Luận án The effects of elevated temperature and hypoxia on the respiratory organs of Pangasianodon hypophthalmus

𝐹
× 𝑡̅ × (
𝑎
𝑝
) × ∑ 𝑃𝑒𝑝,𝑝𝑐,𝑏𝑙 
VSL (mm3) = Vep + Vpc + Vbl 
Vparasites (mm3) = 
1
𝑆𝑆𝐹
×
1
𝐴𝑆𝐹
× 𝑡̅ × (
𝑎
𝑝
) × ∑ 𝑃𝑝𝑎𝑟𝑎𝑠𝑖𝑡𝑒𝑠 
SSF is section sampling fraction; 
AFS is area sampling fraction; 
t is section thickness; 
∑PP are total number of points hitting primary 
filaments 
A/P is area per point 
∑PP,ep,pc,bl are total number of points hitting 
primary filaments, epithelium, pillar cells, and 
blood spaces, respectively; 
a/p is area per point 
∑Pparasites are total number of points hitting 
parasites 
Surface area of secondary lamella SSL 
SSL (mm2) = 2 × 
∑ 𝐼𝑖
(
𝑙
𝑝
)× ∑ 𝑃𝑖
 × 𝑉𝑆𝐿 
∑Ii are total number of intersections between 
test lines and secondary lamellae; 
∑Pi are total number of points hitting 
secondary lamellae 
l/p is length per point 
VSL is volume of secondary lamellae 
The precision of the Cavalieri reference volume was 
estimated using the coefficient of error (CE) 
CE (∑P) = 
√𝑇𝑜𝑡𝑎𝑙 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒 𝑜𝑓 ∑ 𝑃
∑ 𝑃
Total variance of ∑P = Noise + VarSURS(∑area); 
Noise = 0.0724 × (b/√𝑎) × √𝑛 × ∑𝑃; 
VarSURS(∑area) = 
3(𝐴−𝑁𝑜𝑖𝑠𝑒)−4𝐵+𝐶
240
; 
with A=Pi×Pi; B=Pi×Pi+1; C=Pi×Pi+2 
Mean CE = [(CE12+CE22+...+CEn2)/n]1/2; 
CV=SD/Mean 
Noise is the point counting error variance; 
b/√𝒂 is the average profile shape and, in this 
study, the value of 12 was used for secondary 
lamella shape according to the nomogram of 
Gundersen and Jensen (1987); 
n is the number of examined sections; 
∑ 𝑷 is the total number of points hitting the 
observed tissue; 
VarSURS(∑area) is variance of the total 
surface area. 
CV is Coefficient of variance 
Specific growth rate (SGR) of individual fish 
SGR (%) = [(lnW2 lnW1) / t] × 100 
W1 and W2 are the initial and final weight of 
the fish, respectively; 
t is time (days) between W2 and W1 
Statistics 
Data were presented as means ± standard error of the mean (means ± S.E.M). Two-way 
MANCOVA (a multivariate analysis of covariance) (Ramos et al., 2012) was used in this study 
to test for overall effects of temperature, oxygen or parasite volume on gill parameters followed 
by Turkey’s post hoc test to identify differences between samples. For growth performance, 
two-way ANOVA was used to test whether higher temperature and/or hypoxia significantly 
affected on P. hypophthalmus growth performance as well as to compare means amongst 
treatments in each period of samplings. A probability (P) value at the 0.05 level was considered 
as significant. All the statistics applied in this study were performed by using PASW Statistic 
18.0. 
 56 
3.3. Results 
Gill morphometric 
Gill surface area 
There were significant changes in secondary lamellae surface area of the fish with time, 
with temperature and with hypoxia (Fig. 3.1A, B; Table 3.1; Table 3.2). At the beginning of 
the experiment, the respiratory surface area of small fish (12.4±0.8 g) at 27ᵒC in normoxic 
water was 212±22 mm2 g-1. After 5 weeks in 27ᵒC, however, the lamella surface area decreased 
significantly to 12.9±7.9 mm2 g-1. The addition of hypoxia in this group resulted in even smaller 
4.9±2.3 mm2 g-1 although there was no significant difference between these two groups 
(P>0.05). In both 27ᵒC groups at 5 weeks, the gill filaments had undergone a significant change 
in morphology, with the inter lamellae spaces filling almost completely with a cell mass (Fig. 
3.2). In contrast, fish exposed to 33ᵒC, showed surface area at 5 weeks that allometrically scaled 
with size in the hypoxic group (208±29 mm2 g-1) whereas in the normoxic group a reduction to 
39.4±3.5 mm2 g-1 occurred (Fig. 3.1B). 
With long-term exposure (16 weeks) to the same conditions, the surface area of 
lamellae in the low temperature groups was again much lower than those in elevated 
temperature. Highest surface area (160±19 mm2 g-1) was found in 33ᵒC & hypoxia group, 
followed by 133±25 mm2 g-1 in 33ᵒC & normoxia, which was 8.0-fold and 6.5-fold, larger than 
the corresponding areas in the two groups at 27ᵒC (Table 3.1). 
 57 
Fig. 3.1 Respiratory lamella surface area according to body mass of juvenile striped catfish 
(Pangasianodon hypophthalmus) reared in different temperatures and/or hypoxia after 5 week 
exposures (I) and 1 week recovery without hypoxia exposure (II). A shows respiratory lamella surface 
area of individual fish; B shows mean respiratory lamella surface area per fish body mass, with data 
being presented as mean ± S.E.M. Different letters indicate significant difference amongst treatments 
(P<0.05). 
Fig. 3.2 Light micrographs of Pangasianodon hypophthalmus gill filaments from fish in 27ᵒC and 
Normoxia treatment (left) and fish in 33ᵒC and Hypoxia treatment (right). 
Scale bar 50 µm. 
Fish mass (g)
0 20 40 60 80
R
e
s
p
ir
a
to
r y
 l
a
m
e
ll
a
 s
u
rf
a
c
e
 a
re
a
 (
m
m
2
)
0
2000
4000
6000
8000
10000
12000
14000
16000 27oC & 92% (Beginning) 
27oC & 92% (I)
27oC & 35% (I) 
33oC & 92% (I) 
33oC & 35% (I) 
27oC & 92% (II) 
27oC & 92%* (II) 
33oC & 92% (II) 
33oC & 92%* (II) 
Time (weeks)
0 1 2 3 4 5 6
S
u
rf
a
c
e
 a
re
a
/M
a
s
s
 (
m
m
2
 g
-1
)
0
50
100
150
200
250
a a
ac
b
b
b
b
c
A B
 58 
Water – blood diffusion distance 
The harmonic mean barrier thickness of juvenile P. hypophthalmus (12.4±0.8 g) was 
1.67±0.12 µm (Table 3.1). It was almost impossible to measure respiratory diffusion distance 
in the groups of fish after 5 and then 6 weeks of low temperature exposure under either hypoxic 
or normoxic water because their lamellae were filled with ILCM (Fig. 3.2). However, in the 
33ᵒC groups hypoxia caused a significant reduction in harmonic mean barrier thickness from 
1.64±0.16 µm in normoxia to 1.04±0.18 µm in hypoxia (Table 3.1). 
Volumes of gill components 
Table 3.1 presents the volume of gill filaments and the percentages of secondary lamella 
on the filaments as well as the percentages of other components (pillar cells, blood spaces, and 
epithelium) constructing secondary lamella of experimental fish. The total volume of gill 
filaments of juvenile striped catfish P. hypophthalmus (12.4±0.8 g) was around 101.9±5.7 mm3 
(with 8.38±0.37 mm3 g-1) in which secondary lamella accounted for 12.6±1.23% (1.6±0.28% 
pillar cells, 4.9±0.78% blood spaces, and 6.1±0.46% epithelium). The volume of gill filaments 
ranged around 7.3 11 mm3 g-1 fish between the groups of fish exposed to temperature and/or 
hypoxia. This parameter was neither significantly affected by ambient hypoxic water nor the 
interaction between temperature and hypoxia factors, although being affected by temperature 
(P<0.01) (Table 3.2). The volume of secondary lamellae which exposed to water of the fish at 
27ᵒC groups decreased significantly after 5 weeks, to less than 1% of filament volume, and 
similar in the recovery week (without exposure to hypoxia); whereas the lamella volume was 
high in the high temperature group, especially with added hypoxia. At the same time, the total 
blood space volume of secondary lamella in this group (9.59±1.3%) significantly increased by 
a factor of 10 as compared to the other groups, and of 2 as compared to the beginning group; 
however, the epithelium volume decreased twice, with 3.59±0.41% compared to 6.1±0.46% of 
the beginning group. The same pattern was found for the fish exposed long-term to the 
conditions (after 16 weeks), which volume of lamella decreased sharply at 27ᵒC opposite to 
those of the high temperature groups (Table 3.1). During the experiments, parasites were found 
on gills of several fish, however, they did not cause significant effects on any of gill parameters 
observed (P>0.05) (Table 3.2).
 59 
Table 3.1 Lamellae respiratory surface area (mm2 g-1), harmonic mean water-blood barrier thickness (µm), volumes of gill filaments (mm3 g-1), the 
percentages of secondary lamellar volume and lamellae structural component volumes (pillar cells, blood spaces, epithelium) of Pangasianodon 
hypophthalmus exposed to elevated temperature and/or hypoxia. Data are presented as mean ± SEM. 
Experiments Treatments Mass (g) 
Lamella 
Surface 
area/mass 
(mm2 g-1) 
Harmonic 
mean 
water-
blood 
barrier 
thickness, 
τh (µm) 
CV 
Gill 
Filament 
volume/mass 
(mm3 g-1) 
Mean 
CE 
(%) 
% Volume 
Pillar cells 
Mean 
CE 
(%) 
% Volume 
Blood spaces 
Mean 
CE 
(%) 
% Volume 
Epithelium 
Mean 
CE 
(%) 
% Absolute 
volume 
secondary 
lamellae 
exposed to 
water 
Parasites 
volume/mass 
(mm3 g-1) 
No. fish 
infected/
Total 
I 
(0 week) 
Beginning (n=12) 12.4 ± 0.8 212 ± 22 1.67 ± 0.12 0.26 8.4 ± 0.4 0.09 1.6 ± 0.28 0.44 4.9 ± 0.78 0.12 6.1 ± 0.46 0.07 12.6 ± 1.23 0 0/12 
I 
(5 weeks) 
27ᵒC & 92% (n=5) 24.6 ± 2.8 12.9 ± 8.0 - - 11.3 ±1.2 0.03 0.09 ± 0.05 6.8 0.26 ± 0.16 2.1 0.62 ± 0.3 1.56 0.98 ± 0.49 0.15 ± 0.01 5/5 
27ᵒC & 35% (n=5) 25.6 ± 3.6 4.9 ± 2.3 - - 10.1 ± 0.6 0.03 0.04 ± 0.02 43.2 0.16 ± 0.05 7.0 0.34 ± 0.14 5.3 0.53 ± 0.18 0.12 ± 0.02 4/5 
33ᵒC & 92% (n=5) 44.3 ± 5.6 39.4 ± 3.5 1.64 ± 0.16 0.22 8.6 ± 0.5 0.03 0.43 ± 0.07 0.46 1.27 ± 0.23 0.14 1.74 ± 0.33 0.08 3.44 ± 0.6 0.1 ± 0.02 4/5 
33ᵒC & 35% (n=5) 47.9 ± 1.5 208 ± 29 1.04 ± 0.18 0.08 7.5 ± 0.4 0.02 1.5 ± 0.19 0.07 9.59 ± 1.3 0.01 3.59 ± 0.41 0.02 14.67 ± 1.81 0.07 ± 0.01 4/5 
I 
(Recovery 
1 week) 
27ᵒC & 92% (n=5) 28.4 ± 2.4 4.3 ± 0.7 - - 10.1 ± 0.3 0.03 - - - - - - - 0.08 ± 0.02 3/5 
27ᵒC & 35% (n=5) 25.9 ± 2.9 2.9 ± 1.2 - - 10.1 ± 0.9 0.03 - - - - - - - 0.1 ± 0.02 4/5 
33ᵒC & 92% (n=5) 52.5 ± 5.7 53.5 ± 11.1 1.21 ± 0.07 0.14 8.3 ± 0.7 0.02 0.42 ± 0.1 0.46 2.47 ± 0.8 0.06 1.58 ± 0.33 0.07 4.46 ± 1.22 0.1 ± 0.02 4/5 
33ᵒC & 35% (n=5) 54.1 ± 7.6 144 ± 18 0.86 ± 0.12 0.11 7.4 ± 0.4 0.02 0.82 ± 0.21 0.23 6.3 ± 0.8 0.02 2.51 ± 0.18 0.04 9.63 ± 1.12 0.05 ± 0.01 3/5 
II 
(16 weeks) 
27ᵒC & 92% (n=5) 27.6 ± 2.8 20 ± 4 1.17 ± 0.25 0.42 7.8 ± 0.8 0.04 0.13 ± 0.04 8.3 0.58 ± 0.12 0.56 1.36 ± 0.33 0.26 2.06 ± 0.46 0.02 1/5 
27ᵒC & 35% (n=4) 22.4 ± 2 19.2 ± 16.4 - - 7.3 ± 1.4 0.06 - - - - - - - 0 0/4 
33ᵒC & 92% (n=5) 204 ± 36 133 ± 25 1.13 ± 0.12 0.23 7.3 ± 0.4 0.02 0.95 ± 0.27 0.54 6.01 ± 1.6 0.07 2.36 ± 0.17 0.08 9.32 ± 1.89 0 0/5 
33ᵒC & 35% (n=5) 108 ± 11 160 ± 19 0.98 ± 0.07 0.14 9.4 ± 0.8 0.02 0.87 ± 0.22 0.16 4.77 ± 1.36 0.05 2.55 ± 0.91 0.06 8.19 ± 2.41 0 0/5 
 60 
Table 3.2 Evaluation the significant effects of temperature and/or hypoxia and parasites on growth 
and gill parameters of striped catfish Pangasianodon hypophthalmus 
Experiments Variables 
P-value 
Temperature 
factor 
O2 
saturation 
factor 
Interaction of factors 
(Temperature * 
O2 saturation) 
Parasites 
 (covariate) 
I 
(After 5 weeks exposure) 
Lamellae surface area 0.000 0.000 0.000 0.335 
Filament Volume 0.004 0.127 0.937 0.566 
Lamella Volume 0.000 0.000 0.000 0.357 
Total Pillar cells Volume 0.000 0.002 0.001 0.445 
Total Blood spaces Volume 0.000 0.000 0.000 0.590 
Total epithelium volume 0.000 0.084 0.009 0.193 
I 
(Recovery 1 week) 
Lamellae surface area 0.000 - - 0.158 
Filament Volume 0.003 - - 0.266 
Lamella Volume 0.000 - - 0.172 
Total Pillar cells Volume 0.000 - - 0.191 
Total Blood spaces Volume 0.000 - - 0.130 
Total epithelium volume 0.000 - - 0.158 
II 
(16 weeks) 
Lamellae surface area 0.000 0.540 0.427 0.640 
Filament Volume 0.384 0.358 0.182 0.829 
Lamella Volume 0.000 0.650 0.578 0.613 
Total Pillar cells Volume 0.005 0.859 0.616 0.785 
Total Blood spaces Volume 0.006 0.874 0.999 
0.381 
0.862 
Total epithelium volume 0.260 0.829 0.333 
II 
(Growth experiment) 
Growth performance 0.000 0.000 0.006 - 
 61 
Growth performance 
Over the experimental period, it was found that both temperature and ambient hypoxic 
water caused significant effects on striped catfish growth rate (P<0.001 and P<0.01, 
respectively; Table 3.2). During the first 8 weeks of the experiment, some mortality (13-15%) 
was seen in all four treatments. However, from this time onwards, no further mortality 
observed. The fish in the 33ᵒC groups were clearly more active in taking feed, and the fish in 
these groups grew significantly faster (P<0.05) than the 27ᵒC groups (Fig. 3.3). Thus, at 16 
weeks, fish in 33ᵒC had grown from 2.4±0.1 g to 165±9.8 g and 127±6.4 g in normoxia and 
hypoxia respectively; whereas in corresponding 27ᵒC groups, fish grew to only 26.5±1.7 g and 
17.4±2.1 g. With respect to the 33ᵒC growth groups, the fish in normoxic water significantly 
grew faster than those in hypoxic hater (P<0.05), while there was no significant effect of water 
PO2 at 27ᵒC (P>0.05). 
Fig. 3.3 Growth performance of juvenile striped catfish (Pangasianodon hypophthalmus) reared in 
33ᵒC & normoxia (white square), 33ᵒC & hypoxia (black square), 27ᵒC & normoxia (white circle), 
and 27ᵒC & hypoxia (black circle). Data are presented as mean ± S.E.M. Different letters indicate 
significant difference amongst treatments at each sampling period (P<0.05). 
Time (weeks)
0 2 4 6 8 10 12 14 16
F
is
h
 m
a
s
s
 (
g
)
0
20
40
60
80
100
120
140
160
180
200
a
a
b
c
a
a
b
c
a
a
b
c
a
a
b
c
a
a
b
c
 62 
Specific growth rate (SGR) followed the same general patterns with the additional expected 
fall in time declining from ca. 5% to around 2% in the high temperature groups, less than 2% 
for the low temperature groups (Fig. 3.4). It was evident that temperature had a particularly 
large effect on growth rate during the first 10 weeks with the high temperature normoxic group 
attaining an astounding >4% body weight gain per day, which was twice the daily rate seen in 
the two 27ᵒC groups. 
Fig. 3.4 Periodic specific growth rate (SGR) of juvenile striped catfish (Pangasianodon 
hypophthalmus) reared in 33ᵒC & normoxia (crossed column), 33ᵒC & hypoxia (grey column), 27ᵒC 
& normoxia (open column), and 27ᵒC & hypoxia (black column). Data are presented as mean ± 
S.E.M. 
3.4. Discussion 
Gill morphometry and gill remodelling 
In the majority of fish species, the gills represent the primary organ for fish respiration 
but are also central in ionic and osmotic regulation, acid-base regulation, and in nitrogenous 
excretion (Evans et al., 2005; Diaz et al., 2009). The basic structures of the gills of P. 
hypophthalmus gills follow the generalised pattern in teleosts (Wilson and Laurent, 2002; 
Evans et al., 2005; Fernandes et al., 2012). The overall appearance is of large gill arches with 
Time period (weeks)
0-8 10 12 14 16
S
G
R
 (
%
)
0
1
2
3
4
5
33C & Normoxia 
33C & Hypoxia 
27C & Normoxia 
27C & Hypoxia 
 63 
long primary filaments resembling that of an active water breather such as rainbow trout. 
Although the present experiment was not designed to detect mitochondrial rich cells, visual 
inspection revealed that these ionophores are present on the primary lamellae in P. 
hypophthalmus and where present, within the ICLM, but are absent from secondary lamellae. 
In line with previous suggestions that this species has a large branchial surface area 
(Lefevre et al., 2011a, b; Lefevre et al., 2013; Damsgaard et al., 2015a) we measured a large 
branchial surface area in small individuals (12.4±0.8g) (27ᵒC and normoxia) and in fish grown 
for 5 weeks to 47.9±1.5g at 33ᵒC in hypoxia (35% saturation) to be 212±22 mm2 g-1 and 208±29 
mm2 g-1, respectively (Table 3.1, Fig. 3.1). These values are indeed high compared to those of 
other air-breathing fish species and compare well to active water breathing species (Fig. 3.5 
and refs therein). However, the second part of our hypothesis that predicted a reduction in gill 
SA in hypoxia to reduce the risk of oxygen loss was not supported. Thus in experiment 1, 
groups grown at 27ᵒC grew to a size of 20-30g over the five weeks, and developed ICLM in 
both normoxic and hypoxic treatments with secondary lamellae SA of 4-13 mm2g-1, with no 
significant effect of oxygen saturation. Those grown in 33ᵒC on the other hand grew much 
faster to 44-48g and also attained larger gills reaching 39 mm2g-1 in normoxia but 208 mm2g-1 
in hypoxia (Table 3.1, Fig 3.1). The difference in these two groups being in the degree of 
ICLM proliferation (e.g. Fig 3.2). In the second experiment, which ran for 16 weeks instead of 
6 the same growth stimulation was seen in relation to temperature and also the same trend for 
a strong temperature effect on SA. There was also a trend for higher SA in the 33ᵒC in hypoxia 
compared to in normoxia, but here the differences were not significant. Thus we conclude that 
secondary lamellae SA in P. hypophthalmus respond to increased oxygen demand by 
increasing SA, resulting in the highest SA in hypoxia at the highest temperature, and do not 
find support for reductions in SA in hypoxia. In this respect the SA plasticity we see in P. 
hypophthalmus resembles that of the crucian carp (Sollid et al., 2003, 2005; Sollid and Nilsson, 
2006) as well as a variety of other water breathing fish (Matey et al., 2008; Tzaneva and Perry, 
2010; Tzaneva et al., 2011; McBryan et al., 2016). It should be noted that when the ICLM is 
fully developed in this P. hypophthalmus, the apparent respiratory SA is then small even for 
obligate air-breathing fishes (Fig. 3.5). Nilsson (2007) argued that fish gill remodelling could 
be an ancient mechanism, which could be present in many other teleost species but which 
remains largely unexplored. It has since been shown that environmental changes can also 
induce gill morphology modifications in air-breathing fish. Thus the mangrove killifish 
remodels its gills as a result of air-exposure (Ong et al., 2007) and the dwarf gourami remodels 
 64 
its gills as a result of water acidification (Huang and Lin, 2011). In addition to these SA 
differences, the harmonic mean water – blood diffusion barrier of P. hypophthalmus (1.67±0.12 
µm) was thin compared to these water-breathers but was similar to the facultative air-breather 
Boleophthalmus boddarti (1.43 µm; Niva et al., 1981) (Fig. 3.5). 
Branchial surface area (mm
2
 g
-1
)
0 200 400 600 800 1000 1200 1400
W
a
te
r-
b
lo
o
d
 d
if
fu
si
o
n
 d
is
ta
n
c
e
 (
m
)
0
2
4
6
8
10
12
Dogfish
Brown Bullhead
Climbing perch
Arapaima
Arapaima
Icefish Rainbow trout
Stinging catfish
Tench
Wolf fish
SnakeheadWalking catfish
Striped catfish 
Mud Skipper
Horse Mackerel
Striped catfish 
Striped catfish 
Striped catfish 
Striped catfish 
Tuna
Fig. 3.5 Respiratory gill surface area and harmonic mean barrier thickness of several fish species with 
different life-styles and habitats. Icefish, Chaenocephalus aceratus (Hughes, 1972); Rainbow trout, 
Oncorhynchus mykiss (Hughes, 1972); Tench, Tinca tinca (Hughes, 1972); Brown bullhead, Ameiurus 
nebulosus (Hughes, 1972); Dogfish, Scyliorhinus canicula (Hughes, 1972); Wolf fish, Hoplias 
malabaricus (Fernandez et al., 1994; Sakuragui et al., 2003); Climbing perch, Anabas testudineus 
(Hughes et al., 1973); Arapaima, Arapaima gigas (da Costa et al., 2007; Fernandez et al., 2012); 
Snakehead, Channa punctatus (Hakim et al., 1978); Walking catfish, Clarias mossambicus (Maina 
and Maloiy, 1986); Mud Skipper, Boleophthalmus boddarti (Niva et al., 1981); Stinging catfish, 
Heteropneustes fossilis (Hughes and Munshi, 1973); Horse mackerel, Trachurus trachurus (Hughes, 
1966); Tuna, Katsuwonus pelamis (Muir and Hughes, 1969); Striped catfish, Pangasianodon 
hypophthalmus (Present study). Water-breathing species: opened circles; Facultative air-breathing 
species: semi-filled left circles; Obligate air-breathing species: black circles. Note: Five groups of 
striped catfish with the gills embedded by ILCM were not presented because of the lack of diffusion 
distan