ARTICLES

Chinese Science Bulletin © 2009

SCIENCE IN CHINA PRESS

Springer

Study on elastic modulus of individual ferritin ZHANG JinHai, CUI ChengYi & ZHOU XingFei† Department of Physics, Ningbo University, Ningbo 315211, China

The mechanical property of individual ferritin was measured with force-volume mapping (FV) under contact mode of atomic force microscopy (AFM) in this work. The elastic modulus of individual ferritin was estimated by the Hertz mode. The estimated value of the elastic modulus of individual ferritin was about 250－800 MPa under a small deformation. In addition, the elastic modulus of individual ferritin was compared with that of the colloid gold nanoparticle.

The iron-storage proteins ferritins broadly exist in animals, plants and human bodies. The single ferritin molecule is made up of 24 subunits and an iron core. Each subunit is an individual molecule that joins its neighboring subunits through noncovalent interactions. The iron core contains up to 4500 irons atoms. The composition of the iron core is found to be close to that of ferrihydrite[1]. Besides its important biological function, ferritin has been used widely in materials science due to its unique physical and chemical properties, such as magnetic and electrochemical properties[2–5]. For example, ferritin has been used as a catalyst for carbon nanotube growth[6,7] and as a template for synthesis of magnetic and photonic nanostructures[8]. Recently, Shin et al.[9] reported that poly(vinyl alcohol) (PVA) nanofibers containing bimolecular ferritin nanoparticles exhibited the enhancement of elastic modulus as compared to pure PVA nanofibers due to chemical interactions between the ferritin and the PVA matrix. Bhattacharyya et al.[10] found that the novel composite films containing ferritin-functionalized multiwall carbon nanotubes (MWCNTs) were reinforced by hydrogen bonding formed between the ferritin functionalized MWCNTs and the polymeric films. To our knowledge, the mechanical property of individual ferritin has not been studied up to now. Atomic force microscopy (AFM) is a useful tool not only for imaging the topography of surfaces on the www.scichina.com | csb.scichina.com | www.springerlink.com

nanometer scale but also for characterizing the mechanical property in different areas on a sample surface[11–13]. The so-called force-mapping mode is a very sensitive tool for measuring interaction forces, such as adhesion or compression with a lateral resolution of only a few nanometers[14,15]. With this method Schaer-Zammaretti and Ubbink[16] compared the surface stiffness of different bacteria. Moreover, FV mapping has also been extensively used in materials science[17,18]. For example, Shulha et al.[19] quantitatively investigated the compression elastic properties of two kinds of dendritic molecules G3 and G4. In the previous work, we studied the compression properties of a water-soluble conjugated polymer, poly[lithium 5-methoxy-2-(4-sulfobutoxy)-1,4phenylenevinylene] (MBL-PPV)[12]. In this work, we estimated the elastic modulus of individual ferritin with FV mapping, and compared its value with that of the colloid gold nanoparticles.

1 Theories and methods 1.1 Materials and sample preparation The ferritin solution used in our experiment was purReceived July 17, 2008; accepted October 10, 2008 doi: 10.1007/s11434-008-0544-6 † Corresponding author (email: [email protected]) Supported by the National Natural Science Foundation of China (Grant No. 10604034), Natural Science Foundation of Zhejiang Province (Grant No. Y606309), Ningbo Natural Science Foundation (Grant No. 2006A610046) and K. C. Wong Magna Fund in Ningbo University

Chinese Science Bulletin | March 2009 | vol. 54 | no. 5 | 723-726

BIOPHYSICS

ferritin, elastic modulus, atomic force microscopy (AFM)

chased from Sigma (Branch Department in Shanghai, China). The original concentration was about 50－150 mg/mL. Before the experiment it was directly diluted with HEPES (pH=6－7, 25 mmol/L, purchased from Shize Co. in Shanghai) to about 10 ng/μL. 3-aminopropyl-triethoxysilane (APS) used to modify the mica surface was also purchased from Sigma (Branch Department in Shanghai, China). The conductance of Milli-Q water was 18.2 MΩ. Freshly cleaved mica was rinsed with Milli-Q water, and dried with nitrogen gas. Then the mica surface was modified with 1% APS solution and roasted for about 2 h at 120℃. Two droplets of the ferritin solution were put on a pretreated mica surface and were dried in air.

2 Results and discussion Figure 1(a) demonstrates a typical high-resolution topographical image of ferritin on the APS-modified mica surface, similar to the results reported by Hemmersam et al.[21]. From the topographic image, we could see that most of the proteins were well dispersed and the heights were around 6－12 nm, which ensured that the tip was compressed mostly on single ferritin molecules during the FV mapping. Additionally, we could see a few small round aggregates with the height above 15 nm (marked by arrow a) and a few fragments with the heights between 2 and 4 nm (by arrow b). We only collected the data from ferritins with the heights around 7－10 nm.

1.2 Instruments and theories Experiments were carried on a multimode scanning probe microscope (Nanoscope IIIa, Veeco Instruments, Santa Barbara, CA, USA). All tapping mode AFM topographic images were taken with silicon tips NSC11. The FV mappings were taken by the NSC12/Ti-Pt tip with the spring constant about 4.5 N/m (from company) and the length about 130 μm. NSC11 and NSC12/Ti-Pt tips were purchased from MikroMasch Co. FV mapping combined the force measurement and topographic imaging capabilities. Single force curve recorded the force (or deflection of the cantilever) felt by a tip and the corresponding displacement of the piezo as it approached and retracted at one point on a sample surface. Each force curve was measured at a unique position on a sample surface, and force curves from an array of points were combined into a three-dimensional array of force data[19,20]. Therefore the FV data set combined simultaneously the topography information (constant-force height) and the force information. In our experiment, we first obtained the topographic image of a well dispersed single ferritin at the tapping mode. Then FV mapping under the relative trig mode was carried out in the same area at the contact mode. The sensitivity of the piezo was carefully calibrated on a bare mica surface right before the FV mapping. During compression the tip deflection was set 2－3 nm to get a small deformation and the number of pixels (32×32) was conducted on for selected areas (500 nm×500 nm). Tapping mode imaging was repeated after force mapping to ensure preservation of initial surface morphology. In the experiment the temperature was controlled to be about 15℃ and the humidity about 30%. 724

Figure 1 The AFM image of ferritin on the APS mica (a) and the corresponding section analysis image (b).

Figure 2 shows a typical force-volume image. Combining Figure 2(a) and Figure 1(a), we could see that most of the ferritins in the selected scanning area were single ones. Due to the thermal drift within a long time period (more than 1 h) and movement of ferritin by tips during scanning, tapping mode imaging was repeated in the same area immediately after force mapping. Despite some random deviations, the shapes and relative locations of each molecule were almost unchanged before and after FV mapping (data not shown). In Figure 2(b), the FV image is made up of a series of force curves, and a clear correlation can be seen between the force and the

ZHANG JinHai et al. Chinese Science Bulletin | March 2009 | vol. 54 | no. 5 | 723-726

corresponding topographic image. We could easily recognize the single ferritin due to its softness. Typical force-distance curves (Figure 3) on the mica surface and the single ferritin clearly illustrated the different compression behavior during the tip approaching and retracting samples[22]. When the tip initiatively approached a sample it felt little force (a→b). As the tip was close to the sample surface, it deflected downwards from point b mainly due to the Van der Waals interaction. When the tip further approached, the sample was gradually compressed (d→e). We could see that ferritin was softer than both mica and gold particles according to the force curves slope as shown in Figure 3.

Figure 3 Typical force-distance curves for individual ferritin and gold nanoparticles.

Single ferritin force curve data in Figure 3 showed different compression behaviors under low normal load. One interesting feature in these data was the slope in the contact region. There was a deviation between the mica and the ferritin due to elastic deformation of molecules, which can be analyzed assuming a Hertzian contact between the tip and the molecules. Since the elastic modulus of the tip material and the mica was very high,

ARTICLES

2

9 (1 − ν ) 2 F , 16 RE 2 where ν, R, E and d were respectively the Poisson ratio (usually 0.5 for soft sample[23]), tip radius, the elastic modulus and the deformation of protein. We estimated the elastic modulus of 10 ferritins at different compressing stages according to ref. [14]. Figure 4 shows the relationship between the elastic modulus and corresponding deformations. The estimated elastic modulus of single ferritin was about 250－800 MPa under small deformations, which was comparable to the Young’s modulus of single lysozyme’s ((500±200) MPa)[14], and also to single DNA under large radial compression in our previous work[24]. The interactions between a tip and a sample increased rapidly when the tip further approached due to the full compression of the protein shell around the iron core. Usually the elastic modulus of protein will increase with the increase of deformation[19]. It was surprising to us that the value of the elastic modulus was much higher (as shown in Figure 4) during the initial compression (deformation is smaller than 1.0 nm). This was likely because we had ignored the adhesion force between the tip and the sample or the strong attraction at the very beginning when the tip slightly touched the sample in our experiment. Further studies are needed to explore this issue. In our case, we merely estimated the elastic modules of ferritin and gave people an impression how soft it was. Besides, we compared the elastic modulus of ferritin and colloid gold with similar diameter under the same condition. The elastic modulus of gold particles was d3 =

ZHANG JinHai et al. Chinese Science Bulletin | March 2009 | vol. 54 | no. 5 | 723-726

725

BIOPHYSICS

Figure 2 The force-volume image of ferritin with 32×32 pixels in the area 500 nm×500 nm. (a) The topographical image; (b) the FV image.

the only detectable indentation would be the indentation of ferritin adsorbed on the mica. Therefore we neglected the elastic indentation of the tip and the mica in our case[22]. There were several limitations in this model, such as the assumption of isotropic elastic properties and no adhesion forces. In order to quantitatively calculate the elastic modulus of individual ferritin, we prepared three samples, randomly selected several ferritins and selected 10 force curves on single ferritin. We could estimate the force between a tip and a sample: F = k×τ, where k was the cantilever force constant, and τ the deflection value. The interaction between the tip and the sample was predigested by the simple model: the AFM tip was supposed to be a sphere with radius R. Therefore the elastic modulus of individual ferritin could be estimated using the Hertz mode[14]:

estimated to be about several GPa, one order larger than that of ferritin, but still less than that of bulk gold. That might be due to the softness of additive surrounding the gold nanoparticles in the commercial product.

3 Conclusions

Figure 4 ferritins).

1

Depth profile of the elastic modulus of single ferritin (10

We investigated the elastic modulus of individual ferritin using the FV mapping of AFM and compared it with the commercial colloid gold nanoparticles. The compression modulus at small deformations of single ferritin was estimated to be about 250－800 MPa in terms of the Hertz mode. It varied according to the different deformations of ferritin.

Wang Q L, Kong B, Huang H Q. Progress in structural and functional

between transferrin and anti-transferrin by atomic force microscope. Chin Sci Bull, 2006, 51(4): 405－408

study of nanometer protein shell of the ferritin (in Chinese). Prog Chem, 2004, 16(4): 516－519 2

14

Harris J G E, Grimaldi J E, Awschalom D D. Excess spin and the dynamics of antiferromagnetic ferritin. Phys Rev B, 1999, 60(5): 3453－3456

3

15

Awschalom D D, Smyth J F, Grinstein G, et al. Macroscopic quantum

4

16

Tejada J, Zhang X X, del Barco E, et al. Macroscopic resonant tun-

5

17

Xu D, Watt G D, Harb J N, et al. Electrical conductivity of ferritin Yamashita I. Fabrication of a two-dimensional array of nano-particles

croscopy. Langmuir, 1998, 14: 2606－2609 18

using ferritin molecule. Thin Solid Films, 2001, 393(1-2): 12－18 7

8

9

10

19

665－667 Mackle P, Charnock J M, Garner C D. Characterization of the man-

hyperbranched macromolecules at solid surfaces. Macromolecules, 20

2003, 36: 2825－2831 Digital Instruments Veeco Metrology Group. NanoScope Command

copy. J Am Chem Soc, 1993, 115: 8471－8472 Shin M K, Kim S I, Kim S J, et al. Reinforcement of polymeric nan-

21

Hemmersam A G, Rechendorff K, Besenbacher F, et al. pH-de-

Reference Manual Version 5.12 Revision B, 2001

ofibers by ferritin nanoparticles. Appl Phys Lett, 2006, 88:

pendent adsorption and conformational change of ferritin studied on

193901－193903

metal oxide surfaces by a combination of QCM-D and AFM. J Phys 22

Chem C, 2008, 112: 4180－4186 Vinckier A, Semenza G. Measuring elasticity of biological materials

23

by atomic force microscopy. FEBS Lett, 1998, 430: 12－16 Domke J, Radmacher M. Measuring the elastic properties of thin

Bhattacharyya S, Sinturel C, Salvetat J P, et al. Protein-functionalized 113104

11

Li H, Han B S. The observation and cutting DNA molecules stretched

12

on Si surface. Chin Sci Bull, 2003, 48(7): 682－685 Zhou X F, Xu H, Fan X H, et al. Compression of single conjugated-polymer nanoparticles with AFM tips. Chem Lett, 2005, 34:

726

Langmuir, 2001, 17: 3286－3291 Shulha H, Zhai X, Tsukruk V V. Molecular stiffness of individual

ganese core of reconstituted ferritin by X-ray Absorption Spectros-

carbon nanotube-polymer composites. Appl Phys Lett, 2005, 86:

13

Du B, Tsui O K C, Zhang Q, et al. Study of elastic modulus and yield strength of polymer thin films using atomic force microscopy.

Bonard J-M, Chauvin P, Klinke C. Monodisperse multiwall carbon nanotubes obtained with ferritin as catalyst. Nano Lett, 2002, 2(6):

logical and structural data. Ultramicroscopy, 2003, 97: 199－208 Chizhik S A, Huang Z, Gorbunov V V, et al. Micromechanical properties of elastic polymeric materials as probed by scanning force mi-

proteins by conductive AFM. Nano Lett, 2005, 5(4): 571－577 6

interfaces, 2008, 62: 206－213 Schaer-Zammaretti P, Ubbink J. Imaging of lactic acid bacteria with AFM—elasticity and adhesion maps and their relationship to bio-

neling of magnetization in ferritin. Phys Rev Lett, 1997, 79(9): 1754－1757

microscope. Langmuir, 1994, 10: 3809－3814 Gaboriaud F, Parcha B S, Gee M L, et al. Spatially resolved force spectroscopy of bacterial surfaces using force-volume imaging. Bio-

tunneling in magnetic proteins. Phys Rev Lett, 1992, 68(20): 3092－3095

Radmaeher M, Fritz M, Cleveland J P, et al. Imaging adhesion forces and elasticity of lysozyme adsorbed on mica with the atomic force

1488－1491 Zheng Z W, Yang P H, Zeng G C, et al. Probing molecular interaction

polymer films with the atomic force microscope. Langmuir, 1998, 14: 24

3320－3325 Zhou X F, Sun J L, An H J, et al. Radial compression elasticity of single DNA molecules studied by vibrating scanning polarization force microscopy. Phys Rev E, 2005, 71: 062901

ZHANG JinHai et al. Chinese Science Bulletin | March 2009 | vol. 54 | no. 5 | 723-726