Hindawi

Journal of Immunology Research

Volume 2021, Article ID 6630715, 15 pages
https://doi.org/10.1155/2021/6630715

Research Article
Prolactin Increases the Frequency of Follicular T Helper Cells with
Enhanced IL21 Secretion and OX40 Expression in Lupus-Prone
MRL/lpr Mice

Yolanda P. Alemán-García,1 Ricardo M. Vaquero-García,1 Rocio Flores-Fernández,1

Ezequiel M. Fuentes-Pananá,2 Patricia Gorocica-Rosete,3 Alberto Pizaña-Venegas,4
Luis Chávez-Sanchéz,1 Francico Blanco-Favela,1 María V. Legorreta-Haquet,1
and Adriana K. Chávez-Rueda 1
1
UIM en Inmunología, Hospital de Pediatría, CMN Siglo XXI, Instituto Mexicano del Seguro Social, Mexico
2
Unidad de Investigación en Virología y Cáncer, Hospital Infantil de México “Federico Gómez”, Mexico
3
Departamento de Investigación en Bioquímica, Instituto Nacional de Enfermedades Respiratorias “Ismael Cosió Villegas”, Mexico
4
Unidad de Investigación y Bioterio, Instituto Nacional de Enfermedades Respiratorias “Ismael Cosió Villegas”, Mexico

Correspondence should be addressed to Adriana K. Chávez-Rueda; [email protected]

Received 29 December 2020; Revised 12 February 2021; Accepted 23 February 2021; Published 9 March 2021

Academic Editor: Carlo Perricone

Copyright © 2021 Yolanda P. Alemán-García et al. This is an open access article distributed under the Creative Commons
Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work
is properly cited.

Systemic lupus erythematosus is characterized by high levels of IgG class autoantibodies that contribute to the pathophysiology of
the disease. The formation of these autoantibodies occurs in the germinal centers, where there is cooperation between follicular T
helper cells (TFH) and autoreactive B cells. Prolactin has been reported to exacerbate the clinical manifestations of lupus by
increasing autoantibody concentrations. The objective of this study was to characterize the participation of prolactin in the
differentiation and activation of TFH cells, by performing in vivo and in vitro tests with lupus-prone mice, using flow cytometry
and real-time PCR. We found that TFH cells express the long isoform of the prolactin receptor and promoted STAT3
phosphorylation. Receptor expression was higher in MRL/lpr mice and correlative with the manifestations of the disease.
Although prolactin does not intervene in the differentiation of TFH cells, it does favor their activation by increasing the
percentage of TFH OX40+ and TFH IL21+ cells, as well as leading to high serum concentrations of IL21. These results support a
mechanism in which prolactin participates in the emergence of lupus by inducing overactive TFH cells and perhaps promoting
dysfunctional germinal centers.

1. Introduction favors the differentiation of thymocytes [10], increasing the

expression of CD69 and CD25 in activated CD8+ T cells
The neuroendocrine and immune systems are closely interre- [11]. In CD4+ T cells, autocrine PRL is important for main-
lated, as the secretory products of the neuroendocrine system taining the expression of CD69 and CD40L and the secretion
can act on the immune system and vice versa [1]. One exam- of IL2 and IFN-γ [5]. In a CD4+ T cell line, PRL induced T-
ple involves hormones that can regulate the immune system bet transcription through phosphorylation of JAK2 and
[2, 3], such as prolactin (PRL) secreted by the pituitary gland, STAT5 [12]. In addition, hyperprolactinemia has been
and extrapituitary immune system cells, such as T cells [4, 5], detected in many patients with different autoimmune dis-
B cells, antigen presenting cells (APCs) [6], natural killer cells eases [13–15], including systemic lupus erythematosus
[7, 8], and monocytes/macrophages [9]. The immunostimu- (SLE), where it has been associated with disease activity [16,
latory functions of PRL have been previously described. PRL 17], with the concentration of anti-dsDNA antibodies [18],
2 Journal of Immunology Research

anemia, and all types of serositis [19]. SLE is a chronic auto- Mice were housed in a pathogen-free barrier facility and were
immune disease characterized by the presence of autoanti- provided with sterile food and water ad libitum.
bodies targeting DNA, RNA, histones, RNP, Ro, La, etc.
[20]. These antibodies are from the IgG isotope, which form 2.2. Prolactin Hormone. We used murine recombinant PRL
immune complexes that are deposited in any organ, causing (National Hormone and Peptide Program, NIH).
damage. The prevalence of SLE is approximately ninefolds
higher in women than in men, and it increases after puberty 2.3. Antibodies. All cells were labeled with the viability dye
and decreases after menopause [21]. There are well- Ghost Red (Tonbo Bioscience, USA). The antibodies used
established experimental models mimicking many aspects for cell culture were as follows: anti-CD3 (clone 145-2C11)
of SLE, such as the MRL/lpr mouse strain [22]. Raising serum and anti-CD28 (clone 37.51) from Invitrogen, USA; anti-
PRL levels in this strain, we demonstrated that the concentra- IFN-γ (clone XMG12) and anti-IL4 (clone 11B11) from Bio-
tion of IgG isotype anti-dsDNA autoantibodies increased, Legend, USA; anti-TGFβ from Peprotech, USA; cytokines
resulting in earlier and more severe manifestations of the dis- IL6 and IL21 from Miltenyi Biotec, Germany. The antibodies
ease [23, 24]. used for staining were as follows: anti-mouse PRL receptor
In the different mouse models that develop SLE, there is APC (clone T6, Novus Biologicals, USA); anti-CD4 PECy5
an increase in the spontaneous formation of germinal centers (clone GK1.5), anti-Ki-67 Alexa 488 (clone 16A8), anti-
(GCs), which correlates with the beginning of the production IL21 biotin (clone 7H20-I19-M3), and PE-conjugated strep-
of autoantibodies [25, 26]. GCs provide a proper microenvi- tavidin from BioLegend, USA; anti-CD44 PECy5 (clone
ronment for the activation, somatic diversification, and affin- IM7), anti-CD62L PE, (clone MEL-14), anti-BCL6 PE (clone
ity maturation of autoreactive B cells, which occur before the BCL-DWN), anti-CXCR5 PECy7 (clone SPRCL5), and anti-
production of autoantibodies [27, 28]. GC formation PD1 APC (clone J43) from eBioscience, Invitrogen, USA;
depends on the presence of follicular T helper cells (TFH), a anti-ICOS VioGreen (clone 7E.17G9), anti-OX40 PE (clone
specialized subpopulation of CD4 T cells. TFH cells are char- REA625), and anti-AKT PE (clone REA677) from Miltenyi
acterized by their expression of CXCR5, ICOS, PD1, CD154, Biotec, Germany; and anti-STAT1 PE (clone A15158B),
and transcription factor BCL6, in addition to secreting IL21 anti-STAT3 PE (clone 13A3-1), and anti-STAT5 PE (clone
[29–33]. An increase in the frequency of circulating TFH is SRBC2X) from BioLegend.
reported in patients with SLE, having a positive correlation 2.4. Induction of High Prolactin Levels and Assessment of SLE
with autoantibody titers and disease activity [34–37]. Mean- Manifestations. MRL/lpr and C57BL/6 female mice (9-
while, it has been observed that the clinical manifestations of weeks-old) were subcutaneously injected with (i) 200 μg of
the disease decrease upon inhibiting the expression of the metoclopramide (Sigma-Aldrich, US) in 100 μL of PBS, (ii)
IL21 receptor in mouse models [38]. Therefore, dysregula- 0.6 mg/kg of bromocriptine (Santa Cruz Biotechnology,
tion of the TFH response contributes to the production of USA) in 100 μL of PBS, (iii) 100 μL of PBS, or (iv) no treat-
pathogenic autoantibodies and, therefore, to the promotion ment for 6 weeks. Urinary protein levels were assessed semi-
of autoimmune diseases mediated by autoantibodies such quantitatively using reagent strips for urinalysis (Mission,
as SLE [39]. USA). Serum samples obtained at the beginning and at the
Taking into account all aforementioned findings, we end of the experiments were kept at −35°C until they were
designed this study to determine the contribution that PRL assayed for anti-dsDNA antibodies as we have previously
has to the differentiation and activation of TFH cells in the reported [23, 24].
MRL/lpr mice. We found that TFH cells express the long iso-
form of the PRL receptor and promoted STAT3 phosphory- 2.5. Serum IL21 Concentration. For the detection of IL21 in
lation. Furthermore, PRL favors the dysregulation of TFH sera, the commercial Legend Max Mouse IL21 ELISA kit
cells by increasing both their absolute number and their (BioLegend, USA) was used according to the supplier’s
activation. instructions. For each determination, 50 μL of serum was
used. The plate was read in the ELISA reader (Dynatech
MR5000) at 450 nm.
2. Materials and Methods
2.6. Purification of Tnaïve and TFH Cells from the Spleen. Eigh-
2.1. Mice. All studies were approved by the Animal Care teen-week-old mice were euthanized, and spleen cells were
Committee of the Instituto Nacional de Enfermedades collected with cold RPMI supplemented with 2% FBS and
Respiratorias “Ismael Cosio Villegas” and the Hospital de 2 mM EDTA (IBI Scientific, USA). Red blood cells were
Pediatría, Centro Medico Nacional Siglo XXI, IMSS (proto- depleted with lysis buffer (Sigma-Aldrich, USA) and incu-
col numbers R-2016-785-050 and R-2017-785-114), and all bated with anti-CD4 MicroBeads (for T CD4 cells, Miltenyi
mouse measurements were in accordance with the approved Biotec); they were selected with the magnetically activated
guidelines established by Mexico (Norma Oficial Mexicana cell sorting (MACS) system (Miltenyi Biotec, Germany)
NOM-062-ZOO-1999) and the NIH Guide for the Care through positive selection using LS columns (Miltenyi Bio-
and Use of Laboratory Animals. MRL/MpJFASlpr (MRL/lpr) tec). Single-cell suspensions of CD4+ T cells were incubated
mice were purchased from the Jackson Laboratory (Maine, with fluorescently labeled antibodies specific for CD44,
USA), and C57BL/6 mice were purchased from the Instituto CD62L, CXCR5, and PD1 in staining buffer (PBS with 0.5%
Nacional de Ciencias Médicas y Nutrición (CDMX Mexico). BSA) for 20 min at 4°C. Further, the cells were incubated with
Journal of Immunology Research 3

DAPI to select living cells (DAPI−), washed, and Tnaïve using a Foxp3/transcription factor staining buffer set
(CD44−CD62L+) and TFH cells (CXCR5+PD1+) were iso- (eBioscience, USA) or an Intracellular Fixation and Perm-
lated. Cell sorting was performed using a FACS Influx Sorter abilization Buffer Set (eBioscience, USA) for the latter two.
(BD Biosciences). The purity of sorted cells ranged from 95% All FACS data were acquired with an MACSQuant Analyzer
to 98%. 10 flow cytometer (Miltenyi Biotec, Germany) and analyzed
using the FlowJo software (Tree Star, USA).
2.7. RT-PCR for Prolactin Receptor Isoforms. To determine
the expression of PRL receptor isoforms, Tnaïve and TFH cells 2.12. Analysis of STATs and AKT Phosphorylation. TFH cells
from 18-week-old MRL/lpr mice were purified by sorting differentiated in vitro were left to rest for 8 h in medium, then
with a BD Influx Cytometer. Real-time PCR was performed TFH and Tnaïve cells were incubated with PRL (50 ng/mL) for
using the following primers synthesized by Integrated DNA 30 min and fixed with 1x BD Phosflow Lyse/5x FIx Buffer
Technologies (IDT, USA): β-actin (housekeeping control) (BD Biosciences, USA) for 10 min. Cells were permeabilized
5 ′ –GAGGAGGCTCTGGTTCAACA–3 ′ (left) and 5 ′ – with Perm Buffer III from BD Phosflow (BD Biosciences)
CAGTAAATGCCACGAACGAA–3 ′ (right). To determine to determine STAT3, STAT1, STAT5, and AKT phosphory-
the PRL receptor isoforms, three primers were used: com- lation. Cells were washed with FACS buffer and incubated at
mon 5 ′ –AAGCCAGACCATGGATACTGGAG–3 ′ (left), 4°C for 30 min with the antibodies for flow cytometry analy-
long isoform 5 ′ –AGCAGTTCTTCAGACTTGCCCTT–3 ′ sis. Data were acquired using an MACSQuant Analyzer 10
cytometer (Miltenyi Biotec) and analyzed with FlowJo soft-
(right), and short isoform 5 ′ –TTGTATTTGCTTGCAG
ware (Tree Star, USA). To confirm STAT3 activation, TFH
AGCCAGT–3 ′ (right). The samples were run in the LightCy- and Tnaïve cells were preincubated for 30 min in basal
cler II thermal cycler (Roche, Germany) under the following medium alone or with 10 mM of the STAT3 inhibitor (Stat-
conditions: one cycle at 95°C for 15 min, 40 cycles at 95°C for tic, Cell Signaling Technology, USA).
10 s, 61°C for 30 s, and 72°C for 30 s, and one cycle at 72°C for
30 s. The relative expression was analyzed using the 2−ΔΔCt 2.13. Statistical Analysis. The Shapiro–Wilk normality test
method. The murine breast cancer cell line EpH4 1424 was was used to determine the distribution of data. The results
used as a positive control for the expression of the long and were expressed as the mean and standard deviation. Differ-
short PRL receptor isoforms. ences between groups were determined using the ANOVA
test. A p value < 0.05 was considered significant; statistical
2.8. Prolactin Receptor Expression (Protein). CD4+ T cells analysis of the data was performed using the SPSS Statistics
from 9- and 18-week-old mice were isolated from the spleen 27 software.
with the CD4+ T Cell Isolation Kit (Miltenyi Biotec, Ger-
many) and stained with anti-mouse PRL receptor, anti-
CD4, anti-CD44, and anti-CD62L for naïve T cells or anti- 3. Results
CD4, anti-CXCR5, and anti-PD1 TFH cells. 3.1. TFH Cells Increase in Mice That Develop Lupus. Murine
models of SLE spontaneously increase the formation of GC,
2.9. Purification of Tnaïve. Nine-week-old mice were eutha-
which, however, has not been further explored. Seeking for
nized and spleen cells were collected with cold RPMI, and
an explanation for this observation and given that the
blood cells were depleted with lysis buffer. Naïve T cells were
increased GC formation correlates with prodromal SLE fea-
isolated from the spleen using a CD4+ naïve T cell (Tnaïve)
tures, we measured the percentage of TFH cells in splenocytes
Isolation Kit (BioLegend, USA), following the manufac-
of 9- and 18-week-old mice, in which the disease activity was
turer’s instructions.
determined by measuring the concentration of anti-dsDNA
2.10. TFH Differentiation. Tnaïve cells were differentiated to antibodies (IgG) and proteinuria. The 18-week-old MRL/lpr
TFH cells in the presence of the following antibodies and mice showed significantly elevated serum concentrations of
cytokines: anti-CD3, 2.5 μg/mL; anti-CD28, 5 μg/mL; anti- anti-dsDNA antibodies (9:7 ± 3:97 μg/mL) and proteinuria
IFN-γ, 10 μg/mL; anti-IL4, 10 μg/mL; anti-TGFβ, 20 μg/mL; (100 ± 7:56 mg/dL) compared with 9-week-old mice (anti-
IL6, 10 ng/mL; IL21, 10 ng/mL; and with or without dsDNA 0:82 ± 1:18 μg/mL; proteinuria 4:29 ± 6:50 mg/dL).
50 ng/mL of PRL for 48 h at 37°C and 5% CO2. We did not or we barely detected anti-dsDNA antibodies
and proteinuria in the control strain (C57BL/6)
2.11. Flow Cytometry of In Vitro Differentiated TFH Cells. For (Figures 1(a) and 1(b)).
OX40 expression, differentiated TFH cells in vitro or spleno- We found that the 18-week-old MRL/lpr mice had a sig-
cytes from mice that underwent different treatments were nificantly higher percentage of TFH (6:35% ± 1:98%) com-
stained with Ghost Red (viability), anti-CD4, anti-CXCR5, pared with the 9-week-old MRL/lpr mice (0:71% ± 0:43%)
anti-PD1, and anti-OX40. For intracellular IL21, cells were and C57BL/6 mice (9 weeks old, 1:22% ± 0:90%; 18 weeks
incubated with 1x Cell Stimulation cocktail (Invitrogen, old, 0:88% ± 0:08%) (Figures 1(c) and 1(d)). A similar behav-
USA) and 1x Protein Transport Inhibitor cocktail (Invitro- ior was observed with the cell absolute numbers; 18-week-old
gen, USA) for 5 h at 37°C and 5% CO2. Cells were stained MRL/lpr mice had a higher number of TFH cells
with Ghost Red, as well as anti-CD4, anti-CXCR5, and (2:16 ± 0:97 × 106 cells/spleen) compared with 9-week-old
anti-PD1 antibodies. To stain for intracellular proteins MRL/lpr mice (0:40 ± 0:24 × 106 cells/spleen) and C57BL/6
(BCL/6, Ki-67, and IL21), cells were fixed and permeated mice (9 weeks old, 0:26 ± 0:19 × 106 cells/spleen; 18 weeks
4 Journal of Immunology Research

15 ⁎⁎⁎ 150

Ab anti-dsDN A(𝜇g/mL)

Proteinuria (mg/dL)
⁎⁎⁎
10 100

5 50

0 0
9 weeks 18 weeks 9 weeks 18 weeks

C57 BL/6 C57 BL/6

MRL/lpr MRL/lpr
(a) (b)

150k 200k Lymphocytes

Singlets 150k
100k
FSC-H

SSC-A

100k
50k

50k 10.0
0
⁎⁎⁎

Follicular T cells (%)

0
0 30k 60k 90k 120k 0 30k 60k 90k 120k
7.5
FSC-A FSC-A

CD4+ Live 5.0

120k 120k 94.7

90k 90k
2.5
FSC-A

FSC-A

60k 60k

30k 30k 0.0

0 0
C57BL/6 MRL/lpr
103 4
105 103 4
0 10 0 10 105

CD4 Ghost Red

9 weeks
18 weeks
105 105
TFH
TFH
4.39
4.39
104
4
BCL6
PD-1

10

103

103
102
0
0
0 103 104 105 0 103 104 105

CXCR5 CXCR5

(c) (d)
4 15
Follicular T cell ( 106)
106)

⁎⁎⁎
3
Follicular T cell (

10
2
5
1

0 0
C57BL/6 MRL/lpr 0 5 10 15

9 weeks Ab anti-dsDNA
18 weeks r = 0.9238
p < 0.0001
(e) (f)

Figure 1: Continued.
Journal of Immunology Research 5

15

Follicular T cell ( 106)

10

0
5 10 15 20
Age (weeks)

r = 0.9052
p < 0.0001
(g)

Figure 1: TFH cells increase in mice that develop lupus. In 9- and 18-week-old mice of the C57BL/6 and MRL/lpr strains, the following were
determined: (a) concentration of anti-dsDNA antibodies by ELISA and (b) levels of proteinuria using a test strip. (c) Demonstration of the
gating strategy for the flow cytometry analysis of TFH cells. Doublets were excluded by gating on FSC-H×FSC-A lymphocytes which were
identified on the basis of their scatter properties (FSC-A×SSC-A plot), and live cells were gated in the Ghost Red−. The gate of CD4+ T
cells was selected. The CXCR5+PD1+ or CXCR5+BCL6+ population represents TFH cells. (d) Percentage and (e) absolute number of TFH
cells. Each determination was made in eight mice. Pooled data are presented as the mean ± SD; ∗∗∗ p < 0:001 using ANOVA. Pearson’s
correlation between absolute number of TFH cells and (f) anti-dsDNA antibody concentration and (g) age.

old, 0:11 ± 0:01 × 106 cells/spleen) (Figure 1(e)). Therefore, more significant in mice treated with metoclopramide
in lupus-prone MRL/lpr mice, the increased formation of (348:84 ± 52:71 × 106 cells), while the numbers of spleno-
GCs may be at least partially explained by the increased for- cytes in the bromocriptine condition (138:80 ± 25:95 × 106
mation of TFH cells. Indeed, TFH cell numbers correlated with cells) were closer to the 9-week baseline
autoantibody concentrations and with age (Figures 1(f) and (92:27 ± 12:45 × 106 cells) (Figure 3(b)). A similar observa-
1(g)). tion was made for the absolute numbers of CD4+ T cells
and TFH cells, as well as for activated TFH OX40+ cells and
3.2. Tnaïve and TFH Cells Express the Long Isoform of the PRL TFH IL21+ cells. For all these populations, the highest abso-
Receptor. To explore whether the formation of TFH cells may lute numbers were from mice treated with metoclopramide
be influenced by PRL, we determined the expression pattern and the lowest for the bromocriptine condition. CD4+ T cells
of the PRL receptor between lupus-prone and control mice, are composed of the following: metoclopramide 63:58 ±
reporting the expression of the PRL receptor as the fold 6:15 × 106 cells, PBS 44:74 ± 18:71 × 106 cells, and bromo-
change in TFH cells with respect to that of 9-week-old Tnaïve criptine 24:57 ± 1:22 × 106 cells (Figure 3(c)). TFH popula-
cells. We did not find an increase in the PRL receptor expres- tions are composed of the following: metoclopramide
sion in TFH cells of 9- and 18-week-old C57BL/6 mice. On (TFH7:69 ± 2:66; TFH OX40+2:35 ± 0:60; TFH
the contrary, the MRL/lpr strain showed augmented expres- + 6
IL21 0:14 ± 0:07 × 10 cells), PBS (TFH3:77 ± 2:72; TFH
sion, both at 9 (2:85 ± 0:56-fold change) and at 18 weeks of OX40+1:03 ± 0:43; TFH IL21+0:05 ± 0:01 × 106 cells), and
age (3:87 ± 0:33-fold change), with TFH cells of the 18- bromocriptine (TFH2:20 ± 0:59; TFH OX40+0:70 ± 0:11; TFH
week-old mice exhibiting the greatest expression
IL21+0:03 ± 0:01 × 106 cells) (Figures 3(d)–3(f)). Therefore,
(Figures 2(a)–2(c)); we have previously made a similar obser-
the increased number of splenocytes observed in each condi-
vation in B cell splenocytes [24]. We observed that both Tnaïve
tion mirrors the numbers of each of these CD4 populations
and TFH cells of MRL/lpr mice only express the long isoform
that participate in GC formation. We did not observe differ-
of the PRL receptor (Figure 2(d)).
ences in the absolute numbers of these cells between PBS
treated or untreated MRL/lpr mice.
3.3. Prolactin Increases the Absolute Number of TFH OX40+
We determined the expression of BCL6 in TFH cells,
Cells and IL21-Secreting Cells. We previously reported in
observing an increase only in mice treated with metoclopra-
MRL/lpr mice that pharmacologically raising serum PRL
mide (Figure 3(g)). Furthermore, the serum levels of IL21
levels with metoclopramide exacerbates the clinical manifes-
were also more elevated in mice treated with metoclopramide
tations of SLE, with an increase in autoantibody concentra-
(Figure 3(h)). Meanwhile, we did not observe changes in the
tion, as well as proteinuria [23, 24]. To determine whether
numbers of these populations in C57BL/6 mice (Figure S1,
PRL could affect the number of TFH cells, as well as their acti-
Supplementary Materials).
vation in vivo, we treated MRL/lpr mice with metoclopra-
mide (to increase PRL levels), bromocriptine (to decrease
PRL levels), or PBS (Figure 3(a)). We found that the absolute 3.4. Prolactin Does Not Affect the Survival or Differentiation
number of splenocytes spontaneously increased with age, as of TFH Cells. PRL has been reported to increase survival in
it was observed even in MRL/lpr mice treated with PBS (16 immature B cells of mice that develop SLE [40]. We deter-
weeks 235:36 ± 78:21 × 106 cells). Still, this increase was mined whether PRL could favor the survival and/or
6 Journal of Immunology Research

250K 250K

200K Singlets 200K Lymphocytes

150K 150K
FSC-H

SSC-A
100K 100K

50K 50K

0 0
0 50K 100K 150K 200K 250K 0 50K 100K 150K 200K 250K

FSC-A FSC-A

250K 250K
CD4 Live
200K 200K

150K 150K
FSC-A

FSC-A

100K 100K
6
50K 50K

PRL receptor (fold change )

⁎⁎⁎
0 0

0 103 104 105 0 103 104 105 ⁎⁎⁎

4
CD4 Ghost Red

105
Tnaive
2
104
CD62L

103
0
Tnaïve TFH TFH Tnaïve TFH TFH
0
9w 18 w 9w 18 w
0 103 104 105
C57BL6
CD44 MRL/lpr
(a) (b)
MRL/lpr
0.5

TFH 18 weeks
Relative expression (mRNA)

0.4

0.3
TFH 9 weeks
0.10
Tnaïve 9 weeks
0.05

Isotype
0.00
T TFH
0 103 104 105
Short
PRL receptor Long
(c) (d)

Figure 2: PRL receptor expression on Tnaïve and TFH cells. Splenocytes from 9- and 18-week-old C57BL/6 and MRL/lpr mice were stained
with anti-CD4, anti-CXCR5, and anti-PD1 for TFH cells (as shown in Figure 1(c)) and with anti-CD4, anti-CD44, and anti-CD62L for
Tnaïve cells, then cells were stained with an anti-PRL receptor antibody. (a) Demonstration of the gating strategy for the flow cytometry
analysis of Tnaïve cells. Doublets were excluded by gating on FSC-H×FSC-A, lymphocytes were identified on the basis of their scatter
properties (FSC-A×SSC-A plot), and live cells were gated in the Ghost Red−. The gate of CD4+ Tnaïve was selected (CD62L+ CD44−). (b)
Expression of the PRL receptor is reported as the fold change in receptor expression with respect to PRL receptor expression in Tnaïve
cells. (c) Representative histogram of PRL receptor expression in MRL/lpr mouse cells. The measurement was carried out in duplicate in
six mice per group. Pooled data are presented as the mean ± SD; ∗∗∗ p < 0:001 using ANOVA. (d) Tnaïve and TFH cells from 18-week-old
MRL/lpr mice were purified by Sort, and the isoform of the PRL receptor was determined by real-time (RT-) PCR. The murine breast
cancer cell line EpH4 1424 was used as a positive control for the expression of the long and short PRL receptor isoforms (not shown).
Two different experiments were performed; in each experiment, a pool of cells isolated from three mice was used.
Journal of Immunology Research 7

9 weeks old 16 weeks old 500

⁎⁎⁎
400

Splenocytes ( 106)
0 1 2 3 4 5 6 7 (weeks) 300
Treatment
200 ⁎

100

0
9 16 PBS Bromo Meto

Weeks Treatment

(a) (b)
80 15

⁎⁎
CD4+T Cell ( 106)

60

6)
10

TFHcells (
40
5
20

0 0
9 16 PBS Bromo Meto 9 16 PBS Bromo Meto

Weeks Treatment Weeks Treatment

(c) (d)
4000 300
⁎⁎⁎
TFHOX40+ cells ( 103)

⁎⁎⁎
TFHIL21+ cells ( 103)

3000
200
2000
100
1000

0 0
9 16 PBS Bromo Meto 9 16 PBS Bromo Meto

Weeks Treatment Weeks Treatment

(e) (f)
2000 300
⁎⁎⁎
1500
IL21 (pg/ml)
BCL6 (MFI)

200
1000
100
500

0 0
16 weeks PBS Bromo Meto 9 16 PBS Brom Meto

Weeks Treatment

(g) (h)

Figure 3: Metoclopramide increases the absolute number of TFH populations in MRL/lpr mice. Nine-week-old MRL/lpr mice were treated
with metoclopramide (meto), bromocriptine (bromo), or PBS or were left without intervention (left column marked by age in weeks) for 6
weeks. (a) Flow chart of treatment strategy. At the end of the treatment, cells were labeled with anti-CD4, anti-CXCR5, anti-PD1, anti-
OX40, anti-BCL6, or anti-IL21 antibodies. The graphs show the absolute number of (b) splenocytes, (c) CD4+ T cells, (d) TFH cells, (e)
OX40+ TFH cells, and (f) IL21+ TFH cells. (g) Expression of BCL6 in TFH cells. (h) IL21 concentration in serum. For the determination of
TFH IL21+, the cells were stimulated with stimulation cocktail (PMA ionomycin) for 5 h and then stained. MFI: mean fluorescent intensity.
Eight mice per condition were used. Pooled data are presented as the mean ± SD; ∗∗∗ p < 0:001, ∗∗ p < 0:01, and ∗ p < 0:05 using ANOVA.

differentiation of TFH cells as a mechanism to explain their the expression (mean fluorescence intensity, MFI) of BCL6,
increased numbers. For this, we isolated CD4 Tnaïve cells ICOS, and CXCR5 (Figure 4(c)); although a greater differen-
from 9-week-old C57BL/6 and MRL/lpr mice and induced tiation to TFH was observed in cells from MRL/lpr mice com-
TFH differentiation in culture. We did not find differences pared with cells from C57BL/6 mice. When we determined
in the percentage of TFH cells differentiated without PRL the survival of TFH cells after the differentiation assay, we
(C57BL/6 9:26% ± 3:50%; MRL/LPR 13:12% ± 3:26%) and did not find significant differences between the percentage
with PRL (C57/BL6 8:56% ± 2:35%; MRL/lpr 12:50% ± 4:41 of live differentiated TFH cells without PRL (C57BL/6 38:69
%) (Figures 4(a) and 4(b)) nor did we find a difference in % ± 4:83%; MRL/lpr 45:49% ± 4:72%) and with PRL
8 Journal of Immunology Research

Medium
105
TFH
104 0.98

25 103
⁎
TFH CXCR5+PD1+(%)

0
20
0 103 104 105

PD1
15
Differentiated Differentiated+PRL
10
105 105
TFH TFH
5 11.5 12.8
104 104

0 103 103
C57BL/6 MRL/lpr 0 0

Differentiated 0 103 104 105 0 103 104 105

Differentiated+PRL
CXCR5
(a) (b)

600 200 250

200
150
CXCR5 (MFI)

400

ICOS (MFI)
BCL6 (MFI)

150
100
100
200
50
50

0 0 0
Differentiated Differentiated Differentiated Differentiated Differentiated Differentiated
+PRL +PRL +PRL
(c)
80 60
⁎⁎
60
Survival (%)

40
Ki-67 (%)

40
20
20

0 0
C57BL/6 MRL/lpr C57BL/6 MRL/lpr

Differentiated Differentiated
Differentiated+PRL Differentiated+PRL
(d) (e)

Figure 4: Differentiation and survival of TFH cells in the presence of PRL. Tnaïve cells from 9-week-old C57BL/6 and MRL/lpr mice were
purified by MACS and differentiated to TFH in the presence and absence of PRL for 48 h, before staining with a viability marker (Ghost
Red) and anti-Ki-67, anti-CD4, anti-CXCR5, anti-BCL6, and anti-ICOS antibodies. The surface CXCR5+PD1+ population represents TFH
cells. (a) Percentage of differentiation to TFH in vitro. (b) Zebra plot of one representative experiment. (c) Expression of BCL6, CXCR5,
and ICOS (MFI) in TFH cells. (d) Percentage of TFH cell survival. (e) Percentage of proliferation (Ki-67+). Six different experiments were
performed; each experiment was done in triplicate. Pooled data are presented as the mean ± SD; ∗∗ p < 0:01 and ∗ p < 0:05 using ANOVA.

(C57BL/6 36:62% ± 4:87%; MRL/lpr 45:33% ± 5:33%) 3.5. Prolactin Activates TFH Cells. To assess whether TFH cells
(Figure 4(d)). Additionally, the cells differentiated to TFH were more active upon PRL treatment, we measured the
from MRL/lpr mice presented slightly better survival than expression of OX40 and IL21, both molecules serving as acti-
those from C57BL/6 mice. We did not observe differences vation markers of TFH cells. We found that the TFH cells dif-
in the percentages of proliferating cells (Figure 4(e)). ferentiated in the presence of PRL presented a statistically
Journal of Immunology Research 9

2000 ⁎⁎⁎ ⁎⁎⁎

100
⁎⁎⁎

OX40 expression (MFI)

1500 80
⁎⁎⁎

TFHOX40+ (%)
1000 60

40
500
20
0
0
C57BL/6 MRL/lpr
C57BL/6 MRL/lpr
Differentiated
Differentiated
Differentiated+PRL
Differentiated+PRL
MRL/lpr
MRL/lpr
Differentiated
105 105
+ PRL OX40 OX40
70.4 85.1
104 104

103 103
Differentiated
0 0
0 103 104 105
0 103 104 105 0 103 104 105

OX40 Differentiated Differentiated+PRL

(a) (b)
150 20
⁎⁎⁎ ⁎⁎⁎
⁎⁎⁎ 15 ⁎⁎⁎
TFHOX40+ (%)

100
IL21 (MFI)

10
50
5

0 0
C57BL/6 MRL/lpr C57BL/6 MRL/lpr
Differentiated Differentiated
Differentiated+PRL Differentiated+PRL
MRL/lpr
MRL/lpr
Differentiated 105 IL21 105 IL21
+ PRL 104 9.22 104 13.3
103 103
102 102
102 102
Differentiated
0 0
–102 –102
0 103 104 105
100 101 102 103 104 105 100 101 102 103 104 105

Differentiated Differentiated+PRL

(c) (d)

Figure 5: Continued.
10 Journal of Immunology Research

2.5 ⁎ 15
⁎ ⁎⁎⁎

pSTAT3 (fold change)

2.0

pSTAT3 (% )
10
1.5

1.0 5
0.5

0.0 0
Tnaive TFH Medium PRL Stattic+PRL

MRL/lpr MRL/lpr
MRL/lpr
Medium
105 pSTAT3 105 pSTAT3 105 pSTAT3
PRL
104
8.4 104 12.8 104
8.9
Stattic+PRL
MRL/lpr (TFH) 103 103 103

102 102 102

Stattic+PRL
0 0 0
100 101 102 103 104 105 100 101 102 103 104 105 100 101 102 103 104 105

PRL
Medium PRL Stattic+PRL
Medium

0 103 104 105

pSTAT3

(e) (f)

Figure 5: Activation and signaling of differentiated TFH cells in vitro in the presence of PRL. Tnaïve cells from 9-week-old C57BL/6 and
MRL/lpr mice were purified by MACS and differentiated to TFH with or without PRL. (a) Expression of OX40 (MFI) and representative
histograms of OX40 expression in TFH cells from MRL/lpr mice. (b) Percentage and representative Zebra plots of TFH OX40+ cells. (c)
Expression of IL21 (MFI) and representative histogram of IL21 expression in TFH cells from MRL/lpr mice. (d) Percentage and Zebra
plots of TFH IL21+ cells. (e, f) Tnaïve and TFH cells were preincubated for 30 min with the inhibitor of STAT3 (Stattic). For STAT3
inhibition, Tnaïve cells were differentiated to TFH, left to rest for 8 h, and then incubated for 30 min with PRL to subsequently determine
the MFI of pSTAT3 (histogram). For the plot, pSTAT3 is reported as fold change taken as baseline levels found in Tnaïve or TFH cells only
treated with medium. (f) Percentage and Zebra plots of pSTAT3. Six different experiments were performed; each experiment was done in
duplicate. Pooled data are presented as the mean ± SD; ∗∗∗ p < 0:0001 and ∗ p < 0:05 using ANOVA.

significant increase in MRL/lpr mice, determined by both cells (medium: 1.00-fold change, 8:46% ± 1:00%; PRL:
expression (MFI) and percentage of OX40 1.74-fold change, 11:02% ± 1:33%; Stattic: 1.01-fold change,
(1359:88 ± 172:05 MFI; 82:85% ± 4:20%), compared with 7:74% ± 1:06%) (Figure 5(e)). In addition, the PRL activity
the condition without PRL (1138 ± 76:87 MFI; 70:87% ± was more prominent in MRL/lpr TFH cells, since the inhib-
4:07%), likewise, for IL21, with PRL (97:36 ± 4:00 itor significantly reduced pSTAT3 only in the lupus-prone
MFI;11:90% ± 1:12%) versus without PRL (87:43 ± 1:70 strain. We did not observe pSTAT3 in TFH cells from
MFI;9:27% ± 0:61%). On the other hand, we did not observe C57BL/6 mice or in Tnaïve cells from any mice. In addition,
any difference in the C57BL/6 mouse cells (Figures 5(a)– PRL did not induce STAT1, STAT5, and AKT phosphory-
5(d)). Moreover, TFH cells derived from MRL/lpr mice lation in MRL/lpr mice (Figure S2, Supplementary
expressed more OX40 and IL21 than cells derived from Materials).
C57BL/6 mice at baseline.
4. Discussion
3.6. Prolactin Promoted STAT3 Phosphorylation in TFH Cells.
It is known that the long PRL isoforms signal through the The endocrine system produces hormones that regulate dif-
JAK-STAT and PI3K-AKT pathways [41, 42]. We deter- ferent systems, one of them being the immune system [43].
mined the signaling components associated with the PRL The bidirectional interactions between the endocrine and
receptor upon activation with recombinant PRL in TFH cells immune systems play critical roles in the maintenance of
differentiated in vitro. We measured STAT1, STAT3, STAT5, homeostasis. Disturbing mutual communication between
and AKT phosphorylation via flow cytometry. We found that these systems might initiate or exacerbate the development
PRL induced phosphorylation of STAT3 (pSTAT3) only in of a wide variety of diseases, such as autoimmune thyroid dis-
TFH cells derived from MRL/lpr mice and confirmed this ease [44], rheumatoid arthritis [45], Sjögren syndrome [46],
PRL activity with an inhibitor of STAT3 (Stattic). The level and SLE [47]. Patients with SLE, as well as experimental
of pSTAT3 was measured as a fold change (with respect to model mice of the disease (MRL/lpr, NZB/W), show an
TFH cells treated with medium) and percentage of positive increase in serum PRL levels associated with the activity of
Journal of Immunology Research 11

the disease and/or the concentration of IgG autoantibodies upon PRL treatment, explaining why the effect of PRL on
[17, 18, 24]. Furthermore, the activity of lupus has also been Tnaïve cells and their differentiation to TFH was not observed.
associated with an increase in TFH cells [36, 48, 49], a subset The increase in the absolute number of TFH cells and the
of helper CD4 T cells that play a crucial role in the generation expression of BCL6 (MFI) in vivo may be rather due to an
of antibodies. Indeed, dysfunctional TFH cells can activate indirect effect of PRL. PRL could be acting on other cells that
autoantibody-producing B cells that cause SLE [50]. are helping TFH cells to differentiate. For example, it is known
Although these studies support that PRL influences TFH cell that B cells (follicular and marginal zone) express the PRL
function in SLE, with a concomitant rise of autoantibodies, receptor and that this expression increases when PRL con-
the link between PRL and TFH cells is still not clear. centrations rise [24]. On the other hand, it has been reported
In this study, we present new evidence of the importance that IL6 secreted by B cells is important for the differentiation
of PRL in the development of SLE by increasing the absolute of TFH cells [25]; thus, it will be important to demonstrate, in
number of TFH cells, the activation of TFH cells, and IL21 future tests, if PRL can increase IL6 secretion in B cells, thus
secretion in lupus-prone mice. This could favor an uncon- favoring the differentiation of TFH cells.
trolled response of GCs, faulty tolerance, and an increase in It could also be due to the effect that PRL may have on
the production of autoantibodies implicated in the pathogen- other hormones that also influence specific components of
esis of the disease. As it has been observed in the B6.MRL- the immune responses, such as the thyroid-stimulating hor-
Faslpr (B6.lpr) and BXD2 strains, the increase in TFH corre- mone (TSH). The elevated TSH levels increased the
lates positively with total IgG concentration in serum, as well mitogen-induced proliferative response of mouse lympho-
as with anti-dsDNA antibody levels [51, 52]. We demon- cytes [60], as well as the percentage of CD4+ T cells [61]. Fur-
strated here that lupus-prone MRL/lpr mice also presented thermore, serum levels of TSH correlate positively with those
a positive correlation between the absolute number of TFH of PRL [62], and 11.6% of patients with SLE present elevated
cells and the concentration of anti-DNA IgG isotype autoan- levels of TSH [63]. This suggests that in our in vivo tests, the
tibodies, as well as a correlation with age. Increased serum increase in the number of TFH cells in the mice treated with
PRL levels in these SLE-developing mice are associated with metoclopramide may be due both to an indirect effect of
disease exacerbation [24, 53, 54]. Previously, we have also PRL on other cells and to the effect of other hormones such
reproduced the exacerbation of the disease by pharmacolog- as TSH on CD4+ T cells. Therefore, it will be important to
ically raising serum PRL concentrations with metoclopra- study the effect of TSH on TFH cells.
mide [23, 24]. Here, this same treatment induced an It is probable that the effect of PRL directly occurs in
increase in the absolute number of CD4+ T cells and TFH cells that are already differentiated and/or activated where
cells. Conversely, treating mice with an antagonist of the the expression of the receptor is greater. The costimulatory
secretion of PRL (bromocriptine) decreased the absolute roles of PRL in the in vitro activation of T cells and B cells
number of these cells with respect to mice treated with PBS have been previously reported [5, 64]. In addition, PRL
or without treatment; this behavior was only observed in promotes differentiation into CD4+ T-bet+ T cells [12],
lupus-prone mice. This increase in TFH cells may be due to CD4+ Eomes+ T cells [6], and NK cells [8]. TFH cells have
an increase in the differentiation of TFH cells, as the expres- a higher expression of the receptor with respect to Tnaïve
sion of BCL6, the master transcription factor of TFH cells cells. This expression increases with age and with the man-
[55], was increased only in mice treated with metoclopra- ifestations of the disease in mice that develop SLE, as seen
mide. This increase in the absolute number of TFH cells could for T cells from patients with SLE, where the T cells express
give us at least a partial explanation for the association higher levels of the receptor than T cells from healthy
between high levels of PRL and the increase in autoantibodies subjects [65, 66].
of the IgG isotype in patients or mice with SLE, as the uncon- In this work, we demonstrated that TFH cells exclusively
trolled accumulation of TFH cells might activate autoreactive express the long isoform of the PRL receptor, finding that
B cells to produce excessive autoantibodies that cause auto- PRL could participate in signaling through STAT3 in these
immune responses [50, 56]. cells. An extensive body of evidence links STAT3 with auto-
In different reports, it has been shown that PRL is an immune diseases. Most of this evidence is related to the
important factor for both survival and proliferation of differ- capacity of STAT3 to influence the differentiation of lym-
ent cell types [57, 58]. It has been demonstrated that PRL is phoid cells, such as Th17 and Treg CD4+ T cells [67]. Stattic
an important factor for both the survival and proliferation has also been used to delay the onset of disease in MRL/lpr
of early T-cell precursors, such as CD25+CD4−CD8− double mice, reducing the levels of clinical hallmarks of SLE, such
negative cells [10], as well as for the protection of thymocytes as nephritis, renal and skin lesions, proteinuria, and serum
from glucocorticoid-induced apoptosis [59]. However, in this autoantibodies [68, 69]. This increase in the phosphorylation
work, the prosurvival effects of PRL were not observed in TFH of STAT3 when incubating TFH cells with PRL could explain
cells differentiated from mature Tnaïve cells, as happens in the the role of PRL in increasing the percentage (in vitro) and
immature B cells of these mice [40]. In addition, there was no absolute number (in vivo) of IL21-secreting TFH cells (TFH
effect on the differentiation and proliferation of TFH cells, IL21+). This is consistent with the observations that the
despite the fact that Tnaïve and TFH cells expressed the PRL increased IL21 mRNA expression in CD4+ T cells from SLE
receptor; however, receptor expression was lower in Tnaïve patients is dependent on the activation of STAT3 [70, 71]
cells. Furthermore, we did not find evidence of STAT3 activa- and that STAT3 directly binds the IL21 promoter [72, 73].
tion in Tnaïve cells, as this kinase was not phosphorylated Furthermore, in mice treated with metoclopramide, the
12 Journal of Immunology Research

JAK JAK
P

STAT3
P

STAT3
STAT3
P

P STAT3
STAT3
MRL/lpr OX40 IL21

ST
JA
K Germinal center

P
AT

P
3
JA
K

TFH cells

Plasma cells

PRL-R IL21R

Autoantibodies OX40

PRL IL21

Figure 6: Working model of the mechanism of action of PRL in TFH cells from mice that developed SLE. TFH cells from lupus-prone mice
display increased levels of PRL receptor expression restrictive to the long isoform, together with higher levels of serum PRL. This combination
results in the heightened activation of STAT3, as well as the increase in OX40 expression and IL21 secretion, which increase the activation of
TFH cells favoring the formation of germinal centers, generation of autoreactive plasma cells and increased levels of autoantibodies, with
enhanced tissue damage due to the immune complex deposition that characterizes SLE.

serum levels of IL21 were increased. IL21 serves as a “helper” 5. Conclusions

cytokine to stimulate B cells through interacting with IL21R.
IL21 enhances murine B-cell proliferation, IgG class switch- Collectively, our data suggest that PRL acts on TFH cells that
ing, and plasmablast differentiation [74, 75]. Therefore, the express the long isoform of the receptor and could participate
increase in IL21 in lupus-prone mice could favor the genera- in signaling through STAT3. We also observed an increase in
tion of autoreactive plasma cells and the increase in the number and activation of TFH cells that may favor the for-
autoantibodies. mation of GC, interfere with tolerance, and facilitate the gen-
Another effect of PRL on TFH cells was an increase of TFH eration of autoreactive plasma cells and the secretion of
OX40+ cells. OX40 is transiently induced following TCR autoantibodies. Therefore, in future studies, it will be impor-
engagement after antigen (Ag) recognition. Many factors tant to assess the influence of PRL on the GCs, as well as the
are involved in the kinetics of OX40 expression, including interaction of B cells and TFH in an environment featuring
IL21 [76]. IL21 acts in an autocrine way in TFH cells [77]; high levels of PRL, to better understand the role of PRL in
thus, the PRL-dependent increase in the percentage and GC formation and to define the most important steps in
number of TFH IL21+ cells, as well as the serum levels of the pathogenesis of SLE that could be targeted by antagonis-
IL21, could favor an increase in the percentage of activated tic molecules (Figure 6).
TFH cells (OX40+). However, it has also been reported that
STAT3 plays a direct regulatory role in OX40 mRNA expres- Data Availability
sion in CD4+ T cells [78]. Similarly, STAT3 enhances T cell
survival by upregulating OX40, BCL2, and Fas ligand [76]. All data included in this study are available upon request by
Therefore, the PRL-mediated increase of OX40 on TFH cells contact with the corresponding author.
could be a direct effect or mediated through IL21. Further-
more, an increased percentage of OX40-expressing CD4+ T Ethical Approval
cells was found in SLE patients, in which it was an indicator
of disease activity [79], and the OX40L-OX40 axis was also All studies were approved by the Animal Care Committee of
found to contribute to lupus pathogenesis by promoting the the Instituto Nacional de Enfermedades Respiratorias
generation of TFH cells [80]. Therefore, PRL influences the “Ismael Cosio Villegas” and the Hospital de Pediatria, Centro
immune system in SLE exacerbating the activity of the dis- Medico Nacional Siglo XXI, IMSS (protocol number R-2016-
ease by increasing the number of OX40+ TFH cells and acti- 785-050 and R-2017-785-114); all experiments were per-
vating the OX40-OX40L axis. formed in accordance with approved guidelines established
Journal of Immunology Research 13

by Mexico (NOM-062-ZOO-1999) and the NIH Guide for [9] R. Barrett, C. A. Narasimhulu, and S. Parthasarathy, “Adrener-
the Care and Use of Laboratory Animals. gic hormones induce extrapituitary prolactin gene expression
in leukocytes-potential implications in obesity,” Scientific
Reports, vol. 8, no. 1, p. 1936, 2018.
Conflicts of Interest
[10] P. C. Carreno, R. Sacedon, E. Jimenez, A. Vicente, and A. G.
The authors declare no conflict of interest. Zapata, “Prolactin affects both survival and differentiation of
T-cell progenitors,” Journal of Neuroimmunology, vol. 160,
no. 1-2, pp. 135–145, 2005.
Acknowledgments [11] K. Takizawa, S. Kitani, F. Takeuchi, and K. Yamamoto,
We extend our gratitude to the flow cytometry core facility, “Enhanced expression of CD69 and CD25 antigen on human
peripheral blood mononuclear cells by prolactin,” Endocrine
the Coordinación de Investigación en Salud, CMN SXXI,
Journal, vol. 52, no. 5, pp. 635–641, 2005.
IMSS, for instrumental and technical support. This work
[12] A. Tomio, D. J. Schust, K. Kawana et al., “Prolactin can mod-
was supported by Fondo de Investigador en Salud IMSS
ulate CD4+ T-cell response through receptor-mediated alter-
(grant numbers FIS/IMSS/PROT/G16/1594 and FIS/IMS- ations in the expression of T-bet,” Immunology and Cell
S/PROT/G18/1804) and CONACYT (grant number A1-S- Biology, vol. 86, no. 7, pp. 616–621, 2008.
9789). [13] V. V. Borba, G. Zandman-Goddard, and Y. Shoenfeld, “Pro-
lactin and autoimmunity,” Frontiers in immunology, vol. 9,
Supplementary Materials p. 73, 2018.
[14] S. Praprotnik, N. Agmon-Levin, B. S. Porat-Katz et al., “Pro-
Figure S1: absolute number of TFH cells in C57BL/6 mice. lactin's role in the pathogenesis of the antiphospholipid syn-
Figure S2: PRL does not activate AKT, STAT1, and STAT5 drome,” Lupus, vol. 19, no. 13, pp. 1515–1519, 2010.
in MRL/lpr mice. (Supplementary Materials) [15] M. Fojtikova, J. Tomasova Studynkova, M. Filkova et al., “Ele-
vated prolactin levels in patients with rheumatoid arthritis:
References association with disease activity and structural damage,” Clin-
ical and Experimental Rheumatology, vol. 28, no. 6, pp. 849–
[1] N. T. Ashley and G. E. Demas, “Neuroendocrine-immune cir- 854, 2010.
cuits, phenotypes, and interactions,” Hormones and Behavior, [16] W. A. Wan Asyraf, M. S. Mohd Shahrir, W. Asrul et al., “The
vol. 87, pp. 25–34, 2017. association between serum prolactin levels and interleukin-6
[2] A. H. Schneider, A. Kanashiro, S. G. V. Dutra et al., “Estradiol and systemic lupus erythematosus activity,” Reumatismo,
replacement therapy regulates innate immune response in vol. 70, no. 4, pp. 241–250, 2018.
ovariectomized arthritic mice,” International Immunophar-
[17] G. G. Song and Y. H. Lee, “Circulating prolactin level in sys-
macology, vol. 72, pp. 504–510, 2019.
temic lupus erythematosus and its correlation with disease
[3] A. S. Porings, T. Lowin, B. Dufner, J. Grifka, and R. H. Straub, activity: a meta-analysis,” Lupus, vol. 26, no. 12, pp. 1260–
“A thyroid hormone network exists in synovial fibroblasts of 1268, 2017.
rheumatoid arthritis and osteoarthritis patients,” Scientific
[18] J. Yang, Q. Li, X. Yang, and M. Li, “Increased serum level
Reports, vol. 9, no. 1, p. 13235, 2019.
of prolactin is related to autoantibody production in sys-
[4] S. Gerlo, P. Verdood, E. L. Hooghe-Peters, and R. Kooijman, temic lupus erythematosus,” Lupus, vol. 25, no. 5,
“Modulation of prolactin expression in human T lymphocytes pp. 513–519, 2016.
by cytokines,” Journal of Neuroimmunology, vol. 162, no. 1-2,
pp. 190–193, 2005. [19] H. Orbach, G. Zandman-Goddard, M. Boaz et al., “Prolactin
and autoimmunity: hyperprolactinemia correlates with serosi-
[5] K. Chavez-Rueda, J. Hernandez, E. Zenteno, A. Leanos-
tis and anemia in SLE patients,” Clinical Reviews in Allergy and
Miranda, M. V. Legorreta-Haquet, and F. Blanco-Favela,
Immunology, vol. 42, no. 2, pp. 189–198, 2012.
“Identification of prolactin as a novel immunomodulator on
the expression of co-stimulatory molecules and cytokine secre- [20] J. Choi, S. T. Kim, and J. Craft, “The pathogenesis of systemic
tions on T and B human lymphocytes,” Clinical Immunology, lupus erythematosus-an update,” Current Opinion in Immu-
vol. 116, no. 2, pp. 182–191, 2005. nology, vol. 24, no. 6, pp. 651–657, 2012.
[6] C. Zhang, B. J. E. Raveney, H. Hohjoh, C. Tomi, S. Oki, and [21] V. V. Borba, G. Zandman-Goddard, and Y. Shoenfeld, “Exac-
T. Yamamura, “Extrapituitary prolactin promotes generation erbations of autoimmune diseases during pregnancy and post-
of Eomes-positive helper T cells mediating neuroinflamma- partum,” Best Practice & Research. Clinical Endocrinology &
tion,” Proceedings of the National Academy of Sciences of the Metabolism, vol. 33, no. 6, p. 101321, 2019.
United States of America, vol. 116, no. 42, pp. 21131–21139, [22] W. Li, A. A. Titov, and L. Morel, “An update on lupus animal
2019. models,” Current Opinion in Rheumatology, vol. 29, no. 5,
[7] P. Triggianese, C. Perricone, R. Perricone, and C. De Carolis, pp. 434–441, 2017.
“Prolactin and natural killer cells: evaluating the [23] M. V. Legorreta-Haquet, R. Flores-Fernandez, F. Blanco-
neuroendocrine-immune axis in women with primary infertil- Favela et al., “Prolactin levels correlate with abnormal B cell
ity and recurrent spontaneous abortion,” American Journal of maturation in MRL and MRL/lpr mouse models of systemic
Reproductive Immunology, vol. 73, no. 1, pp. 56–65, 2015. lupus erythematosus-like disease,” Clinical & Developmental
[8] D. M. Tufa, T. Shank, A. M. Yingst et al., “Prolactin acts on Immunology, vol. 2013, article 287469, pp. 1–11, 2013.
myeloid progenitors to modulate SMAD7 expression and [24] Y. Ledesma-Soto, F. Blanco-Favela, E. M. Fuentes-Panana
enhance hematopoietic stem cell differentiation into the NK et al., “Increased levels of prolactin receptor expression corre-
cell lineage,” Scientific Reports, vol. 10, no. 1, p. 6335, 2020. late with the early onset of lupus symptoms and increased
14 Journal of Immunology Research

numbers of transitional-1 B cells after prolactin treatment,” Immunology Research, vol. 2016, Article ID 3219017, 11
BMC Immunology, vol. 13, no. 1, p. 11, 2012. pages, 2016.
[25] T. Arkatkar, S. W. Du, H. M. Jacobs et al., “B cell-derived IL-6 [41] N. Binart, A. Bachelot, and J. Bouilly, “Impact of prolactin
initiates spontaneous germinal center formation during sys- receptor isoforms on reproduction,” Trends in Endocrinology
temic autoimmunity,” The Journal of Experimental Medicine, and Metabolism, vol. 21, no. 6, pp. 362–368, 2010.
vol. 214, no. 11, pp. 3207–3217, 2017. [42] P. A. Abramicheva and O. V. Smirnova, “Prolactin receptor
[26] I. G. Luzina, S. P. Atamas, and C. E. Storrer, “Spontaneous for- isoforms as the basis of tissue-specific action of prolactin in
mation of germinal centers in autoimmune mice,” Journal of the norm and pathology,” Biochemistry, vol. 84, no. 4,
Leukocyte Biology, vol. 70, no. 4, pp. 578–584, 2001. pp. 329–345, 2019.
[27] G. L. Moschovakis, A. Bubke, M. Friedrichsen, C. S. Falk, [43] S. Muthusami, B. Vidya, E. M. Shankar et al., “The functional
R. Feederle, and R. Förster, “T cell specific Cxcr5 deficiency significance of endocrine-immune interactions in health and
prevents rheumatoid arthritis,” Scientific Reports, vol. 7, disease,” Current Protein & Peptide Science, vol. 21, no. 1,
no. 1, p. 8933, 2017. pp. 52–65, 2020.
[28] J. M. Tas, L. Mesin, G. Pasqual et al., “Visualizing antibody [44] A. Antonelli, S. M. Ferrari, A. Corrado, A. Di Domenicanto-
affinity maturation in germinal centers,” Science, vol. 351, nio, and P. Fallahi, “Autoimmune thyroid disorders,” Autoim-
no. 6277, pp. 1048–1054, 2016. munity Reviews, vol. 14, no. 2, pp. 174–180, 2015.
[29] J. Shi, S. Hou, Q. Fang, X. Liu, X. Liu, and H. Qi, “PD-1 con- [45] C. Salliot, Y. Nguyen, G. Gusto et al., “Female hormonal expo-
trols follicular T helper cell positioning and function,” Immu- sures and risk of rheumatoid arthritis in the French E3N-EPIC
nity, vol. 49, no. 2, pp. 264–274.e4, 2018. cohort study,” Rheumatology (Oxford), Article ID keab101,
[30] S. Crotty, “T follicular helper cell biology: a decade of discovery 2021.
and diseases,” Immunity, vol. 50, no. 5, pp. 1132–1148, 2019. [46] L. Yang, W. Wei, X. He, Y. Xie, M. A. Kamal, and J. Li, “Influ-
[31] J. Y. Choi, A. Seth, M. Kashgarian et al., “Disruption of patho- ence of hormones on Sjögren's syndrome,” Current Pharma-
genic cellular networks by IL-21 blockade leads to disease ame- ceutical Design, vol. 24, no. 35, pp. 4167–4176, 2018.
lioration in murine lupus,” Journal of Immunology, vol. 198, [47] N. Coronel-Restrepo, I. Posso-Osorio, J. Naranjo-Escobar, and
no. 7, pp. 2578–2588, 2017. G. J. Tobón, “Autoimmune diseases and their relation with
[32] C. Pedros, Y. Zhang, J. K. Hu et al., “A TRAF-like motif of the immunological, neurological and endocrinological axes,”
inducible costimulator ICOS controls development of germi- Autoimmunity Reviews, vol. 16, no. 7, pp. 684–692, 2017.
nal center TFH cells via the kinase TBK1,” Nature Immunol- [48] B. Xu, S. Wang, M. Zhou et al., “The ratio of circulating follic-
ogy, vol. 17, no. 7, pp. 825–833, 2016. ular T helper cell to follicular T regulatory cell is correlated
[33] D. DiToro, C. J. Winstead, D. Pham et al., “Differential IL-2 with disease activity in systemic lupus erythematosus,” Clinical
expression defines developmental fates of follicular versus Immunology, vol. 183, pp. 46–53, 2017.
nonfollicular helper T cells,” Science, vol. 361, no. 6407, [49] J. Y. Choi, J. H. Ho, S. G. Pasoto et al., “Circulating follicular
2018. helper-like T cells in systemic lupus erythematosus: associa-
[34] X. Huang, H. Wu, H. Qiu et al., “The expression of Bcl-6 in cir- tion with disease activity,” Arthritis & Rhematology, vol. 67,
culating follicular helper-like T cells positively correlates with no. 4, pp. 988–999, 2015.
the disease activity in systemic lupus erythematosus,” Clinical [50] A. Seth and J. Craft, “Spatial and functional heterogeneity of
Immunology, vol. 173, pp. 161–170, 2016. follicular helper T cells in autoimmunity,” Current Opinion
[35] A. Makiyama, A. Chiba, D. Noto et al., “Expanded circulating in Immunology, vol. 61, pp. 1–9, 2019.
peripheral helper T cells in systemic lupus erythematosus: [51] Z. Zhang, R. Feng, L. Niu et al., “Human umbilical cord mesen-
association with disease activity and B cell differentiation,” chymal stem cells inhibit T follicular helper cell expansion
Rheumatology (Oxford), vol. 58, no. 10, pp. 1861–1869, 2019. through the activation of iNOS in lupus-prone B6.MRL-Faslpr
[36] C. Liu, D. Wang, Y. Song, S. Lu, J. Zhao, and H. Wang, mice,” Cell Transplantation, vol. 26, no. 6, pp. 1031–1042, 2017.
“Increased circulating CD4(+)CXCR5(+)FoxP3(+) follicular [52] Y. U. Kim, H. Lim, H. E. Jung, R. A. Wetsel, and Y. Chung,
regulatory T cells correlated with severity of systemic lupus “Regulation of autoimmune germinal center reactions in
erythematosus patients,” International Immunopharmacology, lupus-prone BXD2 mice by follicular helper T cells,” PLoS
vol. 56, pp. 261–268, 2018. One, vol. 10, no. 3, article e0120294, 2015.
[37] H. Xu, J. Liu, X. Cui et al., “Increased frequency of circulating [53] E. Peeva, J. Gonzalez, R. Hicks, and B. Diamond, “Cutting
follicular helper T cells in lupus patients is associated with edge: lupus susceptibility interval Sle3/5 confers responsive-
autoantibody production in a CD40L-dependent manner,” ness to prolactin in C57BL/6 mice,” Journal of Immunology,
Cellular Immunology, vol. 295, no. 1, pp. 46–51, 2015. vol. 177, no. 3, pp. 1401–1405, 2006.
[38] J. A. Bubier, T. J. Sproule, O. Foreman et al., “A critical role for [54] G. Recalde, T. Moreno-Sosa, F. Yudica et al., “Contribution of
IL-21 receptor signaling in the pathogenesis of systemic lupus sex steroids and prolactin to the modulation of T and B cells
erythematosus in BXSB-Yaa mice,” Proceedings of the National during autoimmunity,” Autoimmunity Reviews, vol. 17, no. 5,
Academy of Sciences of the United States of America, vol. 106, pp. 504–512, 2018.
no. 5, pp. 1518–1523, 2009. [55] D. Alterauge, J. W. Bagnoli, F. Dahlström et al., “Continued
[39] N. Gensous, M. Charrier, D. Duluc et al., “T follicular helper Bcl6 expression prevents the transdifferentiation of established
cells in autoimmune disorders,” Frontiers in Immunology, Tfh cells into Th1 cells during acute viral infection,” Cell
vol. 9, p. 1637, 2018. Reports, vol. 33, no. 1, p. 108232, 2020.
[40] R. Flores-Fernández, F. Blanco-Favela, E. M. Fuentes- [56] L. Petersone, N. M. Edner, V. Ovcinnikovs et al., “T cell/B cell
Pananá et al., “Prolactin rescues immature B-cells from apo- collaboration and autoimmunity: an intimate relationship,”
ptosis induced by B-cell receptor cross-linking,” Journal of Frontiers in Immunology, vol. 9, p. 1941, 2018.
Journal of Immunology Research 15

[57] C. M. Hyslop, S. Tsai, V. Shrivastava, P. Santamaria, and [72] L. Wei, A. Laurence, K. M. Elias, and J. J. O'Shea, “IL-21 is pro-
C. Huang, “Prolactin as an adjunct for type 1 diabetes immu- duced by Th17 cells and drives IL-17 production in a STAT3-
notherapy,” Endocrinology, vol. 157, no. 1, pp. 150–165, 2016. dependent manner,” The Journal of Biological Chemistry,
[58] H. Y. Lee, Y. T. Heo, S. E. Lee et al., “Short communication: vol. 282, no. 48, pp. 34605–34610, 2007.
retinoic acid plus prolactin to synergistically increase specific [73] L. Ysebrant de Lendonck, F. Eddahri, Y. Delmarcelle et al.,
casein gene expression in MAC-T cells,” Journal of Dairy Sci- “STAT3 signaling induces the differentiation of human
ence, vol. 96, no. 6, pp. 3835–3839, 2013. ICOS(+) CD4 T cells helping B lymphocytes,” PLoS One,
[59] N. Krishnan, O. Thellin, D. J. Buckley, N. D. Horseman, and vol. 8, no. 7, article e71029, 2013.
A. R. Buckley, “Prolactin suppresses glucocorticoid-induced [74] K. Ozaki, R. Spolski, R. Ettinger et al., “Regulation of B cell dif-
thymocyte apoptosis in vivo,” Endocrinology, vol. 144, no. 5, ferentiation and plasma cell generation by IL-21, a novel
pp. 2102–2110, 2003. inducer of Blimp-1 and Bcl-6,” Journal of Immunology,
[60] M. Provinciali, G. Di Stefano, and N. Fabris, “Improvement in vol. 173, no. 9, pp. 5361–5371, 2004.
the proliferative capacity and natural killer cell activity of [75] S. G. Tangye and C. S. Ma, “Regulation of the germinal center
murine spleen lymphocytes by thyrotropin,” International and humoral immunity by interleukin-21,” Journal of Experi-
Journal of Immunopharmacology, vol. 14, no. 5, pp. 865–870, mental Medicine, vol. 217, no. 1, 2020.
1992. [76] J. Willoughby, J. Griffiths, I. Tews, and M. S. Cragg, “OX40:
[61] M. Jaeger, Y. J. E. Sloot, R. T. Horst et al., “Thyrotrophin and structure and function - what questions remain?,” Molecular
thyroxine support immune homeostasis in humans,” Immu- Immunology, vol. 83, pp. 13–22, 2017.
nology, 2021. [77] F. Caprioli, M. Sarra, R. Caruso et al., “Autocrine regulation of
[62] L. K. Sharma, N. Sharma, A. K. Gadpayle, and D. Dutta, “Prev- IL-21 production in human T lymphocytes,” Journal of Immu-
alence and predictors of hyperprolactinemia in subclinical nology, vol. 180, no. 3, pp. 1800–1807, 2008.
hypothyroidism,” European Journal of Internal Medicine, [78] H. Hanieh, K. Masuda, H. Metwally et al., “Arid5a stabilizes
vol. 35, pp. 106–110, 2016. OX40 mRNA in murine CD4+T cells by recognizing a stem-
[63] R. Mader, S. Mishail, M. Adawi, I. Lavi, and R. Luboshitzky, loop structure in its 3’UTR,” European Journal of Immunology,
“Thyroid dysfunction in patients with systemic lupus erythe- vol. 48, no. 4, pp. 593–604, 2018.
matosus (SLE): relation to disease activity,” Clinical Rheuma- [79] M. N. Farres, D. S. Al-Zifzaf, A. A. Aly, and N. M. Abd Raboh,
tology, vol. 26, no. 11, pp. 1891–1894, 2007. “OX40/OX40L in systemic lupus erythematosus: association
[64] C. S. Hau, N. Kanda, Y. Tada, S. Shibata, S. Sato, and with disease activity and lupus nephritis,” Annals of Saudi
S. Watanabe, “Prolactin induces the production of Th17 and Medicine, vol. 31, no. 1, pp. 29–34, 2011.
Th1 cytokines/chemokines in murine Imiquimod-induced [80] C. Jacquemin, N. Schmitt, C. Contin-Bordes et al., “OX40
psoriasiform skin,” Journal of the European Academy of Der- ligand contributes to human lupus pathogenesis by promoting
matology and Venereology, vol. 28, no. 10, pp. 1370–1379, T follicular helper response,” Immunity, vol. 42, no. 6,
2014. pp. 1159–1170, 2015.
[65] M. V. Legorreta-Haquet, K. Chávez-Rueda, E. Montoya-Díaz
et al., “Prolactin down-regulates CD4+CD25hiCD127low/-
regulatory T cell function in humans,” Journal of Molecular
Endocrinology, vol. 48, no. 1, pp. 77–85, 2012.
[66] M. V. Legorreta-Haquet, K. Chávez-Rueda, L. Chávez-Sánchez
et al., “Function of Treg cells decreased in patients with sys-
temic lupus erythematosus due to the effect of prolactin,” Med-
icine (Baltimore), vol. 95, no. 5, article e2384, 2016.
[67] T. Gharibi, Z. Babaloo, A. Hosseini et al., “Targeting STAT3 in
cancer and autoimmune diseases,” European Journal of Phar-
macology, vol. 878, p. 173107, 2020.
[68] L. J. Edwards, M. Mizui, and V. Kyttaris, “Signal transducer
and activator of transcription (STAT) 3 inhibition delays the
onset of lupus nephritis in MRL/lpr mice,” Clinical Immunol-
ogy, vol. 158, no. 2, pp. 221–230, 2015.
[69] Y. Du, W. Zhang, S. Liu, X. Feng, F. Gao, and Q. Liu, “S3I-201
ameliorates tubulointerstitial lesion of the kidneys in MRL/lpr
mice,” Biochemical and Biophysical Research Communications,
vol. 503, no. 1, pp. 177–180, 2018.
[70] M. Nakou, E. D. Papadimitraki, A. Fanouriakis et al., “Inter-
leukin-21 is increased in active systemic lupus erythematosus
patients and contributes to the generation of plasma B cells,”
Clinical and Experimental Rheumatology, vol. 31, no. 2,
pp. 172–179, 2013.
[71] T. K. Rasmussen, T. Andersen, R. O. Bak et al., “Overexpression
of microRNA-155 increases IL-21 mediated STAT3 signaling
and IL-21 production in systemic lupus erythematosus,” Arthri-
tis Research & Therapy, vol. 17, no. 1, p. 154, 2015.