2.2. Brown adipose tissue in mammals

The first time BAT was described as a tissue was almost 450 years ago (1551) when Konrad Gessner found it from hibernating marmots. At that time, the function of the tissue was unclear. The function of BAT as a specialized thermogenic tissue and as the main effector of NST was elucidated in the later half of the 20^th century (Heim & Hull 1966, Smith & Horwitz 1969, Foster & Frydman 1979), which aroused a high interest in the presence and function of this tissue in cold-adapted, newborn and hibernating animals. The species distribution of BAT is still not defined with certainty but, with the exception of domestic pigs (Trayhurn et al. 1989), it seems to be present only in mammals (Klaus et al. 1991, Trayhurn 1993). It is generally believed that BAT developed late in the course of evolution in parallel with the development of homeothermy, and more specifically with the capacity for the facultative thermoregulation by NST (Néchad 1986). This characteristic in adult mammals is considered a key feature in the long-term acclimation to cold. Birds are homeotherms that do not possess BAT (Saarela et al. 1989, 1991) but there is evidence that NST does occur in the skeletal muscles of birds, and that the characteristics of avian NST are probably different from those of the mammalian NST (Duchamp et al. 1999). Some kind of NST may also occur in lower vertebrates, as findings of a thermogenic tissue ‘brain heater’ in various species of fish have suggested (Carey 1982). Within mammals, BAT and NST are not limited to placental mammals but have also been found in marsupials (cf. Nicol et al. 1997).

There are large differences in the ontogenetic development of BAT in mammals between different species and developmental types. In precocial species, such as the rabbit and humans, BAT is well-developed and functional at birth (Néchad 1986, Nedergaard et al. 1986) whereas in altricial species such as the rat, the thermogenic capacity of BAT develops over the first days of life, and it does not develop in the "immature" hamster until three weeks after birth (Hissa 1968, Nedergaard et al. 1986). The cold-induced release of NA from the sympathetic nerve endings is considered necessary for the stimulation of thermogenesis in BAT (Jansky 1973, Nedergaard et al. 1986, Obregón et al. 1989). In precocial species, the onset of NST at birth also requires the severance of the umbilical cord, or the removal of the inhibitory effect of placental prostaglandins PGE₂ and the onset of another prostaglandin, PGI₂ that lungs start to produce when breathing is established (Ball et al. 1995). BAT in precocials develops to a certain degree in utero, as shown by the presence of UCP₁ in adipose tissues of bovine and ovine foetuses (Casteilla et al. 1987, 1989). The sympathetic nervous system obviously contributes the differentiation of foetal BAT without cold stimulus (Casteilla et al. 1994), but also other signals may be involved (Nedergaard et al. 1986). The expression of UCP₁ in the adipose tissues of foetal sheep and cattle is very low but increases sharply at birth (Casteilla et al. 1987, 1989, Trayhurn et al. 1993a), which agrees with the fact that NST in these species peaks at birth (Alexander & Bell 1975, Alexander et al. 1975).

During postnatal development, the BAT in precocial species transforms into a tissue that has the characteristics of WAT, while the amount of UCP₁ decreases (Casteilla et al. 1989, Trayhurn et al. 1993b). These changes are in parallel with the postnatal decrease in the capacity for NST (Gemmel et al. 1972, Alexander et al. 1975) - a change that also occurs in reindeer (Soppela et al. 1986). It has not been clearly established whether the transformation of BAT to WAT is due to the atrophy of brown adipocytes and their replacement by white adipocytes, or due to the transformation of brown adipocytes to white adipocytes. The most important physiological factor maintaining the thermogenic capacity of BAT or the expression of UCP₁ is the cold-induced stimulation of the sympathetic nervous system via adrenergic β₃-receptors (Arch 1989, Himms-Hagen 1991). There is evidence in bovine calves that if raised in cold environments, the disappearance of UCP₁ and its mRNA are delayed (Casteilla et al. 1989). In the adults of precocial species, BAT is usually absent. However, the reconvertibility of BAT is, in principle, possible as indicated by the ß₃-adrenergic stimulation of the adipose tissues in adult dogs (Champigny et al. 1991).

In the altricial species, BAT is present and active throughout the adult age (Himms-Hagen 1989). In addition to ontogenic growth, their BAT has a special growth that has been characterized by the term ‘recruitment’ (Nedergaard et al. 1986), and which results in the higher mitochondrial content of tissue (Lonèar 1991) and a higher amount and activity of UCP₁ in mitochondria (Cannon & Nerdergaard 1985, Klaus et al. 1991) and thus increases the proportional significance of BAT in the metabolism of an animal. The main factor that recruits BAT is chronic or regular sympathetic stimulation (Himms-Hagen 1991). In addition to cold, overeating and a high-fat ‘cafeteria’ diet activate thermogenesis in rodent BAT (DIT, diet-induced thermogenesis) (Rothwell & Stock 1979) and act as a mechanism to burn off excess fat and regulate body weight (Himms-Hagen 1989). Diets containing a high proportion of PUFAs have been found effective in increasing DIT in BAT, and PUFAs have been proposed to stimulate thermogenesis either centrally or peripherally (Nedergaard et al. 1983, Sadurskis et al. 1995, Oudart et al. 1997). Fatty acids are the main fuel for thermogenesis in BAT, and it has been proposed that long-chain fatty acids or their acyl CoAs interact directly with UCP₁ and act as a signal for the activation of thermogenesis (Nicholls et al. 1986, Boss et al. 1998, Lowell & Spiegelman 2000).

BAT has been considered the only tissue in mammals exclusively capable of producing heat by NST. Recent studies have indicated that WAT and other tissues also contain uncoupling proteins (Boss et al. 1998). Three new uncoupling proteins have been found. These have been named UCP₂ (Fleury et al. 1997) or UCP homologue, UCPH (Gimeno et al. 1997), UCP₃ (Vidal-Puig et al. 1997) and UCP₄ (Mao et al. 1999). Of these, UCP₂ and UCP₃ are 73 % identical to each other in their amino acid sequence and both are 56 % identical to UCP₁ (Lowell & Spiegelman 2000). UCP₂ has been detected in several tissues of adult humans, including WAT and skeletal muscles, and in mouse BAT (Fleury et al. 1997, Gimeno et al. 1997), and has been proposed to uncouple oxidative phosphorylation from ATP synthesis in a similar manner as UCP₁ (Fleury et al. 1997, Gimeno et al. 1997). UCP₃ is abundantly present in skeletal muscles in humans, and it is present in both BAT and skeletal muscles in rodents (Vidal-Puig et al. 1997). Neither UCP₂ nor UCP₃ are expressed in avian tissues (Denjean et al. 1999). Recent studies have also shown the expression of UCP₁ in rat bone marrow cell line adipocytes (Marko et al. 1995). These results suggest that tissues other than BAT may also be capable of producing heat through facultative thermogenesis. However, the uncoupling activity of the newly found UCPs has not yet been established. Their quantitative significance for thermogenesis is still controversial, and it is an active area investigation.