BIOLOGY OF REPRODUCTION 85, 243?253 (2011)
Published online before print 12 May 2011.
DOI 10.1095/biolreprod.110.089417
Increasing Age Influences Uterine Integrity, But Not Ovarian Function or Oocyte
Quality, in the Cheetah (Acinonyx jubatus)1
Adrienne E. Crosier,2,4 Pierre Comizzoli,4 Tom Baker,5 Autumn Davidson,5 Linda Munson,3,5
JoGayle Howard,3,4 Laurie L. Marker,6 and David E. Wildt4
Center for Species Survival,4 Smithsonian Conservation Biology Institute, Front Royal, Virginia and Washington, DC
School of Veterinary Medicine,5 University of California-Davis, Davis, California
Cheetah Conservation Fund,6 Otjiwarongo, Namibia
ABSTRACT
Although the cheetah (Acinonyx jubatus) routinely lives for
more than 12 yr in ex situ collections, females older than 8 yr
reproduce infrequently. We tested the hypothesis that repro-
duction is compromised in older female cheetahs due to a
combination of disrupted gonadal, oocyte, and uterine function/
integrity. Specifically, we assessed 1) ovarian response to
gonadotropins; 2) oocyte meiotic, fertilization, and develop-
mental competence; and 3) uterine morphology in three age
classes of cheetahs (young, 2?5 yr, n? 17; prime, 6?8 yr, n? 8;
older, 9?15 yr, n ? 9). Ovarian activity was stimulated with a
combination of equine chorionic gonadotropin and human
chorionic gonadotropin (hCG), and fecal samples were collected
for 45 days before gonadotropin treatment and for 30 days after
oocyte recovery by laparoscopy. Twenty-six to thirty hours post-
hCG, uterine morphology was examined by ultrasound, ovarian
follicular size determined by laparoscopy, and aspirated oocytes
assessed for nuclear status or inseminated in vitro. Although no
influence of age on fecal hormone concentrations or gross
uterine morphology was found (P . 0.05), older females
produced fewer (P , 0.05) total antral follicles and oocytes
compared to younger counterparts. Regardless of donor age,
oocytes had equivalent (P . 0.05) nuclear status and ability to
reach metaphase II and fertilize in vitro. A histological
assessment of voucher specimens revealed an age-related
influence on uterine tissue integrity, with more than 87% and
more than 56% of older females experiencing endometrial
hyperplasia and severe pathologies, respectively. Our collective
findings reveal that lower reproductive success in older cheetahs
appears to be minimally influenced by ovarian and gamete aging
and subsequent dysfunction. Rather, ovaries from older females
are responsive to gonadotropins, produce normative estradiol/
progestogen concentrations, and develop follicles containing
oocytes with the capacity to mature and be fertilized. A more
likely cause of reduced fertility may be the high prevalence of
uterine endometrial hyperplasia and related pathologies. The
discovery that a significant proportion of oocytes from older
females have developmental capacity in vitro suggests that in
vitro fertilization and embryo transfer may be useful for
??rescuing?? the genome of older, nonreproductive cheetahs.
aging, carnivore reproduction, felid, gamete, ovary, ovum, uterus
INTRODUCTION
The number of cheetahs (Acinonyx jubatus) living in nature
has decreased approximately 85% since 1900 [1?3], with the
approximately 10 000 remaining individuals threatened by
continuing loss of habitat and prey base [2, 3]. Thus, having a
viable ex situ (captive) population is important 1) as a hedge
against extinction, 2) for promoting public awareness, and 3)
for studying biological phenomena virtually impossible to
investigate in rare and elusive, free-ranging individuals [4].
However, the cheetah has always been challenging to
reproduce consistently in captivity. Since 1970, approximately
78% of females and approximately 82% of males in North
American institutions have died without reproducing [5, 6]. To
facilitate genetic management of this species in North
American institutions, a group of 56 cooperating institutions
pool animals in a Species Survival Plan (SSP). This population,
currently at approximately 210 adult cheetahs, has never been
self-sustaining, and death rate has exceeded birth rate in 13 of
the last 16 yr [5, 6]. Approximately 85% of adult females in the
current North American population have never reproduced [5].
Of females producing young for the first time, approximately
65% have been 2?5 yr of age [5, 6]. Thirty-seven percent of
females in this contemporary population already are 8 yr of age
or older, with an average life span in captivity of approximately
12 yr [6]. This longevity is at least 6 yr longer than what has
been measured in two wild populations [7, 8]. Thus, not only is
overall reproduction poor, but the advanced age skew is
ominous for this ex situ population?s future. Consequently, a
priority is understanding the impact of age on ovarian, oocyte,
and uterine function. Such information could hold clues to
addressing suboptimal fecundity in cheetahs in general and for
older females that are genetically valuable but still underrep-
resented in the population specifically.
The female cheetah is reproductively active throughout the
year (cycles average 13.6 days in length [9]), although with
unpredictable and variable intervals of ovarian quiescence [10].
This phenomenon has been observed widely on the basis of
direct ovarian observations [11] and endocrine patterns and is
unrelated to animal age, because periods of ovarian shutdown
occur even during prime reproductive years [9]. The species is
an induced ovulator [9], and females ranging from 3 to 12 yr of
age are highly responsive to a simple exogenous gonadotropin
treatment to provoke follicular development and ovulation [12,
13]. Aspirated follicular oocytes have the capacity to achieve
nuclear maturation and fertilize in vitro [14]. Artificial
1Supported by the Ohrstrom Foundation, the William H. Donner
Foundation, Inc., the Smithsonian?s Undersecretary for Science
Endowment, and Friends of the National Zoo.
2Correspondence: Adrienne E. Crosier, Smithsonian Conservation
Biology Institute, Center for Species Survival, 1500 Remount Road,
Front Royal, Virginia 22630. FAX: 540 635 6506;
e-mail: crosiera@si.edu
3Deceased.
Received: 10 November 2010.
First decision: 2 December 2010.
Accepted: 19 April 2011.
 2011 by the Society for the Study of Reproduction, Inc.
eISSN: 1529-7268 http://www.biolreprod.org
ISSN: 0006-3363
243
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
insemination also is more successful in younger (age, 2?5 yr)
than in older sperm recipients [13].
What is known about age effects on mammalian ovarian and
oocyte physiology comes from species distantly related to
felids (for reviews, see [15?18]). Specifically, for primate and
murine models, an accelerated loss of oocytes takes place near
the onset of reproductive senescence, occurring coincidentally
with irregular and protracted cycles [17, 19], decreases in
oocyte number [17, 19] and quality [20], and sharp declines in
fecundity [21, 22]. Circulating follicle-stimulating hormone
(FSH) and luteinizing hormone (LH) concentrations also rise as
the transition to senescence nears in humans [23] and
nonhuman primates [17, 24], whereas estrogen [24, 25] and
progesterone concentrations [25] decrease markedly. More
specific studies have indicated that advancing age influences
both nuclear and cytoplasmic factors critical for oocyte
maturation, fertilization, and embryo development. For exam-
ple, oocytes from older murine and human females experience
more DNA fragmentation, meiotic incompetence, and com-
promised fertilization compared to younger counterparts [20,
26, 27].
Minimal information is available regarding the influence of
age on reproductive function in any carnivore species. For the
domestic dog, a greater number of high-quality oocytes are
recoverable from the ovaries of adult bitches (age, 2?6.5 yr)
compared to gametes recouped from very young (age, 45?60
days) or older (age, .9.5 yr) donors [28]. In colony-reared
domestic cats, aging appears to markedly influence uterine
integrity. In one study, approximately 88% of females older
than 5 yr were experiencing cystic endometrial hyperplasia,
compared to an incidence of only 30% in 2- to 4-yr-old
individuals [29]. Similarly, a survey of 212 wild felids
representing 23 species managed in zoos has revealed more
endometrial hyperplasia in females that are older and that have
a history of nulliparity and/or exposure to exogenous
progestins (for contraceptive purposes) [30].
Our general aim in the present study was to begin
understanding the influence of age on reproductive function
in the cheetah, specifically by exploring if an effect existed on
ovarian function/activity, oocyte quality/developmental com-
petence, and uterine morphology/integrity. We hypothesized
that all three of these reproductive components were being
influenced by the aging process, especially in females aged 9 yr
or older. Our specific objectives were to measure 1) ovarian
responsiveness over time, first noninvasively via fecal steroid
metabolite patterns and then laparoscopically after gonadotro-
pin treatment; 2) oocyte quality, including the ability to achieve
maturation, fertilize, and develop to early embryo stages in
vitro; and 3) uterine status, by evaluating living females via
ultrasonography as well as histological analysis of previously
collected specimens. Thus, our comparative approach targeted
three components of the reproductive system (gonad, gamete,
and uterus) and was bolstered by our unique, special access to a
significant number of individuals of this endangered species
across three distinctive age classes. This included availability
of some cheetahs of wild-caught origin living in Namibia.
MATERIALS AND METHODS
Animals
All animal-related procedures were approved by the National Zoological
Park?s Institutional Animal Care and Use Committee and similar committees of
the Birmingham Zoo, White Oak Conservation Center, Denver Zoological
Gardens, Living Desert, Wildlife Safari, Fossil Rim Wildlife Center, and
Cheetah Conservation Fund (CCF). Adult females (age, 2?15 yr; n? 29) were
managed either singly or in groups at these North American institutions (all
participants within the SSP, n ? 12) or at the CCF (n ? 17). The latter is a
nongovernmental organization near Otjiwarongo, Namibia, dedicated to
studying and conserving this species in that range country. Five females at
the CCF facility were included in the study protocol twice, with a 1-yr interval
between procedures. Cheetah age was estimated by a thorough examination of
tooth wear [31].
Animals were assigned to one of three age category groups: young (age, 2?
5 yr; n ? 17), prime (age, 6?8 yr; n ? 8), or older (age, 9?15 yr; n ? 9).
Justification for this classification scheme was based on information gleaned
from ex situ breeding and studbook records for the North American cheetah
population [6]. Specifically, the majority (;65%) of first-time cheetah
pregnancies was established in females from 2 to 5 yr of age, with
approximately 25% of initial conceptions taking place from 6 to 8 yr of age.
A steep decline occurred in reproductive competence in females that were 9 yr
or older, with less than 5% of first pregnancies occurring during this period [5,
6].
For females in the current study, 100% (17/17) of the young, 75% (6/8) of
the prime, and 44% (4/9) of the older females were nulliparous. For each age
group, we evaluated 1) ovarian function and activity on the basis of endocrine
patterns (assessed noninvasively by measuring estradiol and progestogen
metabolites in feces) and direct ovarian observations of follicle number and
morphology after gonadotropin treatment; 2) gamete quality as determined by
subjective quality grade and objective capacity to achieve nuclear maturation,
fertilization, and early embryo development in vitro; and 3) uterine status on the
basis of gross appearance/morphometry as assessed by ultrasonography.
Additionally, to generate a more detailed understanding of the role of age
and uterine integrity in this species, we conducted a detailed histological
evaluation of fixed tissue from known-age, voucher specimens.
Ovarian Function
Ovarian activity was monitored directly and indirectly. For the former, each
ovary was examined thoroughly by laparoscopy [11, 13] after stimulation with
an established exogenous gonadotropin regimen for this species [12, 13]. In
brief, this treatment involved an i.m. injection of 300 IU of equine chorionic
gonadotropin (eCG) followed 82?86 h later by an i.m. injection of 150 IU of
human chorionic gonadotropin (hCG; Sigma Chemical Co.). Females were
anesthetized using a combination of ketamine HCl (2.0?3.5 mg/kg body wt;
Ketaset; Fort Dodge Laboratories) and medetomidine hydrochloride (22?25 lg/
kg; Domitor; Pfizer, Inc.) or a combination of medetomidine hydrochloride (40
lg/kg), butorphanol (0.3 mg/kg; Dolorex; Intervet, Inc.), and midazolam (0.2
mg/kg; Baxter Healthcare Corp.), all delivered by an i.m. injection 25?30 h
post-hCG. After endotracheal intubation, anesthesia was maintained with
isoflurane gas/oxygen inhalation as previously described [13]. After insuffla-
tion of the abdominal cavity with an ancillary Verres device, a laparoscope
(diameter, 10 mm; Olympus Co.) was inserted along the abdominal midline to
methodically examine both uterine cornua and all aspects of each ovary [11,
13]. A video camera (Olympus Co.) attached to the laparoscope was used to
record ovarian and uterine horn external morphology to allow further
assessments as necessary. The known diameter of the Verres device allowed
measuring the width and length of each ovary, with values then used to
calculate volume for each gonad using the formula (4 3 p 3 0.5H 3 0.5L 3
0.5W)/3, where H is height, L is length, and W is width, and then an overall
combined ovarian volume per animal [32]. The Verres needle also was used to
determine the length, width, and height of each follicle to generate an estimated
surface volume. Because level of vascularization in the follicle wall is
indicative of late preovulatory status and containment of a high-quality oocyte
in domestic cats [33], this metric was scored on a scale of 0 (no evidence of
surface vessels; see, e.g., Fig. 1A) to 3 (expansive, highly evident vascularity;
see, e.g., Fig. 1B), with scores then averaged for each female.
Ovarian function also was assessed in a subset of females (n? 16 [6 young,
6 prime, and 4 older]) by measuring fecal hormone metabolite concentrations
before, during, and after the eCG/hCG provocation. This was accomplished by
collecting freshly voided feces (within 24 h of excretion) once daily (4?5 days/
wk) for 45 days before gonadotropin stimulation and for 30 days after oocyte
recovery. Samples were stored frozen at 208C and then shipped to the
Smithsonian Conservation Biology Institute for analysis. Approximately 0.2 g
of dried fecal powder were boiled in 90% ethanol:10% distilled water [9]. Each
sample was centrifuged at 500 3 g for 20 min, the supernatant recovered, and
the resulting pellet redissolved in 5 ml of 90% ethanol before recentrifugation
(5003 g, 15 min). This secondary supernatant was recovered, pooled with the
first, dried under air, and redissolved in 1 ml of methanol (100%). Fecal
extracts were vortexed, then sonicated for 15 min and stored at 208C until
hormonal analysis.
Enzyme immunoassays (EIAs) for measuring estradiol and progestogen
metabolite concentrations were validated for cheetah feces by demonstrating 1)
parallelism between dilutions of pooled fecal extracts and the respective standard
curve (1.95?500 pg/well, y?1.104x 1.473, r?0.99 for estradiol and 0.78?200
244 CROSIER ET AL.
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
pg/well, y?1.173x 3.153, r?0.99 for progestogens) and 2) significant recovery
of exogenous estradiol (93%) or progestogens (85%) added to extracts before
analysis [34]. Fecal estradiol concentrations were quantified using a polyclonal
antibody assay (no. R4972; C. Munro, Department of Clinical Endocrinology,
University of California, Davis, CA) and associated horseradish peroxidase
ligand. The interassay coefficients of variation (CVs) for two internal controls (n?
46 assays) were 11.2% and 10.0%, and the intra-assay CV was less than 10%.
Progestogen metabolite concentrations were quantified using a monoclonal
antibody assay (no. CL42; Quidel Co.) and associated horseradish peroxidase
ligand [35]. The interassay CVs for two internal controls (n ? 64 assays) were
12.9% and 17.2%, and the intra-assay CV was less than 10%. The EIAs were
performed in 96-well microtiter plates (Nunc-Immuno, Maxisorp Surface; Fisher
Scientific) with assay-specific standards (Steroloids, Inc.) and diluted fecal
extracts assayed in duplicate. Means for steroid hormones were calculated for
individual animals for Days45 through 0 (pre-eCG/hCG), the 5-day peri-eCG/
hCG interval (Day ?1 to ?5, where Day ?1 is the day of eCG injection), and for
Days ?6 to ?16, Days ?17 to ?26, and Days ?27 to ?36. Mean hormone values
were pooled for each of these five intervals within the three age groups.
Oocyte Assessments, Fertilization, and Embryo
Development
At the time of laparoscopy and after quantifying ovarian metrics, oocytes
were aspirated from antral follicles (diameter, 2 mm [11]) with a 22-gauge
needle inserted through the abdominal wall and attached to an aspiration device
[36]. Recovery efficiency (number of oocytes collected/number of follicles
aspirated) for all females combined was 91.7% 6 8.2% (no difference [SEM, P
. 0.05] among age groups). Oocytes were pooled and graded according to
standard criteria for felids [37]. Briefly, grade 1 (excellent) oocytes had a
uniformly dark cytoplasm and multiple layers of expanded cumulus cells (Fig.
1C). Grade 2 (good) oocytes had a similar cytoplasmic appearance, but with
fewer layers of expanded cumulus cells. Grade 3 (average) oocytes had a
slightly vacuolated and nonuniform cytoplasm as well as few (n? 1?3) layers
of cumulus cells. Grade 4 (poor) oocytes were characterized by vacuolated and
inconsistent cytoplasm as well as a lack of cumulus cell investment.
A subset of oocytes (n ? 1?4) from each female was fixed in 2.5%
paraformaldehyde and stored at 48C before microtubule immunostaining [38].
In brief, nonspecific sites within the oocyte were blocked by incubation in 0.5%
solution of Triton X-100 in PBS with 20% fetal calf serum. Oocytes were
incubated overnight at 48C with anti-a-tubulin monoclonal antibody (Sigma
Chemical Co.) diluted 1:2000 in blocking solution and then incubated with
fluorescein isothiocyanate-labeled anti-mouse immunoglobulin G diluted 1:150
for 1 h at 388C. Chromatin was counterstained with 5 lg/ml of Hoechst 33342
(Sigma Chemical Co.) for 5 min at 388C [38]. After mounting on a slide with
Vectashield medium (Vector Laboratories), oocytes were evaluated using
epifluorescence [38]. Figure 1 depicts cheetah oocytes at anaphase I (Fig. 1D),
telophase I (Fig. 1E), and metaphase II (Fig. 1F) of maturation.
To assess fertilization and subsequent development, 282 oocytes (152 from
young; 80 from prime; 50 from older females) were inseminated in vitro with
conspecific spermatozoa that had been collected previously by electro-
ejaculation and cryopreserved [11, 39?41]. Five ejaculates from four
anesthetized [40?42] males were evaluated immediately for seminal volume
and sperm concentration (using a hemocytometer), sperm motility, and forward
progressive status [39, 40]. Samples were diluted, washed, and cryopreserved
in plastic straw containers over liquid nitrogen vapor [41, 42]. Straws were
individually thawed for 10 sec in air followed by 30 sec in a 378C water bath,
evaluated to ensure adequate sperm motility metrics and concentration (see
below), diluted in sterile Ham F10 culture medium (HF10 500 ll/straw; Irvine
Scientific, Santa Ana, CA, and centrifuged (100 3 g, 8 min) [41, 42]. After
aspirating the supernatant and swim-up processing (60 ll of fresh HF10 for 30
min at ambient temperature [42]), approximately 50 ll of sperm suspension
was recovered and the concentration and motility assessed.
Oocytes were inseminated in 50-ll microdrops of HF10 without Hepes
supplemented with pyruvate (1 mM), L-glutamine (2 mM), 10 000 IU/ml of
penicillin, and 100 lg/ml of streptomycin (Sigma Chemical Co.) under mineral
oil (38.58C, 5% CO2 in air) with approximately 4.0 3 10
6 motile sperm/ml
(from one aliquot of thawed ejaculate as processed above). Additionally, from
one to four oocytes per female were incubated in medium without sperm (as
parthenote controls). The sperm motility (average 6 SEM) immediately
postthaw for samples utilized for in vitro insemination of cheetah oocytes was
46.3% 6 2.5%. After swim-up separation, the proportion of motile cells
increased (P , 0.01) to 68.8% 6 1.4%, and the total motile sperm used for
each insemination was 1.9 6 0.53 105. The proportions of sperm with normal
morphology and intact acrosomes were 36.5% 6 3.3% and 30.5% 6 5%,
respectively, after swim-up processing. At approximately 18 h postinsemina-
tion (hpi), cumulus cells and bound spermatozoa were removed by mechanical
stripping. Presumptive zygotes then were transferred to a fresh medium droplet
and cultured in HF10 in the same in vitro environment for an additional 7 days.
The culture medium was not changed during the 7 days. Uncleaved embryos
were removed from culture and fixed at 32 hpi, and the nuclear status was
evaluated [43, 44]. Retrospective analysis of oocyte maturation or successful
fertilization was determined through visualization of the nucleus or imaging of
pronuclei poststaining. In brief, the latter involved evaluation of chromatin
status with epifluorescence microscopy after fixation in ethanol and staining
with Hoechst 33342. At Day 7 of culture (Day 8 postinsemination), embryos
earlier than the blastocyst stage of development were evaluated with
epifluorescence microscopy after fixation in ethanol and staining with Hoechst
33342 to determine cell number [43, 44]. A morula was defined as an embryo
containing 16?50 cells with no apparent blastocoele, and a blastocyst was
defined as an embryo with more than 50 cells and a discernible blastocoele by
Day 7 of culture.
Uterine Morphology and Histology
Uterine status was evaluated by two means, with the first being
ultrasonography during the anesthesia interval for laparoscopic ovarian
observation and oocyte aspiration. The uterine cornua also were observed
directly during laparoscopy, but no morphological differences were apparent.
Ultrasonography is well-established for quantifying ovarian morphology,
ovulation timing, and width and thickness of the uterine body/cornua and for
diagnosing uterine pathologies in felids [45]. For the present study, we used a
Mindray 6600 digital diagnostic imaging system (Shenzhen Mindray
Biomedical) and a linear (5.0/7.5/10-mHz) scanhead to measure uterine body
width, uterine body wall thickness, left and right uterine cornu total width (for a
representative ultrasound image to assess cornu width, see Fig. 2A), and left
FIG. 1. Cheetah ovaries and oocyte quality post-eCG/hCG and
laparoscopic recovery. Ovaries with no follicular vascularization (score
0; A) and high degree of vascularization (score 3; B) are shown. Grade 1
(excellent) oocytes (arrow; C) with uniformly dark cytoplasm and multiple
layers of expanded cumulus cells also are shown, as are oocytes in normal
anaphase I (D), with normal telophase spindle (green) and chromosome
(white arrow; E), and with normal metaphase spindle (green) and
chromosome plate (arrow; F). Original magnification 325 (C), 3400 (D),
and 31000 (E and F).
AGE INFLUENCE ON FEMALE CHEETAH REPRODUCTION 245
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
and right cornu wall thickness (for a representative image to assess wall
thickness, see Fig. 2B).
The second approach for assessing relatedness of age to uterine status
involved a histological assessment of biopsy specimens collected at necropsy
from 1987 to 2008. In total, 118 of these voucher specimens were distributed
across our age categories as follows: young (age, 2?5 yr), n? 21; prime (age,
6?8 yr), n ? 18; and older (age, 9?17 yr), n ? 79. Of the total, 105 samples
(;89%) were derived from cheetahs that were captive-born and managed in the
North American SSP population, with the remaining 13 specimens (;11%)
originating from individuals either free-ranging or captive-held (at CCF) in
Namibia. Additionally, for these 118 voucher specimens, approximately 95%
(20/21) of samples from the young group, approximately 72% (13/18) from
prime group, and approximately 60% (47/79) of samples from the older group
were derived from nulliparous females. As described above, tooth wear [31]
was used to age cheetahs in our Namibian study population. In all cases, each
reproductive tract was evaluated for tissue structure and evidence of
hyperplasia [30], and then two transverse sections of freshly excised uterine
endometrium (and any gross lesions) were fixed in 10% neutral-buffered
formalin. Tissues were embedded in paraffin, sectioned (thickness, 7 lm),
stained with hematoxylin and eosin, and then examined histopathologically as
previously described [30]. The severity of endometrial hyperplasia was
categorized as follows: grade 0, no hyperplastic changes (Fig. 3A); grade 1,
minimal-to-mild proliferative and/or cystic changes in glands or surface
epithelium without an increase in overall endometrial height (Fig. 3B); grade 2,
moderate hyperplastic and/or cystic change with an increased endometrial
thickness up to 2-fold normal (Fig. 3C); and grade 3, severe hyperplastic and/or
cystic changes with increased endometrial thickness more than 2-fold normal
(Fig. 3D) [30]. All tissues were additionally categorized for the presence and
type of severe pathology expressed, including cysts within the endometrium,
endometrial fibrosis, hydrometra, adenomyosis, chronic lymphocytic endome-
tritis, endometrial atrophy, hydrometra with endometrial atrophy, pyometra,
and endometrial polyps [30]. The proportion of cheetahs experiencing
hyperplasia and severe pathologies was plotted across the three age categories.
Statistical Analysis
Data for ovarian morphology (assessed by laparoscopy), hormone
metabolite concentrations, oocyte quality, embryo development, and uterine
morphology (evaluated by ultrasound) were analyzed by Pearson correlations
and general linear model procedures of SAS [46]. All percentage data were
arcsine transformed before analysis. Means were separated using Duncan
multiple-range test. The final model for data on ovarian morphology, including
total follicles and oocytes recovered as well as oocyte quality (maturation stage
at aspiration) included the main effects of female age. The final model for
analysis of hormone concentration data for each time interval (pre-eCG/hCG,
peri-eCG/hCG, Days ?6 to ?16, Days ?17 to ?26, and Days ?27 to ?36)
included the main effect of cheetah age group. For analysis of egg quality, the
proportion of oocytes in metaphase II per female was determined based on total
number of aspirated oocytes. The proportion of fertilized oocytes per female
then was calculated based on total number of metaphase II oocytes for that
individual. The proportion of cleaved embryos was determined based on total
number of fertilized oocytes per female. Finally, the proportion of embryos
reaching 8 and 16 cells was determined on the total number of cleaved embryos
for an individual female. The final model analyzing oocyte fertilization,
FIG. 2. Ultrasound images of an adult
cheetah uterus: uterine cornu total width (A)
and uterine cornu wall thickness (B) after
gonadotropin stimulation (arrows in both
panels indicate uterine cornua; cursors
demarcate portion measured).
FIG. 3. Histological examination of uter-
ine tissues at necropsy. Biopsies were
stained with hematoxylin-eosin and graded
for severity of hyperplasia. A) Cross-section
of a normal (grade 0) cheetah uterus with no
hyperplastic changes. B) Cross-section of a
cheetah uterus with mild (grade 1) cystic
endometrial hyperplasia and minimal-to-
mild proliferative and/or cystic changes in
glands or surface epithelium without an
increase in overall endometrial height. C)
Cross-section of a cheetah uterus with
moderate (grade 2) hyperplastic and/or
cystic change with an increased endome-
trial thickness up to 2-fold normal. D) Cross-
section of a cheetah uterus with severe
(grade 3) hyperplastic and/or cystic changes
with increased endometrial thickness more
than 2-fold normal. Bar ? 500 lm.
246 CROSIER ET AL.
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
cleavage, and late embryo development included the main effects of female age
and the covariate of ovarian volume. The final model for data on uterine
morphology assessed via ultrasound, including the incidence of cysts and
hydrometra, incorporated the main effects of age group and whether a female
had previously given birth.
Proportional data for uterine histology metrics were pooled across age
group and analyzed by Pearson correlations and nonparametric (NPAR1Way)
procedures of SAS [46]. The class variables analyzed included age group,
proven breeding status, and the interval in years from last litter produced to
animal death. Means were considered to be statistically different at P , 0.05,
and results are reported as the mean 6 SEM.
RESULTS
Ovarian Function
Age had no effect (P . 0.05) on average size of left or right
ovary (data not shown) or on total ovarian volume (Table 1).
Follicle vascularization scores were similar (P . 0.05) across
groups, but young cheetahs produced more (P , 0.05) total
follicles (diameter, 2 mm) and a higher average number of
recovered oocytes than older counterparts (Table 1). When
females of all ages were combined, total ovarian volume for a
given individual was correlated to total antral follicle number (r?
0.47, P , 0.01) and total oocytes recovered (r?0.50, P , 0.01).
Interestingly, based on fecal estradiol concentrations, all
females (for all age groups combined) had at least one estrous
cycle during the 6-wk sampling period before gonadotropin
stimulation. Fecal estradiol and progestogen metabolite
concentrations did not differ (P . 0.05) among age groups
before eCG/hCG, during the eCG/hCG peri-delivery interval,
or during the three postovulatory intervals examined (Fig. 4).
For all age groups combined (Fig. 4), estradiol values were
highest (P , 0.05) during the peri-eCG/hCG interval compared
to all other periods, with concentrations of this hormone
already basal by the Day ?6 to Day ?16 period and continuing
thereafter. Progestogen concentrations were basal and similar
(P . 0.05) in all cheetah groups during the pre- and peri-eCG/
hCG intervals, with the first detected increase occurring at
Days ?6 to ?16, with elevations again (and then stability) at
Days ?17 to ?26 and thereafter (Fig. 4).
Oocyte Assessments, Fertilization, and Embryo
Development
When the total of 222, 98, and 77 oocytes recovered from
young, prime, and older cheetahs, respectively, were evalu-
ated, the proportions meeting high-quality grade criteria were
similar (P . 0.05) (Table 1). For all cheetahs combined,
approximately 6% (3/49) of oocytes fixed at the time of
aspiration demonstrated abnormal spindle formation; all of
these oocytes were recovered from two individuals in the
young group.
The number of oocytes inseminated from young, prime, and
older females was 152, 80, and 50 oocytes, respectively, or 282
in total, of which 277 (98.2%) were quality grade 1 or grade 2
and five (1.8%) were grade 3. Because grade 1 and grade 2
felid oocytes have comparable fertilization and developmental
potential in vitro [36], gametes meeting these grades were
pooled for insemination. The five grade 3 oocytes were
recovered from two prime females.
The proportion of oocytes (those assessed at the time of
aspiration combined with those evaluated at 32 hpi) that could
be categorized as already at the metaphase II stage was
comparable (P . 0.05) among age groups and ranged from
46% to 54% (Fig. 5). Likewise, the percentages of oocytes that
were at the metaphase I, germinal vesicle, or degenerate stage
were similar (P . 0.05) (Fig. 5). None of the parthenote
controls was activated following culture in vitro. Overall,
approximately 70% of all metaphase II oocytes cultured with
sperm fertilized in vitro, an incidence of success that was
similar (P . 0.05) among age groups (Table 1). Twelve of the
282 total oocytes (4.3%) inseminated experienced polyspermy,
with no differences (P . 0.05) among groups (young, 3.9%;
prime, 6.3%; older, 2.0%). More than half of all metaphase II
oocytes from each age classification fertilized (mean range,
53.6%?68.4%), with no differences (P . 0.05) among groups
(Table 1). We observed a total of 27 morulae (Fig. 6A) and six
blastocysts (Fig. 6, B and C). Overall, approximately 34% of
fertilized and 41% of cleavage-stage embryos progressed to at
least the 8-cell stage of development in vitro; approximately
24% of all cleaved embryos developed to the morula stage. The
largest proportion of cleaved embryos developing to the morula
stage in vitro occurred in the older compared to the prime and
young groups (Table 1). All blastocysts were produced from
oocytes aspirated from two young females.
Uterine Morphology and Histology
No influence (P . 0.05) of female age was found on any of
the metrics measured via ultrasound (Table 2). Evidence of
endometrial cysts was apparent in the uterine cornua or body of
TABLE 1. Fertilization and embryo development in cheetah oocytes in vitro (means 6 SEM).
Metric
Age group
Young (2?5 yr) Prime (6?8 yr) Older (9 yr)
No. females 17 8 9
Combined ovarian volume (mm3)* 769.5 6 76.7 763.5 6 111.8 644.4 6 105.4
Average follicular vascularization 1.7 6 0.2 1.7 6 0.4 1.8 6 0.3
Average follicle no. 20.6 6 1.6a 17.2 6 2.3a,b 11.6 6 2.2b
Total no. oocytes recovered 222 98 77
Average no. oocytes recovered 16.7 6 1.7a 16.6 6 2.5a 9.6 6 2.4b
Average no. grade 1 and 2 oocytes 11.0 6 2.7 11.9 6 2.8 6.9 6 2.5
Total no. oocytes inseminated in vitro 152 80 50
Metaphase II oocytes fertilized (%) 68.3 6 17.5 67.7 6 17.2 69.2 6 15.7
Fertilized oocytes cleaved (%) 68.4 6 22.4 53.6 6 23.2 63.0 6 19.9
Cleaved oocytes reaching 8 cells (%) 34.7 6 13.6a,b 14.9 6 14.1b 63.3 6 12.0a
Cleaved oocytes reaching morula/blastocyst stage (%) 14.5 6 11.1a 0 6 0a 50.6 6 9.8b
a,b Within rows, values with different superscripts differ (P , 0.05).
* Ovarian volume calculated as [(4 3 p 3 0.5 height 3 0.5 length 3 0.5 width)/3].
 Vascularization score of 0?3, with 0 ? no vascularization and 3 ? highly vascularized.
AGE INFLUENCE ON FEMALE CHEETAH REPRODUCTION 247
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
two cheetahs in the young (ages, 3 and 5 yr), one in the prime
(age, 6 yr), and one in the older (age, 15 yr) group. Age had no
influence (P . 0.05) on prevalence of endometrial cysts in
adult cheetahs, with an overall prevalence of approximately
12% for the study cohort (Table 2). Three of the young (ages,
3, 3, and 5 yr), two of the prime (ages, 6 and 7 yr), and one of
the older (age, 9 yr) females had appreciable fluid (hydrome-
tra) within the uterine cornua (Table 2), but this condition was
unrelated (P . 0.05) to age. Additionally, none of the metrics
displayed in Table 2 differed (P . 0.05) for females on the
basis of pregnancy history; the frequency of occurrence was
comparable between individuals that had given birth versus
those that were nulliparous (data not shown).
Based on the histological assessments, the prevalence of
uterine hyperplasia (all grades combined) increased (P , 0.05)
with age, being 19% for the young group, 50% for the prime
group, and 87% for the older group (Fig. 7). Age also
influenced (P , 0.01) the incidence of grade of hyperplasia;
for example, grade 3 hyperplasia (the most serious and
advanced disease observed for this dataset) was measured in
19% of older cheetahs but in none of the young or prime
females. Of the young cohort, only 19% expressed any form of
hyperplasia, with all being grade 1 (least serious). By contrast,
grade 1 and 2 hyperplasia was measured in 50% and 5.5%,
respectively, of the prime and in 44% and 28%, respectively,
of the older age counterparts. The severe pathologies identified
histologically included cysts within the endometrium, endo-
metrial fibrosis, hydrometra, adenomyosis, chronic lympho-
cytic endometritis, endometrial atrophy, hydrometra with
endometrial atrophy, pyometra, and endometrial polyps. The
prevalence of severe uterine pathologies was highly age-
dependent (P , 0.0001), with anomalies observed in only
approximately 5% of tissues analyzed from the young cheetahs
(n? 1), approximately 22% of tissues from the prime cheetahs
(n ? 4), and approximately 56% of tissues from the oldest
cheetahs (n ? 44) (Fig. 7). By far, the most common
pathologies were adenomyosis (13% of all females), hydro-
metra (12%), and pyometra (7%), and often, these anomalies
occurred in combination. It was not unusual for individuals
from the older group to show multiple pathologies simulta-
3
FIG. 4. Average concentrations of estradiol (A?E) and progestogen (F?J)
metabolites for cheetah females: (A and F) 45 days before eCG/hCG
stimulation, (B and G) the 5-day peri-eCG/hCG interval (Day ?1 [eCG
injection] to Day ?5), (C and H) Day ?6 to Day ?16, (D and I) Day ?17
to Day ?26, and (E and J) Day ?27 to Day ?36. Values are presented as
the mean 6 SEM. Within hormone (estradiol or progestogens) and for all
age groups combined, values with different letters differ (P , 0.05).
TABLE 2. Cheetah uterine metrics as assessed by ultrasonography of
living individuals (means 6 SEM).
Metric
Age group
Young
(2?5 yr)
Prime
(6?8 yr)
Older
(9 yr)
No. females 17 8 9
Uterine body width (mm) 11.1 6 1.4 12.9 6 0.9 11.3 6 0.8
Uterine body wall thickness (mm) 3.7 6 0.5 4.4 6 0.3 4.3 6 0.3
Left cornu wall thickness (mm) 2.8 6 0.3 2.8 6 0.2 2.9 6 0.2
Right cornu wall thickness (mm) 2.7 6 0.3 2.9 6 0.2 3.0 6 0.2
Left uterine cornu width (mm) 7.0 6 0.8 7.2 6 0.6 7.2 6 0.5
Right uterine cornu width (mm) 6.8 6 0.6 7.3 6 0.5 7.1 6 0.4
Incidence of uterine cysts (%) 11.8 12.5 11.1
Incidence of hydrometra (%) 17.7 25.0 11.1
248 CROSIER ET AL.
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
neously at necropsy. For example, of the 79 older females,
seven (;9%) had a combination of hydrometra with
accompanying adenomyosis or endometritis, adenomyosis
with pyometra or fibrosis, and one female with a combination
of polyps, cysts within the endometrium, hydrometra, and
adenomyosis. Again, reproductive history (parity vs. nullipar-
ity) had no influence (P . 0.05) on uterine condition and
integrity. However, the longer the time interval (in years) from
the age a proven female last gave birth to her death, the more
likely (r? 0.68, P , 0.05) that individual expressed a grade 2
or grade 3 hyperplasia as well as severe uterine pathologies (r?
0.57, P , 0.05).
DISCUSSION
Although the influence of aging has been only modestly
studied in animal reproductive science, this topic is particularly
relevant to ex situ collections of rare wildlife species. As
significant advances have been made in husbandry and
geriatric medicine, security populations of endangered species
now live longer, and knowledge regarding reproductive
capacity in the context of life span is needed [30, 47]. This
is especially important for managed populations, in which the
goal is to retain maximal genetic diversity through carefully
planned breeding programs [4, 48]. In such instances, the most
valuable individuals are matched for propagation, whereas
senescent animals are retired from breeding to be used for
public education in zoos. For species relying on such intensive
management, the need also exists to understand which
component (or components) of the reproductive system fails
and at what age, especially in instances when assisted breeding
technologies could help ensure production of young from
??underrepresented?? individuals [4, 49]. In the case of older
female cheetahs, we presumed that there would be a concurrent
decline in ovarian function, oocyte quality, and uterine
competence over time, thereby explaining reduced fecundity
in older individuals. Interestingly, the older cohort, although
producing fewer ovarian follicles and recoverable oocytes, was
surprisingly normal in ovarian function and gamete quality
based on our metrics. The influence of age within the time
spans evaluated had minimal impact on steroidogenesis, grade
of oocyte produced, maturational ability, fertilization, or early
embryogenesis in vitro. In contrast, age had a significant
impact on uterine integrity that, although undetectable using
ultrasonography, was evident from histological assessments of
voucher specimens. Therefore, we concluded that the reduced
fecundity routinely observed in older female cheetahs was
caused largely by a high prevalence of uterine endometrial
hyperplasia and related serious pathologies and not necessarily
by ovarian or gamete dysfunction.
Our approach was thorough in that we compared function-
ality and integrity at three levels?gonad, oocyte, and uterus?
and, despite the species? rarity, across a significant number of
individuals. Our strategy also was bolstered by having
significant biological information regarding the cheetah,
probably the most studied ??wild?? animal in the reproductive
sciences [4]. Besides extensive physiological data on sperm
form and function [40, 41, 50, 51] and endocrine patterns
(gonadal [9, 10] and adrenal [10, 52, 53]), substantial data exist
regarding ovarian morphology, activity, and sensitivity to
exogenous gonadotropins [11?13]. Because the cheetah is an
induced ovulator [9], and due to inexplicable and often
protracted intervals of ovarian quiescence [9], the ovary of
this species generally responds well and consistently to the
eCG/hCG regimen used in the present study [12, 13]. Although
producing only half the number of mature follicles measured in
young or prime counterparts, follicles from older females were
comparable in morphology (including vascularization) and
contained equivalent-quality oocytes and fertilization capacity
(see below). Likewise, no discernible differences in mean
excreted estrogen was found during the peri-gonadotropin
FIG. 5. Incidence of oocyte maturation for cheetahs across three age
groups. Values are presented as the mean 6 SEM.
FIG. 6. Morula (A) and blastocyst (B) produced by IVF (light microscopy)
as well as cheetah blastocyst (C) stained with Hoechst to assess blastomere
number. Bar ? 50 lm.
FIG. 7. Proportion of voucher specimens indicative of uterine endome-
trial hyperplasia and the incidence of severe uterine pathologies across
cheetahs of three age groups. Within category, values with different letters
differ significantly (P , 0.05).
AGE INFLUENCE ON FEMALE CHEETAH REPRODUCTION 249
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
treatment interval (or at other periods), thereby suggesting
normative folliculoestrogenic activity across age groups.
Similarly, luteal function as measured by mean progestogens
was comparable, especially during the postaspiration interval,
indicating that steroid excretion capacity after a gonadotropin
challenge was unaffected by age. Clearly, the ovaries of older
female cheetahs had retained significant sensitivity to both an
FSH and LH stimulus, because follicles were provoked to
develop, and oocytes were recoverable that later achieved
nuclear maturation and fertilization in vitro. The reduction in
follicle number (and, naturally, in recovered oocytes) in the
older group logically may have been due to an age-related,
modest loss in gonadotropin receptivity in the ovary or, more
likely, simply to fewer available, healthy follicles.
Our findings that the ovaries of older cheetahs are
responsive to gonadotropins help explain the observations of
behavioral estrus in aged animals, even individuals up to 12 yr
of age [10]. A portion of the historically poor reproductive
success of this species in captivity has been attributed to subpar
management [10] and/or incompatibility within a breeding pair
[11], but clearly, age also is a factor. Record assessments have
determined that females 9 yr of age or older have produced
only 10% of the litters born since 1980 in accredited North
American institutions [6]. Additionally, less than 5% of
females become pregnant for the first time after the age of 8
yr [6]. Our findings of fewer ovarian follicles and oocytes
recovered in older cheetahs also were compatible with earlier
observations that animals in this age group consistently
produce smaller-size litters (approximately two cubs each)
compared to younger counterparts (approximately four cubs
each) [6]. Of course, this difference also could be related to the
common discovery of a more compromised uterine environ-
ment in older individuals.
Indeed, strong evidence suggests that as the cheetah ages,
the uterus becomes inhospitable, including by as early as 2 yr
(one young individual with hydrometra and endometrial
atrophy). Interestingly, only a modest number of uterine
abnormalities was observed via ultrasonography and even
fewer by direct laparoscopic observations that, of course, only
permitted surface evaluations of the cornua and uterine body.
Retrospective histology of appropriately stored voucher
specimens from necropsied cheetahs (generally housed origi-
nally at the same institutions as the living counterparts) was a
more useful source of information. Clearly, older animals had
significantly more evidence of uterine anomalies than coun-
terpart groups. Even so, half of prime females were
experiencing endometrial hyperplasia, compared to more than
85% of the older group. Histological assessments were even
more enlightening in identifying a complexity of individual and
combined serious pathologies, some of which may have
eventually been diagnosed at a well-advanced stage by
ultrasonography [45]. The types and severity of these
conditions likely prevent implantation and/or sustaining an
early stage nidation [30].
Endometrial hyperplasia has been most thoroughly studied
in humans [54] and, to a limited extent, in companion animals
[55], and in both cases, it is known to lead to endometrial
dysfunction and, possibly, sterility [30]. Endometrial hyper-
plasia in ex situ populations of wild felids has been recognized
for years, but generally related to the protracted use of an
exogenous progestin (melengestrol acetate [MGA]) as a
contraceptive to prevent reproduction in already well-repre-
sented individuals [30, 56, 57]. In one study of 212 zoo-
managed specimens representing 23 felid species, endometrial
hyperplasia was prevalent, with more than 70% of all females
demonstrating some degree of this pathology and almost 60%
of those cases being considered as severe [30]. In the latter
study, cases of moderate-to-severe cystic endometrial hyper-
plasia were found in 85% of individuals implanted with MGA
for a minimum of 6 mo [30]. The causative mechanism is
believed to have been uninterrupted, high-concentration
progestin exposure that induces secretory differentiation of
endometrial epithelial cells [30]. This hyperstimulated increase
in cell number and differentiation causes cystic and adenoma-
tous changes characteristic of hyperplasia (hyperchromasia,
pseudostratification, and a high nuclear:cytoplasmic ratio) [30].
However, our study population of cheetahs was never exposed
to exogenous MGA or another progestin. Furthermore, because
this species is an induced ovulator [9], an opportunity rarely
existed for uterine tissue to be influenced by endogenous
progesterone unless following natural corpus luteum formation
after mating and ovulation opportunities. An alternative
explanation for observations made in the present study was
provocation of hyperplasia and other severe pathologies from
prolonged reproductive cyclicity in the absence of occasional
??protective?? progesterone afforded by ovulation and pregnan-
cy [30]. For example, cheetah females in nature produce a first
litter at approximately 3 yr of age and then subsequent litters
every 2 yr until death [2, 7]. As long as competent males are in
the vicinity, wild cheetah females appear to experience little
time cycling and, rather, are either pregnant, lactating, or
raising young [7, 8]. Therefore, a major difference between the
uterus of cheetahs living in situ versus ex situ is that the former
are more exposed to endogenous progesterone whereas the
latter are experiencing waves of estrogen (associated with
frequent cyclicity) punctuated by bouts of ovarian quiescence
[9]. For cheetahs managed in captivity that fail to mate and
ovulate, this nonprogestogenic environment could continue for
more than 8 yr. The dominant presence of estrogen then may
contribute to a proliferative impact on endothelial tissues, with
one mechanism of action perhaps expressed via an epidermal
growth factor (EGF) pathway [58, 59]. In the uterus, estradiol
induces the EGF receptor to stimulate cellular replication [58],
and specifically, both in vivo (rat) and in cell cultures (human),
estrogen up-regulates uterine EGF receptor numbers and,
therefore, overall cellular growth and proliferation of the
endometrium [59, 60].
Regardless of the causative source and mechanism of action,
the direct impact of aging on increasing the incidence of
insidious pathologies within the uterus likely influences the
capacity to produce offspring, even in well-timed ovulating and
successfully mated or artificially inseminated females. Pre-
suming that a protective effect of progesterone exists in felids,
then practical implications follow, perhaps even offering
suggestions for how to reduce the risk of endometrial
hyperplasia and associated lesions. Munson et al. [30] have
predicted that compared to multiparity, a nulliparous condition
increases the chances for a compromised uterus and, thus,
infertility in felids. Our dataset on uterine histology was
unhelpful in supporting this claim, largely because only a few
of the animals in the young (n ? 1) and prime (n ? 5) age
groups had previously produced young. Nonetheless, if
periodic progesterone exposure is safeguarding to the felid
uterus, it does not protect in the long term. Within our
histological dataset, more than 40% (n? 32) of the older group
was multiparous. Of these, however, 87.5% (n ? 28) had
endometrial hyperplasia, and approximately 44% (n ? 14)
experienced other serious uterine pathologies, including
pyometra, adenomyosis, and endometritis. Yet, it is possible
that duration from last exposure to steroid is important. This
might explain our discovery that cheetahs were more likely to
experience uterine hyperplasia as the interval between last litter
250 CROSIER ET AL.
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
and death increased. Nonetheless, other factors are known to
predispose a carnivore to uterine hyperplasia and associated
pathologies. For example, in the cat, elevated serum estradiol
levels [29] as well as exposure to progestins, advancing age,
and nulliparity [30] predispose individuals to hyperplasia.
Similarly, hyperplasia in canine species is believed to be
related largely to exposure to progestins and advancing age
[61] as well as nulliparity [45], with diestrus and postpartum
metritis also leading bitches to hyperplastic conditions [45].
Regardless, our observations justify the need for more detailed
explorations of steroidal influences on felid uterine integrity.
Meanwhile, these findings also support the practical notion that
more effort be made to achieve pregnancies in young cheetahs
managed in security populations, including achieving all
needed reproduction during the early years (age, ,8 yr).
Because we discovered severe uterine pathologies even in
individuals considered to be of ??prime?? age, it appears prudent
to focus on reproducing cheetahs during the first 5 yr of life,
similar to what occurs in nature [7, 8].
Beyond our primary focus of examining the influence of
animal age on the cheetah?s ovary, oocyte, and uterine
integrity, it was encouraging to determine the possibility of
consistently producing embryos in vitro. Embryo technolo-
gies, including in vitro fertilization (IVF) and embryo
transfer, are not routinely used in the ex situ management of
genetically valuable, rare felid species [4, 49, 62]. Nonethe-
less, the literature contains important examples of how these
collective approaches have been used to produce ??milestone??
births, including in the African wild cat, caracal [63, 64], and
ocelot [62]. Therefore, the biological potential of felid
embryos generated from in vitro maturation/IVF has been
demonstrated, suggesting the potential of this approach for
ramping up animal numbers and ensuring the retention of
genetic diversity in intensively managed populations. IVF can
be challenging in felids, including the cheetah, often related to
a preponderance of pleiomorphic spermatozoa in the ejaculate
[11, 40, 41, 65]. In the only earlier publication, to our
knowledge, involving gamete interaction in culture for the
cheetah [14], 12 females were treated with eCG/hCG and
produced an average of 24 mature follicles and 23 oocytes
recovered per animal. Approximately 90% of these were
classified as mature (based on gross morphology), and 26%
demonstrated evidence of fertilization, with 11 embryos
developing to the 16-cell stage and none beyond. By contrast,
in the present study, we recovered an average of 15 oocytes
per female (across a diverse age group), with approximately
50% achieving maturation, 68% of mature oocytes fertilizing,
and 27 embryos reaching at least the 16-cell stage in culture.
The two studies differed vastly in how oocyte quality was
assessed. In the earlier investigation, oocytes were determined
to be mature on the basis of a simple visual assessment?
specifically, examining for expansion of cumulus oophorus
cells [14]. Some, or even many, of the oocytes incubated with
sperm in that study likely were immature, resulting in low
fertilization and poor embryo development. Our contempo-
rary investigation is strengthened by the ability to retrospec-
tively evaluate nuclear maturation for all oocytes failing to
cleave by 32 hpi using the advanced methods of microtubule
immunostaining or Hoechst staining, thereby allowing
nuclear status assessment of every aspirated oocyte. The
present study also benefited by using high-quality cryopre-
served sperm samples from a limited number of donors.
Fertilization was achieved by all samples from each male,
with no influence of donor on subsequent embryo develop-
ment.
Perhaps most important was our finding of no discernible
differences in oocyte quality, fertilizability, and early
embryogenesis among the three cheetah age groups in the
present study. Additional metrics, such as assessment of
embryo metabolism or live offspring production posttransfer,
would provide even greater insight regarding oocyte quality
and developmental quality in this species. However, we find it
encouraging that we were able to recover an average of nine
oocytes from every cheetah in the 9- to 15-yr age group, of
which almost seven met the highest quality criteria. That the
ability to achieve nuclear maturation and fertilization was no
different than in the young and prime groups also indicated
that older females could produce significant numbers of
embryos in vitro, including those advancing to the morula
stage (in at least four of the nine older females). The capacity
of these embryos to translate into living offspring remains to
be determined, but the availability of viable-appearing
oocytes from older females that mature, interact with sperm,
and form early blastocyst-stage embryos presents interesting
opportunities for management and conservation. For example,
most cheetahs older than 8 yr currently are disregarded as
breeding candidates due to low fertility expectations.
However, more than 35% of cheetahs in the contemporary
North American population are in this age frame, with 85% of
these having never reproduced. Our findings in the present
study suggest the need for a more serious examination of the
potential role of embryo technologies in cheetah manage-
ment?specifically, oocyte recovery and IVF followed by
embryo transfer to already well-represented (and still young)
cheetah recipients. It was noteworthy that embryo production
in the present study was accomplished using conspecific
spermatozoa cryopreserved and thawed using data from recent
cryostudies [41, 42]. Therefore, the source of male germplasm
for future IVF efforts could be sperm from systematic sperm
banking, including from free-living males [49]. Now that it is
apparent viable oocytes are recoverable even from aged
conspecifics, a logical next step is exploring IVF/embryo
transfer as a way to avoid loss of gene diversity, including
transfer from older females that likely will never reproduce by
natural means.
ACKNOWLEDGMENTS
The authors thank the White Oak Conservation Center, Fossil Rim
Wildlife Center, Birmingham Zoo, Denver Zoological Gardens, Living
Desert, CCF, and Wildlife Safari for cooperation. We also thank Drs.
Arthur Bagot-Smith, Karen Terio, and Janine Brown for assistance and
advice and Nicole Presley, Sarah Putman, Jason Ahistus, Marianne de
Jonge, Bonnie Schumann, Michelle Bacon, Elizabeth Lester, and Phil
Randle for technical support.
REFERENCES
1. Marker-Kraus L, Kraus D. The history of cheetahs in Namibia. Swara
1993; 16:8?12.
2. Marker-Kraus L, Kraus D. The Namibian free-ranging cheetah. Environ
Conserv 1995; 21:369?370.
3. Marker-Kraus LL, Kraus D, Barnett D, Hurlbut S. Cheetah Survival on
Namibian Farmlands. Windhoek, Namibia: Cheetah Conservation Fund;
1996.
4. Wildt DE, Swanson W, Brown JL, Sliwa A, Vargas A. Felids ex situ:
Managed programmes, research, and species recovery. In: Macdonald
DW, Loveridge AJ (eds.), Biology and Conservation of Wild Felids.
Oxford: Oxford University Press; 2010:217?236.
5. Cheetah Species Survival Plan (SSP). Silver Spring, MD: American Zoo
and Aquarium Association Cheetah Master Plan, Annual Report, 2010.
6. Marker LL. International Studbook Cheetah (Acinonyx jubatus).
Otjiwarongo, Namibia: Cheetah Conservation Fund; 2008.
7. Caro TM. Cheetahs of the Serengeti Plains: Group Living in an Asocial
Species. Chicago: The University of Chicago Press; 1994.
AGE INFLUENCE ON FEMALE CHEETAH REPRODUCTION 251
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
8. Marker LL, Dickman AJ, Jeo RM, Mills MGL, Macdonald DW.
Demography of the Namibian cheetah (Acinonyx jubatus jubatus). Biol
Conserv 2003; 114:413?425.
9. Brown JL, Wildt DE, Wielebnowski N, Goodrowe KL, Graham LH, Wells
S, Howard JG. Reproductive activity in captive female cheetahs (Acinonyx
jubatus) assessed by fecal steroids. J Reprod Fertil 1996; 106:337?346.
10. Wielebnowski NC, Ziegler K, Wildt DE, Lukas J, Brown JL. Impact of
social management on reproductive, adrenal and behavioral activity in the
cheetah (Acinonyx jubatus). Anim Conserv 2002; 5:291?301.
11. Wildt DE, Brown JL, Bush M, Barone MA, Cooper KA, Grisham J,
Howard JG. Reproductive status of cheetahs (Acinonyx jubatus) in North
American zoos: the benefits of physiological surveys for strategic
planning. Zoo Biol 1993; 12:45?80.
12. Howard JG, Donoghue AM, Barone MA, Goodrowe KL, Blumer ES,
Snodgrass K, Starnes D, Tucker M, Bush M, Wildt DE. Successful
induction of ovarian activity and laparoscopic intrauterine artificial
insemination in the cheetah (Acinonyx jubatus). J Zoo Wildl Med 1992;
23:288?300.
13. Howard JG, Roth TL, Byers AP, Swanson WF, Wildt DE. Sensitivity to
exogenous gonadotropins for ovulation induction and laparoscopic
artificial insemination in the cheetah and clouded leopard. Biol Reprod
1997; 56:1059?1068.
14. Donoghue AM, Howard JG, Byers AP, Goodrowe KL, Bush M, Blumer
E, Lukas J, Stover J, Snodgrass K, Wildt DE. Correlation of sperm
viability with gamete interaction and fertilization in vitro in the cheetah
(Acinonyx jubatus). Biol Reprod 1992; 46:1047?1056.
15. Wu JM, Zellinski MB, Ingram DK, Ottinger MA. Ovarian aging and
menopause: current theories, hypotheses, and research models. Exp Biol
Med 2005; 230:818?828.
16. te Velde ER, Scheffer GJ, Dorland M, Broekmans FJ, Fauser BC.
Developmental and endocrine aspects of normal ovarian aging. Mol Cell
Endocrinol 1998; 25:67?73.
17. Atsalis S, Videan E. Reproductive aging in captive and wild common
chimpanzees: factors influencing the rate of follicular depletion. Am J
Primatol 2009; 71:271?282.
18. Baird DT, Collins J, Egozcue J, Evers LH, Gianaroli L, Leridon H, Sunde
A, Templeton A, Van Steirteghem A, Cohen J, Crosignani PG, Devroey P,
et al. Fertility and aging. Hum Reprod 2005; 11:261?276.
19. de Boer EJ, den Tonkelaar I, te Velde ER, Burger CW, Klip H, van
Leeuwen FE. A low number of retrieved oocytes at in vitro fertilization
treatment is predictive of early menopause. Fertil Steril 2002; 77:978?985.
20. Fujino Y, Ozaki K, Yamamasu S, Ito F, Matsuoka I, Hayaski E, Nakamura
H, Ogita S, Sato E, Inoue M. DNA fragmentation of oocytes in aged mice.
Hum Reprod 1996; 11:1480?1483.
21. Gagliardi C, Liukkonen JR, Phillippi-Falkenstein KM, Harrison RM,
Kubisch HM. Age as a determinant of reproductive success among captive
female rhesus macaques (Macaca mulatta). Reproduction 2007; 133:819?
826.
22. van Asselt KM, Kok HS, Pearson PL, Dubas JS, Peeters PHM, te Velde
ER, van Noord PAH. Heritability of menopausal age in mothers and
daughters. Fertil Steril 2004; 82:1348?1351.
23. Taffe J, Dennerstein L. Time to the final menstrual period. Fertil Steril
2002; 78:397?403.
24. Schramm RD, Paprocki AM, Bavister BD. Features associated with
reproductive aging in female rhesus monkeys. Hum Reprod 2002;
17:1597?1603.
25. Mayer LP, Devine PJ, Dyer CA, Hoyer PB. The follicle-deplete mouse
ovary produces androgen. Biol Reprod 2004; 71:130?138.
26. Djahanbakhch O, Ezzati M, Zosmer A. Reproductive ageing in women. J
Pathol 2007; 211:219?231.
27. Martin RH. Meiotic errors in human oogenesis and spermatogenesis.
Reprod Biomed Online 2008; 16:523?531.
28. Rocha AA, Bastos R, Cunha ICN, Adona PR, Santos JA. Quantity and
quality of oocytes recovered from donor bitches of different ages.
Theriogenology 2006; 66:1465?1467.
29. Perez JF, Conley AJ, Dieter JA, Sanz-Ortega J, Lasley BL. Studies on the
origin of ovarian interstitial tissue and the incidence of endometrial
hyperplasia in domestic and feral cats. Gen Comp Endocrinol 1999;
116:10?20.
30. Munson L, Gardner A, Mason RJ, Chassy LM, Seal US. Endometrial
hyperplasia and mineralization in zoo felids treated with melengestrol
acetate contraceptives. Vet Pathol 2002; 39:419?427.
31. Marker LL, Dickman AJ. Morphology, physical condition, and growth of
the cheetah (Acinonyx jubatus jubatus). J Mammal 2003; 84:840?850.
32. Howard JG, Brown JL, Bush M, Wildt DE. Teratospermic and
normospermic domestic cats: ejaculate traits, pituitary-gonadal hormones,
and improvement of spermatozoal motility and morphology after swim-up
processing. J Androl 1990; 11:204?214.
33. Pelican KM, Spindler RE, Pukazhenthi BS, Wildt DE, Ottinger MA,
Howard JG. Progestin exposure before gonadotropin stimulation improves
embryo development after in vitro fertilization in the domestic cat. Biol
Reprod 2010; 83:558?567.
34. Brown JL, Wasser SK, Wildt DE, Graham LH. Comparative aspects of
steroid hormone metabolism and ovarian activity in felids, measured
noninvasively in feces. Biol Reprod 1994; 51:776?786.
35. Graham L, Schwarzenberger F, Mostl E, Galama W, Savage A. A versatile
enzyme immunoassay for the determination of progesterone in feces and
serum. Zoo Biol 2001; 20:227?236.
36. Spindler RE, Pukazhenthi BS, Wildt DE. Oocyte metabolism predicts the
development of cat embryos to blastocyst in vitro. Mol Reprod Dev 2000;
56:163?171.
37. Wood TC, Wildt DE. Effect of the quality of the cumulus-oocyte complex
in the domestic cat on the ability of oocytes to mature, fertilize, and
develop into blastocysts in vitro. J Reprod Fertil 1997; 110:355?360.
38. Comizzoli P, Wildt DE, Pukazhenthi BS. Effect of 1,2-propanediol versus
1, 2-ethanediol on subsequent oocyte maturation, spindle integrity,
fertilization, and embryo development in vitro in the domestic cat. Biol
Reprod 2004; 71:598?604.
39. Howard JG. Semen collection and analysis in carnivores. In: Fowler ME
(ed.), Zoo and Wild Animal Medicine: Current Therapy III. Philadelphia:
W.B. Saunders Co.; 1993:390?399.
40. Crosier AE, Marker L, Howard J, Pukazhenthi BS, Henghali JN, Wildt
DE. Ejaculate traits in the Namibian cheetah (Acinonyx jubatus): influence
of age, season and captivity. Reprod Fertil Dev 2007; 19:370?382.
41. Crosier AE, Pukazhenthi BS, Henghali JN, Howard J, Dickman AJ,
Marker L, Wildt DE. Cryopreservation of spermatozoa from wild-born
Namibian cheetahs (Acinonyx jubatus) and influence of glycerol on
cryosurvival. Cryobiology 2006; 52:169?181.
42. Crosier AE, Henghali JN, Howard JG, Pukazhenthi BS, Terrell KA,
Marker LL, Wildt DE. Improved quality of cryopreserved cheetah
(Acinonyx jubatus) spermatozoa after centrifugation through Accudenz. J
Androl 2009; 30:298?308.
43. Comizzoli P, Wildt DE, Pukazhenthi BS. Overcoming poor in vitro
nuclear maturation and developmental competence of domestic cat oocytes
during the nonbreeding season. Reproduction 2003; 126:809?816.
44. Comizzoli P, Mermillod P, Cognie Y, Chai N, Legendre X, Mauget R.
Successful in vitro production of embryos in the red deer (Cervus elaphus)
and the silka deer (Cervus nippon). Theriogenology 2001; 55:649?659.
45. Davidson AP, Baker TW. Reproductive ultrasound of the bitch and queen.
Top Companion Anim Med 2010; 24:55?63.
46. SAS Institute. SAS/STAT User?s Guide: Statistics. Release 9.1 ed. 2002.
Cary, NC: Statistical Analysis System Institute, Inc.; 2002.
47. Margulis SW, Atsalis S, Bellem A, Wielebnowski N. Assessment of
reproductive behavior and hormonal cycles in geriatric western lowland
gorillas. Zoo Biol 2007; 26:117?139.
48. Wildt DE, Comizzoli P, Pukazhenthi B, Songsasen N. Lessons from
biodiversity?the value of nontraditional species to advance reproductive
science, conservation, and human health. Mol Reprod Dev 2010; 77:397?
409.
49. Wildt DE, Rall WF, Critser JK, Monfort SL, Seal US. Genome resource
banks: living collections for biodiversity conservation. BioScience 1997;
47:689?698.
50. Durrant BS, Millard SE, Zimmerman DM, Lindburg DG. Lifetime semen
production in a cheetah (Acinonyx jubatus). Zoo Biol 2001; 20:359?366.
51. Wildt DE, O?Brien SJ, Howard JG, Caro TM, Roelke ME, Brown JL,
Bush M. Similarity in ejaculate-endocrine characteristics in captive versus
free-ranging cheetahs of two subspecies. Biol Reprod 1987; 36:351?360.
52. Terio K, Citino SB, Brown JL. Fecal cortisol metabolite analysis for
noninvasive monitoring of adrenocortical function in the cheetah
(Acinonyx jubatus). J Zoo Wildl Med 1999; 30:484?491.
53. Terio KA, Marker L, Munson L. Evidence for chronic stress in captive but
not free-ranging cheetahs (Acinonyx jubatus) based on adrenal morphol-
ogy and function. J Wildl Dis 2004; 40:259?266.
54. Lane BF, Wong-You-Cheong JJ. Imaging of endometrial pathology. Clin
Obstet Gynecol 2009; 52:57?72.
55. Schlafer DH, Gifford AT. Cystic endometrial hyperplasia, pseudoplacenta-
tional endometrial hyperplasia, and other cystic conditions of the canine
and feline uterus. Theriogenology 2008; 70:349?358.
56. Munson L. Contraception in felids. Theriogenology 2006; 66:126?134.
57. Kazensky CA, Munson L, Seal US. The effects of melengestrol acetate on
the ovaries of captive wild felids. J Zoo Wildl Med 1998; 29:1?5.
58. Levin ER. Bidirectional signaling between the estrogen receptor and the
epidermal growth factor receptor. Mol Endocrinol 2003; 17:309?317.
252 CROSIER ET AL.
D
ow
nloaded from
 w
w
w
.biolreprod.org. 
59. Mukku VR, Stancel GM. Regulation of epidermal growth factor receptor
by estrogen. J Biol Chem 1985; 260:9820?9824.
60. Igna-Trowbridge DM, Pimentel M, Teng CT, Korach KS, McLachlan JA.
Cross talk between peptide growth factor and estrogen receptor signaling
systems. Environ Health Perspect 1995; 103:35?38.
61. Moresco A, Munson L, Gardner IA. Naturally occurring and melengestrol
acetate-associated reproductive tract lesions in zoo canids. Vet Pathol
2009; 46:1117?1128.
62. Swanson WF. Application of assisted reproduction for population
management in felids: the potential and reality for conservation of small
cats. Theriogenology 2006; 66:49?58.
63. Pope CE, Gomez MC, Dresser BL. In vitro production and transfer of cat
embryos in the 21st century. Theriogenology 2006; 66:59?71.
64. Pope CE. Embryo technology in conservation efforts for endangered
felids. Theriogenology 2000; 53:163?174.
65. Pukazhenthi BS, Wildt DE, Howard JG. The phenomenon and
significance of teratospermia in felids. J Reprod Fertil Suppl 2001;
57:423?433.
AGE INFLUENCE ON FEMALE CHEETAH REPRODUCTION 253
D
ow
nloaded from
 w
w
w
.biolreprod.org.