Exotic Pumpkinseed Sunfish Lepomis gibbosus (Linnaeus, 1758) in the International Minho River (Iberian Peninsula), and Parasitic Association with Myzobdella lugubris Leidy,

To evaluate the establishment of the emerged species Lepomis gibbosus (Linnaeus, 1758) in the international section of the Minho River (Iberian Peninsula) the abundance was analysed for 7 years. This study also describes the world’s first record of parasitism between the leech Myzobdella lugubris Leidy, 1851 and L. gibbosus. The earliest record of species L. gibbosus , also known as pumpkinseed sunfish, was in the Spanish area of the Minho River basin, in the beginning of the 2000s. It is considered an established species in the estuary, due to its ecological release capacity, which has been increasing over the last few years. Physical condition, age, feeding patterns, and reproduction habits were analysed from July 2014 to October 2015. Individuals sampled appear in a good physical condition (b>3). The oldest individual captured was 4 years old, with L max =18.6cm. Females have a higher gonadosomatic investment than males (F 1;585 =150.43, p<0.01), particularly in July (F=7.39%; M=1.4%), which comply with the values of the hepatosomatic index (F 2;300 =34.23, p<0.01). Breeding season occurs between July/August and sexual maturity is reached at age of 1. Regarding the feeding regime, 18 taxa were identified from gut content. In the first year of life, the pumpkinseed has a preference for insects and crustaceans. In adult age, gastropods are the preference, which can cause trophic competition with native species. Therefore, monitoring studies should continue to follow the population dynamics.


Introduction
In nature, ecosystems are dynamic environments that have evolved over billions of years to reach the actual natural barriers that surround them [1]. However, direct, and indirect anthropogenic impacts can lead to local and global changes by breaking those barriers, and provoking changes in modern species distribution. The spread of invasive species is the main cause of serious disturbances, instigating harmful effects on natural ecosystems worldwide and is one of the main reasons for local biodiversity loss, particularly in aquatic ecosystems [2,3]. Furthermore, the introduction of new pathological organisms (e.g., via associated parasitic species) and 'genetic actuation' (e.g., hybridization and introgression) are other impacts that may affect the native species [4].
The River Minho is in the Iberian Peninsula and its international section separates North Portugal from South Galicia (Spain), over the last 75 km of the 300 km of total length [5]. According to the Portuguese Environment Agency, the Minho River is considered to have good water quality [6]. The ichthyofauna is composed of 50 species [7], and some of them are classified under Natura 2000 Minho River site including diadromous fishes: sea trout (Salmo trutta Linnaeus, 1758), allis shad (Alosa alosa (Linnaeus, 1758), twaite shad (Alosa falax (Lacepède, 1803), sea The pumpkinseed sunfish Lepomis gibbosus (Linnaeus, 1758) is natural from North America, and it was introduced in Europe in 1880 where it had a huge expansion in several countries, accompanying the introduction of other species via sport fishing and as an ornamental species, in the aquariums [9]. The earliest record of Lepomis gibbosus on the Iberian Peninsula was between 1910-1913, identified at Lake Banyoles, northern Spain. In Portugal, it was first recorded in 1977 in the Guadiana River [10]. The introduction of this species on the Iberian Peninsula would have occurred accidentally, at a similar time as the insertion of the largemouth black bass, known as Micropterus salmoides (Lacepède, 1802) [9]. Other examples of accidental introduction of the pumpkinseed sunfish identified in Europe, which also coincided with the introduction of the common carp (Cyprinus carpio Linnaeus, 1758), includes England and Czech Republic [9].
In the Minho River, under the Natura Miño-Minho project (INTERREG 0234_Miño_Minho_1_E), pumpkinseed has been detected in Spanish dams in the early 2000s, while in the Minho River estuary the capture of one specimen was recorded in 2007 [8,11]. This study, therefore, aims to perform the first characterization of the species L. gibbosus in the international Minho River section and to assess the impacts it may cause on established native species. Additionally, association parasitic with the leech ectoparasitic species Myzobdella lugubris Leidy, 1851 is described for the first time in Europe, as well as on a global scale.

Study Area
The Minho River is located in the located in the NW-Iberian Peninsula (SW Europe; Figure 1) and extends approximately 300 km to the Atlantic Ocean. The last 70 km serve as a northwestern border between Portugal and Spain. The tidal influence extends until 40 km upstream comprising the estuarine area of approximately 23 Km 2 , however, saltwater intrusion reaches only the first 25 km [8]. River flow varies seasonally: the maximum flow (2500 m 3 .s -1 ) is associated with periods of intense precipitation in the winter/early spring, and the minimum flow (60 m 3 .s -1 ) occurs in the summer/early autumn, associated with periods of drought [12]. The international section of the Minho River is classified as a Natura 2000 site due to its ecological importance.

Sampling Strategy
Sampling took place between January 2013 and September 2019 and the more species-specific data, such as physiological indices, feeding regime and age analysis comprised samples obtained from July 2014 until August 2015, using five fyke nets (10 mm mesh, 7 m in length with two openings in funnel shape) in Marina da Lenta, checked twice a week. In each sample taken, water parameters were recorded (temperature, pH, salinity, depth, and conductivity), with a YSI 6820 multiparameter probe (USA).

Data collection
All captured individuals were measured for total length (TL) and fork length (FL) (0.1cm) and weighed (0.01g). Stomach contents were collected and subsequently analysed, according to the Knoöpell [14] methodology. After the evisceration and scrutinized through the magnifying glass, each prey item was identifying to the lowest taxonomic level possible, gathering all in seven groups: Crustacea, Monogononta, Mollusca, Insect, Arachnida, Fish and Vegetation [15][16][17]. After the extraction, otoliths were washed in 70% alcohol to remove organic matter [18]. To a proper age analysis, the option followed was the staining technique [19]. Otoliths were placed in gelatin capsules, correctly identified with paper strips, and then filled with resin (Resin EpoThin No. 20-8140 and the catalyst EpoThin No.20.8142, with the proportion 100:39, respectively).
After 9h drying, a transversal axis cut was made in the capsule and glued on a slide [20]. Posteriorly, each side of the capsule was softly polished and stained with toluidine blue (1%) for one hour. To finish the process, the capsule was washed with clean water [21] and EDTA acid was applied for five minutes. To better support the age analysis, scales were also used to read rings, according to Regier [22]. Scales were collected from every individual, with different sizes, and washed with KOH 5%, to remove all the exceeding organic matter. Next step, using a magnifying glass, scales with the focus well defined were selected and measured (software Cell^b -Basic Research imaging software -Olympus).

Data analysis
Abundance was expressed in CPUE (catch per unit effort) represented by the CPUE=n/e/t, where n is the number of individuals captured by fyke nets, e is the number of fyke nets used and t is the unit of time. Using gonads and liver weight, the gonadosomatic index (GSI = gonad weight (g) / body weight (g) x 100) [23] and hepatosomatic (HSI = liver weight (g) / body weight (g)) [24], respectively, serving as indicators of the reproductive season [25]. The length-weight equation W = a L b was used to estimate the relationship between the weight (g) of the fish and its total length (cm). Using the linear regression of the logtransformed equation: log (W) = log (a)+b log (L), the parameters a and b were calculated with a representing the intercept and b the slope of the relationship.
The value of b may deviate from the 3 that represents an isometric growth due to environmental features or fish condition [26]. Prey items were quantified using the relative frequency of occurrence of each food item and expressed as the percentage occurrence of all food organism [27]. Prior to the statistical analyses, normality and homoscedasticity tests were applied, using Shapiro-Wilk and Levene, respectively. Differences between groups were assessed through a distribution-free analysis of variance (One-Way, ANOVA), (Student's t-tests, Two-Sample), (Test X 2 ) using the STATISTICA software, version 8.0 [28]. Variations of physical condition and gender ratio were tested to verify any differences, performing the non-parametric Kruskal-Wallis test.  The abiotic factors registered between 2014 and 2015 are expressed through PCA in Figure 3. According to graphical analysis, temperature, and transparency (Secchi) were the most driving factors, explaining the months distribution, with a clear dissimilarity between summer ones (May, June, July, August, and September) and the winter period (from November 2014 to March 2015). The highest values of transparency were recorded between November and January 2015, with the highest value in December (3.1 ± 0.67m). The highest values for temperature were recorded in June, July, and August, above 22°C. The lowest temperature was recorded in January 2015 (8.86 ± 0.65 °C). Simultaneously, in July 2014, the highest mean values of salinity (0.08 ± 0.03 ups) and the highest mean values of conductivity (0.18 ± 0.07 mS/cm) were observed.

Physical Condition
The relationship between weight-length was calculated based on the data collected from measurements taken in 2013 until 2019, of a total of 12562 individuals from the River Minho ( Figure  4). L. gibbosus population presented a minimum length of 2.    For this time 1076 individuals (8.6% of the total data) were analysed regarding their gender, age, and diet, collected between July 2014 and September 2015. Fulton's factor condition was calculated for both genders of L. gibbosus, showing no significant differences between males and females (F 1;1042 =2.68; p>0.05). However, over the sampling period, both females (F 12;560 =43.10; p<0.01) and males (F 13;457 =28.76; p<0.01) show significant differences. In general, there is a slightly higher k value in females when compared to males, except for November 2014 and April 2015. The highest values reached for both males and females were in August 2014, with k=2.01 and k=2.13, respectively. The month with the lowest physical condition was in April 2015 for females (k=1.43) and in December for males (k=1.43) (Figure 6). For individuals caught in the same months (Aug/2014 to Sep/2015), according to their gender and age (0 to 3), and considering the data normality, there were no significant differences (t-test = 1.64; p>0.05) for physical condition (1.78 ± 0.33 for males and 1.81 ± 0.24 for females) ( Table 3).

Age and Growth
The relation between TL (total length) of pumpkinseed and their age was established essentially through the analysis of scales, expressed in the following equation y =3. According to the total number, pumpkinseed with 1 + were the most collected over the years. In the second year, the mean length increased to 11.9 cm, with a minimum value of 14.  (Table 4). Table 4: Pumpkinseed individuals age structure in the River Minho, from 2013 to 2019. The number of specimens (N) and de mean of length (X) in cm are given, as the total number of individuals captures for each age group. Minimum and maximum length values (in cm) for each year and age group are given in brackets. Total  2014  2015  2016  2017  2018  2019 0 + X 5.5(4.4-6.4) 5.6(4-6.9) 5.7(4.2-6.9) 4.7(2.6-6.9) 6.0(3.3-6.9) 5.9 (2.9-6.9) 6.4 (2.7-6.9) 5.9

Diet composition
Between July 2014 and September 2015 there were 1077 individuals collected, and only 532 were analysed regarding their stomachal content, which 269 (50.6%) individuals had an empty stomach and 263 (49.4%) showed traces or entire prey inside, allowing their identification. In total were identified 18 taxonomic groups, being that 216 individuals had preys possible to identify, which are given in Table 5. The most important item is Diptera (44%) which appears in almost every age group, followed in second by Gastropoda (31.5%) also the preference in ages of 2 + and 3 + , and in third by Cladocera (27.8%), which is highly consumed in age 0 + . • Material Examined: International River Minho, Iberian Peninsula section: Vila Nova de Cerveira municipality (Portugal), 1 adult with 11.5 cm length (NatMIP-CAPe-0002); 2 adult specimens, being one with 15.5 cm and another with 9.5 cm length (NatMIP-CAPe-0008); 3 adult individuals, 13.6, 11.5, 14.

Oceanography & Fisheries Open access Journal
• Diagnosis: L. gibbosus is distinguished from the other species of the family Centrarchidae through particular morphological characteristics, such as the smaller operculum (dimensions similar to those of the eyes), dark colouration with lighter points in reddish or orange tones, 3-4 anal spines, 36-37 scales on the side of the body, the posterior-ventral portion of the dorsal and anal area without a dark mark, small mouth, length of the upper jaw is equal to the diameter of the eye (Figure 10a) [29,30].

Oceanography & Fisheries Open access Journal
• Diagnosis: This species has an elongate greenish to brownish body, tapering anteriorly and enlarged on posterior region, with a little flattening (Figure 11a, b). Rounded anterior oral sucker with two eyespots (Figure 11c); caudal sucker on posterior end shorter than maximal body width (Figure 11d). Median segments with 12 to 14 annuli. Eversible bursa with an atrium projecting as obtuse cone-shaped, carrying an orifice [31,32]. This leech species, and its parasitic association with L. gibbosus, is now recorded for the first time at the International Minho River, as well as in the European continent. Parasitism association with this pumpkinseed species is considered as the first record in the world.

Discussion
The Spanish Minho River basin is highly fragmented with dams and the first record of pumpkinseed results from work under the Natura Miño-Minho project carried out in Ourense area dam reservoir ("albufeira", in Portuguese) [11] which mentions their presence since 2000. Although one specimen was captured in the Minho River estuary [8], the present work records its establishment in tidal freshwater wetlands from 2013 onwards. A possible pumpkinseed downward movement may have occurred after its introduction in the Ourense area. The monitoring programme with fyke nets in the Minho River tidal freshwater wetlands began in 2007 [7] on a continuous basis and it was only from July 2013 that regular catches of L. gibbosus occurred.
Individuals live between eight to ten years [9], preferring water with a slow flow or even totally stopped, and it can live in cooler waters than any other sunfish [33]. Temperatures in Minho River suffer an amplitude of 17°C, from the minimum (8°C) to the maximum (25°C), similarly with the results of Antunes et al. [12], being the parameter with the highest variance throughout each year. This reflects the number of captured individuals in the hottest months, when conditions are also likely to favour pumpkinseed sunfish, resulting in higher densities, greater dispersion [34] and breeding activity, which occurs when temperatures are between 13 to 28ºC [35][36][37].
Besides this, other parameters were considered, although their impact on data distribution is less significant on L. gibbosus Oceanography & Fisheries Open access Journal population, i.e., pH, conductivity, and salinity, which can be supported by this species up to 18.2% [30]. The depth of the sampling site was relatively continuous throughout the other years, staying between 3 to 4 m, as well as for salinity (between 0.02 to 0.11 mg/L), conductivity (±140 mS/cm), and pH (between 6 to 8). A specifical analysis was made between 2014 and 2015 when L. gibbosus start to manifest more regularly in fyke nets. This period seems to reveal a more adaptative response to the Marina da Lenta habitat. After 2015, captures were registered regularly every month, showing a possible stabilization of L. gibbosus population. According to Tola and Infiesta [38], the average length reached by the pumpkinseed in America is around 30 cm, while in the Iberian Peninsula, more precisely in the southwest area individuals have lengths between 8 to 15 cm. In the Minho River estuary, the length average increased over time and reached the maximum value of 18.6 cm in 2016.
Reproduction occurs between the first and third year of life, from April/May until June/August when temperatures reach 16-18°C. The nests are built and maintained (ventilated and protected) by the males. They guard the nests for a period of 3 to 10 days until the young disperse [39]. According to Fox and Keast [40], the L. gibbosus population reaches maturity early and increases reproducibility in adverse environmental conditions, which results in great phenotypic plasticity due to the accumulated threats through the course of the pumpkinseed life history. Native American populations are more studied because most of the Lepomis genus is particularly important as a game species [41], while European populations received relatively little attention, with limited studies on individuals' growth and age. However, the perception of this species as an invasive species has led to an increased need for knowledge.
Most pumpkinseeds in both native and introduced populations have been found to achieve maturity during their third year of life (age 2 to age 3 increment), and would thus be considered adults [2,9]. However, strong juvenile growth and precocious maturity, with a shorter lifespan, appear to be adaptive responses to elevated water temperatures [42]. Indeed, almost all the populations from southern Europe, where the species is considered invasive, exhibit fast juvenile growth and early maturity [43]. In the case of Minho River, the individuals collected and analysed regarding their age structure, between 2014 and 2015, showed superior values of SL for each age group, when compared with other northwestern European populations. These differences resulted mainly from the influence of environmental conditions, such as the temperature that will have a directly proportional correlation and feeding [29].
Although omnivorous, the pumpkinseed tends to feed on prey with greater abundance, with the predation of fish eggs having the most impact on the ecosystem [9]. According to Sadzikowski and Wallace [44], juvenile feeding is mainly composed of microcrustaceans (Cladocera, Copepods) and Chironomidae larvae, while in the adult phase, gastropods are the preferred prey [39]. This type of preference appears to be justified by the morphology of the pharyngeal apparatus and its musculature, specialized in gastropod feeding, which may have resulted from a more primitive benthic feeding [2,9,29]. It was found the most important food items to be Diptera and Gastropoda in older individuals while Cladocera dominated in the younger age class.
In Europe, there are very few natural predators, like sander (Sander lucioperca) and the northern pike (Esox lucius) [9], absents in the River Minho, and his biggest competitor is the bluegill (Lepomis macrochirus) also absent in Portugal [10]. In the Minho River, the potential predators are the otter (Lutra lutra) and the cormorant (Phalacrocorax carbo). In addition to being proven that this species has contributed to the decline of native species [9], changing the ecosystem, it is also proven that they are able to take advantage of human activities for their benefit: the construction of dams provide an upstream environment with less flow variation and less current, which facilitates the construction of nests and the nutrients resulting from agriculture guarantee also better food availability [2].
The parasitic performance of M. lugubris with host fishes, adhering on its fins and other skin surfaces has been observed in other studies [31,32,45,46]. In Canada (Lake St. Clair), for example, these leeches are found on dorsal, pelvic, pectoral, and caudal fins Percina caprodes (Rafinesque, 1818), and Ameiurus nebulosus (Lesueur, 1819) [47], and attached into to pectoral fins or beneath the inferior jaw of M. salmoides [48]. In North Carolina (USA), intense ulcerations of the tongue and buccal cavity were reported in M. salmoides, associated with species M. lugubris [49].
Some studies mention the species Piscicolaria reducta Meyer, 1940, including, for example, its ectoparasitic association with Lepomis cyanellus Rafinesque, 1819, Lepomis macrochirus Rafinesque, 1819 and Lepomis punctatus (Valenciennes, 1831) [50]. Notwithstanding, this leech species may be the same as M. lugubris, and their sequences' GenBank are probable a misidentification because the phylogenetic tree and consensus tree from Maximum Parsimony and Bayesian Inference indicate P. reducta nested within the same clade of M. lugubris [32].

Conclusion
Pumpkinseeds populational dynamics/characterization is now described for Minho River, at the international section area. The species L. gibbosus, being an invasive organism in the Minho River, provoked changes in the aquatic environment and influenced in local abundance of fish species, such as M. salmoides. The leech species M. lugubris is recorded for the first time in European waters; and its parasitic association with L. gibbosus is now noticed as the first record in the world, representing an important parasitic organism on host sunfishes' individuals.