INTRODUCTION

Sex change is a frequent phenomenon in tropical reef fish (Warner, 1984; Adams and Williams, 2001). Change from female to male is defined as protogynous sex change (Munday et al., 2006). It has been reported that 14 families of tropical fish exhibit protogynous sex change (Warner, 1984; Walker et al., 2007). The Serranidae family and its subfamilies are well known as protogynous sex-changing fish (Fishelson, 1970).

Red-belted anthias Pseudanthias rubrizonatus belongs to the Serranidae family and the subfamily Anthiinae (Senou, 2013). Sixty-two species of the Pseudanthias genus are found in subtropical and tropical zones (Nair, 2008). P. rubrizonatus is globally distributed in Japan, Taiwan, Hong Kong, the Gulf of Tonkin, and the West Pacific (Senou, 2013). In Australia, P. rubrizonatus lives around artificial structures located between 82 and 135 m in depth. Its total length ranges from 16.9 to 96.5 mm (Flower and Booth, 2012). In Japan, P. rubrizonatus is distributed from the Izu Islands to Kagoshima. Kagoshima Bay is located in the temperate zone of south Japan (Fig. 1). It is one of the deepest bays in Japan, with a maximal depth of about 200 m (Ishibashi et al., 2008). The timing of protogynous sex change of fish of the genus Pseudanthias has been discussed (Popper and Fishelson, 1973; Suzuki et al., 1978; Yogo, 1985). Although it is known that P. rubrizonatus inhabits Kagoshima Bay, little is known about the conditions of its sex change in the bay.

Fig. 1.
figure 1

Sampling sites at Kamoike Port, Shimomura Port. and the 14th refuge port in Kagoshima, Japan.

Full size image

In this study, we sampled P. rubrizonatus individuals in Kagoshima Bay. We measured their body size and estimated the age at which the sex change occurs. To estimate the timing of the sex change, we examined gonadal histology and estimated age from otoliths, a more accurate method than estimation from scales (Masuda and Noro, 2003). The early stages of gonadal development are little known in several fish species exhibiting sex change (Liu and Sadovy, 2009). In this study, we focused on the growth and gonadal development of P. rubrizonatus and aimed to describe some information about their reproduction in Kagoshima bay.

MATERIALS AND METHODS

Sample Collection

A total of 113 individuals of P. rubrizonatus were collected at the Kamoike Port and around Sakurajima Island in Kagoshima Bay (Fig. 1). The fish were caught either using hooks or directly while scuba diving. The sampling was conducted in July, August, and December 2016 and in May and June 2017. The fork length (FL) and body weight (W) of the fish were measured.

Determination of sex.P. rubrizonatus is dimorphic sexually (Senou, 2013). Therefore, we determined the sex of P. rubrizonatus from sex-specific body coloration. Males have a red band in the middle of the body and long anal and caudal fins (Fig. 2a). Females have no red band and short anal and caudal fins (Fig. 2b).

Fig. 2.
figure 2

Differences in body coloration between male and female red-belted anthias Pseudanthias rubrizonatus: (a) male, the arrow shows the location of the red band; (b) female. Scale bar: 2 cm.

Full size image

Growth Characteristics and Sex Change

We estimated the FL and W of fish exhibiting sex change from the point of intersection between the regression lines of females and males. The value of the intersection point was obtained from a double-logarithmic graph. Regression analysis for females and males was performed. The slopes were analyzed by comparison of two regression slopes (Rao, 1964) to check the existence of the intersection, and the point of intersection between female and male regression lines was determined.

Gonadal Histology

The gonads were fixed overnight in Davidson’s solution at room temperature. Then the gonads were embedded in paraffin and cut into 6- to 10-µm sections and stained with hematoxylin and eosin following the standard method.

The gonadal phase of P. rubrizonatus was classified into four types based on the following histological observations (Suzuki et al., 1978; Bhadari et al., 2003; Liu and Sadovy, 2009): immature to mature ovary (type 1), oocytes and spermatogonia in ovary (type 2), atretic oocytes in testis (type 3), and only testis (type 4).

Observation of Number of Otolith Rings

The otoliths were removed from the cranium and dried in a desiccator. The right otolith was used unless that side was broken, in which case the left otolith was used. The otoliths were embedded in epon resin consisting of Eopk812 (Okenshoji, Japan), DDSA (dodecenylsuccinic anhydride; Okenshoji), MNA (methyl nadic anhydride; Okenshoji Co. Ltd.), and DMP-30 (2,4,6,-tri-dimethyl aminomethyl phenol; TAAB Laboratories Equipment, United Kingdom). First, 24 mL of Eopk812, 11 mL of DDSA, and 16 mL of MNA were mixed with a magnetic stirrer for 30 min. After mixing, 1 mL of DMP-30 was added and mixed for 30 min. The epon resin was polymerized in an oven. The temperature during polymerization was gradually increased stepwise up to 35°C for 8 h, 45°C for 12 h, and 60°C for 48 h. The otoliths were cut along the long axis by a microcutter with a diamond saw (MC-201 type; Maruto Co. Ltd., Japan). The otoliths were placed on grinding plates (9820; Makita, Japan) for polishing. The polished otoliths were coated with a bond. The number of ring marks (outer edge of the opaque zone) was counted under a microscope (M2125; Leica, Germany) with transmitted light on a black background (Fig. 3). The age of the fish was estimated following the methodology for P. rubrizonatus (Fowler and Booth, 2012; Fowler et al., 2015).

Fig. 3.
figure 3

Sectioned otolith of Pseudanthias rubrizonatus and ring marks (arrows). Age, 3 years. Scale bar: 200 μm.

Full size image

RESULTS

Growth Characteristics of Female and Male P. rubrizonatus

The relationship between FL and W of P. rubrizonatus is shown in Fig. 4a. The body weight increases exponentially with FL. FL ranged from 10 to 114 mm, and W ranged from 0.08 to 21.78 g. Based on the double-logarithmic graph (Fig. 4b), the regression lines of male and females were the following: Female: y = 3.07x – 4.96 (R 2 = 0.99), Male: y = 2.23x – 3.28 (R 2 = 0.74).

Fig. 4.
figure 4

Relationship between fork length and body weight of Pseudanthias rubrizonatus (n = 117): (⚫) females; (◆) males.

Full size image

The slopes of the regression lines were significantly different (p < 0.05). Based on the intersection point of these lines, the theoretical values at which sex change occurred were 96.91 mm FL and 14.13 g W.

Histological Observation of Gonads of Fish Undergoing Sex Change

Most fish with FL < 80 mm had immature oocytes and were classified as type 1 (Table 1). According to our observation of gonadal histology, the fish 30 mm FL possessed immature oocytes (Fig. 5a). The oocytes ranged from the chromatin nucleolus to the perinucleolus stage (Fig. 5a). Any cell clearly identified as oocyte was not observed from histological sections of gonads of the fishes smaller than 30 mm FL. Fish with FL 71–90 mm had mature ovaries and were classified as type 2. The oocytes were mainly at the late yolk globule stage. Some migratory nuclei and postovulatory follicles also existed in these ovaries (Fig. 5b). Two of those fish had immature oocytes and included spermatogonia and spermatocytes in the gonads. (Figs. 6a, 6b). Fish with atretic oocytes in the testis were classified as type 3 (Figs. 6c, 6d). Ovarian cavities were also observed with atretic oocytes in the testis (Fig. 6c). Spermatogonia, spermatocytes, and spermatozoa were observed in the testis of males classified type 4 as mature testis. Spermatozoa were observed in the testes of many males in the spawning stage (Fig. 5c). The ovarian cavities were observed in the testes of many males (Fig. 5d). Gonadal transition from type 2 to type 3 was observed in fish with 81, 91, 92, and 99 mm FL and 7, 11, 12, and 14 g W.

Table 1.   Relationship between fork length (FL) and number of fishes with different gonadal stages in Pseudanthias rubrizonatus
Full size table
Fig. 5.
figure 5

Gonads of male and female of Pseudanthias rubrizonatus. (a) oocytes from chromatinnucleolus stage to perinucleolus stage; (b) mature oocytes, vitellogenicoocytes, and postovulatory follicles; (c) mature testis; (d) formation of ovarian cavity at the center of testis as type 4. FOC—ovarian cavity; POF—postovulatory follicle; SC—spermatocyte; SG—spermatogonia; SP—spermatozoa. Scale bar: 50 µm.

Full size image
Fig. 6.
figure 6

Male and female gonads containing cells of different sexes in Pseudanthias rubrizonatus: (a) spermatogonium making cyst at the edge of the ovary; (b) spermatogonia and spermatocyte stage cells in the gonad of a fish with a female body coloration; (c) formation of ovarian cavity; (d) atretic oocytes in the testis of a fish with a male body coloration. AO—atretic oocyte; FOC— ovarian cavity; SC—spermatocyte; SG—spermatogonia. Scale bars: (a, d) 10, (b, c) 25 µm.

Full size image

Age of Individuals at the Time of Sex Change

We counted the number of otolith rings in 110 fish. The relationship between estimated age and measured FL is shown in a scatter plot (Fig. 7). No fish 1 or 6 years old were found. All fish with zero ring marks on the otolith were females. Among fish 2–3 years old, both females and males were observed, and the gonadal stage of these fish was in transition from type 2 to type 3. All fish 4 years of age or older were the males. Since females were up to 3 years old and males appeared after the age of 2 years, it is likely that P. rubrizonatus undergoes sex change at 2–3 years old. The oldest individual was 7 years old.

Fig. 7.
figure 7

Relationship between fork length and age in Pseudanthias rubrizonatus: (⚫) females; (◆) males.

Full size image

DISCUSSION

Based on the analysis of growth characteristics, gonadal histology, and otolith sections, P. rubrizonatus individuals exhibited sex change at 81 to 99 mm FL by the age of 2–3 years (Fig. 6). Based on the regression analysis of FL and W, we estimated the average body length at the time of sex change in P. rubrizonatus as 96.9 mm FL (Figs. 4a, 4b; Table 1). This theoretical size for sex change and the results of histological observation were almost consistent with each other. Conversely, theoretical W (14.1 g for sex change) was out of the range for the measured values. The time of sex change in P. squamipinnis had been estimated with regression analysis of FL and Fin length (Suzuki et al., 1978). The intersection point of the regression lines of females and males was at 91 mm FL (Suzuki et al., 1978). Spermatogonia, spermatocytes, and spermatids were observed in the ovary of fish that were changing sex (Suzuki et al., 1978). These results suggest that the time when this fish species changes its sex could be estimated by regression analysis between FL. The factors of sex change are discussed, but the deciding factor is not known (Suzuki et al., 1978; Shapiro and Boulon, 1982). Our results support the idea that body size is one of the main factors for sex change.

The remnants of an ovarian cavity in the testis are evidence of protogynous sex change (Liu and Sadovy, 2004). Males of fish in the Labridae family had ovarian cavities after sex change (Nakamura et al., 1989). Ovarian cavities were also observed in the testis of Anthias conspicuous (Heemstra, 1973). However, it has been reported that P. squamipinnis has no clear ovarian cavities in the testis. All the fish of male appearance that were changing their body color had only testes, but some fish of female appearance had spermatogonia, spermatocytes, and spermatids with oocytes (Suzuki et al., 1978). In this study, we observed the formation of ovarian cavities in the male testis of P. rubrizonatus. No fish that were changing their body color were found. However, the fish that are male and female in sex specific body color have atretic oocytes, spermatogonia, and spermatocytes. In P. squamipinnis, the fish that were changing their body color were not found in the samples (Suzuki et al., 1978). These results suggest that the change in body coloration occurs after the gonadal sex reversion.

In this study, we expected that the number of males would be more than the number of females. In fact, there were more males than females among the sampled fish (male : female ratio, 3.3 : 1). The sex ratio may be affected by biased fishing sampling (Kendall and Quinn, 2013). Larger males P. squamipinnis are dominant over smaller males (Yogo, 1985). The larger males are more likely to be removed by fishing using hooks. Therefore, bias of the sex ratio by fishing sampling is an issue to be considered in this study. We also conducted sampling by scuba diving in order to collect smaller females. However, the sex ratio did not meet our expectations because medium-sized fish undergoing sex change could not be caught. To resolve this issue, it would be necessary to capture the entire school of fish rather than parts of it.

We estimated the age of the fish by the analysis of otolith sections. This is an accurate method for estimating age (Masuda and Noro, 2003). The estimated age of the fish was from 0 to 7 years, but no 1 or 6 years old fish were caught. All fish at the age 0+ were the females. Many 2-year-old female P. squamipinnis change their sex after spawning (Yogo, 1985). In a feeding experiment with P. squamipinnis (Suzuki et al., 1978), sex change was observed in fish that had already spawned. Yogo (1985) reported that females of P. rubrizonatus matured in the first year and changed sex after spawning in the second year. In the present study, the relationship between sex change and spawning was unclear. However, 2-to-3-year-old females possessed matured oocytes and post-ovulatorty follicles. This result indicated that P. rubrizonatus change sex around 2-to-3-years old.

ACKNOWLEDGMENTS

The authors are grateful to the laboratory members for supporting sample collection. The authors would like to thank Enago (www.enago.jp) for the English language review.

COMPLIANCE WITH ETHICAL STANDARDS

Conflict of interests. The authors declare that they have no conflict of interest.

Statement on the welfare of animals. All applicable international, national, and/or institutional guidelines for the care and use of animals were followed.