The use of magnetite-doped polymeric microspheres in calibrating cell tracking velocimetry

https://doi.org/10.1016/S0165-022X(00)00085-3Get rights and content

Abstract

Continuous magnetic separation, in which there is no accumulation of mass in the system, is an inherently dynamic process, requiring advanced knowledge of the separable species for optimal instrument operation. By determining cell magnetization in a well-defined field, we may predict the cell trajectory behavior in the well-characterized field environments of our continuous separators. Magnetization is determined by tracking the migration of particles with a technique known as cell tracking velocimetry (CTV). The validation of CTV requires calibration against an external standard. Furthermore, such a standard, devoid of the variations and instabilities of biological systems, is needed to reference the method against day-to-day shifts or trends. To this end, a method of synthesizing monodisperse, magnetite-doped polymeric microspheres has been developed. Five sets of microspheres differing in their content of magnetite, and each of approximately 2.7 μm diameter, are investigated. An average gradient of 0.18 T/mm induces magnetic microsphere velocities ranging from 0.45 to 420 microns/s in the CTV device. The velocities enable calculation of the microsphere magnetization. Magnetometer measurements permit the determination of magnetization at a flux density comparable to that of the CTV magnet’s analysis region, 1.57 T. A comparison of the results of the CTV and magnetometer measurements shows good agreement.

Introduction

The separation of subsets of cells depending on their surface expression of specific molecules has broad application in biological and medical practice [1], [2]. To achieve these separations, a host of immunological methods have been developed, including immunofluorescent, immunomagnetic and immunomatrix methods. Due to recent advances in magnetic bead synthesis, immunomagnetic cell separation is gaining rapidly in popularity. Monodisperse colloidal immunomagnetic beads (or ferrofluids), much smaller in size than cells, enable magnetic labeling proportionate to the number of expressed antigens. Thus, distribution in molecule expression gives rise to a proportional distribution in cell mobility when the suspension of cells is exposed to a magnetic field gradient [3]. This variation in mobility is the principal means of sorting cells into two or more fractions in a continuous process [4]. We have devised continuous cell sorters and fractionators at The Cleveland Clinic Foundation, in collaboration with The Ohio State University, which take advantage of mobility distributions [5], [6], [7], [8].

To characterize the efficacy of immunomagnetic labeling as well as to assist the operators of continuous, magnetic cell-sorting devices, there is a need for measuring cell mobilities. The cell tracking velocimetry (CTV) technique, developed at The Ohio State University, measures cell (or particle) mobilities by computing their velocities as they migrate in a magnetic field of constant energy gradient [9]. Currently, this technique is semi-automated and capable of tracking hundreds to thousands of cells (or particles) on an individual basis. The data are then used to generate frequency versus mobility (increment) histograms of immunomagnetically labeled cell populations.

Our initial studies using the CTV technique are promising, with results that are consistent with expectations. However, as with any analytical instrument, calibration of the instrument’s performance against independently known standards is essential. Labeled cells are unsuitable as standards because of their instabilities, the difficulties of the labeling process, the variations from one population to another, and the lack of any means of predicting their magnetizations beforehand.

Ideal calibration particles should have the following properties. (1) They should consist of spheres with a high degree of sphericity, monodisperse in size with diameters similar to our cell models. (2) They should consist of a pure homogeneous material that is stable in aqueous media. (3) The constituent material should be paramagnetic with no saturation magnetization up to about 2 Tesla. (4) The magnetic susceptibilities should be widely tabulated and accurately known, alleviating the need for user testing and the associated uncertainties, and the range of susceptibilities should roughly coincide with that of our labeled cell populations. (5) They should have densities not too dissimilar from that of water, preventing the problems of rapid sedimentation in the CTV system.

Such ideal particles have not been identified; however, some monodisperse magnetic microspheres have been synthesized that meet many of these criteria [11], [12]. These monodisperse microspheres are good candidates for calibration purposes for several reasons. (1) The microspheres have narrow distributions in diameter, in density, and in magnetization, reducing the dispersions in migration and sedimentation velocities; see , , below. (2) The velocities of the four magnetite-containing samples have a range from approximately 0.001 to 1 mm/s, comparable to the range observed for labeled cells. (3) The size and refractive index of the microspheres make visualization easy, without any modification of the CTV apparatus. (4) Compared to cells, they are stable, reproducible, and require no pre-processing.

Section snippets

Theory

In this document, ‘volumetric magnetization’ and ‘magnetization’ are used interchangeably to mean volumetric magnetization. For a particle with inducible magnetization (paramagnetic), M, in an external field, B, the force, Fm, on the particle is given by the gradient of the magnetostatic potential energy, ∇Um, as [13]Fm=−∇Um=12 V∇(M·B)where the particle volume is V. If we assume that the particle is free to rotate to the position of lowest potential energy, so that M and B are parallel, and

Results

Fig. 4 presents magnetization curves for the four silica- and magnetite-coated polystyrene microspheres. Intrinsic magnetization per unit mass of dry sample is determined by the vibrating sample magnetometer in 200 Gauss (15,915 A/m) intervals, from 15,000 to −15,000 Gauss (1.19×106 to −1.19×106 A/m). Two measurements are performed for each microsphere, with average values plotted. The curves are symmetrical about both axes, indicating a lack of hysteresis, which characterizes non-ferromagnetic

A simplified description of the method and its (future) applications

For measuring mobilities of labeled cells for sorting in our continuous cell separation devices, we employ an apparatus and method called cell tracking velocimetry. A new kind of monodisperse magnetic microsphere has been synthesized for the purpose of providing a calibration tool for CTV. The microspheres have a range of magnetizations comparable to those of immunomagnetically labeled cells. The CTV method predicts particle magnetizations that are in good agreement with independent results

Acknowledgements

This work is supported by grants from the NIH (number CA62349 to M.Z.), from the U.S.–Israel Binational Science Foundation (number 96-00486 to S.M. and M.Z.), and the NSF (number BES-9731059 to J.J.C.).

References (23)

  • L. Sun et al.

    Continuous, flow-through immunomagnetic cell sorting in a quadrupole field

    Cytometry

    (1998)
  • Cited by (52)

    • Basic magnetic field configurations

      2007, Laboratory Techniques in Biochemistry and Molecular Biology
      Citation Excerpt :

      The isodynamic field is generated by two pole pieces of hyperbolic cross section (Fig. 4.16) and is limited to a relatively small fraction of the gap volume (Fig. 4.17). The condition F =const is typically maintained to within 1% of its average value over the isodynamic field domain [24, 25]. An ideal quadrupole field is strictly a linear function of spatial coordinates, B =B(r) = αix + βjy + γkz, where α, β, and γ are constants; r = [x,y,z] is the position vector; and i, j, and k are the unit vectors along the Cartesian coordinates axes 0x, 0y and 0z, respectively.

    • New challenges and opportunities

      2007, Laboratory Techniques in Biochemistry and Molecular Biology
      Citation Excerpt :

      A typical leukapheresis sample CD34 analysis result by FCM is shown in Fig. 12.10. Magnetophoretic mobility measurements were made on an instrument referred to as a CTV (reviewed in Chapter 8) [72, 73]. The schematics of CTV are shown in Fig. 8.1.

    View all citing articles on Scopus
    View full text