- Email: [email protected]
Cartilage sulfation inhibitor from rat liver curtails growth of embryonic chicken cartilage in vitro
Cartilage
Sulfation
Inhibitor From Rat Liver Curtails Growth of Embryonic Chicken Cartilage in Vitro
Rena Vassilopoulou-Sellin,
Robert L. Lock II, Caroline 0. Oyedeji, and Naguib A. Samaan
We studied the effect of high MW cartilage sulfation (somatomedin) inhibitors from rat liver on cartilage growth in vitro. Pelvic rudiments from 11-day-old chicken embryos (5.70 mg average weight) were incubated in an organ-culture system with defined tissue-culture medium; after two days (T,,). media were changed and incubation continued for another three days (T2_6).Normal rat serum (10% vol/vol) stimulated cartilage growth (weight change + 1.33 + 0.16 mg, mean t SEM T,, and $ 1.33 r 0.27 mg, T,,). Partially purified cartilage sulfation inhibitors (CSI) caused a weight decrease (- 1.09 * 0.17 mg T,, and -0.44 ? 0.06 mg T,,). Adding inhibitors to cartilage incubations containing normal serum abolished the growth-promoting effect of serum f -0.79 + 0.07 mg T, and -0.40 + 0.08 mg T,,). The growth-curtailing effect of CSI was reversible; after preincubating cartilage with CSI for two days (-0.62 + 0.11 mg TQ2)_ subsequently exposing it to normal serum allowed cartilage growth to resume (+ 1.02 + 0.21 mg T,.,). Cartilages incubated with normal serum and various concentrations of inhibitor exhibited a dose-dependent inhibition of serum-stimulated growth. Cartilage Length was not altered by the inhibitor: cartilage dry:wet weight ratio or protein concentration (pg/mg wet weight) did not differ among groups. Triodothyronine (T, 1 stimulated cartilage growht in a dose-dependent manner as expected. Adding CSI to cartilage incubations containing T, (1.5 nmol/l or 15 nmol/L) completely abolished the growth-stimulating effect of T,. This is the first demonstration that high MW cartilage sulfation inhibitors from rat liver cause a dose-dependent and reversible inhibition of the growth of embryonic chicken cartilage in vitro; their ability to inhibit both serum- and T,-stimulated growth suggests that they exert broad, antianabolic effects. @ 1987 by Grune & Stratton, Inc.
OMATOMEDIN inhibitors obtained from serumle3 and liver preparation? are thought to be antagonists of somatomedin-stimulated anabolic events.‘** Although they have been shown to inhibit basal cartilage metabolic processes, 2B6 they have generally been measured indirectly by the ability of inhibitor-enriched samples to oppose serum-stimulated cartilage sulfation. All previous studies for somatomedin inhibitors have been based on measurements of radiolabeled isotope incorporation and on the demonstration that inhibitors decrease sulfate incorporation into cartilage. Such methods have had several shortcomings: (1) they are indirect measurements of biologic action, (2) the results are very sensitive to any small changes in sulfate concentration, (3) it has not been possible to show that inhibitor action is reversible due to the assay techniques; therefore, there remains concern that the inhibitors may be nonspecific toxins that simply damage the bioassay cartilage, and (4) it has not been possible to demonstrate actual inhibition of cartilage growth. We utilized the embryonic chicken pelvic rudiment system as a method for the measurement of cartilage sulfation/ somatomedin inhibitors. This model has been used extensively by other investigators for the study of growth regulationg-“; because the pelvic rudiments are an integral part of the growing skeleton and permit direct measurement of several growth parameters, we employed this technique to examine the role of inhibitors on cartilage growth and to answer the aforementioned questions. Several cartilage sulfation/somatomedin inhibitors appear to exist based on molecular size determinations.6.7.‘6 The predominant inhibitor in liver extracts is a heat labile, >50,000 MW factor6; it is the actions of this particular CSI fraction that we concentrate on in this report. The present studies demonstrate cartilage growth inhibition by this partially purified hepatic inhibitor preparation. The dose-depen-
S
dent,
reversible
Mefabo//sm,Vol36.
inhibition
No
of both
serum
1 (January). 1987:pp89-94
and
T,-stimulated
cartilage growth indicates action for these inhibitors.
a broad.
direct,
antianabolic
MATERIALS AND METHODS
Prolonged Organ Culture of Pelvic Cartilages The procedure was modified from Burch and Lebovitz9 Briefly, pelvic rudiments from I I-day-old chicken embryos were removed, cleaned of adjacent tissues, measured. weighed, and combined individually with the test substances plus Waymouth’s MB 752/l tissue culture medium and a mixture of streptomycin, penicillin, polymixin, and amphotericin B as previously described’; the mixture (final volume 2 mL) was then placed in 24-well polystyrene tissue culture plates. Five-day incubations were carried out in a humidified, 37OC, 5% CO,:95% room air tissue culture incubator. Serum, CSI, or T,, alone or in combination, were added as described below; after two days (Ts2). each cartilage was removed, blotted dry, measured, weighed, and placed in fresh medium for an additional three-day incubation (T2_s). Groups of 48 cartilages from 24 eggs were studied each week.
Growth Factors and Inhibitors Serum from a pool of normal 150-g rats was added at I %, 5%. or 10% vol/vol as specified in “Results.” T, at 0. I5 nmol, I .5 nmol, or 15 nmol was studied with or without normal serum or CSl. Serum
of Endocrinolo~v, of Texas M. D. Anderson
From the Section University
Department Hospital
of Medicine.
the
and Tumor Institute
at Houston. Supported by the American Diabetes Association Feasibility Grant and the Nancy Carmichael Gift Fund. Address reprint requests to Rena Vassilopoulou-Sellin. MD, Section of Endocrinology, Department of Medicine, the University of Texas M. D. Anderson Hospital and Tumor Institute at Houston, 6723 Bertner Ave. Houston, TX 77030. 8 1987 by Grune & Stratton. Inc. 0026-0495/87/3601-0016$03.00/0
89
90
VASSILOPOULOU-SELLIN
from a pool of hypophysectomized 80-g rats was used at 10% (vol/vol) as a GH-deficient serum control. CSI was prepared as previously described.6 Briefly, weighed, frozen rat livers were homogenized with 4 ~010.1 mol/L ammonium acetate, pH 7.4, and centrifuged; supernates were lyophilized (8 g liver weight yield 1 g lyophilized powder). Lyophilized extract was subjected to gel filtration on Sephadex G-100 eluted with 0.1 mol ammonium acetate, pH 7.4; the fraction previously tested with the rat cartilage bioassay and shown to be enriched in cartilage sulfation inhibitor Kav 0.06 to 0.18 was collected and lyophilized, thus yielding a partially purified CSI preparation. This CSI fraction is similar in molecular weight to the predominant inhibitor-enriched fraction of diabetic rat serum6 and of inhibitor released by the liver of diabetic rats during isolated liver perfusion (unpublished observation). The activity of this material on the hypox rat costal cartilage bioassay has been detailed previously.* Recalibrated for the present series of experiments, each I mg lyophilized powder contains 0.3 mg protein and produces a 50% reduction of basal cartilage sulfation in the bioassay, well within the steep portion of the bioassay doseresponse curve. For the cartilage incubations, 20 mg CSI powder/l mL culture medium was prepared and added at 10% vol/vol (ie, 2 mg/mL), 5% (ie, I mg/mL), or 1% (ie, 0.2 mg/mL) as specified below. During the course of these studies, we found that growth stimulation induced by normal serum was similar at l%, 5%, or 10% serum
Medium
6
Change
1
+1.0
i!l FJ
ET AL
-cs
I
1
I
I
1
0
10%
--Buffer . . . . ... . .. NRS 10%
I
I
2
5
Time (Days) * Reversibility of CSI effect. Cartilages were placed in Fig 2. fresh. different media at T,, (medium change). All media were at 10% vol/vol as in Fig 1. N = 4 for 8UF - NRS and NRS - 8UF; n = 13 for CSI - NRS and NRS - CSI.
+2.0
HYPOX
Buffer
0
cs NRS+CS
concentration and that growth inhibition induced by CSI was similar at 5% or 10% CSI concentration. To avoid duplication, the data in Fig 1 depict growth changes for cartilages incubated with 10% CSI and sera. The data in Figs l-4 are presented as change of wet weight in absolute milligrams. There was no loss of potency for incubated sera or CSI during the in vitro incubation periods, as ascertained by measuring sulfate uptake induced by these media with the hypox rat costal cartilage bioassay. At T,, some cartilages were homogenized in 0.9% NaCl and 0.2% Triton X-100 as described by Burch and Lebovitzy and protein content was measured according to the method of Lowry.” Other cartilages were dehydrated with ETOH and ether, weighed and dry:wet weight ratio determined. Materials
-2.0
I
0
I
I
I
I
2
I
5
Waymouth’s MB 752/l medium was purchased from Grand Island Biological Co (Grand Island, NY). Other chemicals were from Sigma Chemical Co (St Louis). CSI was prepared from liver extracts of Sprague-Dawley rats from Charles River Laboratories (Wilmington, Mass). Fertilized eggs were obtained from Richglo (El Campo, TX).
Time (Days) RESULTS Fig 1. Embryonic chicken pelvic rudiment growth in vitro. Effect on weight by normal rat serum at 10% vol/vol (NRSL hypophysectomized GH-deficient rat serum at 10% vol/vol IHRS), Waymouth’s culture buffer alone (BUF). CSI at 10% vol/vol fie, 2 mg/mL as described in “Materials and Methods”). or NRS 10% plus CSI 10% INRS + CSI). Values are given as mean + SEM fn = 8 to 13 cartilages each medium). lP < 0.05 Y initial weight.
Pelvic Cartilage Growth in Vitro Normal rat serum stimulated a significant increase in pelvic cartilage weight (t 1.33 + 0.16 mg T’s_, and + 1.33 f 0.27 mg Ta.s. Incubation with serum from hypophysectomized GH-deficient rats resulted in a small weight gain,
CSI INHIBITS
91
CARTILAGE GROWH
l-
NRS
J
k
+3.0
?I s lz a
NRS+T3
1SnM
NRS+T3
0.15&l
T3
15nM
T3
1.5nM
+1.0
5
T3 0.15nM
E .-cn s
3cp< .05 vs NRS
-1.0
A p -z .05 vs NRS+CS
I
I
I
0
2
5
Time (Days) Fig 4. CSI effect on T,-stimulated cartilage growth. Cartilages were incubated with T,, 0.15 nmol, or 1.5 nmol, with and without NRS at 10% vol/vol. Also, cartilages were incubated with T,, 1.5 nmol. or 15 nmol, with and without CSI at 10% vol/vol (ie, 2 mg/mL). N = 6 to 9 cartilages in each group. P 4 0.05 for T, v CSI + T, for each T, concentration at all time points.
0
2
5
Time (Days) Fig 3. Dose-dependent inhibition of cartilage growth by increasing concentrations of added CSI. Normal rat serum data at 1% and 5% vol/vol behaved identically and were therefore combined (NRS. n = 14). Top panel, CSI at 1% vollvol (0.2 mg/mL prepared as in “Materials and Methods”) with or without NRS (n = 16). Bottom panel, CSI at 5% vol/vol Il.0 mg/mL as prepared in “Materials and Methods”) with or without NRS (n = 13). lP < 0.05 Y NRS alone: AP < 0.05 Y NRS CSI.
+0.68 t 0.14 mg at T,., only. In contrast, a significant decrease in weight was observed in cartilages incubated with CSl (-1.09 t 0.17 mg To_* and -0.44 + 0.06 mg T2_+ Further, adding CSI to cartilage incubations containing normal serum abolished the growth-stimulating effect of the serum (-0.79 + 0.07 mg To_* and -0.40 * 0.08 mg T,,). Although some growth occasionally occurred in pelvic cartilages incubated with tissue culture buffer alone, overail, in this group of experiments no significant weight increase was demonstrated over the five-day period. Over the five-day incubation period (Table l), exposure to normal serum was associated with a 40% increase in weight and a +0.17 i 0.01 cm increase in length (approximately 25% of initial length). In contrast, exposure to CSI was associated with a 25% decrease in weight (5.89 k 0.47 mg to 4.44 t 0.40 mg T,,) with no change in length. Exposure to buffer alone produced no change in either length or weight.
Protein concentration after five days incubation was comparable among the three groups. At TM, before incubation, pelvic cartilage dry weight was 13 k 2.9 x 10e2 mg/mg wet weight. After five days of incubation, the dry:wet weight ratio was comparable among all groups. This indicates that changes in wet weight are not due to tissue swelling but accurately reflect growth. Because cartilage dehydration precludes further incubation and because the dry:wet weight ratio is stable among groups, wet weights were generally measured in the different incubation conditions. Reversibility of CSI Effect We studied the reversibility of the effects of CSI and serum by examining the responsiveness of the cartilages to changes in the incubation medium. Exposure to normal serum after incubation with buffer alone resulted in a resumption of growth (-0.10 + 0.18 mg T,,., in buffer, followed by +.66 2 .I 3 mg T,, in normal serum). Of particular interest, cartilage growth inhibition due to incubation with CSI (-0.62 2 0.11 mg TO_2) was reversed by subsequent exposure to normal serum (t1.02 _+0.21 mg T,,). In the reverse experiment, growth stimulation by normal serum (+ 1.30 + 0.14 mg T,,) was comparably reversed by subsequent exposure to CSI ( -0.61 t 0.10 mg TM). To explore
further
the tissue
or reversibility,
we also
92
VASSILOPOULOU-SELLIN Table 1. Growth Parameters Parameter
of Chick Pelvic Cartilages
Normal Serum
Length change (O-5 d) (cm)
+0.17
+ 0.01 (11)
Wet weight (0 d) (mg)
5.71
? 0.23 (12)
Wet weight (5 d) (mg)
8.0 & 0.42 (12)
Dry/wet weight ratio (5 d)
0.1 1 & 0.044 (12)
Protein concentration* (g/me wet wt)
5.38
+ 1.91 (17)
ET AL
CSI
-0.02
Buffer
* 0.01
+0.07
(ll)t 5.89 ? 0.47
-
+ 0.02 (lo)t
5.42
(12)t 4.44 f 0.40
f 0.40 (9)t
5.55
+ 0.46
(12) 0.12 i 0.045
0.11
(15) 4.84 ? 0.67
(10) 6.34 ? 1.14
19) 5 0.042
(4)
(14)
Values in parentheses refer to the number of cartilages. *Concentration in cartilage homogenates after five days incubation. tP < 0.05 \I normal serum (mean t SEM).
performed more prolonged cartilage incubation with two medium changes (Table 2). Cartilages were incubated for seven days and media were changed at three and five days. Growth of cartilages exposed to normal serum To.,, then CSI T,.,, then normal serum again TS_? was compared with growth of cartilages exposed to CSI T,, then normal serum T3_*then CSI again TS_,. Incubation with normal serum stimulated cartilage growth each time and incubation with CSI inhibited growth each time in a pattern comparable to Fig 2. Dose-Dependent CSI Effect on Cartilage Growth
To delineate the dose-dependent action of CSI, we conducted dose response studies (Fig 3). Cartilage incubated with 1% CSI did not exhibit growth inhibition until TSd (-0.68 + 0.15 mg), although its weight change was significantly less than that produced by normal serum both at TM and TSd. The growth of cartilage incubated with a combination of serum and 1% CSI was intermediate and significantly less than that induced by serum alone. When the concentration of CSI was higher (5% vol/vol), growth inhibition was significant and more pronounced at both T,, and T,,. Furthermore, when cartilages were exposed to the combination of serum and 5% CSI, the CSI effect predominated and growth was inhibited overall (-0.53 * 0.10 mg T,,_2and -0.40 + 0.05 mg T2_5). CSI Effect of T,-Stimulated Cartilage Growth Triodothyronine has previously been shown to stimulate cartilage growth”-“; the mechanism of action is believed to differ from that of somatomedins. We were, therefore, interested to determine whether CSI opposes somatomedinstimulated growth selectively or whether it can also inhibit T,-stimulated growth. Table 2. Reversibility of CSI Action TO= T3d (Awt = mg) CSI -0.60 NRS +0.98
t 0.25 f 0.30
T, = TM (Awt = mg) NRS +1.29 + 0.32 CSI -0.93 + 0.26
T5= T,, IAwt - mg) CSI -0.61 NRS +0.82
Values are given as mean + SE for all Awt. NRS and CSI are given at 10% concentration each. N refers to the number of cartilages.
+ 0.20 * 0.27
N 10 7
As expected, T, alone stimulated cartilage growth in a dose-dependent fashion at concentrations between 0.15 nmol and 15 nmol. Interestingly, the combination of normal serum and T, produced apparently additive growth stimulation (ie, during the first two days of incubation (T&, cartilage weight increased by + 1.80 * 0.26 mg in the presence of NRS + T, 0.15 nmol relative to +0.32 + 0.06 mg in the presence of Tj 0.15 nmol alone; during the subsequent three days of incubation (T,.,), cartilage weight increased further by + 1.95 * 0.23 mg in the presence of NRS + Ts 0.15 nmol relative to +0.32 + 0.21 mg in the presence of T3 0.15 nmol alone). The CSI-induced cartilage growth inhibition was dissociated from somatomedin action in these studies. Cartilages exposed to CSI plus T, at either 1.5 nmol or 15 nmol exhibited significant growth inhibition throughout the incubation period (ie, during the first two days of incubation (Ta2), cartilage weight increased by +0.80 k 0.10 mg in the presence of T, 1.5 nmol while cartilage weight decreased by -0.87 + 0.17 mg in the presence of combined CSI + T, 1.5 nmol; during the subsequent three days of incubation (TZ_5)r cartilage weight increased further by +0.95 _t 0.34 mg in the presence of T, 1.5 nmol while cartilage weight decreased by -0.16 k 0.04 mg in the presence of CSI + T3 1.5 nmol). In a separate experiment, in cartilages incubated with combined normal serum T3 and CSI, growth inhibition was also seen (ie, -0.33 * 0.07 mg at T,, and +0.04 _t 0.14 mg at T2_5for combined media, n = 4). This finding suggests that CSI has broad antianabolic action and the previously used term “somatomedin inhibitor” is too restrictive. DISCUSSION
Despite accumulating knowledge of the factors that regulate cartilage and skeletal growth, information about naturally occuring brakes on cartilage metabolism remains sparse. The somatomedin inhibitors have been described by several investigators as factors obtained from sera or liver preparations that inhibit somatomedin-stimulated metabolism of cartilage,‘-’ although they may also inhibit directly basai cartilage, muscle, and adipose tissue metabolism.2~5~‘*It has been suggested that poor growth in children with starvation, uncontrolled diabetes, or uremia is due in part to increased levels of circulating somatomedin inhibitors. How-
CSI INHIBITS
93
CARTILAGE GROWTH
ever, it has been very difficult to prove that these factors actually inhibit skeletal growth. In addition, most inhibitor bioassays express the presence of inhibitor relative to somatomedin-stimulated cartilage sulfation; there still remains some doubt in many investigators’ minds whether inhibitors may be somatomedin-binding proteins, which somehow interfere with bioassays of somatomedin activity. Finally, the inability to demonstrate that cartilage sulfation inhibition by these factors can be reversed has created the concern that inhibitors are nonspecific toxins rather than metabolically regulated factors. We felt that resolution of these important issues was not possible because of technical limitations of existing bioassays. In the present study we used a new approach to measure the biologic action of a partially purified high molecular weight (MW) cartilage sulfation/somatomedin inhibitor from rat liver. With the embryonic chicken pelvic rudiment bioassay, we were able to show reversible, dose-dependent inhibition of serum-stimulated and thyroid hormone-stimulated cartilage growth. Recent studies indicate that somatomedin-inhibitory activity in the sera of diabetic,16 starved,’ and uremic rats” consists of several MW components. After gel filtration with Sephadex G-100, pH 7.4, we have also found small MW inhibitors in rat liver extracts and diabetic rat sera6 and in rat liver perfusates (unpublished observation). However, we find a predominant fraction of cartilage sulfation/somatomedin inhibitors at MW >50,000. and it is this inhibitor fraction that we describe in this report. Embryonic chicken cartilage has been used extensively to elucidate the hormonal control of growth stimulation.g-‘s We chose the embryonic chicken pelvic rudiment prolonged organ culture system because it allows the study of different parameters including measurements of actual growth of a growing skeletal organ. It is, therefore, a simple model in which to study the effect of CSI on overall cartilage growth. Although in this study incubation with buffer alone did not support cartilage growth, the stimulation of cartilage
growth by T, and normal serum (in terms of both weight and length increments) is qualitatively consistent with previous reports for this experimental mode1.9*‘4Our data provide the first demonstration that CSI produces inhibition of cartilage growth in a consistent, dose-dependent fashion. The cartilage length did not change, and the rudiments appeared sturdy and viable grossly and under light microscopy (not shown). The change in weight was unaccompanied by changes in dry:wet weight ratio or protein concentration, suggesting that the effect was not due simply to altered hydration of the cartilages. Of critical importance, growth inhibition by CSI is reversible. In previous studies of somatomedin inhibitors using cartilage bioassays,2*4,6 the methodology was such that reversibility could not be tested. The present experiments demonstrate for the first time that cartilage exposed to CSl can recover and resume growth when presented with a stimulatory medium. Serum-stimulated growth was comparable whether the cartilage had been incubated previously with CSI (Fig 2, Table 2) or with normal serum (Fig I). The reversibility of CSI action strongly supports the view that this factor is not simply a tissue toxin. Thyroid hormone has previously been shown to be critically important for normal skeletal development. Researchers believe that the mechanism of growth stimulation by thyroid homone differs from that by somatomedins.‘0’20 CSI clearly inhibits T,-stimulated cartilage growth at all incubation times. The demonstration that CSI blocks TX-stimulated growth eliminates the possibility that it represents a somatomedin-binding protein. Further, the inhibition of growth stimulation by both somatomedins in normal serum and T, alone supports the view that CSl has broad antianabolic tissue action.
ACKNOWLEDGMENT
We thank Rosa M. Quiroz for her expert assistance preparation
with the
of the manuscript.
REFERENCES
I. Bomboy JD Jr, Burkhalter VJ. Nicholson
WE, et al: Similarity of somatomedin inhibitor in sera from starved, hypophysectomized, and diabetic rats: Distinction from a heat-stable inhibitor of rat cartilage metabolism. Endocrinology 112:371-377, 1983 2. Phillips LS, Vassilopoulou-Sellin R, Reichard LA: Nutrition and somatomedin. VIII. The “somatomedin inhibitor” in diabetic rat serum is a general inhibitor” in diabetic rat serum is a general inhibitor of growing cartilage. Diabetes 28:919-924, 1979 3. Phillips LS, Belosky DC, Reichard LA: Nutrition and somatomedin. VI. Somatomedin activity and somatomedin inhibitory activity in sera from normal and diabetic rats. Endocrinology 104:1519~1524, 1979 4. Binoux M, Lassarre C, Seurin D: Somatomedin production by rat liver in organ culture II. Studies of cartilage sulfation inhibitors released by the liver and their separation from somatomedins. Acta Endocrinol 93:83-90, 1980 5. Vassilopoulou-Sellin R, Phillips LS, Reichard LA: Nutrition and somatomedin. Vll. Regulation of somatomedin activity by the perfused rat liver. Endocrinology 106:260-267, 1980 6. Vassilopoulou-Sellin
R, Oyedeji
CO, Samaan
NA: Extraction
of cartilage sulfation inhibitors and somatomedins from rat liver. Endocrinology 114:576-581, 1984 7. Salmon WD Jr, Holladay LA, Burkhalter VJ: Partial characterization of somatomedin inhibitor in starved rat serum. Endocrinology I 12:360-370,1983 8. Phillips LS, Belosky DC, Reichard LA: Nutrition and somatomedin. V. Action and measurement of somatomedin inhibitor(s) in diabetic rat serum. Endocrinology 104: 15 13-l 5 18, 1979 9. Burch WM, Lebovitz HE: Triiodothyronine stimulation of in vitro growth and maturation of embryonic chick cartilage. Endocrinology 111:462-468,1982 IO. Kemp SF, Mutchnick M, Hintz RL: Hormonal control of protein synthesis in chick chrondrocytes: A comparison of effects of insulin, somatomedin C and triiodothyronine. Acta Endocrinol 107:179-184, 1984 Il. Audhya TK, Segen BJ, Gibson KD: Stimulation of proteoglycan synthesis in chick embryo sternum by serum and L-3,5,3’triiodothyronine. J Biol Chem 251:3763-3767, 1976 12. Hall BK: Thyroxine and the development of the tibia in the embryonic chick. Anat Rev 176:49, 1973 13. Garland JT, Jennings J, Levitsky LL, et al: Stimulation of
94
DNA synthesis in isolated chondrocytes by somatomedin. II. Validation of the assay for clinical use and comparison with the stimulation of protein synthesis. J Clin Endocrinol Metab 43:847-85 1, 1976 14. Hall K: Quantitative determination of the sulphation factor activity in human serum. Acta Endocrinol63:338-350, 1970 15. Daughaday WH, Phillips LS, Herington AC: Measurement of somatomedins by cartilage in vitro. Methods Enzymol 37B:93109.1975 16. Phillips LS, Bajaj VR, Fusco AC, et al: Nutrition and somatomedin. XI. Studies of somatomedin inhibitors in rats with streptozotocin-induced diabetes. Diabetes 32: 1117-I 125, 1983
VASSILOPOULOU-SELLIN
ET AL
17. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the folin phenol reagent. J Biol Chem 193:265-274, 1951 18. Phillips LS, Scholz TD: Nutrition and somatomedin. IX. Blunting of insulin-like activity by inhibitor in diabetic rat serum. Diabetes 31:97-104, 1982 19. Phillips LS, Kopple JD: Circulating somatomedin activity and sulfate levels in adults with normal and impaired kidney function. Metabolism 30:1091-1095, 1981 20. Burch WM, Lebovitz HE: In vitro stimulation of alkaline phosphatase activity in immature embryonic chick pelvic cartilages by adenosine 3’,5’-monophosphate. J Cell Biol 93:338-342, 1982
Sulfation
Inhibitor From Rat Liver Curtails Growth of Embryonic Chicken Cartilage in Vitro
Rena Vassilopoulou-Sellin,
Robert L. Lock II, Caroline 0. Oyedeji, and Naguib A. Samaan
We studied the effect of high MW cartilage sulfation (somatomedin) inhibitors from rat liver on cartilage growth in vitro. Pelvic rudiments from 11-day-old chicken embryos (5.70 mg average weight) were incubated in an organ-culture system with defined tissue-culture medium; after two days (T,,). media were changed and incubation continued for another three days (T2_6).Normal rat serum (10% vol/vol) stimulated cartilage growth (weight change + 1.33 + 0.16 mg, mean t SEM T,, and $ 1.33 r 0.27 mg, T,,). Partially purified cartilage sulfation inhibitors (CSI) caused a weight decrease (- 1.09 * 0.17 mg T,, and -0.44 ? 0.06 mg T,,). Adding inhibitors to cartilage incubations containing normal serum abolished the growth-promoting effect of serum f -0.79 + 0.07 mg T, and -0.40 + 0.08 mg T,,). The growth-curtailing effect of CSI was reversible; after preincubating cartilage with CSI for two days (-0.62 + 0.11 mg TQ2)_ subsequently exposing it to normal serum allowed cartilage growth to resume (+ 1.02 + 0.21 mg T,.,). Cartilages incubated with normal serum and various concentrations of inhibitor exhibited a dose-dependent inhibition of serum-stimulated growth. Cartilage Length was not altered by the inhibitor: cartilage dry:wet weight ratio or protein concentration (pg/mg wet weight) did not differ among groups. Triodothyronine (T, 1 stimulated cartilage growht in a dose-dependent manner as expected. Adding CSI to cartilage incubations containing T, (1.5 nmol/l or 15 nmol/L) completely abolished the growth-stimulating effect of T,. This is the first demonstration that high MW cartilage sulfation inhibitors from rat liver cause a dose-dependent and reversible inhibition of the growth of embryonic chicken cartilage in vitro; their ability to inhibit both serum- and T,-stimulated growth suggests that they exert broad, antianabolic effects. @ 1987 by Grune & Stratton, Inc.
OMATOMEDIN inhibitors obtained from serumle3 and liver preparation? are thought to be antagonists of somatomedin-stimulated anabolic events.‘** Although they have been shown to inhibit basal cartilage metabolic processes, 2B6 they have generally been measured indirectly by the ability of inhibitor-enriched samples to oppose serum-stimulated cartilage sulfation. All previous studies for somatomedin inhibitors have been based on measurements of radiolabeled isotope incorporation and on the demonstration that inhibitors decrease sulfate incorporation into cartilage. Such methods have had several shortcomings: (1) they are indirect measurements of biologic action, (2) the results are very sensitive to any small changes in sulfate concentration, (3) it has not been possible to show that inhibitor action is reversible due to the assay techniques; therefore, there remains concern that the inhibitors may be nonspecific toxins that simply damage the bioassay cartilage, and (4) it has not been possible to demonstrate actual inhibition of cartilage growth. We utilized the embryonic chicken pelvic rudiment system as a method for the measurement of cartilage sulfation/ somatomedin inhibitors. This model has been used extensively by other investigators for the study of growth regulationg-“; because the pelvic rudiments are an integral part of the growing skeleton and permit direct measurement of several growth parameters, we employed this technique to examine the role of inhibitors on cartilage growth and to answer the aforementioned questions. Several cartilage sulfation/somatomedin inhibitors appear to exist based on molecular size determinations.6.7.‘6 The predominant inhibitor in liver extracts is a heat labile, >50,000 MW factor6; it is the actions of this particular CSI fraction that we concentrate on in this report. The present studies demonstrate cartilage growth inhibition by this partially purified hepatic inhibitor preparation. The dose-depen-
S
dent,
reversible
Mefabo//sm,Vol36.
inhibition
No
of both
serum
1 (January). 1987:pp89-94
and
T,-stimulated
cartilage growth indicates action for these inhibitors.
a broad.
direct,
antianabolic
MATERIALS AND METHODS
Prolonged Organ Culture of Pelvic Cartilages The procedure was modified from Burch and Lebovitz9 Briefly, pelvic rudiments from I I-day-old chicken embryos were removed, cleaned of adjacent tissues, measured. weighed, and combined individually with the test substances plus Waymouth’s MB 752/l tissue culture medium and a mixture of streptomycin, penicillin, polymixin, and amphotericin B as previously described’; the mixture (final volume 2 mL) was then placed in 24-well polystyrene tissue culture plates. Five-day incubations were carried out in a humidified, 37OC, 5% CO,:95% room air tissue culture incubator. Serum, CSI, or T,, alone or in combination, were added as described below; after two days (Ts2). each cartilage was removed, blotted dry, measured, weighed, and placed in fresh medium for an additional three-day incubation (T2_s). Groups of 48 cartilages from 24 eggs were studied each week.
Growth Factors and Inhibitors Serum from a pool of normal 150-g rats was added at I %, 5%. or 10% vol/vol as specified in “Results.” T, at 0. I5 nmol, I .5 nmol, or 15 nmol was studied with or without normal serum or CSl. Serum
of Endocrinolo~v, of Texas M. D. Anderson
From the Section University
Department Hospital
of Medicine.
the
and Tumor Institute
at Houston. Supported by the American Diabetes Association Feasibility Grant and the Nancy Carmichael Gift Fund. Address reprint requests to Rena Vassilopoulou-Sellin. MD, Section of Endocrinology, Department of Medicine, the University of Texas M. D. Anderson Hospital and Tumor Institute at Houston, 6723 Bertner Ave. Houston, TX 77030. 8 1987 by Grune & Stratton. Inc. 0026-0495/87/3601-0016$03.00/0
89
90
VASSILOPOULOU-SELLIN
from a pool of hypophysectomized 80-g rats was used at 10% (vol/vol) as a GH-deficient serum control. CSI was prepared as previously described.6 Briefly, weighed, frozen rat livers were homogenized with 4 ~010.1 mol/L ammonium acetate, pH 7.4, and centrifuged; supernates were lyophilized (8 g liver weight yield 1 g lyophilized powder). Lyophilized extract was subjected to gel filtration on Sephadex G-100 eluted with 0.1 mol ammonium acetate, pH 7.4; the fraction previously tested with the rat cartilage bioassay and shown to be enriched in cartilage sulfation inhibitor Kav 0.06 to 0.18 was collected and lyophilized, thus yielding a partially purified CSI preparation. This CSI fraction is similar in molecular weight to the predominant inhibitor-enriched fraction of diabetic rat serum6 and of inhibitor released by the liver of diabetic rats during isolated liver perfusion (unpublished observation). The activity of this material on the hypox rat costal cartilage bioassay has been detailed previously.* Recalibrated for the present series of experiments, each I mg lyophilized powder contains 0.3 mg protein and produces a 50% reduction of basal cartilage sulfation in the bioassay, well within the steep portion of the bioassay doseresponse curve. For the cartilage incubations, 20 mg CSI powder/l mL culture medium was prepared and added at 10% vol/vol (ie, 2 mg/mL), 5% (ie, I mg/mL), or 1% (ie, 0.2 mg/mL) as specified below. During the course of these studies, we found that growth stimulation induced by normal serum was similar at l%, 5%, or 10% serum
Medium
6
Change
1
+1.0
i!l FJ
ET AL
-cs
I
1
I
I
1
0
10%
--Buffer . . . . ... . .. NRS 10%
I
I
2
5
Time (Days) * Reversibility of CSI effect. Cartilages were placed in Fig 2. fresh. different media at T,, (medium change). All media were at 10% vol/vol as in Fig 1. N = 4 for 8UF - NRS and NRS - 8UF; n = 13 for CSI - NRS and NRS - CSI.
+2.0
HYPOX
Buffer
0
cs NRS+CS
concentration and that growth inhibition induced by CSI was similar at 5% or 10% CSI concentration. To avoid duplication, the data in Fig 1 depict growth changes for cartilages incubated with 10% CSI and sera. The data in Figs l-4 are presented as change of wet weight in absolute milligrams. There was no loss of potency for incubated sera or CSI during the in vitro incubation periods, as ascertained by measuring sulfate uptake induced by these media with the hypox rat costal cartilage bioassay. At T,, some cartilages were homogenized in 0.9% NaCl and 0.2% Triton X-100 as described by Burch and Lebovitzy and protein content was measured according to the method of Lowry.” Other cartilages were dehydrated with ETOH and ether, weighed and dry:wet weight ratio determined. Materials
-2.0
I
0
I
I
I
I
2
I
5
Waymouth’s MB 752/l medium was purchased from Grand Island Biological Co (Grand Island, NY). Other chemicals were from Sigma Chemical Co (St Louis). CSI was prepared from liver extracts of Sprague-Dawley rats from Charles River Laboratories (Wilmington, Mass). Fertilized eggs were obtained from Richglo (El Campo, TX).
Time (Days) RESULTS Fig 1. Embryonic chicken pelvic rudiment growth in vitro. Effect on weight by normal rat serum at 10% vol/vol (NRSL hypophysectomized GH-deficient rat serum at 10% vol/vol IHRS), Waymouth’s culture buffer alone (BUF). CSI at 10% vol/vol fie, 2 mg/mL as described in “Materials and Methods”). or NRS 10% plus CSI 10% INRS + CSI). Values are given as mean + SEM fn = 8 to 13 cartilages each medium). lP < 0.05 Y initial weight.
Pelvic Cartilage Growth in Vitro Normal rat serum stimulated a significant increase in pelvic cartilage weight (t 1.33 + 0.16 mg T’s_, and + 1.33 f 0.27 mg Ta.s. Incubation with serum from hypophysectomized GH-deficient rats resulted in a small weight gain,
CSI INHIBITS
91
CARTILAGE GROWH
l-
NRS
J
k
+3.0
?I s lz a
NRS+T3
1SnM
NRS+T3
0.15&l
T3
15nM
T3
1.5nM
+1.0
5
T3 0.15nM
E .-cn s
3cp< .05 vs NRS
-1.0
A p -z .05 vs NRS+CS
I
I
I
0
2
5
Time (Days) Fig 4. CSI effect on T,-stimulated cartilage growth. Cartilages were incubated with T,, 0.15 nmol, or 1.5 nmol, with and without NRS at 10% vol/vol. Also, cartilages were incubated with T,, 1.5 nmol. or 15 nmol, with and without CSI at 10% vol/vol (ie, 2 mg/mL). N = 6 to 9 cartilages in each group. P 4 0.05 for T, v CSI + T, for each T, concentration at all time points.
0
2
5
Time (Days) Fig 3. Dose-dependent inhibition of cartilage growth by increasing concentrations of added CSI. Normal rat serum data at 1% and 5% vol/vol behaved identically and were therefore combined (NRS. n = 14). Top panel, CSI at 1% vollvol (0.2 mg/mL prepared as in “Materials and Methods”) with or without NRS (n = 16). Bottom panel, CSI at 5% vol/vol Il.0 mg/mL as prepared in “Materials and Methods”) with or without NRS (n = 13). lP < 0.05 Y NRS alone: AP < 0.05 Y NRS CSI.
+0.68 t 0.14 mg at T,., only. In contrast, a significant decrease in weight was observed in cartilages incubated with CSl (-1.09 t 0.17 mg To_* and -0.44 + 0.06 mg T2_+ Further, adding CSI to cartilage incubations containing normal serum abolished the growth-stimulating effect of the serum (-0.79 + 0.07 mg To_* and -0.40 * 0.08 mg T,,). Although some growth occasionally occurred in pelvic cartilages incubated with tissue culture buffer alone, overail, in this group of experiments no significant weight increase was demonstrated over the five-day period. Over the five-day incubation period (Table l), exposure to normal serum was associated with a 40% increase in weight and a +0.17 i 0.01 cm increase in length (approximately 25% of initial length). In contrast, exposure to CSI was associated with a 25% decrease in weight (5.89 k 0.47 mg to 4.44 t 0.40 mg T,,) with no change in length. Exposure to buffer alone produced no change in either length or weight.
Protein concentration after five days incubation was comparable among the three groups. At TM, before incubation, pelvic cartilage dry weight was 13 k 2.9 x 10e2 mg/mg wet weight. After five days of incubation, the dry:wet weight ratio was comparable among all groups. This indicates that changes in wet weight are not due to tissue swelling but accurately reflect growth. Because cartilage dehydration precludes further incubation and because the dry:wet weight ratio is stable among groups, wet weights were generally measured in the different incubation conditions. Reversibility of CSI Effect We studied the reversibility of the effects of CSI and serum by examining the responsiveness of the cartilages to changes in the incubation medium. Exposure to normal serum after incubation with buffer alone resulted in a resumption of growth (-0.10 + 0.18 mg T,,., in buffer, followed by +.66 2 .I 3 mg T,, in normal serum). Of particular interest, cartilage growth inhibition due to incubation with CSI (-0.62 2 0.11 mg TO_2) was reversed by subsequent exposure to normal serum (t1.02 _+0.21 mg T,,). In the reverse experiment, growth stimulation by normal serum (+ 1.30 + 0.14 mg T,,) was comparably reversed by subsequent exposure to CSI ( -0.61 t 0.10 mg TM). To explore
further
the tissue
or reversibility,
we also
92
VASSILOPOULOU-SELLIN Table 1. Growth Parameters Parameter
of Chick Pelvic Cartilages
Normal Serum
Length change (O-5 d) (cm)
+0.17
+ 0.01 (11)
Wet weight (0 d) (mg)
5.71
? 0.23 (12)
Wet weight (5 d) (mg)
8.0 & 0.42 (12)
Dry/wet weight ratio (5 d)
0.1 1 & 0.044 (12)
Protein concentration* (g/me wet wt)
5.38
+ 1.91 (17)
ET AL
CSI
-0.02
Buffer
* 0.01
+0.07
(ll)t 5.89 ? 0.47
-
+ 0.02 (lo)t
5.42
(12)t 4.44 f 0.40
f 0.40 (9)t
5.55
+ 0.46
(12) 0.12 i 0.045
0.11
(15) 4.84 ? 0.67
(10) 6.34 ? 1.14
19) 5 0.042
(4)
(14)
Values in parentheses refer to the number of cartilages. *Concentration in cartilage homogenates after five days incubation. tP < 0.05 \I normal serum (mean t SEM).
performed more prolonged cartilage incubation with two medium changes (Table 2). Cartilages were incubated for seven days and media were changed at three and five days. Growth of cartilages exposed to normal serum To.,, then CSI T,.,, then normal serum again TS_? was compared with growth of cartilages exposed to CSI T,, then normal serum T3_*then CSI again TS_,. Incubation with normal serum stimulated cartilage growth each time and incubation with CSI inhibited growth each time in a pattern comparable to Fig 2. Dose-Dependent CSI Effect on Cartilage Growth
To delineate the dose-dependent action of CSI, we conducted dose response studies (Fig 3). Cartilage incubated with 1% CSI did not exhibit growth inhibition until TSd (-0.68 + 0.15 mg), although its weight change was significantly less than that produced by normal serum both at TM and TSd. The growth of cartilage incubated with a combination of serum and 1% CSI was intermediate and significantly less than that induced by serum alone. When the concentration of CSI was higher (5% vol/vol), growth inhibition was significant and more pronounced at both T,, and T,,. Furthermore, when cartilages were exposed to the combination of serum and 5% CSI, the CSI effect predominated and growth was inhibited overall (-0.53 * 0.10 mg T,,_2and -0.40 + 0.05 mg T2_5). CSI Effect of T,-Stimulated Cartilage Growth Triodothyronine has previously been shown to stimulate cartilage growth”-“; the mechanism of action is believed to differ from that of somatomedins. We were, therefore, interested to determine whether CSI opposes somatomedinstimulated growth selectively or whether it can also inhibit T,-stimulated growth. Table 2. Reversibility of CSI Action TO= T3d (Awt = mg) CSI -0.60 NRS +0.98
t 0.25 f 0.30
T, = TM (Awt = mg) NRS +1.29 + 0.32 CSI -0.93 + 0.26
T5= T,, IAwt - mg) CSI -0.61 NRS +0.82
Values are given as mean + SE for all Awt. NRS and CSI are given at 10% concentration each. N refers to the number of cartilages.
+ 0.20 * 0.27
N 10 7
As expected, T, alone stimulated cartilage growth in a dose-dependent fashion at concentrations between 0.15 nmol and 15 nmol. Interestingly, the combination of normal serum and T, produced apparently additive growth stimulation (ie, during the first two days of incubation (T&, cartilage weight increased by + 1.80 * 0.26 mg in the presence of NRS + T, 0.15 nmol relative to +0.32 + 0.06 mg in the presence of Tj 0.15 nmol alone; during the subsequent three days of incubation (T,.,), cartilage weight increased further by + 1.95 * 0.23 mg in the presence of NRS + Ts 0.15 nmol relative to +0.32 + 0.21 mg in the presence of T3 0.15 nmol alone). The CSI-induced cartilage growth inhibition was dissociated from somatomedin action in these studies. Cartilages exposed to CSI plus T, at either 1.5 nmol or 15 nmol exhibited significant growth inhibition throughout the incubation period (ie, during the first two days of incubation (Ta2), cartilage weight increased by +0.80 k 0.10 mg in the presence of T, 1.5 nmol while cartilage weight decreased by -0.87 + 0.17 mg in the presence of combined CSI + T, 1.5 nmol; during the subsequent three days of incubation (TZ_5)r cartilage weight increased further by +0.95 _t 0.34 mg in the presence of T, 1.5 nmol while cartilage weight decreased by -0.16 k 0.04 mg in the presence of CSI + T3 1.5 nmol). In a separate experiment, in cartilages incubated with combined normal serum T3 and CSI, growth inhibition was also seen (ie, -0.33 * 0.07 mg at T,, and +0.04 _t 0.14 mg at T2_5for combined media, n = 4). This finding suggests that CSI has broad antianabolic action and the previously used term “somatomedin inhibitor” is too restrictive. DISCUSSION
Despite accumulating knowledge of the factors that regulate cartilage and skeletal growth, information about naturally occuring brakes on cartilage metabolism remains sparse. The somatomedin inhibitors have been described by several investigators as factors obtained from sera or liver preparations that inhibit somatomedin-stimulated metabolism of cartilage,‘-’ although they may also inhibit directly basai cartilage, muscle, and adipose tissue metabolism.2~5~‘*It has been suggested that poor growth in children with starvation, uncontrolled diabetes, or uremia is due in part to increased levels of circulating somatomedin inhibitors. How-
CSI INHIBITS
93
CARTILAGE GROWTH
ever, it has been very difficult to prove that these factors actually inhibit skeletal growth. In addition, most inhibitor bioassays express the presence of inhibitor relative to somatomedin-stimulated cartilage sulfation; there still remains some doubt in many investigators’ minds whether inhibitors may be somatomedin-binding proteins, which somehow interfere with bioassays of somatomedin activity. Finally, the inability to demonstrate that cartilage sulfation inhibition by these factors can be reversed has created the concern that inhibitors are nonspecific toxins rather than metabolically regulated factors. We felt that resolution of these important issues was not possible because of technical limitations of existing bioassays. In the present study we used a new approach to measure the biologic action of a partially purified high molecular weight (MW) cartilage sulfation/somatomedin inhibitor from rat liver. With the embryonic chicken pelvic rudiment bioassay, we were able to show reversible, dose-dependent inhibition of serum-stimulated and thyroid hormone-stimulated cartilage growth. Recent studies indicate that somatomedin-inhibitory activity in the sera of diabetic,16 starved,’ and uremic rats” consists of several MW components. After gel filtration with Sephadex G-100, pH 7.4, we have also found small MW inhibitors in rat liver extracts and diabetic rat sera6 and in rat liver perfusates (unpublished observation). However, we find a predominant fraction of cartilage sulfation/somatomedin inhibitors at MW >50,000. and it is this inhibitor fraction that we describe in this report. Embryonic chicken cartilage has been used extensively to elucidate the hormonal control of growth stimulation.g-‘s We chose the embryonic chicken pelvic rudiment prolonged organ culture system because it allows the study of different parameters including measurements of actual growth of a growing skeletal organ. It is, therefore, a simple model in which to study the effect of CSI on overall cartilage growth. Although in this study incubation with buffer alone did not support cartilage growth, the stimulation of cartilage
growth by T, and normal serum (in terms of both weight and length increments) is qualitatively consistent with previous reports for this experimental mode1.9*‘4Our data provide the first demonstration that CSI produces inhibition of cartilage growth in a consistent, dose-dependent fashion. The cartilage length did not change, and the rudiments appeared sturdy and viable grossly and under light microscopy (not shown). The change in weight was unaccompanied by changes in dry:wet weight ratio or protein concentration, suggesting that the effect was not due simply to altered hydration of the cartilages. Of critical importance, growth inhibition by CSI is reversible. In previous studies of somatomedin inhibitors using cartilage bioassays,2*4,6 the methodology was such that reversibility could not be tested. The present experiments demonstrate for the first time that cartilage exposed to CSl can recover and resume growth when presented with a stimulatory medium. Serum-stimulated growth was comparable whether the cartilage had been incubated previously with CSI (Fig 2, Table 2) or with normal serum (Fig I). The reversibility of CSI action strongly supports the view that this factor is not simply a tissue toxin. Thyroid hormone has previously been shown to be critically important for normal skeletal development. Researchers believe that the mechanism of growth stimulation by thyroid homone differs from that by somatomedins.‘0’20 CSI clearly inhibits T,-stimulated cartilage growth at all incubation times. The demonstration that CSI blocks TX-stimulated growth eliminates the possibility that it represents a somatomedin-binding protein. Further, the inhibition of growth stimulation by both somatomedins in normal serum and T, alone supports the view that CSl has broad antianabolic tissue action.
ACKNOWLEDGMENT
We thank Rosa M. Quiroz for her expert assistance preparation
with the
of the manuscript.
REFERENCES
I. Bomboy JD Jr, Burkhalter VJ. Nicholson
WE, et al: Similarity of somatomedin inhibitor in sera from starved, hypophysectomized, and diabetic rats: Distinction from a heat-stable inhibitor of rat cartilage metabolism. Endocrinology 112:371-377, 1983 2. Phillips LS, Vassilopoulou-Sellin R, Reichard LA: Nutrition and somatomedin. VIII. The “somatomedin inhibitor” in diabetic rat serum is a general inhibitor” in diabetic rat serum is a general inhibitor of growing cartilage. Diabetes 28:919-924, 1979 3. Phillips LS, Belosky DC, Reichard LA: Nutrition and somatomedin. VI. Somatomedin activity and somatomedin inhibitory activity in sera from normal and diabetic rats. Endocrinology 104:1519~1524, 1979 4. Binoux M, Lassarre C, Seurin D: Somatomedin production by rat liver in organ culture II. Studies of cartilage sulfation inhibitors released by the liver and their separation from somatomedins. Acta Endocrinol 93:83-90, 1980 5. Vassilopoulou-Sellin R, Phillips LS, Reichard LA: Nutrition and somatomedin. Vll. Regulation of somatomedin activity by the perfused rat liver. Endocrinology 106:260-267, 1980 6. Vassilopoulou-Sellin
R, Oyedeji
CO, Samaan
NA: Extraction
of cartilage sulfation inhibitors and somatomedins from rat liver. Endocrinology 114:576-581, 1984 7. Salmon WD Jr, Holladay LA, Burkhalter VJ: Partial characterization of somatomedin inhibitor in starved rat serum. Endocrinology I 12:360-370,1983 8. Phillips LS, Belosky DC, Reichard LA: Nutrition and somatomedin. V. Action and measurement of somatomedin inhibitor(s) in diabetic rat serum. Endocrinology 104: 15 13-l 5 18, 1979 9. Burch WM, Lebovitz HE: Triiodothyronine stimulation of in vitro growth and maturation of embryonic chick cartilage. Endocrinology 111:462-468,1982 IO. Kemp SF, Mutchnick M, Hintz RL: Hormonal control of protein synthesis in chick chrondrocytes: A comparison of effects of insulin, somatomedin C and triiodothyronine. Acta Endocrinol 107:179-184, 1984 Il. Audhya TK, Segen BJ, Gibson KD: Stimulation of proteoglycan synthesis in chick embryo sternum by serum and L-3,5,3’triiodothyronine. J Biol Chem 251:3763-3767, 1976 12. Hall BK: Thyroxine and the development of the tibia in the embryonic chick. Anat Rev 176:49, 1973 13. Garland JT, Jennings J, Levitsky LL, et al: Stimulation of
94
DNA synthesis in isolated chondrocytes by somatomedin. II. Validation of the assay for clinical use and comparison with the stimulation of protein synthesis. J Clin Endocrinol Metab 43:847-85 1, 1976 14. Hall K: Quantitative determination of the sulphation factor activity in human serum. Acta Endocrinol63:338-350, 1970 15. Daughaday WH, Phillips LS, Herington AC: Measurement of somatomedins by cartilage in vitro. Methods Enzymol 37B:93109.1975 16. Phillips LS, Bajaj VR, Fusco AC, et al: Nutrition and somatomedin. XI. Studies of somatomedin inhibitors in rats with streptozotocin-induced diabetes. Diabetes 32: 1117-I 125, 1983
VASSILOPOULOU-SELLIN
ET AL
17. Lowry OH, Rosebrough NJ, Farr AL, et al: Protein measurement with the folin phenol reagent. J Biol Chem 193:265-274, 1951 18. Phillips LS, Scholz TD: Nutrition and somatomedin. IX. Blunting of insulin-like activity by inhibitor in diabetic rat serum. Diabetes 31:97-104, 1982 19. Phillips LS, Kopple JD: Circulating somatomedin activity and sulfate levels in adults with normal and impaired kidney function. Metabolism 30:1091-1095, 1981 20. Burch WM, Lebovitz HE: In vitro stimulation of alkaline phosphatase activity in immature embryonic chick pelvic cartilages by adenosine 3’,5’-monophosphate. J Cell Biol 93:338-342, 1982