Biophysical differences between chronic myelogenous leukemic quiescent and proliferating stem/progenitor cells

Abstract

The treatment of chronic myeloid leukemia (CML), a clonal myeloproliferative disorder has improved recently, but most patients have not yet been cured. Some patients develop resistance to the available tyrosine kinase treatments. Persistence of residual quiescent CML stem cells (LSCs) that later resume proliferation is another common cause of recurrence or relapse of CML. Eradication of quiescent LSCs is a promising approach to prevent recurrence of CML. Here we report on new biophysical differences between quiescent and proliferating CD34+ LSCs, and speculate how this information could be of use to eradicate quiescent LSCs. Using AFM measurements on cells collected from four untreated CML patients, substantial differences are observed between quiescent and proliferating cells in the elastic modulus, pericellular brush length and its grafting density at the single cell level. The higher pericellular brush densities of quiescent LSCs are common for all samples. The significance of these observations is discussed.

Keywords: Atomic force microscopy; Cell mechanics; Chronic myeloid leukemia; Personalized medicine; Single cell analysis.

Figure 1

A schematic of interaction of a spherical AFM probe with a cell loosely attached to the bottom of a culture dish used in the described model. The cell body treated as a homogeneous elastic material is surrounded by the pericellular (brush) layer. Note that the assumption of entropic behavior of the brush layer is required only for extraction of the brush layer parameters, the equilibrium length L and grafting density N.

Figure 2

An example of processing of the raw AFM curve (Z vs d) recorded on the surface of the quiescent CML stem cells through the brush model. (a, c) fitting of the different regions of raw Z-d curve with Eq.(2). (b) Forces of interaction between the AFM probe and the brush layer obtained by processing raw Z-d curve for the d=55-90 nm (here open circles correspond to the force-separation data, solid line is the model fitting Eq(4)).(d) the best Hertz fitting for the entire range of the nonzero force. Deviation of the model curves from the experimental data demonstrates the limit of applicability of the described brush model and the reference Hertz model.

Figure 2

An example of processing of the raw AFM curve (Z vs d) recorded on the surface of the quiescent CML stem cells through the brush model. (a, c) fitting of the different regions of raw Z-d curve with Eq.(2). (b) Forces of interaction between the AFM probe and the brush layer obtained by processing raw Z-d curve for the d=55-90 nm (here open circles correspond to the force-separation data, solid line is the model fitting Eq(4)).(d) the best Hertz fitting for the entire range of the nonzero force. Deviation of the model curves from the experimental data demonstrates the limit of applicability of the described brush model and the reference Hertz model.

Figure 2

An example of processing of the raw AFM curve (Z vs d) recorded on the surface of the quiescent CML stem cells through the brush model. (a, c) fitting of the different regions of raw Z-d curve with Eq.(2). (b) Forces of interaction between the AFM probe and the brush layer obtained by processing raw Z-d curve for the d=55-90 nm (here open circles correspond to the force-separation data, solid line is the model fitting Eq(4)).(d) the best Hertz fitting for the entire range of the nonzero force. Deviation of the model curves from the experimental data demonstrates the limit of applicability of the described brush model and the reference Hertz model.

Figure 2

An example of processing of the raw AFM curve (Z vs d) recorded on the surface of the quiescent CML stem cells through the brush model. (a, c) fitting of the different regions of raw Z-d curve with Eq.(2). (b) Forces of interaction between the AFM probe and the brush layer obtained by processing raw Z-d curve for the d=55-90 nm (here open circles correspond to the force-separation data, solid line is the model fitting Eq(4)).(d) the best Hertz fitting for the entire range of the nonzero force. Deviation of the model curves from the experimental data demonstrates the limit of applicability of the described brush model and the reference Hertz model.

Figure 3

Quantitative results for cells collected from four patients. Both G₀ and G1 phases are shown. The average and SEM values are plotted for (a) the Young’s moduli, (b) brush length, (c) grafting density, (d) the total size brush (N*L parameter), (e) and effective molecular volume density (N/L parameter).

Figure 3

Quantitative results for cells collected from four patients. Both G₀ and G1 phases are shown. The average and SEM values are plotted for (a) the Young’s moduli, (b) brush length, (c) grafting density, (d) the total size brush (N*L parameter), (e) and effective molecular volume density (N/L parameter).

Figure 3

Quantitative results for cells collected from four patients. Both G₀ and G1 phases are shown. The average and SEM values are plotted for (a) the Young’s moduli, (b) brush length, (c) grafting density, (d) the total size brush (N*L parameter), (e) and effective molecular volume density (N/L parameter).

Figure 3

Quantitative results for cells collected from four patients. Both G₀ and G1 phases are shown. The average and SEM values are plotted for (a) the Young’s moduli, (b) brush length, (c) grafting density, (d) the total size brush (N*L parameter), (e) and effective molecular volume density (N/L parameter).

Figure 3

Quantitative results for cells collected from four patients. Both G₀ and G1 phases are shown. The average and SEM values are plotted for (a) the Young’s moduli, (b) brush length, (c) grafting density, (d) the total size brush (N*L parameter), (e) and effective molecular volume density (N/L parameter).