|Home | About | Journals | Submit | Contact Us | Français|
The past decade has seen substantial growth in research into how changes in the biomechanical and biophysical properties of cells and subcellular structures influence, and are influenced by, the onset and progression of human diseases. This paper presents an overview of the rapidly expanding, nascent field of research that deals with the biomechanics and biophysics of cancer cells. The review begins with some key observations on the biology of cancer cells and on the role of actin microfilaments, intermediate filaments and microtubule biopolymer cytoskeletal components in influencing cell mechanics, locomotion, differentiation and neoplastic transformation. In order to set the scene for mechanistic discussions of the connections among alterations to subcellular structures, attendant changes in cell deformability, cytoadherence, migration, invasion and tumor metastasis, a survey is presented of the various quantitative mechanical and physical assays to extract the elastic and viscoelastic deformability of cancer cells. Results available in the literature on cell mechanics for different types of cancer are then reviewed. Representative case studies are presented next to illustrate how chemically induced cytoskeletal changes, biomechanical responses and signals from the intracellular regions act in concert with the chemomechanical environment of the extracellular matrix and the molecular tumorigenic signaling pathways to effect malignant transformations. Results are presented to illustrate how changes to cytoskeletal architecture induced by cancer drugs and chemotherapy regimens can significantly influence cell mechanics and disease state. It is reasoned through experimental evidence that greater understanding of the mechanics of cancer cell deformability and its interactions with the extracellular physical, chemical and biological environments offers enormous potential for significant new developments in disease diagnostics, prophylactics, therapeutics and drug efficacy assays.
The links between biomechanics and human diseases have been the subject of considerable scientific research effort for a number of decades. The application of traditional “solid” and “fluid” mechanics concepts from physics and engineering to the study of biological and physiological problems has provided valuable insights into the mechanics of organ function, the mechanical response of tissues, joints and articulating surfaces, and the physical and mechanical processes associated with blood flow through the vasculature (see e.g. Refs. [1–7] and the sources cited therein). Classical “continuum mechanics” approaches have also provided a broad and fundamental framework within which phenomenological responses associated with biological systems could be consistently and quantitatively rationalized. Consequently, such approaches have also been adapted, with appropriate modifications, to model the mechanics of deformation of biological cells, subcellular components such as the cytoskeleton and phospholipid bilayer membrane, and biological molecular networks and attachment systems. Examples of concepts adapted from engineering science to the study of cell mechanobiology include:
Continuum mechanics concepts have thus provided a broad quantitative framework to categorize and rationalize biomechanical processes and mechanisms without specifically and directly addressing the issues of length- and time-scales of particular concern to biochemical mechanistic phenomena at the cellular and molecular levels. In the past decade, however, rapid advances in engineering, technology, physical, life and information sciences and medicine, as well as the growing trend to integrate such advances across multidisciplinary boundaries, have produced spectacular progress in the study of mechanics and physics at the levels of individual cells and biomolecules. These advances have significant implications for cancer research and treatment. Specifically,
In light of these trends and developments, the study of the influence of cellular, subcellular and molecular mechanics on human disease states, including cancer, has emerged as a topic of rapidly expanding scientific interest in recent years. A particular focus of research in this area is to explore the connections among the cell ultrastructure, cellular and cytoskeletal mechanical properties, biological function and human health/disease by specifically addressing the following issues.
These relationships among cell structure, biomechanics and disease states are schematically illustrated in Fig. 1. This structure–property–function–disease paradigm is of particular interest to the growing community of researchers dealing with the mechanics of cancer cells.
A necessary first step in establishing connections among chemobiomechanical pathways in the context of human diseases inevitably requires systematic studies of cell mechanical properties as a function of disease state under conditions that mirror in vivo phenomena. Given the complexity of the mechanisms underlying this process, it is not surprising that comprehensive studies involving detailed cell mechanics experiments that cover the full range of appropriate disease states are available only for a very limited set of disease classes. This situation is further compounded by the fact that the biological cell is a “dynamic” (far from equilibrium) system whose continuously changing chemical and physical characteristics cannot be characterized by “fixed” mechanical properties [13,69,70]. The cell also has a complex “microstructure” that evolves in response to its chemomechanical environment as well as disease state. A typical eukaryotic cell whose complex microstructure, and hence mechanics, can be significantly altered by cancer, is illustrated schematically in Fig. 2.
An example of systematic studies of how disease states influence single-cell and cell-population biomechanics can be found in recent in vitro studies of the effects of the malaria-inducing Plasmodium falciparum parasite on the deformability of human RBCs [68,71–73]. Here the optical tweezers method (described in Section 4) was used to obtain force vs. deformation of Plasmodium-infected RBCs over a force range of 1–200 pN (strains in excess of 100%). These studies have examined the full range of parasite maturation stages during the 48 h time span of the asexual cycle within the RBC at normal physiological and febrile temperatures. Such measurements have also been complemented with microfluidic assays capable of documenting movements of infected RBCs through constricted channels whose inner cross-sections are comparable to those encountered by RBCs circulating in small capillaries . These biomechanical probes of cell deformability have made use of genetic manipulations of specific parasite proteins for altering cell elasticity [67,68]. For this purpose, targeted gene disruption of proteins involved in RBC deformation in Plasmodium has been utilized in conjunction with RBC deformability studies to quantify the contributions of these proteins to cell elasticity.
This paper focuses specifically on the biomechanics and biophysics of cancer cells, and provides an overview of many critical issues pertaining to the deformability of cancer cells with implications for cell signaling, cytoadherence, migration, invasion and metastatic potential. According to a study by the World Health Organization (WHO) , 7.6 million people worldwide died of cancer during 2005 alone. WHO estimates that mortality rates due to cancer will continue to rise worldwide, with 9 million cancer deaths by 2015 and 11.4 million by 2030. Major advances have been made in the past several decades in our understanding of the cellular, molecular, genetic and environmental factors influencing the onset and progression of human cancer, in the technology for diagnosing cancer and in the treatment regimes. Despite such progress, overall cancer death rates in the USA have not improved in the past five decades. Each year, approximately 1.4 million people in the USA are diagnosed with cancer, and 600,000 succumb to one of more than 200 different types of cancer .
Figs. 3 and and44 show, for males and females in the USA, respectively, the age-adjusted mortality rates due to cancer over seven decades since 1930 . Mortality has declined significantly due to changes in food storage practices and possibly Helicobacter pyroli infection rates for certain types of cancer, such as stomach cancer, and to screening for others, such as cervical and colorectal cancers . However, despite major diagnostic and therapeutic advances and discoveries emanating from cancer research, several types of cancer have remained resistant to treatments particularly when the tumors advance to metastasis, due to a variety of reasons, including lifestyle and environmental factors.
The paper is arranged in the following sequence. A brief summary of key points from the biology of cancer cells is presented in Section 2 to set the stage for subsequent discussion. Section 3 addresses subcellular cytoskeletal networks and their influence on cell biomechanics. This is followed, in Section 4, by a brief summary of various experimental techniques used to extract quantitative information on the mechanical properties of cancer cells. This summary also includes a broad survey of available information on the biomechanics and biophysics of cancer cells. Section 5 deals with specific case studies that illustrate how different types of subcellular cytoskeletal networks and proteins influence cell biomechanics for human breast cancer, pancreatic cancer and melanomas. An example of experiments and analytical means to determine the energy of adhesion of murine sarcoma cells is described in Section 6. Section 7 reviews possible mechanical signaling pathways between cancer cells and the extracellular matrix through integrins and focal adhesions as cell mechanotransduction acts concurrently with molecular tumorigenic signal pathways to promote malignant transformations. The effects of drugs used to treat cancer and their role in influencing cellular architecture and disease states are the focus of discussion in Section 8. The paper concludes with Section 9, where a summary of key observations and a discussion of possible avenues for further research are provided.
Cancer is a disease that arises from malfunctioning biological cells. The diseased cells proliferate uncontrollably and disrupt the organization of tissue.
Biopolymeric proteins constitute the essential components of the cytoskeleton of the cell. This internal scaffolding, as described in detail in the next section, determines the shape and mechanical rigidity of the cell. Proteins secreted in regions between cells constitute the extracellular matrix (ECM), which clusters and binds the cells together to form tissues. Integrins are cell surface receptors which create clusters known as focal adhesions that serve as binding locations between the cell surface and the ECM. The structures of the cytoskeleton and the ECM are transformed by cancer. Altered protein structures also change the ability of cancer cells to contract or stretch, by influencing their mechanics of deformation. As a result, the motility of cancer cells can be very different from that of normal cells, causing them to migrate through tissue to different sites in the human body and inducing tumor metastasis.
Cancer cells create their own signals for sustained growth and duplication and transmit them between proteins through a process commonly referred to as signal transduction. They reprogram their growth and division through a control circuit assembled out of proteins. The signals of this circuit are transmitted back and forth in a two-way communication that exists between the ECM and the cell interior, including the cytoplasm and nucleus, through the integrins and focal adhesions that bind the cell to the ECM.
The cellular and molecular mechanisms determining the origins of cancer can be traced to the pioneering discovery of Varmus and Bishop  of the proto-oncogene, which provided clues about the role of molecules and genes in creating cancer in humans. The proto-oncogene discovery led to the realization that the genomes associated with the cells of normal vertebrates possess a gene with the potential, in some instances, to facilitate the transformation of a normal cell into a cancer cell .
Table 1 lists a brief summary of some key observations on the biology of cancer cells which serve to provide a foundation for subsequent discussion of cancer cell mechanics. More comprehensive discussions of cancer cell biology can be found in the book by Weinberg  and the references cited therein. The various steps involved in this invasion–metastasis cascade are shown schematically in Fig. 5.
The cytoskeleton, the internal scaffolding comprising a complex network of biopolymeric molecules, primarily determines the cell’s shape and mechanical deformation characteristics [13,69,70]. In concert with accessory proteins, it also plays an important role in such cellular processes as mechanotransduction, mitosis and migration. The concentrations and molecular architecture of different constituent components of the cytoskeleton determine the overall deformability and mechanical response of the cell along with the chemomechanical environment, which determines the interactions of the cell with other neighboring cells and the ECM. Any alterations to the cell function due to biochemical processes occurring within the human body, invasion of foreign organisms or disease development can significantly alter the mechanical properties of the cell. Consequently, measures of the mechanical properties of cells could be used as indicators of its biological state and could offer valuable insights into the pathogenic basis of diseases, including the possible identification of one disease from another (see e.g. [13,69,71–73,79–82]).
Eukaryotic cells generally contain three distinct types of polymer biomolecules that serve as structural elements in the cytoskeleton of the cell: actin microfilaments, intermediate filaments and microtubules. Fig. 6 is an immunofluorescence image of a 3T3 mouse fibroblast cell showing nuclear DNA, actin microfilaments and alpha-tubulin. An image of the keratin intermediate filament network inside a human pancreas epithelial cancer cell (Panc-1) is shown in a later section.
Fig. 7 schematically shows the basic structure of actin microfilaments, intermediate filaments and microtubules [13,36,70]. The basic structural features, elastic properties and key characteristics of these three biopolymer filaments are described in detail in Tables 2–4. The information contained in these tables illustrates the critical role of these cytoskeletal components in influencing not only cell mechanics, but also the interactions of the cell with its external biochemical environment. As noted in these tables, defects in the structure of the cytoskeleton influence a number of diseases, including different types of tumors. Conversely, targeting the cytoskeleton to modify its structural, mechanical and biochemical functions provides an avenue for therapeutics for cancer.
Measurements of the individual viscoelastic properties of the three major cytoskeletal fibers, actin microfilaments, vimentin intermediate filaments and tubilin (microtubules) have been performed by Janmey et al.  for various concentrations of the biopolymers. They employed a torsional loading apparatus capable of imposing steady and pulsating stresses on the specimens . The stress vs. strain response of the specimens and the storage modulus G′ values as a function of resonance frequency of oscillations were obtained for different concentrations of the biopolymers. The viscoelastic characteristics of the three isolated cytoskeletal polymers were compared with those of gels comprising fibrin protofibrils, which is the basic structural element of blood clots.
Fig. 8 shows the stress–strain characteristics of actin, vimentin, tubulin and fibrin, all at a fixed protein concentration of 2 mg ml−1. It is clear that the microtubule polymer networks exhibit the largest deformability. The tubulin sample completely loses elasticity beyond a strain of approximately 50%. F-Actin is capable of sustaining much higher stresses and withstands deformation better than both vimentin and microtubules, but begins to rupture and flow at a strain of about 20%. Vimentin and fibrin are more deformable than tubulin and F-actin, but they sustain stresses at much larger strains without flowing freely.
The variation of storage modulus G′ of the four biopolymers as a function of concentration is plotted in Fig. 9 on a log–log scale. Actin and fibrin exhibit larger increases in storage modulus with concentration, with the modulus being proportional to the square of the protein concentration .
Although the in vivo biomechanical response of filamental proteins in the cell cytoplasm is likely to be different from that observed in laboratory experiments, the foregoing measurements offer useful insights into the relative roles of different cytoskeletal components in influencing cell biomechanics in both health and disease. The significance of these results can be summarized as follows .
A wide variety of experimental biophysical probes have been used to extract the mechanical properties of cancer cells. This section briefly summarizes broad categories of various experimental methods. This is followed by a detailed survey of experimental results on the mechanics of cancer cells from the literature.
Fig. 10 schematically shows different experimental methods used for biomechanical and biophysical probes of living cells. In this figure, (a)–(c) show three techniques: atomic force microscopy (AFM) [42,47–50], magnetic twisting cytometry (MTC) [34,87,104–106] and instrumented depth-sensing indentation methods [48,51,107, 108]. In these three methods, a portion of the cell surface could be mechanically probed at forces on the order of 10−2−10−6 N and displacements smaller than 1 nm. In AFM, local deformation is induced on a cell surface through physical contact with the sharp tip at the free end of a cantilever. The applied force is then estimated by calibrating the deflection of the cantilever tip, which is detected by a photodiode. MTC entails the attachment of magnetic beads to functionalized surfaces. A segment of the cell surface is deformed by the twisting moment arising from the application of a magnetic field. Elastic and viscoelastic properties of the cell membrane or subcellular components are then extracted from the results through appropriate analysis of deformation.
Fig. 10d–g shows laser/optical tweezers (OT) [14,15,19,71,109–112], mechanical microplate stretcher (MS) [71,113–115], micro-postarray deformation (mPAD) with patterned microarrays that serve as cell substrates [48,116] and micropipette aspiration (MA) [8,11,13,117, 118], respectively. Here forces over the range of 10−12−10−7 N can be induced on the whole cell while submicrometer displacements are monitored optically. With OT, a laser beam is aimed at a high refractive index dielectric bead attached to the cell. The resulting attractive force between the bead and the laser beam pulls the bead towards the focal point of the laser trap. Two beads specifically attached to diametrically opposite ends of a cell could be trapped by two laser beams, thereby inducing relative displacements between them, and hence uniaxially stretching the cell to forces of up to several hundred piconewtons . Another variation of this method involves a single trap, with the diametrically opposite end of the cell specifically attached to a glass plate which is displaced relative to the trapped bead . In the microplate stretcher, force- or displacement-controlled extensional or shear deformation is induced between two functionalized glass plates to the surfaces of which a cell is specifically attached [71,113–115]. In mPAD, a patterned substrate of microfabricated, flexible cantilevers is created and a cell is specifically tethered to the surfaces of these micro-posts. Deflection of these tiny cantilevers due to focal adhesions can then be used to calibrate the force of adhesion . Other patterns, such as discs and spherical islands, can also be created using micro- and nano-fabrication techniques to design different substrate geometries. In MA, a portion of a cell or the whole cell is aspirated through a micropipette by applying suction. Observations of geometry changes along with appropriate analysis then provide the elastic and viscoelastic responses of the cell, usually by neglecting friction between the cell surface and the inside walls of the micropipette [117,118].
Figs. 10g and h illustrate methods with which the substrate deformation (SD) and cytoadherence characteristics of populations of cells could be inferred. Fig. 10h shows a method from which the biomechanical response of populations of cells could be extracted by monitoring the shear resistance of cells to fluid flow . Shear flow experiments involving laminar or turbulent flows are also commonly performed using a cone-and-plate viscometer consisting of a stationary flat plate and a rotating inverted cone. Alternatively, cells could be subjected to forces from laminar flow in a parallel plate flow chamber [13,67,119]. The mechanics of cell spreading, deformation and migration in response to imposed deformation on compliant polymeric substrates to which the cells are attached through focal adhesion complexes is illustrated schematically in Fig. 10i [120,121]. The patterned substrate method shown in Fig. 10f could also be used to probe populations of cells. More detailed descriptions of each of these techniques along with their merits and drawbacks are summarized in Refs. [13,48].
The type of biophysical assay used to probe a particular cell type depends on a variety of factors, including its effective stiffness and time-dependent deformation characteristics, the extent of stretching or contraction it undergoes during mechanical testing, and the force ranges typically achieved during such deformation. Figs. 11a and b summarize the force ranges and length (displacement) scales of relevance, respectively, in select cell and molecular processes of interest in biological systems. Also indicated in these figures are the force and displacement ranges achieved by various biomechanical assays. Fig. 11c illustrates examples of energy scales involved in select biochemical processes of interest in cell and molecular mechanics.
In addition to the above methods, a wide variety of simple methods are commonly used to assess mechanical deformability of biological cells without any quantitative measures of forces on the cells or the resulting deformation response. For example, the time of passage of a large population of cells through a micropore filter of a particular pore size is a common method used to simulate cell motility through microvasculature or size-limiting pores in the human body. Such “micropore filtration methods”  have also been used to investigate tumor cell deformability as a function of cell size, origin, temperature and pH.
Microfluidic and nanofluidic assays [72,123–125], with relatively rigid or compliant channels fabricated from silicon or poly(dimethylsiloxane) (PDMS), respectively, are also employed to simulate the flow of cells through blood vessels. Such assays could be designed to impose a controlled pressure gradient across the channel, and the resulting flow characteristics of a large population of cells could be recorded in terms of entry time into the channel, cell velocity through the channel, relaxation time for the cell to fully recover its original shape upon egress from a constricted channel, etc. In addition, the inner surfaces of the channels could be lined with an endothelial layer or coated with appropriate receptors to simulate more realistic in vivo biochemical environments. The microfluidic assays could also be used as high throughput cell sorters and employed in conjunction with quantitative cell deformability assays, such as optical tweezers to study the elastic and viscoelastic characteristics of cells. An example of the latter is described in Section 5.2 in the context of deformability of mammary epithelial tumor cells.
Table 5 provides a survey of various observations of the mechanical response of different types of cancer cells, the tools employed to extract the biophysical characteristics and available information on the possible effects of cytoskeleton in modulating cell mechanics.
This section examines specific examples of cancer cell mechanics with implications for migration, invasion and metastasis.
The first example deals with the deformability of non-malignant and malignant human breast cancer cells and of the latter malignant cells chemically modified to increase metastatic efficiency. Guck et al.  used a microfluidic optical stretcher, which is a coaxially aligned dual-beam laser tweezers system, to sequentially suspend, trap and deform isolated cells through a contactless microfluidic channel. The net forces from the two beams balance along the beam axis, thereby producing a stable trap. The cells are individually stretched, at forces on the order of 200–500 pN, in fluid suspension at a flow rate of approximately 1 cell min−1.
In this method, the dimensions of major and minor axes, a and b, respectively, of the cell are optically recorded prior to and during stretching in the trap. The difference between the aspect ratio at/bt (i.e. the ratio of the major to the minor axis at time t) at the stretching power and the corresponding aspect ratio at the reference trapping power, ao/bo, are determined to define an approximate measure of normalized deformability, Dt, as
In order to compensate for variations in cell size for different cells in the same population and for different cell conditions or disease states, Eq. (1) is further modified to define a quantitative optical deformability index (ODt):
where Fa and Fb are the integrated stretching forces along the major and minor axes, respectively. The subscripts “ref” and “cell” denote the reference and deformed configurations, respectively. ODt thus provides an approximate measure of the relative cell stiffness or spring constant at any time during cell stretching.
Fig. 12 shows the deformability of nonmetastatic human breast epithelial cells and their malignant counterparts. MCF 10 is a standard model nontumorigenic epithelial cell line derived from benign breast tissue; the cells are immortal, but otherwise normal and noncancerous. MCF 7 is a corresponding human adenocarcinoma cell line in which the cells are nonmotile, nonmetastatic and cancerous. When MCF 7 cells are modified (Mod MCF 7) by treatment with 100 nM 12-O-tetradecanoylphorbol-13-acetate (TPA) for 18 h, an 18-fold increase in the invasiveness and metastatic efficiency is achieved . It is evident from the figure that the cancerous MCF 7 cells are more deformable than the normal MCF 10, and that the meta-static Mod MCF 7 cells are even more deformable than MCF 7 cells. The increased deformability of cancer cells with modified metastatic competence appears to be accompanied by a further reduction in structural strength compared with that of the nonmetastatic cancer cells. These reductions in elastic rigidity appear to arise from a reduction in F-actin concentration of as much as 30% caused during the malignant transformation of the cell  at the small strain levels imposed for the results shown in Fig. 12. At larger strain levels, additional contributions to changes in elasticity could also arise from reorganization in the molecular architecture of keratins, which is a major cytoskeletal component of epithelial cells.
The main assessment basis for pathology for breast cancer is the change in morphology of the suspect tissue. However, access to coherent tissue samples inevitably requires biopsy, and is often performed only when a sufficiently noticeable collection of abnormal cells is available. The microfluidic optical stretcher has the advantage of enabling observation and mechanical probing of isolated cells obtained from exfoliative cytology . This method can also be used for microfluidic sorting of the cells on the basis of deformability, thereby obviating the need for genomic and proteomic manipulations of the cells. A drawback of the method is that the forces imposed on the cell by the optical tweezers are not sufficiently large to promote significant deformability, which is necessary to simulate in vivo conditions encountered by migrating tumor cells as they navigate through size-limiting pores. Furthermore, the direct exposure of the cells to the laser beam during stretching is another limitation of the technique.
Secretion of pancreatic juice into the duodenum in the pancreatic duct and secretion of insulin and glucagons for the regulation of blood glucose levels are two major functions of the human pancreas. Tumor of the pancreas is one of the most lethal forms of cancer in the developed world because of the difficulty in detecting it early and its aggressive propensity for invasion, migration and metastasis. Clinical findings have shown that a bioactive lipid, sphingosylphosphorylcholine (SPC), plays a critical role in the metastatic invasions of gastrointenstinal cancers, including gastric and pancreatic tumors . SPC occurs naturally in human blood components such as blood plasma and high-density lipoprotein particles to promote anti-apoptotic effects.
SPC-induced molecular reorganization of the Panc-1 human epithelial pancreatic cancer cells and the consequent changes in mechanical deformability of the cells have been examined as possible pathways that facilitate easier migration and increased metastatic competence of pancreatic tumor cells [71,115]. Human pancreatic epithelial cancer cells express keratins 7, 8, 18 and 19, and the subline of Panc-1 epithelial tumor cells primarily expresses K8 and K18. When human Panc-1 cancer cells are treated with 10 µM SPC to mimic in vivo conditions, phosphorylation leads to a drastic reorganization of keratin into a dense, ring-like molecular structure in the perinuclear region of the cell within 45 min. This is shown in the immunofluorescence images at the lower left of Fig. 13.
Biomechanical assays of Panc-1 cell deformability with and without SPC treatment have been performed using the microplate mechanical stretcher method [71,113–115], which is shown schematically in Fig. 10e. For this purpose, a Panc-1 cancer cell is placed between two glass plates functionalized with fibronectin to adhere the cell to the surfaces of the microplates. The cell is then mechanically deformed at physiological temperature by repeated tensile loading over force ranges on the order of several hundred nanonewtons in a microplate cell stretcher to determine its deformability over cell displacements on the order of several micrometers. The resultant spring constant for the cell is calculated as an indicator of its effective elastic rigidity.
Fig. 13 also shows relative changes in the Panc-1 cell effective spring constant as a function of time, first without any SPC treatment and then following SPC treatment. When the reorganization of keratin by SPC is complete, the elastic stiffness of the Panc-1 cell is found to decrease by as much as a factor of 3 compared with its pre-SPC-treatment value. On the basis of these results, it has been postulated  that such changes to cell mechanics arising from biochemically induced keratin reorganization could play a critical role in enabling the Panc-1 cancer cell to migrate more easily through size-limiting pores and in facilitating greater metastatic efficiency.
The relatively dominant role of keratin intermediate filaments, over other cytoskeletal filamental biopolymers, in influencing cell deformability was also demonstrated by investigating modifications to cytoskeletal architecture arising from other chemical agents. For example, lysophosphatidic acid (LPA) is known to influence many biological processes, such as platelet activation, mitogenesis, smooth muscle contraction and alteration of the shapes of neurons. These functions are facilitated by the formation of actin stress fibers when the cells are exposed to LPA. If the Panc-1 cancer cell is treated with 10 µM LPA for 15 min, instead of with SPC, microplate mechanical stretch assays do not reveal any significant change in stiffness due to cytoskeletal reorganization [71,115].
More quantitative connections between changes in cell deformability and motility can be obtained with microfluidic channels. The microfabricated devices offer a means to quantify cell motility characteristics in vitro by providing velocity measurements and shape change and recovery observations as a population of cells flow through channels with well-defined geometrical constrictions and controlled pressure gradients. Fig. 14 shows a microfluidic probe used to study Panc-1 cancer cell migration.
Distant metastatic tumor formation is considered to be influenced strongly by the stable adhesion of cancer cells to the small blood vessel walls. Fluid shear forces and the blood environment surrounding the cells during microcirculation, along with the mechanical response of the cell cytoskeleton, can modulate metastatic potential of circulatory cancer cells in host organs. Korb et al.  investigated the role of actin and microtubules in early metastasis in vivo through intravital observation of cytoadherence of colon carcinoma cells within liver microcirculation of rats1 and their invasion into liver parenchyma.
Disruption of tubulin reduced adhesive interactions in vivo, whereas disruption of actin microfilaments increased cell adhesion. During in vitro flow conditions, cytodisruption modulated cell signaling via focal adhesions. These results suggest that changes in cell stiffness and avidity of the cell adhesion molecules (arising from cytoskeletal disruptions) are important for initial adhesive interactions in vivo. However, the broad applicability of these results critically depends on the compatibility of human cells with the immune system of the rat.
As noted in Table 2, the actin microfilament network stabilizes or gels the cell periphery and the plasma membrane. Actin networks exhibit high elastic moduli and also undergo fluidization at large strains. Processes critical to cell locomotion, such as pseudopod extension, require that the cytoplasm change its viscosity at the cell periphery by transforming from a gelled state to a fluid state.
Solation (destruction) and gelation (reconstruction) of actin microfilaments facilitate the formation of protrusions on the surface of motile cells during their locomotion. Experiments by Cunningham et al.  reveal that when dispersed actin filaments are cross-linked and connected to plasma membrane glycoproteins by an actin-binding protein (ABP), which in this case is a homodimer with rod-like 280 kDa subunits with flexible hinges, the cell surface is stabilized. Under these conditions, protrusion activity is organized and locomotion is efficiently achieved. They also show that when cells are ABP-deficient, as in certain human malignant melanoma tumor cell lines, locomotion is impaired and the plasma membrane undergoes circumferential blebbing2 (due to peripheral cytoskeletal instability). If ABP is transfected into such cells, normal locomotion is restored and blebbing is suppressed. ABP transfection also increases the elastic stiffness of the pellets of intact cells by more than a factor of two, from 27 ± 1.1 to 67 ± 4.2 Pa. Since melanoma can be a locally invasive cancer, cell invasion and metastasis and the consequent clinical course of malignancy could be influenced by the migratory efficiency of the cancer cells.
The physics of adhesion between biological cells significantly influences cancer cell motility, invasion and metastasis. However, quantification of cytoadherence is extremely complex, and there are no generally accepted constitutive models or theoretical analysis with which experimental observations of cell adhesion could be systematically rationalized and quantitatively interpreted.
When solid elastic bodies adhere, the underlying mechanics and physics of adhesion could often be analyzed mathematically [146,147]. In general, when two slightly deformable, elastic bodies adhere strongly, the application of a force to separate them causes a finite contact area to develop at the interface prior to the detachment of the surfaces. This behavior is often analyzed by recourse to the Johnson–Kendall–Roberts (JKR) theory , which relates the adhesion energy Wa to the force of detachment Fs through the radii of curvature of the contacting materials. If two spherical solids are in adhesive contact, the JKR theory predicts that
where Rm is the harmonic mean of the radii of the adhering spheres. Here, an example of the application JKR theory is presented for the quantification of adhesion energy between murine sarcoma cells.
Chu et al.  have quantified the energy of adhesion between two murine sarcoma S180 cells through micropipette aspiration experiments by exploiting strong adhesion created by a depletion effect in a highly concentrated dextran solution. These cells do not naturally adhere because of the paucity of superficial adhesion receptors, but strongly attach to each other when forced to touch in the presence of dextran through a pair of micropipettes.
Fig. 15 shows a series of micrographs that illustrate the process of adhesion and separation between the cells. The final separation force between the cells is given by
where ΔP is the average of the aspiration pressures prior to cell separation and R is the inner radius of the micropipette. Since the cells exhibit moderate deformability and finite contact zones during adhesion, it is appropriate to examine the relevance of JKR theory to quantify the energy of adhesion between the cells.
Fig. 16 is a plot of the adhesion energy Wa estimated from Eq. (3) along with experimental measurements for different dextran concentrations. These results can also be compared with experimental measurements of adhesion energies by the depletion of dextran on vesicles . Theoretical analysis of this process  shows that the adhesion energy Wa is related to dextran concentration V by the relation
where kBT is the thermal energy (with kB being the Boltzmann constant and T, the absolute temperature) and m is the size of the monomer. Alternatively, the relation between separation force and adhesion energy can also be extracted from the analysis of spherical shells  to be
Cells within different tissues are held together by the ECM. The onset and advance of cancer leads to distinct changes in the stiffness of the surrounding tissue. Consequently, oncologists usually diagnose cancer by sensing the change in elasticity of tissue by palpation . It has long been known that the mechanical response of the ECM itself can play a critical role in tumor formation; a stiffer ECM can result in a stiffer solid tumor [152,153].
There are also clear connections among single-cell mechanics, ECM mechanics, cell mechanotransduction, oncogenic signaling and tumor formation. Select observations illustrating such connections are given below.
The result is the creation of a closed mechanical signaling loop that acts in concert with the tumorigenic signaling pathways to promote malignant transformation .
Chemotherapeutic agents for cancer are purposely designed to target cell membranes and cytoskeleton so as to induce cytotoxicity and alter cytoadherence. Here, select examples of the effects of chemotherapy on subcellular structures are presented. The effects of chemotherapy on cell mechanics and the attendant consequences for changes in disease states are then described with the specific example of vascular complications arising from chemotherapy.
As noted in Tables 2–4, anti-tumor drugs are often purposely designed to target cancer cell membranes and cytoskeleton as they effect cytotoxic action. The following cancer drugs are used as therapeutics for ovarian, breast and lung cancer, malignant melanoma and leukemias [101,158–162].
Alterations to cell mechanics induced by chemotherapeutic agents can also result in other clinical complications. The following example describes one such disease which is instigated by the initiation of chemotherapy itself.
The accumulation of leukemia cells in the blood vessels of vital organs such as the lungs and the brain is known to result in respiratory failure and intracranial hemorrhage . This life-threatening complication of acute leukemia, known as leukostasis, is caused by unknown pathophysiological mechanisms and its diagnosis is usually confirmed only by autopsy. Chemotherapy, which is often essential to treat leukemia, can by itself be a likely risk factor. In some patients, leukostasis develops soon after the initiation of chemotherapy. Advances in chemotherapy treatment regimens for leukemia patients have resulted in a gradual reduction in mortality; however, the mortality rate for the subset of leukemia patients who develop vascular complications has not improved beyond the typical range of 69–74% . Possible causes of vascular complications have been ascribed to mechanical obstruction of blood vessels in vital organs due to an increase in the number of circulating leukemia cells, increased cell stiffness and increased cytoadherence.
Employing the AFM as a quantitative biophysical probe, Lam et al.  studied the effects of chemotherapy on changes in elastic properties of acute lymphoblastic leukemia (ALL) and acute myeloid leukemia (AML) cells. Samples for the former were obtained from pre-B cell or T-cell precursor cells taken from the peripheral blood of six different, newly diagnosed teenage patients with acute leukemia and with detectable peripheral blasts. Samples for the latter were taken from the peripheral blood or bone marrow of 8- to 78-year-old patients. The AFM chamber was fitted with a PDMS perfusion chamber and a heating stage to facilitate measurements at normal physiological temperature. The ALL and AML cells were exposed to 1 µM dexamethasone and 1 µM daunorubicin, respectively, which are common induction chemotherapeutic agents for the two types of leukemia. Since leukocytes (white blood cells) are nonadherent cells, they were mechanically immobilized in microfabricated wells so as to prevent them from slipping under the AFM cantilever during contact probing. The experiments revealed the following general trends.
Such correlations between cell mechanics and cell death provide a possible rationale for the clinical observation that some patients develop leukostasis soon after chemotherapy is initiated. The results shown in Figs. 18 and and1919 suggest that reduced cell deformability caused by actin network reorganization following chemotherapy initiation might contribute significantly to vascular complications by obstructing microcirculation through the stiffening of leukemia cells during cell death. These results could have some implications for pathophysiology of other diseases where increased stiffness of circulating cells directly affects cell interactions with the endothelium: diabetes, atherosclerosis or autoimmune inflammatory disorders.
Major advances in cell biology and biophysics have provided extraordinary opportunities to examine the links between the mechanics of cells/subcellular structures and cell functions such as mitosis, signaling, mechanotransduction, ribosomal and vesicle transport, and cell locomotion and motility. A significant outcome of such progress is the availability of techniques to measure the force vs. displacement signature of living cells through a variety of sophisticated biomechanical assays. This mechanical signature, in concert with other chemical, biological and genetic pathways, offers unique new perspectives through which the highly complex mechanistic underpinnings of human health and disease at the molecular and cellular levels could be better understood.
This paper has examined critical issues, results and examples of how changes in cell mechanical responses arising from the biochemical alterations of a stiff and ordered cytoskeleton to an irregular biopolymer network is related to the transformation of a mature, postmitotic, normal cell into an immortal, replicating and motile tumor cell. Our review also points to compelling evidence that clear signaling pathways exist among cell deformability, contractility of the extracellular matrix, cell mechanotransduction, release of integrins and focal adhesions, communication between the cell nucleus and the surface for the activation of transcription factors that influence cell proliferation, and tumorigenic processes. An important step in cancer metastasis involves the penetration of epithelial cells through the endothelial layer, which critically depends on cell elasticity and deformability.
With the rapid growth of the nascent field of cell and molecular mechanics of cancer, a number of challenges remain in translating scientific discoveries from in vitro assays to understanding in vivo processes in the human body. Immobilized cells on artificial substrates and enclosed in artificial environments often provide different mechanical responses in the absence of appropriate biochemical signals. Furthermore, different biophysical assays probe different components of the subcellular structures under different stress states to different levels of precision. It is therefore not surprising that conflicting information on the cell mechanical responses are extracted for the same cellular disease processes from different biomechanical probes. This is especially pronounced when large variations and experimental scatter occur in response characteristics and underscores the need for a large sample set to extract statistically significant results. There is also a critical need to develop more detailed computational simulations of cell and molecular mechanics that accurately capture interatomic and intermolecular interactions and cytoskeletal dynamics as the cytoskeletal network is altered by the disease state. Establishing comprehensive connections between cell mechanics on the one hand and chemical and biological cell functions in human health and disease on the other promises to provide novel and powerful developments for cancer diagnostics, prophylactics and therapeutics.
Preparation of this article and related work in the area of cell mechanics received support from the National Institutes of Health/National Institute of General Medical Sciences Grant 1-R01-GM076689-01 and the Advanced Materials for Micro and NanoSystems Program and the Computational Systems Biology Program of the Singapore-MIT Alliance, and support from the National University of Singapore through the Tan Chin Tuan Centennial Overseas Chair, the Global Enterprise for Micromechanics and Molecular Medicine (GEM4), and an Inter-University Research Project. The author thanks members of his research group and collaborators for many insightful discussions on the topics covered in this paper. Preparations of this article also benefited from partial support from the Nano/Bio Mechanics Program of the Engineering Directorate of the US National Science Foundation for the GEM4 2007 Summer School on Cell and Molecular Mechanics in Biomedicine.
Subra Suresh is the Ford Professor of Engineering at the Massachusetts Institute of Technology (MIT), where he holds joint faculty appointments in the Departments of Materials Science and Engineering, Mechanical Engineering, and Biological Engineering. He is also an affiliated faculty member in the Harvard-MIT Division of Health Sciences and Technology, and the Tan Chin Tuan Centennial Professor (Overseas) at the National University of Singapore. He received a Bachelor of Technology degree in first class with distinction from Indian Institute of Technology in Madras, a Master of Science degree from Iowa State University and a Doctor of Science degree from MIT. He holds an honorary doctorate degree in engineering from Sweden’s Royal Institute of Technology. He was the Head of the Department of Materials Science and Engineering at MIT from 2000 to 2006. Prior to joining MIT, he was on the faculty of the Division of Engineering at Brown University from 1983 to 1993. Professor Suresh’s research has focused on the mechanical properties of materials from the atomistic to macroscopic length scales, with the particular aim of establishing connections among the material nano/microstructure, mechanical properties at multiple length scales, and function. His work has pioneered a wide variety of innovative experimental, analytical and computational methods to study different classes of materials in both bulk and thin-film form. His recent work has provided unique new perspectives on how the mechanics of biological cells and molecules in concert with chemical and biological factors influence, and are influenced by, a variety of human diseases. Professor Suresh has been elected to the United States National Academy of Engineering, American Academy of Arts and Sciences, Indian National Academy of Engineering, Indian Academy of Sciences, Academy of Sciences of the Developing World (Trieste, Italy), Royal Spanish Academy of Sciences, and the German Academy of Sciences Leopoldina.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
The first Acta Materialia Gold Medal Lecture was presented by Professor Subra Suresh at the 2006 Fall Meeting of the Materials Research Society in Boston, MA, USA, on November 27, 2006.
1CC531 rat colon carcinoma cells and HT-29 human colon carcinoma cells with comparable tumor characteristics are found to exhibit similar cytoadherence and invasive behavior in liver microcirculation . It is, however, possible that human and rat adhesion molecules are not 100% compatible.
2Cells exposed to stresses usually exhibit morphological changes such as geometrical contraction, flowering, etc. Such changes are often referred to as cell blebbing.
3Mutations that cause EGFR overexpression (so-called upregulation) have been linked to different types of cancer, including lung cancer and glioblastoma multiforme. The identification of EGFR as an oncogene has spurred the development of anticancer drugs that target EGFR; such therapeutics include gefitinib (originally coded 2D1839) and erlotinib for lung cancer, cetuximab for colon cancer and trastuzumab for breast cancer. These drugs use monoclonal antibodies against EGFR or involve protein kinase inhibitors.
4The protein Rho belongs to the Ras superfamily of G-proteins, which relay intracellular signals and regulate cell processes. Rho also controls actin’s association with cell membrane. Many human cancers can be triggered by oncogenic mutations in Ras genes.