Reproduction of invasive Loricariichthys platymetopon Isbrücker & Nijssen , 1979 ( Actinopterygii , Loricariidae ) on the upper Paraná River loodplain

Loricariichthys platymetopon originally occurred in the Uruguay, Paraguay and lower Paraná rivers. With the formation of Itaipu Reservoir in 1982, it invaded and colonized the upper Paraná River. Using the data obtained quarterly at nine sampling sites on the upper Paraná River loodplain, the reproductive phases and the germ cells of L. platymetopon females were characterized. Variation in oocyte diameter, batch fecundity, and relative batch fecundity by standard length and total weight was also veriied, as well as the reproduction sites on the upper Paraná River loodplain. The following stages were recorded: single oogonia, oogonial and oocytes cysts, initial, intermediate and inal primary growth oocytes, full-grown oocytes and maturation oocytes. The reproductive phases were validated by the germ cell attributes and females were recorded in the following phases: developing, spawning-capable, regression and regeneration. Oocyte diameters varied from 100 to 4100 μm. Batch fecundity varied from 372 to 1392 oocytes.g and the relative batch fecundity estimate by length was 16 to 49 oocytes.cm and by weight was 3.6 to 6.4 oocytes.g. The reproductive patterns of L. platymetopon on the upper Paraná River loodplain showed the sites where reproductive activity is more intense (Guaraná, Garças and Fechada lagoons and the Baía River) and sites where it is less intense (Ivinheima and Paraná rivers). Thus, the biotic resistance of the Ivinheima River, due to the integrity of these environments, which is part of the Environmental Protection Area of the Islands and Marshes of the Paraná River, has prevented the reproductive success of this invader.


Introduction
Biological invasion is a major concern for ecologists, since invasive species can cause environmental damage (e.g., loss of native species and loss of ecosystem services), as well as economic harm (PEJCHAR; MOONEY, 2009;SIMBERLOFF et al., 2013).Various hypotheses try to explain invasive success from an environmental perspective and from an invasive species position (BARNEY;WHITLOW, 2008).Understanding these processes is fundamental to the adequate implementation of precautionary and management measures to minimize the impact of these species.
Various problems related to invasive species (e.g., aquatic macrophytes; ish such as carp and tilapia) have been recorded in freshwater aquatic environments (AGOSTINHO et al., 2007;GALLARDO et al., 2016), leading to irreversible problems, since once established it is very dificult to eliminate the species from an environment.There was a massive invasion of 33 ish species in the Paraná River after the suppression of a natural barrier that separated the ish faunas of the lower and upper stretches (sensu ABELL et al., 2008).These species colonized the upper Paraná after the formation of Itaipu Reservoir (JULIO JR. et al., 2009).Among these species, Loricariichthys platymetopon Isbrücker & Nijssen, 1979 was originally distributed in the basins of the Paraguay, Uruguay and lower Paraná rivers (REIS; PEREIRA, 2000) and became one of the most abundant on the upper Paraná River loodplain (DEI TOS et al., 1997;SUZUKI et al., 2000;OLIVEIRA et al., 2001).In addition, this species increased its geographic range and reached several upstream reservoirs in the Paranapanema River (MARCUCCI et al., 2005;SOUTO et al., 2011;CASIMIRO et al., 2017).Skóra et al. (2015) comment that 22 other non-native species were introduced in the upper Paraná River by other vectors, such as ish stocking, aquaculture and release from aquarists, totaling 55 species.
As it is a detritivorous species, L. platymetopon carries out a fundamental role in the nutrient cycling and food web of the sites where it occurs (FUGI et al., 1996;HAHN et al., 2004;TONELLA et al., 2018).It is also prey for larger ish species, birds and mammals and has potential commercial value (QUEROL et al., 1996).
Reproductive strategy is one factor that facilitates the invasive success of a certain species, which can include traits such as length at sexual maturity, life span, spawning month, fecundity, egg diameter and hatching time (WOOTTON, 1984).Loricariichthys platymetopon larvae hatch very developed, in the post-lexion stage Reproduction of invasive Loricariichthys platymetopon on the upper Paraná River loodplain (NAKATANI et al., 2001).Reproductive strategies are largely inheritable traits, linked to the environment, genotype expression and intraspecific spatial and temporal variation related to the inluence of biotic and abiotic factors.These traits are often correlated, and presumably optimized, by natural selection (for revision see MCBRIDE et al., 2015).
The reproductive activity of L. platymetopon can vary from year to year, demonstrating high variability between sites.For example, on the upper Paraná River loodplain, the greatest reproductive activity was from October to January and from October to February in the irst and second period of evaluation, respectively (DEI TOS et al., 1997).These inter-annual variations were related to the intensity and duration of the lood in the reproductive period.Other periods of evaluation for this species on this loodplain showed some reproductive activity from January to August (highly seasonal).More than 30% of the individuals were in reproduction in October, November and December (SUZUKI et al., 2004).Marcucci et al. (2005), for example, demonstrated in Capivara Reservoir in two study periods that this species reproduces between November and February and December and March, coinciding with a higher rainfall.Investigation of this species in Saraiva Lagoon, Ilha Grande National Park (states of Mato Grosso do Sul and Paraná), showed reproductive activity from September to February (BAILLY et al., 2011).Records from a reservoir on Nova Esperança Ranch, Santana Velha, Uruguaiana (state of Rio Grande do Sul) demonstrate reproductive activities of L. platymetopon from November to February (QUEROL et al., 2002).The reproductive strategies of this invader showed alterations in size at irst maturation, spawning period and other strategies, such as male care of the egg mass (conservative for this species).
Despite the large number of articles characterizing reproductive aspects of L. platymetopon, little is known, based on light microscopy, about its germ cells, oocyte development and spawning periods.The spawning sites were evaluated on the upper Paraná River loodplain by Dei Tos et al. (1997) and Suzuki el al. (2004); but after so much time the species may have managed to advance into different sites (e.g., the tributaries).This information is essential to understand how the reproductive framework of the species contributes to its invasive success and high abundance in diverse environments.Thus, this study aims to: i) describe the germ cells of L. platymetopon females; ii) recognize the reproductive phases through the macroscopic and microscopic characteristics of the ovary and gonadosomatic index; iii) verify the variation in oocyte diameter; iv) estimate batch fecundity and relative batch fecundity by standard length and total weight of the population; and v) determine the reproduction sites during the period of investigation on the upper Paraná River loodplain.For this, the reproductive proile of L. platymetopon was traced using several macroscopic and microscopic morphological characteristics of the ovary.Reproductive success was evaluated in different areas of the upper Paraná River loodplain through estimates of fecundity, the gonadosomatic index (GSI) and reproductive phases.

Study area
The samples were collected quarterly (March, June, September and December 2013, 2014 and 2015) (12 samples) on the upper Paraná River loodplain.This stretch of loodplain comprises the only undammed stretch of the Paraná River in Brazil, with the Baía and Ivinheima rivers being fundamental to the maintenance of species biodiversity and loodplain habitat heterogeneity (AGOSTINHO; ZALEWSKI, 1996;REYNALTE-TATAJE et al., 2013).Sampling was carried out at nine sites (Figure 1
Standard length (cm), total weight (g) and weight of the ovaries (g) were recorded for each individual.A sample from each ovary was fixed in Bouin solution for histological studies and a sample from a spawning capable ovary was fixed in 10% buffered formalin to estimate oocyte diameter and fecundity.During the histological procedures, the ovaries were dehydrated in ethanol, embedded in Historesin (Leica) and cross-sectioned (4 μm thick; 2 replicas for staining).The sections were stained using Periodic Acid Schiff (PAS)/ hematoxylin/metanil yellow (QUINTERO-HUNTER et al., 1991) and 0.5 % Toluidine Blue (VIDAL, 1987;MELLO;VIDAL, 1980).The recordings of germ cell development and gonadal development phases were carried out using a microscope coupled to a camera and a Legend: Baía River -1; Ivinheima River -2; Paraná River -3; Guaraná Lagoon -4; Patos Lagoon -5; Garças Lagoon -6; Ressaco do Pau Véio Lagoon -7; Fechada Lagoon -8; Ventura Lagoon -9.
Reproduction of invasive Loricariichthys platymetopon on the upper Paraná River loodplain computer with an image capture program.The calibration of the scales was done using Image-Pro Plus.
The terminology to identify the steps and stages of the germ cells found for L. platymentopon followed Grier et al. (2009) and Quagio-Grassiotto et al. (2011).The reproductive phases were attributed based on the maturity stages of the germ cells according to Brown-Peterson et al. (2011) for marine Perciformes and adapted for Neotropical freshwater ish by Quagio-Grassiotto et al. ( 2013) and Wildner et al. (2013).The gonadosomatic index (GSI) was calculated for each individual and used to evaluate the variation in the different reproductive phases.The index was calculated according to the equation: GSI = GW/TW * 100, where GW = gonad weight and TW = total weight.Three 0.3 g samples were removed from the anterior, middle and posterior region of the right ovary of 18 individuals to estimate fecundity.The oocytes were then separated, measured on a micrometric scale and counted under a stereomicroscope.Batch fecundity (F B ) and relative batch fecundity (number of developing oocytes divided by ish body weight [g -1 ] and standard length [cm -1 ]) were estimated using gravimetric analysis based on the relation between ovary weight and the oocyte density in the ovary (VAZZOLER, 1996;MURUA et al., 2003).The number of eggs spawned per batch (batch fecundity) was estimated based on oocyte size larger than 2500 µm.
The description of the main reproduction sites of the species was done using the female abundance in each reproductive phase found at each sampled site.

Results
A total of 1101 females were sampled.Standard length ranged from 5.3 to 29.5 cm and total weight from 4.2 to 242.8 g.The macroscopic characteristics of the ovary of L. platymetopon showed a hollow, paired, elongated, saccular organ suspended dorsally within the coelom by the mesovarioum.The ovary anatomically varied in appearance at different times in the reproductive cycle.The following phases were recorded: immature, developing, spawning capable, regressing and regenerating.During the reproductive cycle, the GSI ranged from an average of 0.36 to 5.03 (Table 1).
The morphology of the germ cells of the ovary of 255 individuals was examined through light microscopy.
The oogenesis of L. platymetopon was characterized by the development stages oogonial proliferation, primary growth oocytes, secondary growth [vitellogenesis] oocytes and oocyte maturation (Table 2; Figure 2).The oogonia divide and form cell nests in the germinal epithelium.These nests consist of batches of germ cells (oogonia and oocytes) and pre-follicle cells.

Oogonia/Oocytes
The oogonium is surrounded by pre-follicle epithelial cells (Fig. 2A) proliferated by mitosis (Fig. 2B) and a cyst of cells formed in the ovarian stroma (Fig. 2B).The oocytes in the irst meiotic division (Fig. 2C) were recorded in pachytene.

Primary growth
Primary growth was marked by ooplasmatic basophilia and a nucleus with a single nucleolus (Fig. 2D), perinuclear nucleoli (Fig. 2E) and the start of the formation of the cortical alveoli (Fig. 2F).

Secondary growth
Secondary growth (Fig. 2G-J) was characterized by vitellogenesis, in which vitellogenin derived from yolk globules starts to accumulate in the ooplasm and leads to an increase in the oocytes, which, in turn, leads to the formation of initial (Fig. 2G), intermediate (Fig. 2H) and inal (Fig. 2I) secondary growth oocytes and full-grown oocytes (Fig. 2J).

Oocyte maturation
Mature oocyte (Fig. 2K) with the nucleus in the animal pole and detail of the surrounding follicle cells and a theca in the zona pellucida (Fig. 2L).During ovulation the mature oocyte moves from the follicle to the ovarian lumen and becomes an egg, leaving the post-ovulatory follicle complex irregular (Figs.2M-O).Follicle cells and theca cells were observed in the complex.
Reproductive phases and oocyte development based on light microscopy diagnosis were described and recorded (Table 3; Figure 3).

Developing
This phase was characterized by the abundant presence of primary growth oocytes and early secondary growth oocytes and some intermediate secondary growth oocytes (Fig. 3A-F).After regenerating in the beginning of development, some post-ovulatory follicle complexes may be present.

Spawning capable
The presence of primary growth oocytes, early, intermediate and inal secondary growth oocytes, full-grown oocytes, oocyte maturation, atresic follicles and post-ovulatory follicle complex was veriied in this phase (Fig. 3G-I).

Regressing
Primary growth oocytes, early and inal secondary growth oocytes, atresic follicles and several post-ovulatory follicle complexes were recorded (Fig. 3J-L).

Regenerating
The abundant presence of oogonia, oogonial and oocytes nests, and primary growth oocytes was veriied.Post-ovulatory and atresic follicles were also observed in this phase (Fig. 3M-O).Reproduction of invasive Loricariichthys platymetopon on the upper Paraná River fl oodplain Different processes of degeneration of ovarian follicles (atresia) were observed in the secondary growth follicles of L. platymetopon (Figure 2P-R).
The ovary showed germ compartments in the form of lamellae protruding into the internal ovarian cavity.This cavity is continuous with the oviduct.In this species, ovulation in adults occurs in the internal ovarian cavity and the communication of the ovary to the exterior is through the urogenital papilla and is classii ed as cystovarian.
Relative batch fecundity is positively correlated with the total weight and the standard length of L. platymetopon, so that the larger the individual, the higher its fecundity (Figure 5).It should be noted that the total weight possesses a larger coefi cient of determination than the standard length (R 2 = 0.77 and R 2 = 0.68, respectively).that reproduction was more intense in Patos Lagoon and moderate in the Ivinheima River and Garças Lagoon.On the other hand, studies conducted on the loodplain from October/1986 to September/1998 and May/1992 to February/1995 revealed that more than 30% of the individuals reproduced in the Ivinheima and Iguatemi rivers and about 10 to 30% in the Paraná and Baía rivers, in the Corutuba and Cortado channels and in Patos and Garças lagoons (SUZUKI et al., 2004).The results showed greater reproductive activity in Guaraná and Fechada lagoons and the Baía River, guaranteeing reproductive success.On the other hand, reproductive activity was less intense in the Ivinheima and Paraná rivers.This indicates possible factors that can bar the invasion of this species or even factors that can facilitate it.Since access to the sites is similar, it is possible that local factors are responsible for limiting reproductive activity in some environments.A possible explanation is the biotic resistance (HENRIKSSON et al., 2015) of the Ivinheima River, because the diversity at this site is very high and, as a state park, the fauna is protected (since 1998), thus maintaining its original characteristics.
Another factor that explains less reproductive activity in these rivers is that lotic environments may limit male egg-carrying.
Knowledge of these reproductive patterns helps explain how this species has been so successful invading this site.Loricariidae is among the families that possess the greatest number of invasive species in the Neotropical region, so many that its species successfully inhabit the most varied freshwater habitats (ORTEGA et al., 2015;GUBIANI et al., 2018).The reasons for this success are the reproductive behavior and the plasticity found in reproductive parameters like reproduction sites, time of the year and fecundity.In addition to reproducing at various sites, the species cares for its offspring, thus guaranteeing greater reproductive success.
The terminology applied to describe the germ cells of the ovary of Loricariichthys castaneus in Lajes Reservoir (state of Rio de Janeiro), based on light microscopy, was primary oocytes, previtellogenic oocytes, cortical vesicle oocytes and vitellogenic oocytes (DUARTE et al., 2007).The classiications of these authors were based on macroscopic and light microscopy characteristics of the male and female gonads and attributed to the reproductive cycle phases rest, initial maturation, advanced maturation, partially spent/spawned and totally spent/spawned.However, for the same reservoir and species, the following germ cell stages were identiied: oogonia, chromatin nucleolar, perinucleolar, yolk-vesicle formation, vitellogenesis, ripe and post-ovulatory follicles (GOMES et al., 2011).They also recognized immature, maturing, ripe, spawn/ spent and recovering as phases of sexual maturation considering macroscopic and light microscopy aspects of the gonads.As regards Loricariichthys melanocheilus, based on macroscopic evaluation and quantitative indices, recognized phases were: immature, developing, mature and spent for the Ibicuí River (state of Rio Grande do Sul) (ZARDO; BEHR, 2015).In addition, based on the macroscopic evaluation of the gonads of L. platymetopon from Capivara Reservoir in the Paranapanema River, the phases immature, initial maturation, mature, partially-spawned, spent and resting were used (MARCUCCI et al., 2005).
In the upper Paraná River, irst maturation size of L. platymetopon was estimated to be 13.6 cm for males and 14.6 cm for females and all individuals were adults at a standard length of 19.0 cm.Males reached a standard length of 28.2 cm and females 33.6 cm (DEI TOS et  , 1997).Studies conducted in another period on the upper Paraná River loodplain revealed irst maturation size to be 15.7 cm and 14.5 cm for females and males, respectively, and maximum standard length was 33.6 cm (SUZUKI et al., 2004).Studies carried out in the stretch between the Paranapanema and Iguaçu rivers showed that L. platymetopon was common in loodplain lagoons and rare in Itaipu Reservoir and its tributaries (DEI TOS et al., 1997;SUZUKI et al., 2000).According to Suzuki et al. (2000), this species has synchronous oocyte development in batches.The mean oocyte diameter was 2320 µm and relative fecundity (of the batch) was 9.46 oocytes.g - body weight.Studies of L. platymetopon in Capivara Reservoir, situated in the middle stretch of the Paranapanema River, show that the mean diameter of the mature oocyte was 3330 µm, while the maximum diameter was 4140 µm.Absolute fecundity varied from 464 to 850 (mean = 663.95± 141.99 oocytes) (MARCUCCI et al., 2005).These authors found relative fecundity by weight to be between 3.2 and 7.5 oocytes.g - (mean = 4.9; SD ± 1.24 oocytes.g - ).Bailly et al. (2011) recorded the diameter of oocytes varying between 1380 and 2160 µm (mean = 1670) in Saraiva Lagoon.The estimated absolute fecundity varied from 522.9 to 1594.6 oocytes (mean = 962.1;SD ± 382.48 oocytes).Relative fecundity was 7.60 (SD ± 2.25) oocytes.g - and 5.13 (SD ± 0.75) oocytes.cm - .There was little variation between our results and those obtained in the above investigations in different periods and regions, indicating a conservative strategy.This conservative strategy may explain the ease that this species has in colonizing these sites, since all have similar characteristics, both in relation to the low and the history of environmental formation, going from lotic to lentic.Therefore, the probability of invasion success is similar for several environments, since the species also has a consistent physiological and behavioral pattern, as indicated in the results discussed here.
We conclude that evaluation of fecundity by weight and length, associated with information about reproductive areas based on macroscopic and light microscopy evaluation of the ovary, is an efficient management tool to show variations in the reproductive cycle.The individuals of this species are sedentary, possess a detritivorous feeding habit, predominate in lentic environments, generally reproduce in these environments, show synchronous oocyte development, are batch spawners and the males carry the eggs under the lower lip (FUGI et al., 1996;DEI TOS et al., 1997;SUZUKI et al., 2000;HAHN et al., 2004).These ecological characteristics permit their occupation and reproductive success on the loodplain, especially in the lagoons, as these attributes give them competitive advantages over the natives.Thus, the biotic resistance of the Ivinheima River, due to the integrity of this environment, which is part of the Environmental Protection Area of the Islands and Marshes of the Paraná River (Área de Proteção Ambiental das Ilhas e Várzeas do Rio Paraná), has prevented the reproductive success of this invader.
The gonadal development phase of each individual was determined macroscopically, considering the size and shape of the ovary, occupation in the abdominal cavity, evidence of vascularization, color, visualization of the oocytes and l accidity, according toBrown-Peterson et al. (2011) and Quagio-Grassiotto et al. (2013).

FIGURE 1 :
FIGURE 1: Study area and location of the sampling site on the upper Paraná River l oodplain.

FIGURE 2 :
FIGURE 2: Light microphotography representative of stages and steps of germ cells of L. platymetopon females from the upper Paraná River l oodplain.

FIGURE 3 :
FIGURE 3: Light microphotography of the ovarian histology representative of the reproductive phases of L. platymetopon from the upper Paraná River l oodplain.

FIGURE 4 :
FIGURE 4: Frequency of the diameter of the oocytes of L. platymetopon from the upper Paraná River l oodplain.
The frequency of individuals in different reproductive phases per sampling site revealed reproductive success in Guaraná and Fechada lagoons and the Baía River (n = 678), followed by Garças and Pau Véio lagoons (n = 250) (Figure6A).A smaller number of individuals were found in Patos and Ventura lagoons and the Ivinheima and Paraná rivers (n = 173).Light microscopy coni rmed the reproduction areas of this species (Figure6B).

FIGURE 5 :
FIGURE 5: Relation between relative batch fecundity and total weight (A)/standard length (B) of L. platymetopon from the upper Paraná River l oodplain.
Reproduction of invasive Loricariichthys platymetopon on the upper Paraná River loodplain al.

TABLE 1 :
Diagnosis of the reproductive phases of L. platymetopon females from the upper Paraná River loodplain.

TABLE 2 :
Diagnosis of the different stages of oocyte development of L. platymetopon females from the upper Paraná River loodplain.

TABLE 3 :
Diagnosis of the reproductive phases of L. platymetopon females from the upper Paraná River loodplain.

TABLE 4 :
Overview of reproductive strategies of L. platymetopon from the upper Paraná River l oodplain.