Metabolism of Isolated Fat Cells

I. PZ$`E:CTS OF HORMOKES ON GLUCOSE MtiTABOLIShI ASD LIPOl,YSIS

MARTIN RODBELL

From the Laboratory of Nutrition and Endocrinology, National Institute of ArthCtis and dIetabolic Diseases,
            National Institutes of Health, Bethesda 14, Maryland

(Received for publication, August 5, 1963j

Increased attention has been focused on the metabolism of
adipose tissue and its marked sensitivity to various hormones
(for reviews, see Vaughan (1) and Jeanrenaud (2)). In an
attempt to obtain a homogeneous preparation of fat cells, it
was found that if rat adipose tissue is t,reated with collagenase,
fat cells are liberated.  Owing to their high fat content, the fat
cells can be separated from the more dense stromal-vascular cells
by flot,ation.

The metabolism of glucose and the response to various hor-
mones by free fat cells are reported in this paper.

EXPERIMENTAL PROCEIIURES

Male Sprague-Dawley rats (160 to 210 g) were used in these
st,udies and were fed, ad libitum, a high carbohydrate diet (3)
consisting of: ground whole wheat, SSyO; casein, 15y0; whole
milk powder, lo,,,
       w. cottonseed oil, 5yc; and required vitamins
and salts.

Materials-Crystalline zinc insulin was obtained from Eli
Lilly (Lot 288614, 26 units per mg).  ACTHI and Fraction V
bovine albumin were Armour products. Purified TSH was
furnished by Dr. Peter G. Condliffe and contained 6 units per
mg. Dilutions of the hormones were prepared in buffered
albumin on the day of the experiment.  Collagenase prepared
from C. histolylirum was obt,ained from Worthington Biochemical
Corporation (Lot 61424). Glucose-W4C2 was obt.ained from
Xew England Nuclear Corporation.

Preparation of Fat Cells-Plastic or siliconized glass vessels
from Clay-Adams, Inc., were used during the preparation and
incubation of the fat cells.

Rats were killed by dceapit,at,ion, the rpididymal fat pads were
removed and rinsed in 0.857; NaCl Folut,ion, and thin distal
portions from each pad were cut into three pieces.  Up to 1 g of
tissue \vas added to a siliconized 25-ml flask containing 3 ml of
albumin-bicarbonate buffer, 10 mg of collagenase, and 3 pmoles
of glucose per ml.  Incubations were carried out for 1 hour at 37"
in a met,abolic shaker. The bicarbonate buffer solution, pH
7.4, was prepared according to Cohen (4) with half the suggested
concentration of calcium ion and contained 4c/, Fraction V
bovine albumin which had been dialyzed against. the bicarbonate
buffer.  The same buffer was used for t,he met,abolic studies.
The tissue was dispersed into small fragments within 1 hour of

incubat,ion with collagenase.  Fat, cells were liberated from the

1 The abbreviations used are: ACTH, corticotropin; TSH,
thryoid-stimulating hormone.
2 C;lucose-U-W refers to the uniformly or randomly labeled
compound.

tissue fragments by gentle stirring with a rod. Liberation of
the cells was manifested by an increased turbidity in the medium.
Fragments of tissue still remaining after this treat,ment were
removed with forceps.  The suspension of cells was cent,rifuged
in polyethylene centrifuge tubes" for 1 minut.e at, 400 x g.  The
fat cells floated to the surface, and t,he stromal-vascular cells
(capillary, endot'helial, mast, macrophage, and epithelial cells)
were eedimented.  The stromal-vascular cells were removed by
aspiration, and the fat cells were washed by :usl)ending them
in 10 ml of n-arm (37") albumin buffer cont#aining t,he desired
conrent'ration of glucose and centrifuging for 1 minute at 400 x 9.
This procedure was repeated three times. Stromal-vascular
cells here absent, by histological examination, from the fat cell
preparation after three washes.  Fat droplets, which may have
been formed from t,he breakage of t.he fat, cells, floated more
rapidly t)o the surface than the fat cells and were aspirated from
the surface, after gently stirring the cell suspension.

For a Pet of experiments, fat cells were usually obtained from
the pooled adipose t#issue of three rats.  The washed cells ww
suspended in 15 to 20 ml of albumin-bicarbonat,e buffer con-
taining a given concentration of glucose, usually 3 pmoles per
ml.  The triglyceride concentration (fat cell content) was
generally bet,ween 30 and 10 ~molos of triglyceride )equivalent
to about 50 to 60 mg of tissue) per ml of suspension. Just
prior to dispensing t,he cells in the incubation vials, sufficient
glucose-W4C was added to the suspension to give a final specific
activity of about 0.1 PC per pmole of glucose.

The following met,hod was used for measuring and dispensing
the cells.  The cell suspension was swirled to ensure delivery of
uniform suspensions of cells and was immediately drawn up into
18 cm of plastic tubing3 attached to a 2-ml calibrated syringe
(lubricat,ed to maintain a tight fit) and then discharged into
plast.ic counting vials from the Packard Instrument' Company.
The volume of t,ubing was sufficient to contain 0.5 to 1.0 ml of
cell suspension, the usual amounts measured.  With this pro-
cedure, it was possible to dispense, without cell breakage, 0.5 ml
of cell SUSlJenSiOn int,o 40 vials within 3 minutes.

After the addition of hormones, etc., the vials were capped
with rubber serum stoppers fitted with hanging glass wells
purchased from the Kontes Glass Company. The well3 cow
tained cylinders of Whatman K-3. 1 paper rolled from 2- X S-cm
strips. Unless stated otherwise, incubations were carried out,
wit.h shaking, at 37" for 2 hours. The gas phase was 955;

3 h-o. 10 Tri~nsflex tubing, Irvington, Plastic Division, Freehold,
Sew Jersey.

375


376

Metabolism of Isolated Fat Cells.  I             Vol. 239, x0. 2

FIG. 1. Photomicrographs of free fat cells and stromal-vascular  not clearly visible on the photograph, stained nuclei were associ-
cells obtained from collagenase-treated rat epididymal adipose  at,ed with each sphere.  H, fat cells at higher magnification (210X)
tissue.  Suspensions of unfixed cells were stained with methylene  showing nuclei.  C, intact blood vessels and other st,romal-vascu-
blue.  -4, view of fat cells at low magnificat,ion (100X).  Although  lar cells (25X).

Oa-5C1 CO?. At least four replicate flasks were incubated in
each experiment.

Analytical Procedures-Radioactivity was det#ermined in a
Packard model 314 EX liquid scintillation counter. 14C lipids
and COZ were count,ed in a scintillation solution consisting of
0.451 2,5-diphenyloxazole and 0.05% p-bis-2-(phenyloxazolyl)-
benzene in t,oluene.  Bray's scintillation solution (5) was used
for counting 1%.glucose in the incubation medium.

At the end of the incubation period, 0.2 ml of Hyamine-10X
ljurchased from the Packard Insbrument Company was injected
onto the filter paper and 0.25 ml of 1 N sulfuric acid into the cell
suspension.  After t,he flasks were shaken for 15 minutes at room
temperature, the paper strips were transferred to 10 ml of scintil-
lat,ion fluid and counted.  Two drops of methanol were added
to the counting vials to increase the solubility of the Hyamine-
CO2 in the scintillation fluid.

For the determination of 14C lipids, the cell suspensions were
transferred to centrifuge tubes with glass stoppers and ext,racted
wit,h 5 ml of Dole's ex;traction mixture (6). After the mixture
had stood for 15 minutes at room temperature, 3 ml of water
and 3 ml of hesane were added, and the phases were allowed to
separate.  The lower phase was removed by aspiration and t,he
upper phase was washed with 3 ml of water.  Portions of the
upper phase were analyzed for ester content (7) with tripalmitin
as standard, free fatty acids by a slight modification4 of the
method of Dole and Meinertz (S), and total lipid radioact,ivity.
To determine radioactivity in triglyceride fatty acids, 1 ml of the
upper phase was evaporated and the lipid was saponified by
refluxing for 1 hour with 2 ml of ethanolic KOH (1.0 ml of Fatu-
rated KOH per 100 ml of 95'.;h ethanol, freshly prepared).  After
2 ml of wat,er were added, t.he sample nas neutralized to a brom-
cresol green end point and the fatty acids were extracted wit,h 3
ml of hexane.  A 2.0.ml aliquot of the latter was evaporated to
dryness, and 10 ml of scintillation fluid were added to dissolve

the residue.  Fatty acid content was determined by titration (8)
and generally agreed with the ester content of the original lipid
extract.  Since the lipids of adipose tissue are primarily tri-
glycerides (9), the difference in radioacbivity found in fatty acids
and that in total lipid was assumed to represent' radioactivity
in the glycerol moiety and is referred to as glyceride-glycerol.

Glucose in the medium was measured, without deprot,einiza-
tion, by the glucose oxidase procedure (10). The quantity of
glucose carbon converted to COz , glyceride-glycerol, and fat,ty
acids was calculated from the init.ial specific activity of the
glucose in the medium and the quantit.y of radioactivity in the
product,s. Results are expressed as micromoles of glucose per
mmole of triglyceride in the cell suspension.

         o         RESULTS
The low densit.y material released frotn adipose t'issue by

collagenase consisted of spheres which contained a nucleus and
were 50 to 100 M in diameter (Fig. 1.-l and B).  The thin cyto-
plasmic rim usually seen surrounding that fat globule in thin
sections of adipose tiseue (II), could not be seen in the free fat
cell suspensions. The fat cells were freely dispersed in the
medium; there was no clumping of the fat cells (Fig. L4).

The dense material which sedimented after treatment of
adipose tissue with collagenase, contained mast cells, macro-
phages, connective tissue cells, and intact blood vessels and is
referred to as st,romal-vascular cells. A view of this mat.erial
is shown in Fig. 1C.

Fat cells incubated in plastic vessels converted 29% of the
W-glucose in the medium to CO?, glyceride-glycerol, and fatt.y
acids (Table I).  xo change in t,he gross structure of the cells
was observed after 3 hours of incubation.  When the cells were
incubated in glass vessels, however, only 4% of the 14C-glucose
was converted to the measured products.  The loss in metabolic
activit,y was primarily due to the rupture of t,he cells by incuba-
tion in t,he glass vessels.  When the fat cells were homogenized,

4 S. S. Chernick, unpublished procedures.

the fat was released from t,he cells and formed large dronlets


February 1964

M. Rodbell

377

TABLE I

Glucose metabolism by free fat cells and stromal-vascular cells

Fat cells'.
Fat cells,b..
Fat cells, hu-
rnogenizedd.
St romal-vas-
culnr cells'

Incubation

Plastic"
Glassc

Plastic

Plastic

Glucose in medium converted to

co2

Glyceride-
glycerol

           %
9.0 f 0.2  13.2 f 0.5  6.0 * 0.05
2.1  f 0.5  1.2 f 0.3  0.5 f 0.7

oio o

<O.l  lo O

o Fat cells, containing 30 pmoles of triglyceride, were incubated
for 2 hours at 37" in 1.0 ml of albumin-bicarbonate buffer, pH 7.4,
containing 3 rmoles of glucose-U-% per ml.  Values are the mean
of 4 replications f standard error.

b Plastic counting vials, purchased from the Packard Instru-
ment, Company.

c Erlenmeyer flasks, 25.ml rapacity, not siliconized.
d Fat, cells, containing 150 pmoles of triglyceride, were ho-
mogenized at room temperature in a Potter-type glass homogen-
izer fitted with a Teflon plunger.

e Stromal-vascular cells represent the material sedimented at
400 X 9 after 2.5 g of adipose tissue was dispersed withcollagenase,
as described in "Experiment,al Procedure." The cells were
washed with albumin-bicarbonate buffer, and t,he entire amount
was suspended in 1.0 ml of buffer containing 3 rmoles of glucose-
U-1oC per ml. Experimental conditions were the same as those
employed with the fat cells.

which bore no resemblance to t,he fat cell spheres shown in Fig. 1.
Nuclei were not associated with the fat droplets.  The homoge-
nized preparations did not oxidize glucose or synthesize tri-
glycerides from glucose.  These results indicate t,hat the preser-
vation of cellular structure is required for the metabolism of
glucose by the fat cells.

Table I also shows that the stromal-vascular cells incubated in
plastic vessels oxidized less than O.lyO of the medium glucose
and did not synthesize triglycerides. Insulin had no effect, on
the metabolism of glucose by the stromal-vascular cells (see
below).

The quant,ity of glucose-TV% oxidized to CO? was propor-
tional to the amount, of fat cells (Fig. 2) and oxidation proceeded
at, a constant, rate for at least 3 hours (Fig. 3).  Addit.ions of
insulin (1 milliunit per ml) caused a 2.5.fold increase in glucose
osidation, which also proceeded at a constant rate for 3 hours.

The effects of insulin on glucose upt,ake and its metabolism
t,o carbon dioxide, glyceride-glycerol, and fatty acids are shown
in Table II. Insulin increased proportionately glucose upt.ake
and met,abolism. Approximately 50y0 of the labeled glucose
removed from the medium was accounted for a~ carbon dioxide
and triglyceride.  In these experiments, the over-all recovery of
radioacbivity in CO*, glyceride-glycerol, and fatty acids was
reproducible to within lOy0 for a set of six replicate flasks.
Lactic acid and glycogen formation were not measured but
probably accounted for the remainder of the product,s of glucose
metabolism (12).

The oxidation of glucose-W4C by the isolated fat cells w-as
increased by adding insulin in amount,s as small as 10 micro-
units per ml; this was the lowest concentration tested (Fig. 4).
A linear dose response was observed in t.he range of 10 to 100

ic.000~

i

                     :/
16,000-            /

4,OOOb
d
      5    IO      20               40
         pMOLES TRIGLYCERIDE

FIG. 2. Relationship between cell triglyceride content and
amount of glucose-U-"C oxidized by isolated fat cells. Each
point represents the mean of 4 replications. The oertical bar
through each point, represents 1 standard error. Cells were in-
cubated for 2 hours at 37" in 0.5 ml of albumin-bicarbonate buffer,
pH 7.4, containing 3 @moles of glucose-U-W per ml.  Initial spe-
cific activity of medium glucose, 203,000 d.p.m. per Mmole.

16OOC

MINUTES

FIG. 3. Time course of oxidation of glucose-U-WZ bv isolated
fat cells incubated Kth and without insulin. Each po"itnt repre-
sents the mean of 5 experiments f standard error (vertical bars).
0, control; o , insulin (1 milliunit per ml). Incubation condi-
tions same as described in Fig. 2. Fat cell concentration. 31
hmoles of triglyceride (TG) per ml.

TABLE II

  E$ect of insulin added in vitro on uptake and metabolism
      of glucose-U-W by isolated fat cells"
Fat cell suspension, 1 ml (28 pmoles of triglyceride), was in-
cubated for 2 hours at 37" in albumin-bicarbonate buffer, pH 7.4,
containing 3 pmoles of glucose-U-*X per ml; amount of insulin
added was 1 milliunit, per ml.

Conversion of glucose-U-W to

   pnde/mmole higlyceride
3.2 rt 0.04  4.4 + 0.16  2.0 f 0.04
5.4 Z!Z 0.08  G.4 f 0.12  3.6 f 0.02

a Mean values obtained from 6 replications f standard error.


378

Metabolism of Isolated Fat Cells.  I             1-01. 239, so. 2

microunits of insulin per ml at a glucose concentration of 3
pmoles per ml.  Increasing the insulin concentration to 1 and t'o
5 milliunits per ml did not, cause a further increase in glucose
oxidation. The same results were obtained when fatty acid
synthesis was measured.

The effects of increasing t,he medium glucose concentration
from 1 t.o 16.5 pmoles per ml on glucose metabolism are shown
in Fig. 5.  Maximal formation of CO? and fatty acids was ob-
served when the glucose concent,rat,ion was increased from 1 to
6.5 pmoles per ml. Glyceride-glycerol formation was not
changed by increasing glucose concent8rabions. .\t a glucose
concentration of 1 pmole per ml, addit,ion of insulin (1 milliunit
per ml) increased the conversion of glucose to COI, glyceride-

GLUCOSE-L-14C INTO CO,

                            f
1 '~~-+J--
   i


=L Ll `I1
0'6                             I
          IO 40 100      500    1000 5000

          INSULIN MICROUNITS/ML

FIG. 4. Effects of varying insulin concentration on glucose-U-
I'C oxidation by isolated fat cells incubated for 2 hours. Each
point represents the micromoles of glucose carbon oxidized per
mmole of triglyceride (mean of 4 experiments).  The vertical bars
represent 1 standard error.  Conditions same as in Fig. 2.

glycerol, and fatty acids by about 3- to 6-fold. =1 further in-
crease in CO, and fatt,y acid formation occurred in response to
insulin if the medium glucose concentrat,ion was increased from
1 to 3.5 pmoles per ml.  A%dditional glucose in the medium did
not change the amount of CO,, glyceride-glycerol, and fatt,y
acids that w-ere formed in response to insulin.

4
.

LA-
I          6      8     12      5     I8
        pMOLES GLUCOSE/IV'L

FIG. 5. Patterns of glucose-U-14C metabolism obtained by in
creasing glucose concentration in absence and presence of added
insulin.  Four equal portions of fat cells obtained from epididymal
fat pads of 3 rats were washed 3 times with 10 ml of albumin-b-
carbonate buffer containing 1.0, 3.5, 6.25, and 16.5 rmoles of glu
case per ml, respectively, and finally suspended in 10 ml of buffer
containing the same designated glucose concentration.  One
milliliter of the cell suspension (10 rmoles of triglyceride (TG) per
ml) was incubated, in quadruplicate, for 2 hours at 3i".  Insulin
added was 1 milliunit per ml.  Each point represents the mean +
standard error (vertical bars). Solid line, insulin; dashed linr,
control.

GROUP I

GROUP1

O INSULIN ADDED
H CONTROL

CO2 GG    FA      CO, GG    FA    CO, GG    FA

CO, GG    FA

              -CELLS-    -TISSUE-   -CELLS-   ----TISSUE-
FIG. 6. Metabalism of glucose-U-W by fat cells and adipose  means of 4 replications for fat cells and for 3 pieces of adipose t,is-
tissue obtained from fed rats and rats deprived of food.  Fat cells  sue weighing approximately 50 mg each.  T-ertical lines at top of
and pieces of adipose t,issue (45 pmoles of triglyceride per ml) were  each column represent standard error of the mean.  Croup I rz-
incubated in 1 ml of albumin-bicarbonate buffer, pH 7.4, for 2  fers to rats (190 to 210 g) that were not fed overnight; Croup II,
hours at 37".  Glucose concentration was 3 fimoles per ml. In-  rats (160 to 170 g) were fed.  L"G, triglyceride; GG, glyceride-glyc
sulin added was I milliunit, per ml.  The data are expressed as the  erol; FA4, fatty acids.


l`ebrnary 1964

M. Rodbell

379

The effect of insulin on the 1)attern of glucose met,abolism
in intact. adipose tissue and isolated fat cells is shown in Fig. 6.
`I`n-o groups of animals (3 rats per group) were used in this study.
Rats in Group I (190 to 210 g) were maintained on the high
carbohydrate diet for 2 weeks and were deprived of food for
24 hours prior to the experiment. Group II contained ram
(160 to 150 g) which were fed the diet for 5 days. For the
intact, tissue studies, approximately 50 mg of tissue w-ere cut
from the disbal portion of each fat pad and incubated for 2 hours
at 37" in medium rontaining 3 pmoles of glucose-U-W per ml.
The two pieces of tissue from each animal were incubated wit,h
and lvithout added insulin.  The remainder of the adipose tissue
from the three animals in each group was pooled and treated
with collagenase for 1 hour, and the fat cells were isolated in the
usual manner. The fat cells were incubated under the same
conditions as described for bhe intact tissue.
In contrast, to the results obtained v&h the isolated fat cells,
the intact adipose tissues indicated a larger variation in t,he
quantity of glucose converted to COZ, glyceride-glycerol, and
fatty acids. This was particularly not,iceable when insulin
was added to the medium.  Although there appeared to be less
glucose metabolized by the intact, tissue, t,he difference was not
statistically significant.

The "pattern" of gluco~ metabolism (glucose t,o COz, glycer-
ide-glyrerol, and fat,ty acids) in the tissue and free fat, cells was
similar in each group of rats.  The cells and tissue from the fed
rats (Groul) II), had a lower basal level of glucose metabolism
and a larger percentage of conversion of glucose to CO, and
fatty acids in response to insulin than those from fasted ram.
Several other hormones were examined for their effects on
g!ucose metabolism and fatty acid release by free fat cells.  The
effects 011 glucose metabolism are shown in Table III.  Oxytocin,
like insulin, stimulated the formation of COZ, glyceride-glycerol,
and fatty acids from labeled glucose.  The minimal effective
~onccnt,ration for the osyt~ocin effect was 0.2 pg per ml.  hddi-
tion of LIC'l`H, `NH, and epinrphrine also increased the forma-
tion of CO? and glyceride-glyc*erol from glucose, but,, unlike
insulin and osytocin, drprcssed fatty acid synthesis.

I.\l3l.E III

       I
Adaitions       Conversion of glucose-U-"C to"

cot

Glyceride-glycerol   Fatty acids

~rnokirnnlole lri&`lyceride

Now ..,, ..,, 1l.M f 0.18 Il.86 * 0.24 2.40 + 0.12
Insulin, 0.02 fig per

ml 5.13 f 0.24 ~ 7.08 + 0.14 5.70 * 0.09
Osytrlcin, 0.2pg per
1111   :3. tin f 0.X  S.48 It 0.48 3.54 zk 0.2-i
TSH, I .O gg per ml  Z.(i(i f 0.x G.OG + O.GG 1.62 + 0.12
ACTH, 1.2 rg per
ml.  3.48 f 0.42 8.4G f- 0.42 1.50 + 0.12
15pineplrrinc, 1 pg
per ml  ..j -4.14 zk 0.54 7.20 xt 0.54 1.86 f 0.12

`I Mean Z!Z standard error for 4 replications. 1.0 ml of cell sus-
pension (~11 triglyceride = 40 Pnlolesj incubated for 2 hours at
:V" in alhunliI,-~ica~l)onste lmffrr, pH 7.4, containing 3 pmoles of
glurosc-u-`~c per ml.

              TABLE IX-
Change in free fatty acid content of fat cells and medium
     in response to narious hormones

Additions"

Yet change in free
fatty acids*

                             ~eqimvmles triglyceride
iTone.       1.6 & 0.5
Insulin.         -3.4 * 0.5
Osytocin..      -2.0 ItI 0.7c
TSH. ., ,.,.    40.5 z?z 8.0
Epinephrine  
ACTH. . . . . . . . . . . . ..~......     73.8 f 6.0
                              54.0 f 2.0

a Same concentrations of hormones, cells, and incubation condi-
tions as described in Table III.  Initial free fatty acid concentra-
tion in medium, 0.2 req per ml.

*Mean of 4 replications & standard error.  Values are final
minus zero time free fatty acid content.

c Not significantly different from control; >0.05 <O.lO.

ACTH, TSH, and epinephrine stimulated the release of free
fatty acids by the isolated fat, cells (Table IV).  Insulin caused
a net, reduction in the free fatty acid content of the cells and
medium; oxytocin did not have a significant effect.

The cells used for determining the effects of insulin, oxytocin,
and the three Jipolytic hormones on both glucose metabolism
and fatty acid release were obtained from pooled adipose tissue
of three rats.  The amount. of cells obtained from this quantit)
of tissue n-as sufficient to measure each parameter in quadrupli-
cate.

IJISCUBSIOiX

These results indicate that fat cells separated from adipose
tissue maintain the intrinsic metabolic characteristics of this
tissue. The similarit,y in the patterns of glucose metabolism
(CO:, glyceride-glycerol, and fatty acid formation) by both
isolated fat cells and by intact adipose tissue indicates that the
major pathways of glucose metabolism have been preserved in
the isolated cells.  The isolated fat cells accomlt for essentially
all of the glucose medabolism observed in adipose tissue.

The dependence of CO, and fatty acid synthesis on the mc-
dium glucose concentration in isolated fat cells has also been
found in studies with epididymal fat pads (13). Glyceride-
glycerol format,ion was unaffected by increasing the medium
glucose concentration.  This has also been observed nith intact
tissue (13, 14) and probably rcflect,s the relatively decreased
availabilit,y of substrate to the Eml)denMeyerhof pathway as
more substrate becomes available to the other pat.hways. In
this regard, t'he quantity of glucose available for l~l~osl~horylation
int,racellularly It--as st,rongly dependent on the presence of insulin
as evidenced by the finding that an increased glurosc gradient
in the medium was not suflicient, at least at concentrations from
20 to 200 mg/lOO ml, to Cause an elevation of glucose metabolism
to the levels attained with insulin at low concentrations.  Studies
with epididymal adipose tissue have shown that. increasing
glucose concentrations to 2000 mg/lOO ml simulates the effect
of insulin on fat)ty acid synthesis and CO2 formation in t,he
presence of low glucose concentrations (13, 14).

In view of the exquisite sensitivity of rat adipose tissue to
insulin, it was of particular interest. to find that the free fat cells
responded to insulin over the same dose response range as t,hat


380                 Metabolism of Isolated Fat Cells.  I             Vol. 239, No. 2

reported for intact adipose tissue (15, 16). The magnitude of
the insulin effect on glucose metabolism in isolated cells was as
great as that in the tissue. These findings suggest that the
isolated fat cells may be useful for insulin assay studies.

Hormones other than insulin also increased the metabolism
of glucose by the fat cells.  The insulin-like effect of oxytocin
on glucose metabolism may be related to the disulfide bridge
which is common to both hormones, as suggested by Pittman
et al. (17) in bheir studies with adipose tissue.  However, insulin
at' 0.1 the concentration was more effective than oxytocin.  The
effects of lipolytic hormones on glucose metabolism and lipolysis
in the free fat cells were the same as those observed in intact
adipose tissue (ACTH (12,18-21) ; TSH (19,22) ; and epinephrine
(12, 17, 23-27)).

SUMMARY

Studies Deere made on the metabolism of isolated fat cells and
stromal-vascular cells prepared by collagenase treatment of rat
epididymal adipose tissue.

The isolated fat cells metabolized glucose by the same path-
ways as intact adipose tissue and accounted for essentially all
of the glucose metabolism observed in this tissue.  Free fat cells
also maintained the different metabolic characteristics observed
in adipose tissue from fasting and fed rats.

Increasing the medium glucose concentration to a maximal
level of 6.5 pmoles per ml stimulat#ed the formation of COZ and
fatt,y acids from glucose. Insulin caused a further a-fold in-
crease in the uptake of glucose and its utilization to CO*, glycer-
ide-glycerol, and fat,ty acids.  As little as 10 microunits of
insulin per ml stimulated glucose metabolism by free fat cells;
the insulin dose response range was 10 to 100 microunits per ml.

Oxytocin, at 10 times t,he maximal effective concentration of
insulin, caused an insulin-like effect on the metabolism of glucose
by free fat cells. Corticotropin, thyroid-stimulating hormone,
and epinephrine stimulated the conversion of glucose to CO2
and glyceride-glycerol and the release of free fatty acids; they
suppressed fatty acid synt.hesis from glucose.  Insulin, but not
oxytocin, caused a net' disappearance of free fat,ty acids from
medium and cells.

It is concluded that isolated fat cells retain the ability to
metabolize bot.h glucose and triglycerides and respond to several
hormones that have been shown to affect the metabolism of
adipose tissue.

9 cknowledgments-I would like to express my appreciation
for the excellent technical assistance given by Miss Ann Butler
during the course of this study.

              REFERENCES
1. V~~~~~~,M.,J.LipidResearch.2,293 (1961).
2. JEANREKAUD, B., &ldabolism, 16,535 (1961).
Scow. R. 0.. Am. J. Phvsiol.. 173. 199 (19531.

3.
4.

5.
6.
7.


8.


9.


10.


11.
12.


13.


14.


15
16.


17.

18.


19.

20.
21.


22.
23.


24.


25.


26.


27.

COHEN, P. P., in W. WY. U&RI&, R.`H. BURRIS, ~\ND J. F.
STAUFFER (Editors), Manometric technipzces and tissue
metabolism, Ed. 2, Burgess Publishing Co., Minneapolis,
1951, p. 119.

BRAY, 6. A., Anal. Chem., 1, 279 (1960).
DOLE. V. P.. J. Clin. Invest.. 36. 150 119561.

RAPPORT, G. M., AND ALO~ZO; N., .J. B>ol. Chem., 217, 193
(1955).

DOLE, V. P., -4s~ MEIXEFLTZ, H., J. Biol. Chem., 236, 2595
(1960).

HIRSCH, J., FARGUH;IR, J. W., AHREN, E. H., PETERSOK, AI.
L., AND STOFFEL, W., Am. J. Clin. Nutrition, 8, 499 (1960).
FILES, F. W., RUSSELL, J. A., .~ND FSIS, J. N., Clin. Chew.,
7, 289 (1961).

C;ERSH, I., AND STILL,&%. -4., J. Erptl. .I~ed.,Sl, 219 (1945).
LYNN, S. W., MACLEOD, R. M., AND BROWN, R. H., J. Bid.
Chem., 236, 1904 (1960).

JEANRENAUD, B., AND RENOLD, A. E., J. Biol. Chem., 234,
3082 (1959).

LEONARDS, J. R., AND L.~NDE~~, B. R., .lrch. Biochem. Bio-
vhzls., 91, 194 (1960).

RI&GEL&AS, P. ik, Metabolism, 11, 1315 (1962).
BALL, E. G., MARTIN, I>. B., ASD COOPER, O., J. Bid. (`hem..
234, 774 (1959).

PITTMAN, J. A., BOSHELL, B. R., WILLIAMS, B. H., HAMMER,
D.. AND HILL. P.. Biochem. and Biovhus. Research Comm uns..
6, i9 (1961). ' '             I "

CSHILL, G. F., JR., LEBOEUF, B., AND FLINN, R. B., J. Bid.
Chem., 236, 1246 (1960).

LEBOEUF, B., AND CAHILL, G. F., JR., J. Bid. Chem., 236, 41
(1961).

WHITE, J. E., .~ND ENGEL, F. L., J. Clin. Invest., 37,1556 (1958).
RBBEN, M. S., AND HOLLENBERG, C. H., J. Chin. Invest., 38,
1032 (1959).

FREINKEL, N., J. Clin. Invest., 40, 476 (1961).
HAGEN, J. H., AXD BALL, E. G., J. Biol. Chem., 236, 1545
(1960).

STEINBERG. D., V~UGH~X. ht.. END M.4RGOLIS. S., J. Bid.

Chem., 236, PC38 (1960):                

C~ORDON!R. 6.. AND CHERKES. A.. Proc. Sot. Ervtl. Biol. Med..

97, 156 (1958').          , ,
WHITE, J. E., AND ENGEL, F. L., Proc. Sot. E&l. Biol. dfed.,
99, 375 (1958).

WADSTROM, L. B., Xalure, 179, 259 (1957).