вход по аккаунту


In vivo morphometry and functional morphology of brown adipose tissue by magnetic resonance imaging.

код для вставкиСкачать
THE ANATOMICAL RECORD 231293-297 (1991)
In Vivo Morphometry and Functional Morphology of Brown Adipose
Tissue by Magnetic Resonance Imaging
Institute of Human Anatomy and Histology, University of Verona, I-37134 Verona ( A S . ,
C.Z., F.O.) and N M R Research Laboratory, I.R.C.C.S. S. Raffaele, I-20132 Milano
(A.M.B., A.B.), Italy
Brown adipose tissue (BAT) is the main effector of nonshivering
thermogenesis and diet-induced thermogenesis in mammals. Assessment of the
magnitude and perturbations of BAT deposits in the intact, living body would be
of much relevance for quantitative studies of BAT functions, but such studies have
been impossible to date. In this paper i t is shown that magnetic resonance imaging
(MRI) morphometry can provide the means for accurate, repeated determinations
of the volume of BAT deposits in a living animal; moreover, tissue modifications
due to acclimation at different ambient temperatures are revealed in vivo by MRI,
which correlates with histology and ultrastructure. Furthermore, MRI differentiates areas of BAT responsive to acute adrenergic stimulation, thereby giving information on the thermogenetically active tissue in the intact animal. Therefore,
MRI represents a reliable tool for correlative morphological and functional studies
of BAT in the living animal.
The two types of adipose tissue namely, the brown
adipose tissue (BAT) and the white adipose tissue
(WAT), are morphologically and functionally different
(Nedergaard and Lindberg, 1981; Hull and Segall,
1966; Nechad, 1986). In the brown fat cell the lipid
deposit is multivacuolar instead of monovacuolar. In
BAT mitochondria, which are large, numerous, and
have parallel, tightly packed cristae, the oxidation of
fatty acid is uncoupled from the synthesis of ATP,
thereby dissipating energy as heat (Nicholls and
Locke, 1984). Further, BAT is provided with a n exceedingly rich sympathetic innervation (Cottle, 1970); noradrenaline release at nerve terminals is the trigger for
activation of thermogenesis in brown fat cells (Girardier and Seydoux, 1986).
The physiological role of BAT as a source of heat
(nonshivering thermogenesis) is well established in
newborn mammals and the arousing hibernator
(Smith and Hock, 1963; Dawkins and Hull, 1964). In
the adult animal BAT contributes to diet-induced thermogenesis (Rothwell and Stock, 1979) and, possibly, to
cachexia associated with cancer (Brooks et al., 1981).
During cold exposure, BAT blood flow is markedly enhanced and gives significant contribution to thermogenesis (Foster, 19833. In man, a functionally deficient
or hyperactive BAT has been suggested to play a role in
the pathogenesis of obesity (Himms-Hagen, 1984) and
in pirexial forms of cot death (Lean and Jennings,
19891, respectively.
In rodents, large BAT deposits are found in the cervical, interscapular, and axillary regions. In vivo identification of BAT deposits as distinct from WAT is not
feasible by means of X-ray imaging, and thermography
(Rothwell and Stock, 1979) has proven unreliable for
precise localization and determination of BAT deposits
(Dauncey et al., 1983).
Recently we demonstrated that magnetic resonance
imaging (MRI) defines BAT deposits in the living animal (Osculati et al., 1989); moreover, high spatial resolution MRI proved effective in anticipating some cellular characteristics of brown adipocytes at different
ages (Osculati et al., 1991). We now provide evidence
that the magnitude of BAT deposits can be determined
by a combination of MRI and morphometry, and the
dynamics of the tissue investigated as well, thereby
allowing investigation of the functional morphology of
the tissue in vivo.
A total of 15 male rats were used. Animals were
maintained at room temperature and feed ad libitum
unless otherwise specified. The animals were pre-anesthetized with ether, anesthetized with Nembutal (40
mglkg IP), and examined in a n SIS 2001330 imager
spectrometer (SIS Co., Freemont, CAI, equipped with a
4.7 T Oxford magnet with a 33 cm bore. The pulse
sequence timing was optimized for the contrast among
the histological parameters of interest: the pulse sequence consisted of 16 contiguous multislice spin echoes with a TR = 0.3 sec and TE = 0.03 sec. The actual
voxel size was 0.2 x 0.2 x 1.5 mm and 16 transients
were required for a n adequate sensitivity SIN. MRI
pictures were printed at a final magnification of 7 x .
Semi-automatic image analysis of MRI prints was performed in a Videoplan image analyzer (Kontron). Total
cervical and interscapular BAT volumes were calculated according to the following formula: V(obj) = t x
Received January 23, 1991; accepted May 24, 1991.
Address reprint requests to A. Sbarbati, Instituto di Anatomia
Umana ed Istologia, Strada Le Grazie 8, 1-37134 Verona, Italy.
Fig. 1. Representative series (1-6) of contiguous axial MRI pictures encompassing the whole cervical
BAT deposit of a 1-month-old male Wistar rat. The cervical BAT deposit profile is triangular, with its
base resting on the vertebral a r c h star, cervical medulla. Bar = 1 mm.
Ca(sec), where V(obj) is the total BAT deposit volume,
t is the slice width, and Ca(sec) is the C of the areas of
all cross sections of the object. After the last MRI examination, some animals were killed, frozen, and sectioned on planes correspondent to those visualized by
MRI. In other animals fragments of BAT and WAT
were excised, fixed in 2% glutaraldehyde in phosphate
buffer for 2 hours, postfixed in 1% osmium tetroxide for
1 hour, dehydrated, and embedded in Epon-Araldite.
Semithin sections were stained with toluidine blue.
Thin sections were stained with lead citrate and uranyl
acetate and viewed in a Zeiss EM 10 electron microscope.
interscapular BAT deposit of a 2-month-old rat upon
subsequent acclimations in the cold, near the thermoneutral point, and again in the cold (SOC). In the interscapular BAT deposit of 28%-acclimated animals a homogeneously high signal intensity was found (Fig. 2d);
lipid-laden brown adipocytes were prevailing in histological sections of BAT from parallel-acclimated rats
(not shown). Acclimation and re-acclimation in the cold
(Fig. 2c,e) resulted in lower signal intensity in the ventrorostral portion of the deposit, where a majority of
obviously multivacuolar adipocytes are found histologically (Osculati e t al., 1991). The dorsocaudal portion,
with predominantly monovacuolar cells (Osculati et
al., 19911, was less affected. Light and electron microRESULTS
scopy of the ventrorostral IBAT showed marked lipid
In the dorsal region of the r a t two large BAT depos- loss in brown adipocytes of cold acclimated animals in
its, the cervical (CBAT) and the interscapular (IBAT) comparison with warm acclimated controls (Fig. 3).
The ability of MRI to display function-related
are found, which are both visualized by MRI (Figs. 1,2).
The size of BAT deposits in vivo (Table 1) was deter- changes in BAT was studied by experiments of in vivo
mined on prints of MRI pictures encompassing the en- administration of noradrenaline (NA) to rats. Within
tire cervical and interscapular BAT deposits of young minutes from the IP injection of NA the signal intenrats. Resolution was in the mm range and the spin-echo sity from the ventrorostral portion of the rat IBAT (i.e.,
sequence differentiated BAT from surrounding tissues the location of prevailing multivacuolar adipocytes) bewith high contrast. Gravimetry of the same BAT de- comes reduced in comparison with the basal condition
posits obtained a t autopsy showed BAT mass values in (not shown). This effect of NA is enhanced in the cold
acclimated rat (Fig. 4) and completely prevented by
agreement with in vivo measurements (Table 1).
The effect of acclimation at different ambient tem- injection of the adrenergic antagonist propranolol (not
peratures on BAT is visualized noninvasively by mul- shown). Dissection of NA-injected animals after MRJ
tiple MRI examinations of the same, living animal. revealed a dark-brownish BAT with dilated vessels,
Figure 2c- shows representative MRI pictures of the mainly in its ventrorostral portion.
TABLE 1. The size of BAT deposits in vivo'
Calculated volume (mm3)
Wet weight (mg)
46.3 2 1.96
43.2 L 0.80
140.0 2 9.92
133.7 2 6.25
'The volumes of the cervical and interscapular BAT deposits (CBAT
and IBAT, respectively) in vivo were calculated on MRI pictures by
means of image analysis and morphometry. Results are the mean -t
one SD of four I-month-old rats. After MRI procedures animals were
killed with ether, CBAT and IBAT were dissected out free of contaminant white fat, and weighed fresh. Gravimetric and morphometric
figures fall in a very narrow range. Note that the density of BAT is
< 1 because of the high lipid content of the tissue; this accounts for
some of the discrepancy between the two measurements.
Fig. 2. Interscapular BAT (IBAT) deposit of 2-month-old male
Wistar rat. a: axial section; right panel, in vivo MRI picture; left
panel, frozen section of the same animal. C, cutis; W, white adipose
tissue; B, brown adipose tissue; M, muscle. b sagittal section of a
2-month-old male Wistar rat acclimated at room temperature. W,
white adipose tissue; B, brown adipose tissue; M, muscle. c-e: sagittal
MRI pictures of a 2-month-old male Wistar rat sequentially acclimated a t 8, 28 (thermoneutrality), and 8°C. Each acclimation period
was a t least 2 weeks. c, cold acclimation; note a region of lower signal
intensity (arrow) in the rostroventral portion of the IBAT. d, acclimation around the thermoneutrality point; note enlarged size of the
IBAT deposit which shows now homogeneously high signal intensity.
e, the same animal upon re-acclimation in the cold. The IBAT deposit
is reduced in size in comparison with a and b and a rostroventral area
of lower signal intensity (arrow) is again visible. a, x 4; b, x 4.5; c,
x 3.5; d, x 3; e, x 2.3.
Quantification of BAT in the living body would be of
much importance to assess the role of the tissue in the
dynamics of energy balance and body weight. Magnetic
resonance imaging allows precise in vivo definition of
BAT deposits (Osculati et al., 1989). The present paper
demonstrates for the first time that morphometry of
BAT deposits is feasible in vivo by means of MRI, thus
enabling calculation of deposit's volumes. This methodology has the advantages of multiple examinations
in the same subject, multiple planes of section of the
same structure, absence of artifacts due to fixation, embedding, and sectioning. It can be therefore utilized in
prospective studies as well a s in association with conventional morphometric procedures. The main drawback (beside the costs for the basic instrumentation) is
that resolution is relatively low in comparison with
light microscopy; however, when higher spatial resolution is needed, exploitation of high-field magnets coupled with a mini-imaging system can give good results
(Osculati et al., 1991; Sbarbati e t al., 1991).
The lipid content of the brown fat cell roughly mirrors its functional status, being larger in the relatively
quiescent cell and reduced in the actively thermogenetic cell (Senault et al., 1981). In a previous paper we
showed that MRI differentiates areas of prevailing
monovacuolar or multivacuolar brown adipocytes (Osculati et al., 1991). The present work shows that
changes in the intracellular lipid content of BAT associated with acclimation at different ambient temperatures are identifiable in the same animal. Thus, the
present report reveals the in vivo morphological counterpart of different functional states of BAT.
The ability of MRI to display function-related
changes in BAT deposits is also shown by experiments
of in vivo administration of NA to rats. Noradrenaline
administration is a recognized procedure to acutely
stimulate BAT thermogenesis and NA has major effects on BAT blood flow (Foster and Frydman, 1978).
Injection of NA was associated with modification of the
magnetic resonance signal intensity from areas of
IBAT with prevailing multivacuolar adipocytes, such
changes being larger in cold acclimated animals, as
expected on the basis of function studies of BAT in
response to catecholamines (Himms-Hagen, 1986). In
particular, results presented here are consistent with
previous findings on the effect of NA injection on BAT
blood flow which required such a delicate method a s
injection of microspheres (Foster and Frydman, 1978).
Changes in signal intensity of portions of IBAT upon
NA administration are ascribable to modifications of
the tissue waterilipid ratio. Both increased blood flow
and lipolysis contribute in principle to the relative increase of the waterilipid ratio in IBAT during NA stimulation. The effect of NA on BAT vasculature is probably more important, since preliminary experiments
utilizing the chemical shift imaging technique, which
permits separate investigation of the lipid and water
components of tissues, showed undetectable changes of
interscapular BAT lipid signal before and minutes after NA injection (not shown).
Fig. 3
far by the lack of a reliable, safe, in vivo procedure of
BAT definition.
We thank Professor L. Kruger for comments on the
manuscript. The work was supported in part by grant
Programmi di Ricerca Sanitaria Finalizzata, Regione
Marche, Italy.
Fig. 4. Magnetic resonance imaging pictures of a 3-month-old male
Wistar rat, Sagittal sections of the interscapular region. a: cold acclimation (2 weeks a t 8°C); b the same animal 20 min after a single
injection of noradrenaline; note marked decrease of signal intensity of
the interscapular BAT deposit, mainly in its ventrorostral portion.
Bar = 1 mm.
We conclude that MRI is suitable for direct investigation of the volume and functional morphology of
BAT deposits in vivo. The MRI acquisition procedure
presented in this paper allows monitoring, in the same
animal, of the effects on BAT of acute and chronic
treatments (physical or pharmacological). Morphological or biochemical investigations of the brown fat deposit are possible after MRI, since the procedure is
quite safe and non invasive. The size of WAT deposits
could be assessed in parallel with the BAT, thereby
offering information about the behaviour of these two
metabolically “antagonist” tissues. Finally, MRI is, in
principle, suitable for investigation of BAT in nonobese
and obese human subjects; this has been hindered thus
Fig. 3. a: Light microscopy of the ventrorostral portion of the IBAT
deposit. Two-month-old male Wistar rat acclimated at room temperature. Adipocytes with abundant, large lipid droplets are visible. b,c:
Light and electron microscopy of the ventrorostral portion of the interscapular BAT deposit in a cold acclimated animal. Lipid droplets
(stars) are scanty and the cytoplasm is almost completely filled with
mitochondria. a, X 450; b, X 1,300; c, X 8,000.
Brooks, S.L., A.M. Neville, N.J. Rothwell, M.J. Stock, and S. Wilson
1981 Sympathetic activation of brown adipose tissue thermogenesis in cachexia. Biosc. Rep., 1509-517.
Cottle, W.H. 1970 The innervation of brown adipose tissue. In: Brown
Adipose Tissue. 0. Lindberg, ed. American Elsevier Publishing
Company, New York, pp. 155-178.
Dawkins, J.R., and D. Hull 1964 Brown adipose tissue and the response of new-born rabbits to cold. J. Physiol. (Lond.), 172:216238.
Dauncey, M.J., C. Haseler, D.P. Page Thomas, and G. Parr 1983 Influence of meal on skin temperatures estimated from quantitative IR-thermography. Experientia, 392360-862.
Foster, D.O. 1983 Quantitative contribution of brown adipose tissue
thermogenesis to overall metabolism. Can. J. Biochem. Cell.
Biol., 62t618-622.
Foster, D.O., and M.L. Frydman 1978 Nonshivering thermogenesis in
the rat. 11. Measurement of blood flow with microsuheres uoint to
brown adipose tissue as the dominant site of the calorigenesis
induced by noradrenaline. Can. J. Physiol. Pharmacol., 56t110122.
Girardier, L., and J . Seydoux 1986 Neural control of brown adipose
tissue. In: Brown Adipose Tissue. P. Trayhurn and D.G. Nicholls,
eds. E. Arnold, London, pp. 122-151.
Himms-Hagen, J . 1984 Thermogenesis in brown adipose tissue as an
energy buffer. N. Engl. J . Med., 311t1549-1558.
Himms-Hagen, J. 1986 Brown adipose tissue and cold acclimation. In:
Brown Adipose Tissue. P. Trayhurn, and D.G. Nicholls, eds. E.
Arnold, London, pp. 214-268.
Hull, D., and M.M. Segall 1966 Distinction of brown from white adipose tissue. Nature, 212t469-472.
Lean, M.E., and G. Jennings 1989 Brown adipose tissue activity in
pirexial cases of cot death. J. Clin. Pathol., 42r1153-1956.
Nechad, M. 1986 Structure and development of brown adipose tissue.
In: Brown Adipose Tissue. P. Trayhurn and D.G. Nicholls, eds. E.
Arnold, London, pp. 1-33.
Nedereaard, J.. and 0. Lindbere 1981 The brown fat cell. Int. Rev.
citoi., 74tis7-286.
Nicholls, D.G., and R.M. Locke 1984 Thermogenic mechanisms in
brown fat. Physiol. Rev., 64r1-64.
Osculati, F., F. Leclercq, A. Sbarbati, C. Zancanaro, S. Cinti, and K.
Antonakis 1989 Morphological identification of brown adipose
tissue by magnetic resonance imaging in the rat. Eur. J . Radiol.,
Osculati, F., A. Sbarbati, F. Leclercq, C. Zancanaro, C. Accordini, K.
Antonakis, and A. Boicelli 1991 The correlation between magnetic resonance imaging and ultrastructural patterns of brown
adipose tissue. J. Submicrosc. Cytol. Pathol., 23t167-174.
Rothwell, N.J., and M.J. Stock 1979 A role for brown adipose tissue in
diet-induced thermogenesis. Nature, 281 t31-35.
Sbarbati, A., F. Leclercq, C. Accordini, F. Franceschini, and K. Antonakis 1991 Magnetic resonance imaging of the vomeronasal
organ. Chem. Senses, 16t105-108.
Senault, C., G. Cherqui, M. Cadot, and R. Portet 1981 Cold-induced
developmental changes in fat cell size and number in brown adipose tissue of the rat. Am. J. Physiol., 240:E379-E383.
Smith, R.E., and R.J. Hock 1963 Brown fat: Thermogenic effector of
arousal in hibernators. Science, 140t199-200.
Без категории
Размер файла
1 174 Кб
morphology, morphometric, magnetic, imagine, brown, vivo, adipose, tissue, function, resonance
Пожаловаться на содержимое документа