A Histological Study of Ovarian Development in the Giant Red Shrimp Aristaeomorpha foliacea (Crustacea: Decapoda: Aristeidae) from the Southern Tyrrhenian Sea (Western Mediterranean)

The reproductive features of the giant red shrimp, Aristaeomorpha foliacea, were investigated in the southern Tyrrhenian sea by experimental trawl sampling. The annual length-frequency distribution showed a multimodal trend in females, ranging between 16 and 67 mm carapace length (CL), and a unimodal trend in males (18–45 mm CL). Mature males occurred in different proportions all year round, while females displayed seasonal maturity (June—September), with a peak in July. Six oocyte developmental stages were identified, the most advanced of which (Pv, postvitellogenic) had never been described before in this species. Ovary development followed a group-synchronous pattern, with the yolked oocyte stock clearly separated from the reservoir of unyolked oocytes, suggesting that A. foliacea is a total spawner, with determinate fecundity. Based upon histological findings, a revision of macroscopic maturity staging employed in Mediterranean bottom trawl surveys (MEDITS) is proposed.


Introduction
The giant red shrimp Aristaeomorpha foliacea (Risso, 1827) (Decapoda, Aristeidae) is a widespread decapod crustacean in the eastern and western Atlantic, western Pacific, the Indian Ocean, as well as the Mediterranean Sea, showing a preference for muddy bottoms, with aggregation mainly in submarine trenches and canyons along the continental slope from 150 m to 1850 m with a peak in abundance between 300 m and 700 m [1,2].
Together with the other red shrimp Aristeus antennatus (Risso, 1816), A. foliacea represents an important target species of commercial fleet fishing off the southern Tyrrhenian Sea operating over the continental slope, with total landings of the order of one hundred tons [3]. Because of its economic value, many studies have been carried out on the biology, population ecology, and fisheries of this shrimp [4][5][6][7][8][9][10][11][12][13].
Reproductive biology is one of the main concerns in formulating proper management practices for fishery science. In fact, the relationship between size and reproduction events, such as mating and sexual maturity, is crucial to understand the dynamics of a stock. Knowledge of reproductive biology of A. foliacea has mainly come from studies on ovary maturation of fresh specimens by means of macroscopical investigations [10,[13][14][15][16][17][18][19].
On the other hand, few studies examined histology of ovarian development [20][21][22][23], and many aspects, such as the spawning pattern and type of fecundity, remain to be solved. International commissions for fisheries study (i.e., ICES, International Council for the Exploration of the Sea, GFCM, General Fisheries Council for the Mediterranean) strongly recommend standardizing criteria for evaluating sexual maturity in target species. Recently, the need for a common system for classification of maturity stages in Crustaceans has  been highlighted by an Expert Group of ICES (2010), which also pointed out the importance of histology as a tool for obtaining the highest accuracy in these studies. This paper describes for the first time some reproductive features of A. foliacea caught in the southern Tyrrhenian Sea, such as monthly maturity stages by size and length structures by maturity stage. Moreover histological analyses of ovaries were performed in order to validate the empirical, macroscopic scale used in experimental trawl surveys through a detailed description of maturity stages based upon oocyte development. The information provided by this paper is intended to contribute to establishing production and population dynamics models, for a rational exploitation of this important fishery resource.

Biological Data and Reproductive Parameters.
Red shrimps were caught on a monthly basis (September 2008 through August 2009) by commercial trawlers operating in the southern Tyrrhenian Sea, between the Gulf of Patti and Aeolian Islands, in a depth range from 500 to 700 metres ( Figure 1). The sampling was conducted in order to sample all stages of maturity. Specimens (n = 21 163) were sexed and measured for carapace length (CL) to the nearest of 0.01 mm. The overall length frequency distributions (LFDs) of males and females for all year were estimated. The differences between the two sexes were tested by Kolmogorov-Smirnov test (P ≤ 0.001). Body weight (BW) was recorded to the nearest 0.01 g. Gonad weight (GW) of females was recorded to the nearest 0.001 g, and gonadosomatic index (GSI) was estimated as (GW/BW) × 100.
On the basis of the colour and shape of the ovary and testis, each specimen was classified according to the macroscopic maturity scale adopted in MEDITS programme (Table 1). Males maturity was checked by the presence of the spermatophores in the terminal ampoullae of vasa deferentia. Presence or absence of spermatophores in the thelycum of females (inseminated specimens) was also recorded.
The length structures by maturity stage were estimated by analyzing the box-plot representation. Statistical differences between maturity stages and carapace length were tested by Kruskal-Wallis test (P ≤ 0.001).

Ovarian
Histology. At least 10 females of each maturity stage were fixed on board in 10% neutral formalin, after cutting the integument. Once in the laboratory, after washing in running tap water for 4 hours, ovaries were dissected, dehydrated in a graded series of ethanol, embedded in resin, and cut in 5 µm thick sections. Sections were stained with haematoxylin-eosin and observed under a light microscope for histological description of oocyte development and ovarian maturation. Oocyte size was estimated by the imaging software Leica IM1000 v 1.20, by taking the average between the major and minor axis in rounded oocytes, and by measuring the major axis (length) of rectangular oocytes, for the sake of comparison with previous studies [23].

Population Structure.
The population structure of females and males is presented in Figure 2. The size of females ranged from 16 to 67 mm CL, showing a multimodal trend during the year. The most abundant individuals had carapace length generally between 27 and 30 mm.
The males measured between 18 and 45 mm CL and exhibited a unimodal distribution. A significant difference in LFD between males and females was found (P < 0.001).

Gonadal Maturation.
Females of Aristaeomorpha foliacea displayed seasonal maturity. Ovarian maturation started in April (stage 2c = 8%). Mature females (stage 2d) were found from June through September, with a peak value in July (2d = 38%) (Figure 3   throughout the year, while specimens with immature ovaries (stage 1) were caught from February to September. The smallest female with mature ovaries (stage 2d) was observed in July and measured 41 mm CL. Only larger females (CL range: 41-66 mm) contributed to the reproductive pool, which, taking into consideration both maturing (2c) and mature (2d) specimens, represented about 50% of total during the spawning peak.
The occurrence of reproductive males of A. foliacea, characterized by the presence of spermatophores in the terminal ampullae, was constant and very high during the whole sampling period, with values ranging from 59% to 100% ( Figure  3(b)). The size range of mature males ranged from 25 to 45 mm CL.
The size at maturity stage analysis, made on all combined months, indicated, for both sexes, a satisfactory resolution capability of the adopted scale. For females the 2a stage prevailed (36.7%) while for males the 2d stage prevailed (90.4%) ( Table 2).  Relation between median and maturity stages showed significant differences for both sexes (P < 0.001), except between stages 2b and 2c (P = 0.22) and stages 2b and 2e (P = 0.30) for females (Figure 4).

Histology of Ovary Development.
Six oocyte developmental stages could be distinguished from the histological analysis of ovaries ( Figure 5), namely, oogonia (Oo), early primary oocytes (Ep), late primary oocytes (Lp), early vitellogenic oocytes (Ev), late vitellogenic oocytes (Lv), and postvitellogenic oocytes (Pv), whose microscopic features are synthetically reported in Table 3. Post-vitellogenic oocytes, which have never been reported so far in A. foliacea, were found in some specimens macroscopically staged 2d (mature) ( Table 1). These oocytes had the same size and yolk inclusions as the late vitellogenic ones (350-470 µm), but their The Scientific World Journal 5    nucleus was not visible, and they displayed slightly eosinophilic, columnar cristae protruding into the outer cytoplasmic cortex ( Figure 5(e)), strongly resembling penaeid cortical rods [24,25]. For these reasons, such oocytes were considered to have achieved vitellogenesis and started final maturation prior to ovulation. Another type of oocyte was found, that is, Lp oocytes, showing massive vacuolization of cytoplasm ( Figure 6(e)). As these oocytes occurred only in spent specimens, they are likely to be abortive oocytes at an initial stage of atresia (Ao), destined to be resorbed.
The histological description of macroscopic maturity stages employed in MEDITS surveys is reported in Table 4 and Figure 6. Immature ovaries (stage 1) showed an even connective stroma populated by oogonia, early primary oocytes, and occasional late primary oocytes. In all other specimens, the ovarian parenchyma consisted of germinal and maturative zones, the former containing oogonia and early primary oocytes and the latter containing later oocyte stages. Stage 2a (virgin developing) and 2b (recovering) ovaries were histologically identical, with Lp oocytes as the most advanced developmental stage. As ovary maturation proceeded, developing oocytes occupied most of the ovarian parenchyma, leaving only small germinative islets of resting oocytes interspersed between them. Oocyte development in the maturative parenchyma occurred in a synchronous way, as indicated by a single batch of oocytes at the early vitellogenic (Ev) stage in maturing (2c) ovaries and a single batch of oocytes at the late vitellogenic (Lv) or post-vitellogenic (Pv) stage in mature (2d) ovaries. Resting (2e) ovaries were characterized by hyperaemia of ovarian stroma and marked proliferation and hypertrophy of mesodermal (somatic) cells lining tubular units. The latter were populated by atresic oocytes (Ao), among which some residual Pv oocytes could be detected.
The lowest and highest GSI values of females examined in the present study were 0.006 (stage 1 specimen with CL = 25 mm) and 13.87 (stage 2d specimen with CL = 46 mm), respectively. The mean GSI values for each maturity stage are reported in Table 4.

Discussion
The giant red shrimp, Aristaeomorpha foliacea, is among the most prized demersal resources exploited by deep-sea fishery in the southern Tyrrhenian Sea. The importance of this resource to Mediterranean fisheries is supported by the results on the biology and ecology of this species, obtained from more than twenty years of trawl survey programmes The reproductive season of A. foliacea in the studied area, as indicated by presence of mature females (June-September), is comparable with that one reported along the Sardinian coasts [26], but it is shorter than in northern areas of western Mediterranean [16,20], as well as than in eastern Mediterranean [18,19]. On the other hand, the occurrence of mature males (≥50%) in all months, suggesting no evidence of a seasonal maturity cycle, seems to be a constant reproductive feature of A. foliacea all over the Mediterranean. Such a protracted reproductive capacity of males could be related to mating activity. In fact, inseminated females occurred all year round, as in all Mediterranean populations studied [16][17][18][19]. Inseminated females were staged 2b through 2e, suggesting that mating may take place also after oviposition. On the other hand, immature (stage 1) and developing virgins (stage 2a) were never inseminated, suggesting that ovary maturation is triggered by mating. In other areas, resting females bearing spermatophores were reported in proportions ranging from 6% (west Pacific) [22] to 60% (eastern Mediterranean) [19]. However, while ovary maturation has been reported to follow mating by four months in eastern Mediterranean [19], it would seem that it occurs within one month from mating during the first reproductive cycle in our western Mediterranean population. The lower proportion of mature males from April to September was not paralleled by a decreased number of inseminated females in the same months, suggesting the possibility that males are able to mate with multiple females, as hypothesized in Aristeus antennatus [19].
The fully mature female minimum size of A. foliacea in the present study (41 mm CL) is by far the largest ever recorded in the Mediterranean (27-40 mm CL), confirming a westward gradient in minimum size at maturity [18]. The size at maturity stage analysis, made on all combined months, indicated, for both sexes, a satisfactory resolution capability of the adopted scale.
A reliable maturity staging is needed for a correct evaluation of spawning-stock biomass (SSB) (i.e., the biomass of reproducing individuals in a population), which is crucial in 8 The Scientific World Journal stock assessment studies. In fact, it has been demonstrated that macroscopic staging of the gonads may lead to overestimating SSB by up to 35% in fish [27]. For this reason, histological analysis of gonads is recommended to validate macroscopic observations, as well as to standardize criteria for comparing maturity data from different populations [28]. This is particularly true for crustacean species, whose reproductive features are less documented than those of fish.
The GSI values of the different maturity stages were similar to those from other studies, except for the immature stage, which was considerably lower than those reported not only in Taiwan and eastern Mediterranean [19,22], but also in a population of western Mediterranean [21].
The histological observation of ovaries in A. foliacea allowed us to detect a novel oogenic stage, which had not been described in previous studies [21][22][23], namely, the postvitellogenic stage. Such stage was characterized by the presence of cortical cristae, resembling the "cortical rods" of penaeid shrimps [29,30]. Cortical rods are formed after completion of yolk accumulation and are involved in the cortical reaction initiated by contact with sea water [29], by forming a jelly layer around the egg, preventing polyspermy. In penaeid oocytes, which possess a vitelline envelope, cortical rods develop outside the oocyte plasma membrane but under the vitelline membrane [30]. In our study, light microscopy showed that cortical cristae are housed in crypts formed by invaginations of the oolemma, and therefore they are extracellular, as cortical rods.
Ovary maturation in A. foliacea involved a large batch of oocytes developing synchronously inside tubular units, with resting oocytes remaining in the germinative islets between them. Prespawning ovaries were almost completely filled with post-vitellogenic oocytes, to be released at the next spawning, while oogonia and oocytes at the previtellogenic stages remained in the ovary, with the most advanced of them (Lp oocytes) undergoing degeneration after spawning. Such features are indicative of a group-synchronous ovarian developmental pattern [31] and strongly suggest that A. foliacea has a determinate fecundity, that is, fecundity is fixed before the onset of spawning [32]. It is likely that A. foliacea is a total spawner, that is, all yolked oocytes are released in a unique spawning event during each reproductive season, as A. antennatus [33].
For the first time, the histological features of spent ovaries were described. In these ovaries, which was found in August and September, the architecture of ovarian stroma appeared modified, by proliferation and hypertrophy of mesodermal cells, accompanied by hyperaemia, while pre-vitellogenic oocytes (Lp stage) left in the ovary after spawning of mature eggs displayed severe vacuolization of the cytoplasm. This is likely to be an initial phase of atresia, suggesting future resorption of these oocytes, while the oocytes destined to develop in the next reproductive season may be recruited from the batch of early primary oocytes left in the ovary. The occurrence of the "spent" stage in A. foliacea has been debated, since the first observations [21], which lead [22] to hypothesize that females may die after spawning, or move to waters deeper than 600 m, where they may be missed by trawlers. On the other hand, [19] reported 16% spent specimens from July, consistently with the 1-month-advanced spawning in that area. Our results suggest that A. foliacea starts reorganizing the ovary soon after spawning, and this process lasts about 2 months, after which a recovering ovary with no signs of previous spawning or oocyte resorption is found.