Download Growth Hormone Synthesized and Secreted by Human Thymocytes

Vol
Printed
Juurnal
of Chmcal
Endocrinology
and Metahohsm
Copyright
0 1996 by The Endocrine
Society
81, No 7
tr, U.S.A
Growth
Hormone
Synthesized
and Secreted by Human
Thymocytes
Acts via Insulin-Like
Growth
Factor I as
an Autocrine
and Paracrine
Growth
Factor*
PARAMJEET
Department
SABHARWAL
of Medicine,
AND
SUPRIYA
Ohio State University,
VARMA
Columbus,
Ohio 43210
ABSTRACT
There is increasing
evidence
that GH can influence
immune
function and that it is secreted
by lymphocytes.
In the present
study we
investigated
the endogenous
synthesis
and secretion
of GH and insulin-like
growth
factor I (IGF-I)
from human
thymocytes
and evaluated the autocrine/paracrine
effects of GH and IGF-I
on T cell and
thymic
epithelial
cell proliferation.
First, the presence
of thymic
GH
and IGF-I
was detected
by RIA of thymocyte
extracts.
Next, using a
hormonal
enzyme-linked
immunoplaque
assay, we found that thymocytes
secreted GH and IGF-I.
Further,
we documented
the endogenous synthesis
of GH by human
thymocytes
using [35Slmethionine
labeling
followed
by immunoprecipitation,
gel electrophoresis,
and
autoradiography.
We then evaluated
the physiological
role of endogenously
generated
GH and IGF-I.
Using an affinity-purified
GH
polyclonal
antibody,
we observed
a marked
inhibition
(P < 0.04) of
phytohemagglutinin-stimulated
thymocyte
proliferation,
suggesting
an autocrineiparacrine
role for the secreted GH. Further,
we observed
I
N 1930, SMITH (1) reported that hypophysectomized
mice
had atrophic thymus glands. In recent years, reports have
appeared
suggesting
that GH has numerous
effects on the
immune system, including
the thymus (2). For example, GH
reverses thymic atrophy and restores T cell-dependent
functions in aged rats (3) and mice (4). GH also stimulates T cell
proliferation
(5) and c-myc expression
(6) and promotes maturation of rat thymocytes
(7).
GH may act on the immune
system by stimulating
IGF-I
action. In human lymphocyte
cell lines (5) and rat leukocytes
(8, 9), GH can stimulate
insulin-like
growth
factor I (IGF-I)
and/or
IGF-II production.
In addition,
GH and IGF-I stimulate thymulin
production
and thymic epithelial
cell (TEC)
growth (10,ll).
Thymulin,
a thymic hormone,
is secreted by
medullar
TEC. Incubation
with thymosin or TEC can induce
H-antigen in uncommitted
bone marrow rosette-forming
cells
(12, 13).
An increasing
complexity
of endocrine-immune
interactions, however,
is suggested by the observation
that lymphocytes can secrete peptides that are similar or identical to
pituitary
hormones.
We and others have found that ACTH,
significant
(P < 0.001)
increases
in thymocyte
proliferation
in
cultures
stimulated
with
varying
doses of GH and IGF-I.
Also,
conditioned
medium
of human
thymocytes
(1 x 10’ cells) stimulated
with GH for 48 h contained
a significant
(P < 0.001) amount
of
IGF-I.
Thymocyte
proliferation
stimulated
by GH was significantly
(P < 0.01) inhibited
by monoclonal
as well as polyclonal
human
IGF-I
antisera.
Finally,
we studied the paracrine
effect of thymocyte-secreted
GH
on human nrimarv
thvmic
enithelial
cell (TEC) cultures.
A simificant
(P < 0.05) increask
inC”HJthimidine
uptake in TEC culturesafter
GH
addition
was observed,
which was abolished
by GH antiserum.
Polyclonal and monoclonal
IGF-I
antisera
significantly
(P < 0.05) inhibited GH-stimulated
TEC proliferation.
In summary,
human
thymocytes
synthesize
and secrete GH and
IGF-I.
Further,
GH functions
as an autocrine/paracrine
growth
factor
in the human
thymus
via locally
synthesized
IGF-I.
(J Clin Endocrinol Metab 81: 2663-2669, 1996)
GH, PRL, LH, and LH-releasing
hormone-like
peptides are
secreted by human immune cells and serve as comitogens
in
lymphoproliferation
(10-22). These latter observations
raise
the possibility
that the GH influence on thymic function may
be produced
by the local synthesis of GH and IGF-I.
In the present study we investigated
the endogenous
synthesis and secretion of GH and IGF-I from human thymocytes and their roles as autocrine/paracrine
growth
factors.
Materials
and
Methods
Cell culture
Thymic tissues (n = 16) were obtained
from subjects (ages 10 days to
3 yr) undergoing
cardiovascular
surgery
for congenital
heart disease.
The tissues were cut into small pieces, and the cells were then gently
teased apart and passed through
a stainless steel mesh (40-pm mesh) to
remove
clumps
of cells and connective
tissue. The mononuclear
cells
were separated
by centrifugation
on a Ficoll-Hypaque
density gradient.
Thymocytes
were washed
twice, suspended
in serum-free
AIM-V
medium (Life Technologies,
Gaithersburg,
MD), and incubated
at 37 C in
5% CO,. The viability
of the cells (always
~95%)
was determined
by
trypan
blue exclusion.
The experiments
mentioned
below were performed on cells obtained
from more than one thymus as indicated
in the
individual
figure legends.
Received
June 21, 1995. Revision
received
December
20, 1995.
Accepted
January
1, 1996.
Address
all correspondence
and requests
for reprints
to: Dr.
Paramjeet
Sabharwal,
M.D., Johns Hopkins
Hospital,
Department
of
Surgery,
665 Blalock,
600 North
Wolfe Street, Baltimore,
Maryland
21287.
* This work was supported
by NIH General Clinical
Research Center
Grant MOl-RR-0034.
Presented
in part at the 75th Annual Meeting
of The
Endocrine
Society, June 10, 1993.
GH and IGF-I
RIA
Thymocytes
obtained by the above-mentioned
procedure
were homogenized in a lysing buffer (0.25 mol/L sucrose and 0.05 mol/L Tris-HCl,
pH
7.6; 1 X 10” cells/ml).
Cell extracts were centrifuged
at 10,000 X s for 30
min at 4 C. Then, supernatants
were separated
and filtered to remove
insoluble particles. The supematants
were assayed by RlA for GH (NIDDK,
National
Pituitary
Hormone
Program)
and IGF-1. The IGF-I assay was
supplied in a kit form by Nichols institute Diagnostics
(San Juan Capistrano,
2663
2664
SAEZIARWAL
CA). Before assay, the samples were extracted with an acid-ethanol
solution
to remove IGF-I-binding
proteins. The intra- and interassay
coefficients
of
variation
for these assays were less than 10%.
Hormonal
enzyme-linked
immunoplaque
assay
GH was measured
by a sensitive
hormonal
enzyme-linked
immunoplaque
assay developed
in our laboratory
(16). In brief, the nitrocellulose membrane
filter bottom of microtiter
plates were first coated with
a monoclonal
antibody
to human
GH (1:lOOO; Fitzgerald
Industries,
Acton, MA) in phosphate-buffered
saline (PBS)-BSA
to prevent
nonspecific binding.
Excess antibody
was washed using PBS; thereafter,
100
WL of a suspension
(1 x 105/mL)
of thymocytes
were plated on the
antibody-coated
nitrocellulose.
Plates were incubated
for 48 h at 37 C.
Thereafter,
cells were removed,
and the plates were washed
with PBS
(twice, 200 PL). A second antibody,
goat antihuman
GH (BioSpacific,
Emerysville,
CA) at a final dilution
of l:lOOO, was added to the wells and
incubated
at 37 C for 4 h. The wells were then washed with PBS-Tween
20 (0.05%; eight times, 200 PL). Horseradish
peroxidase-conjugated
rabbit antigoat
IgG was added to the wells and incubated
for 1 h at 37 C.
The wells were then washed
with PBS-Tween
20 (0.05%; eight times,
2OOkL) and TBS (twice, 200~L).
The addition
of an enzyme
substrate
solution,
4-chloro-1-naphthol,
to the wells provided
visualization
of
GH-specific
violet plaques.
IGF-1 secretion
from thymocytes
was also
detected with this assay; the only difference
was the monoclonal
capture
antibody
(3D1/2/1,
NIDDK)
and the polyclonal
detect antibody
(UB3189, NIDDK).
For specificity
of the assay upon coating the nitrocellulose
membrane
with normal
rabbit serum (NRS) instead of monoclonal
antibody
or replacing
the polyclonal
antibody
with NRS, no significant
plaque formation
was observed.
Omission
of monoclonal
or polyclonal
antibody
also prevented
subsequent
plaque formation.
Western
blot and immunoprecipitation
Thymocytes
cell extracts were prepared
by lysis in 50 mmol/L
TrisHCl (pH 8.0), 150 mmol/L
NaCl, 1% (vol/vol)
Triton X-100, and 0.02%
(wt/wt)
NaN, containing
phenylmethanesulfonylfluoride
(100 pg/mL)
and aprotinin
(1 pg/mL).
Samples were dissolved
by heating for 5 min
at 100 C in 0.3 mmol/L
Tris-HCl,
pH 6.8, with 4% SDS, 10% glycerol,
and
5% dithiothreitol.
Samples were subjected
to SDS-PAGE
using a 4%
stacking gel and a 14% separating
gel (1.0 X 130 X 160 mm). Proteins
were transferred
from gels to nitrocellulose
membranes
in a Millipore
semidry
graphite
transblot
apparatus
at 0.2 milliamperes/gel
for 2.0 h.
Immunostaining
was achieved
with an alkaline
phosphatase
immunoblot assay. Antiserum
dilutions
were as follows:
affinity-purified
antihuman GH (BioSpacific;
cross-reactivity
<O.Ol% for PRL, TSH, FSH, and
LH), l:lO,OOO; and rabbit antigoat
IgG conjugated
with alkaline
phosphatase,
1:3,000. GH-like
molecules
were visualized
by reaction
with
nitro blue tetrazolium
chloride
and 5-bromo-4-chloro-3-indolyl
phosphate p-toluidine
salt.
Thymorte
proteins
(5 x 10’ cells/ml)
were radiolabeled
with 150
pCi/mL
J3-Slmethionine
for 4 h in methionine-free
RPM1 1640 medium
(Sigma Chemical Co., St. Louis, MO). Thymocytes
were dissolved
in l-fold
concentrated
immunoprecipitation
buffer [50 mmol/L
Tris-HCl
(pH 8.0),
150 mmol/L
NaCl, 0.02% NaN, 100 pg/mL
phenylmethylsulfonylfluoride,
1 pg/mL
aprotinin
1% Triton X-100, and ddH,OJ.
Samples were precleaned
for 1 h at room temperature
with 50 wL/mL
protein A (Life Technologies).
After centrifugation
(6000 X g for 2 min) to remove
protein
A-gel, the
supematants
were reacted overnight
at 4 C with 50 pL/mL
antihuman
GH
(1 mg/mL)
affinity-purified
antibody
(BioSpacific,
Emerysville,
CA), and
samples were incubated
with protein A-gel for 2 h before centrifugation
(6000 X 8 for 2 min) at room temperature.
The pellets were washed five
times with immunoprecipitation
buffer, boiled for 2 min in SDS-sample
buffer with 5% mercaptoethanol,
and placed in SDSPAGE.
Gels were fixed
overnight
in 10% acetic acid with 25% methanol
and dried for 30 min, then
exposed to Kodak X-Omat
AR5 film (Eastman Kodak, Rochester,
NY) at
-20 C for 12-96 h before developing.
Blastogenic
JCE & M . 1996
Vol81~No7
AND VARMA
0.5 &i [“Hlthymidine
(6.7 Ci/mmol;
ICN Radiochemicals,
Irvine, CA) was
added to each well; 18 h later, cells were harvested
onto glass fiber filters
with a pHD cell harvester.
The filters were air-dried,
and incorporated
radioactivity
was measured
by liquid scintillation
spectroscopy.
TEC
Thymus
fragments
were explanted
in Costar 12.well plastic plates
(-10
fragments/well;
Costar,
Cambridge,
MA)
with Eagle’s culture
medium
(Life Technologies)
and 10% human AB serum (Sigma). Cultures were kept at 37 C, and the medium
was changed twice a week (16).
Rings of TEC were visible at the explant
periphery
within
4-7 days,
reaching
a diameter
of 0.5-l cm after lo-15 days. Cultures
were rinsed
4 times with fresh medium
before further
use. Cultures
were then
trypsinized,
and TEC were obtained.
Lymphocytes
(nonadherent
cells
before trypsinization)
were not present in these cultures,
as determined
by light microscopy.
TEC were then washed, counted,
and resuspended
in serum-free
Eagle’s Medium
at a concentration
of 2 x 1Oi cells/ml.
One hundred
microliters
of cell suspension
were added per well in a
96-well
microtiter
plate and incubated
with 100 ILL concentrated
thymocyte-conditioned
medium
(1 X lO’/mL
human
thymocytes
were
cultured
in serum-free
medium
for 48 h, and the supernatant
was concentrated
IO-fold).
To other TEC cultures,
we added increasing
doses of
anti-GH
antiserum
and NRS and evaluated
DNA synthesis, as described
in the previous
section.
Statistical
evaluation
Data were analyzed
by one- and two-way
ANOVA
using the JMP
Statistical
Analysis
Package,
which is the SAS product
for Macintosh
(SAS Institute,
Cary, NC). P 5 0.05 was considered
significant.
GH in extracts
Increasing
[‘2511GH from
pituitary
GH,
pituitaryand
amounts
of thymic tissue extracts displaced
GH antiserum
parallel
to that produced
by
suggesting
immunologic1
similarity
between
thymocyte-derived
GH (Fig. 1).
Secretion of GH and IGF-I by human
by the immunoplaque
assay
thymocytes
detected
After finding GH in extracts of human thymus, we investigated GH and IGF-I secretion from thymocytes.
Human
100,
assays
Experiments
were run in triplicate
in lOO-FL volumes
using 96.well
tissue culture
plates.
To evaluate
DNA
synthesis
in thymocytes,
Thymocytes
were incubated
with various stimulants
for 72 h, after which
of human
Results
thymocytes
GtI
FIG. 1. Parallel
(rig/ml)
displacement
of [ i2”IlGH
by 25, 50, 100, and 150 FL
of thymocyte
cell extract
(--I
and pituitary
GH suggested
that a
GH-like
peptide
was present
in human
thymocytes.
TIIYMIC
2665
GH AND IGF-I
thymocytes produced homogeneously stained immunoreactive GH and IGF-I plaques (Fig. 2). In contrast, when nonhormone-secreting Nb, cells were plated instead of thymocytes, specific GH and IGF-I plaque formation was
completely abolished. Omission of either the monoclonal or
polyclonal antibody also prevented subsequent plaque formation (data not shown).
30.OK.II
21.5K14.3K-
Synthesis
of GH by human
thymocytes
To determine whether the secreted GH was being synthesized by the thymocytes, they were cultured in medium
containing [35S]methionine, and the supernatant was concentrated and immunoprecipitated with affinity-purified
goat antihuman GH. The immunoprecipitate was separated
according to molecular size, using SDS-PAGE gels under
reducing conditions. Autoradiograms of the gels showed a
distinct band at 23 kDa (Fig. 3). A similar 23-kDa (Fig. 4)
specific GH band was visualized using Western blot analysis,
whereas substitution of primary antibody with normal rabbit
serum did not produce a 23-kDa form.
Effect of GH on thymocyte
GH functions
as an autocrine
46K
-
30K
-
4
21.5Km
14.3K-
proliferation
To evaluate the physiological role of GH on human thymocytes, various dosesof human GH (NIDDK hGH I-3) were
incubated with human thymocytes, which enhanced proliferation, as demonstrated by a significant (P < 0.01) increase
in thymidine incorporation (Fig. 5). A biphasic dose-responsecurve was produced, such that a low concentration of
GH stimulated T cell proliferation, but the stimulatory responseto low concentrations of GH was attenuated at higher
GH concentrations.
Secreted
2
FIG. 3. Autoradiogram
showing a 23-kDa band immunoprecipitated
from a [35S]methionine-labeled
human thymocyte protein extract.
Lane 1, Nonspecific bands precipitated using NRS; lane 2, specific
23-kDa band from an immunoprecipitate
using affinity-purified anti-GH antibody.
1
growth
factor
Phytohemagglutinin @‘HA)-stimulated thymidine incorporation into human thymocytes was inhibited (P < 0.04)by GH
antisera both NIDDK anti-hGH-2 (1:lOOO)and BioSpacific 05784405(l:lOOO)], but not by NES (Fig. 6). This finding suggested
,\’
1
\
2.3
FIG. 4. Immunoblot using GH antiserum, demonstrating the presence of a 23-kDa band from a hGH pituitary preparation (NIDDK
hGH l-3; lane 1) and from an extract of human thymocytes (lane 2).
When NRS, instead of GH antiserum, was used to blot proteins from
thymocyte extract, no band was visualized (lane 3).
30000
u
I::
20000
=:z
f$js
5
IF
10000
”
0
GH(ug/ml)
FIG. 5. The effects of varying doses of GH (0.01-10 pg/rnW on thymocyte proliferation. Increases in thymidine uptake were significant
(P < 0.001) for all doses compared to that by unstimulated thymocytes. Results represent the mean ? SEM of triplicate cultures. Similar
results were obtained in experiments performed on cells obtained
from six different thymic tissues.
that mitogen-stimulated thymocytes were secreting GH that
was acting as a comitogen for thymocyte proliferation.
Effect of recombinant
FIG. 2. The
observed in
leased from
as a control
secretion of IGF-I and GH from human thymocytes, as
an immunoplaque assay. The IGF-I (A) and GH (C) recells after 48 h of incubation are indicated. Nb, cells used
did not release GH (B) or IGF-I.
IGF-I
on thymocyte
proliferation
To examine the role of IGF-I in thymic function, it was
added to thymocyte cultures. We found that recombinant
IGF-I (Fig. 8) significantly (P < 0.01) enhanced thymocyte
proliferation in a biphasic manner.
2666
SAENARWAL
JCE & M.
Vol81.No
AND VARMA
1996
7
PHA alone
V
plus antlGH(Blo)
e
plus antiGH(NIH)
---+--
plus
NRS
FIG. 6. Effect of GH antiserum
on thymocytes
stimulated
with various
doses
of PHA. PHA stimulation
of thymidine
incorporation
into human
thymocyte
cultures
(- - -1 was significantly
(P <
0.04) decreased
by both NIDDK
antiGH-2
(l:lO,OOO;
-1
and BioSpacific
057-B4405
(1:lOOO; -) GH antiserum.
NRS (- -1 was used as a negative
control.
Results
represent
the mean 2 SEM of
triplicate
cultures.
Similar
results were
obtained
in experiments
performed
with cells from two individual
thymic
PHA
FIG. 7. Effects
of GH and thymocyte-conditioned
medium
on TEC.
GH (1 kg/mL)
and thymocyte-conditioned
medium
(100 &/well)
significantly
(P < 0.05) enhanced
TEC proliferation,
which was inhibited
by affinity-purified
GH antiserum
057-B4405
(l:lO,OOO).
Anti-GH
plus GH (1 pg) was used as a specificity
control.
Counts were 1906 ?
562, whereas
anti-GH
(l:lO,OOO)
alone inhibited
the baseline
counts
(from 1521 5 167 to 1092 ? 47). Results
represent
the mean t SEM
of triplicate
cultures.
Similar
results
were obtained
in experiments
repeated
with TEC from two different
thymic
tissues.
IGF-I
in thymocyte
cell extracts
A role for IGF-I in thymic function was further supported
by its presence in thymocyte extracts. Human thymocyte
extracts and recombinant IGF-I produced parallel displacement of IGF-I label from IGF-I antibody (Fig. 9).
Conditioned
a paracrine
medium from human thymocytes
growth factor for TEC
functions
as
GH and conditioned media from thymocytes, when added
to the thymic epithelial monolayers, significantly stimulated
(P < 0.05) mitogenesis, an effect inhibited by GH antiserum
(Fig. 7). This finding suggested another physiological role for
(ug/ml)
IGFI
(ug,ml)
FIG. 8. The effects of varying
doses of IGF-I
on thymocyte
proliferation. Note the significant
(P < 0.001) increase
in thymidine
uptake
and a biphasic
dose-response
curve.
Results
represent
the mean i
SEM of triplicate
cultures.
Similar
results
were obtained
in experiments repeated
with three different
thymuses.
GH in the thymus in addition to promoting thymocyte
proliferation.
Autocrine
and paracrine
effects of hGH on thymocytes
thymic epithelial
cells uia IGF-I
and
To determine whether the lymphoproliferative action of
GH is mediated by IGF-I, we stimulated thymocytes with
increasing doses of recombinant GH (Genentech, South San
Francisco, CA), and a significant (P < 0.001) increase in
medium IGF-I concentrations was observed (Fig. 10). Recombinant GH-stimulated [3H]thymidine incorporation into
human thymocytes was significantly (P < 0.01) inhibited by
monoclonal as well as polyclonal anti hIGF-I antisera (Fig.
11). In addition, monoclonal and polyclonal anti-hIGF-I an-
THYMIC
2667
GH AND IGF-I
J
60 50 40 30 20 -
04
I
1
10 -
0 I
,
10
100
I
‘.““‘I
1000
IGF-1
10000
(pg/ml)
FIG. 9. The parallel
displacement
of [ 126111GF-I from IGF-I antibody
produced
by 25, 50, 100, 150, 200, 250, and 300 PL thymocyte
cell
extract
C-O-) and IGF-I (-•) suggested
that an IGF-I-like
peptide was
present
in thymocytes.
To determine
the true IGF-I
value from the
graph for a given number
of microliters
of cell extract,
the number
needs to be multiplied
by 225, the dilution
factor
for acid-ethanol
extraction.
1
0
1’1000000
I
1
IGF-I
100000
antisera
I
1
10000
I
1
*
1000
dilutions
FIG. 11. Thymidine
incorporation
into human
thymocytes
produced
by 1 wg/mL GH (-W-J was significantly
(P < 0.01) inhibited
by monoclonal (-A-) and polyclonal
c-m-1 IGF-I
antisera.
NRS was used as a
negative
control.
Results
represent
the mean -C SEM of triplicate
cultures.
Similar
results
were obtained
in experiments
repeated
with
TEC from three different
tissues.
250
0
0.1
1
10
100
GH(ug/ml)
FIG. 10. Effects of various
doses of GH on IGF-I
secretion
by human
thymocytes.
IGF-I
secretion
is significantly
(P < 0.001) enhanced
in
a dose-dependent
manner.
Results
represent
the mean t- SEM of two
experiments.
IGF-I
secreted
by unstimulated
cells is considered
as
100% (100 X 5 ng/lOs
cells); maximum
IGF-I
secreted
was 227.7
ng/lOs cells stimulated
by 100 pg/mL
GH.
tisera significantly inhibited hGH-stimulated thymic epithelial monolayers, as indicated by a decrease in thymidine
incorporation (Fig. 12).
Discussion
In this study we investigated the synthesis and secretion
of GH by human thymocytes and its role as an autocrine/
paracrine thymic growth factor. First the presence of thymic
GH and IGF-I was detected by RIA of thymocyte extracts.
Then, using an immunoplaque assay developed in our laboratory (16) to detect the secretion of immunoreactive peptides from human immune cells, we determined the secretion
of GH and IGF-I by human thymocytes. Also, the active
synthesis of GH was demonstrated by finding immunoprecipitable 35S-labeled23-kDa GH from cell lysates, and a similar 23-kDa band was observed when thymic tissue extracts
were subjected to gel electrophoresis followed by immunoblotting. These results suggested that an immunoreactive
FIG. 12. Effect of anti-IGF-I
antiserum
on GH (1 Fg/mL)-stimulated
TEC. Both polyclonal
[anti-IGF-I(P)]
and monoclonal
[anti-IGF-I(M)1
anti-IGF-I
antisera
(final dilution,
1:lOOO) significantly
U’ < 0.05)
inhibited
GH-stimulated
TEC mitogenesis,
but NRS did not (15,303
+- 550). Results
represent
the mean
2 SEM of triplicate
cultures.
Similar
results were obtained
in experiments
repeated
with TEC from
two different
thymuses.
GH with a mol wt similar to that of pituitary GH was being
secreted by thymocytes.
Several studies have recently suggested that GH or IGF-1
influences thymic function. GH as well as IGF-I have been
reported to increase thymic weight in aged rats (3), hypophysectomized rats (23), diabetic rats with thymic atrophy (24),
and a transgenic mouse model (25).
In humans with GH deficiency, cellular immune defects
have been inconsistently reported before and after GH therapy (23, 25-28). Our laboratory has reported that in acromegaly, a condition of excessive GH secretion, macrophage
function was enhanced (29). Also in acromegalic patients,
levels of thymulin, a peptide product of TEC, are elevated
(10).
To evaluate the physiological relevance of thymocyte-secreted GH, we added GH as well as IGF-I to thymocyte
cultures and noted that they promoted cellular proliferation.
The proliferative responsesto GH and IGF-I were biphasic,
consistent with the fact that low concentrations of antigen
stimulate T cells where as high concentrations causeanergy,
similar to what was observed for IGF-I in lymphoblast cell
lines (5,30,31). The mitogenic responsewas greater with GH
2668
SABHARWAL
than with IGF-I stimulation.
As there is an inhibitory
feedback mechanism by which immunoreactive
IGF-I secreted by
lymphocytes
inhibits
immunoreactive
GH synthesis
by
themselves
and by the surrounding
lymphocytes
(32), the
greater mitogenic
response to GH than to IGF-I stimulation
may be due to this feedback inhibition.
We also observed that
the stimulatory
response to low concentrations
of GH and
IGF-I was attenuated
at higher concentrations;
therefore, GH
stimulation
increases IGF-I synthesis in the microenvironment of the lymphocyte,
which
stimulates
mitogenesis,
where, as upon adding IGF-I in the culture medium,
high
IGF-I concentrations
are achieved quickly, which instead of
stimulating
the mitogenesis,
attenuates it.
When PHA-stimulated
thymocytes
were incubated
with
GH antiserum,
a significant
inhibition
of thymidine
uptake
was observed, suggesting
that the endogenously
secreted
GH functioned
as an autocrine
mitogen
in thymocyte
proliferation.
An autocrine mitogenic
role for GH in peripheral
blood mononuclear
cells has also been proposed
(15, 16).
To determine
whether
the cellular
proliferative
effect of
GH was mediated
via locally generated
IGF-I, thymocyte
cultures were stimulated
with varying doses of rGH. A significant (P < 0.01) increase in medium
IGF-I concentrations
was observed after treatment with GH. Similarly,
Murphy
et
al. and others (10, 22,30,33)
have shown that GH increases
IGF-I messenger
ribonucleic
acid in the thymus. Also, the
addition of monoclonal
and polyclonal
IGF-I antisera to these
thymocyte
cultures significantly
(P < 0.01) inhibited
thymidine incorporation,
thus further supporting
the hypothesis
that GH acts via locally generated
IGF-I.
Using primary
TEC cultures, we further investigated
the
possible paracrine
role of GH secreted by thymocytes.
Conditioned medium from thymocyte cultures and recombinant
GH stimulated
the proliferation
of TEC, which was inhibited
by GH and IGF-I antisera. Similarly,
researchers
(10) have
reported that exogenous
GH stimulates
thymulin
and IGF-I
production
as well as the proliferation
of TEC. Collectively,
these findings suggest that endogenously
synthesized
thymic GH and IGF-I play significant
roles as growth factors for
both thymocytes
and TEC.
A functional
role for GH in thymus and other immune
tissues has been suggested
by recent studies of GH and
cytokine receptors.
It has been established
that numerous
interleukins
as well as GH and PRL receptors have common
amino acid sequences in the important
ligand-binding
regions of their extracellular
receptor domains and, therefore,
have been referred to as the GH/PRL
cytokine receptor family (34). Also, in addition
to GH, other hormones
and neuropeptides
have been found in the thymus (20,35,36).
Other
than acting as comitogens,
are there other potential
roles for
these agents in T cell maturation?
Of interest is a recent report
by Li et al. (7) that examined
the role of GH in rat thymic
maturation.
They noted that the progression
of immature
double negative T cells, CD4- CDB-, to double positive T
cells, CD4+ CDB+, was influenced
by GH (7).
It has also been suggested that the intrathymic
expression
of these peptides may serve as important
mediators
of positive and negative selection
of T cells (37). For example,
neuropeptides
such as GH, IGF-I, LH, and arginine
vasopressin, which
we and others have found in the human
JCE & M . 1996
Volt31 . No 7
AND VARMA
thymus, might serve as self-antigens
representing
other homologous
neuropeptides.
These peptides could promote the
negative selection of autoreactive
thymic cells produced
during recombination
events that establish T cell receptor structure. Hence, GH and other peptides
may be involved
in
various immunological
events that determine
the ability of
T cells to recognize
self- from foreign antigens.
We conclude
that human
thymocytes
synthesize
and
secrete a GH-like peptide. In addition,
thymic GH acts as an
autocrine
growth factor, modulating
T cell proliferation
via
locally synthesized
IGF-I, and also functions as a paracrine
factor, modulating
TEC growth.
Acknowledgments
We appreciate the human GH RIA kits and
clonal antihuman
IGF-I antibodies
provided
by
mone and Pituitary
Program.
We thank W. B.
valuable
guidance,
and Thomas
M. O’Dorisio,
mations
of IGF-I in various
samples.
monoclonal
and polythe NIH National
HorMalarkey,
M.D., for his
M.D., for making
esti-
References
1 Smith PE. 1930 Effect of hypophysectomy
upon the involution
of the thymus
in the rat. Anat Rec. 47119-126.
2 Kelley
KW. 1991 Growth
hormone
in immunobiology.
In: Ader R, Felten DL,
Cohen N, eds. Psychoneuroimmunology,
2nd ed. New York: Academic
Press;
377-402.
3 Kelley
KW, Brief S, Westly
HJ, et al. 1986 GH, pituitary
adenoma
cells can
reverse
thymic
aging rats. Proc Nat1 Acad Sci USA. 83:5663-5667.
4 Goya RG, Gagnerault
MC, Leite De Moraes
MC, Savino
W, Dardenne
M.
1992 III oioo effects of growth
hormone
on thymus function
in aging mice. Brain
Behav Immun.
6:341-354.
5 Geffner
ME, Bersch N, Lippe BM, Rosenfeld
RG, Hintz RL, Golde DW. 1990
Growth
hormone
mediates
the growth
of T-lymphoblast
cell lines via locally
generated
insulin-like
growth
factor-l.
J Clin Endocrinol
Metab.
71:464-469.
6 Berczi
I, Nagy E, deToldeo
SM, Matusik
RJ, Friesen
HG. 1991 Pituitary
hormones
regulate
c-myc and DNA synthesis
in lymphoid
tissue. J Immunol.
146:2201-2206.
7. Li YM, Brunke
DL, Dantzer
R, Kelley
KW. 1992 Pituitary
epithehal
cell
implants
reverse the accumulation
of CD4-CDB-lymphocytes
in thymus glands
of aged rats. Endocrinology.
30:2703-2709.
8. Baxter JB, Blalock
JE, Weigent
DA. 1991 Characterization
of immunoreactive
insulin-like
growth
factor-1
from leukocytes
and its regulation
by growth
hormone.
Endocrinology.
129:1727-1734.
9. W&gent
DA, Baxter JB, Blalock
JE. 1992 The production
of growth
hormone
and insulin-like
growth factor-1 by the same subpopulation
of rat mononuclear
leukocytes.
Brain Behav Immun.
6:365-376.
10. Timsit J, Savino
W, Safieh
B, et al. 1992 Growth
hormone
and insulin-like
growth
factor-1
stimulate
hormonal
function
and proliferation
of thymic
epithelial
cells. J Clin Endocrinol
Metab.
75:183-188.
11. Goya RG, Gagnerault
MC, Sosa YE, Bevilacqua
JA, Dardenne
M. 1993 Effects
of GH and thyroxine
on thymulin
secretion
in aging rats. Neuroendocrinology.
58:338-343.
12. Bach JF, Dardenne
M, Goldstein
AL, Guha A, White
A. 1971 Appearance
of
T-cell markers
in bone marrow
rosette
forming
cells after incubation
with
thymosm,
a thymic
hormone.
Proc Nat1 Acad Sci USA. 68:2734-2737.
13. Papiernik
M, Nabarra
B, Bach J-F. 1975 Irl uitro culture
of functional
human
thymic
epithelium.
Clin Exp Immunol.
19:281-287.
A, Singh
A, Krai T, Solomon
S. 1989 The immune-hypothalamic14. Bateman
pituitary-adrenal-axIs.
Endocr
Rev. 10:92-112.
15. Weigent
DA, Baxter
JB, Wear WE, Smith
LR, Bost KL, Blalock
JE. 1988
Production
of immunoreactive
growth
hormone
by mononuclear
leukocytes.
FASEB J. 2:2812-2818.
16. Varma
S, Sabharwal
P, Sheridan
JF, Malarkey
WB. 1993 Growth
hormone
secretion
by human
peripheral
blood
mononuclear
cells detected
by an enzyme-linked
immunoplaque
assay. J Clin Endocrinol
Metab. 76:49-53.
17. Binder
G, Revskoy
S, Gupta D. 1994 In viva growth hormone
gene expression
in neonatal
rat thymus
and bone marrow.
J Endocrinol.
140:137-143.
Pellegrini
I,
Lebrun
JJ,
Ali
S,
Kelly
PA.
1992
Expression
of
prolactin
and its
18.
receptor
in human
lymphoid
cells. Mel Endocrinol.
6:1023-1031.
19. Sabharwal
P, Glaser
R, Lafuse
W, et al. 1992 Prolactm
synthesized
and
secreted
by human
peripheral
blood mononuclear
cells: an autocrine
growth
factor for lymphoproliferation.
Proc Nat1 Acad Sci USA. 89:7713-7716.
20. Sabharwal
P, Varma
S, Malarkey
WB. 1992 Human
thymocytes
secrete lu-
THYMIC
21
23
24.
00.
00.
00.
25.
26.
27.
teinizing
hormone:
an autocrine
regulator
of T-cell proliferation.
Biochem
Biophys
Res Commun.
182:1187-1192.
Varma
S, Sabhanval
I’, Malarkey
WB. 1992 Human
splenocytes
secrete
LHRH,
which
inhibits
lymphocyte
proliferation.
Prog NeuroEndocrinImmunol. 5187-191.
Yamada
M, Hato F, Kinoshita
Y, Tominaga
K, Tsuji Y. 1994The
indirect
participation
of growth
hormone
in the thymocyte
proliferation
system. Cell
Mol Biol. 40:111-21.
Rapaport
R, Oleske
J, Ahdieh
H, Solomon
S, Delfaus
C, Denny
T. 1986
Suppression
of immune
function
in growth
hormone-deficient
children
durmg
treatment
with human growth
hormone.
J Pediatr.
109:434-439.
Froesch
ER, Guler HP, Schmid
C, Binz K, Zapf J. 1990 Therapeutic
potential
of insulin-like
growth
factor 1. Trends Endocrinol
Metab.
2254-260.
Montgomery
DW, LeFevre
JA, Ulrich
ED, Adamson
CR, Zukoski
CF. 1990
Identification
of prolactin-like
proteins
synthesized
by normal
murine
lymphocytes.
Endocrinology.
127~2601-2603.
Gular
HP, Zapf J, Scheiwiller
E, Froesch
ER. 1988 Recombinant
human
insulin-like
growth factor I stimulates
growth
and has distinct
effects on organ
size in hypophysectomized
rats. Proc Nat1 Acad Sci USA. 85:4889-4893.
Davila
DR, Brief S, Simon J, Hammer
RE, Brinster
RL, Kelley
KW. 1987 Role
of growth hormone
in regulating
T-dependent
immune
events in aged, nude,
and transgenic
rodents.
J Neurosci
Res. 18:108-116.
Kiess W, Doers H, Eisl E, Butenandt
0, Belohradsky
BH. 1986 Lymphocyte
subsets and natural
killer activity
in growth hormone
deficiency.
N Engl J Med.
314-321.
Gupta
S, Frikiz
SM, Nova1 NS. 1983 Immunological
studies
in patients
with
isolated
growth
hormone
deficiency.
Clin Exp Immunol.
54:87-90.
Petersen
BH, Rapaport
R, Henry
DP, Huseman
C, Moore
WV. 1990 Effects
of treatment
with biosynthetic
human growth
hormone
(GH) on peripheral
blood lymphocyte
population
and function
in growth
hormone-deficient
children. J Clin Endocrinol
Metab.
70:1756-1760.
GH AND
IGF-I
2669
28. Abbasi
V, Bellanti
JA. 1985 Humoral
and cell-medlated
immunity
in growth
hormone
deficient
children.
Effect of therapy
with human growth
hormone.
Pediatr
Res. 19:2YY-303.
29. Sabhrawal
P, Zwilling
B, Glaser
R, Malarkey
WB. 1992 Cellular
immunity
in
patients
with acromegaly
and prolactinomas.
Prog NeuroEndocrinlmmunol.
5:120-125.
30. Palmer JM, Wallis
M. 1989 Human
growth hormone
stimulates
somatomedin
C/insulin-like
growth
factor I production
by the human lymphoid
cell line,
IM-9. Mel Cell Endocrinol.
63:167-173.
31. Johnson
EW, Jones LA, Kozak
RW. 1992 Expression
and function
of insulin
like growth
factor receptors
on anti CD3 activated
human
T-lymphocytes.
J Immunol.
148:63-71.
32. Baxter JB, Blalock
JE, Weigent
DA. 1991 Characterization
of immunoreactive
insulin-like
growth
factor-1 from leukocyte
and its regulation
by growth
hormone. Endocrinology.
129:1727-1734.
33. Murphy
LJ, Bell GI, Duckwaith
ML, Freiesen
HG. 1987 Identification,
characterization
and regulation
of a rat complementary
deoxyribonucleic
acid which
encodes
insulin
like growth
factor-l.
Endocrinology.
121:
684-691.
34. Kelly PA, Djiane
J, Post&Vinay
MC, Edery M. 1981 The prolactin/growth
hormone
receptor
family.
Endocr
Rev. 12;3:235-251.
35. Greenen
V, Legros JJ, Franchimont
P, Baudrihaye
M, Defresne
MP, Boniver
J. 1986 The neuroendocrine
thymus:
coexistence
of oxytocin
and neurophysin
in the human
thymus.
Science.
232:508-511.
36. Geenen
V, Achour
I, Robert
F, et al. 1993 Evidence
that insulin-like
growth
factor 2 (lGF2) is the dominant
thymlc
peptidt
of the insulin
superfamily.
Thymus.
21:1115-127.
37. Geenen
V, Robert
R, Martens
H, et al. 19Yl Biosynthesis
and paracrine/
cryptocrine
actions of ‘self’ neurohypophysial-related
peptides
in the thymus.
MoI Cell Endocrinol.
76:C27<31.