Expression of the IFNAR1 chain of type 1 interferon receptor in benign cells protects against progression of acute leukemia

Abstract

Type I interferons (IFN) were widely used for leukemia treatment. These cytokines act on cell surface receptor consisting of the IFNAR1/2 chains to induce anti-tumorigenic effects. Given that levels of IFNAR1 can be regulated by phosphorylation-driven ubiquitination and degradation that undermines IFN signaling and anti-tumorigenic effects, we sought to determine the importance of IFNAR1 downregulation in progression of acute leukemia. Using knock-in mice deficient in downregulation of IFNAR1 we uncovered that IFNAR1 expression in stromal benign cells functions to protect against progression of leukemia. We discuss putative mechanisms of this regulation and potential of therapeutic targeting of IFNAR1 downregulation to treat leukemia.

Introduction

While type I interferons (IFN) were widely used in therapy against diverse types of leukemia in the past, development of more potent and less toxic targeted therapies have largely replaced IFN as a standard of care. However, rapid development of resistance to targeted therapies prompt a renewed interest in IFN use including such as a part of combination or maintenance therapy ^1–5. All effects of IFN on cells require adequate expression of its cognate receptor that consists of IFNAR1 and IFNAR2 chain ^6,7. Levels of this receptor are tightly regulated by a number of mechanisms including phosphorylation-dependent ubiquitination, endocytosis and lysosomal degradation of IFNAR1 ^8–10. Activation of the mechanisms accelerating downregulation of IFNAR1 and subsequent suppression of IFN signaling may play a critical role not only in development and progression of diverse malignancies but also in resistance to IFN therapy ⁷.

Alterations of IFN signaling were indeed found in many oncologic diseases including leukemia ^7,11–16. Furthermore, previous studies clearly demonstrated an important role of IFNAR1 in leukemia. For example, ablation of Ifnar1 significantly stimulated leukemogenesis induced by RUNX1-ETO9a fusion oncogene ¹⁷ and by v-Abl oncogene ¹⁸.

Anti-tumorigenic functions of IFN are mediated by their direct anti-proliferative and pro-apoptotic effects on malignant cells but also by indirect mechanisms. The latter are targeted to benign cells and include inhibition of angiogenesis and stimulation of the anti-tumor immunity⁶. Importantly, factors produced in tumor cells can downregulate IFNAR1 in either malignant¹⁹ or benign cells ²⁰. Previous findings that Bcr-Abl fusion oncogene can signal to downregulate the IFNAR1 chain of IFN receptor ²¹ prompted us to determine the role of IFNAR1 in malignant and benign cells in defensive mechanisms against acute leukemia. We found that inability to downregulate IFNAR1 in the benign stromal cells impedes the progression of Bcr-Abl-harboring leukemia.

Results

Our previous studies demonstrated that hyperactive fusion oncogene Bcr-Abl expressed in non-leukemic cells is capable of stimulating the phosphorylation-dependent ubiquitination and downregulation of IFNAR1 ²¹. Thus, we sought to determine whether expression of Bcr-Abl can affect the levels of IFNAR1 in primary bone marrow cells from wild type mice or mice that harbor the knocked-in Ifnar1^S526A allele, which expresses mutant IFNAR1 resistant to ubiquitination and degradation. These mice were previously shown to display a normal development and, when unchallenged with tumor or inflammatory conditions, did not differ from wild type littermates in organs weight, frequencies of specific blood cell types or in serum biochemical parameters ^22,23. We isolated bone marrow cells from donor mice, transduced these cells with a retroviral vector for expression of p210^{Bcr-Abl 24} (at ~20–30% transduction efficiency) and examined the resulting levels of IFNAR1 and ability of these cells to induced leukemia when transplanted into lethally irradiated recipient wild type mice (Figure 1A).

Figure 1. Retroviral transfer of Bcr-Abl downregulates IFNAR1 in both wild type (WT) and SA bone marrow cells.

(A) Schematics of the experimental design. Interleukin-7–primed bone marrow (BM) cells from WT and SA mice were transduced with pMSCV-P210^BCR-ABL-IRES-GFP, and then transplanted into lethally irradiated WT recipients.

(B) Representative FACS analysis and quantification of the cell surface IFNAR1 levels in p210^BCR-ABL transformed (GFP+) WT or SA BM (n=4 in each group,P=0.0004 and 0.0001 respectively).

(C) Analysis of the white blood cells (WBC) counts in the recipient mice at the indicated time after bone marrow transplantation. Mean values (n=5 in each group) ±SEM are shown.

(D) Kaplan-Meier survival analysis of recipient wild type mice that received p210^Bcr-Abl transduced WT or SA bone marrow cells (P=0.69).

Flow cytometry analysis of the cell surface IFNAR1 levels demonstrated that transduction of p210^Bcr-Abl robustly downregulated IFNAR1 levels in the GFP⁺ wild type bone marrow cells (Figure 1B). Intriguingly, a similar extent of downregulation was also observed in the GFP-positive bone marrow from the knock-in mice that harbor phosphorylation/ubiquitination-deficient Ifnar1^S526A allele (henceforth termed “SA”). These results suggest that forced expression of Bcr-Abl in primary bone marrow cells can trigger IFNAR1 downregulation in an ubiquitination-independent manner. When transplanted into irradiated wild type recipient mouse, p210^Bcr-Abl-harboring wild type bone marrow cells caused a rapid development of leukemia leading to a massive increase in the number of white blood cells (Figure 1C) and development of the terminal disease (Figure 1D). These results are consistent with previously published observations in this model ^24–27. Notably, these phenotypes were observed regardless of whether donor bone marrow cells were derived from wild type or from SA mice (Figure 1C-D). Given this lack of difference in IFNAR1 downregulation and leukemia progression, it appears that this bone marrow transplantation model is not suitable for delineating the role of the cell-autonomous downregulation of IFNAR1 in the malignant cells in progression of leukemia.

Thus, we have sought to investigate the importance of downregulation of IFNAR1 in benign stromal cells. To this end, we used the injection of cells from an established leukemic cell line (whose IFNAR1 status remained unchanged) into the host mice differing in their ability to downregulate IFNAR1 (WT or SA). For our experiments, we selected a syngeneic p185^Bcr-Abl-GFP-harboring leukemic cells isolated from the v-Abl oncogene-induced robust leukemia mouse model because these cells rapidly induce leukemia and the disease progression might be affected by the IFNAR1 status ¹⁸. We injected these cells into either wild type or SA mice and compared the rates of leukemia progression (Figure 2A).

Figure2. Stromal IFNAR1 levels determine progression of acute leukemia.

(A) Schematics of the experimental design. Equal numbers (3×10⁵ per mouse) of leukemic cells generated from the pMSCV-p185^BCR-ABL-IRES-GFP transgenic mouse model were injected in WT and SA mice, which were then observed for leukemia development and progression.

(B) Representative FACS analysis plots showing the percentage of leukemic cells (GFP⁺) in WT or SA bone marrow 10 days post injection. Mean values (n=5 in each group) ±SEM are shown. P=0. 0003

(C) Representative FACS analysis plots showed percentage of leukemic cells (GFP⁺) in WT or SA bone marrow 20 days post leukemic cells injection. Mean values (n=5 in each group) ±SEM are shown. P=0. 0025

Monitoring the number of GFP-positive leukemic cells revealed a notable accumulation of these cells in the bone marrow of wild type recipient mice within 10 days after inoculation (Figure 2B). The progression of the disease in wild type mice was very rapid and GFP-positive cells constituted approximately 80% of all bone marrow cells in wild type mice at day 20 (Figure 2C). Remarkably, SA mice that initially received the same amount of leukemic GFP-positive cells, exhibited a dramatic delay in accumulation of these cells in bone marrow at both 10 and 20 days after inoculation (Figures 2B-C) suggesting that expression of stabilized IFNAR1 in benign cells can inhibit the progression of leukemia.

To further examine this possibility, we compared the number of white blood cells in wild type and SA mice inoculated with leukemic cells. As seen from Figure 3A, a significantly lower number of leukocytes was found in the blood of SA mice. Similarly, under these conditions, SA mice displayed a lower spleen weight compared to WT mice (Figure 3B). Consistent with these results, the mortality in SA mice was significantly delayed. While wild type mice had to be euthanized on humane grounds because of their moribund appearance by day 24 after leukemic cells inoculation, the earliest lethality in SA mice occurred on day 25 and some of these mice survived past day 50 (Figure 3C). These results suggest that downregulation of IFNAR1 in non-leukemic benign stromal cells promotes progression of leukemia.

Figure3. Progression of leukemia is delayed in SA mice.

(A) Analysis of the white blood cells (WBC) counts 20 days post leukemic cells injection. Mean values (n=5 in each group) ±SEM are shown. P=0.0028

(B) Analysis of the spleen mass 20 days post leukemic cells injection. Mean values (n=5 in each group) ±SEM are shown. P=0.0004

(C) Kaplan-Meier survival analysis of WT or SA mice injected with p185^BCR-ABL leukemic cells (n=10 in each group). The difference in survival is significant (P=0.0042), as determined by the log-rank test

Discussion

Previous data showed that expression of Bcr-Abl in HeLa cells can stimulate the ubiquitination and degradation of wild type IFNAR1 but not SA mutant ²¹. Thus, we attempted to use bone marrow from wild type and SA mice in a model of leukemia induced by a retroviral Bcr-Abl expression in bone marrow followed by its transplantation ²⁴. However, under these experimental condition, transduction of bone marrow with high titer of Bcr-Abl retrovirus downregulated the receptor in both wild type and SA cells (Figure 1). Lower retroviral titer allowed us to achieve disparate levels of receptor (downregulated in wild type but not SA bone marrow); however, this titer was insufficient to induce leukemia (data not shown). This technical problem precluded us from investigating the cell autonomous role of IFNAR1 downregulation in leukemia development and progression.

Nevertheless, data presented here suggest that expression of IFNAR1 in benign cells plays a defensive role against leukemia progression. Our data demonstrating the role of IFNAR1 in controlling leukemia are consistent with previous observations revealing that conventional ablation of Ifnar1 stimulates leukemogenesis induced in two different mouse models ^17,18. Intriguingly, while the role of natural killer (NK) cells in controlling the Abelson oncogene-driven disease has been proven ²⁸, conditional acute ablation of Ifnar1 in NK cells did not accelerate leukemia progression ¹⁸. Based on our current studies, we believe that the importance of IFNAR1 signaling in NK cells might be underappreciated because downregulation of IFNAR1 within benign stromal compartment (as shown for cytotoxic T lymphocytes²⁰) may partially mimic Ifnar1 genetic ablation thereby minimizing the phenotypic differences between Ifnar1-null and wild type animals. Thus, it would be of interest to assess whether a localized stabilization in IFNAR1 levels within NK cells may inhibit leukemia progression. Future use of tissue-specific and inducible SA knock-in animals in models in which IFNAR1 is targeted for accelerated degradation will likely emphasize the importance of IFN signaling in systems comparing wild type mice with total or tissue-specific receptor knockout.

It remains to be determined how IFNAR1 downregulation in the benign cells can accelerate the progression of leukemia. Current ongoing studies are focused at the mechanisms related to providing a proper niche for the leukemic stem cells and potentially undermining the immune response against the leukemic cells. Additional studies are expected to delineate the mechanism of IFNAR1 downregulation of benign cells and determine whether targeting these mechanisms can be utilized for leukemia therapies. While the role of protein kinase D2 in downregulation of IFNAR1 in response to Bcr-Abl and IFN signaling ^21,29,30 is shown, it remains to be seen whether this kinase contributes to IFNAR1 regulation in benign tissues of acute leukemia patients.

An alternative pathway that involves a concerted action of p38α and casein kinase 1α (CK1α) in IFNAR1 downregulation has been also described ⁷. Recent genetic studies proved a critical role of this pathway in control of IFNAR1 levels and signaling in some cultured cells and the intestinal tissues ^31,32. Furthermore, given that CK1α mutations are found in myelodysplastic syndrome ³³, and therapeutic targeting CK1α by either catalytic inhibitors ^33,34 or lenalidomide that induces ubiquitination and degradation of CK1α ³⁵ provides an impetus for investigating potential genetic alterations of CK1α and the efficacy of CK1α targeting in acute leukemia.

Role of the p38α stress activated protein kinase that controls IFNAR1 stability upstream of CK1α was also demonstrated ^36,37. Recently, a proof of principle studies demonstrated a therapeutic value of targeting p38α or its downstream kinase MK2 in leukemia and showing that inhibiting this pathway can increase the anti-leukemic efficacy of Smac-mimetics ³⁸. Given that Smac mimetics are known to act in a manner dependent on IFN signaling ³⁹, determining the impact of IFNAR1 stabilization by p38 inhibition on progression of leukemia is warranted.

Materials and Methods

Mice

All experiments with animals were carried out under the protocols 803995 and 804470 approved by the IACUC of The University of Pennsylvania. Eight-ten week old C57BL/6 littermate mice including wild type (Ifnar1^+/+) mice and Ifnar1^tm1.1Syfu (Ifnar1^SA) mice ^22,23 were used.

Flow cytometry

Bone marrow cells were suspended in FACS buffer (PBS, 1%BSA) and blocked with anti-mouse CD16/32 antibodies for 20 min, then stained with anti-IFNAR1-PE (Biolegend) for 30min. All samples were analyzed using LSR Fortessa flow cytometer (BD Biosciences) and data were analyzed with FlowJo (Tree Star).

Bone marrow transduction and transplantation

Induction of leukemia by transplanting Bcr-Abl-expressing bone marrow cells was carried out as previously described ²⁴. Briefly, WT or SA mice BM cells were harvested and cultured in IMDM media and 15% FBS supplemented with 10 ng/ml murine IL-7 (Genzyme). After 3 days culture, viable cells were put in a 6-well plate and spin infected with MSCV-p210^BCR-ABL-IRES/GFP retrovirus. Twenty four hr later, the second round of spin infection was performed for another 24 hr. These cells were collected, washed with PBS and the percentage of GFP-positive cells was analyzed by FACS to determine overall transduction efficiency (usually 20–30%). These cells were then injected into the tail-vein of lethally irradiated (9Gy) WT recipient mice at 1×10⁶ cells per mouse. For transplantation of P185^Bcr-Abl leukemic cells (described in ref ¹⁸), 3×10⁵ of these cells were injected into the tail veins of WT or SA recipient mice. Moribund mice were sacrificed and were examined for white blood cell counts, spleen weight and presence of leukemic cells by FACS analysis.

Statistics

The presented quantified experiments were evaluated based on the average of at least three independent experiments that performed in triplicate and calculated as mean values ± SEM.

We used unpaired two-tailed t test to assess mean difference. The Kaplan-Meier method was used for the survival curves, and log-rank test was used to compare the survival curves. The nominal P value of <0.05 is used as statistical significant threshold, and considered as follows: ∗P < 0.05; ∗∗P < 0.01; and ∗∗∗P < 0.001.