Introduction
Molecular genetics show great potential for use in con-
servation biology and wildlife management. One such
application involves the development of species-specific
genetic markers that can be detected noninvasively from
shed integument, regurgitated items, or defecation pro-
ducts (H?ss et al. 1992; Taberlet & Bouvet 1992; Constable
et al. 1995; Gerloff et al. 1995). These markers can then be
used to identify and survey elusive species or individuals.
Endangered San Joaquin kit foxes Vulpes macrotis mutica in
California?s San Joaquin Valley are often sympatrically
distributed with the native gray fox Urocyon cinereoar-
genteus, the introduced red fox Vulpes vulpes, the coyote
Canis latrans, and the domestic dog Canis familiaris. The fox
species may be secretive and difficult to detect with standard
census methods (Orloff 1992), especially at low population
densities. In addition, fox scats found in the field cannot be
reliably identified to species on the basis of appearance or
morphometrics (Halfpenny 1986; Clifton 1992). In order to
improve detection of endangered kit fox and therefore
enhance conservation and management strategies, we
developed a system of restriction enzyme digestion of
PCR-amplified mitochondrial DNA that can be easily and
reliably amplified from scats found during routine field-
work.
Materials and Methods
Fox tissue or DNA samples (n = 21) and scats (n = 10) were
obtained from field sites, museum tissue collections, and
zoos. The identities of the tissue and DNA samples were
known (four gray fox, four red fox and 13 kit fox), but the
10 scat samples were provided to the laboratory as
?unknowns? in order to test the reliability of our extraction
and diagnosis techniques in a blind trial, and were sent to
the lab in numbered paper bags. Collection localities were
known from all samples. San Joaquin kit foxes were from
11 different groups within the San Joaquin Valley,
California. Red and gray foxes were collected in both Los
Angeles and Orange counties, California.
S H O R T  C O M M U N I C A T I O N
A noninvasive method for distinguishing among canid
species: amplification and enzyme restriction of DNA
from dung
E .  P A X I N O S , *  C .  M C I N T O S H ,  K .  R A L L S  and R .  F L E I S C H E R
Department of Zoological Research, National Zoological Park, Smithsonian Institution, Washington, DC 20008, USA, *Brown
University, Department of Ecology and Evolutionary Biology, Providence, RI 02912, USA
Abstract
Endangered San Joaquin kit foxes Vulpes macrotis mutica can be sympatrically distrib-
uted with as many as four other canids: red fox, gray fox, coyote and domestic dog. Canid
scats are often found during routine fieldwork, but cannot be reliably identified to
species. To detect and study the endangered kit fox, we developed mitochondrial DNA
markers that can be amplified from small amounts of DNA extracted from scats. We
amplified a 412-bp fragment of the mitochondrial cytochrome-b gene from scat samples
and digested it with three restriction enzymes. The resulting restriction profiles discrim-
inated among all five canid species and correctly identified 10 ?unknown? fox scats to
species in blind tests. We have applied our technique to identify canids species for an
environmental management study and a conservation study. We envision that our proto-
col, and similar ones developed for other endangered species will be greatly used for
conservation management in the future.
Keywords: kit fox, canid, endangered species, PCR amplification, noninvasive DNA sampling
Received 9 July 1996; revision received 14 October 1996; revision accepted 27 November 1996
Molecular Ecology 1997, 6, 483?486
? 1997 Blackwell Science Ltd
Correspondence: E. Paxinos, Molecular Genetics Lab, National
Zoological Park, Smithsonian Institution, 3001 Block Connecticut
Ave NW, Washington DC 20008, USA. Tel.: +1-202-673-4781, Fax:
+1-202-673-4648. E-mail: NZPGL110@SIVM.SI.EDU
DNA from tissue samples was isolated by standard
methods of Proteinase K digestion, phenol?chloroform
extraction, and DNA precipitation in ethanol (Sambrook et
al. 1989). DNA from scats was isolated with two different
methods under quasi-clean room conditions to prevent
contamination of potentially suboptimal material with the
more robust DNA of fresh tissue samples (P??bo 1990;
Cooper 1994). Although there is DNA throughout the scat
in the form of shed epithelial cells, scats remain unprotect-
ed from sun and rain for unknown lengths of time prior to
collection, and may sustain oxidative and hydrolytic dam-
age (Lindahl 1993). In the first method, after removal of
rodent bones and visible plant material, about 0.5?1.0 g of
fox scat was minced with a sterile razor blade and washed
twice with 0.5 M EDTA (pH 8.0) via inversion followed by
centrifugation. The samples, including extraction controls
(reagents without template), were incubated with rotation
for 12?24 h at 55 ?C in 12 mL extraction buffer each [0.01 M
NaCl, 0.01 M Tris-HCl, 0.1 M EDTA (pH 8.0), 1 mg/mL
Proteinase K, 10 mg/mL DDT, and 1% SDS], then ex-
tracted three times with phenol and once with
Chloroform : Isoamyl alcohol (24 : 1). The aqueous phase
was filtered through Centriprep 30? (Amicon) concen-
trators and rinsed twice via centrifugal dialyses with 2 mL
deionized sterile water. Retentate was brought to 100 m L,
and divided into aliquots to provide backup stocks in the
event of post-extraction contamination of ?working?
stocks. Two microlitres retentate was used for each sub-
sequent PCR amplification reaction.
Although organic extraction of DNA is reliable, it is
unattractive for forensic applications due to long turn-
around time, number of steps and extra sample transfers
which increase the probability of cross-contamination. We
therefore modified a Chelex? 100 resin (Bio-rad) method
(Walsh et al. 1991) for DNA isolation from scat samples.
Scat material (? 0.1?0.3 m g) was suspended in 500 m L 5%
Chelex? 100 in sterile water in screw-cap tubes. Samples
were then boiled in a waterbath for 7 min, vortexed at full
speed, boiled another 7 min, and finally centrifuged 5 min
at 14 000 g. Extraction controls (all reagents, no scat mate-
rial) were made concurrently throughout the procedure
with the samples. The supernatant was removed as a
source of template DNA and aliquoted into three sterile
Eppendorf tubes. Four microlitres was used for each sub-
sequent PCR amplification.
Using published mitochondrial cytochrome-b
sequences from three fox species (Geffen et al. 1992), and
the coyote (Gotelli et al. 1994), we designed a canid specif-
ic primer, CanidL1 (5? -AATGACCAACATTCGAAA-3? ) as
well as a previously described primer, H15149 (Kocher et
al. 1989) to amplify a 412-bp fragment of this gene. Final
amplification reagents in 100 m L volumes were: 1X reac-
tion buffer (Perkin-Elmer), 2.5 mM MgCl2, 200 m M each
dNTP, 1.7 mg/mL Fraction-V BSA, 2 units Taq polymerase
(Perkin-Elmer), and 1 m M each primer. The reactions for
scat extracts as well as extract and PCR controls (reaction
components without template) were cycled 35 times fol-
lowing an initial hot start using the following profile: 94 ?C
for 1 min, 52 ?C for 1 min, and 72? for 1.5 min. There was
no evidence of contamination of extraction or amplifica-
tion reagents. Products amplified from the DNA isolated
from the 10 scats, as well as the 21 known tissue samples
were gel purified via electrophoresis in TAE buffered 2%
NuSieve? agarose, then excised, and eluted (Wizard
Prep?, Promega). PCR products were sequenced on both
strands using an ABI 373 automated DNA sequencer
(using PRISM? Ready Reaction DyeDeoxy Terminator
Cycle Sequencing chemistry: ABI). Sequences from tissues
with known identity were aligned to reveal diagnostic
restriction endonuclease cut sites and percentage sequence
divergence between species.
Although sequencing is the most reliable method, it is
expensive and time consuming. We therefore developed a
system of restriction endonuclease digestion of the ampli-
fied region that easily distinguishes between canid
species. Cytochrome-b amplification products from DNA
isolated from scats were subjected directly to restriction
digestion by three endonucleases following manufacturers?
recommended conditions. Digest products were visualized
with ethidium bromide staining following electrophoresis
in TBE buffered 2% NuSieve agarose.
Results
Chelex? extractions of scat samples were comparable in
quality and reliability in producing PCR amplification
products to more expensive organic methods. The Chelex?
method was preferred because it eliminated the need of
laborious centrifugal dialyses to remove detergents, salts
and enzymes from extracts, and involved fewer steps and
sample transfers, thereby minimizing the potential for
contamination.
Alignment of 378-bp of cytochrome-b sequence of the
three fox species (total n = 13) revealed uncorrected
sequence differences of 12.0% for kit vs. red, 14.4% for red
vs. gray, and 14.7% for kit vs. gray foxes (Fig. 1).
Restriction sites diagnostic for each fox species were pre-
sent in the aligned sequences. Restriction endonuclease
digestion of the PCR products with two enzymes (AluI
and HinfI) resulted in profiles that distinguished each of
the fox species from each other, and all of the foxes from
domestic dogs and coyotes (Fig. 2a and b). A third enzyme
(TaqI) also distinguished the three fox species from each
other (Fig. 2c). However, the kit fox, coyote, and domestic
dog could not be distinguished with this enzyme, as these
three species do not have a TaqI restriction site. ?Correct?
species-specific digest profiles were found for every indi-
vidual amplified (seven gray, eight red, and 16 kit foxes)
? 1997 Blackwell Science Ltd, Molecular Ecology, 6, 483?486
484 E .  P A X I N O S  E T  A L .
including the 10 scat samples (four gray, three red, three
kit foxes). Restriction digestion proved highly reliable for
the identification of fox species from their scats.
Discussion
We present above a simple method of isolating DNA from
canid scats and a protocol for PCR amplifying a small frag-
ment of DNA from the scat DNA. Further, we show that
digestion with a set of three restriction enzymes provides
clean and reliable markers for identification of five sym-
patric canids to species. The ability to identify canid
species present at a given locality by mitochondrial DNA
markers amplified from scats has been useful for biologists
conducting surveys of San Joaquin kit foxes. In the last
year, we successfully identified unknown canid scats to
species for two separate conservation programmes: the
San Joaquin Endangered Species Recovery Planning
Program (Fresno, CA) and PRC, Environmental
Management, Inc. (Denver, CO). Our results provided
these organizations with species determination unavail-
able by other means. The Chelex? and restriction enzyme
methods we adapted have proven much less time consum-
ing and expensive but are equal in reliability to trapping
programmes for the identification of canids.
Coyotes, red foxes and domestic dogs may prey upon
kit foxes (Ralls & White 1995) and all canids may scavenge
? 1997 Blackwell Science Ltd, Molecular Ecology, 6, 483?486
G E N E T I C  M E T H O D S  F O R  D I S T I N G U I S H I N G  A M O N G  S Y M P A T R I C  C A N I D  S P E C I E S 485
Fig. 1 378 bp of cytochrome-b sequence
from three sympatric fox species.
Enzyme-restriction sites are underlined
and identified.
Fig. 2 Restriction enzyme digested cytochrome-b amplification
products of five canid species electrophoresed in a 2.0% agarose
gel: coyote (C), domestic dog (i.e. greyhound; D), gray fox (G), kit
fox (K), and red fox (R). DNA size marker (M) is HincII-digested
f x 174 [fragment sizes from the top are 1057 bp, 770 bp, 612 bp,
495 bp, 392 bp (345?341?335 bp combined) (297?291 bp com-
bined), 210 bp, and 162 bp]. (a) AluI digests. (b) HinfI digests. (c)
TaqI digests. The second lane of each fox species is a digestion of
a product amplified from scat. Coyote and domestic dog had
identical patterns for AluI.
food from the carcass of another canid species. Thus DNA
from more than one species could conceivably be ampli-
fied from a single scat, and would be evident as ?hybrid?
restriction digest patterns or ambiguous sequence. Such a
result would indicate that both species were present at the
locality where the scat was collected. Their identities could
be explicitly resolved using a restriction enzyme which
cuts only one species? amplification product, leaving all
others intact. When electrophoresed through agarose, the
uncut product could be isolated from agarose and then
identified using restriction enzymes, or sequencing.
Acknowledgements
We thank Cathleen Cox, Dave Garcelon, Rob Grimmins, Dave
Erler, Bert Paluch, Robert Wayne, the San Joaquin Endangered
Species Recovery program and PRC Environmental Services, Inc.
for samples. The Friends of the National Zoo and the Abbott and
Witherspoon Funds of the Smithsonian Institution provided
financial support. Gary Nunn and two anonymous referees pro-
vided useful comments on a previous draft of this manuscript.
References
Clifton SD (1995) Comparison of the diets of the San Joaquin kit
fox, gray fox and red fox from Merced county, California,
based on scats. In: Endangered and Sensitive Species of the San
Joaquin Valley, California (eds Williams DF, Byrne S, Rado TA),
pp. 374 California Energy Communication, Sacramento.
Cooper A (1994) DNA from museum specimens. In: Ancient DNA
(eds Hermann B, Hummel S) Springer-Verlag, New York.
Constable JJ, Packer C, Collins SA et al. (1995) Nuclear DNA from
primate dung. Nature, 373, 393.
Geffen E, Mercure A, Girman DJ et al. (1992) Phylogenetic rela-
tionships of the fox-like canids: mitochondrial DNA restriction
fragment, size and cytochrome-b sequence analyses. Journal of
Zoology, 228, 27?39.
Gerloff U, Schl?tterer C, Rassmann K et al. (1995) Amplification of
hypervariable simple sequence repeats (microsatellites) from
extremental DNA of wold living Bonobos (Pan Paniscus).
Molecular Ecology, 4, 515?518.
Gotelli D, Sillero-Zubiri C, Applebaum GC et al. (1994) Molecular
genetics of the most endangered canid: the Ethiopian wolf
Canis simensis. Molecular Ecology, 3, 301?312.
Halfpenny J (1986) A Field Guide to Mammal Tracking in North
America. Johnson Books, Boulder, CO.
H?ss M, Kohn M, P??bo S et al. (1992) Excrement analysis by PCR.
Nature, 359, 199.
Kocher TD, Thomas WK, Meyer A et al. (1989) Dynamics of mito-
chondrial DNA evolution in animals: amplification and
sequencing with conserved primers. Proceedings of the Natural
Academy of Sciences USA, 86, 6196?6200.
Lindahl T (1993) Instability and decay of the primary structure of
DNA. Nature, 362, 709?715.
Orloff SG (1995) Survey techniques for the San Joaquin kit fox
(Vulpes macrotis mutica). In: Endangered and Sensitive Species of
the San Joaquin Valley, California (eds Williams DF, Byrne S,
Rado TA), pp. 185?197 California Energy Communication,
Sacramento.
P??bo S (1990) Amplifying ancient DNA. In: PCR Protocols: A
Guide to Methods and Applications (eds Innis, MA, Gelfand, DH,
Sninsky, JJ, White, TJ). Academic Press, New York.
Sambrook J, Fritsch EF, Maniatis T (1989) Molecular Cloning: A
Laboratory Manual, 2nd edn. Cold Spring Harbour Laboratory
Press, New York.
Ralls K, White PJ (1995) Predation on San Joaquin kit foxes by
larger canids. Journal of Mammalogy, 76, 723?729.
Taberlet P, Bouvet J (1992) Bear conservation genetics. Nature, 358,
197.
Walsh PS, Metzger DA, Higuchi R (1991) Chelex? 100 as a medi-
um for simple extraction of DNA for PCR-based typing from
forensic material. BioTechniques, 10, 506?513.
Eleni Paxinos is a graduate student at Brown University and a
Smithsonian Pre-doctoral fellow. She uses DNA from suboptimal
material, including subfossil bones, scat, and museum specimens,
for population and phylogenetic analyses. Carl McIntosh is a
biotechnician at the Molecular Genetics Lab with interests in mol-
ecular systematics. Katherine Ralls is a research zoologist at the
National Zoo who conducts field research on San Joaquin kit
foxes. Robert Fleischer heads the Molecular Genetics Lab at the
National Zoo and uses molecular genetic analyses of organisms 
in studies of population genetics, systematics, and animal 
behaviour.
? 1997 Blackwell Science Ltd, Molecular Ecology, 6, 483?486
486 E .  P A X I N O S  E T  A L .