Comparative Studies of the Thyroid Function in Teleosts of Tropical and Temperate Habitats

Marine & Lenhart (1910) observed a hyperplastic condition of the thyroid tissue in some trout in hatcheries which was correlated with sluggishness, altered pigmentation and fatness. This condition suggested an altered metabolism. Feeding experiments with mammalian thyroid tissue (Etkin, Root & Mofshin, 1940; Smith & Everett, 1943 ; Hasler & Meyer, 1942; Root & Etkin, 1937) failed to alter oxygen consumption of teleosts. Only a few individuals showed a rise in oxygen consumption when fed teleostean thyroid tissue (Smith & Matthews, 1947). Thyroid hormone production in mammals is depressed by thiourea. Matthews & Smith (1947) however, were unable to demonstrate a depression of oxygen consumption in Fundulus by the use of this drug.

As growth and development of fish may be determined to some extent by the presence of thyroid hormone, experiments to elucidate this effect have been carried out. Some observers have reported a positive effect on giving thyroid hormone (Grobstein & Bellamy, 1939; Goldsmith, Nigrelli, Gordon, Charipper & Gordon, 1944; Hopper, 1952; Vivien & Gaiser, 1952), and others (Smith & Everett, 1943) the reverse. Using thiourea, Smith, Sladek & Kellner (1953) recorded a depressive effect on the growth of Lebistes, but they attributed this to a slight toxic effect on the drug.

These contradictory results prompted the work reported below. If the thyroid mechanism regulates metabolism in the teleostean fish, it might be expected that a study of such fish at two temperature extremes would throw some light on the problem. A comparison of the effects of temperature change on thyroid activity, as exemplified by structural changes of the thyroid, and on growth and development afforded a promising approach. Fish from temperate and tropical climes were selected and these were studied under different environmental temperature conditions. As seasonal variations in the thyroid gland of many teleosts are known to occur it was necessary to extend observations over a whole year (Lieber, 1936; Hoar, 1939; Buchmann, 1940).

Two species of fish were selected; the temperate form Phoxinus laevis (Phoxinus phoxinus) and the tropical form Lebistes reticulatus. In both, the thyroid gland consisted of typical follicles scattered along the ventral aorta with occasional invasion into the adjacent muscle bands.

Phoxinus laevis, rarely longer than 6 cm. is found in lakes and quiet streams in England. It can tolerate a temperature range from that of ice-covered water to 23° C. though it has been taken from water at a temperature of 31° C. on the continent of Europe. Spawning occurs from May to July (Frost, 1943). Supplies were obtained from Lake Windermere except during January 1952 when they came from Surrey as the fish retire to deep water in Lake Windermere during the winter.

The fish were kept in large glass jars containing 2 1. of well-aerated tap water or in solutions made up with tap water. The jars were immersed in water-baths whose temperature could be controlled to ± 1° C. The fish were fed with enchytraeid worms and horse flesh twice weekly and the solutions changed after feeding. Mortality was negligible.

Lebistes reticulatus is a small tropical teleost living within a temperature range of 15–35° C. with an optimum of 25° C. The females are viviparous, and are capable of producing a brood every month. As the ‘pregnancies ‘affect the thyroid gland structure (Stolk, 1951) the fish, which were bred locally, were chosen so that each experimental batch contained both young fry and older fish. The fish were kept in a litre of tap water or solution made up with tap water. The tap water was allowed to stand for at least a week before use so as to reduce the amount of chlorine present. The glass jars containing the fish were kept in water-baths at a constant temperature. Aeration was not necessary. Feeding was daily and solutions were changed twice a week. Some deaths inevitably resulted from the change.

The most suitable concentration of thiourea used as an anti-thyroid drug was determined during a preliminary study on Lebistes. It was found that a concentration of 0·05 g. thiourea per 100 ml. water was the most concentrated solution which caused no deaths during 7 days.

The temperature levels used were: Phoxinus—3°, 15, 20 and 25° C.; Lebistes—20, 25 and 30^o C. Twenty fish were used for each experiment, ten in thiourea solution and ten in tap water as controls at each temperature level. Phoxinus controls could not be kept alive at 25° C. One fish was killed from each group when the water was changed twice weekly. Histological examination of the thyroid gland was carried out on each specimen. Fixation was in Bouin’s fluid. Sections were stained with Masson’s trichrome stain.

An estimate of the degree of thyroid activity is based frequently on the structural changes found in the gland such as : (a) alterations in the height of the follicular epithelial cells ; (b) changes in the amount of colloid found in the vesicles ; (c) the altered position of the nucleus of the follicular cell. The changes produced in the mammalian thyroid gland by the administration of thyrotrophic hormone (TSH), which stimulates production of thyroid hormone (TH), are identical with those found when anti-thyroid drugs such as thiourea are given. The latter drug inhibits the production of TH and consequently the concentration of TH in the blood falls which in turn is the adequate stimulus for increased production of TSH. Thus the morphological signs of increased thyroid activity are not necessarily related to the rate of TH production. In fact, the degree of stimulation of the thyroid gland is demonstrated by these signs, the extent of morphological change depending upon the relative levels of TSH and TH. If the thyroid gland is producing large quantities of TH, inhibition of this production will cause a change in the TSH : TH ratio with a consequent marked and rapid change in gland structure. If the gland is producing little TH, inhibition of production of TH will cause only a small disturbance in the TSH : TH ratio, hence the resultant stimulation of TSH production will be small and structural changes will not be marked. The apparent changes in thyroid activity seen in histological material can thus be used as indices of an alteration in the TSH : TH ratio. Various methods have been devised to measure the degree of activity manifested by the thyroid gland in response to TSH. The most general is to measure the height of the follicular epithelium. Cell height increases with a rise in the rate of production of TSH. This method, however, is only reliable when under control conditions, the follicular cells are reasonably even in height. In teleostean fish, cell height is very variable even in the same follicle. Small follicles have a lower epithelium than large follicles. When activity increases the fish thyroid tends to show proliferation of the follicular epithelium with consequent formation of new follicles. The method of measuring thyroid activity suggested by Uhlenhuth et al. (1945), and used in a study of the salamander thyroid gland, was adopted. This consists of determining on sections of the thyroid gland the relative areas of the follicular epithelium and follicular colloid. In the two fish species used here the amount of colloid present was inversely proportional to the height of the follicular epithelium. Stained slides were projected on to squared paper. The outlines of the outside of the follicles were traced and also the outlines of the inner edge of the epithelium. The area represented by the follicles was cut off and the paper weighed. The area representing the cells was then cut off and weighed. The ratio of ‘cell weight’ to ‘total follicular weight’ was then expressed as a percentage, indicating thyroid activity. It was found that when the ratio, calculated from the actual areas of the follicles and follicular cells, was compared with that determined by the weighing technique there was an agreement within ± 1 %. The mean ratio from twenty-five follicles was determined for each specimen.

The advantages of this method of assessing the state of activity of the thyroid gland are: (1) irregularities of the follicles and cell proliferations are taken into account; (2) the amount of colloid present is also taken into account in the ratio; (3) the value of the ratio differs less between large and small follicles than the difference in activity as estimated by differences in cell height in follicles of different sizes.

The chief disadvantage of this method is that there is a lower limit below which measurement is inaccurate. In very small follicles there is relatively little colloid and hence the ratio appears high. This however is cancelled out by measuring a large number of large and small follicles. It is clear that the ratio can be disturbed by two factors : (a) changes in the height of the epithelium, and (b) changes in the amount of colloid formed. The ratio, although a purely arbitrary method of assessing the level of thyroid activity, certainly will show direction of change even if the amount shown by variations in the ratio may not be an exact reflexion of the altering levels of thyroid activity. This method also enables the average size of the follicles to be determined, the number of follicles less than a certain arbitrary size to be estimated and the percentage of different staining colloids within the follicles to be determined.

The experiments described above were carried out over a period of 5 weeks, and repeated at intervals throughout the year. The experiments were divided into two groups : (a) controls, (b) fish exposed to thiourea. Environmental temperatures used were: 3, 15, 20^o C. in the control group; 3, 15, 20 and 25^o C. in the thiourea group.

The results found in the period April to May are given in Table 1 and Fig. 1 A, B. This period may be regarded as representative. The mean values of the control group show clearly that thyroid activity does not increase as the environmental temperature is lowered. It will be noted in the thiourea group that at 3° C. the highest level of activity (63·9) was attained only after 5 weeks. At 15^o C. a 95·7 activity ratio was attained on the 28th day. A level of activity of 90 was reached at 20° C. between the 10th and 14th day. At 25° C. this same level was reached between the 7th and 10th day.

Similarly the experiments were divided into two groups : (a) control, (b) exposed to thiourea. Environmental temperatures used were : 20, 25 and 30^o C.

The results found in a representative period (April-May) are given in Table 1 and Fig. 1, C, D. The results again indicate that thyroid activity increases with increased environmental temperature. The mean value of the control groups (61 2 activity ratio at 30° C., 61·1 at 25° C., 48·3 at 20° C.) demonstrates this. The thiourea-treated groups behaved in a like manner ; the fish at 30^o C. taking between 7 and 10 days to reach a thyroid activity ratio of 90, between 10 and 14 days at 25° C. and between 17 and 21 at 20° C., to reach the same activity ratio.

It has already been noted that Phoxinus could not be kept alive at 25^o C., unless treated with thiourea. No attempt at gradual acclimatization to this increased temperature was made, but when treated with thiourea the fish survived indefinitely and appeared normal. Hence thiourea treatment must affect the thermal death-point. To test this Phoxinus, already treated with thiourea for 3 days, were subjected to a rise in temperature of 10° C. (i.e. to 33° C.) over 2 days. They survived indefinitely at this increased temperature and appeared normal, the controls dying under such conditions between 23 and 24° C.

Lebistes has a higher thermal range but this could also be extended by thiourea treatment.

The thermal level of response to thiourea is higher in Lebistes (20° C.) than in Phoxinus (3° C.).

These experiments were designed to determine if the possible seasonal variation in thyroid activity was related to the external temperature to which the fish was subjected, or if there was an inherent rhythm associated with the breeding cycle. Phoxinus has a yearly breeding cycle which could be related to annual variations in thyroid activity. Lebistes, on the other hand, breeds monthly throughout the year.

As each of the above experiments was repeated at intervals throughout the year, similarly treated specimens could be compared for any 2-monthly period. A complete record was obtained except for the month of March.

The results of the control experiments with Phoxinus showed that there were two peaks in thyroid activity, one during April-May and the other during late August-September which were common to all three environmental temperatures (Fig. 2 A).

Lebistes showed no distinct variation common to all temperatures (Fig. 2B).

Eggs were obtained from Lake Windermere in May. A similar technique was again used and the experiments were divided into two groups : (a) control, (6) immersed in 0·05 % thiourea. There were approximately 600 specimens in each. Half the volume of solution was changed thrice weekly. The eggs were placed in the fluid in shallow glass troughs and aerated vigorously. The majority hatched out after 3 days. Three days later feeding was commenced. Protozoa culture was given three times a day for the first month, with small amounts of a dried fry food and living Daphnia at the end of the third week ; the Protozoa culture was then discontinued and the other food was given twice daily. Specimens were fixed at first each day, with gradually lengthening intervals as the fry became larger. Very small specimens required double embedding.

The results show that the gland develops as a solid mass of tissue from the ventral region of the pharynx between the first and second gill slits. After 13 days, the control group show a few follicles with low epithelium and abundant green-staining colloid. Those treated with thiourea have a high epithelium, small amounts red-staining colloid and an increased number of follicles. However, although thiourea treatment produces marked changes in the thyroid structure, growth of the fish is not permanently affected. There was a slight difference in the average size of the fish in the two groups between the 15 and 35 days after hatching (thiourea, average length 8·5 mm., control, 9-5 mm.). This was only temporary and the average length became the same in both groups and reached 11 mm. 3 months after hatching. The thiourea-treated fish still showed marked thyroid hyperplasia.

A similar technique was used, but as Lebistes is viviparous a third group was included : (a) control ; (b) thiourea-treated from birth ; (c) fry from a pregnant female previously kept in thiourea for 3 weeks before birth, the treatment of the fry being continued afterwards.

The fry may have a yolk-sac at birth or if more completely developed, may be without one. Any differences were noted. Specimens were killed daily and treated as before for histological examinations.

There was no size difference between fry treated with thiourea for 4 weeks and normal fry of the same age (average length 6·5 mm. at birth, 8·5 mm. at 4 weeks). There was a differentiated thyroid in all fry at birth, with the control fish having well-developed follicles, low epithelium cells and abundant colloid. All thiourea-treated fry showed a hyperplastic thyroid and colloid absent or reduced.

The modem conception of the action of anti-thyroid drugs is based almost entirely on work with avian and mammalian thyroids. It is generally accepted that there is a partial or complete blockage of thyroid hormone synthesis. Histological changes take place in the teleostean gland, after thiourea treatment, which are analogous to those occurring in mammals and birds after similar treatment. The experiments outlined here show that thiourea treatment causes a change in the temperature range of the two species of teleosts used. This is a definite physiological effect, and justifies the conclusion that thyroid hormone production has been reduced by this method.

When the fish are placed in thiourea solution a definite time interval is required for the drug to exert its maximum effect at any given temperature. The time taken to reach maximum response to thiourea diminishes with an increase in environmental temperature. Although some glands give ratios of 100, which means that the area occupied by the follicular cells is equal to the total area of the follicle, it is preferable to accept a ratio of 90 as indicating an arbitrary ‘maximum’ level of activity. The time taken to reach this level of activity can be determined for each temperature level and for each month of the year at the same level. When this is done for a specific monthly period (e.g. April to May) the results show that there is strikingly little variation in the ratio of the total follicular area to cell area in the control groups of both species at any given temperature, over a period of 3–5 weeks. On the other hand, the ratio for the thiourea-treated fish gradually increases over the same time interval (see Table 1 and Fig. 1). The mean control values indicate a depression of thyroid activity with lowered environmental temperatures. This is emphasized by the varying speed of reaction to thiourea which is more rapid at high temperatures than at low. If cold, as well as thiourea, had stimulated the gland the reverse would have been found. These results therefore show clearly that the thyroid gland is stimulated to increased activity on exposure to higher environmental temperatures. Obviously these results are the reverse of those to be expected were the thyroid gland acting to maintain some uniformity of heat production.

The arbitrary standard set up allows a comparison between states of activity at different times of the year. If a true seasonal variation occurs, it should be apparent at all those levels of environmental temperature at which the thyroid has been shown to be reactive. It is shown in Table 1 that, with the exception of Phoxinus at an environmental temperature of 3° C., such a reactive condition obtains. Phoxinus shows a seasonal variation during thiourea treatment at 15, 20 and 25° C. At 3 and 15° C., although a ratio of 90 is not attained during most months, a higher level of activity is reached at the spring and summer peaks. The peak activity is reached in early spring (April to May) and late summer (August to September). During the winter the lowest activity is found from December to February. Bullough (1939) has shown that there is an increase in gonad size in Phoxinus twice during the year; September to October and in the spring. The thyroid activity peaks precede the increase in gonad size, and it may be assumed that there is some relationship between the two.

Lebistes shows an indefinite variation. Although a peak activity is apparent at 20° C. during July to August it is not present at the other environmental temperatures. It is therefore not a true seasonal peak which would be present at all environmental temperatures. As the breeding cycle is monthly it appears that seasonal variation is correlated with an annual breeding cycle.

No effect of thiourea treatment on growth could be demonstrated.

1. Normal and thiourea-treated Phoxinus and Lebistes show a depression of thyroid activity by low environmental temperatures and stimulation by high.

2. Thiourea treatment affects the thermal range in both species.

3. Phoxinus shows a seasonal variation in thyroid activity, correlated with breeding periods. This is absent in Lebistes.

4. No effect of the thyroid on growth could be demonstrated.

My grateful thanks are due to Prof. Spaul for his constant help and encouragement throughout this work. Also I should like to thank Dr Beattie for assistance with the MS.