Tetraethyl orthosilicate (TEOS), Hexadecyltrimethylammonium bromide (CTAB), Sodium carbonate, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT), Oxalic acid, Diethylene triaminepentaacetic acid (DTPA), Ethylenediaminetetraacetic acid (EDTA), Sodium hydroxide, 4,7,13,16,21,24-Hexaoxa-1,10-diazabicyclo[8.8.8] (K2.2.2), Potassium carbonate, Human serum, HEPES buffer solution (1 M in H2O), Ethanol (ACS grade), Methanol (Sigma-Aldrich, HPLC grade), and Ammonium hydroxide (28~30% in water). Deferoxamine mesylate (USP, DFO), Phosphate buffered saline (Gibco, PBS, pH 7.4), Trifluoroacetic acid (DAEJUNG, TFA), BCA (Bicinchoninic acid) assay kit, Sodium dodecyl sulfate (SDS)-polyacrylamide gel, Lithium dodecyl sulfate (LDS) buffer, Enhanced chemi-luminescence (ECL) kit, Polyvinylidene difluoride (PVDF) membrane (Thermo-Fisher), Dulbecco’s modified eagle medium (Thermo-Fisher, DMEM), Fetal bovine serum (Thermo-Fisher, FBS), and Zr-resin (TRISKEM), Yttrium metal foil 0.64 mm ~99.5% (Alfa Aesar), Hydrochloric acid trace metal grade (Thermo-Fisher) were obtained and used without further purification. All animal experiments were progressed according to institutional animal care and use committee (IACUC) guidelines provided by Korea Atomic Energy Research Institute.

Preparation of hollow mesoporous silica nanospehers

SNPs were synthesized by Stӧber method52. Briefly, 10 ml of ammonia solution and 6 ml of distilled water (D.W) were added in 250 ml of ethanol at 30 °C to which 40 ml of TEOS was added rapidly. Reactants were stirred constantly for two hours following which it was centrifuged and thoroughly washed with ethanol and water.

Mesoporous silica coated SNPs (SNP@mSNPs) were further synthesized using 0.15 g of CTAB and dissolved in 30 ml of water following which 30 ml of ethanol and ammonia solution were added. This solution was homogenised by stirring it for 30 min and 0.1 g of SNPs dispersed in 20 ml of distilled water was added. Next, 0.25 mL of TEOS was rapidly added and then stirred for 6 h. The reactant was centrifuged, washed and re-dispersed in 20 ml of water. Finally, HMSNs were synthesized by re-deposition method. The SNP@mSNPs dispersion was stirred for 12 h to which 0.46 g of sodium carbonate was added and stirred at 80 °C for 10 h after which it was suspended in mixture of methanol(16): HCl (1) solution and refluxed for 24 h at 80 °C. The product was centrifuged and washed by D.W.53.


NPs were recorded on a Perkin-Elmer FT-IR and prepared by classical technique KBr pellet. Removal of moisture and free CTAB from HMSNs were monitored thermogravimetrically (model 2950, TA Instruments, USA). Morphology and size of NPs were measured by scanning electron microscope (FE-SEM S-4200, Hitachi, Japan), transmission electron microscope (TEM, JEM 2011, Jeol, Japan) and bio-TEM (Tecnai G2 spirit Twin, FEI, USA). XRD pattern of the prepared HMSN powder was conducted using an X-ray diffractometer (Phillips X’pert MPD diffractometer, Almelo, Netherlands) with copper K-α target (40 kV, 30 mA). Data was collected at the 2-theta angle ranging from 3 to 90 with a scanning speed of 0.4 deg/min. To determine the hydrodynamic sizes and zeta potentials of SNPs, HMSN, Rm-SNPs and Rm-HMSN, as prepared suspensions were diluted (350 ppm) with D.W. The modified surface dependent hydrodynamic size and surface charge were determined using Zetasizer Nano ZS (Malvern instruments Ltd).

Characterization of cell membrane proteins

CD47 proteins on the Rm-HMSNs were quantified using Western blotting analysis. Characterization of the Rm proteins, emptied RBCs and Rm-HMSN were treated with LDS lysis buffer. Then, the samples were denatured at 85 °C for 2 min and proteins of Rm and Rm-HMSN were quantitated using BCA assay kit. 25 µg of each protein was added into each well in a 4–12% sodium dodecyl sulfate-polyacrylamide gel. The gel was run at 200 V for 22 min and polyacrylamide gel was stained by coomassie brilliant blue according to the provided protocol before imaging. After separation of protein by SDS-PAGE, proteins transferred onto a PVDF membrane. After blocking the membranes with milk blocking buffer, the membrane was incubated using CD47 primary antibody (1:500) and IgG secondary antibody (1:1,000) was incubated for 3 h. Expressed CD47 protein onto Rm and Rm-HMSN was detected for the luminescence using an ECL kit by Chemi-doc (iBright FL-1000 Imager, Thermo-Fisher Scientific)

Labelling efficiency of 89Zr-labelled SNP and HMSN

SNP and HMSN were dispersed in a HEPES buffer (pH 7.5; 0.1 M, 0.5 mL) at various concentrations (15, 50, 100, 200, 400 and 1,000 µg/mL). The pH of 89Zr-oxalate and 89Zr-chloride (~20 MBq) was adjusted to 7.5–8 using a Na2CO3 solution to which dispersed nanoparticles solution was added. After stirring at room-temperature for 24 hours, labelling efficiency was confirmed by radio thin-layer chromatography (radio-TLC) using 50 mM DTPA as the mobile phase29.

Red blood cell membrane camouflaged 89Zr-HMSN and 89Zr-SNPs

RBCs were isolated using the blood of CT-26 bearing female Balb/c mouse withdrawn by cardiac puncture (<10% of circulating blood volume, re-use the blood drawn from same mice to avoid immune barrier). Mouse blood was collected into the 500 μL syringe (24 G) containing 8 μL of 0.5 M EDTA and was centrifuged (1500 rpm, 5 min, 4 °C) to separate the serum, buffy coat, white blood cell and platelet layers sequentially. Separated RBCs were washed with ice cold 1 × PBS. The purified RBCs were hemolyzed by 0.25 × PBS treatment for extraction of Rm then washed by centrifugation (13,500 rpm, 5 min) with ice 1 × PBS. The Rm was re-suspended in 1 × PBS and sonicated for 3 min at power of 130 W. After sonication the Rm fragments were extruded (Avanti, mini-extruder) sequentially through 400 nm and 200 nm polycarbonate (PC) membrane filters. The prepared Rm and 89Zr labelled NPs were mixed and extruded by 200 nm of PC membrane for at least 10 passes. Rm coated 89Zr-NPs were centrifuged and washed to remove uncoated Rm.

Cell culture and animal tumor modelling

All cancer cell lines (CT-26, KB and A549) were cultured in DMEM and supplemented with 10% fetal bovine serum. Cell lines were maintained at 37 °C in a humidified atmosphere containing 5% CO2. CT-26 transplanted mice were prepared with female Balb/c mice. The cancer cell lines (5 × 106 cells) were injected in the right thigh. The tumors were allowed to grow for 2–3 weeks before ex vivo and in vivo animal experiment.

In vitro; cell viability

1 × 104 cells/well were cultured in micro well plates and exposed to different weight of nanoparticles (0–50 µg/mL) and added 10 µL of MTT reagent to each well, including controls and incubated for 48 hr at 37 °C until a purple coloured formazan. It was dissolved in 100 µL of detergent to all wells. Then, supernatant was measured at 570 nm using a microplate reader model.

In vitro; cellular uptake and internalization

The cellular uptake and internalization of Rm-89Zr-HMSNs was measured with CT-26, KB and A549 cells. The cancer cells were sub-cultured in 24 well plates (5 × 105 cells/well) and incubated for 0.5, 1, 4, 24 and 48 hours. After incubation, the supernatant was removed and the cell pellets were washed by cold PBS to remove the unbound Rm-89Zr-HMSNs. Trypsin solution was used to harvest NPs from the internalized cancer cells. Following this, the cancer cell pellets were treated with 0.1 M sodium citrate for 5 min to remove surface bound Rm-89Zr-HMSNs. The radioactivity of unbounded Rm-89Zr-HMSNs and cells were measured by gamma counter and results were calculated as the accumulation ratio (%) ± standard deviation (s.d.).

Ex vivo; biodistribution

Biodistribution studies were performed in compliance with the animal experimental guidelines and ethics approved by Korea Atomic Energy Research Institute (IACUC-2015-004). For biodistribution studies using the Rm-89Zr-HMSN, 1.85~2.59 MBq of Rm-89Zr-HMSN was injected by tail i.v in mice (n = 3 for each groups). Mice were dissected at defined time points: 1, 24, 48, 72 and 96 hours after i.v injection and the blood was collected from the heart and other organs viz. heart, lung, liver, spleen, stomach, intestine, pancreas, kidney, muscle, fat, bone, skin, tail, brain and tumor. The collected blood and organs were counted for accumulated Rm-89Zr-HMSN per organ weight relative to the injected Rm-89Zr-HMSN by measuring the radioactivity with a gamma counter. The results of biodistribution were expressed as the percentage of injected dose per gram (% ID/g).

In vivo; small animal PET images

CT-26 bearing mice were used to identify pharmacokinetic pathways and tumor accumulation. The mice were anesthetized by exposing it to 1.5~2% isoflurane in oxygen. PET images were obtained at 1, 2, 24 and 48, 72 and 144 h of interval after intravenous (i.v) injection through tail vein in mice with of 18F-HMSN, 68Ga-HMSN, Rm-89Zr-SNPs, 89Zr-HMSN and Rm-89Zr-HMSN.

In vivo; blood clearance

Once under anesthesia, Normal Balb/c mice (n = 3 per group) were injected intravenously via the tail vein with ~3.7 MBq 89Zr-HMSNs and Rm-89Zr-HMSNs. Blood samples (100 μL) were collected from the saphenous vein at a series of time points (1, 4, 24, 48 and 54 hour) into a γ-counting tube. Blood radioactivity was measured in a γ-counter and expressed as a percentage of the injected activity per mL (% IA/mL). The blood radioactivity at each time point was calculated including the decay correction.