Differential in vivo biodistribution of 131 I-labeled exosomes from diverse cellular origins and its implication in the theranostic application

Exosomes are critical mediators of intercellular crosstalk and regulator of cellular/tumor microenvironment. Exosomes have great prospects for clinical application as theranostic and prognostic probe. Nevertheless, the advancement of the exosomes research has been thwarted by limited knowledge elucidating the most efficient isolation method and their in vivo trafficking. Here we have showed that combination of two size-based methods using 0.20 µm syringe filter and 100k centrifuge membrane filter followed by ultracentrifugation method yields a greater number of uniform exosomes. We also demonstrated the visual representation and quantification of differential in vivo distribution of radioisotope 131I-labelled exosomes from diverse cellular origins, e.g., tumor cells with or without treatments (HET0016 and GW2580), myeloid-derived suppressor cells and endothelial progenitor cells. We also determined that the distribution was dependent on the protein/cytokine contents of the exosomes. The applied in vivo imaging modalities can be utilized to monitor disease progression, metastasis, and exosome-based targeted therapy. Abbreviations bFGF basic fibroblast growth factor CSF1R colony stimulating factor 1 receptor CT computed tomography CTLA4 cytotoxic T-lymphocyte-associated protein 4 EGF epidermal growth factor EMT epithelial to mesenchymal transition EVs extracellular vesicles EPCs endothelial progenitor cells FasL Fas ligand G-CSF granulocyte-colony stimulating factor GM-CSF granulocyte-macrophage colony-stimulating factor HGF hepatocyte growth factor HSP heat shock protein ICAM-1 intercellular adhesion molecule 1 IFN-gamma interferon gamma IL – 1beta interleukin-1 beta IL – 1ra interleukin-1 receptor antagonist IL – 2 interleukin-2 IL – 4 interleukin-4 IL – 6 interleukin-6 IL – 7 interleukin-7 IL – 10 interleukin-10 IL – 12 interleukin-12 IL – 13 interleukin-13 IL – 17 interleukin-17 KC keratinocyte-derived chemokine LIX lipopolysaccharide-induced CXC chemokine M-CSF macrophage colony-stimulating factor MCP-1 monocyte chemoattractant protein 1 MDC macrophage-derived chemokine MDSCs myeloid derived suppressor cells MFP mammary fat pad MIP-1α macrophage-inflammatory protein-1alpha MMP-2 matrix metalloproteinase-2 MRI magnetic resonance imaging NIS sodium iodide symporter NTA nanoparticle tracking analysis PET positron emission tomography PF-4 platelet factor 4 RANTES regulated on activation, normal T cell expressed and secreted ROIs region of interest SDF-1α stromal cell-derived factor-1 SEM standard error of the mean SPECT single-photon emission computed tomography SCF stem cell factor TAMs tumor-associated macrophages TEM transmission electron microscopy TIMP 2 tissue inhibitors of metalloproteinases 2 TLPC thin layer paper chromatography TME tumor microenvironment TNF-α tumor necrosis factor-α TSLP thymic stromal lymphopoietin UC ultracentrifugation VEGF-A vascular endothelial growth factor A VEGFR2 vascular endothelial growth factor receptor 2.


Background
Exosomes are highly heterogeneous [1] endosomal origin lipid bilayered membranous vesicles (30-150 nm) produced by all cell types [2] and released into the biological fluids or cell culture medium. The initial concept of exosomes as "garbage bag" has been changed drastically as now exosomes are thought to be an integral part of intercellular communication. At any given time point, exosomes can contain all known bioactive constituents of a cell, including proteins, lipids and nucleic acid [3][4][5][6][7]. Besides their physiological functions such as maintenance of cellular-stemness, immunity [8,9], tissue homeostasis, protein clearance [10], signaling [11,12]; exosomes contribute to the pathophysiology of several diseases.
The tumor cell-derived exosomes (TDEs) are capable of altering the tumor microenvironment (TME) by stimulating the secretion of growth factors and cytokines from TME associated cells. Thus TDEs play an imperative role in epithelial to mesenchymal transition (EMT) [3], immune escape [13], tumorigenesis, tumor growth, angiogenesis, invasion, cancer stemness, tumor drug resistance [14]. TDEs arbitrate tumorassociated immune suppression [15] and trigger pre-metastatic niche formation, acclimated to cancer-cell seeding prior the arrival of the first cancer cells [16]. Exosomes have immense prospects for clinical application as a diagnostic marker, monitoring treatment response and disease progression. Recently, exosome-based therapeutic delivery system is drawing tremendous attention owing to the distinctions between exosomes and synthetic nano-carriers. Nevertheless, few unsettled queries still affect the possible application of exosomes e.g., most efficient and reproducible approach for large-scale production of quality exosomes in a shorter time, the loading efficiency of therapeutic agents and precise detection of exosome biodistribution.
Despite the intense research for understanding the biological and pathophysiological functions of exosomes, only few studies have scrutinized exosome biodistribution. A breakthrough to investigate in vivo distribution and track the exosomes is immensely desired for safe and effective clinical application.
Yet, to track their whereabouts, only few effective methods are reported that mostly adopted fluorescent and bioluminescence imaging either labeling them with lipophilic membrane dye or manipulating them to exhibit a membrane reporter. Contrarily, the utilization of nuclear medicine imaging techniques, such as single-photon emission computed tomography (SPECT) or positron emission tomography (PET) that are non-invasive imaging, can be combined with anatomical imaging such as, computed tomography (CT) or magnetic resonance imaging (MRI) for exosome localization can be barely found in the published articles. These nuclear imaging techniques have indisputable advantages over fluorescent and P a g e 4 o f 1 8 bioluminescence imaging, owing to their excellent sensitivity for the deeper tissues and quantitative measurement potential of the clinical grade labeling radioisotopes ( 99m Tc,131 I,111 In-oxine).
Here, we proposed a simpler and quicker isolation technique by a combination of size based and ultracentrifugation method. We also demonstrated the visual representation and in vivo distribution of exosomes, isolated from tumor cells with or without treatment; myeloid-derived suppressor cells (MDSCs) and endothelial progenitor cells (EPCs) in metastatic breast cancer animal models by SPECT/CT. We also showed that differential biodistribution is related to the protein/cytokine contents of the collected exosomes.

Methods
Description of cell lines, nanoparticle tracking analysis, flow cytometry, isolation of MDSCs and EPCs, protein quantification, thin layer paper chromatography, quantitative analysis of radioactivity, ex vivo gamma activity and statistical analysis are described in Supplementary Material.

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 hours and dark for 12 hours. Food and water were offered ad libitum. All efforts were made to ameliorate the suffering of animals. CO2 with secondary method was used to euthanize animals for tissue collection.

Exosome isolation
Exosomes were isolated from the culture supernatant of 4T1 and AT3 tumor cell lines. Briefly, 5×10 6 tumor cells were plated in 175cm 2 flasks and grown overnight with 10% FBS complete media in normoxia (20% oxygen). The media was removed and replenished by exosome free complete media.
Exosomes were depleted from the complete media by ultracentrifugation for 70 minutes at 100,000x g using a Ultracentrifuge (Beckman Coulter) and SW28 swinging-bucket rotor. Then the cells were treated with control (DMSO), colony stimulating factor-1 receptor (CSF1R) antagonist (GW2580, 1µM) and 20-HETE synthase inhibitor (HET0016, 100µM) in hypoxia (1% oxygen) for 48 hours. The cell culture supernatant was centrifuged at 1500 rpm for 10 minutes to get rid of cell debris. We employed five P a g e 5 o f 1 8 different methods as follows -1) ultracentrifugation only by initial step with 10,000x g for 30 minutes followed by two steps with 100,000x g for 70 minutes each, 2) size-based method by passing through 0.20 µm syringe filter (Corning, USA) followed by centrifugation with 100k membrane tube (Pall Corporation, USA) at 3900 rpm for 30 minutes, 3) Combination of two steps of size-based method by passing through 0.20 µm syringe filter and centrifugation with 100k membrane tube at 3900 rpm for 30 minutes followed by single step of ultracentrifugation at 100,000x g for 70 minutes, 4) combination of one step size-based method by passing through 0.20 µm and single ultracentrifugation at 100,000x g for 70 minutes and 5) commercially available density gradient separation by total exosome isolation reagent (Invitrogen™, USA). The reagent was added to the culture supernatant sample and incubated overnight at 4° C. The precipitated exosomes were recovered by centrifugation at 10,000x g for 60 minutes.

Tumor model
Both 4T1 and AT3 cells expressing the luciferase gene were orthotopically implanted in syngeneic BALB/c and C57BL/J6 mice, respectively (Jackson Laboratory, USA). All the mice were between 5-6 weeks of age and weighing 18-20g. Animals were anesthetized using a mixture of Xylazine (20mg/Kg) and Ketamine (100 mg/Kg). Either 50,000 4T1 cells or 100,000 AT3 cells in 50µL matrigel (Corning, USA) were injected into the right mammary fat pad.

Radiolabeling of exosomes using Iodine-131 ( 131 I)
Isolated TDEs were labeled by Pierce™ Iodination Beads (Thermo Scientific™). In short, 4-5 iodination beads were cleaned with sterile normal saline and allowed to air dry. The beads were then added directly to 5 mCi of 131 I solution (Cardinal Health, Inc.) and then incubated at room temperature. After 5 minutes, exosomes resuspended in PBS were added to the reaction tube and incubated at room temperature for 30 minutes. To stop the iodination reaction, the beads were taken out from the reaction tube. To get rid of free 131 I, the labeled exosomes were washed and centrifuged with extra PBS using a 100k membrane tube at 3900 rpm for 10 minutes.

In vivo SPECT/CT imaging
After the intravenous injection of 350±50 µCi of 131 I-labeled exosomes in 100µL into the tail vein, all animals underwent SPECT-CT scanning. During the whole procedure, the animals were anesthetized and maintained using a combination of 1.5% isoflurane and 1 L/min medical oxygen flow and their body was immobilized in an imaging chamber to restrain movements. Whole body CT followed by SPECT-imaging was acquired by a nanoScan 4-headed micro-SPECT-CT scanner (Mediso, USA). The image acquisitions P a g e 6 o f 1 8 were commenced 3 hours after the injection of 131 I-labeled exosomes. The reconstructed image size was 205×205×205µm.

Protein array
Exosomal proteins were evaluated for the expression profiles of 44 factors in duplicate by mouse cytokine antibody array (RayBiotech, Inc.). 500μg of protein sample was loaded to the membrane and the chemiluminescent reaction was detected by using LAS-3000 imaging machine (Fuji Film, Japan). All signals emitted from the membrane were normalized to the average of 6 positive control spots of the corresponding membrane using ImageJ software.

Results
Optimizing the exosome isolation method.
To optimize exosome isolation method, we collected exosomes from 4T1 cell culture supernatant using five different techniques-1) ultracentrifugation (UC) only, 2) size based only (0.20 µm and 100k membrane), 3) combination of size based (0.20 µm and 100k membrane) and UC, 4) combination of single size based (0.20 µm) and UC and 5) density gradient separation. Among the different methods employed, density gradient separation technique (#5) yielded most concentrated exosomes with 1.1x10 11 particles/mL, while (#1) UC alone and its (#4) combination with size based (0.20 µm) yielded the lowest concentration of 1.8x10 10 particles/mL and 1.7x10 10 particles/mL, respectively ( Figure 1A). Combination of size based (0.20 µm and 100k) followed by UC (#3) yielded at a concentration of 4.2x10 10 particles/mL. However, after the isolation process, there was visible sedimentation of co-isolated impurities and polymer or reagent along with the exosomes isolated by density gradient separation. Mean diameter of exosomes isolated by (#1) UC alone was 149.7±64.3 nm, (#3) combination of size based (0.20 µm and 100k membrane) and UC was 126.7±52.9 nm and (#5) density gradient method was 135.1±70.8 nm ( Figure 1B). Size distribution curve of density gradient separation showed a wider base with a thick tail extending towards smaller size ( Figure 1C). We also examined the common exosomal markers, CD9 and CD63, in all the samples by flow cytometry (Figure 1D). There was no significant difference between density gradient separation and the combination of size based and UC technique.
Transmission electron microscopy (TEM) images for the exosomes isolated by a combination of size based and UC method showed normal morphology of exosomes without any distortion ( Figure 1E). by SPECT/CT, and the reconstructed images were analyzed by ImageJ. We also injected free 131 I in tumor-bearing mice (free I-131) to determine the uptake of free 131 I to the tumors. We observed an ample amount of radioactive intensity after 3 hrs at the primary tumor site and metastatic site (lung) in the animals that received 131 I-labeled TDEs ( Figure 2D). There was almost no radioactivity in the tumor area, and negligible radioactivity in the lungs of the group injected with free 131 I. The group of animals without any tumor and injected with radiolabeled TDEs showed almost no radioactivity in the mammary fat pad but plenty of radioactivity in the lungs (Figure 2E and 2F). We observed notable radioactivity in thyroid and stomach. Radioactivity was also high in the bladder for the renal clearance.

Distribution of MDSCs, EPCs and HEK293-derived exosomes in primary and metastatic site.
Next, we wanted to see the biodistribution of exosomes derived from other cell types that play crucial role in tumor progression and metastasis. As a non-cancerous, non-specific cell line, we used human embryonic kidney 293 (HEK293) cells. MDSCs were collected from the spleen of tumor-bearing mice with more than 99% purity, and EPCs were isolated from the bone marrow of normal mice with more than 85% purity ( Figure 3A). TDEs (tumor exo) were collected from the cell culture supernatant. Mean diameter of isolated exosomes from HEK293 cells (HEK293 exo), MDSCs (MDSC exo) and EPCs (EPC exo) was 97.6 nm, 131.1 nm, and 140.1 nm respectively ( Figure 3B and 3C). Three hours after injecting the 131 I-labeled exosomes intravenously in tumor-bearing animals, exosomes from all groups accumulated in the primary breast tumor and metastatic site in the lungs except the HEK293 exo ( Figure 3D).
Interestingly, exosomes from the EPCs were abundantly located in the primary tumor site, and ample P a g e 8 o f 1 8 amount of exosomes from the MDSCs were visualized more in the metastatic site-lungs than any other groups ( Figure 3E and 3F).

Ex-vivo gamma activity measurement of a different organ.
After the final scan, we euthanized the animals, measured the weight and emitted gamma activity of individual harvested organs. While most of the organs as a whole showed negligible radioactivity, only the tumor and the liver retained a significant load of radioactivity in all the groups (Supplemental Figure   S1A). Interestingly, lungs from the animals injected with MDSC-exo showed considerably higher radioactivity than the other groups. For the EPC-exo, the primary tumor showed notably higher gamma activity than the other organs. We also calculated the radioactivity per milligram of the weight of an individual organ that demonstrated similar changes of radioactivity as of the whole organ. The non-tumor bearing group, injected with 131 I labeled tumor exosomes showed gamma activity mostly in the liver (Supplemental Figure S1B).

Distribution of exosomes from treated tumor cells.
To explore further, we wanted to investigate whether exosomes collected from tumor cells treated with drugs that decrease cancer growth and metastasis, would distribute differently in vivo. Previously our group showed that HET0016 and GW2580 treatment decreases lung metastasis of breast tumor-bearing mice [18,19] and HET0016 decreases neovascularization, tumor growth with increased survival in glioblastoma model [18,19]. GW2580, a selective small molecule kinase inhibitor of CSF1R and HET0016 is a highly selective inhibitor of CYP4A in the arachidonic acid pathway, that have been shown to decrease recruitment of tumor-infiltrating myeloid cells and decreased tumor-associated macrophages (TAMs) polarization towards M2 macrophages [20]. There was no significant differences in exosome size and marker (CD9 and CD63) after GW2580 and HET0016 treatment of cells ( Figure 5A, 5B and 5C).
After labeling with 131 I, the exosomes were injected in tumor-bearing mice and in vivo imaging was acquired by SPECT/CT after 3 hours, 24 hours and 48 hours. There was increased radioactivity at the primary tumor site in all the groups ( Figure 5D). Though there was an increased localization of the HET exo in the primary tumor, it was not statistically significant ( Figure 5E). However, there was an ascertainable decline of radioactivity in the lungs of GW exo, and HET exo groups compared to that of control exo groups ( Figure 5F).

Distribution in other organs and clearance
In addition to the tumor and lungs, we also measured the radioactivity in the other organs and clearance over time (Supplemental Figure S2). After 3 hours, the highest percentage of radioactivity was observed in the urinary bladder area that was subsequently cleared off. Considerable amount of radioactivity was noticed in the stomach, which was scarcely detectable in subsequent scanning. Among the organs, liver, lungs, and tumor retained the perceptible amount of radioactivity over the time. Furthermore, thyroid gland in all the groups showed no substantial change of radioactivity in all three scans (3, 24 and 48 hrs), suggesting no dissociation of the 131 I from the exosomes that may eventually increase the free 131 I uptake by the thyroid.

Cytokine array of tumor-derived exosomal protein content after treatment.
Finally, because of the differential biodistribution of exosomes isolated from tumor cells with or without treatment, we wanted to see if the treatment of 4T1 tumor cells with GW2580 and HET0016 affects the protein contents of exosomes. We extracted the exosomal proteins from the 4T1 cells without any treatment (control), GW2580 treated cells, and HET0016 treated cells. We performed cytokine array analysis to evaluate the changes of 44 cancer-related factors for these three samples. Among the angiogenesis factors, VEGFR2, ICAM-1, bFGF were significantly reduced in exosomes from the treated P a g e 1 0 o f 1 8 cells than the control cells (Figure 6). From the chemotactic factors, level of KC and macrophage-derived chemokine (MDC) was significantly downregulated in GW2580 exo and HET0016 exo respectively.

Discussion
To expedite the research and clinical application of the exosomes, it is fundamental that exosomes are explicitly and more effectively isolated from a wide spectrum of impurities e.g., virus, proteins, apoptotic bodies. Until now, mainly five isolation techniques have been developed by exploiting a particular trait, such as their size, density and surface markers [21]. Here, we compared five different isolation techniques and implemented a modified reproducible protocol of isolating quality exosomes more efficiently for downstream experiments in a shorter time. UC only method takes approximately 4-5 hrs to isolate exosomes from the culture supernatant, and density gradient-based technique needs long incubation (overnight) time. Although size-based methods (alone or in combination) need shortest time, it is not possible to pellet down the exosomes from the media and subsequent wash. Our optimized method by a combination of size based using 0.20 µm syringe filter followed by 100k centrifuge membrane filter and ultracentrifugation method requires less than 2 hrs, and with the NTA we demonstrated that it yields greater amount of exosomes with uniform size. Also, exosomes isolated by our optimized method showed a comparable percentage of common exosome markers, CD9 and CD63. Using 0.20 µm syringe filter allows passing of vesicle size <200 nm, ensuring the exclusion of other larger extracellular vesicle types.
Following 100k membrane centrifugation makes sure of getting rid of most of the protein impurities that could be co-isolated with the exosomes by other methods. Although, density gradient separation techniques yielded more exosomes, the size distribution curve was not uniform, and there were visible impurities along with isolated exosomes. We did not compare the immune-affinity capture based isolation method as it separates only the specific subgroup of exosomes that possesses the antigen of interest.
A few articles have reported the tumor targeting and metastatic tumor behavior of exogenously administered exosomes. While most of them adopted either fluorescence imaging [22][23][24][25][26]  Noteworthy to mention that, there was almost no accumulation of TDEs in normal MFP, but they accumulated even in the normal lungs, implying the propensity of the breast TDEs for the future metastatic site. Thyroid and stomach showed significant radioactivity due to the presence of sodiumiodide symporter (NIS), which is an intrinsic plasma membrane glycoprotein, actively arbitrates iodide transport into the thyroid follicular cells [29] and several extra-thyroidal tissues, e.g., salivary glands, gastric mucosa and lactating mammary glands [29,30]. However, the NIS expression level is lower in extra-thyroidal tissues than in thyroid tissues, and longstanding retention of iodide does not occur in these tissues [31]. That is why initial high radioactivity in the stomach was almost undetectable in subsequent scanning.
MDSCs and EPCs are critical components of TME, have differential functions in respect to tumor growth On the other hand, we noticed a maximal load of accumulation of EPCs derived exosomes in the primary tumor than any other groups, which could be due to their neovascularization effects in the TME.  In summary, we favorably demonstrated a simple, rapid, high yielding exosome isolation technique utilizing a combination of size based and ultracentrifugation method. As per our knowledge, we are the first group to demonstrate the exact visualization and quantification of radioisotope-labeled exosomes from different cell types with or without treatment, in the primary solid tumors and metastatic area, which are dependent on the protein/cytokine contents of the exosomes. Our imaging technique and quantification of exosomes could be applied for potential metastatic site prediction, monitoring tumor progression and targeting efficacy of exosome-based therapy, thus unlocking a theranostic potential for these exosomes. .       Quantitative data are expressed in mean ± SEM. *P<.05, **P<.01, ***P<.001, ****P<.0001. n = 2.