- Email: [email protected]
Encapsulated cells: an atomic force microscopy study
ARTICLE IN PRESS
Biomaterials 25 (2004) 3655–3662
Encapsulated cells: an atomic force microscopy study Meng Yua, Albena Ivanisevica,b,* a
Department of Biomedical Engineering, Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA b Department of Chemistry, Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA Received 27 February 2003; accepted 10 October 2003
Abstract Two types of cells—human platelets and spore cells—were encapsulated in polymer shells by adsorbing polyanions and polycations in a stepwise fashion. The encapsulated cells were attached to gold and silicon surfaces and their morphological and adhesion properties were studied in air using tapping mode atomic force microscopy (AFM). The roughness of the encapsulated cells increased upon the addition of a new polymer layer. The increase in roughness can be attributed to the formation of a shell around the cells, which is stabilized by electrostatic interactions, as well as to the drying effects associated with the immobilization and dehydration of the cells. Trigger mode was used to perform the force imaging and map out the adhesion characteristics of the cells. Systematic ‘‘maps’’ of the adhesion properties of the encapsulated cells to clean and amine terminated AFM tips were collected. The adhesion force data for the different tips and encapsulated cells showed dependence not only on the number and thickness of the polymer layers, but also on the interactions between these layers. The encapsulated cells’ morphology and roughness characteristics remained intact over a substantial storage period. This stability and adhesion properties make them suitable building blocks for the design and construction of biomimetic templates where AFM is used as the primary tool to do the fabrication. r 2003 Elsevier Ltd. All rights reserved. Keywords: Platelets; Spores; Polymers; Atomic force microscopy; Adhesion
1. Introduction The development of methods to position, manipulate and image single molecules is important from a fundamental as well as applied standpoint. Basic understanding of how single molecules, and in particular biologically relevant ones, take advantage of their unique properties such as elasticity and biochemical host–guest recognition can help in designing strategies to protect and regenerate biological scaffolds. Furthermore, one can utilize this information to fabricate improved biocompatible devices. A promising route to make this a reality is the usage of surface templates to mediate the building of complex structures via a bottom-up strategy [1]. The identification, fabrication and evaluation of biomimetic templates holds great potential for a variety of applications [2]. The engineering of biomimetic surfaces that can act as sensors for the *Corresponding author. Department of Biomedical Engineering and Department of Chemistry, Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA. E-mail address: [email protected] (A. Ivanisevic). 0142-9612/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2003.10.061
purpose of enhancing recognition processes and achieving great control over desired morphology, electronic and mechanical properties, is an extremely valuable investment for future versatile systems. The atomic force microscope (AFM) [3] is a commercial tool that is used to investigate the properties of biomimetic templates. In morphological studies it can serve as a tool to acquire images with nanometer resolution by rastering a sharp tip across a surface. In addition, this technique can be utilized to measure adhesion forces between the tip modified with a certain chemical functionalities and a sample surface. The adhesion properties of a given material play a crucial role in sub-micron and nanometer scale surface modification and self-assembly processes [4]. A number of applications, such as AFM-based lithographic fabrication protocols, directly depend upon properties such as adhesion and diffusion [1,5]. In order to achieve robustness and reproducibility during the fabrication process, it is essential to have a fundamental understanding of the properties of different materials. AFM force measurements can be applied as a quantitative tool to record and compare attractive and repulsive
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3656
forces [6]. The force maps one can generate allow to elucidate mechanical properties such as adhesion and elasticity [7]. Unlike a lot of conventional hard materials, biomaterials can be fragile and thus easily damaged during a number of surface immobilization procedures. Therefore it is essential to extract the mechanical properties of these materials prior to incorporating them in micro- and nano-fabrication protocols that produce complex surface templates. Using AFM to analyze biomaterials [8] in terms of their adhesion properties is advantageous compared to other methods due to its sensitivity often quoted in the range of 1012 N. In addition, one can position the tip on a given surface coordinate with a precision of a few nanometers. In this study, we report the morphological and adhesion properties of encapsulated cells. To create functional structures we followed a procedure reported by Lvov et al. [9] for bovine platelets. We use two types of cells: (i) human platelets, blood cells in the human body; and (ii) Bacillus subtilus (B. subtilus) spore cells. To encapsulate the cells in charged polymer shells, we took advantage of the fact that their surfaces are covered with different kinds of glycoproteins [10]. By tuning the pH of the solution they are dispersed in, one can treat them as charged micro-structures and functionalize them with nanometer size biomimetic shells [9]. The modification of the two types of cells we used was done using the layer-by-layer (LbL) [11], approach by choosing proper polycations and polyanions (Chart 1). This approach allows us to terminate the cells’ surfaces with a given chemical functionality and has been adapted in modifying a wealth of organic, inorganic and biological structures [12]. In this paper, we systematically map out the properties of the encapsulated cells using AFM. We plan to incorporate the results of this study in the development of methods to pattern these functional surfaces via scanning lithography approaches [1,5].
2. Materials and methods 2.1. Materials 3-Aminopropyltriethoxysilane (99%), poly(diallyldimethylammonium chloride) (PDDA, 20 wt% in water, PDDA *
H2C
H3C
N
+
n
*
*
CH
CH2
Cl CH3
SO3 Na+
Chart 1.
2.2. Silicon surface preparation Silicon pieces ð1 1 cm2 Þ were cut from a 400 wafer and cleaned with DI water, EtOH and MeOH using an ultrasonic bath. Each piece was incubated in a piranha solution (mixture of sulfuric acid (51%) and hydrogen peroxide (30%) in a 3:1 ratio, v/v) for 25 min at room temperature. The surface was rinsed with DI water and EtOH several times and dried under a stream of nitrogen. Subsequently, the substrates were incubated in a home-built chamber in the presence of vapors of 3aminopropyltriethoxysilane and water for 40 min. To perform this vapor-phase silanization we followed a method described by Bein et al. [13]. Each surface was baked at 120 C in a laboratory oven for 30 min. Contact angle measurements were done before and after the silanization to verify the treatment. These measurements matched the literature [14]. 2.3. Gold surface preparation A silicon wafer with 500 nm oxide layer was coated with 10 nm Ti followed by 30 nm Au using a thermal evaporator. The evaporation was performed at an evaporation rate of 1 nm/s and a vacuum pressure of 3 107 Torr: After the evaporation the wafer was cut into 1 1 cm2 pieces. Each piece was incubated in 1 mm MHA solution in EtOH for 24 h in order to create a selfassembled monolayer on the surface. Contact angle measurements of these substrates matched the literature [15,16]. 2.4. Human platelets extraction
PSS CH2
average Mw 100,000–200,000), poly(sodium 4-styrenesulfonate) (PSS, average Mw B 70,000), 16-mercaptohexadecanoic acid (MHA), hydrogen peroxide (30%), sulfuric acid, HPLC grade methanol (MeOH) and ethanol (EtOH) were purchased from Aldrich. Phosphate-buffered saline tablets (PBS) and 1% paraformaldehyde were purchased from Sigma. All chemicals were used as received. Deionized (DI) water (18.0 MO cm) was obtained from a Millipore Milli-Q system. B. subtilis (sus-1A-8) water-suspension at 108 CFU were obtained from Raven Biological Laboratories (NE). Silicon wafers with 500 nm of thermally evaporated oxide were fabricated by WaferNet, Inc, CA. Solutions of 2 mg/ml PDDA and 3 mg/ml PSS were prepared in PBS at pH 7.4.
n
*
Platelets were purified from fresh human blood using a platelet-rich plasma (PRP) method [17]. Fresh human whole blood was centrifuged at 66g for 10 min at 25 C to obtain PRP (supernatant). PRP was immediately fixed in 1% paraformaldehyde solution for 2 h. Fixed platelets were purified by centrifugation at 640g for
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
5 min at 25 C and re-suspended in PBS solution (pH 7.4). The final concentration of the platelets was the same as the concentration of PRP. 2.5. Encapsulated cells We followed a modified LbL procedure described by Lvov et al. [9]. Briefly, cells were incubated with shaking in a 2 mg/ml PDDA solution for 20 min at room temperature. Three washing cycles with PBS solution were done by centrifugation at 640g for 5 min at 25 C to remove the excess polymers. Through the same procedure, a negatively-charged PSS layer was deposited over the PDDA layer. The last two steps were repeated to achieve a desired number of polymer layers. 2.6. Attachment of cells to surfaces Prior to the tapping mode imaging, cells and encapsulated cells were attached to solid substrates in the following fashion: Cells terminated with a PDDA layer were attached to gold surfaces modified with MHA. Cells terminated with a PSS layer were attached to silanized surfaces. Uncoated cells were attached to silanized surfaces. In each case, a drop of the appropriate type of cells was incubated on top of a given surface for 2 min and subsequently washed several times with PBS buffer, pure water and dried under a stream of nitrogen. 2.7. AFM imaging and force evaluation Tapping mode images and force curves were collected with a Multimode Nanoscope IIIa (Digital Instruments, Santa Barbara, CA) using clean tips and 3-aminopropyltriethoxysilane modified tips. All experiments were done with silicon tapping mode tips purchased from Digital Instruments, model number RTESP7 with a length of 125 mm, frequency 300 kHz, a spring constant of 20 N/m and a rotated tip. Tips were cleaned with a piranha solution, EtOH solution and then coated with 3-aminopropyltriethoxysilane using a procedure by Bein et al. [13] described under the silicon surface preparation section. Force curves were measured on uncoated cells and encapsulated cells with up to four polymer layers. For each layer, at least five cells were evaluated with no less than 30 measurements for each individual threshold value (see definition below). The same tip was used to collect force curve data for all types of cells (uncoated and encapsulated) for each data set shown in Figs. 6 and 7 (see below). 3. Results and discussion We carried out all experiments in tapping mode in air in order to investigate the morphology, adhesion
3657
properties and robustness of the encapsulated cells. Furthermore, we performed this study in order to assess if encapsulated cells are suitable substrates for nanolithographic approaches [1,5]. In one of these lithographic methodologies, Dip-Pen Nanolithography, the patterning is executed in air using the water meniscus on the tip as a transport media [18]. In another approach, nanografting, high loading forces (mostly in the range of 5–50 nN) are applied to remove molecules from a surface [5]. We chose to use platelets because prior studies have reported them as robust cells that can withstand loading forces of up to 3 nN [19]. Furthermore, B. subtilus cells were utilized in the encapsulation strategy because they are even more robust, have a very thick protein coat and thus have the potential to serve as ‘‘sturdy construction material’’. Moreover, encapsulated microcapsules have been studied with AFM and loaddeformation profiles revealed that such structures can become stiffer at higher loads [20]. The importance of understanding the fundamental properties of polymer encapsulated structures and how they can be used in micro- and nano-construction has been recognized by others [21]. In this study, we detail the properties of the two types of encapsulated cells we prepared, and draw conclusions that are of significant importance when trying to design and execute template fabrication methodologies that incorporate them. 3.1. AFM morphological studies of uncoated and encapsulated cells Images of the two types of cells prior to the coating procedure and subsequently after the encapsulation with polymers were obtained in tapping mode by scanning a bare AFM tip over their surfaces. Fig. 1 shows a representative image from each cell type where one can see the expected shapes and sizes of the platelet and spore cells, part A and B, respectively. The complex surface morphology of the two cell membranes can be seen in these high-resolution AFM images. One observes a small number of artifacts in these images, a finding consistent with reports by others [22]. The artifacts are due to the height of the cells we are imaging, are dependent on the type of cantilever used in the experiment and are more evident in images with scan sizes between 3 and 5 mm. Furthermore, one can get more information regarding the constituents of the cell membrane from the phase images recorded simultaneously with the height ones, shown on the right side of Fig. 1. The different kinds of proteins that make up the cell membrane contribute to the well-pronounced phase shift changes as one moves the tip across the cells to acquire the image. The origins of the phase shift can be linked to several factors [23]. For the experiments presented here the factors that will contribute the most to the differences in phase images will be the interaction
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3658
1 µm
(A)
0.5 µm
(B) Fig. 1. Tapping mode AFM images of: (A) a dried uncoated platelet cell collected in air at room temperature. The scan rate was 2 Hz; (B) dried uncoated B. subtilus cells collected under the same conditions as in (A) and prepared as described in the text. The pictures on the left and right side represent the height and phase images, respectively. The height and phase images were collected simultaneously under the same set of conditions.
0.5 µm
(A)
between the tip and the cell membrane and the composition and properties of the surface (in this case the cell membrane) that is being imaged. In our morphological studies, the phase images provided greater contrast and allowed to make better qualitative comparisons. Upon encapsulation of the cells with PDDA and PSS layers, their morphology changes dramatically, Figs. 2 and 3. The topography images (on the left side of each figure) reveal that as the number of polymer layers increases, the height and surface roughness increase as well. The roughness data is summarized in Fig. 4 (see a detailed discussion below). In addition, the phase images reveal the consistency of polymer coverage around a given cell. The height increase in the images presented in Figs. 2 and 3 is expected due to sequential adsorption of polyanions and polycations. Compared with the phase images of bare cells, on the right side of each figure, the encapsulated cells exhibit a larger number of artifacts. The dramatic increase of the number of artifacts can be qualitatively related to the increase in height of the cells being imaged. As the number of layers increases, the phase images reveal a number of domains in the coatings. Others have recorded similar effects on templated red blood cells [24]. The effect we and others have observed can also be due to fact that the encapsulated cells get dehydrated during the drying procedure [25]. The dehydration can cause the
0.5 µm
(B)
1 µm
(C)
1 µm
(D)
Fig. 2. Tapping mode AFM images of encapsulated dried platelets collected in air, at room temperature and a scan rate of 2 Hz. The polymer layers were built via sequential adsorption of oppositely charged polymers, as described in the text, and images were obtained for human platelets coated with (A) one polymer layer of PDDA; (B) two polymer layers composed of PDDA/PSS; (C) three polymer layers composed of PDDA/PSS/PDDA; and (D) four polymer layers composed of PDDA/PSS/PDDA/PSS. The left part of each AFM microscopy is the height image and the right part is the phase image. These two images were collected simultaneously under the same set of conditions.
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3659
1 µm
1 µm
(B)
(A)
1 µm
(C)
0.5 µm
(D)
Fig. 3. Tapping mode AFM images of encapsulated dried platelets collected in air at room temperature and a scan rate of 2 Hz. The polymer layers were built via sequential adsorption of oppositely charged polymers, as described in the text, and images were obtained for B. subtilus cells coated with (A) one polymer layer of PDDA; (B) two polymer layers composed of PDDA/PSS; (C) three polymer layers composed of PDDA/PSS/PDDA; and (D) four polymer layers composed of PDDA/PSS/PDDA/PSS. The left part of each AFM microscopy is the height image and the right part is the phase image. These two images were collected simultaneously under the same set of conditions.
Fig. 4. Surface roughness statistics extracted for encapsulated (A) platelets and (B) B. subtilus cells. The data was compiled from images collected in tapping mode, in air, at room temperature, as described in the text.
appearance of many folds that dominate in the phase images of the individual encapsulated structures. In addition, when collecting high-resolution images under ambient conditions, depending of the time taken to setup the experiment, we noticed that some of the spore cells started to germinate even when they were encapsulated in polymers, Fig. 3b and c. Other researchers have done a comprehensive study to demonstrate that yeast cells after encapsulation with polymers preserve their metabolic activities and are able to divide [26]. We evaluated the surface roughness of uncoated cells and cells encapsulated with up to four polymer layers, Fig. 4. We report the roughness in terms of their root
mean square (RMS) values we extracted from height images. RMS is the deviation in the height of the evaluated object and can be used to represent the roughness of a certain surface. For each type of cells, RMS values were collected for four individual cells and averaged to describe the roughness of that structure. The consistent roughness increase is expected from the fact that more folds are likely to form as one dries cells with a greater number of polymer layers. Uncoated and encapsulated cells immobilized on surfaces were evaluated after 1, 10, and 30 day(s) from the day of their initial surface attachment. The morphology and roughness characteristics of the encapsulated platelets
ARTICLE IN PRESS 3660
M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
remained intact over that period of time. The encapsulated spore cells were observed to germinate over a very short period of time if they were not refrigerated. 3.2. Adhesion characteristics The adhesion data we present in this study was extracted from a series of force curves. We briefly review the anatomy of a force curve [27], prior to presenting and discussing our data. In these measurements, the tip is moved towards the surfaces and a number of approach-and-retract cycles are collected. The cantilever starts out by not touching the surface until it reaches the point where it feels a repulsive force which causes it to bend. As the piezo continues to retract, the cantilever bends in the opposite direction due to attraction to the surface. Upon further retraction of the piezo the tip breaks free from the surface attraction and rebounds upward. The most important part of this measurement for the studies we report, is the point at which the tip separates from the surface, which is used to quantify the rupture force necessary to break the two apart or the adhesion between the tip and surface. Generally, in cases where thin soft surfaces are studied, the separation between the surface and the tip takes longer, and the adhesion force is larger due to the bigger contact area with the tip [28]. Force curves were collected with a Digital Instruments Multimode Nanoscope IIIa AFM. We chose to perform the force collection in tapping mode to eliminate the repulsive force between the tip and the sample surface. Prior to the force curves collection an image of the cell was obtain in tapping mode and then the force curve data was collected at the center of a given cell. Clean tips and 3-aminopropyltriethoxysilane modified tips (see the materials and methods section) were utilized in this part of the study. In addition, force evaluation was done in trigger mode with assigned positive threshold values in order to precisely control the repulsive force and extract quantitative information.
Fig. 5 contrasts two different threshold values collected on the same cell. The figure also illustrates the dependence of adhesion force on a chosen threshold value when all other parameters in the experiment were kept constant. Thus, one can use different threshold values to either minimize or increase the contact area between the tip and the cell surface. Two types of experiments were done in a serial fashion for a given type of cell: (i) collection of force curves with clean a tip; and (ii) collection of force curves with a 3-aminopropyltriethoxysilane modified tip. For each type of tip, force curves were recorded on uncoated cells and on cells coated with up to four polymer layers. For all cell types we collected force curves at 10 different positive triggering thresholds. In two instances, the number of threshold values we used was less than 10, due to the fact that the threshold limit was reached. We swept the thresholds from 10 to 100 nm with 10 nm steps and collected no less than 30 measurements for each threshold on a selected cell. For each type of uncoated and polymer encapsulated cell we studied, five different cells were evaluated and the average adhesion force for each threshold was obtained, Figs. 6 and 7. All the data presented in the individual part of Figs. 6 and 7 (e.g. Fig. 6a) was collected with the same tip. The adhesion data we collected is presented in Figs. 6 and 7. For all cell types we studied, regardless of whether they were uncoated or encapsulated in polymer layers, an increased adhesion at higher threshold values was recorded. This is as a result of an increase in the contact area of the tip with the surface. Our results indicate that, despite the soft nature of these materials one can control the adhesion between the tip and the surface and, with properly chosen threshold values, minimize the depth of tip penetration in air. Our results are in agreement with previous studies where the properties of uncoated platelets were studied and relatively large loading forces were used [19]. At lower threshold values one can avoid indentation of the bare and encapsulated cell. Therefore, if one uses low loading
Fig. 5. Force vs. distance curves collected in air at room temperature at a frequency of 2 Hz. The curves were collected over a dried platelet coated with four-layers of polymers, as described in the text, using trigger mode at a threshold value of (A) 40 nm and (B) 60 nm.
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3661
Fig. 6. Adhesion force (nN) vs. threshold (nm) data for encapsulated dried platelets collected in air at room temperature and a frequency of 2 Hz, using trigger mode and (A) a clean silicon tip; and (B) a tip modified with 3-aminopropyltriethoxysilane. The numbers next to each curve are used to indicate the following: 0—uncoated platelets; 1—platelets encapsulated with one polymer layer of PDDA; 2—platelets encapsulated with two polymer layers composed of PDDA/PSS; 3—platelets encapsulated with three polymer layers composed of PDDA/PSS/PDDA; 4—platelets encapsulated with four polymer layers composed of PDDA/PSS/PDDA/PSS.
Fig. 7. Adhesion force (nN) vs. threshold (nm) data for encapsulated dried B. subtilus cells collected in air at room temperature and a frequency of 2 Hz, using trigger mode and (A) a silicon clean tip; and (B) a tip modified with 3-aminopropyltriethoxysilane. The numbers next to each curve are used to indicate the following: 0—uncoated spore cells; 1—spore cells encapsulated with one polymer layer of PDDA; 2—spore cells encapsulated with two polymer layers composed of PDDA/PSS; 3—spore cells encapsulated with three polymer layers composed of PDDA/PSS/PDDA; 4—spore cells encapsulated with four polymer layers composed of PDDA/PSS/PDDA/PSS.
forces, these structures can be suitable for lithographic approaches such as Dip-Pen Nanolithography [1], where one delivers molecules from the AFM tip without compromising the integrity of the underlying surface and minimal adhesion of the tip to the surface is advantageous. Moreover, our data indicates that it is easy to determine the threshold value for a given cell type at which one begins to detect large adhesion forces between the tip and the encapsulated cell surface. At these threshold values, lithographic approaches such as nanografting and nanoshaving [5], will be appropriate and can be used to remove a given type of polymer from the encapsulated cell membrane. Furthermore, we were able to increase the adhesion force as we increased the number of threshold increments. However, the linearity was worse in cases when a tip functionalized with an amine was used. This observation is important when designing coating strategies for AFM tips to be used in nanofabrication.
Our systematic adhesion data presented in Figs. 6 and 7 indicate that there is no simple explanation one can use to describe the trends with respect to the nature of the outer layer of a given cell, number of layers surrounding the cell and expected interaction with a specific chemical functionality at the end of the AFM tip. The complexity of the morphological and mechanical properties of similar organic and inorganic encapsulated structures is documented in the literature [20,21,27,29–33]. Zhu and co-workers have used pulsed-force-mode to differentiate between surface adhesion properties of gold surfaces coated with PSS based on ‘‘differences of adhesive forces of the tip/ sample and the tip/surface couples’’ [27]. Through the pulsed-force-mode technique they were able to increase the roughness limit for investigating polymers to 6.1 nm. Due to the dehydration effects discussed above a number of our structures are characterized by much higher roughness values (Fig. 4). In other studies,
ARTICLE IN PRESS 3662
M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
researchers have suggested the possibility of forming an electrostatically stabilized 3D net around the encapsulated structures [20,21]. Such a 3D net can play a substantial role in properties such as adhesion and can be one of the reasons for our data trends. Separating the different kinds of effects that contribute to the overall complexity of the data is not trivial. Experiments to gain insight into such mechanisms might include: changing the nature of the alternating polymer layers, in terms of chemical functionalities and molecular weight [12]; exploring different solvent systems during the polyelectrolyte assembly procedure that are known to effect the shell permeability [34]; and encapsulation of ‘‘hard materials’’, such as nanoparticles in between the polymer layers [9]. The extent of these studies requires a separate manuscript. Our studies allow us to map the properties of encapsulated cells in sufficient detail to enable us to design lithographic approaches that utilize an AFM tip to pattern onto soft, biomimetic surfaces. The products from such lithographic methodologies have the potential to be used in medical applications for diagnostics and therapeutics, catalysis studies and to participate in the construction of bioelectronic devices.
4. Summary In this work, we systematically evaluated human platelets and B. subtilus spore cells coated with different numbers of polymer layers. Their adhesion characteristics with two different types of AFM tips were mapped. The roughness of the encapsulated cells increased as the number of adsorbed polymer layers increased and the cells were dried onto a solid surface. The encapsulated cells show adhesion characteristics that are dependent not only on the number, chemical nature and thickness of the layers, but also on the interactions between those layers. Explaining the data simply in terms of the number of coating cycles or depth of tip penetration is an oversimplification. One can design nano-lithographic approaches using scanning probe techniques based on the adhesion and morphological studies reported in this work. Furthermore, the encapsulated structures are robust which together with their other characteristics qualifies them as a good choice for the design and construction of templated biomimetic architectures.
based upon work supported by NASA under award No. NCC 2-1363. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Aeronautics and Space Administration.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]
Acknowledgements
[31]
We thank Tim Miller for help with the thermal evaporator and Sharon Snyder for collecting the blood. Bindley Biosciences Center at Purdue’s Discovery Park is acknowledged for start-up funds. This material is
[32] [33] [34]
Mirkin CA. MRS Bull 2001;26:535–8. Niemeyer CM. Angew Chem Int Ed 2001;40:4128–58. Binnig G, Quate CF, Gerber C. Phys Rev Lett 1986;56:930–3. Langer R. Accounts Chem Res 2000;33:94–101. Liu GY, Xu S, Qian Y. Accounts Chem Res 2000;33:457–66. Leng YS, Jiang SY. J Am Chem Soc 2002;124:11764–70. Lehenkari PP, Charras GT, Nesbitt SA, Horton MA. Expert Rev Mol Med 2000;1–19. Grandbois M, Dettmann W, Benoit M, Gaub HE. J Histochem Cytochem 2000;48:719–24. Ai H, Fang M, Jones SA, Lvov YM. Biomacromolecules 2002;3:560–4. McGregor JL, Clemetson KJ, James E, Luscher EF, Dechavannne M. Biochim Biophys Acta 1980;599:473–83. Decher G. Science 1997;227:1232–5. Caruso F. Adv Mat 2001;13:11–22. Kurth DG, Bein T. Langmuir 1993;9:2965–73. Oh SJ, Cho SJ, Kim CO, Park JW. Langmuir 2002;18:1764–9. Dubois LH, Nuzzo RG. Annu Rev Phys Chem 1992;43:437–63. Noble-Luginbuhl AR, Nuzzo RG. Langmuir 2001;17:3937–44. Choi J, Bogdanski D, Koller M, Esenwein SA, Muller D, Muhr G, Epple M. Biomaterials 2003;24:3689–96. Piner RD, Zhu J, Xu F, Hong S, Mirkin CA. Science 1999;283:661–3. Radmacher M, Fritz M, Kacher CM, Cleveland JP, Hansma PK. Biophys J 1996;70:556–67. Lulevich VV, Rudtchenko IL, Sukhorukov GB, Vinogradova OI. J Phys Chem B 2003;107:2735–40. Lulevich VV, Radtchenko IL, Sukhorukov GB, Vinogradova OI. Macromolecules 2003;36:2832–7. Velegol SB, Pardi S, Li X, Velegol D, Logan BE. Langmuir 2003;19:851–7. Noy A, Sanders CH, Vezenov DV, Wong SS, Lieber CM. Langmuir 1998;14:1508–11. Leporatti S, Voigt A, Mitlohner R, Sukhorukov G, Donath E, Mohwald H. Langmuir 2000;16:4059–63. Gao C, Leporatti S, Moya S, Donath E, Mohwald H. Langmuir 2001;17:3491–5. Diaspro A, Silvana D, Krol S, Cavalleri O, Gliozzi A. Langmuir 2002;18:5047–50. Zhu M, Akari S, Mohwald H. Nano Lett 2001;1:569–73. Radmacher M, Tillmann RW, Fritz M, Gaub HE. Science 1992;257:1900–5. Lukkari J, Salomaki M, Aaritalo T, Loikas K, Laiho T, Kankare J. Langmuir 2002;18:8496–502. Nonogaki T, Xu S, Kugimiya S, Sato S, Miyata I, Yonese M. Langmuir 2000;16:4272–8. Kim DW, Blumstein A, Kumar J, Samuelson LA, Kang B, Sung C. Chem Mater 2002;14:3925–9. Sasaki T, Ebina Y, Tanaka T, Harada M, Watanabe M. Chem Mater 2001;13:4661–7. Khopade AJ, Caruso F. Langmuir 2002;18:7669–76. Lvov Y, Antipov AA, Mamedov A, Mohwald H, Sukhorukov GB. Nano Lett 2001;1:125–8.
Biomaterials 25 (2004) 3655–3662
Encapsulated cells: an atomic force microscopy study Meng Yua, Albena Ivanisevica,b,* a
Department of Biomedical Engineering, Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA b Department of Chemistry, Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA Received 27 February 2003; accepted 10 October 2003
Abstract Two types of cells—human platelets and spore cells—were encapsulated in polymer shells by adsorbing polyanions and polycations in a stepwise fashion. The encapsulated cells were attached to gold and silicon surfaces and their morphological and adhesion properties were studied in air using tapping mode atomic force microscopy (AFM). The roughness of the encapsulated cells increased upon the addition of a new polymer layer. The increase in roughness can be attributed to the formation of a shell around the cells, which is stabilized by electrostatic interactions, as well as to the drying effects associated with the immobilization and dehydration of the cells. Trigger mode was used to perform the force imaging and map out the adhesion characteristics of the cells. Systematic ‘‘maps’’ of the adhesion properties of the encapsulated cells to clean and amine terminated AFM tips were collected. The adhesion force data for the different tips and encapsulated cells showed dependence not only on the number and thickness of the polymer layers, but also on the interactions between these layers. The encapsulated cells’ morphology and roughness characteristics remained intact over a substantial storage period. This stability and adhesion properties make them suitable building blocks for the design and construction of biomimetic templates where AFM is used as the primary tool to do the fabrication. r 2003 Elsevier Ltd. All rights reserved. Keywords: Platelets; Spores; Polymers; Atomic force microscopy; Adhesion
1. Introduction The development of methods to position, manipulate and image single molecules is important from a fundamental as well as applied standpoint. Basic understanding of how single molecules, and in particular biologically relevant ones, take advantage of their unique properties such as elasticity and biochemical host–guest recognition can help in designing strategies to protect and regenerate biological scaffolds. Furthermore, one can utilize this information to fabricate improved biocompatible devices. A promising route to make this a reality is the usage of surface templates to mediate the building of complex structures via a bottom-up strategy [1]. The identification, fabrication and evaluation of biomimetic templates holds great potential for a variety of applications [2]. The engineering of biomimetic surfaces that can act as sensors for the *Corresponding author. Department of Biomedical Engineering and Department of Chemistry, Purdue University, 500 Central Drive, West Lafayette, IN 47907, USA. E-mail address: [email protected] (A. Ivanisevic). 0142-9612/$ - see front matter r 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.biomaterials.2003.10.061
purpose of enhancing recognition processes and achieving great control over desired morphology, electronic and mechanical properties, is an extremely valuable investment for future versatile systems. The atomic force microscope (AFM) [3] is a commercial tool that is used to investigate the properties of biomimetic templates. In morphological studies it can serve as a tool to acquire images with nanometer resolution by rastering a sharp tip across a surface. In addition, this technique can be utilized to measure adhesion forces between the tip modified with a certain chemical functionalities and a sample surface. The adhesion properties of a given material play a crucial role in sub-micron and nanometer scale surface modification and self-assembly processes [4]. A number of applications, such as AFM-based lithographic fabrication protocols, directly depend upon properties such as adhesion and diffusion [1,5]. In order to achieve robustness and reproducibility during the fabrication process, it is essential to have a fundamental understanding of the properties of different materials. AFM force measurements can be applied as a quantitative tool to record and compare attractive and repulsive
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3656
forces [6]. The force maps one can generate allow to elucidate mechanical properties such as adhesion and elasticity [7]. Unlike a lot of conventional hard materials, biomaterials can be fragile and thus easily damaged during a number of surface immobilization procedures. Therefore it is essential to extract the mechanical properties of these materials prior to incorporating them in micro- and nano-fabrication protocols that produce complex surface templates. Using AFM to analyze biomaterials [8] in terms of their adhesion properties is advantageous compared to other methods due to its sensitivity often quoted in the range of 1012 N. In addition, one can position the tip on a given surface coordinate with a precision of a few nanometers. In this study, we report the morphological and adhesion properties of encapsulated cells. To create functional structures we followed a procedure reported by Lvov et al. [9] for bovine platelets. We use two types of cells: (i) human platelets, blood cells in the human body; and (ii) Bacillus subtilus (B. subtilus) spore cells. To encapsulate the cells in charged polymer shells, we took advantage of the fact that their surfaces are covered with different kinds of glycoproteins [10]. By tuning the pH of the solution they are dispersed in, one can treat them as charged micro-structures and functionalize them with nanometer size biomimetic shells [9]. The modification of the two types of cells we used was done using the layer-by-layer (LbL) [11], approach by choosing proper polycations and polyanions (Chart 1). This approach allows us to terminate the cells’ surfaces with a given chemical functionality and has been adapted in modifying a wealth of organic, inorganic and biological structures [12]. In this paper, we systematically map out the properties of the encapsulated cells using AFM. We plan to incorporate the results of this study in the development of methods to pattern these functional surfaces via scanning lithography approaches [1,5].
2. Materials and methods 2.1. Materials 3-Aminopropyltriethoxysilane (99%), poly(diallyldimethylammonium chloride) (PDDA, 20 wt% in water, PDDA *
H2C
H3C
N
+
n
*
*
CH
CH2
Cl CH3
SO3 Na+
Chart 1.
2.2. Silicon surface preparation Silicon pieces ð1 1 cm2 Þ were cut from a 400 wafer and cleaned with DI water, EtOH and MeOH using an ultrasonic bath. Each piece was incubated in a piranha solution (mixture of sulfuric acid (51%) and hydrogen peroxide (30%) in a 3:1 ratio, v/v) for 25 min at room temperature. The surface was rinsed with DI water and EtOH several times and dried under a stream of nitrogen. Subsequently, the substrates were incubated in a home-built chamber in the presence of vapors of 3aminopropyltriethoxysilane and water for 40 min. To perform this vapor-phase silanization we followed a method described by Bein et al. [13]. Each surface was baked at 120 C in a laboratory oven for 30 min. Contact angle measurements were done before and after the silanization to verify the treatment. These measurements matched the literature [14]. 2.3. Gold surface preparation A silicon wafer with 500 nm oxide layer was coated with 10 nm Ti followed by 30 nm Au using a thermal evaporator. The evaporation was performed at an evaporation rate of 1 nm/s and a vacuum pressure of 3 107 Torr: After the evaporation the wafer was cut into 1 1 cm2 pieces. Each piece was incubated in 1 mm MHA solution in EtOH for 24 h in order to create a selfassembled monolayer on the surface. Contact angle measurements of these substrates matched the literature [15,16]. 2.4. Human platelets extraction
PSS CH2
average Mw 100,000–200,000), poly(sodium 4-styrenesulfonate) (PSS, average Mw B 70,000), 16-mercaptohexadecanoic acid (MHA), hydrogen peroxide (30%), sulfuric acid, HPLC grade methanol (MeOH) and ethanol (EtOH) were purchased from Aldrich. Phosphate-buffered saline tablets (PBS) and 1% paraformaldehyde were purchased from Sigma. All chemicals were used as received. Deionized (DI) water (18.0 MO cm) was obtained from a Millipore Milli-Q system. B. subtilis (sus-1A-8) water-suspension at 108 CFU were obtained from Raven Biological Laboratories (NE). Silicon wafers with 500 nm of thermally evaporated oxide were fabricated by WaferNet, Inc, CA. Solutions of 2 mg/ml PDDA and 3 mg/ml PSS were prepared in PBS at pH 7.4.
n
*
Platelets were purified from fresh human blood using a platelet-rich plasma (PRP) method [17]. Fresh human whole blood was centrifuged at 66g for 10 min at 25 C to obtain PRP (supernatant). PRP was immediately fixed in 1% paraformaldehyde solution for 2 h. Fixed platelets were purified by centrifugation at 640g for
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
5 min at 25 C and re-suspended in PBS solution (pH 7.4). The final concentration of the platelets was the same as the concentration of PRP. 2.5. Encapsulated cells We followed a modified LbL procedure described by Lvov et al. [9]. Briefly, cells were incubated with shaking in a 2 mg/ml PDDA solution for 20 min at room temperature. Three washing cycles with PBS solution were done by centrifugation at 640g for 5 min at 25 C to remove the excess polymers. Through the same procedure, a negatively-charged PSS layer was deposited over the PDDA layer. The last two steps were repeated to achieve a desired number of polymer layers. 2.6. Attachment of cells to surfaces Prior to the tapping mode imaging, cells and encapsulated cells were attached to solid substrates in the following fashion: Cells terminated with a PDDA layer were attached to gold surfaces modified with MHA. Cells terminated with a PSS layer were attached to silanized surfaces. Uncoated cells were attached to silanized surfaces. In each case, a drop of the appropriate type of cells was incubated on top of a given surface for 2 min and subsequently washed several times with PBS buffer, pure water and dried under a stream of nitrogen. 2.7. AFM imaging and force evaluation Tapping mode images and force curves were collected with a Multimode Nanoscope IIIa (Digital Instruments, Santa Barbara, CA) using clean tips and 3-aminopropyltriethoxysilane modified tips. All experiments were done with silicon tapping mode tips purchased from Digital Instruments, model number RTESP7 with a length of 125 mm, frequency 300 kHz, a spring constant of 20 N/m and a rotated tip. Tips were cleaned with a piranha solution, EtOH solution and then coated with 3-aminopropyltriethoxysilane using a procedure by Bein et al. [13] described under the silicon surface preparation section. Force curves were measured on uncoated cells and encapsulated cells with up to four polymer layers. For each layer, at least five cells were evaluated with no less than 30 measurements for each individual threshold value (see definition below). The same tip was used to collect force curve data for all types of cells (uncoated and encapsulated) for each data set shown in Figs. 6 and 7 (see below). 3. Results and discussion We carried out all experiments in tapping mode in air in order to investigate the morphology, adhesion
3657
properties and robustness of the encapsulated cells. Furthermore, we performed this study in order to assess if encapsulated cells are suitable substrates for nanolithographic approaches [1,5]. In one of these lithographic methodologies, Dip-Pen Nanolithography, the patterning is executed in air using the water meniscus on the tip as a transport media [18]. In another approach, nanografting, high loading forces (mostly in the range of 5–50 nN) are applied to remove molecules from a surface [5]. We chose to use platelets because prior studies have reported them as robust cells that can withstand loading forces of up to 3 nN [19]. Furthermore, B. subtilus cells were utilized in the encapsulation strategy because they are even more robust, have a very thick protein coat and thus have the potential to serve as ‘‘sturdy construction material’’. Moreover, encapsulated microcapsules have been studied with AFM and loaddeformation profiles revealed that such structures can become stiffer at higher loads [20]. The importance of understanding the fundamental properties of polymer encapsulated structures and how they can be used in micro- and nano-construction has been recognized by others [21]. In this study, we detail the properties of the two types of encapsulated cells we prepared, and draw conclusions that are of significant importance when trying to design and execute template fabrication methodologies that incorporate them. 3.1. AFM morphological studies of uncoated and encapsulated cells Images of the two types of cells prior to the coating procedure and subsequently after the encapsulation with polymers were obtained in tapping mode by scanning a bare AFM tip over their surfaces. Fig. 1 shows a representative image from each cell type where one can see the expected shapes and sizes of the platelet and spore cells, part A and B, respectively. The complex surface morphology of the two cell membranes can be seen in these high-resolution AFM images. One observes a small number of artifacts in these images, a finding consistent with reports by others [22]. The artifacts are due to the height of the cells we are imaging, are dependent on the type of cantilever used in the experiment and are more evident in images with scan sizes between 3 and 5 mm. Furthermore, one can get more information regarding the constituents of the cell membrane from the phase images recorded simultaneously with the height ones, shown on the right side of Fig. 1. The different kinds of proteins that make up the cell membrane contribute to the well-pronounced phase shift changes as one moves the tip across the cells to acquire the image. The origins of the phase shift can be linked to several factors [23]. For the experiments presented here the factors that will contribute the most to the differences in phase images will be the interaction
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3658
1 µm
(A)
0.5 µm
(B) Fig. 1. Tapping mode AFM images of: (A) a dried uncoated platelet cell collected in air at room temperature. The scan rate was 2 Hz; (B) dried uncoated B. subtilus cells collected under the same conditions as in (A) and prepared as described in the text. The pictures on the left and right side represent the height and phase images, respectively. The height and phase images were collected simultaneously under the same set of conditions.
0.5 µm
(A)
between the tip and the cell membrane and the composition and properties of the surface (in this case the cell membrane) that is being imaged. In our morphological studies, the phase images provided greater contrast and allowed to make better qualitative comparisons. Upon encapsulation of the cells with PDDA and PSS layers, their morphology changes dramatically, Figs. 2 and 3. The topography images (on the left side of each figure) reveal that as the number of polymer layers increases, the height and surface roughness increase as well. The roughness data is summarized in Fig. 4 (see a detailed discussion below). In addition, the phase images reveal the consistency of polymer coverage around a given cell. The height increase in the images presented in Figs. 2 and 3 is expected due to sequential adsorption of polyanions and polycations. Compared with the phase images of bare cells, on the right side of each figure, the encapsulated cells exhibit a larger number of artifacts. The dramatic increase of the number of artifacts can be qualitatively related to the increase in height of the cells being imaged. As the number of layers increases, the phase images reveal a number of domains in the coatings. Others have recorded similar effects on templated red blood cells [24]. The effect we and others have observed can also be due to fact that the encapsulated cells get dehydrated during the drying procedure [25]. The dehydration can cause the
0.5 µm
(B)
1 µm
(C)
1 µm
(D)
Fig. 2. Tapping mode AFM images of encapsulated dried platelets collected in air, at room temperature and a scan rate of 2 Hz. The polymer layers were built via sequential adsorption of oppositely charged polymers, as described in the text, and images were obtained for human platelets coated with (A) one polymer layer of PDDA; (B) two polymer layers composed of PDDA/PSS; (C) three polymer layers composed of PDDA/PSS/PDDA; and (D) four polymer layers composed of PDDA/PSS/PDDA/PSS. The left part of each AFM microscopy is the height image and the right part is the phase image. These two images were collected simultaneously under the same set of conditions.
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3659
1 µm
1 µm
(B)
(A)
1 µm
(C)
0.5 µm
(D)
Fig. 3. Tapping mode AFM images of encapsulated dried platelets collected in air at room temperature and a scan rate of 2 Hz. The polymer layers were built via sequential adsorption of oppositely charged polymers, as described in the text, and images were obtained for B. subtilus cells coated with (A) one polymer layer of PDDA; (B) two polymer layers composed of PDDA/PSS; (C) three polymer layers composed of PDDA/PSS/PDDA; and (D) four polymer layers composed of PDDA/PSS/PDDA/PSS. The left part of each AFM microscopy is the height image and the right part is the phase image. These two images were collected simultaneously under the same set of conditions.
Fig. 4. Surface roughness statistics extracted for encapsulated (A) platelets and (B) B. subtilus cells. The data was compiled from images collected in tapping mode, in air, at room temperature, as described in the text.
appearance of many folds that dominate in the phase images of the individual encapsulated structures. In addition, when collecting high-resolution images under ambient conditions, depending of the time taken to setup the experiment, we noticed that some of the spore cells started to germinate even when they were encapsulated in polymers, Fig. 3b and c. Other researchers have done a comprehensive study to demonstrate that yeast cells after encapsulation with polymers preserve their metabolic activities and are able to divide [26]. We evaluated the surface roughness of uncoated cells and cells encapsulated with up to four polymer layers, Fig. 4. We report the roughness in terms of their root
mean square (RMS) values we extracted from height images. RMS is the deviation in the height of the evaluated object and can be used to represent the roughness of a certain surface. For each type of cells, RMS values were collected for four individual cells and averaged to describe the roughness of that structure. The consistent roughness increase is expected from the fact that more folds are likely to form as one dries cells with a greater number of polymer layers. Uncoated and encapsulated cells immobilized on surfaces were evaluated after 1, 10, and 30 day(s) from the day of their initial surface attachment. The morphology and roughness characteristics of the encapsulated platelets
ARTICLE IN PRESS 3660
M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
remained intact over that period of time. The encapsulated spore cells were observed to germinate over a very short period of time if they were not refrigerated. 3.2. Adhesion characteristics The adhesion data we present in this study was extracted from a series of force curves. We briefly review the anatomy of a force curve [27], prior to presenting and discussing our data. In these measurements, the tip is moved towards the surfaces and a number of approach-and-retract cycles are collected. The cantilever starts out by not touching the surface until it reaches the point where it feels a repulsive force which causes it to bend. As the piezo continues to retract, the cantilever bends in the opposite direction due to attraction to the surface. Upon further retraction of the piezo the tip breaks free from the surface attraction and rebounds upward. The most important part of this measurement for the studies we report, is the point at which the tip separates from the surface, which is used to quantify the rupture force necessary to break the two apart or the adhesion between the tip and surface. Generally, in cases where thin soft surfaces are studied, the separation between the surface and the tip takes longer, and the adhesion force is larger due to the bigger contact area with the tip [28]. Force curves were collected with a Digital Instruments Multimode Nanoscope IIIa AFM. We chose to perform the force collection in tapping mode to eliminate the repulsive force between the tip and the sample surface. Prior to the force curves collection an image of the cell was obtain in tapping mode and then the force curve data was collected at the center of a given cell. Clean tips and 3-aminopropyltriethoxysilane modified tips (see the materials and methods section) were utilized in this part of the study. In addition, force evaluation was done in trigger mode with assigned positive threshold values in order to precisely control the repulsive force and extract quantitative information.
Fig. 5 contrasts two different threshold values collected on the same cell. The figure also illustrates the dependence of adhesion force on a chosen threshold value when all other parameters in the experiment were kept constant. Thus, one can use different threshold values to either minimize or increase the contact area between the tip and the cell surface. Two types of experiments were done in a serial fashion for a given type of cell: (i) collection of force curves with clean a tip; and (ii) collection of force curves with a 3-aminopropyltriethoxysilane modified tip. For each type of tip, force curves were recorded on uncoated cells and on cells coated with up to four polymer layers. For all cell types we collected force curves at 10 different positive triggering thresholds. In two instances, the number of threshold values we used was less than 10, due to the fact that the threshold limit was reached. We swept the thresholds from 10 to 100 nm with 10 nm steps and collected no less than 30 measurements for each threshold on a selected cell. For each type of uncoated and polymer encapsulated cell we studied, five different cells were evaluated and the average adhesion force for each threshold was obtained, Figs. 6 and 7. All the data presented in the individual part of Figs. 6 and 7 (e.g. Fig. 6a) was collected with the same tip. The adhesion data we collected is presented in Figs. 6 and 7. For all cell types we studied, regardless of whether they were uncoated or encapsulated in polymer layers, an increased adhesion at higher threshold values was recorded. This is as a result of an increase in the contact area of the tip with the surface. Our results indicate that, despite the soft nature of these materials one can control the adhesion between the tip and the surface and, with properly chosen threshold values, minimize the depth of tip penetration in air. Our results are in agreement with previous studies where the properties of uncoated platelets were studied and relatively large loading forces were used [19]. At lower threshold values one can avoid indentation of the bare and encapsulated cell. Therefore, if one uses low loading
Fig. 5. Force vs. distance curves collected in air at room temperature at a frequency of 2 Hz. The curves were collected over a dried platelet coated with four-layers of polymers, as described in the text, using trigger mode at a threshold value of (A) 40 nm and (B) 60 nm.
ARTICLE IN PRESS M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
3661
Fig. 6. Adhesion force (nN) vs. threshold (nm) data for encapsulated dried platelets collected in air at room temperature and a frequency of 2 Hz, using trigger mode and (A) a clean silicon tip; and (B) a tip modified with 3-aminopropyltriethoxysilane. The numbers next to each curve are used to indicate the following: 0—uncoated platelets; 1—platelets encapsulated with one polymer layer of PDDA; 2—platelets encapsulated with two polymer layers composed of PDDA/PSS; 3—platelets encapsulated with three polymer layers composed of PDDA/PSS/PDDA; 4—platelets encapsulated with four polymer layers composed of PDDA/PSS/PDDA/PSS.
Fig. 7. Adhesion force (nN) vs. threshold (nm) data for encapsulated dried B. subtilus cells collected in air at room temperature and a frequency of 2 Hz, using trigger mode and (A) a silicon clean tip; and (B) a tip modified with 3-aminopropyltriethoxysilane. The numbers next to each curve are used to indicate the following: 0—uncoated spore cells; 1—spore cells encapsulated with one polymer layer of PDDA; 2—spore cells encapsulated with two polymer layers composed of PDDA/PSS; 3—spore cells encapsulated with three polymer layers composed of PDDA/PSS/PDDA; 4—spore cells encapsulated with four polymer layers composed of PDDA/PSS/PDDA/PSS.
forces, these structures can be suitable for lithographic approaches such as Dip-Pen Nanolithography [1], where one delivers molecules from the AFM tip without compromising the integrity of the underlying surface and minimal adhesion of the tip to the surface is advantageous. Moreover, our data indicates that it is easy to determine the threshold value for a given cell type at which one begins to detect large adhesion forces between the tip and the encapsulated cell surface. At these threshold values, lithographic approaches such as nanografting and nanoshaving [5], will be appropriate and can be used to remove a given type of polymer from the encapsulated cell membrane. Furthermore, we were able to increase the adhesion force as we increased the number of threshold increments. However, the linearity was worse in cases when a tip functionalized with an amine was used. This observation is important when designing coating strategies for AFM tips to be used in nanofabrication.
Our systematic adhesion data presented in Figs. 6 and 7 indicate that there is no simple explanation one can use to describe the trends with respect to the nature of the outer layer of a given cell, number of layers surrounding the cell and expected interaction with a specific chemical functionality at the end of the AFM tip. The complexity of the morphological and mechanical properties of similar organic and inorganic encapsulated structures is documented in the literature [20,21,27,29–33]. Zhu and co-workers have used pulsed-force-mode to differentiate between surface adhesion properties of gold surfaces coated with PSS based on ‘‘differences of adhesive forces of the tip/ sample and the tip/surface couples’’ [27]. Through the pulsed-force-mode technique they were able to increase the roughness limit for investigating polymers to 6.1 nm. Due to the dehydration effects discussed above a number of our structures are characterized by much higher roughness values (Fig. 4). In other studies,
ARTICLE IN PRESS 3662
M. Yu, A. Ivanisevic / Biomaterials 25 (2004) 3655–3662
researchers have suggested the possibility of forming an electrostatically stabilized 3D net around the encapsulated structures [20,21]. Such a 3D net can play a substantial role in properties such as adhesion and can be one of the reasons for our data trends. Separating the different kinds of effects that contribute to the overall complexity of the data is not trivial. Experiments to gain insight into such mechanisms might include: changing the nature of the alternating polymer layers, in terms of chemical functionalities and molecular weight [12]; exploring different solvent systems during the polyelectrolyte assembly procedure that are known to effect the shell permeability [34]; and encapsulation of ‘‘hard materials’’, such as nanoparticles in between the polymer layers [9]. The extent of these studies requires a separate manuscript. Our studies allow us to map the properties of encapsulated cells in sufficient detail to enable us to design lithographic approaches that utilize an AFM tip to pattern onto soft, biomimetic surfaces. The products from such lithographic methodologies have the potential to be used in medical applications for diagnostics and therapeutics, catalysis studies and to participate in the construction of bioelectronic devices.
4. Summary In this work, we systematically evaluated human platelets and B. subtilus spore cells coated with different numbers of polymer layers. Their adhesion characteristics with two different types of AFM tips were mapped. The roughness of the encapsulated cells increased as the number of adsorbed polymer layers increased and the cells were dried onto a solid surface. The encapsulated cells show adhesion characteristics that are dependent not only on the number, chemical nature and thickness of the layers, but also on the interactions between those layers. Explaining the data simply in terms of the number of coating cycles or depth of tip penetration is an oversimplification. One can design nano-lithographic approaches using scanning probe techniques based on the adhesion and morphological studies reported in this work. Furthermore, the encapsulated structures are robust which together with their other characteristics qualifies them as a good choice for the design and construction of templated biomimetic architectures.
based upon work supported by NASA under award No. NCC 2-1363. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Aeronautics and Space Administration.
References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30]
Acknowledgements
[31]
We thank Tim Miller for help with the thermal evaporator and Sharon Snyder for collecting the blood. Bindley Biosciences Center at Purdue’s Discovery Park is acknowledged for start-up funds. This material is
[32] [33] [34]
Mirkin CA. MRS Bull 2001;26:535–8. Niemeyer CM. Angew Chem Int Ed 2001;40:4128–58. Binnig G, Quate CF, Gerber C. Phys Rev Lett 1986;56:930–3. Langer R. Accounts Chem Res 2000;33:94–101. Liu GY, Xu S, Qian Y. Accounts Chem Res 2000;33:457–66. Leng YS, Jiang SY. J Am Chem Soc 2002;124:11764–70. Lehenkari PP, Charras GT, Nesbitt SA, Horton MA. Expert Rev Mol Med 2000;1–19. Grandbois M, Dettmann W, Benoit M, Gaub HE. J Histochem Cytochem 2000;48:719–24. Ai H, Fang M, Jones SA, Lvov YM. Biomacromolecules 2002;3:560–4. McGregor JL, Clemetson KJ, James E, Luscher EF, Dechavannne M. Biochim Biophys Acta 1980;599:473–83. Decher G. Science 1997;227:1232–5. Caruso F. Adv Mat 2001;13:11–22. Kurth DG, Bein T. Langmuir 1993;9:2965–73. Oh SJ, Cho SJ, Kim CO, Park JW. Langmuir 2002;18:1764–9. Dubois LH, Nuzzo RG. Annu Rev Phys Chem 1992;43:437–63. Noble-Luginbuhl AR, Nuzzo RG. Langmuir 2001;17:3937–44. Choi J, Bogdanski D, Koller M, Esenwein SA, Muller D, Muhr G, Epple M. Biomaterials 2003;24:3689–96. Piner RD, Zhu J, Xu F, Hong S, Mirkin CA. Science 1999;283:661–3. Radmacher M, Fritz M, Kacher CM, Cleveland JP, Hansma PK. Biophys J 1996;70:556–67. Lulevich VV, Rudtchenko IL, Sukhorukov GB, Vinogradova OI. J Phys Chem B 2003;107:2735–40. Lulevich VV, Radtchenko IL, Sukhorukov GB, Vinogradova OI. Macromolecules 2003;36:2832–7. Velegol SB, Pardi S, Li X, Velegol D, Logan BE. Langmuir 2003;19:851–7. Noy A, Sanders CH, Vezenov DV, Wong SS, Lieber CM. Langmuir 1998;14:1508–11. Leporatti S, Voigt A, Mitlohner R, Sukhorukov G, Donath E, Mohwald H. Langmuir 2000;16:4059–63. Gao C, Leporatti S, Moya S, Donath E, Mohwald H. Langmuir 2001;17:3491–5. Diaspro A, Silvana D, Krol S, Cavalleri O, Gliozzi A. Langmuir 2002;18:5047–50. Zhu M, Akari S, Mohwald H. Nano Lett 2001;1:569–73. Radmacher M, Tillmann RW, Fritz M, Gaub HE. Science 1992;257:1900–5. Lukkari J, Salomaki M, Aaritalo T, Loikas K, Laiho T, Kankare J. Langmuir 2002;18:8496–502. Nonogaki T, Xu S, Kugimiya S, Sato S, Miyata I, Yonese M. Langmuir 2000;16:4272–8. Kim DW, Blumstein A, Kumar J, Samuelson LA, Kang B, Sung C. Chem Mater 2002;14:3925–9. Sasaki T, Ebina Y, Tanaka T, Harada M, Watanabe M. Chem Mater 2001;13:4661–7. Khopade AJ, Caruso F. Langmuir 2002;18:7669–76. Lvov Y, Antipov AA, Mamedov A, Mohwald H, Sukhorukov GB. Nano Lett 2001;1:125–8.