Review
Stiffness of normal and pathological erythrocytes studied by means of atomic force microscopy

https://doi.org/10.1016/j.jbbm.2005.11.003Get rights and content

Abstract

During recent years, atomic force microscopy has become a powerful technique for studying the mechanical properties (such as stiffness, viscoelasticity, hardness and adhesion) of various biological materials. The unique combination of high-resolution imaging and operation in physiological environment made it useful in investigations of cell properties. In this work, the microscope was applied to measure the stiffness of human red blood cells (erythrocytes). Erythrocytes were attached to the poly-l-lysine-coated glass surface by fixation using 0.5% glutaraldehyde for 1 min. Different erythrocyte samples were studied: erythrocytes from patients with hemolytic anemias such as hereditary spherocytosis and glucose-6-phosphate-dehydrogenase deficiency patients with thalassemia, and patients with anisocytosis of various causes. The determined Young's modulus was compared with that obtained from measurements of erythrocytes from healthy subjects. The results showed that the Young's modulus of pathological erythrocytes was higher than in normal cells. Observed differences indicate possible changes in the organization of cell cytoskeleton associated with various diseases.

Abbreviations

AFM
atomic force microscopy
RBC
red blood cell
G6PD
glucose-6-phosphate-dehydrogenase
PBS
phosphate-buffered saline

Keywords

Erythrocyte stiffness
Hemolytic anemias
Atomic force microscopy (AFM)

1. Introduction

Atomic force microscopy (AFM) has become a fundamental tool in several fields of research, such as surface science, biochemistry, and biology. It allows high-resolution topography studies of materials, as well as obtaining mechanical information about biological systems in their natural environment [1]. AFM uses a sharp probe (tip) mounted at the end of a flexible cantilever that deflects when interacting with the sample surface. The sample can be scanned in X and Y directions by means of a piezoelectric scanner. Changes in the height (Z direction) due to the tip interactions with substrate are detected by a setup consisting of a laser and a position sensitive detector (i.e. photodetector).
AFM has been used to study the topography and mechanical properties of living and fixed cells [2], [3]. AFM investigations of living cells and cellular structures provided important information about the viscoelastic properties of the cell membrane and of the cell cytoskeleton organization. Information about micromechanical properties is important for cellular systems since it helps to understand cell architecture and its functions. For example, AFM was used to monitor the influence of chitosan on stiffness of normal and cancer cells [4], where the increase of Young's modulus was observed in cancer cells treated with different types of chitosan preparations. Furthermore, the AFM was also applied to trace certain dynamic cellular processes as cell growth, exocytosis and endocytosis [5]. The first application of AFM in medical diagnostics presented by Zachèe in 1992 [6] where changes in shape of RBCs were observed in patients after splenectomy.
Red blood cells (RBCs, erythrocytes) are a major component of blood. Their function is to transport oxygen to all parts of the body since they contain hemoglobin, a protein able to carry oxygen. The RBC membrane consists of various lipid and protein components. About 52% of the membrane mass consists of proteins, 40% – lipids and 8% – carbohydrates. RBCs are disk-shaped and biconcave and their mean diameter is approximately 7 μm [7].
Hemoglobin is a hetero-tetramer, consisting of two alpha (141 residues present in human hemoglobin) and two beta (146 residues) chains. Hemoglobin is found in erythrocytes where it is responsible for binding oxygen in the lung and transporting the oxygen throughout the body where it is used in aerobic metabolic pathways. Although the secondary and tertiary structures of various hemoglobin chains are similar, reflecting extensive homology in amino acid composition, variations in amino acid composition that do exist impart marked differences in the ability of hemoglobin to transport oxygen. In addition, the quaternary structure of hemoglobin leads to physiologically important allosteric interactions between the chains [8].
Anemia is characterized by abnormally low number of erythrocytes in blood and/or decreased amount of hemoglobin and sometimes also by its altered structure. Anemia may be caused by variety of pathologic conditions, including such congenital diseases as hereditary spherocytosis, thalassemia and glucose-6-phosphate-dehydrogenase (G6PD) deficiency.
Hereditary spherocytosis is characterized by abnormal shape of RBCs [9]. Erythrocytes in patients with hereditary spherocytosis have pathological cell membrane and cytoskeleton. They are smaller than normal RBC. From molecular point of view, hereditary spherocytosis is characterized by a deficiency of certain proteins as α- and β-spectrin, ankyrin, protein 4.1 that are structural proteins of RBC cytoskeleton. Deficiency of these proteins results in cytoskeleton disruption or reorganization that is manifested by the membrane instability forcing the cell to occupy the smallest volume i.e. a sphere. Therefore, the rounded erythrocytes (spherocytes) have impaired deformability and are more prone to hemolysis than normal RBCs [10].
Thalassemias are a complex group of hereditary blood diseases [11]. Thalassemias involve impaired production of α- or β-chains of hemoglobin. There are two basic types of thalassemias: α-thalassemia linked with mutations of chromosome 16 and β-thalassemia where chromosome 11 is implicated. They are named after the pathological hemoglobin chains –α or β. Thalassemias are associated with changes of shape of RBCs due to alterations of α- or β-hemoglobin chains that cause certain cytoskeleton modifications [12].
G6PD deficiency is a hereditary disease characterized by one or more defects in the gene coding the enzyme, glucose-6-phosphate dehydrogenase [13]. It can cause hemolytic anemias varying in severity from severe persistent to very mild intermittent anemia or even be asymptomatic. G6PD is present in all human cells but is particularly important to RBCs since it is involved in the production of reduced glutathione which protects erythrocytes against oxidative stress [14]. Thus, when G6PD is defective, the oxidative injury to RBCs may cause hemolysis.
The aim of this work was to study elastic properties of erythrocytes from patients with different types of anemias using AFM and to compare results with those obtained in normal cells. Additional comparison was performed for anisocytic erythrocytes since the alteration of the erythrocyte shape can be reflected by changes of Young's modulus. Erythrocytes were attached to the poly-l-lysine-coated glass surface by fixation using 0.5% glutaraldehyde for 1 min. Quantitative characterization of each type of pathological erythrocytes was performed by the analysis of Young's modulus values which can be related to the cytoskeleton properties. These results suggest a possible new analytical approach to erythrocyte pathologies.

2. Materials and methods

2.1. Immobilization of erythrocytes on a glass surface

Erythrocytes were collected from complete blood samples taken from children treated at the Department of Haematology on outpatient basis. The RBCs were attached to the glass surface which was previously modified with poly-l-lysine (Super Frost + glass slides, Sigma) solution. Poly-l-lysine assures the attachment of living erythrocytes on the glass surface due to the electrostatic interactions between negative charged erythrocytes and the positively charged poly-l-lysine surfaces. The following procedure of immobilization of cells onto glass was used. First, the fresh blood was diluted in phosphate-buffered saline (PBS, Sigma) in the ratio of 1:1. Next, the solution was placed on the glass substrate (two drops of blood solution) for 45 min. Afterwards, one drop of 0.5% glutaraldehyde solution was added for 1 min. Finally, such pretreated cells were washed with PBS in order to remove the unbounded cells and glutaraldehyde.

2.2. Atomic force microscopy

Erythrocyte's stiffness measurements were carried out using AFM (type CP, Veeco) working in contact mode and equipped with a “liquid cell” setup. The standard silicon nitride cantilevers with the spring constant of 0.03 N/m were used. The tip radii were checked by using a standard TGT01 silicon grating (NMDT, Moscow) and were in the range from 40 nm to 50 nm (Fig. 1). All measurements were performed in phosphate-buffered solution (PBS) at room temperature.
  1. Download: Download full-size image

Fig. 1. (A) Image of AFM tip obtained on the standard TGT01 and (B) a profile along scan line showing the three images for the same scanning tip.

2.3. Cell stiffness determination

Local elastic properties of erythrocytes were quantitatively determined from the force versus distance curves. The force curves were collected from the central part of the erythrocyte (Fig. 2). For each measurement around 200–300 force curves were recorded.
  1. Download: Download full-size image

Fig. 2. (A) The three-dimensional topographical image of RBC measured by AFM. (B) The relationships between the force and indentation depth obtained for the same RBC at different locations.

The Young's modulus was calculated using the Hertz model describing the elastic deformation of the two bodies in contact under load. This theory was extended by Sneddon assuming an appropriate shape of indenter deforming elastic half space [15]. It characterizes the relationship between the applied force and the indentation depth [16]. When the shape of the AFM tip may be approximated by a paraboloid, the force as a function of indentation is described by the following equation:(1)F=43REΔz32where Δz is the indentation depth, R is the tip radius of curvature, E′ is the reduced Young's modulus of the tip-sample system defined as(2)1E=1μtip2Etip+1μsample2Esamplewhere Etip, Esample are Young's moduli of the tip and the sample, μtip, μcell are Poisson ratios. If Esample ≤ Etip (in the case of Si3N4 tips, the Young's modulus is 150 GPa [17]), E′ can be simplified as:(3)E=Esample1μsample2For soft tissues, the Poisson ratio ranges from 0.490 to 0.499 [18]. The Poisson ratio used in present study was assumed to be 0.5.

3. Results

3.1. Determination of the indentation depth

The quantitative determination of the elastic properties of a particular material can be obtained from the relationship between the applied force F and the indentation depth Δz using Eq. (1). When force curve is measured on a hard substrate (like glass or mica), the cantilever deflection is proportional to the relative sample position resulting in a linear slope in the part of the curve for which the tip and the sample are in contact. When soft samples like erythrocytes are investigated, the recorded cantilever deflection as a function of the relative sample position is not linear due to RBC deformation. Fig. 3 presents force curves recorded on erythrocyte (Fig. 3A) and on PLL-modified glass surface (Fig. 3B). The latter one was used as a calibration curve, since there was no permanent sample deformation observed.
  1. Download: Download full-size image

Fig. 3. Typical AFM force curves obtained on (A) erythrocyte and (B) glass surface modified with PLL.

The indentation produced by pushing AFM tip was determined by subtracting the reference curve, measured on the glass, from the curve recorded for erythrocyte (Fig. 4A). The difference between the deflection of the cantilever detected on the hard and on the soft sample is a measure of the indentation depth. Typical force-versus-indentation curves obtained for the investigated erythrocytes from patients with anisocytosis (a), hereditary spherocytosis (b), G6PD deficiency (c) are presented in Fig. 4B. The force-versus-indentation curves obtained for normal erythrocytes (d) indicated the largest compliance of healthy RBCs.
  1. Download: Download full-size image

Fig. 4. (A) Determination of the force-versus-indentation curves. The straight line corresponds to curves measured on hard, non-deformable surface (glass coverslip) and the nonlinear curve – to compliant surface (erythrocytes). (B) The force-versus-indentation curves (dots are data and lines are fits) obtained for erythrocytes coming from blood samples of patients with anisocytosis (a), hereditary spherocytosis (s), G6PD deficiency (d). The control, normal cells are presented by line (n).

3.2. Young's modulus determination

The Young's modulus was determined by fitting Eq. (1) to each force-versus-indentation curve. The indentation depth range varied from 50 to 500 nm. Depths values larger than 500 nm were omitted due to the expected influence of the hard substrate on which cells were immobilized (typical erythrocyte thickness varied from 600 nm at the central region to around 1.2 μm at the RBC edge).
The presence of the internal structure of erythrocytes suggests that the calculated Young's modulus can vary revealing its distinct value as a function of the indentation depth. The example of the dependence between the Young's modulus and the indentation depth is presented in Fig. 5. All curves showed the plateau at large indentation depth indicating the linear dependence between the applied load and the produced deformation. Pathological erythrocytes reached the plateau faster for indentation depth around 200 nm, while for normal erythrocytes Young's modulus became constant at 300 nm. The saturated values of the modulus are most probably due to the influence of the tension of external cell membrane dominating for larger indentation depth.
  1. Download: Download full-size image

Fig. 5. The example of the dependence of the Young's modulus as a function of the indentation depth for the measured types of the erythrocytes: normal, hereditary spherocytosis, anisocytosis, and G6PD deficiency.

Since erythrocytes come from patients with different types of anemias and the range of alterations in the structure of the erythrocytes was not known, the same indentation depth value was chosen. The value of 200 nm was selected, taking into account the results shown in Fig. 5 where elasticity of erythrocytes from a patient with hereditary spherocytosis was already reaching the plateau. In other cases, plateau was observed for larger indentation depths.
The average values of Young's modulus for different types of anemias were obtained by fitting the Gaussian distribution to the histogram (Fig. 6A–D). The bin size of each histogram was 5 kPa and errors were estimated as the half width of the Gaussian peak. The measured Young's moduli were: 26 ± 7 kPa, 43 ± 21 kPa, 40 ± 24 kPa and 90 ± 20 kPa, for normal, hereditary spherocytosis, thalassemia, and G6PD deficiency, respectively.
  1. Download: Download full-size image

Fig. 6. Histograms of the Young's modulus determined for RBCs from (A) normal, (B) hereditary spherocytosis, (C) thalassemia and (D) G6PD deficiency blood samples. The calculations of the Young's modulus were performed at the indentation depth of 200 nm.

The observed two peaks in histogram for erythrocytes from patients with anisocytosis (Fig. 7) indicated that in the blood samples two populations of erythrocytes were present – normal and pathological ones. The presence of normal erythrocytes was pointed out by the Young's modulus values which was close to that determined for other normal erythrocytes while the pathological ones had three times larger values.
  1. Download: Download full-size image

Fig. 7. Histograms of the Young's modulus determined for erythrocytes showing anisocytosis. The observed two peaks of erythrocytes were attributed to the presence of two RBC populations in blood samples: normal and pathological cells.

3.3. Shape of erythrocytes

The observed alterations of erythrocyte's Young's modulus suggested that also the shape of erythrocyte may change. Therefore, the analysis of their shape was performed but only in case of hereditary spherocytosis that the change of cell shape was observed (Fig. 8).
  1. Download: Download full-size image

Fig. 8. The surface topography of the (A) spherocyte and (B) erythrocyte from a patient with G6PD deficiency measured in air in contact mode.

4. Discussion

The main goal of this work was the determination the erythrocyte's stiffness in blood samples taken from patients with various anemias such as hereditary spherocytosis, thalassemia and G6PD deficiency. Stiffness of erythrocytes from samples with observed anisocytosis was also measured.
The obtained Young's moduli of all above types of erythrocytes were generally larger than that obtained for normal erythrocytes. The most dramatic change was observed for erythrocytes with G6PD deficiency, where the calculated Young's modulus was more than 3 times larger than in normal cells (26 ± 7 kPa and 90 ± 20 kPa, respectively). The obtained Young's modulus values were from 10 to 100 times larger (depending on the studied case) than these obtained for other cells [4]. This can be explained by the fact that during the erythrocytes' immobilization, the glutaraldehyde was used as an agent allowing cell deposition. However, we assume that 0.5% concentration of glutaraldehyde and short time of fixation (1 min) does not screen completely the alterations of cell stiffness occurring during pathological state.
It has already been reported in many publications that the Young's modulus determined using AFM reflects alterations in cell cytoskeleton [19], [20]. Therefore, in order to explain the obtained results the erythrocytes cytoskeleton was taken into account. During indentation of erythrocytes, the AFM tip monitors the response from cell membrane and from cytoskeleton. Both structures are connected to each other. Therefore, it is difficult to separate their response to deformation and, in consequence, any alterations occurring in these structures can influence the overall stiffness of erythrocytes.
The observed increase of Young's modulus for erythrocytes of patients with hereditary spherocytosis can be explained as follows. The erythrocyte cytoskeleton in this disease is altered by changes of spectrin structure, particularly of its α- and â-chains, or by spectrin deficiency, or by its lack of ability to bind ankyrin. These structural changes are due to a gene mutation [10]. These abnormalities of RBC cytoskeleton can significantly influence its elastic properties.
Erythrocytes from patients with thalassemia had also larger Young's modulus than normal RBCs. Such increase can be attributed to the molecular anomaly in hemoglobin structure involving the changes of α- and β-chains [11]. The obtained Young's modulus distribution was not symmetric what may indicate the certain population of erythrocytes with large alterations in the structure of cytoskeleton. Since the measured erythrocytes come from blood samples taken from patients, we may expect such variations of the Young's modulus. In the limit, we can expect the bimodal Young's modulus distribution as it was obtained in case of anisocytosis (see Fig. 7).
The largest differences of Young's modulus were obtained for cells from patients with G6PD deficiency. This condition causes impaired metabolism of RBCs, and particularly ATP production, which might be the main reason for the change of elasticity and shape of erythrocytes [21]. It has already been reported that the reorganization of the actin cytoskeleton and the cell energy metabolism are correlated and the reorganization of actin filaments can be associated with altered ATP metabolism [22], [23]. Therefore, the increase of the Young's modulus may be attributed to the impaired ATP metabolism (toward inhibition of the cell's glycolytic activity).
Various anemias are often associated with anisocytosis i.e. the presence of populations of erythrocytes of different diameters. This may also be due to erythrocyte cytoskeleton changes which might be reflected by variations of Young's modulus determined using AFM. The two observed populations of erythrocytes, those with the modulus close to normal ones and those with its larger values, confirmed that not all erythrocytes present in blood sample were pathological.

5. Conclusions

The AFM measurement of the cellular stiffness can be employed as a method of a direct detection of physical properties of erythrocyte cell membrane and cytoskeleton. It demonstrates the capability to detect changes of erythrocyte membrane and/or cytoskeleton that can be correlated with structural alterations typical for certain blood diseases such as hereditary spherocytosis, thalassemia, and G6PD deficiency. The possible use of AFM to detect two populations of erythrocytes in blood samples from patients with anisocytosis suggests that it can be applied not only to the characteristic cell stiffness but also to detect the pathological erythrocytes in a whole population of RBCs.

Acknowledgements

This work was supported by the Grant 3 T11E 033 26 from the Committee for Scientific Research (MNII) of Poland, realized in years 2004–2006.

References

View Abstract