Growth Hormone and Insulin Binding to Human Liver D. CARR* AND H. G. FRIESEN Department of Physiology, Faculty of Medicine, University of Manitoba, Winnipeg, Canada ABSTRACT. Specific binding of 125I-hGH to human liver was found in autopsy specimens from 12 of 15 patients. Specific binding was studied using a new technique employing 20 fx. thick 'microslices' cut on a cryostat. About 0.5 to 1 mg of tissue per assay tube makes feasible the study of small samples. The range of specific binding was 1.4 to 11.7% of 80,000 cpm 125I-hGH added expressed per mg dry weight of tissue. Specific binding was also demonstrable in homogenates and crude membrane preparations from liver. No correlation was seen between 125 I-hGH binding and age, sex, or pathology in

T

HE RECOGNITION that rabbit liver contains receptors for growth hormone (1) is of interest partly because of the corollary that liver may be a primary target tissue of growth hormone and also because rabbit liver has been used successfully as the tissue source in a radioreceptor assay (RRA) for growth hormone. The rabbit liver receptors do not distinguish between growth hormones from several mammalian species and, in this respect, they differ from the receptors found in certain strains or cultured human leukocytes (2,3) which will bind human growth hormone (hGH) but not nonprimate growth hormones. These human cell lines have been used in a radioreceptor assay specific for human growth hormone, but this assay requires the availability of the specific cell lines which have been found to possess the receptors and also the specialized facilities for maintaining them in culture. An assay with similar specificity, but using an alternative tissue, might be useful and, in view of a previous report of production of 'sulfation factor' by liver perfused with growth hormone (4), it seemed natural to test human liver, obtained at

Manitoba,

the series of patients studied. No specific binding of 125 I-hGH was observed in lung, adrenal, spleen, or kidney, although all the tissues bound 125I-insulin. Of several species of growth hormone tested, only primate GH displaced l25I-hGH from its binding sites in human liver. No displacement was seen with ovine or human prolactin or with insulin. Primate placental lactogens had only 0.5-1.0% potency of native hGH in displacing 125I-hGH from human liver. Ungulate placental extracts, however, were equipotent with hGH in this respect. (J Clin Endocrinol Metab 42: 484, 1976)

autopsy, as this material is more generally available. The commonly used technique in examining a tissue for polypeptide hormone receptors is to make a preparation containing the plasma membranes by a process of homogenization followed by differential centrifugation. In practice, this needs at least 1.0 g of tissue, which is not a problem when studying autopsy specimens of liver in ample supply. Nevertheless we used this model system to investigate the use of frozen sections cut from blocks of whole tissue on a cryostat as a way of identifying the presence of receptors in small samples. Materials and Methods

Human tissues were obtained from 15 subjects at autopsy and kept at - 2 0 C until used. The cause of death and the time which elapsed before the specimens were obtained are listed in Table 1. As indicated, most of the autopsies were performed within 9 hours of death. Approximately 1.0 g of liver was homogenized with a Polytron PT-10 homogenizer for 1 minute at maximum speed in 10 ml of ice-cold Tris HC1 buffer (25 mM, pH 7.6) followed by 10 strokes in a 15 ml glass tissue grinder (Pyrex brand 7727). Aliquots of 0.1 ml of the homogenate were used for binding studies. Received April 11, 1975. The second type of binding study involved * Present address is: North Tees General Hospital, Hardwick, Stockton-on-Tees, Cleveland TS19 8PE, U.K. the use of 'microslices' of liver which were cut 484

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 18:38 For personal use only. No other uses without permission. . All rights reserved.

485

GROWTH HORMONE BINDING TO HUMAN LIVER TABLE 1. Specific binding of l25I-hGH and l25I-insulin to human livers % Specific binding of I25I-hGH Subject No.

Sex

Age

Time to autopsy1

Clinical features

Slices2

Homogenate3

% Specific binding of 125I-insulin Slices2

1

M

61

9

Myocardial infarct (MI)

2.5

2.8

10.8

2

M

79

4V2

Alcoholic cirrhosis

1.4



15.4

3

M

56

4

Metastatic carcinoma of bronchus

0

0.3

5.9

4

M

62

3y2

Asthma (5 yrs on prednisone)

0

0.9

13.5

5

F

69

2*4

Ruptured aortic aneurysm

5.3

5.3

8.3

6

M

72

3

MI

11.7

9.6

8.4

7

M

74

3

MI

4.5

3.6

9.2

8

M

63

9

MI

4.4

6.3

5.7

9

M

23

IVi

Head injury

1.4

0.3

7.4

10

M

54

6

MI

2.4

1.7

6.9

11

M

22

2

Head injury

3.1

2.0

7.8

12

F

50

5*4

MI

5.6

6.7

4.7

13

M

20

13

Multiple injuries

4.5

8.0

5.4

14

M

24

Volvulus

0

0

15

F

MI

5.4

6.8

6 weeks 68

7

11.0 —

1

Hours from time of death to removal of liver at autopsy. Expressed per mg dry weight of liver. 3 Expressed per 5 mg wet weight of liver.

2

on a cryostat. In this case, specimens which were immediately frozen after collection were not allowed to thaw prior to cutting. A block of frozen liver tissue was trimmed to an even crosssection (e.g., a cylinder 2 to 5 mm diameter made with a cork-borer) and cut at 20 \x intervals to obtain a series of consecutive slices of almost identical weight. One or more slices were placed in 10 x 75 mm glass tubes kept at - 2 0 C to maintain the slices in the frozen state until binding studies were performed. On occasion, slices from each block were also weighed on pretared 12 mm aluminum pans using an electrobalance (Cahn 4100). The slices were allowed to dry completely on the pans prior to weighing. In general, 4 slices (20 /A) with a total weight of 0.5 to 1.0 mg were added to each assay tube. Particulate fractions of the liver specimens, homogenized in 0.3M sucrose, were also prepared

using a centrifugation procedure identical to that used for isolating fractions from rabbit livers (1). The 15,000 x g and 100,000 x g pellets were resuspended and diluted in Tris HC1 buffer (25 mM, pH 7.6) to give a protein concentration of 2 mg/ml determined by the method of Lowry et al. (5). As a considerable portion of binding activity was found to be present in the 15,000 x g pellets, a higher yield particulate receptor was obtained subsequently for use in the RRA, by centrifuging at 300 x g for 10 minutes to remove debris, and then proceeding directly to 100,000 x g for 1 hour to obtain the remaining particulate material in a single fraction. Similar procedures were used to prepare 'microslices' from kidney, muscle, and adrenal, and particulate fractions of kidney, adrenal, lung, and spleen from some of the subjects studied.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 18:38 For personal use only. No other uses without permission. . All rights reserved.

Iodination of hGH IZ5

I-hGH was prepared by the lactoperoxidase method (6) using 1 mCi of Na I25I (New England Nuclear) and 5 /xg of hGH. Unreacted iodide and damaged hormone were separated from intact 125 I-hGH by gel filtration on a Sephadex G-100 column (1.5 x 50 cm) using Tris HC1 (25 mM, pH 7.6) as the buffer. The specific activity of the 125 I-hGH was 40-90 Hormones used hGH preparation HS 1648E (NIH), 2.0 IU/mg was used for iodination and standards. Other hormones used in testing the specificity of binding were as follows: monkey GH (mGH) NIH-M945A; bovine GH (bGH) NIH-B17; ovine GH (oGH) NIH 0743B; canine GH (dGH) NIH-D1001A; porcine GH (pGH) NIH-P526B; ovine prolactin (oPRL) NIH-PS10; monkey placental lactogen (mPL) 1 and 2 (7), ovine placental lactogen (oPL) (8), human placental lactogen (hPL) (9), and human prolactin (hPRL) (10) prepared in this laboratory; and porcine insulin (lot no. 1119 Connaught Laboratories Ltd.). Determination conditions

JCE & M • 1976 Vol 42 • No 3

CARR AND FRIESEN

486

of specific binding:

incubation

Incubations were carried out in 10 x 75 mm glass test tubes using an incubation volume of 0.5 ml made up with Tris HC1 buffer (25 mM, pH 7.6) containing 0.1% bovine serum albumin (BSA). All dilutions of hormone standards and iodinated hormones were made in this buffer. The Mg++ concentration was always 20 mM except in those experiments where the effect of differing Mg++ and Ca ++ concentrations were being studied. 50,000-100,000 cpm of 125I-hGH were added to each tube. Unlabelled hormones were added in a volume of 0.1 ml. Where homogenates or particulate fractions of tissues were being tested, these were added in a volume of 0.1 ml. In each experiment, the non-specific adherence of iodinated hormone to the test tube was assessed by including a number of control tubes containing iodinated hormone and buffer only. The incubations were usually carried out at 4 C for 24 hours or occasionally at 24 C for 6 hours. The reaction was stopped by diluting with 3 ml of sodium acetate buffer (25 mM, pH 5.4) at 4 C, centrifuging at 2000 x g and 4 C for 20 minutes, then decanting and draining off the

supernatant. The tubes containing the pellets were counted in an LKB automatic gamma counter. Calculation of specific binding The percentage of specific binding of l25I-hGH to the tissue was calculated from the formula: Specific binding = (CPM bound in the absence of unlabelled hGH-CPM bound in the presence of a large excess [2 /x.g/ml] of unlabelled hGH) x 100/total CPM of I25I-hGH added. The per cent of specific binding was corrected for the amount of tissue used and is quoted per mg dry weight for the microslices and per 5 mg wet weight for the crude homogenates. For particulate fractions, the specific binding was first estimated per 20 /u,g protein and then expressed per 5 mg wet weight of the original tissue. Specific binding of

i2s

I-insulin

Porcine insulin was iodinated with chloramine T to a specific activity of approximately 100 liCU/jig, as previously described (11). The specific binding of 125I-insulin was estimated in exactly the same way as for 125I-hGH, but conducting all tests at 4 C for 24 hours and using 3 ml Tris HC1 (0.025M, pH 7.6 with 0.1% BSA) to dilute the reaction mixture prior to centrifugation. The specificity of binding was tested by incubating tissue with 125I-insulin in the absence and presence of unlabelled insulin, hGH, bGH, and oPRL, each at a concentration of 2 jug/ml.

Results 1. Specific binding of 125I-hGH to human livers using microslices, homogenates and particulate fractions (Table 1) Specific binding of hGH was demonstrable in 12 of the 15 human liver samples tested. The specific binding of these 12 samples ranged from 1.4% to 11.7% with a mean value of 4.4% per mg dry weight of tissue using the microslice technique. There was good correlation between the specific binding of 125I-hGH to microslices and to homogenates (Fig. 1, r = 0.89) from indi-

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 17 November 2015. at 18:38 For personal use only. No other uses without permission. . All rights reserved.

487

GROWTH HORMONE BINDING TO HUMAN LIVER

vidual livers. Of the total 125I-hGH bound, only 50-60% could be displaced even by high concentrations of unlabelled hGH. This was observed with all the livers, tested as microslices or by the other techniques. The significance, if any, of the residual 40-50% 'non-specific' binding is unknown. Table 2 shows the variation in yields of protein in the 15,000 x g and 100,000 x g particulate fractions of 8 livers with the corresponding specific binding of hGH by these fractions expressed per 5 mg weight of tissue. The proportion of binding of the crude homogenate which is contained in the 100,000 x g fraction is variable and does not correlate as well with the binding by the microslices. There was no correlation of specific binding of 125I-hGH with age, sex, or pathology in the limited series of cases studied. 2. Specific binding of human tissues tested

125

I-hGH to other

No significant specific binding (

Growth hormone and insulin binding to human liver.

Specific binding of 125I-hGH to human liver was found in autopsy specimens from 12 to 15 patients. Specific binding was studied using a new technique ...
945KB Sizes 0 Downloads 0 Views