Content uploaded by Joycelyn Cagatin Jumawan
Author content
All content in this area was uploaded by Joycelyn Cagatin Jumawan on Jun 06, 2014
Content may be subject to copyright.
1Biology Department, Caraga State University, Butuan City, Philippines
2Institute of Biology, University of the Philippines-Diliman, Diliman, Quezon City, Philippines
3Institute of Environmental Science and Meteorology, University of the Philippines-Diliman, Diliman,
Quezon City, Philippines
*Corresponding author: joycejumawan@gmail.com
Embryonic and larval development of the
suckermouth sailfin catfish Pterygoplichthys pardalis
from Marikina River, Philippines
38
EurAsian Journal of BioSciences
Eurasia J Biosci 8, 38-50 (2014)
http://dx.doi.org/10.5053/ejobios.2014.8.0.4
Siluriformes exhibit diverse reproductive
strateg
ies and most studies are just focused on late
embryonic and larval development (Adriaens and
Vandewalle 2003). Nonetheless, knowledge of the
crucial phases of the life history of invasive species is
vital in order to understand their developmental
strategies and identify the advantageous
ontogenetic features for invasion and
establishment. The management of invasive alien
fish species with known high fecundity potential and
high success rate of establishment in invaded
environments requires knowledge of its early life
stages, such as the timing of embryogenesis and
organogenesis and the shift from endogenous to
exogenous feeding (Godinho et al. 2003).
The armoured suckermouth sailfin catfish
Pterygoplichthys pardalis (Castelnau, 1855) is a
native of South America (Weber 2003), but has been
introduced to the Philippines through aquarium
trade. It has invaded many freshwater systems of
the country along with another hypostomine
loricariid, Ptergoplichthys disjunctivus. These two
species, popularly known in the country as “janitor
fish”, are not regarded as important commercial
fishes because of their hard body
armour, very little
meat, propensity to compete for food resources,
and their potential to bioaccumulate heavy metals in
polluted environments (Chavez et al. 2006, Lam and
Su 2009, Jumawan et al. 2010a). Nonetheless, the
hardy nature of this genus, its capacity to down
Received: January 2014
Received in revised form: April 2014
Accepted: April 2014
Printed: May 2014
INTRODUCTION
Abstract
Background: There is little information about the early development of this invasive fish species in
order to understand its early life history and developmental strategies towards invasion.
Material and Methods: Female Pterygoplichthys pardalis were induced to spawn using human
chorionic gonadotropin (HCG) so as to study the developmental stages from fertilization until yolk
resorption.
Results: The females subjected to a single dose of HCG responded positively to treatment (97%)
with higher fertilization success (88%) compared to the untreated females (21%). Nonetheless, the
HCG-induced fertilized eggs had a low hatching success (49%), while from the free-living embryos
successfully hatched, a high number (90%) survived to become juveniles. Embryonic development
in P. pardalis was completed 168 h and 30 min after fertilization, with the total yolk resorption
completed on the 8th day post hatching, during which the suckermouth gradually shifted from
rostral to ventral position to commence the loricariid algae-scraping feeding mode.
Conclusions: Pterygoplichthys pardalis does not undergo a true larval metamorphosis between the
free-living embryo and the juvenile stage and a definitive adult phenotype is developed directly.
These results provided basic, yet essential information on the early developmental features of this
invasive species whose spawning and early developmental strategies were difficult to observe in the
field. Implications of some ontogenetic features in this species with regards to invasion are also
discussed.
Keywords: Development, embryogenesis, invasion, larvae, morphology, Pterygoplichthys pardalis.
Jumawan JC, Herrera AA, Vallejo JrB (2014) Embryonic and larval development of the suckermouth
sailfin catfish Pterygoplichthys pardalis from Marikina River, Philippines. Eurasia J Biosci 8: 38-50.
http://dx.doi.org/10.5053/ejobios.2014.8.0.4
Joycelyn Cagatin Jumawan1*, Annabelle Aliga Herrera2, Benjamin Vallejo Jr3
©EurAsian Journal of BioSciences
regulate metabolism during periods of scarcity of
food (German et al. 2010), its tolerance to poor
water conditions and its ability to breathe air under
hypoxic water conditions (Armbruster 1998) enabled
this sh to invade and successfully establish itself
even in disturbed freshwater systems.
The seasonality of reproduction and the gonad
features of the Pterygoplichthys spp. population in
Marikina River has been previously described,
showing the females to be highly fecund and the
oocytes exhibiting features associated with parental
care (Jumawan et al. 2010b). However, the burrow-
spawning and nest-guarding nature involved during
the critical period of embryogenesis in these fishes
was found difficult to replicate under laboratory
conditions. To date, there are no available studies
describing the early development of P. pardalis and
P. disjunctivus in
their original habitat for comparison
with species thriving in non-native environments.
The present study will serve as a baseline reference
of the early embryogenesis, larval development and
organogenesis of the invasive P. pardalis through in-
vitro fertilization.
Strip method for artificial fertilization
The artificial breeding protocol of P. pardalis
adapted the procedure by Tan-Fermin et al. (2008)
with some modifications. All experiments were
carried out in triplicates. Collection and
experimentation were performed during the peak
spawning months (July to September) of the
Pterygoplichthys spp. population in Marikina River.
Because of the absence of defined sexual
dimorphism in this fish, large and mature P. pardalis
weighing 350-500 g were collected. Females were
selected based on their full and heavy body, gravid
abdomen and reddish, swollen vent. To identify the
sex of the fish, pressure on the abdomen to extrude
oocytes and the use of cannula were also attempted
for most samples. Males were selected based on a
streamlined body and flat abdomen. A total of 18
females and 18 males were utilized for this study.
Prior to hormonal induction, all females were
anesthetized in a 200 ppm 2-phenoxyethanol
(Merck, Germany) bath for 4 min. The female P.
pardalis received an intramuscular injection of
human chorionic gonadotrophin (HCG) (Argent
Chemicals, Philippines) at 4 IU/g body weight
(injection volume: 1μL/g BW). The hormone-injected
females were then placed separately in 1 m x 0.5 m
plastic tanks containing de-chlorinated tap water to
a depth of 0.4 m each. Females untreated with HCG
served as controls and were placed in a separate
tank.
Approximately 14-18 h after HCG administration,
the females were checked for ovulation by applying
pressure to the abdomen to confirm ovulation. Eggs
from ovulated females were then stripped in a dry
plastic basin. At about the same time, males were
anesthetized, sacrificed by a sharp blow to the head
and had their testes removed. Milt were collected
after maceration of the testes and then immediately
diluted with 0.9% NaCl to obtain milt solution. The
milt solution was poured into a bowl containing the
stripped eggs and mixed for 30-60 sec using a
feather. Approximately 5 mL of tap water was added
to the bowl to ensure fertilization. After 2 min of
gentle stirring, the fertilized eggs were transferred
to a plastic strainer and rinsed with running water
for about 1 min to remove excess milt. Fertilized
eggs were immediately transferred to a 45x30x30
cm aerated plastic aquarium for incubation. The
aquaria were provided with partial shade by using a
1x1 m black, plastic polyethylene bag to simulate
the darkened burrows in the field. The eggs were
examined 10-15 min after gamete mixing to check
for blastodisc formation. Unfertilized eggs were
carefully removed from the aquaria using fine
forceps.
Initial observations showed that the fertilized
eggs had a low hatching success rate when
incubated in de-chlorinated tap water (10-15%) or
rain water (10-20%), but had an improved hatching
rate when using a mix of natural river and tap water.
Hence, the subsequent trials used a mix of aerated
Marikina River water and tap water as an incubation
medium at 1:1 ratio. Prior to mixing the water
medium, river water and tap water were analyzed
for hardness (CaCO3/L), chloride (mg/L), calcium
Jumawan et al.
39
EurAsian Journal of BioSciences 8: 38-50 (2014)
MATERIALS AND METHODS
(mg/L) and magnesium (mg/L) levels. The physical
and chemical parameters of the mixed water in the
aquaria, such as temperature (°C), pH, salinity and
dissolved oxygen (D.O.), were recorded throughout
the experiment.
Embryogenesis, larval ontogenesis and
biometry
For each of the 3 replicate plastic aquaria, twenty
developing eggs were observed at 10-30 min
intervals until completion of cleavage, 3 times per
day until hatching. Hatchlings were documented
twice daily, until the yolk was fully resorbed. The
developmental stages were divided into embryo and
free-living embryo stages. The embryonic stage
occurs inside the chorion and ends in hatching. The
free-living embryo stage is characterized by the
nutritive contribution of the yolk sac and the stage
ends when the free-living embryo becomes capable
of exogenous feeding after the yolk has been
consumed (Geerinckx et al. 2008).
Three subsamples of fertilized eggs were
collected daily until yolk resorption and were xed
in Bouin’s fluid for histological purposes.
Additionally, three more fertilized eggs and
hatchlings were collected until yolk resorption and
were directly fixed in 4% neutral formaldehyde for
biometry using a digital caliper (accuracy 0.001 mm;
Control Company, USA), following the parameters
described in the study of Guimaraes-Cruz et al.
(2009) (Fig. 1).
The fertilization rate (number of eggs with
blastodisc/total oocytes x 100), hatching rate
(number of hatched eggs/number of fertilized eggs
x 100) and survival rate (number of surviving
juveniles/total number of hatched embryos x 100)
were recorded. Three replicate runs from
fertilization to hatching were conducted.
Analysis of variance (ANOVA) was used to
compare the mean values of the morphometric
variables according to each stage of larval
development, and between oocyte mean diameter
values for HCG-induced and non-HCG-induced
samples. All tests were conducted in a 0.05
signicance level using Graphpad Prism 5®.
Description of fertilized eggs
The adult female P. pardalis were administered
HCG a day after samples were obtained from the
field. Difficulty in distinguishing between males and
females was noted due to lack of defined sexual
dimorphism as well as difficulty in extruding oocytes
and milt from the vent after the application of
considerable abdominal pressure and cannulation.
Table 1 shows the physico-chemical parameters of
the water used for embryo incubation. Oocyte
extrusion was performed 18 h after the exposure to
HCG. Approximately 200-250 oocytes were extruded
from the ovaries of a single female exposed to HCG
due to difficulty of handling and applying pressure in
the abdomen of the fish. The females injected once
with HCG responded positively to the treatment
(97% success rate) and hydrated, producing nearly
uniform sized (2-3 mm), transparent-yellow oocytes
after the ovary was stripped. In contrast, the oocytes
from females not exposed to HCG were mostly
opaque yellow with occasional pre-vitellogenic
oocytes (<1 mm) along with larger vitellogenic
oocytes. Adhesion of oocytes was observed
immediately after any excess milt was washed off.
The HCG-injected females had a higher fertilization
success (88.3%) compared to the untreated females
(20.9%). The fertilized eggs from HCG-injected
females had low hatching success (48.6%), with
mortalities during early stage somitogenesis.
From the successfully hatched embryos, a high
number survived to become free-living embryos up
to the termination of experiments on the 8th day post
hatching (8 dph) (Table 2). The diameter of the eggs
did not differ significantly between the two
treatment groups. No change in the diameter of the
fertilized oocytes (3.2±0.23 mm) was also observed
from the onset of fertilization until the pre-hatching
stage, although a very small perivitelline space was
formed surrounding P. pardalis eggs a few minutes
after fertilization.
Embryonic development
The observations of P. pardalis were divided into
two periods: (1) before hatching (embryo), and (2)
between hatching and yolk sac depletion (free-living
40
Jumawan et al.
EurAsian Journal of BioSciences 8: 38-50 (2014)
RESULTS
embryo). The main events in embryogenesis and
their respective times of observation are
summarized in Table 3 and Figs. 2-4. The fertilized
eggs of P. pardalis show meroblastic cleavage in
which the blastoderm is restricted to a small area at
the animal pole. The first segmentation that divided
the blastodisc into two blastomeres occurred within
5 to 15 min after fertilization. Subsequent
successive cleavage until 64 blastomeres was
observed until it was completed after 5 h with
blastomeres very much decreased in size.
Blastomeres very fine in appearance eventually
flattened at the animal pole at 14-15 h post
fertilization. The spread of the blastoderm was
evident, covering the yolk with the embryo body
becoming more elongated, while the head and tail
ends of the embryo can be clearly seen at 20 h.
Finally, full yolk invasion and closure of the
blastopore was observed at 24 h.
Differentiation of the embryo
Pre-hatching organogenesis in P. pardalis
commenced with the formation of the notochord
and observations of the cranial-caudal portions of
the embryo (29 h), as well as the first somites and
the optic vesicles (36 h). Somitogenesis was
observed starting 2 days post fertilization (2 dpf)
Jumawan et al.
41
EurAsian Journal of BioSciences 8: 38-50 (2014)
Fig. 1. Lateral view of a P. pardalis free-living embryo (4 dph;
14.2 mm TL).
TL: Total length; SL: Standard length; PAD: Pre-anal distance; HL:
Head length; HH: Head height; SNL: Snout length; ED: Eye diameter;
YSL: Yolk sac length; YSH: Yolk sac height, BH: Body height.
Fig. 2. Stages of embryonic development in P. pardalis.
(a) Early stage fertilized egg with distinct invagination at the animal
pole before blastodisc formation; (b) Blastodisc stage; (c) Cleavage
at 2-cell stage; (d) Cleavage at 4-cell stage; (e) Cleavage at 8-cell
stage; (f) Cleavage at 64 blastomeres; (g) High blastula stage; (h)
Low blastula; (i) Early gastrula.
Scale bar: 1 mm.
Fig. 3. Stages of embryonic and larval development in P.
pardalis.
(a) Late gastrula: formation of the embryonic shield (es); (b)
Somites development (black arrow); (c) Late Neurula stage; f:
Forebrain; m: Midbrain; h: Hindbrain, oc: Otic capsule; (d-e) Rostral
location of suckermouth (*), Note egg sac removed; (f) P. pardalis at
6 dpf; (g) Newly hatched embryo (7 pf); (h) Note the ventral
position of the suckermouth in (g).
Scale bar: 1 mm.
and subsequent somite formation enabled tail
movement, although the embryo was largely
confined within the perivitelline space. No
corresponding somite formation for each hour
during the onset of embryogenesis in P. pardalis was
noted because of the difficulty in turning the
strongly adhesive eggs to locate the somites
without causing mechanical injury to the developing
embryos. The time when the pericardial cavity was
formed was not determined; however, the pulsating
heart and blood circulation was visible as a very faint
stream of capillary network in the pericardial cavity
at 36 h after fertilization. Ectodermal thickening to
form the lens of the eyes was observed starting at
36 h, with the eye lens fully formed at 49 to 50 h post
fertilization. The suckermouth was rostrally
positioned inside the membrane (Fig. 3d-e, Fig. 4a).
Tail movement (72 h post fertilization) was
observed before the formation of the vitelline
circulatory system. Movement in the gill cavity and
blood circulation in the gill arches was noted
beginning at 5 dpf. Hatching was observed 167-168
h post fertilization. The caudal tail region was
observed detaching from the yolk mass with a
subsequent increase in body movement, causing the
rupture of the chorion and the emergence of the
embryo from the capsule.
Free-living embryo development
Details of development and average body
measurements are reflected in Tables 4 and 5.
Histologic sections of larvae at different days post
hatching are shown in Figs. 5-7.
Hatchling
Newly hatched, free-living embryos or eleuthe-
rembryos (Balon 1986) have a mean total length of
7.86±0.12 mm, still containing a large amount of yolk
(Fig. 5a). The newly hatched larvae had already
ventrally located the suckermouth and maxillary
barbels and were able to attach to the sides of the
glass substrata with their suckermouth, with water
inflow for the sucking action passing through the
furrows of the maxillary barbels, and attachment
sometimes assisted by body and tail movements
(Fig. 3h). Embryos initially had a rostrally located
42
Jumawan et al.
EurAsian Journal of BioSciences 8: 38-50 (2014)
Table 1. Some physico-chemical features of the water medium used in this study.
*Data analyzed only once, through the Research and Analyti cal Servi ces Laboratory (RASL), Natural Sciences Research Institute (NSRI);
Remaining data monitored once in every 3 days for 15 days. Values are listed as means±S.E.
Table 2. Percent (%) response of P. pardalis to HCG-injection at 24ºC
*P<0.001; Values are listed as means±S.E.
mouth, which gradually shifted into the ventral
location during hatching. Dorsal and caudal fins were
observed. The bodies of the newly hatched free-
living embryos were transparent, except for the
onset of dendritic pigmentations initially found
interspersed finely on the head and on the edges of
the snout. The eye diameter was small (0.51±0.02
mm). The dorsal and caudal ns were well
recognizable. Serial sections of the digestive system
during the day of hatching showed that the gut is
tubular and closed at both ends, while the intestine
is composed of simple cubic epithelium. The first
outlines of the gill arches supported by blood
vessels were observed in the gill cavity. A tubular
heart was observed.
3-d old free- living embryo
A continued reduction of yolk sac and an increase
in body length (TL 12.41±0.05 mm) were observed
along with increasing pigmentation in the retina and
increasing complexity of the gills. Increasing
dendritic chromatophore pigmentation was
observed in the outlines of the head and dorsal side
of the body. The digestive system was characterized
by a simple striated border in the intestine (Fig. 5f).
The cranial kidney was now well developed with
readily recognizable glomeruli tufts within the
network of reticular fibers (Fig. 5e). The spleen could
be observed on the left lateral-dorsal-right lateral
part of the borderline between the gut and the
kidney. The liver could be observed ventrally located
in the cranial region with the hepatic parenchyma
very homogenous in appearance.
5-d old free-living embryo
The average length of a 5-day old free-living
embryo was about 14.15±0.04 mm TL with the yolk
sac length becoming gradually reduced (3.22±0.03
mm). Increasing pigmentation all throughout the
body was observed with dendritic chromatophores
reaching the lateral region caudally to the pectoral
n (Fig. 4c-d). The yolk was visible, although less
compact near the digestive system. The intestines
became looped and lengthened, with the intestinal
epithelium exhibiting a complex columnar
arrangement of mucous and goblet cells. Intestinal
Jumawan et al.
43
EurAsian Journal of BioSciences 8: 38-50 (2014)
Table 3. Main events of the embryonic development of P. pardalis and their respective times (mean) after HCG-induced
fertilization at 24ºC. n: 192 developing eggs/stage.
content (residue) was observed in this stage.
Nephric ducts in the cranial kidney were very visible;
the encephalon was compact with associated cells
(Fig. 6e). The heart was compartmentalized with a
very visible atrium (Fig. 6f).
7-d old free-living embryo
The average length of 7-day old free-living
embryo was about 14.84±0.34 mm TL with the yolk
sac length becoming gradually reduced (2.32±0.09
mm). Increased pigmentation all throughout the
body including the ventral part previously occupied
by the yolk sac (Fig. 4e). Full pigmentation of the
eyes observed (Fig. 7c). Increased use of the
suckermouth to anchor body in the glass aquaria was
observed in larvae. The gills were more developed
with elongated filaments and gill lamellae with
intense vascularization. Gas bladder was observed
with simple squamous epithelium (Fig. 7d). The heart
presented two compartments. Small intestine
mucosa can be seen, including microvilli, columnar
epithelium and muscularis mucosa. The head kidney
had visible renal tubules and extensive hemato-
poetic tissue.
8-d old free-living embryo
The average total length of free-living embryo
was 6.95±0.20 mm TL and the yolk was fully
resorbed. The body had become opaque with the
accumulation of pigments all throughout its surface
as chromatophores became more numerous and
darker, maintaining the same distribution pattern.
Undifferentiated gonad with primordial germ cells
was observed (Fig. 7e), while a true larval stage
(Balon 1986, 1999) was not observed. P. pardalis
underwent a direct transition from a free-living
embryo with large yolk into a juvenile without
undergoing a true larval stage when the yolk was
fully consumed at 8 dph. Except for the absence of
hardened armour covering the body and the
abdominal pattern distinct for P. pardalis, an adult-
like appearance was observed at the moment the
44
Jumawan et al.
EurAsian Journal of BioSciences 8: 38-50 (2014)
Fig. 4. Free-swimming embryo development in P. pardalis.
(a) Embryo with yolk sac removed. Note the rostral position of the
mouth; (b) Development at 3 dph; (c-d) Development at 5 dph; Note
the capacity of the ventrally positioned mouth for attachment in
the glass substrata in; (c) Body pigmentation in (d); (e)
Pigmentation in the ventral position of the 7 dph free-living
embryo.
Fig. 5. Organogenesis in P. pardalis.
(a) Newly hatched free-living embryo; medial portion; (b) 2dph,
cranial portion; (c) 3 dph, Retina (*) and its layers; (d) 3 dph, cranial
portion; (e) 3 dph, Digestive system; (f) 3 dph, intestine with simple
striated border. N: Nostrils; E: Encephalon; H: Heart; B: Barbells; CK:
Cranial kidney; GB: Gas bladder; I: intestine; y: Yolk; G: Ganglionar
cell; IN: Inner nuclear layer; ON: Outer nuclear layer; YG: ocular
globe; GA: gill arches.
Scale bars: a,d,e: 200 μm; c,f: 20 μm; b: 40 μm.
Jumawan et al.
45
EurAsian Journal of BioSciences 8: 38-50 (2014)
Table 5. Body measurements* (mm) of P. pardalis from hatching until yolk resorption.
TL: Total length; SL: Standard length; PAD: Pre-anal distance; HL: Head length; HH: Head height; SNL: Snout length; ED: Eye diameter; YSH:
Yolk sac height; YSL: Yolk sac length; BH: Body height. Values are listed as means±S.E.; n= 212
Table 4. Main morphologic events occurring during the development of P. pardalis from hatching until yolk resorption (TL,
mm).
last yolk was consumed at 8 dph.
Biometric parameters registered a gradual
increase in all values except for the decrease in the
yolk sac length and height as the free-living embryo
neared and completed the yolk resorption
externally (Table 5). A full yolk resorption in the
juvenile P. pardalis was observed 336 h or 14 dpf at
24°C. All the oocytes observed for fertilization until
yolk resorption in juveniles developed synchro-
nously.
Most invasive loricariids do not reproduce
spontaneously when reared under laboratory
conditions (Alfaro et al. 2008). The present study
provides baseline information regarding the early
stages development of P. pardalis, a highly invasive
loricariid, whose early life history has not been
studied so far.
The reproduction and spawning behavior of the
janitor fish P. pardalis is difficult to observe, as they
are known to spawn in burrows with males guarding
the fertilized clutch. What is surprising in this
experiment, however, is the tendency of the eggs to
hatch well in river water characterized to be hardy,
with higher calcium and magnesium levels compared
with tap water.
The water quality of key areas along Marikina
River where spawning colonies of P. pardalis are
abundant showed that the river is highly eutrophic
and turbid, and had overall high conductivity,
nitrate, ammonia and phosphate levels. Fertilized
clutches from the river also have a higher hatching
46
Jumawan et al.
EurAsian Journal of BioSciences 8: 38-50 (2014)
Fig. 7. Organogenesis in P. pardalis.
(a) 6 dph; gas bladder and kidney; (b) 6dph, intestine; (c) 7 dph,
cranial portion; (d) 7 dph, digestive system; (e) 8 dph,
undifferentiated gonad; (f) 8 dph, segmentation of the notochord.
B: Barbels; GB: Gas bladder; CK: Cranial kidney; I: Intestine; Y: Yolk.
G: Ganglionar cell; IN: Inner nuclear layer. ON: Outer nuclear layer;
YG: Ocular globe; GA: Gill arches; H: Heart; PGC: Primordial germ
cell; N: Notochord; DF: Dorsal fin; M: Muscle.
Scale bars: c,d,f: 200 μm; b,e: 20 μm.
DISCUSSION
Fig. 6. Organogenesis in P. pardalis.
(a) 3 dph, cranial portion; (b) 3 dph, spleen and kidney; (c) 4 dph,
dorsal nervous tube; (d) 4 dph, digestive system; (e) 5 dph,
encephalon; (f) 5 dph, heart. E: Encephalon; H: Heart; B: Barbells; K:
Kidney; I: Intestine; y: Yolk; G: Ganglionar cell; IN: Inner nuclear
layer; ON: Outer nuclear layer; YG: Ocular globe; GA: Gill arches; S:
Stomach; SP: Spleen, N: Notochord; M: Myomeres; NC:
Neurocranium.
Scale bars: a: 200 μm; b,d: 40 μm; c,e,f: 40 μm.
Jumawan et al.
47
EurAsian Journal of BioSciences 8: 38-50 (2014)
rate if incubated in river water. It is not clear how the
nest guarding nature of the males for this species
actually influences the successful hatching because
of the difficulty of observing this process in the field.
Makeshift and darkened plastic aquaria to replicate
burrows in the field in the preliminaries of this
experiment failed to encourage spawning under
laboratory conditions. The low success rate in
hatching for P. pardalis under artificially-induced
spawning conditions may be attributed to the
absence of parental care during the experiment or
to the preference of eggs for river water
environments.
Eggs of P. pardalis are highly adhesive, allowing
the formation of tight clutches of eggs. Teleost eggs
can be non-adhesive, weakly adhesive or strongly
adhesive. Rizzo et al. (2002) point out that adhesive
eggs are often large, laid in smaller numbers and
associated with the sedentary nature of the species
or with parental care. The zona radiata of
Pterygoplichthys spp. in Marikina River was thin
(mean 4.56 μm), while its granulosa layer was thick
(mean 2.81 μm) (Jumawan and Herrera 2014). A thick
granulosa layer may provide better adhesion of
eggs, while the thin zona radiata may be
compensated by the nest guarding nature of the
males of this species, as was the nature of some nest
guarding loricariids (Suzuki et al. 2000).
It was apparent in the results that although
artificially fertilized eggs had a low hatching rate, all
hatched P. pardalis free-living embryos successfully
survived and had resorbed their yolk at 8 dph. A
large endogenous supply of yolk nutrients enhances
survival during the period when feeding structures
are still developing (Geerinckx et al. 2008), while also
enabling the free-living embryo to avoid an
intermediate larval stage and the cost of
metamorphosis since a denitive adult phenotype
was developed directly (Balon 1986).
Although hardened armour covering the body
was not observed when the last yolk was consumed
at 8 dph, hardening of the head and dorsal structure
towards development of the armoured covering of
the body was eventually observed 30 days post yolk
resorption in juveniles (data not shown).
Types of feeding (exogenous, endogenous),
nutrient supply (altricial, precocial) and life history
models (indirect, direct development) were used as
basis in the description of embryos (Balon 1986,
1999). Embryos with indirect development are a
consequence of poor vitellogenesis and depend
entirely on an endogenous nutrient supply, as eggs
are altricial in nature (Balon 1986). The short embryo
period in this type of development appears to be
extended by the larva period that feeds
exogenously prior to the formation of the denitive
phenotype. Fish taxa characterized by direct
development have
a prolonged embryo period due
to a large endogenous supply as eggs are precocial
(large amount of yolk) that ultimately enables the
embryo to develop directly into a definitive
phenotype (Balon 1986). Embryos develop for
longer but directly into juveniles that are able to
compete in the adult habitat (Balon 1999). Direct
development occurs more frequently in the
reproductive guilds of guarders and bearers as these
fish groups exhibit parental care such as site
selection, egg deposition and nest guarding (Balon
1986). A true larva requires some tissues and
structures very different to those in the denitive
organism, and so, has to be remodeled through the
process of metamorphosis (Balon 1999). Fish species
with direct development lack the necessary cost of
forming temporary organs through metamorphosis.
The large yolk (mean 3.3 mm) in P. pardalis is an
advantage in this context, as it requires none or little
external nutrients to develop into a definitive
phenotype. It has been proposed that the larger and
more advanced an individual at the onset of
exogenous feeding, the better its chances of
surviving (Balon 1999). This scenario improves
competitiveness in P. pardalis even in its early stages
and could be an advantage for invasion (Balon 1986).
The absence of a true larval stage in P. pardalis is
similar to most loricariids (Geerinckx et al. 2008). It is
however important to note that a slight variation in
measurements may be caused by the xative used in
this study (4% neutral formaldehyde), as xatives
may have some dehydrating effects contrary to the
measurement of fresh samples.
The eggs of P. pardalis contain a large amount of
48
Jumawan et al.
EurAsian Journal of BioSciences 8: 38-50 (2014)
evenly distributed yolk, hence, is classified as
telolecitic (Ribeiro et al. 1995, Marques et al. 2008).
P. pardalis has a meroblastic cleavage pattern
restricted to the animal pole, which is common in
teleosts. Blastopore closure occurred 24 h after
fertilization, indicating fertilization success. The
embryonic development of P. pardalis lasted 7 days
(168 h, 30 min), which is long compared with the
duration of embryogenesis in other siluriforms: 45 h,
50 min at 24ºC in R. aspera (Perini et al. 2009), 21 h,
20 min at 23ºC in Pimelodus maculates (Luz et al.
2001) and 18 h at 27ºC in P. corruscans (Marques et
al. 2008).
Long embryonic periods are known to be
associated with non-migratory species having large
eggs and with those that display parental care
(Sargent et al. 1987), features that both fit the
natural history of P. pardalis. Invasive species
exhibiting parental care may be considered an
advantage in the context of ensuring egg survival
until the juvenile period. By females precisely
selecting a safe and protected site for egg
deposition and males guarding the nest, the
progress of the early developmental stages in P.
pardalis becomes ensured. This protection makes
the early stages difficult to observe in the wild as
they are well hidden from predators and other
threats, emerging only when post-y
olk sac juveniles
are capable of feeding and the definitive phenotype
ensures the adult form.
A pigmentation of the retina observed early
during hatching in P. pardalis could be associated
with the need to develop a functional visual system
before the first feeding, typical in some fishes (Hall
et al. 2004). The undifferentiated gonad located
between the cranial kidney and the digestive tract in
P. pardalis already contains a cluster of primordial
germ cells (PGCs) at 8 dph. PGCs in the
undifferentiated gonad were also observed at 5 dph
in
R. aspera and at 13 dpf in Pimephales promelas
(Uguz 2008).
The shift of the suckermouth from rostral to
ventral position long before the hatching stage is a
common feature for most loricariids. This shift is of
more advantage to P. pardalis as newly hatched free-
living embryos already had a ventrally flattened
mouth despite the large yolk sac during hatching. On
the contrary, A. cf triradiatus, a relative loricariid,
was noted to have a rostro-ventrally positioned
suckermouth during hatching, but had to w
ait for 2-
4 days for the yolk sac to resorb for the transition to
a ventrally flattened mouth (Geerinckx et al. 2008).
The newly hatched P. pardalis in this study was able
to attach immediately to the substrate with their
suckermouth. This observation w as also reported for
Sturisoma aureum (Riehl and Patzner 1991) and A. cf.
triradiatus (Geerinckx et al. 2008). This feature may
be essentially an advantage for loricariid species
where hatchlings leave the shelter immediately
(Suzuki et al. 2000). In the case of an invasive species
such as P. pardalis, accidental release of eggs and
juveniles may result in assured higher survival rates
in the wild.
The SL of P. pardalis during the time of hatching
(7.86±0.12 mm) and the large yolk sac during the
same time fall within the size range for most
loricariids (6-8 mm) (Riehl and Patzner 1991,
Nakatani et al. 2001, Geerinckx et al. 2008, Perini et
al. 2009). The morphometric development in P.
pardalis, which reflected a gradual increase in all
parameters with decreasing yolk size, is also
observed in A. cf triradiatus (Geerinckx et al. 2008), R.
aspera (Perini et al. 2009) and Lophiosilurus alexandri
(Guimaraes-Cruz et al. 2009), all of which exhibited a
high degree of allometric growth. Geerinckx et al.
(2008) hypothesized that loricariid hatchlings often
exhibit rapid allometric growth in the snout and
remarkable lip transformation because it is a
necessity and advantageous for the suckermouth to
attach to the substratum for scraping and sucking
food.
In conclusion, this study pointed out several
baseline features of the early life strategies in P.
pardalis that may be of advantage to its biotic spread
potential: (1). The propensity of the embryo to thrive
in polluted water; (2) the adhesiveness of the eggs
allowing for higher hatching success rate, further
contributing to the nest guarding feature of males;
(3) the already ventrally positioned suckermouth
and sucking capacity of the free living embryo,
allowing for higher survival potential once left out of
parental care due to its substratum scraping capacity
Jumawan et al.
49
EurAsian Journal of BioSciences 8: 38-50 (2014)
for food and attachment; and (4) the absence of true
larval metamorphosis between the free-living
embryo and the juvenile stage due to the large
supply of yolk. Information on the early life history
strategies of P. pardalis from its original habitat
would be essential for comparison with the non-
native counterparts in order to determine possible
developmental plasticity of the fish in invaded water
systems.
JC Jumawan is grateful to the Commission on
Higher Education- Science and Engineering Grants
(CHED-SEGS) for the dissertation grant and to the
Philippine Kidney Transplant Institute (PKDF)
Histology laboratory for the histological
preparations. The authors thank Drs PJ Denusta and
LMB Garcia for the technical help in the in-vitro
experiment.
ACKNOWLEDGEMENTS
Adriaens D, Vandewalle, P (2003) Embryonic and larval development in catfishes. In: Arratia G, Kapoor AS, Chardon M,
Diogo R (eds.), Catfishes, Enfield, NH, Science Publishers, 639-666.
Alfaro RM, Fisher JP, Courtenay W, Martinez CR, Mendoza A, Gallardo CE, Torres PA, Osorio PK, Balderas SM (2008)
Armoured catfish (Loricariidae) trinational risk assessment. In: Orr R (ed.), Trinational Risk Assessment Guidelines
for Aquatic Alien Invasive Species, Commission on Environmental Cooperation, Montreal, 25-37.
Armbruster JW (1998) Modifications of the digestive tract for holding air in loricariid and scolopacid catfishes. Copeia
1998 (3): 663-675.
Balon EK (1986) Types of feeding in the ontogeny of fishes and the life-history model. Environmental Biology of Fishes
16: 11-24. http://dxdoi.org/10.1007/BF00005156
Balon EK (1999) Alternative ways to become juvenile or a definitive phenotype (and on some persisting linguistic
offenses). Environmental Biology of Fishes 56: 17-38.
Chavez HM, Casao EA, Villanueva EP, Paras MP, Guinto MC, Mosqueda MB (2006) Heavy metal and microbial analyses
of the janitor fish (Pterygoplichthys spp.) in Laguna de Bay, Philippines. Journal Environmental Science and
Management 9(2): 31-40.
Geerinckx T, Verhaegen Y, Adriaens D (2008) Ontogenetic allometries and shape changes in the suckermouth
armoured catfish Ancistrus cf. triradiatus Eigenmann (Loricariidae, Siluriformes), related to suckermouth
attachment and yolk sac size. Journal of Fish Biology 72: 803-814.
http://dxdoi.org/ 10.1111/j.1095-8649.2007.01755.x
German DP, Neuberger DT, Callahan MN, Lizardo NR, Evans DH (2010) Feast to famine: The effects of food quality and
quantity on the gut structure and function of a detritivorous catfish (Teleostei: Loricariidae). Comparative
Biochemistry and Physiology 155: 281-293.
Godinho HP, Santos JE, Sato Y (2003) Ontogenese larval de cinco especies de peixes do Sao Francisco. In: Godinho HP,
Godinho AL (eds.), Aguas, Peixes e Pescadores do Sao Francisco das Minas Gerais, Belo Horizonte: CNPq/PADCT,
Editora PUC Minas,133-148.
Guimaraes-Cruz RJ, Santos JE, Sato Y, Veloso-Junior VC (2009) Early development stages of the catfish Lophiosilurus
alexandri Steindachner, 1877 (Pisces: Pseudopimelodidae) from the Sao Francisco River basin, Brazil. Journal of
Applied Ichthyology 25: 321-327.
Hall TE, Smith P, Johnston IA (2004) Stages of embryonic development in the Atlantic cod Gadus morhua. Journal of
Morphology 259: 255-270.
Jumawan JC, Hererra AA, Jacinto SD (2010b) Length-weight and gonado-morphometric characterization in the janitor
fish Pterygoplichthys disjunctivus (Weber, 1991) from Marikina River, Philippines. In: Dolores R (eds.), Transactions of
the National Academy of Science & Technology (NAST-Philippines) 10 July 2010, Manila, 32: 80.
Jumawan JC, Herrera AA (2014) Ovary Morphology and Reproductive Features of the Female Suckermouth Sailfin
Catfish, Pterygoplichthys disjunctivus (Weber 1991) from Marikina River, Philippines. Asian Fisheries Science 27: 75-89.
REFERENCES
50
Jumawan et al.
EurAsian Journal of BioSciences 8: 38-50 (2014)
Jumawan JC, Salunga TL, Catap ES (2010a) Lipid peroxidation and cadmium and lead accumulation in the vital organs
of the suckermouth armoured catfish Pterygoplicthys Gill 1858 from Marikina River. Journal of Applied Science in
Environmental Sanitation 5 (4): 375-390.
Lam JC, Su GL (2009) Arsenic and mercury concentrations of the waters and Janitor fish (Pterygoplichthys spp.) in
Marikina River, Philippines. Journal of Applied Science in Environmental Sanitation 4(3): 187-195.
Luz RK, Reynalte-Tataje DA, Ferreira AA, Zaniboni-Filho E (2001) Desenvolvimento embrionario e estagios larvais do
mandi-amarelo Pimelodus maculatus. Boletim do Instituto de Pesca 27: 49-55.
Marques C, Nakaghi LSO, Faustino F, Ganeco LN, Senhorini JA (2008) Observation of the embryonic development in
Pseudoplatystoma coruscans (Siluriformes: Pimelodidae) under light and scanning electron microscopy. Zygote 16:
333-342. http://dxdoi.org/ 10.1017/S0967199408004838
Nakatani K, Agostinho AA, Baumgartner G, Bialetzki A, Sanches PV, Makrakis MC, Pavanelli CS (2001) Eggs and larvae
of freshwater fishes: development and identification manual. Editora da Universidade Estadual de Maringa 378.
Perini N, Sato Y, Rizzo E, Bazzoli N (2009) Biology of eggs, embryos and larvae of Rhinelepis aspera (Spix & Agassiz, 1829)
(Pisces: Siluriformes). Zygote 18: 159-171. http://dxdoi.org/ 10.1017/S0967199409990165
Ribeiro CR, Leme dos Santos HS, Bolzan, AA (1995) Estudo comparativo da embriogenese de peixes osseos (Pacu,
Piaractus mesopotamicus, Tambaqui, Colossoma macropomum e hibrido Tambacu). Revista Brasiliera Biologia 55: 65-
78.
Riehl R, Patzner RA (1991) Breeding, egg structure and larval morphology of the catfish Sturisoma aureum
(Steindachner) (Teleostei, Loricariidae). Journal of Aquariculture and Aquatic Sciences 6: 1-6.
Rizzo E, Sato Y, Barreto BP, Godinho HP (2002) Adhesiveness and surface patterns of eggs in neotropical freshwater
teleosts. Journal of Fish Biology 61: 615-632. http://dxdoi.org/10.1006/jfbi.2002.2085
Sargent RC, Taylor PD, Gross MR (1987) Parental care and evolution of egg size in fishes. American Naturalist 121: 32-
46.
Suzuki HI, Agostinho AA, Winemiller KO (2000) Relationship between oocyte morphology and reproductive strategy in
loricariid catfishes of the Parana River, Brazil. Journal of Fish Biology 57: 791-807.
http://dxdoi.org/ 10.1111/j.1095-8649.2000.tb00275.x
Tan-Fermin JD, Fermin AC, Bombeo RF, Evangelista AD, Catacutan MR, Santiago CB (2008) Breeding and seed
production of the Asian catfish Clarias macrocephalus (Gunther). Aquaculture Extension Manual 40, Southeast Asia
Fisheries Development Center, Iloilo.
Uguz C (2008) Histological evaluation of gonadal differentiation in fathead minnows (Pimephales promelas). Tissue and
Cell 40: 299-306. http://dxdoi.org/.10.1016/j.tice.2008.02.003
Weber C (2003) Loricariidae-Hypostominae (Armored catfishes). In: Reis RE, Kullander SO, Ferraris CJ Jr. (eds.), Checklist
of the Freshwater Fishes of South and Central America, Edipucrs, Puerto Alegre.
