Vol. 168, No. 2, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

April 30. 1990 Pages 396-401

ImmnoreactiveGrowthHonmne (GH) Secretion byEIunranLyr@ocybes:

AugrssntedRe1eas.e bymenous GH

Naoki Hattori, Akira Shimatsu, Masahiko Sugita, Shunichi Kumagai,

andHirooImura

Second Division, Department of Internal Medicine,

Kyoto University Faculty of Medicine, Kyoto 606, Japan

Received February 26, 1990

Peripheral blood mononuclear cells (PBMCs) from normal adults secreted small amounts of human growth hormone (GH;0.2-0.6 pg/105 cells/7 days culture) as measured by a highly sensitive enzyme immunoassay. Stimulation of PBMCs with phytohemagglutinin (PHA) consistently showed a 4-6 fold in- crease in GH secretion. Transformed B-lymphocytes by Epstein-Barr virus also secreted GH (O-8-4.8 pg/5 x lo* cells/7 days culture). GH secreted by lymphocytes comigrated with pituitary GH on an Ultrogel AcA44 column. Addition of GH during the culture augmented endogenous GH secretion from PHA-stimulated PBMCs. GH-releasing hormone and a somatostatin analogue, SMS 201-995, did not affect GH secretion from non-stimulated and PHA- stimulated PBMCs. These findings suggest that both T and B lymphocytes secrete immunoreactive GH in a different manner from that in the anterior pituitary. o 1990 Academic Pres*, Inc.

Recent studies indicate that bidirectional communication exists be-

tween the neuroendocrine and immune systems (1). Neuroendocrine hormones

have immunoregulatory effects and cytokines have neuroendocrine effects.

It has been shown that lymphocytes produce neuroendocrine peptides such as

ACTH, endorphin (2), TSH (3) and prolactin (4). Prolactin/growth hormone

(GH)-related mRNA species were also detected in mitogen-stimulated lym-

phocytes (5). Very recently, Weigent et al. (6) have provided evidence

that mononuclear leukocytes can synthesize and secrete GH in vitro. GH

produced by lymphocytes is similar to pituitary GH in terms of

bioactivity, antigenecity, and molecular weight. In the present study, we

describe the quantitative analysis of GH secretion by T and B lymphocytes

using a highly sensitive enzyme immunoassay (EIA) for GH (7).

Venous blood was taken after an overnight fast from normal adult volunteers. Plasma GH levels determined by EIA at the time of blood withdrawal ranged from 0.18-l-5 ug/L. Peripheral blood mononuclear cells (PBMCS) were isolated by standard Ficoll-Hypaque gradient centrifugation (8) and suspended in culture medium (RPM1 1640) containing 10% fetal calf serum (Filtron, Australia) and antibiotics. PBMCs (1 x 10'cells in 200 pL)

ooo6-291x.90 $1.50 Copyright 6 1990 by Academic Press, Inc. All rights of reproduction in any form reserved. 396

Vol. 168, No. 2, 1990 BlOCHEMlCAL AND BIOPHYSKAL RESEARCH COMMUNICATIONS

were incubated in triplicate at 37OC in 5% COz with the following additives: 0.2% phytohemagglutinin (PHA;Difco, Detroit, MI), 1% pokeweed mitogen (PWM;Sigma, St. Louis, MO), 20 mg/L lipopolysaccharide (LPS; Difco, Detroit, MI), human pituitary GH (Sumitomo Pharmaceuticals, Tokyo, Japan), GH-releasing hormone (GHRH;Peptide Institute, Osaka, Japan), a somatostatin analogue (SMS 201-995;Sandos Pharmaceuticals, Bazel, Switzerland). After incubation, the cells were removed by centrifugation and the culture supernatant was used for the assay. Epstein Barr (EB) virus-transformed B cell lines were established from normal subjects, as previously described (9).

Gel chromatography The culture supernatant of PHA-stimulated T cells or EB virus-

transformed B cells was concentrated lo-30 fold with polyethylene glycol. The concentrated samples (0.5-1.0 mL) or GH standards in culture medium were applied onto an Ultrogel AcA44 column (0.9 x 70 cm) and eluted with 0.01 mol,/L phosphate buffer containing 0.1 mol/L NaCl, 0.1% NaNa, and 0.1% bovine serum albumin at 4'C. The fractions of 0.8 mL were collected and assayed for GH. The column was calibrated by markers with various molecular weights.

Assays GH was measured by a highly sensitive enzyme immunoassay (EIA) as pre-

viously described (6). In brief, samples (100 pL) were incubated with anti-GH IgG-coated polystyrene balls at 37OC for 6h. After washing with saline, the polystyrene balls were incubated with anti-GH Fab' peroxidase conjugate at 4 'C or 16h and at 20°C for 6h. After washing with saline, peroxidase activity bound to the balls was assayed using a fluorescence method. A spectrofluorophotometer (Shimadzu RF-540, Kyoto, Japan) was used to measure the fluorescence intensity. The detection limit was 1 x lO"ug/L. The intra- and inter-assay coefficients of variation were 6.0% and 8.6%, respectively.

Statistical analysis The data were expressed as means?SEM. Student's t-test and one way

analysis of variance in combination with Duncan's new multiple range test were used for statistical analysis as appropriate.

RESULTS

PBMCs (1 x 10' cells/well) cultured for 7 days secreted small amounts

of immunoreactive GH (0.2-0.6 pg/well) into the media, as shown in Fig.1.

Stimulation of PBMCs with PHA (0.0125-0.2%), a T-cell mitogen, resulted in

a dose-related increase in GH secretion. PWM cl%), a B-cell mitogen, also

increased GH secretion about two fold, while addition of LPS (20 mg/L) did

not stimulate GH secretion. EB virus transformed B cell lines (5 x 10'cells

/well) cultured for 7 days secreted variable amounts of GH (0.8-4.8

pg/well).

Fig.2a shows the time-course of GH secretion from non-stimulated and

PHA-stimulated lymphocytes. GH was not detected immediately after the

incubation. Small amounts of GH (0.2-0.6 pg/well) were secreted after the

incubation for 6h and remained unchanged during 9 days of culture without

PHA. GH secretion from PBMCs was stimulated by PHA after the 3rd to 5th

day and continued to increase till the 9th day. As shown in Fig.Zb, PHA-

stimulated GH secretion was dependent on the number of cultured cells.

397

Vol. 168, No. 2, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

stimulated PHA PWM LPS EB-V transformed Time B cell lines

xl05cellshel

Figure 1. Secretion of immunoreactive GH from non-stimulated, mitogens (PHA, PWM, LPSJ-stimulated lymphocytes, and three EB virus-transformed B cell lines during the culture for 7 days.

Figure 2. (a) Time course of GH secretion from non-stimulated (triangle) and PHA-stimulated (circle) lymphocytes in 2 normal subjects. The number of cells was 1 x lo5 /well. (b) The relationship of GH secretion from PHA- stimulated lymphocytes with the cell numbers in the 7 days culture. Mean?SE of GH levels in triplicate wells are shown.

To investigate whether GH secretion was regulated, exogenous GH (lo-

100 x lo-' pq/L) was added in the culture media and incubated for I days

with PBMCs. GH was stable in media without cells (mean recovery:99.3%). We

calculated the amounts of secreted GH by subtracting the amounts of added

GH from the recovered GH levels. The addition of GH significantly in-

creased GH secretion from PHA-stimulated lymphocytes in a dose-related

manner (Table 1). The cell numbers were not changed by the addition of

exogenous GR. GHRH (lo- 'O-10*' M) and/or SMS 201-995 (10“ -10s6 M) did not

affect the GH secretion from non-stimulated and PHA-stimulated lymphocytes

(data not shown).

A dilution curve of GH produced by lymphocytes was parallel with the

standard curve of GH (Fiq.3). Immunoreactive GH was mainly eluted at the

position of 22K GH on gel chromatography (Fiq.4). Minor portions of big

forms were also observed as in the case of pituitary GH.

Table 1. Effects of exogenous GH on GH secretion from lymphocytes

Added GH (~lO-~pg/L) Secreted GH (pg/well) non-stimulated PHA-stimulated

0 0.2 kO.1 2.4t0.3

10 0.3 + 0.4 2.6tO.3 50 0.0 to.9 5.2+0.8*

100 0.6 20.7 9.lf1.2Df*

Values show mean rSE of secreted GH in sextuplicate. * PCO.02 vs. secreted GH from PHA-stimulated lymphocytes

in the absence of exogenous GH. ** PCO.01 vs. secreted GH from PHA-stimulated lymphocytes

in the absence of exogenous GH. ***p<o.o5 vs. secreted GH from PHA-stimulated lymphocytes

in the presence of exogenous GH (50 x lO"ug/L).

398

Vol. 168, No. 2, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

+ 1 10 100 loo0 GH (x~O-~ pg/L) 0 4

Pituitary I

T-CM 3

Fraction Number

Figure 3. Dilution curve of immunoreactive GH secreted by lymphocytes (0) and standard curve of GH (0 > by EIA.

Figure 4. Gel filtration profiles of standard pituitary GH (left column), immunoreactive GH secreted by PHA-stimulated T cells (middle column) and EB virus-transformed B cells (right column) on Ultrogel AcA 44. Arrows in- dicate the position of monomeric GH.

DISCUSSION

In the present study, we have confirmed the recent observation by

Weigent et al. (6) that GH was secreted from lymphocytes in vitro, and ex-

tended it in a quantitative manner using a highly sensitive enzyme im-

munoassay for GH. Moreover, we demonstrated that GH secretion from T- and

B-lymphocytes can be induced by mitogens (PHA or PWM). These findings are

in good agreement with the previous report that prolactin/GH related mRNA

species were induced in concanavalin A-stimulated lymphocytes (5).

Small amounts of GH was released by non-stimulated lymphocytes after

the incubation for 6h. The GH released in the culture media was not a con-

taminant from serum GH, since it was not detected immediately after the

incubation. Kiess et al. (10) reported that GH bound to the receptors on

lymphocytes dissociated within several hours. It is possible, therefore,

that membrane-associated or internalized GH is released from lymphocytes

during the short incubation. Another explanation is that lymphocytes in-

itiate the production of GH immediately after the incubation, since GH

mBNA was detected in non-stimulated lymphocytes 2h after the incubation

(6).

Weigent et al. reported that more than 50% of radioactivities of GH-

associated labelled amino acids were eluted as high molecular weight forms

(>300,000) which were converted to a lower molecular weight form after

reduction. We found that lymphocyte-derived GH was mainly eluted at the

399

Vol. 168, No. 2, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

position of monomeric GH (22,000). Minor portions of big forms of GH were

observed but no apparent peak of GH-GH binding protein complex as seen in

plasma (11) or urine (12) was present in the conditioned media.

The present study showed that GH secretion from lymphocytes was aug-

mented by exogenous GH administration in a dose-related manner. GH can act

to inhibit its own secretion from the anterior pituitary (13). GHRH and

somatostatin, well known regulators of pituitary GH secretion, did not af-

fect GH release from the lymphocytes. These findings suggest that the

regulation of GH secretion from lymphocytes is different from that in the

pituitary gland.

The mechanism and significance of augmented GH release by GH remains

to be determined. Although GH acutely causes down-regulation of GH recep-

tors (14,15), continuous exposure of GH resulted in restoration of the

receptors on IM-9 lymphocytes (14) and even in up-regulation of liver GH

receptors (16). It is possible that GH secreted by lymphocytes may act on

lymphocytes as an autocrine/paracrine growth factor, since GH induces lym-

phocyte proliferation in vitro (17) and modulates mitogen-induced

proliferation (18). Further studies are required to clarify the

physiological significance of GH production by lymphocytes.

This work was supported in part by grants from the Ministry of Education, Science, Culture, the Ministry of Health and Welfare, the Foun- dation for Growth science in Japan, and by Nordisk HGH Research Award.

1. Blalock, J.E. (1989) Physiol. Rev. 69, l-32. 2. Smith, E.M. and Blalock, J.E. (1981) Proc. Natl. Acad. Sci. USA. 78,

7530-7533. 3. Smith, E.M., Phan. M., Kruger, T.E., Coppenhaver, D.H. and Blalock,

J.E. (1983) Proc. Natl. Acad. Sci. USA. 80, 6010-6013. 4. Montgomery, D.W., Zukoski, C.F., Shah, G.N., Buckley, A-R., Pacholczyk,

T. and Russell, D.H. (1987) Biochem. Biophys. Res. Corn. 145, 692-698. 5. Hiestand, P-C., Mekler, P., Nordmann, R., Grieder, A. and Permmongkol,

C. (1986) Proc. Natl. Acad. Sci. USA. 83, 2599-2603. 6. Weigent, D.A., Baxter, J-B., Wear, W-E., Smith, L-R., Bost, K.L. and

Blalock, J.E. (1988) FASEB J. 2, 2812-2818. 7. Hattori, N., Kato, Y., Murakami, Y., Hashida, S., Ishikawa, E., Mohri,

Z. and Imura H. (1988) J. Clin. Endocrinol. Metab. 66, 727-732. 8. Umehara, H., Kumagai, S., Ishida, H., Suginoshita, T., Maeda, M. and

Imura H. (1988) Arthritis Rheum. 31, 401-407. 9. Pope, J.H. (1979) In The Epstein-Barr virus (M.A. Epstein and B.G.

Achong, Eds.), pp. 205-230. Springer-Verlag, Berlin. 10. Kiess, W. and Butenandt, 0. (1985) J. Clin. Endocrinol. Metab. 60,

740-746. 11. Leung, D-W., Spencer, S-A., Cachianes, G., Hammonds, R-G., Collins,

C ., Henzel, W-J., Barnard, R., Waters, M.J. and Wood, W-1. (1987) Nature 330, 537-543.

400

Vol. 168, No. 2, 1990 BIOCHEMICAL AND BIOPHYSICAL RESEARCH COMMUNICATIONS

12. Hattori, N., Shimatsu, A., Kato, Y. and Imura, H. (1990) Kidney Int. in press.

13. Abe, H., Molitch, M.E., Van Wyk, J.J. and Underwood, L.E. (1983) Endocrinology 113, 1319-1324.

14. Lesniak, M.A. and Roth, J. (1976) J. Biol. Chem. 251, 3720-3729. 15. Messina, J.L., Eden, S. and Kostyo, J.L. (1985) Am. J. Physiol. 249,

E56-E62. 16. Baxter, R.C., Zaltsman, 2. and Turtle, J.R. (1984) Endocrinology 114,

1893-1901. 17. Astaldi, A., Yalcin, B.. Meardi, G., Burgio, G.R., Merolla, R. and

Astaldi, G. (1973) Blut 26, 74-81. 18. Kiess, W., Holtmann, H., Butenandt, 0. and Eife, R. (1983) Eur. J.

Pediatr. 140, 47-50.

401