Magnetite nanoparticles have been produced by a previously reported co-precipitation method (see Materials and Methods) [²⁶,²⁷]. Application of a silica layer was achieved by following a method reported by Philips and co-workers [²⁸]. Briefly, a colloidal solution of magnetite nanoparticles in tetramethylammonium hydroxide (TMAH) was treated with sodium silicate in order to deposit a thin layer of silica on the surface of the oxide particles (Figure 1). In a separate step, a carboxylic acid protoporphyrin (protoporphyrin IX) was reacted with 3-aminopropyltriethoxysilane (3-APTES) under inert conditions in the presence of the carbodiimide coupling agent (EDCI) to form an amide bond. This modified porphyrin was then reacted with the silica coated particles, to form a stable colloidal suspension. In this case the silica shell was necessary to provide an effective barrier between the particle core and the fluorescent porphyrin, preventing the risk of quenching.

Elemental analysis of the PMNC shows a content of C - 9.31, H - 1.21 and N - 1.51%. This clearly confirms the presence of conjugated porphyrin molecules at the surface of nanoparticles.

The presence of the porphyrin on the surface of the silica-coated magnetite nanoparticles was also confirmed by Fourier Transform Infrared Infra-red (FTIR) spectroscopy (Figure 2). FTIR spectra of the silica coated magnetite nanoparticles before and after reaction with the protoporphyrin were recorded in KBr. The silica coated particles (shown in heavy black circles above) have stretches at 579, 1087, 1625 and 3286 cm^-1 corresponding to the Fe-O, Si-O-Si, and water stretches respectively. The stretch found at 947 cm^-1 is due to the presence of Fe-O-Si bonds in the sample. For the nanocomposite material, again Fe-O and water stretches are found at 579, 1621 and 3286 cm^-1 respectively. The Fe-O-Si stretch is this time masked by the NH wagging from the porphyrin centred at 901 cm^-1. NH stretches are found at 1473 cm^-1. Aromatic and aliphatic stretches associated with the porphyrin are found at 2500 and 2945 cm^-1 respectively. FTIR spectra (Figure 2) of the porphyrin nanocomposite material showed stretches for NH due to the presence of the porphyrin at 901 and 1473 cm^-1. Aromatic and aliphatic stretches associated with the porphyrin are found at 2500 and 2945 cm^-1 respectively. A core-shell-like morphology is apparent from TEM images taken of the silica coated nanoparticles. The average primary particle size is calculated to be 9.3 ± 1 nm. The silica coating on the particles has an average thickness of 3 nm, however upon reaction with the 3-APTES modified protoporphyrin IX, the coating thickness is found to slightly increase to 4 nm.

Absorbance spectra of the original porphyrin and the PMNC were taken in tetrahydrofuran, THF (Figure 3). Peaks at 406, 505, 538, 577 and 632 nm were observed for a solution of porphyrin. Emission (λ_ex, 406 nm) and excitation spectra (λ_em was 633 nm) have also been recorded, with significant broadening witnessed for the PMNC when compared to the original porphyrin. This can be attributed to the scattering of the incoming light by the PMNC suspension, but gives further indication of the binding of the porphyrin to the nanoparticle surface.

Dynamic Light Scattering (DLS) studies of silica coated magnetite nanoparticles have shown a very broad size distribution with an average hydrodynamic diameter of 131 nm, due to aggregation of nanoparticles. However, silica coated porphyrin functionalised magnetic nanoparticles have demonstrated relatively narrow distribution with an average hydrodynamic diameter of 818 nm (Figure 4).

TEM images taken of the silica coated nanoparticles reveal the presence of a core-shell structure (Figure 5). The silane coating on the particles has an average thickness of 3 nm and the particles are found to aggregate in group of approximately ten particles. Upon reaction with the 3-aminopropyltriethoxysilane modified protoporphyrin, the coating thickness is found to slightly increase to 4 nm.

Magnetisation measurements of original non-coated magnetite nanoparticles, silica coated magnetite nanoparticles and silica-coated porphyrin functionalised magnetite nanoparticles as expected, have shown a gradual reduction in saturation magnetisation values (58.4 Am²/kg for pristine magnetite nanoparticles, 35.2 Am²/kg for silica coated magnetite nanoparticles and 27.5 Am²/kg for the corresponding porphyrin conjugates) with an increasing amount of non-magnetic material, i.e. silica and porphyrin conjugates (Figure 6).

The combination of both magnetic and fluorescent properties in one nanocomposite can provide additional benefits. For example, these nanocomposites could serve as multimodal assays for in vitro- and vivo- bioimaging applications such as MRI and fluorescence microscopy. They can also be used as bimodal agents for anticancer therapy, encompassing photodynamic and hyperthermic capabilities. Fluorescent-magnetic nanocomposites may also be utilised as a multimodal diagnostic and therapeutic tool, which could be used, for example, to identify, diagnose and simultaneously treat various diseases. Other potential exciting applications of these nanocomposites include cell tracking, cytometry and magnetic separation, which could be easily controlled and monitored using fluorescent microscopy and MRI.

Previously, the cytotoxic effects of poly(ethylene glycol)-co-fumarate (PEGF)-coated magnetite nanoparticles and their capability for changing cell culture medium compositions have been investigated. The primary mouse connective tissue cells (L929) were used for this purpose. Toxicity levels and changes in the cell culture medium compositions were determined using the MTT assay and a UV/vis spectroscopy method, respectively. As compared to the conventional method, a more reliable method for determining the cytotoxicity of nanoparticles for in vitro applications is found [²⁹]. It also has been demonstrated that the incorporation of anticancer drugs such paclitaxel into PLGA nanoparticles in nanoparticles strongly enhances the cytotoxic effect of the drug as compared to Taxol, this effect being more relevant for prolonged incubation times [³⁰,³¹].

In our studies THP-1 (human acute monocytic leukaemia) cells were cultured in the presence of phorbol 12-myristate 13-acetate (PMA) for 3 days to initiate monocyte to macrophage differentiation. Half the volume of complete culture media was then replaced with serum free media to "starve" the cells, allowing for an enhanced up-take of the PMNC. Figure 7 shows a confocal image of a THP-1 cell after incubation with the PMNC for 3 hours. The particles are localised in the cytoplasm, as detected by confocal imaging, and the image is representative of the broader cell population.

During the experiments we have found that following an exposure time of only 482 ms, the PMNC begins to photobleach, and after only 4 seconds, the fluorescence intensity has greatly decreased (Figure 8). This process is highly undesirable as it makes the generation, capture and interpretation of the image extremely difficult and does not allow for a full examination of the cells to determine precisely the nature of the nanocomposite localisation.

In an effort to alleviate the photobleaching, β-mercaptoethanol (β-ME) was employed to counteract this problem, and was added once the cells were fixed. Control experiments performed with THP-1 cells in the presence of β-ME show that no deleterious effects are witnessed and the cells retain their size, shape and membrane integrity (Figure 9). Addition of β-ME to cells cultured with the PMNC prevented this photobleaching effect and resulted in a stronger fluorescence signal compared to those without β-ME (see Figures 9 and 10). Following the same cell preparation and co-incubation times, we can clearly see not only an increased resistance to photobleaching, but the addition of β-ME also resulted in an enhanced and extended fluorescence emission. This improved photostability is due to the β-ME acting as a scavenger for singlet oxygen, which is traditionally attributed to the photobleaching of porphyrin based dyes.

To further exemplify the enhancement in fluorescent emission intensity due to the presence of the β-ME, pseudocoloured images of the cells were generated (Figure 10). The pseudocoloured images reflect the integrated fluorescence intensity levels across the cytoplasmic and nuclear compartments of the THP-1 cells. Background fluorescence appears blue and predominates for those cells without any nanocomposite added (panels 10 c and d) but also for cells cultured with the PMNC alone (panel 10 a). A dramatic change is witnessed upon addition of the β-ME (panel 10 b). The PMNC is no longer quenched so dramatically and readily emits with substantial emission increases recorded after 4 seconds.

To verify that the increased emission is due to the PMNC and not as a result of increased cellular autofluorescence, control experiments were also conducted. Panels 10 c and d show untreated and β-ME treated THP-1 cells respectively and clearly illustrate that the level of background fluorescence remains constant throughout the experiment. This proves that autofluorescence does not contribute to the enhanced levels of emission witnessed for those cells cultured with the PMNC in the presence of β-ME.

In order to probe the maximum fluorescence intensity of the PMNC as a function of excitation wavelength and also to monitor the fluorescence intensity across the cell compartments, a series of fluorescence images were recorded over varying wavelengths together with emission spectral mapping of the cell. Cells were cultured with the PMNC with and without the presence of β-ME. Lambda scans recorded of cells cultured without β-ME show a noted decrease in brightness and the combined image is noticeably lower in fluorescence intensity. This is shown quantitatively in the fluorescence emission spectra, which show an almost 20% decrease in fluorescence intensity (Figures 11 and 12).

Finally, to compare the results a series of similar control experiments have been performed on "non-starved" THP-1 cells, which were incubated with serum free media. Confocal and composite images of the cells incubated with fluorescent magnetic nanoparticles have been taken at different periods of time 24 hrs, 48 hrs and 5 days (Figures 13, 14, 15 and 16). The results have demonstrated that there was no cellular uptake of the fluorescent nanocomposites but only binding of the PMNC to the outer cell membranes.

THP-1 cells, a human acute monocytic leukaemia cell line, were obtained from the American Type Culture Collection (ATCC, LGC Promochem, U.K.). Cells were cultured in RPMI 1640 medium with 25 mM HEPES (Gibco, Invitrogen Ltd., Ireland), supplemented with heat-inactivated foetal bovine serum (FBS) (9%), 200 mM/ml L-glutamine (0.9%, 1.8 mM/ml), 10,000 U/ml penicillin (0.9%, 90 U/ml), and 10 mg/ml streptomycin (0.9%, 0.09 mg/ml) at 37°C and 5% CO₂ content.

THP-1 cells were seeded into an 8-well chamber slide, with a final concentration of 2 × 10⁴ cells per well. To induce monocyte to macrophage differentiation, THP-1 cells were cultured in 100 ng/ml phorbol 12-myristate 13-acetate (PMA) for 3 days at 37°C in 5% CO₂. Half the volume of media was then replaced by serum free media (RPMI 1640 with 25 mM HEPES) and the cells were incubated for 2 hours at 37°C in 5% CO₂. The PMNC was then added at a final concentration of 0.25 mg/ml. To another well, β-mercaptoethanol β-ME was added. A control experiment with the Protoporphyrin IX dissolved in mixture of water and tetrahydrofuran (THF) was carried out in parallel using the same conditions, with a final porphyrin concentration of 0.119 mg/ml. Fluorescence images were recorded using a Nikon TE3000 inverted microscope. Confocal imges were collected using a 63× oil immersion objective, laser excitation 561 nm and LP 575 nm filter on a Zeiss LSM Meta-510 confocal microscope.

UV-Vis absorption spectra were recorded using a Cary 50 UV-Vis spectrophotometer. Fluorescence measurements were performed on a Cary Eclipse spectrophotometer. FT-IR spectra were recorded KBr using a Perkin Elmer Spectrum One FT-IR spectrophotometer. TEM images were taken on a Hitachi H-7000 at a beam voltage of 100 kV.