Ultrasmall Paramagnetic Iron Oxide Nanoprobe ... - ACS Publications

Subscriber access provided by University of Texas Libraries

Article

Ultra-small paramagnetic iron oxide nanoprobe targeting EGFR for in vivo MR imaging of hepatocellular carcinoma Yan Chen, Quan Zhou, Xue Li, Fa Wang, Kevin Heist, Rork D. Kuick, Scott Owens, and Thomas Wang Bioconjugate Chem., Just Accepted Manuscript • DOI: 10.1021/acs.bioconjchem.7b00501 • Publication Date (Web): 03 Oct 2017 Downloaded from http://pubs.acs.org on October 4, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Bioconjugate Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Bioconjugate Chemistry

Ultra-small paramagnetic iron oxide nanoprobe targeting EGFR for in vivo MR imaging of hepatocellular carcinoma Yan Chen1, Quan Zhou2, Xue Li1, Fa Wang1, Kevin Heist3, Rork Kuick4, Scott R. Owens5, Thomas D. Wang1,2,6* 1

Department of Internal Medicine, University of Michigan, Ann Arbor, MI 48109, United States

2

Department of Biomedical Engineering, University of Michigan, Ann Arbor, MI 48109, United

States 3

Department of Radiology, University of Michigan, Ann Arbor, MI

4

Department of Biostatistics, University of Michigan, Ann Arbor, MI 48109, United States

5

Department of Pathology, University of Michigan, Ann Arbor, MI 48109, United States

6

Department of Mechanical Engineering, University of Michigan, Ann Arbor, MI 48109, United

States

Corresponding Author *

Thomas D. Wang, M.D., Ph.D.

Professor of Medicine, Biomedical Engineering, and Mechanical Engineering H. Marvin Pollard Collegiate Professor of Endoscopy Research Division of Gastroenterology, University of Michigan 109 Zina Pitcher Pl. BSRB 1522 Ann Arbor, MI 48109-2200 Office: (734) 936-1228 Fax: (734) 647-7950 Email: [email protected]

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Hepatocellular carcinoma (HCC) is a common worldwide cancer that is rising rapidly in incidence. MRI is a powerful non-invasive imaging modality for HCC detection but lack of specific contrast agents limits visualization of small tumors. EGFR is frequently overexpressed in HCC and is a promising target. Peptides have fast binding kinetics, short circulatory half-life, low imaging background, high vascular permeability, and enhanced tissue diffusion for deep tumor penetration. We demonstrate a peptide specific for EGFR labeled with an ultra-small paramagnetic iron oxide (UPIO) nanoparticle with 3.5 nm dimensions to target HCC using T1weighted MRI. We modified the hydrophobic core with a water dispersible oleic acid coating and capped with PEGylated phospholipids DSPE-PEG and DSPE-PEG-Mal. The EGFR peptide is attached via thioether-mediated conjugation of a GGGSC linker to the maleimide-terminated phospholipids. On in vivo MR images of HCC xenograft tumors, we observed peak nanoprobe uptake at 2 hours post-injection followed by a rapid return to baseline by ∼24 hours. We measured significantly greater MR signal in tumor with the targeted nanoprobe versus scrambled peptide, blocked peptide, and Gadoteridol. Segmented regions on MR images support rapid renal clearance.

No significant difference in animal weight, necropsy, hematology, and

chemistry was found between treatment and control groups at one month post-injection. Our nanoprobe based on an EGFR specific peptide labeled with UPIO designed for high stability and biocompatibility showed rapid tumor uptake and systemic clearance to demonstrate safety and promise for clinical translation to detect early HCC.

Keywords: peptide, ultra-small, iron oxide nanoparticle, MRI, EGFR


Page 2 of 39

Page 3 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Hepatocellular carcinoma (HCC) is one of the most common cancers worldwide, and is responsible for >500,000 deaths per year.1 The global incidence is rising rapidly as a result of widespread chronic infection with hepatitis B and C and the increasing prevalence of fatty liver.2,3

Magnetic resonance (MR) imaging is sensitive to anatomic changes in tissue

microstructure and is accurate for detecting large tumors in cirrhotic livers.4 This imaging modality has excellent soft tissue contrast with sub-millimeter resolution, and can be used to monitor tumor size non-invasively over time. However, regenerative nodules are difficult to distinguish from HCC in cirrhotic livers,5 and small lesions 100 nm), they can be taken up by macrophages and sequestered in the reticuloendothelial (RES) system.10 Ultra-small paramagnetic iron oxide nanoparticles (UPIO) that are 66% of HCC tumors,15 and in a variety of other cancers.16 EGFR is also a common therapeutic target.17,18 We have previously developed a peptide that is specific for domain 2 of EGFR.19 This extracellular ligand-binding domain that can be developed for targeted imaging with MR. By comparison with bulky monoclonal antibodies, this 7 amino acid peptide is much smaller in size and lower in molecular weight. These properties provide improved pharmacokinetic properties for diagnostic imaging by reaching peak uptake within hours rather than days and result in short circulatory half-lives that reduce imaging background.20 Also, smaller dimensions allow for greater vascular permeability and enhanced tissue diffusion to produce deeper tumor penetration and better access to tumor targets.21,22 Here, we aim to evaluate the preclinical validity of QRH*-UPIO using HCC as a model by assessing the performance and safety of this nanoprobe to enhance T1weighted contrast in MR images and to demonstrate specific detection of HCC xenograft tumors in vivo.


Page 4 of 39

Page 5 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Results

Nanoprobe synthesis We synthesized the 7 amino acid peptide QRHKPRE (black), hereafter QRH*, for specific binding to EGFR, Fig. 1A. A GGGSC linker (blue) separates the peptide from the nanoparticle to prevent steric hindrance. The scrambled sequence PEHKRRQ, hereafter PEH*, was similarly conjugated for use as control. Mass spec analysis was performed to confirm the expected molecular weight for QRH* and PEH*, Fig. S1A,B. The PEGylated phospholipids DSPE-PEG and DSPE-PEG-Mal have hydrophobic and hydrophillic regions, Fig. 1B. We modified the surface of ultra-small iron oxide (UIO) nanoparticles with DSPE-PEG and DSPE-PEG-Mal by embedding the fatty acid chains into the hydrophobic part of oleic acid to stabilize and form PEGylated UPIO nanoparticles, Fig. 1C,D. QRH* is attached to the surface of UPIO via a thioether-mediated conjugation at the C-terminus with DSPE-PEG-Mal, hereafter QRH*-UPIO.

Fig. 1. Synthesis of EGFR targeted nanoprobe. A) EGFR specific peptide QRHKPRE (black) is attached to a GGGSC linker (blue).

B) The hydrophobic and hydrophillic regions of

PEGylated phospholipids DSPE-PEG and DSPE-PEG-Mal are shown. C) UIO nanoparticles



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 39

functionalized with oleic acid are coated with the PEGylated phospholipids to form D) UPIO and labeled with QRH* by reacting maleimide (Mal) with the cysteine in the linker to form E) QRH*-UPIO.

Nanoprobe characterization We used atomic force microscopy (AFM) to demonstrate minimal nanoparticle aggregation and to characterize QRH*-UPIO morphology, Fig. S2A. We measured a mean (±SD) diameter of 3.5±0.2 nm on height profile, Fig. S2B, and a mean zeta potential of ξ = -13.4±0.2 mV for UPIO and -37.5±4.3 mV for QRH*-UPIO. Because QRH* is negatively charged, this result supports successful coating of the UPIO surface with QRH*.

UIO nanoparticles were

synthesized and stabilized using oleic acid. AFM images and height profiles of UIO showed a much smaller mean size of 1.4±0.2 nm, Fig. S2C,D. We monitored the modification process using X-ray photoelectron spectroscopy (XPS), and found Fe and O to be uniformly distributed in UIO, Fig. S3A,B. The elements P and S were found abundantly in phospholipids and peptide, respectively, and were identified in UPIO, and QRH*-UPIO, respectively, Fig. S3C,D. We also identified characteristic stretching and bending modes for QRH*-UPIO and UIO using Fourier transform infrared (FTIR) spectra, Fig. S4A,B.

These results further support successful

conjugation of the QRH* peptide to UPIO.

Nanoprobe optimization We varied the molar ratio of DSPE-PEG-Mal to total phospholipid with surface densities of 0, 25%, 50%, 75%, and 100%, and determined the value that optimized binding of QRH*-UPIO to human SK-Hep1 HCC cells. We stained cells with Prussian blue and a nuclear fast red counterstain, Fig. 2A-F, and found that 25% DSPE-PEG-Mal provided the highest iron content,


Page 7 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 2G.

We used QRH*-UPIO and PEH*-UPIO with 25% density for all subsequent

experiments. We measured the longitudinal (r1) relaxivity with different ratios between UIO and DSPE-PEG to optimize QRH*-UPIO for use as a T1 MR contrast agent with Gadoteridol, a gadolinium-based contrast agent used clinically, as control.

We found that a ratio of 1:5

achieved a maximum relaxivity of r1 = 0.362 mM-1s-1 with correlation R2 = 0.99, Fig. 2H.

Fig. 2. Optimized targeting with EGFR peptide. Prussian blue stain show presence of iron in SK-Hep1 human HCC cells incubated with QRH*-UPIO synthesized using different percentages of DSPE-PEG-Mal, including A) control (no nanoparticle), B) 0%, C) 25%, D) 50%, D) 75%, and E) 100%. G) Quantified intensities show greatest binding with 25% DSPE-PEG-Mal. H) Ratio of 1:5 for UIO to DSPE-PEG achieves peak relaxivity of r1 = 0.362 mM-1/s-1 with R2 = 0.99. Final product at ratio of 1:20 was not stable. We measured the intensity for each ratio on 4 slides with 3 replicate measures per slide. We fit an ANOVA model to log-transformed data



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

with terms for 6 conditions and 4 slides, and compared pairs of conditions after averaging replicates.

Nanoprobe validation in vitro

Prussian blue stain We observed significantly greater iron accumulation with SK-Hep1 human HCC cells using QRH*-UPIO versus PEH*-UPIO, Fig. 3A,B. No iron was seen with SK-Hep1 cells alone (no nanoprobe), Fig. 3C. Minimal staining was seen with HepG2 human HCC cells for each condition, Fig. 3D-F. We quantified the iron content, and found a significantly greater mean value for QRH*-UPIO versus either PEH*-UPIO or control, Fig. 3G. Western blot shows EGFR expression for either cell, Fig 3H.


Page 8 of 39

Page 9 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 3. Specific binding of targeted nanoprobe in vitro. On Prussian blue stain, A) QRH*UPIO shows strong binding to cell surface (arrow) of SK-Hep1 human HCC cells by comparison with B) PEH*-UPIO and C) control (no nanoprobe). D-F) Minimal binding is seen with HepG2 cells. G) Quantified image intensities showed significantly greater binding for QRH*-UPIO than for either PEH*-UPIO or control. The difference between QRH*-UPIO and either control or PEH*-UPIO was significantly larger for SK-Hep1 than for the same differences using HepG2 cells. We fit an ANOVA model with terms for 6 groups and 4 slides per cell type to logtransformed data after averaging 3 replicate measures for each slide. H) Western blot shows EGFR expression level for each cell line. I) On MR images of SK-Hep1 cells, binding by



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 39

QRH*-UPIO produces increased intensity and decreased T1 relaxation time in comparison with PEH*-UPIOs, control (no nanoprobe), and QRH* blocked.

MR imaging of SK-Hep1 human HCC cells We used MR to further validate nanoprobe binding to SK-Hep1 cells. After 10 min for incubation, the mean T1 relaxation time (sec) for cells incubated with QRH*-UPIO was greatly shortened (9.1±2.3) compared with cells incubated with either PEH*-UPIO (13.5±8.7) or control (14.1±6.8), Fig 3I. After adding unlabeled QRH* to compete with QRH*-UPIO for binding to EGFR, the T1 relaxation time (13.8±5.7) was comparable with control. We incubated SK-Hep1 and HepG2 cells with increasing concentrations of QRH*-UPIO, and found that cell viability was only marginally affected to support low cytotoxicity, Fig. S5.

Nanoprobe validation with MRI in vivo We evaluated EGFR expression in human HCC xenograft tumors with T1-weighted MR images acquired at 0.5, 2, 4, and 24 hours after systemic administration of 5 mg/kg [Fe] QRH*UPIO, Fig. 4A-D. We observed peak tumor uptake of nanoprobe at 2 hours post-injection, Fig. 4E.

By comparison, PEH*-UPIO showed reduced signal from tumor.

We also injected

unlabeled QRH* (7 mM, 200 µL) 20 min prior to QRH*-UPIO to compete for binding to EGFR and found decreased signal. We used Gadoteridol for control. The mean value for QRH*-UPIO was significantly greater than that for PEH*-UPIO, QRH* block, and Gadoteridol at 2 hours post-injection, Fig. 4F.


Page 11 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 4. MR images of HCC xenograft tumor in vivo. T1-weighted images were collected before (pre) and at 0.5, 2, 4, and 24 hours after intravenous injection of A) QRH*-UPIO and B) PEH*-UPIO to evaluate uptake in HCC tumors (red ovals). C) Unlabeled QRH* was injected 20 min prior to QRH*-UPIO to compete for binding (block). D) Gadoteridol was used as control. E) Quantified intensities show peak uptake at 2 hours post-injection. F) Normalized intensity with QRH*-UPIO at 2 hours is significantly greater than that for PEH*-UPIO, QRH* (block), and Gadoteridol. We normalized the MR intensities by dividing each value by that at time 0



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 39

(pre-injection) for the same mouse. P-values were obtained using ANOVA model with terms for 16 means (4 time points for 4 groups) on log transformed data.

Immunohistochemistry After

completion

of

imaging,

we

euthanized

the

animals,

and

performed

immunohistochemistry (IHC) to validate EGFR expression in HCC tumor sections.

We

observed strong staining by HCC xenograft tumor cells (arrow), Fig. 5A. Histology (H&E) for HCC is shown, Fig. 5B. By comparison, only a few lightly stained hepatocytes (arrow) can be appreciated in normal liver, Fig. 5C. We also stained tissue sections with Prussian blue to detect the presence of QRH*-UPIO. At 2 hours post-injection, scattered iron can be seen in HCC and kidney (arrow), Fig. 5D,E. Normal liver showed minimal staining, Fig. 5F. No staining was seen in HCC, kidney, or liver at 30 days post-injection, Fig. 5G-I.


Page 13 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 5. Ex vivo validation. A) EGFR expression (arrow) was validated on section of SK-Hep1 HCC xenograft tumor after completion of imaging. B) Histology (H&E) shows a nest of tumor cells (arrow) with large irregular round nuclei and infiltrating blood vessels lined with flattened endothelial cells (arrowhead). C) IHC of normal liver shows faint staining on hepatocyte (arrow) surrounding central vein (arrowhead). On Prussian blue stain, the strong microscopic presence of iron (arrow) can be seen in D) HCC and E) kidney but only faintly in F) normal liver at 2 hours. G-I) At 30 days post-injection, no staining of iron is seen.

Nanoprobe Toxicity We evaluated the biodistribution of QRH*-UPIO following intravenous injection over time in non-tumor bearing mice using T1-weighted MR, Fig. 6A-D. We observed peak signal in liver, bladder, kidney, and spleen at 0.5 hours with return to baseline by ∼24 hours, Fig. 6E. Bladder and kidney produced significantly higher signal at 0.5 hours than either liver or spleen to support renal clearance. We examined the single dose toxicity of QRH*-UPIO by injecting 10mg/kg [Fe] of QRH*-UPIO in n = 5 tumor-free mice in either a treatment or control group. All mice appeared healthy by visual inspection, and no significant difference in the weight gain was observed between groups for >30 days post-injection, Fig S6. All labs for complete blood count and chemistries were in the normal range, Table S1. All mice survived to the scheduled necropsy. We euthanized the animals, and performed anatomic pathology on the heart, liver, spleen, lung, and kidney.

There were no signs of test article-related findings at terminal

necropsy in either the treatment or control group, Fig. S7A-J.



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 39

Fig. 6. Systemic clearance of nanoprobe. T1-weighted MR intensity on images of A) liver (arrow), B) bladder (circle), C) kidney (boxes), and D) spleen (chevron) in tumor-free mice after intravenous injection of QRH*-UPIO are shown over time. We observed peak MR signal at 0.5 hours post injection with return to baseline (pre-injection) by ∼24 hours. Bladder and kidney produced significantly higher signal than either liver or spleen at 0.5 and 2 hour. We normalized the MR intensities by dividing each value by that at time 0 (pre-injection) for the same mouse and tissue. We then fit an ANOVA model with terms for 16 means (4 tissues, at times 0.5, 2, 4, 24 hours) to log-transformed data. Discussion Here, we report the MR imaging performance of the QRH*-UPIO nanoprobe consisting of a peptide specific for EGFR and labeled with a ∼3.5 nm diameter UPIO. The small size and low molecular weight of this targeting ligand provides a number of pharmacokinetic advantages for detection of early HCC tumors.

We modified the nanoparticle surface with PEGylated

phospholipids to improve stability and prevent aggregation, and adjusted the ratio between iron oxide and phospholipid to achieve peak relaxivity for T1-weighted MR contrast. Our images in


Page 15 of 39

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


human HCC xenografts tumors support good retention of magnetic properties after surface modification. We observed peak uptake at 2 hours post-injection with low imaging background. This targeted nanoprobe produced significantly higher MR signal in tumor by comparison with scrambled peptide, blocked peptide, and Gadoteridol control. The scrambled peptide provides the same surface properties and size albeit with a different affinity to EGFR. The nanoprobe cleared within ∼24 hours, and showed minimal biodistribution outside of the tumor to suggest a safe toxicity profile. Our results demonstrate promise for a safe and effective T1 contrast agent that has potential to expand the clinical utility of MRI and improve precision diagnostics by targeting detection of early HCC. We used a simple strategy to synthesize the QRH*-UPIO nanoprobe for in vivo imaging with MRI. We coated the nanocore with oleic acid to stabilize the paramagnetic iron oxide,23 and then capped the surface with a monolayer of functional phospholipids (DSPE-PEG and DSPE-PEG-Mal) to mask the hydrophobic layer.24 By creating a hydrophilic surface, we were able to conjugate the EGFR peptide to the maleimide group at the end of DSPE-PEG-Mal for targeted detection of HCC. We carefully controlled the surface chemistry to reduce non-specific cell uptake by macrophages and to minimize nanoparticle aggregation as characterized by the monodisperse size distribution found with

Ultrasmall Paramagnetic Iron Oxide Nanoprobe ... - ACS Publications

Recommend Documents