Monocytes engineered with iSNAP inhibit human B‐lymphoma progression

Abstract Monocytes are important regulators for the maintenance of homeostasis in innate and adaptive immune system and have been reported to play important role in cancer progression. CD47‐SIRPα recognition is a coinhibitory immune signal to inhibit phagocytosis in monocytes and macrophages and has been well‐known as the “Don't eat me” signal. By using an approach of integrated sensing and activating proteins (iSNAPs), we have rewired the CD47‐SIRPα axis to create iSNAP‐M which activates pathways in engineered human monocytes (iSNAP‐MC). The mRNA expression levels of the monocyte/macrophage markers CD11b, CD14, and CD31 are upregulated in iSNAP‐monocytes (iSNAP‐MC). With PMA induction, the iSNAP‐MC‐derived macrophages (iSNAP‐MΦ) showed upregelation in CD86 and CD80, but not CD206. TNFα expression and secretion were also increased in iSNAP‐MΦ. Furthermore, the injection of iSNAP‐MC into mice bearing human B‐lymphoma tumors led to the suppression of tumor progression. Therefore, the engineered monocytes, via blockage of coinhibitory immune signals by rewiring CD47‐SIRPα axis, can be applied to suppress target tumors for cancer immunotherapy.


| INTRODUCTION
Human innate and adaptive immune system maintains homeostasis and prevents carcinogenesis. Innate immunity also plays a crucial role for the activation of adaptive immunity. Monocytes are innate immune cells which can migrate into tissues and differentiate into macrophages in response to inflammatory stimulations and chemotaxis. 1 Macrophages are phagocytosis cells and present antigens to activate T cells through the binding of major histocompatibility complex II and T-cell receptors. Macrophages can also engulf tumor cells by phagocytosis and further activate T cell response via antigen presenting. 2 Signal regulatory protein α (SIRPα) is a membrane glycoprotein highly expressed on macrophages, serving as a coinhibitory signal that interacts with CD47 and inhibits phagocytosis; so called the "Don't eat me" signal. 3 SHP1 and SHP2 are downstream tyrosine phosphatases of SIRPα activation that are involved in multiple signaling pathways to modulate macrophage phenotype. 4,5 Systemic injection of macrophages coated by antibodies for tumor targeting and for SIRPα blocking can repress tumor growth. 6 Therefore, the CD47-SIRPα axis in macrophages is considered as a crucial target for antitumor immunotherapy.
When monocytes are differentiated to macrophages inside tissues, they are further activated. Polarization of activated macrophages into proinflammatory state (M1) and anti-inflammatory state (M2) has been reported to occur via different cytokine stimulations in vitro. 7 M1 macrophages secrete TNFα, IL-6, IL-8 and activate Th1 response, whereas M2 macrophages secrete IL-10 and activate Th2/Th17 response. In vivo, macrophages have more diverse and plastic responses to changes in local microenvironment. 8 Tumorassociated macrophages (TAMs) have been reported to promote tumor progression and develop M2 polarization in local tumor microenvironment. 9,10 In fact, a high degree of TAM infiltration has been reported to correlate with poor diagnosis in clinical studies. Consistent with this finding, switching M2 to M1 phenotype of TAMs can inhibit tumor growth. 11 Therefore, manipulating macrophage phenotypes in tumors can be an appealing therapeutic strategy for solid tumors.
In the last few decades, engineered macrophages with the blockage of CD 47-SIRPα recognition have shown promising efficacy to reduce tumor growth in non-Hodgkin lymphoma (NHL), acute lymphoblastic leukemia, acute myeloid leukemia, and myeloma. 12 Previously, we engineered integrated sensing and activating proteins (iSNAPs) that are capable of rewiring the signaling of the inhibitory SIRPα proteins into activating pathways in macrophages. 13 Overexpressing this iSNAPs in RAW264.7 macrophage cell lines and primary bone marrow-derived macrophages 13 enhanced the phagocytic ability of these engineered macrophages.
In this study, we introduced the engineered iSNAP in human THP1 monocytes to generate iSNAPs derived monocytes (iSNAP-MC) and macrophages (iSNAP-MΦ) and demonstrated that the engineered iSNAP-MC/MΦ have enhanced efficacy in suppressing human B-lymphoma in vivo.

| The establishment of iSNAP-MC
The genetic construct of iSNAP-M was described in our previous study 13 Table S1. All values were normalized with the human housekeeping gene GAPDH and expressed as mean ± SD. Statistical analysis was performed by double tailed t-test. Significance was determined from p < 0.05.

| Flow cytometry
The cells were collected and washed with PBS (Sigma) twice. Then, the cells were suspended in 100 μl PBS and stained with human CD86 (BD Pharmingen) and CD206 (BD Pharmingen) for 1 h on ice.
After PBS washing twice, the cell surface markers were analyzed by flow cytometry (BD Accuri™ C6 Plus).

| Cytokine array and ELISA assay
iSNAP-MC cells were induced to differentiate into macrophages for 48 h by the treatment of 20 ng/m PMA (ab120297; Abcam). The cell conditioned medium was collected and centrifuged to remove cell debris at 1000g for 5 min. The supernatant was collected and stored in À80 C. Cell secretion profile was analyzed with a cytokine array (ARY005B; R&D). Secretions of TNFα and IL-10 in the conditioned medium were quantified by using ELISA kits (DY210-05 and DY217B-05; BD Pharmingen). All values are expressed as mean ± SD and analyzed statistically by the double tailed t-test. Significance was determined from p < 0.05.

| Animals
Female and male nude mice age 4-6 weeks were purchased from Jackson Laboratories. The animals were housed in a biosafety vivarium and fed an autoclaved laboratory diet. At the conclusion of the study, the mice were euthanized with CO 2 inhalation, which was confirmed with cervical dislocation. All animal experiments were approved by Institutional Animal Care and Utilization Committee of the UCSD Institutional Animal Care and Use Committee (IACUC S14009).

| Treatment plan
The mice were randomized into three groups: No-treatment, and injections of THP-1 or iSNAP-MC. The "No-treatment" control mice were administered intratumorally with PBS. Mice receiving treatments were administered with THP-1 or iSNAP-MC cells (1 Â 10 6 ) via intratumoral injection. Tumor sizes were measured with a caliper weekly for 3 weeks. IVIS imaging was performed after intraperitoneal injection of D-luciferin (LUCK-1G; Gold Biotechnology) and acquired by using a Xenogen IVIS 200 system. Tumor volume was calculated with the formula of volume = (width 2 Â length)/2. 14

| Immunostaining
The tumors were collected and fixed with 4% paraformaldehyde.
After dehydration and paraffin embedding, tumor samples were sectioned for immunostaining. Anti-YFP antibody (MBS833304; MyBioSource) was used to detect injected iSNAP-MC in the tumors.

| Human IL-1β ELISA
Blood sampling in mice was performed after 1 week of the iSNAP-MC treatment. After sitting for 30 min in room temperature, blood samples were spin down at 2000g for 10 min. Then, the serum was collected and stored in À80 C. Human IL-1Β in the serum was quantified by using ELISA kits (Thermo Fisher Scientific).

| Statistical analysis of tumor growth
We modeled the tumor growth by the 2/3 power law, which assumes that the tumor growth occurs at the surface of a three dimensional solid tumor. 15 The tumor growth rate at time t was computed as (V(t)(  transmigration. [16][17][18] The result further showed that the expression of iSNAP-M in iSNAP-MC is at levels markedly higher than that of the endogenous SIRPα (Figure 2d).  Figure S1). Thirty days after the Toledo cell inoculation, when the tumor growth is observed to reach 5 mm in diameter, we started to treat the tumors by intratumoral injection of iSNAP-MC, confirmed by YFP immunostaining ( Figure S2). The protocol of treatment plan is shown in Figure 5a, and the tumor volume was measured at 7, 14, and 21 days by caliper and IVIS imaging. As In solid tumors and leukemia, there are many ongoing clinical trials of genetically engineered T cells. 30 While intravenous injection is a conventional route in clinics, the efficiency of trafficking to tumors is an important issue for solid tumors. Although we have not tested directly in the current study, the engineered monocytes, with their trafficking capability, can be intravenously injected to target tumors that can not be reached by local injection. Our approach using the engineered monocytes can hence be readily extended to different types of cancers where CD47 is highly expressed, including acute leukemia, NHL, colorectal, and ovarian cancers. 31 In fact, SIRPα-blocked macrophages primed with tumor targeting antibody have been reported to traffic to solid tumor after intravenous injection. 6 Accordingly, intravenous injection of iSNAP-MC will be further tested to investigate the efficiency of trafficking to tumors and efficacy of tumor repressing in the future.  While nanoparticles can be administered to inhibit CSF1R and SHP2 signaling to enhance the phagocytic ability of macrophages in vivo, the materials of these nanoparticles, for example, phosphatidyl choline, have not been approved by FDA for intratumoral or subcutaneous injections. 37 In contrast, cell-based therapy utilizing immune cells, including macrophages, T cells, and natural killer cells, have been well-established for clinical applications with long term effects. [38][39][40] As such, compared to nanoparticle-based drug delivery systems targeting macrophages, genetically engineered cells, including macrophages, are more biocompatible with clinical practices and have longer-term effect.

| iSNAP-MC-derived macrophages exhibit M1 phenotype
Genetics can be further designed to enhance the tumor infiltration and antitumor cytokine release of these engineered cells.
Therefore, genetically engineered macrophages/monocytes should have tremendous potentials to be translational toward clinical medicine in the future.
Antibody-dependent cellular phagocytosis of macrophages has been reported to be involved in cancer immunity. 41 In this study, we have demonstrated that iSNAP-MC treatment can inhibit tumor progression of human B-lymphoma, it would be warranted to conduct further investigations to examine whether iSNAP-MC treatment can inhibit progression in other kinds of tumors which do not respond to antibody drugs and do not have specific markers to target. In fact, the engineering of macrophages is a rapidly advancing field for cancer immunotherapy. 42 In clinical trials, blockage of CD47-SIRPα recognition with antibody drugs has also been reported and is under ongoing testing. 43 In clinical settings, CAR T-cell therapy with T-cytotoxic cells expressing chimeric antigen receptors has shown efficacy in cancer treatment. 44 In this study, we established engineered human monocytes via overexpression of engineered-SIRPα to rewire the "Don't eat me" signal. As such, our findings pave a solid ground for applying reengineered monocytes or macrophages toward cancer immunotherapy. Hongquan Xu: Formal analysis (equal); methodology (equal). Shu Chien: Conceptualization (equal); writingreview and editing (equal).

CONFLICT OF INTERESTS
Yingxiao Wang is a scientific cofounder of Cell E&G and Acoustic Cell Therapy, Inc. Other authors declare no conflict of interests.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.