Magnetic Iron Nanocubes Effectively Capture Epithelial and Mesenchymal Cancer Cells
Dhananjay Suresh, Shreya Ghoshdastidar, Abilash Gangula, Soumavo Mukherjee, Anandhi Upendran, Raghuraman Kannan
MMagnetic Iron Nanocubes E ff ectively Capture Epithelial andMesenchymal Cancer Cells Dhananjay Suresh, ⊥ Shreya Ghoshdastidar, ⊥ Abilash Gangula, Soumavo Mukherjee, Anandhi Upendran,and Raghuraman Kannan * Cite This:
ACS Omega − Read Online
ACCESS
Metrics & More Article Recommendations * s ı Supporting Information
ABSTRACT:
Current methods for capturing circulating tumorcells (CTCs) are based on the overexpression of cytokeratin (CK)or epithelial cell-adhesion molecule (EpCAM) on cancer cells.However, during the process of metastasis, tumor cells undergoepithelial-to-mesenchymal transition (EMT) that can lead to theloss of CK/EpCAM expression. Therefore, it is vital to develop acapturing technique independent of CK/EpCAM expression onthe cancer cell. To develop this technique, it is important toidentify common secondary oncogenic markers overexpressed ontumor cells before and after EMT. We analyzed the biomarkerexpression levels in tumor cells, before and after EMT, and foundtwo common proteins human epidermal growth factor receptor 2(Her2) and epidermal growth factor receptor (EGFR) whose levels remained una ff ected. So, we synthesized immunomagnetic ironnanocubes covalently conjugated with antibodies of Her2 or EGFR to capture cancer cells irrespective of the EMT status. Thenanocubes showed high speci fi city (6 − ’ blood to develop an e ff ectivetreatment plan. ■ INTRODUCTION
Isolation of circulating tumor cells (CTCs) from the blood ofcancer patients and analyzing them enables the clinician topredict the disease status, drug resistance, and the selection ofappropriate therapy. Food and Drug Administration (FDA)-approved CellSearch is currently used for the detection ofCTCs in a variety of metastatic tumor types to predict theoverall survival and progression-free survival in patients. Thissystem utilizes magnetic microbeads coated with an antibody(Ab) speci fi c to epithelial cell-adhesion molecule (EpCAM)for the enrichment of CTCs from the patient ’ s blood. Eventhough these magnetic beads are bene fi cial and widely used inclinics, the system has several drawbacks. Notably, thesystem detects only EpCAM-positive CTCs and fails tocapture tumor cells with no epithelial markers. The tumor cells lose cytokeratin (CK) and EpCAM whileundergoing epithelial-to-mesenchymal transition (EMT), aprocess that occurs during metastases.
In fact, the loss ofthese epithelial markers makes CTCs elastic, aiding cellmovement through the extracellular matrix of a tumor leadingto metastasis (Figure 1).
Importantly, the transitioned CTCswith no epithelial markers provide crucial information aboutthe metastasis and e ff ective treatment options. − Therefore,it is vital to develop a capturing technique independent of CK/ EpCAM expression on the cancer cell. The technique based onmicro fl uidics for sorting cells has been developed for capturingmesenchymal cells. Attempts have been made to increase thee ffi ciency of the micro fl uidic system by combining withimmunomagnetic beads. The ideal system to e ffi cientlycapture the EMT transitioned cells is still lacking.One possible way to capture the cells with high e ffi ciency isto develop immunomagnetic beads that are selective, incapturing both epithelial and mesenchymal cancer cells, andpowerful by selectively removing the cells from the milieu. Todevelop a selective immunomagnetic bead, we need to identifythe common biomarkers overexpressed on tumor cells beforeand after EMT. Therefore, we analyzed the biomarkerexpression levels in tumor cells, before and after EMT, andfound two common proteins human epidermal growth factorreceptor 2 (Her2) and epidermal growth factor receptor(EGFR), whose levels remained una ff ected. On the other Received:
June 8, 2020
Accepted:
August 21, 2020
Published:
September 9, 2020
Article http://pubs.acs.org/journal/acsodf © 2020 American Chemical Society https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega − This is an open access article published under an ACS AuthorChoice License, which permitscopying and redistribution of the article or any adaptations for non-commercial purposes. D o w n l o a d e d v i a . . . on O c t ob e r , a t : : ( U T C ) . S ee h tt p s :// pub s . ac s . o r g / s h a r i nggu i d e li n e s f o r op ti on s on ho w t o l e g iti m a t e l y s h a r e pub li s h e d a r ti c l e s . and, to develop a powerful immunomagnetic bead, we needparticles with high magnetic moment. Traditional sphere-shaped magnetic beads have a magnetic moment of 5 − − The magnetic moment can be increased bydecreasing the size of beads or changing the spherical shape toa cube.
For example, nanosized spherical particles ( ∼ Therefore, in thepresent study, we have developed smaller sized nanocubesattached with biomarkers expressed in EMT cells and studiedtheir e ffi cacy in cell capture.Brie fl y, we followed a two-pronged approach for theisolation of cancer cells. First, we synthesized paramagnetic20 nm iron oxide nanocubes (FeNCs) with a high magneticmoment of 65 emu/g. Second, we conjugated antibodies to theparticles to obtain immunomagnetic iron nanocubes. We choseHer2 (ERBB2) and EGFR (ERBB1) antibodies for function-alizing the nanocubes as they play critical roles in regulating Figure 1. (a) Schematic illustration of the migration of tumor cellsafter undergoing EMT and (b) regulators and markers of the EMTprocess in tumor cells and metastatic abilities.
Figure 2. (a) Schematic diagram of the preparation of iron oxide nanocube constructs; magnetic properties for FeNC (b) superconductingquantum interference device (SQUID) hysteresis; (c) zero- fi eld curve; (d) schematic illustration of iron nanocubes conjugated to the antibody;and (e) transmission electron microscopy (TEM) images of (i) hydrophilic FeNC, (ii) PEGylated FeNC, (iii) FeNC − Her, and (iv) FeNC − EGF.
ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 MT.
It is anticipated that their expression levels oftenremain una ff ected in the cancer cells. − Using the cube-shaped nanoparticles, we developed a highly e ffi cientimmunomagnetic platform, functionalized with antibodies,for the isolation of cancer cells with or without epithelialmarkers. We avoided the conventional strategy of using gold orsilica shell over the magnetic particles, which can potentiallydecrease the magnetic moment of the fi nal construct. Instead, we employed poly(maleic anhydride)-based ring-opening strategy to generate carboxylic functional groups thatallow covalent conjugation with biomarker-speci fi c antibodies,as well as increase the hydrophilicity and solubility of theFeNC.We analyzed the expression levels of biomarkers that areretained and overexpressed on the membrane of cancer cellsbefore and after EMT. The expression levels of Her2 or EGFRwere con fi rmed to remain una ff ected, before and after EMTstimulation in cancer cells. Targeting these biomarkers,una ff ected by EMT, would allow us to capture cancer cellsirrespective of the expression of epithelial antigens. − Beforethe EMT process is fully complete, there is an intermediarystage known as partial EMT that may express both EMTregulators such as Her2/EGFR and also epithelial antigens. At this stage, both Her2 and EGFR are activated, resulting inoverexpression in the cell surface. However, epithelial antigenlevels may fl uctuate depending on the environment. Therefore, combining an anti-epithelial marker and anti-Her2/EGFR markers with nanocubes may present an e ffi cientapproach to capture all CTC subpopulations in a given sample.In this article, we present the results of the following studies:(i) synthesis and characterization of FeNC functionalized withHer2 or EGFR antibodies; (ii) evaluation of the expressionlevels of biomarkers in cancer cells before and after EMT; (iii)determination of FeNC ’ s ability to capture epithelial andmesenchymal cancer cells; and (iv) evaluation of epithelialmarkers in captured cells. ■ RESULTS AND DISCUSSION
Synthesis and Characterization of Magnetic Nano-cubes.
We synthesized FeNC by thermal decomposition (295 ° C) of iron − acetylacetonate complex that generates Fe ionsthat are reduced by polyol to produce nanocrystalline seeds(Figure 2a). By carefully maintaining the growth conditions,we obtained uniform nanocubes with an edge length of 20 nm.The cubic shape imparted a higher magnetic moment of 65emu/g (Figure 2b,c), relative to previously known sphericalnanoparticles, which showed 40 −
50 emu/g.
In compar-ison, magnetic beads of size 0.2 − −
25 emu/g. Commercially availableDynabeads show a magnetic moment of 12 emu/g. Themagnetic moment data obtained for FeNC synthesized in thisstudy is similar to previously reported iron nanocubes. Thesynthetic protocol used in this study facilitates the transition offerromagnetic to paramagnetic FeNC crystals; consequently,the nanoparticles are in a nonmagnetized form in the absenceof an externally applied fi eld. This characteristic propertyenhances the utilization of FeNCs in cell capture and selectiveisolation. Additionally, nanocubes exhibit high surface area,excellent binding capacity, and higher analytical sensitivity ascompared to traditional spherical nanoparticles. As synthesized, the FeNCs coated with oleic acid arehydrophobic and ine ffi cient in capturing cells under aqueousconditions. Therefore, we modi fi ed the surface of the particles with hydrophilic molecules to impart water solubility. Recentstudies have utilized amphiphilic molecules such as poly-(maleic anhydride- alt -1-octadecene) (PMAOD) for makingnanoparticles water-soluble (Figure 2a). − Importantly,PMAOD adds functional − COOH groups to the nanoparticlesfor conjugating with antibodies. We replaced the oleic acidon the surface of FeNC with PMAOD and, subsequently,hydrolyzed the rings of maleic anhydride to obtain carboxylicgroups (Figure 2a). Further, to prevent the interparticleattraction of FeNCs, which lead to agglomerates, weintroduced “ poly(ethylene glycol) (PEG) ” molecules as astabilizing linker. The chosen PEG linker “ NH -PEG-COOH ” (molecular weight (MW) 2000) increased the interparticledispersity (Figure 2a,d). Additionally, as the PEG linkercontains terminal carboxylic groups, the total available carboxylgroups per nanoparticle remain unchanged before and afterPEGylation. We used N -(3-dimethylaminopropyl)- N ′ -ethyl-carbodiimide hydrochloride (EDC) − N -hydroxysuccinimide(NHS) coupling agent for conjugating PEGylated FeNCswith antibody (Ab) of our choice (Figure 2a,d). It is importantto ascertain that the speci fi city of the antibody is retained afterconjugating with PEGylated FeNC. We chose the “ antiglobinand globin ” to validate the speci fi city by enzyme-linkedimmunosorbent assay (ELISA), wherein we conjugated theantibody of globin with PEGylated FeNC and globin as ananalyte (Figure S1). The ELISA data showed that the FeNC − Ab − globin construct displayed a linear correlation toward theantigen concentration, con fi rming that the antibody retains itsspeci fi city upon conjugation. Based on the encouraging results,we conjugated FeNC with Her2 or EGFR antibodies forevaluating their ability to capture cancer cells that overexpressthe respective receptors.As a next step, we characterized all of the synthesizednanoparticles using conventional analytical techniques. TEMimages of hydrophilic FeNC, PEGylated FeNC, and FeNC − Ab constructs showed the uniform shape of the particles(Figures 2e and S2). FeNC − Ab constructs are stable inaqueous solutions and showed a hydrodynamic size of ∼ ζ potential (Table 1 and Figure S3). Biomarker Selection for Cell Capture.
The overex-pressed growth factor receptors, Her2 and EGFR, areconsidered to be the essential drivers for cell proliferationand regulation of EMT. − To identify if Her2 or EGFRcan be used as suitable biomarkers for targeting and isolatingcancer cells after EMT, we investigated the expression levels ofgrowth receptors in multiple cancer cells by Western blot(WB) analysis. In this study, we selected three cancer cell lines,namely, A549, HCC827, and MCF-7. We isolated the lysatesand performed WB analysis and the results are shown in Figure3a. The data showed that A549 expresses high levels of both
Table 1. Size and ζ Potential Characteristics for the As-Synthesized Hydrophilic FeNC, PEGylated FeNC, andAntibody-Attached, FeNC − Her and FeNC − EGF construct hydrodynamic size( d , nm) ζ (mV) TEM edge length(nm)hydrophilicFeNC 148 −
63 20PEGylated FeNC 142 − − Her 171 − − EGF 175 − ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 er2 and EGFR, with Her2 showing 1.6 times higherexpression than EGFR; HCC827 expresses high EGFR (3.9times higher than Her2 relatively) and low Her2, whereasMCF-7 expresses high levels of Her2 and no expression ofEGFR. The expression levels agreed with previously publishedstudies. − Therefore, for Her2-based capturing of cells, weused A549 as a positive and HCC827 as a negative control. Onthe other hand, for EGFR-based cell capture, we usedHCC827 as a positive and MCF-7 as a negative control.
Expression Levels of Protein Markers in EMT-Stimulated Cancer Cells.
As a next step, we arti fi ciallystimulated EMT in our positive target cell lines (A549 andHCC827) to monitor the changes in expression levels of EMTmarkers, epithelial antigens, Her2, and EGFR proteins. Weused TGF β -1 and EGF to induce EMT in both A549 andHCC827. As shown in Figure 3b, within 72 h, the cells losttheir epithelial adhesion on the substrate and transformed intoan invasive elongated state con fi rming EMT stimulation. Thelysates from these EMT-stimulated cells were then analyzedusing WB (Figure 3c). The data showed increased expressionlevels of vimentin in both the cell lines after the induction ofEMT (increase of 171% in A549 and 53% in HCC827). Theupregulation of vimentin is characteristic of mesenchymalcells, and the result further validates successful EMTstimulation in cells. As shown in previous studies, we also observed a decrease in expression levels of most cytokeratins(CK; −
31% in A549 and −
12% in HCC827) and EpCAMepithelial markers ( −
73% in A549 and −
82% in HCC827) inthe EMT-induced cells.
A slight decrease in Her2expression levels ( − − Thus, targeting these two biomarkers allowsthe cell capture to be more e ffi cient and independent of theirEMT status (CK and EpCAM levels). Magnetic Nanocubes for Capturing Epithelial CancerCells. We fi rst investigated the selectivity of the FeNC − Herand FeNC − EGF to capture epithelial cancer cells. As per ourWB analysis, A549 showed a high expression of both Her2 andEGFR. Therefore, we hypothesized that the FeNC − Her orFeNC − EGF would be able to capture A549 cells selectively;while HCC827 showed only the EGFR expression, wepostulated that only FeNC − EGF would be able to captureHCC827 cells. To test the hypotheses, we prepared a mixture
Figure 3. (a) Western blot images showing the native expression levels of EGFR and Her2 in A549, HCC827, MCF-7 cells, and its correspondingband densitometry plot. (b) Morphological changes during EMT conversion for 0 −
72 h in A549 cells. After EMT conversion, the cells appearmore elongated in shape compared to untreated (Figure S4). Images were taken using bright- fi eld microscopy at a 10 × magni fi cation, and the scalebar represents 200 μ m. (c) Western blot images showing changes in the expression levels of proteins in A549 and HCC827 cells before and afterinduction of EMT, and its corresponding band densitometry plot. ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 f A549 and HCC827 in equal proportions and independentlylabeled with di ff erent color dyes (for example, A549 with a redmembrane dye, while HCC827 with a blue nuclear dye).Subsequently, we used FeNC − Ab constructs to capture cellsfor isolation by magnetic separation. The cells were thencounted by fl uorescence and identi fi ed by the correspondingcolor (Figure 4; see the Methods section).To isolate a maximum number of cells, we optimized twocrucial parameters: antibody coating on FeNC and particulateconcentrations of FeNC required for isolation. For evaluatingthe fi rst parameter, we used FeNC with varying degrees of Abon the surface and treated with the sample containing 100 cellsof each cell line (Figure 5). We anticipated that increased Abconcentrations (1 × , 6 × , and 20 × Ab) would increase thespeci fi city of cell capture. That is, FeNC conjugated with ahigh concentration of herceptin (anti-Her2) should captureA549 cells with a higher degree of speci fi city. Similarly, FeNCconjugated with high concentrations of cetuximab (anti-EGFR) should capture both A549 and HCC827 with nopreference for either one. As shown in Figure 5, the increase inAb coating directly correlated with the number of cellscaptured. The data showed that the e ffi ciency of FeNC − Herto capture A549 cells increased from 4- to 8-fold, while forFeNC − EGF to capture A549 or HCC827 cells increased from 4.3- to 7-fold or 3- to 9-fold, respectively. These resultsindicate the high selectivity of FeNC in capturing cancer cellsbased on the target receptor expression. For evaluating thesecond parameter, we treated di ff erent concentrations ofFeNC − Her or FeNC − EGF or FeNC (control) with a cellmixture of 10 cells per cell line, and the results are presentedin Figure 5. In this experiment, we performed cell capture withvarying amounts of NPs, and the results are presented inFigure 5. The data for FeNC − Ab showed a proportionalincrease in capture (2- to 3-fold) with an increase in particleconcentration. The data also showed a 3-fold increase incapture selectivity for FeNC − Ab constructs as compared tothat for the FeNC control at higher concentrations. The resultsindicated that a minimum concentration of 2 × particles/mL would be required to achieve high sensitivity in capturingcells without increasing nonspeci fi c binding counts. Together,the data established the ability of FeNC − Ab to preciselycapture cancer cells based on their target expression.
Magnetic Nanocubes for Capturing Epithelial CancerCells in Serum.
As a next step, we evaluated the ability ofantibody-conjugated FeNC constructs to capture cancer cellsspiked in serum to simulate clinical sampling conditions. Inthis study, we used FeNC (control), FeNC − Her, and FeNC − EGF for selective capturing of A549 and HCC827 cellsindependently. The cells were spiked in serum isolated frompig blood. Both FeNC − Her and FeNC − EGF successfullycaptured cells according to the receptor expression levels inA549 and HCC827 (Figure 6). As expected, the FeNC controlshowed negligible capture (Figure 6). Our results showed thatFeNC − Her was highly speci fi c and e ffi cient in capturing A549cells than HCC827 cells as compared to FeNC − EGF (92 vs55%), while FeNC − EGF was more active in detectingHCC827 cells. The fold-change between FeNC − Her andFeNC − EGF in capturing A549 cells was about 3-fold, whilethe speci fi city of FeNC − EGF over FeNC − Her in capturingHCC827 was almost 4-fold. These data suggested that FeNC − Her was an ideal candidate for capturing A549, while FeNC − EGF was a candidate to capture HCC827 cells. Additionally,the data indicated that magnetic nanocubes were stable incapturing cells in serum and exhibit similar speci fi city andselectivity as that of aqueous solutions. Such studies with theHer/EGFR-targeted capture based on heterogeneous expres-sion in cells have been previously demonstrated. Subsequently, we conducted a speci fi city test for FeNC − Herand FeNC − EGF in capturing positive target cells relative tocontrol from a cell mixture in serum. For this experiment, weused samples spiked with an A549 − HCC827 1:1 mixture(from 10 to 10 cells per cell line), wherein A549 was a Figure 4.
Steps in the immunomagnetic FeNC-mediated capture and isolation of cancer cells.
Figure 5.
Capture selectivity and sensitivity of nanoparticles withincreased (a, b) antibody coating on the surface and (c, d) particleconcentrations on a 1:1 mixture of A549 − HCC827 cells. Cellsseparated using immunomagnetic FeNC exhibit high speci fi city andsensitivity. ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 ositive control for FeNC − Her, and HCC827 was a negativecontrol. Similarly, we tested a sample spiked with HCC827 − MCF-7 mixture (from 10 to 10 cells per cell line), whereinHCC827 was a positive control for FeNC − EGF, and MCF-7was a negative control. Our results indicated a successfulcapture and isolation of cells by the constructs over theircontrols (Figure 7). Data showed a speci fi city di ff erence of 9-and 6-folds for FeNC − Her and FeNC − EGF, respectively, incapturing positive cells instead of negative control cells at thelowest cell spike. In both cases, the nonspeci fi c capture byFeNC (control) was insigni fi cant (<2%) compared to that ofFeNC − Ab constructs. Similar results in a previous study withmagnetic liposomes showed higher speci fi city when EGFR wastargeted compared to EpCAM. Another group showed thattargeting Her2 and EGFR can complement and enhanceepithelial marker-based targeting methods. In summary, thedata demonstrated the ability of FeNC to capture cancer cellsbased on the antibody − receptor speci fi city to oncogenicsurface receptors such as Her2 or EGFR in serum. Magnetic Nanocubes for Capturing MesenchymalCancer Cells in Serum.
To test the ability of FeNC − Her andFeNC − EGF in capturing mesenchymal A549 and HCC827cells, we spiked the EMT-induced cells in serum (along withnegative HCC827 or MCF-7 epithelial cells) and analyzed thecapture counts (Figure 8). We followed the fl uorescentlabeling strategy as described above to count the cells.FeNC − Her showed excellent capturing ability with A549cells before and after EMT induction (Figure 8). In fact,FeNC − Her showed higher e ffi ciency in capturing most of thecells after EMT. Similarly, FeNC − EGF also showedremarkable capturing ability before and after EMT conversionof HCC827 cells. If the sample contains ≤
10 cells, weobserved background noise and more signi fi cant deviations inthe capture. However, data was consistent within the rangetested (10, 50, and 100 spiked cells per each type). The dataestablished the capability of FeNC − Her and FeNC − EGF incapturing cancer cells irrespective of their EMT status. Theseresults are in agreement with a previous study where targetingmesenchymal cells by EMT-marker was 2.1-fold e ffi cient thantargeting epithelial markers. In another study, the magneticbeads targeting EGFR were found to be 80% more e ffi cientthan targeting epithelial markers in a low EpCAM expressingcell line. Together, the data strengthens the use of growthreceptors such as Her2 or EGFR to capture mesenchymal cells.
Biomarker Characterization of Captured Cells.
As a fi nal step, we investigated the surface epithelial marker status ofthe captured cells. We performed fl uorescent immunostainingusing an anti-CK antibody (tagged with Alexa Flour 647) inthe epithelial and mesenchymal captured cells (Figure 9a).Then, we analyzed the di ff erence between the total cellscaptured and the cells that showed positive-CK staining(Figure 9b,c). A549 cells captured by FeNC − Her showed a62.5% decrease in cells expressing CK after EMT stimulation( p < 0.047). Similarly, HCC827 cells captured by FeNC − EGFshowed a 75% decrease in cells expressing CK after inductionof EMT ( p < 0.005). The drastic decrease in epithelial marker Figure 6.
Demonstration of (a, b) the receptor-based capture using10 or 10 cells and (c) construct speci fi city to A549 vs HCC827 inserum solution. Figure 7.
Speci fi city capture tests between (a, b) FeNC − Her andFeNC (nonspeci fi c capture) for A549 − HCC827 mixed spikedsamples and (c, d) FeNC − EGF and FeNC (nonspeci fi c capture)for HCC827 − MCF-7 mixed spiked samples.
Figure 8.
Cell capture before and after induction of EMT in (a, b)A549 − HCC827 mixture and (c, d) HCC827 − MCF-7 mixture byFeNC − Her and FeNC − EGF, respectively. EMT in A549 andHCC827 cells were stimulated, respectively.
ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 xpression in cells con fi rmed that the captured cells are ofmesenchymal type. Further, we showed that magneticnanocubes capture and allow the detection of cytokeratinstatus in the captured cells. The ability to characterizebiomarkers in captured cells is an essential step in developingtechnologies for clinical applications. − In summary, ourresults demonstrate that FeNC − Her and FeNC − EGFconstructs could successfully capture, isolate, and allow thedetection of epithelial marker status in captured cells. ■ CONCLUSIONS
We developed a novel magnetic nanocube-based tumor-cellcapture technique for separating epithelial and/or mesenchy-mal cells. In this study, we have shown that surface markerssuch as Her2 and EGFR are una ff ected by the EMT state andare potential biomarkers for capturing CTCs. The magneticnanocubes can help in isolating speci fi c cancer cells based ontheir marker and allow the detection of EMT and non-EMTpopulation subsets within a given pool of cells. The results ofthis study underscore the importance of using growth receptorsas a secondary target for capturing CTCs. As the tumor isalways heterogeneous, the proposed method provides anopportunity to capture CTCs that express variable biomarkers.The results present in this study can aid in developing next-generation techniques for CTC isolation in clinics. ■ METHODS
Materials.
We purchased iron(III) acetylacetonate, oleicacid, benzyl ether, poly(maleic anhydride- alt -1-octadecene)(PMAOD), N -(3-dimethylaminopropyl)- N ′ -ethylcarbodiimidehydrochloride (EDC), N -hydroxy sulfosuccinimide sodium salt(sulfo-NHS), 2-( N -morpholino)ethane sulfonic acid (MES),bovine serum albumin (BSA), MS-SAFE protease andphosphatase inhibitor, Triton X-100, Tween-20, and phos-phate-bu ff ered saline (PBS) from Sigma-Aldrich. Hexane,toluene, and chloroform were purchased from Acros Organics.Sodium hydroxide (NaOH), acetone, sucrose, sodium boratebu ff er, Hoechst dye, CellMask deep red membrane dye, PierceECL Western blotting substrate, Pierce BCA protein assay kit,and 3,3 ′ ,5,5 ′ -tetramethylbenzidine (TMB) were purchasedfrom Thermo Fisher Scienti fi c. β -Actin, vimentin, EpCAM,cytokeratin primary antibody kit (7, 8, 17, 18, 19, PAN), EGFR, Her2, and both antimouse and antirabbit secondaryhorseradish peroxidase (HRP)-linked immunoglobulin G(IgG) antibodies were purchased from Cell Signaling.Carboxylic-poly(ethylene glycol)-amine was custom-synthe-sized from Nanocs and Laysan Bio. 10 × Tris-bu ff ered saline(TBS), 10 × Tris/glycine/sodium dodecyl sulfate (SDS) bu ff er,4 −
15% 10-well 50 μ L ready Mini-Protean TGX, Precision PlusProtein dual-color standard ladder, and supported nitro-cellulose membrane (0.2 μ m) were purchased from Bio-Rad.Recombinant human EGF (carrier-free) and Alexa Fluor 647tagged anticytokeratin PAN antibody were purchased fromBioLegend. Recombinant human transforming growth factor β -1(TGF β -1) and EGF were purchased from Novoprotein.Amicon 0.5 mL of 100 kDa regenerated cellulose ultracelcentrifugal fi lter units were purchased from Merck MilliporeLtd. Cetuximab (Eli Lilly and Company) and herceptin(Genetech Inc.) were purchased from the hospital pharmacy. Instrumentation.
High-resolution JEOL transmissionelectron microscope (TEM) was used to measure and recordthe images of nanoparticles. The hydrodynamic size and ζ potential of the nanoparticles were measured using a MalvernZetasizer Nano-ZS (Malvern Panalytical). The centrifugationwas performed on a 5424 Eppendorf and refrigerated RC 6+Sorvall centrifuge (Thermo Fisher Scienti fi c). The pH wasmeasured using a Seven Compact Mettler Toledo pH meterequipped with an InLab Micro electrode. We performed fl uorescence imaging, fl uorescence, and UV − vis absorptionmeasurements using a Cytation 5 cell imaging multi-modereader (BioTek Instruments Inc.). Gel electrophoresis wasperformed on a Bio-Rad Mini-PROTEAN Tetra system, andblots were transferred using a Genscript e-blot transfer system.We performed Western blot imaging and acquisition usingImage Lab version 5.2.1 software on a ChemiDoc XRS systemfrom Bio-Rad. Solvent extraction was performed on a Bu ̈ chirotavapor R-124 attached to a water bath and water-cooleddistillation column. Finally, SQUID measurements wereperformed on a Quantum Design MPMS 3 magnetometer(Quantum Design). Synthesis of Iron Oxide Nanocubes.
We synthesizediron oxide nanocubes (FeNCs) using a high-temperaturepolyol reduction process in an organic phase. Brie fl y, amodi fi ed protocol based on previous studies was used forsynthesizing particles. In this procedure, iron(III)acetylacetonate (7.06 g, 20 mmol) was added to a mixture ofoleic acid (1.129 g, 12.69 mL) and benzyl ether (104 g, 99.71mL) in a three-neck 200 mL round-bottom fl ask (RBF)attached to a re fl ux condenser and a Schlenk line. This solutionwas degassed using Argon (20 kPa) for 1 h under constantstirring (400 rpm), and the reaction was kept under argonthroughout the process. After initial degassing, the reactiontemperature was increased to 295 ° C at a heating rate of 20 ° C/min under constant stirring (500 rpm). The reaction colorturned jet black from reddish brown after 30 min, and at thisstage, the solution temperature decreased to 100 ° C. Next, thesolution was then quickly cooled on ice and a mixture (80 mL)of hexane, toluene, and acetone with a ratio of 1:1:2 wasadded. This solution was stirred and sonicated to removeparticles stuck to the stir bar. The solution was then transferredto a 500 mL beaker with a seal and allowed to precipitate for15 h. The top solution was decanted, and the remainingsolution containing the precipitate was centrifuged at 20 000 g for 1 h (15 ° C). The pellet was then resuspended and washedseveral times using a mixture of hexane and toluene (ratio of
Figure 9. (a) Steps for EMT stimulation and magnetic separation;and staining and detection of cytokeratin in cells captured by (b)FeNC − Her and (c) FeNC − EGF.
ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 :1) until a slight clear supernatant formed. For each wash, thetube was kept over a neodymium magnet (pull force of 250 lb)for 15 min followed by the slow removal of the supernatantusing a 10 mL pipette. This solution was then concentrated inhexane and washed fi ve times (15 000 g for 15 min) until aclear supernatant formed. Finally, the particles (20 mg yield)were resuspended in 4 mL of hexane and stored at roomtemperature. Hydrophilic Conversion of Iron Oxide Nanocubes.
The iron nanocubes synthesized using the polyol processresulted in the formation of hydrophobic particles. To convertthem to hydrophilic particles for use in aqueous andphysiological solutions, we replaced the organic coating witha functional amphiphilic polymer by modifying a previouslypublished protocol. Brie fl y, synthesized FeNCs (in hexane)were injected into a rapidly stirring (500 rpm) poly(maleicanhydride- alt -1-octadecene) (1.2 g, average molecular weight30 000 −
50 000) solution dissolved in chloroform (100 mL) ina one-neck 200 mL RBF. After 2 h of stirring, the solvent wasremoved using a rotary evaporator (100 rpm, 55 ° C) attachedto a standard vacuum line protected by a liquid nitrogen cooltrap, until a wet sludge was formed. The sludge was quicklymixed with 50 mL of 0.1 M NaOH solution and dispersed bysonicating for 1 h. The solution was then stirred (500 rpm) for15 h at room temperature to hydrolyze and open the maleicanhydride rings. The solution was washed twice with 0.1 MNaOH solution at 20 000 g (15 ° C) for 1 h, followed bywashing twice with a three-layer sucrose density gradientcentrifugation (30, 20, and 10% sucrose). The pellet wasresuspended and concentrated in 4 mL of 10 mM MES bu ff erand washed once. As a next step, 100 kDa centrifugal fi lterswere equilibrated with MES bu ff er and used for removingexcess polymer (6000 g for 6 min). The solution was then againsubject to a two-layer sucrose density gradient centrifugation(30 and 10% sucrose) at 15 000 g for 15 min, followed by sixwashes with DI water. The FeNC pellet was then dispersed in5 mL water and stored at room temperature. Synthesis of PEGylated Nanocubes.
To enhance thestability and solubility in aqueous solutions, the carboxylgroups of hydrophilic particles were activated using EDC/sulfo-NHS and conjugated with NH -PEG-COOH. Brie fl y, theparticles were dissolved in 0.1 M MES bu ff er (pH 4.5)containing 2 mg EDC and 2 mg sulfo-NHS and kept forshaking (850 rpm) for 1 h at 28 ° C. The particles were thencentrifuged (15 000 g for 15 min), and the pellet was dispersedin a 50 mM sodium borate bu ff er containing 10 mg carboxylic-poly(ethylene glycol)-amine (COOH-PEG-NH , MW 2000; 5 μ mol) and 5 mg EDC (26 μ mol). This solution was keptshaking (850 rpm) for 15 h at 22 ° C. After the reaction,PEGylated FeNCs were washed with water two times (15 000 g for 15 min), suspended in DI water, and stored at roomtemperature. PEGylated FeNCs were used for conjugationwith antibodies. Antibody Conjugation.
We conjugated the antibody(Ab) to the hydrophilic PEGylated FeNCs by the traditionalEDC/sulfo-NHS cross-linking procedure. Brie fl y, 100 μ L (2 × particles/mL) of FeNCs was washed with 0.1 M MESbu ff er (pH 4.5) prior to being activated. The pellet wasdispersed in 0.1 M MES bu ff er (400 μ L; pH 4.5) with 3 mgEDC (16 μ mol) and 4 mg sulfo-NHS (18 μ mol) at 28 ° C for3.5 h (shaking at 800 rpm). After activation, the particles werecentrifuged (15 000 g for 15 min), dispersed in 100 μ L 1 × PBS,and mixed with the antibody solution of either 4 μ L of globin − Ab (10 mg/mL), 200 μ L of herceptin (2 mg/mL), or 200 μ Lof cetuximab (2 mg/mL). The reaction was kept in a shaker(800 rpm) for 15 h at 22 ° C and, upon completion, washedtwice with 1 × PBS (15 000 g for 15 min). The fi nal solutionwas dispersed in 1 mL of 1 × PBS and stored at 4 ° C. TheFeNC − Ab − globin was used for estimating the proteinconjugation e ffi ciency using the known globin protein, whileFeNC conjugated to cetuximab and herceptin was used totarget EGFR and Her2 in cell studies. These particles werelabeled as FeNC − EGF and FeNC − Her, respectively. Due tothe ready availability of globin − Ab, we were able to estimatethe conjugation e ffi ciency using ELISA and anticipate that theconjugation e ffi ciency of antibodies of EGF and Her to FeNCwould be similar to that of globin − Ab to FeNC.
Characterization of Magnetic Nanocubes.
We charac-terized FeNC using TEM and dynamic light scattering (DLS)technique. For TEM, 8 μ L of 1/10th diluted FeNCs wasplaced over a CF300-Cu carbon fi lm-based copper grid and air-dried for 10 min in a hot air oven (40 ° C). TEM imaging forpolymer-coated and PEGylated FeNCs showed a uniform sizeof 20 nm. For DLS measurements, 1 mL of 1/10th dilutedFeNC solution was placed in a disposable UV-grade cuvetteand analyzed. The magnetic properties were further inves-tigated using the superconducting quantum interference device(SQUID) analysis. For this analysis, 5 mg of the lyophilizedFeNC power was loaded in sealed SQUID sample holders. Themagnetic moment was recorded using the temperaturesweeping mode under zero- fi eld-cooled (ZFC) conditionsfrom 2 to 300 K at 1000 Oe. The corrected magnetic momentwas then determined by a previously published procedure. ELISA.
To determine the conjugation e ffi ciency of theantibody bound to FeNC, we used an indirect enzyme-linkedimmunosorbent assay (ELISA). For this experiment, theFeNC − Ab − globin construct was used to evaluate thespeci fi city to globin antigen; the reaction-supernatant wasused to quantify the amount of conjugated globin antibody.PEGylated FeNCs were used as the negative control, and PBSwas considered as the blank. Brie fl y, 100 μ L of globin solution(0.1, 1, 10, and 100 μ g/mL, respectively) was added to wells ofa fl at-bottom sterile 96-well plate and left undisturbed for 15 hat 4 ° C. The wells were then washed with a wash bu ff er (0.5%Tween-20) followed by blocking with 200 μ L of 2% BSAsolution for 30 min at room temperature. As a next step, thewells were washed with a wash bu ff er prior to the addition ofsamples (100 μ L). After 2 h, at 37 ° C, the sample solutionswere washed with a wash bu ff er. Subsequently, 50 μ L of the0.32 μ g/mL secondary antibody (goat antirabbit IgG-HRP)was then added to the wells and incubated for 30 min at 37 ° Cfollowed by two washes with a wash bu ff er. Finally, 50 μ L ofthe substrate solution (TMB peroxidase) was added to thewells and the reaction was stopped by adding 0.1 M HClsolution. The plates were then read by a Cytation 5 absorbanceplate reader at 450 nm, and the absorption intensity wascorrelated to the concentration of the globin antibodyconjugated to FeNC.
SQUID Magnetic Analysis.
We evaluated the magneticproperties of FeNCs using superconducting quantum interfer-ence device (SQUID) equipped with a 7 T magnet. In thisprocess, Josephson junctions were used to measure a change inthe magnetic fl ux using a coil with a known inductance. For themagnetometry analysis, 5 mg of pure lyophilized FeNCs wasloaded in a VSM sample holder. Measurements were madeusing a fi ve-quadrant magnetic moment with varying external ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 agnetic fi eld analysis ( H max of 5 T at T = 2 K). Zero- fi eld-cooled magnetic susceptibility and fi eld-cooled warmingmeasurements were performed using a sweep mode from 2to 300 K at 1000 Oe. Using this analysis, the blockingtemperature was estimated. Raw data was then corrected forsample shape and radial o ff sets and graphs were plotted. Isolation of Blood Serum.
Pig serum was obtained froman investigator at the University of Missouri for the spikingstudies. Protocols in accordance with the ethical animal-handling regulations as approved by the University of Missouriwere followed. Five milliliters of blood was collected from pigsin a sealed test tube. The clotted blood was centrifuged (100 g for 10 min at 4 ° C), and the serum was isolated and stored at 4 ° C. The serum was used for cancer cell spiking experiments.
Cell Culture.
A549, HCC827, and MCF-7 human cancercell lines (ATCC) were grown in the Roswell Park MemorialInstitute (RPMI) 1640 medium (obtained from Gibco BRL).The media was supplemented with 4.5 g/L D -glucose, 25 mM N -(2-hydroxyethyl)piperazine- N ′ -ethane sulfonic acid(Hepes), 0.11 g/L sodium pyruvate, 1.5 g/L sodiumbicarbonate, 2 mM L -glutamine, 10% heat-inactivated fetalbovine serum (FBS; Atlanta Biologicals), and 1% penicillin/streptomycin antibiotic solution. The cells were cultured in ahumidi fi ed atmosphere of 95% air and 5% CO at 37 ° C(Thermo Scienti fi c). EMT Conversion.
To induce epithelial-to-mesenchymaltransition (EMT) arti fi cially in cells, 120 ng of TGF β (20 ng/mL) and 60 ng of EGF (10 ng/mL) protein were mixed in 6mL serum-free media and treated with A549 cells. Treatmentwas performed at 60% con fl uence in a T25 fl ask for 72 h in aCO incubator. Flasks were monitored every 24 h using abright- fi eld microscope, and images were obtained. At the endof the conversion, the fl ask was washed once with fresh RPMImedia containing 10% FBS and twice with cold 1 × PBS. Thecells were then dislodged and used for magnetic capture.
Capture and Detection of Cancer Cells.
To capture anddetect spiked cancer cells, we developed a robust magneticextraction process. In this experiment, the cells selected forcapture and detection were either A549 (Her2+++, EGFR++)or HCC827 (EGFR+++), while control cells were eitherHCC827 (EGFR+++) or MCF-7 (Her2+++), respectively.For capturing the cells of interest, A549 or HCC827 cells withor without EMT treatment (induced by TGF β -1 and EGF)were dislodged from the T25 fl ask (70% con fl uence) and usedfor spiking experiment. For all experiments, target cellssuch as A549 or HCC827 were live-stained red, while controlcells such as HCC827 or MCF-7 were live-stained blue. Theaddition of control cells allows us to test the speci fi city of theconstruct in each study. Additionally, all capture experimentsrelied on live cells, while cytokeratin detection was performedon fi xed cells. Brie fl y, the cells were suspended in 1 × PBS andincubated with 40 μ L of Hoechst (blue) dye or 40 μ L ofCellMask (deep red stain) for 20 min at 37 ° C. The cells werewashed with 1 × PBS twice to remove excess dye (2000 rpmfor 5 min). The cells were counted, and cell stock solution wasprepared (10 /10 or 10 cells). Two hundred microliters ofcell suspension (10 /10 or 10 cells) was mixed with 200 μ Lof control cells (10 /10 or 10 cells, respectively) and placedin a tube. Equal amounts of each target and control cell werealso placed in separate tubes as nontreated control. For testing fi xed concentration, 200 μ L of the antibody-functionalizedFeNC construct (FeNC − Her or FeNC − EGF; 2 × particles/mL) was added to the tube with cells and incubated at 37 ° C for 2 h (gently shaken every 5 min). For evaluatingvarying concentrations, either the surface − antibody concen-tration on nanoparticles (1 × (0.4 mg/reaction), 6 × , and 20 × )or nanoparticle concentration (1 × or 2 × particles/mL) was changed. Following the treatment, the tubes wereplaced (0.1 cm spacing) next to an N50-grade neodymiumblock magnet (with a surface fi eld of 4833 G, a dipole moment(m) of 75.6 A m , and a permeance coe ff . P c of 1.25) for 30min. The supernatant was carefully removed, and the capturedpellet was washed twice with 500 μ L of 1 × PBS using magneticseparation. The FeNC − Ab − cell solution was then resus-pended in 200 μ L of 1 × PBS and transferred to a fl at-bottom96-well plate. To prevent any agglomeration, a hydrophobiclow retention 200 μ L tip was used to quickly mix and transferwithout immersing the tip fully in solution. Cell counting ofstained cells was performed using an automated fl uorescence-based cell counter using the Cytation 5 imaging plate reader. Western Blot.
To perform Western blotting (WB), A549,HCC827, or MCF-7 cells (1 × ) with or without EMTtreatment at 70% con fl uence were seeded in six-well plates.The cells were treated in serum-free media for a period of 72 hfollowed by whole-cell lysate preparation using lysis bu ff ercontaining 1 × phosphatase inhibitor cocktail. Proteins wereseparated by 4 −
15% polyacrylamide gel electrophoresis(PAGE) and were transferred onto a nitrocellulose membranefor protein blot analysis. Membranes were washed with 1 × TBST, blocked with a 5% BSA blocking bu ff er containing 0.5%normal goat serum (NGS) for 2 h, and incubated with aprimary antibody overnight on a shaker at 4 ° C. Blots werethen washed before and after incubation with a secondaryantibody and developed with the chemiluminescence system.Densitometry analysis was performed using Image Studio Litesoftware (LI-COR Biosciences).
Detection of Cytokeratin.
To detect epithelial markerstatus, the cells captured by antibody-functionalized FeNCconstructs (FeNC − Her or FeNC − EGF) were stained(Hoechst dye) and fi xed by adding 500 μ L of 4%paraformaldehyde for another 10 min at 37 ° C. Next, 500 μ L of 0.1% Triton X-100 was added and incubated for 5 minafter staining and fi xing. The rest of the steps were the same forthe FeNC treatment. After cell capture and washing once with1 × PBS, 500 μ L of 5% FBS solution in 1 × PBS was added as ablocking agent for 30 min. After the removal of a blockingagent, 400 μ L of Alexa Fluor 647 anticytokeratin antibody (2 μ g/mL) solution diluted in 1 × PBS cells was added to the tubeand incubated for 2 h at room temperature. The cells werethen washed twice with 1 × PBS and then resuspended in 200 μ L of 1 × PBS. This solution was then quickly transferred to a96-well plate for imaging and fl uorescence-based cell countingusing the Cytation 5 imaging plate reader. Statistics.
All statistics were performed using Minitab. Theaverage values from triplicates were calculated and used forstatistical analysis. Statistical signi fi cance was evaluated usingStudent ’ s t test, and p < 0.05 was considered signi fi cant. ■ ASSOCIATED CONTENT * s ı Supporting Information
The Supporting Information is available free of charge athttps://pubs.acs.org/doi/10.1021/acsomega.0c02699.ELISA plot for FeNC − Ab − globin; TEM images ofFeNC in di ff erent stages; DLS and ζ potential of FeNC; ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 nd microscopy images of untreated A549 cells for aperiod of 0 −
72 h (Figures S1 − S4) (PDF) ■ AUTHOR INFORMATION
Corresponding Author
Raghuraman Kannan − Department of Bioengineering andDepartment of Radiology, University of Missouri, Columbia,Missouri 65212, United States; orcid.org/0000-0003-1980-3797; Email: [email protected]
Authors
Dhananjay Suresh − Department of Bioengineering, Universityof Missouri, Columbia, Missouri 65212, United States; orcid.org/0000-0003-3490-5636
Shreya Ghoshdastidar − Department of Bioengineering,University of Missouri, Columbia, Missouri 65212, UnitedStates
Abilash Gangula − Department of Radiology, University ofMissouri, Columbia, Missouri 65212, United States
Soumavo Mukherjee − Department of Bioengineering,University of Missouri, Columbia, Missouri 65212, UnitedStates
Anandhi Upendran − Department of Medical Pharmacology & Physiology and Institute of Clinical and Translational Science,University of Missouri, Columbia, Missouri 65212, UnitedStates
Complete contact information is available at:https://pubs.acs.org/10.1021/acsomega.0c02699
Author Contributions ⊥ D.S. and S.G. contributed equally to this work. D.S. formedthe initial hypothesis and detection method. D.S. and R.K.conceived the experiments. D.S. and S.G. synthesized andcharacterized the samples. D.S., S.G., and A.G. handled theantibody staining and magnetic separation. D.S. and S.G.handled the cellular imaging. D.S., S.M., and S.G. carried outthe Western blotting. D.S. and A.U. handled the blood andserum isolation. D.S. performed data analysis and S.G. carriedout the statistics. D.S. and R.K. wrote the manuscript.
Notes
The authors declare no competing fi nancial interest. ■ ACKNOWLEDGMENTS
R.K. acknowledges Michael J and Sharon R BuksteinEndowment for fi nancial support of this work. ■ REFERENCES (1) Schneck, H.; Gierke, B.; Uppenkamp, F.; Behrens, B.;Niederacher, D.; Stoecklein, N. H.; Templin, M. F.; Pawlak, M.;Fehm, T.; Neubauer, H. EpCAM-Independent Enrichment ofCirculating Tumor Cells in Metastatic Breast Cancer.
PLoS One , , No. e0144535.(2) Kowalik, A.; Kowalewska, M.; Go ́ z ́ dz ́ , S. Current approaches foravoiding the limitations of circulating tumor cells detection methods-implications for diagnosis and treatment of patients with solid tumors. Transl. Res. , , 58 − J. Thorac. Dis. , , 183 − ’ universal CTC-chip ’ . Oncol. Rep. , , 77 − Oncotarget , , 24677 − ̀ , A. M.; Cortesi, E.;Gazzaniga, P. Circulating tumour cells lacking cytokeratin in breastcancer: the importance of being mesenchymal. J. Cell. Mol. Med. , , 1066 − Breast Cancer Res. , , No. R46.(8) Maheswaran, S.; Haber, D. A. Circulating tumor cells: a windowinto cancer biology and metastasis. Curr. Opin. Genet. Dev. , ,96 − ̀ , F.; Minotti, L.; Baldassari, F.; Crudele, F.;Cook, W. J. J.; Di Leva, G.; d ’ Adamo, A. P.; Gasparini, P.; Volinia, S.Heterogeneity in Circulating Tumor Cells: The Relevance of theStem-Cell Subset.
Cancers , , No. 483.(10) La Thangue, N. B.; Kerr, D. J. Predictive biomarkers: aparadigm shift towards personalized cancer medicine. Nat. Rev. Clin.Oncol. , , 587 − J. Natl. Cancer Inst. , , 1054 − J. Natl. Cancer Inst. , , 339 − Curr.Pharm. Biotechnol. , , 810 − AIP Adv. , , No. 025023.(15) Mayo, C.; Ortega, F. G.; Gime ́ nez-Capita ́ n, A.; Molina-Vila, M.A.; Serrano, M. J.; Viteri, S.; Costa, C.; Gasco ́ , A.; Bertran-Alamillo, J.;Karachaliou, N.; et al. CK-coated magnetic-based beads as a tool toisolate circulating tumor cells (CTCs) in human tumors. Transl. LungCancer Res. , , 65 − ̈ ber method for His-tagged proteinsseparation. J. Colloid Interface Sci. , , 156 − J.Phys.: Conf. Ser. , , No. 012009.(18) Kuroda, C. S.; Maeda, M.; Nishibiraki, H.; Matsushita, N.;Handa, H.; Abe, M. In Styrene Coated Iron Nanobead for Medical Use ,INTERMAG ASIA 2005: Digests of the IEEE InternationalMagnetics Conference; IEEE, 2005; pp 457 − J.Nanomater. , , 1 − ff ey, K. R. Investigation of GMI in CoSiB andNiFe tri-layer thin fi lm structures. INTERMAG Asia 2005. Digests ofthe IEEE International Magnetics Conference 2005 ; IEEE, 2005.(21) Namanga, J.; Foba, J.; Ndinteh, D. T.; Yufanyi, D. M.; Krause,R. W. M. Synthesis and Magnetic Properties of a SuperparamagneticNanocomposite “ Pectin-Magnetite Nanocomposite ” . J. Nanomater. , , 1 − Sci.Rep. , , No. 9894. ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735
23) Chung, S. S.; Giehl, N.; Wu, Y.; Vadgama, J. V. STAT3activation in HER2-overexpressing breast cancer promotes epithelial-mesenchymal transition and cancer stem cell traits.
Int. J. Oncol. , , 403 − Br. J. Cancer , , 134 − J. Biomed. Sci. , , No. 98.(26) Gupta, P.; Srivastava, S. K. HER2 mediated de novo productionof TGF β leads to SNAIL driven epithelial-to-mesenchymal transitionand metastasis of breast cancer. Mol. Oncol. , , 1532 − Oncogene , , 4702 − Cancer Res. , , 9066 − Chem. Commun. , , 10452 − J. Mater. Chem. , , 2629.(31) Vergara, D.; Simeone, P.; Franck, J.; Trerotola, M.; Giudetti, A.;Capobianco, L.; Tinelli, A.; Bellomo, C.; Fournier, I.; Gaballo, A.;et al. Translating epithelial mesenchymal transition markers into theclinic: Novel insights from proteomics. EuPa Open Proteomics , , 31 − β -cateninpathway and promotes EMT-like phenotype in trastuzumab-resistantHER2-overexpressing breast cancer cells. Mol. Cancer Res. , ,1597 − ı ́ guez, M.; Garc ı ́ a-Puche, J. L.;Delgado-Rodriguez, M.; Sole ́ , F.; et al. EMT and EGFR in CTCscytokeratin negative non-metastatic breast cancer. Oncotarget , ,7486 − ACS Nano , , 3080 − ACSOmega , , 8010 − O @Au superparamagnetic nanoparticles: in vitro bioactivity. Nanotechnology , , No. 315102.(37) Shtykova, E. V.; Huang, X.; Gao, X.; Dyke, J. C.; Schmucker, A.L.; Dragnea, B.; Remmes, N.; Baxter, D. V.; Stein, B.; Konarev, P. V.;et al. Hydrophilic Monodisperse Magnetic Nanoparticles Protected by an Amphiphilic Alternating Copolymer. J. Phys. Chem. C , ,16809 − New J. Chem. , , 2051.(39) Di Corato, R.; Quarta, A.; Piacenza, P.; Ragusa, A.; Figuerola,A.; Buonsanti, R.; Cingolani, R.; Manna, L.; Pellegrino, T. Watersolubilization of hydrophobic nanocrystals by means of poly(maleicanhydride-alt-1-octadecene). J. Mater. Chem. , , 1991.(40) Bronstein, L. M.; Shtykova, E. V.; Malyutin, A.; Dyke, J. C.;Gunn, E.; Gao, X.; Stein, B.; Konarev, P. V.; Dragnea, B.; Svergun, D.I. Hydrophilization of Magnetic Nanoparticles with ModifiedAlternating Copolymers. Part 1: The Influence of the Grafting. J.Phys. Chem. C , , 21900 − ́ n ̃ ez-Lozano, R.; Lumbreras, R.; Gonza ́ lez-Rodríguez, V.; Delgado-García, A.; Jime ́ nez-Hoyuela, J. M.; de LaCueva-Me ́ ndez, G. Partial PEGylation of superparamagnetic ironoxide nanoparticles thinly coated with amine-silane as a source ofultrastable tunable nanosystems for biomedical applications. Nano-scale , , 812 − β Int. J. Mol. Med. , , 113 − β and matrix-regulated epithelialto mesenchymal transition. Biochim. Biophys. Acta, Gen. Subj. , , 2621 − Oncogene , , 1436 − Sci.Rep. , , No. 7406.(46) Kuai, J.-H.; Wang, Q.; Zhang, A.-J.; Zhang, J.-Y.; Chen, Z.-F.;Wu, K.-K.; Hu, X.-Z. Epidermal growth factor receptor-targetedimmune magnetic liposomes capture circulating colorectal tumor cellsefficiently. World J. Gastroenterol. , , 351 − ACS Nano , , 8816 − J. Circ. Biomarkers , , No. 1849454418782617.(49) Vila, A.; Abal, M.; Muinelo-Romay, L.; Rodriguez-Abreu, C.;Rivas, J.; Lo ́ pez-Lo ́ pez, R.; Costa, C. EGFR-Based Immunoisolation asa Recovery Target for Low-EpCAM CTC Subpopulation. PLoS One , , No. e0163705.(50) Wallander, M. L.; Geiersbach, K. B.; Tripp, S. R.; Layfield, L. J.Comparison of reverse transcription-polymerase chain reaction,immunohistochemistry, and fluorescence in situ hybridizationmethodologies for detection of echinoderm microtubule-associatedproteinlike 4-anaplastic lymphoma kinase fusion-positive non-smallcell lung carcinoma: implications for optimal clinical testing. Arch.Pathol. Lab. Med. , , 796 − ̈ m, H.; Johansson, L.; Jirstro ̈ m, K.; Jo ̈ nsson, M.;Jo ̈ nsson, P.; Planck, M. Immunohistochemistry in the differentialdiagnostics of primary lung cancer: an investigation within theSouthern Swedish Lung Cancer Study. Am. J. Clin. Pathol. , ,37 − ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735
ACS Omega −23735 agnetic nanospheres for rapid, efficient capture and sensitivedetection of circulating tumor cells. ACS Nano , , 941 − Lab Chip , , 78 − J. Am. Chem.Soc. , , 454 − J. Solid StateChem. , , 334 − ACS Omega http://pubs.acs.org/journal/acsodf
Article https://dx.doi.org/10.1021/acsomega.0c02699
ACS Omega −23735