The effectiveness of a novel treatment of TIM‐3(−) NK cells infusion in murine models of immune‐mediated bone marrow failure

Abstract Background T‐cell immunoglobulin and mucin‐containing domain (TIM)‐3 exerts its inhibitory effect on NK cells and participates in the immune pathogenesis of SAA. In this study, we aimed to explore a novel treatment method of TIM‐3(+) NK or TIM‐3(−) NK cell infusion in combination with immunosuppressive therapy for bone marrow failure (BMF)/aplastic anemia (AA) mice. Methods BMF/AA mouse model was constructed. The TIM‐3 expression and functional molecules on TIM‐3(+) and TIM‐3(−) NK cells of the BMF group, total body irradiation (TBI) group, and normal control (NC) group mice were detected by flow cytometry. After treatment, the general condition, whole blood cell and bone marrow cell (BMC) count, and immune condition of mice from each group were compared. Results TIM‐3 expression in the peripheral blood NK cells of BMF mice was significantly lower than that of the TBI and NC group mice. TIM‐3(−) NK cells expressed more NKG2D receptors than TIM‐3(+) NK cells. The levels of P‐Akt and PI3K in TIM‐3(−) NK cells were higher than those in TIM‐3(+) NK cells. On the 17th day after BMF induction, the weight, peripheral whole blood cell count, and BMC count of BMF mice decreased significantly compared with that of the NC group mice. The therapeutic effect in the TIM‐3(−) NK cell treatment group was better than that in the TIM‐3(+) NK cell treatment and CsA treatment groups. Concurrently, the ratio of CD4+T and CD8+T cells of BMF mice was significantly lower than that of the NC group mice. The therapeutic effect in CsA + TIM‐3(−) NK group was more significant than that of the CsA treatment and the CsA + TIM‐3(+) NK groups. Conclusions In this study, we found that the general condition, peripheral whole blood cell and BMC count, and immune status of BMF mice improved significantly after CsA + TIM‐3(−) NK cell treatment. These results may provide further insights into the immune pathogenesis of SAA and novel therapeutic ideas for improving SAA treatment.


| INTRODUC TI ON
Aplastic anemia (AA) is a bone marrow failure (BMF) disorder resulting in bone marrow hypocellularity and peripheral pancytopenia. Abnormal immune system activation plays a key role in AA pathogenesis, and immunosuppressive therapy (IST) using antithymocyte globulin (ATG) and cyclosporine A (CsA) can effectively treat the disease. 1,2 The currently recognized AA pathogenesis includes a complex immune network where a proportion of natural killer (NK) cells that suppress the negative regulatory factors of immune response is significantly decreased. [3][4][5][6] Even with the application of IST and the new drug eltrombopag (ELT), 30% of patients still have poor treatment effects, 10% of effective patients have a recurrence possibility, and 16% of patients have treatmentrelated death because of drug-related side effects. 7,8 Therefore, further exploration of the role of NK cells and their subgroups in SAA pathogenesis can be helpful in discovering novel therapeutic targets and improve therapeutic efficacy. [9][10][11] In our previous study, we reported a decrease in NK cell numbers in patients with SAA, which played a protective role in the onset of SAA. [9][10][11] We also found low T-cell immunoglobulin and mucin-containing do-

| Construction of BMF mouse model
Specific-pathogen-free grade and 8 weeks old CB6F1 mice (strain code: 303), except for those in the normal control (NC) group, were exposed to a sublethal TBI dose of 5Gy before lymph node cell infusion. Lymph node cells were infused from the groin, brachial plexus, and axilla of C57BL/6 mice. CB6F1 mice were injected with lymphocytes (5 × 10 7 cells) through the caudal vein to induce BMF. The successfully constructed AA mouse model survived for more than 17 days. On the 17th day, mice were euthanized and blood samples were collected for further experiments.
The labeled protein bands were detected using Super ECL Plus detection reagent. All protein levels were normalized to GAPDH.

| Weight, whole blood cell and BMC count and bone marrow biopsy histology in mice
Blood was collected from the retro-orbital sinus, and BMCs were obtained by femoral cavity flushing. A whole blood cell count was performed using an automatic blood cell analyzer (MEK-722, Nihon Kohden). We fixed the mice femurs with 4% paraformaldehyde overnight and sectioned them for paraffin embedding. Following this, hematoxylin-eosin staining was performed, and slides were observed under a light microscope (Nikon Diaphot, Tokyo, Japan) with a 10× objective lens.

| The immune condition of mice
Blood was collected from the retro-orbital sinus, and BMCs were obtained by femoral cavity flushing. Many circulating CD4 + T and CD8 + T cell subsets in mice were identified by FCM.

| Statistical analyses
Data are presented as mean ± SD from at least three independent experiments. The significance between means was determined with multiple t-tests using the HolmSidak method, one-way analysis of variance (ANOVA) when Gaussian distribution was assumed, and Kruskal-Wallis test when Gaussian distribution was not assumed.
Multiple pairwise comparison tests were performed between the control and BMF groups. Statistical analyses were performed using GraphPad Prism version 6.0 software and Statistical Package for the Social Sciences 21.0 statistical software. The p-values of <.05 were considered statistically significant.

| TIM-3 expression in NK cells of AA mice was significantly lower than that of the TBI and NC group mice
TIM-3 expression in the peripheral NK cells of AA mice was 12.21 ± 10.06%, which was significantly lower than that of the TBI (19.24 ± 8.52%, p < .01) and NC group mice (23.52 ± 11.17%, p < .01) ( Figure 1). There was no statistically significant difference in TIM-3 expression in NK cells between TBI and NC group mice (p > .05).
Compared with the NC group, TIM-3 expression in NK cells in AA mice was similar to those in patients with AA, as in both cases, TIM-3 expression was lower than that of the NC group mice.

| TIM-3(−) NK cells in AA mice highly expressed activating receptors than TIM-3(+) NK cells
The expression of functional molecule inhibitory receptor NKG2A

| The general condition of AA mice improved significantly after TIM-3(−) NK cell infusion and CsA + TIM-3(−) NK cell infusion treatment
There was no body weight difference before TBI and lymphocyte infusion among all groups (Table 1, Figure 4A, and all p > .05). The NC and TBI group mice gained slight weight on the 17th day, whereas the weight of mice in the AA group was significantly lower than the basal weight on the 17th day compared with the NC group (p < .01).
On the 17th day, compared with the AA group, CsA treatment    c Compare with CsA treatment group (p < .05).

| The whole blood cell count of AA mice improved significantly after TIM-3(−) NK cell infusion and CsA + TIM-3(−) NK cell infusion treatment
The whole blood cell count on the 17th day after modeling is shown in Table 2. Pairwise comparison showed that compared with the NC group, the whole blood cell count in the AA group was significantly   However, after the treatment, the blood cell counts of the mice in each group were still lower than those in the NC group, indicating that it would take longer for the mice to return to normal whole blood cell count.

| BMC count and histopathology of AA mice improved significantly after TIM-3(−) NK cell infusion and CsA + TIM-3(−) NK cell infusion treatment
BMC count in the AA group was significantly lower than that in the NC group, indicating that the modeling was successful. A pairwise comparative ANOVA showed that the BMC count of the CsA treat-

| The immune condition of AA mice recovered after the treatment of TIM-3(−) NK cell reinfusion
The CD4 + T cell count, CD8 + T cell count and CD4 + T/CD8 + T ratio on day 17 after modeling is shown in Table 3. Pairwise comparison showed that compared with the NC group, the CD4 + T cell count and CD4 + T/CD8 + T ratio in the AA group was significantly decreased

| DISCUSS ION
To further explore the immune pathogenesis of SAA and improve the therapeutic effect, the present study conducted mouse modeling experiments on the basis of previous studies. 9

AUTH O R CO NTR I B UTI O N S
All authors contributed to the study's conception and design.
Material preparation, data collection and analysis were performed

CO N FLI C T O F I NTE R E S T S TATE M E NT
All authors report no conflicts of interest. The authors alone are responsible for the content and writing of this article.

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

I N FO R M E D CO N S E NT
Informed consent was obtained from all individual participants included in the study.