Expression of CXCR4 on T-cell subsets and Plasma IL-17 Concentrations in Patients with Aplastic Anaemia

Acquired aplastic anaemia (AA) is caused by T-cells migrating to and attacking bone marrow (BM) in response to chemokines (e.g., CXCR4). We investigated CXCR4 expressions on circulating T-cell subsets, plasma IL-17A concentrations, and their correlations with AA manifestations. We enrolled 71 patients with acquired AA (36 severe AA cases [SAA] and 35 non-severe AA cases [NSAA]) and 42 healthy volunteers. We used flow cytometry and ELISA to measure circulating CD4+ and CD8+ T-cells, their CXCR4 expressions, and plasma IL-17A concentrations. Compared to the healthy controls, SAA patients had fewer peripheral CD4+ T-cells, more CD8+ T-cells, and a significantly decreased CD4+/CD8+ ratio which was positively correlated with AA manifestations. Patients with SAA or NSAA had higher proportions of CD4+CXCR4+ and CD8+CXCR4+ T-cells, which were negatively correlated with haemoglobin concentrations and absolute neutrophil counts. Patients with SAA or NSAA had higher plasma IL-17A concentrations, which were negatively correlated with AA manifestations and the CD4+/CD8+ ratio. IL-17A concentrations showed a very week correlation with CD4+CXCR4+ T-cells frequencies, and no correlation with CD8+CXCR4+ T-cells frequencies. Aberrant CXCR4 expression may allow circulating T-cells, especially CD8+ T-cells, to infiltrate BM during AA progression. Elevated IL-17A concentrations may contribute to AA progression outside of the CXCR4-SDF-1α axis.

The present study aimed to investigate the expression of CXCR4 on peripheral blood CD4 + and CD8 + T-cells, the plasma concentration of IL-17A, and their relationships in patients with aAA. Our results suggest that enhanced CXCR4 expression on CD4 + T-cells was positively correlated with increased plasma IL-17A concentrations in patients with AA. In addition, the level of CXCR4 expression on T-cells and IL-17A concentration were both positively correlated with AA severity.

Methods
This prospective study was performed in accordance with a protocol that was approved by the Ethics Committees of the Chinese Human Genome Project and the West China Hospital. All participants provided their written informed consent.

Participants
We enrolled 42 healthy volunteers (19 men and 23 women) and 71 patients with newly diagnosed aAA (32 men and 39 women) from the Hematology Department of the West China Hospital (Sichuan University) between July 2015 and April 2016. All patients were diagnosed and classified based on the Guidelines for the Diagnosis and Management of Aplastic Anemia 18 . The disease was considered severe AA (SAA) if the patient had pancytopenia with ≥2 of the following parameters: a neutrophil count of <0.5 × 10 9 /L, a platelet count of <20 × 10 9 /L, and a reticulocyte count of <20 × 10 9 /L with hypocellular bone marrow. We excluded patients with congenital AA and other haematological or autoimmune diseases, as well as patients with infections [including hepatitis B virus (HBV), hepatitis C virus (HCV), human immunodeficiency virus (HIV), Epstein-Barr virus (EBV) and tuberculosis (TB) infection]. The characteristics of the patients and healthy controls are shown in Table 1.

Blood samples
All participants provided a 3-ml fasting blood sample, which was collected into a BD Vacutainer tube containing sodium heparin at 8:00-9:00 AM. The whole blood was used for flow cytometry. Plasma was obtained after centrifugation and stored at −80 °C for the cytokine testing.

Enzyme-linked immunosorbent assay (ELISA)
The IL-17A level was determined using a specific human IL-17A Platinum ELISA kit (Cat#BMS2016; Bender Med Systems, Burlingame, USA). The limit and sensitivity of detection for the ELISA kit are 1.6-100 pg/ml and 0.5 pg/ml, respectively. Protocol recommended by manufacturer was followed. All samples were measured in duplicate. Results are expressed as pg/ml.

Statistical analysis
Summary statistics (number and percentage or median and interquartile range [IQR]) were used to describe the participants' baseline characteristics. Numerical results were analysed using the IBM SPSS software (version 20.0; IBM Corp., Armonk, NY). The significance level was set at 5% for all statistical tests. The data were initially analysed using analysis of variance or the Kruskal-Wallis H test. If a significant result was observed, the Student-Newman-Keuls or Mann-Whitney tests were used to detect inter-group differences. Spearman's correlation coefficient was used to test the correlations between pairs of two continuous variables.   Table 1. Characteristics of the patients with aplastic anaemia and the healthy controls. Data are presented as mean (standard deviation) for most variables, although sex is presented as n (%). SAA: severe aplastic anaemia; NSAA: non-severe aplastic anaemia.

Frequencies of circulating CXCR4 + T-cell subsets in patients with AA and healthy controls.
Gating on the CD4 + or CD8 + T-cells allowed us to calculate the frequencies of circulating CD4 + CXCR4 + or CD8 + CXCR4 + T-cells (Fig. 3A). Figure 3B showed the representative flow cytometric histograms of CXCR4 expression on CD4 + and CD8 + T-cells for three groups. As shown in Fig. 4A, the frequency of CD4 + CXCR4 + T-cells was significantly higher in the patients with SAA (83.2 ± 7.6%) and NSAA (81.0 ± 9.1%), compared to the healthy controls (76.8 ± 10.1%) (P < 0.05). However, there was no significant difference between the two AA groups (P > 0.05). Significant differences in the frequencies of CD8 + CXCR4 + T-cells were observed when we compared all three study groups (P < 0.01, Fig. 4B). The SAA group had the highest frequency of CD8 + CXCR4 + T-cells (89.1 ± 7.7%), which was followed by that in the NSAA group (82.4 ± 12.8%) and the control group (74.6 ± 16.8%).

Discussion
AA is primarily driven by the aberrant regulation of CD8 + T-cells and CD4 + Th1 cells, which facilitates T-cell sequestration to the BM and its subsequent destruction. In patients with aAA, activated circulating CD8 + T-cells are a lymphocyte subset that inhibit hematopoiesis 19 , and 28% of patients with AA have an elevated percentage of activated CD8 + T-cells in their BM 20 . In the present study, patients with severe AA had a significantly lower CD4 + /CD8 + T-cell ratio, which was the result of fewer CD4 + T-cells and more CD8 + T-cells. Furthermore, the ratio decreased with disease severity and was positively correlated with the patients' manifestations (including haemoglobin concentrations, platelet counts, and absolute neutrophil counts). These findings support the potentially critical role of CD8 + T-cells in the pathogenesis of aAA.
The mechanisms that govern the sequestration of pathogenic T-cells to the BM remain unclear. However, the CXCR4 chemokine receptor facilitates cellular chemotaxis in response to its natural ligand (SDF-1α), which is strongly expressed in the BM 8,21 . Aberrantly elevated CXCR4 concentrations have been reported in numerous autoimmune conditions, and may provide a means for malignant cells to target SDF-1α-enriched sites, such as the BM [14][15][16][17]22 . The only study to demonstrate CXCR4's critical role in facilitating pathogenic T-cells homing to the BM was performed in a mouse model of AA 17 , and data from patients with AA remain scarce. However, we used flow cytometry to evaluate circulating CD4 + and CD8 + T-cells from patients with aAA, and observed that high CXCR4 expression was associated with disease severity. These findings suggest that the interaction between CXCR4 and SDF-1α may provide a mechanism for sequestering pathogenic T-cells to the BM during AA progression. In addition, the percentage of CD8 + CXCR4 + T-cells was significantly higher than that of CD4 + CXCR4 + T-cells in patients with SAA, and was negatively correlated with absolute neutrophil counts. These findings suggest that CD8 + CXCR4 + T-cells may also play a critical role in AA exacerbation, which is indirectly supported by previous findings of high CXCR4 expression on BM-infiltrating CD8 + T-cells. Because IL-17A is a pro-inflammatory cytokine, its dysregulation is associated with various autoimmune disorders, such as type 1 diabetes, rheumatoid arthritis, and Sjögren's syndrome [23][24][25] . Furthermore, our data support previous findings that patients with active AA had elevated IL-17 concentrations 15,16 , and suggest that IL-17 may also play a role in AA progression. Further analyses revealed a moderate negative correlation between plasma IL-17A concentration and the CD4 + /CD8 + T-cell ratio, and negative correlation between IL-17A concentrations and the patients' clinical manifestations of AA. However, the correlation between IL-17A concentrations and CXCR4 expressions on circulating CD4 + or CD8 + T-cells was very weak or absent. When considered together, these results suggest that IL-17A may contribute to AA progression outside of the CXCR4-SDF-1α axis.
However, further studies of BM from patients with AA are needed to validate our findings. Actually, evaluating the role of CXCR4 and IL-17A on outcome after therapy, which need more evidences from further BM assessment and prospective analyses, is of great value in assessing efficacy of clinical management of AA. In addition, CCR4 + CCR6 + T cells were recently validated to produce IL-17 and expressed the Th17-specific transcription factor RORC, establishing a link between CD4 + T cell trafficking potential and immunologic function 26 . The co-expression of CCR6 in AA patients can be of interest. And, all these will be our future research contents and directions. To elucidate potential therapeutic approaches, these studies should evaluate the full array of chemokine receptors and adhesion molecules that can facilitate the sequestration of pathogenic CD4 + and CD8 + T-cells to the BM during AA progression.

Conclusions
Ours is the first study to reveal that aberrant CXCR4 expression may help sequester circulating T-cells, and especially CD8 + T-cells, to the BM during aAA progression. Elevated IL-17A concentrations may also help mediate the abnormal CD4 + /CD8 + T-cell ratio, but were not significantly associated with CXCR4 expression.