season of lemurs. Boulet et al., suggest that a strong olfactory
222
signal of relatedness during breeding seasons would be crucial for preventing inbreeding
223
[12]. This manner of discrimination can be treated as a communication mechanism
224
where prior to identification, the animal must be able to receive relevant signals from
225
other individuals and distinguish appropriately between them [41].
226
provided the individual odor cues vary. Therefore, studying body volatiles may enable
227
one to better understand how animals use kin-biased behavior in their interactions within
228
breeding populations [42].
This can occur
M AN U
SC
RI PT
218
When considering relationships in horses, breed lineages need to be queried rather
230
than just assessing direct descendants such as parent-offspring or siblings. Such closely
231
line bred pedigrees as seen in today’s domestic horses could group horses that might not
232
share a direct relationship through a sire or dam. Even though all sire and dam pairs were
233
not known in the study, VOC profiles were able to demonstrate relatedness between
234
horses with known lineages. Although the closely related horses had similar chemical
235
components, it cannot be said that horses from the same kin group share the exact odor
236
phenotype.
237
compared to distant kin indicated a graded or continuous relationship of the volatilome
238
within kin groups when using relative abundances rather than a discrete presence/absence
239
[5]. In a study on rodents, males recognized similarities in the odors of brothers when
AC C
EP
TE D
229
The greater similarity detected between odor profiles of half siblings
11
ACCEPTED MANUSCRIPT
compared to an unrelated male [5]. In this study similarities were observed between the
241
odors of double cousins and cousins demonstrating that kinship from siblings to cousins
242
is reflected in odor similarity. It can also be postulated that similar to wild birds [43] and
243
primates [12], horses may be recognizing similar odor profiles and responding to
244
individuals based on the degrees of genetic relatedness and to identify kin/non-kin.
RI PT
240
It has been suggested that individual identification is one of the most important
246
cues used in vertebrate chemical communication [44]. The evaluation of relatedness may
247
be crucial in socio-sexual behavior of harem/herd animals like horses, especially if left to
248
random mating instead of artificial selection now used in domestic horse breeding
249
practices. Odor profiles can provide an array of information about the horse’s social,
250
reproductive, or health status and as was seen in this study, the animal’s identity, breed,
251
and kinship. It has been established that VOCs may play an important part in identifying
252
individuals, establishing dominance [45], signaling sexual readiness [11], facilitating
253
mate choice for genetically dissimilar individuals [46] and inbreeding avoidance [47].
254
The results in this study indicate that individual odor profiles could play a key role in
255
signaling individual characteristics, relatedness and that these volatile cues may possibly
256
be used for kin recognition in horses.
M AN U
TE D
EP
AC C
257
SC
245
258
COMPETING INTERESTS
259
The authors have no competing interests.
260
12
ACCEPTED MANUSCRIPT
Figure 1. The nMDS plot with similarity analysis overlay showed a tight grouping of
262
replicate samples from each horse. The overlapping or adjacent clusters of replicates for
263
each horse indicated that the volatiles obtained from hair samples were constant and
264
reproducible. The grouping of replicates also individualizes horses based on their VOC
265
profiles.
RI PT
261
SC
266
Figure 2. The nMDS plot with similarity analysis overlay grouped the horses according to
268
the two breeds included in study. Closed circles are Appaloosa horses while closed
269
triangles depict Quarter Horses.
270
grouping away from the 17 Quarter Horses demonstrating breed specific signal in VOC
271
profiles.
The nMDS plot shows the six Appaloosa horses
TE D
272
M AN U
267
Figure 3. The nMDS plot with similarity analysis overlay indicated that the stable in
274
which the horses were kept had little influence on the VOC profiles for each horse.
275
Horses grouped closer to related individuals rather than horses that were housed in the
276
same stables. This demonstrated little to no effect of housing environment on the VOC
277
profiles.
AC C
278
EP
273
279
Figure 4. A hierarchical cluster analyses and heatmap of the volatiles that differentiated
280
Appaloosa and Quarter Horse breeds, as well as each individual. The X-axis represents a
281
grouping of related and unrelated horses, while Y-axis are the 47 compounds contributing
282
to differences in breeds. White represents no difference in relative abundance and dark
13
ACCEPTED MANUSCRIPT
283
red represents maximum relative abundance. The heat map indicated a clear difference in
284
relative ratios rather
285
individualization of VOC for horses.
presence absence
of
compounds
contributing to
RI PT
than
286
Table 1. Information on the horses used in this study. Data on sex, relatedness, stable and
288
breed were noted and assessed in the study.
SC
287
289 REFERENCES
291
1. Wedekind, C., & Penn, D. (2000). MHC genes, body odours, and odour preferences.
292
Nephrology Dialysis Transplantation, 15(9), 1269-1271. doi:10.1093/ndt/15.9.1269
293
2. Shirasu, M., & Touhara, K. (2011). The scent of disease: Volatile organic compounds
294
of the human body related to disease and disorder. Journal of Biochemistry, 150(3),
295
257-266. doi:10.1093/jb/mvr090
TE D
296
M AN U
290
3. Eklund, A. (1997). The major histocompatibility complex and mating preferences in wild house mice (Mus domesticus). Behavioral Ecology, 8(6),
298
doi:10.1093/beheco/8.6.630
300 301
4. Todrank, J., Heth, G., & Johnston, R. E. (1998). Kin recognition in golden hamsters:
AC C
299
630-634.
EP
297
Evidence
for
kinship
odours.
Animal
Behaviour,
55(2),
377-386.
doi:10.1006/anbe.1997.0611
302
5. Busquet, N., & Baudoin, C. (2005). Odour similarities as a basis for discriminating
303
degrees of kinship in rodents: Evidence from Mus spicilegus. Animal Behaviour,
304
70(5), 997-1002. doi:10.1016/j.anbehav.2004.12.023
14
ACCEPTED MANUSCRIPT
6. Rajakaruna, R., Brown, J., Kaukinen, K., & Miller, K. (2006). Major
306
histocompatibility complex and kin discrimination in Atlantic salmon and brook
307
trout. Molecular Ecology, 15(14), 4569-4575. doi:10.1111/j.1365-294x.2006.03113.x
308
7. Olsson, M., Madsen, T., Nordby, J., Wapstra, E., Ujvari, B., & Wittsell, H. (2003).
309
Major histocompatibility complex and mate choice in sand lizards. Proceedings.
310
Biological
311
doi:10.1098/rsbl.2003.0079
/
the
Royal
Society,
270
Suppl
2,
S254-6.
SC
Sciences
RI PT
305
8. Zelano, B., & Edwards, S. V. (2002). An MHC component to kin recognition and
313
mate choice in birds: Predictions, progress, and prospects. The American Naturalist,
314
160(S6), S225-S237. doi:10.1086/342897
M AN U
312
9. Cutrera, A. P., Fanjul, M. S., & Zenuto, R. R. (2012). Females prefer good genes:
316
MHC-associated mate choice in wild and captive Tuco-tucos. Animal Behaviour,
317
83(3), 847-856. doi:10.1016/j.anbehav.2012.01.006
TE D
315
10. Kean, E. F., Chadwick, E. A., & Müller, C. T. (2015). Scent signals individual
319
identity and country of origin in otters. Mammalian Biology-Zeitschrift Für
320
Säugetierkunde, 80(2), 99-105. doi:10.1016/j.mambio.2014.12.004
322 323
11. Charpentier, M. J., Boulet, M., & Drea, C. M. (2008). Smelling right: The scent of
AC C
321
EP
318
male lemurs advertises genetic quality and relatedness. Molecular Ecology, 17(14),
3225-3233. doi:10.1093/beheco/ari034
324
12. Boulet, M., Charpentier, M. J., & Drea, C. M. (2009). Decoding an olfactory
325
mechanism of kin recognition and inbreeding avoidance in a primate. BMC
326
Evolutionary Biology, 9, 281-2148-9-281. doi:10.1186/1471-2148-9-281
15
ACCEPTED MANUSCRIPT
328 329 330
13. Wyatt, T. D. (2003). Pheromones and animal behaviour: Communication by smell and taste. Cambridge University Press. doi:10.1017/cbo9780511615061 14. Saslow, C. A. (2002). Understanding the perceptual world of horses. Applied Animal
RI PT
327
Behaviour Science, 78(2), 209-224. doi:10.1016/s0168-1591(02)00092-8
15. Kiley-Worthington, M. (1997). The behaviour of horses: In relation to management
332
and training. Applied Animal Ethology, 9(1), 99. doi:10.1016/0304-3762(82)90184-5
333
16. Hothersall, B., Harris, P., Sörtoft, L., & Nicol, C. J. (2010). Discrimination between
334
conspecific odour samples in the horse (Equus caballus). Applied Animal Behaviour
335
Science, 126(1), 37-44. doi:10.1016/j.applanim.2010.05.002
M AN U
SC
331
17. Péron, F., Ward, R., & Burman, O. (2014). Horses (Equus caballus) discriminate
337
body odour cues from conspecifics. Animal Cognition, 17(4), 1007-1011. doi:
338
10.1007/s10071-013-0717-9
TE D
336
18. Stahlbaum, C. C., & Houpt, K. A. (1989). The role of the flehmen response in the
340
behavioral repertoire of the stallion. Physiology & Behavior, 45(6), 1207-1214.
341
doi:10.1016/0031-9384(89)90111-x
EP
339
19. Kimura, R. (2001). Volatile substances in feces, urine and urine-marked feces of feral
343
horses. Canadian Journal of Animal Science, 81(3), 411-420. doi:10.4141/a00-068
344
20. Ainslie, T., & Ledbetter, B. (1980). The body language of horses. William Morrow,
345
AC C
342
New York.
346
21. Krueger, K., & Flauger, B. (2011). Olfactory recognition of individual competitors by
347
means of faeces in horse (Equus caballus). Animal Cognition, 14(2), 245-257.
348
doi:10.1007/s10071-010-0358-1
16
ACCEPTED MANUSCRIPT
22. Feist, J. D., & McCullough, D. R. (1976). Behavior patterns and communication in
350
feral horses. Zeitschrift Für Tierpsychologie, 41(4), 337-371. doi:10.1111/j.1439-
351
0310.1976.tb00947.x
352 353
RI PT
349
23. Rubenstein, D. I., & Hack, M. A. (1992). Horse signals: The sounds and scents of fury. Evolutionary Ecology, 6(3), 254-260. doi:10.1007/bf02214165
24. McDonnell, S. (2003). Practical field guide to horse behavior: The equid ethogram.
355
Journal of Equine Veterinary Science, 23(1), A1. doi:10.1016/s0737-0806(03)70087-
356
2
M AN U
SC
354
357
25. Ransom, J. I., & Cade, B. S. (2009). Quantifying equid behavior--A research
358
ethogram for free-roaming feral horses. Publications of the United States Geological
359
Survey, p 26.
26. Curran, A. M., Rabin, S. I., Prada, P. A., & Furton, K. G. (2005). Comparison of the
361
volatile organic compounds present in human odor using SPME-GC/MS. Journal of
362
Chemical Ecology, 31(7), 1607-1619. doi:10.1007/s10886-005-5801-4
TE D
360
27. Curran, A. M., Ramirez, C. F., Schoon, A. A., & Furton, K. G. (2007). The frequency
364
of occurrence and discriminatory power of compounds found in human scent across a
365
population determined by SPME-GC/MS. Journal of Chromatography B, 846(1), 86-
AC C
366
EP
363
97. doi:10.1016/j.jchromb.2006.08.039
367
28. Curran, A. M., Prada, P. A., & Furton, K. G. (2010). The differentiation of the
368
volatile organic signatures of individuals through SPME‐GC/MS of characteristic
369
human
370
doi:10.1111/j.1556-4029.2009.01236.x
scent
compounds.
Journal
of
Forensic
Sciences,
55(1),
50-57.
17
ACCEPTED MANUSCRIPT
29. Penn, D. J., Oberzaucher, E., Grammer, K., Fischer, G., Soini, H. A., Wiesler, D.,
372
Novotny, M.V., Dixon, S.J., Xu, Y., & Brereton, R. G. (2007). Individual and gender
373
fingerprints in human body odour. Journal of the Royal Society, Interface, 4(13), 331-
374
340. doi:10.1098/rsif.2006.0182
30. Pence, H. E., & Williams, A. (2010). ChemSpider: An online chemical information resource.
377
doi:10.1021/ed100697w
378 379 380
Journal
of
Chemical
Education,
87(11),
1123-1124.
SC
376
31. Clarke, K. R., & Gorley, R. N. (2015). PRIMER v7: User manual/tutorial PRIMER-
M AN U
375
RI PT
371
E, Plymouth, 296pp.
32. Clarke, K. R. (1993). Non ‐ parametric multivariate analyses of changes in
381
community
382
doi:10.1111/j.1442-9993.1993.tb00438.x
Australian
Journal
of
Ecology,
18(1),
117-143.
TE D
structure.
33. Ma, W., & Klemm, W. (1997). Variations of equine urinary volatile compounds
384
during the oestrous cycle. Veterinary Research Communications, 21(6), 437-446.
385
doi:10.1023/A:1005859419574
EP
383
34. Mozūraitis, R., Būda, V., Kutra, J., & Borg-Karlson, A. (2012). p-and m-cresols
387
emitted from estrous urine are reliable volatile chemical markers of ovulation in
388 389
AC C
386
mares.
Animal
Reproduction
Science,
130(1),
51-56.
doi:10.1016/j.anireprosci.2011.12.008
390
35. Palagi, E., & Dapporto, L. (2006). Beyond odor discrimination: Demonstrating
391
individual recognition by scent in Lemur catta. Chemical Senses, 31(5), 437-443.
392
doi:10.1093/chemse/bjj048
18
ACCEPTED MANUSCRIPT
36. Setchell, J. M., Vaglio, S., Abbott, K. M., Moggi-Cecchi, J., Boscaro, F., Pieraccini,
394
G., & Knapp, L. A. (2011). Odour signals major histocompatibility complex genotype
395
in an old world monkey. Proceedings. Biological Sciences / the Royal Society,
396
278(1703), 274-280. doi:10.1098/rspb.2010.0571
RI PT
393
37. Gassett, J., Wiesler, D., Baker, A., Osborn, D., Miller, K., Marchinton, R., &
398
Novotny, M. (1997). Volatile compounds from the forehead region of male white-
399
tailed deer (Odocoileus virginianus). Journal of Chemical Ecology, 23(3), 569-578.
400
doi:10.1023/b:joec.0000006397.63647.5b
M AN U
SC
397
38. Singer, A. G., Beauchamp, G. K., & Yamazaki, K. (1997). Volatile signals of the
402
major histocompatibility complex in male mouse urine. Proceedings of the National
403
Academy of Sciences of the United States of America, 94(6), 2210-2214.
404
doi:10.1073/pnas.94.6.2210
TE D
401
39. Celerier, A., Huchard, E., Alvergne, A., Fejan, D., Plard, F., Cowlishaw, G.,
406
Raymond, M., Knapp, L.A., & Bonadonna, F. (2010). Detective mice assess
407
relatedness in baboons using olfactory cues. The Journal of Experimental Biology,
408
213(Pt 9), 1399-1405. doi:10.1242/jeb.038349
410
40. Davies, M. J. (2008). Chemical communication in the European otter, Lutra lutra.
AC C
409
EP
405
Unpublished University of Hull
411
41. Tang-Martinez, Z. (2001). The mechanisms of kin discrimination and the evolution of
412
kin recognition in vertebrates: A critical re-evaluation. Behavioural Processes, 53(1-
413
2), 21-40. doi:10.1016/s0376-6357(00)00148-0
19
ACCEPTED MANUSCRIPT
414
42. Beecher, M. D. (1989). Signaling systems for individual recognition: An information
415
theory
416
3472(89)80087-9
Animal
Behaviour,
38(2),
248-261.
doi:10.1016/S0003-
RI PT
approach.
43. Bonadonna, F., & Sanz-Aguilar, A. (2012). Kin recognition and inbreeding avoidance
418
in wild birds: The first evidence for individual kin-related odour recognition. Animal
419
Behaviour, 84(3), 509-513. doi:10.1016/j.anbehav.2012.06.014
421
44. Wilson, E. O. (1970). Chemical communication within animal species. Chemical Ecology, 9, 133-155. doi:10.1016/b978-0-12-654750-4.50013-x
M AN U
420
SC
417
422
45. Marty, J. S., Higham, J. P., Gadsby, E. L., & Ross, C. (2009). Dominance, coloration,
423
and social and sexual behavior in male drills Mandrillus leucophaeus. International
424
Journal of Primatology, 30(6), 807-823. doi:10.1007/s10764-009-9382-x 46. Setchell, J. M., Vaglio, S., Moggi-Cecchi, J., Boscaro, F., Calamai, L., & Knapp, L.
426
A. (2010). Chemical composition of scent-gland secretions in an old world monkey
427
(Mandrillus sphinx): Influence of sex, male status, and individual identity. Chemical
428
Senses, 35(3), 205-220. doi:10.1093/chemse/bjp105
EP
TE D
425
47. Charpentier, M., Peignot, P., Hossaert-McKey, M., Gimenez, O., Setchell, J. M., &
430
Wickings, E. J. (2005). Constraints on control: Factors influencing reproductive
431 432
AC C
429
success in male mandrills (Mandrillus sphinx). Behavioral Ecology, 16(3), 614-623. doi:10.1093/beheco/ari034
20
ACCEPTED MANUSCRIPT
Table 1.
Sex
Known Dam
Dam
Sire
Related horses
Male
Dam 5
Sire 1
H4, H5
H2
Male
Dam 4
Sire 6
H3
H3
Female
Dam 4
Sire 7
H2
H4
Female
Dam 9
Sire 1
H1, H5
H5
Male
Dam 10
Sire 1
H1, H4
H6
Male
Dam 11 Sire 11
H7
Female
H8
Unknown
M AN U
Dam 1
Dam 12
Sire 8
H8, H9, H10
Male
Dam 1
Sire 2
H7, H9, H10
H9
Female
Dam 1
Sire 3
H7, H8, H10, H12
H10
Female
Dam 1
Sire 4
H7, H8, H9
H11
Female
Sire 9
H12
H12
Male
H13
Female
H14
Male
H15
Male
H16
Female
Dam 6
TE D
Dam 2
Dam 2
Sire 3
H11
Dam 7
Sire 10
Unknown
Dam 3
Sire 2
Unknown
Dam 19 Sire 18
H16, H17
Dam 8
Sire 5
H15, H17
AC C
EP
Dam 3
Stable 1 Stable 1 Stable 1 Stable 2 Stable 2 Stable 2 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 3 Stable 3 Stable 3 Stable 4 Stable 4 Stable 2 Stable 5 Stable 5
Breed
Appaloosa
SC
H1
Stable
RI PT
Identifier
H17
Female
Dam 8
Sire 5
H15, H16
H18
Female
Dam 17 Sire 16
Unknown
H19
Female
Dam 18 Sire 17
Unknown
H20
Male
Dam 13 Sire 12
Unknown
H21
Female
Dam 14 Sire 13
Unknown
H22
Male
Dam 15 Sire 14
Unknown
Quarter Horse
ACCEPTED MANUSCRIPT
Dam 16 Sire 15
Female
Stable 5
Unknown
AC C
EP
TE D
M AN U
SC
RI PT
H23
ACCEPTED MANUSCRIPT
Figure 1. The nMDS plot with similarity analysis overlay showed a tight grouping of replicate samples from each horse. The overlapping or adjacent clusters of replicates for each horse indicated that the volatiles obtained from hair samples were constant and
RI PT
reproducible. The grouping of replicates also individualizes horses based on their VOC
AC C
EP
TE D
M AN U
SC
profiles.
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12
H13 H14 H15 H16 H17 H18 H19 H20 H21 H22 H23
ACCEPTED MANUSCRIPT
Figure 2. The nMDS plot with similarity analysis overlay grouped the horses according to the two breeds included in study. Closed circles are Appaloosa horses while closed triangles depict Quarter Horses.
The nMDS plot shows the six Appaloosa horses
RI PT
grouping away from the 17 Quarter Horses demonstrating breed specific signal in VOC
AC C
EP
TE D
M AN U
SC
profiles.
ACCEPTED MANUSCRIPT
Figure 3. The nMDS plot with similarity analysis overlay indicated that the stable in which the horses were kept had little influence on the VOC profiles for each horse. Horses grouped closer to related individuals rather than horses that were housed in the
RI PT
same stables. This demonstrated little to no effect of housing environment on the VOC
AC C
EP
TE D
M AN U
SC
profiles.
ACCEPTED MANUSCRIPT
Figure 4. A hierarchical cluster analyses and heatmap of the volatiles that differentiated Appaloosa and Quarter Horse breeds as well as each individual. The X-axis represents a grouping of related and unrelated horses, while Y-axis are the 47 compounds contributing
RI PT
to differences in breeds. White represents no difference in relative abundance and dark red represents maximum relative abundance. The heat map indicated a clear difference in relative ratios rather
than
presence absence
AC C
EP
M AN U
TE D
Quarter Horse
compounds
SC
individualization of VOC for horses.
of
contributing to
Appaloosa
ACCEPTED MANUSCRIPT
HIGHLIGHTS Volatile organic compounds from hair samples are stable and reproducible over time
•
Volatile organic compounds collected from hair are able to differentiate horse breeds
•
Odor profiles are unique between cohorts of related and unrelated horses
•
Odor profiles show evidence and degree of kinship between individuals
•
Odor fingerprints are a product of qualitative and quantitative compound patterns
AC C
EP
TE D
M AN U
SC
RI PT
•
PII:
S0737-0806(17)30431-8
DOI:
10.1016/j.jevs.2017.05.013
Reference:
YJEVS 2331
To appear in:
Journal of Equine Veterinary Science
Received Date: 4 May 2017 Revised Date:
26 May 2017
Accepted Date: 29 May 2017
Please cite this article as: Deshpande K, Furton Kg, Mills Dek, The Equine Volatilome: Volatile Organic Compounds as Discriminatory Markers, Journal of Equine Veterinary Science (2017), doi: 10.1016/ j.jevs.2017.05.013. 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
SPME-GC-MS analysis of hair samples of domestic horses determines kinship as well as
AC C
EP
TE D
M AN U
SC
RI PT
provides evidence of breed specific signals based on volatile organic compound profiles.
ACCEPTED MANUSCRIPT
Title - THE EQUINE VOLATILOME: VOLATILE ORGANIC COMPOUNDS AS DISCRIMINATORY MARKERS
RI PT
Authors - KETAKI DESHPANDE 1,2, KENNETH G. FURTON 1,3 & DE ETTA K. MILLS* 1,2
International Forensic Research Institute; 2 Department of Biological Sciences;
3
Department of Chemistry and Biochemistry, Florida International University, Miami,
SC
1
M AN U
FL 33199 USA
Corresponding Author - DeEtta K. Mills, Florida International University. 11200 SW 8th Street, OE 167, Biological Sciences, Miami, FL 33199, United States.
ABSTRACT
TE D
Phone: 305-348-7410, Fax Number: 305-348-1986, E-mail: [email protected]
Background: Animals rely on subtle signals perceived between individuals that convey
EP
information such as sex, reproductive status, individual identity, ownership, competitive ability as well as health status. These cues have important influences on behaviors that
AC C
are vital for reproductive success, such as parent–offspring attachment, recognition of relatedness, mate choice and territorial marking.
Objective: This study investigates
individual odor profiles as discriminatory markers possibility affecting equine behavior. Methodology: This study investigated the volatile compounds in horses using solid phase microextraction gas chromatography-mass spectrometry and hair samples. Study design: Two horses breeds, Appaloosa (n=6) and Quarter Horses (n=17), were used as a model to
ACCEPTED MANUSCRIPT
assess the volatilome that may be used to identify individuals and kin recognition. Results: A total of 187 volatiles were identified and demonstrated that the volatilome carries information on genetic relatedness. Interestingly, apart from individualization and
RI PT
kinship, the volatiles detected were also able to discriminate between breeds. Main limitations: The study is limited exclusively two horse breeds and should be expanded to include other domestic and wild horses. Additionally, compounds were identified based
SC
on NIST 98 mass spectral library. Conclusions: The results of this study support the individual odor hypothesis suggesting that each individual possesses a unique scent that
M AN U
acts as a characteristic or odor fingerprint. Such qualitative and quantitative patterns have not been explored in domestic or wild animals and there is a need to understand more about the influence of odor on behavior and interactions.
AC C
EP
hair
TE D
Key Words – Kin; breed-specific; domestic; volatilome; Volatile organic compounds;
ACCEPTED MANUSCRIPT
ABSTRACT
2
Background: Animals rely on subtle signals perceived between individuals that convey
3
information such as sex, reproductive status, individual identity, ownership, competitive
4
ability as well as health status. These cues have important influences on behaviors that
5
are vital for reproductive success, such as parent–offspring attachment, recognition of
6
relatedness, mate choice and territorial marking.
7
individual odor profiles as discriminatory markers possibility affecting equine behavior.
8
Methodology: This study investigated the volatile compounds in horses using solid phase
9
microextraction gas chromatography-mass spectrometry and hair samples. Study design:
10
Two horse breeds, Appaloosa (n=6) and Quarter Horses (n=17), were used as a model to
11
assess the volatilome that may be used to identify individuals and kin recognition.
12
Results: A total of 187 volatiles were identified and demonstrated that the volatilome
13
carries information on genetic relatedness. Interestingly, apart from individualization and
14
kinship, the volatiles detected were also able to discriminate between breeds. Main
15
limitations: The study is limited exclusively two horse breeds and should be expanded to
16
include other domestic and wild horses. Additionally, compounds were identified based
17
on NIST 98 mass spectral library. Conclusions: The results of this study support the
18
individual odor hypothesis suggesting that each individual possesses a unique scent that
19
acts as a characteristic or odor fingerprint. Such qualitative and quantitative patterns
20
have not been explored in domestic or wild animals and there is a need to understand
21
more about the influence of odor on behavior and interactions.
RI PT
1
AC C
EP
TE D
M AN U
SC
Objective: This study investigates
1
ACCEPTED MANUSCRIPT
Key Words – Kin; breed-specific; domestic; volatilome; Volatile organic compounds;
23
hair
24
INTRODUCTION
RI PT
22
Olfactory cues function in the animal kingdom to distinguish between kin and
26
predators, detect food sources and even environmental toxins. The core body odors
27
(volatilome) are emitted volatile organic compounds (VOCs) that are the end products of
28
metabolism [1]. Factors such as genetics, diet, environment, the microbiome present on
29
the body, and even exogenous materials modulate the VOCs and can contribute to an
30
individual’s odor profile. Moreover, disease processes, such as infection, parasite load, or
31
endogenous metabolic disorders, can influence an individual’s odor profile by producing
32
different volatiles or by changing the proportions of VOCs that are normally produced
33
[2]; thus, they convey information about an individual’s metabolic or physiological
34
status. The olfactory differentiation for mating preferences and conspecific identification
35
or kin recognition has been studied across various taxonomic groups including rodents [3,
36
4,5] fish [6], Swedish sand lizards [7], birds [8], Tuco-tucos [9], otters [10] and lemurs
37
[11,12]. Although most of these studies did not focus on the individual components of the
38
odor, they led to the conclusion that individuals have a distinct body-odor type, which is
39
influenced by their inherited MHC alleles, and plays a pivotal role in kin recognition,
40
mate selection and identification of dissimilar or similar individuals [13].
M AN U
TE D
EP
AC C
41
SC
25
Horses, like all mammals, can recognize and discriminate chemical signals, which
42
provide essential information for individual and herd survival and greatly influence their
43
social behavior [14]. The large olfactory bulbs in a horse’s brain exhibit a convoluted
2
ACCEPTED MANUSCRIPT
surface, while the large size of the olfactory epithelium suggest that olfactory information
45
is vital to horses. Additionally, horses exhibit a well-developed vomeronasal organ that
46
is receptive to nonvolatile, large, species-specific molecules found in body secretions
47
[14]; hence, they have a highly developed olfactory capacity. Observational studies of
48
domestic and wild/feral horses have described how horses recognize each other on the
49
basis of body odors, urine, and feces [15,16,17]. Recognition at the individual level
50
guides the horse’s response based on previous experiences and determines the outcome of
51
the interactions. Close proximity mutual sniffing has been seen during horse greetings
52
and sexual advances that include blowing air on the face, standing parallel and sniffing
53
the neck and under another’s belly [18]. Horses often sniff excrement (dung piles) [19]
54
that allows horses to recognize other individuals [20,21] and to differentiate the sex of the
55
horse by its feces [18]. In fact, stallions create fecal piles known as stud piles and
56
repeatedly return to them to defecate as a territorial marking behavior [22,23]. Similar
57
practices are seen when the harem stallion covers an area with his urine where the harem
58
female horses have previously urinated or defecated [24,25]. Such greetings and scent
59
marking displayed by feral and domestic horses, suggest that odor is crucial in social
60
encounters and it is used to gather useful information from the chemical cues [20].
SC
M AN U
TE D
EP
AC C
61
RI PT
44
Curran et.al have proposed that individuals emanate “primary odor” VOCs that
62
are unique and stable over time regardless of diet or environmental factors and has
63
demonstrated using SPME/GC-MS that comparison of the presence and ratios of these
64
primary odor VOCs, individual humans can be differentiated [26,27,28]. Penn et al. have
65
proposed an “individual odor hypothesis” also suggesting that each individual possesses a
3
ACCEPTED MANUSCRIPT
unique VOC fingerprint [29]. The objective of the present study was to investigate odor
67
profiles from domestic horses, Equus caballus. Evidence for equine body volatiles that
68
are presumably used for individual discrimination is lacking; therefore, this study
69
examined the VOC profiles of domestic horses to determine the components of an
70
individual’s odor and whether these profiles can reliably indicate a degree kinship.
RI PT
66
METHODS AND MATERIALS
73
Sample Collection
M AN U
72
SC
71
Triplicate hair samples were donated by the horse owners and plucked from the
75
manes of 23 domestic horses (Table 1). Mane hair samples (each ≈10 mg) were collected
76
from each horse using sterile disposable gloves. The sample were immediately placed
77
into 10-mL glass clear, screw top vials (Supelco, Bellefonte, PA), sealed and allowed to
78
equilibrate for 24 h prior to SPME extraction. The usages of specific horse fly sprays,
79
feed, bathing routines were noted for all samples.
80
sampling and use of fly spray was avoided with the exception of Horse H5, who suffered
81
from insect bite hypersensitivity and was under treatment. Empty 10-mL vials (Supelco)
82
were exposed to the stable’s atmosphere for environmental VOC collection at each
83
sampling time. All the horses included in this study were fed the same feed and managed
84
in a similar fashion.
85
Solid-Phase Microextraction (SPME)- Gas Chromatography-Mass Spectrometry (GC-
86
MS) Procedures
Horses were not bathed before
AC C
EP
TE D
74
4
ACCEPTED MANUSCRIPT
Solid-Phase Microextraction (SPME) coupled with GC-MS is a sampling
88
technique that involves the use of a fiber coated with an extracting phase, that can capture
89
different analytes from samples. The GC works on the principle that a mixture will
90
separate into individual substances when heated, while mass spectrometry identifies
91
compounds by the mass of the analyte molecule. This enables the identification of VOCs,
92
which are secondary volatile metabolites and contribute to an individual’s unique odor
93
profile. Divinylbenzene/Carboxen on Polydimethylsiloxane (CAR/DVB on PDMS) 50/30
94
µm fibers (Supelco) were used to extract the VOCs from the headspace of the vials
95
containing the hair samples and environmental blanks. Fiber exposure was conducted at
96
room temperature for 12 h. The samples were separated and analyzed by GC-MS using
97
an Agilent 6890 GC with a 5973 MS. A HP5-MS column, with helium as the carrier gas
98
at flow rate of 1.0 mL/min for the separation of the analytes was used. The extracted
99
VOCs were desorbed in the injection port at 250°C for 10 min in splitless mode. The
100
temperature program was: initial oven temperature of 40°C for 5 min, 10°C per minute
101
ramp to a final temperature of 250°C, and a final hold for 2 min for a total run time of 32
102
min. The mass spectrometer used was an HP 5973 MSD with a quadrupole analyzer in
103
full scan mode. The compounds were identified using the NIST 98 mass spectral library.
104
The criterion for the identification of compounds was based on the quality of the detected
105
peak, set at greater than or equal to 40%. Additionally, compounds detected in all three
106
replicates at the same retention time were considered as part of an individual’s
107
volatilome.
108
Statistical Analysis
AC C
EP
TE D
M AN U
SC
RI PT
87
5
ACCEPTED MANUSCRIPT
Prior to statistical analyses, data were reduced by removing compounds present in
110
the environmental blanks and in Pyranha fly spray based on the MSDS description.
111
Further, compound names were searched on ChemSpider [30] to assess if they were
112
related to equine husbandry practices or presence in the environment.
113
commonly associated with ointments and plant/feed related compounds were removed
114
from the analysis. The relative ratio of each VOC’s abundance was calculated for all
115
VOCs and, subsequently transformed using square-root transformation.
116
Spearman Rank Correlation
RI PT
109
M AN U
SC
Compounds
117
Each horse’s VOC profile was statistically evaluated to determine the similarity
118
between each replicate. The replicate profiles for each horse were then averaged to
119
produce a single representative VOC profile.
120
correlated, in a pair wise manner, to the rest of the horses using Spearman rank
121
correlations.
122
Multivariate Analyses of Volatiles
TE D
Each individual VOC profile was
A non-metric multidimensional scaling (nMDS) plot is a visualization of the
124
pattern of proximities based on similarities or distances among the objects. The analysis
125
was performed in order to identify similarity/differences of (i) replicates from each horse
126
(ii) horse breeds and (iii) kinship. The analyses were performed using Primer-E ver 7
127
software [31] to create a similarity matrix among variables using the Bray-Curtis
128
similarity coefficient. The relative ratios were used for hierarchical cluster analysis
129
(HCA) using the complete linkage method. In order to observe percent similarity, the
AC C
EP
123
6
ACCEPTED MANUSCRIPT
cluster analysis output was overlaid on the nMDS ordination plots, indicated by circles
131
grouping the data.
132
Differences in composition of compounds within groups were tested using an Analysis of
133
Similarities (ANOSIM) [32]. Similarity Percentage analysis (SIMPER) was performed to
134
determine which compounds influenced the discrimination observed between the
135
individuals, breeds and kin. Heatmaps were generated to visualize the contribution of top
136
47 compounds obtained in SIMPER analyses. The heatmap is based on relative ratio of
137
compounds where the regions of white represent absence of a compound, blue represents
138
low relative abundance while dark red indicates higher relative abundance with highest
139
value of 0.52.
M AN U
SC
RI PT
130
140 RESULTS
TE D
141
The hair samples revealed compounds from various functional groups: alkanes,
143
alcohols, hydrocarbons, aldehydes, alkenes, amines/amides, ketones, esters, and indols
144
(n=187).
Three VOC with varying abundances were detected in 100% of horses
145
sampled;
Nonanal,
146
Dibromo-2-Methyl- Undecane. Ninety-eight compounds were specific to Quarter Horses;
147
however, no one compound was common across all 17 Quarter Horses.
148
Spearman Rank Correlations
Fluoren-9-Ol,
3,6-Dimethoxy-9-(2-Phenylethynyl)-,
and
1,2-
AC C
149
EP
142
Among horses of known relationships (parent-offspring, full sibling sharing dam
150
and sire, half sibling sharing dam or sire), correlation coefficient of ≥ 0.7 (S1) was
151
observed. These values demonstrate that, though there are qualitative similarities with
7
ACCEPTED MANUSCRIPT
presence/absence of VOCs between horses, the difference in relative abundance ratios
153
allowed for a higher percentage of discrimination between related and unrelated horses.
154
Multivariate Analysis
RI PT
152
The nMDS plot (Figure 1) with a stress value of 0.15 and 80% similarity shows
156
the VOC replicates for each horse and indicated that VOC from hair samples were stable
157
over time and reproducible. In support of the ‘individual odor hypothesis’, there was a
158
significant difference among each horse’s profile (R = 0.90; P < 0.001) in comparison to
159
other horses. The nMDS plot with a stress value of 0.15 (Figure 2) using the averages
160
showed a clear separation of Appaloosa and Quarter Horse breeds (R=0.814; P < 0.001).
161
The 40% similarity grouping was observed for related horses while a 20% similarity was
162
seen between horse breeds.
M AN U
SC
155
The nMDS plot (Figure 3) with a stress value of 0.15 indicated that the sampling
164
site (stables) had minimal effect on grouping of the horses when related individuals are
165
included in the study. This suggests that animal husbandry practices did not have an
166
influence on the VOC profiles. Overall, plots showed that VOC profiles are different for
167
each horse and consequently, one can individualize them. The HCA clearly grouped the
168
Appaloosas separate from the Quarter Horses demonstrating breed specific patterns.
169
Although clustering of related horses was quite evident, clustering of distant relationships
170
was also observed for horses H15 – H16 – H17, were H15 is the uncle to H16 and H17
171
from the dam’s lineage.
AC C
EP
TE D
163
172
SIMPER analyses of the dominant 47 compounds were able to discriminate the
173
Appaloosa and Quarter Horse breeds (S2). The average dissimilarity between the breeds
8
ACCEPTED MANUSCRIPT
was 77.03%. Using these 47 compounds, the heatmap (Figure 4) indicated the relative
175
ratio differences that provide individual uniqueness and the clustering of the two breeds.
176
The heatmap revealed that, rather than just a simple presence or absence of compounds,
177
the relative ratio of compounds differed across the horses sampled.
RI PT
174
178 DISCUSSION
SC
179
The potential role of olfaction in horses is known by a relatively small number of
181
published studies concentrated primarily on the role of odor in horse mating behavior and
182
sex identification. Such studies have focused on volatile chemical markers from feces
183
and urine samples that fluctuate with ovulation or the estrus cycle in mares [33,34] and
184
can elicit a response from stallions. The current study targeted odor profiles from hair
185
samples in order to learn whether particular individuals have a specific odor profile.
186
Cumulative VOC profiles considering presence/absence of compounds along with
187
relative ratios can successfully individualize horses supporting the “individual odor
188
hypothesis” [29]. As has been seen in vertebrate species like lemurs [35], mandrills [36],
189
deer [37], and now the odor profiles in horses, a high percentage of chemical compounds
190
were shared among profiles of all horses. However, the significant differences between
191
each individual suggested that the uniqueness in the chemical profile depends more on
192
the concentration of compounds and on complex interactions between compounds than
193
on the simple presence or absence of specific chemicals compounds [28,38]. Using
194
headspace SPME method Curran et al., further showed that a combination of the relative
195
ratios of compounds and the presence of differing compounds allows for the
AC C
EP
TE D
M AN U
180
9
ACCEPTED MANUSCRIPT
chromatographic distinction among individuals [27]. Curran et al. found that compounds
197
were present in differing ratio patterns between the males and females, thus indicating
198
qualitative similarities among individuals with quantitative differences [28].
RI PT
196
In a study conducted by Celerier et al., it was shown that mice were able to detect
200
baboons based on individual odor differences and were able to perceive a higher odor
201
similarity between related baboons than between unrelated baboons. The ability of mice
202
to discriminate other mammal species based on their odor suggests that odors may play a
203
role in both the signaling of individual characteristics and of relatedness among
204
individuals [39]. This supports the observed variation in volatiles between horses and
205
was indicative of an individual-specific odor that could aid in individual identification,
206
which may influence the behavioral responses. In this study, it is interesting to observe
207
that volatiles patterns differentiated between breeds even when stabled in different
208
locations. Horses that were housed at the same location but not related demonstrated that
209
housing and management appeared to have little influence on their volatile profiles. This
210
is contrary to the findings in otters housed in the same location where the similarity was
211
attributed to common diet as a potential explanation [40]. Grouping of horses with
212
unknown relationships may be due to variables that were not investigated in this study,
213
for example, age, current reproductive or health status of the individuals.
M AN U
TE D
EP
AC C
214
SC
199
From an evolutionary standpoint, kin recognition aids in parental care, kin
215
altruism, inbreeding avoidance, and maintenance of optimal outbreeding.
216
choice, most animals mate with unrelated or distant relatives. This innate selection is
217
thought to be evolutionarily favored, as it should improve their inclusive fitness and
Given a
10
ACCEPTED MANUSCRIPT
decrease the effects of inbreeding on the population [12]. In lemurs it was observed that
219
chemical cues coded genetic relatedness within and between sexes. Additionally, these
220
chemicals showed a seasonal pattern demonstrating variation in accordance to the
221
competitive breeding season of lemurs. Boulet et al., suggest that a strong olfactory
222
signal of relatedness during breeding seasons would be crucial for preventing inbreeding
223
[12]. This manner of discrimination can be treated as a communication mechanism
224
where prior to identification, the animal must be able to receive relevant signals from
225
other individuals and distinguish appropriately between them [41].
226
provided the individual odor cues vary. Therefore, studying body volatiles may enable
227
one to better understand how animals use kin-biased behavior in their interactions within
228
breeding populations [42].
This can occur
M AN U
SC
RI PT
218
When considering relationships in horses, breed lineages need to be queried rather
230
than just assessing direct descendants such as parent-offspring or siblings. Such closely
231
line bred pedigrees as seen in today’s domestic horses could group horses that might not
232
share a direct relationship through a sire or dam. Even though all sire and dam pairs were
233
not known in the study, VOC profiles were able to demonstrate relatedness between
234
horses with known lineages. Although the closely related horses had similar chemical
235
components, it cannot be said that horses from the same kin group share the exact odor
236
phenotype.
237
compared to distant kin indicated a graded or continuous relationship of the volatilome
238
within kin groups when using relative abundances rather than a discrete presence/absence
239
[5]. In a study on rodents, males recognized similarities in the odors of brothers when
AC C
EP
TE D
229
The greater similarity detected between odor profiles of half siblings
11
ACCEPTED MANUSCRIPT
compared to an unrelated male [5]. In this study similarities were observed between the
241
odors of double cousins and cousins demonstrating that kinship from siblings to cousins
242
is reflected in odor similarity. It can also be postulated that similar to wild birds [43] and
243
primates [12], horses may be recognizing similar odor profiles and responding to
244
individuals based on the degrees of genetic relatedness and to identify kin/non-kin.
RI PT
240
It has been suggested that individual identification is one of the most important
246
cues used in vertebrate chemical communication [44]. The evaluation of relatedness may
247
be crucial in socio-sexual behavior of harem/herd animals like horses, especially if left to
248
random mating instead of artificial selection now used in domestic horse breeding
249
practices. Odor profiles can provide an array of information about the horse’s social,
250
reproductive, or health status and as was seen in this study, the animal’s identity, breed,
251
and kinship. It has been established that VOCs may play an important part in identifying
252
individuals, establishing dominance [45], signaling sexual readiness [11], facilitating
253
mate choice for genetically dissimilar individuals [46] and inbreeding avoidance [47].
254
The results in this study indicate that individual odor profiles could play a key role in
255
signaling individual characteristics, relatedness and that these volatile cues may possibly
256
be used for kin recognition in horses.
M AN U
TE D
EP
AC C
257
SC
245
258
COMPETING INTERESTS
259
The authors have no competing interests.
260
12
ACCEPTED MANUSCRIPT
Figure 1. The nMDS plot with similarity analysis overlay showed a tight grouping of
262
replicate samples from each horse. The overlapping or adjacent clusters of replicates for
263
each horse indicated that the volatiles obtained from hair samples were constant and
264
reproducible. The grouping of replicates also individualizes horses based on their VOC
265
profiles.
RI PT
261
SC
266
Figure 2. The nMDS plot with similarity analysis overlay grouped the horses according to
268
the two breeds included in study. Closed circles are Appaloosa horses while closed
269
triangles depict Quarter Horses.
270
grouping away from the 17 Quarter Horses demonstrating breed specific signal in VOC
271
profiles.
The nMDS plot shows the six Appaloosa horses
TE D
272
M AN U
267
Figure 3. The nMDS plot with similarity analysis overlay indicated that the stable in
274
which the horses were kept had little influence on the VOC profiles for each horse.
275
Horses grouped closer to related individuals rather than horses that were housed in the
276
same stables. This demonstrated little to no effect of housing environment on the VOC
277
profiles.
AC C
278
EP
273
279
Figure 4. A hierarchical cluster analyses and heatmap of the volatiles that differentiated
280
Appaloosa and Quarter Horse breeds, as well as each individual. The X-axis represents a
281
grouping of related and unrelated horses, while Y-axis are the 47 compounds contributing
282
to differences in breeds. White represents no difference in relative abundance and dark
13
ACCEPTED MANUSCRIPT
283
red represents maximum relative abundance. The heat map indicated a clear difference in
284
relative ratios rather
285
individualization of VOC for horses.
presence absence
of
compounds
contributing to
RI PT
than
286
Table 1. Information on the horses used in this study. Data on sex, relatedness, stable and
288
breed were noted and assessed in the study.
SC
287
289 REFERENCES
291
1. Wedekind, C., & Penn, D. (2000). MHC genes, body odours, and odour preferences.
292
Nephrology Dialysis Transplantation, 15(9), 1269-1271. doi:10.1093/ndt/15.9.1269
293
2. Shirasu, M., & Touhara, K. (2011). The scent of disease: Volatile organic compounds
294
of the human body related to disease and disorder. Journal of Biochemistry, 150(3),
295
257-266. doi:10.1093/jb/mvr090
TE D
296
M AN U
290
3. Eklund, A. (1997). The major histocompatibility complex and mating preferences in wild house mice (Mus domesticus). Behavioral Ecology, 8(6),
298
doi:10.1093/beheco/8.6.630
300 301
4. Todrank, J., Heth, G., & Johnston, R. E. (1998). Kin recognition in golden hamsters:
AC C
299
630-634.
EP
297
Evidence
for
kinship
odours.
Animal
Behaviour,
55(2),
377-386.
doi:10.1006/anbe.1997.0611
302
5. Busquet, N., & Baudoin, C. (2005). Odour similarities as a basis for discriminating
303
degrees of kinship in rodents: Evidence from Mus spicilegus. Animal Behaviour,
304
70(5), 997-1002. doi:10.1016/j.anbehav.2004.12.023
14
ACCEPTED MANUSCRIPT
6. Rajakaruna, R., Brown, J., Kaukinen, K., & Miller, K. (2006). Major
306
histocompatibility complex and kin discrimination in Atlantic salmon and brook
307
trout. Molecular Ecology, 15(14), 4569-4575. doi:10.1111/j.1365-294x.2006.03113.x
308
7. Olsson, M., Madsen, T., Nordby, J., Wapstra, E., Ujvari, B., & Wittsell, H. (2003).
309
Major histocompatibility complex and mate choice in sand lizards. Proceedings.
310
Biological
311
doi:10.1098/rsbl.2003.0079
/
the
Royal
Society,
270
Suppl
2,
S254-6.
SC
Sciences
RI PT
305
8. Zelano, B., & Edwards, S. V. (2002). An MHC component to kin recognition and
313
mate choice in birds: Predictions, progress, and prospects. The American Naturalist,
314
160(S6), S225-S237. doi:10.1086/342897
M AN U
312
9. Cutrera, A. P., Fanjul, M. S., & Zenuto, R. R. (2012). Females prefer good genes:
316
MHC-associated mate choice in wild and captive Tuco-tucos. Animal Behaviour,
317
83(3), 847-856. doi:10.1016/j.anbehav.2012.01.006
TE D
315
10. Kean, E. F., Chadwick, E. A., & Müller, C. T. (2015). Scent signals individual
319
identity and country of origin in otters. Mammalian Biology-Zeitschrift Für
320
Säugetierkunde, 80(2), 99-105. doi:10.1016/j.mambio.2014.12.004
322 323
11. Charpentier, M. J., Boulet, M., & Drea, C. M. (2008). Smelling right: The scent of
AC C
321
EP
318
male lemurs advertises genetic quality and relatedness. Molecular Ecology, 17(14),
3225-3233. doi:10.1093/beheco/ari034
324
12. Boulet, M., Charpentier, M. J., & Drea, C. M. (2009). Decoding an olfactory
325
mechanism of kin recognition and inbreeding avoidance in a primate. BMC
326
Evolutionary Biology, 9, 281-2148-9-281. doi:10.1186/1471-2148-9-281
15
ACCEPTED MANUSCRIPT
328 329 330
13. Wyatt, T. D. (2003). Pheromones and animal behaviour: Communication by smell and taste. Cambridge University Press. doi:10.1017/cbo9780511615061 14. Saslow, C. A. (2002). Understanding the perceptual world of horses. Applied Animal
RI PT
327
Behaviour Science, 78(2), 209-224. doi:10.1016/s0168-1591(02)00092-8
15. Kiley-Worthington, M. (1997). The behaviour of horses: In relation to management
332
and training. Applied Animal Ethology, 9(1), 99. doi:10.1016/0304-3762(82)90184-5
333
16. Hothersall, B., Harris, P., Sörtoft, L., & Nicol, C. J. (2010). Discrimination between
334
conspecific odour samples in the horse (Equus caballus). Applied Animal Behaviour
335
Science, 126(1), 37-44. doi:10.1016/j.applanim.2010.05.002
M AN U
SC
331
17. Péron, F., Ward, R., & Burman, O. (2014). Horses (Equus caballus) discriminate
337
body odour cues from conspecifics. Animal Cognition, 17(4), 1007-1011. doi:
338
10.1007/s10071-013-0717-9
TE D
336
18. Stahlbaum, C. C., & Houpt, K. A. (1989). The role of the flehmen response in the
340
behavioral repertoire of the stallion. Physiology & Behavior, 45(6), 1207-1214.
341
doi:10.1016/0031-9384(89)90111-x
EP
339
19. Kimura, R. (2001). Volatile substances in feces, urine and urine-marked feces of feral
343
horses. Canadian Journal of Animal Science, 81(3), 411-420. doi:10.4141/a00-068
344
20. Ainslie, T., & Ledbetter, B. (1980). The body language of horses. William Morrow,
345
AC C
342
New York.
346
21. Krueger, K., & Flauger, B. (2011). Olfactory recognition of individual competitors by
347
means of faeces in horse (Equus caballus). Animal Cognition, 14(2), 245-257.
348
doi:10.1007/s10071-010-0358-1
16
ACCEPTED MANUSCRIPT
22. Feist, J. D., & McCullough, D. R. (1976). Behavior patterns and communication in
350
feral horses. Zeitschrift Für Tierpsychologie, 41(4), 337-371. doi:10.1111/j.1439-
351
0310.1976.tb00947.x
352 353
RI PT
349
23. Rubenstein, D. I., & Hack, M. A. (1992). Horse signals: The sounds and scents of fury. Evolutionary Ecology, 6(3), 254-260. doi:10.1007/bf02214165
24. McDonnell, S. (2003). Practical field guide to horse behavior: The equid ethogram.
355
Journal of Equine Veterinary Science, 23(1), A1. doi:10.1016/s0737-0806(03)70087-
356
2
M AN U
SC
354
357
25. Ransom, J. I., & Cade, B. S. (2009). Quantifying equid behavior--A research
358
ethogram for free-roaming feral horses. Publications of the United States Geological
359
Survey, p 26.
26. Curran, A. M., Rabin, S. I., Prada, P. A., & Furton, K. G. (2005). Comparison of the
361
volatile organic compounds present in human odor using SPME-GC/MS. Journal of
362
Chemical Ecology, 31(7), 1607-1619. doi:10.1007/s10886-005-5801-4
TE D
360
27. Curran, A. M., Ramirez, C. F., Schoon, A. A., & Furton, K. G. (2007). The frequency
364
of occurrence and discriminatory power of compounds found in human scent across a
365
population determined by SPME-GC/MS. Journal of Chromatography B, 846(1), 86-
AC C
366
EP
363
97. doi:10.1016/j.jchromb.2006.08.039
367
28. Curran, A. M., Prada, P. A., & Furton, K. G. (2010). The differentiation of the
368
volatile organic signatures of individuals through SPME‐GC/MS of characteristic
369
human
370
doi:10.1111/j.1556-4029.2009.01236.x
scent
compounds.
Journal
of
Forensic
Sciences,
55(1),
50-57.
17
ACCEPTED MANUSCRIPT
29. Penn, D. J., Oberzaucher, E., Grammer, K., Fischer, G., Soini, H. A., Wiesler, D.,
372
Novotny, M.V., Dixon, S.J., Xu, Y., & Brereton, R. G. (2007). Individual and gender
373
fingerprints in human body odour. Journal of the Royal Society, Interface, 4(13), 331-
374
340. doi:10.1098/rsif.2006.0182
30. Pence, H. E., & Williams, A. (2010). ChemSpider: An online chemical information resource.
377
doi:10.1021/ed100697w
378 379 380
Journal
of
Chemical
Education,
87(11),
1123-1124.
SC
376
31. Clarke, K. R., & Gorley, R. N. (2015). PRIMER v7: User manual/tutorial PRIMER-
M AN U
375
RI PT
371
E, Plymouth, 296pp.
32. Clarke, K. R. (1993). Non ‐ parametric multivariate analyses of changes in
381
community
382
doi:10.1111/j.1442-9993.1993.tb00438.x
Australian
Journal
of
Ecology,
18(1),
117-143.
TE D
structure.
33. Ma, W., & Klemm, W. (1997). Variations of equine urinary volatile compounds
384
during the oestrous cycle. Veterinary Research Communications, 21(6), 437-446.
385
doi:10.1023/A:1005859419574
EP
383
34. Mozūraitis, R., Būda, V., Kutra, J., & Borg-Karlson, A. (2012). p-and m-cresols
387
emitted from estrous urine are reliable volatile chemical markers of ovulation in
388 389
AC C
386
mares.
Animal
Reproduction
Science,
130(1),
51-56.
doi:10.1016/j.anireprosci.2011.12.008
390
35. Palagi, E., & Dapporto, L. (2006). Beyond odor discrimination: Demonstrating
391
individual recognition by scent in Lemur catta. Chemical Senses, 31(5), 437-443.
392
doi:10.1093/chemse/bjj048
18
ACCEPTED MANUSCRIPT
36. Setchell, J. M., Vaglio, S., Abbott, K. M., Moggi-Cecchi, J., Boscaro, F., Pieraccini,
394
G., & Knapp, L. A. (2011). Odour signals major histocompatibility complex genotype
395
in an old world monkey. Proceedings. Biological Sciences / the Royal Society,
396
278(1703), 274-280. doi:10.1098/rspb.2010.0571
RI PT
393
37. Gassett, J., Wiesler, D., Baker, A., Osborn, D., Miller, K., Marchinton, R., &
398
Novotny, M. (1997). Volatile compounds from the forehead region of male white-
399
tailed deer (Odocoileus virginianus). Journal of Chemical Ecology, 23(3), 569-578.
400
doi:10.1023/b:joec.0000006397.63647.5b
M AN U
SC
397
38. Singer, A. G., Beauchamp, G. K., & Yamazaki, K. (1997). Volatile signals of the
402
major histocompatibility complex in male mouse urine. Proceedings of the National
403
Academy of Sciences of the United States of America, 94(6), 2210-2214.
404
doi:10.1073/pnas.94.6.2210
TE D
401
39. Celerier, A., Huchard, E., Alvergne, A., Fejan, D., Plard, F., Cowlishaw, G.,
406
Raymond, M., Knapp, L.A., & Bonadonna, F. (2010). Detective mice assess
407
relatedness in baboons using olfactory cues. The Journal of Experimental Biology,
408
213(Pt 9), 1399-1405. doi:10.1242/jeb.038349
410
40. Davies, M. J. (2008). Chemical communication in the European otter, Lutra lutra.
AC C
409
EP
405
Unpublished University of Hull
411
41. Tang-Martinez, Z. (2001). The mechanisms of kin discrimination and the evolution of
412
kin recognition in vertebrates: A critical re-evaluation. Behavioural Processes, 53(1-
413
2), 21-40. doi:10.1016/s0376-6357(00)00148-0
19
ACCEPTED MANUSCRIPT
414
42. Beecher, M. D. (1989). Signaling systems for individual recognition: An information
415
theory
416
3472(89)80087-9
Animal
Behaviour,
38(2),
248-261.
doi:10.1016/S0003-
RI PT
approach.
43. Bonadonna, F., & Sanz-Aguilar, A. (2012). Kin recognition and inbreeding avoidance
418
in wild birds: The first evidence for individual kin-related odour recognition. Animal
419
Behaviour, 84(3), 509-513. doi:10.1016/j.anbehav.2012.06.014
421
44. Wilson, E. O. (1970). Chemical communication within animal species. Chemical Ecology, 9, 133-155. doi:10.1016/b978-0-12-654750-4.50013-x
M AN U
420
SC
417
422
45. Marty, J. S., Higham, J. P., Gadsby, E. L., & Ross, C. (2009). Dominance, coloration,
423
and social and sexual behavior in male drills Mandrillus leucophaeus. International
424
Journal of Primatology, 30(6), 807-823. doi:10.1007/s10764-009-9382-x 46. Setchell, J. M., Vaglio, S., Moggi-Cecchi, J., Boscaro, F., Calamai, L., & Knapp, L.
426
A. (2010). Chemical composition of scent-gland secretions in an old world monkey
427
(Mandrillus sphinx): Influence of sex, male status, and individual identity. Chemical
428
Senses, 35(3), 205-220. doi:10.1093/chemse/bjp105
EP
TE D
425
47. Charpentier, M., Peignot, P., Hossaert-McKey, M., Gimenez, O., Setchell, J. M., &
430
Wickings, E. J. (2005). Constraints on control: Factors influencing reproductive
431 432
AC C
429
success in male mandrills (Mandrillus sphinx). Behavioral Ecology, 16(3), 614-623. doi:10.1093/beheco/ari034
20
ACCEPTED MANUSCRIPT
Table 1.
Sex
Known Dam
Dam
Sire
Related horses
Male
Dam 5
Sire 1
H4, H5
H2
Male
Dam 4
Sire 6
H3
H3
Female
Dam 4
Sire 7
H2
H4
Female
Dam 9
Sire 1
H1, H5
H5
Male
Dam 10
Sire 1
H1, H4
H6
Male
Dam 11 Sire 11
H7
Female
H8
Unknown
M AN U
Dam 1
Dam 12
Sire 8
H8, H9, H10
Male
Dam 1
Sire 2
H7, H9, H10
H9
Female
Dam 1
Sire 3
H7, H8, H10, H12
H10
Female
Dam 1
Sire 4
H7, H8, H9
H11
Female
Sire 9
H12
H12
Male
H13
Female
H14
Male
H15
Male
H16
Female
Dam 6
TE D
Dam 2
Dam 2
Sire 3
H11
Dam 7
Sire 10
Unknown
Dam 3
Sire 2
Unknown
Dam 19 Sire 18
H16, H17
Dam 8
Sire 5
H15, H17
AC C
EP
Dam 3
Stable 1 Stable 1 Stable 1 Stable 2 Stable 2 Stable 2 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 1 Stable 3 Stable 3 Stable 3 Stable 4 Stable 4 Stable 2 Stable 5 Stable 5
Breed
Appaloosa
SC
H1
Stable
RI PT
Identifier
H17
Female
Dam 8
Sire 5
H15, H16
H18
Female
Dam 17 Sire 16
Unknown
H19
Female
Dam 18 Sire 17
Unknown
H20
Male
Dam 13 Sire 12
Unknown
H21
Female
Dam 14 Sire 13
Unknown
H22
Male
Dam 15 Sire 14
Unknown
Quarter Horse
ACCEPTED MANUSCRIPT
Dam 16 Sire 15
Female
Stable 5
Unknown
AC C
EP
TE D
M AN U
SC
RI PT
H23
ACCEPTED MANUSCRIPT
Figure 1. The nMDS plot with similarity analysis overlay showed a tight grouping of replicate samples from each horse. The overlapping or adjacent clusters of replicates for each horse indicated that the volatiles obtained from hair samples were constant and
RI PT
reproducible. The grouping of replicates also individualizes horses based on their VOC
AC C
EP
TE D
M AN U
SC
profiles.
H1 H2 H3 H4 H5 H6 H7 H8 H9 H10 H11 H12
H13 H14 H15 H16 H17 H18 H19 H20 H21 H22 H23
ACCEPTED MANUSCRIPT
Figure 2. The nMDS plot with similarity analysis overlay grouped the horses according to the two breeds included in study. Closed circles are Appaloosa horses while closed triangles depict Quarter Horses.
The nMDS plot shows the six Appaloosa horses
RI PT
grouping away from the 17 Quarter Horses demonstrating breed specific signal in VOC
AC C
EP
TE D
M AN U
SC
profiles.
ACCEPTED MANUSCRIPT
Figure 3. The nMDS plot with similarity analysis overlay indicated that the stable in which the horses were kept had little influence on the VOC profiles for each horse. Horses grouped closer to related individuals rather than horses that were housed in the
RI PT
same stables. This demonstrated little to no effect of housing environment on the VOC
AC C
EP
TE D
M AN U
SC
profiles.
ACCEPTED MANUSCRIPT
Figure 4. A hierarchical cluster analyses and heatmap of the volatiles that differentiated Appaloosa and Quarter Horse breeds as well as each individual. The X-axis represents a grouping of related and unrelated horses, while Y-axis are the 47 compounds contributing
RI PT
to differences in breeds. White represents no difference in relative abundance and dark red represents maximum relative abundance. The heat map indicated a clear difference in relative ratios rather
than
presence absence
AC C
EP
M AN U
TE D
Quarter Horse
compounds
SC
individualization of VOC for horses.
of
contributing to
Appaloosa
ACCEPTED MANUSCRIPT
HIGHLIGHTS Volatile organic compounds from hair samples are stable and reproducible over time
•
Volatile organic compounds collected from hair are able to differentiate horse breeds
•
Odor profiles are unique between cohorts of related and unrelated horses
•
Odor profiles show evidence and degree of kinship between individuals
•
Odor fingerprints are a product of qualitative and quantitative compound patterns
AC C
EP
TE D
M AN U
SC
RI PT
•