Endangered species in small habitat patches can possess high genetic diversity: the case of the Tana River red colobus and mangabey

Abstract

We used mtDNA sequence data from the Tana River red colobus and mangabey to determine how their population genetic structure
was influenced by dispersal and habitat fragmentation. The colobus and mangabey are critically endangered primates endemic
to gallery forests in eastern Kenya. The forests are a Pliocene–Pleistocene refugium that has recently undergone significant
habitat loss and fragmentation due to human activities. We expected both primates to exhibit low levels of genetic diversity
due to elevated genetic drift in their small populations, and to show a strong correspondence between genetic and geographic
distance due to disruption of gene flow between forests by habitat fragmentation. Additionally, because mangabey females are
philopatric, we expected their mtDNA variation to be homogeneous within forest patches but to be heterogeneous between patches.
In contrast, colobus have a female-biased dispersal and so we expected their mtDNA variation to be homogeneous within and
between forest patches. We found high levels of haplotype and nucleotide diversity as well as high levels of sequence divergence
between haplotype groups in both species. The red colobus had significantly higher genetic variation than the mangabey did.
Most of the genetic variation in both primates was found within forest fragments. Although both species showed strong inter-forest
patch genetic structure we found no correspondence between genetic and geographic distances for the two primates. We attributed
the high genetic diversity to recent high effective population size, and high sequence divergence and strong genetic structures
to long-term habitat changes in the landscape.

KeywordsMtDNA-Climate change-Africa-Genetic diversity-Conservation

Full text

Page 1

RESEARCH ARTICLE
Endangered species in small habitat patches can possess high
genetic diversity: the case of the Tana River red colobus
and mangabey
David N. M. Mbora•Mark A. McPeek
Received: 29 April 2009/Accepted: 8 February 2010
? Springer Science+Business Media B.V. 2010
Abstract
River red colobus and mangabey to determine how their
population genetic structure was influenced by dispersal
and habitat fragmentation. The colobus and mangabey are
critically endangered primates endemic to gallery forests in
eastern Kenya. The forests are a Pliocene–Pleistocene
refugium that has recently undergone significant habitat
loss and fragmentation due to human activities. We
expected both primates to exhibit low levels of genetic
diversity due to elevated genetic drift in their small pop-
ulations, and to show a strong correspondence between
genetic and geographic distance due to disruption of gene
flow between forests by habitat fragmentation. Addition-
ally, because mangabey females are philopatric, we
expected their mtDNA variation to be homogeneous within
forest patches but to be heterogeneous between patches. In
contrast, colobus have a female-biased dispersal and so we
expected their mtDNA variation to be homogeneous within
and between forest patches. We found high levels of hap-
lotype and nucleotide diversity as well as high levels of
sequence divergence between haplotype groups in both
species. The red colobus had significantly higher genetic
variation than the mangabey did. Most of the genetic
We used mtDNA sequence data from the Tana
variation in both primates was found within forest frag-
ments. Although both species showed strong inter-forest
patch genetic structure we found no correspondence
between genetic and geographic distances for the two pri-
mates. We attributed the high genetic diversity to recent
high effective population size, and high sequence diver-
gence and strong genetic structures to long-term habitat
changes in the landscape.
Keywords
Genetic diversity ? Conservation
MtDNA ? Climate change ? Africa ?
Introduction
The current genetic diversity of any species has been
influenced by many factors in the past. Past changes in
climate are known to be a major influence on the distri-
bution of populations and therefore their genetic structure
(Hewitt 2000). In Africa, climate change caused major
shifts in faunal assemblages during a time interval lasting
5.3 million years during the Pliocene–Pleistocene epochs
(deMenocal 2004; Bobe and Behrensmeyer 2004). The
genetic imprint of these habitat changes is evident in sev-
eral mammals in East Africa; buffalo (Heller et al. 2008),
elephant (Okello et al. 2008), hippopotamus (Okello et al.
2005) and baboons (Storz et al. 2002a, b). During that
interval, lower temperatures and increased aridity in East
Africa reduced and fragmented tropical forests and left
them as isolated patches along major rivers and on high
elevation (Bobe and Behrensmeyer 2004). These forest
refugia have provided important habitat for forest depen-
dent non-human primates (Fleagle 1999). Thus, the current
population genetic structure of primates endemic to these
forests should reflect their histories in these refugia.
D. N. M. Mbora ? M. A. McPeek
Department of Biological Sciences, Dartmouth College,
Hanover, NH 03755, USA
e-mail: Mark.McPeek@Dartmouth.edu
D. N. M. Mbora
Tana River Primate National Reserve, P O Box 4, Hola, Kenya
D. N. M. Mbora (&)
Department of Biology, Whittier College, PO Box 634, Whittier,
CA 90608, USA
e-mail: David.Mbora@Dartmouth.edu; dmbora@yahoo.com;
dmbora@Whittier.edu
123
Conserv Genet
DOI 10.1007/s10592-010-0065-0

Page 2

Furthermore, primates endemic to these forests are now
vulnerable to further loss of genetic diversity because of
additional forest reduction and fragmentation caused by
human activities (Mace and Balmford 2000).
The Tana River forests of southeastern Kenya are an
example of forest fragments whose origin dates back to the
increasing aridity of the Pliocene–Pleistocene interval in
East Africa (Bobe and Behrensmeyer 2004). Botanical
surveys of the Tana River forests suggest that evergreen
forests were once more continuous in Africa (Hamilton
1981). In particular, 12% of tree species in the Tana are
from the Guinea-Congolian region indicating an earlier
period of continuous rain forest across the African continent
(Medley 1992). The forests occupy the lower floodplain of
the Tana River and are of great conservation importance.
They are part of the east African coastal forests global
biodiversity hotspot (Myers et al. 2000) and support a high
diversity of rare plant and animal species (Andrews et al.
1975). In particular, they provide the only known habitat of
two endemic primates: the Tana River red colobus (Pro-
colobus rufomitratus) and the Tana mangabey (Cercocebus
galeritus). Both species are critically endangered (Hilton-
Taylor 2000) and ranked among the IUCN’s top 25 most
endangered primates (Grubb et al. 2003; Mittermeier et al.
2007). It is estimated that the population of the colobus is
less than 1000 individuals and that of the mangabey does
not exceed 2000 individuals (Butynski and Mwangi 1994).
In addition to the natural forest fragmentation caused by
the meandering of the river in its old stage, recent human
activities have further reduced and fragmented the forests
causing precipitous declines in the primate populations, and
extinctions in several of the fragments (Mbora and Meikle
2004). Thus, these forests offer a natural setting to study the
effects of forest loss and fragmentation on population
genetic structure of endemic, endangered forest primates.
The population structure seen in mitochondrial DNA
can be particularly useful in understanding the effects of
forest loss and fragmentation on population genetic struc-
ture of forest primates. Mitochondrial DNA is maternally
inherited (Gyllensten et al. 1985), lacks recombination
(Hayashi et al. 1985) and exhibits rapid sequence evolution
(Brown et al. 1979). Consequently, any mtDNA lineages
that diverge in populations (e.g. in forest fragments) are
independent clones that rapidly accumulate divergent sets
of mutations through time. Thus, in species where females
are philopatric, there should exist little or no variation
within, and much variation between populations, e.g. in
many macaque species (Macaca spp.; Melnick and Hoelzer
1992) and vervet monkeys (Cercopithecus aethiops aethi-
ops; Shimada 2000). In contrast, female dispersal should
lead to much differentiation within and less differentiation
between populations; e.g. in the hamadryas baboons (Papio
hamadryas hamadryas; Hapke et al. 2001).
We analyzed the population structures of mtDNA vari-
ation (NADH dehydrogenase subunit 4, ND4, gene) of the
Tana River red colobus and mangabey to determine how
they are influenced by the pattern of dispersal and habitat
fragmentation. We expected both primates to exhibit low
levels of genetic diversity due to genetic drift in their rel-
atively small populations, and for populations that were
geographically close to one another to be more genetically
similar because of greater gene flow (Wright 1978; Kimura
and Weiss 1964). Mangabey females are largely philopatric
(Kinnaird 1992), while red colobus females disperse on
attaining sexual maturity (Marsh 1979). Thus, we expected
the mtDNA variation in the mangabey to be relatively
homogeneous within forest patches but to be heterogeneous
between forest patches. Conversely, we expected the
mtDNA variation of the red colobus to be relatively
homogeneous within and between forest patches due to
female dispersal in this species (Marsh 1979).
Materials and methods
Study area and species
The study area comprises approximately 26 km2of gallery
forest occurring in scattered patches of various sizes on
both sides of the Tana River in eastern Kenya (Fig. 1;
Mbora and Meikle 2004). This area encompasses the entire
distribution range of the Tana River red colobus and
mangabey. These forests exist in an arid environment with
an annual total rainfall of less than 400 mm. Forest is
created and maintained by groundwater, and by periodic
flooding of the river (Hughes 1990). The depth of the water
table drops off rapidly from the edge of the river and limits
the lateral extent of the forests to about 1 km on either side
(Hughes 1990). The intervening matrix is mainly cultivated
land, riparian grassland and dry shrubs.
We mapped the gallery forest using aerial photographs
taken in 1994 and 1996, and selected 12 forest patches as
study sites. We chose forest patches so that approximately
equal forest area was sampled east and west of the Tana
River, and inside and outside the Tana River Primate
National Reserve (TRPNR) to capture the range of habitat
conditions within the floodplain (Fig. 1). We surveyed
each study forest to determine the number of resident
groups of colobus and mangabeys, and identified a subset
of groups within each forest for detailed studies of group
size, age and sex composition over time. We systemati-
cally selected social groups that were easy to locate and
to identify using ‘‘marker’’ animals. Since 2001, we have
periodically surveyed the forests and monitored all these
study groups (Mbora and Meikle 2004; Mbora, unpub-
lished data).
Conserv Genet
123

Page 3

The Tana colobus and mangabey are of similar body
size but their behavioral ecologies and life history strate-
gies are quite different (Kinnaird 1992; Marsh 1979). The
red colobus is a specialist frugivore with limited vagility. It
is almost exclusively arboreal and lives in relatively small
social groups that exhibit high site fidelity (Marsh 1981). A
canopy dweller, the colobus depends on a diet of mainly
leaves obtained from a limited number of canopy tree
species (Marsh 1981; Mbora and Meikle 2004). Thus, it
was relatively easy to locate and observe colobus groups, to
maintain contact with them and to determine their group
composition while they were in the canopy. In contrast, the
mangabey is a dietary generalist that is mostly terrestrial
and highly vagile. It lives in much larger social groups and
its diet comprises seeds and ripe fruit from a variety of tree
species, and substantial amounts of animal prey (Kinnaird
1992). Mangabeys are quite skittish, and it was necessary
to get groups well habituated to human presence in order to
determine their size and composition and to obtain fecal
samples for mtDNA analysis. Consequently, detailed
observation of mangabeys focused on fewer social groups
than in the colobus.
Collection of fecal samples and DNA extraction
In 2004 and 2005, from July to September, we collected
fecal samples from study groups of colobus and mangabeys
by following them, on separate days, from 0600 h to
1130 h, and then from 1500 h until nightfall. Upon
observing an animal defecating, we collected a sample of
the feces using a sterile collecting stick while wearing latex
gloves. We aimed to take only a single sample from any
particular animal, but to sample as many individuals from
each social group and forest fragment as possible. Colobus
feces are usually deposited in distinct pellets so we just
collected 1–3 pellets depending on size. The mangabey
does not produce its feces in distinct pellets so we extracted
a sample of the feces from the outermost part of the dung
bolus. In either case, the sample was placed into a tube
containing 30 ml of 100% ethanol and labeled with a
permanent marker to indicate the date, species identity, and
coded to identify the troop and forest. The ethanol and
sample were then mixed by inversion without shaking. The
goal was to maintain the bolus form of the sample in order
to avoid losing target cells along with the ethanol
Fig. 1 Study area indicating the location and distribution of study
forests in southeastern Kenya for the colobus (a) and mangabey (b).
The frame on the top half of the map shows the approximate location
of the Tana River Primate National Reserve. The Roman numeral
codes identify study forests adjacent to them, and the major villages
are named
Conserv Genet
123

Page 4

supernatant in the next step. After 36 h, we carefully
poured off the ethanol with the tube loosely capped, and
transferred the remaining solid material into a new-labeled
tube containing silica for drying and storage (Nsubuga
et al. 2004). The second tube was also labeled with a
permanent marker as above. The samples were stored at a
cool temperature in a tent in the field, and at -80?C after
arrival in the laboratory.
Approximately %200 mg of the fecal sample was
extracted using the QIAamp DNA Stool kit (Qiagen)
according to the manufacturer’s instructions with minor
modifications (Nsubuga et al. 2004). The dried samples
were vortexed in 1.6 ml of ASL buffer and left overnight
(12–16 h) in an agitator at 25?C. The intermediate steps
followed the manufacturer’s protocol, but we included an
incubation step of 20 min followed by centrifugation for
2 min (Nsubuga et al. 2004) in the final step of the pro-
cedure where buffer AE elutes the DNA.
DNA samples used for positive controls in PCR reac-
tions and numts diagnostics were extracted from tissue
following standard phenol extraction methods (Dowling
et al. 1990). Colobus DNA was extracted from two indi-
viduals of black and white colobus (Colobus guereza) liver
tissue donated by Dr. Cathi Lehn (American Zoo and
Aquarium Association’s Biomaterials Banking Advisory
Group), a tissue sample of Procolobus badius donated by
Dr. Nelson Ting (University of Iowa). Mangabey DNA was
extracted from muscle tissues preserved in ethanol. The
muscle was acquired in Tana River following a fatal attack
on a mangabey by an unidentified bird of prey in one of the
forests in August 2005. In addition, total genomic DNA of
two Uganda red colobus (Procolobus tephrosceles) was
donated by Dr. TL Goldeberg (University of Illinois,
Urbana-Champaign).
We used several measures to avoid cross contamination
of samples, and contamination of our samples with con-
centrated DNA sources. In particular, all the laboratory
procedures reported here were performed in a section of the
laboratory dedicated to the analyses of DNA from the two
primates; this laboratory does not work with any other
species of primates or other vertebrates. We worked on the
samples from the two study species sequentially rather than
concurrently; we first worked on the colobus and then
mangabey samples.
Genotyping and sequencing
We successfully amplified and sequenced DNA from 53
colobus individuals (53 stool samples) from 10 forests, and
36 mangabey individuals (35 stool samples and 1 tissue
sample) from 6 forests (Fig. 1; Table 1). The following
primer pair was used at a concentration of 5 mM;
STRETCHM (50-RCTTGCGTTGAGGCGTTCTG, H11
196) and ND4#1 (50-CTTCTAACACTRACCGCCTGACT,
L10952). The primers amplify the NADH dehydrogenase
subunit 4 (ND4) of Procolobus badius corresponding to
site 10203-11580 of the mitochondrial genome (accession
no. DQ355301). We used 1 ll of the eluate from the
extraction procedure as template in a 50 ll polymerase
chain reaction (PCR) containing HotStarTaq DNA poly-
merase, PCR buffer with 3 mM MgCl2and 400 lM each
dNTP (Qiagen). We performed a hot start PCR cycle in an
MJ Research PTC-200 Peltier thermal cycler under the
following conditions: an activation step at 95?C for
15 min; followed by 45 cycles at 95?C for 30 s, 57.9?C for
30 s and 72?C for 1 min, and a final extension step at 72?C
for 10 min. Each reaction was replicated at least three
times and included positive and negative controls, and the
Table 1 Attributes of the study
populations of Tana River red
colobus and mangabey
aHaplotype groups as
identified in Fig. 2
Forest nameColobusMangabey
Population
ID
Individuals
genotyped
Haplotypesa
found
Population
ID
Individuals
genotyped
Haplotypesa
found
Wenje EastI6A
Makere EastII7A, B
Guru-McheleloI5A, B, D, E, FIII4A
CongolaniII8A, C, E, F IV8A
Sifa EastIII8A, C, D, EV7A, B
Lalafitu IV9A, B, C, D, F
MnaziniV2D, FVI4A, B
KinyaduVI 7A, D, E, G
Bubesa VII3C, F
Sailoni West VIII3C
Hewani East IX2 A, D
Home forest X5A, F
Conserv Genet
123

Page 5

success of the PCR was assessed by electrophoresis of 5 ll
of the product on a 1.5% agarose gel.
We purified the PCR products using the QIAquick PCR
purification kit (Qiagen) following the manufacture’s pro-
tocol, and cloned them using pGEM?-T easy Vector Sys-
tems (Promega Corporation). We prepared overnight
cultures of cells containing pGEM?easy Vector by picking
individual ampicillin-resistant colonies from fresh plates,
inoculating 2 ml of LB broth containing 100 lg/ml ampi-
cillin and shaking samples overnight at 37?C. We harvested
the bacterial cells by centrifugation and purified single
stranded DNA by extraction and precipitation using
GenEluteTMPlasmid Miniprep Kit (Sigma) following the
manufacturer’s protocol.
The bacterial clones from the above procedures were
sequenced with an ABI PrismTM3100 genetic analyzer
(Applied Biosystems) using a BigDye Terminator Cycle
Sequencing Kit (Applied Biosystems). Sequences were
then aligned using SEQUENCHERTM(version 4.5) and
verified for accuracy. In addition, we translated all the
nucleotide sequences into protein sequences to determine if
they contained any missense mutations or internal stop
codons using DnaSP v. 4.10 (Rozas et al. 2003) and MEGA
version 3.1 (Kumar et al. 2004). On translation of the
nucleotide sequences into protein sequences, we did not
find any missense mutations or internal stop codons
(GenBank accession numbers FJ881863-FJ882004). Thus,
all our sequences appear to be valid functional mtDNA.
Data analyses
We treated all sequences from fecal samples collected from
the same forest patch as comprising a population. To
compute the mtDNA sequence variation of the two species,
we calculated the haplotype diversity, nucleotide diversity
(p), and the proportion of nucleotide polymorphisms (h) for
each species (Nei 1987) using DnaSP v. 4.10 (Rozas et al.
2003). To examine the relationships between haplotypes
detected in the each species in the landscape, we computed
a minimum spanning network between haplotypes using
ARLEQUIN 2.0 (Excoffier et al. 2005), and then used the
connection lengths between samples (Operational Taxo-
nomic Units) to draw a diagram of the minimum spanning
network of haplotypes. We then mapped the distribution of
haplotypes in the forests (Fig. 1; Table 1).
We constructed a neighbor joining phylogenetic trees
using MEGA version 3.1 (Kumar et al. 2004) using
MODELTEST (Posada and Crandall 1998) to determine
the appropriate nucleotide substitution model for the data
set. To investigate the possibility of a past bottleneck in
both species, we conducted an analysis of pairwise
sequence mismatch distributions (Rogers 1995) using
ARLEQUIN 2.0 (Excoffier et al. 2005). The sequence
mismatch distributions in a population that has experienced
a population bottleneck should be smooth and have a peak,
whose position identifies the time of the bottleneck (Har-
pending 1994).
We conducted two analyses to elucidate the role of
current habitat fragmentation in shaping the population
structure of the mtDNA variation among populations
(forest patches) for each of the two species. First, we
conducted an analysis of molecular variance (AMOVA) to
determine how mtDNA variation was partitioned among
and within populations (Excoffier et al. 1992). Second, we
calculated the genetic distance between populations as
pairwise Fst values (Weir and Cockerham 1984) and
measured geographic distances between populations as
linear centroid-to-centroid distances between forests using
ArcMap GIS. We then tested for the correspondence
between geographic and genetic distance using a mantel
test (Mantel 1967) and linear regression analyses in the
R-package (Casgrain et al. 2005).
Results
We found high levels of genetic diversity in both monkeys,
but the colobus had significantly greater levels of genetic
variability than the mangabey. We identified 34 haplotypes
among the 53 red colobus sequences, and 18 haplotypes
among the 36-mangabey sequences (Table 2; Fig. 2). In
addition, when we compared metrics that account for dif-
ferences in numbers of sequences for each species, we also
found that red colobus had significantly greater haplotype
and nucleotide diversity than mangabeys (Table 2). Com-
parison of the minimum spanning networks among haplo-
types highlights the major difference that underlies these
differences in genetic diversity. The red colobus haplotypes
form seven distinct groups that are each separated from the
next closest group in the network by 17–24 nucleotide
substitutions (the average distance between adjacent
groups was 20.5 mutational steps (Fig. 2a). The mangabey
network contained only two such haplotype groups sepa-
rated by 30 mutational steps from one another (Fig. 2b).
Although they differed in the overall levels of genetic
diversity, the two species showed simpler patterns of spa-
tial population structure. Five of the seven-haplotype
groups identified in the red colobus were represented in
four or more populations; two groups were found in six
populations (Table 1; Fig. 1a). Similarly, the mangabey
network had one diverse haplotype group that was widely
distributed among the various populations, and a second
haplotype group, that was found in only three populations
(Table 1; Fig. 1b).
The sequence mismatch distributions exhibited a smooth
distribution with a peak (Table 3). However, the smooth
Conserv Genet
123