Full text of DOI:10.1007/BF03343834

J. Endocrinol. Invest. 24: 147-151, 2001
The effect of prolactin and bromocriptine on human
1
peripheral immune status
P. Kadioglu*, O. Açbay*, G. Demir**, N. Gazioglu***, and S. Gundogdu*
*University of Istanbul, Cerrahpasa Medical Faculty, Department of Internal Medicine, Division of
Endocrinology, **Oncology and ***Department of Neurosurgery, Istanbul, Turkey
ABSTRACT. The aim of this study was to deter-
mine the influence of elevated serum prolactin
(PRL) levels on the peripheral lymphocyte subsets
in patients with hyperprolactinemia. For this pur-
pose we studied 20 hyperprolactinemic patient
lymphocyte subsets by flow cytometry on their
hyperprolactinemic state and after their serum
prolactin concentration was normalized with
bromocriptine (BC) alone or BC and surgery. We
observed decreased absolute numbers and
percentage of Natural Killer (p=0.0009 and
0.0001, respectively) and CD3/CD25 lymphocy-
tes (p= 0.009 and 0.002) in hyperprolactinemic
patients, compared to 8 sex- and age-matched
normal controls. There was no correlation be-
tween PRL levels and CD16/56 and CD3/CD25
numbers (p=0.72 and 0.33, respectively). We did
not find any significant difference in absolute num-
bers (p=0.95) and percentage (p=0.84) of B-lym-
phocytes of hyperprolactinemic patients, as com-
pared with normal controls. We did not detect any
increase in absolute cell numbers of CD16/CD56
(p=0.21) and CD3/CD25 (p=0.61) of BC-treated
patients when compared to their hyperpro-
lactinemic state. We demonstrated an increase in
CD8-cells (p=0.0173) and a decrease in CD4/CD8
ratio (p=0.036) in hyperprolactinemic patients
treated with BC. There was also an increase in the
number of activated T-cells (CD3/HLA DR) in this
group, compared to normal controls and the
hyperprolactinemic state of the same patients
(p=0.04). In conclusion, elevated PRL levels do not
lead to an “overstimulation” of the B-cells, but de-
teriorate the cytotoxic function.
(J. Endocrinol. Invest. 24: 147-151, 2001)
©2001, Editrice Kurtis
INTRODUCTION
mic patients, whereas Matera et al. (3) were unable
to reproduce these findings. On the other hand,
Cross et al. showed an increase in B-cell activity in
hyperprolactinemia (4). In other studies, it was
shown that the inhibition of T-cell function by hy-
perprolactinemia was more prominent than the
stimulatory effect on B-cell functions in supraphys-
iologic PRL levels (5, 6). In animal models of au-
toimmune diseases, administration of dopaminergic
agents suppressed the postpartum exacerbation of
systemic lupus erythematosus (7), autoimmune
uveitis (8) and collagen-induced arthritis (9).
Prolactin receptors have been shown on the mem-
branes of white blood cells. The immunosuppres-
sive agent cyclosporine A blockades these recep-
tors and competes with PRL for binding to them.
This may be one explanation of the immunosup-
pressive action of cyclosporin A (10).
Involvement of the neuroendocrine system in im-
mune regulation was first observed in hypophy-
sectomized rats in 1930. Initial observations indi-
cated that hypophysectomized rats present thymic
atrophy and lymphopenia. This effect was reversed
by the reintroduction of prolactin (PRL) to these rats
(1). Thereafter, a number of studies have focused
on the potential immunomodulatory roles of pro-
lactin. Gerli et al. (2) demonstrated reduced Natu-
ral Killer (NK) cytotoxic activity in hyperprolactine-
1Presented in part as an abstract at the 4th European Congress of Endo-
crinology, Sevilla, Spain, May 9-13, 1998.
Supported by the Research Fund of The University of Istanbul. Project
no. 843/190496.
The aim of this study was to determine the influ-
ence of elevated serum prolactin levels and the ef-
fect of a dopaminergic agent, bromocriptine (BC)
on the peripheral immune status in patients with
prolactinoma.
Key-words: Prolactine, bromocriptine, immunomodulation.
Correspondence: Dr. Pinar Kadioglu, Fulya cad. Inci ap. 27/30, 81620
Mecidiyekoy Istanbul, Turkey.
E-mail: atgsk@escortnet.com
Accepted July 24, 2000.
147

P. Kadioglu, O. Acbay, G. Demir, et al.
PATIENTS AND METHODS
Patients selection
Peripheral lymphocyte subsets were studied by flow
cytometry in all subjects during hyperprolactinemia
and after their serum PRL levels were normalized by
BC or surgery and BC. Samples were immediately an-
alyzed by cell counter (Cell Counter Microdiff. 18.
Coulter Elect. Miami, U.S.A.) and by flow cytometer
(Coulter EPICS XL MCL Coulter Elect. Miami, U.S.A.).
Whole blood rather than Ficoll separated mononu-
clear cells was used for flow cytometric analysis. One
hundred mμl of blood was pipetted to each tube, 10
mμl of isotype controls (DAKO inc.) and immuno-
flourescent antibodies (DAKO inc.) were added
and incubated for 30 minutes at 4 C in the dark.
Q-Prep (Coulter Elect. Miami, U.S.A.) was used for
dye fixation and erythrocyte lysing. Two-color
analysis (FITC/PE) was performed for CD2/CD25,
CD3/HLA-DR, CD2/CD8, CD16/56, CD4/CD45RA,
and CD4/CD45RO.
Twenty patients with hyperprolactinemia were ad-
mitted to the study after having given written in-
formed consent. The subjects were considered as 3
different groups as follows:
- group 1: 20 patients with hyperprolactinemia;
- group 2: 8 age- and sex-matched normal controls;
- group 3: 14 patients of group 1, whose serum PRL
levels were normalized with BC alone or
with surgery and BC. In these patients
all the tests were repeated at the end of
first month of the treatment.
Study design
The first examination included clinical examination,
and evaluation of PRL, SGOT, SGPT, TSH levels. Also
a history of drug use which may produce hyperprolac-
tinemia was obtained. PRL levels were measured in
plasma samples collected every 20 min for three times.
PRL levels were determined with ELISA (Merck,
Darmstadt, Germany; normal: female<18 ng/ml,
Table 1 - Prolactin (PRL) levels (ng/ml) and lymphocyte subsets distribution (absolute number, ab/mm3) of patients in their hyper-
prolactinemic state (group 1), normal controls (group 2) and after bromocriptine (BC) therapy (group 3).
Group 1
no.=20
Group 2
no.=8
Group 3
no.=14
PRL
193.5
(62.48-628.00)
14.00
(11.25-19.25)
15.22
(3.40-42.92)
CD 16/56
CD19
157.70
566.00
132.82
(107.25-330.63)
(503-707.25)
(131.1-488.4)
303.35
(191.25- 1059.50)
313.00
(254.25-408.50)
573.55
(396.65- 1288.80)
CD3/25
CD3
12.70
98.00
21.00
(6.90-36.65)
(43.50-310.25)
(8.10-37.99)
1608.35
(971.18- 2091.60)
1466.50
(1302.50-1603.25)
1849.5
(1641.93- 2723.00)
CD4/45 RA
CD4/45 RO
CD4
331.20
464.50
562.80
(255.53- 618.45)
(405.25-544.75)
(494.10- 654.94)
342.00
(245.55- 761.30)
580.00
(500.25-643.5)
500.50
(423.67- 796.40)
943.45
(579.75- 1505.25)
1027.00
(912-1083.75)
1149.58
(1112.40- 1347.50)
CD8
552.60
(382.65- 803.03)
610.00
(536-756.25)
872.62
(696.59- 1088.50)
CD4/CD8
CD3/HLA DR
Leucocyte
Lymphocyte
1.41
1.63
1.29
(1.18- 3.02)
(1.40-2.08)
(1.24-1.71)
122.60
(37.83- 589.40)
107.00
(88.75-347.25)
489.71
(283.52- 682.88)
6750
(4835-8975)
6400
(5200-7900)
9300
(7600- 12800)
2450.00
(1500-3025)
2100.00
(1900-2100)
2700
(2336.90-3450.00)
Values are median and interquartile ranges.
148

Hyperprolactinemia and immune system
male<17 ng/ml). Serial dilutions were performed,
if necessary, to give an accurate evaluation of PRL
levels.
and CD16/56 and CD3/25 cell numbers (p=0.72 and
0.33, respectively). We also could not determine any
correlation between disease duration and CD16/56
and CD3/25 numbers (p=0.414 and p=0.48, re-
spectively).
Statistical analysis
Statistical analysis was done by Wilcoxon test in
paired groups (group 1 and group 3), and by Mann-
Whitney-U in non-paired groups. The Pearson prod-
uct-moment correlation coefficient was used to
evaluate the degree of correlation between all pa-
rameters. Significance was accepted if p<0.05.
Calculations were made using SPSS software (SPSS,
Inc., Evanston, IL).
BC-treated patients (BC-patients), when compared
to normal controls, had significantly higher total
leukocytes (p=0.018), total lymphocytes counts
(p=0.015), CD3 (p=0.04), CD3/HLA DR (p=0.01)
and CD8 cells (p=0.01), but decreased CD16/56
cell numbers (p=0.002) and IL-2 expression on ac-
tivated T-cells (p=0.002). There was also no in-
crease in absolute cell numbers of CD16/56
(p=0.21) and CD3/CD25 (p=0.61) of BC-treated pa-
tients when compared to their hyperprolactinemic
state. On the other hand, an increased expression
of HLA-DR on T-lymphocytes (CD3/HLA DR) in this
group was observed, when compared to normal
controls and hyperprolactinemic state of the same
patients before BC treatment (p=0.004 and 0.003
respectively). We demonstrated an increase in CD8-
cells (p=0.017) and a decrease in CD4/CD8 ratio
(p=0.036) (Fig. 3) in hyperprolactinemic patients,
after treatment with BC.
RESULTS
Twenty subjects had hyperprolactinemia [14 wom-
en, 6 men, median age: 34 yr (range: 28-37)]. The
cause of hyperprolactinemia was pituitary mi-
croadenoma in 13 patients, macroadenoma in 6,
and empty sella in one patient. The median dura-
tion of the hyperprolactinemia was 24 months
(range: 9-120 months) and the mean dose of BC
required for normalization of PRL was 7.5 3.64
mg/day.
Prolactin levels and peripheral lymphocyte subsets
distribution of patients is shown on Table 1.
We observed decreased absolute numbers and per-
centage of NK (CD16/56) (p=0.0009 and 0.0001, re-
spectively) (Fig. 1) and decreased expression of in-
terleukin-2 (IL-2) on T-lymphocytes (CD3/CD25)
(p=0.009) (Fig. 2) in hyperprolactinemic patients,
compared to 8 sex, and age-matched normal con-
trols. There was no correlation between PRL levels
DISCUSSION
An optimal amount of PRL must be present for
maximal immune function. Either low or high lev-
els of PRL may result in immunocompromise.
Conflicting data regarding the effect of PRL on
the immune system does exist, however. Muk-
herjee et al. (11) studied the in vitro effect of PRL
1400
1200
1000
800
450
400
350
300
250
200
150
100
600
400
200
0
*
**
**
*
50
0
Hyperprolactinemic Healthy After bromocriptine
state controls therapy
Hyperprolactinemic Healthy After bromocriptine
state controls therapy
Fig. 1 - Comparison of median absolute numbers of Natural
Killer cells (CD16/56) between hyperprolactinemic patients
(no.=20), normal controls (no.=8) and after bromocriptine ther-
apy (no.=14). Results were expressed as the median and in-
terquartile range. *p=0.0009; **p=0.002 (compared to normal
controls by Mann-Whitney-U test).
Fig. 2 - A decrease was observed in expression of interleukin-2
(IL-2) on T-lymphocytes (CD3/CD25) in hyperprolactinemic pa-
tients, and after bromocriptine-therapy as compared with con-
trols. Results were expressed as the median and interquartile
ranges. *p=0.009; **p=0.002 (by Mann-Whitney-U test).
149

P. Kadioglu, O. Acbay, G. Demir, et al.
5
4.5
4
Morikawa et al. (15) demonstrated that BC has a
lowering effect on IL-2 production from T-cells,
but BC poorly influenced the expression of IL-2
receptors. It was also remarkable that the BC
causes an increase in total leukocyte and lym-
phocyte counts in our study due to increment in
suppressor T-cells. BC also activates suppressor
T-cells, which is characterized by increased
CD3/HLA DR expression.
3.5
3
*
2.5
2
1.5
1
The absence of increased markers of B-cells such
as CD19 (p=0.95) indicates that PRL in high concen-
tration does not have any effect on B-cells.
The present results confirm the suppressor effect
of hyperprolactinemia on NK-cell numbers, and the
T-cell activation markers. On the other hand, the
data demonstrates that supraphysiological PRL lev-
els have no effect on B-cell numbers.
0.5
0
Hyperprolactinemic Healthy After bromocriptine
state controls therapy
Fig. 3 - CD4/CD8 ratio in hyperprolactinemic patients, after treat-
ment with bromocriptine and in normal controls. The bars rep-
resent the median levels. 75th and 25th levels were also shown.
*p=0.036 compared to CD4/CD8 ratio in normal controls.
In conclusion, we suggest that hyperprolactine-
mia does not have any immune stimulatory effect
on B-cells, but it has a suppressive effect on T-
and NK-cells. These findings give insight on a
suppressive role of prolactin on cytotoxic func-
tions of immune system, rather than a stimulato-
ry effect on it.
on splenocytes from ovariectomized rats and
showed that PRL induced the formation of inter-
leukin-2 cell surface receptors. On the other hand
Koller et al. (12) investigated several immune pa-
rameters in nine patients with hyperprolactine-
mia. They did not demonstrate any changes in in-
terleukin-2 receptor, CD45RO, and HLA-DR ex-
pression of CD4 or CD8 cells in the patients with
prolactinoma, but they showed an increased
CD4/CD8 ratio, which remained high after treat-
ment and did not seem to correlate with serum
prolactin concentrations. Vidaller et al. (13) stud-
ied four patients with tumoral hyperprolactinemia
and showed that hyperprolactinemia was associ-
ated with a decrease in suppressor cell function,
in the production of IL-2 and in the response to
mitogens. In addition, PRL and hyperprolactine-
mia seem to suppress NK-cell activity. Gerli et al.
(2) and Honorati et al. (14) have demonstrated re-
duced NK-cell activity in human hyperprolactine-
mic patients compared with BC-treated prolacti-
noma patients and healthy control subjects.
Matera et al. (3) were unable to reproduce these
findings, however.
Our data support the decreasing effect of PRL on
the NK-cell numbers and the IL-2 receptor ex-
pression on T-cells. The lack of any correlation
between PRL levels and these differences sug-
gests that PRL may have suppressor effects on T-
cells in high concentrations, however because of
a non PRL-mediated effect, expected correlation
may not be found. Although treatment with BC
normalized PRL levels, NK-cell numbers and IL-2
expression were not changed. These findings sug-
gest that BC also has an immune suppressive ef-
fect independent of its PRL lowering effect.
REFERENCES
1. Cooke N.E.
Prolactin: Basic physiology.
In: DeGroot L.J. (Eds), Endocrinology.
Saunders, Philadelphia, 1995, vol. 1, p. 368.
2. Gerli R., Rambotti P., Nicoletti I., Orlandi S.,
Migliorati G., Riccardi C.
Reduced number of natural killer cells in patients
with pathological hyperprolactinemia.
Clin. Exp. Immunol. 1986, 64: 399-406.
3. Matera L., Ciccarelli E., Cesano A., Veglia F., Miola
C., Camanni F.
Natural killer activity in hyperprolactinemic patients.
Immunopharmacology 1989, 18: 143-146.
4. Cross R.J., Campbell J.L., Roszman T.L.
Potentation of antibody responsiveness after the
transplantation of a syngeneic pituitary gland.
J. Neuroimmunol. 1989, 25: 29-35.
5. Dineen J.K., Kelly J.D.
The suppression of rejection of nippostrongylus
brasiliensis in lacting rats: the nature of immunologic
defect.
Immunology 1972, 22: 1-12.
6. Kelly J.D., Dineen J.K.
The suppression of rejection of nippostrongylus
brasiliensis in Lewis strain rats treated with ovine
prolactin.
Immunology 1973, 24: 551-558.
150

Hyperprolactinemia and immune system
7. McMurray R., Keisler D., Kanuckel K., Izui S., Walker
S.E.
12. Koller M., Kotzmann H., Clodi M., Riedl M., Luger A.
Effect of elevated serum prolactin concentrations
on the immunophenotype of human lymphocytes,
mitogen-induced proliferation and phagocytic ac-
tivity of polymorphonuclear cells.
Prolactin influences autoimmune disease activity in
the female B/W mouse.
J. Immunology 1991, 147: 3780-3787.
Eur. J. Clin. Invest. 1997, 27: 662-666.
8. Palestine A.G., Muellenberg-Coulombre C.G., Kim
M.K., Gelato M.C., Nussenblatt R.B.
13. Vidaller A., Llorente L., Larrea F., Mendez J.P.,
Alcocer-Varela J., Alarcon-Segovia D.
Bromocriptine and lowe dose cyclosporine in the treat-
ment of experimental autoimmune uveitis in the rat.
J. Clin. Invest. 1987, 79: 1078-1081.
T-cell dysregulation in patients with hyperpro-
lactinemia: Effect of bromocriptine treatment.
Clin. Immunol. Immunopathol. 1986, 38: 337-343.
9. Whyte A., Williams R.O.
Bromocriptine suppresses postpartum exacerbation
of collagen-induced arthritis.
Arthritis Rheum. 1988, 31: 927-928.
14. Honorati M.C., Travaglini P., De Bernardi B.,
Scorza R., Hu C., Togni M.E., Radelli L., Vanoli M.,
Zanussi C.
NK activity in patients with pathological hyperpro-
lactinemia: effect of bromocriptine.
Int. J. Neurosci. 1990, 51: 303-305.
10. Reber P.M.
Prolactin and immunomodulation.
Am. J. Med. 1993, 95: 637-644.
15. Morikawa K., Oseko F., Morikawa S.
Immunosuppressive activity of bromocriptine on hu-
man T lymphocyte function in vitro.
11. Mukherjee P., Mastro A.M., Hymer W.C.
Prolactin induction of interleukin-2 receptors on rat
splenic lymphocytes.
Clin. Exp. Immunol. 1994, 95: 514-518.
Endocrinology 1990, 126: 88-94.
151