Abstract |
The immune system has evolved to protect the body from foreign invading pathogens. To accomplish this
critical role, T lymphocytes must discriminate between self and non-self; this property translates into immune
recognition and elimination of infectious invaders while leaving host tissues intact. This highly selective
process occurs through several complicated mechanisms. The first step is the elimination of self-reactive cells
during T cell development in the thymus, aiming at a T cell repertoire that is ‘self-tolerant’. However, this
process is incomplete and autoreactive T cells that escape ‘central’ tolerance are controlled by peripheral
mechanisms (‘peripheral’ tolerance). Autoimmunity results when either central or peripheral mechanisms fail.
Among the mechanisms involved in safeguarding of self-tolerance, B7/CD28 membrane receptors are crucial
for fine-tuning of T cell function. Programmed death-1 (PD-1) is an inhibitory lymphocyte receptor that has
recently emerged as a key player in induction and maintenance of tolerance. Compared to other B7 receptors,
PD-1 has a broader role in regulation of immune responses considering its expression on activated T- and Blymphocytes,
the wide expression pattern of its ligands (PD-L1, PD-L2) on both lymphoid and non-lymphoid
parenchymal cells, and its distinct mechanism of action. The role of PD-1 in tolerance is further highlighted by
findings in PD-1-deficient mice which develop strain-specific autoimmune phenotypes, and in chronic viral
infections which correlate with upregulation of PD-1 on virus-specific ‘exhausted’ T cells. In humans, a
possible role for PD-1 in immune regulation is indicated by genetic-association studies which have shown
certain polymorphisms of the PD-1 gene to confer increased risk for various autoimmune diseases. We sought
to examine the role of PD-1 and PD-1 ligands in regulation of T cell function in systemic lupus erythematosus
(SLE), the prototype of autoimmune disorder. SLE is characterized by hyper-active T cells and aberrant T cellmediated
autoantibody production and end-organ tissue injury.
We first performed PCR-based restriction fragment length analysis in 289 SLE patients and 256 matched
healthy controls and confirmed the previously reported association with the PD1.3 (+7146G/A) single
nucleotide polymorphism (SNP) (risk A allele associated with odds ratio 2.23 [95% confidence interval 1.55–
3.38] for SLE). PD1.3 resides in an enhancer-like domain in intron 4 of the PD-1 gene, and the A allele has
been shown to disrupt a RUNX1 binding site. To address the putative regulatory effects of PD1.3 SNP in PD-1
transcription, we performed transient transfection experiments in Jurkat T cells with reporter constructs
expressing the firefly luciferase gene under the control of the SLE-associated A allele or the wild-type G allele
of PD1.3. Over-expression of RUNX1 resulted in increased luciferase activity that was significantly higher with
the G allele than with the A allele (by 36 ± 4%), indicating that PD1.3A may alter PD-1 expression. Indeed,
two SLE patients homozygous for PD1.3 (A/A) had diminished PD-1 expression, assessed by flow cytometry,
on peripheral blood CD4+ CD25+, CD4+ CD69+ and CD4+ HLA-DR+ T cell subsets compared to heterozygous
(G/A) and wild-type (G/G) SLE patients. PD1.3 A/A patients also had defective induction of PD-1 on activated
CD4+ T cells following activation with suboptimal –but not optimal– concentration of anti-CD3/anti-CD28
antibodies, associated with impaired PD-1–mediated suppression of T cell proliferation and IFN-γ production
under suboptimal concentrations of PD-L1.Fc.
To further characterize the role of PD-1 in regulation of immune responses in the context of SLE, we
performed autologous mixed lymphocyte reaction (AMRL) experiments, which is an ex vivo model of
autoreactivity against apoptotic self-antigens. During AMLR, regulatory circuits suppress effector T cells, and
only a small degree of cell proliferation ensues. Compared to healthy controls, SLE patients had defective
induction of PD-1 on AMLR CD4+ CD25+ and CD4+ CD69+ T cells; abnormalities were more pronounced in
patients carrying the PD1.3A polymorphism. In contrast, PD-L1 expression on AMLR CD14+ monocytes was
comparable between SLE patients and healthy controls. In accordance, blockade of PD-1/PD-L1 interactions
with monoclonal antibodies caused significant increase in AMLR T cell proliferation in healthy controls but not
in SLE patients.
5
Since PD-1 is involved in regulation of effector T cells at the site of inflammation through interaction with PDL1-
expressing activated parenchymal cells, we examined the expression of PD-1/PD-L1 in renal biopsy
samples from patients with lupus nephritis by immunohistochemistry. Weak-to-moderate PD-1 staining was
detected in the glomeruli in 8 of 13 samples (62%) from patients with lupus nephritis compared with 0 of 9
(0%) control samples; PD-1 expression correlated with CD3 T cell staining. PD-L1 was detected in the renal
tubules of both SLE patients (10 of 15, 67%) and controls (5 of 9, 56%), indicating a potential role for PD-
1/PD-L1 in regulation of local immune inflammatory responses in SLE patients.
Despite its expression in the peripheral blood and inflamed tissues of SLE patients, the PD-1/PD-L1 pathway
could be adversely affected by the lupus inflammatory microenvironment. To test this hypothesis, purified
CD4+ T cells from healthy controls were cultured in serum from non-homologous controls or active SLE
patients and the effect of PD-1 crosslinking on T cell proliferation was assessed. At suboptimal PD-1
activation, incubation with SLE serum resulted in decreased suppression of T cell proliferation compared with
normal serum (6.7 ± 3.6% versus 16.3 ± 2.7%); optimal PD-L1.Fc concentrations were able to overcome this
effect and resulted in profound SLE T cell inhibition.
In summary, this study provides circumstantial evidence for an important role for the inhibitory PD-1/PD-L1
pathway in regulation of T cell function in human SLE. Importantly, our experiments indicate aberrant
expression and/or function of PD-1 in lupus as a result of both direct (PD1.3 SNP, AMLR) and indirect
(inflammatory microenvironment) effects. The expression of PD-1/PD-L1 in the affected tissues and during
AMLR suggests a role of this pathway in maintenance of peripheral T cell tolerance and supports its rationale
for its manipulation as a novel therapeutic option in SLE.
|