The X-ray diffraction pattern of the SEM imaged nanowires (Fig. 1), showed a single Ni phase pattern (Fig. 2) (powder diffraction, X-rays wavelength of CuKa (λ = 1.5418 Å), 2-θ values in agreement with JCPDS No:04-0850). The room-temperature magnetization curve of the nickel nanowires is shown in Fig. 3. The saturation magnetization is 50.7 Am²kg^-1, which is significantly less than the value for bulk nickel (55.4 Am²kg^-1). The EDAX analysis indicated an O/Ni ratio of 1:15, hence it seems likely that the surface of the wires is coated with a layer of nickel oxide which is approximately 3–4 nm thick, with a certain amount of Al₂O₃. The absence of NiO and Al₂O₃ peaks in X-ray diffraction pattern can be explained by the amorphous nature of the oxides. The magnetic moment, m of a typical wire is 1.6 × 10^-13 A m².

The presence of the nanowires made it easy to separate the cells containing a wire from those which did not. The separation experiment was carried out as described above with the MC3T3-E1 and UMR-106 cells. As a result, more than 70 % and 60 % separation purity, defined as the cells captured with nanowires divided by the total number of captured cells, was achieved for MC3T3-E1 cells and UMR cells respectively, which is consistent with the work by Hultgren et. al. 7. The separated cells were then replated and cultured at physiological condition till confluency, up to 3 days, and then divided for further experimental separation. All the primary cells and cell lineages with internalised magnetic nanowires responded with a total cell survival higher than 95 % (live/death viability/cytotoxicity kit, Molecular Probes, USA) up to 5 days after separation.

From the magnetically-separated cell-nanowire colonies single cell manipulation and imaging was achieved by scanning electron and fluorescent microscopy. Figure 4 shows the osteoblast cell line with each cell containing one or more wires. An image taken with the scanning electron microscope (Fig 5) shows a nanowire inside the cell. There, the left cell extends out towards the right cell and uses the wire main axis as an alignment guidance. At the same time, due to the difference in mechanical properties between nanowire and cell, the nanowire introduces an anisotropical stiffening contribution to the cellular internal structures.

To investigate this mechanical re-arrangement in cell cytoplasm evoked by nanowire internalization, fluorescent microscopy was used. To show this difference in cell activity, MC3T3-E1 osteoblast single cell with and without internalised nanowire were co-cultured and stained with living mitochondria and lysosome fluorescent trackers. Figure 6 clearly show the difference in cell organelles distribution. While the normal cell has a clear mitochondrial fluorescent staining due to normal cell activity but no lysosome staining, the cell with a nanowire has a completely different response.

A combined staining for lysosomes and mitochondria localized around the nanowire indicates that the internal organelles inside the cytoplasm respond to the nanowire. In parallel to that localized mitochondrial staining also shows lamellipodia extensions due to cell tethering and re-alignment. This could be associated with a nanowire-induced cell stiffening response.

We showed that after separation and replating to confluence half of the cells inside each field of view contain wires and they are clearly distinguishable, Figure 7. Here, it was important to show that UMR106 cells have the same size as the nanowires and that we succeeded in the internalization process. This was achieved by the external magnetic field stimulation of the nanowires and subsequent orientation of small colonies of UMR106, as shown in Fig. 7B,7C.

To extend this finding to a larger statistical population, 5 batches, and 100 fields of view were analyzed for each group studied. In total several thousand cells were counted. The quantitative results achieved for the osteosarcoma cells were similar to those reported for the primary MSCs.

Marrow Stromal (MS) cells were successfully separated and isolated in small colonies to further investigate the internal cell cytoskeleton re-arrangement. Small colonies and single MS cells were stimulated and recorded by phase microscopy. Figure 8 shows both adherent cells and floating cells with internalized nickel nanowires.

The MS cells were time lapse imaged for a total period of 12-hours under phase contrast microscope, which was set so that there was a progressive heating of the culture medium, which led successively to cell detachment from the substrate and subsequent programmed cell death [see Additional file 1]. The video shows floating cells ingesting a nanowire, as well as adherent cells containing a nanowire detaching from the substrate, and eventually undergoing programmed cell death. The dead cells subsequently release part of the cytoplasm and the internalised nanowire.

The alignment experiments of primary MS cells were aimed to achieve orientation in an external magnetic field. This was achieved for MS cells in suspension where manipulation and cell-to-cell bridging was reported [see Additional file 2]. These were carried out in parallel to the UMR106 osteosarcoma cells. As mentioned for the UMR106, 5 batches and 100 fields of view were analyzed for each group of study. In total, several thousand cells were counted. The histograms shown in Figure 9 and 10, plot the angular distribution of the nanowires and cells, in the absence and presence of the 100 mT magnetic field. The data are fitted to a Gaussian distribution.

While it can be seen that there is a clear orientation of the wires in the applied field, the orientation of the cells themselves is barely significant. Several experiments were carried out to investigate the extent of cell alignment versus the anisotropic orientation of the wires.

Nickel nanowires were grown in alumina membranes (Whatman, UK). The membranes used were 20 mm in diameter and 60 μm thick, with 200 nm parallel-pores spaced by 300 nm. Firstly a gold electrode was deposited on the back of the membrane 13. Then, nickel was deposited into the pores by electroplating from a NiSO₄ bath, at a potential of -0.9 to -1.0 V relative to an Ag, AgCl/KCl reference electrode 13. The nickel wires were then separated from the membrane by dissolving the alumina membrane in 3 M·l^-1 NaOH 14. The nanowires were then characterized by scanning electron microscopy (Fig. 1), X-ray diffraction (Fig. 2), X-ray fluorescence and vibrating-sample magnetometry (VSM).

In this study three cell lines were used: primary cells from adult rat marrow stromal cells (MSC), MC3T3-E1 osteoblasts cell line and rat osteosarcoma cells (UMR106) (ATCC, USA). Every cell type used in this study was cultured under a different environment and culture medium, in line with the specific cell-culture requirements.

Primary MSCs were cultured and incubated at 37°C, 5 % CO₂ and 95 % relative humidity, to confluency with DMEM supplemented with 2 % penicillin/streptomycin (Gibco, UK), 10 % foetal Bovine Serum (FBS; Gibco, UK), 0.5 % l-Glutamine (Gibco, UK), 0.5 % Glutamax (Gibco, UK), and 1 % non-essential amino acids (Gibco, UK).

For the UMR 106 cell line, all samples were cultured and incubated as above with DMEM (30-2002 ATCC, USA) supplemented with 10 % foetal bovine serum, 100 U/ml penicillin and 100 μg/ml streptomycin.

Whereas, MC3T3-E1 were cultured in alpha minimum essential medium (α-MEM: Sigma-Aldrich, UK) containing 10 % foetal bovine serum (FBS; Gibco, UK), 100 U/ml penicillin and 100 μm/ml streptomycin, 2 mM L-glutamine (Gibco, UK) and 0.5 % 100 mM sodium pyruvate (Sigma-Aldrich, UK). The osteoblasts were kept in an incubator at 37°C in a 95 % air 5 % CO₂ environment. Subculture was routinely performed at ~80 % confluence, using 0.25 % trypsin.

Nickel nanowires were firstly washed 5 times in deionized water (Ω <= 30 M ohm) and rinsed in different PBS baths. The nanowires were then stabilized for 2 hours into the specifics cell culture medium used for each cell line with a density of 10¹¹ particles/ml and then dispersed by 5 minutes ultrasonic agitation immediately prior to addition. Cell medium-nanowire solution was then added at a final concentration of 10⁶ particles/ml to adhering and suspended cells. Adhering cells were incubated for 30 minutes to promote the internalization of the wires and then incubated overnight. Suspended cells were incubated for 30 minutes and then exposed to magnetic field. After each magnetic exposure, cells on coverslips were subsequently fixed or stained as required for the imaging technique adopted.

Each experiment was repeated several times as part of a larger study to address the extent of nanowires internalization and cell yield.

To characterize the ingestion of nanowires into the cells several imaging and immunofluorescence techniques were used. These were optimized for each cell type to examine the interaction between the cells and the nanowires. In this study all the three cell types were selectively stained for tracking active mitochondria and acidic organelles in live cells to highlight the cell activity during internalization.

The cell staining and preparation was carried out by staining live cells for cell tracking. In this study two cytoskeleton markers were Myto- and Lyso-tracker (Molecular Probes, USA). In brief, after having prepared a 1 mM stock solution of each tracker, a final concentration diluted in the respective growth medium (described above) of 50 nM for Myto-tracker and 50 nM for Lyso-tracker was respectively used to stain samples. Single and double staining on adherent and suspended live cell were carried out for 30 min for Lyso-tracker and 3 min for Myto-tracker; samples were then washed with fresh warm medium and imaging were taken. Subsequent, to live imaging the stained cells were permeabilised and fixed in 4.1 % paraformaldehyde in PBS for 30 min. These concentrations were kept as low as possible in order to reduce possible overloading and artefacts.

Phase contrast, bright-field and time lapse imaging microscopy were carried out with the use of a Zeiss inverted microscope (Zeiss Axiovert 200 M, Germany) connected to a cooled CCD camera (Zeiss Axiocam HRm, Germany).

SEM imaging was carried out on adherent MC3T3-E1 and MSC cells with internalized wires. All cell samples were prepared by fixing each glass coverslip slide with 3 % Glutaraldehyde in 0.1 M Sodium Cacodylate buffer (pH 7.2). The primary fixation was carried out for 1 hour at room temperature. Samples were then washed 6 times for 1 hour with either 0.05 M Phosphate or 0.1 M Cacodylate buffer to remove any un-reacted Glutaraldehyde from the samples before rinse and dehydration was carried out. The samples were then quickly rinsed with 50 % ethanol; dehydrated with 10 %, 30 %, 50 %, 70 %, and 95 % ethanol for 10 min each; and dehydrated with 100 % ethanol twice for 15 min each. The samples were then gold sputter-coated for high resolution imaging.

The ingestion of nanowires into the cells was examined by time lapse contrast microscopy. It was possible to examine the interaction of the cells with the nanowires by programming the Zeiss microscope to acquire an image every 15 minutes for 18 hrs (as reported in the supplemental material video). There, the light was programmed to heat the medium so that both adherent and floating cells could be examined.

The cell separation set-up was made of two Nd₂Fe₁₄B permanent magnets of 10 mm diameter and 15 mm length in antiparallel configuration at opposite sides of a 10 ml falcon tube for cell suspension. The surface magnetic field of the magnets of 0.6 T was measured by hall sensor gaussmeter (Hirst, UK). The two magnets fixed on either side of the tube together produced a magnetic field gradient up to 100 T/m.

Cells cultured with nanowires were dissociated with warmed trypsin plus EDTA4Na solution (Sigma-Aldrich, UK) for 5 min. From each full batch half was put into the magnetic separation set-up, and the other half was replated with 3 ml of fresh medium to estimate the amount of cell binding with nanowires pre-separation.

The magnetic separation was carried out for 5 minutes in order to obtain the complete sedimentation of those cells without nanowires. These last, after separation were removed and plated in 3 ml fresh medium in a new Petri dish. After having removed the magnets, the cells with nanowires were dispersed into 3 ml of fresh medium and replated in a new Petri dish.

Complete cell adhesion was achieved in less then 1 h of incubation at physiological conditions for all the batches and then cell counting was carried out in 2–3 rounds to minimize systematic and operator errors. Images were captured using an Olympus BX41 microscope (Japan) with magnifications of 10×, 20×, and 40× connected to a cooled CCD camera (Image II, USA). All images were acquired and stored by using Analysis imaging software (Analysis, Germany).

For the nanowire and cell alignment test, a uniform magnetic field of 100 mT was applied by a cylindrical Halbach permanent magnet made of segments of Nd₂Fe₁₄B (Magnetic Solutions, Ireland). The inner and outer radii of the magnet were 106 and 156 mm, respectively. Cells with Ni nanowires were co-cultured overnight before the test. The duration of the magnetic field exposure was chosen to be 1 hr, 24 hrs or 48 hrs inside an incubator at physiological conditions (T = 37°C, CO₂ = 5 % and RH = 95 %). Subsequently, images were taken under phase contrast as described above. The degree of alignment of the wires and cell morphology to the direction of the field was conducted by using Scion Image software (Scion Corporation, USA). Both cell and nanowire counting were carried out under blind conditions to reduce systematic errors.