Magnetic Flow Cytometry

A substantial part of biomedical, chemical and pharmaceutical applications relies on magnetic nanomaterials in combination with multifunctional polymers, lipids or proteins. For example, liposomes or emulsions loaded with magnetic nanoparticles are capable to respond to external stimuli like magnetic fields, which allows the remote control over the chemical species [Sab08]. Magnetic nanoparticles are widely used for chemotherapies [Ale05], in vivo magnetic resonance imaging (MRI) [Doe08], and as labels in immunological tests, i.e. magnetic enzyme-linked immunosorbent assays (ELISAs) [Kou&Ste02].

High throughput biological assays and screenings require a tool which is able to analyze, manipulate, and sort objects, e.g. liposome capsules, emulsions or micro-gels containing known doses of magnetic nanoparticles associated with medications or reagents. A promising route relies on the implementation of magnetically-labeled biochemical species in combination with magnetic sensors [Smi12]. So far, progress is restrained to the mere sensing and counting of magnetic objects [Mön11], while advanced and quantitative analysis, which is the main advantage of e.g. optical flow cytometry [Kru&Nol06], has not yet been explored with magnetic sensors. Apart from detection and analysis, sorting of species is invaluable in diagnostic devices. The sorting of biochemical objects has been dominated by conventional fluorescence-activated cell sorting techniques.

We demonstrated for the first time a magnetic emulsion analyzer (MEA) based on giant magnetoresistive (GMR) sensors which is capable of multiparametric studies and sorting [Lin13]. GMR sensors are highly sensitive elements which are used as an important component in hard disk drives to detect very weak magnetic fields from magnetic bits. Taking advantage of their high sensitivity, we implanted them in the heart of our MEA to as a key element in probing ferrofluid droplets of different size and concentrations of magnetic nanoparticles produced with volumes ranging from 20 nl up to 400 nl – ideally suited for culturing monoclonal cell populations and monitoring bacteria growth. Multiparametric analysis as well as magnetic sorting of ferrofluid droplets of different size are performed to demonstrate the potential of GMR sensors for magneto-cytometry studies. This work as well as the research topic "Rolled up magnetic sensors for fluidics" are carried out in collaboration with Dr. Larysa Baraban (chair "Material Science and Nanotechnology" of the TU Dresden).

Figure 1: (a) Concept of GMR-based droplet analyzer for multiparametric analysis and sorting. Sorting of small (b1) and large (b2) droplets at the T-junction. (c) Magnetoresistive characterization of GMR sensor. Inset: Photograph of the patterned sensor prepared for measurements in 4-point configuration. (d) Real time detection of a train containing small and large ferrofluid emulsion droplets. Multiparametric density plots for the detection of magnetic droplets which are produced with (e) different volume of ferrofluid droplets (concentration: 15 mg/ml) and (f) different concentrations of ferrofluid in droplets (volume: 160 nl). Circles are guide to the eyes.

The concept of MEA is presented in Figure 1a. Water-in-oil emulsion droplets of various sizes with encapsulated ferrofluids are produced in a T-junction geometry. The key component is the GMR sensor, which is composed of a Py/Cu (Py = Ni₈₁Fe₁₉) layer stack coupled at the 2nd antiferromagnetic maximum. The magnetoresistive characteristics (GMR curve and sensitivity) are shown in Figure 1c. A GMR ratio of 11% and sensitivity of up to 26 T^-1 in a magnetic field of 0.9 mT is achieved. The high sensitivity of the GMR sensor and the hysteresis-free sensor response under cycling magnetic field are crucial for the magneto-cytometric detection.

First, we demonstrate the capability of MEA to detect and examine ferrofluid droplets of different sizes and concentrations. Each detection event is represented by a peak (Figure 1d) characterized by its amplitude and the full-width at half maximum (FWHM). These characteristics serve as fingerprints for a ferrofluid droplet, flow conditions and emulsification procedure. For instance, in the case of a fixed sensor-droplet separation, the amplitude is related to the size and concentration of the magnetic material in a droplet, whereas the FWHM contains information about the spread in speed and size of a droplet.

Therefore, a multiparametric density plot can be use to analyze the as-produced droplets in terms of size and magnetic content by using amplitude and FWHM as parameters. As an example, multiparametric density plots for droplets which are produced with different volumes and concentrations are shown in Figure 1e and 1f, respectively. Each distinct cloud represents droplets produced under a specific flow condition, which can be used to examine the flow parameters inversely. Droplets with increasing volumes are characterized by an increasing displacement of cloud center towards higher amplitude, while when droplets are longer than the sensor dimension, the information can still be extracted from FWHM (Figure 1e). For droplets produced with different ferrofluid concentrations, droplets clouds are shifted towards higher amplitude while FWHM maintains constant for the same droplet volumes (Figure 1f).

The magnetic droplet analyzer can be extended towards sorting emulsions with respect to a specific property addressed in a multiparametric analysis. For demonstration purposes, we performed sorting of droplets of different sizes, which provide various readouts of the GMR sensor (Figure 1d). Small and big droplets were produced in a continuous mineral oil flow using two separated T-junctions. The small droplets (volume: 20 nl) cause an average voltage change of about 7 µV, while the big droplets (volume: 500 nl) provide an average positive voltage change of about 30 µV. Droplets of different size can be sorted into two separate channels via controlling the opening of an isolation valve when droplets are crossing the sensor (Figure 1-b1, b2). It is foreseeable that various types of emulsion droplets can be sorted with this tool by implementing additional sorting criteria, e.g. dose of the encapsulated magnetic nanoparticles. The demonstration of magnetic-activated sorting opens up promising perspectives for drug design and screening, which could be beneficial to the pharmaceutical industry.

References:

[Sab08] Sabaté, R. et al., International Journal of Pharmaceutics 347, 156 (2008).
[Ale05] Alexiou, C. et al., Journal of Magnetism and Magnetic Materials 293, 389 (2005).
[Doe08] Doerr, A., Nature Methods 5, 668 (2008).
[Kou&Ste02] Kourilov, V. & Steinitz, M., Analytical Biochemistry 311, 166 (2002).
[Smi12] Smith, E. J. et al., Lab Chip 12, 1917 (2012).
[Mön11] Mönch, I. et al., ACS Nano 5, 7436 (2011).
[Kru&Nol06] Krutzik, P. O. & Nolan, G. P., Nature Methods 3, 361 (2006).
[Lin13] Lin, G. et al., Scientific Reports 3, 2548 (2013).

Read more on Rolled-up Sensors for Fluidics