Membrane deformation of unfixed erythrocytes in air with time lapse

investigated by tapping mode atomic force microscopy

Yong Chen a,b, Jiye Cai a,1,*

a Department of Chemistry, Jinan University, Guangzhou 510632, People’s Republic of Chinab Department of Microbiology and Immunology, University of Illinois, 835 S. Wolcott Avenue, RE704, Chicago, IL 60612, USA

Received 5 May 2005; received in revised form 19 November 2005; accepted 19 November 2005

Abstract

Estimation of the time of death is one of the most important problems for forensic medicine and law. Physical and chemical postmortem

changes are evaluated together while estimating the time of death. The pattern analysis of antemortem and postmortem bloodstains is one of the

important parameters for forensic science, and cellular changes of blood cells can be useful for the quantitative assessment of the time of death. In

this study, by successively investigating erythrocytes exposed in air on mica for 5 days using tapping mode atomic force microscopy (TM-AFM),

we observed deformation of whole cell and membrane surface of unfixed erythrocytes with time lapse. We found that the time-dependent cellular

changes occurred after exposure of erythrocytes in air for several days. At 0.5 days of exposure, fissures and cell shrinkage were observed. At

2.5 days of exposure, the emergence of nanometer-scale protuberances were observed and these protuberances increased in number with

increasing time. The changes of cell shape and cell membrane surface ultrastructure can be used to estimate the time of death. Futhermore, smear-

induced abnormal erythrocytes and immunostained erythrocytes were observed here. The need for more precise research is indicated, such as the

correlation of membrane changes to intervals of less than 0.5 day of air exposure, and use of various substrates in addition to mica, including glass,

metals, fabrics, among others, on which the bloodstains appear in crime scenes. The results of this research demonstrate the efficacy of AFM as a

potentially powerful analytical tool in forensic science.

q 2005 Elsevier Ltd. All rights reserved.

PACS: 07.79.L

Keywords: Atomic force microscopy (AFM); Tapping mode; Erythrocyte; Forensic examination; Bloodstains

1. Introduction

Until now, the most widely used methods for the forensic

examination of evidence in crime scenes were optical and

electron microscopy. The recent development of atomic force

microscopy (Friedbacher and Harald, 1999; Dufrene, 2003;

Frederix et al., 2003; Horber and Miles, 2003; Ikai and Afrin,

2003; Rounsevell et al., 2004; Santos and Castanho, 2004;

Hansma et al., 2004; Chen et al., 2005) gives an opportunity to

complement or even replace the classical instruments used in

this field. For example, Smith quantitatively analyzed the outer

surfaces of human hair by AFM (Smith, 1998), and more

0968-4328/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.micron.2005.11.011

* Corresponding author. Tel.: C1 312 413 7892; fax: C1 312 996 5725.

E-mail addresses: dr_yongchen@hotmail.com (Y. Chen), tjycai@jnu.edu.

cn (J. Cai).1 Tel.: C86 20 8522 3569; fax: C86 20 8522 3569.

recently Gurden et al. classified the AFM images of human hair

and even presented an algorithm for the potential automatic

analysis of them (Gurden et al., 2004). In order to validate the

documentary evidence in many different types of litigation,

forensic examination of line crossings in documents by AFM

have been made by Kasas and co-workers (Kasas et al., 2001).

Bersellini and collaborators (Bersellini et al., 2001) have used

AFM in the research on latent fingerprints on metallic surfaces.

For humans, the size and shape of their erythrocytes are

important indicators of well being. Therefore, erythrocytes

have been extensively studied by AFM over the years because

of their relatively simple structure and ease of isolation. Much

research has been done on erythrocytes with AFM, including

normal erythrocytes (Nowakowski et al., 2001; O’Reilly et al.,

2001), pathological erythrocytes (Zachee et al., 1994),

tumorous erythrocytes (Chen et al., 2002), infected erythro-

cytes (Nagao et al., 2000), traumatized erythrocytes (Girasole

et al., 2001; Ohta et al., 2002), and the membrane cytoskeleton

of erythrocytes (Swihart et al., 2001; Liu et al., 2005). All this

research indicates that AFM is a powerful and mature

Micron 37 (2006) 339–346

www.elsevier.com/locate/micron

Y. Chen, J. Cai / Micron 37 (2006) 339–346340

technique to study cellular changes and membrane surface

ultrastructure of erythrocytes. The pattern analysis of ante-

mortem and postmortem bloodstains is one of the important

parameters for forensic science, and cellular changes of

erythrocytes can be useful for the quantitative assessment of

the time of death. AFM can be utilized to study antemortem

and postmortem bloodstains. Using a calibration curve,

investigators can pin down the age of a bloodstain with much

greater precision than simple visual examination affords. Even

so, there are few reports on the application of AFM in

postmortem bloodstains, and on the variation processes of cells

in postmortem bloodstains.

In this study, the morphological changes of unfixed

erythrocytes in air over the lapse of time, after introduction

to substrate, were observed by TM-AFM. The aim is to

establish if this new technique can be a useful tool for the

examination of antemortem or postmortem bloodstains and for

the estimation of the time of death.

2. Materials and methods

A fresh human blood sample was obtained and mixed with

an anticoagulant (EDTA). The freshly extracted human blood

was diluted in phosphate buffered saline (PBS: 0.2 mg/ml

KH2PO4, 0.2 mg/ml KCl, 2.9 mg/ml Na2HPO4$12H2O,

8.0 mg/ml NaCl, pH 7.4; 80 times). Five microliters aliquots

of this solution and erythrocytes immunolabelled with

R-phycoerythrin (R-PE)—conjugated IgG to human glyco-

phorin A (Becton Dinckison Biosciences Pharmingen, San

Diego, CA) were deposited on the freshly cleaved mica

surfaces and were then air-fixed rapidly by vigorous manual

gesticulation. The film which was produced was examined

under the optical microscope.

Atomic force microscopy was performed in air on the blood

film using a commercial AFM (AutoProbe CP, Thermomicro-

scopes) in tapping mode. The mica carrying the blood film was

mounted onto the XY stage of the AFM and the integral video

camera was used to locate the regions of interest. Micro-

fabricated silicon cantilevers (Park Scientific Instruments) with

a force constant of approximately 2.8 N/m were used. Repeated

scanning of the same red blood cells confirmed that no physical

damage occurred during imaging. The blood film was kept on

the XY stage of AFM for 5 days (humidity: 75%, temperature:

25 8C. Humidity was externally controlled via two dehumidi-

fiers and monitored with a hygrometer.), and the same cell was

scanned every 0.5 day. For high resolution purposes 15 mm!15 mm area AFM images were obtained at a line scan rate of

0.5 Hz. These images were stable and reproducible during

repeated scanning.

Before quantification was carried out, the AFM images were

planar leveled using the software (Thermomicroscopes Proscan

Image Processing Software Version 2.1) provided with the

instrument. Using the line analysis function of the software, the

average width and the average height (i.e. thickness) of the red

blood cells and protuberances on cell were determined. Using

the particle analysis function of the software, the surface area

and the volume of red blood cell were determined.

3. Results

Fig. 1 is a relatively low-resolution image of single and

multiple erythrocytes. Both optical microscopy and electron

microscopy can observe a single, whole cell which is

approximately several microns in size, but AFM has distinct

advantages, such as clearer images, easy sample preparation,

extensive environment (in air or liquid) of sample to escape from

the damage of reagents, low-temperature, and ultrahigh vacuum

in electron microscopy, and so on. Smearing is a general

technique used to make cell films for optical microscopic

observation. However, it is not a good method for AFM

observation, for smearing will flatten erythrocytes (Fig. 1b and

e) and even convert them into echinocytes (Fig. 1c and f).

Erythrocytes from patients with various diseases have been

studied by AFM in our previous publication (Chen et al., 2002).

Normal red blood cells (RBCs) are biconcave in shape with an

approximate thickness of 2–3 mm and a diameter of around

7 mm. Fig. 1a is the AFM image ofmultiple normal erythrocytes

in 90 mm!90 mm scanning area. In Fig. 1d, the RBC is standard

biconcave in shapewith an approximate thickness of 1 mmand a

diameter of 8 mm.Because erythrocytes were on a solid support,

the thickness measurement was less than that of regular RBCs.

The size and shape of the central depression is variable, and

depending on its deformation. In Fig. 1d, the central depression

is 516 nm according to line analysis. As observed in a

conventional SEM, the blood cells have a tendency to stick to

each other or form stacks. In Fig. 1g, a biconcave shaped red

blood cell sticks to a spherical blood cell.

Figs. 2–10 illustrate the AFM three-dimentional topo-

graphic images of the same erythrocyte on mica at 0.5 day

exposure interval in air. After approximately 5 days exposure

in air and scanning many times by tip, the erythrocyte still kept

its biconcave shape, and the conditions of the cell environment

did not change, but the shapes were different at various

exposure intervals. After 0.5 days of exposure, a fissure

appeared in the erythrocyte (Fig. 3), another fissure appeared

after 1 day later (Fig. 4). After 1.5 days, there were more

fissures. These fissures were around 350 nm wide and 90 nm

deep, which are larger than the sizes of needle of the cantilever.

In AFM technique, a needle with a sharp ultralever tip is

scanned over the cell surface. The typical radius of curvature of

tip used in our experiment is w10 nm, and the diameter of the

needle where it is 90 nm away from the top end of the tip,

is less than 90 nm. So it could be concluded that these fissures,

as well as those in Fig. 1g, were not caused by the AFM needle.

It must be mentioned that the parallel lines (shown by

arrowheads) in Fig. 1g, 2, 4, and others were not caused by

scratching of the AFM needle, but by artifacts of the imaging

system since they dont’t appear in Figs. 3 and 6–10.

Another of the time dependent cellular changes of the

erythrocyte was shrinkage of cell. For example, the part of the

cell indicated by the arrow in Fig. 3 sank after 0.5 day, also

shown by the arrow in Fig. 4. Fig. 5 illustrates four large

fissures in the membrane surface of the erythrocyte after

1.5 days of exposure in air, but only two large fissures remained

3 days later. The other two large fissures shown by the two

Fig. 1. AFM images of single and multiple normal erythrocytes, and abnormal erythrocytes from smear samples. (a) AFM image of multiple normal erythrocytes,

scanning area: 90 mm!90 mm; (b and c) AFM images of multiple abnormal erythrocytes of various degrees, scanning area: 90 mm!90 mm; (d) AFM images and

surface profiles obtained for biconcave shaped erythrocyte; scanning area: 15 mm!15 mm, max. height 1 mm, width about 8 mm; (e and f) AFM images and surface

profiles of flatter and flatter erythrocytes caused by smear technique; scanning area: 15 mm!15 mm; (g) three-dimensional AFM image of two blood cells sticking to

each other; scanning area: 17 mm!17 mm.

Y. Chen, J. Cai / Micron 37 (2006) 339–346 341

arrows in Fig. 5 disappeared. That is, the part of the erythrocyte

with the other two large fissures has sunk.

Fig. 11 indicates the changes of AFM-measured volume and

surface area of the same erythrocyte based on Figs. 2–10. It is

clear that the AFM-measured surface area decreased but the

AFM-measured volume did not change much within the time

lapse period of 5 days. It is understandable that the sinkage of

the cell may be the main reason for the results, that is,

Fig. 2. AFM image of an erythrocyte introduced to the mica for 0 day. Fig. 4. AFM image of the same erythrocyte introduced to the mica for 1 day.

Y. Chen, J. Cai / Micron 37 (2006) 339–346342

additional cell surface were not detectable by AFM once they

were beneath the cell.

Figs. 7–10 also show that, after a 2.5-days exposure of

erythrocytes in air, many protuberances appeared in the

membrane. Fig. 12a is the three-dimensional topographic

AFM image of the concave part of the erythrocyte 2 days after

exposure in air, and Fig. 12b is the three-dimensional

topographic AFM image of the same part of the cell 2.5 days

Fig. 3. AFM image of the same erythrocyte introduced to the mica for 0.5 day.

after exposure in air. In the center of the two figures, there are

some large particles with a diameter of approximately several

100 nmwhichmay be dustwhichwas not seen on the surfaces of

other erythrocytes. The difference between the two figures is

the appearance ofmany protuberances in Fig. 12b. These images

of the protuberances were enlarged in order to be seen more

clearly in Fig. 12c. There are large protuberances with an

approximate height of 8–10 nmand a diameter of approximately

Fig. 5. AFM image of the same erythrocyte introduced to the mica for 1.5 day.

Fig. 6. AFM image of the same erythrocyte introduced to the mica for 2 day. Fig. 8. AFM image of the same erythrocyte introduced to the mica for 3 day.

Y. Chen, J. Cai / Micron 37 (2006) 339–346 343

50 nm, and some small protuberances with an approximate

height of 2–3 nm and a diameter of approximately 20 nm. There

are w47 protuberances in 1 mm!1 mm surface area. It is very

clear that the cell surface become more rough from Fig. 2 to

Fig. 10, especially in Fig. 10, where the protuberances are

obvious, implying that the appearance of these protuberances is

time-dependent.

Another phenomenon worth paying attention to is the

change of fluidity of the erythrocyte. In the first 2 days

Fig. 7. AFM image of the same erythrocyte introduced to the mica for 2.5 day.

(Figs. 3–6), the shape of the whole erythrocyte changed

dramatically, and even projected to a great extent at some

positions (as shown by the asterisks in Figs. 4 and 5). However,

after 2 days, the overall shape of the erythrocyte didn’t change

any more (Figs. 7–10), implying that the fluidity of the

erythrocyte became weaker with the loss of water. Further-

more, the period (2–2.5 days) when the cell stopped projecting

dramatically is consistent with the period (2–2.5 days) when

the protuberances on cell surface started appearing, implying

Fig. 9. AFM image of the same erythrocyte introduced to the mica for 4 day.

Fig. 10. AFM image of the same erythrocyte introduced to the mica for 4.5 day.

Y. Chen, J. Cai / Micron 37 (2006) 339–346344

that most of the water in the erythrocyte was lost by this point

and the cell membrane had attached to the cytoskeleton

network allowing transmembrane and peripheral proteins to

form those protuberances.

It is notable that the protuberances appearing on the

membrane surface of the erythrocyte look like antibodies that

were used to label the membrane surface (Fig. 12). It was very

difficult for AFM to detect single antibodies on the membrane

surface because they are similar in size to membrane proteins.

Therefore, R-PE proteins were used to increase the size of the

antibodies. In Fig. 12e, the R-PE—conjugated IgG antibodies

to human glycophorin A antigen, with a diameter of 40–60 nm

which is consistent with our previous study (Chen et al., 2004),

are shown clearly.

In the large scan area, the overall morphology of the cells is

visible, but detailed structural features on the cell surface are

Fig. 11. AFM-measured volumes and surface areas of the erythrocyte in Figs.

2–10 extracted from the images by the AFM software. It is clear that the AFM-

measured surface area of the erythrocyte decreased with time lapse although

the AFM-measured volume had little change.

not resolved. As depicted in Fig. 12, by zooming in on the

surface of the erythrocyte to a smaller scan size, 3 mm!3 mmand 0.6 mm!0.6 mm, the fine structure of membrane surface of

the erythrocyte can often be resolved. With a 0.6 mm!0.6 mmscan size, the smallest features on the erythrocyte that can be

resolved are approximately 3 nm.

4. Discussion

In certain situations, bloodstain analysis can be considered a

supplemental tool for determining the postmortem interval

(PMI). Bloodstain pattern analysis is often useful in establish-

ing and reconstructing the sequence of events or mechanisms

that caused blood flow. It is important to remember, however,

that a body can release blood either while alive or after death,

due to gravity. Analysis of this ‘static aftermath’ may assist in

determining some time factors, such as exposure to air, heat

and humidity, when making any time of death estimations

using bloodstain analysis.

Like other tissue cells, blood cells also lose their normal

morphology because of postmortem autolysis and putrefaction,

and are unidentifiable in the last period. The changing process

through normal morphology to the unidentifiable period can be

a useful criterion for estimating postmortem interval. In our

paper, after exposure in air for several days, some morpho-

logical changes of an erythrocyte could be observed.

Firstly, fissures on the membrane surface of the erythrocyte

appeared after 0.5 day of exposure (Fig. 3), and more fissures

appeared with the lapse of time (Figs. 4–10). The difference of

local microenvironments may be one of the reasons, which

may result in repeated dehydration in some areas of the

erythrocyte but in relatively stable conditions in other areas,

just as the fissions of earth surface and human skin. On the

other hand, a progressive stiffening of the membrane due to

oxidation of the cell may occur and promote formation of the

fissions. Second, after exposure in air, because of gravity of

itself, the erythrocyte stabilized on mica would sink step by

step in different degrees correlated to exposure times. The

arrows in Figs. 3–5 show sinking parts of the cell. Thirdly,

protuberances at the nanometer scale emerged after 2.5 days

exposure (Fig. 12), and the number of the emerging

protuberances was time-dependent. There are at least two

possible causes to explain these protuberances. The hemo-

globin in cytoplasm flows out through dehydration-induced

holes in the membrane. With the water in the membrane (about

20% of the total membrane weight) evaporated, the membrane

becomes thinner and exposes these integral membrane

proteins, such as band three protein, glycophorin A, and

others, or the cytoskeleton proteins. In fact, there has been

some literature reporting AFM imaging of the intracellular

membrane skeleton from the extracellular surface of cells

(Takeuchi et al., 1998; Christian et al., 1997). In Takeuchi’s

paper, a sunken membrane model was given, which stated that,

‘during drying, the membrane sinks with the loss of water into

the cell where there is no support by the membrane skeleton,

leaving behind the membrane skeleton and the membrane

attached to it’ (Takeuchi et al., 1998). Because scanning was

Fig. 12. Comparison of the three-dimensional AFM images of the concave center part of the erythrocyte in (a) Fig. 2 and (b) Fig. 7; scanning area: 3 mm!3 mm. In b,

the white circles indicate some protuberances. These large particles in center maybe dust; (c) enlargements of the same part of b, left two-dimensional image, right

three-dimensional image; scanning area: 0.6 mm!0.6 mm. The central parts of erythrocytes labeled without (d) and with (e) R-PE—conjugated mouse monoclonal

antibodies to human glycophorin A antigen; insets: the enlargements of the parts of the boxed parts in d and e; scanning size: 3 mm!3 mm (d and e); 1 mm!1 mm

(insets).

Y. Chen, J. Cai / Micron 37 (2006) 339–346 345

done in contact mode in their study, the depression of the AFM

tip on the membrane during scanning should be considered

more or less one of the reasons why they obtained such

resolution of the cytoskeleton network. Besides the reasons

mentioned above for interpretations of the fissions and

protuberances, other factors shall be considered, most

important two of which are pH value and salt concentration.

Actually, with repeated dehydration of the cell membrane, the

pH value and salt concentration will continually change, and

this will affect cell morphology to some extent.

All these changes in cell shape and cell membrane surface

ultrastructure will be useful for estimating the time of death.

But more precise research is indicated, such as the correlation

of membrane changes to shorter intervals of less than 0.5 day

exposure in air, and the use of various substrates in addition

to mica, including glass, metals, fabrics, among others, on

which the bloodstains appear in crime scenes.

In addition, the speed of all postmortem changes varies

widely in dependence of the conditions of storage during the

postmortem interval. Because of many factors, it is quite

difficult to report the exact time of death. Although, there are

many methods to estimate the time of death in forensic

medicine, none of them is reliable enough to be used alone.

Therefore, many of the available methods are applied together

and the most plausible time of death is estimated. Optical

microscopy is the traditional method to observe shape changes

of cells, and only the whole cell, not membrane surface

ultrastructure, can be observed. Because of its high resolution

and fast imaging, AFM cannot only investigate promptly

a multitude of cells (Fig. 1a–c) and a single whole cell

Y. Chen, J. Cai / Micron 37 (2006) 339–346346

(Figs. 1d–g–10), but also membrane surface ultrastructure

(Fig. 12), such as the protuberances at the nanometer scale

emerging on membrane surfaces. Observation of the mem-

brane surface ultrastructure may be useful for more accurate

estimation of the time of death. Therefore, AFM technique will

be a potentially powerful assistant tool in forensic science.

5. Conclusions

Morphological changes in a whole erythrocyte and of the

erythrocyte membrane surface ultrastructure, unfixed on mica

substrate, exposed in air over a 5-day period, were observed by

tapping mode atomic force microscopy (TM-AFM). The

changes include the appearance of fissures, shrinkage of

cells, and especially emergence of protuberances at the

nanometer scale on cell membrane surfaces. All the changes

of cell shape and cell membrane surface ultrastructure were

time-dependent and could help to improve estimation of the

time of death. However, this present study is quite preliminary,

more in-depth research will be needed to provide a calibration

curve for some kind of morphological marker of time lapsing in

order to realize as precise estimation of the time of death as

possible.

Acknowledgements

This research project is supported by national 973 program

of China (No. 2001CB510101), national natural science

foundation of China (No. 62078014) and key program of

national natural science foundation of China (No. 30230350).

We’d like to give many thanks to Zahida Ali, Department of

Microbiology and Immunology, University of Illinois at

Chicago, for her helpful corrections of the revised paper.

References

Bersellini, C., Garofano, L., Giannetto, M., Lusardi, F., Mori, G., 2001.

Development of latent fingerprints on metallic surfaces using electro-

polymerization processes. J. Forensic Sci. 46, 871–877.

Chen, Y., Cai, J., Zhao, J., 2002. Diseased red blood cells studied by atomic

force microscopy. Int. J. Nanosci. 1, 683–688.

Chen, Y., Cai, J., Xu, Q., Chen, Z.W., 2004. Atomic force bio-analytics of

polymerization and aggregation of phycoerythrin-conjugated immuno-

globulin G molecules. Mol. Immunol. 41, 1247–1252.

Chen, Y., Cai, J., Zhao, T., Wang, C., Dong, S., Luo, S., Chen, Z.W., 2005.

Atomic force microscopy imaging and 3-D reconstructions of serial thin

sections of a single cell and its interior structures. Ultramicroscopy 103,

173–182.

Christian, L.G., Eric, L., Marie-Cecile, G., Eric, F., Jean-Pierre, G., 1997.

Simultaneous imaging of the surface and the submembraneous cytoskeleton

in living cells by tapping mode atomic force microscopy. Life Sci. 320,

637–643.

Dufrene, Y.F., 2003. Recent progress in the application of atomic force

microscopy imaging and force spectroscopy to microbiology. Curr. Opin.

Microbiol. 6, 317–323.

Frederix, P.L., Akiyama, T., Staufer, U., Gerber, C., Fotiadis, D., Muller, D.J.,

Engel, A., 2003. Atomic force bio-analytics. Curr. Opin. Chem. Biol. 7,

641–647.

Friedbacher, G.F., Harald, F., 1999. Classification of scanning probe

microscopies. Pure Appl. Chem. 71, 1337–1357.

Girasole, M., Cricenti, A., Generosi, R., Congiu-Castellano, A., Boumis, G.,

Amiconi, G., 2001. Artificially induced unusual shape of erythrocytes: an

atomic force microscopy study. J. Microsc. 204, 46–52.

Gurden, S.P., Monteiro, V.F., Longo, E., Ferreira, M.M., 2004. Quantitative

analysis and classification of AFM images of human hair. J. Microsc. 215,

13–23.

Hansma, H.G., Kasuya, K., Oroudjev, E., 2004. Atomic force microscopy

imaging and pulling of nucleic acids. Curr. Opin. Struct. Biol. 14, 380–385.

Horber, J.K., Miles, M.J., 2003. Scanning probe evolution in biology. Science

302, 1002–1005.

Ikai, A., Afrin, R., 2003. Toward mechanical manipulations of cell membranes

and membrane proteins using an atomic force microscope: an invited

review. Cell Biochem. Biophys. 39, 257–277.

Kasas, S., Khanmy-Vital, A., Dietler, G., 2001. Examination of line crossings

by atomic force microscopy. Forensic Sci. Int. 119, 290–298.

Liu, F., Mizukami, H., Sarnaik, S., Ostafin, A., 2005. Calcium-dependent

human erythrocyte cytoskeleton stability analysis through atomic force

microscopy. J. Struct. Biol. 150, 200–210.

Nagao, E., Nishijima, H., Akita, S., Nakayama, Y., Dvorak, J.A., 2000. The cell

biological application of carbon nanotube probes for atomic force

microscopy: comparative studies of malaria-infected erythrocytes.

J. Electron Microsc. (Tokyo) 49, 453–458.

Nowakowski, R., Luckham, P., Winlove, P., 2001. Imaging erythrocytes under

physiological conditions by atomic force microscopy. Biochim. Biophys.

Acta 1514, 170–176.

Ohta, Y., Okamoto, H., Kanno, M., Okuda, T., 2002. Atomic force microscopic

observation of mechanically traumatized erythrocytes. Artif. Organs 26,

10–17.

O’Reilly, M., McDonnell, L., O’Mullane, J., 2001. Quantification of red blood

cells using atomic force microscopy. Ultramicroscopy 86, 107–112.

Rounsevell, R., Forman, J.R., Clarke, J., 2004. Atomic force microscopy:

mechanical unfolding of proteins. Methods 34, 100–111.

Santos, N.C., Castanho, M.A., 2004. An overview of the biophysical

applications of atomic force microscopy. Biophys. Chem. 107, 133–149.

Smith, J.R., 1998. A quantitative method for analysing AFM images of the

outer surfaces of human hair. J. Microsc. 191, 223–228.

Swihart, A.H., Mikrut, J.M., Ketterson, J.B., Macdonald, R.C., 2001. Atomic

force microscopy of the erythrocyte membrane skeleton. J. Microsc. 204,

212–225.

Takeuchi, M., Miyamoto, H., Sako, Y., Komizu, H., Kusumi, A., 1998.

Structure of the erythrocyte membrane skeleton as observed by atomic

force microscopy. Biophys. J. 74, 2171–2183.

Zachee, P., Boogaerts, M., Snauwaert, J., Hellemans, L., 1994. Imaging uremic

red blood cells with the atomic force microscope. Am. J. Nephrol. 14,

197–200.