Immunomodulatory and direct activities of ropeginterferon alfa‐2b on cancer cells in mouse models of leukemia

Abstract Although ropeginterferon alfa‐2b has recently been clinically applied to myeloproliferative neoplasms with promising results, its antitumor mechanism has not been thoroughly investigated. Using a leukemia model developed in immunocompetent mice, we evaluated the direct cytotoxic effects and indirect effects induced by ropeginterferon alfa‐2b in tumor cells. Ropeginterferon alfa‐2b therapy significantly prolonged the survival of mice bearing leukemia cells and led to long‐term remission in some mice. Alternatively, conventional interferon‐alpha treatment slightly extended the survival and all mice died. When ropeginterferon alfa‐2b was administered to interferon‐alpha receptor 1–knockout mice after the development of leukemia to verify the direct effect on the tumor, the survival of these mice was slightly prolonged; nevertheless, all of them died. In vivo CD4+ or CD8+ T‐cell depletion resulted in a significant loss of therapeutic efficacy in mice. These results indicate that the host adoptive immunostimulatory effect of ropeginterferon alfa‐2b is the dominant mechanism through which tumor cells are suppressed. Moreover, mice in long‐term remission did not develop leukemia, even after tumor rechallenge. Rejection of rechallenge tumors was canceled only when both CD4+ and CD8+ T cells were removed in vivo, which indicates that each T‐cell group functions independently in immunological memory. We show that ropeginterferon alfa‐2b induces excellent antitumor immunomodulation in hosts. Our finding serves in devising therapeutic strategies with ropeginterferon alfa‐2b.


| INTRODUC TI ON
Interferon-alpha is a pleiotropic cytokine that was first identified as a viral replication inhibitor. Nevertheless, the IFNα function has since been extended to cancer suppression. 1-3 IFNα is now approved for the treatment of both solid and hematologic tumors. [4][5][6][7][8] The underlying mechanisms of the IFNα function in tumor suppression are currently a widely studied subject, and it has long been thought that IFNα suppresses tumor development through its direct functions in tumor cells. 9,10 Also, IFNα may attenuate tumor progression by activating host immune cells. [11][12][13][14] Despite these findings, it remains unclear whether IFNα exerts its antitumor effects through stimulation of the host immune system, by a direct effect on tumor cells, or both. 15 Ropeginterferon alfa-2b (ropeg) is a monopegylated IFNα that was developed for the treatment of myeloproliferative neoplasms.
Contrary to other PEG-IFN compounds, ropeg comprises a single positional isomer resulting in an extended elimination half-life, enabling less frequent dosing (every other week or monthly during maintenance therapy), and shows promising results for the treatment of patients with polycythemia vera. [16][17][18] Despite increasing clinical evidence, the mechanism by which ropeg treatment controls tumor cells remains unclear. In this study, we used an immunocompetent mouse model mimicking acute leukemia and showed its antitumor mechanism from both the direct and indirect effects of ropeg.

| Ex vivo and in vivo experiments
Mice, leukemia cell lines, flow cytometry analysis, next-generation sequencing analysis, drug reagents, and study design are described in Appendix S1. Animal experiments were approved by Osaka City University Animal Ethics Committee and performed according to the institutional guidelines of Osaka City University.

| Statistical analysis
P values were calculated using a two-tailed Student's t test, nonparametric Mann-Whitney U test, or Kruskal-Wallis test. The log-rank test was used for the analysis of survival. P values of <0.05 were considered statistically significant. All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University), which is a graphical user interface for R (The R Foundation for Statistical Computing). 19

| Ropeginterferon alfa-2b decreases the leukemia burden in vivo and leads to the long-term survival of mice
To assess the in vivo efficacy of ropeg for leukemia, we utilized BA-1, a murine B-cell leukemia cell line that shows aggressive development, leading to death with infiltration of the bone marrow, blood, and spleen after injection into normal C57BL/6 mice without prior treatment (data not shown). To investigate the overall therapeutic potential of ropeg treatment regarding leukemia, we inoculated C57BL/6 mice with BA-1 cells via tail vein injection ( Figure 1A). A significant reduction in the tumor burden in the peripheral blood was observed, especially in the ropeg-treated group ( Figure 1B). We next evaluated the effects of ropeg treatment on the survival of the mice. Ropeg treatment led to longterm remission in four of eight (50%) mice, whereas all mice in the rIFNα treatment group died at approximately 40 days, which demonstrated the potential of stronger antitumor effect of ropeg in vivo in comparison with rIFNα in leukemia treatment.
Increasing the dose of rIFNα did not contribute to prolonged survival ( Figure S1). The antitumor effect of ropeg was also tested in a tumor model using EL4, a murine T-cell leukemia cell line ( Figure S2A). Ropeg treatment significantly delayed tumor growth of EL4 tumors, whereas rIFNα treatment did not prevent tumor growth ( Figure S2B).

| Ropeginterferon alfa-2b increases apoptosis and cell cycle arrest of leukemic cells in a time-dependent manner
We intended to evaluate the direct cytotoxic effect of ropeg and rIFNα on the proliferation of l BA-1 cells in vitro to elucidate the mechanism of leukemia rejection. Both ropeg and rIFNα suppressed BA-1 growth, which was more pronounced in the ropegtreated group (Figure 2A). The same result was observed in EL4 ( Figure S3).
The apoptosis of BA-1 cells was significantly increased after exposure to ropeg and rIFNα. In terms of total cells, the ropeg group showed an increase in early-and late-apoptotic cells in a timedependent manner, whereas the ratio of early and late apoptotic cells was most remarkable in the rIFNα group after 24 hours ( Figure 2B).

Ropeg significantly increased early and late apoptotic cells in com-
parison with rIFNα and control 72 hours after the start of coculture (Table 1). Regarding viable cells, ropeg induced an increase in the MFI of PE-annexin over time, whereas rIFNα peaked at 24 hours and did not increase thereafter ( Figure 2C). IFNα induces apoptosis in malignant cells via activation of caspase-3. 20, 21 We confirmed that ropeg promoted a time-dependent increase in caspase-3 expression, which was not observed with rIFNα ( Figure 2D). Additionally, cell cycle analyses were performed under the same culture conditions described in the apoptosis assay. When BA-1 cells were cocultured with ropeg or rIFNα, the G1 phase of the cells increased ( Figure 2E).
The increase in the G1 phase after 72 hours was more remarkable in the ropeg group ( Table 1). IFNα has been shown to induce p21, which inhibits the active cyclin D/CDK4 complex, resulting in G1 phase arrest. [22][23][24] We found that exposure to rIFNα induces an increase in p21 expression in BA-1 cells, which peaks out at 24 hours, whereas exposure to ropeg induces an increase in its expression over time ( Figure 2F).
When type I IFN binds to a receptor consisting of Ifnar, STAT1 undergoes tyrosine phosphorylation by the JAK-Stat pathway. Thus, we checked whether ropeg promoted p-STAT1. The MFI of p-STAT1 was increased via coculture with ropeg or rIFNα. Additionally, the MFI of p-STAT1 increased in a time-dependent manner in the ropegtreated group, which was in line with the time-dependent effects observed in the apoptosis and cell cycle assays ( Figure 2G). We administered rIFNα continuously every 24 hours in the medium cul-

| Loss of indirect effect attenuates the antileukemic effect of ropeg in vivo
Based on the direct cytotoxic effects of ropeg for BA-1 cells confirmed in vitro, we next investigated the direct effect on leukemia cells in vivo ( Figure 3A). Both ropeg and rIFNα therapies significantly enhanced the survival of Infar1 −/− mice bearing BA-1 cells, in which these therapies do not act systemically. Ropeg treatment led to a survival benefit in comparison with rIFNα treatment.
Although ropeg treatment led to the long-term remission of leukemia in some immunocompetent mice, all Ifnar1 −/− mice died within 17-24 days after tumor inoculation ( Figure 3B). The direct action of ropeg on EL4 cells did not inhibit tumor cell growth in Ifnar1 −/− mice ( Figure S4A,B). These findings demonstrate that the immune system plays a prominent role in the ropeg-mediated antitumor effect in vivo.

| Indirect antitumor effects of ropeg significantly inhibit leukemia cell growth in vivo and lead to the long-term survival of mice
Based on these findings, we next investigated whether the tumorsuppressive effects and prolonged survival of leukemia-bearing mice, which we observed in ropeg-treated wild-type mice, could be F I G U R E 1 Ropeg treatment in BA-1 cell-injected mice. A, A schematic illustration of the therapy protocol for in vivo experiments. B, Single-cell suspensions of blood at 7, 14, 21, 28, 35, and 42 d after tumor inoculation were analyzed via flow cytometry to determine the fraction of GFP+ cells (BA-1 cells). Error bar indicates the mean and SE. Statistical significance was determined using Student's t test with comparisons indicated by brackets (**p < 0.01). C, Survival data in C57BL/6 mice were plotted in a Kaplan-Meier survival curve, and statistical significance was calculated with the log-rank test. No treatment (n = 11), rIFNα-treated (n = 8), and ropeg-treated C57BL/6 (n = 8) mice. Data were derived from at least three independent experiments (*p < 0.05; **p < 0.01) confirmed based on indirect effect alone. To confirm this, we generated BA-1/Ifnar1 −/− by deleting the Ifnar1 gene using the CRISPR/ Cas9 system, on which ropeg did not have a direct antitumor effect. (Figure 4A,B). The indirect antitumor effect of ropeg was evaluated by inducing BA-1/Ifnar1 −/− leukemia in wild-type C57BL/6 mice ( Figure 4C). Ropeg treatment significantly prolonged the survival of mice in comparison with the rIFNα-treated and untreated groups ( Figure 4D). Furthermore, ropeg treatment introduced long-term remission in four of the six (67%) treated mice, which was confirmed in ropeg-treated mice with normal BA-1. These results showed that

| Changes in splenic lymphocytes after the administration of ropeg
These results suggested that ropeg treatment exerts a more critical antitumor effect than the conventional IFNα due to the immunomodulation in mice. Thus, we investigated the proportion and the absolute number of spleen lymphocytes after the administration of ropeg or rIFNα in comparison with drug-free mice.
As shown in Figure 5A, we characterized the lymphocyte sub-  Note: BA-1 cells were exposed to compounds at the indicated concentrations for 72 h before analysis was performed. The data presented correspond to the mean ±standard error of the mean of three independent experiments. All experiments were performed at least three times with similar results.

| Depletion of either CD4 + or CD8 + T cells eliminates the antitumor effect of ropeg in vivo
Based on these findings, we then aimed to identify which immune cell types were crucial for the survival benefit that we observed in treated wild-type mice. We therefore evaluated the efficacy of ropeg in C57BL/ 6 mice bearing BA-1 cells that were depleted of CD4 + , CD8 + T cells, NK1.1 + cells, and CD19 + B cells, respectively ( Figure 6A). CD4 + or CD8 + T-cell depletion resulted in a significant loss of therapeutic efficacy ( Figure 6B). Conversely,  Figure 6B). These data show that the treatment efficacy of ropeg is dependent on adaptive immunity, with CD4 + and CD8 + T cells being essential mediators of the antitumoral immune response.

| Immunological memory is induced and T cells play a central role in mice that overcome leukemia after ropeg treatment
Next, we evaluated whether long-lasting immunological memory was established in ropeg-treated mice that had survived the BA-1 challenge. Surviving mice were rechallenged with BA-1 cells at least 100 days after the first BA-1 cell inoculation and compared with tumor-inoculated control mice ( Figure 7A). Surviving mice withstood the BA-1 rechallenge in all cases ( Figure 7B). This suggested that mice that overcame leukemia with ropeg treatment acquired an antitumor immune memory, and we further investigated whether this was an acquired immunity specific to BA-1 cells. BA-1-surviving mice with ropeg treatment produced a greater percentage of CD4 + or CD8 + T cells producing cytotoxic cytokines compared with BA-1-naïve mice after the injection of BA-1 cells. Conversely, T cells from mice that had overcome BA-1 leukemia with ropeg treatment did not respond to intravenous infusion of EL4 cells ( Figure 7C,D).
These results suggest that ropeg treatment induces tumor-specific

| DISCUSS ION
Immunotherapy using IFNα has been shown to rely on two mechanisms, namely a direct cytotoxic or cytostatic effect on the tumor cells via the induction of apoptosis and cell cycle arrest and the induction of a cellular antitumoral immune response. 32,33 Although ropeg, which is a novel long-acting pegylated IFNα, has been confirmed to have superior results in myeloproliferative neoplasms, the mechanism remains unclear. 16,18,34 In this study, we showed that ropeg provided better survival benefit in comparison with conven- Wild-type mice treated with ropeg rejected Ifnar1 −/− leukemic cells.
These results indicate that long-term survival after the induction of remission requires an antitumor immune response after the administration of ropeg. As a factor explaining these phenomena, we confirmed a numerical increase in CD8 + T cells, CD4 + T cells, and B cells in ropeg-treated mice, which are key components of antitumor adaptive immunity. 37 Especially, cytotoxic CD8 + T cells play a pivotal role in providing effective antigen-specific immunity against tumors, 38,39 and CD4 + T cells primarily mediate antitumor immunity by providing croenvironment. 45 We also found that ropeg treatment induces an cytokines, whereas they did not respond to EL4 inoculation. These results suggest that ropeg treatment induces both tumor-specific memory CD4 + and CD8 + T cells, which independently have the potential to prevent the host from tumor relapse.
In summary, ropeg treatment induces an immune cell-mediated response in preclinical cancer models, leads to long-term survival by enhancing adoptive immunity in T cells, and finally establishes an immunological memory that is protective against tumor relapse.
This research will help inform future clinical research that seeks to develop the most effective strategies to implement this highly effective therapeutic agent in the treatment of cancer patients.

E TH I C A L S TATEM ENT
Animal experiments were approved by the Osaka City University Animal Ethics Committee and performed according to the institutional guidelines of Osaka City University.

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.