Regulatory, Integrative and Comparative Physiology

Cold-induced changes in thyroid function in a poikilothermic mammal, the naked mole-rat

Rochelle Buffenstein, Ryan Woodley, Cleopatra Thomadakis, T. Joseph M. Daly, David A. Gray


Cold acclimation induces very divergent responses in thyroid function in reptiles and mammals reflective of their different thermoregulatory modes. Naked mole-rats, unlike other small mammals, are unable to effectively employ endothermy and are operatively poikilotherms. We therefore investigated changes in their thyroid status with chronic cold exposure. Under simulated burrow conditions, free thyroxine (T4; 0.39 ± 0.09 ng/dl) and thyroid stimulating hormone (TSH; 1.12 ± 0.56 μIU/ml) levels fell within the reptilian range, one order of magnitude lower than mammalian levels. However, cold induced typical mammalian responses: free T4levels (0.55 ± 0.09 ng/dl) and thyroid follicular cell height were significantly greater. Although TSH levels (1.28 ± 0.83 μIU/ml) were not significantly elevated, thyrotrophs exhibited ultrastructural signs of increased secretory activity. Low thyroid hormone concentrations may contribute substantially to the unusual thermoregulatory mode exhibited by naked mole-rats.

  • follicular cell height
  • thyroxine
  • thyrotrophs
  • thyroid-stimulating hormone

across all phylogenetic groups, organisms adjust their physiological responses to overcome the direct effects of prolonged exposure to varying environmental temperatures (21, 27, 28). In response to extreme cold, terrestrial vertebrates may employ winter dormancy (7, 13, 18) or defend body temperature (Tb) by behavioral thermoregulation, augmented insulation, thyroid hormone secretion, and enhanced heat production (14, 43). The precise reaction may be limited by physiological or morphological constraints impeding physiological adjustments. Thyroid hormones play a pivotal role in this regard and are critical in the central regulation of body temperature (Tb), stimulating thermogenesis, and regulating cellular metabolism (41).

The thyroid gland is structurally conserved in all vertebrates (38). Although there are gross morphological differences across the phylum, follicular structure and function are similar. Responses of the thyroid gland to environmental cues are, however, considerably different across the vertebrate classes. In most cases, homeotherms generally increase thyroid activity at low temperatures (9, 17), whereas in poikilotherms, thyroid activity is higher during exposure to warmer conditions (5, 12, 38).

Most small mammals respond to cold exposure by augmenting thyroid-mediated processes (9, 17). These increase basal metabolism and heat production (45, 48). Poikilotherms on the other hand, generally show a decrease in both thyroid activity and metabolism in the cold (5, 12). There are, however, reports of reptiles that maintain higher levels of thyroid activity during cool periods (11, 29, 30, 32).

Naked mole-rats exhibit an unusual mode of thermoregulation in that they are endothermic and capable of employing nonshivering thermogenesis (20) and also poikilothermic, in that Tb tracks ambient temperature (Ta) (6). Their unusual mode of thermoregulation is considered highly suited to a strictly subterranean existence (4). Subterranean rodents live in an environment where both gas and heat exchange are markedly restricted. In response to these environmental stresses, most subterranean rodents exhibit low metabolic rates and relatively low Tb levels (35). Naked mole-rats have taken this trend to the extreme and exhibit the lowest mass-specific metabolic rate for mammals (6, 16, 31, 46,47). Indeed, metabolic rates of these hairless mammals are similar to that reported for reptiles of similar body mass (16) and, not surprisingly, with such low rates of heat production and limited insulation, naked mole-rats are unable to regulate Tb and are effectively poikilothermic (6).

Given the anomalous and unique thermoregulatory mode of naked mole-rats, we questioned whether thyroid function would follow typical mammalian or reptilian trends in response to cold acclimation. We proposed the following competing set of hypotheses for investigation:1) thyroid gland activity and hormone secretion are elevated in response to prolonged cold exposure, thus conforming to mammalian norms, or 2) being poikilothermic, the naked mole-rat exhibits decreased thyroid activity in response to prolonged cold exposure. In testing this set of competing hypotheses we measured Tb, free thyroxine (T4), and thyroid-stimulating hormone (TSH) concentrations, as well as the distribution and ultrastructure of pituitary thyrotrophs and thyroid gland histomorphology in two groups of naked mole-rats housed for at least 18 mo at either 25 or 30°C.


Animal Care and Maintenance

Non-breeding pairs of naked mole-rats were housed in standard laboratory rat cages; only the males were used for this study. Perspex tubing was provided to simulate burrows, and sawdust served as bedding. Animals were fed an ad libitum diet of apple and sweet potato and received a high-energy, protein-rich, and vitamin-supplemented cereal (Pronutro, Becketts).

Thermal Acclimation

Naked mole-rats were housed at a Ta of 25 ± 1°C for a period exceeding 18 mo. A control group was maintained at its preferred housing temperature (30 ± 1°C), which is identical to that experienced by mole-rats in their natural habitat (24) and lies within their thermoneutral zone (6). At both housing temperatures, relative humidity was maintained >60%. The lower Ta (25°C) was selected as this represents a significant cold stress for these poorly insulated mammals, of comparable magnitude as housing other small mammals at considerably lower Ta levels of ∼5°C (47).

Body Temperature Measurements

A calibrated copper-constantan thermocouple was inserted into the rectum to a depth of ±2 cm, and the Tb was recorded to within 0.1°C via a digital display (model Bat-12, Physitemp).

Thyroid Function

Thyroid function was assessed by both histological and endocrinological methods. Animals were anesthetized by halothane (Fluothane, Zeneca) inhalation and killed by cardiac exsanguination.

Thyroid hormone analysis.

Blood from nine cold-acclimated naked mole-rats and six control animals was collected in heparinized syringes and immediately centrifuged at 3,000 rpm for 5 min. The plasma fraction was stored at −70°C for later analysis of TSH and free T4 levels in our laboratory using commercial immunoassay kits and methodologies supplied by the manufacturer (Nichols Institute Diagnostics, San Juan Capistrano, CA). These radioimmunoassays are based on the human form of the hormones, i.e., they use anti-human hormone antibodies, and, in the case of T4, involved equilibrium dialysis at 37°C (33,34). We assumed that in the case of the peptide hormone (TSH), its structure in the naked mole-rat is sufficiently similar to that of the human type to produce a high degree of cross-reaction with the three monoclonal antibodies supplied.

Histological investigations.

The thyroid glands of three naked mole-rats from each group were carefully dissected out, fixed in 10% buffered Formalin, and routinely processed for light microscopy. Serial sections (5 μm) were cut in the transverse plane and stained with hematoxylin and eosin. A flexible image processing system (Council for Scientific and Industrial Research) linked to a Leitz Laborlux II light microscope was used for the determination of follicular cell height. For each lobe, a ×400 magnification area was selected for the most active and least active follicles and the follicular cell height of eight well-nucleated cells, with distinct cell borders, was counted in all four compass planes. This procedure was followed for every tenth section.

Pituitary Thyrotrophic Function


The pituitary glands of six naked mole-rats from each group were removed, fixed in Bouins fixative, routinely processed, and embedded in paraffin wax. Serial sections of each pituitary were cut at 5 μm through the entire gland. Sections from each animal were selected from five levels of the gland, from the ventral to dorsal surface, and immunostained for TSH cells (thyrotrophs) using a modification of the streptavidin-biotin technique. Sections were dewaxed in xylene and treated with 3% hydrogen peroxide in methanol for 20 min. They were then incubated in 10% rabbit serum for 30 min, followed by incubation with rabbit anti-rat TSH (UCB Pharmaceuticals) at a concentration of 1:4,000 in PBS for 1 h. After rinsing in PBS, sections were incubated with a biotinylated goat anti-rabbit IgG (1:200; Amersham, Weil) for 40 min. After further rinsing in PBS, sections were incubated with a streptavidin-biotin peroxidase complex (1:300; Amersham) for 20 min. All incubations were conducted on a hot plate at 37°C. The peroxidase complex was revealed with 3,3′-diaminobenzidene hydrochloride (Merck, Johannesburg, South Africa). Immunocytochemistry controls for each batch of sections stained were as follows:1) the primary and secondary antisera were replaced with horse serum or buffer and 2) increasing concentrations of TSH were absorbed with a 1:8,000 dilution of anti-TSH for the absorption control.

Electron microscopy.

The pituitary glands from three naked mole-rats from each group were fixed in 2.5% gluteraldehyde-PBS solution. The tissue was postfixed in 1% osmium tetroxide, dehydrated in increasing concentrations of alcohol, and cleared in propylene oxide. The tissue was then infiltrated and embedded in epon-avaldite resin and polymerized at 60°C for 48 h. The thin-semithin serial sectioning technique was used to cut 0.1-μm sections followed by 1-μm sections. Resin was removed in the sections by placing them into sodium ethoxide. These sections were then stained with the streptavidin-biotin technique to demonstrate TSH cells. Gold sections were mounted on grids and routinely stained with uranyl acetate and lead citrate. TSH cells that were immunostained on semithin sections were then matched on the gold sections and viewed with a JEDL 100S transmission electron microscope at 80 kV.

Statistical Analysis

Data are presented as means and SD. Data sets were analyzed using unpaired t-tests. Results were considered significant for P < 0.05.



Cold-acclimated animals had significantly lower Tblevels (31.1 ± 1.7°C) than the control group (33.1 ± 0.6°C; P = 0.0281), but the temperature differential (Tb − Ta) was much greater (P < 0.01) for the cold-acclimated animals (6.1°C) compared with that of the control group (3.1°C).

Endocrine Status

Concentrations of TSH varied considerably within each group and were not significantly different (P > 0.05) between the two groups (Table 1). Cold-acclimated animals had 40% higher (P < 0.01) free T4levels than the control group (Table 1).

View this table:
Table 1.

Chronic cold-induced changes in TSH, T4, and thyroid morphology in naked mole-rats

Thyroid Gland Histology

Follicular cell height was significantly greater in the cold-acclimated animals for both the active (P < 0.0001) and the inactive areas (P < 0.0001) of both lobes (Table 1). Furthermore, the thyroid gland of the cold-acclimated animals was distinguished by dense areas of follicular cells with reduced colloid and was more extensively vascularized (Figs. and 1 and 2).

Pituitary Thyrotrophic Immunocytochemistry

Distribution of thyrotrophs within the adenohypophysis appeared similar in both groups and appeared to be more concentrated in the middle region of the gland (Fig.3 A), with fewer cells in the more ventral and dorsal regions (Fig. 3 B). The cells ranged from circular to spindle shaped, all with eccentrically placed nuclei. They were often found paired or in clusters closely associated with capillaries. Incubation with rabbit serum or buffer as replacement for rabbit anti-rat TSH showed no specific staining. The TSH antigen abolished staining at a concentration of 20 μg/ml.

Fig. 3.

Immunocytochemical detection (streptavidin-biotin) of adenohypophyseal thyrotrophs: middle region of the adenohypophysis (A) and ventral/dorsal region (B) of the gland. Scale bar (10 mm) is equivalent to 18 μm.

Pituitary Thyrotrophic Electron Microscopy

Immunoidentified thyrotrophs in cold-acclimated animals had a large nucleus that was eccentrically situated (Fig.4 A). The membrane-bound secretory granules were clustered mainly toward the periphery of the cell, close to the plasmalemma (Fig. 4 A). Thyrotrophs in the control animals also had large eccentrically placed nuclei, but the secretory granules were scattered throughout the cytoplasm (Fig.4 B).

Fig. 4.

Ultrastructure of immunoidentified thyrotrophs in the adenohypophysis of cold-acclimated (30°C) naked mole-rats showing peripherally located secretory granules (A) and thermoneutral-acclimated (30°C) control naked mole-rats with more evenly dispersed granules (B). Scale bar (10 mm) is equivalent to 1 μm.


Thyroid hormones have a major influence on metabolism, lipogenesis, and lipolysis (22, 39, 43), facilitating the plasticity of phenotypic responses to changing ambient conditions (22, 44). Although the thyroid gland is structurally conserved in all vertebrates, responses of the thyroid gland and concomitant thyroid hormone concentrations are considerably different across the phylum. Generally, the proportion of free T4 to total T4 remains relatively constant (<0.10%), with the fraction unbound in rodents accounting for 0.09% of the total concentration (40). Free T4 levels in the naked mole-rat were considerably lower than those reported for a wide variety of vertebrate species and are one order of magnitude lower than for other mammals (10, 40). Of course, it should be borne in mind that we used a clinical assay kit to measure T4 on the assumption that the mole-rat T4 behaves, under the conditions of the assay, the same way as human hormone. We believe that this is a reasonable assumption and that the low levels of T4 cannot totally be explained by an inappropriate measurement technique. Because most of the T4 data in the literature of poikilotherm thyroid activity are expressed as total T4, for comparative purposes we estimated the total T4 by assuming that the fraction unbound was the same as in other rodents. Assuming that 0.09% of T4 remains unbound (i.e., free T4; Ref. 40), total T4concentrations in the naked mole-rat would range between 4.3 (control) and 6.1 ng/ml (cold acclimated). These estimates of total T4 naked mole-rat values, in keeping with their poikilothermic mode of thermoregulation, are in a similar range to those reported for reptiles (23, 26) such as Chinese cobras (2–16 ng/ml; Ref. 2), iguanid lizards (1–5 ng/ml; Ref. 25), and western fence lizards (3.3–54.2 ng/ml; Ref. 26). Such low levels of thyroid activity reflect in part the extremely low metabolic rates of these anomalous mammals (6, 31), which is also more in keeping with reptilian rates (16). Although T4 is largely regarded as the principal circulating thyroid hormone, triiodothyronine (T3) is the more active metabolite and is produced in target tissues using the circulating supply of available T4 (38). Although T3 levels were not measured in this study and are involved in the physiological activities attributed to the thyroid gland, it is highly likely that the low levels of T4 would also lead to low T3concentrations and that these also would be correlated with the low metabolic rates exhibited by naked mole-rats.

TSH levels were similarly found to be lower than those of other small mammals (19). As the structure of TSH in naked mole-rats is currently unknown, this could be attributed to poor homology in the amino acid sequence and thus diminished cross reactivity with human TSH antibodies resulting in an underestimated hormone concentration. However, the fact that the free T4 levels are also very low supports our premise that the measured naked mole-rat TSH levels reflect real values: low free T4 levels measured cannot be attributed to poor cross-reactivity, because T4has an identical structure in all species so far studied. That these two hormone concentrations follow similar trends and are both low suggests that the values we obtained for TSH are indeed indicative of modest thyrotropic activity. Furthermore, the small concentrations of both these hormones correlate well with the very low basal metabolism of these mammals (6, 16, 31) and no doubt contribute substantially to the markedly reduced cellular heat production and poikilothermic profile of Tb with acute changes in Ta.

Contrary to most poikilotherms that show decreased thyroid activity in the cold, naked mole-rats conform to general mammalian trends by increasing thyroid activity, with a concomitant elevation in free T4 serum concentration. The 40% increase in free T4 levels is of similar magnitude to the observed changes in basal metabolic rate (50%, 47), suggesting that the increase in thyroid activity is intimately linked to the change in basal heat production and the concomitant increased temperature differential (Tb − Ta). Thyroid responses of other small mammals to prolonged cold exposure vary considerably with the extent of other physiological and morphological changes and also with the intensity of cold exposure and rate of change in ambient conditions (17). Whereas thyroid activity in mammals is inversely dependent on Ta, in contrast, the activity of the reptilian thyroid gland is directly dependent on Ta(12). Exceptions to these generalizations have been noted, and indeed some species of winter-active reptiles respond in a similar manner to naked mole-rats and show greater thyroid activity in winter than in summer (1, 29, 30).

Changes in thyroid gland activity are regulated by the hypothalamic-pituitary axis, thus one would expect a concomitant change in TSH concentration and anterior pituitary activity. The absence of a significant increase in TSH levels in the cold may be explained by the pulsatile nature of its release (3) and the likelihood that the peptide has a relatively short half-life (42). These contributing factors resulted in a large variation in TSH levels and similar mean values. However, thyrotrophs of cold-acclimated animals had numerous secretory granules concentrated at the periphery of the cell, suggesting enhanced secretory capacity and/or frequency (Fig. 4 A). Similar thyrotrophic morphology has been noted in hypothyroid mice where TSH concentrations are raised in response to the lack of inhibitory regulation via negative feedback mechanisms (36, 37). In the latter case, the secretory granules had either disappeared or were located along the cell periphery (36). Other evidence of increased TSH production is provided by changes in TSH-mediated thyroid gland morphology. Follicular cell height, an indicator of thyroid activity (15), was significantly greater in the cold-acclimated animals (Table 1, Figs. 1 and 2).

Fig. 1.

Thyroid gland histology of thermoneutral-acclimated (30°C) control naked mole-rats (A) and cold-acclimated (25°C) naked mole-rats showing increased follicular cell height with cold-acclimation (B). Sections were stained with hematoxylin and eosin. Scale bar (10 mm) is equivalent to XX μm.

Fig. 2.

Thyroid gland histology of thermoneutral-acclimated (30°C) control naked mole-rats (A) and cold-acclimated (25°C) naked mole-rats showing increased thyroid activity (B). Sections were stained with hematoxylin and eosin. Scale bar (10 mm) is equivalent to 12 μm.

Despite showing a typical mammalian response to cold exposure by augmenting thyroid hormone concentrations and presumably thyroid-mediated metabolic activities, these increases failed to completely restore Tb to the levels of controls at simulated burrow temperatures. This is attributed to the high thermal conductivity and poor insulatory properties of the naked mole-rat integument, which lacks a functional hypodermis (6, 8). Nevertheless, the temperature differential (Tb − Ta) for the cold-acclimated animals (6.1°C) was twice that of the thermoneutral-acclimated control animals (3.1°C) and this is attributed to increased thyroid-mediated thermogenesis. Maintenance of an elevated temperature differential may be important to this poikilothermic mammal, where physiological function (e.g., gut activity, pregnancy) and locomotion are temperature dependent. The 40% elevation in free T4 levels may be intimately linked to the 50% increase in basal metabolism and the lowering of the apparent thermoneutral zone (47). As such, minimal metabolism can be maintained over a lower range of Ta levels, thus facilitating an enormous energy saving.


Naked mole-rats, similar to all other animals studied to date (21, 27, 28), acclimate to cooler conditions and exhibit a suite of morphological and physiological changes that counteract the challenges imposed by these environmental conditions. This rodent, however, remains an anomaly of confounding contradictions and does not obey one particular thermoregulatory dogma. Rather, it shows both typical mammalian and reptilian responses to prolonged cold exposure as well as atypical features of each. Naked mole-rats conform to typical small mammal profiles and increase thyroid gland activity and basal metabolism. Despite these typical endothermic mechanisms, even after prolonged exposure to a 5°C drop in Ta, heat generation is inadequate to maintain Tb. This inability to regulate Tb may reflect limitations to endothermic metabolism imposed by other rate-limiting physiological systems or may reflect inadequate thyroid function. Elucidating the physiological constraints evident in their extreme responses to fluctuating environmental conditions may also explicate some of the mechanisms impeding sustained endothermy, which would otherwise only be apparent when more tolerant organisms are exposed to severely challenging situations. Furthermore, this unusual mammal may also provide new opportunities for the study of changes in other physiological mammalian systems with variable body temperature (such as may occur during clinical hypothermia). The reason for this is that, unlike every other mammal, known body temperature can be set at any desired level, in the same manner that it can in reptiles.

In conclusion, thyroid hormone concentrations are of a similar order of magnitude to those reported for other poikilotherms. Nevertheless, these rodents exhibit typical mammalian cold-acclimation responses with increased thyroid activity and basal metabolism under cooler conditions.


The authors thank Wilson Shai of the Central Animal Service of the University of the Witwatersrand for expert and meticulous care of the animals. Sincere thanks also to Alison Mortimer and Sherrie Rogers for expert technical skills in the preparation of the histological sections and photomicrographs, respectively.


  • Address for reprint requests and other correspondence: R. Buffenstein, Dept. of Biology, City Univ. of New York, 138th St. at Convent Ave., New York, NY 10031 (E-mail:rochelle{at}

  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract