BIOLOGY OF REPRODUCTION 67, 247–255 (2002)
Comparative Testis Morphometry and Seminiferous Epithelium Cycle Length
in Donkeys and Mules1
Elizabeth S. Neves, Hélio Chiarini-Garcia, and Luiz R. França2
Laboratory of Cellular Biology, Department of Morphology, Institute of Biological Sciences,
Federal University of Minas Gerais, Belo Horizonte, Minas Gerais, Brazil 31270-901
ABSTRACT
The mule (Equus mulus mulus) is a sterile hybrid domestic
animal that results from the breeding of a male donkey (Equus
asinus) to a female horse (Equus caballus). Usually, spermato-
genesis in mules does not advance beyond spermatocytes. In the
present study, we performed a comparative and more accurate
morphometric and functional investigation of the testis in don-
keys and mules. Due to the smaller testis size, lower seminif-
erous tubule volume density, and fewer germ cells, the total
length of seminiferous tubules in mules was significantly smaller
than in donkeys. However, the percentage of seminiferous tu-
bules containing germ cells (spermatogonia and spermatocytes)
in mules was approximately 95%. The total number of Sertoli
cells per testis observed in donkeys and mules was very similar.
However, the total number of Leydig cells in mules was approx-
imately 70% lower than in donkeys. At least in part, this differ-
ence was probably related to the lower number of germ cells
present in mule seminiferous tubules. Although spermatogenesis
in mules did not advance beyond secondary spermatocytes/new-
ly formed round spermatids, germ cell associations in the sem-
iniferous epithelium and pachytene spermatocytes nuclear vol-
ume in donkeys and mules were similar. The duration of sper-
matogenesis was estimated using intratesticular injections of tri-
tiated thymidine. Each spermatogenic cycle in donkeys lasted
10.5 days. A similar value was found in mules (;10.1 days).
Considering that the entire spermatogenic process takes ap-
proximately 4.5 cycles to be completed, its total duration in
donkeys was estimated to last 47.2 days. The results found for
mules suggest that the mechanisms involved in the determina-
tion of testis structure and function are probably originated from
donkeys. Also, the data found for mules suggest that their sem-
iniferous tubules are able to sustain complete spermatogenesis.
In this regard, this species is a potential model for transplants
of germ cells originated from donkeys and horses or other large
animals.
apoptosis Leydig cells, Sertoli cells, spermatogenesis
INTRODUCTION
The male mule (Equus mulus mulus, 63 chromosomes)
is a sterile domestic animal that results from the breeding
of a male donkey (Equus asinus, 62 chromosomes) to a
female horse (Equus caballus, 64 chromosomes) [1]. As the
number of metacentric and acrocentric autosomal chromo-
somes in horses and donkeys diverges approximately 35%,
1
The scholarship awarded to E.S.N. from the Brazilian Foundation (CAPES)
is fully appreciated. Financial support from the Minas Gerais State Foun-
dation (FAPEMIG) and the Brazilian Research Council (CNPq) is gratefully
acknowledged.
2
Correspondence. FAX: 55 31 34992780; e-mail: lrfranca@icb.ufmg.br
Received: 28 November 2001.
First decision: 12 December 2001.
Accepted: 6 February 2002.
Q 2002 by the Society for the Study of Reproduction, Inc.
ISSN: 0006-3363. http://www.biolreprod.org
247
the main cause of sterility in mules is probably related to
the homologous chromosome pairing failure that occurs
during meiosis [2, 3].
Because Sertoli cells are the first cell type known to
Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024
differentiate in the gonad, they are believed to act as the
organizing center of the male gonad [4]. Also, there is
strong genetic evidence that the pre-Sertoli cell expresses
the testis-determining factor (Sry) [4], which is located in
the short arm of the Y chromosome [5, 6]. This aspect
makes the mule and donkey an interesting model on which
to perform comparative studies related to testis structure
and function in both species, mainly when it is considered
that the Sry shows great interspecific variability [7].
Spermatogenesis is a cyclic and highly coordinated pro-
cess in which diploid spermatogonia differentiate into ma-
ture haploid spermatozoa. This highly organized process
encompasses different cell associations called stages. The
sequence of events that occurs from the disappearance of a
given cellular association to its reappearance constitutes the
cycle of seminiferous epithelium, while the interval re-
quired for one complete series of cellular associations to
appear at one point within the tubule is called duration of
the cycle of the seminiferous epithelium [8].
Spermatogenesis is one of the most productive self-re-
newing systems in the body, lasting from 30 to 75 days in
mammals [9]. Although it is not established yet which
genes regulate the duration of spermatogenesis, recent work
has demonstrated that the spermatogenic cycle length is un-
der the control of germ cell genotype [10].
The general organization of spermatogenesis is essen-
tially the same in all mammals [11]. However, there are
some specific characteristics concerning the types and the
number of spermatogonial generations and the morpholog-
ical characteristics of germ cells present at the various stag-
es of spermatogenesis among different species [9, 12]. The
major criteria for stage identification resides in the mor-
phological characteristics of spermatids, in particular, their
nucleus and acrosomic system [8, 9, 13]. With this method,
the number of stages and the features used for the classi-
fication scheme will vary between species and even among
different investigators concerning the same species [14].
Another method, the tubular morphology system, is based
on the shape and location of spermatid nuclei, presence of
meiotic divisions, and overall seminiferous epithelium com-
position. This method yields 8 stages of the cycle for all
species, is less arbitrary, and is a simpler method to char-
acterize the stages of the cycle [15].
Although the basic structure of the testis is highly con-
served among vertebrates [4], specific characteristics of the
testis structure might be found for a particular species. In
this regard, quantitative data can be used to answer impor-
tant questions about the testis function and to provide a
more complete understanding of spermatogenesis [9, 16].
Many quantitative investigations of spermatogenesis re-
quire identification of the stages of seminiferous epithelium
248 NEVES ET AL.
cycle and the knowledge of the cycle length [15]. There
are few reports in the literature concerning the spermato-
genic process and testis morphometry in donkeys [17, 18]
and testis structure in mules [17]. In these studies, the sem-
iniferous epithelium cycle length and quantitative investi-
gation of Sertoli and Leydig cells were not performed. The
present work shows a more accurate morphometric analysis
of the testis and the spermatogenic cycle length in donkeys
and mules. A possible role of donkey Y chromosome in
determining important structural and functional aspects of
mule testis is also investigated.
MATERIALS AND METHODS
Animals
Five sexually mature donkeys (Equus asinus; 4–11 yr of age) and 9
young adult mules (Equus mulus mulus; 2.5–3.5 yr of age) were utilized
in the present work. The investigated donkeys belonged to the Pêga breed.
They were donated by the Brazilian Pêga Donkey Association and were
being utilized as breeders. Although reproduction in this species is not
seasonal [19, 20], the animals were orchiectomized during the equine-
breeding period, which is from September to February in the southern
hemisphere. All mules investigated were offspring of Pêga donkeys, and
six of nine were fathered by the same jack. Before surgery, all animals
received i.v. injections of 0.8 ml of Sedivet (Boehringer De Angell, U.K.)
per 100 kg of body weight. All surgical procedures were performed by a
veterinarian and followed approved guidelines for the ethical treatment of
animals.
Thymidine Injections and Tissue Preparation
To estimate the duration of seminiferous epithelium cycle, animals re-
ceived intratesticular injections of tritiated thymidine (thymidine [methyl-
3H], specific activity 82.0 Ci/mmol; Amersham Life Science, Little Chal-
font, Buckinghamshire, U.K.). The injection of 150 mCi of tritiated thy-
midine was performed in three different regions (50 mCi in 0.5 ml) of
each testis (near the epididymal head, body, and tail) using a hypodermic
needle. For each animal, the right or left testis was excised at different
time intervals after injections so that each time interval had two different
animals. These intervals were 1 h and 7 and 14 days in donkeys and 1 h
and 10 days in mules. The latter interval chosen for mules was based on
the preliminary estimation of the cycle length found for donkeys, which
took into account labeling in preleptoene/leptotene spermatocytes. After the
testes and epididymis were removed, the blood was cleared by introducing
0.9% saline with heparin (125 IU/L) via a needle through the testicular artery.
Subsequently, they were perfused-fixed by gravity-fed perfusion with 4%
buffered glutaraldehyde for 25–30 min. After perfusion, testes were
trimmed out from the epididymis and weighed and cut longitudinally by
hand with a sharp knife. Tissue samples, measuring 1- to 3-mm thickness,
were taken near the tunica albuginea and close to the locations of the
thymidine injections. Testis fragments were routinely processed and em-
bedded in plastic (glycol methacrylate). Sections of 4-mm thickness were
obtained for light microscopic investigations and were subsequently placed
on glass slides and stained with toluidine blue.
To perform autoradiographic analysis, unstained testis sections (4 mm)
were dipped on the autoradiography emulsion (Kodak NTB-2; Eastman
Kodak Company, Rochester, NY) at 458C. After drying for approximately
1 h at 258C, sections were placed in sealed black boxes and stored in a
refrigerator at 48C for 4–6 wk. Subsequently, testis sections were devel-
oped in Kodak D-19 solution at 158C according to Bundy [21] and stained
with toluidine blue. Analyses of these sections were performed by light
microscopy to detect the most advanced germ cell type labeled at different
periods postthymidine injection. Cells were considered labeled when four
to five or more grains were present over the nucleus in the presence of
low-to-moderate background.
Stages of the Seminiferous Epithelium Cycle in Donkeys
and Histological Evaluation of the Seminiferous Tubules
in Mules
Stages of the cycle in donkeys were characterized based on the shape
and location of spermatid nuclei, presence of meiotic divisions, and overall
seminiferous epithelium composition [15, 22, 23]. This method provides
8 stages of the seminiferous epithelium cycle. The relative stage frequen-
cies were determined from the analysis of 400 seminiferous tubule cross-
sections per animal (n 5 5) at 4003 magnification. Both testes were an-
alyzed for each animal. The histological sections utilized were those that
presented better quality and more tubular cross-sections.
Although spermatogenesis in mules rarely advances beyond the mei-
otic phase, the same criteria for stage classification was utilized for this
species. However, the stages of the cycle in this species were characterized
mainly based on the association of different types of primary spermato-
cytes and spermatogonia [23, 24]. Seminiferous tubules in mules were also
classified according to the presence of lumen and apoptosis and the pres-
ence of meiotic figures and/or secondary spermatocytes in the seminifer-
ous epithelium. Two hundred to 400 seminiferous tubule cross-sections
were evaluated per animal (n 5 9) at 4003 magnification. Both testes
were analyzed for each animal. Because the presence of germ cells in
mule seminiferous tubules was highly variable, stage frequencies were not
estimated for this species.
Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024
Length of the Seminiferous Epithelium Cycle
The duration of the spermatogenic cycle in donkeys was estimated
based on the stage frequencies and the most advanced germ cell type
labeled at different periods postthymidine injections. In mules, the duration
of the spermatogenic cycle took into account the most advanced germ cell
labeled in the interval between the two different periods of injections uti-
lized, according to the criteria utilized by França et al. [10].
Morphometry of the Testis in Donkeys and Mules
The tubular diameter was measured at 4003 magnification using an
ocular micrometer calibrated with a stage micrometer. At least 30 tubular
profiles that were round or nearly round, chosen randomly, were measured
for each animal. The volume densities of various testicular tissue com-
ponents were determined by light microscopy using a 441-intersection grid
placed in the ocular of the light microscope. Fifteen fields chosen random-
ly (6 615 points) were scored for each animal at 4003 magnification. Ar-
tifacts were rarely seen and were not considered in the total number of
points utilized to obtain volume densities. Points were classified as one of
the following: seminiferous tubule, tubular lumen, Leydig cell, blood and
lymphatic vessels, and connective tissue. The volume of each component
of the testis was determined as the product of the volume density and
testis volume. For subsequent morphometric calculations, the specific
gravity of testis tissue was considered to be 1.0. To obtain a more precise
measure of testis volume, 10.8% and 20.4% of the testis capsule plus
mediastinum in donkeys and mules [25], respectively, was excluded from
the testis weight. The total length of seminiferous tubule (meters) was
obtained by dividing seminiferous tubule volume by the squared radius of
the tubule times the p value [26].
Pachytene Spermatocyte Nuclear Volume
Primary spermatocyte nuclear volume grows markedly during meiotic
prophase [27]. The magnitude of this growth is variable among different
species [16]. To investigate comparatively the evolution of nuclear volume
in pachytene during spermatogenesis in both donkeys (n 5 3) and mules
(n 5 5), 30 nuclear diameters from early (stage 5), middle (stage 8), and
late (stage 2) pachytene primary spermatocytes were measured. These
pachytenes were associated with type A spermatogonia (early), type A and
B spermatogonia and preleptotene spermatocytes (middle), and type A
spermatogonia and zygotene (late). Pachytene nuclear volume was ex-
pressed in mm3 and was obtained by the formula (4/3)pR3, where R 5
nuclear diameter/2.
Cell Counts and Cell Numbers
Sertoli cells nucleoli were counted in 20 round or nearly round semi-
niferous tubule cross-sections, chosen at random, for each animal. These
counts were corrected for section thickness and nucleolus diameter ac-
cording to Abercrombie [28], as modified by Amann [29]. Twenty Sertoli
cell nucleoli diameter were measured for each animal. The total number
of Sertoli cells per testis in donkeys and mules was determined from the
corrected counts of Sertoli cell nucleoli per tubule cross-section and the
total length of seminiferous tubules, according to Hochereau-de Reviers
and Lincoln [30].
Leydig cell individual volume was obtained from nucleus volume and
the proportion between nucleus and cytoplasm. Because the Leydig cell
nucleus in donkeys and mules is spherical, its nucleus volume was ob-
tained from the knowledge of the mean nuclear diameter. For this purpose,
 TESTIS FUNCTION IN DONKEYS AND MULES 249

TABLE 1. Biometric and morphometric data in donkeys and mules (mean 6 SEM).

Parameter Donkey (n 5 5) Mule (n 5 9)

Age (months) 95 6 15 33 6 2
Body weight (kg) 262 6 20a 264 6 8b
Testis weight (g) 196 6 9 44 6 5c
Seminiferous tubule volume density (%) 84.2 6 1.1 60.9 6 2.6c
Seminiferous tubule diameter (mm) 222 6 6 127 6 4c
Total length of seminiferous tubule per testis (meters) 3806 6 209 1704 6 186c
Leydig cell
Volume density (%) 3.1 6 0.4 5.0 6 1.0
Volume in the testis parenchyma (ml) 5.4 6 0.9 1.9 6 0.5c
Nuclear diameter (mm) 7.2 6 0.2 7.4 6 0.2
Individual volume (mm3) 1351 6 125 1547 6 193
Cytoplasm volume (mm3) 1155 6 108 1327 6 177
Nucleus volume (mm3) 196 6 19 220 6 18
Number per testis (3109) 4.2 6 0.9 1.2 6 0.2c

Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024

Sertoli cell nucleolar diameter (mm) 2.6 6 0.06 2.1 6 0.04c
Sertoli cell number per testis (3109) 5.0 6 0.4 4.9 6 0.5
a n 5 4 animals.
b n 5 8 animals.
c Statistically significant (P , 0.05).

30 nuclei showing evident nucleolus had their diameters measured for each
animal. Leydig cell nuclear volume was expressed in mm3 and was ob-
tained by the same formula mentioned for pachytene. To calculate the
proportion between nucleus and cytoplasm, a 441-point square lattice was
placed over the sectioned material at 4003 magnification. One thousand
points over Leydig cells were counted for each animal. The number of
Leydig cells per testis in donkeys and mules was estimated from the Ley-
dig cell individual volume and the volume occupied by Leydig cells in
the testis parenchyma.
Statistical Analysis
All data are presented as the mean 6 SEM. Student t-test and analysis
of correlation and variance (Newman-Keuls test) were done using the pro-
gram STATISTICA for windows (StatSoft, Inc., Tulsa, OK). All values
for volume densities were subjected to arcsine transformation prior to anal-
ysis. The significance level considered was P , 0.05.
RESULTS
Biometric Data and Testis Morphometry
In spite of the age difference observed between donkeys
and mules utilized in the present work (;8 and 2.8 yr,
respectively), animals from both species investigated were
adult. As shown in Table 1, their mean body weights were
very similar (P . 0.05). However, testis size in donkeys
was almost 5-fold higher (P , 0.05) than in mules. Also,
the values found for seminiferous tubule volume density,
tubular diameter, and total length of seminiferous tubule
were significantly higher in donkeys (P , 0.05; Table 1).
Histological Evaluation of the Seminiferous Tubules
in Mules
The analysis of at least 200 seminiferous tubule cross-
sections per animal showed that approximately 70% of the
tubules presented patent lumens (Fig. 1, a and c). In ap-
proximately 25%, only vacuoles of different sizes and
shapes were observed in the epithelium (Fig. 1b). Tubules
containing no lumen comprised 5% of the tubular profiles
investigated (Fig. 1a). Basically, the same trend was ob-
served when the germ cell type present in the epithelium
was analyzed. In this regard, 75% of the tubules presented
spermatogonia and spermatocytes (Fig. 1d), 20% showed
only spermatogonia (Fig. 1c), while 5% were Sertoli cell-
only tubules (Fig. 1b). Also, meiotic figures and/or second-
ary spermatocytes were observed in 5% of the tubules re-
corded (Fig. 1e). Early spermatids were rarely seen and
were present in only 1 animal investigated. Because all
seminiferous tubules in donkeys presented normal sper-
matogenesis, the abovementioned analysis performed for
mules was not done for donkeys.
Approximately 45% of the tubules investigated showed
apoptotic cells in the epithelium. In almost 95% of them,
these cells were located in the middle and apical regions
of the epithelium. Presumably, they were germ cells (sper-
matocytes).
Stages of the Seminiferous Epithelium Cycle in Donkeys
and Mules
The 8 stages of the cycle in donkeys, characterized ac-
cording to the tubular morphology system, were very sim-
ilar to those already described for horses [23]. For this rea-
son, they will be only briefly described in the caption of
Figure 2.
Although spermatogenesis in mules, investigated in the
present study, did not advance beyond the spermatocyte
phase, it was possible to characterize 8 different germ cell
associations or stages in this species. Except for the pres-
ence of spermatids, these cell associations were basically
the same as described in donkeys.
The data obtained for nuclear volume of pachytene pri-
mary spermatocytes measured at early, middle, and late
phases of development were, respectively, as follows: for
donkeys, 267 6 1, 453 6 2, 511 6 4 mm3; and for mules,
272 6 10, 398 6 21, 483 6 14 mm3. In both species,
significant growth (P , 0.05) of pachytene nucleus volume
occurred from early to middle phase and from middle to
late phase. However, comparing both species, the size of
the pachytene nucleus was similar (P . 0.05) at the dif-
ferent phases analyzed.
Relative Stage Frequencies in Donkeys
The mean percentage of each stage of the seminiferous
epithelium cycle in donkeys was as follows: stage 1, 18.1
6 1.2; stage 2, 9.1 6 0.3; stage 3, 5.8 6 1.0; stage 4, 19.3
6 0.5; stage 5, 8.9 6 0.5; stage 6, 22.6 6 1.4; stage 7, 6.3
6 0.6; and stage 8, 9.9 6 0.5. As can be noted, stage 6
was the most frequent, while stage 3 presented the lowest
frequency. The frequencies of premeiotic (stage 1 to stage
250 NEVES ET AL.
FIG. 1. Seminiferous tubules/cords in
mules. a) Seminiferous tubules (ST) depict-
ing evident lumen (asterisk), a group of
seminiferous tubules delimited by arrow-
heads, and seminiferous cords (SC) with
no lumen. b) Seminiferous cords (SC) con-
taining only vacuoles (asterisk) and Sertoli
cells (S) are observed. c) Seminiferous tu-
bules with patent lumen and containing
only Sertoli cells (S) and spermatogonia
(G). d) In other cross-sections of seminifer-
ous tubules (ST), primary spermatocytes
(C), apoptotic cells (Ap), type A spermato-
gonia (G), and Sertoli cells (S) are ob-
served. e) A seminiferous tubule (ST) cross-
section shows meiotic figures (M) and sec-
ondary spermatocytes (II). Bar 5 50 mm
(a); bar 5 15 mm (b–e).
3), meiotic (stage 4), and postmeiotic (stage 5 to stage 8)
stages were 33%, 19.3%, and 47.7%, respectively. As men-
tioned in Materials and Methods, stage frequencies were
not obtained for mules.
Seminiferous Epithelium Cycle Length in Donkeys
and Mules
The most advanced labeled germ cell types observed at
different time periods investigated after thymidine injec-
tions in donkeys are shown in Table 2 and Figure 3. One
hour after injection, the most advanced labeled germ cells
were identified as preleptotene spermatocytes or cells in
transition from preleptotene to leptotene. These cells were
present at stage 1 and were located in the basal compart-
ment (Fig. 3a). All 8 stages of the seminiferous epithelium
cycle showed labeled germ cells that had the typical mor-
phological characteristics of the different spermatogonial
cell types present in these stages.
The most advanced germ cell type labeled 7 days after
thymidine injection was pachytene spermatocyte at stage 7
(Fig. 3b), while labeled meiotic figures and secondary sper-
matocytes present at stage 4 were observed 14 days after
injection (Fig. 3c).
Based on the most advanced labeled germ cell type ob-
served at each time period investigated postthymidine in-
jection in donkeys (Figs. 3 and 4) and the stage frequencies,
the mean duration of the seminiferous epithelium cycle was
estimated to be 10.5 6 0.4 days (Table 2). Durations of the
various stages of the cycle were determined taking into ac-
count the cycle length and the percentage of occurrence of
each stage (Fig. 4). The shortest stage was stage 3 (0.61
days), while the longest stage was stage 6 (2.37 days). Con-
sidering that approximately 4.5 cycles are necessary for the
Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024
spermatogenic process to be completed, the total length of
spermatogenesis was estimated as being 47.2 days. The ap-
proximate duration of the spermatogonial phase was 15.7
days, while the life spans of primary spermatocytes and
spermatids in donkeys were 14 and 15.5 days, respectively.
The meiotic divisions, corresponding to the duration of
stage 4, lasted 2.02 days.
In mules, the most advanced germ cell types labeled at
1 h and approximately 10 days after injections were pre-
leptotene/leptotene spermatocytes associated with pachy-
tene spermatocytes and pachytene spermatocytes associated
with preleptotene/leptotene, respectively (Figs. 3, d–e, and
5). This cell association was characteristic of stage 1 de-
scribed for donkeys. The mean size of pachytene sper-
matocyte nuclei present in this cell association at 1 h (440
6 20 mm3) and 10 days (455 6 29 mm3) after thymidine
injection was very similar (P . 0.05). Based on the label-
ing pattern found, the duration of the spermatogenic cycle
in mules was estimated as being 10.1 days, which is the
same value found for donkeys when the cycle length was
estimated considering the initial labeling point in prelep-
totene/leptotene (Table 2).
Leydig and Sertoli Cells
Leydig cell volume density and all parameters related to
Leydig cell size were similar in donkeys and mules (P .
0.05). However, the volume occupied by Leydig cells in
the testis parenchyma and the number of Leydig cells per
testis were, respectively, about 3-fold and 3.5-fold higher
in donkeys (P , 0.05; Table 1). Leydig cell nuclear volume
showed significant correlation with tubular diameter in don-
keys (r 5 0.96) and mules (r 5 0.78).
Although the nucleolar diameter was approximately 25%
 TESTIS FUNCTION IN DONKEYS AND MULES 251

FIG. 2. Stages 1–8 of the seminiferous

epithelium cycle based on the tubular
morphology system. a) Stage 1 shows
pachytene primary spermatocytes (P), pre-
leptotene spermatocytes (Pl), round sper-
matids (R), and Sertoli cell (S). b) Stage 2
presents type A spermatogonia (A), lepto-
tene spermatocytes (L), pachytene sper-
matocytes (P), elongating spermatids (E),
and Sertoli cells (S). c) Stage 3 contains
type A spermatogonia (A), zygotene sper-
matocytes (Z), diplotene spermatocytes
(D), elongate spermatids (E), and Sertoli
cell (S). d) Stage 4 shows type A spermato-
gonia (A), zygotene spermatocytes (Z),
meiotic figure (M), secondary spermato-
cytes (II), newly formed round spermatids
(R), elongate spermatids (E), and Sertoli

Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024

cells. e) Stage 5 contains type A spermato-
gonia (A), pachytene spermatocytes (P),
round spermatids (R), elongate spermatids
(E), and Sertoli cells (S). f) Stage 6 presents
type B1 spermatogonia (B1), pachytene
spermatocytes (P), round spermatids (R),
elongate spermatids (E), and Sertoli cell
(S). g) Stage 7 shows type A spermatogo-
nia (A), type B2 spermatogonia (B2), pachy-
tene spermatocytes (P), round spermatids
(R), elongate spermatids (E), Sertoli cells
(S), and residual bodies (Rb). h) Stage 8
shows type A spermatogonia (A), pachy-
tene spermatocytes (P), round spermatids
(R), elongate spermatids (E), Sertoli cells
(S), and residual bodies (Rb). Bars 5 15
mm.

TABLE 2. Length (days) of seminiferous epithelium cycle in donkeys (mean 6 SEM).

Cycle length based Cycle length based

Most advanced germ Number of cycles on labeling in on intermediate
Time after injection cell type labeleda Stage of the cycle traversed preleptotene/leptotene labeling point
65 min Pl/L 1 — — —
7.01 daysb P 7 0.720 9.74 —
14.01 daysb M/II 4 1.342 10.44 11.25
Mean duration of the cycle (days)c 10.48 6 0.44 days
a Pl/L, Preleptotene/leptotene primary spermatocytes; P, pachytene primary spermatocytes; M, meiotic figures from first and second meiotic division; II,
secondary spermatocytes.
b Total time after thymidine injection minus 65 min.
c Based on Pl/L and intermediate labeling.
252 NEVES ET AL.

FIG. 3. Most advanced labeled germ cell

type found after different time periods fol-
lowing intratesticular injections of tritiated
thymidine. a) In donkeys, 1 h after injec-
tion, preleptotene/leptotene spermatocytes
(arrows) at stage 1. b) Seven days after in-
jection, pachytene spermatocytes (arrows)
at stage 7. c) Fourteen days after injection,
secondary spermatocytes (arrows) at stage
4. d) In mules, 1 h after injection, prelep-
totene/leptotene spermatocytes (arrows) as-
sociated with pachytene spermatocytes (P)
at stage 1, and e) approximately 10 days
after injection, pachytene spermatocytes
(arrows) associated with preleptotene (Pl)
at stage 1. Bars 5 10 mm.

Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024

bigger (P , 0.05) in donkeys (Table 1), the total number
of Sertoli cells per testis in donkeys and mules was virtually
the same (P . 0.05).
DISCUSSION
This is the first study to perform a more accurate and
comparative morphometric and functional investigation of
the testis in donkeys and mules. The results suggest that
common factors, probably originated from donkeys, are in-
volved in the determination of testis structure and regula-
tion of the number of Sertoli cells per testis and the sper-
matogenic cycle length in both species, while signals re-
sulting from the interaction between Sertoli cells and germ
cells could be responsible for proliferation of adult-type
Leydig cells.
Due to the smaller testis size, lower seminiferous tubule
volume density, and fewer germ cells in mules, the total
length of seminiferous tubules in this species was signifi-
cantly smaller than in donkeys. The percentage of seminif-
erous tubules containing germ cells (spermatogonia and
spermatocytes) in mules investigated in the present study
was almost 95%. This figure was 2–3 times higher than the
results found in the literature [3, 17]. However, while we

FIG. 4. Diagram showing the most ad-

vanced germ cell type labeled at the eight
stages of seminiferous epithelium cycle in
donkeys at different time periods (1 h and
7 and 14 days) following tritiated thymi-
dine injections. The Roman numerals indi-
cate the spermatogenic cycle. The space
given to each stage is proportional to its
frequency and duration. A, Type A sper-
matogonia; B, type B spermatogonia; Pl,
preleptotene spermatocytes; L, leptotene;
Z, zygotene; P, pachytene; D, diplotene; II,
secondary spermatocytes; R, round sper-
matids; E, elongate spermatids.
 TESTIS FUNCTION IN DONKEYS AND MULES
analyzed 9 mules, Chandley et al. [3] and Hernández-Jáur-
egui and Monter [17] investigated only 1 and 3 animals,
respectively. Apart from the sample size, this difference
found was also probably related to the good quality of tis-
sue fixation and embedding, which allowed us to discrim-
inate precisely spermatogonia from Sertoli cells, and the
extensive morphometric evaluation (analysis of thousands
of tubule cross-sections) performed in the present work.
Although germ cell apoptosis is a common finding in
hybrids, these animals are not necessarily sterile or infertile
[31–33]. According to the literature, hybrid sterility might
have several causes [32]. The most obvious and observed
in mules is autosomal pairing failure [3] that occurs due to
differences in parental chromosome numbers [1]. Genic in-
compatibility between parents [32] and sexual chromosome
pairing failure might also cause sterility [34]. In agreement
with other studies [3, 17], we rarely observed germ cells
more advanced than late pachytene spermatocytes in mule
seminiferous tubules. Studies developed in our laboratory
[25] and by Hernández-Jáuregui and Monter [17] showed
that Leydig cell ultrastructure was apparently normal in
mules. This suggests that Leydig cell steroidogenic capacity
is probably not affected in this species. Also, Sertoli cell
ultrastructure as well as Sertoli cell barrier integrity inves-
tigated by transmission electron microscopy and lanthanum
were apparently normal in mules [25]. Besides that, epidid-
ymal duct epithelium in mules showed no morphological
abnormalities [17, 35]. All these findings suggest strongly
that sterility in mules is mainly due to chromosome pairing
failure [3]. The fact that Sertoli cell nucleolar diameter is
decreased in mules, compared with donkeys, is probably
related to the functional status of this cell due to the pres-
ence of fewer germ cells.
The total number of Sertoli cells per testis was very sim-
ilar in donkeys and mules and in horses investigated during
the breeding season [36, 37]. According to the literature,
the Sertoli cell is the major controller of testis development
and efficiency of spermatogenesis [4, 38–40]. The total
number of Sertoli cells present in the testis is determined
during the prepubertal period [39, 41], being regulated
mainly by FSH and thyroid hormones [39, 42] and not be-
253
FIG. 5. Diagram showing the most ad-
vanced germ cell type labeled at the eight
stages of seminiferous epithelium cycle in
mules at the two time periods (1 h and 10
days) following tritiated thymidine injec-
tions. The Roman numerals indicate the
spermatogenic cycle. A, Type A spermato-
gonia; B, type B spermatogonia; Pl, pre-
leptotene spermatocytes; L, leptotene; Z,
zygotene; P, pachytene; D, diplotene; II,
secondary spermatocytes; R, round sper-
matids; E, elongate spermatids. Apoptosis
in the seminiferous epithelium: moderate
(light gray), intense (dark gray). Absence of
the expected germ cell (stripes).
Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024
ing dependent on the presence of germ cells [43]. To our
knowledge, no data exist in the literature showing what
chromosomes or genes might be intrinsically involved in
the regulation of Sertoli cell proliferation. However, as it is
known that Sry present in the Y chromosome induces an
increase in proliferation of the coelomic epithelial cells that
give rise to Sertoli cell precursors [4], it could be speculated
that the mechanisms involved in Sertoli cell proliferation
in mules and donkeys were similar.
The total number of Leydig cells found for donkeys in
the present work was similar to adult horses studied during
the breeding season and with comparable testis size [26,
44]. Leydig cell proliferation occurs continuously from
birth to adulthood in pigs [41] and horses [26], being sig-
nificantly correlated with testis weight, tubular diameter, to-
tal length of seminiferous tubules, and germ cell number
per tubule cross-section [41]. Compared with donkeys, the
number of Leydig cells in mules investigated in the present
study was approximately 70% lower. This difference could
be related to the mean age of the mules, which were youn-
ger than the donkeys, and the lower number of germ cells
present in mules [41]. The significant and positive corre-
lation between Leydig cell nuclear volume and tubular di-
ameter found for donkeys and mules suggests the occur-
rence of cross-talk between the seminiferous tubules and
Leydig cells [39, 45, 46].
The cell composition of the seminiferous tubule epithe-
lium and the germ cells’ morphological characteristics at
the eight stages of the seminiferous epithelium cycle in
donkeys and mules were basically the same as character-
ized for horses [47]. However, the individual relative fre-
quencies of several stages of the seminiferous epithelium
cycle in donkeys studied in the present work were consid-
erably different from the frequencies found for the African
donkey (Equus asinus africanus) [18] and horses [47]. In
contrast, when the stages were grouped in premeiotic and
postmeiotic phases of spermatogenesis, similar percentages
were observed for donkeys and horses [18, 47]. These find-
ings support the hypothesis made in recent publications [16,
48, 49], which suggests that this aspect of spermatogenesis
254 NEVES ET AL.
might be phylogenetically determined among members of
the same mammalian family.
Although strain or breed differences can be found in the
literature among members of the same species [9, 50, 51],
the length of the spermatogenic cycle has been generally
considered to be constant for a given species [50]. Recent
studies, using spermatogonial transplants from rats to mice,
demonstrated that the spermatogenic cycle length is under
the control of the germ cell genotype [10]. To our knowl-
edge, there are no data in the literature regarding the sper-
matogenic cycle length in sterile hybrids. The spermato-
genic cycle length found for donkeys and mules in the pre-
sent work was very similar. However, the cycle length val-
ues obtained in these species were considerably lower than
that found in stallions (12.2 days) [47]. Given that the du-
ration of spermatogenesis is not necessarily the same for
species closely related [9, 16, 48, 49], it is considered that
the cycle length is not phylogenetically determined in mam-
mals. Perhaps hybrids or genetically modified animals
could be used as a model to more thoroughly study this
important and fascinating aspect of spermatogenesis.
In conclusion, the results found for mules suggest that
the mechanisms involved in the determination of testis
structure and function probably originated from donkeys.
Also, the data found for mules suggest that their seminif-
erous tubules are able to sustain complete spermatogenesis,
making this species a potential model for transplants of
germ cells originated from donkeys, horses, or other large
animals.
ACKNOWLEDGMENT
Technical help from Rubens Miranda is highly appreciated.
REFERENCES
1. Benirschke K, Brownhill E, Beath MM. Somatic chromosomes of the
horse, the donkey and their hybrids, the mule and the hinny. J Reprod
Fertil 1962; 4:319–326.
2. Chandley AC, Jones RC, Dott HM, Allen WR, Short RV. Meiosis in
interspecific equine hybrids. I. The male mule (Equuss asinus 3 E.
caballus) and hinny (E. caballus 3 E. asinus). Cytogenet Cell Genet
1974; 13:330–341.
3. Chandley AC, Short RV, Allen WR. Cytogenetic studies of three
equine hybrids. J Reprod Fertil Suppl 1975; 23:365–370.
4. Capel B. The battle of the sexes. Mech Dev 2000; 92:89–103.
5. Palmer MS, Sinclair AH, Berta P, Ellis NA, Goodfellow PN, Abbas
NE, Fellous M. Genetic evidence that ZFY is not the testis-determin-
ing factor. Nature 1989; 342:937–939.
6. Pelliniemi LJ, Frojdman K, Paranko J. Embryological and prenatal
development and function of Sertoli cells. In: Russell LD, Griswold
MD (eds.), The Sertoli Cell. Clearwater, FL: Cache River Press; 1993:
88–113.
7. Short RV. The testis: the witness of the mating system, the site of
mutation and the engine of desire. Acta Paediatr Suppl 1997; 422:3–7.
8. Leblond CP, Clermont Y. Definitions of the stages of the cycle of the
seminiferous epithelium in the rat. Ann N Y Acad Sci 1952; 55:548–
573.
9. Russell LD, Ettlin RA, Sinha-Hikim AP, Clegg ED. Histological and
Histopathological Evaluation of the Testis. Vienna, IL: Cache River
Press; 1990.
10. França LR, Ogawa T, Avarbock MR, Brinster RL, Russell LD. Germ
cell genotype controls cell cycle during spermatogenesis in the rat.
Biol Reprod 1998; 59:1371–1377.
11. Sharpe RM. Regulation of spermatogenesis. In: Knobil E, Neill JD
(eds.), The Physiology of Reproduction. New York: Raven Press;
1994: 1363–1434.
12. De Rooij DG, Russell LD. All you wanted to know about spermato-
gonia but were afraid to ask. J Androl 2000; 21:776–798.
13. Hess RA. Quantitative and qualitative characteristics of the stages and
transitions in the cycle of the rat seminiferous epithelium: light mi-
croscopic observations of perfusion-fixed and plastic-embedded testes.
Biol Reprod 1990; 43:525–542.
14. Hess RA, Schaeffer DJ, Eroschenko VP, Keen JE. Frequency of the
seminiferous epithelium in the rat. Biol Reprod 1990; 43:517–524.
15. Berndtson WE. Methods for quantifying mammalian spermatogenesis:
a review. J Anim Sci 1977; 44:818–833.
16. França LR, Russell LD. The testis of domestic animals. In: Martı́nez
F, Regadera J (eds.), Male Reproduction. A Multidisciplinary Over-
view, 1st ed. Madrid: Churchill Livingstone; 1998: 197–219.
17. Hernández-Jáuregui P, Monter HM. Fine structure of mule testes: light
and electron microscopy study. Am J Vet Res 1977; 38:443–447.
18. Nipken C, Wrobel KH. A quantitative morphological study of age-
related changes in the donkey testis in the period between puberty and
senium. Andrologia 1997; 29:149–161.
19. Gastal MO, Henry M, Beker AR, Gastal EL, Gonçalves A. Sexual
behaviour of donkey jacks: influence of ejaculatory frequency and
season. Theriogenology 1996; 46:593–603.
20. Kreuchauf A. Reproductive physiology in the jackass. Anim Res Dev
Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024
1984; 20:51–78.
21. Bundy DC. Photographic emulsions and processing. In: Stumpf WE,
Solomon HF (eds.), Autoradiography and Correlative Imaging, 1st ed.
San Diego: Academic Press; 1995: 49–57.
22. Courot M, Hochereau-de Reviers MT, Ortavant R. Spermatogenesis.
In: Johnson AD, Gomes WR, Vandemark NL (eds.), The Testis, 1st
ed. New York: Academic Press; 1970: 339–432.
23. Johnson L. Spermatogenesis. In: Cupps PT (ed.), Reproduction in
Domestic Animals, 3rd ed. New York: Academic Press, 1991:
173–219.
24. Van Haaster LH, de Rooij DG. Spermatogenesis is accelerated in the
immature Djungarian and Chinese hamster and rat. Biol Reprod 1993;
49:1229–1235.
25. Neves ES. Comparative study of the testis structure and spermatogenic
process in donkeys (Equus asinus) and mules (Equus mulus mulus)
[in Portuguese]. Belo Horizonte, Brazil: Federal University of Minas
Gerais; 2001. Dissertation.
26. Johnson L, Neaves WB. Age-related changes in the Leydig cell pop-
ulations, seminiferous tubules and sperm production in stallions. Biol
Reprod 1981; 24:703–712.
27. França LR, Ye S-J, Ying L, Sandberg M, Russell LD. Morphometry
of rat germ cells during spermatogenesis. Anat Rec 1995; 241:181–
204.
28. Abercrombie M. Estimation of nuclear populations from microtome
sections. Anat Rec 1946; 94:238–248.
29. Amann RP. Reproductive capacity of dairy bull. III. The effect of
ejaculation frequency, unilateral vasectomy and age on spermatogen-
esis. Am J Anat 1962; 110:49–67.
30. Hochereau-de Reviers MT, Lincoln GA. Seasonal variation in the his-
tology of the testis of the red deer, Cervus elephus. J Reprod Fertil
Suppl 1978; 54:209–213.
31. McCool CJ. Spermatogenesis in Bali cattle (Bos sondaicus) and hy-
brids with Bos indicus and Bos taurus. Res Vet Sci 1990; 48:288–
294.
32. Borondi PM, Rogatcheva MB, Zhelezova AI, Oda S. Chromosoma
pairing in inter-racial hybrids of the house musk sherew (Suncus mu-
rinus, Insectivora, Soricidae). Genome 1998; 41:79–90.
33. Switonski M, Stranzinger G. Studies of synaptonemal complexes in
farm mammals—a review. J Hered 1998; 88:473–480.
34. Forejt J. Hybrid sterility in the mouse. Trend Genet 1996; 12:412–
417.
35. Arrighi S, Romanello MG, Domeneghini C. Morphological exami-
nation of epididymal epithelium in the mule (E. hinnu) in comparison
with parental species (E. asinus and E. caballus). Histol Histopathol
1991; 6:325–337.
36. Johnson L, Nguyen HB. Annual cycle of the Sertoli cell population
in adult stallions. J Reprod Fertil 1986; 76:311–316.
37. Johnson L, Tatum ME. Temporal appearance of seasonal changes in
number of Sertoli cells, Leydig cells, and germ cells in stallions. Biol
Reprod 1989; 40:994–999.
38. Hess RA, Cooke PS, Bunick D, Kirby JD. Adult testicular enlarge-
ment induced by neonatal hypothyroidism is accompanied by in-
creased Sertoli cell and germ cell number. Endocrinology 1993; 132:
2607–2613.
39. Orth JM. Cell biology of testicular development in fetus and neonate.
In: Desjardins C, Ewing LL (eds.), Cell and Molecular Biology of the
Testis, 1st ed. New York: Oxford University Press; 1993: 3–42.
40. França LR, Hess RA, Cooke PS, Russell LD. Neonatal hypothyroid-
ism causes delayed Sertoli cell maturation in rats treated with pro-
 TESTIS FUNCTION IN DONKEYS AND MULES
pylthiouracil: evidence that the Sertoli cell controls testis growth. Anat
Rec 1995; 242:57–69.
41. França LR, Silva VA Jr, Chiarini-Garcia H, Garcia SK, Debeljuk L.
Cell proliferation and hormonal changes during postnatal development
of the testis in the pig. Biol Reprod 2000; 63:1629–1636.
42. Cooke PS, Hess RA, Kirby JD. A model system for increasing testis
size and sperm production: potential application to animal science. J
Anim Sci 1994; 72(suppl 3):43–54.
43. França LR, Bartke A, Borg KE, Cecim M, Fadden CT, Yagi A, Russell
LD. Sertoli cells in testes containing or lacking germ cells: a com-
parative study of paracrine effects using the W (c-kit) gene mutant
mouse model. Anat Rec 1994; 240:225–232.
44. Johnson L, Thompson DL Jr. Effect of seasonal changes in Leydig
cell number on the volume of smooth endoplasmic reticulum in Ley-
dig cells and intratesticular testosterone content in stallions. J Reprod
Fertil 1987; 81:227–232.
45. Habert R, Lejeune H, Saez JM. Origin, differentiation and regulation
of fetal and adult Leydig cells. Mol Cell Endocrinol 2001; 179:47–
74.
255
46. Russell LD, Warren J, Debeljuk L, Richardson LL, Mahar PL, Way-
mire KG, Amy SP, Ross AJ, Macgregor GR. Spermatogenesis in
bclw-deficient mice. Biol Reprod 2001; 65:318–332.
47. Swierstra EE, Gebauer MR, Pickett BW. Reproductive physiology of
the stallion. I. Spermatogenesis and testis composition. J Reprod Fertil
1974; 40:113–123.
48. França LR, Becker-Silva SC, Chiarini-Garcia H. The length of the
cycle of seminiferous epithelium in goats (Capra hircus). Tissue Cell
1999; 31:274–280.
49. Paula TAR, Chiarini-Garcia H, França LR. Seminiferous epithelium
cycle and its duration in capybaras (Hydrochoerus hydrochaeris). Tis-
sue Cell 1999; 31:327–334.
50. Clermont Y. kinetics of spermatogenesis in mammals: seminiferous
epithelium cycle and spermatogonial renew. Physiol Rev 1972; 52:
198–236.
51. França LR, Cardoso FM. Duration of spermatogenesis and sperm tran-
sit time through the epididymis in the Piau boar. Tissue Cell 1998;
Downloaded from https://academic.oup.com/biolreprod/article/67/1/247/2683877 by guest on 04 April 2024
30:573–582.