Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Radionuclides transform chemotherapeutics into phototherapeutics for precise treatment of disseminated cancer

Radionuclides transform chemotherapeutics into phototherapeutics for precise treatment of... ARTICLE DOI: 10.1038/s41467-017-02758-9 OPEN Radionuclides transform chemotherapeutics into phototherapeutics for precise treatment of disseminated cancer 1,5 2 2 1 1 Nalinikanth Kotagiri , Matthew L. Cooper , Michael Rettig , Christopher Egbulefu , Julie Prior 2 1 1 2 1 1 Grace Cui , Partha Karmakar , Mingzhou Zhou , Xiaoxia Yang , Gail Sudlow , Lynne Marsala , 2 2 1 1 2 Chantiya Chanswangphuwana , Lan Lu , LeMoyne Habimana-Griffin , Monica Shokeen , Xinming Xu , 2 2 2 2 1,3,4 Katherine Weilbaecher , Michael Tomasson , Gregory Lanza , John F. DiPersio & Samuel Achilefu Most cancer patients succumb to disseminated disease because conventional systemic therapies lack spatiotemporal control of their toxic effects in vivo, particularly in a compli- cated milieu such as bone marrow where progenitor stem cells reside. Here, we demonstrate the treatment of disseminated cancer by photoactivatable drugs using radiopharmaceuticals. An orthogonal-targeting strategy and a contact-facilitated nanomicelle technology enabled highly selective delivery and co-localization of titanocene and radiolabelled fluorodeox- yglucose in disseminated multiple myeloma cells. Selective ablation of the cancer cells was achieved without significant off-target toxicity to the resident stem cells. Genomic, proteomic and multimodal imaging analyses revealed that the downregulation of CD49d, one of the dimeric protein targets of the nanomicelles, caused therapy resistance in small clusters of cancer cells. Similar treatment of a highly metastatic breast cancer model using human serum albumin-titanocene formulation significantly inhibited cancer growth. This strategy expands the use of phototherapy for treating previously inaccessible metastatic disease. 1 2 Department of Radiology, Washington University School of Medicine, St. Louis, MO 63110, USA. Department of Medicine, Washington University School of Medicine, St. Louis, MO 63110, USA. Department of Biochemistry and Molecular Biophysics, Washington University School of Medicine, St. Louis, MO 63110, USA. Department of Biomedical Engineering, Washington University, St. Louis, MO 63105, USA. Present address: James L Winkle College of Pharmacy, University of Cincinnati, Cincinnati, OH 45267, USA. Nalinikanth Kotagiri and Matthew L. Cooper contributed equally to this work. Correspondence and requests for materials should be addressed to S.A. (email: [email protected]) NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ost deadly cancers are associated with metastatic Results spread , requiring systemic treatment strategies with Contact-facilitated drug delivery via VLA-4-targeted nanomi- Mchemotherapeutic drugs and radiation therapy. Second celles. Targeted delivery of a radionuclide and a drug is necessary generation systemic therapies rely on targeting precise molecular to enable co-localization in the same or adjacent cell for signatures of cancer or invoke immune responses against certain subsequent activation and therapy. Once internalized by a target epitopes specific to cancer. While immensely promising, on-going cell, the radionuclide which essentially behaves as a point source clinical trials indicate that these strategies are often associated of photoelectronic energy, can excite or stimulate photoactive with life-threatening on-target, off-tumour toxicities . For many materials in its vicinity (Fig. 1a). Previous studies have demon- cancers, bone marrow is invariably involved as the point of origin strated the modularity afforded by this approach in treating 3 10 or a distant metastatic niche . The microenvironment of the bone cancer cells or subcutaneous solid tumours or mice using dif- 11,12 marrow is laden with hematopoietic stem cells and progenitors, ferent radionuclides and photosensitizer combinations . making it a highly challenging niche for selective cancer cell Unlike subcutaneous solid tumour xenografts, which do not killing and a difficult terrain for emerging systemic therapeutics. recapitulate the physiopathology of human cancer and are trea- Moreover, in advanced stages of disseminated cancers, patients table by conventional PT, most disseminated tumour models often present with extremely low total lymphocyte counts. As present a different set of challenges because they are embedded in such, they are stratified as severely immunocompromised and the complex and protective microenvironment of the bone 4–7 13 carry the risk of poor prognosis and low overall survival rates . marrow . As a result, it would require more effective targeting These patients are typically unsuitable candidates for and delivery strategies to maximize cell death. We selected mul- existing systemic therapies and emerging immunotherapies. tiple myeloma (MM), an incurable plasma cell dyscrasia that Photodynamic therapy or phototherapy (PT) can offer high predominantly affects the bone marrow, spleen, and bones as the spatiotemporal precision and control of tumour killing through a representative orthotopic disseminated tumour model (Fig. 1a) . combination of direct cytotoxicity, immune-stimulatory, and We also used PyMT-BO1 cancer cell line derived from transgenic antiangiogenic mechanisms . Therefore, PT could serve as an PyMT cancer cells as a highly aggressive metastatic breast cancer effective therapeutic platform and a viable option for model (see below). disseminated cancers, offering an alternative treatment for Titanocene (TC) was used in this study as the photosensitizer the chemotherapy-refractory disease. However, the limited for several reasons, including its UV light excitability and penetration of external light has confined PT to the treatment of responsiveness to low radiance of CR ; biodegradability with surface accessible lesions. In addition, a priori knowledge significantly low cellular footprint post therapy; ease of human of tumour location is a prerequisite for initiating PT, which often translation due to its safety profile in phase 2 clinical trials ; and is indeterminate in the case of disseminated tumours. small size and lipophilicity, allowing integration into lipid-based An alternative approach that delivers light or stimulate vehicles and incorporation into cell membranes post targeting. In light-sensitive drugs within tissues and inside cells in vivo could addition to harvesting CR luminescence, the metal centre can also facilitate the treatment of PT-inaccessible systemic and metastatic interact with radiation particles to further stress cells. However, cancers. Clinically relevant radiopharmaceuticals are reliable two fundamental challenges to TC and similar photoactive drugs, sources of Cerenkov radiation (CR) for cancer imaging .A transvascular delivery to tumour cells and cellular localization, decaying radionuclide could excite materials through, including have remained unaddressed in the context of CRIT. In our the direct interaction of electron and positron emission with previous study, we used transferrin (Tf) to deliver TC to tumour 11 17,18 matter, particularly metals; the emission of ultraviolet-blue light . Tf has only two binding pockets for TC . In the cells emitted by beta (β) particles, known as CR, to generate cytotoxic docking process, the cyclopentadienyl (Cp) ligands of TC can be reactive oxygen species (ROS); chemiluminescent reaction when displaced, leaving the Ti(IV) ion alone as the predominant ambient ionizing radiation excites bulk water; and emission of γ component that binds to the pockets . Because both photo- photons after the annihilation event. For simplicity, we group all activation of Ti(IV) ion and oxidation of Cp ligand to peroxyl these effects as Cerenkov radiation-induced therapy (CRIT). radical contribute to the cytotoxicity of TC, Tf-mediated Therefore, a critical component of the study is to efficiently transport of TC would potentially lower the therapeutic efficacy harvest the diverse potential effects of radionuclides to stimulate of Tf-TC. spatiotemporal cell death in the presence of photosensitizers. ROS-mediated damage to lipid membranes is a primary mode Many drugs possess photoactive properties, but the absence of a of action in PT . Given the short half-lives and small diffusion depth-independent photoelectronic energy source has confined distance of some ROS, the mode of delivery of the drug to the their use as chemotherapeutics, preventing therapy enhancement target cell and its proximity to the cell membrane are important through a complementary phototherapeutic effect. considerations for effective therapy. There is also growing In this study, we hypothesize that CR-mediated conversion of evidence that therapeutic efficacy of PT can be enhanced by light-sensitive drugs to phototherapeutic agents will induce cell selective delivery of hydrophobic photoactive drugs to the plasma death through pathways distinct from the ground state drug (che- membrane compared to receptor-mediated endocytotic uptake . motoxicity) and in a highly selective fashion for the treatment of The contact-facilitated delivery of drugs to the plasma membrane diverse cancer phenotypes. Using multiple myeloma (MM) and by lipid vehicles serves this purpose efficiently. Although metastatic breast cancer models in mice, we demonstrate that liposomal formulations can deliver drugs to cells through this incorporating unmodified and pristine hydrophobic light-sensitive mechanism, conventional liposomes have an average diameter of drugs in tumour-targeted lipid nanomicelles or human serum 100 nm (for unilamellar vesicles) and 0.5–5 μm (for multilamellar albumin (HSA) nanoparticles, selectively deliver the agents in vesicles) , which exceeds the physiologic upper limit of 60 nm disseminated cancer cells. Subsequent in vivo administration of a pore size for transvascular transport of macromolecules to flow radiopharmaceutical for CRIT inhibits the proliferation of across capillary walls of bone marrow . To deliver pristine TC to disseminated multiple myeloma and aggressive metastatic breast the plasma membrane of MM cells, we used nanoscale cancer cells in mice. Our treatment strategy transforms che- unilamellar phospholipid micelles, also known as nanomicelles motherapeutics to spatiotemporally photoactivatable drugs using (NM), as a carrier vehicle. The NM have an average diameter of clinically relevant radiopharmaceuticals and expands the use of ≤15 nm, which is ideal for targeting the bone marrow interstitial phototherapy for treating previously inaccessible metastatic disease. space . The upregulation of a key adhesion molecule, VLA-4 2 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE Bone Stromal cell Stem cell (6) (2) VLA4 (5) (4) Glut Cancer cell (3) Radiopharmaceuticals Targeted (1) nanomicelles Blood vessel bc LLP2A Phospholipid micelle d e Titanocene Fig. 1 Orthogonal cancer targeting strategy using nanomicelles. a Schematic of the process of photoactivation of Titanocene in disseminated cancer cells in the bone marrow microenvironment. The various phases are numbered: 1. Administration of targeted NM-TC; 2. The targeted NM enter the bone marrow from the vasculature and bind to α4β1 receptor on the cancer cells and subsequently deliver the drug to the cell; 3. Administration of radiopharmaceuticals 18 18 ( FDG), which is typically 1.5–2 h after phase (1); 4. FDG enters the cancer cells through the overexpressed Glut transporters on cancer cells; 5. Once the drug and radiopharmaceutical are co-localized in the cancer cells, the former is photoactivated by the latter through CR leading to cell death (6). Notice that since the other vital cells in the bone marrow, such as stem cells and stromal cells, do not express the combination of α4β1 and glut receptors essential for the treatment to work, they would largely remain unaffected causing minimal off-target toxicity. b Schematic of phospholipid NM with VLA-4 homing ligands. c TEM image of micelles alone. Scale bar, 100 nm. Inset: single micelle. Scale bar, 10 nm. d Schematic of phospholipid NM encapsulating TC with VLA-4 homing ligands. e TEM image of micelle incorporated with TC in the membrane. Scale bar, 100 nm. Inset: single NM-TC. Scale bar, 10 nm NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 3 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 Table 1 Size distribution of the nanomicelles Sample Hydrodynamic diameter (nm) Polydispersity index Zeta potential (mV) Nanomicelle + LLP2A 11.9 ± 0.5 0.217 −0.81 TC nanomicelle + LLP2A 14.7 ± 2 0.241 2.12 of TC in rodent blood. The NM in circulation remained intact in vivo, until cleared or destroyed. However, the micelles have a Table 2 Metal (Ti) and TC content in nanomicelles and HSA limited half-life and must reach their target early before elimination. Sample Average Ti content (μg/ Average TC content 20 μL) (mg/mL) TC nanomicelle 0.52 0.192 VLA-4-targeted nanomicelles delivers TC to MM-avid organs. TC-HSA 21.65 5.613 The selectivity of LLP2A to MM cells and the serum stability of the NM in delivering the TC was determined by in vivo biodistribution analysis. Using inductively coupled plasma optical emission spectrometry, we determined the Ti metal content (α β integrin), in MM provides an attractive target for precision 4 1 24,25 ex vivo in organ samples from an orthotopic disseminated MM1. imaging and therapy . Human MM1.S cell line is widely used S/SCID model. We compared the biodistribution of NM-TC, to study MM in rodents. Screening of the MM human cell line, Tf-TC and MKT4, a water soluble analogue of TC that was used MM1.S, using anti-CD49d (α4) and CD29 (β1) antibodies 16,27 in phase 1 and phase 2 clinical trials at 90 min post injection. showed a ≥95% expression level of VLA-4 (Supplementary The choice of 90 min time point is based on rat PK data, which Figure 1). We loaded the NM with LLP2A (Supplementary showed a t of 123 min in rats but the rate of clearance after 90 1/2 Figure 2a), a small molecule peptidomimetic that binds VLA-4 min approached stasis, probably representing the contribution of with an exceptionally high affinity (IC50 = 2 pM) . LLP2A was an intraversation process of drug from tissues to blood (Fig. 2a). synthesized on a solid support, followed by conjugation to Although this time point is expected to be shorter in mice, we phospholipids (DSPE) engrafted with polyethylene glycol (PEG) chose 90 min for the mouse study to ensure that the blood chains to improve circulation in blood (see “Materials and concentration of TC is sufficiently low, to prevent potential Methods' section for details; Supplementary Figure 2b, c). The systemic toxicity, but not too late when the amount in tumour NM were generated as a microfluidized suspension containing tissue is small. In mice administered with NM-TC, the highest Ti LLP2A-PEG-DSPE and TC. Control NM that excluded the concentrations were found in skeletal tissue and spleen, homing ligand LLP2A or TC were also prepared (Fig. 1b, c). which typically house MM cells, with relative values of 115 ± 7 An average size distribution of NM with and without TC was −1 and 52 ± 9.5 μgg , respectively (Fig. 2b). In comparison, the 14.7 ± 2 nm and 11.9 ± 0.5 nm, respectively, with an average uptake of MKT4 was lower in tumour sites, with values of 53 ± 9 polydispersity index of 0.2 (Table 1). −1 −1 and 16 ± 4 μgg , for skeletal tissue and spleen, respectively We successfully loaded 0.19 mg mL of TC in the NM (Fig. 2b). Similarly, the accumulation of TC in these tissues for (Table 2). Based on the full-width half-maximum of the NM size −1 mice treated with Tf-TC was only 27.5 ± 6 and 14 ± 1 μgg , distribution (about 15 nm), the volume of NM, and the net respectively. These results demonstrate the advantage of using concentration of TC per volume of NM using inductively coupled NM to deliver TC to MM target organs. Previous studies have plasma optical emission spectrometry, we determined the average suggested that the cylopentadienyl rings in TC, which assists in number of TC per NM as 3 (range, 2–5). The incorporation of TC stabilizing the Ti(IV) ion in a monomeric form, are lost in MKT4 in the lipid layer was confirmed by electron microscopy (Fig. 1d, 17,18 and Tf-TC . Thus, sequestration of TC in the hydrophobic e). The metallic titanium (Ti) centre in TC rendered the vesicles region of NM may help stabilize the drug and minimize rapid loss electron dense in contrast to the control vesicles without TC. from target tissues. Upon addition into the NM, TC incorporated in the interface between the lipid and the hydrophilic layers, as evidenced by electron microscopy (Fig. 1d, e). Probably, the hydrolysis of TC CRIT inhibits tumour growth in disseminated MM mouse dichloride to the dihydroxyl derivative in aqueous medium model. We used an FDA approved and clinically employed created an amphiphilic structure, favouring the orientation of the radiopharmaceutical, FDG (t = 109.8 min), as a source of 1/2 two cyclopentadienyl and dihydroxyl moieties toward the 28 photoelectronic energy . The radiopharmaceutical, which is 29,30 hydrophobic core and the outer hydrophilic segment, respec- currently the gold standard for clinical imaging of MM , tively. Incorporation of LLP2A did not destabilize the NM and targets metabolically active tumours via the glucose transporter the presence of unnatural amino acids conferred protease (GLUT1) protein. By using an orthogonal-targeting GLUT1 and resistance and high plasma stability on the nanosystem . 18 VLA-4 strategy to, respectively, deliver the FDG and NM-TC to the MM cells, we aimed to minimize the potential saturation or In vivo pharmacokinetics of VLA-4-targeted nanomicelles.In depletion of the targeted receptors. In healthy subjects, FDG vivo pharmacokinetic (PK) profile of the NM-TC was studied in uptake is low in the bone marrow and spleen, but significantly naive rats. A plasma half-life of 123 min was obtained after higher in malignancy, inflammation or after administration of systemic administration (Fig. 2a). We performed the PK in rats hematopoietic growth factors . Using small animal positron instead of mice to obtain sufficient blood sample for serial emission tomography (PET) of MM in mice, we found more than measurements of TC concentration in the same animal. twofold uptake of FDG in bones compared to naive mice Otherwise, the small volume of blood in mice would require us to (Fig. 2c–i). pool samples from different mice, masking inter-specimen The performance of CRIT in a disseminated MM1.S/ SCID variability. Although the PK value in mice are expected to be mouse model was tested. Based on the biodistribution data, shorter than rats, the information allowed us to estimate half-life sequential tail vein injections of NM and then FDG were spaced 4 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE ab 0.05 NM-TC 0.04 MKT4 Tf-TC 0.03 0.02 0.01 0.00 0 0 204060 80 100 120 140 Time (min) ** 20 * cd ef Multiple Fore limb myeloma Naive Multiple Hind limb myeloma Naive Spine Max Fore limb Min g h Multiple myeloma Naive Spine Hind limb Fig. 2 Monitoring nanomicelles biodistribution and spread of multiple myeloma in vivo. a Pharmacokinetics of NM-TC in rats using coupled plasma optical emission spectrometry. Half-life is 123.4 min. b Comparison of biodistribution in mice of targeted NM-TC and pristine TC in vivo showing highest uptake 18 18 and retention in bones and spleen, characteristic of multiple myeloma, 2 h post injection. FDG-PET images showing increased uptake of FDG in mouse forelimbs, spine, and hind limbs of mice with multiple myeloma (c, e, g) compared to naive mice (d, f, h, i). Comparison of standard uptake values (SUV) of FDG in multiple myeloma vs. naive mice in various bones. Values are means ± s.e.m. *P < 0.05, **P < 0.01. n = 5 mice for each of the pharmacokinetics study in rats; and biodistribution study in mice 90 min apart to activate TC in tumours. Treatment was repeated confined at random sites within the major bones, particularly the four times at an interval of 1 week, and the disease progression vertebrae (Fig. 3b, e). These localized cancer cells continued to grow, was monitored weekly by bioluminescence imaging (BLI; Fig. 3a). albeit at a slow rate. The surviving cancer cells were subsequently A week interval was chosen for treatment for several reasons that extracted from the mice and reintroduced into a fresh group of naive include the need to allow the mice to fully recover from the SCID mice to determine response to when treated with CRIT. 18 18 treatment; account for full decay cycle of FDG; consider However, BLI (Fig. 4a) and FDG-PET did not show noticeable logistical reasons such as tail vein recovery; and allow sufficient differences between the treated and untreated groups, suggesting the time for imaging time points between treatment sessions. In the cells were resistant to CRIT. These CRIT-resistant MM1.S (MM1. CRIT-RES control groups consisting of untreated mice or those treated with S ) cells were harvested and analysed for the expression either NM or FDG alone (Fig. 3b, c), we observed an levels of GLUT1, α4and β1 integrins to determine whether uptake exponential increase in the BLI signal over several weeks, of FDG by GLUT1 or α4β1 binding of the NM were compro- demonstrating the systemic progression of the disease and mised. GLUT1 mRNA (Fig. 4b) or β1 cell surface expression indicating the primary involvement of the spleen and skeletal (Fig. 4c, d) analyses did not demonstrate significant difference 18 CRIT-RES tissues. In contrast, mice treated with NM-TC and FDG showed between the parental MM1.S and the MM1.S cells. How- CRIT-RES a conspicuous decrease in the disease progression, suggesting the ever, the MM1.S cells expressed lower cell surface α4than effective targeting and response of MM1.S to CRIT. Survival parental MM1.S cells (Fig. 4e, f). Flow cytometry analysis demon- studies revealed a significant advantage of the CRIT over the strated that LLP2A-Cy5, which selectively binds VLA-4 with high 32 CRIT-RES control groups with 50% surviving up to about 90 days compared affinity , did not internalize in the MM1.S cells compared to about 62 days for the control groups (Fig. 3d). Correlative to the parental MM1.S cells (Fig. 4g, h). These results suggest that 18 CRIT-RES FDG-PET imaging confirmed the lower tumour burden in the MM1.S cells had downregulated the expression of α4 CRIT-treated mice compared to the control groups (Fig. 3e, f). (CD49d), possibly impairing the binding of the LLP2A functiona- The mice were killed after they developed hind limb paralysis lized NM to some MM cells. Unlike in vitro studies where static resulting from spinal cord and spinal vertebral involvement. The incubation of nanoparticles can abrogate specificbinding of treated mice eventually succumbed to cancer due to the remnant receptor-targeted materials, the in vivo dynamics and the relatively MM cells that could not be completely eradicated by CRIT. small number of these resistance cells in the initial tumour popu- lation could have favoured the homing of NM-TC to the VLA-4 positive cells in mice. As a result, CRIT could have preserved a CRIT selects for α4-deficient multiple myeloma cells in vivo. subclone of MM1.S with low α4 that was present at low frequency in Residual cancer cells that escaped treatment appeared focal and the injected cells. Thus, targeting VLA-4-rich cancer cells selects for NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 5 | | | Brain Kidney Liver Muscle Heart Skin Spleen Lung Bone MM mice Naive mice Dose%/g blood Titanium content (µg/g) SUV (tissue/muscle) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 Cell injection CRIT CRIT CRIT CRIT (Days) 1 5 12 19 26 Untreated FDG NM CRIT Week 2 100,000 80,000 60,000 Week 4 40,000 20,000 Week 6 cd Untreated FDG NM CRIT –50 –100 –150 ** –200 0 3 7 13 20 27 34 41 48 40 60 80 100 120 Time (days) Time (days) ef CRIT Untreated ** Max Min FS F S Fig. 3 Response of multiple myeloma to CRIT. a Timeline of treatment. b Bioluminescence imaging of representative multiple myeloma-bearing mice in different treatment groups—untreated, FDG, NM controls and CRIT. All images are dorsal images and on the same scale. The images of control groups appear saturated on week 6 in comparison to CRIT. c Change in bioluminescence intensity as a result of treatment compared to untreated control. The intensity consistently remains lower than untreated controls during the treatment and beyond. d Comparison of survival of different treatment groups showing a twofold increase in survival in treated mice compared to control groups. **P < 0.01. e FDG-PET images of MM mice before and after treatment showing lower tumour burden in the latter. F: frontal view, S: sagittal view. Boxes denote tumour region. f SUV values of the treatment group were lower than untreated controls. **P < 0.01. n = 15 mice for CRIT, n = 10 mice for untreated control and n = 5 mice for NM-TC alone and FDG alone treated mice 6 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | Untreated CRIT Radiance (p/s/cm /sr) % fold change in bioluminescence intensity from control mice Percent survival SUV (tumor/muscle) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE the subset of MM1.S cells with low CD49d expression levels and low CRIT preserves normal hematopoietic stem cells. An important levels of the activated conformation of VLA-4. A potential approach consideration during CRIT is to preserve and sustain the long- to achieving complete eradication of MM cells is to identify com- term viability of the hematopoietic stem cells and progenitor cell plementary biomarkers that allow the delivery of TC-loaded NM to population in the bone marrow. Clonogenic assays of normal all MM1.S cells or through the use of combination therapy that bone marrow progenitor cells extracted from mice treated with more effectively targets and eliminates both CRIT-responsive MM1. CRIT did not reveal a significant change in colony-forming units CRIT-RES Sand MM1.S cells in vivo. (CFU) compared to the control groups (Fig. 4i). Competitive ab ns 1.5 Untreated CRIT 1.0 0.5 0.0 02 468 Time (weeks) cd e f Parental Resistant Parental Resistant 100 100 100 80 80 80 60 60 60 40 40 40 40 20 20 20 28.12 99.92 0 0 0 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 10 10 10 10 10 10 10 10 10 10 10 10 0 10 10 10 10 CD29-APC CD49d-PD594 g h i Parental Resistant 100 100 15,000 ns 80 80 60 60 10,000 40 40 20 20 84.15 6.60 0 0 0 1 2 3 0 1 2 3 0 10 10 10 10 0 10 10 10 10 LLP2A-Cy5 jk CD45.2 CRIT + 60 CD45.1 Engraftment analysis Day 60 CD45.1/2 Day 0 PBS NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 7 | | | Parental Resistant Untreated NM FDG CRIT Untreated CRIT % of max % of max Photons/s % of max CFU/10 BM % CD 45.2 BM Relative Glut1 mRNA expression ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 bone marrow repopulation experiments showed that there was no tumour cells disseminated to most major organs, including the detrimental effect of CRIT on the primitive hematopoietic stem vertebrae and lower limbs. Quantitative analysis of the Ti con- cell compartment . The 2-month hematopoietic reconstitution tents in tissue samples from blood and lower limbs by inductively of mice transplanted with bone marrow from wild-type vs. CRIT- coupled plasma mass spectrophotometry showed the highest treated mice was not significantly different, suggesting that there accumulation of the metal in the limbs at 3 h post injection of was no obvious reduction in engrafting hematopoietic stem cell HSA-TC (Fig. 5d). The concentration of Ti in the lower limbs population after treatment with CRIT (Fig. 4j, k). decreased gradually over time (p < 0.05). At 24 h, the Ti content from the HSA-TC in the limbs was indistinguishable from the background content. We had to subtract the background Ti from HSA-TC nanoparticles deliver drug to metastatic breast cancer. untreated mice to determine the contribution of HSA-TC because We extended the use of CRIT in a metastatic breast cancer model. most mouse feedstock contains Ti products. PyMT-BO1 cell line is a highly aggressive breast cancer cell line derived from transgenic PyMT breast cancer mice . Previous reports have demonstrated that rapidly proliferating tumour cells HSA-TC inhibits growth of metastatic breast cancer in mice. actively internalize albumin for use as a source of nitrogen and We explored the feasibility of using CRIT to inhibit tumour energy, partially accounting for the formulation of drugs in growth in the highly metastatic PyMT-BO1 GFP/Luc breast 34–36 albumin or its nanoparticles for drug delivery to tumours . cancer model in C57B6 mice. Our biodistribution data indicate Thus, formulation of hydrophobic drugs, such as TC in albumin, that the accumulation of TC in the cancer-homing organs is will not only enhance solubilization in an aqueous medium, but highest at about 3 h post injection of HSA-TC. FDG (50–60 µL; also mediate delivery to tumours. We mixed TC (80 mg) in an 800 µCi per mouse) was administered intraperitoneally at 2 h aqueous solution (16 mL) containing 0.5% human serum albumin after the intravenous administration of HSA-TC. We chose (HSA) for 6 h before lyophilizing the entire mixture (see 'Mate- intraperitoneal instead of intravenous route for FDG injection rials and Methods' section). We used a low concentration of HSA to maintain consistency across the experiments because of the in this formulation to prevent potential immunogenic response in difficulty of finding viable tail veins in the same mouse for mice. The lyophilized product was reconstituted in 0.9% saline multiple injections of both HSA-TC and FDG. Comparison of immediately before use. Using inductively coupled plasma mass different routes of FDG administration in mice determined that spectrometry, we determined the concentration of TC in the the SUV of FDG injected intraperitoneally in tumours is opti- −1 reconstituted sample as 5.6 mg mL (Table 2). The correspond- mal at about 1 h post injection, and is similar to intravenous route −1 ing concentration of HSA in the formulation was 8.95 mg mL , 37,38 at closer to this time point . We hypothesized that adminis- as determined by using a protein assay. Dynamic light scattering tering the radionuclide at about 2 h after injection of the HSA-TC (DLS) measurements indicate that the HSA-TC nanoparticles (100 µL of 0.6 mg per 20 g mouse) will achieve maximum accu- were fairly monodispersed with an average size of 12–15 nm mulation of both CRIT effectors in tumours after 3 h. Three (Supplementary Figure 3). About 1.5% of the particles formed treatment cycles of HSA-TC and FDG were administered large aggregate clusters of 100 nm, creating a bimodal distribution 2 days apart starting from day 2 after initiating the metastatic that skewed the z-average diameter (120 nm) and polydispersity disease when the tumours are observable by BLI (Supplementary index (0.278). Electron microscopy size measurement correlated Figure 5). Groups with no treatment, treatment with HSA-TC with the DLS results, showing mostly monodispersed HSA-TC alone and FDG alone served as controls. Whole-body luciferase nanoparticles of 10–15 nm in diameter (Supplementary Figure 4), activity from day 2 to day 9 were analysed for tumour cell pro- along with few aggregates of 30–90 nm. The HSA nanoparticles liferation. Compared to the untreated mice (Fig. 6a), a small allowed us to load high concentration of the hydrophobic TC per decrease in BLI signal was observed in the HSA-TC (Fig. 6b) and volume of aqueous solution for subsequent delivery to metastatic FDG (Fig. 6c)-treated mice compared to the untreated cohort. breast cancer. However, the pattern of tumour growth in all the three control groups was similar. In contrast, the CRIT group showed sig- Biodistribution of HSA-TC in metastatic breast cancer model. nificant tumour stasis, with a few focused cluster of tumour cells Intracardiac injection of PyMT-BO1 cells stably transfected with that did not respond to the therapy (Fig. 6d, e). The slow growth GFP-firefly luciferase in mice-induced bone metastases, especially and focal nature of the CRIT-resistant cells suggests that this to the lower limbs and other major organs (Fig. 5a, b). By day 10, therapeutic method can transform metastatic cancer into a sur- the tumour burden was very high, requiring immediate killing gical disease. Whereas PyMT-BO1 model is an excellent model (Supplementary Figure 5). For the biodistribution study, we for the rapid evaluation of drugs or treatment methods, all the reconstituted the lyophilized HSA-TC in saline and administered animals eventually died or were killed between day 9 and 12 due 100 µL of 0.6 mg per 20 g mouse. Both the non-invasive in vivo to the aggressiveness of this model. The fast death cycle prevents (Fig. 5b) and the ex vivo (Fig. 5c) BLI analysis showed that the longitudinal evaluation of each animal, which is needed to obtain Fig. 4 CRIT selects for CD49d cells in MM model. a Bioluminescence intensity plot showing resistant nature of MM cells extracted from treated cohort CRIT-RES (MM1 ) upon rechallenging with CRIT in fresh mice. b No difference in GLUT1 mRNA expression was observed between parental MM1.S cells and CRIT-RES resistant MM1.S cells as assessed by qRT-PCR. ns not significant. c–h No difference in expression of CD29 was observed between MM1.S cells (c) CRIT-RES or MM1.S cells (d) following treatment with CRIT in vivo. MM1.S stopped responding to CRIT by downregulating expression of VLA-4 subunit CD49d (Resistance = 28.12% CD49d+) (f) relative to parental cells injected into mice at the beginning of the experiment (parental = 99.92% CD49d+) (e), resulting in reduced binding of the VLA-4-targeting ligand LLP2A on resistant cells (h) (LLP2A+ = 6.6%) compared to parental MM1.S (g) (LLP2A+= 84.15%). i No significant difference in colony-forming units of progenitor stem cells was observed between untreated, control and treated mice. j To determine if CRIT reduced engraftment of haematologic cells in vivo, we assessed BM repopulation following CRIT treatment. Bone marrow from treated mice or PBS-treated controls were mixed with congenic B6.CD45.1/2 at a ratio of 1:1 before infusion of 1 × 10 total BM cells into lethally irradiated (TBI 1100 cGy) B6.CD45.1 recipients. k Percentage of cells derived from treated donor BM (CD45.2) were calculated as a percentage of total donor BM (CD45.2 + CD45.1/2). BM from CRIT-treated mice effectively repopulated recipients (n = 5 per group) 8 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | 6 ×10 ×10 ×10 NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE ab c d Liver 200 Lung 400 200 Blood * Lower limbs 300 150 Kidney Spleen 200 100 Lower limb Spine 100 50 136 24 Naive PyMT-B01 –50 Time (h) Naive PyMT-B01 Fig. 5 Dissemination of metastatic breast cancer cells and biodistribution of TC in mice. a In vivo BLI non-tumour-bearing C57BL/6J mice. b In vivo BLI of metastatic breast cancer 10 days post intracardiac injection of PyMT-BO1 GFP/Luc in C57BL/6J mice. c Ex vivo BLI of metastatic tumour burden 10 days post tumour initiation. The left panel are tissues obtained from a mouse and the right panel are tissues obtained from b mouse. d Inductively coupled plasma mass spectrometry analysis of Ti content in blood samples and the lower limbs, where tumour burden is high. The Ti content was background- corrected from untreated mice; *P < 0.05. Studies were performed with n = 5 mice per each group ab c d e 2×10 Untreated HSA-TC 08 FDG 1×10 CRIT 5×10 Untreated HSA-TC FDG CRIT Fig. 6 Representative BLI of PyMT-BO1 GFP/Luc metastatic breast cancer cells in C57B6. a Untreated C57B6 mouse bearing highly metastatic PyMT-BO1 −1 cancer. Accumulation in the lower limbs were predominant. b Mouse treated with 30 mg kg of HSA-TC nanoparticles. c Mouse treated with 800 µCi of 18 18 FDG. d Mouse treated with a combination of HSA-TC and 800 µCi FDG. e Quantification of whole-body luminescence in CRIT-treated mice compared 18 18 to untreated, HSA-TC treated or FDG-treated controls (*P values are 0.038, 0.23 and 0.017 for CRIT, HSA-TC alone and FDG alone, respectively). BLI and data analysis were performed on day 9 after initiation of PyMT-BO1 metastasis in mice. Studies were performed with n = 5 mice per each group reliable survival plots. Future studies will explore different models association of both therapeutic components in vital organs. These and establish the mechanism of therapy resistance in this cancer findings expand the potential use of PT for treating previously cell line. PT-inaccessible metastatic, infectious and cardiovascular diseases. In summary, we have successfully demonstrated the applica- tion of PT for treating disseminated malignancies using VLA-4- Methods targeted NM and HSA-TC nanoparticles, activated by radio- Synthesis and characterization of VLA-4-targeted titanocene micelles. The pharmaceuticals. Integral to this strategy is the availability of a VLA-4 ligand, LLP2A (Supplementary Figure 2a), was prepared on solid support wide range of radionuclides for clinical PET imaging and using standard fluorenylmethyloxycarbonyl (Fmoc) peptide synthesis as reported preclinical Cerenkov luminescence imaging to further monitor previously . Starting with Rink Amide resin, serial Fmoc deprotection cycles were 39,40 and guide treatment response . We demonstrated a strategy to achieved with 20% piperidine in dimethylformamide (DMF) and coupling of amino acids was performed with hydroxy-benzotriazole (HOBt) and 1,3-diiso- rescue abandoned light-sensitive drugs with poor therapeutic propylcarbodiimide (DIC) in DMF at 25 °C. The crude product was cleaved from outcomes such as TC and some FDA-approved drugs with the resin with a mixture of 95% trifluoroacetic acid (TFA): 2.5% water: 2.5% inherent photoactivity into precision phototherapeutics. In triisopropylsilane and precipitated with cold diethyl ether. The product was pur- addition, clinical biochemistry parameters and histopathologic ified by RP-HPLC and characterized by analytical HPLC and ES-MS: calculated mass for LLP2A is 1084.27 Da; observed mass is 1085 Da (Supplementary Fig- assessment of vital organs in the treated and untreated controls ure 2b). To incorporate LLP2A into nanomicelles, LLP2A was dissolved in ethanol were similar (Supplementary Figure 6 and Supplementary and mixed with 2-iminothiolane in methanol. After reacting for 2 h at 25 °C, Figure 7). The brain, heart, liver and kidneys were of particular polyethylene glycol -phosphatidylethanolamine (PEG-PE) was added to the interest because FDG naturally accumulates in these organs mixture and incubated for another 2 h. The product was purified using 3000 Da MWCO dialysis tubing to dialyse off the free LLP2A, and lyophilized to give a because of their high glucose utilization and elimination white solid (LLP2A-PEG-PE). The product was characterized by analytical HPLC pathways. The absence of off-target toxicity to normal hemato- and ES-MS: average calculated mass for LLP2A-PEG-PE is about 4100 Da; poietic stem cells may favour the translation of this approach in observed mass is about 4065 Da (Supplementary Figure 2c). The phospholipid/ the clinic as either a standalone therapy or as a combination with polysorbate 80 micelles were prepared as a microfluidized suspension comprising other therapies, including chemotherapy, where the suppression 20% polysorbate Tween 80 (v/v), a 2.0% (w/v) surfactant commixture and 1.7% (w/ v) glycerine in filtered MilliQ Nanopure water. The surfactant co-mixture were 0 or of the bone marrow and the risk of pancytopenia may not be 2 mole% TC and 0 or 0.15 mole% of LLP2A-PEG-PE, with the remainder as anticipated to be greater than those patients receiving che- phosphatidylcholine (>98% purity, NOF America). The surfactant components motherapy without CRIT. Our results suggest that the sequential were dried from organic solvent into a film, resuspended in nanopure water, and administration of the NM-TC and radionuclide minimizes the combined with polysorbate 80 and glycerine mixtures, followed by sonication at 4 ° NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 9 | | | Untreated HSA-TC FDG HSA-TC+FDG Radiance (p/s/cm /sr) Radiance (p/s/cm /sr) Radiance (p/s/cm /sr) BLI photon flux (photons/s) Titanium content (µg/g) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 C for 3 min and then microfluidized (LV1, Microfluidics, Inc) at 20,000 psi for five and snared with two ligatures, one proximal one distal, for introducing the passes. The micelles were filtered with a 0.2 µm filter into sterile serum vials, catheter. The wound is wetted periodically with lidocaine (2%)/bupivicaine (2.5%) preserved under inert gas, capped and crimp-sealed before storage at 4 °C for and saline to maintain analgesia. The distal tie is tightened to ligate the artery, and subsequent use. The concentration of TC in TC nanomicelles was determined by the proximal tie is made into a Potter’s knot. A silastic catheter is inserted through inductively coupled plasma optical emission spectrometry (PerkinElmer Optima an arteriotomy and secured in place with the ligatures. The catheter is then flushed 5300). For each measurement, 20 µL reconstituted TC nanomicelle (n = 3) was with heparinized saline. Blood is sampled via a carotid catheter. Blood samples digested with concentrated nitric acid in Mars 6 Microwave Digestion System were ~150 μL and flushed with equivolume of saline. The blood samples were (CEM Corporation). Microwave power was ramped up to 200 °C for 20 min fol- digested by microwave digestion in HNO /H O under pressure and the con- 3 2 2 −1 lowed by a hold for 20 min at the same temperature. All external calibration centration of TC (mg mL ) was determined by inductively coupled plasma optical −1 standards of 0.01, 0.1, 1.0, 10, 50, 100 and 200 µgL were within 10% of expected emission spectrometry and reported as percentage of injected dose per gram of −1 concentration. Relative standard deviation measurements were less than 5% for all tissue (ID g ). standards and the samples. Dynamic light scattering measurements were acquired with a Zeta Plus Zeta Potential Analyzer (Brookhaven Instruments Corporation, Holtsville, NY) Biodistribution of NM-TC in mice. The NM-TC at a dose of 50 μL per mouse equipped with a 633 nm laser. Three measurements were conducted in deionized were administered intravenously in MM1.S (n = 5) tumour-bearing mice. A one- water for each sample with at least 10 runs and each run lasting 10 s. All sizes time dose was administered 30 days post cell injection of MM1.S cells. Tf-TC and reported were based on intensity average. For negative staining and electron MKT4 were prepared as described in the literature . Briefly, MKT4 was prepared microscopy analysis of NM, samples were allowed to absorb onto Formvar/carbon- by adding fivefold molar excess of mannitol (500 mg) to TC (100 mg) and a 19-fold coated copper grids for 10 min. Grids were washed two times for 1 min each in molar excess of sodium chloride (450 mg) in 50 mL water. The solution was mixed dH O and stained with 1% aqueous uranyl acetate (Ted Pella Inc., Redding CA) for in a shaker at 25 °C for 4 h before lyophilizing. For intravenous injections, MKT4 −1 1 min. Excess liquid was gently wicked off and grids were allowed to air dry. was reconstituted in saline at a dose of 0.25 mg kg . To prepare Tf-TC, fivefold Samples were viewed on a JEOL 1200EX transmission electron microscope (JEOL molar excess of TC was added to human apo-Tf and incubated in a shaker for 2 h USA, Peabody, MA) equipped with an AMT 8-megapixel digital camera at 25 °C. A working stock of TC was initially prepared in DMSO due to low (Advanced Microscopy Techniques, Woburn, MA). solubility of TC in water and aqueous buffers. The mixture was then dialysed overnight against DPBS using a 3000 Da molecular weight cutoff (MWCO) Slide- A-Lyzer MINI Dialysis Devices to remove excess TC. MKT4 and Tf-TC were Synthesis and characterization of HSA-TC. TC (80 mg) was added in an aqueous −1 administered intravenously at a dose of 0.25 mg kg (normalized to the TC solution (16 mL) containing 0.5% HSA. The mixture was shaken moderately at 560 content in NM) in MM1.s-SCID mice (n = 5). The mice were then killed 90 min oscillation per min in IKA KS 130 basic plate shaker for 6 h at room temperature, post injection and the organs were dissected. The organs and blood were processed followed by lyophilization in Thermo Fischer SAVANT RVT5105 refrigerated by microwave digestion in HNO /H O under pressure and the TC measurements 3 2 2 vapour trap lyophilizer to obtain the HSA-TC as dry powder. The lyophilized −1 were measured in mg mL by inductively coupled plasma optical emission spec- product was reconstructed in 0.9% saline immediately before use. The con- −1 −1 trometry and adjusted to %ID g tissue. centration of the TC in the reconstituted injection was 5.6 mg mL (characterized by inductively coupled plasma optical emission spectrometry and that of HSA in −1 the reconstituted solution was 8.95 mg mL (determined by Bio-Rad Quick Start Biodistribution of HSA-TC in mice. A similar procedure described above was used Bradford Protein Assay kit). The size and dispersity of the HSA-TC was confirmed −1 in the HSA-TC study. HSA-TC (50 µL per mice, 30 mg kg ) was injected intra- by DLS (Malvern Zetasizer nano Series) and electron microscopy (JEOL TEM-1400 venously in C57B6 mice on day 10 post intracardiac injection of PyMT-BO1 GFP/ electron microscope). Luc breast cancer cells. In addition to the untreated group, each group of five mice were killed at 1, 3, 6 and 24 h post injection after BLI of the living animal. The Cell culture. All cell lines underwent STR profiling and tested for mycoplasma organs and blood were harvested and imaged ex vivo using BLI to estimate the contamination. MM1.S-Luc cells were cultured in RPMI1640 medium containing tumour distribution. To determine uptake of the Ti distribution, blood and lower 10% heat-inactivated FBS and 2 mercaptoethanol (50 μM final) and 1× of all of the limbs at the different time points were were digested by microwave in HNO /H O 3 2 2 −1 following: penicillin/streptomycin (100 μgmL final), sodium pyruvate (1 mM under pressure and analysed by inductively coupled plasma mass spectrometry. −1 final), non-essential amino acids, HEPES (10 mM final) and L-glutamine. A similar The Ti content in each tissue were measured in mg mL after subtraction of condition was used to culture PyMT-BO1 cells in DMEM media plus 10% FBS. background Ti from the untreated cohorts. Tumour models in mice. All animal studies were conducted in compliance with 18 18 FDG-PET imaging. FDG-PET imaging was performed on MM mice 30 days the guidelines established by the Animal Studies Committee at Washington Uni- post cell injection (n = 4) along with naive mice (n = 4). The mice were fasted for 6 versity in St. Louis, Missouri. Fox Chase SCID Beige mice (4-week, female) were h before each scan. After anesthetizing the mice with 1.5–2% isoflurane and oxy- purchased from Charles River laboratories for developing the disseminated MM −1 gen, 0.19 mCi (7.03 MBq) 0.1 mL of FDG was administered intravenously. A 10- model. C57BL/6J mice (6-week, female) were purchased from the Jackson min transition scan was performed just before the 10 min emission at 1 h post Laboratory for developing the metastatic breast cancer model. MM1.S-Luc cells injection using a MicroPET-Inveon MultiModality scanner (Siemens Preclinical (1 × 10 in 100 µL per mouse) were injected intravenously for the MM model and Solutions, Erlangen, Germany). The animals were placed on the microCT in the PyMT-BO1-GFP-Luc cells (1 × 10 in 50 µL per mouse) were injected in the left same position to obtain anatomical imaging and co-registered to the microPET ventricular chamber for the breast cancer model. BLI was used to monitor cell image. The data were analysed using Inveon Research Workstation software, by viability and tumour burden. manually drawing three-dimensional ROI from PET images using CT anatomical guidelines. The activity associated with tumour was measured and maximum SUVs Bioluminescence imaging. Ex vivo and in vivo bioluminescence imaging of MM1. −1 were calculated using SUV = ([mCi mL ] x [animal weight (g)]/[injected dose S-Luc in SCID mice and PyMT-BO1 GFP/Luc in C57BL/6J was performed on an (mCi)]). IVIS Lumina (PerkinElmer, Waltham, MA; Living Image 4.3, 5 min to 1 s exposure, bin2–8, FOV12.5 cm, f/stop1, open filter). Image analysis was performed using Living Image 2.6 software. For in vitro imaging, optimal bioluminescence for In vivo CRIT in disseminated MM1.S-luc/SCID mouse model. After MM1.S cell subsequent in vivo studies was determined by plating 1 × 10 cells in a black 24- injection, considered day 1, mice were imaged weekly using BLI for 7–9 weeks. well plate. Fixed regions of interest (ROIs) were drawn on each well and images Treatment was initiated on day 5. Mice were administered 50 μL of NM-TC (0.25 8 −1 −2 −1 captured after an exposure time of 10 s. A radiance >1× 10 photons s cm sr −1 −1 18 mg kg TC) intravenously followed by 29.6 MBq 0.1 mL of FDG 90 min later, was considered as the threshold. For in vivo imaging, mice were injected intra- also administered intravenously (n = 15). Control mice (n = 5 per group) were −1 peritoneally with D-luciferin (150 mg kg in PBS; Gold Biotechnology, St. Louis, administered with NM-TC or FDG alone. A total of four treatment cycles at an MO) and imaged after anaesthetizing with isoflurane (2% vaporized in O ). Total interval of 1 week were given per animal, where a cycle refers to an administration −1 −2 −1 photon flux (photons s cm sr ) was measured from fixed ROIs over the entire of NM-TC and FDG. BLI and survival of these groups of mice were tracked along mouse. with untreated controls (n = 10) on a weekly basis. Food was withheld from mice for 6 h before administering FDG and kept in a dark, lead-shielded room post Pharmacokinetics of NM-TC in rats. For pharmacokinetic analysis, the NM at a injection. The weight and any physical signs of distress were also monitored closely. −1 dose of 25 μLkg was administered intravenously to rats (n = 5) and serial blood The mice were killed by cervical dislocation after anaesthesia with 5% isoflurane samples were drawn at 1, 5, 15, 30, 60, 90, 120 and 1440 min for metal analysis. The when there was a loss of >20% total body weight. For rechallenging studies on CRIT-RES PK in rats is done to allow enough sample for serial measures in the same animal. resistant MM1 cells (MM1 ), the bone marrow (BM) was harvested from 4 −1 −2 −1 Otherwise, blood samples must be pooled, masking inter-specimen variability. treated mice when whole-body radiance exceeded 5 × 10 photons s cm sr . Briefly, animals are weighed, anesthetized and shaved on the ventral side of the BM was harvested by flushing the femoral shaft with PBS and reinjecting the cells neck. The surgical area is cleaned and wiped free of hair before surgery. Following into a fresh set of Fox Chase SCID Beige mice. The treatment schedule followed stabilization, the incision site is anesthetized, and an incision made just off midline was similar to MM1.S described above. BLI was used to track and compare to the trachea through the skin. The carotid artery is exposed by blunt dissection treatment response in a CRIT group (n = 10) and an untreated group (n = 10). 10 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE In vivo CRIT in metastatic breast cancer PyMT-BO1 model. A similar procedure Data availability. The data that support the findings of this study are available described above was followed in the HSA-TC CRIT study. On day 2 post intra- from the corresponding author on reasonable request. cardiac injection of PyMT-BO1-GFP-Luc cells in C57BL/6J mice, BLI was used to confirm tumour viability and engraftment. The mice were stratified into the fol- Received: 23 May 2017 Accepted: 21 December 2017 lowing groups: untreated control, HSA-TC only, FDG only and combination of HSA-TC and FDG cohorts. Each group has five mice weighing between 18 and 20 g. All mice were fasted 3 h before commencing the treatment on day 2. HSA-TC −1 (100 µL per mice, 30 mg kg was injected intravenously in the HSA-TC and HSA- TC + FDG groups. This was followed by intraperitoneal (i.p.) injection of the 18 18 FDG (800 µCi in 50 µL saline containing 0.02% ethanol) in the FDG-only group and the HSA-TC + F FDG group after 2 h from the drug injection. This References treatment regimen was performed every 2 day for 3 times and BLI images were 1. Steeg, P. S. Targeting metastasis. Nat. Rev. Cancer 16, 201–218 (2016). captured first on each occasion before the therapy. Tumour response was quan- 2. Lee, D. W. et al. Current concepts in the diagnosis and management of cytokine tified using both whole body and then lower limb signals from BLI. release syndrome. Blood 124, 188–195 (2014). 3. Sipkins, D. A. et al. In vivo imaging of specialized bone marrow endothelial microdomains for tumour engraftment. Nature 435, 969–973 RNA analysis. Bone marrow was flushed from femurs using cold PBS and single (2005). cell suspensions were prepared by filtering through a 40 μm nylon strainer. Red 4. Hasenclever, D. & Diehl, V. A prognostic score for advanced Hodgkin’s disease. blood cells were lysed using Red Blood Cell Lysing Buffer (Sigma-Aldrich) and International prognostic factors project on advanced Hodgkin’s disease. N. CRIT-RES washed with cold PBS. MM1 cells were sorted by fluorescent-activated cell Engl. J. Med. 339, 1506–1514 (1998). CRIT-RES sorting for GFP. RNA was extracted from MM1 cells using the Qiagen 5. De Angulo, G., Yuen, C., Palla, S. L., Anderson, P. M. & Zweidler-McKay, P. A. RNAeasy mini kit according to the manufacturer’s instructions. cDNA was syn- Absolute lymphocyte count is a novel prognostic indicator in ALL and AML: thesized using qRT-PCR was performed on the Applied Biosystems StepOnePlus implications for risk stratification and future studies. Cancer 112, 407–415 Real-Time System (Thermo Fisher) using pre-designed TaqMan Gene Expression (2008). Assays (Life Technologies; 18S RNA Hs99999901 and glut1 Hs00892681) 6. Wilcox, R. A. et al. The absolute monocyte and lymphocyte prognostic score according to the manufacturer’s instructions. The ΔΔCT method was used to predicts survival and identifies high-risk patients in diffuse large-B-cell calculate changes in fold expression and results were analysed with GraphPad lymphoma. Leukemia 25, 1502–1509 (2011). Prism 6. 7. Ege, H. et al. Prediction of survival using absolute lymphocyte count for newly diagnosed patients with multiple myeloma: a retrospective study. Br. J. Haematol. 141, 792–798 (2008). Flow cytometry. Red blood cells in the flushed BM were lysed using Red Blood 8. Spring, B. Q. et al. A photoactivable multi-inhibitor nanoliposome for tumour Cell Lysing Buffer (Sigma-Aldrich) and washed with cold PBS. Cell subsets were control and simultaneous inhibition of treatment escape pathways. Nat. evaluated by flow cytometry by re-suspending the cells in staining buffer (PBS Nanotechnol. 11, 378–387 (2016). supplemented with 0.5% bovine serum albumin and 2 mM EDTA) and incubating 9. Shaffer, T. M., Pratt, E. C. & Grimm, J. Utilizing the power of Cerenkov light for 30 min at 4 °C with pre-titrated saturating dilutions of the following with nanotechnology. Nat. Nanotechnol. 12, 106–117 (2017). fluorochrome-labelled monoclonal antibodies: CD49d-PE/Dazzle594 (clone 9F10; 10. Hartl, B. A., Hirschberg, H., Marcu, L. & Cherry, S. R. Activating photodynamic Biolegend), CD29-APC (clone TS2/16; Biolegend), human CD45-Alexa Fluor 700 therapy in vitro with cerenkov radiation generated from yttrium-90. J. Environ. (clone HI30; Biolegend), CD138-BV421 (clone B-B4; Biolegend) and murine Pathol. Toxicol. Oncol. 35, 185–192 (2016). CD45-BV510 (clone 30-F11; BD Biosciences). To measure LLP2A binding, cells 11. Kotagiri, N., Sudlow, G. P., Akers, W. J. & Achilefu, S. Breaking the were washed and resuspended in Hanks Balanced Salt Solution (HBSS; Lonza) 2+ 2+ depth dependency of phototherapy with Cerenkov radiation and low-radiance- containing Ca (1.26 mM), Mg (0.81 mM) and 0.1% bovine serum albumin responsive nanophotosensitizers. Nat. Nanotechnol. 10, 370–379 (Sigma-Aldrich). Aliquots of cells were left untreated or treated with 10 nM BIO5192 (Tocris) at 4 °C for 15 min. Samples were then incubated with or without (2015). −1 LLP2A-Cy5 (0.41 µg µL ) for 30 min at 4 °C and washed twice with HBSS.1 Dead 12. Kamkaew, A. et al. Cerenkov radiation induced photodynamic therapy using −1 cells were excluded from these assays by staining with 2 mg mL 7-amino- chlorin e6-loaded hollow mesoporous silica nanoparticles. ACS Appl. Mater. actinomycin D (Molecular Probes, Eugene, OR) for 5 min before analysis. Samples Interfaces 8, 26630–26637 (2016). were analysed on a Beckman Coulter Gallios flow cytometer and data were ana- 13. Pantel, K., Alix-Panabieres, C. & Riethdorf, S. Cancer micrometastases. Nat. lysed using FlowJo software (TreeStar, Ashland, OR). Statistical comparisons of Rev. Clin. Oncol. 6, 339–351 (2009). flow cytometry data were performed using an unpaired two-tailed Student's t test 14. Ghobrial, I. M. Myeloma as a model for the process of metastasis: implications (GraphPad Prism). P values ≤0.05 were considered significant. for therapy. Blood 120,20–30 (2012). 15. Su, X. et al. Antagonizing Integrin beta3 Increases Immunosuppression in Cancer. Cancer Res. 76, 3484–3495 (2016). Colony-forming unit assay. For enumeration of blood colony-forming cells 16. Kröger, N., Kleeberg, U. R., Mross, K., Edler, L. & Hossfeld, D. K. Phase II (CFU), aliquots of bone marrow cells were incubated in duplicate in commercially clinical trial of titanocene dichloride in patients with metastatic breast cancer. available growth factor-supplemented methylcellulose medium for mouse CFU-C Oncol. Res. Treat. 23,60–62 (2000). (Stem Cell Technologies, Vancouver, BC) as described. Blood CFU were enum- 17. Guo, M., Sun, H., McArdle, H. J., Gambling, L. & Sadler, P. J. Ti(IV) uptake and erated after 6–8 days of culture. release by human serum transferrin and recognition of Ti(IV)-transferrin by cancer cells: understanding the mechanism of action of the anticancer drug titanocene dichloride. Biochemistry 39, 10023–10033 (2000). Competitive bone marrow transplant. C57BL/6 mice (B6.CD45.2) were treated 18. Buettner, K. M., Snoeberger, R. C. 3rd, Batista, V. S. & Valentine, A. M. with either CRIT (n = 5) or PBS control (n = 5). BM was harvested on day 40 by Pharmaceutical formulation affects titanocene transferrin interactions. Dalton flushing the femoral shaft with PBS. Bone marrow from treated mice was mixed Trans. 40, 9580–9588 (2011). with congenic BM (B6.CD45.1/2) at a ratio of 1:1 before infusion of 1 × 10 total 19. Girotti, A. W. Photosensitized oxidation of membrane lipids: reaction BM cells into lethally irradiated (TBI 1100 cGy) B6.CD45.1 recipients via lateral tail pathways, cytotoxic effects, and cytoprotective mechanisms. J. Photochem. vein injection. Peripheral blood was analysed by flow cytometry at week 8+ to Photobiol. B 63, 103–113 (2001). quantify engraftment of donor BM. Percentage of cells derived from treated 20. Kim, J., Santos, O. A. & Park, J.-H. Selective photosensitizer delivery into donor BM (CD45.2) was calculated as a percentage of total donor BM (CD45.2 + plasma membrane for effective photodynamic therapy. J. Control. Release 191, CD45.1/2). 98–104 (2014). 21. Torchilin, V. P. Recent advances with liposomes as pharmaceutical carriers. Nat. Rev. Drug Discov. 4, 145–160 (2005). Statistical analysis. Statistical significance was measured by Student’s t test using 22. Sarin, H. Physiologic upper limits of pore size of different blood capillary types GraphPad Prism software (GraphPad, San Diego, CA). Kaplan–Meier survival and another perspective on the dual pore theory of microvascular permeability. curves were plotted using GraphPad Prism software. Unless noted otherwise, all J. Angiogenesis Res. 2,1–19 (2010). values are means and error bars are standard deviations. For animal studies, sample 23. Soodgupta, D. et al. Small molecule MYC inhibitor conjugated to integrin- size estimates depend on the effect size (mean difference between untreated and targeted nanoparticles extends survival in a mouse model of disseminated treatment groups/SD) of the outcome. For effect size of 2.1 and using a two-sided t multiple myeloma. Mol. Cancer Ther. 14, 1286–1294 (2015). test, typically five per group were needed with 80% power to detect a significant 24. Olson, D. L., Burkly, L. C., Leone, D. R., Dolinski, B. M. & Lobb, R. R. Anti- difference at a type I error rate of 0.05. There were no exclusion criteria, except for alpha4 integrin monoclonal antibody inhibits multiple myeloma growth in a the breast cancer study where the use of female mice was needed to recapitulate the murine model. Mol. Cancer Ther. 4,91–99 (2005). histopathology of female breast cancer. There was no blinding of investigators in this research, including animal study. NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 11 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 25. Kiziltepe, T. et al. Rationally engineered nanoparticles target multiple myeloma Acknowledgements cells, overcome cell-adhesion-mediated drug resistance, and show enhanced Funding for this project was in part by grants from the US National Institutes of Health efficacy in vivo. Blood Cancer J. 2, e64 (2012). (U54CA199092, NIBIB R01 EB008111, NCI R01 CA194552, NCI R01 CA152329 and 26. Peng, L. et al. Combinatorial chemistry identifies high-affinity peptidomimetics NIH P50 CA094056), the Department of Defence Breast Cancer Research Program against [alpha]4[beta]1 integrin for in vivo tumor imaging. Nat. Chem. Biol. 2, (W81XWH-16-1-0286), and the Alvin J. Siteman Cancer Research Fund (11-FY16-01). 381–389 (2006). We thank the staff of Siteman Cancer Center Small Animal microPET Facility for 27. Korfel, A. et al. Phase I clinical and pharmacokinetic study of titanocene assistance with imaging and therapy studies; Wandy Beatty for assistance with the dichloride in adults with advanced solid tumors. Clin. Cancer Res. 4, 2701–2708 electron microscope imaging; Suellen Greco for clinical chemistry and H&E images; (1998). Patty Wurm for coupled plasma optical emission spectrometry analysis and Deepti 28. Mitchell, G. S., Gill, R. K., Boucher, D. L., Li, C. & Cherry, S. R. In vivo Soodgupta for flow cytometry analysis. Cerenkov luminescence imaging: a new tool for molecular imaging. Philos. Trans. A Math. Phys. Eng. Sci. 369, 4605–4619 (2011). Author contributions 29. Bredella, M. A., Steinbach, L., Caputo, G., Segall, G. & Hawkins, R. Value of N.K. and S.A. conceived study; N.K., M.L.C. and S.A. designed the research; N.K., M.L.C., FDG PET in the assessment of patients with multiple myeloma. Ajr. Am. J. M.R., J.P., G.S., L.M., K.W., C.C., L.L., L.H., C.E., P.K., M.Z. and X.S. performed the Roentgenol. 184, 1199–1204 (2005). research; G.C. and X.Y. generated the NM and performed pharmacokinetic and bio- 30. Cavo, M. et al. Role of 18F-FDG PET/CT in the diagnosis and management of distribution analysis; M.S., K.W., G.L. and S.A. contributed materials; S.A. supervised multiple myeloma and other plasma cell disorders: a consensus statement by overall study; N.K., M.L.C., M.T., G.L., J.D., C.E., P.K. and S.A. wrote and edited the the International Myeloma Working Group. Lancet Oncol. 18, e206–e217 manuscript. (2017). 31. Agool, A. et al. Radionuclide imaging of bone marrow disorders. Eur. J. Nucl. Med. Mol. Imaging 38, 166–178 (2011). Additional information 32. Soodgupta, D. et al. Ex vivo and in vivo evaluation of overexpressed VLA-4 in Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- multiple myeloma using LLP2A imaging agents. J. Nucl. Med. 57, 640–645 017-02758-9. (2016). 33. Ford, C. E., Hamerton, J. L., Barnes, D. W. & Loutit, J. F. Cytological Competing interests: The authors declare no competing financial interests. identification of radiation-chimaeras. Nature 177, 452–454 (1956). 34. Thota, R., Pauff, J. M. & Berlin, J. D. Treatment of metastatic pancreatic Reprints and permission information is available online at http://npg.nature.com/ adenocarcinoma: a review. Oncology 28,70–74 (2014). reprintsandpermissions/ 35. Al-Hajeili, M., Azmi, A. S. & Choi, M. Nab-paclitaxel: potential for the treatment of advanced pancreatic cancer. Onco Targets Ther. 7, 187–192 Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in (2014). published maps and institutional affiliations. 36. Kratz, F. Albumin as a drug carrier: design of prodrugs, drug conjugates and nanoparticles. J. Control. Release 132, 171–183 (2008). 37. Kim, C. et al. Comparison of the intraperitoneal, retroorbital and per oral routes for F-18 FDG administration as effective alternatives to intravenous Open Access This article is licensed under a Creative Commons administration in mouse tumor models using small animal PET/CT studies. Attribution 4.0 International License, which permits use, sharing, Nucl. Med Mol. Imaging 45, 169–176 (2011). adaptation, distribution and reproduction in any medium or format, as long as you give 38. Fueger, B. J. et al. Impact of animal handling on the results of 18F-FDG PET appropriate credit to the original author(s) and the source, provide a link to the Creative studies in mice. J. Nucl. Med. 47, 999–1006 (2006). Commons license, and indicate if changes were made. The images or other third party 39. Thorek, D. L., Ogirala, A., Beattie, B. J. & Grimm, J. Quantitative imaging of material in this article are included in the article’s Creative Commons license, unless disease signatures through radioactive decay signal conversion. Nat. Med. 19, indicated otherwise in a credit line to the material. If material is not included in the 1345–1350 (2013). article’s Creative Commons license and your intended use is not permitted by statutory 40. Bernhard, Y., Collin, B. & Decreau, R. A. Redshifted Cherenkov radiation for regulation or exceeds the permitted use, you will need to obtain permission directly from in vivo imaging: coupling Cherenkov radiation energy transfer to multiple the copyright holder. To view a copy of this license, visit http://creativecommons.org/ forster resonance energy transfers. Sci Rep 7, 45063 (2017). licenses/by/4.0/. 41. Pan, D. et al. A strategy for combating melanoma with oncogenic c-Myc inhibitors and targeted nanotherapy. Nanomedicine 10, 241–251 (2015). © The Author(s) 2018 12 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Nature Communications Springer Journals

Radionuclides transform chemotherapeutics into phototherapeutics for precise treatment of disseminated cancer

Download PDF
12 pages

Loading next page...
 
/lp/springer-journals/radionuclides-transform-chemotherapeutics-into-phototherapeutics-for-4W00rL3KCg

References (45)

Publisher
Springer Journals
Copyright
Copyright © 2018 by The Author(s)
Subject
Science, Humanities and Social Sciences, multidisciplinary; Science, Humanities and Social Sciences, multidisciplinary; Science, multidisciplinary
eISSN
2041-1723
DOI
10.1038/s41467-017-02758-9
pmid
29348537
Publisher site
See Article on Publisher Site

Abstract

ARTICLE DOI: 10.1038/s41467-017-02758-9 OPEN Radionuclides transform chemotherapeutics into phototherapeutics for precise treatment of disseminated cancer 1,5 2 2 1 1 Nalinikanth Kotagiri , Matthew L. Cooper , Michael Rettig , Christopher Egbulefu , Julie Prior 2 1 1 2 1 1 Grace Cui , Partha Karmakar , Mingzhou Zhou , Xiaoxia Yang , Gail Sudlow , Lynne Marsala , 2 2 1 1 2 Chantiya Chanswangphuwana , Lan Lu , LeMoyne Habimana-Griffin , Monica Shokeen , Xinming Xu , 2 2 2 2 1,3,4 Katherine Weilbaecher , Michael Tomasson , Gregory Lanza , John F. DiPersio & Samuel Achilefu Most cancer patients succumb to disseminated disease because conventional systemic therapies lack spatiotemporal control of their toxic effects in vivo, particularly in a compli- cated milieu such as bone marrow where progenitor stem cells reside. Here, we demonstrate the treatment of disseminated cancer by photoactivatable drugs using radiopharmaceuticals. An orthogonal-targeting strategy and a contact-facilitated nanomicelle technology enabled highly selective delivery and co-localization of titanocene and radiolabelled fluorodeox- yglucose in disseminated multiple myeloma cells. Selective ablation of the cancer cells was achieved without significant off-target toxicity to the resident stem cells. Genomic, proteomic and multimodal imaging analyses revealed that the downregulation of CD49d, one of the dimeric protein targets of the nanomicelles, caused therapy resistance in small clusters of cancer cells. Similar treatment of a highly metastatic breast cancer model using human serum albumin-titanocene formulation significantly inhibited cancer growth. This strategy expands the use of phototherapy for treating previously inaccessible metastatic disease. 1 2 Department of Radiology, Washington University School of Medicine, St. Louis, MO 63110, USA. Department of Medicine, Washington University School of Medicine, St. Louis, MO 63110, USA. Department of Biochemistry and Molecular Biophysics, Washington University School of Medicine, St. Louis, MO 63110, USA. Department of Biomedical Engineering, Washington University, St. Louis, MO 63105, USA. Present address: James L Winkle College of Pharmacy, University of Cincinnati, Cincinnati, OH 45267, USA. Nalinikanth Kotagiri and Matthew L. Cooper contributed equally to this work. Correspondence and requests for materials should be addressed to S.A. (email: [email protected]) NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 1 | | | 1234567890():,; ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ost deadly cancers are associated with metastatic Results spread , requiring systemic treatment strategies with Contact-facilitated drug delivery via VLA-4-targeted nanomi- Mchemotherapeutic drugs and radiation therapy. Second celles. Targeted delivery of a radionuclide and a drug is necessary generation systemic therapies rely on targeting precise molecular to enable co-localization in the same or adjacent cell for signatures of cancer or invoke immune responses against certain subsequent activation and therapy. Once internalized by a target epitopes specific to cancer. While immensely promising, on-going cell, the radionuclide which essentially behaves as a point source clinical trials indicate that these strategies are often associated of photoelectronic energy, can excite or stimulate photoactive with life-threatening on-target, off-tumour toxicities . For many materials in its vicinity (Fig. 1a). Previous studies have demon- cancers, bone marrow is invariably involved as the point of origin strated the modularity afforded by this approach in treating 3 10 or a distant metastatic niche . The microenvironment of the bone cancer cells or subcutaneous solid tumours or mice using dif- 11,12 marrow is laden with hematopoietic stem cells and progenitors, ferent radionuclides and photosensitizer combinations . making it a highly challenging niche for selective cancer cell Unlike subcutaneous solid tumour xenografts, which do not killing and a difficult terrain for emerging systemic therapeutics. recapitulate the physiopathology of human cancer and are trea- Moreover, in advanced stages of disseminated cancers, patients table by conventional PT, most disseminated tumour models often present with extremely low total lymphocyte counts. As present a different set of challenges because they are embedded in such, they are stratified as severely immunocompromised and the complex and protective microenvironment of the bone 4–7 13 carry the risk of poor prognosis and low overall survival rates . marrow . As a result, it would require more effective targeting These patients are typically unsuitable candidates for and delivery strategies to maximize cell death. We selected mul- existing systemic therapies and emerging immunotherapies. tiple myeloma (MM), an incurable plasma cell dyscrasia that Photodynamic therapy or phototherapy (PT) can offer high predominantly affects the bone marrow, spleen, and bones as the spatiotemporal precision and control of tumour killing through a representative orthotopic disseminated tumour model (Fig. 1a) . combination of direct cytotoxicity, immune-stimulatory, and We also used PyMT-BO1 cancer cell line derived from transgenic antiangiogenic mechanisms . Therefore, PT could serve as an PyMT cancer cells as a highly aggressive metastatic breast cancer effective therapeutic platform and a viable option for model (see below). disseminated cancers, offering an alternative treatment for Titanocene (TC) was used in this study as the photosensitizer the chemotherapy-refractory disease. However, the limited for several reasons, including its UV light excitability and penetration of external light has confined PT to the treatment of responsiveness to low radiance of CR ; biodegradability with surface accessible lesions. In addition, a priori knowledge significantly low cellular footprint post therapy; ease of human of tumour location is a prerequisite for initiating PT, which often translation due to its safety profile in phase 2 clinical trials ; and is indeterminate in the case of disseminated tumours. small size and lipophilicity, allowing integration into lipid-based An alternative approach that delivers light or stimulate vehicles and incorporation into cell membranes post targeting. In light-sensitive drugs within tissues and inside cells in vivo could addition to harvesting CR luminescence, the metal centre can also facilitate the treatment of PT-inaccessible systemic and metastatic interact with radiation particles to further stress cells. However, cancers. Clinically relevant radiopharmaceuticals are reliable two fundamental challenges to TC and similar photoactive drugs, sources of Cerenkov radiation (CR) for cancer imaging .A transvascular delivery to tumour cells and cellular localization, decaying radionuclide could excite materials through, including have remained unaddressed in the context of CRIT. In our the direct interaction of electron and positron emission with previous study, we used transferrin (Tf) to deliver TC to tumour 11 17,18 matter, particularly metals; the emission of ultraviolet-blue light . Tf has only two binding pockets for TC . In the cells emitted by beta (β) particles, known as CR, to generate cytotoxic docking process, the cyclopentadienyl (Cp) ligands of TC can be reactive oxygen species (ROS); chemiluminescent reaction when displaced, leaving the Ti(IV) ion alone as the predominant ambient ionizing radiation excites bulk water; and emission of γ component that binds to the pockets . Because both photo- photons after the annihilation event. For simplicity, we group all activation of Ti(IV) ion and oxidation of Cp ligand to peroxyl these effects as Cerenkov radiation-induced therapy (CRIT). radical contribute to the cytotoxicity of TC, Tf-mediated Therefore, a critical component of the study is to efficiently transport of TC would potentially lower the therapeutic efficacy harvest the diverse potential effects of radionuclides to stimulate of Tf-TC. spatiotemporal cell death in the presence of photosensitizers. ROS-mediated damage to lipid membranes is a primary mode Many drugs possess photoactive properties, but the absence of a of action in PT . Given the short half-lives and small diffusion depth-independent photoelectronic energy source has confined distance of some ROS, the mode of delivery of the drug to the their use as chemotherapeutics, preventing therapy enhancement target cell and its proximity to the cell membrane are important through a complementary phototherapeutic effect. considerations for effective therapy. There is also growing In this study, we hypothesize that CR-mediated conversion of evidence that therapeutic efficacy of PT can be enhanced by light-sensitive drugs to phototherapeutic agents will induce cell selective delivery of hydrophobic photoactive drugs to the plasma death through pathways distinct from the ground state drug (che- membrane compared to receptor-mediated endocytotic uptake . motoxicity) and in a highly selective fashion for the treatment of The contact-facilitated delivery of drugs to the plasma membrane diverse cancer phenotypes. Using multiple myeloma (MM) and by lipid vehicles serves this purpose efficiently. Although metastatic breast cancer models in mice, we demonstrate that liposomal formulations can deliver drugs to cells through this incorporating unmodified and pristine hydrophobic light-sensitive mechanism, conventional liposomes have an average diameter of drugs in tumour-targeted lipid nanomicelles or human serum 100 nm (for unilamellar vesicles) and 0.5–5 μm (for multilamellar albumin (HSA) nanoparticles, selectively deliver the agents in vesicles) , which exceeds the physiologic upper limit of 60 nm disseminated cancer cells. Subsequent in vivo administration of a pore size for transvascular transport of macromolecules to flow radiopharmaceutical for CRIT inhibits the proliferation of across capillary walls of bone marrow . To deliver pristine TC to disseminated multiple myeloma and aggressive metastatic breast the plasma membrane of MM cells, we used nanoscale cancer cells in mice. Our treatment strategy transforms che- unilamellar phospholipid micelles, also known as nanomicelles motherapeutics to spatiotemporally photoactivatable drugs using (NM), as a carrier vehicle. The NM have an average diameter of clinically relevant radiopharmaceuticals and expands the use of ≤15 nm, which is ideal for targeting the bone marrow interstitial phototherapy for treating previously inaccessible metastatic disease. space . The upregulation of a key adhesion molecule, VLA-4 2 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE Bone Stromal cell Stem cell (6) (2) VLA4 (5) (4) Glut Cancer cell (3) Radiopharmaceuticals Targeted (1) nanomicelles Blood vessel bc LLP2A Phospholipid micelle d e Titanocene Fig. 1 Orthogonal cancer targeting strategy using nanomicelles. a Schematic of the process of photoactivation of Titanocene in disseminated cancer cells in the bone marrow microenvironment. The various phases are numbered: 1. Administration of targeted NM-TC; 2. The targeted NM enter the bone marrow from the vasculature and bind to α4β1 receptor on the cancer cells and subsequently deliver the drug to the cell; 3. Administration of radiopharmaceuticals 18 18 ( FDG), which is typically 1.5–2 h after phase (1); 4. FDG enters the cancer cells through the overexpressed Glut transporters on cancer cells; 5. Once the drug and radiopharmaceutical are co-localized in the cancer cells, the former is photoactivated by the latter through CR leading to cell death (6). Notice that since the other vital cells in the bone marrow, such as stem cells and stromal cells, do not express the combination of α4β1 and glut receptors essential for the treatment to work, they would largely remain unaffected causing minimal off-target toxicity. b Schematic of phospholipid NM with VLA-4 homing ligands. c TEM image of micelles alone. Scale bar, 100 nm. Inset: single micelle. Scale bar, 10 nm. d Schematic of phospholipid NM encapsulating TC with VLA-4 homing ligands. e TEM image of micelle incorporated with TC in the membrane. Scale bar, 100 nm. Inset: single NM-TC. Scale bar, 10 nm NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 3 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 Table 1 Size distribution of the nanomicelles Sample Hydrodynamic diameter (nm) Polydispersity index Zeta potential (mV) Nanomicelle + LLP2A 11.9 ± 0.5 0.217 −0.81 TC nanomicelle + LLP2A 14.7 ± 2 0.241 2.12 of TC in rodent blood. The NM in circulation remained intact in vivo, until cleared or destroyed. However, the micelles have a Table 2 Metal (Ti) and TC content in nanomicelles and HSA limited half-life and must reach their target early before elimination. Sample Average Ti content (μg/ Average TC content 20 μL) (mg/mL) TC nanomicelle 0.52 0.192 VLA-4-targeted nanomicelles delivers TC to MM-avid organs. TC-HSA 21.65 5.613 The selectivity of LLP2A to MM cells and the serum stability of the NM in delivering the TC was determined by in vivo biodistribution analysis. Using inductively coupled plasma optical emission spectrometry, we determined the Ti metal content (α β integrin), in MM provides an attractive target for precision 4 1 24,25 ex vivo in organ samples from an orthotopic disseminated MM1. imaging and therapy . Human MM1.S cell line is widely used S/SCID model. We compared the biodistribution of NM-TC, to study MM in rodents. Screening of the MM human cell line, Tf-TC and MKT4, a water soluble analogue of TC that was used MM1.S, using anti-CD49d (α4) and CD29 (β1) antibodies 16,27 in phase 1 and phase 2 clinical trials at 90 min post injection. showed a ≥95% expression level of VLA-4 (Supplementary The choice of 90 min time point is based on rat PK data, which Figure 1). We loaded the NM with LLP2A (Supplementary showed a t of 123 min in rats but the rate of clearance after 90 1/2 Figure 2a), a small molecule peptidomimetic that binds VLA-4 min approached stasis, probably representing the contribution of with an exceptionally high affinity (IC50 = 2 pM) . LLP2A was an intraversation process of drug from tissues to blood (Fig. 2a). synthesized on a solid support, followed by conjugation to Although this time point is expected to be shorter in mice, we phospholipids (DSPE) engrafted with polyethylene glycol (PEG) chose 90 min for the mouse study to ensure that the blood chains to improve circulation in blood (see “Materials and concentration of TC is sufficiently low, to prevent potential Methods' section for details; Supplementary Figure 2b, c). The systemic toxicity, but not too late when the amount in tumour NM were generated as a microfluidized suspension containing tissue is small. In mice administered with NM-TC, the highest Ti LLP2A-PEG-DSPE and TC. Control NM that excluded the concentrations were found in skeletal tissue and spleen, homing ligand LLP2A or TC were also prepared (Fig. 1b, c). which typically house MM cells, with relative values of 115 ± 7 An average size distribution of NM with and without TC was −1 and 52 ± 9.5 μgg , respectively (Fig. 2b). In comparison, the 14.7 ± 2 nm and 11.9 ± 0.5 nm, respectively, with an average uptake of MKT4 was lower in tumour sites, with values of 53 ± 9 polydispersity index of 0.2 (Table 1). −1 −1 and 16 ± 4 μgg , for skeletal tissue and spleen, respectively We successfully loaded 0.19 mg mL of TC in the NM (Fig. 2b). Similarly, the accumulation of TC in these tissues for (Table 2). Based on the full-width half-maximum of the NM size −1 mice treated with Tf-TC was only 27.5 ± 6 and 14 ± 1 μgg , distribution (about 15 nm), the volume of NM, and the net respectively. These results demonstrate the advantage of using concentration of TC per volume of NM using inductively coupled NM to deliver TC to MM target organs. Previous studies have plasma optical emission spectrometry, we determined the average suggested that the cylopentadienyl rings in TC, which assists in number of TC per NM as 3 (range, 2–5). The incorporation of TC stabilizing the Ti(IV) ion in a monomeric form, are lost in MKT4 in the lipid layer was confirmed by electron microscopy (Fig. 1d, 17,18 and Tf-TC . Thus, sequestration of TC in the hydrophobic e). The metallic titanium (Ti) centre in TC rendered the vesicles region of NM may help stabilize the drug and minimize rapid loss electron dense in contrast to the control vesicles without TC. from target tissues. Upon addition into the NM, TC incorporated in the interface between the lipid and the hydrophilic layers, as evidenced by electron microscopy (Fig. 1d, e). Probably, the hydrolysis of TC CRIT inhibits tumour growth in disseminated MM mouse dichloride to the dihydroxyl derivative in aqueous medium model. We used an FDA approved and clinically employed created an amphiphilic structure, favouring the orientation of the radiopharmaceutical, FDG (t = 109.8 min), as a source of 1/2 two cyclopentadienyl and dihydroxyl moieties toward the 28 photoelectronic energy . The radiopharmaceutical, which is 29,30 hydrophobic core and the outer hydrophilic segment, respec- currently the gold standard for clinical imaging of MM , tively. Incorporation of LLP2A did not destabilize the NM and targets metabolically active tumours via the glucose transporter the presence of unnatural amino acids conferred protease (GLUT1) protein. By using an orthogonal-targeting GLUT1 and resistance and high plasma stability on the nanosystem . 18 VLA-4 strategy to, respectively, deliver the FDG and NM-TC to the MM cells, we aimed to minimize the potential saturation or In vivo pharmacokinetics of VLA-4-targeted nanomicelles.In depletion of the targeted receptors. In healthy subjects, FDG vivo pharmacokinetic (PK) profile of the NM-TC was studied in uptake is low in the bone marrow and spleen, but significantly naive rats. A plasma half-life of 123 min was obtained after higher in malignancy, inflammation or after administration of systemic administration (Fig. 2a). We performed the PK in rats hematopoietic growth factors . Using small animal positron instead of mice to obtain sufficient blood sample for serial emission tomography (PET) of MM in mice, we found more than measurements of TC concentration in the same animal. twofold uptake of FDG in bones compared to naive mice Otherwise, the small volume of blood in mice would require us to (Fig. 2c–i). pool samples from different mice, masking inter-specimen The performance of CRIT in a disseminated MM1.S/ SCID variability. Although the PK value in mice are expected to be mouse model was tested. Based on the biodistribution data, shorter than rats, the information allowed us to estimate half-life sequential tail vein injections of NM and then FDG were spaced 4 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE ab 0.05 NM-TC 0.04 MKT4 Tf-TC 0.03 0.02 0.01 0.00 0 0 204060 80 100 120 140 Time (min) ** 20 * cd ef Multiple Fore limb myeloma Naive Multiple Hind limb myeloma Naive Spine Max Fore limb Min g h Multiple myeloma Naive Spine Hind limb Fig. 2 Monitoring nanomicelles biodistribution and spread of multiple myeloma in vivo. a Pharmacokinetics of NM-TC in rats using coupled plasma optical emission spectrometry. Half-life is 123.4 min. b Comparison of biodistribution in mice of targeted NM-TC and pristine TC in vivo showing highest uptake 18 18 and retention in bones and spleen, characteristic of multiple myeloma, 2 h post injection. FDG-PET images showing increased uptake of FDG in mouse forelimbs, spine, and hind limbs of mice with multiple myeloma (c, e, g) compared to naive mice (d, f, h, i). Comparison of standard uptake values (SUV) of FDG in multiple myeloma vs. naive mice in various bones. Values are means ± s.e.m. *P < 0.05, **P < 0.01. n = 5 mice for each of the pharmacokinetics study in rats; and biodistribution study in mice 90 min apart to activate TC in tumours. Treatment was repeated confined at random sites within the major bones, particularly the four times at an interval of 1 week, and the disease progression vertebrae (Fig. 3b, e). These localized cancer cells continued to grow, was monitored weekly by bioluminescence imaging (BLI; Fig. 3a). albeit at a slow rate. The surviving cancer cells were subsequently A week interval was chosen for treatment for several reasons that extracted from the mice and reintroduced into a fresh group of naive include the need to allow the mice to fully recover from the SCID mice to determine response to when treated with CRIT. 18 18 treatment; account for full decay cycle of FDG; consider However, BLI (Fig. 4a) and FDG-PET did not show noticeable logistical reasons such as tail vein recovery; and allow sufficient differences between the treated and untreated groups, suggesting the time for imaging time points between treatment sessions. In the cells were resistant to CRIT. These CRIT-resistant MM1.S (MM1. CRIT-RES control groups consisting of untreated mice or those treated with S ) cells were harvested and analysed for the expression either NM or FDG alone (Fig. 3b, c), we observed an levels of GLUT1, α4and β1 integrins to determine whether uptake exponential increase in the BLI signal over several weeks, of FDG by GLUT1 or α4β1 binding of the NM were compro- demonstrating the systemic progression of the disease and mised. GLUT1 mRNA (Fig. 4b) or β1 cell surface expression indicating the primary involvement of the spleen and skeletal (Fig. 4c, d) analyses did not demonstrate significant difference 18 CRIT-RES tissues. In contrast, mice treated with NM-TC and FDG showed between the parental MM1.S and the MM1.S cells. How- CRIT-RES a conspicuous decrease in the disease progression, suggesting the ever, the MM1.S cells expressed lower cell surface α4than effective targeting and response of MM1.S to CRIT. Survival parental MM1.S cells (Fig. 4e, f). Flow cytometry analysis demon- studies revealed a significant advantage of the CRIT over the strated that LLP2A-Cy5, which selectively binds VLA-4 with high 32 CRIT-RES control groups with 50% surviving up to about 90 days compared affinity , did not internalize in the MM1.S cells compared to about 62 days for the control groups (Fig. 3d). Correlative to the parental MM1.S cells (Fig. 4g, h). These results suggest that 18 CRIT-RES FDG-PET imaging confirmed the lower tumour burden in the MM1.S cells had downregulated the expression of α4 CRIT-treated mice compared to the control groups (Fig. 3e, f). (CD49d), possibly impairing the binding of the LLP2A functiona- The mice were killed after they developed hind limb paralysis lized NM to some MM cells. Unlike in vitro studies where static resulting from spinal cord and spinal vertebral involvement. The incubation of nanoparticles can abrogate specificbinding of treated mice eventually succumbed to cancer due to the remnant receptor-targeted materials, the in vivo dynamics and the relatively MM cells that could not be completely eradicated by CRIT. small number of these resistance cells in the initial tumour popu- lation could have favoured the homing of NM-TC to the VLA-4 positive cells in mice. As a result, CRIT could have preserved a CRIT selects for α4-deficient multiple myeloma cells in vivo. subclone of MM1.S with low α4 that was present at low frequency in Residual cancer cells that escaped treatment appeared focal and the injected cells. Thus, targeting VLA-4-rich cancer cells selects for NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 5 | | | Brain Kidney Liver Muscle Heart Skin Spleen Lung Bone MM mice Naive mice Dose%/g blood Titanium content (µg/g) SUV (tissue/muscle) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 Cell injection CRIT CRIT CRIT CRIT (Days) 1 5 12 19 26 Untreated FDG NM CRIT Week 2 100,000 80,000 60,000 Week 4 40,000 20,000 Week 6 cd Untreated FDG NM CRIT –50 –100 –150 ** –200 0 3 7 13 20 27 34 41 48 40 60 80 100 120 Time (days) Time (days) ef CRIT Untreated ** Max Min FS F S Fig. 3 Response of multiple myeloma to CRIT. a Timeline of treatment. b Bioluminescence imaging of representative multiple myeloma-bearing mice in different treatment groups—untreated, FDG, NM controls and CRIT. All images are dorsal images and on the same scale. The images of control groups appear saturated on week 6 in comparison to CRIT. c Change in bioluminescence intensity as a result of treatment compared to untreated control. The intensity consistently remains lower than untreated controls during the treatment and beyond. d Comparison of survival of different treatment groups showing a twofold increase in survival in treated mice compared to control groups. **P < 0.01. e FDG-PET images of MM mice before and after treatment showing lower tumour burden in the latter. F: frontal view, S: sagittal view. Boxes denote tumour region. f SUV values of the treatment group were lower than untreated controls. **P < 0.01. n = 15 mice for CRIT, n = 10 mice for untreated control and n = 5 mice for NM-TC alone and FDG alone treated mice 6 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | Untreated CRIT Radiance (p/s/cm /sr) % fold change in bioluminescence intensity from control mice Percent survival SUV (tumor/muscle) NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE the subset of MM1.S cells with low CD49d expression levels and low CRIT preserves normal hematopoietic stem cells. An important levels of the activated conformation of VLA-4. A potential approach consideration during CRIT is to preserve and sustain the long- to achieving complete eradication of MM cells is to identify com- term viability of the hematopoietic stem cells and progenitor cell plementary biomarkers that allow the delivery of TC-loaded NM to population in the bone marrow. Clonogenic assays of normal all MM1.S cells or through the use of combination therapy that bone marrow progenitor cells extracted from mice treated with more effectively targets and eliminates both CRIT-responsive MM1. CRIT did not reveal a significant change in colony-forming units CRIT-RES Sand MM1.S cells in vivo. (CFU) compared to the control groups (Fig. 4i). Competitive ab ns 1.5 Untreated CRIT 1.0 0.5 0.0 02 468 Time (weeks) cd e f Parental Resistant Parental Resistant 100 100 100 80 80 80 60 60 60 40 40 40 40 20 20 20 28.12 99.92 0 0 0 0 1 2 3 0 1 2 3 0 1 2 3 0 1 2 3 0 10 10 10 10 10 10 10 10 10 10 10 10 0 10 10 10 10 CD29-APC CD49d-PD594 g h i Parental Resistant 100 100 15,000 ns 80 80 60 60 10,000 40 40 20 20 84.15 6.60 0 0 0 1 2 3 0 1 2 3 0 10 10 10 10 0 10 10 10 10 LLP2A-Cy5 jk CD45.2 CRIT + 60 CD45.1 Engraftment analysis Day 60 CD45.1/2 Day 0 PBS NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 7 | | | Parental Resistant Untreated NM FDG CRIT Untreated CRIT % of max % of max Photons/s % of max CFU/10 BM % CD 45.2 BM Relative Glut1 mRNA expression ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 bone marrow repopulation experiments showed that there was no tumour cells disseminated to most major organs, including the detrimental effect of CRIT on the primitive hematopoietic stem vertebrae and lower limbs. Quantitative analysis of the Ti con- cell compartment . The 2-month hematopoietic reconstitution tents in tissue samples from blood and lower limbs by inductively of mice transplanted with bone marrow from wild-type vs. CRIT- coupled plasma mass spectrophotometry showed the highest treated mice was not significantly different, suggesting that there accumulation of the metal in the limbs at 3 h post injection of was no obvious reduction in engrafting hematopoietic stem cell HSA-TC (Fig. 5d). The concentration of Ti in the lower limbs population after treatment with CRIT (Fig. 4j, k). decreased gradually over time (p < 0.05). At 24 h, the Ti content from the HSA-TC in the limbs was indistinguishable from the background content. We had to subtract the background Ti from HSA-TC nanoparticles deliver drug to metastatic breast cancer. untreated mice to determine the contribution of HSA-TC because We extended the use of CRIT in a metastatic breast cancer model. most mouse feedstock contains Ti products. PyMT-BO1 cell line is a highly aggressive breast cancer cell line derived from transgenic PyMT breast cancer mice . Previous reports have demonstrated that rapidly proliferating tumour cells HSA-TC inhibits growth of metastatic breast cancer in mice. actively internalize albumin for use as a source of nitrogen and We explored the feasibility of using CRIT to inhibit tumour energy, partially accounting for the formulation of drugs in growth in the highly metastatic PyMT-BO1 GFP/Luc breast 34–36 albumin or its nanoparticles for drug delivery to tumours . cancer model in C57B6 mice. Our biodistribution data indicate Thus, formulation of hydrophobic drugs, such as TC in albumin, that the accumulation of TC in the cancer-homing organs is will not only enhance solubilization in an aqueous medium, but highest at about 3 h post injection of HSA-TC. FDG (50–60 µL; also mediate delivery to tumours. We mixed TC (80 mg) in an 800 µCi per mouse) was administered intraperitoneally at 2 h aqueous solution (16 mL) containing 0.5% human serum albumin after the intravenous administration of HSA-TC. We chose (HSA) for 6 h before lyophilizing the entire mixture (see 'Mate- intraperitoneal instead of intravenous route for FDG injection rials and Methods' section). We used a low concentration of HSA to maintain consistency across the experiments because of the in this formulation to prevent potential immunogenic response in difficulty of finding viable tail veins in the same mouse for mice. The lyophilized product was reconstituted in 0.9% saline multiple injections of both HSA-TC and FDG. Comparison of immediately before use. Using inductively coupled plasma mass different routes of FDG administration in mice determined that spectrometry, we determined the concentration of TC in the the SUV of FDG injected intraperitoneally in tumours is opti- −1 reconstituted sample as 5.6 mg mL (Table 2). The correspond- mal at about 1 h post injection, and is similar to intravenous route −1 ing concentration of HSA in the formulation was 8.95 mg mL , 37,38 at closer to this time point . We hypothesized that adminis- as determined by using a protein assay. Dynamic light scattering tering the radionuclide at about 2 h after injection of the HSA-TC (DLS) measurements indicate that the HSA-TC nanoparticles (100 µL of 0.6 mg per 20 g mouse) will achieve maximum accu- were fairly monodispersed with an average size of 12–15 nm mulation of both CRIT effectors in tumours after 3 h. Three (Supplementary Figure 3). About 1.5% of the particles formed treatment cycles of HSA-TC and FDG were administered large aggregate clusters of 100 nm, creating a bimodal distribution 2 days apart starting from day 2 after initiating the metastatic that skewed the z-average diameter (120 nm) and polydispersity disease when the tumours are observable by BLI (Supplementary index (0.278). Electron microscopy size measurement correlated Figure 5). Groups with no treatment, treatment with HSA-TC with the DLS results, showing mostly monodispersed HSA-TC alone and FDG alone served as controls. Whole-body luciferase nanoparticles of 10–15 nm in diameter (Supplementary Figure 4), activity from day 2 to day 9 were analysed for tumour cell pro- along with few aggregates of 30–90 nm. The HSA nanoparticles liferation. Compared to the untreated mice (Fig. 6a), a small allowed us to load high concentration of the hydrophobic TC per decrease in BLI signal was observed in the HSA-TC (Fig. 6b) and volume of aqueous solution for subsequent delivery to metastatic FDG (Fig. 6c)-treated mice compared to the untreated cohort. breast cancer. However, the pattern of tumour growth in all the three control groups was similar. In contrast, the CRIT group showed sig- Biodistribution of HSA-TC in metastatic breast cancer model. nificant tumour stasis, with a few focused cluster of tumour cells Intracardiac injection of PyMT-BO1 cells stably transfected with that did not respond to the therapy (Fig. 6d, e). The slow growth GFP-firefly luciferase in mice-induced bone metastases, especially and focal nature of the CRIT-resistant cells suggests that this to the lower limbs and other major organs (Fig. 5a, b). By day 10, therapeutic method can transform metastatic cancer into a sur- the tumour burden was very high, requiring immediate killing gical disease. Whereas PyMT-BO1 model is an excellent model (Supplementary Figure 5). For the biodistribution study, we for the rapid evaluation of drugs or treatment methods, all the reconstituted the lyophilized HSA-TC in saline and administered animals eventually died or were killed between day 9 and 12 due 100 µL of 0.6 mg per 20 g mouse. Both the non-invasive in vivo to the aggressiveness of this model. The fast death cycle prevents (Fig. 5b) and the ex vivo (Fig. 5c) BLI analysis showed that the longitudinal evaluation of each animal, which is needed to obtain Fig. 4 CRIT selects for CD49d cells in MM model. a Bioluminescence intensity plot showing resistant nature of MM cells extracted from treated cohort CRIT-RES (MM1 ) upon rechallenging with CRIT in fresh mice. b No difference in GLUT1 mRNA expression was observed between parental MM1.S cells and CRIT-RES resistant MM1.S cells as assessed by qRT-PCR. ns not significant. c–h No difference in expression of CD29 was observed between MM1.S cells (c) CRIT-RES or MM1.S cells (d) following treatment with CRIT in vivo. MM1.S stopped responding to CRIT by downregulating expression of VLA-4 subunit CD49d (Resistance = 28.12% CD49d+) (f) relative to parental cells injected into mice at the beginning of the experiment (parental = 99.92% CD49d+) (e), resulting in reduced binding of the VLA-4-targeting ligand LLP2A on resistant cells (h) (LLP2A+ = 6.6%) compared to parental MM1.S (g) (LLP2A+= 84.15%). i No significant difference in colony-forming units of progenitor stem cells was observed between untreated, control and treated mice. j To determine if CRIT reduced engraftment of haematologic cells in vivo, we assessed BM repopulation following CRIT treatment. Bone marrow from treated mice or PBS-treated controls were mixed with congenic B6.CD45.1/2 at a ratio of 1:1 before infusion of 1 × 10 total BM cells into lethally irradiated (TBI 1100 cGy) B6.CD45.1 recipients. k Percentage of cells derived from treated donor BM (CD45.2) were calculated as a percentage of total donor BM (CD45.2 + CD45.1/2). BM from CRIT-treated mice effectively repopulated recipients (n = 5 per group) 8 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | 6 ×10 ×10 ×10 NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE ab c d Liver 200 Lung 400 200 Blood * Lower limbs 300 150 Kidney Spleen 200 100 Lower limb Spine 100 50 136 24 Naive PyMT-B01 –50 Time (h) Naive PyMT-B01 Fig. 5 Dissemination of metastatic breast cancer cells and biodistribution of TC in mice. a In vivo BLI non-tumour-bearing C57BL/6J mice. b In vivo BLI of metastatic breast cancer 10 days post intracardiac injection of PyMT-BO1 GFP/Luc in C57BL/6J mice. c Ex vivo BLI of metastatic tumour burden 10 days post tumour initiation. The left panel are tissues obtained from a mouse and the right panel are tissues obtained from b mouse. d Inductively coupled plasma mass spectrometry analysis of Ti content in blood samples and the lower limbs, where tumour burden is high. The Ti content was background- corrected from untreated mice; *P < 0.05. Studies were performed with n = 5 mice per each group ab c d e 2×10 Untreated HSA-TC 08 FDG 1×10 CRIT 5×10 Untreated HSA-TC FDG CRIT Fig. 6 Representative BLI of PyMT-BO1 GFP/Luc metastatic breast cancer cells in C57B6. a Untreated C57B6 mouse bearing highly metastatic PyMT-BO1 −1 cancer. Accumulation in the lower limbs were predominant. b Mouse treated with 30 mg kg of HSA-TC nanoparticles. c Mouse treated with 800 µCi of 18 18 FDG. d Mouse treated with a combination of HSA-TC and 800 µCi FDG. e Quantification of whole-body luminescence in CRIT-treated mice compared 18 18 to untreated, HSA-TC treated or FDG-treated controls (*P values are 0.038, 0.23 and 0.017 for CRIT, HSA-TC alone and FDG alone, respectively). BLI and data analysis were performed on day 9 after initiation of PyMT-BO1 metastasis in mice. Studies were performed with n = 5 mice per each group reliable survival plots. Future studies will explore different models association of both therapeutic components in vital organs. These and establish the mechanism of therapy resistance in this cancer findings expand the potential use of PT for treating previously cell line. PT-inaccessible metastatic, infectious and cardiovascular diseases. In summary, we have successfully demonstrated the applica- tion of PT for treating disseminated malignancies using VLA-4- Methods targeted NM and HSA-TC nanoparticles, activated by radio- Synthesis and characterization of VLA-4-targeted titanocene micelles. The pharmaceuticals. Integral to this strategy is the availability of a VLA-4 ligand, LLP2A (Supplementary Figure 2a), was prepared on solid support wide range of radionuclides for clinical PET imaging and using standard fluorenylmethyloxycarbonyl (Fmoc) peptide synthesis as reported preclinical Cerenkov luminescence imaging to further monitor previously . Starting with Rink Amide resin, serial Fmoc deprotection cycles were 39,40 and guide treatment response . We demonstrated a strategy to achieved with 20% piperidine in dimethylformamide (DMF) and coupling of amino acids was performed with hydroxy-benzotriazole (HOBt) and 1,3-diiso- rescue abandoned light-sensitive drugs with poor therapeutic propylcarbodiimide (DIC) in DMF at 25 °C. The crude product was cleaved from outcomes such as TC and some FDA-approved drugs with the resin with a mixture of 95% trifluoroacetic acid (TFA): 2.5% water: 2.5% inherent photoactivity into precision phototherapeutics. In triisopropylsilane and precipitated with cold diethyl ether. The product was pur- addition, clinical biochemistry parameters and histopathologic ified by RP-HPLC and characterized by analytical HPLC and ES-MS: calculated mass for LLP2A is 1084.27 Da; observed mass is 1085 Da (Supplementary Fig- assessment of vital organs in the treated and untreated controls ure 2b). To incorporate LLP2A into nanomicelles, LLP2A was dissolved in ethanol were similar (Supplementary Figure 6 and Supplementary and mixed with 2-iminothiolane in methanol. After reacting for 2 h at 25 °C, Figure 7). The brain, heart, liver and kidneys were of particular polyethylene glycol -phosphatidylethanolamine (PEG-PE) was added to the interest because FDG naturally accumulates in these organs mixture and incubated for another 2 h. The product was purified using 3000 Da MWCO dialysis tubing to dialyse off the free LLP2A, and lyophilized to give a because of their high glucose utilization and elimination white solid (LLP2A-PEG-PE). The product was characterized by analytical HPLC pathways. The absence of off-target toxicity to normal hemato- and ES-MS: average calculated mass for LLP2A-PEG-PE is about 4100 Da; poietic stem cells may favour the translation of this approach in observed mass is about 4065 Da (Supplementary Figure 2c). The phospholipid/ the clinic as either a standalone therapy or as a combination with polysorbate 80 micelles were prepared as a microfluidized suspension comprising other therapies, including chemotherapy, where the suppression 20% polysorbate Tween 80 (v/v), a 2.0% (w/v) surfactant commixture and 1.7% (w/ v) glycerine in filtered MilliQ Nanopure water. The surfactant co-mixture were 0 or of the bone marrow and the risk of pancytopenia may not be 2 mole% TC and 0 or 0.15 mole% of LLP2A-PEG-PE, with the remainder as anticipated to be greater than those patients receiving che- phosphatidylcholine (>98% purity, NOF America). The surfactant components motherapy without CRIT. Our results suggest that the sequential were dried from organic solvent into a film, resuspended in nanopure water, and administration of the NM-TC and radionuclide minimizes the combined with polysorbate 80 and glycerine mixtures, followed by sonication at 4 ° NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 9 | | | Untreated HSA-TC FDG HSA-TC+FDG Radiance (p/s/cm /sr) Radiance (p/s/cm /sr) Radiance (p/s/cm /sr) BLI photon flux (photons/s) Titanium content (µg/g) ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 C for 3 min and then microfluidized (LV1, Microfluidics, Inc) at 20,000 psi for five and snared with two ligatures, one proximal one distal, for introducing the passes. The micelles were filtered with a 0.2 µm filter into sterile serum vials, catheter. The wound is wetted periodically with lidocaine (2%)/bupivicaine (2.5%) preserved under inert gas, capped and crimp-sealed before storage at 4 °C for and saline to maintain analgesia. The distal tie is tightened to ligate the artery, and subsequent use. The concentration of TC in TC nanomicelles was determined by the proximal tie is made into a Potter’s knot. A silastic catheter is inserted through inductively coupled plasma optical emission spectrometry (PerkinElmer Optima an arteriotomy and secured in place with the ligatures. The catheter is then flushed 5300). For each measurement, 20 µL reconstituted TC nanomicelle (n = 3) was with heparinized saline. Blood is sampled via a carotid catheter. Blood samples digested with concentrated nitric acid in Mars 6 Microwave Digestion System were ~150 μL and flushed with equivolume of saline. The blood samples were (CEM Corporation). Microwave power was ramped up to 200 °C for 20 min fol- digested by microwave digestion in HNO /H O under pressure and the con- 3 2 2 −1 lowed by a hold for 20 min at the same temperature. All external calibration centration of TC (mg mL ) was determined by inductively coupled plasma optical −1 standards of 0.01, 0.1, 1.0, 10, 50, 100 and 200 µgL were within 10% of expected emission spectrometry and reported as percentage of injected dose per gram of −1 concentration. Relative standard deviation measurements were less than 5% for all tissue (ID g ). standards and the samples. Dynamic light scattering measurements were acquired with a Zeta Plus Zeta Potential Analyzer (Brookhaven Instruments Corporation, Holtsville, NY) Biodistribution of NM-TC in mice. The NM-TC at a dose of 50 μL per mouse equipped with a 633 nm laser. Three measurements were conducted in deionized were administered intravenously in MM1.S (n = 5) tumour-bearing mice. A one- water for each sample with at least 10 runs and each run lasting 10 s. All sizes time dose was administered 30 days post cell injection of MM1.S cells. Tf-TC and reported were based on intensity average. For negative staining and electron MKT4 were prepared as described in the literature . Briefly, MKT4 was prepared microscopy analysis of NM, samples were allowed to absorb onto Formvar/carbon- by adding fivefold molar excess of mannitol (500 mg) to TC (100 mg) and a 19-fold coated copper grids for 10 min. Grids were washed two times for 1 min each in molar excess of sodium chloride (450 mg) in 50 mL water. The solution was mixed dH O and stained with 1% aqueous uranyl acetate (Ted Pella Inc., Redding CA) for in a shaker at 25 °C for 4 h before lyophilizing. For intravenous injections, MKT4 −1 1 min. Excess liquid was gently wicked off and grids were allowed to air dry. was reconstituted in saline at a dose of 0.25 mg kg . To prepare Tf-TC, fivefold Samples were viewed on a JEOL 1200EX transmission electron microscope (JEOL molar excess of TC was added to human apo-Tf and incubated in a shaker for 2 h USA, Peabody, MA) equipped with an AMT 8-megapixel digital camera at 25 °C. A working stock of TC was initially prepared in DMSO due to low (Advanced Microscopy Techniques, Woburn, MA). solubility of TC in water and aqueous buffers. The mixture was then dialysed overnight against DPBS using a 3000 Da molecular weight cutoff (MWCO) Slide- A-Lyzer MINI Dialysis Devices to remove excess TC. MKT4 and Tf-TC were Synthesis and characterization of HSA-TC. TC (80 mg) was added in an aqueous −1 administered intravenously at a dose of 0.25 mg kg (normalized to the TC solution (16 mL) containing 0.5% HSA. The mixture was shaken moderately at 560 content in NM) in MM1.s-SCID mice (n = 5). The mice were then killed 90 min oscillation per min in IKA KS 130 basic plate shaker for 6 h at room temperature, post injection and the organs were dissected. The organs and blood were processed followed by lyophilization in Thermo Fischer SAVANT RVT5105 refrigerated by microwave digestion in HNO /H O under pressure and the TC measurements 3 2 2 vapour trap lyophilizer to obtain the HSA-TC as dry powder. The lyophilized −1 were measured in mg mL by inductively coupled plasma optical emission spec- product was reconstructed in 0.9% saline immediately before use. The con- −1 −1 trometry and adjusted to %ID g tissue. centration of the TC in the reconstituted injection was 5.6 mg mL (characterized by inductively coupled plasma optical emission spectrometry and that of HSA in −1 the reconstituted solution was 8.95 mg mL (determined by Bio-Rad Quick Start Biodistribution of HSA-TC in mice. A similar procedure described above was used Bradford Protein Assay kit). The size and dispersity of the HSA-TC was confirmed −1 in the HSA-TC study. HSA-TC (50 µL per mice, 30 mg kg ) was injected intra- by DLS (Malvern Zetasizer nano Series) and electron microscopy (JEOL TEM-1400 venously in C57B6 mice on day 10 post intracardiac injection of PyMT-BO1 GFP/ electron microscope). Luc breast cancer cells. In addition to the untreated group, each group of five mice were killed at 1, 3, 6 and 24 h post injection after BLI of the living animal. The Cell culture. All cell lines underwent STR profiling and tested for mycoplasma organs and blood were harvested and imaged ex vivo using BLI to estimate the contamination. MM1.S-Luc cells were cultured in RPMI1640 medium containing tumour distribution. To determine uptake of the Ti distribution, blood and lower 10% heat-inactivated FBS and 2 mercaptoethanol (50 μM final) and 1× of all of the limbs at the different time points were were digested by microwave in HNO /H O 3 2 2 −1 following: penicillin/streptomycin (100 μgmL final), sodium pyruvate (1 mM under pressure and analysed by inductively coupled plasma mass spectrometry. −1 final), non-essential amino acids, HEPES (10 mM final) and L-glutamine. A similar The Ti content in each tissue were measured in mg mL after subtraction of condition was used to culture PyMT-BO1 cells in DMEM media plus 10% FBS. background Ti from the untreated cohorts. Tumour models in mice. All animal studies were conducted in compliance with 18 18 FDG-PET imaging. FDG-PET imaging was performed on MM mice 30 days the guidelines established by the Animal Studies Committee at Washington Uni- post cell injection (n = 4) along with naive mice (n = 4). The mice were fasted for 6 versity in St. Louis, Missouri. Fox Chase SCID Beige mice (4-week, female) were h before each scan. After anesthetizing the mice with 1.5–2% isoflurane and oxy- purchased from Charles River laboratories for developing the disseminated MM −1 gen, 0.19 mCi (7.03 MBq) 0.1 mL of FDG was administered intravenously. A 10- model. C57BL/6J mice (6-week, female) were purchased from the Jackson min transition scan was performed just before the 10 min emission at 1 h post Laboratory for developing the metastatic breast cancer model. MM1.S-Luc cells injection using a MicroPET-Inveon MultiModality scanner (Siemens Preclinical (1 × 10 in 100 µL per mouse) were injected intravenously for the MM model and Solutions, Erlangen, Germany). The animals were placed on the microCT in the PyMT-BO1-GFP-Luc cells (1 × 10 in 50 µL per mouse) were injected in the left same position to obtain anatomical imaging and co-registered to the microPET ventricular chamber for the breast cancer model. BLI was used to monitor cell image. The data were analysed using Inveon Research Workstation software, by viability and tumour burden. manually drawing three-dimensional ROI from PET images using CT anatomical guidelines. The activity associated with tumour was measured and maximum SUVs Bioluminescence imaging. Ex vivo and in vivo bioluminescence imaging of MM1. −1 were calculated using SUV = ([mCi mL ] x [animal weight (g)]/[injected dose S-Luc in SCID mice and PyMT-BO1 GFP/Luc in C57BL/6J was performed on an (mCi)]). IVIS Lumina (PerkinElmer, Waltham, MA; Living Image 4.3, 5 min to 1 s exposure, bin2–8, FOV12.5 cm, f/stop1, open filter). Image analysis was performed using Living Image 2.6 software. For in vitro imaging, optimal bioluminescence for In vivo CRIT in disseminated MM1.S-luc/SCID mouse model. After MM1.S cell subsequent in vivo studies was determined by plating 1 × 10 cells in a black 24- injection, considered day 1, mice were imaged weekly using BLI for 7–9 weeks. well plate. Fixed regions of interest (ROIs) were drawn on each well and images Treatment was initiated on day 5. Mice were administered 50 μL of NM-TC (0.25 8 −1 −2 −1 captured after an exposure time of 10 s. A radiance >1× 10 photons s cm sr −1 −1 18 mg kg TC) intravenously followed by 29.6 MBq 0.1 mL of FDG 90 min later, was considered as the threshold. For in vivo imaging, mice were injected intra- also administered intravenously (n = 15). Control mice (n = 5 per group) were −1 peritoneally with D-luciferin (150 mg kg in PBS; Gold Biotechnology, St. Louis, administered with NM-TC or FDG alone. A total of four treatment cycles at an MO) and imaged after anaesthetizing with isoflurane (2% vaporized in O ). Total interval of 1 week were given per animal, where a cycle refers to an administration −1 −2 −1 photon flux (photons s cm sr ) was measured from fixed ROIs over the entire of NM-TC and FDG. BLI and survival of these groups of mice were tracked along mouse. with untreated controls (n = 10) on a weekly basis. Food was withheld from mice for 6 h before administering FDG and kept in a dark, lead-shielded room post Pharmacokinetics of NM-TC in rats. For pharmacokinetic analysis, the NM at a injection. The weight and any physical signs of distress were also monitored closely. −1 dose of 25 μLkg was administered intravenously to rats (n = 5) and serial blood The mice were killed by cervical dislocation after anaesthesia with 5% isoflurane samples were drawn at 1, 5, 15, 30, 60, 90, 120 and 1440 min for metal analysis. The when there was a loss of >20% total body weight. For rechallenging studies on CRIT-RES PK in rats is done to allow enough sample for serial measures in the same animal. resistant MM1 cells (MM1 ), the bone marrow (BM) was harvested from 4 −1 −2 −1 Otherwise, blood samples must be pooled, masking inter-specimen variability. treated mice when whole-body radiance exceeded 5 × 10 photons s cm sr . Briefly, animals are weighed, anesthetized and shaved on the ventral side of the BM was harvested by flushing the femoral shaft with PBS and reinjecting the cells neck. The surgical area is cleaned and wiped free of hair before surgery. Following into a fresh set of Fox Chase SCID Beige mice. The treatment schedule followed stabilization, the incision site is anesthetized, and an incision made just off midline was similar to MM1.S described above. BLI was used to track and compare to the trachea through the skin. The carotid artery is exposed by blunt dissection treatment response in a CRIT group (n = 10) and an untreated group (n = 10). 10 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | | NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 ARTICLE In vivo CRIT in metastatic breast cancer PyMT-BO1 model. A similar procedure Data availability. The data that support the findings of this study are available described above was followed in the HSA-TC CRIT study. On day 2 post intra- from the corresponding author on reasonable request. cardiac injection of PyMT-BO1-GFP-Luc cells in C57BL/6J mice, BLI was used to confirm tumour viability and engraftment. The mice were stratified into the fol- Received: 23 May 2017 Accepted: 21 December 2017 lowing groups: untreated control, HSA-TC only, FDG only and combination of HSA-TC and FDG cohorts. Each group has five mice weighing between 18 and 20 g. All mice were fasted 3 h before commencing the treatment on day 2. HSA-TC −1 (100 µL per mice, 30 mg kg was injected intravenously in the HSA-TC and HSA- TC + FDG groups. This was followed by intraperitoneal (i.p.) injection of the 18 18 FDG (800 µCi in 50 µL saline containing 0.02% ethanol) in the FDG-only group and the HSA-TC + F FDG group after 2 h from the drug injection. This References treatment regimen was performed every 2 day for 3 times and BLI images were 1. Steeg, P. S. Targeting metastasis. Nat. Rev. Cancer 16, 201–218 (2016). captured first on each occasion before the therapy. Tumour response was quan- 2. Lee, D. W. et al. Current concepts in the diagnosis and management of cytokine tified using both whole body and then lower limb signals from BLI. release syndrome. Blood 124, 188–195 (2014). 3. Sipkins, D. A. et al. In vivo imaging of specialized bone marrow endothelial microdomains for tumour engraftment. Nature 435, 969–973 RNA analysis. Bone marrow was flushed from femurs using cold PBS and single (2005). cell suspensions were prepared by filtering through a 40 μm nylon strainer. Red 4. Hasenclever, D. & Diehl, V. A prognostic score for advanced Hodgkin’s disease. blood cells were lysed using Red Blood Cell Lysing Buffer (Sigma-Aldrich) and International prognostic factors project on advanced Hodgkin’s disease. N. CRIT-RES washed with cold PBS. MM1 cells were sorted by fluorescent-activated cell Engl. J. Med. 339, 1506–1514 (1998). CRIT-RES sorting for GFP. RNA was extracted from MM1 cells using the Qiagen 5. De Angulo, G., Yuen, C., Palla, S. L., Anderson, P. M. & Zweidler-McKay, P. A. RNAeasy mini kit according to the manufacturer’s instructions. cDNA was syn- Absolute lymphocyte count is a novel prognostic indicator in ALL and AML: thesized using qRT-PCR was performed on the Applied Biosystems StepOnePlus implications for risk stratification and future studies. Cancer 112, 407–415 Real-Time System (Thermo Fisher) using pre-designed TaqMan Gene Expression (2008). Assays (Life Technologies; 18S RNA Hs99999901 and glut1 Hs00892681) 6. Wilcox, R. A. et al. The absolute monocyte and lymphocyte prognostic score according to the manufacturer’s instructions. The ΔΔCT method was used to predicts survival and identifies high-risk patients in diffuse large-B-cell calculate changes in fold expression and results were analysed with GraphPad lymphoma. Leukemia 25, 1502–1509 (2011). Prism 6. 7. Ege, H. et al. Prediction of survival using absolute lymphocyte count for newly diagnosed patients with multiple myeloma: a retrospective study. Br. J. Haematol. 141, 792–798 (2008). Flow cytometry. Red blood cells in the flushed BM were lysed using Red Blood 8. Spring, B. Q. et al. A photoactivable multi-inhibitor nanoliposome for tumour Cell Lysing Buffer (Sigma-Aldrich) and washed with cold PBS. Cell subsets were control and simultaneous inhibition of treatment escape pathways. Nat. evaluated by flow cytometry by re-suspending the cells in staining buffer (PBS Nanotechnol. 11, 378–387 (2016). supplemented with 0.5% bovine serum albumin and 2 mM EDTA) and incubating 9. Shaffer, T. M., Pratt, E. C. & Grimm, J. Utilizing the power of Cerenkov light for 30 min at 4 °C with pre-titrated saturating dilutions of the following with nanotechnology. Nat. Nanotechnol. 12, 106–117 (2017). fluorochrome-labelled monoclonal antibodies: CD49d-PE/Dazzle594 (clone 9F10; 10. Hartl, B. A., Hirschberg, H., Marcu, L. & Cherry, S. R. Activating photodynamic Biolegend), CD29-APC (clone TS2/16; Biolegend), human CD45-Alexa Fluor 700 therapy in vitro with cerenkov radiation generated from yttrium-90. J. Environ. (clone HI30; Biolegend), CD138-BV421 (clone B-B4; Biolegend) and murine Pathol. Toxicol. Oncol. 35, 185–192 (2016). CD45-BV510 (clone 30-F11; BD Biosciences). To measure LLP2A binding, cells 11. Kotagiri, N., Sudlow, G. P., Akers, W. J. & Achilefu, S. Breaking the were washed and resuspended in Hanks Balanced Salt Solution (HBSS; Lonza) 2+ 2+ depth dependency of phototherapy with Cerenkov radiation and low-radiance- containing Ca (1.26 mM), Mg (0.81 mM) and 0.1% bovine serum albumin responsive nanophotosensitizers. Nat. Nanotechnol. 10, 370–379 (Sigma-Aldrich). Aliquots of cells were left untreated or treated with 10 nM BIO5192 (Tocris) at 4 °C for 15 min. Samples were then incubated with or without (2015). −1 LLP2A-Cy5 (0.41 µg µL ) for 30 min at 4 °C and washed twice with HBSS.1 Dead 12. Kamkaew, A. et al. Cerenkov radiation induced photodynamic therapy using −1 cells were excluded from these assays by staining with 2 mg mL 7-amino- chlorin e6-loaded hollow mesoporous silica nanoparticles. ACS Appl. Mater. actinomycin D (Molecular Probes, Eugene, OR) for 5 min before analysis. Samples Interfaces 8, 26630–26637 (2016). were analysed on a Beckman Coulter Gallios flow cytometer and data were ana- 13. Pantel, K., Alix-Panabieres, C. & Riethdorf, S. Cancer micrometastases. Nat. lysed using FlowJo software (TreeStar, Ashland, OR). Statistical comparisons of Rev. Clin. Oncol. 6, 339–351 (2009). flow cytometry data were performed using an unpaired two-tailed Student's t test 14. Ghobrial, I. M. Myeloma as a model for the process of metastasis: implications (GraphPad Prism). P values ≤0.05 were considered significant. for therapy. Blood 120,20–30 (2012). 15. Su, X. et al. Antagonizing Integrin beta3 Increases Immunosuppression in Cancer. Cancer Res. 76, 3484–3495 (2016). Colony-forming unit assay. For enumeration of blood colony-forming cells 16. Kröger, N., Kleeberg, U. R., Mross, K., Edler, L. & Hossfeld, D. K. Phase II (CFU), aliquots of bone marrow cells were incubated in duplicate in commercially clinical trial of titanocene dichloride in patients with metastatic breast cancer. available growth factor-supplemented methylcellulose medium for mouse CFU-C Oncol. Res. Treat. 23,60–62 (2000). (Stem Cell Technologies, Vancouver, BC) as described. Blood CFU were enum- 17. Guo, M., Sun, H., McArdle, H. J., Gambling, L. & Sadler, P. J. Ti(IV) uptake and erated after 6–8 days of culture. release by human serum transferrin and recognition of Ti(IV)-transferrin by cancer cells: understanding the mechanism of action of the anticancer drug titanocene dichloride. Biochemistry 39, 10023–10033 (2000). Competitive bone marrow transplant. C57BL/6 mice (B6.CD45.2) were treated 18. Buettner, K. M., Snoeberger, R. C. 3rd, Batista, V. S. & Valentine, A. M. with either CRIT (n = 5) or PBS control (n = 5). BM was harvested on day 40 by Pharmaceutical formulation affects titanocene transferrin interactions. Dalton flushing the femoral shaft with PBS. Bone marrow from treated mice was mixed Trans. 40, 9580–9588 (2011). with congenic BM (B6.CD45.1/2) at a ratio of 1:1 before infusion of 1 × 10 total 19. Girotti, A. W. Photosensitized oxidation of membrane lipids: reaction BM cells into lethally irradiated (TBI 1100 cGy) B6.CD45.1 recipients via lateral tail pathways, cytotoxic effects, and cytoprotective mechanisms. J. Photochem. vein injection. Peripheral blood was analysed by flow cytometry at week 8+ to Photobiol. B 63, 103–113 (2001). quantify engraftment of donor BM. Percentage of cells derived from treated 20. Kim, J., Santos, O. A. & Park, J.-H. Selective photosensitizer delivery into donor BM (CD45.2) was calculated as a percentage of total donor BM (CD45.2 + plasma membrane for effective photodynamic therapy. J. Control. Release 191, CD45.1/2). 98–104 (2014). 21. Torchilin, V. P. Recent advances with liposomes as pharmaceutical carriers. Nat. Rev. Drug Discov. 4, 145–160 (2005). Statistical analysis. Statistical significance was measured by Student’s t test using 22. Sarin, H. Physiologic upper limits of pore size of different blood capillary types GraphPad Prism software (GraphPad, San Diego, CA). Kaplan–Meier survival and another perspective on the dual pore theory of microvascular permeability. curves were plotted using GraphPad Prism software. Unless noted otherwise, all J. Angiogenesis Res. 2,1–19 (2010). values are means and error bars are standard deviations. For animal studies, sample 23. Soodgupta, D. et al. Small molecule MYC inhibitor conjugated to integrin- size estimates depend on the effect size (mean difference between untreated and targeted nanoparticles extends survival in a mouse model of disseminated treatment groups/SD) of the outcome. For effect size of 2.1 and using a two-sided t multiple myeloma. Mol. Cancer Ther. 14, 1286–1294 (2015). test, typically five per group were needed with 80% power to detect a significant 24. Olson, D. L., Burkly, L. C., Leone, D. R., Dolinski, B. M. & Lobb, R. R. Anti- difference at a type I error rate of 0.05. There were no exclusion criteria, except for alpha4 integrin monoclonal antibody inhibits multiple myeloma growth in a the breast cancer study where the use of female mice was needed to recapitulate the murine model. Mol. Cancer Ther. 4,91–99 (2005). histopathology of female breast cancer. There was no blinding of investigators in this research, including animal study. NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications 11 | | | ARTICLE NATURE COMMUNICATIONS | DOI: 10.1038/s41467-017-02758-9 25. Kiziltepe, T. et al. Rationally engineered nanoparticles target multiple myeloma Acknowledgements cells, overcome cell-adhesion-mediated drug resistance, and show enhanced Funding for this project was in part by grants from the US National Institutes of Health efficacy in vivo. Blood Cancer J. 2, e64 (2012). (U54CA199092, NIBIB R01 EB008111, NCI R01 CA194552, NCI R01 CA152329 and 26. Peng, L. et al. Combinatorial chemistry identifies high-affinity peptidomimetics NIH P50 CA094056), the Department of Defence Breast Cancer Research Program against [alpha]4[beta]1 integrin for in vivo tumor imaging. Nat. Chem. Biol. 2, (W81XWH-16-1-0286), and the Alvin J. Siteman Cancer Research Fund (11-FY16-01). 381–389 (2006). We thank the staff of Siteman Cancer Center Small Animal microPET Facility for 27. Korfel, A. et al. Phase I clinical and pharmacokinetic study of titanocene assistance with imaging and therapy studies; Wandy Beatty for assistance with the dichloride in adults with advanced solid tumors. Clin. Cancer Res. 4, 2701–2708 electron microscope imaging; Suellen Greco for clinical chemistry and H&E images; (1998). Patty Wurm for coupled plasma optical emission spectrometry analysis and Deepti 28. Mitchell, G. S., Gill, R. K., Boucher, D. L., Li, C. & Cherry, S. R. In vivo Soodgupta for flow cytometry analysis. Cerenkov luminescence imaging: a new tool for molecular imaging. Philos. Trans. A Math. Phys. Eng. Sci. 369, 4605–4619 (2011). Author contributions 29. Bredella, M. A., Steinbach, L., Caputo, G., Segall, G. & Hawkins, R. Value of N.K. and S.A. conceived study; N.K., M.L.C. and S.A. designed the research; N.K., M.L.C., FDG PET in the assessment of patients with multiple myeloma. Ajr. Am. J. M.R., J.P., G.S., L.M., K.W., C.C., L.L., L.H., C.E., P.K., M.Z. and X.S. performed the Roentgenol. 184, 1199–1204 (2005). research; G.C. and X.Y. generated the NM and performed pharmacokinetic and bio- 30. Cavo, M. et al. Role of 18F-FDG PET/CT in the diagnosis and management of distribution analysis; M.S., K.W., G.L. and S.A. contributed materials; S.A. supervised multiple myeloma and other plasma cell disorders: a consensus statement by overall study; N.K., M.L.C., M.T., G.L., J.D., C.E., P.K. and S.A. wrote and edited the the International Myeloma Working Group. Lancet Oncol. 18, e206–e217 manuscript. (2017). 31. Agool, A. et al. Radionuclide imaging of bone marrow disorders. Eur. J. Nucl. Med. Mol. Imaging 38, 166–178 (2011). Additional information 32. Soodgupta, D. et al. Ex vivo and in vivo evaluation of overexpressed VLA-4 in Supplementary Information accompanies this paper at https://doi.org/10.1038/s41467- multiple myeloma using LLP2A imaging agents. J. Nucl. Med. 57, 640–645 017-02758-9. (2016). 33. Ford, C. E., Hamerton, J. L., Barnes, D. W. & Loutit, J. F. Cytological Competing interests: The authors declare no competing financial interests. identification of radiation-chimaeras. Nature 177, 452–454 (1956). 34. Thota, R., Pauff, J. M. & Berlin, J. D. Treatment of metastatic pancreatic Reprints and permission information is available online at http://npg.nature.com/ adenocarcinoma: a review. Oncology 28,70–74 (2014). reprintsandpermissions/ 35. Al-Hajeili, M., Azmi, A. S. & Choi, M. Nab-paclitaxel: potential for the treatment of advanced pancreatic cancer. Onco Targets Ther. 7, 187–192 Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in (2014). published maps and institutional affiliations. 36. Kratz, F. Albumin as a drug carrier: design of prodrugs, drug conjugates and nanoparticles. J. Control. Release 132, 171–183 (2008). 37. Kim, C. et al. Comparison of the intraperitoneal, retroorbital and per oral routes for F-18 FDG administration as effective alternatives to intravenous Open Access This article is licensed under a Creative Commons administration in mouse tumor models using small animal PET/CT studies. Attribution 4.0 International License, which permits use, sharing, Nucl. Med Mol. Imaging 45, 169–176 (2011). adaptation, distribution and reproduction in any medium or format, as long as you give 38. Fueger, B. J. et al. Impact of animal handling on the results of 18F-FDG PET appropriate credit to the original author(s) and the source, provide a link to the Creative studies in mice. J. Nucl. Med. 47, 999–1006 (2006). Commons license, and indicate if changes were made. The images or other third party 39. Thorek, D. L., Ogirala, A., Beattie, B. J. & Grimm, J. Quantitative imaging of material in this article are included in the article’s Creative Commons license, unless disease signatures through radioactive decay signal conversion. Nat. Med. 19, indicated otherwise in a credit line to the material. If material is not included in the 1345–1350 (2013). article’s Creative Commons license and your intended use is not permitted by statutory 40. Bernhard, Y., Collin, B. & Decreau, R. A. Redshifted Cherenkov radiation for regulation or exceeds the permitted use, you will need to obtain permission directly from in vivo imaging: coupling Cherenkov radiation energy transfer to multiple the copyright holder. To view a copy of this license, visit http://creativecommons.org/ forster resonance energy transfers. Sci Rep 7, 45063 (2017). licenses/by/4.0/. 41. Pan, D. et al. A strategy for combating melanoma with oncogenic c-Myc inhibitors and targeted nanotherapy. Nanomedicine 10, 241–251 (2015). © The Author(s) 2018 12 NATURE COMMUNICATIONS (2018) 9:275 DOI: 10.1038/s41467-017-02758-9 www.nature.com/naturecommunications | | |

Journal

Nature Communications – Springer Journals

Published: Jan 18, 2018

There are no references for this article.