You are on page 1of 12

THE ANATOMICAL RECORD 300:265–276 (2017)

Morphology and Morphometry of the


Ovaries and Uteri of the Amazonian
Freshwater Stingrays
(Potamotrygonidae: Elasmobranchii)
MARIA ISABEL DA SILVA,* MARIA INES ^ BRAGA DE OLIVEIRA,
OSCAR TADEU FERREIRA DA COSTA, AND WALLICE PAXIUBA  DUNCAN
Department of Morphology, Federal University of Amazonas, Manaus, Brazil

ABSTRACT
The reproductive biology of South American freshwater stingrays
(family Potamotrygonidae) is still poorly studied compared to other
marine species. In the present study, we examined the gross anatomy and
histology of six species of potamotrygonids from the Amazon basin and
described the structural asymmetry of the ovaries and their relationship
between ovarian and uterine fecundities. Stereological techniques were
used to quantify the volume of ovarian and epigonal organ tissue associ-
ated with the left and right sides of the Potamotrygon wallacei, a recently
described species, locally known as the cururu ray. This species presented
ovarian asymmetry; the left epigonal organ-ovary complex was 55 times
larger than the right side. The right side was composed of, volumetrically,
7.3% ovarian tissue and 92.7% epigonal organ tissue whereas the left side
was 51.2% of ovarian tissue and 48.8% epigonal organ tissue. In all spe-
cies, six types of follicles were identified in both right and left ovaries.
Uteri were symmetrical and the fecundity ratio between the right and
left sides was 0.9:1.1, respectively. Despite the volumetric difference of
ovarian tissue between the two sides, the uterine fecundity shows that
both ovaries are functional and that ovarian fecundity alone is not an
accurate measure to determine the reproductive potential of freshwater
stingrays. Anat Rec, 300:265–276, 2017. V C 2016 Wiley Periodicals, Inc.

Key words: reproductive biology; potamotrygonid; folliculogen-


esis; epigonal organ-ovary complex; trophonemata

The modern chondrichthyans are primarily marine


and estuarine, with exception of the family Potamotrygo-
nidae, which are obligatorily freshwater batoids adapted Grant sponsor: Amazonas State Research Support Foundation
to South American aquatic systems (Carvalho et al., (FAPEAM) and National Council of Technological and Scientific
2003; Rosa et al., 2010). These stingrays evolved from Development (CNPq); Grant number: FAPEAM-PPP No. 389/
marine ancestors during Paleocene-Miocene marine 2012, FAPEAM-Universal No. 209/2012, and CNPq-Universal
No. 484374/2011-7.
incursions (Carvalho et al., 2004; Lovejoy et al., 2006;
*Correspondence to: Maria Isabel da Silva, Department of
Adnet et al., 2014).
Morphology, Federal University of Amazonas, 6200 Av. General
The family Potamotrygonidae exhibits some unique Rodrigo Octavio – Coroado I, Manaus/AM 69080-900, Brazil. Tel.:
physiological and morphological traits as a result of 155 92 3305-4231 E-mail: mariaisa21silva@gmail.com
their conspicuous evolutionary history. For example, Received 29 February 2016; Revised 15 August 2016;
these freshwater stingrays osmoregulate through elec- Accepted 23 August 2016.
trolytes similar to freshwater teleost (Wood et al., 2002; DOI 10.1002/ar.23501
Duncan et al., 2009). In addition, they have low urea Published online 14 October 2016 in Wiley Online Library
levels, salt-secreting rectal gland degeneration and (wileyonlinelibrary.com).

C 2016 WILEY PERIODICALS, INC.


V
266 DA SILVA ET AL.

Fig. 1. The collected species of freshwater stingrays from the Potamotrygonidae family: Paratrygon aier-
eba [A], Potamotrygon orbignyi [B], P. scobina [C], P. motoro [D], P. schroederi [E] and P. wallacei [F].

modified electroreceptive organs (Thorson et al., 1983). orbignyi, P. schoroederi, P. scobina and P. wallacei, recent-
However, most of their life history still corresponds to ly described species populary known as the cururu ray,
their marine ancestors: low population growth rate, late shows that both uteri store embryos and, consequently,
maturation, low fecundity and a long gestation period are functional (Charvet-Almeida et al., 2005).
(Carrier et al., 2004; Charvet-Almeida et al., 2005). The association between the epigonal organ and the
Despite the general similarities with marine relatives, ovary is a unique characteristic in elasmobranchs (Lutton
the reproductive biology of potamotrygonids is still poorly and Callard, 2008). The two tissues are inwardly con-
characterized (Charvet-Almeida et al., 2005). This family nected without delimitations, which does not allow their
has approximately 30 described species belonging to four differentiation (Serra-Pereira et al., 2011). It is common
genera: Heliotrygon, Paratrygon, Plesiotrygon e Potamo- that all elasmobranchs present asymmetric ovaries
trygon (Rosa and Lasso, 2014). Because of their diverse (Wourms et al., 1988). Consequently, the epigonal organ
shape, skin patterns and coloration these stingrays are might also be asymmetric. However, little is known about
targeted in fisheries for the international ornamental fish this two-organs structure and its tissue distribution on
trade (Rosa et al., 2010). Therefore, knowledge about the left and right sides.
their reproductive aspects becomes extremely important This study presents a detailed description of the gross
to the conservation and management of these species. anatomy and stereology-based morphometry of ovaries
Freshwater stingrays are aplacental viviparous with and uteri of six species of potamotrygonids from the
histotrophy. Structures called trophonemata work as pla- Amazon basin. The structural asymmetry of ovaries,
cental analogues responsible for the production of the ovarian and uterine fecundities are also characterized.
uterine milk that feeds the embryos (Wourms et al.,
1988). The right ovary is described as rudimentary and MATERIAL AND METHODS
non-functional, because of its smaller size and lack of ova
Area of Capture and Sampling
production (Wourms, 1977; Thorson et al., 1983). In con-
trast, for the Colombian species Potamotrygon magdale- Twelve adult female stingrays were caught using bottom
nae, both ovaries and uteri were described as functional longline during the dry season, with permission of Chico
(Teshima and Takeshita, 1992). Ovarian asymmetry in P. Mendes Institute for Biodiversity Conservation (ICMBio,
magdalenae was later reported suggesting that its right the Brazilian environmental agency, license No. 22055-4).
ovary is atrophied similar to other potamotrygonid rays The specimens were measured, weighed and identified
(Ramos-Socha and Grijalba-Bendeck, 2011). Recently, (Fig. 1) according to Rosa and Lasso (2014): Paratrygon
detailed descriptions of the morphology of the reproduc- aiereba (N 5 1), Potamotrygon orbignyi (N 5 2) and
tive system for this species confirmed that both ovaries Potamotrygon scobina (N 5 1) were collected in the Rio
are functional (Pedreros-Sierra et al., 2016). To date, it is Jutaı (38280 5100 S/668040 0800 W). Potamotrygon motoro were
not known if both ovaries of other potamotrygonids are collected in both Solim~oes River (N 5 1, 38170 600 S/
functional. Data for uterine fecundity in Plesiotrygon iwa- 608110 900 W) and Rio Negro (N 5 2, 08580 3000 S/628550 2600 W).
mae, Paratrygon aiereba, Potamotrygon motoro, P. Potamotrygon schroederi (N 5 1) and Potamotrygon
REPRODUCTIVE MORPHOLOGY OF POTAMOTRYGONID RAYS 267
wallacei “cururu ray” (N 5 4) were also collected in the Rio P < 0.05. Calculations of the error coefficient for tissue
Negro. All animals were anaesthetized and euthanized by volume variance was based on Gundersen et al. (1999):
spinal contusion in accordance with the institutional
pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
guidelines for protection of animal welfare (Ethical Com- Varnoise 1 VarSURS
mittee of Animal Experimentation, CEAA-UFAM protocol CE5 P
No. 070/2012).
Pi
Where Varnoise is the inaccuracy of the square lattice
Tissue Collection and Microscopy grid calculated by the shape of the average profile, Var-
Ovaries and uteri were dissected and photographed for SURS is the sums of variance areas
P through a systematic
anatomical analyses. The organs were fixed in cold and buff- uniform random sampling and Pi is the total number
ered glutaraldehyde 2.5% solution (pH 7.4) for 24 hours and of points counted over the projected surface area of epi-
then preserved in 70% ethanol. Ovarian and uterine fecun- gonal organ-ovary complex.
dity were estimated by counting the number of visible fol-
licles and intrauterine embryos. In some cases, the uterine RESULTS
fecundity was counted after natural abortions, a process Anatomy and Fecundity of Ovaries and Uteri
caused by stress due to fishing. Ovaries and uteri were
Freshwater stingrays presented two ovaries and two
weighed, measured and visible follicles were removed from
uteri located respectively at the superior and inferior
the ovaries to measure their diameter. Tissue samples were
area of the visceral cavity, parallel to the vertebral col-
dehydrated in alcohol and embedded in historesin (Techno- umn (Fig. 2A). Ovaries were pale-yellow, elongated, flat
vit 7100). Five mm-thick sections were cut using a Leica RM and asymmetrical structures with the left gonad bigger
2125RT microtome and stained with Toluidine Blue (0,1%), than the right (Fig. 2B). Macroscopically, the association
Alcian Blue pH 2.5 and Periodic Acid-Shiff (PAS). Finally, of the ovary and epigonal organ was not visible as both
the images were captured using an Olympus BX51 optic organs appeared as a single mass where bright-yellow
microscope coupled with a digital camera and the morpho- vitellogenic follicles develop during the reproductive
metric data was measured using the standard features with- season (Fig. 2C). Ovarian fecundity was also unequal
in ImageJ software (version 2.0) (Schindelin et al., 2015)”. (Table 1). For all species, independent of their reproduc-
tive stage, the left ovary presented a fecundity of 3-20
Stereology follicles with a diameter of 0.93 6 0.47 cm while the
For the species Potamotrygon wallacei, stereological right ovary did not present any visible follicles, except
P. motoro which had only one follicle with a diamter of
analyses were performed to estimate the volumetric tis-
2.5 cm (Fig. 2C).
sue ratio between ovary and epigonal organ. Full epigonal
The uteri were white, cylindrical and individual struc-
organ-ovary complexes were histologically processed, as
tures located above the kidneys with no asymmetry
described above. The volume was estimated using the
between the right and left sides (Fig. 2D). In adult Pota-
Cavalieri Principle (Howard and Reed, 2010). Briefly, the
motrygon wallacei, the mass and dimensions of the uteri
length of the immersed tissue in each historesin block
were: mass 5 2.26 g, length 5 2.52 6 0.15 cm and width 5
was measured and transected by ten equidistantly spaced 1.37 6 0.17 cm. internally, each organ had a muscular
sections (T), beginning at a random starting point within sphincter controlling the opening for the vaginal canal
the first interval. From each section, one five mm-thick and numerous projections called trophonemata arising
histological section was obtained resulting in ten total from the uterine wall (Fig. 2E). In pregnant specimens,
representative sections of the entire organ. The sections the embryos were found in contact with the trophone-
were stained with toluidine blue (0.1%) and photo- mata and immersed in a uterine liquid. The total uterine
graphed. Over every captured image, a square lattice grid fecundity varied from 2 to 3 embryos in the cururu
of points was superimposed by the free software IMOD ray, but reached up to 4 in larger species such as
(Kremer et al., 1996). The points over epigonal and ovary Potamotrygon motoro and Paratrygon aiereba (Table 1).
tissues were counted separately. The point calculations Additionally, preliminary data of uterine fecundity of 27
for ovaries were the sum of points from the germinal epi- Potamotrygon wallacei showed a proportion of 0.9:1.1
thelium, corpus luteum, follicles and ovarian stroma, pre- right:left embryos. Therefore, the average fecundity was
sented as relative volumes. Absolute volume calculation P 1.82 embryos per female (data not shown).
of ovary and epigonal tissue was P obtained as Vovary 5 Pi
ðovaryÞ3 ap 3T and Vepigonaltissue 5 PPiðepigonalÞ3 ap 3T, in Microscopic Description of Ovaries
which V is the respective volume, Pi is the total num-
ber of points falling over ovarian or epigonal tissue, ap is The ovarian tissue was found externally to the epigo-
the area represented by each point in the square lattice nal organ tissue, however both tissues were continuous
grid and T is the distance between sequenced sections lacking a divisible membrane. The epigonal tissue was
(Howard and Reed, 2010). distinguished by a dense aggregation of granulocytes
and lymphocytes while the ovarian tissue had a simple
cuboidal epithelium composed of the vascularized con-
Statistical Analysis
nective tissue of the stroma, which gives support to ovar-
Morphometric and stereological results were presented ian follicles during their development (Fig. 3B). The
as mean 6 standard deviation. Left and right organs were specimens were found in different reproductive stages.
compared in tests for normality and homogeneity and sta- However, both ovaries presented the following compo-
tistical significance by Mann-Whitney U test with nents: germinal epithelium, ovarian stroma, oogonia
268 DA SILVA ET AL.

TABLE 1. Ovarian and uterine fecundity (minimum-


maximum) of Paratrygon aiereba, Potamotrygon
orbignyi, P. scobina, P. motoro, P. schoroederi and P.
wallacei
Ovarian
Fecundity
Uterine
Species DiW R L Fecundity
Paratrygon aiereba 84 0 0 4
Potamotrygon orbignyi 31–41 0 3–5 2
Potamotrygon schroederi 28 0 4 np
Potamotrygon scobina 65 0 18 np
Potamotrygon motoro 39–39 0–1 0–20 4
Potamotrygon wallacei 21–23 0 0–6 2–3

DiW 5 disc width (mm); np 5 non-pregnant specimens, R 5 right;


L 5 left.

submucosa was vascularized and composed of loose con-


nective tissue supporting the mucosa layer. The muscu-
lar layer was thick and organized in two layers of
smooth muscles: longitudinal and transversal fibers.
Finally, the serosa layer coated the organ with a loose
and vascularized connective tissue.
We observed four different structures of the trophone-
mata among the specimens: (1) For non-pregnant speci-
mens of P. orbignyi, P. motoro and P. schroederi, the
cylindrical cells from the epithelium were laterally posi-
tioned, presenting nuclei at the basal pole and small
capillaries at the lamina propria (Fig. 5C). (2) For preg-
nant specimens of Paratrygon aiereba and Potamotrygon
wallacei, the epithelium was cuboidal and organized in
numerous short invaginations forming secretory crypts.
Between the crypts, loose connective tissue supported
wide capillaries located at the surface of the trophone-
mata (Fig. 5D). (3) In the non-pregnant specimen P. sco-
bina the secretory crypts were elongated and a thin core
of tissue surrounded the crypts giving support to
capillaries directed to the surface of the trophonemata
(Fig. 5E). (4) In the non-pregnant specimen of Potamo-
Fig. 2. Anatomy of the female reproductive system in freshwater trygon wallacei, the crypts were in anastomosis and the
stingrays. [A] Anatomical scheme of the reproductive tract. [B,C] Ana- epithelium was breaking down and peeling off (Fig. 5F).
tomical picture of the epigonal organ-ovary complex of Potamotrygon Trophonemata tissue did not have positive staining for
wallacei with lack of follicles and Potamotrygon motoro with follicles in
PAS or Alcian blue (data not shown).
both sides. [D] External and [E] internal anatomy of uteri of Potamotry-
gon motoro, with trophonemata (white arrows). Vertebral column (V);
Right ovary (RO); Left ovary (LO); Right oviductal gland (ROG); Left Stereological Analysis of Ovary-Epigonal
oviductal gland (LOG); Right uterus (RU); Left uterus (LU).
Complex
The stereological analysis for all specimens of the
cells and follicles in different stages of development.
Potamotrygon wallacei showed unequal volumetric
Components such as the mature follicle, atretic follicle
proportions for ovarian-epigonal organ association. In
and corpus luteum were only found during mature
a comparative analysis between the right and left ova-
stages and during pregnancy. For each ovarian compo-
ries, the two-tissue complex was asymmetrical for the
nent, the principal morphological characteristics were
total and for each individual tissue volume (Fig. 6A).
described in Table 2 and illustrated in Figures 3 and 4.
The right side had a larger proportion of epigonal tis-
sue, 92.7% (0.227 mm3/g 6 0.046) compared to ovarian
Microscopic Description of Uteri tissue, 7.3% (0.018 mm3/g 6 0.013) (Fig. 6B,C). On the
The uterine wall was composed of four concentric other hand, the left side consisted of 48.8% (0.953
layers: mucosa, submucosa, muscular and serosa (Fig. mm3/g 6 0.465) epigonal organ tissue, while 51.2%
5A,B). The mucosa was formed by villi-like extensions (0.983 mm3/g 6 0.468) was ovarian tissue. For ovarian
called trophonemata. These structures were composed of components (Fig. 6B,C), the right ovary had 90% of
an axis of loose connective tissue rich in fibroblasts, germinal epithelium (P < 0.05). In contrast, the left
smooth muscle cells, capillaries and coated by an exter- ovary presented a majority (62%) of the volume of the
nal simple layer of cylindrical epithelium. The corpus luteum.
REPRODUCTIVE MORPHOLOGY OF POTAMOTRYGONID RAYS 269

Fig. 3.
270 DA SILVA ET AL.

DISCUSSION pattern found in studies with Mustelus schmitti and


Histological Organization of the Female Sympterygia acuta (Dıaz-Andrade et al., 2009; Galındez
Reproductive Tract et al, 2014).
Records about the presence of corpora lutea in elasmo-
In all of the studied specimens, both ovaries contained branchs are dated, and there is a lack of information
the same ovarian components described in other elasmo- about the characteristics and developmental aspects of
branchs: germinal epithelium, ovarian stroma, oogonia this structure for both elasmobranchs and particularly for
cells and follicles at different stages of development, freshwater stingrays. This study describes the histological
atretic follicle and corpus luteum (Hamlett et al, 1999; characteristics of the corpora lutea during its development
Dıaz-Andrade et al., 2009; Dıaz-Andrade et al., 2011; and regression. Our results correspond to the recent
Serra-Pereira et al., 2011; Galındez et al., 2014; Wehitt description of corpora lutea in P. magdalenae by Pedreros-
et al., 2015; Pedreros-Sierra et al., 2016). Sierra et al. (2016). We observed multiple of these struc-
The germinal epithelium is organized on a single layer tures in the left ovary in different stages of development
of cuboid cells, similar to other marine and freshwater and regression, suggesting that non-simultaneous, multi-
viviparous stingrays, but lacking cilia that are present ple ovulations in this group might provide a long lasting
in marine oviparous rays (Hamlett et al., 1999; Dıaz- hormone secretion during pregnancy.
Andrade et al., 2009; Dıaz-Andrade et al., 2011; The phenomenon of atresia in rays occurs in all stages
Pedreros-Sierra et al., 2016). Oogonia of viviparous rays of follicular development and consists of the disorganiza-
are described as small cells, 20 lm in diameter, with tion, regression, degradation and cellular death of a folli-
large, evident nuclei, basophilic cytoplasm, located in cle during the absorbance of the yolk (Baccari et al.,
nests under the germinal epithelium and found only in 1992; Dıaz-Andrade et al., 2011; Galındez et al., 2014;
immature and maturing specimens (Dıaz-Andrade et al., Wehitt et al., 2015; Pedreros-Sierra et al., 2016). From
2009; Dıaz-Andrade et al., 2011; Wehitt et al., 2015). primordial to previtellogenic follicles, we observed that
These same characteristics, but with oogonia occurring the atresia was characterized by follicular epithelium
in different reproductive phases, were observed in the disorder, cellular fragmentation and cytoplasm regres-
present study. It is possible that this finding corresponds sion. In vitellogenic follicles, the amount of yolk
to a high, lifelong reproductive potential for this group decreased in the ovulum cytoplasm, vascularization
of potamotrygonids. increased inside the follicular epithelium and the folded
In Batoids, folliculogenesis is usually described as follicular epithelium tended to collapse. The vasculariza-
four follicular stages: primordial follicle, primary follicle, tion inside the follicular epithelium becomes important
previtellogenic follicle and vitellogenic follicle (Hamlett to yolk evacuation when blood infiltrates bring macro-
et al, 1999; Dıaz-Andrade et al., 2011; Wehitt et al., phages to accelerate cell degeneration and the formation
2015; Pedreros-Sierra et al., 2016). In this study, five fol- of a scar-like structure (Craik, 1978; Galındez et al.,
licular stages were identified, with two subtypes for 2014; Wehitt et al., 2015; Pedreros-Sierra et al., 2016).
vitellogenic follicles: primordial follicle, primary follicle, In contrast to what was described in the placentary
previtellogenic follicle, I and II vitellogenic follicle and viviparous shark Sphyrna blocliii (Appukuttan, 1978) and
mature follicle. During folliculogenesis, the follicles in the aplacentary viviparous shark Mustelus antarcticus
increased exponentially in diameter (Table 2). The follic- (Storrie et al., 2009), we observed that freshwater sting-
ular cells changed from a simple to a pseudostratified rays do not have uterine compartments. The uterine body
layer composed of cylindrical and globose cells (Serra- was composed of four concentric tissue layers (Hamlett
Pereira et al., 2011; Galındez et al., 2014). PAS staining et al., 1996; Storrie et al., 2009; Galındez et al., 2010; Col-
showed the presence of neutral glycoproteins both inside onello et al., 2013; Dıaz-Andrade et al., 2013; Wehitt
globose cells and the surrounding yolk vesicles. Studies et al., 2015; Pedreros-Sierra et al., 2016), in which the
suggest that these glycoproteins are vitelline precursors two orthogonal muscular layers guarantee uterine multi-
for yolk synthesis and the appearance of the theca and directional expansion (Galındez et al., 2010). The mucosa
its vascularization also influence yolk storage inside the morphology changes in accordance with the female repro-
ovulum through uptake of hormonal signals (Prisco ductive stages (Hamlett et al., 1996; Storrie et al., 2009;
et al., 2002; Prisco et al., 2007; Galındez et al., 2014; Galındez et al., 2010; Colonello et al., 2013; Dıaz-Andrade
Wehitt et al., 2015). et al., 2013; Pedreros-Sierra et al., 2016). We observed
The zona pellucida formation begins during the early major changes in the mucosa with the appearance of
stages of folliculogenesis with varied thickness during secretory cells and an increase of vascularization. Howev-
development (Dıaz-Andrade et al., 2009; Galindez et al, er, Colonello et al. (2013) described the full uterine
2014). In this study, we observed maximum zona pellu- dynamic from the pre-ovulation until post-partum stage
cida thickness in previtellogenic follicles, similar to the in the ray Myliobatis goodie. Based on their description

Fig. 3. Folliculogenesis and its cellular components. [A] Nest of oogo- vitellogenic follicle with folded follicular epithelium (black arrows). [E]
nia (Oo) under the germinal epithelium (blach arrow) stained with tolui- External (et) and internal (it) theca expanding into the follicular epitheli-
dine blue. [B] Epigonal organ tissue (EPO) surrounding the follicles and um folds (black arrow) with the zona pellucida skirting along the folds
characterization of follicular epithelium (black arrows), zona pellucida (asterisk) stained with toluidine blue. [F] Mature follicles of Potamotry-
(asterisk) and theca (t) in formation on primordial follicle (1), primary folli- gon orbignyi visible to the naked eye. [G] Follicular epithelium folds
cle (2) and previtellogenic follicle (3) stained with toluidine blue. [C] I compressed and united (black arrow) because of the great amount of
vitellogenic follicle surrounded by the theca (t) with complete zona pel- stored yolk (y) in the ovum cytoplasm stained with toluidine blue. [H]
lucida (asterisk) and follicular epithelium composed by cylindrical (white Glycoproteic granules surrounding the yolk vesicles (y) and inside the
arrow) and globose (black arrow) cells stained with toluidine blue. [D] II globose cells (black arrow) stained with Periodic Acid-Schiff.
REPRODUCTIVE MORPHOLOGY OF POTAMOTRYGONID RAYS 271
TABLE 2. Description and measurements (mean 6 standard deviation) of ovarian components
Component Characteristics Measurements
Germinal Single layer of cuboidal cells from where the oogonias arise (Fig. 3A). –
epithelium
Oogonia Rounded shape cell with a large central nucleus and homogenous cyto- D: 8.21 6 4.16 mm
plasm, found in a cluster under the germinal epithelium (Fig. 3A). ZPt: absent
Primordial Oocytes with a large and conspicuous nucleus, a homogenous cytoplasm and D: 43.34 6 10.05 mm
follicle a discontinuous simple layer of flat follicular cells. This cellular type is ZPt: absent
located between the ovarian epithelium and stroma (Fig. 3B).
Primary Oocytes with a cuboidal follicular layer where large rounded globose cells D: 88.20 6 24.24 mm
follicle appear. The formation of the zona pellucida began at this follicular stage ZPt: 5.54 6 1.29 mm
between the follicular layer and cytoplasm. This cellular type is located in
the ovarian stroma attached by concentric external layers of connective
tissue that compose the follicular theca (Fig. 3B).
Previtellogenic Oocyte with multiple follicular layers composed of globose cells and other D: 196.01 6 30.09 mm
follicle cells varying from cuboidal to cylindrical shape. The cytoplasm starts to ZPt: 7.57 6 1.25 mm
become granular and the follicular theca is evident and attached to the
ovarian stroma (Fig. 3B).
I Vitellogenic Oocyte with a pseudostratified follicular epithelium, cytoplasm presenting D: 295.55 6 67.31 mm
follicle yolk granules and tiny zona pellucida (Fig. 3C). ZPt: 2.56 6 0.52 mm
II Vitellogenic Oocyte visible to the naked eye with follicular layer projecting folds inward- D: 0.36 6 0.11 cm
follicle ly toward the follicle (Fig. 3D). These folds were filled by vascularized ZPt: present, but too
connective tissue expanding from the internal theca (Fig. 3E). The cyto- tin to measure in
plasm was filled by yolk granules organized in vesicles. light microscopy.
Mature follicle Bright yellow oocyte with oval shape and visible to the naked eye (Fig. 3F). D: 0.93 6 0.47 cm
In this follicular final stage, the cytoplasm is fully filled by yolk vesicles ZPt: present, but too
that compressed the expanded folds of the follicular layer (Fig. 3G). Yolk tin to measure in
vesicles were surrounded by the same glycoproteins found in the cyto- light microscopy.
plasm of globose cells, as evidenced PAS coloration (Fig. 3H). This follicle
is ready to be ovulated and impregnated, a principal characteristic for
mature females.
Corpus luteum Component present only during pregnancy and described in three different –
developing phases:
Phase 1: the residual follicular layer is conserved with the folds close to
each other and tending to merge tensioned by “bridge” structures (Fig.
4A-B). There are no globose cells and the follicular cells become cylindri-
cal with nuclei positioned at the apical pole giving to this whole structure
a glandular appearance (Fig. 4B).
Phase 2: The follicular folds are completely merged forming a large, oval
and vascularized corpus luteum (Fig. 4C). The nuclei in the cylindrical
cells maintain apical position, but the cytoplasm of these cells shows
white coloration (Fig. 4D).
Phase 3: Represents the devolution of the corpus luteum. This structure
was extremely vascularized, disorganized and with degenerated cells (Fig.
4E-F).
Atretic follicle Oocyte in all described follicular stages identified by irregular conformation –
of the cytoplasm. For follicles in advanced maturation such as vitellogenic
follicle, it was observed with high vascularization inward toward the folds
and presence of blood infiltration. The follicular folds were close to each
other and tended to collapse (Fig. 4G-H).

D 5 diameter; ZPt 5 zona pellucida thickness.

and the recent studies of Pedreros-Sierra et al. (2016) in with 90% of the total volume while on the left side ova-
P. magdalenae, we were able to describe four types of ry and epigonal organ volume was 50% for both tissues.
trophonemata, each corresponding, respectively, to the Interestingly, the right ovary was mostly composed of ger-
reproductive stages: pre-ovulation, post-ovulation, mid minal epithelium and some atretic follicles while the left
gestation and post-partum, respectively. ovary was composed primarily of the corpora lutea and
some atretic follicles. These results suggest that in the
Ovarian asymmetry and its implications for analyzed specimens ovulation occurred only in the left
reproductive potential of freshwater stingrays side. However, the presence of germinal epithelium and
atretic follicles in the right side indicates that folliculo-
Ovarian and epigonal organ asymmetry was observed genesis was occurring as well.
in all potamotrygonid species studied. In the cururu ray, For viviparous elasmobranchs, reproductive fecundity
the right ovary was 55 times smaller than the left ovary. is quantified by the counting of visible follicles in the
The volumetric distribution of both ovarian and epigonal ovary and the number of embryos inside the uteri
organ tissue was also unequal between right and left (Musick et al., 2000). In a study with seven species of
sides. On the right side, the epigonal organ predominated potamotrygonids, Charvet-Almeida et al. (2005) showed
272 DA SILVA ET AL.

Fig. 4. Evolution of the corpus luteum and follicular atresia stained coloration and the cells still present apical nuclei (asterisk). [E,F]
with toluidine blue. [A,B] Phase 1: after ovulation the follicular epitheli- Phase 3: the vasculature increases (black arrow) and the corpus
um folds tend to get closer by having bridge-like structures (black luteum cells are fragmented (asterisk). [G,H] Follicular atresia and yolk
arrow) and cellular nuclei are positioned on the apical pole (asterisk). absorbance; Vasculature increases inside the folds (black arrow) bring-
[C,D] Phase 2: the folds are connected, the structure has lighter ing tissue infiltration (white arrow) and cells are fragmentated.
REPRODUCTIVE MORPHOLOGY OF POTAMOTRYGONID RAYS 273

Fig. 5. Uterine histology stained with toluidine blue. [A,B] Uterine layers: mucosa (m) with trophonemata
(t), submucosa (sm), transversal (mst) and longitudinal (msl) muscular and serosa (s). [C,F] Structures of
trophonemata with characterization of epithelial cells (asterisk) and capillaries (black arrows) in different
reproductive stages.
274 DA SILVA ET AL.

Fig. 6. Stereology of ovary and epigonal organ. [A] Volumetric proportion of ovaries and epigonal
organs for the right and left sides. On the right side, the volume of ovarian tissue is significantly smaller
than the volume of epigonal tissue: P 5 0.029 (Mann-Whitney, asterisk). [B,C] Volumetric percentage of
ovary-epigonal organ association and ovarian components in the right and left side, respectively.

that the ovarian fecundity varied from 1-13 mature fol- chambers with equal uterine functionality and fecundity
licles while uterine fecundity varied from 1-16 embryos (Charvet-Almeida et al., 2005), there is no longer strong
in total. However, these data were not separately exam- evidence to classify the right ovary as non-functional.
ined from right and left sides of the ovaries or uteri. In Because its ova often develop into embryos in the right
our study, the number of mature follicles varied from 1- uterus, we determined the right ovary to be functional in
20. Follicles were present in the ovaries of specimens potamotrygonids. Similarly, in P. magdalenae, although
before, during and after pregnancy with a right:left the right ovary is smaller and produces smaller follicles
ovarian fecundity proportion of 0:6. In Potamotrygon compared with the left ovary, corpora lutea were observed
wallacei, the uterine fecundity varied from 2-4. Although in both ovaries indicating that ovulation occurs eventually
the right:left uterine fecundity is difficult to estimate in both functional gonads (Pedreros-Sierra et al., 2016).
due to spontaneous abortion after capture, we were able We believe that the size-reduction of right ovaries might
to estimate the right:left uterine fecundity as 0.9:1.1 for be due to anatomical pressures from the spiral valve posi-
this species. It was observed that the uterine fecundity tioned over the gonad, limiting space for its development.
in freshwater stingrays was smaller than ovarian fecun- This hypothesis, however, still requires investigation.
dity, as reported by Charvet-Almeida et al. (2005). Inter- Additionally, we investigated which fecundity (ovarian
estingly, the fertility of the ovaries and uteri (right and or uterine) actually represents the real reproductive
left) did not match; ovarian fecundity was unequal potential in this group. For the ovarian fecundity, the
between right and left ovaries, but uterine fecundity excessive follicular production does not seem to promote
seemed slightly equilibrated between both uteri. any trade-off to the animal but instead wastes energy.
In this study, the contradiction between ovarian and However, it is possible to hypothesize reasonable trade-
uterine fecundity plays an important role in the under- offs in this situation if: (1) the large number of ova bal-
standing of ovarian functionality. According to Hamlett ances the chances of failure on fecundation and/or fetus
and Koob (1999), Carrier et al (2004) and Lutton et al development; (2) these species have a biennial reproduc-
(2005), the ovarian fecundity in elasmobranchs varies tive cycle. Neither of these hypotheses were confirmed
according to the reproductive status; both oviparous for this group, but studies with Dasyatis centroura and
elasmobranchs ovaries are functional while viviparous Dasyatis imbricate found gestation periods equal to or
species have only one functional ovary with the other shorter than 6 months, suggesting bi-annual cycles
being rudimentary. Being viviparous, potamotrygonids (Struhsaker, 1969; Devadoss, 1978). Potamotygon
are described with a non-functional right ovary motoro, P. orbignyi and Potamotrygon wallacei also pre-
(Wourms, 1977; Thorson et al., 1983). However, studies sent short gestation periods (Charvet-Almeida et al.,
with Potamotrygon magdalenae have shown that both 2005) and these species were found with both mature
ovaries are functional, with the left producing more ova ova and embryos at the same time. Thus, it might be
(Teshima and Takeshita, 1992; Pedreros-Sierra et al., possible that the extra number of ovarian follicles during
2016). In our studied species, we observed that the right pregnancy will be directed to ovulation after the birth of
ovary has a reduced size and partial or no ovarian activ- embryos to the beginning of a new gestation. Fahy et al.
ity because of its reduced or absent follicular production. (2007) made similar observations in Urobatis jamaicen-
However, observing that the uteri are non-connected sis and argued that the maximization of ova production
REPRODUCTIVE MORPHOLOGY OF POTAMOTRYGONID RAYS 275
might be a strategy to increase brood survivorship. Stud- Dıaz-Andrade MC, Lopez-Cazorla A, Galındez EJ. 2013. Histological
ies with captive specimens of P. motoro and P. circularis remarks of the uterus of Sympterygia acuta (Garman, 1877) and
(synonymous of P. motoro, Potamotrygon constellata or P. Sympterygia bonapartii (M€ uller & Henle, 1841) (Chondrichthyes;
orbignyi) have also observed more than one reproductive Rajidae). Int J Morphol 31:864–872.
Duncan WP, Costa OTF, Ara ujo MLG, Fernandes MN. 2009. Ionic
cycle in a year (Thorson et al., 1983). However, this is a
regulation and Na1–K1-ATPase activity in gills and kidney of the
topic that requires additional data and more analysis freshwater stingray Paratrygon aiereba living in white and black-
before its confirmation. waters in the Amazon Basin. J Fish Biol 74:956–960.
In summary, the ovarian and uteri components of Fahy DP, Spieler RE, Hamlett WC. 2007. Preliminiary observations
potamotrygonid rays are similar to other elasmobranchs. on the reproductive cycle and uterine fecundity of the yellow
In all potamotrygonid species, we identified five follicu- stingray, Urobatis jamaicensis (Elasmobranchii: Myliobatiformes:
lar phases in both ovaries. In the left ovary, we observed Urolophidae) in Southesast Florida, USA. Raffles Bull Zool 14:
multiple corpora lutea in different stages. In fact, all 131–139.
Potamotrygonid rays presented ovarian asymmetry. In Galındez EJ, Dıaz Andrade MC, Moya AC, Estecondo S. 2010. Mor-
Potamotrygon wallacei, the left epigonal organ-ovary phological changes in the pregnant uterus of the smooth hound
dogfish Mustelus schmitti springer, 1939 (Gatuzo) (Condrichthyes,
complex was 55 times larger than the right side, but the Triakidae). Microscopic study and phylogenetic reproductive
right ovary was mostly composed by germinal epitheli- implications. Int J Morphol 28:1003–1010.
um. At present, our collected data support the function- Galındez E, Dıaz Andrade M, Estecondo S. 2014. Morphological
ality of the right ovary and we strongly recommend indicators of initial reproductive commitment in Mustelus
future analysis of the left/right quantification for ovarian schmitti (Springer 1939) (Chondrichthyes, Triakidae): Folliculo-
and uterine fecundity in order to obtain a better under- genesis and ovarian structure over the life cycle. Brazilian J Biol
standing of organ asymmetry and fecundity distribution. 74:154–163.
Gundersen HJ, Jensen EB, Kieu K, Nielsen J. 1999. The efficiency
of systematic sampling in stereology–reconsidered. J Microsc Mar
LITERATURE CITED 193:199–211.
Hamlett WC, Musick JA, Eulitt AM, Jarrell RL, Kelly MA. 1996.
Adnet S, Gismondi RS, Antoine PO. 2014. Comparisons of dental mor- Ultrastructure of uterine trophonemata, accommodation for utero-
phology in river stingrays (Chondrichthyes: Potamotrygonidae) lactation and gas exchange in the southern stingray, Dasyatis
with new fossil from the middle Eocene of Peruvian Amazonia americana. Can J Zool Can Zool 74:1417–1430.
rekindle debate on their evolution. Naturwissenschaften 101:33–45. Hamlett WC, Koob T. 1999. Female reproductive system. In:
Appukuttan KK. 1978. Studies on the developmental stages of ham- Sharks, skates and rays: The biology of elasmobranch fishes.
merhead shark Sphyrna (Eusphyrna) blochii from the Gulf of Hamlett WC, editor. Baltimore: The John Hopkins University
Mannar. Indian J Fish 25:41–51. Press. p 398–443.
Baccari GC, Minucci S, Dimatteo L, Chieffi G. 1992. Ultrastructural Hamlett WC, Jezior M, Spieler R. 1999. Ultrastructural analysis of
investigation of the corpora-atretica of the electric ray, Torpedo folliculogenesis in the ovary of the yellow spotted stingray, Urolo-
marmorata. Gen Comp Endocrinol 86:72–80. phus jamaicensis. Ann Anat 181:159–172.
Carrier JC, Pratt HL Jr, Castro JI. 2004. Reproductive biology of Howard CV, Reed MG. 2010. Unbiased stereology: Three-
elasmobranchs. In: Carrier JC, Musick JA, Heithaus MR, editors. dimensional measurement in microscopy. 2nd ed. New York:
Biology of sharks and their relatives. Boca Raton: CRC Press. p Springer-Verlag.
269–286. Kremer JR, Mastronarde DN, McIntosh JR. 1996. Computer visual-
Carvalho MR, Lovejoy NR, Rosa RS. 2003. Family Potamotrygoni- ization of three-dimensional image data using IMOD. J Struct
dae (river stingrays). In: Reis RE, Kullander SO, Ferraris CJ Jr, Biol 116:71–76.
editors. Check list of the freshwater fishes of South and Central Lovejoy NR, Albert JS, Crampton WGR. 2006. Miocene marine
America. Porto Alegre: Edipucrs. p 22–28. incursions and marine/freshwater transitions: Evidence from Neo-
Carvalho MR, Maisey JG, Grande L. 2004. Freshwater stingrays of tropical fishes. J South Am Earth Sci 21:5–13.
the Green River Formation of Wyoming (Early Eocene), with the Lutton BV, George JST, Murrin CR, Fileti LA, Callard IP. 2005. The
description of a new genus and species and an analysis of its phy- elasmobranch ovary. In: Reproductive biology and phylogeny of
logenetic relationships (Chondrichthyes: Myliobatiformes). Bull chondrichthyes: Sharks, batoids and chimaeras. Hamlett WC, edi-
Am Mus Nat Hist 284:1–136. tor. Enfield, NH: Science Publishers. p 237–281.
Charvet-Almeida P, Ara ujo MLG, Almeida MP. 2005. Reproductive Lutton BV, Callard IP. 2008. Morphological relationships and leuko-
aspects of freshwater stingrays (Chondrichthyes: Patamotrygoni- cyte influence on steroid production in the epigonal organ-ovary
dae) in the Brazilian Amazon basin. J Northw Atl Fish Sci 35: complex of the skate, Leucoraja erinacea. J Morphol 269:620–629.
165–171. Musick JA, Burgess G, Caillet GM, Camhi M, Fordham S. 2000.
Colonello JH, Christiansen HE, Cousseau MB, Macchi GJ. 2013. Management of sharks and their relatives (Elasmobranchii). Fish-
Uterine dynamics of the southern eagle ray Myliobatis goodei eries 25:9–13.
(Chondrichthyes: Myliobatidae) from the Southwest Atlantic Pedreros-Sierra TM, Arrieta-Prieto DM, Mejıa-Falla PA. 2016.
Ocean. Ital J Zool 80:187–194. Reproductive system of females of the Magdalena River endemic
Craik JC. 1978. Plasma levels of vitellogenin in the elasmobranch stingray Potamotrygon magdalenae: Anatomical and functional
Scyliorhinus canicula L. (lesser spotted dogfish). Comp Biochem aspects. J Morphol 277:680–697.
Physiol B 60:9–18. Prisco M, Romano M, Ricchiari L, Limatola E, Andreuccetti P. 2002.
Devadoss P. 1978. Maturation and breeding habit of Dasyatis An ultrastructural study on the vitellogenesis in the spooted ray
(Amphotistius) imbricatus (Schneider) at Porto Novo. Indian J Torpedo marmorata. Gen Comp Endo 128:171–179.
Fish 25:29–34. Prisco M, Liguoro A, Ricchiari L, Del Giudice G, andreuccetti P.
Dıaz-Andrade MC, Galındez E, Estecondo S. 2009. The ovary of the 2007. Oogenesis in the spotted ray Torpedo marmorata. Rev Fish
bignose fanskate Sympterygia acuta Garman, 1877 (Chondrich- Biol Fish 17:1–10.
thyes, Rajidae) in the Bahıa Blanca estuary, Argentina: Morpholo- Ramos-Socha HB, Grijalba-Bendeck M. 2011. Bioecologıa de la raya
gy and reproductive features. Braz J Biol 69:405–413. de agua dulce Potamotrygon magdalenae (Dumeril, 1865) (Mylio-
Dıaz-Andrade MC, Galındez EJ, Lopez Cazorla A. 2011. Ovarian batiformes) en la cienaga de Sabayo, Guaimaral, Colombia. Rev
foliculogenesis in the smallnose fanskate Sympterygia bonapartii UDCA Act Div Cient 14:109–118.
(M€uller & Henle, 1841) (Chondrichthyes, Rajidae). Int J Morphol Rosa RS, Charvet-Almeida P, Quijada CCD. 2010. Biology of the
29:174–181. South American Potamotrygonid stingrays. In: Shark and their
276 DA SILVA ET AL.

relatives II: Biodiversity, adaptive physiology and conservation. Struhsaker P. 1969. Observations on the biology and distribution of
In: Carrier JC, Musick JA, Heithaus MR, editors. New York: Tay- the thorny stingray, Dasyatis centroura (Pisces, Dasyatidae). Bull
lor & Francis Group. p 241–286. Mar Sci 19:456–481.
Teshima K, Takeshita K. 1992. Reproduction of the freshwater
Rosa RS, Lasso CA. 2014. Biogeografıa de las rayas de agua dulce
stingray, Potamotrygon magdalenae taken from the Magdalena
(Potamotrygonidae) de Am erica del Sul. In: IX. Rayas de agua
River System in Colombia, South America. Bull Seikai Nat Fish
dulce (Potamotrygonidae) de Suram erica. Parte I. Colombia, Ven-
Res Inst 70:11–27.
ezuela, Ecuador, Peru, Brasil, Guyana, Surinam y Guyana Fran-
Thorson TB, Langhammer JK, Oetinger MI. 1983. Reproduction
cesa: diversidad, bioecologia, uso y conservaci on. In: Lasso CA,
and development of the South American freshwater stingrays,
Rosa RS, S anchez-Duarte P, Morales-Betancourt MA, Agudelo- Potamotrygon circularis and P. motoro. Environ Biol Fishes 9:3–
Cordoba E, editors. Bogot a: Instituto de Investigaci on de los 24.
Recursos Biologicos Alexander von Humboldt. p 39–47. Wehitt A, Di Giacomo EE, Galındez EJ. 2015. The female reproduc-
Schindelin J, Rueden CT, Hiner MC, Eliceire KW. 2015. The ImageJ tive system of Zearaja chilensis (Guichenot, 1848) (Chondrich-
ecosystem: An open platform for biomedical image analysis. Mol thyes, Rajidae). Gametogenesis and microscopic validation of
Reprod Dev 82:518–529. maturity criteria. Int J Morphol 33:309–317.
Wood CM, Matsuo AYO, Gonzalez RJ, Wilson RW, Patrick ML, Val
Serra-Pereira B, Figueiredo I, Gordo LS. 2011. Maturation of the
AL. 2002. Mechanisms of ion transport in Potamotrygon, a steno-
gonads and reproductive tracts of the thornback ray Raja clavata,
haline freshwater elasmobranch native to the ion-poor black-
with comments on the development of a standardized reproduc-
waters of the Rio Negro. J Exp Biol 205:3039–3054.
tive terminology for oviparous elasmobranchs. Mar Coast Fish 3:
Wourms JP. 1977. Reproduction and development in chon-
160–175.
drichthyan fishes. Amer Zool 17:379–410.
Storrie MT, Walker TI, Laurenson LJ, Hamlett WC. 2009. Gesta- Wourms JP, Grove BD, Lombardi J. 1988. The maternal-embryonic
tional morphogenesis of the uterine epithelium of the gummy relationship in viviparous fishes. In: Fish physiology. Hoar WS,
shark (Mustelus antarcticus). J Morphol 270:319–336. Randall DJ, editors. San Diego: Academic Press. p 1–134.