Life on the edge: incubation behaviour and physiological performance of squat lobsters in oxygen-minimum conditions

Abstract

Squat lobsters frequently inhabit low-oxygen waters, facing particular physiological challenges. Pleuroncodes monodon inhabits one of the most extreme oxygen minimum zones in the world’s oceans at low temperatures, but avoids high temperature hypoxic waters. The present study aimed to determine whether the maternally dependent reproductive outcome is compromised under realistic oxygen/temperature conditions (normoxia and 0.7 mg l-1 at 11 and 15°C) and to evaluate some potential metabolic bases. Females incubated for a significantly longer time at low as compared to high temperatures, but reproductive success was only compromised under hypoxic conditions. Brood viability and synchrony were affected by temperature and its interaction with oxygen concentration (especially under hypoxic conditions and 15°C). Non-viable larvae were hatched at hypoxia-15°C, and larvae hatched in hypoxia-11°C did not survive until moulting. Under normoxic conditions, ventilation of the brood mass decreased with advancing embryo development, but remained high or increased under hypoxic conditions, especially at high temperatures. After releasing their broods, females from all treatments had developing oocytes in their ovaries, but the proportion of oocytes in secondary vitellogenesis was larger at 15°C. The diameter of oocytes in secondary vitellogenesis was significantly smaller in hypoxia treatments. Oxygen consumption of ovigerous P. monodon was generally higher at 15°C, especially at normoxia, and their critical point was significantly larger at 15°C. Under hypoxic conditions, ovigerous females compensated their energetic requirements using anaerobic pathways (increase of pyruvate kinase:citrate synthase ratio and lactate). This suggests that this and other species living in hypoxic waters might suffer severe challenges in a warming ocean.

Full text

MARINE ECOLOGY PROGRESS SERIES
Mar Ecol Prog Ser
Vol. 623: 51–70, 2019
https://doi.org/10.3354/meps12984 Published July 30
© Inter-Research 2019 · www.int-res.com*Corresponding author: byannice@gmail.com
Life on the edge: incubation behaviour and
physiological performance of squat lobsters in
oxygen-minimum conditions
María de los Ángeles Gallardo1, 2, 3, Isis Rojas4, 5, Katherina Brokordt3, 5,
Gustavo Lovrich6, Valentina Nuñez7, Kurt Paschke8,9, Martin Thiel2,3,7,
Beatriz Yannicelli2,7,10,*
1Programa de Doctorado en Biología y Ecología Aplicada, Universidad Católica del Norte, Larrondo 1281,
1781421 Coquimbo, Chile
2Millennium Nucleus Ecology and Sustainable Management of Oceanic Island (ESMOI), Larrondo 1281,
1781421 Coquimbo, Chile
3Centro de Estudios Avanzados en Zonas Áridas, Larrondo 1281, 1781421 Coquimbo, Chile
4Programa Cooperativo Doctorado en Acuicultura, Universidad Católica del Norte (UCN), Larrondo 1281,
1781421 Coquimbo, Chile
5Laboratorio de Fisiología y Genética Marina (FIGEMA), Departamento de Acuicultura, Facultad de Ciencias de Mar,
Universidad Católica del Norte, Coquimbo, Chile
6Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET)-Centro Austral de Investigaciones
Científicas (CADIC), Houssay 200, V9410CAB Ushuaia, Tierra del Fuego, Argentina
7Departamento de Biología Marina, Facultad de Ciencias del Mar, Universidad Católica del Norte, Campus Guayacán,
Larrondo 1281, 1781421 Coquimbo, Chile
8Instituto de Acuicultura, Universidad Austral de Chile, Casilla 1327, 5480000 Puerto Montt, Chile
9Centro FONDAP de Investigación en Dinámica de Ecosistemas Marinos de Altas Latitudes (IDEAL), Casilla 1327,
5480000Puerto Montt, Chile
10Centro Universitario Regional Este, Rocha, Universidad de la República, 9, 27000 Rocha, Uruguay
ABSTRACT: Squat lobsters frequently inhabit low-oxygen waters, facing particular physiological
challenges. Pleuroncodes monodon inhabits one of the most extreme oxygen minimum zones in
the world’s oceans at low temperatures, but avoids high temperature hypoxic waters. The present
study aimed to determine whether the maternally dependent reproductive outcome is compro-
mised under realistic oxygen/temperature conditions (normoxia and 0.7 mg l−1 at 11 and 15°C)
and to evaluate some potential metabolic bases. Females incubated for a significantly longer time
at low as compared to high temperatures, but reproductive success was only compromised under
hypoxic conditions. Brood viability and synchrony were affected by temperature and its interac-
tion with oxygen concentration (especially under hypoxic conditions and 15°C). Non-viable larvae
were hatched at hypoxia-15°C, and larvae hatched in hypoxia-11°C did not survive until moulting.
Under normoxic conditions, ventilation of the brood mass decreased with advancing embryo
development, but remained high or increased under hypoxic conditions, especially at high tem-
peratures. After releasing their broods, females from all treatments had developing oocytes in
their ovaries, but the proportion of oocytes in secondary vitellogenesis was larger at 15°C. The
diameter of oocytes in secondary vitellogenesis was significantly smaller in hypoxia treatments.
Oxygen consumption of ovigerous P. monodon was generally higher at 15°C, especially at nor-
moxia, and their critical point was significantly larger at 15°C. Under hypoxic conditions, oviger-
ous females compensated their energetic requirements using anaerobic pathways (increase of
pyruvate kinase:citrate synthase ratio and lactate). This suggests that this and other species living
in hypoxic waters might suffer severe challenges in a warming ocean.
KEY WORDS: Pleuroncodes monodon · Hypoxia · Temperature · Incubation behaviour ·
Physiological performance · Reproductive potential
Resale or republication not permitted without written consent of the publisher
This authors' personal copy may not be publicly or systematically copied or distributed, or posted on the Open Web,
except with written permission of the copyright holder(s). It may be distributed to interested individuals on request.

Mar Ecol Prog Ser 623: 51– 70, 2019
1. INTRODUCTION
The oxygen concentration in ocean waters is nor-
mally above 8 mg O2l−1 (Talley et al. 2011), but in
certain areas of fjords, basins, marginal seas and oxy-
gen minimum zones (OMZs) sensu stricto, oxygen
concentrations can drop down to 1.4 mg O2l−1 and to
completely anoxic conditions. Oxygen depletion rep-
resents a restriction for most animals (Levin 2003,
Vaquer-Sunyer & Duarte 2011, Gilly et al. 2013),
because it might limit oxygen uptake rates, and
therefore, limit the aerobic energy production rates
that allow the organism to meet the basal energy
demands for structural integrity and survival.
Uptake and internal oxygen delivery are depend-
ent on concentration gradients (Willmer et al. 2005),
so organisms that permanently inhabit OMZs are
challenged to enhance energy provision under low
oxygen conditions. A common characteristic of orga -
nisms in OMZs is their capacity to regulate oxygen
uptake rates, so they remain largely independent
from environmental oxygen concentrations, down to
very low levels (Seibel 2011). Ectothermic organisms
that regulate their oxygen uptake rates are known as
oxyregulators (Pörtner & Grieshaber 1993). Different
adaptations contribute to that end (e.g. high ventila-
tion abilities and/or large gill area to organism vol-
ume, among others; Seibel 2011). Nevertheless, as
environmental oxygen drops, there is a concentra-
tion, known as critical oxygen pressure (Pcrit), below
which basal uptake rates are no longer attainable
leading to the onset of anaerobic metabolism (Gries -
haber et al. 1994). In general, while the onset or
upregulation of anaerobic metabolic pathways below
critical oxygen tensions is another common feature
among OMZ organisms, its contribution to meet total
energetic demands is modest, allowing bursts of
activity or temporal exposure to extreme conditions
rather than long-term endurance (Seibel et al. 2018).
Finally, the reversible metabolic depression, that is,
the reduction of total basal metabolic demand, is
another common feature of organisms temporarily
exposed to extreme hypoxia (Seibel & Childress
2013, Seibel et al. 2018).
Oxygen requirements of marine animals increase
with higher temperatures, and hypoxia tolerance of
marine ectotherms narrows when the temperature
increases beyond the optimum of aerobic perform-
ance (Pörtner & Farrell 2008). Therefore, the Pcrit is
variable and depends on both biological conditions
(species, sex, reproductive status and ontogeny) and
environmental conditions (Whiteley & Taylor 2015).
Interactive effects between hypoxia and temperature
influence the reproductive success of marine ecto-
therm invertebrates and consequently their fitness
(Newell & Northcroft 1967, Newell & Branch 1980,
Grieshaber et al. 1994). Negative effects of hypoxia
at high temperatures have been described for game-
togenesis, number and quality of sperm and egg,
reproductive behaviour, duration of development,
hatching (Fernández et al. 2006, Wu 2009) and larval
survival (Yannicelli et al. 2012).
Brood incubation is a reproductive strategy with an
associated energetic cost that might affect female
body condition (Bosch & Slattery 1999, Fernández et
al. 2000). For instance, in subtidal brachyuran crabs,
oxygen consumption of incubating females display-
ing active brood care was significantly higher than
that of non-brooding females, and it also varied
among females carrying embryos at different devel-
opmental stages (Baeza & Fernández 2002). In addi-
tion, maternal expenditure in movements/behav-
iours for oxygen supply to the embryo mass showed a
positive relationship with temperature and the
frequen cy of events associated with parental care
(Brante et al. 2003). For brooding decapod crus-
taceans, abdominal flapping, pereopod probing and
pleopod movements are common behaviours related
to oxygen supply to egg masses (Fernández et al.
2002, Fernández & Brante 2003, Baeza et al. 2016).
Oxygen consumption increases as eggs develop and,
in several crustacean species that inhabit normoxic
and/or temporarily hypoxic environments, they oxy-
conform (decrease oxygen consumption as environ-
mental oxygen decreases). Hypoxia can occur in the
centre of the brood mass, with strong gradients in
normoxic environments (Fernandez & Brante 2003).
Nevertheless, there is a lack of information about
incubation in crustacean species that live in the OMZ
(permanent natural hypoxia) and how temperature/
oxygen interaction affects gametogenesis, embryo
development and metabolic rates, active brood care
and metabolism (aerobic and anaerobic) and general
condition of incubating females.
Some of the most common crustacean species re-
ported from low-oxygen waters around the world are
squat lobsters from the taxa Galatheoidea and Chi-
rostyloidea (Lovrich & Thiel 2011). Amongst them, red
squat lobsters Pleuroncodes monodon (Gala theoidea:
Munididae) from the continental shelf of the Hum-
boldt Current System can inhabit the most extreme
conditions (Gallardo et al. 2004). This species has his-
torically sustained a valuable fishery in central Chile
(30−37° S) (Palma 1994). It is distributed in the eastern
Pacific from Mexico to Chile (Franco-Meléndez 2012)
and has a wide phenotypic plasticity with a pelagic
52
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
(6−25° S) and a benthic adult form (25−37°S) (Haye et
al. 2010). Benthic populations are often associated
with water temperatures from 11− 12°C and oxygen
concentrations below 3.00 and down to 0.16 ml l−1
(Gallardo et al. 2017). Ovigerous females (OFs) of
benthic populations inhabit more oxygenated waters,
and the peak reproductive activity coincides with the
seasonal weakening of the OMZ influence over the
continental shelf (Gallardo et al. 2017). The reproduc-
tive period extends over several months, and females
produce new broods within a few days after larval
hatching under normoxic conditions in laboratory en-
vironments (Thiel et al. 2012). To date, 2 contrasting
hypotheses have been proposed to explain the shal-
lower distribution of benthic OFs during the repro-
ductive season: coastward migration to favour larval
survival after hatching in richer, more oxygenated
and protected areas (Palma & Arana 1997), and the
potential benefit of brood and (pa rallel) gonad devel-
opment under more oxygenated conditions (Gallardo
et al. 2017). Meanwhile, pelagic populations inhabit
mainly waters above the oxycline, at latitudes where
low oxygen water temperatures normally reach 15−
16°C (Gallardo 2017). In cold waters, P. monodon sur-
vives extreme hypoxia for a few hours (Kiko et al.
2015), and the oxygen tension corresponding to the
Pcrit of routine metabolic rates of adults of the close
relative P. planipes doubles as temperatures rise from
11 to 20°C (Quetin & Childress 1976). This suggests
that under high temperatures/hypoxia, process of
high energy-demand could be li mi ted during the ma-
ternal phase. However, this has never been assessed
at temperature/oxygen conditions relevant and real-
istic for P. monodon.
In the present study, we aimed to determine
whether female reproduction is compromised under
any realistic oxygen/temperature condition and to
evaluate some potential metabolic bases. Theoretical
considerations and organism distribution suggest
that hypoxia, especially at high temperatures, could
exert a significant effect (1) on the current reproduc-
tive success of OFs through its effect on embryo
development until hatching and larval viability, and
(2) on future reproduction, through its modulation of
female post-hatching condition and gametogenesis.
Viable populations do not develop at oxygen tensions
that remain permanently below the Pcrit (routine
metabolism), so if our hypothesis is true, only oxygen
uptake by brooding females (but not by males or
non-brooding females) should show a de crease at
natural hypoxia values, and oxygen uptake rate by
the egg mass should be efficient even below that
level.
2. MATERIALS AND METHODS
In order to accomplish our objective, we conducted
both long- and short-term incubation experiments.
We maintained reproductive females throughout the
incubation period in hypoxic/normoxic waters (0.7
and 8.0 mg O2l−1, respectively) at different tempera-
tures (11 and 15°C) (in a fully factorial experimental
design) to describe embryo developmental success
together with brooding behaviour, final condition of
both female and egg mass, gonad development and
metabolic potential of Pleuroncodes monodon fe -
males after larval hatching (long-term incubation
experiment). This experiment served to compare
brooding behaviour, embryo and gonadal develop-
ment as well as final female and embryo physiologi-
cal and proximate composition conditions among
treatments. In a short-term experiment, we evaluated
the rate of oxygen consumption and metabolic poten-
tial of OFs, brood masses, non-reproductive females
and males (each individually) at different oxygen
concentrations and temperatures. The duration of
this experiment depended on the change of slope in
respiration curves and served to determine Pcrit of
each sex/stage in each experimental condition.
2.1. Long-term experiment
2.1.1. Collection and maintenance
Adult specimens of P. monodon were collected in
northern-central Chile at 30° S, by modified Agassiz
trawl at depths ranging from 125 to 150 m. Squat lob-
sters were immediately placed in a plastic container
for transportation to the laboratory at the Oceanolab,
Universidad Católica del Norte, Coquimbo, Chile.
Adults were collected in September 2015 for the
treatments at 11°C and in January 2016 for the treat-
ments at 15°C. In order to avoid problems associated
with previous breeding/environmental influence on
initial female condition, non-brooding females were
acclimated (high oxygen concentrations, at 13−15°C,
and food ad libitum for 15−30 d). Males were then
introduced into a common female aquarium (3 m3) to
mate, and approximately 48 h later, once the brood
had been extruded, 32 healthy females with all their
appendages and of similar size range (Table 1) were
placed in individual containers (20 × 20 × 20 cm; 8 l)
to begin long-term incubations. Experimental condi-
tions were (Table 1) hypoxia (H: 0.7 ± 0.1 mg O2l−1)
and normoxia (N: 8.0 ± 1.0 mg O2l−1) at 2 near con-
stant temperatures (11 ± 0.5°C and 15 ± 0.8°C). Here-
53
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
after, treatments will be referred as H11, H15, N11
and N15.
2.1.2. Experimental design and set-up
Carapace length (CL, mm) and initial mass (g) were
measured just before the animals were placed in indi-
vidual aquaria under controlled conditions. Females
were maintained in the experiment until the larvae
hatched, the brood was lost, or the female died. Near
constant conditions were kept in the sealed individual
aquaria using a flow-through system. Seawater was
filtered (<0.5 µm and UV-sterilized) and brought to
stock tanks in the controlled temperature room where
it acquired the experimental temperature. Normoxic
water was gently flushed to the individual aquaria di-
rectly from the stock tanks. Hypoxic water was pre-
pared at 0.7 mg O2l−1 with a range between 0.5 and
1mg O
2l−1,which was achieved by bubbling 8 hol -
ding tanks (50 l each) with a carbon dioxide−nitrogen
mix (0.5−999 ppm) manufactured by INDURA (www.
indura.cl). This mixture was used to maintain the pH
similar between oxygen treatments and avoid possible
bias of contrasting pH values in the experimental
treatments. The final mix was estimated based on pre-
liminary trials and on the literature (Torres et al. 2013).
Once the desired oxygen concentration was reached,
water from the hypoxic tanks was isolated from the
surrounding air with a floating lid. Water was injected
into each aquarium with a pump system at a flow rate
of 0.0512 l min−1 for 15°C and 0.0466 l min−1 for 11°C.
Eight OFs maintained in individual aquaria were used
for each treatment (H11, N11, H15 and N15). OFs
were fed every second day with a paste prepared with
flake food for fishes and water; during each feeding,
each female received a portion of about 0.125 g (wet
mass). Aquarium water change and removal of food
leftovers was done the day after feeding.
Each aquarium had a chemical optical (non-
invasive) oxygen sensor spot, specific for each oxy-
gen treatment. The O2sensor used for the hypoxia
treatment was a PSt6 sensor (Pre-
Sens) with a measurement range
between 0 and 1.8 mg O2l−1
(±0.010 µmol), and the one used for
the normoxia treatment was a PSt3
sensor (PreSens) with a measurement
range between 0 and 45 mg O2l−1
(±0.14 µmol). Oxygen concentration
and temperature in each aquarium
were monitored using a Fibox 4 (Pre-
Sens) every 2 h (between 13:00 and
21:00 h). Fibox 4 uses a fibre optic compatible with
the oxygen sensor spot (Fig. 1).
2.1.3. Embryo development time and index of
brooding success
Embryo samples (15−25 embryos) were taken ran-
domly from every brood every 4 d to determine the
development stages of the embryos following Palma
(1994), who defined 5 embryo stages, although in this
study, stage V (pre-zoea) was not considered. Stage I
is characterized by yolk evenly distributed through-
out the embryo. In stage II, cell differentiation is ini-
tiated and the yolk is yellowish and grainy. Stage III
is defined by the appearance and pigmentation of the
ocular structures. In stage IV, the eyes appear com-
pletely pigmented and positioned obliquely in the
anterior region of the embryo with developed struc-
tures. Stage IV is also characterized by the presence
of red chromatophores in the dorsal portion of the
embryo. We also identified a non-developing embryo
category, which is an egg that had arrested develop-
ment, and appeared completely white (Fig. S1 in the
Supplement at www.int-res.com/articles/ suppl/ m623
p051 _ supp. pdf).
The total time of development was estimated as that
from the beginning of incubation until we observed
the first hatched larvae. Assumptions of normal distri-
bution and homogeneity of variances were checked
using QQ plots and Levene’s test (library ‘car’ 2.1-6;
https:// r-forge. r-project. org/ projects/ car/ v2.1-6), re-
spectively. A 2-way ANOVA was applied to evaluate
the effect of temperature and oxygen conditions, and
their interaction, on the total time of embryo develop-
ment; statistical significance was set at α= 0.05. Fe-
males that died during the experiment were excluded
from this statistical analysis.
In order to determine the brooding success of incu-
bating females, we registered 12 observations and
tabulated 0/1 binary variables (Table 2) for each fe-
male. For the assigned binary value, 1 was considered
54
Treatment n CL (mm) Temp. (°C) Oxygen conc.
Initial Final (mg O2l−1)
H11 8 8 27.9 ± 3.2 11.1 ± 0.4 0.7 ± 0.4
N11 8 7 29.7 ± 2.5 10.8 ± 0.3 7.9 ± 1.7
H15 8 4 34.3 ± 1.8 15.4 ± 0.9 0.8 ± 0.6
N15 8 8 34.4 ± 1.1 15.2 ± 0.9 6.8 ± 1.5
Table 1. Number of initial and final replicates. Mean ± SD size (CL: carapace
length) of ovigerous female squat lobsters by treatment, and temperature and
oxygen conditions in the long-term experiment. H: hypoxia; N: Normoxia
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
as favourable and 0 as not favourable to reproductive
success. All females were included to construct an in-
dex of brooding success (Table S1 in the Supplement),
and females that died before larval hatching received
a value of 0 for the related variables. A generalized
linear mixed model (GLMM) test was done to
estimate reproductive success for each treatment. Af-
ter backward elimination, the best model was chosen
using Akaike’s information criterion (AIC). The ex-
plained variance (D2or pseudo R2) was calculated fol-
lowing the formula proposed by Zuur et al. (2009).
This analysis was done using R statistical software (R
Core Team 2015).
2.1.4. Behavioural analysis of brooding females
In order to determine the behaviour of incubating
fe males, these were video-recorded every 4 d for
approximately 30 min using a GoPro Hero video
55
Fig. 1. Set up for the long-term experiment to describe embryo developmental success together with brooding behaviour, final
condition of both female and egg mass, gonad development and metabolic potential of Pleuroncodes monodon females after
larval hatching
Criterion Favourable (1) Not favourable (0)
Non-developing embryos in the embryo mass at any time of development Absence Presence
Total number of developmental stages present at the last embryo If 1 or 2 stages If ≥3 stages
mass observation
Most frequent stage in embryo mass prior to (2 or 3 d before) If most frequent If most frequent
larval release is >stage III is <stage III
Presence or absence of infections in embryo mass at any time Absence Presence
Percentage of embryo stages I−III in embryo mass at the observation prior <50% >50%
to (2 or 3 d before) larval release
Most frequent stage in embryo mass at larval release Otherwise If most frequent is <IV
Presence or absence of incompletely hatched larvae in embryo mass Absence Presence
Presence or absence of swimming larvae Presence Absence
Presence or absence of dead larvae on the second day after hatching Absence Presence
Presence or absence of dead larvae before the first moult Absence Presence
Female survival before larval release Alive Dead
Female survival after hatching Alive Dead
Table 2. Criteria used to build the index of brooding success in squat lobsters. The responses of each criterion are binary:
favourable (1) and not favourable (0)
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
camera placed in front of each aquarium. Videos
were taken only at night to avoid disturbance in the
laboratory. Based on preliminary observations, 3
common behaviours were identified: (1) pleopod
movements, (2) abdominal flapping and (3) pereopod
probing (Fig. S2 in the Supplement). These behav-
iours were previously described for brachyuran crabs
(Baeza & Fernández 2002) and the Caribbean spiny
lobster (Baeza et al. 2016).
‘Pleopod movements’ involve moving the embryo
mass up and down or side to side using the pleopods.
These movements are visible as individual move-
ments of the entire embryo mass, and so each coordi-
nated beat of the pleopods corresponds to 1 pleopod
movement, i.e. 1 event. In order to analyse pleopod
movements, each 30 min recording was divided into
non-effective time and effective time of video (ETV).
The ETV was the time that allowed for the adequate
observation of this female behaviour (i.e. female fac-
ing the camera). The number of pleopod movements
min−1 of ETV was calculated on videos with ETVs
longer than 10 min (33 videos out of 107).
‘Abdominal flapping’ involves moving the ab -
domen up and down, either entirely or partially,
when females are standing on their pereopods, ele-
vated above the substratum. Each abdominal flap,
which can occur once or several times in sequence, is
counted individually as an event.
In the state ‘pereopod probing’, females use the 5th
pair of pereopods to probe and slightly shake the
embryo mass. This state begins when the female
starts to touch the embryo mass with its 5th pair of
pereopods, and ends when the female pulls the pere-
opods back out of the embryo mass. In this case, the
duration of pereopod probing was measured over
each observation period. Abdominal flaps and pereo-
pod probing were visible throughout the observation
period, but pleopod movements were only visible
when the females were facing the camera. To ana-
lyse abdominal flapping, the number of events min−1
of video was counted. In case of pereopod probing,
the duration of the state was quantified as the per-
centage of time during the 30 min of observation.
Behaviour was evaluated throughout embryo mass
development (Table S2 in the Supplement). As devel-
opment time changed with temperature and oxygen
concentration, comparisons among treatments must
rely on comparison of similar developmental stages of
the brood mass. Stage I corresponds to the first obser-
vation (Day 4) of incubating females, stage II and
stage III were considered when > 50 % of the brood
mass was in stage II or III, respectively, and stage IV
when >50% of the embryo mass was in stage IV or V.
Statistical analyses of incubating behaviour were
made excluding dead females: 2 OFs from treatment
H15 and 1 from N11. In order to evaluate the effects
of extrinsic (oxygen and temperature) and intrinsic
(embryo developmental stage) factors, and their
interaction, on the presence/absence of a specific in -
cu bating behaviour at any time of each video record-
ing (pleopod movements, abdominal flapping, pereo-
pod probing) among OFs, a GLMM analysis with a
binomial link function was conducted. All variables
(extrinsic and intrinsic) were integrated in an addi-
tive and additive-multiplicative model to determine
the significance level for each one; AIC and D2were
also calculated.
The frequency of the pleopod movements, and
their relationship with experimental factors (categor-
ical variables), was analysed using a GLMM with
gamma distribution link function. Models with addi-
tive and additive-multiplicative terms were tested to
define the best model, and the explained variance
was calculated. The frequency of abdominal flaps,
and their relationship with experimental factors, was
analysed using a general linear model (GLM), where
the number of flaps was a continuous variable
(gamma distribution link function). After backward
elimination, the best model was chosen using the
stepwise AIC, and the D2was calculated.
The percentage of pereopod probing (total time
of state in 30 min of observation) was analysed fol-
lowing the suggestions proposed by Zar (2010) for
populations with distributions that strongly differ
from normal and with different distributions and
variances, which consists of reporting the mean
and variance for each treatment. All analyses were
done using the R statistical software (R Core Team
2015).
2.1.5. Biochemical analyses
Biochemical traits of females were measured on
the abdominal muscle at the end of the incubation
period. They consisted of the determination of proxi-
mate composition (total proteins, carbohydrates and
lipids), apparent specific activity of the key enzymes
citrate synthase (CS) and pyruvate kinase (PK),
respectively associated with aerobic and anaerobic
metabolism, and the metabolic end product lactate.
Additionally, as an indicator of potential cell stress,
levels of the 70 kD heat-shock protein (HSP70) were
determined in female gills.
To quantify total carbohydrates and lipids, female
abdominal muscles were dried to a constant mass at
56
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
60°C. Dry tissues were then pulverized in a mortar
and homogenized with deionized water at a propor-
tion (w/v) of 1:1 for carbohydrates and 4:1 for lipids
and proteins. The phenol-sulphuric acid method
described by Dubois et al. (1956) was used for total
carbohydrate determinations. Its concentration was
determined in a spectrophotometer (Variant Cary
UV) at 490 nm using a solution of glycogen in deion-
ized water at a concentration of 50 µg ml−1 as a stan-
dard. The procedure used by Mann & Gallager
(1985) was followed for lipid determinations. A dou-
ble extraction with chloroform-methanol and further
purification with NaCl was performed. The sample
was read in a spectrophotometer at 520 nm using a
solution of cholesterol in chloroform-methanol (1:2)
at a concentration of 0.8 mg ml−1 as a standard.
Total protein was quantified in 0.03 g (wet weight)
of abdominal muscle and gill (for HSP70 determina-
tions) samples, following Brokordt et al. (2015). Tis-
sues were homogenized in 150 µl of homogenization
buffer (32 mM Tris-HCl at pH 7.5, 2% SDS, 1 mM
EDTA, 1 mM Pefabloc and 1 mM protease inhibitor
cocktail; Sigma). The homogenate was incubated for
5 min at 100°C, then re-suspended in 100 µl of
homogenization buffer and re-incubated at 100°C for
5 min. The homogenate was centrifuged at 10 600 × g
for 20 min. Total protein was quantified in an aliquot
of the supernatant with a Micro-BCA kit using a
microplate spectrophotometer EPOCH (BioTek).
HSP70 was measured in the gill tissue of females by
ELISA following Brokordt et al. (2015), and validated
in previous tests by comparing ELISA results with im-
munoprobing of Western blots. Western-blot analyses
(using the same antibodies described later) showed
only 1 band at the level of HSP70. Total protein (30 µg
ml−1) was diluted in 0.05 M carbonate-bicarbonate
buffer at pH 9.6, and 50 µl of sample per well were in-
cubated in an ELISA plate overnight at 4°C with 3
blanks (containing buffer only) and various concen-
trations of cognate HSP70 (H8285, Sigma) to generate
a standard curve. The plate was washed twice with
phosphate-buffered saline (PBS; 200 µl well−1). Next,
200 µl of blocking buffer (PBS + 5% skim milk) were
added to each well and incubated for 2 h. The wells
were washed again with PBS. Subsequently, 100 µl of
the primary antibody (polyclonal mono-specific anti-
epitope) that recognizes the inducible and constitutive
forms of HSP70, developed in immunized mice with a
synthetic peptide epitope (Group of Immunological
Markers on Aqua tic Organisms, Catholic University
of Valparaiso), diluted 1:400 in blocking buffer +
0.05% Tween-20, were added to each well, and the
plate was incubated overnight at 4°C. The plate was
then washed 4 times with PBS, incubated with goat
anti-mouse IgG (Thermo Scientific) secondary anti-
body, diluted in blocking buffer + 0.05% Tween-20
for 2 h at 25°C and washed again 4 times with PBS.
Next, 100 µl of substrate solution (10 mg o-phenylene-
diamine dihydrochloride in 25 ml of 0.05 M citrate
phosphate buffer) were added, followed by incubation
of the plate for 30 min at 25°C. Finally, the plate was
read at 450 nm in a microplate spectrophotometer
(EPOCH, BioTek). The absorbance of the sample was
corrected by the mean absorbance of the blanks and
divided by a conversion factor that was estimated
from a linear regression curve of cognate HSP70. The
calculated result was the concentration of HSP70 in
µg mg−1 total protein.
The PK:CS activity ratio per female was calculated
in order to evaluate the relative predominance of aer-
obic or anaerobic pathways, considering that CS rep-
resents the potential for aerobic metabolic pathways,
while PK activity represents the potential for fermen-
tative pathways. For CS and PK determination, sam-
ples of abdominal muscle were homogenized on ice
in 10 volumes of homogenizing buffer (50 mM imi -
dazole-HCl, 2 mM EDTA-Na2, 5 mM EGTA, 1 mM
di thio threitol, 0.1% Triton X-100, pH 6.6 or 7.2,
respectively for PK and CS). The homogenates were
cen trifuged at 4°C for 15 min at 600 × g. Conditions
for enzyme assays were adapted from those used by
Brokordt et al. (2000) as follows (all concentrations in
mM) for CS: Tris-HCl 75, oxaloacetate 0.3 (omitted
for the control), DTNB 0.1, acetyl CoA 0.2, pH 8.0;
and for PK: imidazole−HCl 50, MgSO413, KCl 100,
phosphoenolpyruvate 5 (omitted for the con trol),
ADP 5, NADH 0.2, excess lactate dehydrogenase, pH
6.6. Enzyme activities were measured at controlled
room temperature (20°C) using a micro plate spec-
trophotometer (EPOCH, BioTek) to follow the ab sor -
bance changes at 412 nm to detect the transfer of sul-
phydryl groups from CoASH to DTNB for CS, and
that of NAD(P)H at 340 nm for PK. The molar extinc-
tion coefficients used for DTNB and NAD(P)H were
13.6 and 6.22, respectively. All assays were run in
duplicate, and the specific activities are expressed in
international units (IU, µmol of substrate converted to
product per min) per gram of abdominal muscle
mass. Finally, quantitative determination of lactate in
samples of abdominal muscle was achieved with an
enzymatic colorimetric kit (Spinreact) following the
manufacturer’s instructions.
Statistical analyses of all biochemical traits were
made only for females that completed their brooding
period until hatching. QQ plots and Levene’s test
(library ‘car’ 2.1-6) were used to check for normal
57
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
distribution and homogeneity of variances. A 2-way
ANOVA was done in order to determine if oxygen
and temperature conditions, or their interactions,
explained the variance of proteins, lipids, carbohy-
drates, PK, CS, PK:CS ratio and lactate in abdominal
muscle, and HSP70 levels in gills of incubating
females. Tukey’s HSD test was used in each case to
find specific differences among treatments.
2.1.6. Ovary development
Differences in gonad development between fe -
males reared in the different treatments were evalu-
ated by comparing the relative frequency of oocytes
in primary and secondary vitellogenesis in each
gonad, and also by comparing the turning size of
oocytes from primary to secondary vitellogenesis.
After larval hatching, female ovaries were fixed in
Davidson’s solution for 24 h and transferred to 70%
ethanol, dehydrated and stained with haematoxylin-
eosin. A 5 µm slice of each gonad (through its centre)
was prepared for optical microscope analysis. First,
oocytes were identified as being in primary or sec-
ondary vitellogenesis following the descriptions of
Kronenberger et al. (2004) (for Galathea intermedia)
and Moreno et al. (2012), where oocytes in secondary
vitellogenesis differ from those in primary vitellogen-
esis by the presence of granular yolk (Kronenberger
et al. 2004) (Fig. S3 in the Supplement). The number
of oocytes in each stage was counted across the his-
tological preparation. The diameters of oocytes that
presented visible nuclei were then measured to the
nearest 0.01 µm at 50× magnification, after categoriz-
ing them into primary or secondary vitellogenesis. A
G-test was conducted to compare if the proportion of
oocytes in primary or secondary vitellogenesis was
homogeneous between treatments (Sokal & Rohlf
1995), and afterwards a pairwise G-test was applied
using the R statistical software (R Core Team 2015).
As a measure of oocyte development between treat-
ments, we compared the size at which 50 and 95% of
oocytes had reached the secondary vitellogenesis
stage. First, the size frequency distribution of oocytes
in primary and secondary vitellogenesis was built for
each treatment. Following a multi-model comparison
ap proach, the minimal size of secondary oocytes was
estimated as the size at which 50 and 95% of ob -
served oocytes had developed the granulated yolk.
Hypoxia vs. normoxia treatments were compared for
each temperature independently. The difference in
the AIC (Zuur et al. 2009) resulting from adjusting a
single binomial model to all data at a temperature
(with 2 parameters) vs. 2 models at each temperature
(1 for each oxygen level, 4 parameters total) was used
to choose the best model (Johnson & Omland 2004).
2.2. Short-term experiment
Short-term incubations were conducted to evaluate
whether the oxygen consumption rates of adults in dif-
ferent reproductive conditions responded similarly to
environmental conditions. Ovigerous females (OFs,
carrying their brood), females after brood removal
(FABRs), embryo mass (or brood), non-reproductive fe-
males (or non-OFs) and males were tested. Organisms
were collected in October 2016 and maintained as in
the long-term experiment. Nine individuals of each
adult type and 9 embryo masses were evaluated under
each condition by placing them in individual closed
chambers (8 l aquaria). Before the experiment, adults
were placed in individual aquaria at the experimental
temperature in the controlled temperature room and
starved for 24 h. Since in closed respirometry, the time
to reach hypoxia might influence the rate of oxygen
consumption even in starved organisms, 2 starting oxy-
gen conditions were considered for the 24 h acclimation
and experimental conditions: 8.0 and 2.5 mg O2l−1. Hy-
poxic tensions were accomplished as described for the
long-term experiment. During the acclimation period,
water flow was kept through the aquaria with the cor-
responding initial temperature and oxygen conditions.
Measurements started when the flow was shut down
and the first oxygen determination was made. Non-in-
trusive mea surements of oxygen concentration in the
sealed aquaria were conducted with Fibox 4 (PreSens)
every hour until oxygen concentrations of 3.0 mg O2l−1
(for normoxia treatment) and 0.05 mg O2l−1 (for
hypoxia treatment) were reached in the chambers.
Oxygen sensors and spots for the hypoxia/nor-
moxia treatments were the same as in the long-term
experiment. Wet masses of individuals or broods
were measured after the experiment. The embryo
masses were removed from the OFs by using a deli-
cate paintbrush, and placed inside 100 ml Winkler
bottles. Measurements were done with non-invasive
respirometry using spots previously calibrated on
each bottle without a period of acclimation, since the
Winkler bottles were filled with the same water the
embryos had been exposed to when attached to their
mothers. After brood removal, females were subject
to the same procedure of acclimation and measure-
ments as they had previously experienced as OFs.
Data were analysed independently for each stage/
sex. Two control chambers with only water with
58
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
identical initial conditions as those of the experimen-
tal treatments were run in parallel with each stage/
sex trial.
Respiration rates were expressed as mg O2h−1
ind.−1 and mg O2h−1 g−1 (wet weight). The slope of
the oxygen concentration over time was used for the
estimation and referred to the mean oxygen concen-
tration over the period of constant consumption (lin-
ear decrease). The average consumption from the
control chamber was subtracted. In all cases, con-
sumption was fairly constant except at very low oxy-
gen levels. Since mass-specific consumption rate is
dependent on individual mass, an ANCOVA was
applied on adult rates to evaluate the effect of oxy-
gen concentration (continuous variable) and temper-
ature (categorical variable) with individual mass as a
co-variable. Previously, the significance of individual
mass as the explanatory variable was checked as
well as the differences between treatment-mass
slopes with a homogeneity of slopes models. In the
case of significant interactions between the co-vari-
able with any other factor, a separate slope model
was conducted instead of an ANCOVA. To evaluate
brood consumption rate dependency on oxygen con-
centration (continuous variable) and temperature
(categorical variable), a GLM was applied.
For OFs (no change in slope was observed for the
other experimental groups), the Pcrit for each tempera-
ture was estimated by adjusting a polynomial regres-
sion to oxygen consumption as a function of oxygen
concentration. Pcrit corresponds to a change of slope
in the response variable as a function of the independ-
ent variable. The regression was fitted with library
‘ggplot2’ (version 2.2.1; http:// cran. r-project. org/
web/packages/ ggplot2/), function ‘geom_smooth’ and
method ‘LOESS’ (local polynomial re-
gression fitting). This allowed us to
determine the breakpoint (Pcrit), rep-
resented by a change of slope in the
response variable as a function of the
independent variable.
Further biochemical determina-
tions of CS, PK and lactate were con-
ducted for muscle tissue of FABRs,
OFs, non-reproductive females and
males and HSP70 in FABR gills. Tis-
sues were frozen at −80°C immedi-
ately after the experiment. The enzy-
matic analyses (CS, PK and PK:CS
ratio) and lactate concentration were
performed following the same me tho -
dology described for the long-term
experiment.
3. RESULTS
3.1. Long-term experiment: reproductive traits
3.1.1. Embryo development time and index of
brooding success
Embryo development time was only affected sig-
nificantly by temperature (p < 0.0001, Table 3).
Development time was longer at 11°C (mean ± SD =
41.9 ± 3.2 d) than at 15°C (mean = 24.4 ± 2.8 d)
(Fig. 2A) regardless of oxygen conditions. The index
of brooding success was also significantly affected
59
Embryo development time df MS Fp
Temperature 1 2074.7 247.826 < 0.0001
Oxygen 1 0.9 0.102 0.753
Oxygen × Temperature 1 28.6 3.417 0.077
Index of brooding success Estimate SE Zp
(Best model: AIC = 49.5, D2= 0.54, residual deviance = 39.5, df = 27)
Model 1.77071 0.14586 12.139 <0.0001
Oxygen −0.06596 0.20977 −0.314 0.753
Temperature −121109 0.30448 −3.978 < 0.0001
Oxygen × Temperature 0.95326 0.38058 2.505 0.012
Table 3. Results of the 2-way ANOVA and generalized linear model conduc -
ted to test the effect of temperature and oxygen on squat lobster embryo
development time and index of brooding success, respectively. Significant
values (p < 0.05) are highlighted in bold
Fig. 2. Mean and SD of (A) embryo development time and
(B) index of brooding success of Pleuroncodes monodon
under different conditions of oxygen and temperature.
Index of brooding success is a composite variable based on
characteristics describing brood development: the higher
the index, the better embryo development (for details, see
Table 2 and Table S1). Different letters indicate significant
differences at p < 0.05
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
by temperature and decreased from 11 to 15°C
(Table 3). A significant interaction between tempera-
ture and oxygen conditions was also evident; the
index was significantly lower in hypoxia than in nor-
moxia at 15°C but not at 11°C (Table 3, significant
interaction term; Fig. 2B). The presence of undevel-
oped embryos in the brood mass was more frequent
at high temperature, especially under hypoxic condi-
tions (Table S1). Asynchrony of brood mass develop-
ment as indicated by the presence of early stages at,
or immediately before, the beginning of hatching,
was more frequent at 15 than at 11°C. Three or more
developmental stages were more frequently ob ser -
ved at the time of hatching in brood masses in
hypoxia at high temperature. The larvae of the
hypoxia treatments were not viable: at 11°C, larvae
did not reach the first moult because they died within
a few days after hatching, while at 15°C, hatching
was incomplete and swimming larvae were absent
(Table S1).
3.1.2. Incubating behaviour
The number of females that dis-
played pleopod movements (Ta ble S2)
depended on the developmental stage
of embryos and its interaction with
temperature. At 15°C, the number of
females that bore eggs in stage III and
performed the pleopod movements
was significantly higher in hypoxic
conditions (Table 4, Table S2). This
behaviour was sometimes displayed
by females simultaneously with pe -
reopod probing. In females with
advanced developmental stages of
embryos, the frequency of pleopod
movements was higher in hypoxia
than in normoxia at both temperatures
(p = 0.037 for stage IV; Table 5, Fig. 3).
Abdominal flapping behaviour oc -
curred among females independently
from the studied factors (Table 4).
Nevertheless, for females that dis-
played this type of behaviour, abdom-
inal flaps were more frequent at 15°C
than at 11°C. At 11°C, the frequency
of abdominal flapping was constant
throughout incubation and independ-
ent of oxygen concentration. Abdom-
inal flapping became less frequent
with ad vancing embryo development
in N15, while in hypoxia it remained constant. In
females with stage IV embryos, abdominal flapping
per minute was an order of magnitude higher in H15
than in any other condition (Table 5, Fig. 3).
The occurrence of pereopod probing decreased as
egg mass developed (Fig. 3). However, due to the
interaction between developmental stage and tem-
perature, at advanced stages (especially stage III),
pereopod probing was more frequent among females
at 15°C than among those at 11°C (Table 4). In the
early stages of incubation, there was no particular
pattern in the duration of this activity among OFs.
3.1.3. Biochemical analyses
Proximate composition of female muscle as well as
embryos were largely unaffected by the experimen-
tal conditions. Nevertheless, carbohydrate content in
60
Estimate SE Zp
Pleopod movement
(Best model: AIC = 111, D2= 0.13, residual deviance = 93.2, df = 97)
Model 26.992 10.898 2.477 0.013
Stage II −20.517 11.318 −1.813 0.069
Stage III −28.730 11.963 −2.401 0.016
Stage IV −16.637 11.806 −1.409 0.159
Temperature 15°C −0.1352 13.755 −0.098 0.9227
Stage II × Temperature 15°C 26.072 18.323 1.423 0.153
Stage III × Temperature 15°C 37.941 18.685 2.030 0.042
Stage IV × Temperature 15°C 22.164 18.605 1.191 0.234
Abdominal flapping
(Best model: AIC = 150, D2= 0.27, residual deviance = 136, df = 99)
Model 0.7049 0.5354 1.316 0.188
Oxygen N −0.1052 0.4687 −0.224 0.822
Stage II −0.6586 0.6037 −1.091 0.275
Stage III −0.9848 0.5926 −1.662 0.097
Stage IV −10.512 0.6301 −1.668 0.095
Temperature 15°C 0.8466 0.4780 1.771 0.077
Pereopod probing
(Best model: AIC = 13092, D2= 0.42, residual deviance = 925, df = 97)
Model 2.115 1.167 1.813 0.069
Stage II −2.804 1.251 −2.240 0.025
Stage III −3.433 1.356 −2.531 0.011
Stage IV −4.276 1.644 −2.600 0.009
Temperature 15°C −0.392 1.575 −0.249 0.803
Stage II × Temperature 15°C 1.355 1.672 0.810 0.417
Stage III × Temperature 15°C 5.017 1.983 2.530 0.011
Stage IV × Temperature 15°C 6.716 2.580 2.603 0.009
Table 4. Occurrence of different behaviours in ovigerous females of Pleuron-
codes monodon kept under hypoxia (H) and normoxia (N) at 11 and 15°C, until
hatching of larvae. Number of ovigerous females that showed each observed
behaviour is given in Table S2 in the Supplement. Significant values (p < 0.05)
are highlighted in bold
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
the abdominal muscle of incubating
females of Pleuroncodes monodon in -
creased with temperature (p < 0.0001;
Table S3 in the Supplement, Fig. 4).
In the ab do minal muscle of incubat-
ing fe males, neither PK, CS nor their
ratio de pen ded on experimental con-
ditions (Fig. 5). Lactate concentration
in the abdominal muscle of incubat-
ing fe males of P. monodon increased
only at H15 (Fig. 5). Statistical differ-
ences were found between tempera-
ture treatments in HSP70 levels in
gills of incubating females. HSP70
levels were greater at 11°C than at
15°C (p < 0.02; Table S3), but post hoc
analysis did not show differences
among treatments (Fig. 5).
3.1.4. Ovary development
The frequency of oocytes in second-
ary vitellogenesis was significantly
different among treatments (G= 42.1;
61
Estimate SE Zp
Pleopod movement
(Best model: AIC = 71.5, D2= 0.30)
Model 0.038686 0.012903 2.998 0.003
Normoxia −0.005477 0.016333 −0.335 0.737
Stage II −0.009779 0.015814 −0.618 0.536
Stage III −0.021789 0.013967 −1.560 0.119
Stage IV −0.021188 0.014161 −1.496 0.135
Normoxia × Stage II 0.017013 0.023555 0.722 0.470
Normoxia × Stage III 0.040189 0.025997 1.546 0.122
Normoxia × Stage IV 0.084291 0.040323 2.090 0.037
Abdominal flapping
(Best model: AIC = 44.5, D2= 0.47)
Model 40.184 0.8320 4.830 <0.0001
Temperature 15°C −26.666 0.6949 −3.837 0.0003
Normoxia −0.8597 0.5930 −1.450 0.153
Stage II −0.0543 0.8332 −0.065 0.948
Stage III −0.4789 0.6678 −0.717 0.476
Stage IV −0.2064 0.8766 −0.235 0.815
Normoxia × Stage II 10.121 11.041 0.917 0.363
Normoxia × Stage III 18.883 10.038 1.881 0.065
Normoxia × Stage IV 96.981 42.570 2.278 0.027
Table 5. Frequency of different behaviours in ovigerous females of Pleuron-
codes monodon kept under hypoxia (H) and normoxia (N) at 11 and 15°C,
until hatching of larvae. Number of ovigerous females that showed each
observed behavior is given in Table S2. Significant values (p < 0.05) are
highlighted in bold
Fig. 3. Frequency of pleopod movements (PM) and abdominal flaps (AF) counted in effective time of video (ETV) during
30 min of observation of ovigerous female Pleuroncodes monodon. Pereopod probing (PP) was quantified as the percentage of
time that this behaviour occurred during the 30 min of observation. Bar: median; box: interquartile range (IQR): whiskers:
min./max. values ≤1.5 × IQR below/above box respectively; dots: outliers
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
df = 3; p < 0.0001), with the highest proportion
attained at N15 (Table 6B). The size at which 50 and
95% of oocytes I turned into oocytes II showed an
important difference in AIC (larger than that corre-
sponding to p < 0.05) between normoxia and hypoxia
at both temperatures (Table 6C; Fig. S4 in the Sup-
plement). Vitellogenesis II was reached at a diameter
1/5 smaller in hypoxia than in normoxia.
3.2. Short-term experiment:
rate of oxygen consumption
Considering individual mass as a co-variable, oxy-
gen consumption rates depended on oxygen concen-
tration and temperature for most adult categories
considered (Fig. 6, Table 7). Consumption rates of
adults increased with temperature. OFs showed the
highest mean respiration rate at 15°C (as compared
with other treatments). Data dispersion allowed us to
calculate the Pcrit only for OFs, which was higher at
15°C than at 11°C (~2 and 0.9 mg O2l−1, respectively;
62
Fig. 4. Contents (µg mg−1 of abdominal muscle) of total (A)
carbohydrates, (B) lipids and (C) proteins in incubating
ovigerous females of Pleuroncodes monodon were main-
tained for the duration of embryo mass development at dif-
ferent conditions of temperature and oxygen (hypoxia: 0.7 ±
0.1 mg O2l−1; and normoxia: 8.0 ± 1.0 mg O2l−1). Data are
means ± SE, different letters indicate statistical differences
at p < 0.05
Fig. 5. Mean ± SD (A) pyruvate kinase (PK) apparent spe-
cific activity per g, (B) citrate synthase (CS) apparent spe-
cific activity per g, and (C) PK:CS ratio in abdominal muscle;
and mean ± SD (D) lactate concentration and (E) heat-shock
protein 70 (HSP70) levels in gill tissue of incubating females
of Pleuroncodes monodon at different conditions of temper-
ature and oxygen
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
Fig. 6). This Pcrit observed from respiration rates was
accompanied by a higher concentration of lactate at
15°C observed for OFs (Fig. 7D). In the case of
females (non-OF and FABR), the effect of oxygen on
respiration rates was marginally significant, and for
males, the effect of oxygen was observed only at
15°C (Fig. 6, Table 7).
Broods, on the other hand, are oxyconformers, as
they responded linearly to oxygen concentration and
independently from temperature (Fig. 6). Broods
were in stage II of development, so a large part of the
eggs had energetic reserves rather than larval tissue.
Because the brood mass was <10 % of the carrying
female biomass, and oxygen consumption rates per
brood are similar to those of OF (female+brood),
brood mass respiration should comprise less than 5%
of such measurements. Enzymatic activity, but espe-
cially the PK:CS ratio (the higher the ratio, the
greater the relative capacity for anaerobic activity)
responses to short-term exposure to low oxygen con-
centration at the 2 temperatures contrasted with the
lack of differences in the long-term experiment,
although the range of values was consistent. In OFs,
CS increased with oxygen at 15°C (Fig. 7). At low
oxygen levels, PK measurements of females at 15°C
were concentrated in the higher range of values.
Nevertheless, the dispersion throughout higher oxy-
gen levels was large. PK:CS ratios showed a de -
creasing trend from low to high oxygen concentra-
tions in both ovigerous and non-ovigerous females
(Fig. 7C,G); males did not show any trend for enzyme
activities individually or for their rate (Fig. 7K). PK
and CS showed the lowest values in OFs (Fig. 7). In
contrast, the largest concentration of lactate was
observed for OFs at 15°C and mainly at low oxygen,
as compared with the other adult categories, where it
was similar at both temperatures (Fig. 7D,H,L). Lac-
tate was independent of temperature and oxygen for
the other adults.
4. DISCUSSION
4.1. Brooding success
The results of the long-term experiment showed
that the reproductive process and reproductive suc-
cess of the species is modulated by the interaction of
environmental oxygen concentration and tempera-
ture. The range of conditions considered are realistic
63
(A) (B)
Treatment Percentage of secondary oocytes G= 42. 1 df = 3 p < 0.0001
H11 26.9 H11 N11 H15
N11 30.8 N11 0.00425 − –
H15 22.5 H15 0.03554 2.20 × 10−5 −
N15 33.6 N15 0.00033 2.20 × 10−5 0.521
(C)
Treatment Model Size (µm) Parameter logL AIC AICc
Hypoxia 11°C vs. 1 model 50% 235 2 −1152 2308 2307
Normoxia 11°C 95% 287
2 models 50% H 201 4 −843 1695 1690
95% H 228
50% N 248
95% N 288
Hypoxia 15°C vs 1 model
Normoxia 15°C 50% 245 2 −342.5 693 688.3
95% 280
2 models 50% H 205 4 −222 453.9 449.2
95% H 224
50% N 249
95% N 270
Table 6. Ovary development in Pleuroncodes monodon females in the long-term experiment. (A) Percentage of oocytes in sec-
ondary vitellogenesis (VII). (B) Analysis of frequency G-test; values in bold indicate significant differences between treat-
ments. Treatments: H11 (H15): hypoxia at 11°C (15°C); N11 (N15): normoxia at 11°C (15°C). (C) Model fitting for comparison
of sizes at which 50% and 95 % of oocytes I turned into oocyte II
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
across the latitudinal and depth distribution of Pleu-
roncodes monodon (Gallardo 2017), since this is a
bentho-pelagic species, as are many other gala -
theids. Successful populations might occupy benthic,
pelagic or bentho-pelagic habitats, so both the actual
range of temperature/dissolved oxygen values expe-
rienced by the populations as well as those clearly
constraining their habitat use (vertical distribution)
are relevant. Hypoxia under low temperature is char-
acteristic of southern benthic populations (Gallardo
et al. 2017), while high temperature normoxia char-
acterizes Peruvian pelagic habitats, and hypoxia/
high temperatures characterize the vertical distribu-
tion limit of the species (Gutiérrez et al. 2008). Con-
tinuous hypoxic conditions and high temperature
during brood incubation were detrimental for both
the female and brood hatching success. Even if
females survived chronic exposure to warm tempera-
tures and hypoxia until the end of brood carrying,
they did not produce viable larvae. OFs that en -
counter hypoxia for short periods of time at high tem-
peratures (15°C) may supplement their metabolic
costs using anaerobic pathways, but during long
brood incubation, this capacity apparently decreases.
64
Fig. 6. Rates of oxygen consumption in the short-term exper-
iment: ovigerous females, females after brood removal
(FABR), broods, non-ovigerous females and males of Pleu-
roncodes monodon exposed to different oxygen concentra-
tion and 2 temperatures (11°C in grey and 15°C in black). In
ovigerous females, the curves were constructed by the
‘LOESS’ (local polynomial regression fitting) method. The
curves were used to detect the breakpoint (Pcrit), repre-
sented by a change of slope in the response variable as a
function of the independent variable. Pcrit was higher at 15°C
than at 11°C (~2 vs. 0.7 mg O2l−1, dotted and dash-dotted
lines, respectively)
Estimate SE WaldStat p
Ovigerous females
Intercept −2.152 0.351 37.662 0.000
Temperature −0.044 0.015 8.514 0.003
log (Weight) 0.057 0.024 5.376 0.020
Oxygen −0.108 0.052 4.315 0.037
Females after brood removal
Intercept −3.060 0.429 50.858 <<0.001
11°C × Weight −0.062 0.026 5.561 0.018
15°C × Weight −0.009 0.019 0.232 0.630
11°C × DO 0.105 0.040 6.815 0.009
15°C × DO 0.073 0.028 6.615 0.010
Broods
Intercept −4.152 0.208 398.659 <<0.001
Oxygen 0.386 0.053 52.580 <<0.001
Temperature 0.146 0.104 1.979 0.159
Non-ovigerous females
Intercept −2.369 0.180 173.907 0.000
Temperature −0.032 0.008 14.687 0.000
log (Weight) 0.024 0.011 4.538 0.033
Oxygen −0.100 0.052 3.780 0.051
Males
Intercept −2.729 0.324 70.750 0.000
Temperature −0.035 0.011 10.088 0.001
log (Weight) 0.059 0.023 6.434 0.011
Oxygen −0.364 0.071 25.985 0.000
Table 7. Separate slope models analysing the oxygen con-
sumption of Pleuroncodes monodon in the short-term exper-
iment, in different components of the population. DO: dis-
solved oxygen. Significant values (p < 0.05) are highlighted
in bold
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
Embryo development time shortened by almost
50% at 15°C compared to 11°C, but brood develop-
ment synchrony and brood success were also lower.
Developmental times coincided with those previ-
ously reported by Thiel et al. (2012) for the same spe-
cies at ~11°C, and fit well within expected values for
the squat lobster group. The duration of incubation
has an exponential negative relationship with tem-
perature for squat lobsters between 9 and 21°C (Thiel
& Lovrich 2011), and lengthening of development
is steepest for the range of temperatures studied
herein. Populations of P. monodon are found in a
temperature range from 11−21°C, so we evaluated
the development response to temperature over the
lower half of the range found throughout its latitudi-
nal distribution. Nonetheless, in natural conditions,
temperatures above 18°C do not coincide with hy -
poxia, and there are no previous observations on the
effect of combined temperature−oxygen conditions
on development time for squat lobsters or other crus-
taceans that permanently inhabit low oxygen zones.
Our results show that oxygen concen-
tration did not have an effect on devel-
opment time to first hatching. The
non-significant effect of oxygen on
development time of P. monodon
broods confirms that embryonic devel-
opment of P. mo nodon is substantially
shorter than that previously estimated
from field surveys (90−120 d, Palma &
Arana 1997) for populations inhabit-
ing low temperature hypoxic waters.
Interestingly, oxygen concentrations
did have an effect on the index of
brooding success at hypoxia 15°C.
Changes in asynchrony and success of
development caused by temperature
at levels close to the upper limit of
the environmental temperature range
have been found in broods or capsules
of other marine organisms that keep
embryos packed during embryo de -
velopment (Fernández et al. 2006).
Nevertheless, in other studies, hyp -
oxic conditions did show a strong
effect on development synchrony and
the proportion of undeveloped em -
bryos across temperatures (Steer et al.
2002). Besides external oxygen con-
centration, developmental asynchrony
is re lated to oxygen gradients within
crustacean embryo masses (Fernán-
dez et al. 2002) that might be over-
come by active brood ventilation by the female (Fer-
nández et al. 2003). When asynchrony is observed,
respiration rates of embryos are significantly affected
by oxygen concentrations (Fernández & Brante
2003), and something similar occurs in the centre of
the brood under hypoxic conditions. The internal
oxygen gradient within the brood develops as a bal-
ance between embryo consumption and diffusion
rates in the centre of the embryo mass (Fernández et
al. 2000). As the re sponse of oxygen consumption
rates to environmental oxygen con centration at low
temperature is narrower at 11°C than at 15°C,
embryos would be less affected at low temperature
hypoxia, especially during the last developmental
stages.
4.2. Parental care behaviour
Brooding success in squat lobsters is related to
parental care, improving offspring growth and sur-
65
Fig. 7. Apparent specific activity of enzyme and lactate concentration in mus-
cle of ovigerous females, non-ovigerous females and males of Pleuroncodes
monodon at different conditions of temperature and oxygen. (A,E,I) Pyruvate
kinase (PK) apparent specific activity per g. (B,F,J) Citrate synthase (CS)
apparent specific activity per g. (C,G,K) PK:CS ratio. (D,H,L) Lactate concen-
tration per g
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
vival (Thiel & Lovrich 2011). Frequency of parental
care behaviours may be a response to oxygen de -
mand of embryos (Baeza & Fernández 2002), since
oxygen consumption varies within the brood mass,
not only through development of embryos, but also
with oxygen distribution (partial pressure) within the
brood. Also, high temperatures are associated with
more intense incubating behaviours, and this is fur-
ther intensified during embryo development (Brante
et al. 2003).
The 3 common incubating behaviours described for
P. monodon in the present study are pleopod move-
ments, abdominal flapping, and fifth-pair pereopod
probing, similar to what has been shown for anomu-
ran (Pohle 1989) and brachyuran crabs (Fernández et
al. 2000, Baeza & Fernández 2002, Fernández &
Brante 2003). Pleopod movements are used to venti-
late the embryo mass, and are thought to provide
oxygen to the embryo mass (Fernández et al. 2000,
Baeza et al. 2016). The frequency of pleopod fanning
is greater in females carrying late-stage embryos,
which commonly have higher oxygen requirements
(Baeza et al. 2016). Indeed, for P. monodon under
hypoxic conditions, pleopod movements were more
frequent than under normoxia, and probably repre-
sent the attempts of females to better ventilate the
brood mass and provide oxygen to the developing
embryos. Nonetheless, there was no link between
pleopod movements and embryo developmental
stage, but there was an interaction between oxygen
and temperature, indicating that females are sensi-
tive to oxygen concentrations in the brood mass.
Abdominal flapping is an energy-expensive beha -
viour, which could imply a higher effort for OFs
under hypoxic conditions (Fernández & Brante 2003),
even more so when temperatures rise. This might
mean that there were strong costs associated with
oxygen provisioning to the brood, especially in OFs
of P. monodon under hypoxia at 15°C. Abdominal
flapping commonly results in increased oxygen
availability in the centre of the embryo mass (Baeza
& Fernández 2002). Abdominal flapping has also
been described for Munida gregaria (Dellatorre &
Barón 2008) and Panulirus argus (Baeza et al. 2016).
Baeza & Fernández (2002) affirmed that the fre-
quency of abdominal flaps was significantly higher in
females carrying late-stage embryos than in females
carrying early-stage embryos. Females of Pleuron-
codes monodon exhibit frequent abdominal flapping
throughout embryonic development, maintaining
high levels in hypoxic conditions at 15°C in later
embryonic stages. Possibly, under this particular con-
dition, OFs spend more energy because embryos
require more oxygen. Our measurements of brood
oxygen consumption were undertaken for eggs in
stage II, when embryos comprise a small fraction of
overall egg mass/volume; oxyconformity might con-
tribute to diminish intra-brood oxygen gradients.
Nevertheless, as development proceeds oxygen con-
sumption increases in crustacean eggs, and it is nec-
essary for development. This is congruent with the
higher incidence of maternal care behaviours at later
developmental stages. Pereopod probing is used for
grooming and removal of dead embryos from the
embryo mass (Förster & Baeza 2001) and does not
affect oxygen availability in the centre of the brood
(Fernández & Brante 2003, Dellatorre & Barón 2008).
Baeza et al. (2016) found that pereopod probing fre-
quency increased in later embryo stages, but in the
present study, pereopod probing diminished for N11
and H15 in later-stage embryos.
Female condition is affected by the level of paren -
tal care investment, especially at high temperatures
at low oxygen concentrations. Increase in lactate in
the abdominal muscle of females by the end of the
incubation period indicates their reliance on anaero-
bic metabolism to sustain abdominal flapping for
ventilation and is consistent with the lowest carbohy-
drate levels among the 4 treatments. It has been sug-
gested that additional effort in incubating behaviours
under stressful conditions could cause the reduction
of the female’s fitness (Brante et al. 2003). In this
study, cellular stress, as indicated by HSP70 levels,
were similar among females incubating under differ-
ent conditions of temperature and oxygen. However,
lowest carbohydrate levels found in abdominal mus-
cle of incubating females maintained at hypoxia at
low temperature can be attributed to direct costs of
parental care, since carbohydrates are the main
macro molecule that provides energy for muscle con-
traction activity (Jimenez & Kinsey 2015). Among
treatments, higher routine metabolic rates with in -
creasing temperature might lead to a reduced aero-
bic scope of muscle activity, especially in hypoxia.
Observed levels of lactate and carbohydrate concen-
trations at the end of brooding suggest that female
mortality, growth and future broods could be com-
promised in this environmental regime.
4.3. Potential for producing a subsequent brood
Temperature has an important effect on ovary
development as in all other developmental rates due
to increased metabolism. The mean size of oocytes in
secondary vitellogenesis is similar in females that
66
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones
completed egg maturation at 11 and 15°C, even
though developmental time at 11°C is almost twice
as long (~42 d) than at 15°C (~25 d). In normoxia,
females start ovary maturation immediately after egg
extrusion, and gonad maturation advances in parallel
to egg mass development. The sooner the ovary
completes secondary vitellogenesis after completing
brood incubation, the sooner females can produce a
new brood. Successive carrying events with few days
in between have been observed in the laboratory,
where some female P. monodon produced up to 5
broods (in normoxic conditions) during the annual
reproductive season (Thiel et al. 2012). Continuous
exposure to cyclic hypoxia appears to suppress the
number of broods a female grass shrimp is capable of
producing (Brown-Peterson et al. 2008). In our case,
the diameter of oocytes turning into secondary vitel-
logenesis differed among treatments, and the larger
ones appeared in normoxia at both 11 and 15°C. On
the other hand, the frequency of oocytes in vitelloge-
nesis II was lowest in the 2 hypoxic treatments, which
implies that exposure to hypoxia during early gonad
development might affect the size, thus the egg ener-
getic reserves, and number of eggs of the future
brood. This is also consistent with the increased de -
mands of energy for maternal care and lower upper
limit of energy acquisition in this condition. Consis-
tently, the largest effect was observed on oocyte II
proportions at 15°C, and females in hypoxia at 15°C
had the lowest aerobic metabolic potential as meas-
ured by CS activity.
In nature, OFs are distributed according to oxygen
conditions in a range from 0.7−1.42 mg O2l−1 at 11°C
(Gallardo et al. 2017). The observed field threshold
coincides with our experimental results, indicating
that embryonic development starts to be affected at
and below those oxygen levels. Hypoxic conditions
at higher temperatures are detrimental to reproduc-
tive success and female survival, and therefore tend
to be avoided by incubating females. In the field, at
latitudes where the OMZ is associated with higher
temperatures, P. monodon is found in its pelagic form
above the oxycline. Late first maturity (2 to 3 yr old)
in typical cold-water populations, together with lar -
ger embryo masses at first extrusion (number of
embryos proportional to size), would be a disadvan-
tage at high temperatures, where internal dissolved
oxygen gradients could be less marked in smaller
embryo masses typical of smaller (younger, pelagic)
forms. Reproduction onset at smaller (younger) size,
small broods and higher brooding frequency (as is
known for pelagic populations at high temperatures;
Gutiérrez et al. 2008) could be advantageous above
15°C in normoxic conditions, while hypoxia at that
temperature should be avoided due to the high asso-
ciated incubation costs and the risk of reproduction
failure.
Overall, our results show that sub-thermocline
conditions observed in northern Chile and southern
Perú (hypoxia 15°C) can severely suppress the re -
productive potential of P. monodon. In fact, at those
latitudes (10−22° S), OFs remain above the oxycline,
in normoxic warm waters. The pelagic adults attain
smaller sizes than the ones in benthic cold hypoxic
waters (Franco-Meléndez 2012), so the restrictive
effects of 15°C hypoxia observed in our experiment
could be overcome. The pelagic population is in
fact a very pro ductive one, as indicated by the very
high ob served biomasses (Gutiérrez et al. 2008). In
central-Chile (30−37° S), the reproductive period
tends to coincide with more oxygenated bottom
waters, and females are found preferentially above
0.5 ml l−1 (Gallardo et al. 2017). Nevertheless fast
(hours), intense (over 50% change in oxygen satu-
ration) and persistent deoxygenation events over
scales of days might occur during the season, and
the short- vs. long-term differences in enzymatic
activity would be more pronounced during those
periods. Non-OFs could tolerate these episodic
events, although if the escape response of OFs is
not entrained fast enough, they might survive, but
conditions are detrimental for egg and ovarian
development. Video recordings of benthic popula-
tions at about 30° S have shown OFs in extremely
hypoxic conditions lying on their backs, reducing
all other activity except abdominal flapping (J. Sel-
lanes unpubl. data).
4.4. Implication for population dynamics
The observed effects (here on reproductive fe -
males) can influence not only their reproductive
potential but also their growth. Differences in female
and male growth rates have been repeatedly re por -
ted after annual stock assessments (Roa & Tapia
1998). Lower growth rates in females are expected
due to energy allocation to gonad vitellogenesis (Bas-
cur et al. 2018) (extruded egg mass might represent
up to 5% of female body mass), especially during
years when environmental conditions allow for an
extended reproduction period, and eventually more
carrying events per individual females. In addition, if
important (extended and/or intense) hypoxic events
occur during incubation, growth would be further
compromised since other costs of maternal care are
67
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
enhanced under such conditions together with a
lower maximum energy processing capacity, due to
hypoxic limits on oxygen consumption above routine
metabolic rate.
Our results point to a new way to re-evaluate
population assessments in view of environmental
variability (dissolved oxygen, temperature), causing
interannual variability in female mortality, length of
the reproductive period and fecundity, among others.
Temperature and oxygen concentration thus signifi-
cantly influence the reproductive success of P. mon-
odon in its natural environment. These effects could
possibly explain large density-independent recruit-
ment failures among the squat lobster populations.
The definition of hypoxic conditions is related to
species-specific abilities to withstand low dissolved
oxygen levels, and similarly, oxygen concentration
might impose ‘hypoxic’ restrictions at high tempera-
tures. Hypoxia thresholds might also vary between
different ontogenetic stages and reproductive condi-
tions. We conclude that environmental oxygen cycles
in southern-central Chile and episodic temperature
oscillations might affect brood development, as well
as long-term survival of OFs, and are the missing fac-
tor to explore the relationship between these condi-
tions and interannual recruitment success.
Acknowledgements. This study was supported by FONDE-
CYT project 1140845 to B.Y., M.T. and K.B. We thank Dr.
Marcel Ramos and his project FONDECYT 1140832 for
complementary financial support. CONICYT Becas Chile
21110922 and "Beca Postdoctorado Universidad Católica del
Norte N°0003" provided funds for M.A.G. Our gratitude goes
to the team of the FIGEMA laboratory (Ana Mercado,
William Farías and Katherine Jeno) and of the OCEANOLAB
laboratory (Paula Oyarce, Jaime Garcia, Sebastian Rojas and
María Valladares). We also thank the team of the Biological
Collections Room (SCB-UCN) from the Universidad Católica
del Norte (Dr. Javier Sellanes, Jorge Aviles and Jan Tapia),
the crew of the scientific vessel ‘Stella Maris II’, the
laboratory assistants (Sergio Fuentes and Oscar Pino), the
Production Laboratory at UCN (Germán Pizarro, Germán
Lira and Matías Perez) and Andrés Gonzalez for supporting
statistical analyses. Finally, we thank Hanna Francis for Eng-
lish editing.
LITERATURE CITED
Baeza JA, Fernández M (2002) Active brood care in Cancer
setosus (Crustacea: Decapoda): the relationship between
female behaviour, embryo oxygen consumption and the
cost of brooding. Funct Ecol 16: 241−251
Baeza JA, Simpson L, Ambrosio LJ, Mora N, Guéron R, Chil-
dress MJ (2016) Active parental care, reproductive per-
formance, and a novel egg predator affecting reproduc-
tive investment in the Caribbean spiny lobster Panulirus
argus. BMC Zool 1: 6
Bascur M, Guzmán F, Mora S, Espinoza P, Urzúa A (2018)
Temporal variation in the fatty acid composition of ovi -
gerous females and embryos of the squat lobster Pleu-
roncodes monodon (Decapoda: Munididae). J Mar Biol
Assoc UK 98(Spec Issue 8): 1977−1990
Bosch I, Slattery M (1999) Costs of extended brood protec-
tion in the Antarctic sea star, Neosmilaster georgianus
(Echinodermata: Asteroidea). Mar Biol 134: 449−459
Brante A, Fernández M, Eckerle L, Mark F, Pörtner HO,
Arntz W (2003) Reproductive investment in the crab
Cancer setosus along a latitudinal cline: egg production,
embryo losses and embryo ventilation. Mar Ecol Prog Ser
251: 221−232
Brokordt K, Himmelman J, Guderley H (2000) Effect of
reproduction on escape responses and muscle metabolic
in the scallop Chlamys islandica Muller 1776. J Exp Mar
Biol Ecol 251: 205–225
Brokordt K, Pérez H, Herrera C, Gallardo A (2015) Repro-
duction reduces HSP70 expression capacity in Argo -
pecten purpuratus scallops subject to hypoxia and heat
stress. Aquat Biol 23: 265−274
Brown-Peterson NJ, Manning CS, Patel V, Denslow ND,
Brouwer M (2008) Effects of cyclic hypoxia on gene
expression and reproduction in a grass shrimp, Palae-
monetes pugio. Biol Bull (Woods Hole) 214: 6−16
Dellatorre FG, Barón PJ (2008) Multiple spawning and
length of embryonic development of Munida gregaria in
northern Patagonia (Argentina). J Mar Biol Assoc UK 88:
975−981
Dubois M, Gilles K, Hamilton J, Rebers P, Smith F (1956)
Colorimetric method for determination of sugars and
related substances. Anal Chem 28: 350−356
Fernández M, Brante A (2003) Brood care in Brachyuran
crabs: the effect of oxygen provision on reproductive
costs. Rev Chil Hist Nat 76: 157−168
Fernández M, Bock C, Pörtner HO (2000) The cost of being
a caring mother: the ignored factor in the reproduction of
marine invertebrates. Ecol Lett 3: 487−494
Fernández M, Pardo LM, Baeza JA (2002) Patterns of oxy-
gen supply in embryo masses of brachyuran crabs
throughout development: the effect of oxygen availabil-
ity and chemical cues in determining female brooding
behavior. Mar Ecol Prog Ser 245: 181−190
Fernández M, Ruiz-Tagle N, Cifuentes S, Pörtner HO, Arntz
W (2003) Oxygen-dependent asynchrony of embryonic
development in embryo masses of brachyuran crabs.
Mar Biol 142: 559–565
Fernández M, Pappalardo P, Jeno K (2006) The effects of
temperature and oxygen availability on intracapsular
development of Acanthina monodon (Gastropoda: Muri-
cidae). Rev Chil Hist Nat 79: 155−167
Förster C, Baeza JA (2001) Active brood care in the Anomu-
ran crab Petrolisthes violaceus (Decapoda: Anomura:
Porcellanidae):
grooming of brooded embryos by the
fifth pereiopods. J Crustac Biol 21: 606−615
Franco-Meléndez M (2012) Breeding behavior and sex ratio
variation of Pleuroncodes monodon (Crustacea: Gala -
theidae) off the Peruvian coast. Cienc Mar 38: 441−457
Gallardo MA (2017) Ecofisiología e historia de vida de Pleu-
roncodes monodon y su relación con la variabilidad
espacio temporal de la ZMO del Pacífico Oriental. PhD
thesis, Universidad Católica del Norte-Universidad de La
Serena, Coquimbo
Gallardo MA, González López AE, Ramos M, Mujica A,
Muñoz P, Sellanes J, Yannicelli B (2017) Reproductive
68
Author copy

Gallardo et al.: Incubation of squat lobster in oxygen minimum zones 69
patterns in demersal crustaceans from the upper bound-
ary of the OMZ off north-central Chile. Cont Shelf Res
141: 26−37
Gallardo VA, Palma M, Carrasco FD, Gutiérrez D, Levin LA,
Cañete JI (2004) Macrobenthic zonation caused by the
oxygen minimum zone on the shelf and slope off central
Chile. Deep Sea Res II 51: 2475−2490
Gilly WF, Beman JM, Litvin SY, Robison BH (2013) Oceano-
graphic and biological effects of shoaling of the oxygen
minimum zone. Annu Rev Mar Sci 5: 393−420
Grieshaber MK, Hardewig I, Kreutzer U, Pörtner HO
(1994) Physiological and metabolic responses to hyp -
oxia in in vertebrates. Rev Physiol Biochem Pharmacol
125: 43−147
Gutiérrez M, Ramirez A, Bertrand S, Móron O, Bertrand A
(2008) Ecological niches and areas of overlap of the squat
lobster ‘munida’ (Pleuroncodes monodon) and anchoveta
(Engraulis ringens) off Peru. Prog Oceanogr 79: 256−263
Haye PA, Salinas P, Acuña E, Poulin E (2010) Heterochronic
phenotypic plasticity with lack of genetic differentiation
in the southeastern Pacific squat lobster Pleuroncodes
monodon. Evol Dev 12: 628−634
Jimenez AG, Kinsey ST (2015) Energetics and metabolic
regulation. In: Chang E, Thiel M (eds) The natural his-
tory of the Crustacea, Vol 4: Physiology. Oxford Univer-
sity Press, New York, NY, p 389−417
Johnson JB, Omland KS (2004) Model selection in ecology
and evolution. Trends Ecol Evol 19: 101−108
Kiko R, Hauss H, Dengler M, Sommer S, Melzner F (2015)
The squat lobster Pleuroncodes monodon tolerates
anoxic ‘dead zone’ conditions off Peru. Mar Biol 162:
1913−1921
Kronenberger K, Brandis D, Türkay M, Storch V (2004)
Functional morphology of the reproductive system of
Galathea intermedia (Decapoda: Anomura). J Morphol
262: 500−516
Levin LA (2003) Oxygen minimum zone benthos: adaptation
and community response to hypoxia. Oceanogr Mar Biol
Annu Rev 41: 1−45
Lovrich GA, Thiel M (2011) Ecology, physiology, feeding
and trophic role of squat lobsters. In: Poore G, Ahyong S,
Taylor J (eds) The biology of squat lobster. CSIRO Pub-
lishing, Melbourne, p 183–222
Mann R, Gallager SM (1985) Physiological and biochemical
energetics of larvae of Teredo navalis L. and Bankia
gouldi (Bartsch) (Bivalvia: Teredinidae). J Exp Mar Biol
Ecol 85: 211−228
Moreno JE, Méndez CA, Meruane JA, Morales MC (2012)
Descripción histológica y caracterización de los estados
de madurez gonadal de hembras de Cryphiops caemen-
tarius (Molina, 1782) (Decapoda: Palaemonidae). Lat Am
J Aquat Res 40: 668−678
Newell R, Branch GM (1980) The influence of temperature
on the maintenance of metabolic energy balance in mar-
ine invertebrates. Adv Mar Biol 17: 329−396
Newell RC, Northcroft HR (1967) A re interpretation of the
effect of temperature on the metabolism of certain mar-
ine invertebrates. J Zool 151: 277−298
Palma S (1994) Distribución y abundancia de larvas de lan-
gostino colorado Pleuroncodes monodon frente a la costa
de Concepción, Chile. Investig Mar 22: 13−29
Palma S, Arana P (1997) Aspectos reproductivos del langos -
tino colorado (Pleuroncodes monodon H. Milne Edwards,
1837), frente a la costa de Concepción, Chile. Investig
Mar 25: 203−221
Pohle GW (1989) Gill and embryo grooming in lithodid
crabs: comparative functional morphology based on Lith-
odes maja. In:
Felgenhauer BE, Watling L, Thistle AB
(eds) Functional morphology of feeding and grooming in
Crustacea. A.A. Balkema, Rotterdam, p 75−94
Pörtner H, Farrell A (2008) Physiology and climate change.
Science 322: 690−692
Pörtner HO, Grieshaber MK (1993) Critical PO2(s) in oxy-
conforming and oxyregulating animals: gas exchange,
metabolic rate and the mode of energy production. In:
Bicudo J (ed) The vertebrate gas transport cascade:
adaptations to environment and mode of life. CRC Press,
Boca Raton, FL, p 330−357
Quetin LB, Childress JJ (1976) Respiratory adaptations of
Pleuroncodes planipes to its environment off Baja Cali-
fornia. Mar Biol 38: 327−334
R Core Team (2015) R: a language and environment for sta-
tistical computing. R Foundation for Statistical Comput-
ing, Vienna
Roa R, Tapia F (1998) Spatial differences in growth and sex-
ual maturity between branches of a large population of
the squat lobster Pleuroncodes monodon. Mar Ecol Prog
Ser 167: 185−196
Seibel BA (2011) Critical oxygen levels and metabolic sup-
pression in oceanic oxygen minimum zones. J Exp Biol
214: 326−336
Seibel BA, Childress JJ (2013) The real limits to marine life:
a further critique of the respiration index. Biogeosciences
10: 2815−2819
Seibel BA, Luu BE, Tessier SN, Towanda T, Storey KB (2018)
Metabolic suppression in the pelagic crab, Pleuroncodes
planipes, in oxygen minimum zones. Comp Biochem
Physiol B Biochem Mol Biol 224: 88−97
Sokal RR, Rohlf FJ (1995) Biometry. The principles and prac-
tice of statistics in biological research. WH Freeman and
Company, New York, NY
Steer MA, Moltschaniwskyj NA, Gowland FC (2002) Tem-
poral variability in embryonic development and mor -
tality in the southern calamary Sepioteuthis australis: a
field assessment. Mar Ecol Prog Ser 243: 143−150
Talley LD, Pickard GL, Emery WJ, Swift JH (2011) Descrip-
tive physical oceanography: an introduction, 4th edn.
Pergamon Press, Oxford
Thiel M, Lovrich GA (2011) Agonistic behavior and repro-
ductive biology of squat lobsters. In: Poore G, Ahyong S,
Taylor J (eds) The biology of squat lobsters. CSIRO Pub-
lishing, Melbourne, p 223−247
Thiel M, Espinoza-Fuenzalida N, Acuña E, Rivadeneira MM
(2012) Annual brood number and breeding periodicity of
squat lobsters (Decapoda: Anomura: Galatheidae) from
the continental shelf of the SE Pacific — implications for
fisheries. Fish Res 129−130: 28−37
Torres R, Manriquez PH, Duarte C, Navarro JM, Lagos NA,
Vargas CA, Lardies MA (2013) Evaluation of a semi-
automatic system for long-term seawater carbonate che -
mistry manipulation. Rev Chil Hist Nat 86: 443−451
Vaquer-Sunyer R, Duarte CM (2011) Temperature effects on
oxygen thresholds for hypoxia in marine benthic organ-
isms. Glob Change Biol 17: 1788−1797
Whiteley NM, Taylor ETW (2015) Responses to environmental
stresses: oxygen, temperature, and pH. In: Chang ES, Thiel
M (eds) The natural history of the Crustacea, Vol 4: Physiol-
ogy. Oxford University Press, New York, NY, p 320−358
Willmer P, Stone G, Johnston I (2005) Environmental physi-
ology of animals, 2nd edn. Blackwell Publishing, Carlton
Author copy

Mar Ecol Prog Ser 623: 51– 70, 2019
Wu RSS (2009) Effects of hypoxia on fish reproduction
and development. In: Richards JG, Farrell A, Brauner
C (eds) Fish physiology, Vol 27. Hypoxia. Elsevier,
London, p 79−141
Yannicelli B, Castro L, Parada C, Schneider W, Colas F,
Donoso D (2012) Distribution of Pleuroncodes monodon
larvae over the continental shelf of south-central Chile:
field and modeling evidence for partial local retention
and transport. Prog Oceanogr 92−95: 206−227
Zar J (2010) Biostatistical analysis, 5th edn. Prentice Hall,
Upper Saddle River, NJ
Zuur AF, Ieno EN, Walker N, Saveliev AA, Smith GM (2009)
Mixed effects models and extensions in ecology with
R. Springer, New York, NY
70
Editorial responsibility: Inna Sokolova,
Rostock, Germany
Submitted: April 26, 2018; Accepted: April 30 2019
Proofs received from author(s): July 16, 2019
Author copy