Primary adult human bone cells do not respond to tissue (continuum) level strains

J Orthop Sci (2001) 6:295–301

Instructional lecture Primary adult human bone cells do not respond to tissue (continuum) level strains Richard A. Brand1, Clark M. Stanford2, and Daniel P. Nicolella3 1 Department of Orthopaedic Surgery, The University of Iowa Hospitals and Clinics, 01029 John Pappajohn Pavilion, 200 Hawkins Drive, Iowa City, IA 52242, USA 2 Dows Institute for Dental Research, The University of Iowa, Iowa City, IA 52242, USA 3 Southwest Research Institute, 6220 Culebra Road, P.O. Drawer 28510, San Antonio, TX 78228-0510, USA

Abstract Bone adapts to its mechanical environment, and, since the late 1800s, investigators have presumed that this adaptation relates to strain magnitude. Indeed, overwhelming evidence supports the view that either strain or some strainrelated quantity stimulates bone adaptation or remodeling. Virtually all investigators, implicitly or explicitly, assume that the level of strain magnitude responsible for bone adaptation is that measured by strain gauges in vivo (i.e., 100–2500 microstrain) and that bone cells are directly deformed by strained matrix. We present evidence that bone cell deformation in this range does not cause bone adaptation. First, bone cells in vitro typically do not respond to average (continuum) levels of strain magnitude. Second, bone cells in vitro do respond to fluid flow-induced shear stresses in these ostensible physiological ranges. Third, in vivo strain magnitudes presumed to stimulate remodeling reflect only averages, and not local peaks, which are 2–15 times higher. Thus, we hypothesize that sensing cells do not respond to levels of strain presumed to be physiological. Key words Bone · Cells · Strain · Continuum · Remodeling

Introduction The regularity of cortical and trabecular architectures and their reproducible responses to altered loading (e.g., exercise) or geometry (e.g., malunited fractures) strongly argue for a formal (mathematical) relationship between the mechanical environment and bone adaptation. (Bone adaptation in this communication will relate to the changes in normal adult bone as a

Offprint requests to: R.A. Brand Received: January 9, 2001 Presented at the 15th Annual Orthopaedic Research Meeting of the Japanese Orthopaedic Association, Kyoto, Japan, September 29, 2000.

result of changes in the mechanical environment. We will specifically exclude those changes in adult injury and repair; normal skeletal development, in which genetic mechanisms play such a major role that the relative role of mechanics is more difficult to discern, will also be excluded.) Wolff (1892)58 postulated the existence of mathematical laws governing bone “remodeling,” and, while he did not specifically formulate them, he did suggest that they related to principal compressive and tensile stresses (i.e., magnitude of load). In attempting to identify particular aspects of the result of skeletal loadings (i.e., stress/strain parameters) controlling tissue adaptation, subsequent investigators have assumed that tissues “seek” (i.e., remodel to achieve) some measurable “optimum” state of stress or strain (e.g., “attractor state”), and further, implicitly or explicitly, have postulated that the attractor state arises from peak loads,9,27 “averaged” strains,19 or “values which cause fatigue microdamage”.13,21 Cowin (1987)13 further noted other quantities which might account for time: “time averaged values . . . , . . . amplitudes of the oscillatory components (of load) . . . , and (values) dependent on the time rates of change . . .”. Realizing that the relationship between tissue adaptation and loading must involve some “cumulative effects of loading,” Whalen and Carter56 and others (e.g., Huiskes et al.,27 and Levenston et al.35) formulated remodeling rules based upon summing a number of discrete peak loads during some given number of occurrences of similar activities, assuming that the entire loading history influenced bone maintenance and/or remodeling. However, Cowin13 noted, “The precise aspect of the strain history sensed by bone tissue is an open question.” We will use “strain” as a general term for the mechanical stimulus for bone adaptation, but not in a strict mathematical way. Rather, we refer to some (potentially measurable) deformation, because “strain”

296

R.A. Brand et al.: Bone cells do not respond to tissue level strains

becomes problematic to measure at the cellular and subcellular levels (the definition of strain remains the same regardless of scale). We use “strain” rather than “stress”, because it is generally presumed that matrix and cell deformation relates to remodeling, rather than to stress; we easily envision deformation “receptors” or mechanisms, but not stress mechanisms. We also make no implications regarding mechanism, realizing that whole bone deformation results not only in local deformation at the cell level but also in fluid flow, streaming potentials, and other physical phenomena, which might result in cell response. The recognition that strain history, rather than merely strain magnitude at a given time, initiates remodeling arises from experimental evidence suggesting that tissues “consider” temporal aspects of the stimulus; for example, bones respond very differently to static loads over time than to equivalent (i.e., in magnitude) dynamic loads.11,34 Turner52 postulated three “rules” of bone remodeling: “(1) It is driven by dynamic, rather than static, loading. (2) Only a short duration of mechanical loading is necessary to initiate an adaptive response. (3) Bone cells accommodate to a customary mechanical loading environment, making them less responsive to routine loading signals.” Further, Gross et al.22 demonstrated that the strain distribution in the bone at the instant of peak strains differs from the distribution of strain at other times; therefore, if sub-maximal strains initiate bone remodeling, the distribution of added bone will differ substantially from that predicted from the location of peak strains. Perhaps more importantly, some regions of bone habitually experience low strains while others experience high strains, yet maintain spatial concordance.2 We could reasonably presume, then, that stress/strain distribution (and therefore remodeling stimulus) would intimately depend upon the temporal characteristics of the stimulus, in addition to magnitude.3,13 However, there is little question that magnitude is one important feature of the mechanical environment. The question is, what strain magnitudes stimulate bone adaptation? Most investigators have presumed that these levels are those experimentally measured via strain gauges on the external surface of bones, i.e., 50–4000 microstrain for most activities for various animals,50 and up to 2500 microstrain in humans.8,20 Furthermore, most investigators presume that the stimulus arises from direct deformation of bone cells through their connections to the matrix. We propose three arguments to explain why deformation of bone cells, as a result of their connections to locally deformed tissue under “physiological” (continuum) loading, may not stimulate bone adaptation: (1) Bone cells in vitro typically do not respond to

average levels of strain magnitude, but do respond to supra-physiological strain magnitudes. (2) Bone cells in vitro do respond to fluid flow-induced shear stresses in presumed physiological ranges. (3) In vivo strain magnitudes presumed to stimulate remodeling reflect only averages, and not local peaks, which are 2–15 times higher.

Bone cells in vitro typically do not respond to average levels of strain magnitude Many studies document proliferative and/or expressive responses of dynamically strained osteoblast-like cells in culture (e.g., references 5, 23, 25, 38, 40) and bone organ cultures (e.g., references 15, 45). It is important to emphasize four points about these studies: first, the strain magnitudes in these studies were either not well characterized and/or were done at strain magnitudes at or above the gross strain failure of bone. Second, the majority of these studies demonstrate mechanical stimulation of intracellular processes, but not the production of matrix proteins, so the ultimate implications of the intracellular processes for added bone mass are uncertain. Third, most authors ignore lack of a response to a given set of conditions, and, rather, focus on those responses that are different from those in controls; however, in reviewing these papers, we found that a large portion of responses were no different from those in unstrained controls. Fourth, for matters of convenience, most studies use transformed cell lines or cells from very young animals (e.g., newborn rats or chicks). These types of cells appear to be “programmed” to replicate and respond, because they are from rapidly growing tissues and may not reflect cell behavior in normal adult cells. Furthermore, few critical stimulus characteristics (e.g., number of cycles, strain level, duration of cycles, timing of application) were explored, and the studies are quite varied in their mechanical and biological conditions. Nonetheless, these studies do document the responsiveness of osteoblast-like cells to strain under certain in vitro conditions. Brighton et al.4 reported rat calvarial cells strained at physiological levels (i.e., 400 microstrain, a level experienced in vivo by bone as a tissue), with a modest (10%–35%) increase in DNA production, while showing decreased expression of collagen, noncollagenous protein, proteoglycan, and alkaline phosphatase. Jones et al.,31 on the other hand, demonstrated that periosteal-derived osteoblasts, but not Haversianderived osteoblasts, responded (proliferation) to 3000 microstrain, but not to 300 and 10 000 microstrain (i.e., consistent with a window); in periosteal cells, 1 cycle at 3000 microstrain significantly stimulated collagen

R.A. Brand et al.: Bone cells do not respond to tissue level strains

synthesis as much as 30 cycles (i.e., consistent with a trigger to cycle number). It should be emphasized that these and many other experiments in the literature used either immature animal-derived bone-like cells, or transformed cell lines. Furthermore, when responses have been elicited, they usually reflect the stimulation or suppression of intracellular responses, rather than the expression of extracellular matrix proteins. Whether these intracellular responses will translate into the production of structural proteins (required for tissue adaptation) is unknown. We have been unable to demonstrate matrix protein responses of mature human osteoblast-like cells in culture to substrate deformations up to 3000 microstrain.51 Similarly, Jones et al.31 were unable to show responses of bovine cortical bone at 300–3000 microstrain, although they demonstrated increases in proliferation of periosteal cells within those same ranges. Neidlinger-Wilke et al.39,40 reported proliferative responses of mature human osteoblast-like cells to strains of 10 000 to 88 000 microstrain. Recently that same group reported proliferative responses of human osteoblast-like cells to 1000 microstrain, but they did not study the production of structural matrix proteins.32 While our results and those of others may appear contradictory, we must emphasize that cell populations and culture conditions differed; one can never know whether differences occur merely owing to the choice of cells and culture conditions. The findings, nonetheless, suggest that mature human osteoblast-like cells produce structural proteins only at presumably supraphysiological levels of strain magnitude, while not doing so at those levels previously considered physiological (i.e., at continuum tissue levels: the level measured with strain gauges).

Bone cells in vitro do respond to fluid flow-induced shear stresses in presumed physiological ranges Bone is a porous medium (marrow spaces, Haversian canals, lacunae, and canaliculi), and, given such porosities, Zeng et al.59 estimated that the physiological loading of limbs resulted in fluid shear stress in the range of 6–30 dynes/cm2 (0.6–3.0 Pa). A number of studies document the responsiveness of osteoblast-like cells to constant or intermittent hydrostatic loads6,46 or fluid-generated shear stress in the ranges of 5– 100 dynes/cm2.12,28,33,43 (While the dynamic loading of bones leads to instantaneously high pressures in bone,30 we are uncertain of the physiological implications of experiments with sustained hydrostatic pressures.) Pulsing flow (as would occur in vivo with physiological loading) results in greater responses than steady or oscillating flow.29 We note, however, that studies

297

looking at osteoblast-like cell response to shear stress typically considered intracellular responses, and not the production of matrix proteins.

Magnitude of normal strain in vitro varies considerably about the tissue continuum average The magnitudes of strains recorded by strain gauges are, in essence, averaged over the region of the gauge, generally at least on the length scale of a millimeter. These are also the typical “continuum” level strains which would be predicted by finite element analysis (FEA) of a whole bone. However, Harrigan et al.24 have suggested that, “Within three to five trabeculae of an (implant) interface a continuum model is suspect.” Given trabecular widths of 50–150 micrometers, and assuming similar intertrabecular spaces, this means that assuming a solid region below 500 µm would result in questionable strain predictions. This is because a continuum model assumes a solid, homogeneous material for each element (for contemporary models of whole bone typically in the order of many millimeters size). Such models do not account for the structural and material irregularities below the level modeled (i.e., trabecular struts and plates, osteonal structure, Haversian canals, canaliculi, etc.). We hasten to add that this is not an inherent limitation of the FEA, but, rather, a limitation of the way in which it is applied: individual trabeculae can and have been modeled,55 but even these models do not account for microstructural irregularities within the trabeculae. Attempting to build such a model from the trabecular level to whole bone would be computationally intractable at this point. As would be the case in any engineering material, the irregularities in cortical and trabecular bone undoubtedly create “stress risers”: local defects around which local stresses and strains are considerably higher than those some distance from the defects. For example, the stress at the lateral aspect of a hole at the center of a plate of homogeneous material under uniform applied uniaxial stress will be three times the average stress at some distance away from the hole.36 Furthermore, strains near the tips of cracks in high-performance aerospace materials have been shown to be many times the strain in the material away from the crack,16,57 a finding consistent with engineering fracture mechanics predictions of a singularity at the crack tip. In bone, material irregularities, such as osteons, osteocyte lacuna, and canaliculi, as well as cracks from accumulated microdamage, create stress risers. Estimates of the increases in stress about such irregularities range from 2–3 times,55 to 10 times,41,42 or even 300 times26 the continuum or average level. (It is obvious that bone does not, under physiological

298

R.A. Brand et al.: Bone cells do not respond to tissue level strains

circumstances, continuously experience macroscopic strains beyond its ultimate (fatigue) strain. Local strains can, and do, reach levels several times greater than the macroscopic fracture strain. Prior to the coalescence of damage into a fatal crack, the undamaged material locally surrounding a crack or defect acts to constrain the material and prevent failure.17 The macroscopic failure strain is an average strain over the area of the strain measurement device. In general, especially in a heterogeneous material such as bone, there are local material regions that are much less deformed than others, and vice versa. These local strains might in pathological situations.) Recently, in fatigue loaded cortical bone specimens, microcracks were observed to initiate frequently at osteocyte lacunae, lending evidence to the hypothesis that lacunae act as stress concentrators.47 Thus, it is clear that strain in the matrix around some osteocytes is considerably above the levels currently considered by most authors, and likely is larger about microcrack tips than about lacunae.

Discussion Based upon three observations, the lack of responsiveness of bone cells to strain levels up to the high physiological range, the responsiveness of bone-like cells to fluid flow-induced shear stress in the physiological range, and the presence of relatively high strain regions around lacunae, we propose that bone cells do not respond to direct deformation of bone tissue as result of their connections to the adjacent matrix. This proposal does not, however, exclude the ability to respond to fluid-flow generated shear stress, which also deforms the cells, or the ability to respond to levels of matrix strain well above that at a tissue continuum level. Several issues immediately require discussion. First, strain magnitude alone does not determine cell or tissue response.3 Numbers of cycles, frequency, strain rate, and duty cycle all impact on the response.49,54 In fact, low magnitude (i.e., less than 100 microstrain), high frequency (i.e., 30 Hz) signals maintain bone mass in a model normally resulting in disuse osteopenia.44 Most investigators have implicitly assumed that strain magnitude applies to those signals with the dominant frequencies of activities of daily living, in the range of 0.5 to 2.0 Hz, and with reasonable daily cycle numbers. We explicitly intend the same with the proposed hypothesis. However, we do not intend to imply that activities in the dominant frequency domain necessarily dominate bone adaptation. Second, the presence of high strain in the region of a lacuna or microcrack does not necessarily imply that the cell is experiencing that same level of strain. Cells are

not rigidly attached to surrounding matrix, and it is likely the cells do not experience the same strain as the matrix in vivo. In vitro, Neidlinger-Wilke et al.40 demonstrated that cells experienced 85%–90% of their substrate strains. This may or not apply to cells in vivo (we are unaware of any studies documenting bone cell deformation in vivo). Therefore, we refer to local bone matrix strain adjacent to the lacunae, rather than cell strain. However, even if cell strain in vivo is a small fraction of matrix strain, presumably, cells in regions of supra-continuum levels of matrix strain experience considerably greater deformation than cells in continuum levels of matrix strain. Third, if the matrix is regularly subjected to levels of strain that are up to ten times the continuum levels in specific regions, one would suppose some evidence of fatigue damage. Repetitive high level (2500 microstrain, 10 000 cycles, 2 Hz) experimental loading of dog radii caused increased numbers of microcracks and remodeling events. Although the in vivo fatigue limits of bone are not clearly known, in one experiment, dead bone failed after 10 000 cycles at approximately 4000 microstrain.10 Another experiment suggests that the ultimate in vitro strain levels are in the range of 20 000– 60 000 microstrain.14 However, these strain measurements were performed on a macroscopic scale where the strains thus measured are an average over the length scale of the strain-measuring device (i.e., strain gauge or extensometer). Within this gauge area, there may be very localized regions of high strain that do not necessarily result in ultimate bone failure. Rather, these localized regions are constrained from failing by the surrounding material.17 Thus, while not resulting in ultimate failure of the bone, local defects and microdamage may result in substantially increased matrix strains that are greater than reported macroscopic ultimate tissue strains. Such increased levels of local strains could, therefore, be acting on nearby osteocytes, thus triggering a possible remodeling response. In fact, in cortical bone specimens loaded in fatigue at strains at or just above the in-vivo strain gauge-measured strains (2500–3500 microstrain), the initial appearances of microcracks were “adjacent to or touching osteocytes”, suggesting high strain levels around osteocytes.47 Clearly, bone does not grossly fatigue under normal circumstances. However, if the microscopic levels of strain are in the range of 20 000 to 30 000 microstrain (i.e., ten times the continuum average), one would suppose a clear source of microdamage, and perhaps a stimulus for remodeling. In point of fact, experimental studies have shown that bone resorption spaces and remodeling are associated with both linear microcracks and areas of diffuse bone damage.1,7,37 Fourth, tissues clearly adapt to their mechanical environment, and it is likely that matrix2 and cells3 adapt

R.A. Brand et al.: Bone cells do not respond to tissue level strains

similarly. Based upon empirical evidence, Turner52 proposed three rules for bone adaptation: “(1)It is driven by dynamic, rather than static loading. (2) Only a short duration of mechanical loading is necessary to initiate an adaptive response. (3) Bone cells accommodate to a customary mechanical loading environment, making them less responsive to routine loading signals.” This third rule is similar to that proposed by Brand and Stanford,3 who suggested that cells might alter either their connections to the matrix, or their cytoskeleton, to adapt to a new mechanical environment. For example, cells in a high-strain region might develop loose or few connections, while those in a low-strain region might develop tight or many connections. Fifth, any high-strain regions surrounding lacunae would affect only osteocytes, and not cortical or trabecular surface cells (although such local stress risers also occur on bone or implant surfaces). We presume that osteocytes act as sensing cells, which, in turn, stimulate either Haversian remodeling within the substance of bone, or osteoblastic-osteoclastic activity on the surface of bone. In interpreting the literature, one should distinguish cell source. Most of the experiments stretching bone-like cells have been performed on transformed cell lines, or cells from the surface of embryonic, fetal, or newborn bones (most commonly, the calvarium). We question whether or not the responses of these cells are relevant to the hypothesis explored in this article. Rather, we suspect that only the limited experiments in “osteocyte-enriched” cultures (e.g., the approach of Robey and Termine48) from adult humans or animals are relevant. A number of authors report the proliferative responses of human osteocyte-like cells in mixed cultures (See Fig. 1). Jones et al.31 reported reproducible responses in these cells only at 10 000 microstrain, an identical magnitude of strain at which Neidlinger-Wilke et al.40 reported consistent increases in proliferation. It is important to note that both of these groups used nonconfluent cultures, while the other authors noted in Fig. 1 used confluent cultures. It is obvious that nonconfluent cultures (i.e., those with cells not subject to contact inhibition) would be more “primed” to proliferate. The other studies noted in Fig. 1 report either negative responses or inconsistent proliferative responses between cultures from differing subjects. We interpret these data as suggesting that human osteocytelike cells do not proliferate at levels of strain typically measured with strain gauges. When considering the intracellular and matrix protein responses of primary human osteocyte-like cells (from the authors and conditions noted in Fig. 1), we observe virtually no reproducible responses in the studies reporting the effects of prostaglandin E2 (PGE2),18 lactate dehydrogenase,40 bone sialoprotein

299

Fig. 1. Proliferative responses in adult human osteocyte-like bone cells reported in the literature. Data at 800, 1600, 2000, and 3200 microstrain in confluent cultures from Fermor et al.18 Data at 1000 microstrain in confluent cultures from Stanford et al.51 Data at 300, 3000, and 10 000 microstrain in nonconfluent cultures from Jones et al.31 Data at 10 000, 24 000, 53 000, and 88 000 microstrain in nonconfluent cultures from Neidlinger-Wilke et al.40 We interpret these data as suggesting that human osteocyte-like cells are not responsive at levels of strain typically measured with strain gauges

(BSP), osteocalcin, osteopontin, or alkaline phosphatase.51 We thus conclude these cells are not very responsive. Turner et al.53 questioned whether continuum-level strain gradients (rather than strain magnitude) governed trabecular bone adaptation, and, using FEM (and single-time single event peak loading conditions), concluded that gross (organ level) trabecular density (not necessarily orientation) could be predicted by continuum level strain gradients. Their analysis does not relate strictly to cortical bone, although their iterative algorithm predicted dense bone in the cortical regions. Over a large area, continuum level strains or strain gradients might predict trabecular density, but such mathematical prediction does not imply mechanism. Nonetheless, the suggestion by Turner and colleagues that strain gradients might be mechanistically related to bone cell stimulation makes intuitive sense: high strain gradients, such as those around lacunae or canaliculi, might cause greater fluid flow through an essential porous medium, thus stimulating cell responses. We thus make three arguments suggesting that bone cells sensing low-frequency mechanical signals are not responding to direct deformation via direct matrix connections at continuum level strains, although they may be responding to very high local strains well above tissue continuum level strains, or cell deformation resulting in fluid flow-induced shear stress. These arguments are: (1) Bone cells in vitro typically do not respond to average levels of strain magnitude; (2) bone cells in vitro do respond to fluid flow-induced shear

300

R.A. Brand et al.: Bone cells do not respond to tissue level strains

stresses in presumed physiological ranges; and (3) invivo strain magnitudes presumed to stimulate remodeling reflect only averages, and not local peaks, which are two to ten times higher. These three pieces of evidence, along with other evidence on the fatigue limits of bone, suggest that supra-continuum levels of strain routinely occur in some regions of bone and stimulate bone adaptation, perhaps owing to local microdamage or fluid flow. Acknowledgments. Supported in part by NIA Grant AG15197, The Roy J. Carver Charitable Trust, and by the Department of Orthopaedic Surgery and Dows Institute for Dental Research, The University of Iowa.

19.

20.

21. 22.

23.

24.

25.

References 1. Bentolila V, Boyce TM, Fyhrie DP, et al. Intracortical remodeling in adult rat long bones after fatigue loading. Bone 1998;23:275–81. 2. Brand RA. Autonomous informational stability in connective tissues. Med Hypotheses 1992;37:107–14. 3. Brand RA, Stanford CM. How connective tissues temporally process mechanical stimuli. Med Hypotheses 1994;42:99–104. 4. Brighton CT, Strafford B, Gross SB, et al. The proliferative and synthetic response of isolated calvarial bone cells of rats to cyclic biaxial mechanical strain. J Bone Joint Surg Am 1991;73:320–31. 5. Buckley MJ, Banes AJ, Jordan RD. The effects of mechanical strain on osteoblasts in vitro. J Oral Maxillofac Surg 1990;48:276– 82; discussion 282–3. 6. Burger EH, Veldhuijzen JP. Influence of mechanical factors on bone formation, resorption, and growth in vitro. In: Burger EH, Veldhuijzen JP, editors. Bone growth-B. Boca Raton: CRC; 1993. 7. Burr DB, Martin RB, Schaffler MB, et al. Bone remodeling in response to in vivo fatigue microdamage. J Biomech 1985;18:189– 200. 8. Burr DB, Milgrom C, Fyhrie D, et al. In vivo measurement of human tibial strains during vigorous activity. Bone 1996;18:405– 10. 9. Carter DR. Mechanical loading history and skeletal biology. J Biomech 1987;20:1095–109. 10. Carter DR, Spengler DM, Frankel VH. Bone fatigue in uniaxial loading at physiological strain rates. IRCS Medical Science: Biomedical Technology: Connective Tissue, Skin, and Bone: Surgery and Transplantation 1977;5:592. 11. Chamay A, Tschantz P. Mechanical influences in bone remodeling. Experimental research on Wolff’s law. J Biomech 1972;5:173– 80. 12. Chen NX, Ryder KD, Pavalko FM, et al. CA(21) regulates fluid shear-induced cytoskeletal reorganization and gene expression in osteoblasts. Am J Physiol Cell Physiol 2000;278:C989–97. 13. Cowin SC. Bone remodeling of diaphyseal surfaces by torsional loads: theoretical predictions. J Biomech 1987; 20:1111–20. 14. Crowninshield RD, Pope MH. The response of compact bone in tension at various strain rates. Ann Biomed Eng 1974;2:217–25. 15. Dallas SL, Zaman G, Pead MJ, et al. Early strain-related changes in cultured embryonic chick tibiotarsi parallel those associated with adaptive modeling in vivo. J Bone Miner Res 1993;8:251–9. 16. Davidson DL, Lankford J. Fatigue crack tip strains in 7075-T6 by stereoimaging and their use in crack growth models. ASTM STP 1983;811:371–99. 17. Davidson DL, McClung RC. Local constraint near fatigue cracks in alloys and particulate composites. Int J Fracture 1997;84:81–98. 18. Fermor B, Urban J, Murray D, et al. Proliferation and collagen synthesis of human anterior cruciate ligament cells in vitro: effects

26.

27.

28.

29.

30.

31.

32.

33.

34. 35.

36. 37. 38.

39.

40.

of ascorbate-2-phosphate, dexamethasone and oxygen tension. Cell Biol Int 1998;22:635–40. Frost HM. The mechanostat: a proposed pathogenic mechanism of osteoporoses and the bone mass effects of mechanical and nonmechanical agents. Bone Miner 1987;2:73–85. Frost HM. A brief review for orthopedic surgeons: fatigue damage (microdamage) in bone (its determinants and clinical implications). J Orthop Sci 1998;3:272–81. Fyhrie DP, Carter DR. A unifying principle relating stress to trabecular bone morphology. J Orthop Res 1986;4:304–17. Gross TS, McLeod KJ, Rubin CT. Strain energy density as a function of time: characterizing the skeleton’s functional strain history. Trans Orthop Res Soc 1991;16:420. Harell A, Dekel S, Binderman I. Biochemical effect of mechanical stress on cultured bone cells. Calcif Tissue Res 1977;22(Suppl):202–7. Harrigan TP, Jasty M, Mann RW, et al. Limitations of the continuum assumption in cancellous bone. J Biomech 1988; 21:269–75. Hasegawa S, Sato S, Saito S, et al. Mechanical stretching increases the number of cultured bone cells synthesizing DNA and alters their pattern of protein synthesis. Calcif Tissue Int 1985;37:431– 6. Hollister SJ, Brennan JM, Kikuchi N. A homogenization sampling procedure for calculating trabecular bone effective stiffness and tissue level stress. J Biomech 1994;27:433–44. Huiskes R, Weinans H, Grootenboer HJ, et al. Adaptive bone-remodeling theory applied to prosthetic-design analysis. J Biomech 1987;20:1135–50. Hung CT, Pollack SR, Reilly TM, et al. Real-time calcium response of cultured bone cells to fluid flow. Clin Orthop 1995:256–69. Jacobs CR, Yellowley CE, Davis BR, et al. Differential effect of steady versus oscillating flow on bone cells. J Biomech 1998;31: 969–76. Jendrucko RJ, Hyman WA, Newell PH Jr, et al. Theoretical evidence for the generation of high pressure in bone cells. J Biomech 1976;9:87–91. Jones DB, Nolte H, Scholubbers JG, et al. Biochemical signal transduction of mechanical strain in osteoblast-like cells. Biomaterials 1991;12:101–10. Kaspar D, Seidl W, Neidlinger-Wilke C, et al. Dynamic cell stretching increases human osteoblast proliferation and CICP synthesis but decreases osteocalcin synthesis and alkaline phosphatase activity. J Biomech 2000;33:45–51. Klein-Nulend J, Burger EH, Semeins CM, et al. Pulsating fluid flow stimulates prostaglandin release and inducible prostaglandin G/H synthase mRNA expression in primary mouse bone cells. J Bone Miner Res 1997;12:45–51. Lanyon LE, Rubin CT. Static vs dynamic loads as an influence on bone remodelling. J Biomech 1984;17:897–905. Levenston ME, Beaupre GS, Schurman DJ, et al. Computer simulations of stress-related bone remodeling around noncemented acetabular components. J Arthroplasty 1993;8:595– 605. Malvern LE. Introduction to the mechanics of a continuous media. Englewood Cliffs, NJ: Prentice-Hall, 1969. p. 537–539. Mori S, Burr DB. Increased intracortical remodeling following fatigue damage. Bone 1993;14:103–9. Murray DW, Rushton N. The effect of strain on bone cell prostaglandin E2 release: a new experimental method. Calcif Tissue Int 1990;47:35–9. Neidlinger-Wilke C, Stalla I, Claes L, et al. Human osteoblasts from younger normal and osteoporotic donors show differences in proliferation and TGF beta release in response to cyclic strain. J Biomech 1995;28:1411–8. Neidlinger-Wilke C, Wilke HJ, Claes L. Cyclic stretching of human osteoblasts affects proliferation and metabolism: a new experimental method and its application. J Orthop Res 1994;12:70–8.

R.A. Brand et al.: Bone cells do not respond to tissue level strains 41. Nicolella DP, Lankford J, Jepsen KJ, et al. Correlation of physical damage development with microstructure and strain localization in bone. Am Soc Mech Eng 1997;35:311–12. 42. Nicolella DP, Lankford J, Nicholls AE. Direct measurement of microstructural strain in cortical bone. Trans Orthop Res Soc 1997;22:39. 43. Owan I, Burr DB, Turner CH, et al. Mechanotransduction in bone: osteoblasts are more responsive to fluid forces than mechanical strain. Am J Physiol 1997;273:C810–5. 44. Qin YX, Rubin CT, McLeod KJ. Nonlinear dependence of loading intensity and cycle number in the maintenance of bone mass and morphology. J Orthop Res 1998;16:482–9. 45. Rawlinson SC, Mosley JR, Suswillo RF, et al. Calvarial and limb bone cells in organ and monolayer culture do not show the same early responses to dynamic mechanical strain. J Bone Miner Res 1995;10:1225–32. 46. Reed C, Athanasiou DA, Constantinides G, et al. Influence of constant hydrostatic pressure on osteoblast-like cells. J Dent Res 1993;72SI:276. 47. Reilly GC. Observations of microdamage around osteocyte lacunae in bone. J Biomech 2000;33:1131–4. 48. Robey PG, Termine JD. Human bone cells in vitro. Calcif Tissue Int 1985;37:453–60. 49. Robling AG, Burr DB, Turner CH. Partitioning a daily mechanical stimulus into discrete loading bouts improves the osteogenic response to loading [in process citation]. J Bone Miner Res 2000;15:1596–602 [MEDLINE record in process].

301

50. Rubin CT, Lanyon LE. Dynamic strain similarity in vertebrates; an alternative to allometric limb bone scaling. J Theor Biol 1984;107:321–7. 51. Stanford CM, Welsch F, Kastner N, et al. Primary human bone cultures from older patients do not respond at continuum levels of in vivo strain magnitudes. J Biomech 2000;33:63–71. 52. Turner CH. Three rules for bone adaptation to mechanical stimuli. Bone 1998;23:399–407. 53. Turner CH, Anne V, Pidaparti RM. A uniform strain criterion for trabecular bone adaptation: do continuum-level strain gradients drive adaptation? J Biomech 1997;30:555–63. 54. Turner CH, Owan I, Takano Y. Mechanotransduction in bone: role of strain rate. Am J Physiol 1995;269:E438–42. 55. Van Rietbergen B, Muller R, Ulrich D, et al. Tissue stresses and strain in trabeculae of a canine proximal femur can be quantified from computer reconstructions [corrected and republished in J Biomech 1999;32:443–51]. J Biomech 1999;32:165–73. 56. Whalen RT, Carter DR, Steele CR. Influence of physical activity on the regulation of bone density. J Biomech 1988;21:825–37. 57. Williams DR, Davidson DL, Lankford J. Fatigue-crack-tip plastic strains by the stereoimaging technique. Exp Mech 1980;20:134– 9. 58. Wolff J. Das Gezetz der Transformation der Knochen. Berlin: Verlag von August Hirschwald; 1892. 59. Zeng Y, Cowin SC, Weinbaum S. A fiber matrix model for fluid flow and streaming potentials in the canaliculi of an osteon. Ann Biomed Eng 1994;22:280–92.