Aptamer/magnetic nanoparticles decorated with fluorescent gold nanoclusters for selective detection and collection of human promyelocytic leukemia (HL-60) cells from a mixture

See supporting information, available online at stacks.iop.org/NANO/31/025605/mmedia.

Fe₃O₄ NPs were synthesized according to the literatures with some modifications [36]. Briefly, 4.0 g of iron(III) nitrate and 1.0 g of iron(II) chloride were added to 100 ml of deionized water in a three-necked flask under N₂ atmosphere. Then, 16.0 g of trisodium citrate salt was added to the solution in one portion under mechanical stirring. The temperature of the reaction was raised to 80 °C and a solution of sodium hydroxide (2.0 M) was added quickly. The pH of the solution was checked frequently to stay at 9.0. The reaction mixture was stirred for 1 h, followed by cooling down at room temperature. The synthesized Fe₃O₄ NPs were easily separated from the solution using a magnet and washed three times with deionized water. For silica coating, 0.2 g Fe₃O₄ NPs were redispersed in a mixture of 100 ml ethanol and 0.5 mL ammonium hydroxide in a round-bottom flask. Then, 800 μL of tetraethyl orthosilicate (TEOS) was added slowly at room temperature under stirring. After passing 1 h, deionized water (1.0 mL) was added to the mixture. The product was magnetically separated, washed three times and dried overnight.

Preparation of the GNCs with MPS (MPS = (γ-Mercaptopropyl)trimethoxysilane) shell was done according to the following procedure. First, 10 μL of tetrachloroaurate(III) was added to 5.0 mL of methanol and then 13 μL of MPS was added under continuous stirring at room temperature. Amount of 3.0 mg of sodium borohydride was dissolved in 2.0 mL of methanol and added in one portion to the above mixture. It was then allowed completing the reaction for 1 h. The final reddish-brown solution was magnetically separated to remove the unreacted materials and redispersed in PBS^- buffer (1.0 ml, 10 mM, pH 7.4).

First, the freshly prepared GNCs were conjugated to the Fe₃O₄@SiO₂ NPs. For this purpose, 3.0 mL of the GNCs@MPS solution (200 μg mL⁻¹) was added to 3.0 mL of 16 μg mL⁻¹ Fe₃O₄@SiO₂ aqueous solution. Subsequently, (polyethylene glycol) dimethacrylate (PGD) was added to the mixture. This solution was stirred for 2 h to complete the reaction, after which the Fe₃O₄/GNCs nanoparticles were magnetically collected, washed with deionized water, and finally redispersed in PBS^- buffer (1.0 mL, 10 mM, pH 7.4). Immobilization of thiol-functionalized KH1C12 aptamer on the surface of Fe₃O₄/GNCs was done according to the following procedure. For aptamer activation, 15.0 μL (5.0 μM) of the aptamer solution (in PBS^-, 10 mM, pH 7.4) was incubated with 100 μL of 1.0 mg mL⁻¹ DTT at 30 °C for 30 min. To conjugate aptamer with the nanoprob, the activated aptamer solution was mixed with the nanoprobe and thermal initiator V50 (1 mg) and they were incubated at 55 °C for 5 h. The Fe₃O₄/GNCs/Aptamer nanoparticles were magnetically collected and the concentration of unreacted aptamer (in the supernatant) was measured by spectrophotometry.

MTS assay was employed to study the cytotoxicity of the synthesized nanoprobe. The HL-60 cancer cells were cultured in a 96-well plate to reach about 70% confluency. Before addition of the nanoprobe to cancer cells, the culture medium was aspirated and replaced with 150 μL of fresh medium, followed by cell incubation for 30 min. Then, the nonoprobe (1–4 nM) was incubated with the cells for 48 h. Untreated cells and doxorubicin (DOX, 7.5 μM) were also considered as negative and positive controls, respectively. After the nonoprobe incubation, an aliquot of 15 μL MTS solution was added to each well and incubated for a further 4 h at 37 °C. To estimate the cell viability, the absorbance of each well was recorded with a microplate reader at 490 nm.

To investigate specific internalization of KH1C12 aptamer-functionalized nanoprobe into the cancer cells, flow cytometry analysis was performed. HL-60 and HepG2 cells were cultured in a 12-well plates for 24 h (i.e. 2.5 × 10⁵ cells well⁻¹), followed by their incubation with 75 nM of the nonoprobe for 2 h at 37 °C. Then, the cells were centrifuged at 1800 rpm for 7 min. After removing the supernatant of each well, the cells were redispersed in PBS^–. Internalization of the nanoprobe was evaluated by monitoring the fluorescence of GNCs in 627 nm.

HL-60 and HepG2 cells were grown in a 24-well plate for 24 h (i.e. 2 × 10⁴ cells well⁻¹). Then, the nanoprobe (60 nM) was added to each well and incubated for 2 h at 37 °C. The cells were washed three times with PBS^– to remove the excess nanoprobes in advance.

A 96-well plate was prepared by dispensing into each well a range of 10–200 cells of HL-60 along with 20 μL of a 450 nM of Fe₃O₄/GNCs/Aptamer nanoprobes. The final volume in each well reached to 150 μL. The experiments also included a negative control (the wells containing the culture medium along with untreated cells) and each concentration was repeated three times. The plate was placed in the incubator at 37 °C for 2 h. The emission of the fluorescent nanoprobes was monitored at 627 nm in refer to the emission of culture medium and the background emission of the cells.

Different concentrations of the nanoprobe (0.01, 0.02, 0.03, 0.04, and 0.05 mM) were prepared in deionized-water. T₂-weighted images were obtained under a 1.5 T clinical MRI scanner (Siemens MAGNETOM Trio Tim) at room temperature. The relative light intensity of the spots was compared using ImageJ software.

A facial approach was used to prepare multi-functional nanoprobe with simultaneous fluorescent and magnetic properties for selective detection and collection of suspended cancer cells from a mixture. In this regard, as shown in scheme 1, GNCs were generated in an ethanolic solution of MPS to produce GNCs@MPS, which were then mixed with the aqueous solution of Fe₃O₄@SiO₂ in the presence of PGD at room temperature. Based on a core–shell assembly, thiol-modified KH1C12 aptamer was conjugated to the nanoprobe via thiol-en click chemistry. The operative-enhanced Fe₃O₄/GNCs nanoprobe with aptamer exposed fluorescence and high specificity toward target proteins expressed on the surface of HL-60 cancer cells. Therefore, the Fe₃O₄/GNCs/Aptamer nanoprobe was applied to selective detection of the leukemia cell line HL-60 from a mixture.

Zoom In Zoom Out Reset image size

The highly dispersed magnetic NPs were prepared in an aqueous solution based on our previous report [37]. As shown in figures 1(A), (C), highly monodispersed magnetic NPs with an average diameter of 17 ± 4 nm were synthesized with a narrow size distribution as shown in the histogram. This synthesis route produces a gram-scale of hydrophilic Fe₃O₄ NPs that can be redispersed in water [38]. The FESEM image (figure 1(B)) approves the spherical morphology of the Fe₃O₄ NPs. Silica-coated Fe₃O₄ NPs (Fe₃O₄@SiO₂) were prepared by sol-gel reaction with an organosilicon coupling agent (TEOS) [39]. Figures 1(D), (E) shows TEM and FESEM images of the Fe₃O₄@SiO₂ NPs. In figure 1(D), silica shell can be seen as a translucent region around the Fe₃O₄ NPs [40]. According to our previous experiment [37] the thick silica shell can reduce the magnetic property of the Fe₃O₄ NPs and also increases the overall particle size. So, in this experiment, the silica shell thickness was adjusted to be 3 ± 1 nm and the mean particle diameter of the prepared Fe₃O₄@SiO₂ NPs is 20 nm with a narrow size distribution range (figure 1(F)).

Zoom In Zoom Out Reset image size

The GNCs@MPS nanoclusters, as the fluorescent component of the nanoprobe, were synthesized in a facial method. The TEM image and particle size dispersity of the synthesized GNCs are given in figures 1(G), (I). According to the histogram, the nanoclusters demonstrate a narrow size distribution range with a mean particle size of about 3.5 ± 0.5 nm. Fluorescent property of the synthesized GNCs was also studied and the result indicates an emission peak at around 627 nm (λ_ex = 520 nm) (figure 1(H)), which is consistent with previous reportes [24, 41, 42]. To estimate fluorescence quantum yield (QY) of the GNCs, rhodamine 6 G (QY: 0.95 in ethanol) was used as the standard. The results show a QY of about 1.1% for the GNCs at room temperature.

In situ generation of the gold nanoparticles in the presence of high affinity ligands results in formation of small particles, because of the growth quenching of the primary particles by the ligands keeping the size dependent optical property [43]. It is well known that thiols can lead to the formation of GNCs due to the soft-soft interactions between Au and S atoms [18, 43]. In addition, the thiol-capped GNCs represent a long-time emission which reduces auto-fluorescence interferences in biological experiments through time-resolved luminescence techniques [44]. An intermediate of Au(I)-S complex is formed after addition of H[AuCl₄] complex to the MPS solution [45]. In this case, thiol group in the MPS partially reduces the Au(III) ions. Then, to complete the formation of the GNCs, a strong reducing agent (such as NaBH₄) is added to the reaction medium. Conjugation of MPS with the GNCs was evaluated by FT-IR spectroscopy. Figure S1 (supporting information) shows the spectra of MPS and GNCs@MPS. In the MPS spectrum, the presence of two bands at 1112 and 985 cm⁻¹ are assigned to the Si–OH and Si–O–Si groups, respectively [45]. The band of 2500 cm⁻¹ is attributed to the H–S stretching vibration of MPS [46], which is not observed in the GNCs@MPS spectrum confirming the successful conjugation of MPS with the GNCs [45].

Next, to the Fe₃O₄@SiO₂ NPs, a solution of the freshly prepared GNCs was added at room temperature to synthesize the desired nanoprobe through silanization reaction between Si–OH groups of MPS and TEOS [45]. Finally, the PGD was used to functionalize the surface of the nanoprobe for further conjugations. A slight decrease in the fluorescent intensity of the prepared nanoprobe can be seen in figure 1(H) which is related to the quenching effect of the Fe₃O₄ NPs due to some intrinsic properties of the Fe₃O₄ NPs [47]. Figures 2(A), (C) show the TEM image and the size distribution histogram of the Fe₃O₄/GNCs nanoprobe and illustrate that majority of the nanoprobes have particle size of about 26 nm. From the TEM image, it can be clearly seen that some darkened GNCs with a particle diameter of about 3 nm are surrounded on the surface of Fe₃O₄ core NPs. In the TEM image, some aggregated of interconnected particles are appeared. The FESEM image of the Fe₃O₄/GNCs nanoprobe (figure 2(B)) shows that the gold shell decorated on the surface of the Fe₃O₄ NPs and consists of a few adjacent GNCs [48]. The XRD method was used to characterized the crystalline structure of Fe₃O₄/GNCs (figure 2(D)). The lattice spacing estimated from the diffraction peaks located at 30.2, 35.4, 43.5, 53.4, 57.1 and 62.7° well matches with the [220], [311], [400], [422], [511] and [440] Fe₃O₄ planes, respectively. The peaks of the gold crystals are clearly observed at 38.1, 44.2, 64.6, 77.5 and 81.8°, which are assigned to the [111], [200], [220], [311] and [222] gold planes, respectively [33]. The results obtained from XRD patterns indicate the existence of Fe₃O₄ and gold nanocrystals in the nanoprobe structure [33]. A broad peak from θ = 23.66 to 28.99° can be related to the amorphous silica confirming the presence of the silica in the nanoprobe structure [40]. The EDAX data (figure 2(E)), indicate the existence of Cu, Fe, Au, O, C, Si and S elements in the Fe₃O₄/GNCs structure, demonstrating the formation of the nanoprobe. The Cu element can be related to the copper grid used for the preparation of TEM sample. In addition, the magnetic property of nanoprobe (figure 2(F)) was investigated using VSM method. There is not any hysteresis loop in the magnetization curves, indicating a superparamagnetic behavior of Fe₃O₄/GNCs nanoprobes. The VSM curves also show that the magnetization of Fe₃O₄/GNCs decreased from 50 to 30 emu g⁻¹ after wrapping with silica and GNCs layers, respectively [28]. This result shows that the nanoprobe is covered with non-magnetic silica and GNCs layers. In this study, ICP-OES technique was employed to quantitate the metal composition of the Fe₃O₄/GNCs nanoprobe. According to the obtained results, the molar ratio of GNCs:Fe₃O₄ in nanoprobe composition is 54:1. At this optimized ratio, the surface of the Fe₃O₄ NPs is covered uniformly with GNCs.

Zoom In Zoom Out Reset image size

Due to the optical and magnetic properties of the nanoprobe, in the next step, the nanoprobe was functionalized with aptamer for cell imaging and separation. Conjugation of thiol-functionalized KH1C12 aptamer was done via a fast click reaction between the thiol group of the aptamer and the double bond in PGD. Figure S2 shows the two-dimentional structure of the aptamer predicted by bioinformatics software (mfold). First, the conjugation of PGD with the nanoprobe was studied by FT-IR spectroscopy. The strong vibrational bands of PGD are clearly seen at 1622, 1588 cm⁻¹ and 955 cm⁻¹, which are assigned to C=O, C=C and C=C–H of PGD, respectively (figure S3). After conjugation of aptamer with the Fe₃O₄/GNCs nanoprobes via PGD, a new peak at 625 cm⁻¹ appears in the spectrum. This shows that the aptamer was successfully conjugated to the surface of the nanoprobe.

Before the biological application, stability of the prepared Fe₃O₄/GNCs/Aptamer was evaluated using dynamic light scattering in different pH conditions. According to the literature reports, the extracellular pH (6.2–6.9) is more acidic than intracellular pH (7.12–7.65) [37]. So, the stability of the nanoprobe was examined at three different pHs (6.2, 7.4 and 7.6) in PBS^– medium. As can be seen in figure S4, the size of Fe₃O₄/GNCs/Aptamer nanoprobe does not change significantly at these three pHs.

The outstanding application of the Fe₃O₄/GNCs/Aptamer nanoprobe is selective cell separation from a mixture. For this purpose, the biocompatibility of nanoprobe is very important and must be evaluated. In the next step, we explored these features of the Fe₃O₄/GNCs/Aptamer nanoprobe using two cell lines before they were used for in vitro separation application. MTS assay was performed to measure the biocompatibility of the nanoprobe. After incubation with Fe₃O₄/GNCs/Aptamer nanoprobe for 24 h, an MTS viability assay of HL-60 and HepG2 cells was carried out. From figure 3(A) it is observed that the cell viability does not change significantly after treatment with Fe₃O₄/GNCs/Aptamer. These results suggest that the prepared Fe₃O₄/GNCs/Aptamer nanoprobe has excellent cytocompatibility at the tested concentrations.

Zoom In Zoom Out Reset image size

Selectivity of the nanoprobe was investigated by flow cytometry analysis. In this study, HepG2 was selected as a KH1C12 negative cell and the two cell lines were treated with Fe₃O₄/GNCs/Aptamer nanoprobe and the fluorescence of GNCs was used to evaluate the nanoprobe uptake by both cancer cells. According to the obtained histograms (figure 3(B)), in comparison to the untreated cells (Unt cells), a shift toward higher far-red fluorescent intensities in FL-3 channel can be seen for HL-60 as KH1C12 positive cells, while for HepG2 cells, the fluorescent signals generated by nanoprobe and the untreated cells are similar. The results show that Fe₃O₄/GNCs/Aptamer could be selectively taken up by the HL-60 cancer cells. The results obtained from MTS and flow cytometry analyses show that the Fe₃O₄/GNCs/Aptamer nanoprobe is cytocompatible and can recognize the desired cancer cells specifically. ICP-OES was employed to quantitate the amount of nanoparobe taken up by the HL-60 and HepG2 cells. As shown in figure S5 the uptake of the nanoprobe in HL-60 cells is about 10.4 times higher than that in HepG2 cells.

Fluorescent microscopy imaging was utilized to demonstrate the selective uptake of Fe₃O₄/GNCs/Aptamer nanoprobe in a mixture containing equal number of HL-60 and HepG2 cells using fluorescent property of GNCs. Bright-field images of the cells (figure 4(A)) show no changes in the morphology illustrating the biocompatibility of the prepared nanoprobes which is in accordance with the MTS data [41]. The dark-field images show that the fluorescence intensity of the Fe₃O₄/GNCs/Aptamer nanoprobe in the desired cells (HL-60) is significantly higher than that of the control cells (HepG2). Next, for both cancer cell suspensions, the cells were separated by a magnet. Then, both separated and remained cancer cells were characterized using fluorescent imaging. The nanoprobe bound only to HL-60 cells which shows high red emission and not to the HepG2 cells. This experiment also confirms the selectivity of the nanoprobe.

Zoom In Zoom Out Reset image size

To evaluate the sensitivity of our multifunctional nanoprobe with the assistance of fluorescent GNCs, we introduced different concentrations of HL-60 cells to the prepared nanoprobe and monitored their fluorescent intensity and sensitivity by increasing the number of cells. As shown in figure 4(B), the fluorescent intensity enhances along with increasing the number of cells. As the nanoparticles enter to the cells, they may encounter with the biomolecules inside of cells that are able to form corona around them [49, 50]. Corona coating could stabilize nanoparticles and prevents their aggregation that leads to self-quenching of GNCs [51]. Therefore, the fluorescent intensity rises as the number of cells increases. Our result also clearly shows that the sensitivity of the fluorescence assay is about 10 cells.

It is known that most biological environments do not show magnetic background. It provides the opportunity to detect desired targets in biological media using the magnetic nanoparticles. The MRI pictures are shown in figure 4(C). Obviously, the contrast of images declines linearly as nanoprobe concentration increases, demonstrating the potential ability of nanoprobe as a T₂-based contrast agent in MRI applications [41].