Generation of Novel Diagnostic and Therapeutic Exosomes to Detect and Deplete Protumorigenic M2 Macrophages

Given their protumorigenic function and prevalence in most malignant tumors with lower survival; early detection, and intervention of CD206‐positive M2 macrophages may boost the clinical outcome. To determine in vivo distribution of M2 macrophages, 111In‐oxine‐based radiolabeling of the targeted exosomes is adopted. When these radiolabeled targeted exosomes are injected into breast tumor‐bearing mice, exosomes accumulate at the periphery of the primary tumor, metastatic foci in the lungs, spleen, and liver. Ex vivo quantification of radioactivity also shows similar distribution. Injecting DiI dye‐labeled exosomes into the same mice shows adherence of exosomes to the CD206‐positive M2 macrophages on ex vivo fluorescent microscopy imaging. In addition, these engineered exosomes are utilized to carry the Fc portion of IgG2b with the intention of augmenting antibody‐dependent cell‐mediated cytotoxicity. It is demonstrated that M2 macrophage targeting therapeutic exosomes deplete M2 macrophages both in vitro and in vivo, and reduce tumor burden, increasing survival in a metastatic breast cancer model.


Introduction
Exosomes have emerged as potential tools for a drug delivery system that can target specific tissues or cells. Recently, the DOI: 10.1002/adtp.201900209 therapeutic application of exosomes has shown promising results as novel therapeutic vehicles in cancer immunotherapy and suicide therapy, as well as delivery of RNA-interference and drugs. [1][2][3][4][5] Exosomes have clear advantages over synthetic nanoparticles like liposomes as a vehicle because of their improved biocompatibility, low toxicity and immunogenicity, permeability, stability in biological fluids, and ability to accumulate in the tumor with higher specificity. [2,3,[6][7][8][9] Exosomes can be engineered to express targeting peptides or antibodies on their surface for precise targeted therapeutics delivery. [10][11][12][13][14][15][16] Despite the exponential growth of chemotherapeutics and other targeted therapies for the treatment of cancer, there have been few successes for solid tumors. Thus, instead of focusing on the tumor cell alone, treatment strategies have been extended toward other cell types within the tumor microenvironment (TME). Increased infiltration of tumor associated macrophages (TAMs) correlates with tumor stage and poor survival. [17,18] In addition to repolarization of macrophages, therapeutic depletion might be an attractive approach.
CD206-positive M2 macrophages are shown to have a pivotal role in the dissemination of breast cancer cells and prognosis. [19,20] M2 macrophages participate in immune suppression, epithelial to mesenchymal transition, invasion, angiogenesis, tumor progression, and subsequent metastasis foci formation. Investigators have utilized monoclonal antibody (mAb) against CD206 or multimannose analog diagnostic imaging compounds that target the lectin domain of CD206 as imaging agents for detecting M2 macrophages in the TME or draining lymph nodes. [21,22] In recent year, investigators have identified a peptide sequence CSPGAKVRC that binds specifically to CD206+ macrophages in the tumors and sentinel lymph nodes in different tumor models. [22] Generation of exosomes that uniquely bind to the receptor expressed by TAMs will enable the design of rational therapies that specifically target TAMs, ideally leaving normal macrophages unaffected.
Antibody-dependent cell-mediated cytotoxicity (ADCC) is a non-phagocytic mechanism by which most leucocytes (effector cells) can kill antibody-coated target cells in the absence of complement and without major histocompatibility complex (MHC). [23] Targeted therapy utilizing mAbs has instituted immunotherapy as a robust new tool to fight against cancer. As mAb therapy has revolutionized treatment of several diseases, ADCC has become more applicable in a clinical context. Clinical trials have demonstrated that many mAbs perform somewhat by eliciting ADCC. [24] Antibodies serve as a bridge between Fc receptors (FcR) on the effector cell and the target antigen on the cell that is to be killed. There has not been any report of engineered targeted exosomes inducing ADCC. In the proposed model of engineered exosomes along with CD206 binding peptide, we conjugated Fc portion of the mouse IgG2b that could potentially be recognized by FcR on the effector cells and stimulate the ADCC events.

Determination of Specificity of Precision Peptide In Vivo
To assess in vivo targeting potential, rhodamine-labeled precision peptide (red) was injected intravenously (IV) in metastatic syngeneic murine breast cancer (4T1) bearing Balb/c mice. Three h after injection, all animals were euthanized, and lungs, spleen, and tumors were collected for immune-histochemical analysis. Frozen sections from the collected tissues were stained for CD206 (fluorescein, FITC) and counter stained with DAPI. The targeting peptide accumulated in CD206-positive macrophages in tumors, spleen, and lungs (Figure 1a).

Generation of CD206-Positive M2 Macrophage-Specific Exosomes
To confer targeting potentiality, we fused precision peptide for CD206-positive TAMs, to the extra-exosomal N-terminus of murine Lamp2b, a protein found freely in exosomal membranes (Figure 1b). A 6XHis tag in the C-terminus of the protein was added for confirming the expression of the recombinant protein and luciferase was used as a reporter gene. Plasmid encoding the Lamp2b construct was transfected into the HEK293 cells before exosome purification ( Figure 1c). Positively selected cells showed strong luciferase activity in vitro following the addition of luciferin substrate while non-transfected HEK293 cells did not show any activity (Figure 2a). Induction of precision peptide in transfected cells was confirmed by agarose gel electrophoresis showing single band of amplified DNA at the level of 150 bp, corresponding to the targeting peptide ( Figure 2b). 6XHis tag was strongly expressed in engineered exosomes compared to exosomes from non-transfected HEK293 cells and 4T1 tumor cells, based on Western blots with anti-6XHis tag antibody (Figure 2c).
After successfully generating the engineered exosomes, we analyzed their concentration and size distribution by nanoparticle tracking analysis (NTA). There was no significant difference in size distribution between the engineered exosomes compared to those from non-transfected HEK293 cells (Figure 2d). The mean diameter of engineered exosomes was 92.2 ± 4.6 nm and HEK293 cells-derived exosomes was 106.3 ± 14 nm (Figure 2e). Transmission electron microscopic (TEM) images for engineered exosomes showed characteristic round morphology and size without any deformity (Figure 2f).

Targeting Potential of CD206-Positive M2 Macrophage-Specific Exosomes
To assess targeting ability of the engineered exosomes, we differentiated mouse RAW264.7 macrophages toward M2 macrophages by treating them with IL-4 and IL-3 in vitro. Without the treatment (with IL-4 and IL-13) approximately 65% of RAW264.7 cells were positive for CD206 by flow cytometry ( Figure S1, Supporting Information). Following the treatment, almost all the RAW264.7 cells were positive for CD206 by in vitro immunofluorescence staining ( Figure S2, Supporting Information). Then we co-cultured the cells with DiI-labeled (red) engineered exosomes for 4 h followed by immunofluorescence staining for CD206-positive cells (FITC) and DAPI for nuclei. Microscopic images showed that engineered exosomes were attached and internalized by the CD206-positive M2 macrophages (Figure 3a).
Next, we evaluated whether the binding is mediated by CD206. We co-cultured DiI-labeled engineered exosomes with CD206-negative normal mouse embryonic fibroblasts (MEF), and RAW264.7 cells with or without anti-CD206 peptide treatment. After 4 h of incubation, immunofluorescence staining was done for CD206-positive cells (FITC) and DAPI for nuclei. Microscopic images showed that while engineered exosomes were not bound or taken up by the MEF, they were bound to the CD206positive RAW264.7 cells (Figure 3b). Binding of DiI-labeled engineered exosomes was attenuated by treatment with anti-CD206 peptide.
To confirm the targeting efficiency of the engineered exosomes in vivo, we injected same DiI-labeled engineered exosomes in 4T1 tumor-bearing Balb/c mice. After three h of IV injection, mice were euthanized, and tumor, spleen, and lungs were collected for frozen sectioning. Immunofluorescence staining was done for CD206-positive cells (FITC) and DAPI for nuclei ( Figure 3c). Fluorescence microscopic images showed that DiIlabeled engineered exosomes were co-localized with the CD206positive M2 macrophages. Interestingly, the red-colored engineered exosomes spared the white pulp or germinal centers of the splenic follicle that accommodate T-and B-lymphocytes, implying these lymphocytes were not targeted by the engineered exosomes ( Figure 3d).

Detection and Quantification of In Vivo Distribution of CD206-Positive M2 Macrophages Targeting Exosomes
To investigate the validity of engineered exosomes as an imaging probe to determine the distribution of M2 macrophages, we used FDA approved clinically relevant SPECT scanning and la-beling with indium-oxine ( 111 In-oxine) according to our previous study. [25] We used 111 In-oxine-labeled non-engineered control exosomes (HEK293 exo) in metastatic (4T1) mouse breast cancer models, and engineered exosomes (M2-targeting exo) expressing precision peptide treated with either vehicle or clodronate liposome (Clophosome-A) 24 h before the IV administration of 111 In-oxine-labeled exosomes and SPECT studies. Clophosome-A is composed of anionic lipids and depletes more than 90% macrophages in spleen after a single intravenous injection. [26,27] Clophosome-A is not approved for human studies, and it is for experimental use only.
Similar to the previously mentioned 131 I-labeled exosomes, [28] prior to IV injection into mice for biodistribution, we checked the labeling efficiency of 111 In-oxine to the engineered exosomes and serum stability of binding by thin layer paper chromatography (TLPC). While approximately 92% of the free 111 In-oxine alone moved from the spotted point in the bottom to the top half of the TLPC paper (Figure 4a), more than 98% of 111 In-oxine-bound to the engineered exosomes remained at the bottom, indicating that 111 In-oxine was bound to the engineered exosomes with very little dissociation of the 111 In-oxine from the exosomes (Figure 4b). After labeling with 111 In-oxine we also evaluated serum stability of the binding through incubating the labeled exosomes with 20% FBS for 1 h and 24 h. TLPC showed that >92% of 111 In-oxine was still bound to exosomes after 1 and 24 h (Figure 4c).
All animals underwent CT followed by SPECT scanning at 3 h after IV administration of 111 In-oxine-labeled exosomes. The group injected with 111 In-oxine-labeled HEK293 exo did not show any radioactivity or localization of exosomes in tumor, lung, and spleen ( Figure 4d). Significant amount of exosomes was localized in these organs of animals injected with 111 In-oxine-labeled M2targeting exo. Surprisingly, there was an overt accumulation of M2-targeting exo in lymph nodes and bones. As Clophosome-A treatment depleted macrophages, the treated group demonstrated significantly decreased accumulation of M2-targeting exo in tumor, lung, and spleen compared to the untreated group. Additionally, we also created 3D surface plot of lungs and tumors of above-mentioned groups using ImageJ software ( Figure 4e). Consistent with the previous findings, there was almost no radioactivity or exosome accumulation in lungs and tumor of animals injected with HEK293 exo. While accumulation of M2-targeting exo in lungs and tumor was conspicuously high, their localization was considerably attenuated by prior Clophosome-A injection. In the tumor, M2-targeting exo localized only at the M2 macrophage prevalent rim of the tumor.
Activity in different organs including primary and metastatic sites (lungs) was quantified to determine the percent injection dose (%ID). Estimated radioactivity demonstrated significant amount of exosomes were localized in tumor, lungs and spleen of vehicle-treated animals injected with 111 In-oxine-labeled M2targeting exo compared to other two groups ( Figure 4f).
Following the scan, animals were euthanized, and radioactivities of different organs were determined as reported previously. [29,30] Alike in vivo, ex vivo quantification of radioactivity also showed substantially higher radioactivity in lungs, spleen, and tumor of animals injected with 111 In-oxine-labeled M2-targeting exo ( Figure 4g). We observed notable radioactivity in the kidneys and bladder after 3 h of i.v. injection in all the groups, it is due to the fact that exosomes along with radioactive isotopes were excreted through the kidneys into the urine ( Figure  S3, Supporting Information).

Generation of CD206-Positive M2 Macrophage-Targeting Therapeutic Exosomes
Following the confirmation of targeting potential of engineered exosomes for diagnostic purpose, we utilize the exosomes as therapeutic carriers. We conjugated Fc portion of mouse IgG2b next to the targeting precision peptide with a small linker with the purpose of inducing ADCC (Figure 5a,b). Identical to the previous construct, 6XHis tag and luciferase were incorporated as reporter genes.
Positively selected cells showed strong luciferase activity in vitro following addition of luciferin substrate while nontransfected HEK293 cells did not show any activity (Figure 5c).  111 In-oxine to engineered exosomes was validated as shown by a lower percentage of 111 In-oxine (free, dissociated) measured in the top of the paper, compared to the amount remaining in the bottom, which represented the 111 In-oxine-labeled exosomes. c) Serum stability of 111 In-oxine bound engineered exosomes was higher compared with the small amount of free 111 In-oxine disengaged from the bound exosomes. d) In vivo SPECT/CT images (coronal view) after 3 h of intravenous injection showed significant accumulation of M2-targeting exo in tumor, lung, spleen, lymph node, and bones. 111 In-oxine-labeled non-targeting exosomes (HEK293 exo) and CD206-positive M2 macrophage targeting exosomes (M2-targeting exo) were injected into the 4T1 tumorbearing mice. One group was treated with Clophosome to deplete macrophages. Yellow and green arrows denote lymph node and bone metastasis, respectively. e) 3D surface images showing M2-targeting exo are profoundly distributed in both lung and tumor area compared to the group injected with HEK293 exo and pretreated with Clophosome. Yellow arrow indicates the tumor center. f) Quantification of in vivo radioactivity in lungs, spleen, and tumor. g) Ex vivo radioactivity quantification in lungs, spleen, and tumor. Quantitative data are expressed in mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. n = 3.
We confirmed the presence of Fc portion of mouse IgG2b on the surface of the exosomes by flow-cytometry using FITCconjugated anti-mouse IgG2b antibody, that showed ≈52% of engineered therapeutic exosomes express Fc portion of mouse IgG2b ( Figure 5d). We next analyzed concentration and size distribution of the engineered therapeutic exosomes by NTA ( Figure 5e). The mean diameter of engineered exosomes was significantly larger than the non-engineered exosomes (Figure 5f). TEM images for engineered therapeutic exosomes showed distinctive round morphology and size without any distortion (Figure 5g). Flow-cytometric analysis of common exosome markers for the engineered exosomes showed ≈48% positive for CD9 and ≈40% positive for CD63 (Figure 5h).

Induction of Cytotoxicity and Depletion of M2 Macrophages by Engineered Therapeutic Exosomes
To ascertain the capacity of therapeutic exosomes for instigating ADCC, we treated the CFSE-labeled (green) RAW264.7 macrophages with non-therapeutic CD206-positive cell-targeting exosomes (LAMP-206 exo) or CD206-positive cell-targeting therapeutic exosomes (LAMP-206-IgG2b exo), and without any exosome treatment (control) for 48 h in presence of normal mouse splenic mononuclear cells. Fluorescent microscopic analysis showed most of the cells treated with LAMP-206-IgG2b exo were either dead or floating (Figure 6a). Likewise, measured fluorescence intensity of the cells treated with LAMP-206-IgG2b exo was significantly lower compared to the control or LAMP-206 exo treated cells (Figure 6b).
To evaluate whether the engineered therapeutic exosomes can also deplete CD206+ M2 macrophages in vivo, we treated normal Balb/c mice with single, two and three doses of therapeutic exosomes or without treatment (control). We harvested the spleens of the mice for flow-cytometric analysis. Remarkably, we observed a dose-dependent decrease in M2 macrophage population by therapeutic IgG2b exosome treatment compared to control (Figure 6c,d). There was no significant difference in both CD8 and CD4 population between therapeutic exosome treated and untreated group, which indicates engineered therapeutic exosomes do not affect the T cell population (Figure 6e,f).

Treatment with Engineered Therapeutic Exosomes Prevent Tumor Growth and Early Metastasis Increasing Survival
Furthermore, we wanted to determine in vivo distribution of the precision peptide after therapeutic exosome treatment in mouse tumor model to see if the treatment can attenuate distribution of the peptide in M2 macrophage prevalent areas. We implanted tumor cells subcutaneously on the flanks of mice. After 3 weeks of tumor growth we treated one group of mice with engineered therapeutic exosomes for one week (3 doses), and another group without treatment. We conjugated 6-hydrazinopyridine-3carboxylic acid (HYNIC) with the precision peptide and labeled with technetium-99m (99mTc). We injected 99mTc-labeled peptide into both groups of mice and after 3 h CT followed by SPECT images were acquired. Reconstructed images and quantification displayed significant diminution of precision peptide distribution in tumor and spleen of the group treated with therapeutic exosome compared to untreated group (Figure 7a,b).
Finally, we investigated whether depletion of M2 macrophages by therapeutic exosomes can prevent tumor growth and metastasis, and increase survival of tumor-bearing animals. From day 8, after orthotopic implantation of the tumor cells one group of mice was treated with engineered therapeutic exosomes and another group without any treatment (control). Total 6 doses (3 doses per week) of engineered exosomes were injected intravenously for 2 weeks. Tumor growth was monitored by optical imaging every week. We found slower growth of tumor (photon intensities) in engineered therapeutic exosome treated mice compared to the control group (Figure 7c,d). Additionally, control group presented with early metastatic foci after week 4 compared to the treated group, treated group did not show any metastasis even after week 6. Furthermore, survival was prolonged in the group treated with therapeutic exosomes compared to the control group ( Figure 7e). These data validated the therapeutic efficacy of the engineered exosomes in depleting CD206-positive M2 macrophages and subsequently averting tumor growth and metastasis.

Discussion
In recent years, several pioneers have explored the possibility of using exosomes as drug delivery vehicles. Owing to their defined size and natural function, exosomes appear ideal candidates for theranostic nanomedicine application. [23] When compared to the administration of free drugs or therapeutics, exosomes have certain advantages such as improved stability, solubility, and in vivo pharmacokinetics. [24] Exosomes can potentially increase circulation time, [31] preserve drug therapeutic activity, increase drug concentration in the target tissue or cell to augment therapeutic efficacy, [32] while simultaneously reducing exposure of healthy tissues to reduce toxicity. [33] Since they are nanosized and carry cell surface molecules, exosomes can cross various biological barriers, [34] that might not be possible with free drugs or targeting agents. One of the concerning factors for determining in vivo distribution in tumor model was enhanced permeability and retention (EPR) effect by which nanoparticles tend to concentrate in tumor tissue much more than they do in normal tissues. Although, only a fraction (0.7% median) of the total administered nanoparticle dose is usually able to reach a solid tumor, which might give false positive signals of exosome distribution. Surprisingly, we did not observe any retention of radioactivity for free 111 In-oxine, and non-targeted or non-cancerous exosomes (HEK293 exo). This implies that our demonstration of exosome biodistribution and targeted therapy is not an EPR effect, rather the exosomes were directed toward target organs by over-expressed precision peptide on their surface.
Many mechanisms have been implemented to boost the antitumor activities of therapeutic antibodies, including extended half-life, blockade of signaling pathways, activation of apoptosis and effector-cell-mediated cytotoxicity. Here we propose to target Fc gamma-receptor (Fc R) based platform to deplete of M2 macrophages. The direct effector functions that result from Fc R triggering are phagocytosis, ADCC, and induction of inflammation; also, Fc R-mediated processes provide immune-regulation and immunomodulation that augment T-cell immunity and finetune immune responses against antigens. With respect to IgG2b, part of the most potent IgG subclasses can bind specifically into Fc RIII (KD = 1.55 × 10 −6 ) and IV (KD = 5.9 × 10 −8 ) to activate Fc Rs. [35,36] Peptibodies containing myeloid-derived suppressor cells (MDSC)-specific peptide fused with Fc portion of IgG2b was able to deplete MDSCs in vivo and retard tumor growth of a lymphoma mouse model without affecting proinflammatory immune cells types, such as dendritic cells. [37] This plasticity of effector and immune-regulatory functions offers unique opportunities to apply Fc R-based platforms and immunotherapeutic regimens for vaccine delivery and drug targeting against infectious and non-infectious diseases. [38] Investigators have used tumor cells, dendritic cells (DCs), mesenchymal stem cells (MSCs), MDSCs, endothelial progenitor cells (EPCs), neural stem cells (NSCs), and other cell types to generate engineered and non-engineered exosomes for both imaging and therapeutic purpose. [8,10,15,[39][40][41] We have also used tumor cells, MDSCs, EPCs, and NSCs derived exosomes in our previous and ongoing studies. [29,41,42] Tumor cell-derived exosomes carry antigens and elicit immunogenic reaction, therefore, these exosomes have been used in studies for tumor vaccination. [4,5,10] On the other hand, both MSCs and MDSCs derived exosomes have shown to be immune suppressive. [43][44][45] EPC-derived exosomes may enhance neovascularization in the tumors. [46,47] Therefore, using these cells to generate engineered exosomes to carry CD206 targeting peptide may initiate unwanted effects Figure 7. Treatment of 4T1 tumor-bearing animals with therapeutic engineered exosomes prevent tumor growth and metastasis, and improve survival by depleting M2 macrophages. a,b) Reconstructed and co-registered in vivo SPECT/CT images (coronal view) and quantification of subcutaneous syngeneic tumor-bearing animals (on the flank) injected with the 99mTc-labeled precision peptide after 3 h. Group treated with therapeutic exosomes showed lesser level of radioactivity in tumor (yellow arrow) and spleen compared to untreated control group. Quantitative data are expressed in mean ± SEM, *p < 0.05. n = 3. c) Optical images of 4T1 tumor-bearing animals treated with engineered therapeutic exosomes (lower panel) or without treatment (control), showing decreased tumor growth in treated animals compared to control group. Metastatic foci in control group was detected (yellow arrows) as early as fourth week, whereas no metastasis was detected in treated animals after 6 weeks. d) Quantification of optical density of the tumor area also showed decreased tumor growth in treated group compared to control group. Quantitative data are expressed in mean ± SEM. n = 3. e) Kaplan-Meier plot showing prolonged survival of the mice treated with therapeutic engineered exosomes. Log-rank test (Mantel-Cox) was applied to determine the significance of differences among the groups (p = 0.02).
of immune activation, immune suppression, or neovascularization. Moreover, in vitro growth of MSCs, NSCs, and EPCs may be limited due to cell passage number. Ideal cell to generate engineered exosomes should have the following criteria: 1) Nonimmunogenic, 2) unlimited cell passage capacity without changing their characteristics, 3) abundant production of exosomes both in normal and strenuous conditions, 4) cells that can easily be genetically modified. HEK 293 cell is ideal for the production of engineered exosomes. These cells have been extensively used by the biotechnology industry to produce FDA (food and Drug Administration) approved therapeutic proteins and viruses for gene therapies. [48,49] Exosomes derived from these cells show no immune activation or suppression following long-term injections in animal models. [50] We used HEK293 cells to generate our engineered exosomes to carry precision peptide to target CD206+ M2 macrophages.
In conclusion, our study has demonstrated that exosomes targeting M2 macrophages could be utilized effectively to diagnose, monitor, and prevent tumor growth and metastasis for better survival. The study provides novel insights for efficient exosomebased targeting of TME cells.

Experimental Section
Ethics Statement: All the experiments were performed according to the National Institutes of Health (NIH) guidelines and regulations. The Institutional Animal Care and Use Committee (IACUC) of Augusta University (protocol # 2014-0625) approved all the experimental procedures. All animals were kept under regular barrier conditions at room temperature with exposure to light for 12 h and dark for 12 h. Food and water were offered ad libitum. All efforts were made to ameliorate the suffering of animals. CO 2 with a secondary method was used to euthanize animals for tissue collection.
Cell Lines: 4T1, a murine mammary carcinoma cell line from a BALB/cfC3H mouse, was originally obtained from the American Type Tissue Culture Collection (ATCC), and modified by Dr. Hasan Korkaya (Augusta University) to express the luciferase gene reporter. Exosome Isolation: Exosomes were isolated from the culture supernatants of 4T1, HEK293 cells and transfected HEK293 cells. Briefly, 5 × 10 6 cells were plated in 175 cm 2 flasks and grown overnight with 10% FBS complete media in normoxia (20% oxygen). The media was removed and replenished with exosome-free complete media. Exosomes were depleted from the complete media by ultracentrifugation for 70 min at 100 000 × g using an ultracentrifuge (Beckman Coulter) and SW28 swinging-bucket rotor. The cells were then grown for 48 h under normoxic condition. The cell culture supernatant was centrifuged at 700 × g for 15 min to get rid of cell debris. To isolate exosomes, a combination of two steps of size-based method was employed by passing through 0.20 µm syringe filter and centrifugation with 100k membrane tube at 3200 × g for 30 min followed by a single step of ultracentrifugation at 100 000 × g for 70 min (as described in a previous publication by the authors). [28] Nanoparticle Tracking Analysis: Nanoparticle tracking analysis (NTA) was performed using ZetaView, a second-generation particle size instrument from Particle Metrix for individual exosome particle tracking as described previously. [28] This is a high performance integrated instrument equipped with a cell channel, which is integrated into a "slide-in" cassette and a 405-nm laser. Samples were diluted in 1X PBS between 1:100 and 1:2000 and injected in the sample chamber with sterile syringes (BD Discardit II, NJ, USA). All measurements were performed at 23°C and pH 7.4. As measurement mode, used were 11 positions with 2 cycles, and for analysis parameter, maximum pixel 200 and minimum 5 were used. ZetaView 8.02.31 software and Camera 0.703 µm per px were used for capturing and analyzing the data.
Flow Cytometry: The common exosome markers, mouse-specific anti-CD9 FITC, and anti-CD63 APC antibody (Biolegend, San Diego, CA, USA) were used to label exosomes at 4°C for 30 min. Flow cytometry samples were acquired using Accuri C6 flow cytometer (BD Biosciences) with the threshold set at 10 and analyzed by BD Accuri C6 software. For the in vivo flow cytometric analysis, the fresh tissue collected was disseminated into single cells, filtered through a 70 µm cell strainer, and spun at 1200 rpm for 15 min. The pellet was resuspended in 1% BSA/PBS, and incubated with LEAF blocker in 100 µL volume for 15 min on ice to reduce non-specific staining. The single cells were then labeled to detect the macrophage and immune cell populations using fluorescence conjugated antibodies such as CD3, CD4, CD8, CD206, F4/80, and IgG2b. All antibodies were mouse specific and the samples were acquired using Accuri C6 flow cytometer (BD Biosciences).
Tumor Model: 4T1 cells expressing the luciferase gene were orthotopically implanted in syngeneic Balb/c (Jackson Laboratory, Main USA). All the mice were between 5-6 weeks of age and weighing 18-20 g. Animals were anesthetized using a mixture of Xylazine (20 mg kg −1 ) and Ketamine (100 mg kg −1 ) administered intraperitoneally. Hair was removed for the right half of the abdomen by using hair removal ointment, and then abdomen was cleaned by povidone-iodine and alcohol. A small incision was made in the middle of the abdomen, and the skin was separated from the peritoneum using blunt forceps. Separated skin was pulled to the right side to expose the mammary fat pad and 50 000 4T1 cells in 50 µL Matrigel (Corning, NY, USA) were injected. Tumor growth was monitored every week. In vivo, optical images were obtained every week to keep track of primary tumor and metastasis development by injecting 100 µL of luciferin (dose 150 mg kg −1 ) intraperitoneally followed by the acquisition of bioluminescence signal by spectral AmiX optical imaging system (Spectral instruments imaging, Inc., Tucson, AZ). The photon intensity per mm per s was determined by Aura imaging software by Spectral Instruments Imaging, LLC (version 2.2.1.1). The animals were anesthetized using an isoflurane vaporizer chamber (2.5% Iso: 2 ± 3 L min −1 O 2 ) and maintained under anesthesia (2% with oxygen) during the procedure.
Radiolabeling of Exosomes Using Indium-111: Exosomes were labeled with In-111-oxine using an optimized method of labeling. [25] In brief, exosomes (fresh or thawed) were washed with normal saline, reconstituted at 12 billion exosomes mL −1 , incubated with 1mCi of In-111-oxine in normal saline for 30 min at room temperature. Then free from bound In-111 will be separated using Amicon ultra centrifugal filters with a cut-off value of 100 kDa for 30 min at 3200 × g at 20°C. Serum challenge studies were used to determine any dissociation over 24 h, which was determined by TLPC.

Thin Layer Paper Chromatography for Radiolabeling Efficacy and Stability:
3 mm Whatman cellulose chromatography paper was cut into 1 × 8 cm small pieces. The bottom spotted point was made by 5 µL of each sample followed by submerging the bottom part of each piece (fluid level remained below the spotted point) into the eluent consisting of 100% methanol and 2 m sodium acetate solution (1:1). Then the pieces were allowed to remain upright until the eluent reaches the top part. The pieces were cut into the top and bottom halves and were subsequently put in the glass tubes for the measurement of emitted gamma activity by Perkin-Elmer Packard Cobra II Auto-Gamma. Total radioactivity was calculated by combining the activity from top and bottom halves. To determine the percent dissociation of bound 111 In-oxine from exosomes, labeled exosomes were challenged with serum at 37°C up to 24 h. At different time points, free 111 In-oxine, and serum challenged labeled exosomes were tested using TLPC as described above to determine the percent of bound versus free 111 In-oxine.
In Vivo SPECT/CT Imaging of 111 In-Oxine-Labeled Exosomes: After the intravenous injection of 350 ± 50 µCi of 111 In-oxine-labeled exosomes in 100 µL into the tail vein of the mice, whole body SPECT images were acquired using a previously published protocol with a dedicated 4-headed NanoScan, high-sensitivity microSPECT/CT 4R (Mediso, Boston, MA, USA) fitted with high-resolution multipinhole (total 100) collimators. The microSPECT has a wide range of energy capabilities from 20 to 600 keV, with a spatial resolution of 275 µm. The images were obtained using 60 projection images with 60 s per projection, with a medium field of view. Attenuation was corrected using concurrent computed tomographic (CT) images, and then the images were reconstructed with low iteration and low filtered back-projection. The image acquisitions were commenced 3 h after the injection of 111 In-oxine-labeled exosomes. During the whole procedure, the animals were anesthetized and maintained using a combination of 1.5% isoflurane and 1 L min −1 medical oxygen flow and their body was immobilized in an imaging chamber to restrain movements. Throughout the scanning their body temperature was maintained at 37°C and breathing was monitored.
Quantitative Analysis of Radioactivity in Individual Organ: Reconstructed analyze formatted file was used in ImageJ (Wayne Rasband, National Institutes of Health, USA) version 1.51a for both CT and SPECT analysis. The primary tumor, a metastatic site in the lungs and other organs were identified by orthogonal, dorsal, and ventral views from the resliced stack images. Z stack images were created from the CT and SPECT of the individual organ for depth and anatomical accuracy of the organ. Total radioactivity was determined by the sum of the values of the pixels (Raw-IntDen) in the selected region of interest (ROIs) around the organs. The activity in the individual organ was expressed in percent of activity in the whole body (total radioactivity dose).
Ex Vivo Quantification of Gamma Activity of Individual Organ: After the final scan, animals were euthanized, and their organs were harvested and weighed. Emitted gamma radiation from each organ was measured by Perkin-Elmer Packard Cobra II Auto-Gamma after transferring them into the individual glass tube.
Determination of Specificity of Precision Peptide In Vitro and In Vivo: Biotinylated precision peptide (Biotin-CSPGAKVRC) was custom synthesized by a commercial vendor (GeneScript, Piscataway, NJ) using standard peptide synthesis and biotin was attached to the N-terminus. For both in vitro and in vivo studies, biotinylated peptide was labeled with rhodamine using rhodamine-tagged streptavidin utilizing standard protocol for labeling supplied by the vendor (ThermoFisher Scientific). Rhodamine-labeled peptide was used in in vitro studies to determine the specific uptake to CD206 sites on RAW 264.7 cells with or without blocking CD206 receptor using a CD206 blocking peptide (Cat # MBS823969, mybiosource.com). All cells were preincubated with anti-CD44 antibody to block non-specific phagocytosis. All cells were stained for CD206 (fluorescein, FITC) and counter stained with DAPI.
For in vivo specificity, rhodamine labeled peptide (red) was injected intravenously (IV) in metastatic syngeneic murine breast cancer (4T1) bearing Balb/c mice. 3 h after IV administration, all animals were euthanized, and lungs, spleen, and tumors were collected for immunohistochemical analysis. Frozen sections from the collected tissues were stained for CD206 (fluorescein, FITC) and counter stained with DAPI.

Labeling of Conjugated-Precision Peptide with Tc99m
: Hydrazine nicotinamide (HYNIC)-conjugated M2-targeting precision peptide was custom synthesized by a commercial vendor (GeneScript, Piscataway, NJ) using standard peptide synthesis. Then, 250 µg of HYNIC-M2-targeting conjugated peptide was radio labeled with 99m-Tc-pertechnetate in the presence of a solution containing tricine (14.4 mg mL −1 : Acros Organics) and stannous chloride (0.5 mg mL −1 : Acros Organics) in oxygen free condition (air was purged by N 2 ). Following this step, the mixture was centrifuged to remove the unconjugated peptide using 1K centrifugal filter at 3200 × g for 15 min. The amount of radiolabeled peptide was detected using a dose calibrator (CRC-25R: Capintec, Inc.). A dose of approximately 300 µCi of radiolabeled peptide was injected per animal.
Construction for Overexpressing CD206+ M2 Macrophage Targeting Peptide and Fc Portion of Mouse IgG2b on the Exosome Surface: Two different lentiviral vector constructs were made by third party vendor (VectorBuilder, Inc., TX, USA), which were used to generate engineered exosomes in HEK293 cells. CD206+ M2 macrophage targeting peptide and Fc portion of mouse IgG2b along with mouse LAMP2b protein were custom designed and inserted into third-generation lentivirus vector (eBiosciences). QIAquick Gel Extraction Kit (Qiagen, Valencia, CA, USA) and Plasmid Midi Kit (Qiagen, Valencia, CA, USA) were used to extracting the plasmid DNA.

Biogenesis of Engineered Exosomes Expressing Precision Peptide and Fusion Protein:
For the lentiviral production, 1 × 10 6 HEK293TN cells were seeded in a 100 mm culture dish. At 70-75% of confluency, after removing the old media, the cells were supplemented with lentivirus producing plasmids and the targeting cloning plasmid in the presence of Opti-mem and Lipofectamine2000. After 24 h, the culture supernatants containing virus particles were collected, followed by centrifugation and filtration through 0.45 µm PVDF membrane to get rid of the cell debris. For the transfection using lentivector, 500 000 HEK293 cells were seeded in a 100 mm culture dish. At 70-75% of confluency, after removing the old media, the cells were supplemented with transfection cocktail containing regular media, lentivirus, and polybrene. The cells were expanded and subsequently selected with 300 µg mL −1 neomycin for 4 weeks. The transfection of selected cells was confirmed by luciferase activity of the cells following the addition of luciferin. After collecting the supernatant from 6 × 10 6 transfected HEK293 cell cultures incubated for 48 h in a T175 flask with exosomes free media, the supernatant was centrifuged at 700 × g for 15 min to remove cell debris. Then it was filtered through a 0.20 µm PVDF (low protein attachment) membrane and centrifuged using Amicon ultra centrifugal filters with a cut-off value of 100 kDa for 30 min at 3200 × g followed by a final washing step with ultracentrifugation at 100 000 × g for 70 min.
Labeling of Exosomes with DiI: DiI-labeled exosomes were used to demonstrate targeting efficiency of the engineered exosomes both in vitro and in vivo. Following isolation, exosomes were resuspended in 1 mL of DiI working solution (final concentration 5 µm mL −1 in PBS). After 30 min of incubation at 37°C, free DiI was removed by two centrifugation wash steps with PBS using 100k membrane tubes.
Immunofluorescent Staining of Adherent Cell Cultures: 18 × 18-1 glass coverslips were soaked in 100% ethanol for sterilization followed by washing in PBS and then each of them was transferred to each well of 6 wellplates. 300 000 RAW264.7 cells were seeded and incubated overnight. Then the adherent cells were treated with DiI-labeled exosomes (20 µL containing approximately 3 × 10 8 exosomes) and incubated for 4-6 h. After that, media with exosomes was removed and the cells were rinsed twice with PBS. Cells were fixed with 3% paraformaldehyde for 15 min followed by washing with PBS. Cells were covered with blocking solution and incubated for 20-30 min at room temperature. Blocking solution was gently flicked away and appropriate antibody (Alexa 488 anti-mouse CD206 antibody) diluted in blocking solution (1:100) was added. After 2 h of incubation the antibody was removed and the cells were washed with PBS followed by counter staining with DAPI for nuclear stain. After final wash step, the coverslips were transferred for mounting on slides using ProLong Gold Antifade mounting media (Invitrogen).
Determination of Specificity of Engineered Exosomes In Vitro and In Vivo: In vitro studies: Raw264.7 (CD206+ cells) and mouse embryonic fibroblast (MEF, CD206− cells) were used as model cells for in vitro studies of CD206 specificity for engineered exosomes. The anti-CD44 antibody was used before adding the exosomes to block the non-specific uptake of added exosomes by the process of phagocytosis. Both Raw264.7 and MEF cells, grown in small tissue culture petri dish, were treated with anti-CD44 antibody to block phagocytosis, and then these cells were incubated with fluorescent dye DiI labeled engineered and control exosomes collected from HEK293 cells with or without CD206 blocking peptide (Cat # MBS823969, mybiosource.com). CD206 blocking peptide was used to determine the specificity of the engineered exosomes expressing precision CD206 targeting peptide to target CD206 sites. Cells were stained with an anti-CD206 antibody plus FITC tagged secondary antibody. High-resolution fluorescent microscopy images were obtained.
In Vivo Studies Using DiI Labeled Exosomes: For in vivo specificity studies, Balb/c mice bearing 4T1 tumors were used, which were treated with either vehicle or anionic clodronate liposome (Clophosome-A) 24 h before the administration of control or engineered exosomes. Clophosome-A composed of anionic lipids, which deplete more than 90% macrophages in spleen after a single intravenous injection. [26,27] Clophosome-A is not approved for human studies, and it is for experimental use only. Orthotopic breast cancer was developed by injecting 50 000 cells in the fat pad of right lower breast. Untreated animals were used as a positive control, and Clophosome-A treated animal were used as negative control. 24 h after the treatment (5 week old tumor-bearing animals), the mice were used to determine the accumulation of IV administered DiI labeled control and engineered exosomes in the tumors, spleen, liver, and lungs. 3 h after IV administration of exosomes the organs were harvested with proper perfusion. Half of the tumors and organs including lymph nodes were fixed, and sectioned for immunohistochemical studies. Immunohistochemistry was conducted to determine the accumulation of DiI labeled exosomes in CD206+ and CD206− cells.
Immunofluorescent Staining of Frozen Sections: Harvested tissues (tumor, spleen, and Lungs) from the animals were transferred to 30% sucrose and 3% paraformaldehyde solution. 10 µm thick sections were prepared and collected on to prewarmed slides, and allowed to dry at least for a day. Sections were covered with ≈200 µL of blocking solution and were placed in the humidity box for 20-30 min at room temperature. Blocking solution was gently flicked away and appropriate primary antibodies diluted in blocking solution was added. The slides were incubated in humidity box overnight at 4°C. Then the slides were washed twice at least 5 min per wash. Secondary antibodies diluted in blocking solution was added to the sections and incubated at room temperature for two h in humidity box or overnight at 4°C. Then the slides were washed twice at least 5 min per wash followed by counter stain with DAPI for nuclear stain. After final wash step, slides were mounted with ProLong Gold Antifade mounting media (Invitrogen) and with an 18 × 18-1 glass coverslips.
Western Blot: Cells and tissues were processed for protein isolation using Pierce RIPA buffer (Thermo Scientific, USA). Protein concentrations were estimated with Pierce, BCA protein assay kit (Thermo Scientific, USA), and separated by standard Tris/glycine/SDS gel electrophoresis. Membranes were blocked with Odyssey Blocking buffer (LI-COR, Lincoln, NE) for 60 min at room temperature and incubated with primary antibody against 6XHis tag (BioLegend, cat # 362602, 1:500) antibody followed by horseradish peroxidase-conjugated secondary antibody (1:5000). The blot was developed using a Pierce Super Signal West Pico Chemiluminescent substrate kit (Thermo Scientific, USA). Western blot images were acquired by Las-3000 imaging machine (Fuji Film, Japan).
Use of Engineered Exosomes Carrying Fusion Protein as Therapeutic Probes: In vitro studies to assess phagocytosis and cytotoxicity using exosome-Fc-mIgG2b complex: CFSE-stained Raw264.7 was converted to M2 macrophages using IL4 and IL-13 and MEF co-cultured with splenocytes at different ratios. 24 h after co-culture, engineered exosomes carrying Fc-mIgG2b were added to the co-culture, and the studies were repeated at least three times for reproducibility and there was multiple replicate at each time.
Statistical Analysis: Quantitative data were expressed as mean ± standard error of the mean (SEM) unless otherwise stated, and statistical differences between more than two groups were determined by analysis of variance (ANOVA) followed by multiple comparisons using Tukey's multiple comparisons test. Comparison between two samples was performed by Student's t-test. GraphPad Prism version 8.2.1 for Windows (GraphPad Software, Inc., San Diego, CA) was used to perform the statistical analysis. The survival of the animals were analyzed following different treatments. Log-rank test (Mantel-Cox) was applied to determine the significance of differences among the groups. A significance level of 5% ( = 0.05) was used and for a power of 80% (the chance of detecting a significant difference if there is any), the sample size required for the experiments were between 3 and 4 animals per group. The same sample size also was valid for a 90% power calculation. For this reason, the sample size was fixed to n = 3 or n = 4 as mentioned in the methodology. Differences with p < 0.05 were considered significant (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Supporting Information
Supporting Information is available from the Wiley Online Library or from the author.