Significance paragraph:
1. The structure, expression, and signaling pathway of PD-1 were
discussed in this article.
2. The importance of PD-1 in regulating several autoimmune diseases
reflects how manipulating this molecule can effectively treat some
autoimmune diseases.
Introduction
Two mechanisms adjust the activation of T-cells, peptide and MHC linkage
to the T cell receptor (TCR) and co-stimulatory signals (1). Several
co-stimulatory receptors are responsible for the regulation of positive
and negative T cell responses. CTLA 4 and PD-1 (which are also called
PDCD1 and CD279 are the negative regulators of immune responses (2). PD1
confronts positive signals through the T cell receptor (TCR) and CD28 by
emitting its ligands: PDL-1 and PDL-2 (3). PD1 and PDL1 are found on B
cells, T cells, macrophages, and several dendritic cell (DCs) types.
The bondage of PD-1 to its ligand, PD-L1, results in the control of
self-reactivity, induction of tolerance, and prevention of autoimmunity
(4). Therefore, disruption in this pathway can hugely impact host
physiology, and PDCD1-deficient mice showed precipitated autoimmunity.
This supports the critical role of PD-1 in controlling T cell responses
(5).
In this review, our main focus is on the physiological role of PD1 and
its ligands in immunotherapy. To provide a context for these studies, we
first explain the structure and expression of PD-1, then briefly review
PD-1-mediated signaling and its role in autoimmunity. Preclinical and
clinical findings indicate that modulation of PD-1 is a promising
strategy for the treatment of autoimmune diseases such as rheumatoid
arthritis, type 1 diabetes, and lupus erythematous.
Structure, expression, and ligands of programmed cell death
1 (PD-1)
Two signals activate T cells: The primary signal is provided by TCR,
which is an incomplete one, and the second one is provided by a
costimulatory receptor, that strengthens the responses. An important
group of these costimulatory receptors is provided by ligation between
the T cell membrane proteins CD28 and its ligands B7.1 (CD 80) and B7.2
(CD86) which plays a major role in T lymphocyte stimulation (6). This
family contains an increasing number of diverse proteins classified by
the single-chain glycoprotein, which gives rise to two extracellular
Ig-like domains, along with a transmembrane section and an
intracytoplasmic region (7). Among B7 members, PD1 has been assuming
importance recently.
PD-1 is categorized in the extended CD28/CTLA-4 family of T cell
regulators; however, Its domain has 21-23 % sequence similarity with
CTLA-4 (8). PD1 is a type I transmembrane glycoprotein and consists of
an IgV-type extracellular domain, a trans-membrane region, and an
intracellular tail (9). The extracellular domain stands for an
immunoglobulin variable region and the tail includes an immunoreceptor
tyrosine-based inhibitory motif (ITIM) and several individual
phosphorylation sites in immunoreceptor tyrosine-based switch motif
(ITSM) (10). PD-1(CD279) consists of 268 amino acids encoded by the
PDCD1 gene.
In humans, the PD-1 gene is located on chromosome 2 and has
approximately 9 Kb size, and consists of 5 exons. Exon 1 encodes a short
signal sequence, exon 2 encodes an Ig domain, the third exon contains
stalk and transmembrane domains, and the fourth exon encodes a short 12
aa sequence, which marks the beginning of the cytoplasmic domain. The
C-terminal intracellular residues and a long 3 UTR are contained in exon
five (11). Nielsen et al. cloned four variants of PD-1 mRNA
transcripts, including PD-1∆ex2, PD-1∆ex3, PD-1∆ex2,3, and PD-1∆ex2,3,4,
and the full-length isoform (12). upon activation of T cells with
anti-CD28 and anti-CD3, all variants are expressed (12).
PD-1 is expressed on myeloid cells, NK cells, T, and B cells (13). NK
cells with a more stimulated phenotype express larger amounts of PD1 in
comparison to NK cells with a fatigued phenotype (14). Stimulated T
cells express PD-1 on their surface and emit interferons that encourage
the expression of PD-L1 in several tissues In a mouse model of
cytomegalovirus infection (MCMV), endogenous glucocorticoids have been
confirmed to join microenvironmental signals to trigger PD-1 expression,
including tissue-specific cytokines and neuroendocrine (15). PD-1
expression is regulated by multiple mechanisms.
Several factors are involved in the regulation of PD-1 expression.
Cytokine-induced STAT family of proteins and insulators (CTCF) bind to
DNA on the PDCD-1 gene (16, 17). At least 8 - cis-regulatory elements
regulate PDCD-1 expression, and two conserved regions (CR-B and CR-C)
are involved in PDCD-1 activation (18, 19). In the regulation of
expression, CR-C is more important than CR-B. These regulatory sites are
located at the upstream region of the transcription start site (TSS),
and multiple transcription factors bind to these sites and regulate the
PDCD1 gene expression (20, 21). A CR-B contains an AP-1 binding site,
while CR-C has a binding region for the nuclear factor of activated T
cells (NFAT)c1, interferon-stimulated response element (ISRE), NF-κB,
and FoxO1 (19). The immune activation of PD-1 is dependent on its gene
strains, for example, Pdcd1 gene polymorphisms are associated with the
induction of autoimmune diseases such as systemic lupus erythematosus
(SLE), rheumatoid arthritis (RA), ankylosing spondylitis (AS), type 1
diabetes mellitus (T1DM), and multiple sclerosis (MS) (22).
Central and peripheral tolerance are two main procedures to control the
progression of autoimmune diseases via regulating self-reactive T cells
(23). The balance between stimulatory and inhibitory signals is critical
for defensive immune responses, self-tolerance, and promotion of T cell
homeostasis (24, 25). PD-1 reduces the interfaces between dendritic
cells with self-reactive T cells and stimulation and development of
auto-reactive T cells. Therefore, PD-1 induces tissue tolerance via
suppression of self-reactive T cell activation in conjunction and
protects against immune-mediated tissue injury (26).
Regulatory T (Treg) cells are an important subtype of T cells that have
a key role in maintaining tolerance by regulating insufficient responses
and immune cell activation (27). PD-1 and PD-L1 are highly expressed on
Tregs, and unlike other subtypes of T cells, engagement of PD-1 on Tregs
increases the number of these cells. That is the reason why this pathway
is important in maintaining development and immune responses, and is in
charge of functional responses of Treg cells (28).
On the other side, continuous and persistent formation of PD-1 and its
ligands are common during long-time infections and malignancy (29).
Although PD-1 is critical in the maintenance of peripheral T-cell
tolerance, it can limit anti-tumor and anti-viral responses. Indeed, the
blockage of the PD1 pathway in these disorders causes an increase of T
cells and a reduction in tumor and virus load burden (30). As an
example, malignant tumors mainly express PD-L1 on their surface, and the
bondage of PD-L1 to PD-1 inhibits the expansion of active T cells and
causes a drop in anti-tumor immunity (25, 31). PD-1 is likely to disrupt
protective immune responses; therefore, in severe viral infections and
malignant T cells are more dysfunctional (32, 33)
Ligands of PD1
PD-L1, a type 1 transmembrane protein of 290 aa, is encoded by the Cd274
gene on chromosome 9 and consists of seven exons (34). The first exon
encodes the second, third and fourth, fifth, and sixth exons contain the
5′ untranslated regions (5′-UTR), signal sequence, an IgV-like domain,
IgC-like domains, the transmembrane and the intracellular domains,
respectively. The intracellular domain residues plus the 3′ UTR are
encoded by the last exon (34). In brief, the intracellular
domain PD-L1 is short and conserved with no known function (35). A
spliced isoform of PD-L1 was reported in humans, lacking an IgV-like
exon that cannot bind PD1 (36). Whereas, spliced variants of murine
PD-L1 have not been reported.
PD-L1 is mainly expressed on antigen-presenting cells (APC) by
interferon γ (IFN-γ), and non-immune cells such as epithelial cells
(37). Various processes can regulate the expression of PD-L1, such as
gene transcription, post-transcriptional and post-translational
modifications, and exosomal transport. Therefore, we should expand our
knowledge about the regulation process of PD-L1 expression to introduce
immune checkpoint blockers with high efficacy and advance immunotherapy
strategies (38). PD-L1 or CD274 is located on chromosome 9p24.1, which
amplification and translocation in this region upregulate PD-L1
expression, resulting in increased immune escape as demonstrated for
gastric adenocarcinoma, primary mediastinal large B cell lymphoma,
classical Hodgkin lymphoma, squamous cell carcinoma, and NSCLC (38, 39).
Interestingly, Janus kinase 2 (JAK2) is also located on chromosome 9p
and can regulate the expression of PD-L1 that mutation and amplification
of the JAK lead to upregulation of PD-L1 (40). Elevated activity of the
JAK2/STAT (signal transducers and activators of the transcription)
signaling pathway increases PD-L1 expression (41). A DNA double-strand
breaks (DSBs) activates STAT1 and STAT3 signaling which requires
ataxia-ATM (telangiectasia mutated)/(ATR) ATM- and Rad3-related/(Chk1)
checkpoint kinase 1 kinases, resulting in PD-L1 upregulation (42, 43).
On the other hand, PD-L1 3′-UTR disruption involves in various human
cancers through cancer-immune evasion (44). The CD274 3′ UTR disruption
by CRISPR (clustered regularly interspaced short palindromic
repeats)/Cas9 increase PD-L1 expression and leads to immune (44, 45).
Together, these results demonstrated that genomic alterations have an
essential role in elevated PD-L1 levels in the number of cancers, and
these alterations can be used as targeting biomarkers to improve the
therapeutic efficacy of anti-PD-1/PD-L1 (46). chromatin modifiers, such
as epigenetic regulations, have a critical role in the expression of
PD-L1. Recently, blocking PD-L1 signaling using BET inhibitors reduced
PD-L1 expression and showed high therapeutic potential in cancer
(37,38). Histone deacetylase 3 (HDAC3) is also another repressor of
PD-L1 and HDAC inhibitors can upregulate the PD-L1 expression (47, 48).
Therefore, a combination of PD-1/PD-L1 blockade with HDAC inhibition may
augment immunotherapies. Similarly, Lu et al. showed that
tri-methylation of histone H3 on lysine 4 (H3K4me3) also induces the
expression of PD-L1 cancer cells such as pancreatic cancer (49, 50).
Another possible epigenetic regulator of PD-L1 is zeste homolog 2
(EZH2), which catalyzes H3K4me3 and is positively related to PD-L1 in
breast cancer (51, 52). The inhibition of the (ADP-ribose) polymerase 1
(PARP1) can upregulate PD-L1 levels via negative regulation of EZH2 (53,
54). These findings prove that the expression of PD-L1 is epigenetically
regulated by several mechanisms (55).
PD-L2 or CD273 encoded by the Pdcd1 Ig2 gene is a type I transmembrane
protein and is next to Cd274 with seven exons. Three PD-L2 splicing
variants are identified, including the first form eliminating the exon 3
encoding IgV-like domain with no potential to bind PD-1, the second one
is created without the IgC-like domain, and the third form loses the
IgC-like domain (56). The expression of PD-L2 is more restricted than
PD-L1 expression and is expressed on antigen-presenting cells (APCs),
including monocytes, macrophages, dendritic cells, and some B cells
(57). PD-L2– and PD-L2+ B1 cells
have similar proliferative patterns and immunoglobulin production, but
PD-L2+ B1 cells are enriched for the expression of
VH1/VH12 genes (58). The expression of PD-L1 depends on TLR4 and STAT1
and PD-L2 up-regulation can induce by IL-4Rα and STAT6 (59).
Interestingly, T helper (Th) 1 cell up-regulate PD-L1 expression,
whereas up-regulates PD-L2 (59). Little is known about the
transcriptional regulation of PD-L2 and more studies are needed.
Both CD28 and IL-2, though indirectly, encourage initial activation
signals. PD1 induces a delay in this pathway; hence, T-cell
proliferation and existence get disturbed reliably (60). After CD28
binding to CD80/CD86 enhances IL-2 production by induction of T cell
responses (60, 61). CD28 receptor activates the Src family tyrosine
kinases, Fyn and Lck, in lymphocytes (62). Tyrosine motifs of CD28 are
phosphorylated by Fyn and Lck and lead to the activation of
phosphatidylinositol 3-kinase (PI3K) and Akt protein kinase, which
induce glycolytic metabolism (63). In contrast, IL-2 promotes the PIM1
or PIM2 expression, through the JAK/STAT signaling pathway with the same
effect on glycolytic metabolism (63). In brief, costimulation of CD28
can induce the activation of various intracellular pathways, which lead
to T cell responses, including cell survival, cell proliferation, and
cytokine production (64).
CTLA-4 and PD-1 can suppress the activation of T cells by distinct but
synergistic mechanisms. Both of these inhibitory receptors suppress
glucose metabolism and Akt phosphorylation (65). Akt increases thymocyte
survival, cell growth, and proliferation in T lymphocytes (66). Despite
the similar inhibitory mechanism, the cytoplasmic domain of CTLA-4 and
PD-1 have significant differences. The cytoplasmic domain of CTLA-4
interacts with protein phosphatase 2A (PP2A) and tyrosine phosphatase
SHP-2 (64, 67). PD-1 encodes both of immunoreceptor tyrosine-based
switch motif (ITSM) and immunoreceptor tyrosine-based inhibitory motif
(ITIM) (68, 69). The inhibitory function of PD-1 is mainly depending on
ITSM which recruits tyrosine phosphatases SHP-1 and SHP-2 for signal
transduction (60, 70). The expression of SHP-1 is mainly related to
hematopoietic lineage, whereas the exact role of SHP-2 is not fully
clear (71, 72). Both SH-2 domains of SHP-2 are necessary for SHP-2
interaction and enzymatic activation and the absence of even one of the
SH2 domains prevents interaction with PD-1 (73). PD-1 phosphorylation by
TCR-proximal Src family tyrosine kinases enables SHP-2 to form a linkage
with two phosphorylated ITSM-Y248 residues via its N-SH2 and C-SH2
domains (74).
PTEN (Phosphatase and tensin homolog) is the main negative controlling
mechanism of the PI3K/Akt (75). PTEN via its phosphatase activity
converts PIP3 to inactive PIP2 which PIP3 is essential for the AKT
phosphorylation (76). the phosphorylation of T382, S380, and T383 within
the tail of PTEN regulates stability and activity (77). The cytoplasmic
domain of PTEN is phosphorylated by CK2 (Casein Kinase 2) which leads to
inhibiting PTPN phosphatase activity (76). PD‐1 inhibits TCR activation
signals by inhibiting the CK2 function and dephosphorylation and
activation of the PTEN (78).
PD-1 not only suppresses the induction of Bcl-xL, but also brings an end
to the proliferation of transcription factors correlated with effector
cells, such as GATA-3, T-bet, and Eomesodermin (79). Stimulation of
Foxp3 expression in T cells and induction of inhibitory function of Treg
cells is the main pathway by which tolerance is encouraged by PD1 (80).
During chronic viral infections, the up-regulation of PD-1 on exhausted
CD8+ T cells is associated with decreased TNF-α, IL-2, IFN-γ, and
perforin production (81). It is shown that PD-1 ligation makes T
lymphocytes much less sensitive to TCR-mediated signals (82). Given that
PD-1 regulate T cell exhaustion, blocking the PD-1/PD-Ls ligation can
reduce apoptosis, reinvigorate exhausted CD8 T cells, and reprogram
exhausted CD8 T cells into durable memory T cells (83, 84). Coexpression
of T-cell Ig- and mucin-domain-containing molecule–3 (Tim-3) 3 and PD-1
is related to more CD8 T-cell exhaustion (85, 86).
PD-1/PD-L1 pathway is associated with various diseases, such as cancer,
stroke, neurological disorders, and autoimmunity (87). For example, the
development of lupus diseases has been shown in
C57BL/6(B6)-PD-1-/- congenic mice (88). For autoimmune
diseases immunotherapy to be successful, the PD-1/PD-L1 pathway can be
considered an effective target. Recently, several studies have been done
on specific molecules and pathways that have vital roles in triggering
immunological tolerance. Nevertheless, the death toll of autoimmune
diseases associated with PD1 is still high. The PD-1 signaling pathway
is briefly shown in figure 1.
The PD1 pathway and autoimmune
diseases
PD-1 in Type 1 diabetes
Type-1 diabetes (T1D) is an outcome of severe β-cell damage in
pancreatic islets. Type-1 diabetes leads to hyperglycemia accompanied by
similar issues of clinical diabetes (89). PD1 sustains tolerance by
mechanisms that are separate from those of CTLA-4. First of all, PD-1
deficiency leads to tissue-specific injury slowly, dependent on the
genetic background, whereas CTLA-4 deficit culminates in rapid
multi-organ inflammation and death, disregarding genetic (90). Secondly,
while CTLA-4 blockade ameliorates the antitumor immune responses
associated with the creation of autoimmune responses, the blockade of
the PD-L1/PD-1 pathway is distinct from that (91). Thirdly, a lower
T-cell motility is brought about in PD-1/PD-L1 blockade; however, in
CTLA-4 blockade is not (25). Finally, PD-1/PD-L1 blockade swiftly
accelerates diabetes regardless of age. By stark contrast, CTLA-4
blockade stimulates diabetes in neonates (92).
PD-L1 is expressed on β-cells of islets that produce insulin. It may
also act as an inhibitory factor for regulating self-reactive T cells in
the marked tissue (89, 93). Deficiency or blockade of PD-1/PDL1 lead to
the development of autoimmune diabetes in NOD mice (94). T1DM is
initiated in NOD mice by MHC class I-dependent T cell responses and
subsequent pancreatic β cell destruction (95). NOD islets showed a
decreased capacity to upregulate B7-H1 expression, which is induced by
inflammatory cytokines (96). These results suggested that NOD islets
cannot upregulate the expression of B7-H1 immunoregulatory molecules in
response to pro-inflammatory cytokines (96). In NOD mice, PDL1-mediated
regulation of diabetogenic CD4+ and CD8+ T cells is necessary for the
development of diabetes (97). This was approved by CD4+ and CD8+
TCR–transgenic mice; a study showed an elevated number of T cells
following the blockade of PD-L1 (98, 99). Th1 and Th2 imbalance is also
associated with developing autoimmune diabetes in NOD mice, but the
protective effect of PD-1 is not related to inducing IL-4 and
suppressing IFN-γ producing cells (99).
As previously mentioned, PD-L1 induces tolerance and controls pathogenic
self-reactive effector T cells using two binding partners, PD-1 and B7-1
(100). Two anti–PD-L1 Abs with different blocking activities,
anti–PD-L1 mAb 10F.9G2 blocking both PD-1/PD-L1 and PD-L1/B7-1
interactions and anti–PD-L1 mAb 10F.2H11 blocking only PD-L1/B7-1
interactions, can accelerate diabetes in NOD mice, but single-blocker
10F.2H11 mAb is more important at promoting diabetes in older (101).
Therefore, PD-1 pathway blockage in NOD mice can both improve the
disease-related breakdown of tolerance causing T1D in a particular organ
and limit T cell numbers and functions (102).
Anti-PD-1 and anti-PD-L1 monoclonal antibodies can also cause T1D in
human subjects, and diabetes can be an adverse effect of anti-PD-1/PD-L1
immunotherapy (103, 104). For example, treatment with pembrolizumab, a
humanized IgG4 anti-PD-1 Ab, lead to the induction of autoimmune
diabetes in the Chinese population (105). Onset diabetes is also
reported following the treatment with nivolumab or pembrolizumab, which
monitoring glycemia was not predictive of autoimmune diabetes occurrence
(106). One of the reasons for the rapid onset of diabetes is possibly
because of the uncontrolled activation of T cells (107). T cell
infiltration with the low expression level of PD-1 into pancreatic
islets initiates β-cell injury and T1D in the patients (108). The
fulminant onset of diabetes in the affected subjects is partially
related to Human leukocyte antigen (HLA) DR4 and T1D-related
autoantibodies (109). HLA-DRB1*09:01–HLA-DQB1*03:03 also showed a
significant and strong association with T1DM (110). In contrast,
HLA-DRB1*15:02 is mainly protective for type 1 diabetes (111). Sun et
al. showed that the progression of T2D is related to the expression of
PD-1 on NK cells and that more investigation into this is needed (112).
Another important reason for the induction of diabetes after anti-PD-1
therapy is the increase in C-reactive protein (CRP) levels (113). These
results show the importance management of immune-related adverse events
(irAEs) in subjects during the treatment with immune checkpoint
inhibitors to reduce mortality and morbidity rate (114). Hickmott et al.
suggested HbA1c (hemoglobin A1C) and plasma glucose level monitoring
before and during PD-1/PD-L1 therapy (115).
MicroRNAs, a small non-coding RNA, have critical roles in various cell
functions such as development, differentiation, cell cycle, and
apoptosis (116). Some microRNAs are related to the expression of PD-L1
in HSPCs (hematopoietic stem cells) and T1D pathogenesis, which PD-L1
expression can be used as an effective and safe immunotherapy option
(117). Ben Nasr et al. revealed that PD-L1 pharmacological restoration
or genetic overexpression in progenitor cells and HSPC can reverse T1D
(117).
The PD-1 pathway in systemic lupus
erythematosus
Systematic lupus erythematosus (SLE) is the most common type of lupus
and mainly affects women of child-bearing age (118). In this autoimmune
disease, self-tissues are targeted and destroyed by the immune responses
that cause serious inflammation (119, 120). Type I interferon (IFN) and
Toll-like receptor (TLR) signaling are involved pathogenic factors in
SLE (121).
T follicular helper (Tfh) and regulatory B (Breg) cells are associated
with SLE pathogenesis, in which Breg cells are the negative regulator of
immune responses during autoimmune diseases (122). In SLE, The frequency
of Breg cells and the production of IL-10 increase in SLE flares,
whereas decrease during disease remission. CXCR5+ PD-1+ Tfh cells
produce IL-21 and expand in SLE. Yang et al. demonstrated that IL-21 can
induce IL-10 production during the development of Breg cells (122).
IL-21 in the presence of lipopolysaccharide (LPS) and PIB can also
promote the development of CD19+ IL-10+ B cells (122). IL-21 can also
potently promote the differentiation of CD11chiT-bet+ B cells into
Ig-secreting autoreactive plasma cells (123). CXCR3- PD-1 +CD4+ T cells
associate with SLE disease severity through B-cell-help for the
production of autoantibodies (124). B cells migrate to target tissues
that correlate with SLE clinical manifestations. The high serum level of
autoantibodies against PD-1 in SLE patients is related to disease
activity that can break tolerance (125). TIM-3 (T cell immunoglobulin
mucin-3) binds to Galectin-9 and is a negative regulator of T
cell-mediated immune responses. Interestingly, the co-expression of PD-1
and TIM-3 on NK cells in SLE patients is related to disease severity and
activity and has a role in SLE pathogenesis (126).
The number of CD4+CD25+ T cells decreases in clinically active SLE
patients (127). Decreased frequency of CD4+CD25+T cells in patients with
active SLE is inversely associated with disease activity and serum
anti-double-stranded DNA levels (128). The suppressive capacity of Treg
cells also decreases in SLE patients, in which PD-1 expression is
required for the regulatory function of CD4+ CD25+ Foxp3+ Treg cells to
control autoimmune responses (129). Low PD-1 expression on Treg cells is
shown in SLE patients. Kristjansdottir et al. suggested that low PD-1
expression is related to the PD-1.3A allele (130). Polymorphisms in PD-1
have revealed associated risks in SLE development, whereas PD-L1 or
PD-L2 have not been showing associated risks (131). PD1.5 polymorphisms
are significantly related to SLE susceptibility, while PD1.6 is
protective (132, 133).
In vivo treatment with anti–PD-1 mAb decreased the frequency of f CD4+
PD-1+ T cells in the kidney of NZB/W (New Zealand Black 3 New Zealand
White) F1 mice and accelerated lupus-like nephritis and mortality rate
(134). These results suggest that the expression of PD-1 is critical for
the induction of immune tolerance mediated by CD8+ Foxp3+ T cells that
inhibit both pathogenic T and B cells (135).
The Programmed Cell Death 1 pathway in Rheumatoid
Arthritis
Rheumatoid arthritis (RA), an immune-mediated inflammatory disorder of
the synovium, is characterized by hyperplasia and mononuclear cells
infiltration to the synovial membrane, and secretion of proteases
resulting in adjacent bone and destruction of articular cartilage (136).
Macrophages, synovial fibroblasts, and autoreactive CD4+ T cells are
involved in RA pathogenesis (137). there is no doubt that Treg cells
maintain self-tolerance by controlling the activation of autoreactive T
cells (136). An aberrant proportion of CTLA-4 and PD-1 have been related
to the persistent activation of autoreactive T cells in RA (138). The
low expression level of PD-1 on CD4+ and CD8+ T cells is related to the
disease activity score of RA (139). The emergence of inflammatory
arthritis irAEs in patients treated with PD-1/PD-L1 immunotherapy
highlight the important role of this pathway in the regulation of immune
responses (140).
The high concentration of soluble PD-1 seen in SF and serum specimens is
correlated with TNF-α and rheumatoid factor (RF) in SF and the clinical
relevance of RA (138). Wan et al. suggested the expression of an
alternative splice variant (PD-1∆ex3) that has an extracellular domain
of PD-1 without the membrane-spanning domain can affect the regulatory
effect of full-length PD-1 on T cell activation (138). Therefore,
overexpression of these soluble factors can block the inhibitory effect
of costimulatory molecules (CTLA-4 and PD-1). Therefore, agonistic PD1
antibody-based therapeutic options can show efficacy in the treatment of
RA (120). Besides CTLA-4 and PD-1, TIM-3 is also involved in the immune
dysregulation of RA (141).
Studies in Collagen II (CII)-induced arthritis (CIA), a murine
experimental model of RA, approved the protective effect of PD1 or PD-L1
in RA (142, 143). The protective role of PD-1 in RA was shown by
inducing CIA in PD-1–deficient mice (144). In this way, B7-H1-/-
(PD-L1) mice showed exacerbated arthritis and elevated CD4+ T-cell
responses. In B7-H1-/- mice the number of CD4+PD-1hi CXCR5hiICOShi CD62Llo Tfh cells with
elevated expression of IL-21, Bcl6, the apoptosis-inducer molecule FasL
increase, whereas the frequency of CD38+CD138+ plasma cells and antibody responses decrease
(145). Hamel et al. suggested B7-H1 expression on non-T and non-B cell
signals through PD-1 on T cells required B-cell antibody responses and
development of arthritis (145). It is also demonstrated that B7-H1
expression is essential for B-cell survival by regulating the activation
of Tfh cells (145). PD-1+TOX+BHLHE40+ population of T cells expressed
genes supporting B cells activation and driving inflammation by the
production of CXCL13 and IL-21, and activation of myeloid cells by
production of GM-CSF and TNF (145). A PD-1+TOX+EOMES+ population of CD4+
T cells expressed attracting genes for myeloid cells (145).
Extracellular Vesicles (EVs) are present in RA patients and transfer
both the PD-1 receptor and microRNAs associated with T cell inhibition
(TIM-3, CTLA-4, and IRF9) to cells in the microenvironment (146).
Greisen et al. suggested EVs are involved in the development of RA
chronicity, favoring the progression of T cell inhibition and T cell
exhaustion (146).
Therefore, the administration of PDL-Ig could significantly reduce the
CIA severity through the inhibition of cell proliferation and
anti-inflammatory actions (147).
The Programmed Cell Death 1 pathway in Multiple Sclerosis
Multiple sclerosis is an autoimmune chronic demyelinating disease of the
central nervous system (CNS) (148). MS is an immune-mediated disorder
that autoreactive T cells that play an important role in the
immunopathological process (148). Experimental autoimmune
encephalomyelitis (EAE), an animal model of MS, is also a T
cell-mediated autoimmune disease of the CNS that has pathological and
clinical features with MS (149). PD-1/PD-L1 pathway maintains immune
tolerance and inhibits autoimmunity by delivering a negative regulatory
signal (150, 151). The elevated severity of EAE e in PDL-1−/− and
PD-1−/− mice is related to a high production level of pro-inflammatory
cytokines TNF, IL-6, IL-17, and IFN-γ (152). PD-1 down-modulation in
Tregs alters Treg-mediated immune suppression, induces the activation of
T cells, and promotes the EAE (153). It is demonstrated that signals
delivered by the B7-H1-Ig fusion protein could control Th17
differentiation and CNS autoimmunity (154).
Astrocytes and microglia cells can control brain-infiltrating antiviral
T-lymphocytes through the upregulation of PD-L1 (155). Brain endothelial
cells control immune responses and T cell migration into the CNS via
PD-L2 expression (156). Downregulation or absence of the endothelium
PD-L2 in MS lesions is correlated with infiltrating T cells (156, 157).
PD-1 polymorphism is related to defect inhibition of T cell activation
and progression of MS (158). The presence of the PD-1.5 T allele is
associated with the initial manifestations of MS, such as pyramidal
signs and diplopia (159). Decreased number and functionally exhausted
Treg cells in MS are due to the high expression level of PD-1 (129,135).
Targeting the PD-1/PD-L1 has therapeutic potential in a variety of
cancers, but PD1/PD-L1 axis blockade in the brain remains controversial
(139). Krakauer et al. suggested low expression levels of CCR4, PD-1,
and L-selectin in MS can be normalized by treatment with interferon-β
(138).
Inflammatory bowel disease
Inflammatory bowel diseases (IBD), including Crohn’s disease (CD) and
ulcerative colitis (UC), are inflammatory, chronic disorders with
similar features like abdominal pain, weight loss, and diarrhea (141).
Treg cells prevent both innate immune and T cell-mediated responses in
the intestinal lamina propria through IL-10 production and CTLA-4 and
PD-1 expression (142,143). Totsuka et al. suggested that the CD4+CD25–
PD-1+ T cells represent a substantial amount of Treg cells and inhibit
the development of colitis (144). It is demonstrated that
CD44highCD62L–CD4+IL-7Rahigh effector-memory
(TEM)-like cells in colitic mice are exhausted and convert into
cytokine-non-producing Treg cells (145). Blockade of the PD-1/PD-L1 can
restore exhausted T cells ability to undergo proliferation, kill
infected cells, secrete cytokines, and reduce viral load (146).
Interestingly, blockade of the CTLA-4 has no beneficial effect on either
T-cell function or infection control (146).
Colitis is the most frequent irAEs related to immune checkpoint
inhibitors. Features of anti-CTLA-4 colitis are similar to anti-PD-1
colitis, with increased crypt atrophy/dropout and apoptosis (152). After
anti-PD-1 therapy, colitis generally occurs later and is widely
variable, but anti-PD-1 irAE is different and less common than
anti-CTLA-4 (147,148,150). For example, severe colitis after PD-1
blockade with nivolumab is related to Th1 responses, CRP, and IL-6
production (149). Park et al. suggested that PD-1 deficiency alters gut
microbiota and protects the development of experimentally induced
colitis (151). Therefore, it is critical to understand the other
treatment options to minimize toxicity and mortality rate.
Anti-PD1 therapy and autoimmune diseases
The popularity of anti-PD1 therapy has been increasing in recent years.
Immune checkpoint monoclonal antibodies (mAb) target PD-1 and improve
antitumor immune responses (160). There are two IgG4 monoclonal
antibodies named Pembrolizumab (Kytruda) and nivolumab (Opdivo) that
target the PD-1 molecule (161). According to recent studies, T1DM
patients have a considerable reduction in PD-1 expression on CD4+ T
cells, this is subject to the progression of T1DM by T cell activation
(162). Therapy pembrolizumab precipitates type 1 diabetes since it
reactivates CD8+ T cell clone and blocks the PD-1 pathway (163). In the
aftermath of PD-1 pathway blockade, PD-L1 molecules lose the ability to
bind to the PD-1 receptors on autoreactive T cells since pembrolizumab
has blocked them (164). Anti-PD1 therapy has also similar results. With
PD-1 reduction, autoreactive T-cells get stimulated and a wide range of
autoimmune responses appear (165).
On the other hand, since malignant cells circumvent the immune system by
either expressing PD-L1 or PD-L2, anti-PD-L1 and anti-PD-L2 checkpoint
suppressors are invented (166). These anti-checkpoints obstruct the
pathway of PD-1 and thus restore the role of T-cells (167). Blockage of
this pathway leads to both survival of cancer-sensitive T cells and
autoreactive T cells. For instance, T cells that fight against
pancreatic islet cells are survived. These cells cause T1DM (168).
Immune therapy has a range of immune-related adverse events (irAEs)
(169). Pneumonitis, colitis, hepatitis, dermatitis, nephritis,
pancreatitis, vitiligo, rash, pruritus, and endocrinopathies are all
common side effects of anti-PD1 therapy. As well, diabetes mellitus type
1 is induced by targeting PD-1, though it happens less commonly (170).
Tumors are prone to evade the immune system by building an
immunosuppressive tumor microenvironment or activating the inhibitory
pathways that suppress tumor-specific T cell responses (171). IrAEs are
more probable in patients who are sensitive or predisposed to the
progression of autoimmunity. IrAEs occur before, after, or during PD-1
inhibitor therapy (172). Moreover, patients who have one autoimmune
disease are more susceptible to developing a second one (173). Whether
immune checkpoint inhibitor therapy in patients with autoimmune disease
is efficient or not goes beyond discussion since the benefits derived
from this method far outweigh the disadvantages of irAEs (174).
Additionally, Addison’s disease is reported to develop in 0 % - 8% of
patients undergoing anti-PD-1 therapy (175). Nevertheless, there are
several future perspectives. Although PD-1 antibodies rose to popularity
after they intervened in cancer therapy, their treatment efficacy varies
in each cancer and each patient. Generally, there are many unsolved
challenges in anti-PD-1 therapy: First, clarification of how PD-1
antibodies cause anti-tumor impacts, second, biomarker progression to
have a clear prediction of the side effects, and third, further
developments in combination therapy.