PD‐1 mRNA expression in peripheral blood mononuclear cells as a biomarker for different stages of primary gouty arthritis

Abstract There is currently a lack of biomarkers to assist the diagnosis and prediction of primary gouty arthritis (PG). Therefore, we evaluated the clinical value of programmed cell death protein 1 (PD‐1) mRNA expression in peripheral blood mononuclear cells (PBMCs) of patients with PG. This study included 36 patients with acute phase PG (APPG), 48 with non‐acute phase PG (NAPPG), 42 with asymptomatic hyperuricemia (AH) and 79 normal controls (NCs). PD‐1 mRNA expression levels were detected by qRT‐PCR. PD‐1 mRNA expression was statistically analysed by ANOVA or t tests, while correlations between PD‐1 mRNA and clinical variables were assessed using Pearson correlation tests. Receiver operator characteristic (ROC) curve analysis was used to evaluate the diagnostic value of PD‐1 in different PG stages. PD‐1 mRNA expression was significantly lower in patients with APPG than that in NAPPG, AH and NCs (P < 0.01). Correlation analysis revealed that PD‐1 mRNA levels correlated negatively with T‐score (r = −0.209, P < 0.01). ROC curve analysis showed that serum uric acid (SUA), PD‐1 mRNA and both combined displayed higher diagnostic value in patients with PG, NAPPG and APPG compared to that in NCs and patients with non‐PG arthritis (NPG). Moreover, ROC curve analysis showed that SUA and PD‐1 mRNA had good diagnostic value in APPG, with the greatest diagnostic power when combined. PD‐1 mRNA could be a clinical auxiliary diagnostic biomarker for APPG, and the combined use of PD‐1 mRNA and SUA is better than that of SUA alone.


| INTRODUC TI ON
The most common form of arthritis is primary gouty arthritis (PG), a group of inflammatory diseases in which high serum uric acid (SUA) levels (hyperuricemia) cause joint damage and urate accumulation in the joint capsule, bursa, cartilage, bone and other tissues. 1 Recent epidemiological studies have shown that the hyperuricemia is widespread, while the prevalence of PG has also increased year by year, reaching 2.49% in developed countries, entering the list of Common and Multiple Global Diseases. [2][3][4] According to the disease progression, PG is usually divided into acute phase PG (APPG) and non-acute phase PG (NAPPG) by clinicians. Currently, PG is diagnosed by comprehensively evaluating the overall condition of the patient according to the latest and most authoritative diagnostic criteria for PG, the 2015 American Rheumatology Association or European Association of Rheumatoid Arthritis Classification Standards. 5 However, to make an accurate treatment plan during clinical diagnosis, the collection of clinical symptoms, blood index tests and imaging examinations during different stages of PG is required. Since the clinical diagnosis of PG can be more complicated at different stages, SUA levels are often used to assist PG diagnosis; however, it is difficult to differentiate between asymptomatic hyperuricemia, NAPPG and APPG using SUA levels since they lack diagnostic specificity. 6 Therefore, it is necessary to investigate other potential biomarkers in the blood that can assist the differential diagnosis of patients with APPG and NAPPG.
The pathogenesis of PG is complex, and it involves genetic, environmental and immune factors. 7 Previous studies have suggested that the mechanism includes the deposition of urate crystals in the tissues around joints, which causes mononuclear macrophages to become chemotactic and phagocytose the urate crystals, releasing a variety of pro-inflammatory cytokines and a large number of inflammatory mediators. These then recruit more mononuclear macrophages to the surrounding tissues where urate crystals are deposited, amplifying the inflammatory cascade. 8  provides the first signal, while the second results from the interaction between antigen-presenting cells (APCs) and T cell surface co-stimulatory molecules, such as the B7/CD28 family and the tumour necrosis factor/tumour necrosis factor receptor (TNF/TNFR) superfamily. 10 Programmed cell death protein 1 (PD-1) is a co-stimulatory molecule belonging to the B7/CD28 family, which are mainly expressed in mononuclear cells and are critical immunosuppressive molecules involved in inflammation. 10,11 PD-1 is known to play a negative regulatory role in the immune response and has significant immunosuppressive effects in inflammation. [12][13][14] Recent studies on the role of PD-1 in the immune mechanisms, diagnosis and treatment of tumours have shown that it has promising future applications, [15][16][17][18][19][20][21] while others have demonstrated that it has good auxiliary diagnostic value in rheumatoid arthritis, tumours and other complications. 22 It has been shown that the level of immune molecules secreted by mononuclear cells are closely correlated with inflammatory immunity in patients with PG, 23,24 while studies have confirmed that small RNAs play critical regulatory roles in the occurrence and development of APPG. 25 Although numerous studies have demonstrated that PD-1 can be widely used for clinical diagnosis and as an auxiliary diagnostic biomarker, 26,27 the role of PD-1 in APPG has not yet been studied. Therefore, we investigated whether PD-1 is an immunosuppressive molecule secreted by monocytes, which is involved in APPG pathogenesis, and, more importantly, whether PD-1 could be an auxiliary diagnostic biomarker for APPG.

| Diagnostic criteria for asymptomatic hyperuricemia
Patients were diagnosed with asymptomatic hyperuricemia according to the 2013 Consensus on the Treatment of Hyperuricemia and Gout Standards by Experts from the Endocrinology Branch of the Chinese Medical Association 28 : SUA levels of >420 μmol/L (7.0 mg/ mL) in males and >360 μmol/L (6 mg/mL) in females with no patient specimens meeting the diagnostic criteria for gout.

| Diagnostic criteria for PG
The T-score for the type of joint affected, characteristics of the attack, disease onset, clinical evidence of gout, blood uric acid levels and related imaging examinations were scored according to the 2015 American Rheumatology Association/European Anti-Rheumatic Alliance Gout Classification Standards. 5 Patients with T-scores ≥8 can be diagnosed as having PG. Patients with acute inflammation were classified as having APPG, while patients without acute inflammation were classified as having NAPPG.

| Exclusion criteria
Patients were excluded from this study based on the following criteria 29 : (a) patients with PG were excluded if they had experienced pseudogout symptoms, including acute suppurative arthritis, rheumatoid arthritis, calcium dihydrogen pyrophosphate deposition, ankylosing spondylitis and psoriasis, or diseases such as arthritis, osteoarthritis and bone tumours; (b) patients with asymptomatic hyperuricemia were excluded if they had experienced other diseases that cause elevated blood uric acid levels, had taken drugs that can cause elevated blood uric acid levels and had a clear gout, a malignant tumour, and severe kidney disease diagnosis, or had insufficiency or severe systemic diseases (such as cachexia).

| Specimen collection
All patients enrolled in this study were instructed to ingest no food or water for 12 hours to ensure that they had empty stomachs when their peripheral venous blood was collected in an EDTA anticoagulant tube early the next morning. The EDTA anticoagulated whole blood samples were used to isolate and extract PBMCs and detect PD-1 mRNA expression.

| PBMC isolation
Peripheral blood mononuclear cells were isolated within 6 hours of venous blood collection to ensure lymphocyte survival. The density gradient separation technique was used to isolate mononuclear cells from the blood at 18-20°C using Ficoll-Paque PLUS reagent according to the manufacturer's instructions.

| Quantification of PD-1 mRNA expression by qRT-PCR
Total RNA was extracted from the isolated mononuclear cells using

| Instruments and reagents
Ficoll-PaqueTM PLUS reagent was purchased from Sweden, RNA extraction kits were purchased from Invitrogen (Karlsruhe, Germany), reverse transcription cDNA kits were purchased from Takara-Bio (Tokyo, Japan), and PCR kits were purchased from Kangcheng Biotech (Shanghai, China). An Ultramicro Ultraviolet/ visible Spectrophotometer was obtained from SimpliNano (New York, USA), a Hema9600 Gene Amplifier was obtained from Hema (Zhuhai, China), and a Rotor-Gene Q qRT-PCR analyser was obtained from Invitrogen.

| Statistical analysis
Clinical data were analysed using two-tailed independent t tests or

| RE SULTS AND D ISCUSS I ON
PD-1 expression in PBMCs has been associated with good clinical outcome in inflammatory diseases. Due to the increasing prevalence of PG and the lack of a useful diagnostic biomarker that can distinguish between different stages, we investigated whether PD-1 is involved in APPG pathogenesis and, more importantly, whether PD-1 could be an auxiliary diagnostic biomarker for APPG. In this study, we present data indicating the potential of PD-1 mRNA as a biomarker associated with PG, with the strongest association with APPG, predictive value for the development of PG and progression of APPG.
First, we analysed the clinical data obtained from the patients included in this study. TG, Chol, FBG, SUA, WBC, lymphocyte and T-score of APPG were significantly higher in patients with APPG than in NCs (P < 0.05), whereas FBG, WBC, lymphocyte and T-score were significantly higher in patients with APPG than in patients with AH (P < 0.05). Moreover, TG, FBG, SUA and T-score of patients with NAPPG were significantly higher than those in NCs (P < 0.05), whereas T-score of patients with NAPPG was significantly higher than that of patients with AH (P < 0.01; Table S2). The majority of patients with NAPPG and APPG displayed characteristics of first metatarsophalangeal joint involvement at symptom onset, whereas those with NAPPG had typical recurrent attacks, and most of those with APPG experienced one typical attack. SUA levels were 0.48-0.60 mmol/L in patients with NAPPG but were generally ≥0.60 mmol/L in those with APPG. Furthermore, urate deposition was observed in most joints or bursa in patients with NAPPG, but not in those with APPG (Table S3). The serum uric acid concentration in patients with primary gouty arthritis in the acute phase was high, but the T-score did not fluctuate too much (Figure 1). The results showed that the typical recurrence of inflammation in patients with primary gouty arthritis in the acute phase was associated with elevated serum uric acid.
From a clinical point of view, we believe that the observed ability of PD-1 mRNA to predict PG development is of key importance. We used qRT-PCR to detect PD-1 mRNA expression in the PBMCs of patients with different stages of PG. We observed that PD-1 mRNA expression was significantly lower in patients with NAPPG and APPG than in NCs (P = 0.0001 and P = 0.0001, respectively). In addition, it was significantly lower in patients with APPG than in those with NAPPG (P = 0.0062). Thus, PD-1 mRNA could be an auxiliary diagnostic biomarker for APPG. Moreover, PD-1 mRNA expression levels progressively decreased from NCs to patients with AH, NAPPG and APPG and were significantly lower in patients with APPG than in those with NAPPG and AH and NCs. They were significantly lower in the NAPPG group than in the NC group (P < 0.01; Figure 2).
These results indicate that PD-1 mRNA expression is significantly down-regulated in APPG. This effect may be due to malfunction and  Table 1), which is the classification criterion for diagnosing patients with PG. These results indicate that PD-1 mRNA is related to the clinical diagnosis of PG and could be used as an auxiliary diagnosis biomarker for PG.
Finally, we evaluated whether PD-1 mRNA could be an alternative to SUA as a biomarker for PG. Therefore, we performed ROC curve analysis to determine the diagnostic value of PD-1 mRNA levels in patients with different stages of PG. PD-1 mRNA expression was significantly lower in the PG group than in the NC group (P < 0.001; Figure 3A), with ROC curve analysis showing that SUA (AUC: 0.900, P < 0.05) has a better diagnostic performance than PD-1 mRNA (AUC: 0.818, P < 0.05), whereas the combination of both biomarkers displayed the best diagnostic performance for PG (AUC: 0.925, P < 0.05; Figure 3D). PD-1 mRNA expression was also significantly lower in the NAPPG group than in the NCs group (P < 0.001; Figure 3B), with ROC curve analysis revealing that SUA (AUC: 0.865, P < 0.05) had a better diagnostic performance than PD-1 mRNA (AUC: 0.749, P < 0.05) but their combination had the best diagnostic performance for NAPPG (AUC: 0.882, P < 0.05; Figure 3E). In addition, PD-1 mRNA levels were significantly lower in the APPG group than in the AH group (P < 0.001; Figure 3C), with ROC curve analysis revealing that SUA (AUC: 0.947, P < 0.05) displayed better diagnostic performance than PD-1 mRNA (AUC: 0.910, P < 0.05), whereas the combination of both biomarkers had the best diagnostic performance for APPG (AUC: 0.972, P < 0.05; Figure 3F). These results suggest that both SUA and PD-1 mRNA can be used as auxiliary diagnostic biomarkers for patients with different stages of PG, but that SUA is more effective than PD-1 mRNA when diagnosing different stages of PG. Moreover, the diagnostic performance was better in patients with APPG.
Next, we compared the diagnostic value of PD-1 mRNA levels when differentiating between patients with different PG stages and those with asymptomatic hyperuricemia. We evaluated whether the AUC of the ROC analysis differs between the two biomarkers. PD-1 mRNA expression was significantly lower in the APPG group than in the AH group (P = 0.0012; Figure 4C). ROC curve analysis showed that PD-1 mRNA (AUC: 0.967, P < 0.05) had better diagnostic performance than SUA (AUC: 0.749, P > 0.05) and the two markers combined (AUC: 0.923, P < 0.05; Figure 4D) for APPG. Thus, these results suggest that while both SUA and PD-1 mRNA can be used as auxiliary diagnostic biomarkers for APPG, PD-1 mRNA has better diagnostic performance than SUA. Since SUA levels change during  Figure 5D). These findings suggest that PD-1 mRNA is a promising biomarker for the differential diagnosis of non-PG and different stages of PG. In addition, PD-1 mRNA expression was lower in the NAPPG group than in the NPG group (P = 0.0489; Figure 5B), with ROC curve analysis revealing that PD-1 mRNA (AUC: 0.785, P < 0.05) displayed worse diagnostic performance than SUA (AUC: 0.804, P < 0.05) for NAPPG, while their combination displayed the best diagnostic performance (AUC: 0.812, P < 0.05) ( Figure 5E).
These findings indicate that compared to SUA, PD-1 mRNA is less able to distinguish between patients with non-PG and those with NAPPG. Moreover, PD-1 mRNA levels were significantly lower in the APPG group than in the NPG group (P = 0.0002; Figure 5C Despite these significant findings, this study has some limitations.
First, the sample size used in this study was relatively small; thus, our experimental data require verification by large-scale, multi-centre and multiethnic studies. Second, we cannot exclude the possibility that patients with PG were incorrectly classified, which would affect our conclusion that PD-1 mRNA can be used as an auxiliary diagnostic biomarker of APPG, or that other underlying diseases and genetic factors affected the outcome in these patients. Third, we used no other methods to verify PD-1 levels, such as flow cytometry to detect PD-1, ELISA to detect serum sPD-1 levels or Western blotting to detect PD-1 expression. However, since the qRT-PCR method has previously been used to identify auxiliary diagnostic biomarkers 33,34 successfully, we are confident that our methods are adequate for this preliminary study.
Importantly, the current study has significant strengths. This study is the first to examine the relationship between PD-1 mRNA and PG and is the first to identify that PD-1 mRNA expression levels were lower in the PBMCs of patients with APPG, thus could be used as an auxiliary diagnostic biomarker for APPG. Moreover, the combination of PD-1 mRNA and SUA is more effective than SUA alone

ACK N OWLED G EM ENTS
The authors would like to thank the volunteers who participated in this study, the healthcare professionals from The Third Affiliated Hospital of Guangzhou University of Chinese Medicine and Hai Zhu Qu Jiang Hai Jie Community Health Service Center.

CO N FLI C T O F I NTE R E S T
The authors disclose no conflict of interests.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.

O RCI D
Jing Su https://orcid.org/0000-0002-7704-8593 F I G U R E 5 Diagnostic performance of PD-1 mRNA in different stages of PG compared to patients with non-PG. A-C, Forest scatterplot: The qRT-PCR assay was performed to verify the expression levels of PD-1 mRNA in the NPG group, PG group, NAPPG group and APPG group. D-F, ROC curve analysis of SUA and PD-1 mRNA and the combination of both biomarkers. NPG: NCs + AH; PG: NAPPG + APPG