Patterns of energy allocation to reproduction in three Amazonian fish species

The study considered the influence of the hydrological cycle and gonadal development on the accumulation and use of energy in three fish species from an Amazonian flooded area. Fishes were sampled over a 24 hour period at monthly intervals between July 2004 and June 2005 using gillnets of different mesh sizes. Body cavity fat and gonadosomatic indices were determined, as well as energy content of gonads and muscles. Amongst the studied species, different means of energy allocation for reproduction were found: Acestrorhynchus falcirostris allocate energy from body cavity fat to its gonads; Pygocentrus nattereri uses mainly energy accumulated in the muscles for the process of gonadal maturation; and Hoplosternum littorale uses energy accumulated in their muscles and body cavity fat for reproductive processes. It is quite clear that the flood pulse regulates the gain and use of the energy reserves in fishes from the Amazonian floodplain.


Introduction
Patterns of energy allocation in fishes may change in accordance to the degree of sexual maturity of the individual, as timing of reproductive activity as well as maturation of the gonads require materials obtained from ingested food and energetic reserves deposited in various parts of the body (Tytler & Calow, 1985).Amongst fish species, there are differences in the acquisition, accumulation and allocation of energy reserves, given variations in habitat, trophic category, nature and origin of food, as well as the reproductive strategy used by the species in question (Bennemann et al., 1996;Huntingford et al., 2001;Fiorin et al., 2007).
Studies that quantify the energy allocations of organisms have been frequently used to evaluate the magnitude of the allocation for reproduction (Huntingford et al., 2001;Jonsson & Jonsson, 2005).Other studies have demonstrated that fish inhabiting areas with high seasonal variation in water level exhibit differences in the storage and allocation of energy reserves throughout the year (Carvalho, 1979;Junk, 1985;Finstad et al., 2003;Eggleton & Schramm, 2004;Dourado & Benedico-Cecílio, 2005).Such variations in energy allocation can lead to modifications of the physiological state of the fish, which can be detected by measuring biological parameters such as the index of body cavity fat, gonadosomatic index, and the energetic density of somatic and reproductive tissues.These estimates are useful for investigating the life history strategies of fish that inhabit areas with seasonal fluctuations (Chellappa et al., 1995;Encina & Granado-Lorencio, 1997;Penczak et al., 1999;Hoffmayer et al., 2006).
In the Amazon, rivers exhibit a marked seasonal fluctuation in water level.This seasonality exerts a strong influence on the nutrient balance, trophic interactions, and energy cycle of aquatic environments, and consequently, over the aquatic and terrestrial fauna and flora (Junk et al., 1989).These periodical variations exert a strong selective pressure, with the entire biota developing physiologic, anatomic, and/or behavioural mechanisms to survive in these conditions (Val & Almeida-Val, 1999).Main events on fishes' lives have been associated with the flood pulse (Goulding et al., 2003) and this constitutes an endless source of investigation on biological patterns, fact that stimulated this work.
There are no studies that quantify the allocation of energy for use in reproductive processes in seasonal environments in the Amazon basin.Thus, the hypothesis of this study considers there are differences in the strategy of energy allocation among fish species from the Amazon floodplain.The objective of this study was to identify the mode of energy used throughout a hydrological and reproductive cycle of three fish species in the flooded forest of the Amazon.

Study area
This study was conducted in a flooded area of Central Amazon, Amazonas State, Brazil, at the confluence of the Amazonas and Negro Rivers, between the geographic coordinates 3º08' -3º14'S and 59º53' -59º58'W (Fig. 1), named Catalão.This region corresponds to a low-lying land, which, depending upon the hydrological period, it floods to form a large lake continuous with the rivers or forms an isolated lake separated from the rivers.
The hydrological cycle at the study site was defined according to Bittencourt & Amadio (2007), as consisting of four hydrological periods (Fig. 2).Rising: water entering the river system and water levels increasing, from mid December to beginning of March, between 20-26 m above sea level.High: water-level equal to or above 26 m above sea level, from beginning of March to end of July.Receding: water leaving the river system and water levels falling, from end of July to end of October, between 26-20 m above sea level.Low: waterlevel equal to or below 20 m above sea level, from end of October to mid December.

Fish Sampling
Fishes were sampled monthly from July 2004 to June 2005, using a set of 10 gillnets with mesh sizes ranging from 30 to 120 mm between opposite knots, measuring 10 x 1.5-3.5 m.Nets were left in the water for 24 h and fish were removed every 6 hours.The few fish alive after being removed from the nets were anaesthetized with benzocaine (1 g 10 l -1 ) and were all placed in plastic bags, stored in coolers with ice, and brought to the Laboratory at INPA, for measurements.
Standard length (L s -mm), total weight (W t -g), gonad weight (W g -g) and body cavity fat weight (W f -g) were recorded for each fish.Vouchers have been deposited in the Fish Collection of the Instituto Nacional de Pesquisas da Amazônia (A. falcirostris INPA 25652, H. littorale INPA 25614 and P. nattereri INPA 25615).There was no sampling of P. nattereri during the low-water period.Table 1 shows number of individuals, mean standard length and total weight of each species.
Gonad and body cavity fat development were calculated using the Gonadosomatic Index (I G = 100 .W g / W t ) and Body Cavity Fat Index (I F = 100 .W f / W t ) respectively.Only female specimens were used for this study as their gonadal stages were readily recognized.Development stages of gonads were determined based on the degree of vascularization, the percentage of the abdominal cavity occupied by ovaries, and presence, size and colour of oocytes (Vazzoler, 1996;Mazzoni & Iglesias-Rios, 2002;Santos et al., 2005).The following classification was used: immature, maturing, ripe, and spawned/recovering.For the determination of energy content, a minimum of 2 g (wet weight) of muscle and gonad tissue was collected, to ensure a minimum 0.5 g dry weight sample.Samples were freeze-dried, ground to a fine homogeneous powder in a spherical mill and burned in a calorimetric bomb (Parr model 1261).To obtain the necessary sample weight for analysis in the macro-bomb, gonads in maturing and spawned/ recovering stages were pooled from individuals captured in the same period.For each species, an average of 10 individuals at each gonadal stage was considered; due to the low weight of the gonads of immature individuals, it was not possible to quantify their energy content.

Statistical Analyses
Variation in energy content of muscles and gonads, I G and I F , stage of gonadal development and hydrological periods of the year, within and between species, was analyzed by means of Analysis of Variance (ANOVA) followed by a Tukey's post hoc test for contrast between means (Zar, 1999).

Variation in I F , I G and caloric content of muscles and gonads throughout the hydrologic and reproductive cycle
Acestrorhynchus falcirostris -there were significant differences in I F in some stages of gonadal development (Fig. 3a) and throughout the hydrologic cycle (Fig. 3d).Individuals with immature (0.58% + 0.33) and ripe gonads (0.66% + 0.13) exhibited a lower amount of body cavity fat, while the highest values were found in individuals with spawned/recovering (1.99% + 0.12) and maturing gonads (1.66% + 0.19) (ANOVA, p < 0.01).Throughout the hydrologic cycle, the highest values of I F occurred during receding (2.23% + 0.11) and low waters (1.74% + 0.29) while the lowest values occurred during rising (0.67% + 0.13) and high waters (0.70% + 0.09).The receding and low water periods were significantly different from the rising and high water periods (ANOVA, p < 0.01).
In contrast to I F , I G was highest in ripe individuals (5.65% + 0.44), being significantly different (ANOVA, p < 0.01) from other stages (Fig. 3b).When analyzed for the hydrologic cycle, I G exhibited an inverse pattern to I F again.The highest values of I G occurred during rising (2.62 % + 0.56) and high waters (3.80% + 0.53; Fig. 3e), while the lowest values occurred during receding (0.60% + 0.08) and low waters (0.35% + 0.03) (ANOVA, p < 0.01).Most ripe individuals were sampled during the rising and high water periods whereas individuals with maturing and recovering gonads predominated during the receding and low water periods (Fig. 4a).
Significant differences between the values of muscle energy content were observed throughout the reproductive cycle.Individuals with maturing and ripe gonads exhibited lower amounts of energy in their muscles (18.1 kJ g -1 dw + 2.2 and 16.2 kJ g -1 dw + 0.3 respectively) than those with immature and spawned/recovering gonads (19.4 kJ g -1 dw + 0.6 and 18.9 kJ g -1 dw + 1.5 respectively), (ANOVA; p < 0.01) (Fig. 6c).

Discussion
Most reproductive activity of the three species occurred during rising waters.The highest energy content in the gonads, the highest frequency of ripe females and the highest gonadosomatic indices were observed during the same period.Other studies on the reproductive biology of Amazonian fish species have identified rising and high waters as the periods of the year during which the majority of species reproduce (Fernandes & Merona, 1988;Vazzoler & Amadio, 1990;Isaac et al., 2000).The energy acquired by fish through feeding is used for reproduction, metabolism, and growth, or is accumulated in the form of energy reserves (Jobling, 1994).In general, immature individuals exhibited lower concentrations of muscular energy and body cavity fat.In juveniles, most of the energy is allocated to metabolism and somatic production, leaving low levels of energy in their bodies (Tytler & Calow, 1985).Conversely, adults store large amounts of energy to support reproductive processes (Wootton, 1985).For A. falcirostris, significant differences in energy content of gonads among maturing, ripe and spawned/recovering individuals were observed, whereas the energy content of their muscles were not significantly different.These results suggest that this species does not use energy stored in muscle during gonadal development, that is, there is a direct flow of energy from its body cavity fat reserves to the gonads.Similarly, Dória & Andrian (1997), investigating the patterns of variation in energy content of somatic (muscle and liver) and reproductive (gonads) tissues of Schizodon borellii (Boulenger, 1900) and Pimelodus maculatus (La Cèpede, 1803) of the Paraná River floodplain, showed that the highest energy content of gonads with oocytes already in the process of vitellogenesis, did not promote reduction of the somatic reserves, suggesting that in this phase of the reproductive process, visceral fat acts as the energy source necessary to support gonadal development.
The lowest indices of muscular energy observed in ripe females of H. littorale in the period of greatest reproductive activity (rising waters), indicate that, beyond reserves of body cavity fat, it can also utilize energy accumulated in the muscles to support the reproductive process.Encina & Granado-Lorencio (1997), studying the seasonal variation in the energy content of Leuciscus pyrenaicus (Günther, 1868), registered a decline in the amount of energy in somatic tissues, as a result of an increase in reproductive tissues, which, according to the authors, demonstrates a transfer of energy reserves from the somatic tissues to the gonads.Previous research relates that muscles represent an important energy store for some fish species (Eliassen & Vahl, 1982;van Dijk et al., 2005).
Acestrorhynchus falcirostris and Hoplosternum littorale showed a gradual increase in gonadosomatic index correlated with a decrease in the index of body cavity fat.This inverse relationship between I G and I F observed for these species demonstrates the allocation of energy accumulated in the form of fat for gonadal development, which requires energy for oocyte maturation.According to Zwolinski et al. (2001), significant amounts of body cavity fat are used by fish to supply the high energy cost required for gonadal development.
Pygocentrus nattereri showed the lowest energy values in muscles of individuals with ripe gonads, suggesting a mobilization of muscular energy for gonadal development.For this species, a different means of using body cavity fat was observed as its highest cavity fat indices were registered during the period of reproduction, suggesting that body cavity fat is not used for the gonadal maturation processes.
Moreover, P. nattereri had a lower gonadosomatic index than either A. falcirostris or H. littorale.According to Junk (1985), species that have relatively low I G produce smaller numbers of oocytes and due to environmental factors such as food availability, species with this type of life history, including Cichla ocellari (Bloch & Schneider, 1801), Plagioscion squamosissimus (Heckel, 1840), and Astronotus ocellatus (Agassiz, 1831), do not have variable body lipid quantities throughout reproductive process.
Fish recruitment does not depend upon energetic resources allocated to reproductive process only, but also on the synchrony between reproductive timing and site.The remarkable flood pulse regime present in the Amazon region determine biological conditions in the life cycle of fishes such as reproductive activity and energy allocation patterns.

Fig. 1 .
Fig. 1.Map of the Catalão region, located near Manaus, Amazonas State, showing the sampling site.

Fig. 2 .
Fig. 2. Fluctuation of monthly average water level of the Negro River during the period of July 2004/ June 2005, showing the four phases of the hydrological cycle.Source: Bittencourt & Amadio (2007).

Table 1 .
Number of individuals (N), standard length variation (L s ) in mm, mean standard length with standard deviations (L s ) and mean total weight (W t ) for each species.