0013-7227/92/1312-0772$03.00/0 Endocrinology Copyright 0 1992 by The Endocrine
Vol. 131, No. 2 Prrnted in U.S.A.
Society
Growth Hormone Increases Rat Adipocytes: Correlation Carbohydrate Metabolism* YAEL
SCHWARTZ?,
Department
of
HIROSHI
Physiology,
University
YAMAGUCHI, of
Massachusetts
AND
Intracellular with Actions H. MAURICE
Medical
Free Calcium on
in
GOODMAN
School, Worcester, Massachusetts
01605
ABSTRACT Adipocytes that have been preincubated for 3 h or more in hormonefree medium respond to GH with a transient insulin-like increase in glucose metabolism, followed by a period of refractoriness to further insulin-like stimulation. Adipocytes freshly isolated from normal rats and adipocytes that were exposed to GH in the first hour of a 4-h incubation period are refractory to this insulin-like effect. Because earlier studies revealed a relationship between refractoriness and cellular calcium, we examined the effects of GH on the intracellular calcium concentration ([Ca”],) under a variety of conditions in which sensitivity to insulin-like stimulation or refractoriness is known to be affected. A dual nitrogen laser imaging microscope with computerassisted image processing to measure fluorescence changes in individual adipocytes loaded with furaAM was used to measure [Ca’+],. After prolonged incubation in uitro, resting [Ca+“], was 120 & 6 nM and remained unchanged for more than 1 h after the addition of 100 rig/ml human GH (hGH), which doubled the rate of incorporation of [3-3H] glucose into triglycerides in this interval. Lipogenesis declined in the second hour, and [Ca”li slowly increased, reaching 324 + 49 nM by the end of the third hour (P < 0.05). When added with GH, actinomycinD, an inhibitor of RNA synthesis, caused the accelerated rate of
lipogenesis to persist for at least 5 h and blocked the delayed increase in resting [Ca+‘],. Resting [Ca”li in freshly isolated, and hence refractory, adipocytes was 342 + 34 nM and declined to 112 + 11 nM concomitant with acquisition of insulin-like sensitivity to GH. The addition of 100 rig/ml hGH to these cells at the beginning of incubation, under conditions known to sustain refractoriness prevented the decline in resting [Ca’+], and enabled them to exhibit a further acute increase in [Ca”li in response to a second exposure to hGH in the fourth hour. When added 60 min after GH, actinomycin-D blocked the ability to raise [Ca”li acutely in response to GH, but did not interfere with GH’s action to sustain resting [Ca”],, which remained at about 300 nM. The concentration of GH needed to increase [Ca”li acutely in refractory cells is about 6-fold higher than that needed to maintain resting [Ca’“li. Differences in the time of sensitivity to actinomycin-D and dose dependency suggest that sustaining resting [Ca”li and production of the acute increase in [Ca”+], are separate phenomena. Because the acute increase in [Ca’+],, but not resting [Ca”li, correlates with refractoriness, it is suggested that the acute increase in [Ca”li may be responsible for preventing expression of the insulin-like effect in refractory cells. (Endocrinology 131: 772-778, 1992)
H PRODUCES a complex series of effects on carbohydrate metabolism in adipocytes. Fat cells or tissue segments that have been deprived of GH for at least 3 h exhibit a transient insulin-like response that includes accelerated uptake and utilization of glucose (l-3). The insulinlike response begins to subside after about an hour despite continued stimulation with GH, and by about 3 h glucose metabolism is restored to its basal rate (4). At this time the cells are refractory to GH, and even very high concentrations cannot initiate a second insulin-like response. Termination of the insulin-like response and induction of refractoriness are separate, though related, events that depend upon temporally separate periods of RNA synthesis (5,6). When added simultaneously with GH, actinomycin-D blocks the synthesis of RNA and proteins needed for termination of the insulinlike response, which consequently persists for many hours (4). When treatment with actinomycin is delayed for 1 h after GH treatment, the insulin-like response terminates as usual, but the adipocytes remain responsive to insulin-like stimu-
lation by GH (6). Refractoriness does not change the number or affinity of GH receptors (7) and does not interfere with their mediation of responses other than the insulin-like effect. Freshly isolated adipocytes are refractory to the insulin-like actions of GH (2, 3), presumably as a result of recent exposure to GH in viva, but not to its lipolytic actions (5, 8). Although treatment of these cells with GH during the first hour of incubation does not elicit an insulin-like response, it prolongs refractoriness and blocks the emergence of sensitivity to insulin-like stimulation that would otherwise become evident after about 3 h (1, 3). Finally, adipocytes that are refractory to insulin-like stimulation promptly respond to GH with an increase in their intracellular calcium concentrations ([Ca”],) (9). Earlier studies from this laboratory indicated that refractoriness or responsiveness to insulin-like stimulation by GH is closely related to [Ca’+], (3). Blockade of calcium channels with verapamil, inhibition of calmodulin with calmidazolium or trifluoropirazine, or depletion of cellular calcium by incubation in calcium-free medium restored insulin-like sensitivity in refractory cells. Conversely, sensitive cells became refractory when incubated with the calcium ionophore A23187 in the presence of 2.5 mM extracellular calcium (3). Direct measurement of [Ca2+li in individual fat cells using
G
Received February 24, 1992. Address requests for reprints to: Dr. Maurice Goodman, Department of Physiology, University of Massachusetts School of Medicine, 55 Lake Avenue North, Worcester, Massachusetts 01605. * This work was supported by Grant DK-19392 from the NIH. t Supported by Grant DK-070302 from the NIH.
772
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES
[Ca”]
the calcium-sensitive fluorescent dye furarevealed that resting levels of [Ca2+li were about twice as high in refractory cells as in sensitive cells. Furthermore, [Ca2+]i promptly increased even further when GH was added to refractory cells, but remained unchanged when GH was added to sensitive cells (9). The present study extends these observations on the actions of GH and examines the correlation among changes in [Ca2+],, actinomycin sensitivity, and refractoriness.
IN ADIPOCYTES
the eyepiece and resolved into 512 X 240 pixels with an image processor (Recognition Technology, Westborough, MA). Data processing-was performed using a NEC Power Mate I comuuter. ICa2+l, was estimated bv comparison “with a standard calibration curve constructed from the fluorescence ratios measured in the cytosol of refractory and sensitive adipocytes that were made permeable to calcium by treatment with 1 P’M ionomycin and equilibrated with extracellular calcium ranging from 10-4-10-8 M, as described by Williams et nl. (12).
Determination Materials Preparation
and Methods
of adipocytes
Adipocytes were prepared from the epididymal fat of 160. to 180-g rats of the CD strain (Charles River Breeding Laboratories, Kingston, NY) according to a modified (3) method of Rodbell (10). The rats were maintained in the vivarium at constant temperature (18 C) and lighting (lights on between 0600-1800 h) for at least 1 week before study and were fed a standard laboratory diet (Purina 5008, Ralston-Purina, St. Louis, MO) and water ad libitum. All procedures were in accordance with protocols approved by the Institutional Animal Care and Use Committee. For short term incubations (l-4 h), adipocytes were prepared from the pooled epididymal fat of two to eight rats. The minced tissues were digested for 20 min with with 1 mg/ml collagenase (lot 143710, type A, Boehringer Mannheim Biochemicals, Indianapolis, IN). The cells were then collected by centrifugation and washed four times with KrebsRinger bicarbonate buffer with 0.1 rnM glucose (KRBG) containing 1% (wt/vvol) BSA (Metrix, fraction IV, Rehe; Chemical Cd., Phoenix,-AZ). The cells were refractory to GH at this time. To render them sensitive to insulin-like stimulation, they were incubated without GH for at least 3 h. To preserve the original refractoriness of freshly isolated cells, 100 rig/ml recombinant methionyl human GH (hGH; Genentech Corp., San Francisco, CA) were added for the first hour of incubation. The cells were then washed and incubated for 2 h without hormone, as previously described (3). Examination of the time course of responses to GH requires prolonged incubation in vitro to permit development of insulin-like sensitivity (at least 3 h) before the 4-6 h needed for expression of the effects of GH. These studies were most conveniently periormed by allowing sensitivity to develou during an overnight incubation and adding GH the following morning. For the overnightlncubation, each milliliter of packed adipocytes (-2 X lo7 cells) was suspended in 50 ml Dulbecco’s Minimal Essential Medium-l% BSA, which contained 10 rnM HEPES buffer, 5 pg/ml ascorbic acid, 1% fetal calf serum, penicillin (20 U/ml,) and streptomycin (20 mg/ml; Gibco, Grand Island, NY) and placed in a 250ml sterile culture flask. The flasks were incubated in a tissue culture hood at 37 C in an atmosphere of 95% air and 5% COz. The following morning, the cells were washed three times and suspended in KRBG1% BSA. I
Measurement
of [Ca2+lL
[Ca’*], was determined in individual adipocytes using the hexakis(acetoxymethy1) ester (AM) of the calcium-sensitive fluorescent dye fura(11) (Molecular Probes, Eugene OR), as previously described (9). After incubation with 15.0 ELM furafor 20 min, a 20.~1 aliquot of a 1:lO suspension of dye-loaded cells was pipetted into a Plexiglass perifusion chamber mounted on the stage of a Leitz compound microscope (Leitz, Rockleigh, NJ). The buoyant cells were held in place by adhering to a Cell-Tak-coated coverslip (Collaborative Research, Bedford, MA) and were perifused with KRBG at a flow rate of 1 ml/min. The buffer inlet of the perifusion chamber was equipped with a sidearm, which permitted addition of hormone to the perifusate. The adipocytes were sequentially illuminated from below with 3-nsec pulses of light (337 or 380 nm) delivered every 33 msec through a bifurcated quartz fiber extending from a nitrogen laser and a tunable dye laser (Laser Science, Cambridge, MA). Images were recorded with a GVC CCD video camera (GBC Camera, New York, NY) mounted over
173
of lipogenesis
Incorporation of o-[3-3H]glucose (New England Nuclear, Boston, MA) into lioid served as an index of insulin-like responsiveness to GH and was assessed bv a modification (13) of the method of Moody et al. (14). Briefly, 500 ~1 of a suspension (1:30, vol/vol) of adipocytes in KRB buffer that contained 0.1 rnM o-13-3Hlalucose (SA, 0.8 uCi/uM; New England Nuclear Corp., Boston, MA) an’dvvarious‘concentrations of hGH were incubated in 6-ml polyethylene minivials for 1 h at 37 C under an atmosphere of 95% 02-5% CO>. Triplicate aliquots of cells were studied for each experimental condition The incuba’tion was terminated by the addition oi5 ml/vial toluene scintillator fluid (0.3 e 1.4-bis12(4-methvl-5-uhenvloxazolvl)l benzene and 5.0 g 2,5-diphenyloxazole/liter toluene), followed by vigorous mixing by vortex for 30 set to disrupt the fat cells. The vials were permitted to sit for 1 h to allow extraction of the lipids into the toluene phase before counting in a Packard Tri-Carb 4530 o-counter (Packard, Downers Grove, IL). ~
Statistical
”
L
~
,
I
i
,
Ia
analyses
For statistical evaluation, each experiment consisting of a separate population of adipocytes pooled from 2-8 rats was considered to be a single observation; each experiment was repeated at least 6 times. For analyses of [Ca’+],, 15-30 adipocytes were scanned for each experimental condition in each cell population. Data were analyzed by analysis of variance for repeated measures (15, 16). Statistical significance was assessed by multiple pairwise t tests, using the Bonferonni adjustment to correct for additive type I errors due to multiple comparisons (17).
Results Time course of the response to hGH Adipocytes that had been incubated overnight (-15 h) in the absence of hormones were used to study the time course of responses to GH. The cells were collected by centrifugation, resuspended in KRBG at a concentration of 1:30, and transferred to two flasks for incubation without or with 100 rig/ml hGH. Measurements of incorporation of glucose into lipid and [Ca*+]i were made in parallel on aliquots of cells from the same population. For measurement of lipogenesis, aliquots were taken immediately after the addition of GH and every hour thereafter for 4 h. Triplicate aliquots were incubated for 60 min in the presence of [3-H3]glucose. Aliquots of cells were also taken after 1, 2, 3, and 4 h and incubated with furaAM for 20 min to load them with the dye. The data are shown in Fig. 1 using bar graphs (Fig. 1A) to designate lipogenesis measured over the entire hour and points (Fig. 1B) to indicate [Ca2+]i measured for a few seconds in each cell during the latter half of the first, second, third, or fourth hour. GH nearly doubled the incorporation of 3H into lipids in the first hour. This effect declined during the second hour and disappeared by the third hour. In cells that were incubated without GH, [Ca’+]i remained constant at about 100
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES [Ca’+l IN ADIPOCYTES
774 7000-
Endo. Vol131.
1992 No 2
A
n q
**
v) ii? k! !i!
BASAL GH
*
v) G
6000
Y $F
5000
2% 3a
*
*
4000 E
1
2
HOURS
3
AFTER
4
5
$2 l-0” zz-
3000
5
1000
2000
hGH
0 -
BASAL
-----a----~
GH
ACTINOMYCIN
n q
BASAL GH
350
l&T m
HOUk
’
&TH HOGR m +
+
6
* * *
300
L---
4
0
iiOURS
:FTEFl
h:H
FIG. 1. Time course of the response to hGH. Adipocytes were pooled from the epididymal fat of 8 rats/experiment and were incubated overnight in Dulbecco’s Modified Eagle’s Medium-lo% fetal calf serum. The cells were washed 3 times and resuspended in fresh KRBG-1% BSA for incubation with or without 100 rig/ml hGH. Aliquots of cells were removed at the times indicated for measurement (in triplicate) of lipogenesis (A) and [Ca”], (B). Each bar in A and point in B represent the mean for 9 independent experiments. [Ca”], was measured in 1725 cells/experimental point. Brackets indicate the SEM. *, P < 0.05; **, P < 0.01.
nM during the entire incubation period. In GH-treated cells, [Cazfli remained at low resting levels throughout the first hour and increasedsomewhat in the latter half of the second hour coincident with the decline in lipogenesis. [Ca*+]i was significantly increased in the third and fourth hours, when the insulin-like responsewas no longer evident. Effects of actinomycin-D
To examine the dependency of these events on RNA and protein synthesis, adipocytes were collected after overnight incubation and treated as described above, except that half of the cells were incubated in the presence of actinomycinD. Lipogenesisand [Ca’+]i were determined during the first and fifth hours after hGH treatment. In the first hour, hGH once again nearly doubled the rate of incorporation of 3H into lipids and produced no change in [Ca’+]i regardless of whether actinomycin-D was present (Fig. 2). In the fifth hour, the incorporation of 3H into lipids had returned to
50 0 1ST HOUR m
+
5TH HOUR m
+
FIG. 2. Effects of actinomycin-D on the responses of adipocytes to hGH. Adipocytes prepared from the epididymal fat of 8 rats/experiment were treated as described in Fig. 1. Lipogenesis (A) and [Ca’+li (B) were measured in the first and fifth hours after the addition of 100 ng/ ml hGH and/or 5 fig/ml actinomycin-D. Each bar represents the results of 9 experiments, and brackets indicate the SEM. [Ca”li was measured in 18-25 cells/experiment. *, P < 0.01; **, P < 0.05; t**, P < 0.001.
control levels in cells that were incubated with GH alone, but remained significantly elevated in cells that were incubated with GH plus actinomycin-D. In agreement with the findings illustrated in Fig. lB, [Ca’+]i was nearly 3 times higher in cells treated with hGH alone. In the cells that were incubated with hGH plus actinomycin-D, [Ca”]i remained unchanged. These results indicate that both the termination of the insulin-like response and the rise in [Ca’+]i require RNA and protein synthesis. Ku’+],
in freshly isolated cells
In earlier studies (9) refractoriness was associated with high [Ca’+]i, and sensitivity to insulin-like stimulation was associated with low [Ca’+]i. Because these measurements were made after prolonged (4 h or longer) incubation, it was
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES
[Ca’+]
unclear whether refractoriness results from an increase in [Ca2+li or whether sensitivity results from a decrease in [Ca’+],. We, therefore, measured [Ca’+], as early as possible after cell isolation in a effort to determine “normal” [Ca2+],. Isolating adipocytes and loading them with furaAM require nearly 90 min from the time the tissues are excised from the donor rats. [Ca2+], measured at this time was in excess of 300 nM (Fig. 3), similar to concentrations in adipocytes that had been treated with GH 4 h earlier (Figs. 1 and 2). [Ca2+11 remained above about 300 nM during the second hour of incubation in the absence of hormones and declined significantly thereafter, consistent with the time needed for acquisition of sensitivity to insulin-like stimulation (l-3). Although freshly isolated adipocytes were refractory to insulin-like stimulation by GH (data not shown) and had higher resting [Ca’+], than sensitive cells, they, nevertheless, responded acutely to hGH with a gradual increase in [Ca2+]i over the next 6 min (Fig. 4). In contrast, [Ca’+J remained unchanged after the addition of hGH to cells that were 400 -
is C .: H
1
HOURS
2
3
4
POST-ISOLATION
3. Effects of duration of incubation on resting [Ca”]; isolated adipocytes. Adipocytes prepared from the epididymal rats were pooled for each experiment. Each bar represents for 8 experiments (19-27 cells/experiment); brackets indicate
FIG.
*,
P < 0.01;
**,
in freshly fat of 2 the mean the SEM.
P < 0.001.
IN ADIPOCYTES
175
preincubated for 3-4 h in the absence of GH and were, therefore, sensitive to insulin-like stimulation. Figure 4 shows representative 8-min scans of single adipocytes acutely treated with 500 rig/ml hGH immediately after isolation (upper panel) or 4 h later (lower panel). Similar data were obtained from three other cells studied in the first 2 h after excision and from three cells studied in the fourth hour. Effects of GH and actinomycin-D
in refractory
adipocytes
From the results of the experiments illustrated in Figs. 1 and 4 it appears that GH has two effects on [Ca’+]i: a delayed increase in resting [Ca’+]i produced in sensitive cells (see Fig. l), and an acute increase in [Ca2+]i produced only in refractory cells (see Fig. 4). It is possible that the acute and delayed effects of GH on [Ca’+], are manifestations of the same actinomycin-sensitive event, and that the acute effect can only be expressed when [Ca2+]i is above some minimal threshold concentration. Alternatively, sensitization to the acute response to GH may be a separate event from that which maintains resting [Ca2+]i. To distinguish between these possibilities, we delayed the addition of actinomycin-D until 60 min after hGH and determined its effects on [Ca’+]i when GH was added 3 h later (Fig. 5). In adipocytes incubated for 4 h in the absence of hGH, [Ca2+]i declined to about 100 nM regardless of whether actinomycin was present. [Ca*+]i was maintained at approximately 300 nM when 100 rig/ml hGH were added in the first hour and nearly doubled in response to 500 rig/ml hGH added in the fourth hour. Treatment with actinomycin-D 60 min after hGH did not interfere with maintenance of resting [Ca’+]i, but prevented the acute rise in [Ca2+]i in response to the second addition of hGH. These data suggest that different actinomycin-sensitive events may be responsible for maintenance of resting [Ca2+li and sensitization to the acute action of GH to increase [Ca’+]i. In addition, the onset of the acute rise in [Ca’+]i is too rapid to result from induced RNA and protein synthesis, suggesting yet a third phenomenon. The dose dependencies
is 5 T N 9
.-
500 400 300 200 100 n
hGH at tim; hGH at 180 Actinomycin
0 min
-
+ -
+ + -
” -
+ -
+ +
+
+
+
5. Effects of GH and actinomycin-D on [Ca”+], in refractory adipocytes. Adipocytes isolated from the epididymal fat of 2 rats/ experiment were incubated for 4 h in the presence or absence of 100 rig/ml hGH. Actinomycin-D, when present, was added after 1 h. Where indicated, 500 rig/ml hGH were added after 3 h, and [Ca’+], was measured 20-30 min later. Each bar represents the mean of 7 experiments (19-27 cells/experiment). Brackets indicate the SEM. t, P < 0.01; N.S., P > 0.05.
FIG.
SECONDS FIG. 4. Acute changes in [Ca”li in response to 500 rig/ml hGH in individual cells treated as described in Fig. 3. hGH was added in the time designated by the darkened box during the first hour (upperpane2) or third hour postisolation (lower panel).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES [Ca*‘l IN ADIPOCYTES
776
Endo. 1992 Vol131.No2
1
of each of these phenomena are shown in Figs. 6-8. To determine the minimal concentration of hGH needed to maintain resting [Ca’+]i, freshly isolated adipocytes were incubated with various concentrations of hGH for 60 min, washed, and reincubated for an additional 2 h before loading with fura-2. [Ca’+]i was indistinguishable from control values in adipocytes that were preincubated with 1 or 3 rig/ml hGH, but was maintained at about 250 nM in cells that were exposed to 10 rig/ml hGH or more (Fig. 6). The same procedure was followed to determine the concentration dependency of the process that enabled adipocytes to exhibit an acute responseto hGH, except that the cells were tested for their ability to exhibit an acute responseto 500 rig/ml hGH in the fourth hour. The acute increase in [Ca’+]i was seen only in cells that had been preincubated with hGH at concentrations of 10 rig/ml or higher, and 30 rig/ml were required for full expression of the response(Fig. 7). These data are expressedas the increment in [Ca’+]i because they are
I.0
.i
hGH
100
10.00
(nglml)
FIG. 6. Concentration-response relationship for maintaining resting [Ca”]{ in normal adipocytes incubated for 4 h in uitro. Freshly isolated adipocytes from 2 rats/experiment were incubated with the indicated concentrations of hGH for 60 min, washed, and reincubated for an additional 2 h before loading with furaAM. Each point represents the mean of 6 independent experiments (18-23 cells/experiment); brackets indicate the SEM. *, P < 0.05. 400
f 5 .-
600500-
:
2004
"'.".
"""'I .l
""".
"""'I
1
10
hGH
(nglml)
100
"'"m 1000
FIG. 8. Concentration-response relationship for production of the acute increase in [Ca*+h. Adipocytes prepared from 2 rats/experiment were incubated for 60 min with 100 rig/ml hGH, washed 3 times with fresh KRBG-1% BSA, and reincubated for 2 h in the absence of hormone. They were then loaded with furaAM and challenged with the indicated concentrations of hGH. [Ca2+li was measured 20-30 min later. Each point represents the mean of 6 experiments (18-23 cells/ experiment); brackets indicate the SEM. t, P < 0.01; tt, P < 0.001.
superimposed on a changing baseline (seeFig. 6). For determination of the concentration dependency of the acute increasein [Ca2+]i,adipocytes were incubated with 100 rig/ml hGH for the first hour, washed, and reincubated for 2 h in the absence of hormone. The cells were then loaded with fura- and incubated for 20 min with various concentrations of hGH before measurement of [Ca’+]i. The lowest concentration of hGH needed to produce even a small increase in [Ca’+]i was 25 rig/ml, and the effect was maximal at 125 ng/ ml (Fig. 8). The EDs0for maintenance of [Ca”]i (-5 rig/ml, asestimated by inspection of the curve) and that for enabling cells to exhibit and acute responseto hGH (-7 rig/ml) are, thus, quite similar, while 6- to g-fold higher concentrations of hGH were needed to elicit the acute increase in [Ca2+]i (E&or -40 rig/ml). Discussion
1
hGH
(ngiml)
FIG. 7. Concentration-response relationship for sensitizing adipocytes to the acute increase in [Ca*+]; produced by subsequent treatment with hGH. Adipocytes prepared from 2 rats/experiment were preincubated with the indicated concentrations of hGH for 60 min, washed 3 times with fresh KRBG-1% BSA, and reincubated for 2 h in the absence of hormone. They were then loaded with furaAM and challenged with 500 rig/ml hGH. Data are expressed as increments (A) in [Ca”li. Each point represents the mean of 6 independent experiments (18-25 cells/ observation); brackets indicate the SEM. t, P < 0.01; tt, P < 0.001.
Adipocytes appear to require frequent stimulation with GH to maintain their resting [Ca2+]iin the range of 200-300 nM. When deprived of GH for more than 3 h, [Ca2+],falls to a steady state concentration of about 100 nM, even though maintaining an even steeper transmembrane calcium gradient is likely to be more costly for cellular energy. The frequency of stimulation with GH required to maintain [Ca’+]i at the higher resting level appears to be about every 3-4 h and, thus, is similar to the normal frequency of pulsatile secretion of GH in rats (18). Maintenance of [Ca’+]i by GH requires ongoing synthesis of RNA and proteins that may be ion channels, ion pumps, or regulators of such membrane components. Because there are multiple Ca2+ exchangers, pumps, and channels in adipocytes (19-21), and because [Ca2+]iis also determined in part by [Na+]i and [H+]i as well as the concentration of other ions, each of which is regulated by more than one process,the range of possibleGH-dependent proteins is large. In addition to maintaining resting lCa2+],,GH produces an
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES
[Ca”‘]
acute increase in [Ca’+]i that begins within a few seconds and persistsfor at least 20 min (9) and for as long as 1 h in some cells. This acute increase in [Ca”]i is the most rapid effect of GH that has been reported. We know little about its mechanism, except that the profile of change in [Ca*+]i produced by GH differs from that produced by oxytocin (9, 22, 23), which is thought to act by way of an inositol trisphosphate second messengersystem (23). The acute increasein [Ca2+liproduced by GH has thus far only been seen when resting [Ca2+li was 200 nM or higher, but merely maintaining [Ca’+]i at this level is not sufficient to permit expression of the acute response. Some additional GH-dependent factor(s) seemsto be required, as suggestedby the experiment shown in Fig. 5, in which delayed addition of actinomycin blocked expression of a subsequent acute responseeven though resting [Ca’+]i was above 250 nM. The concentration of GH needed to increase[Ca’+]i acutely is considerably higher than that needed to maintain resting [Ca’+]i and is similar to concentrations achieved during physiological pulsesof secretion (18). It might be expected, therefore, that in vim adipocyte [Ca*+]i may oscillate between about 200 nM and about 600 nM. However, at least at high concentrations (500 jLJ/ml), insulin blocked the acute effect of GH on [Ca2+li(9). We do not yet know whether physiological concentrations of insulin are sufficient to block this responseto GH. Unfortunately, the rather long time (-1.5 h) needed to prepare cells for measurement of [Ca’+]i precludes obtaining a meaningful profile of adipocyte [Ca*+]i in vim using current methods. In previous studies (9) we found that refractoriness to the insulin-like effects of GH is associatedwith a relative increase in [Ca2+]i,but could not determine whether the change in resting [Ca*+]i, the acute increase in [Ca*+]i, or both were critical determinants of refractoriness. The present results provide additional insight into the relationship between [Ca*+]iand expressionof the insulin-like response.The effect of GH on resting [Ca*+]i has a dose-dependency that correlates closely with the dose-dependency for induction of refractoriness in both normal adipocytes (7) and segmentsof adipose tissue from hypophysectomized rats (24), while the acute effect has a dose-dependency resembling that of the insulin-like effect (7). The effect on resting [Ca*+]i resembles the effect on the termination of the insulin-like responsein both its time course and the brevity of its sensitivity to blockade by actinomycin-D (5). In earlier experiments, the addition of actinomycin-D 60 min after GH did not prevent the insulin-like responsefrom terminating, but blocked the onset of refractoriness (5). We now find that resting [Ca*+]i is maintained under these circumstances,but the acute rise in [Ca*+]i in responseto subsequent stimulation with GH is blocked. Becauseadipocytes with the higher resting [Ca”]i are refractory to insulin-like stimulation only when the acute increasein [Ca2+liis present, it is likely that the acute rise in [Ca2+liis the immediate causeof refractoriness, but we cannot rule out a contribution from the resting [Ca”]i. The mechanismsby which GH increases[Ca*+]i acutely, and by which elevated [Ca’+]i blocks expression of the insulin-like response are unknown. We also do not yet know
IN ADIPOCYTES
777
whether the effects of GH on calcium in adipocytes are representative of effects of GH in other target tissues and whether increased [Ca2+]iplays a role in the hyperinsulinemia, decreased insulin sensitivity, and hypertension that frequently accompany excessive secretion of GH. The demonstrated relationship between GH and [Ca*+]i may provide new insights and new experimental approaches to these questions. References 1. Goodman HM, Coiro V 1981 Induction of sensitivity to the insulinlike action of growth hormone in normal rat adipose tissue. Endocrinology 108:113-119 2. Eden S, Schwartz J, Kostyo JL 1982 Effects of preincubation on the ability of rat adipocytes to bind and respond to growth hormone. Endocrinology 111:1505-1512 3. Schwartz Y, Goodman HM 1990 Refractoriness to the insulin-like effects of growth hormone depends upon calcium. Endocrinology 127:170-176 4. Goodman HM 1968 Growth hormone and the metabolism of carbohydrate and lipid in adipose tissue. Ann NY Acad Sci 148:419440 5. Goodman HM, Tai LR, Chipkin SR 1990 The isoquinoline sulfonamide inhibitors of protein phosphorylation, H-7, H-8, and HA1004, also inhibit RNA synthesis: studies on responses of adipose tissue to growth hormone. Endocrinology 126:441-450 6 Goodman HM 1981 Separation of early and late responses of adipose tissue to growth hormone. Endocrinology 109:120-129 7. Grichting G, Levy LK, Goodman HM 1983 Relationship between binding and biological effects of human growth hormone in rat adipocytes. Endocrinology 113:1111-1120 8. Fain TN. Kovacev VP. Scow RO 1965 Effect of growth hormone and dexamethasone on lipolysis and metabolism ilisolated fat cells of the rat. J Biol Chem 240:3522-3529 9. Schwartz Y, Goodman HM, Yamaguchi H 1991 Refractoriness to growth hormone is associated with increased intracellular calcium in rat adipocytes. Proc Nat1 Acad Sci USA 88:6790-6794 10. Rodbell M 1964 Metabolism of isolated fat cells. I. Effects of hormones on glucose metabolism and lipolysis. J Biol Chem 239:375-380 11. Grynkiewicz G, Poenie M, Tsien RY 1985 A new generation of Ca’+ indicators with greatly improved fluorescence properties. J Biol Chem 260:3440-3450 12. Williams DA, Fogarty KE, Tsien RY, Fay FS 1985 Calcium gradients in single smooth muscle cells revealed by the digital imaging microscope using fura-2. Nature (Lond) 318:558-561 13. Smal J, Closset J, Hennen G, De Meyts P 1987 Receptor binding properties and insulin-like effects of human growth hormone and its 20-kDa variant in rat adipocytes. J Biol Chem 262:11071-11079 14. Moody AJ, Stan MA, Stan M 1974 A simple free fat cell bioassay for insulin. Horm Metab Res 6:12-16 15. Winer Bl 1962 Statistical Principles in Experimental Design, ed 2. McGrawLHill, New York _ 16. Dixon WJ, Brown MB, Engleman L, Frane JW, Hill MA, Jennrich RI, Toporek JD 1985 BMPD Statistical Software. University of California Press, Berkeley 17. Snedecor GW. Cochran WG 1980 Statistical Methods, ed 7. Iowa State University Press, Ames 18. Tannenbaum GS, Martin JB 1976 Evidence for an endogenous u&radian rhythm governing growth hormone secretion in the rat. Endocrinology 98:562-570 19. Pershadsingh HA, Lee L-Y, Snowdowne KW 1989 Evidence for a sodium/calcium exchanger and voltage-dependent calcium channels in adipocytes. FEBS Lett 244:89-92 20. Pershadsingh HA, Landt M, McDonald JM 1980 Calmodulinsensitive ATE’-dependent Ca*+ transport across adipocyte plasma membranes. 1 Biol Chem 255:8983-8986 21. Black BL, Jarrett L, McDonald JM 1981 The regulation of endo-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
778
GH INCREASES
[Ca”]
plasmic reticulum calcium uptake of adipocytes by cytosolic calcium. J Biol Chem 256:322-329 22. Kelley KL, Deeney JT, Corkey BE 1989 Cytosolic free calcium in rat adipocytes. J Biol Chem 264:12754-12757 23. Uto A, Arai H, Ogawa Y 1991 Reassessment of fura- and the ratio
IN ADIPOCYTES
Endo. Voll31.
method for determination of intracellular Ca” concentrations. Cell Calcium 12:29-37 24. Goodman HM 1984 Biological activity of bacterial derived human growth hormone in adipose tissue of hypophysectomized rats. Endocrinology 114:131-135
Recertification in Endocrinology, Diabetes, and Metabolism Endocrinology, Diabetes, and Metabolism Board of the American Board of Internal Medicine Drs. Martin 1. Surks (Chairman), Glenn D. Braunstein, John D. Brunzell, A. Kreisberg, D. Lynn Loriaux, and Lawrence G. Raisz
Philip E. Cryer, Willa A. Hsueh, Robert
Since 1990, the duration of validity of certification by the American Board of Internal Medicine (ABIM) has been limited to 10 yr. This policy was adopted because medical information changes rapidly and the public needs assurance that certified internists have maintained their skills and kept their knowledge up to date. Time-limited certification and recertification are obligations of an accountable profession. The ABIM
recently has completed
and adopted
1992 No 2
plans for a comprehensive
recertification
program.
Entry into Recertification: Diplomates may attempt recertification only in disciplines in which they were previously certified. Certified endocrinologists may seek recertification at any time after initial certification in endocrinology, diabetes, and metabolism only, in internal medicine only, or in both. Diplomates can allow a time-limited certificate in internal medicine to expire without jeopardizing eligibility for recertification in endocrinology, diabetes, and metabolism. However, expiration of a certificate in internal medicine or endocrinology means the ABIM no longer recognizes an individual as certified in that discipline. Certificates in internal medicine or endocrinology, diabetes, and metabolism issued before 1990 are not time-limited and are valid for life; individuals holding such certificates are eligible to seek recertification without placing existing certificates at risk. Please note that all successful candidates for recertification will be issued a certificate with diabetes in the title: Endocrinology, Diabetes, and Metabolism. The Recertification Process: The recertification process consists of three steps: documentation of clinical competence, completion of the self-evaluation process, and success on a proctored, written final examination. Dual Recertification: The Board anticipates that most endocrinologists will seek recertification in internal medicine as well as endocrinology, diabetes, and metabolism and has, therefore, developed an efficient process for dual recertification that does not change the standards required for recertification in each discipline. For both the self-evaluation processes and the final examinations, substitution of required modules for self-selected modules can reduce the total number of modules required for recertification in internal medicine and endocrinology to six self-evaluation process modules (three in general internal medicine, three in endocrinology) and four final examination modules (two in general internal medicine, two in endocrinology). The score for each final examination will be determined independently. Thus, an individual can be successful in becoming recertified in endocrinology, diabetes, and metabolism but not in internal medicine, even when the processes are undertaken concurrently. Schedule of Availability: This comprehensive recertification program will become available in 1995. The self-evaluation process will be available continuously beginning in 1995. Final examinations will be available annually beginning in 1996. Diplomates who would like to become recertified before this time can take a regularly scheduled certification examination for recertification credit (interim voluntary recertification).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
Vol. 131, No. 2 Prrnted in U.S.A.
Society
Growth Hormone Increases Rat Adipocytes: Correlation Carbohydrate Metabolism* YAEL
SCHWARTZ?,
Department
of
HIROSHI
Physiology,
University
YAMAGUCHI, of
Massachusetts
AND
Intracellular with Actions H. MAURICE
Medical
Free Calcium on
in
GOODMAN
School, Worcester, Massachusetts
01605
ABSTRACT Adipocytes that have been preincubated for 3 h or more in hormonefree medium respond to GH with a transient insulin-like increase in glucose metabolism, followed by a period of refractoriness to further insulin-like stimulation. Adipocytes freshly isolated from normal rats and adipocytes that were exposed to GH in the first hour of a 4-h incubation period are refractory to this insulin-like effect. Because earlier studies revealed a relationship between refractoriness and cellular calcium, we examined the effects of GH on the intracellular calcium concentration ([Ca”],) under a variety of conditions in which sensitivity to insulin-like stimulation or refractoriness is known to be affected. A dual nitrogen laser imaging microscope with computerassisted image processing to measure fluorescence changes in individual adipocytes loaded with furaAM was used to measure [Ca’+],. After prolonged incubation in uitro, resting [Ca+“], was 120 & 6 nM and remained unchanged for more than 1 h after the addition of 100 rig/ml human GH (hGH), which doubled the rate of incorporation of [3-3H] glucose into triglycerides in this interval. Lipogenesis declined in the second hour, and [Ca”li slowly increased, reaching 324 + 49 nM by the end of the third hour (P < 0.05). When added with GH, actinomycinD, an inhibitor of RNA synthesis, caused the accelerated rate of
lipogenesis to persist for at least 5 h and blocked the delayed increase in resting [Ca+‘],. Resting [Ca”li in freshly isolated, and hence refractory, adipocytes was 342 + 34 nM and declined to 112 + 11 nM concomitant with acquisition of insulin-like sensitivity to GH. The addition of 100 rig/ml hGH to these cells at the beginning of incubation, under conditions known to sustain refractoriness prevented the decline in resting [Ca’+], and enabled them to exhibit a further acute increase in [Ca”li in response to a second exposure to hGH in the fourth hour. When added 60 min after GH, actinomycin-D blocked the ability to raise [Ca”li acutely in response to GH, but did not interfere with GH’s action to sustain resting [Ca”],, which remained at about 300 nM. The concentration of GH needed to increase [Ca”li acutely in refractory cells is about 6-fold higher than that needed to maintain resting [Ca’“li. Differences in the time of sensitivity to actinomycin-D and dose dependency suggest that sustaining resting [Ca”li and production of the acute increase in [Ca”+], are separate phenomena. Because the acute increase in [Ca’+],, but not resting [Ca”li, correlates with refractoriness, it is suggested that the acute increase in [Ca”li may be responsible for preventing expression of the insulin-like effect in refractory cells. (Endocrinology 131: 772-778, 1992)
H PRODUCES a complex series of effects on carbohydrate metabolism in adipocytes. Fat cells or tissue segments that have been deprived of GH for at least 3 h exhibit a transient insulin-like response that includes accelerated uptake and utilization of glucose (l-3). The insulinlike response begins to subside after about an hour despite continued stimulation with GH, and by about 3 h glucose metabolism is restored to its basal rate (4). At this time the cells are refractory to GH, and even very high concentrations cannot initiate a second insulin-like response. Termination of the insulin-like response and induction of refractoriness are separate, though related, events that depend upon temporally separate periods of RNA synthesis (5,6). When added simultaneously with GH, actinomycin-D blocks the synthesis of RNA and proteins needed for termination of the insulinlike response, which consequently persists for many hours (4). When treatment with actinomycin is delayed for 1 h after GH treatment, the insulin-like response terminates as usual, but the adipocytes remain responsive to insulin-like stimu-
lation by GH (6). Refractoriness does not change the number or affinity of GH receptors (7) and does not interfere with their mediation of responses other than the insulin-like effect. Freshly isolated adipocytes are refractory to the insulin-like actions of GH (2, 3), presumably as a result of recent exposure to GH in viva, but not to its lipolytic actions (5, 8). Although treatment of these cells with GH during the first hour of incubation does not elicit an insulin-like response, it prolongs refractoriness and blocks the emergence of sensitivity to insulin-like stimulation that would otherwise become evident after about 3 h (1, 3). Finally, adipocytes that are refractory to insulin-like stimulation promptly respond to GH with an increase in their intracellular calcium concentrations ([Ca”],) (9). Earlier studies from this laboratory indicated that refractoriness or responsiveness to insulin-like stimulation by GH is closely related to [Ca’+], (3). Blockade of calcium channels with verapamil, inhibition of calmodulin with calmidazolium or trifluoropirazine, or depletion of cellular calcium by incubation in calcium-free medium restored insulin-like sensitivity in refractory cells. Conversely, sensitive cells became refractory when incubated with the calcium ionophore A23187 in the presence of 2.5 mM extracellular calcium (3). Direct measurement of [Ca2+li in individual fat cells using
G
Received February 24, 1992. Address requests for reprints to: Dr. Maurice Goodman, Department of Physiology, University of Massachusetts School of Medicine, 55 Lake Avenue North, Worcester, Massachusetts 01605. * This work was supported by Grant DK-19392 from the NIH. t Supported by Grant DK-070302 from the NIH.
772
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES
[Ca”]
the calcium-sensitive fluorescent dye furarevealed that resting levels of [Ca2+li were about twice as high in refractory cells as in sensitive cells. Furthermore, [Ca2+]i promptly increased even further when GH was added to refractory cells, but remained unchanged when GH was added to sensitive cells (9). The present study extends these observations on the actions of GH and examines the correlation among changes in [Ca2+],, actinomycin sensitivity, and refractoriness.
IN ADIPOCYTES
the eyepiece and resolved into 512 X 240 pixels with an image processor (Recognition Technology, Westborough, MA). Data processing-was performed using a NEC Power Mate I comuuter. ICa2+l, was estimated bv comparison “with a standard calibration curve constructed from the fluorescence ratios measured in the cytosol of refractory and sensitive adipocytes that were made permeable to calcium by treatment with 1 P’M ionomycin and equilibrated with extracellular calcium ranging from 10-4-10-8 M, as described by Williams et nl. (12).
Determination Materials Preparation
and Methods
of adipocytes
Adipocytes were prepared from the epididymal fat of 160. to 180-g rats of the CD strain (Charles River Breeding Laboratories, Kingston, NY) according to a modified (3) method of Rodbell (10). The rats were maintained in the vivarium at constant temperature (18 C) and lighting (lights on between 0600-1800 h) for at least 1 week before study and were fed a standard laboratory diet (Purina 5008, Ralston-Purina, St. Louis, MO) and water ad libitum. All procedures were in accordance with protocols approved by the Institutional Animal Care and Use Committee. For short term incubations (l-4 h), adipocytes were prepared from the pooled epididymal fat of two to eight rats. The minced tissues were digested for 20 min with with 1 mg/ml collagenase (lot 143710, type A, Boehringer Mannheim Biochemicals, Indianapolis, IN). The cells were then collected by centrifugation and washed four times with KrebsRinger bicarbonate buffer with 0.1 rnM glucose (KRBG) containing 1% (wt/vvol) BSA (Metrix, fraction IV, Rehe; Chemical Cd., Phoenix,-AZ). The cells were refractory to GH at this time. To render them sensitive to insulin-like stimulation, they were incubated without GH for at least 3 h. To preserve the original refractoriness of freshly isolated cells, 100 rig/ml recombinant methionyl human GH (hGH; Genentech Corp., San Francisco, CA) were added for the first hour of incubation. The cells were then washed and incubated for 2 h without hormone, as previously described (3). Examination of the time course of responses to GH requires prolonged incubation in vitro to permit development of insulin-like sensitivity (at least 3 h) before the 4-6 h needed for expression of the effects of GH. These studies were most conveniently periormed by allowing sensitivity to develou during an overnight incubation and adding GH the following morning. For the overnightlncubation, each milliliter of packed adipocytes (-2 X lo7 cells) was suspended in 50 ml Dulbecco’s Minimal Essential Medium-l% BSA, which contained 10 rnM HEPES buffer, 5 pg/ml ascorbic acid, 1% fetal calf serum, penicillin (20 U/ml,) and streptomycin (20 mg/ml; Gibco, Grand Island, NY) and placed in a 250ml sterile culture flask. The flasks were incubated in a tissue culture hood at 37 C in an atmosphere of 95% air and 5% COz. The following morning, the cells were washed three times and suspended in KRBG1% BSA. I
Measurement
of [Ca2+lL
[Ca’*], was determined in individual adipocytes using the hexakis(acetoxymethy1) ester (AM) of the calcium-sensitive fluorescent dye fura(11) (Molecular Probes, Eugene OR), as previously described (9). After incubation with 15.0 ELM furafor 20 min, a 20.~1 aliquot of a 1:lO suspension of dye-loaded cells was pipetted into a Plexiglass perifusion chamber mounted on the stage of a Leitz compound microscope (Leitz, Rockleigh, NJ). The buoyant cells were held in place by adhering to a Cell-Tak-coated coverslip (Collaborative Research, Bedford, MA) and were perifused with KRBG at a flow rate of 1 ml/min. The buffer inlet of the perifusion chamber was equipped with a sidearm, which permitted addition of hormone to the perifusate. The adipocytes were sequentially illuminated from below with 3-nsec pulses of light (337 or 380 nm) delivered every 33 msec through a bifurcated quartz fiber extending from a nitrogen laser and a tunable dye laser (Laser Science, Cambridge, MA). Images were recorded with a GVC CCD video camera (GBC Camera, New York, NY) mounted over
173
of lipogenesis
Incorporation of o-[3-3H]glucose (New England Nuclear, Boston, MA) into lioid served as an index of insulin-like responsiveness to GH and was assessed bv a modification (13) of the method of Moody et al. (14). Briefly, 500 ~1 of a suspension (1:30, vol/vol) of adipocytes in KRB buffer that contained 0.1 rnM o-13-3Hlalucose (SA, 0.8 uCi/uM; New England Nuclear Corp., Boston, MA) an’dvvarious‘concentrations of hGH were incubated in 6-ml polyethylene minivials for 1 h at 37 C under an atmosphere of 95% 02-5% CO>. Triplicate aliquots of cells were studied for each experimental condition The incuba’tion was terminated by the addition oi5 ml/vial toluene scintillator fluid (0.3 e 1.4-bis12(4-methvl-5-uhenvloxazolvl)l benzene and 5.0 g 2,5-diphenyloxazole/liter toluene), followed by vigorous mixing by vortex for 30 set to disrupt the fat cells. The vials were permitted to sit for 1 h to allow extraction of the lipids into the toluene phase before counting in a Packard Tri-Carb 4530 o-counter (Packard, Downers Grove, IL). ~
Statistical
”
L
~
,
I
i
,
Ia
analyses
For statistical evaluation, each experiment consisting of a separate population of adipocytes pooled from 2-8 rats was considered to be a single observation; each experiment was repeated at least 6 times. For analyses of [Ca’+],, 15-30 adipocytes were scanned for each experimental condition in each cell population. Data were analyzed by analysis of variance for repeated measures (15, 16). Statistical significance was assessed by multiple pairwise t tests, using the Bonferonni adjustment to correct for additive type I errors due to multiple comparisons (17).
Results Time course of the response to hGH Adipocytes that had been incubated overnight (-15 h) in the absence of hormones were used to study the time course of responses to GH. The cells were collected by centrifugation, resuspended in KRBG at a concentration of 1:30, and transferred to two flasks for incubation without or with 100 rig/ml hGH. Measurements of incorporation of glucose into lipid and [Ca*+]i were made in parallel on aliquots of cells from the same population. For measurement of lipogenesis, aliquots were taken immediately after the addition of GH and every hour thereafter for 4 h. Triplicate aliquots were incubated for 60 min in the presence of [3-H3]glucose. Aliquots of cells were also taken after 1, 2, 3, and 4 h and incubated with furaAM for 20 min to load them with the dye. The data are shown in Fig. 1 using bar graphs (Fig. 1A) to designate lipogenesis measured over the entire hour and points (Fig. 1B) to indicate [Ca2+]i measured for a few seconds in each cell during the latter half of the first, second, third, or fourth hour. GH nearly doubled the incorporation of 3H into lipids in the first hour. This effect declined during the second hour and disappeared by the third hour. In cells that were incubated without GH, [Ca’+]i remained constant at about 100
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES [Ca’+l IN ADIPOCYTES
774 7000-
Endo. Vol131.
1992 No 2
A
n q
**
v) ii? k! !i!
BASAL GH
*
v) G
6000
Y $F
5000
2% 3a
*
*
4000 E
1
2
HOURS
3
AFTER
4
5
$2 l-0” zz-
3000
5
1000
2000
hGH
0 -
BASAL
-----a----~
GH
ACTINOMYCIN
n q
BASAL GH
350
l&T m
HOUk
’
&TH HOGR m +
+
6
* * *
300
L---
4
0
iiOURS
:FTEFl
h:H
FIG. 1. Time course of the response to hGH. Adipocytes were pooled from the epididymal fat of 8 rats/experiment and were incubated overnight in Dulbecco’s Modified Eagle’s Medium-lo% fetal calf serum. The cells were washed 3 times and resuspended in fresh KRBG-1% BSA for incubation with or without 100 rig/ml hGH. Aliquots of cells were removed at the times indicated for measurement (in triplicate) of lipogenesis (A) and [Ca”], (B). Each bar in A and point in B represent the mean for 9 independent experiments. [Ca”], was measured in 1725 cells/experimental point. Brackets indicate the SEM. *, P < 0.05; **, P < 0.01.
nM during the entire incubation period. In GH-treated cells, [Cazfli remained at low resting levels throughout the first hour and increasedsomewhat in the latter half of the second hour coincident with the decline in lipogenesis. [Ca*+]i was significantly increased in the third and fourth hours, when the insulin-like responsewas no longer evident. Effects of actinomycin-D
To examine the dependency of these events on RNA and protein synthesis, adipocytes were collected after overnight incubation and treated as described above, except that half of the cells were incubated in the presence of actinomycinD. Lipogenesisand [Ca’+]i were determined during the first and fifth hours after hGH treatment. In the first hour, hGH once again nearly doubled the rate of incorporation of 3H into lipids and produced no change in [Ca’+]i regardless of whether actinomycin-D was present (Fig. 2). In the fifth hour, the incorporation of 3H into lipids had returned to
50 0 1ST HOUR m
+
5TH HOUR m
+
FIG. 2. Effects of actinomycin-D on the responses of adipocytes to hGH. Adipocytes prepared from the epididymal fat of 8 rats/experiment were treated as described in Fig. 1. Lipogenesis (A) and [Ca’+li (B) were measured in the first and fifth hours after the addition of 100 ng/ ml hGH and/or 5 fig/ml actinomycin-D. Each bar represents the results of 9 experiments, and brackets indicate the SEM. [Ca”li was measured in 18-25 cells/experiment. *, P < 0.01; **, P < 0.05; t**, P < 0.001.
control levels in cells that were incubated with GH alone, but remained significantly elevated in cells that were incubated with GH plus actinomycin-D. In agreement with the findings illustrated in Fig. lB, [Ca’+]i was nearly 3 times higher in cells treated with hGH alone. In the cells that were incubated with hGH plus actinomycin-D, [Ca”]i remained unchanged. These results indicate that both the termination of the insulin-like response and the rise in [Ca’+]i require RNA and protein synthesis. Ku’+],
in freshly isolated cells
In earlier studies (9) refractoriness was associated with high [Ca’+]i, and sensitivity to insulin-like stimulation was associated with low [Ca’+]i. Because these measurements were made after prolonged (4 h or longer) incubation, it was
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES
[Ca’+]
unclear whether refractoriness results from an increase in [Ca2+li or whether sensitivity results from a decrease in [Ca’+],. We, therefore, measured [Ca’+], as early as possible after cell isolation in a effort to determine “normal” [Ca2+],. Isolating adipocytes and loading them with furaAM require nearly 90 min from the time the tissues are excised from the donor rats. [Ca2+], measured at this time was in excess of 300 nM (Fig. 3), similar to concentrations in adipocytes that had been treated with GH 4 h earlier (Figs. 1 and 2). [Ca2+11 remained above about 300 nM during the second hour of incubation in the absence of hormones and declined significantly thereafter, consistent with the time needed for acquisition of sensitivity to insulin-like stimulation (l-3). Although freshly isolated adipocytes were refractory to insulin-like stimulation by GH (data not shown) and had higher resting [Ca’+], than sensitive cells, they, nevertheless, responded acutely to hGH with a gradual increase in [Ca2+]i over the next 6 min (Fig. 4). In contrast, [Ca’+J remained unchanged after the addition of hGH to cells that were 400 -
is C .: H
1
HOURS
2
3
4
POST-ISOLATION
3. Effects of duration of incubation on resting [Ca”]; isolated adipocytes. Adipocytes prepared from the epididymal rats were pooled for each experiment. Each bar represents for 8 experiments (19-27 cells/experiment); brackets indicate
FIG.
*,
P < 0.01;
**,
in freshly fat of 2 the mean the SEM.
P < 0.001.
IN ADIPOCYTES
175
preincubated for 3-4 h in the absence of GH and were, therefore, sensitive to insulin-like stimulation. Figure 4 shows representative 8-min scans of single adipocytes acutely treated with 500 rig/ml hGH immediately after isolation (upper panel) or 4 h later (lower panel). Similar data were obtained from three other cells studied in the first 2 h after excision and from three cells studied in the fourth hour. Effects of GH and actinomycin-D
in refractory
adipocytes
From the results of the experiments illustrated in Figs. 1 and 4 it appears that GH has two effects on [Ca’+]i: a delayed increase in resting [Ca’+]i produced in sensitive cells (see Fig. l), and an acute increase in [Ca2+]i produced only in refractory cells (see Fig. 4). It is possible that the acute and delayed effects of GH on [Ca’+], are manifestations of the same actinomycin-sensitive event, and that the acute effect can only be expressed when [Ca2+]i is above some minimal threshold concentration. Alternatively, sensitization to the acute response to GH may be a separate event from that which maintains resting [Ca2+]i. To distinguish between these possibilities, we delayed the addition of actinomycin-D until 60 min after hGH and determined its effects on [Ca’+]i when GH was added 3 h later (Fig. 5). In adipocytes incubated for 4 h in the absence of hGH, [Ca2+]i declined to about 100 nM regardless of whether actinomycin was present. [Ca*+]i was maintained at approximately 300 nM when 100 rig/ml hGH were added in the first hour and nearly doubled in response to 500 rig/ml hGH added in the fourth hour. Treatment with actinomycin-D 60 min after hGH did not interfere with maintenance of resting [Ca’+]i, but prevented the acute rise in [Ca2+]i in response to the second addition of hGH. These data suggest that different actinomycin-sensitive events may be responsible for maintenance of resting [Ca2+li and sensitization to the acute action of GH to increase [Ca’+]i. In addition, the onset of the acute rise in [Ca’+]i is too rapid to result from induced RNA and protein synthesis, suggesting yet a third phenomenon. The dose dependencies
is 5 T N 9
.-
500 400 300 200 100 n
hGH at tim; hGH at 180 Actinomycin
0 min
-
+ -
+ + -
” -
+ -
+ +
+
+
+
5. Effects of GH and actinomycin-D on [Ca”+], in refractory adipocytes. Adipocytes isolated from the epididymal fat of 2 rats/ experiment were incubated for 4 h in the presence or absence of 100 rig/ml hGH. Actinomycin-D, when present, was added after 1 h. Where indicated, 500 rig/ml hGH were added after 3 h, and [Ca’+], was measured 20-30 min later. Each bar represents the mean of 7 experiments (19-27 cells/experiment). Brackets indicate the SEM. t, P < 0.01; N.S., P > 0.05.
FIG.
SECONDS FIG. 4. Acute changes in [Ca”li in response to 500 rig/ml hGH in individual cells treated as described in Fig. 3. hGH was added in the time designated by the darkened box during the first hour (upperpane2) or third hour postisolation (lower panel).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES [Ca*‘l IN ADIPOCYTES
776
Endo. 1992 Vol131.No2
1
of each of these phenomena are shown in Figs. 6-8. To determine the minimal concentration of hGH needed to maintain resting [Ca’+]i, freshly isolated adipocytes were incubated with various concentrations of hGH for 60 min, washed, and reincubated for an additional 2 h before loading with fura-2. [Ca’+]i was indistinguishable from control values in adipocytes that were preincubated with 1 or 3 rig/ml hGH, but was maintained at about 250 nM in cells that were exposed to 10 rig/ml hGH or more (Fig. 6). The same procedure was followed to determine the concentration dependency of the process that enabled adipocytes to exhibit an acute responseto hGH, except that the cells were tested for their ability to exhibit an acute responseto 500 rig/ml hGH in the fourth hour. The acute increase in [Ca’+]i was seen only in cells that had been preincubated with hGH at concentrations of 10 rig/ml or higher, and 30 rig/ml were required for full expression of the response(Fig. 7). These data are expressedas the increment in [Ca’+]i because they are
I.0
.i
hGH
100
10.00
(nglml)
FIG. 6. Concentration-response relationship for maintaining resting [Ca”]{ in normal adipocytes incubated for 4 h in uitro. Freshly isolated adipocytes from 2 rats/experiment were incubated with the indicated concentrations of hGH for 60 min, washed, and reincubated for an additional 2 h before loading with furaAM. Each point represents the mean of 6 independent experiments (18-23 cells/experiment); brackets indicate the SEM. *, P < 0.05. 400
f 5 .-
600500-
:
2004
"'.".
"""'I .l
""".
"""'I
1
10
hGH
(nglml)
100
"'"m 1000
FIG. 8. Concentration-response relationship for production of the acute increase in [Ca*+h. Adipocytes prepared from 2 rats/experiment were incubated for 60 min with 100 rig/ml hGH, washed 3 times with fresh KRBG-1% BSA, and reincubated for 2 h in the absence of hormone. They were then loaded with furaAM and challenged with the indicated concentrations of hGH. [Ca2+li was measured 20-30 min later. Each point represents the mean of 6 experiments (18-23 cells/ experiment); brackets indicate the SEM. t, P < 0.01; tt, P < 0.001.
superimposed on a changing baseline (seeFig. 6). For determination of the concentration dependency of the acute increasein [Ca2+]i,adipocytes were incubated with 100 rig/ml hGH for the first hour, washed, and reincubated for 2 h in the absence of hormone. The cells were then loaded with fura- and incubated for 20 min with various concentrations of hGH before measurement of [Ca’+]i. The lowest concentration of hGH needed to produce even a small increase in [Ca’+]i was 25 rig/ml, and the effect was maximal at 125 ng/ ml (Fig. 8). The EDs0for maintenance of [Ca”]i (-5 rig/ml, asestimated by inspection of the curve) and that for enabling cells to exhibit and acute responseto hGH (-7 rig/ml) are, thus, quite similar, while 6- to g-fold higher concentrations of hGH were needed to elicit the acute increase in [Ca2+]i (E&or -40 rig/ml). Discussion
1
hGH
(ngiml)
FIG. 7. Concentration-response relationship for sensitizing adipocytes to the acute increase in [Ca*+]; produced by subsequent treatment with hGH. Adipocytes prepared from 2 rats/experiment were preincubated with the indicated concentrations of hGH for 60 min, washed 3 times with fresh KRBG-1% BSA, and reincubated for 2 h in the absence of hormone. They were then loaded with furaAM and challenged with 500 rig/ml hGH. Data are expressed as increments (A) in [Ca”li. Each point represents the mean of 6 independent experiments (18-25 cells/ observation); brackets indicate the SEM. t, P < 0.01; tt, P < 0.001.
Adipocytes appear to require frequent stimulation with GH to maintain their resting [Ca2+]iin the range of 200-300 nM. When deprived of GH for more than 3 h, [Ca2+],falls to a steady state concentration of about 100 nM, even though maintaining an even steeper transmembrane calcium gradient is likely to be more costly for cellular energy. The frequency of stimulation with GH required to maintain [Ca’+]i at the higher resting level appears to be about every 3-4 h and, thus, is similar to the normal frequency of pulsatile secretion of GH in rats (18). Maintenance of [Ca’+]i by GH requires ongoing synthesis of RNA and proteins that may be ion channels, ion pumps, or regulators of such membrane components. Because there are multiple Ca2+ exchangers, pumps, and channels in adipocytes (19-21), and because [Ca2+]iis also determined in part by [Na+]i and [H+]i as well as the concentration of other ions, each of which is regulated by more than one process,the range of possibleGH-dependent proteins is large. In addition to maintaining resting lCa2+],,GH produces an
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
GH INCREASES
[Ca”‘]
acute increase in [Ca’+]i that begins within a few seconds and persistsfor at least 20 min (9) and for as long as 1 h in some cells. This acute increase in [Ca”]i is the most rapid effect of GH that has been reported. We know little about its mechanism, except that the profile of change in [Ca*+]i produced by GH differs from that produced by oxytocin (9, 22, 23), which is thought to act by way of an inositol trisphosphate second messengersystem (23). The acute increasein [Ca2+liproduced by GH has thus far only been seen when resting [Ca2+li was 200 nM or higher, but merely maintaining [Ca’+]i at this level is not sufficient to permit expression of the acute response. Some additional GH-dependent factor(s) seemsto be required, as suggestedby the experiment shown in Fig. 5, in which delayed addition of actinomycin blocked expression of a subsequent acute responseeven though resting [Ca’+]i was above 250 nM. The concentration of GH needed to increase[Ca’+]i acutely is considerably higher than that needed to maintain resting [Ca’+]i and is similar to concentrations achieved during physiological pulsesof secretion (18). It might be expected, therefore, that in vim adipocyte [Ca*+]i may oscillate between about 200 nM and about 600 nM. However, at least at high concentrations (500 jLJ/ml), insulin blocked the acute effect of GH on [Ca2+li(9). We do not yet know whether physiological concentrations of insulin are sufficient to block this responseto GH. Unfortunately, the rather long time (-1.5 h) needed to prepare cells for measurement of [Ca’+]i precludes obtaining a meaningful profile of adipocyte [Ca*+]i in vim using current methods. In previous studies (9) we found that refractoriness to the insulin-like effects of GH is associatedwith a relative increase in [Ca2+]i,but could not determine whether the change in resting [Ca*+]i, the acute increase in [Ca*+]i, or both were critical determinants of refractoriness. The present results provide additional insight into the relationship between [Ca*+]iand expressionof the insulin-like response.The effect of GH on resting [Ca*+]i has a dose-dependency that correlates closely with the dose-dependency for induction of refractoriness in both normal adipocytes (7) and segmentsof adipose tissue from hypophysectomized rats (24), while the acute effect has a dose-dependency resembling that of the insulin-like effect (7). The effect on resting [Ca*+]i resembles the effect on the termination of the insulin-like responsein both its time course and the brevity of its sensitivity to blockade by actinomycin-D (5). In earlier experiments, the addition of actinomycin-D 60 min after GH did not prevent the insulin-like responsefrom terminating, but blocked the onset of refractoriness (5). We now find that resting [Ca*+]i is maintained under these circumstances,but the acute rise in [Ca*+]i in responseto subsequent stimulation with GH is blocked. Becauseadipocytes with the higher resting [Ca”]i are refractory to insulin-like stimulation only when the acute increasein [Ca2+liis present, it is likely that the acute rise in [Ca2+liis the immediate causeof refractoriness, but we cannot rule out a contribution from the resting [Ca”]i. The mechanismsby which GH increases[Ca*+]i acutely, and by which elevated [Ca’+]i blocks expression of the insulin-like response are unknown. We also do not yet know
IN ADIPOCYTES
777
whether the effects of GH on calcium in adipocytes are representative of effects of GH in other target tissues and whether increased [Ca2+]iplays a role in the hyperinsulinemia, decreased insulin sensitivity, and hypertension that frequently accompany excessive secretion of GH. The demonstrated relationship between GH and [Ca*+]i may provide new insights and new experimental approaches to these questions. References 1. Goodman HM, Coiro V 1981 Induction of sensitivity to the insulinlike action of growth hormone in normal rat adipose tissue. Endocrinology 108:113-119 2. Eden S, Schwartz J, Kostyo JL 1982 Effects of preincubation on the ability of rat adipocytes to bind and respond to growth hormone. Endocrinology 111:1505-1512 3. Schwartz Y, Goodman HM 1990 Refractoriness to the insulin-like effects of growth hormone depends upon calcium. Endocrinology 127:170-176 4. Goodman HM 1968 Growth hormone and the metabolism of carbohydrate and lipid in adipose tissue. Ann NY Acad Sci 148:419440 5. Goodman HM, Tai LR, Chipkin SR 1990 The isoquinoline sulfonamide inhibitors of protein phosphorylation, H-7, H-8, and HA1004, also inhibit RNA synthesis: studies on responses of adipose tissue to growth hormone. Endocrinology 126:441-450 6 Goodman HM 1981 Separation of early and late responses of adipose tissue to growth hormone. Endocrinology 109:120-129 7. Grichting G, Levy LK, Goodman HM 1983 Relationship between binding and biological effects of human growth hormone in rat adipocytes. Endocrinology 113:1111-1120 8. Fain TN. Kovacev VP. Scow RO 1965 Effect of growth hormone and dexamethasone on lipolysis and metabolism ilisolated fat cells of the rat. J Biol Chem 240:3522-3529 9. Schwartz Y, Goodman HM, Yamaguchi H 1991 Refractoriness to growth hormone is associated with increased intracellular calcium in rat adipocytes. Proc Nat1 Acad Sci USA 88:6790-6794 10. Rodbell M 1964 Metabolism of isolated fat cells. I. Effects of hormones on glucose metabolism and lipolysis. J Biol Chem 239:375-380 11. Grynkiewicz G, Poenie M, Tsien RY 1985 A new generation of Ca’+ indicators with greatly improved fluorescence properties. J Biol Chem 260:3440-3450 12. Williams DA, Fogarty KE, Tsien RY, Fay FS 1985 Calcium gradients in single smooth muscle cells revealed by the digital imaging microscope using fura-2. Nature (Lond) 318:558-561 13. Smal J, Closset J, Hennen G, De Meyts P 1987 Receptor binding properties and insulin-like effects of human growth hormone and its 20-kDa variant in rat adipocytes. J Biol Chem 262:11071-11079 14. Moody AJ, Stan MA, Stan M 1974 A simple free fat cell bioassay for insulin. Horm Metab Res 6:12-16 15. Winer Bl 1962 Statistical Principles in Experimental Design, ed 2. McGrawLHill, New York _ 16. Dixon WJ, Brown MB, Engleman L, Frane JW, Hill MA, Jennrich RI, Toporek JD 1985 BMPD Statistical Software. University of California Press, Berkeley 17. Snedecor GW. Cochran WG 1980 Statistical Methods, ed 7. Iowa State University Press, Ames 18. Tannenbaum GS, Martin JB 1976 Evidence for an endogenous u&radian rhythm governing growth hormone secretion in the rat. Endocrinology 98:562-570 19. Pershadsingh HA, Lee L-Y, Snowdowne KW 1989 Evidence for a sodium/calcium exchanger and voltage-dependent calcium channels in adipocytes. FEBS Lett 244:89-92 20. Pershadsingh HA, Landt M, McDonald JM 1980 Calmodulinsensitive ATE’-dependent Ca*+ transport across adipocyte plasma membranes. 1 Biol Chem 255:8983-8986 21. Black BL, Jarrett L, McDonald JM 1981 The regulation of endo-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
778
GH INCREASES
[Ca”]
plasmic reticulum calcium uptake of adipocytes by cytosolic calcium. J Biol Chem 256:322-329 22. Kelley KL, Deeney JT, Corkey BE 1989 Cytosolic free calcium in rat adipocytes. J Biol Chem 264:12754-12757 23. Uto A, Arai H, Ogawa Y 1991 Reassessment of fura- and the ratio
IN ADIPOCYTES
Endo. Voll31.
method for determination of intracellular Ca” concentrations. Cell Calcium 12:29-37 24. Goodman HM 1984 Biological activity of bacterial derived human growth hormone in adipose tissue of hypophysectomized rats. Endocrinology 114:131-135
Recertification in Endocrinology, Diabetes, and Metabolism Endocrinology, Diabetes, and Metabolism Board of the American Board of Internal Medicine Drs. Martin 1. Surks (Chairman), Glenn D. Braunstein, John D. Brunzell, A. Kreisberg, D. Lynn Loriaux, and Lawrence G. Raisz
Philip E. Cryer, Willa A. Hsueh, Robert
Since 1990, the duration of validity of certification by the American Board of Internal Medicine (ABIM) has been limited to 10 yr. This policy was adopted because medical information changes rapidly and the public needs assurance that certified internists have maintained their skills and kept their knowledge up to date. Time-limited certification and recertification are obligations of an accountable profession. The ABIM
recently has completed
and adopted
1992 No 2
plans for a comprehensive
recertification
program.
Entry into Recertification: Diplomates may attempt recertification only in disciplines in which they were previously certified. Certified endocrinologists may seek recertification at any time after initial certification in endocrinology, diabetes, and metabolism only, in internal medicine only, or in both. Diplomates can allow a time-limited certificate in internal medicine to expire without jeopardizing eligibility for recertification in endocrinology, diabetes, and metabolism. However, expiration of a certificate in internal medicine or endocrinology means the ABIM no longer recognizes an individual as certified in that discipline. Certificates in internal medicine or endocrinology, diabetes, and metabolism issued before 1990 are not time-limited and are valid for life; individuals holding such certificates are eligible to seek recertification without placing existing certificates at risk. Please note that all successful candidates for recertification will be issued a certificate with diabetes in the title: Endocrinology, Diabetes, and Metabolism. The Recertification Process: The recertification process consists of three steps: documentation of clinical competence, completion of the self-evaluation process, and success on a proctored, written final examination. Dual Recertification: The Board anticipates that most endocrinologists will seek recertification in internal medicine as well as endocrinology, diabetes, and metabolism and has, therefore, developed an efficient process for dual recertification that does not change the standards required for recertification in each discipline. For both the self-evaluation processes and the final examinations, substitution of required modules for self-selected modules can reduce the total number of modules required for recertification in internal medicine and endocrinology to six self-evaluation process modules (three in general internal medicine, three in endocrinology) and four final examination modules (two in general internal medicine, two in endocrinology). The score for each final examination will be determined independently. Thus, an individual can be successful in becoming recertified in endocrinology, diabetes, and metabolism but not in internal medicine, even when the processes are undertaken concurrently. Schedule of Availability: This comprehensive recertification program will become available in 1995. The self-evaluation process will be available continuously beginning in 1995. Final examinations will be available annually beginning in 1996. Diplomates who would like to become recertified before this time can take a regularly scheduled certification examination for recertification credit (interim voluntary recertification).
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 22 November 2015. at 04:00 For personal use only. No other uses without permission. . All rights reserved.
