Variations in the Plasma Levels of Thyroid Hormones and Testicular Activity in the Male Air-breathing Catfish (Clarias gariepinus) over the Annual Cycle

The purpose of this study is to evaluate seasonal variations in thyroxine (T4) and triiodothyronine (T3) levels and testicular activity in the male air-breathing catfish Clarias gariepinus. Plasma T4 levels increased from March to a peak in July and then decreased throughout autumn, declining to their lowest level in January. Plasma T3 levels also varied seasonally, increasing from March to a peak in June. The increase in plasma levels of thyroid hormones coincided with an increase in the testicular activity of the fish. The gonadosomatic index (GSI) peaked in May and then declined slowly. In February and March, the testes were characterized by a new generation of spermatogenic cells. Mature spermatozoa appeared in April and became dominant in the testes during the summer months. Testicular activity decreased gradually with the gradual decrease in water temperature and day-length during the autumn season. Decreased testicular activity during the winter months was associated with low GSI values. The patterns of seasonal change in the levels of thyroid hormones as well as testicular activity were similar to those in water temperature and natural day-length.

There are published reports of seasonal changes in the serum levels of thyroid hormones in fish (Burke and Leatherland, 1983;Cyr et al. 1988a;Leatherland andSonstegard, 1980, 1981).However, studies that have examined seasonal changes in the plasma levels of thyroid hormones in teleosts are limited to a small number of species, mostly salmonids (Eales and Fletcher, 1982;Pickering and Christie, 1981).The results of these studies vary considerably by species.In fish species, circulating levels of thyroid hormones and/or thyroid activity are influenced by gonadal steroids (Cyr et al., 1988b;MacLatchy and Eales, 1988), which show circannual rhythms in their secretions closely related to breeding cycle and seasonal variations in ambient temperature (Chakraborti and Bhattacharya, 1984;Cyr et al., 1988a;Lamba et al., 1983).Furthermore, the reproductive cycle of fish has been the object of several studies in temperate zones (Hatakeyama and Akiyama, 2007;Pinillos et al., 2003;Prat et al., 1990).Comparatively little information is available for tropical and subtropical species (Bhattacharya et al., 2005;Chakraborti and Bhattacharya, 1984).Changes in gonadal activity are reportedly influenced by temperature and photoperiod (Bhattacharya et al., 2005;Koya and Kamiya 2000;Rodriguez et al., 2001;Shimizu 2003).Reproductive characteristics have been reported for Clarias gariepinus in a temperate zone (Cek and Yilmaz, 2007;Yalcin et al., 2001).Cyclical changes in the gonads have also been examined in a few closely related species, including the Natal mountain catfish Amphilius natalensis (Marriott et al., 1997) and freshwater catfish Mystus montanus (Arockiaraj et al., 2004).However, it is important to mention that these studies did not extend over the course of an entire year, and no simultaneous observations were made of the levels of thyroid hormones and testicular activity.Thus, the objective of the present study is to examine the annual pattern of thyroid hormone levels and testicular activity in the male catfish C. gariepinus (Burchell) maintained under natural climatic conditions in Shillong, India.This report describes monthly changes in the plasma levels of thyroid hormones (thyroxine, T 4 ; triiodothyronine, T 3 ) and testicular activity with particular reference to water temperature and natural day-length/photoperiod.

Fish, experimental design and sampling
All experiments were conducted on the adult male, Clarias gariepinus that breed in India during monsoon, and live in shallow rivers, ponds and muddy places.C. gariepinus can survive in low water levels with little oxygen.Specimens (body weight: 90-100 g; body length: 23-27 cm) were purchased from the local fish suppliers.Fish were maintained in clear plastic tubs and acclimatized at least for 15 days in the laboratory under natural climatic conditions at Shillong (25°30′ N, 91°52′ E; 1450 m ASL: minimum water temperature varied from 4 -6° C and maximum water temperature varied from 23 -25° C).During acclimatization, the fish were fed daily ad libitum with minced earthworms and commercial fish feed (Tokyu).The water was changed every day.The water temperature was recorded daily, and the monthly average of the natural day-length (photoperiod) of Shillong was taken from the Central Meteorological Department in Meghalaya.
After acclimatization, blood samples and testes were collected from the fish regularly on the 14th/15th day of each month at an interval of 30 days throughout the year.The sampling was performed at 11:00 AM and was consistent throughout the experiment.To collect blood samples, we removed 4 fish from each group by a hand net and weighed them.Blood samples were collected from the postcaudal region in numbered heparinized centrifuge tubes.Simultaneously, testes were collected, weighed, and fixed in Bouin's fluid for histological study.Body and testes mass of each fish were used for calculating the gonadosomatic index (gonad mass x 100/total body mass).To obtain plasma, we centrifuged the blood samples for 15 min at 3000 xg by using a REMI (Model R23) centrifuge, separated the plasma, and stored it at -10°C in a refrigerator for the measurement of T 4 and T 3 concentrations.

Measurement of plasma levels of thyroxine (T 4 ) and triidothyronine (T 3 )
The plasma levels of T 4 and T 3 were measured with the help of radioimmunoassay kits RIAK5/5A for T 4 (sensitivity 0.625 ng/ml, based on 93.67% B/Bo intercept) and RIAK4/4A for T 3 (sensitivity 0.0375 ng/ml, based on 90.77% B/Bo intercept).Radioimmunoassay kits were obtained from the Division of Radiopharmaceutical Operations, Board of Radiation and Isotope Technology (BRIT), Mumbai.Radioimmunoassay (RIA) for T 4 and T 3 were conducted following the manufacturer's protocols with slight modifications where the hormone-free serum was replaced by hormone-free fish plasma (Gupta and Premabati, 2002).The RIA was validated using hormone-free fish plasma and different concentrations of standard solutions of T 3 (range, 0.0375-2.4ng/ml) and T 4 (range, 0.625-20 ng/ml).The hormone-free fish plasma used in the assay was prepared by two cycles of the addition of dextran-coated charcoal to the pooled fish plasma, continuous stirring for 6 hours and centrifugation.Polyethylene glycol (PEG) solutions (12% for T 3 and 22% for T 4 ) were used to separate the bound and free fractions of T 4 and T 3 in their respective RIA.The intra-and inter-assay variations were found to be, on average, less than 3.5% and 6.5% for T 3 and T 4 , respectively.The radioactivity in the bound fraction was counted with the help of a well-type gamma counter (Electronic Corporation of India, Hyderabad).The concentrations of total T 4 and total T 3 were expressed as ng/ml of plasma.

Histological study of testes
After the testes had been in the fixative for 24-48 hours, they were removed from the fixative and were washed for 24 hours with running water.This was followed by a dehydration period of 20 minutes in each of 30%, 50%, 70%, 90%, and 90% and twice in a 100% ethanol range.After the dehydration process, the tissue was cleared in xylene and then embedded in paraffin wax.The hardened blocks were trimmed, mounted on tissue chucks and sectioned on a microtome into 6 μm sections.Selected sections were mounted on clean glass slides, floated and stretched on a hot-plate and placed on a warming tray to dry and adhere.After dewaxing in xylene, the sections were hydrated in decreasing grades of ethanol and then stained with hematoxylin.After washing excess of hematoxylin, the sections were dehydrated in increasing grades of ethanol and stained with eosin.The stained sections were then further dehydrated, cleared in xylene and mounted with cover slips using DPX.The sections were examined under Zeiss Microscope and photographed.

Statistical analyses
The data were statistically analyzed with the help of analysis of variance (ANOVA).A probability of P < 0.05 was considered as significant.

Seasonal variations in the plasma levels of T 4 and T 3
The data are presented in Figure 2. Seasonal variations were observed in the plasma levels of both T 4 and T 3 in male C. gariepinus.The plasma levels of T 4 were low in January (winter), increased gradually beginning in March, and reached a peak in July (summer).Thereafter, the plasma levels of T 4 declined gradually to their minimum in January.There was a significant increase from the preceding months in the plasma levels of T 4 in fish sampled during March, June, and July.By September and November, the plasma levels of T 4 had fallen significantly compared to the preceding months.As was the case for T 4 levels, the plasma levels of T 3 were also low in January and increased gradually to their highest levels in June (summer).Thereafter, the plasma levels of T 3 declined gradually, reaching their lowest level in January.The plasma levels of T 3 increased significantly from the preceding months in March and June.Furthermore, there was a significant decrease from the preceding months in the plasma levels of T 3 in fish sampled during September, November, and January.Regression analysis indicated a positive correlation between water temperature and the plasma levels of T 4 (r = 0.92) and T 3 (r = 0.85), as well as between natural day-length/photoperiod and the plasma levels of T 4 (r = 0.91) and T 3 (r = 0.87) (Figure 1).

Seasonal variations in the gonadosomatic index (GSI)
The GSI was low in January and February, increased gradually in March and April (spring).The GSI was highest in May, remained high in June, and thereafter declined gradually to its minimum value in January.The GSI in April and May increased significantly from the preceding months (Figure 3).Regression analysis indicated a positive correlation between water temperature and the GSI (r = 0.86) as well as between photoperiod and the GSI (r = 0.90) (Figure 1).

Seasonal variations in the histology of the testes
Histological observations revealed the presence of only spermatogonia in the testes in January (Plate A).In February, the seminiferous tubules of the testes contained spermatogonia and spermatocytes (Plate B).In March, these tubules increased in diameter and contained lower numbers of spermatogonia, spermatocytes and small populations of spermatozoa (Plate C).Whereas the number of mature spermatozoa increased in April, the number of spermatogonia decreased considerably and the seminiferous tubules were much enlarged, indicating active spermatogenesis (Plate D).In May and June, the seminiferous tubules were filled with mature spermatozoa and smaller numbers of spermatogonia and spermatocytes (Plates E and F).In July, the tubules were filled with spermatozoa (Plate G).However, the spermatozoa declined gradually in August and September (Plates H and I).In October and November (autumn), there was marked reduction in the diameter of the seminiferous tubules, and the tubules contained degenerating spermatozoa along with a few scattered spermatogonia (Plates J and K).In December, the seminiferous tubules contained small numbers of spermatogonia, but residual spermatozoa were also still observed and there was a reduction in the diameter of the tubules (Plate L).

Discussion
Seasonal variations in the levels of thyroid hormones have been reported in a number of temperate zone fish species, such as Oncorhynchus masou, Salmo gairdneri, Pseudopleuronectes americanus, Dentex dentex, and Gadus morhua (Bau and Parent, 2000;Comeau et al., 2000;Cyr et al., 1988aCyr et al., , 1998;;Eales and Fletcher, 1982;Leloup and De-Luze, 1985;Nishikawa et al., 1979;Pavlidis et al., 2000;Sower et al., 1992;Stryjek-Kaminska et al., 1988).Seasonal variations in the levels of thyroid hormones have also been reported in the tropical freshwater perch Anabas testudineus (Chakraborti and Bhattacharya, 1984).However, these studies were performed during a particular season of the year, and no simultaneous observations were made of the seasonal changes in the circulating levels of thyroid hormones and testicular activity.
In the present results, the plasma levels of thyroid hormones and testicular activity in C. gariepinus exhibit prominent and parallel seasonal rhythms.The plasma levels of thyroid hormones and testicular activity (as indicated by the GSI and histology) were lowest during the winter months when the water temperature was low and natural day-length was short.Thereafter, the hormone levels as well as the testicular activity increased gradually with the gradual increase in water temperature and day-length during the spring season.The plasma levels of thyroid hormones and testicular activity were highest during the summer months when the water temperature was comparatively higher and the natural day-length longer.The patterns of seasonal change in the levels of the thyroid hormones as well as testicular activity in C. gariepinus were observed to be similar to those of water temperature and natural day-length.There was also a significant positive correlation between the plasma levels of the thyroid hormones and both water temperature and day-length (Figure 1).These results indicate that prominent seasonal rhythms in the levels of thyroid hormones and testicular activity in C. gariepinus coincide with seasonal changes in water temperature and day-length.Although the possibility of completely dependent or independent control of temperature and photoperiod on the plasma levels of thyroid hormones and testicular activity cannot be discounted, their distinct and parallel seasonal rhythm is consistent with previous reports that environmental temperature and photoperiod synchronize and/or regulate thyroid and reproductive seasonality in the fish.Furthermore, the plasma levels of the thyroid hormones showed a similar pattern of seasonal change as testicular activity/maturation in C. gariepinus.Several reports have suggested that thyroid hormones may be required to initiate gonadal development and maintain the active reproductive state of fish (Sullivan et al., 1995;Swapna and Senthilkumaran, 2007;Tagawa et al., 1994).Therefore, the present observation of the similarity in the pattern of seasonal change in the plasma level of thyroid hormones and testicular maturation seems to suggest that thyroid hormones are involved in maintaining the active reproductive state of C. gariepinus.However, further study is required to determine the causal relationships between the seasonally varying thyroid hormones and the testicular activity systems in this fish in a tropical/subtropical region.Furthermore, it has been reported that the metabolic rate and the rate of respiration of several vital tissues of fish increase and decrease with reproductive activity over the annual time scale (Lynshiang, 1998) and that thyroid hormones reportedly stimulate the rate of respiration/oxidative metabolism in several fish species (Eales and Maclatchy, 1989;Gupta and Thapliyal, 1991;Lynshiang and Gupta, 2000).It thus seems that the observed increase in the levels of thyroid hormones during the spring and summer seasons in C. gariepinus ensures an adequate energy supply for meeting the increased energy demands associated with reproductive behavior and associated activities.
Little information is available on the reproductive cycle of C. gariepinus in tropical/subtropical zones or in temperate regions, and none of the existing studies have described testicular maturation over each month of the year.Some reproductive characteristics of C. gariepinus in temperate regions have been reported under natural conditions (Yalcin et al., 2001) as well as under laboratory conditions (Cek and Yilmaz, 2007).However, it is important to mention that these studies were not conducted over the course of an entire year and hence could not delineate the pattern in the annual rhythm of the testicular activity/maturation of the fish.Therefore, here we have presented findings on the testicular activity/maturation of C. gariepinus in a tropical/subtropical region for each month of the year.Observations on GSI and testicular histology revealed a distinct seasonal variation in the testicular activity of C. gariepinus over the annual time scale.We found that increasing testicular activity of the fish coincided with increasing water temperature and natural day-length.Spermatogenesis started to increase in February and March.Mature spermatozoa were found from April through September, and a rapid decline in testicular activity was observed from October through November (autumn).Quiescent testes were found in December and January (winter).The gradual increase in the GSI in March corresponded to the commencement of basal spermatogenesis.Thus, the prominent and gradual increase in the GSI from March to its peak in May corresponded to the enlargement of testicular tubules due to active spermatogenesis.Furthermore, the seminiferous tubules increased in diameter from March to September.It is important to mention that mature spermatozoa were observed in the seminiferous tubule from April (water temperature, above 20°C; day-length, more than 12 h) to September (water temperature, above 21.5°C;day-length, more than 12.23 h).These results indicate that the breeding phase of the fish (April to September) is prolonged in tropical/subtropical regions.Regression analysis also revealed a significant positive correlation between the GSI and both water temperature and day-length (Figure 1).Thus, water temperature and day-length seem to play an important role in shaping the annual breeding cycle as well as regulating testicular activity in C. gariepinus.Temperature and photoperiod have also been reported to regulate seasonal changes in the reproductive functions of Catla catla, Fundulus heteroclitus, Dicentrarchus labrax (Bhattacharya et al. 2005;Dey et al., 2005;Rodriguez et al., 2001;Shimizu, 2003), Pagrus auratus (Bloch and Schneider) (Scott and Pankhurst, 1992), and Heteropneustes fossilis (Lamba et al., 1983).Furthermore, it has also been reported that increasing water temperature and day-length were favorable for gonadal development in Labeo rohita (Singh et al., 2005).These reports, when taken together with the present findings, indicate that seasonal changes in the environmental water temperature and natural day-length are primary cues for altering testicular activity and play major roles in the regulation of breeding/reproduction of the fish.It is a well-established fact that testosterone and 11-ketotestosterone are important for regulating spermatogenesis in fish (Borg, 1994;Cavaco, 2005;Cavaco et al., 1998;Dubious et al., 2001;Goos et al., 1986;Miura et al., 1991).It has been shown that the plasma levels of testosterone and 11-ketotestosterone are low during the resting/non-reproductive period and high during the nesting/maturation period (Gazola and Borella, 1997;Sisneros et al., 2004).Furthermore, seasonal changes in plasma and serum gonadal steroid hormone levels have been documented in a number of fish species (Dye et al., 1986;Harmin et al., 1995;Liley et al., 1986;Mayer et al., 1990;Pavlidis et al., 1994;Prat et al., 1990;Scott and Sumpter, 1989;Scott et al., 1980;Thorarensen et al., 1995).Therefore, there is a strong possibility that the observed prominent seasonal changes in testicular activity might be related to seasonal variations in the sex steroids of C. gariepinus.Further investigations into the monthly concentrations of sex steroids in C. gariepinus would shed more light on seasonal changes in testicular activity/maturation.

Conclusion
In conclusion, male C. gariepinus displayed distinct seasonal changes in plasma levels of thyroid hormones and testicular activity over the annual cycle.The increase in plasma levels of thyroid hormones coincided with an increase in the testicular activity of the fish.Although the possibility of completely dependent and/or independent thyroid control of testicular activity cannot be discounted and further study on C. gariepinus is needed, the distinct and parallel seasonal rhythms seen here are consistent with previous reports that thyroid hormones are progonadal.The patterns of seasonal change in the levels of thyroid hormones as well as testicular activity were similar to those in water temperature and natural day-length.The breeding phase of the fish also seems to be timed to favorable water temperature (>20°C) and natural day-length (>12 h).

Figure 3 .
Figure 3. Monthly variations in the gonadosomatic index (GSI) GSI values are expressed as mean ± S.E.; n = 4. a Differs significantly from the preceding month: P < 0.05.