Insulin administration induces capillary growth in brown adipose tissue of heat-exposed rats

Camp. Biochem. Physiol.Vol. 103A, No. 4, pp. 673-678, 1992

0300-9629192$5.00 + 0.00

Printedin Great Britain

c> 1992 PergamonPressLtd

INSULIN ADMINISTRATION INDUCES CAPILLARY GROWTH IN BROWN ADIPOSE TISSUE OF HEAT-EXPOSED RATS HITOSHI YAMASHITA,* NOBORU SATo,t MINORU SUZUKIJ

MIKIOYAMAMOTO,~ Yuzo SATO,$KUNIAKI and HIDEKI OHNO*

NAKANISHI,(/

YOSHIAKI HABARA?

Departments of *Hygiene, fBiochemistry and //Pathology, National Defense Medical College, 3-2 Namiki, Tokorozawa 359, Iapan (Tel.: 0429-95-1211; Fax: 0429-95-0638); tTsukuba Research- Laboratories, Ninoon Glaxo Ltd. 43 Wadai. Tsukuba 300-42: GResearch Center of Health. Phvsical Fitness and Suorts. Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464; YLaboratory of Cellular Metabolism, Department of Molecular Physiology, National Institute for Physiological Science, Meidaiji-cho, Okazaki 444,

Japan (Received I3 April 1992; accepred 22 May 1992)

Abstract-l.

An apparent effect of insulin administration on enlargement of interscapular brown adipose tissue (BAT) was found in heat-exposed rats, but not in warm-adapted or cold-acclimated rats. 2. BAT extracts from the heat-acclimated/insulin-treated (HI) rats notably increased the capillary growth in an in vitro angiogenesis model in which microvascular fragments and myofibroblastic (Mf) cells isolated from lipid tissues were grown in co-culture, although a direct effect of insulin was not high. 3. BAT extracts from the HI rats stimulated the production of endothelial cell growth factor and collagen by Mf cells. 4. It is probable that an increased angiogenic activity contributes to the capillary growth and tissue growth in BAT of HI rats.

MATERIALS AND METHODS

INTRODUCTION

Animals

It has been shown that brown adipose tissue (BAT) is the main site of non-shivering thermogenesis and possesses an extremely rich vascular system for the export of the heat produced (reviewed by N&had, 1986). Himms-Hagen (1986) has also reported that BAT grows and regresses according to the actual or anticipated need for its thermogenic function and the amount depends not only on the species concerned but also on the age and prior environmental and dietary circumstances of the animal. The most effective and consistent way for an animal to increase the amount of its BAT is by acclimation to living at a temperature below its thermoneutral zone. Conversely, BAT mass is significantly smaller in heatacclimated (HA) rats than in warm-adapted (WA) rats (Habara and Kuroshima. 1983). In this study, however. an apparent effect of insulin treatment was noted only in HA rats in neither WA nor cold-acclimated (CA) rats, resulting in a remarkable gain in weight of interscapular BAT. On the other hand, it is known that angiogenesis is important in various physiologic or pathologic processes such as embryogenesis, lipid tissue development, wound healing, inflammation and solid tumor growth (Wasserman. 1965; Folkman. 1985). The enlargement of BAT also appeared to depend on the increase of capillary density, since the supply of oxygen and nutrients to BAT cells is limited by the distance from individual capillary units. The present study was, therefore. undertaken to investigate the relationship between angiogenesis and BAT enlargement under insulin treatment in heat environment.

(‘OPIA,ill1J

t

A total of 170 male Wistar strain rats (7-weeks-old) weighing 150-180 g at the beginning of the experiments were divided into two groups. The first group (N = 91) was reared at ambient temperature of 34°C with a relative humidity of 40% for 2 weeks in an artificial climatic room with lights on from 7 am to 7 pm daily in individual cages. These rats were referred to as HA rats. They had free access to the standard laboratory diet (Oriental MF. Oriental Yeast Co. Ltd. Tokyo) and tap water. The animals were cared for in accordance with the guiding principles in the care and use of animals based upon the Helsinki Declaration. Another group (N = 79). made up of control (WA) rats, was kept at the thermoneutral temperature of 25 C. Half of the rats from each group were subcutaneously (interscapular portion) injected with 3.62 nmol (equivalent to 0.5 I.U.) insulin Novo Lente MC (porcine/bovine) (Novo Industry, Copenhagen)/l25pl of saline per IOOg body weight for I3 days and the rest were treated with physiological saline as previously described (Habara, 1989). After cervical dislocation. the interscapular BAT was removed and dissected free from all recognizable white adipose tissue and other connective tissue and then weighed. The extracts of BAT from each group were prepared by homogenization in 2 ml phosphate-buffered saline (pH 7.4)/g tissue with a Potter Elvehjem homogenizer followed by centrifugation.

The effect of the BAT extracts from each group on capillary growth was estimated using an in ci/ru angiogenesis model as previously demonstrated in our group (Sato et ol.. 1987b). In brief. the microvascular fragments and myofibroblastic (Mf) cells prepared from rat epididymal fat pads were co-cultured in 12-well culture plates with 1.5 ml of Medium 199 (M-199) supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 U/ml penicillin G. 673

674

HITOSHI

YAMASHITA

100 pg/ml streptomycin and 0.25 pg/ml amphotericin B (Gibco Laboratories, Grand Island, NY) at 37°C in an atmosphere of 5% CO, in air. Mf cells grew actively and formed a monolayer, after which endothelial cells began to sprout from the microvessel fragments and link together to form long anastomosing cords extending over the Mf

cell monolayer. The cord-forming cells could be stained with anti-factor VIII antibody and electron-microscopically found to possess many characteristics as endothelial cells with the formation of tubular structure (Sato et al., 1987b). The culture medium was replaced after Mf cells were nearly confluent with fresh medium (5% FBS/M-199) containing the BAT extracts from each group, insulin, insulin-like growth factor I or insulin-like growth factor II (recombinant IGF-I and -II, Kabigen AB, Stockholm). After 4 days, the sprouting of endothelial cell cords from the microvessel fragments was photographically recorded (magnification 10x). The cellular cord density per defined area was determined by superimposing a transparent rectangular grid (6.9 mm spacing) and then counting the total number of intersections between cellular cords and grids, as previously described (Sato it al.. 1987a). Measurement qf‘ cell growth

The effects of BAT extracts from HA/insulin-treated (HI) rats were studied in the following experiment, because only the BAT extracts markedly increased capillary growth in uirro, as shown in Fig. 1.Since the growth of endothelial and Mf cells was essential to angiogenesis in our model, we began by testing the effects of the BAT extracts on bovine capillary endothelial (BCE) and Mf cells to examine their mode of action. BCE and Mf cells were isolated from bovine adrenal cortex and from rat epididymal fat pads, respectively. as described by Folkman et al. (1979) and Sato er al.

et

al.

(1987b). The cells were inoculated with I x IO4 cells/well into 24-well culture plates containing 0.5 ml of Dulbecco’s modified Eagle’s medium (DMEM) or M-199 supplemented with 10% FBS and antibiotics. The cells were cultured at 37°C in an atmosphere of 5% CO, in air. On the day after inoculation of the cells, the culture medium was replaced with fresh medium (5% FBS) containing the BAT extracts from HI rats. On days 4-5, the cells were harvested by trypsinization and enumerated using a particle counter model PC-607 (Erma Inc.. Tokyo). Measurement of collagen

synthesis

Collagen synthesis was measured according to the method of Tsuruoka el al. (1988). Confluent Mf cells were cultured in six-well culture plates containing 2 ml of 1% FBS/DMEM in the presence or absence of BAT extracts. After 3 days, the cells were washed three times with 2 ml DMEM, and then cultured in 0.9 ml DMEM containing 5Opg each of ascorbic acid and fi-aminopropionitrile and labelled for 24 hr with 5 PCi of L-[2,3-‘Hlproline. Next, 0. I ml of 2 N NaOH was added to each well and then the proteins in the medium, extracellular matrix and cells were collected together by scraping with a rubber policeman and by pipetting. The proteins of the lysate were precipitated by adding 0.2ml of 50% TCA containing 5% tannic acid on ice. Precipitable proteins were washed three times with 10% TCA containing 1% tannic acid, once with cold acetone. and then dissolved in 1 ml of 0.05 N NaOH. The solution was divided into two equal portions and added to an equal volume of 0.2 M Tris-HCl buffer (pH 7.3) containing 100 mM CaCl, and 5 mM N-ethyhnaleimide, and the mixture was incubated for 5 hr at 37°C with or without 25 units of purified collagenase form I11 (Advance Biofactures Co.. NY). TCA was then added to the mixture to a final concentration of 10% and proteins were precipitated. The radioactivity of proteins in the supcrnatant solubilized by collagenase was measured by a liquid scintillation counter. Mf cell-induced collagen synthesis was expressed as the radioactivity present in collagenase-treated proteins minus the radioactivity present in collagenase-untreated proteins. RESULTS

Effkct of heat acclimation BA T enlargement

and insulin administration

on

As shown in Table 1, an apparent effect of insulin treatment was noted only in HA rats, resulting in a remarkable gain in BAT mass of HI rats in terms of weight or weight per unit body weight about 3-fold of those from HA/saline-treated (HS) rats. Ejkcr

of BAT

extract

on capillary

growth

in vitro

The addition of the BAT extracts from all the groups resulted in an increased length of capillary cords (Fig. I). As well as the increase in mass of BAT, however, the effect of the BAT extracts from HI rats was considerably greater than those from other 0 0.17 0.67 BAT extract Fig. I. Induction

1.67 (%)

of in t+rro capillary growth by BAT

The BAT extracts from warm-adapted/salinetreated (WS) (0). warm-adapted/insulin-treated (WI) (a), heat-acclimated/saline-treated (I-IS) (A) and heat-acclimated/insulin-treated (Hl) (A) rats were added at the indicated concentrations to each well. After 4 days of culture, the density of capillary cords was measured. The mean control value (without extracts) was set to 100% and the density in the other cultures was expressed relative to this value. Mean values are shown i SEM (IV = 5 -7). extracts.

Bodv

GKlllp

BAT

wciuht

‘(Xl L

WS

(40)

216.4 i

3.3

WI

(391

218.6 f 2.x

HS

(37)

1x9.1 _tz

HI

(54)

18X.7 & 1.7

I

(1%)

(rng:

93.6 i

II

100g

bof_yy~v~~&ht)

5Y.2 + 4.7

124.3 _t K 9 131.x*

weirht

7

7.0

270.9 :i 2x.1*

61.3 i_ 5 6 50.2 + 4.

I

I48 7 + 16.1*

Capillary growth in BAT of heat-exposed rats

?g. 2. Photomicrographs of capillary cords developed from rat microvascular fragments. The microascular fragments and myofibroblastic cells were co-cultured in 5% FBS/M-199 without extract (A) or with 1.67% extract (B) of BAT from HI rats for 4 days, x25.

groups. In particular, when cultures were exposed to the medium contained 1.67% solution of the BAT extracts from HI rats, the capillary growth was markedly enhanced (9.6-fold) compared to the control value (Figs 1 and 2). In the presence of 1.67% extract, many cords sprouted from the fragments and then elongated, branched out, and anastomosed to each other forming a capillary-like network over the Mf cell layer (Fig. 2B), whereas the cords hardly elongated in the control culture (Fig. 2A). Moreover, a direct effect of insulin on capillary growth was observed at the range of 10~8-10~6M.

although the degree was not high (about 140% of the control) (Fig. 3). On the other hand, there was no effect of IGF-I or -II on capillary growth (data not shown). Effect of BAT Mf cells

extracts

on the growth

of BCE and

To investigate the mechanism of capillary growth by the BAT extracts from HI rats, the effect of the BAT extracts on the growth of BCE and Mf cells was examined. The BAT extracts stimulated BCE and Mf cell growth to about 150% of the control (Figs 4

HITOSHI YAMASHITA et al.

616

_I

0 -::i

0

10-8

Insulin(M)

Fig. 3. Induction of Insulin was added at well. After 4 days of was measured. Mean

l-k

10-7

in vitro capillary growth by insulin. the indicated concentrations to each culture, the density of capillary cords values are shown k SEM (N = 5-6).

0 0.5 BATextract

11

to-6

2.5 (%)

Fig. 4. Effect of BAT extract from HI rats on the growth of BCE cells. The cells were cultured in 5% FBS/DMEM in the presence or absence of the BAT extract. On day 4, the cells were detached by trypsin treatment and then counted in a particle counter model PC-607 (Erma Inc., Tokyo). Mean values are shown k SEM (N = 3).

and 5). Interestingly, the medium conditioned with Mf cells in the presence of the BAT extracts markedly increased BCE cell number in a dose-dependent manner (Fig. 6). In particular, the conditioned

1

Extract(%)

0

0

CM

-

+

0125 +

05 +

125 +

25 +

Fig. 6. EfTect of BAT extract from HI rats on the secretion of endothelial cell growth factor by Mf cells. Mf cells were seeded into six-well culture plates in 10% FBS/M-199. After the cultures had reached confluency, the monolayers were cultured in 10% FBS/M-199 in the presence or absence of the BAT extract at the concentration of 0.25, 1.0, 2.5 and 5.0% for 3 days, after which the conditioned medium (CM) was collected. Meanwhile, BCE cells were inoculated at a density of 5 x IO’ cells/well in 24-well culture plate in 10% FBS/DMEM. On the day after inoculation, the medium was replaced with 0.25 ml DMEM and 0.25 ml CM of Mf cells. After 5 days, BCE cells were counted as described in Fig. 4. Mean values are shown + SEM (N = 3).

medium (CM) including the 2.5% extracts BCE cell growth by 9.4-fold. Collagen synthesis

enhanced

by klf cells

We examined whether the BAT extracts from HI rats affect the collagen synthesis by Mf cells, because the cells produce an extracellular matrix component which contributes to tubular formation by endothelial cells (Madri et al., 1983; Sato et al., 1987b). After a 3-day exposure of the extracts to Mf cells, the collagen synthesis increased in a dose-dependent manner (up to about 150% of the control, Fig. 7).

r

0 BAT extract Fig. 5. EtTect of dAT extract

(“/.)

of Mf cells. The cells were cultured in 5% FBSiM-199 in the presence or absence of the BAT extract. On day 5. these cells were counted as described in Fig. 4. Mean values are shown A SEM (N = 2). from

HI rats on the growth

025

125

BAT extract

25 (‘1.)

Fig. 7. Effect of BAT extract from HI rats on the synthesis of collagen by Mf cells. The collagen synthesis in the presence or absence of the BAT extract was measured as described in Materials and Methods. Mean values are shown + SEM (N = 3).

Capillary growth

in BAT

DISCUSSION

We have recently demonstrated that insulin administration strongly stimulates the growth of BAT in heat-exposed rats and that the enlargement of BAT is mainly due to the anabolic effects of insulin because of the increased levels of DNA, RNA, protein and triglyceride (Yamashita et al., 1992). The increased mass of BAT appeared to depend on an increase in capillary density, since the supply of oxygen and nutrients to BAT cells is limited by the distance from individual capillary units. Actually, in the active BAT, at least one third of its surface area is in contact with the surface of endothelial cells (Aherne and Hull, 1966). So, it seemed likely that a number of growth factors including the growth factor(s) for capillary endothelial cells are intimately concerned in the unusual enlargement of BAT. In the present study, therefore, we investigated the effect of heat acclimation and insulin administration on enlargement of BAT from the viewpoint of angiogenesis using cell culture. At first, the effect of the BAT extracts from each group on capillary growth was estimated using an in vitro angiogenesis model. In consequence, the BAT extracts from HI rats markedly increased capillary growth compared to those from other groups, suggesting an increased angiogenic activity in BAT of HI rats. On the other hand, the direct effect of insulin on capillary growth was not sufficient, although insulin directly stimulates the growth of various cells in vitro (Nagarajan and Anderson, 1982; Conover et ul., 1985). Likewise, no effect of IGF-I or -II on capillary growth was observed, despite the growth-promoting effect of IGF-I in oivo (Schoenle et al., 1982) and the expression of IGF-II mRNA in BAT of rats (Yamashita et al., 1992). Benzi and Girardier (1986) have revealed that at thermoneutrality (28-32°C) hypoglycemia after a single administration of insulin induces an increase in blood supply to BAT of rats, but not at room temperature (22 ‘C). Moreover, systemic insulin sensitivity appeared to be greater in HI and HS rats than in the respective control rats (unpublished data). Thus, one may speculate that the increased insulin sensitivity under heat environment is concerned in capillary and tissue growth in BAT, in spite of the weak effect in z,itro. We next showed that the BAT extracts from HI rats stimulate the growth of BCE and Mf cells. In addition, the CM from Mf cells treated with the BAT extracts strikingly enhanced BCE cell growth compared to that from untreated Mf cells. These results suggest that the BAT extracts also stimulated the production of endothelial cell growth factor by Mf cells, which produce a kind of endothelial cell growth factor (Mf-ECGF) (Sato et al., 1987b), and that the effect of the BAT extracts on capillary growth arose mainly through Mf cells rather than a direct effect of the BAT extracts for capillary endothelial cells. On the other hand, Mf cells closely akin to those from white adipose tissue were found among cultured BAT cells (data not shown). Dobson et aI. (1990) have recently demonstrated that cultured 3T3 adipocytes secrete a differentiation-dependent angiogenesis factor (I-butyrylglycerol). Moreover, endothelial cells produce basic fibroblast growth factor

of heat-exposed rats

677

(bFGF) and platelet-derived growth factor (PDGF) (Schweigerer et al., 1987; Dicorleto et al., 1983). The BAT extracts may thus contain several ECGFs such as MEECGF and bFGF, PDGF, or stimulating factors for the production of ECGFs. From the fact that BAT is made up of various cell types including endothelial cells, perivascular mesenchymal cells, preadipocytes and Schwann cells (Bukowiecki et al., 1982) and that the mRNAs of IGF-II and hepatocyte growth factor (HGF) are present in BAT (Yamashita et al., 1992), it may be that BAT enlargement is coordinately controlled by some of these growth factors. It has been shown that extracellular matrix components such as collagen contribute significantly to tubular formation by capillary endothelial cells (Madri et al., 1983; Sato et al., 1987b). In our study the BAT extracts from HI rats enhanced the collagen synthesis by Mf cells, supporting an angiogenic stimulatory effect of the BAT extracts. Viewed together, these data seem to lead us to conclude that BAT extracts from HI rats stimulate angiogenesis in vitro by activating both endothelial and Mf cells. It is probable that an increased angiogenie activity contributes to the capillary growth and tissue growth in BAT of HI rats, although the reason why insulin administration results in the BAT enlargement only in HA rats remains to be explained. At all events, further investigation regarding angiogenesis in BAT would provide valuable information to understand the mechanism of BAT enlargement. Acknowledgements-The authors thank Mr Masahiko Segawa for his excellent technical assistance. This work was supported in part by a grant-in-aid for scientific research from the Daiwa Stock Foundation on Health and Welfare.

REFERENCES

Aherne W. and Hull D. (1966) Brown adipose tissue and heat production in the newborn infant. J. Pa/ho/. Bacferiol. 91, 223-234. Benzi R. H. and Girardier L. (1986) The response of adipose tissue blood flow to insulin-induced hypoglycemia in conscious dogs and rats. Pytigers Arch. 406, 3744. Bukowiecki L., Collet A. J., Follea N., Guay G. and Jahjah L. (1982) Brown adipose tissue hyperplasia: a fundamental mechanism of adaptation to cold and hyperphagia. Am. J. Physiol. 242, E3533E359. Conover C. A., Hintz R. L. and Rosenfeld R. G. (1985) Comparative effects of somatomedin C and insulin on the metabolism and growth of cultured human fibroblasts. J. Cell. Physiol. 122, 133-141. Dicorleto P. E. and Bowen-Pope D. F. (1983) Cultured endothelial cells produce a platelet-derived growth factorlike protein. Proc. natn. Acad. Sci. U.S.A. SO. 1919-1923. Dobson D. E., Kambe A., Block E., Dion T.. Lu H., Castellot Jr. J. J. and Spiegelman B. M. (1990) I-Butyrylglycerol: a novel angiogenesis factor secreted by differentiating adipocytes. Cell 61, 2233230. Folkman J., Haudenschild C. C. and Zetter B. R. (1979) Long-term culture of capillary endothelial cells. Proc,. nain. Acad. Sri. U.S.A. 76, 5217-5221. Folkman J. (1985) Tumor angiogenesis. Adr. Cancer Res. 43, 175-203. Habara Y. (1989) Augmentation of secretagogue-induced amylase secretion in pancreatic acini of heat-exposed rats. J. Physiol. 413, 91.-105.

HITOSH~ YAMASHITA el ul

678

Habara Y. and Kuroshima A. (1983) Changes in glucagon and insulin contents of brown adipose tissue after temperature acclimation in rats. Jpn J. Ph~3iol. 33, 661-665. Himms-Hagen J. (1986) Brown adipose tissue and coldacclimation. In Brokt*n Adipose Tissue (Edited by Trayhurn P. and Nicholls D. G.), pp. 214-268. Edward Arnold, London. Madri J. A., Williams S. K.. Wyatt T. and Mezzio C. (1983) Capillary endothelial cell cultures: phenotypic modulation by matrix components. J. Cell Biol. 97, 153-165. Nagarajan L. and Anderson W. B. (1982) Insulin promotes the growth of F9 embryonal carcinoma cells apparently by acting through its own receptor. Biorhwn. hioph~,.~. Res. Commun.

106, 974-980.

N&had M. (1986) Structure and development of brown adipose tissue. In Brown .-ldipo.w Tissue (Trayhurn P. and Nicholls D. G.), pp. I 30. Edward Arnold, London. Sato N., Fukudo K., Nariuchi H. and Sagara N. (1987a) Tumor necrosis factor inhibiting angiogenesis in IGrro. J. n&n. Cancer Inst. 79, 1383--l 391. Sato N., Sawasaki Y., Senoo A., Fuse Y.. Hirdno Y. and Goto T. (1987b) Development of capillary networks from rat microvascular fragments in ritro: the role of myofibroblastic cells. Microraw. Rev. 33, 194 210.

Schoenle E., Zapf J.. Humbel R. E. and Froesch E. R. (1982) Insulin-like growth factor I stimulates growth in hypophysectomized rats. Nature 296, 252-253. Schweigerer L., Neufeld G.. Friedman J., Abraham J. A.. Fiddes J. C. and Gospodarowicz D. (1987) Capillary endothelial cells express basic fibroblast growth factor, a mitogen that promotes their own growth. Nature 325, 257-259. Tsuruoka N., Sugiyama M.. Tawaragi Y.. Tsujimoto M., Nishihara T.. Goto T. and Sato N. (1988) Inhibition of in vitro angiogenesis by lymphotoxin and interferon-y. Biorhem. hiophJx Res. Commun. 155, 429. 435. Wasserman F..(1965) The development of adipose tissue. In Handbook of Phv.Golopv, Vol. 5 (Edited bv Renold A. E. and Cahill G.-F.). $J. 87-100. ‘Am. Physiol. Sot.. Washington D.C. Yamashita H.. Sato N.. Yamamoto M., Gasa S.. Nagasawa J.. Sato Y.. Habara Y., lshikawa M., Segawa M. and Ohno H. (1992) Is there an intimate interplay between temperature-acclimation and exogenous insulin?-with special reference to the participation of brown adipose tissue. In Mrdicine und Sport Science (Edited by Sato Y . Poortmans J. R.. Hashimoto 1. and Oshida Y.) Vol. 37. pp. 237 242. Karger. Basel.