Biotechnol. Rag. 1992, 8, 572-575
572
Altered Distribution of Mitochondria and Actin Fibers in 3T3 Cells Cultured on Microcarriers Tracey V. du Laney, Robert S. Cherry,* Justine A. Coppinger,t and George A. Truskeyt Center for Biochemical Engineering, Duke University, Durham, North Carolina 27708-0276
T h e mitochondria and actin fibers of 3T3fibroblasts cultured on microcarriers in spinner flasks were visualized using fluorescent stains. In contrast to cells grown on planar surfaces under static or steady laminar flow conditions, cells exposed to higher levels of turbulent agitation do not form actin stress fibers. Greater agitation also leads to a more diffuse appearance of the mitochondria and a wider distribution of them throughout the cytoplasm. This response may indicate damaged mitochondria, as similar results have been reported for chemical toxins.
Introduction Despite the important role of fluid motion and mixing in bioreactor design, there has been relatively little work on the effect of nonlethal fluid stresses on the kinds of cells that might be grown commercially. In particular, little is known about the detailed physiology and functioning of cells grown on microcarriers, an especially complex fluid mechanical environment because of the relatively high volume fraction of beads and the presence of fluid turbulence. The effects of long-term exposure to shear stress have most frequently been investigated by subjecting endothelial cells to steady laminar shear stresses at levels typical of those in vivo. Many responses have been reported, including elongation and orientation parallel to the flow (Ives et. al., 1983; Levesque and Nerem, 19851, formation of actin stress fibers parallel to the flow (Franke et al., 1984;Wechezak et al., 19851,and changes in concentrations of proteins and signaling compounds. Similar results for orientation and altered urokinase expression have been reported for human embryonic kidney cells, a cell type not exposed to shear stresses in vivo (Stathopoulos and Hellums, 1985). In addition, under laminar flow, endothelial cell DNA synthesis and the rate of cell doubling decrease as the shear stress increases (Ziegler et al., 1990). Decreases in DNA synthesis have also been reported for mouse NIH 3T3fibroblast cells subjected to shear stresses of 20 dyn/cm2 (Vanhee et al., 1991). The response to transient flow is more complex. Step changes in the shear stress can lead to transient effects such as greatly increased prostacyclin production (Frangos et al., 1988) and elevated internal free calcium concentrations (Ando et al., 1988), while continued oscillatory stresses cause increased histamine metabolism (Skarlatos and Hollis, 1987). Furthermore, while exposure to 1.5 dyn/ cm2 of laminar shear stress has no effect, turbulent flow a t the same mean shear stress causes an increase in DNA synthesis (Davies et al., 1986). To examine cell behavior under bioreactor conditions, we have grown 3T3 fibroblasts on microcarriers in spinner flasks at several agitation levels and recorded the distribution of fluorescently labeled mitochondria and actin
* Corresponding author; also in the Department of Mechanical Engineering. t Present address: Washington University, St. Louis, MO 63130. Also in the Department of Biomedical Engineering.
*
87567938/92/300&0572$03.00/0
fibers in those cells at several different times during the course of batch cultures. Effects on the mitochondria may indicate changes in the energy status of the cells, while changes in the cytoskeleton may affect the strength of attachment, extent of spreading, cell motility, and possibly cell growth (Vasilev, 1985).
Procedures Cells and CultureConditions. 3T3 fibroblasts (ATCC CCL92, Rockville, MD) were cultured in a 37 "C, 5% CO2 incubator in 75 cm2 T-flasks using Dulbecco's Modified Eagle's Medium (DMEM) supplemented with 10% calf serum, 100 units/mL penicillin, and 100 pglmL streptomycin (all from Sigma). Cytodex I1 microcarriers (Pharmacia) were prepared and seeded with cells according to the manufacturer's procedures in 150-mLCorning spinner flasks. After 24 h at 50 rpm, the agitation speed was then changed to 60,100, or 150 rpm for each experiment. To create an unstirred control, cells were seeded onto 25-mm glass cover slips. If cell-bearing microcarriers were transferred to an unstirred flask as a more directly comparable control, many of the cells tended to crawl off the beads and onto the flask surface. Those that remained on the microcarrier surface would not be representative and were not analyzed. Static controls grown directly on plastic tissue culture flasks (Corning)gave good agreement with those on glass cover slips. The entire series of experiments was performed twice with similar results each time. In similar experiments not detailed here, Chinese hamster ovary (CHO) cells showed the same trends as the 3T3 cells under the same conditions. To measure growth rates, cells were seeded onto microcarriers and counted daily for 5 days using 0.1 w t % crystal violet in 0.1 M citric acid to stain the nuclei and burst the cells. Static controls were counted on a hemocytometer after trypsinization. Staining Procedures. The procedure for using rhodamine 123, a mitochondrial-specific cationic fluorescent dye, to visualize the mitochondria was modified from a published procedure (Johnson et al., 1980). Rhodamine 123 (Sigma) was dissolved in deionized water to a concentration of 1mg/mL and then diluted in phosphatebuffered saline (PBS) to make a stock solution of 10 pg/ mL. To dye cells, 1 mL of microcarrier/cell suspension was placed into a small glass test tube. Rhodamine 123 stock solution (15 pL) was added, and the sample was incubated at 37 "C for 30 min. The supernatant was then
0 1992 American Chemical Society and Amerlcan InstRute of Chemical Engineers
57a
Bbtecfmo/. pmg.. 1992. Vol. 8. No. 6
decanted, and the cells were rinsed twice with 1 mL of warm medium to remove unabsorbed dye. The cells, still on the microcarriers, were then resuspended in 1mL of fresh medium. Actin fibers in fixed, permeabiliid cella were stained with fluorescein-labeled (FITC) phalloidin using a procedure modified from that of Barak et al. (1980). On each day it was used, concentrated FITC-phalloidin solution in DMSO (Sigma) was evaporated and rediluted (1:20) in PBS solution (pH 7.3). To stain the cells, 0.5 mL of microcarrier suspension was decanted, washed twice with 0.5 mL of PBS, and fiied hy adding 1 mL of 3.75% formaldehyde. After 10 min, the cells were rinsed twice
a
withPBSandthenextradedwith0.5mLof-2O0Cacetone for 3-5 min. The acetone was removed, and 200 pL of FITC-phalloidin in PBS was added. After 20 min, 0.5 mL of PBS was added and mixed, and the supernatant was decanted. The cells were then rinsed twice with PBS. To view the cells dyed with either rhodamine 123 or FITC-phalloidin, 10 pL of prepared sample was pipetted onto a clean slide, and 10pL of Fluoromount-G mounting medium (Fisher Biotech) was added. A cover slip was placed gently on the drop, and the edges of the cover slip were sealed with clear fingernail polish to minimize evaporation. Although the cells on the microcarrierswere visually distinct if the focus was shifted up and down, taking sharp photographs of the cells was complicated by the curvature of the microcarriers; those presented are representative of several hundred microcarriers observed at each condition. Quantitativeresults were not attempted because (1)in the cells grown on microcarriers,there we?e no well-defined structures whose number or geometry could he measured and (2) each sample of cells took up FITC-phalloidinto a different extent, making comparison of fluorescent intensities meaningless as a measure of F-actin content. While the rhodamine 123 labeled mitochondria more consistently, its intensity reflects the combined effects of the size, number, and transmembrane potential of the organelles. Transmission electron microscopy of fixed cells sectioned through the microcarrier might be one way of obtaining more information about the details of these changes.
Results and Discussion Actin Fiber Staining. Static growth conditions yielded the most highly defined fibers of all growth conditions tested. All cells showed extensive fiber formation a t 24 h, and no changes were seen after that up to 72 h (Figure la). The fibers were oriented parallel to the long axis of the cell, with the brightest fluorescence a t the edges. In the stirred microcarrier experiments a t all agitation levels, theFITC-phalloidin-stained actin fluorescedevenly throughout the cytoplasm, with a slightly higher density around the nucleus. After 24 h, the cells were still spreading and had clearly defined pseudopods with brightly stained edges, indicating the presence of locally high amounts of F-actin. The orientation and location of the few visible fibers varied from cell to cell, as did their length and prominence. After 4 days most cell still showed no distinct fibers at all, just an overall dull glow (Figure lh). A t higher speeds of 100 and 150 rpm, the overall brightness of the cells increased. The cella were elongated in random directions. Elongation with orientation in the direction of a steady laminar shear stress takes about 2 h insubconfluent3T3 cells (TruskeyandPirone, 1990).This relatively long response time implies thatdirectionaleffects should not be seen in a microcarrier system where the direction and magnitude of fluid forces on the beads
b
Figure I. Epifluoresrent imaprof FII'C phalloidin-stninedactin of three 3T3 rclls grown under static conditions fur 24 h 1% top).
Extensive fiber formation is evident. Epifluorescent image of FITC-phalloidin-sutined actin uf a 3T3 cell grown on a microcarrier at 150 rpm fnr 48 h (b. bottomJ. Fiber formation is not evident, yet the whole cell glows dully. Srale harn are 30 urn.
fluctuate on a time scale measured in parts of seconds (Cherry and Kwon, 1990). Elongation is typical behavior for confluent fibroblasts even in static cultures, and we cannot determine in this work whether turbulent agitation caused any increased elongation independent of orientation. The overall trend of these results is that exposure to increasing levels of turbulence does not induce the formation of prominent actin stress fibers in the cells, aa occurs in response to steady laminar shear stress or aa seen in the controls of this study. Static endothelial cell cultures also exhibit short, thin stress fibers within the celland thin bandsalongthecircumferenceand whichare reported to be linked to endothelial contractility, tensegrity, and tissue permeability (Gotlieb et al., 1991). The overall F-actin density of cells grown on microcarriers is qualitatively much lower than in cells grown in static cultureand is reminiscent of migrating cells whichexhibit fewfibersanddiffuseactinstaining(CouchmanandRees, 1979). Mitochondrial Staining. Static growth of 3T3 cella produced highly fluorescent mitochondria concentrated around the nucleus with many points and threads of individual mitochondria visible at 24 h (Figure 2a). The only effect of time in culture was that the mitochondria spread throughout the cytoplasm. The mitochondria of cells stirred at 60 rpm for 24 h tended to be concentrated in a ring around the nucleus, which wasclearlyvisibleasadarkcircle. In approximately halfofthecells, themitochondriaappearedassmall bright spots, while the other cells showed a more uniform
Blotechno.
574
a
pros..
1992, Vol. 8. No. 6
Table I. Growth Rates under Different Agitation Conditions stirrer speed (rpm) static
growth rata (b-l)
fin ..
100 150
b
0.042 0.019 0.008 4.006
a reduced net growth rate and a range of significant structural, and presumably functional, changes in these cells. The low density of actin fihers relative to static culture may he due to continual stimulation and realignment of the cytoskeleton by fluctuating fluid forces. Disruption of the cytoskeleton can inhibit cell division and protein expreasion (Burridgeet al., 1988). Cytoskeletaldisruption might also alter the structure and distribution of mitochondria, as observed here. Alternatively, these mitochondrial and cytoskeletal effects may be independent consequences of a single more immediate response to agitationforcessuchaselevated intracellularcalciumlevels (Ando et al., 1988). The same type of change to a wider distribution and less well-defined individual shape has also been noted in mitochondria damaged by exposure to chemidtoxins (Lachowiezet al., 1989;Kohen et al., 1990) and may he related to poor cell growth and to cell death.
Conclusions
Figure 2. Epifluorescent image of rhodamine 123 stained mitochondria of 3T3 cells grown under static conditions for 24 h (a, top), Points and threads of mitochondria are visible throughout the cell, with a higher density around the nucleus. Epifluorescent image of rhodamine 123stained mitochondria of 3T3 cells grown on a microcarrier at 60 rpm for 24 h (b, bottom). The nuclei of all cells are evident; however, few punctate mitochondria are visible in the dull glow throughout the cytoplasm of the cells. Scale bars are 30 Irm. fluorescent glow (Figure 2h). After 5 days of culture, the glow decreased in intensity and the mitochondria were more spread throughout the cytoplasm, although the nucleus was still evident. In contrast, the mitochondria of cells stirred at 100 and 150rpmgenerallyfluorescedmorestronglyand werelarger, more distinct, and somewhat less concentrated close to the nucleus than those of cells maintained a t 60 rpm throughout 120 h in culture. With more time in culture, the mitochondria spread throughout the cell as occurred at 60 rpm. Some mitochondria still appeared as small spots surrounding the nucleus, hut their glow was not as distinct because of the stronger overall glow throughout the cytoplasm. GrowthRates. The exponential phase growth rates of 3T3 cells under different conditions are listed in Table I. Agitation speeds of 60,100, and 150 rpm clearly had an increasingly deleteriouseffect on cell growth. This reduced net growth rate might he caused by either a reduced rate of cell division or an increased death rate (Cherry and Papoutsakis, 1989). Cells that die detach from the microcarrier, so that the results reported here are for those cells which are still somewhat healthy. These remaining cells, however, are exactly the ones of interest in practical applications. Although we have not demonstrated a causal relationship between either possibility and the observed changes in the cytoskeleton and the mitochondria, the conditions of stirred microcarrier cultures do lead to both
Except for the special case of endothelial cells, the ways in which fluid forces might affect the normal biological and biochemical processes of cultured animal cells have not been explored. This work provides preliminary evidence that the fluctuating external physical forces imposed by agitation affect the normal physiology of a line of non-endothelial cells by reducing the formation of an organized actin cytoskeleton and altering the appearance and distribution of mitochondria. The functioning of anchorage-dependent cells is known to be related to the extent of cell spreading, and presumably therefore the structure of the cytoskeleton, and alterations in the mitochondria potentially have major implications for the availability of energy in the cell. These effects may not be independent since other work has indicated a relation betweencytoskeletal structure (microtubulesin particular) and the shape and distribution of mitochondria (Johnson etal., 1980). A better understanding of these changes and the mechanism by which fluid forces incite them could suggestways to improve the large-scale culturingof animal cells.
Acknowledgment Support for this work was provided to G.A.T. by the Whitaker Foundation and to R.S.C. and J.A.C. by the Carolina-Ohio Science Education Network, a cooperative program among seven private universities to encourage undergraduate students to enter graduate school in the sciences.
Literature Cited Ando, J.; Komatosuda, T.; Kamiya, A. Cytoplasmic Calcium Response to Fluid Shear Stress in Cultured Vascular Endcthelial cells. In Vitro Cell. Deu. Bid. 1988,24,871-877. Barak, L. S.; Yocum, R. R.; Nothnagel, E. A,; Webb, W. W. FluoreseenceStainingoftheActinCytoakeleton inLiving Cella with 7-Nitrobenz-2-oxa-l,3-diazolephallacidin. Proc. Natl. Acad. Sci. U.S.A. 1980, 77,980-984. Burridge, K.; Fath, K.; Kelly, T.; Nuckolls, G.; Turner, C. F d Adhesions: Transmembrane Junctions between the Extra-
Biotechnol.
nag., 1992, Vol. 8, No. 6
cellular Matrix and the Cytoskeleton. Annu. Rev. Cell Biol. 1988,4,4a7-525. Cherry, R. S.; Papoutsakis, E. T. Growth and Death Rates of Bovine Embryonic Kidney Cells in Turbulent Microcarrier Bioreactors. Bioprocess Eng. 1989,4,81-89. Cherry, R. S.; Kwon, K.-Y. Transient Shear Stresses on a Suspension Cell in Turbulence. Biotechnol. Bioeng. 1990,36, 563-571. Couchman, J. R.; Rees, D. A. The Behaviour of Fibroblasts Migratingfrom Chicken Heart Explants: Changes in Adhesion, Locomotion and Growth, and in the Distribution of Actomyosin and Fibronectin. J. Cell Sci. 1979,39,149-165. Davies, P. F.; Remuzzi, A.; Gordon, E. J.; Dewey, C. F., Jr.; Gimbrone, M. A., Jr. Turbulent Fluid Shear Stress Induces Vascular EndothelialCellTurnover in Vitro. Proc. Natl. Acad. Sci. U.S.A. 1986,83,2114-2117. Frangos, J. A.; McIntire, L. V.; Eskin, S. G. Shear Stress Induced Stimulation of Mammalian Cell Metabolism. Biotechnol. Bioeng. 1988,32,1053-1060. Franke, R. P.; Graefe, M.; Schnitller, H.;Seiffge, D.; Dreckhahn, D. Induction of Human Vascular Endothelial Stress Fibers by Fluid Shear Stress. Nature 1984,307,648-649. Gotlieb, A.; Langille, B.; Wong, M.; Kim, D. Structure and Function of the Endothelial Cytoskeleton. Lab. Invest. 1991, 65 (2),123-137. Ives, C. L.; Eskin, S. G.; McIntire, L. V.; DeBakey, M. E. The Importance of Cell Origin and Substrate in the Kinetics of Endothelial Cell Alienment in ResDonse to Steady Flow. Trans. Am. SOC.Art$. Org. 1983,2$ 269-274. Johnson, L.; Walsh, M.; Chen, L. Localization of Mitochondria in Living Cells with Rhodamine 123. Proc. Natl. Acad. Sci. U.S.A. 1980, 77,990-994. Kohen, E.; Kohen, C.; Prince, J.; Schachtsschabel,D.; Hirschberg, J.; Morliere, P.; Santus, R.; Dubertret, L.; Shapiro, B. Bioregulatory Mechanisms at the Level of Cell Organelle Interactions: Microspectrofluorometric in Situ Studies. J. Biotechnol. 1990,13, 1-28. ~~~~~~
~~~~~
575
Lachowiez, R. M.; Clayton, B.; Thallman, K.; Dix, J. A.; Van Buskirk, R. G. Rhodamine 123 as a Probe of in Vitro Toxicity in MDCK Cells. Cytotechnology 1989,2,203-211. Levesque, M. J.; Nerem, R. M. The Elongation and Orientation of Cultured Endothelial Cells in Response to Shear Stress. J . Biomech. Eng. 1985,107,341-347. McIntire, L.V.; Frangos, J. A.; Rhee, B. G.; Eskin, S. G.; Hall, E. R. The Effect of Fluid Mechanical Stress on Cellular Arachidonic Acid Metabolism. Ann. N . Y. Acad. Sci. 1987, 516,513-524. Skarlatos,S. I.; Hollis, T. M. Cultured Bovine Aortic Endothelial Cells Show Increased Histamine Metabolism when Exposed to Oscillatory Shear Stress. Atherosclerosis 1987,64,55-61. Stathopoulos, N. A.; Hellums, J. D. Shear Stress Effects on Human Embryonic Kidney Cells in Vitro. Biotechnol. Bioeng. 1985,27,1021-1026. Truskey, G. A.; Pirone, J. S. The Effect of Fluid Shear Stress upon Cell Adhesion to Fibronectin-Treated Surfaces. J. Biomed. Mater. Res. 1990,24,1333-1353. Vanhee, C.; Nerem, R. M.; Girard, P. R. Influence of FluidInduced Shear Stress on Cell Proliferation in NIH 3T3 Fibroblasts. American Society of Mechanical Engineers, Bioprocess Engineering Symposium, BED-Vol. 21,1991,pp 27-28. Vasilev, J. M. Spreading of Non-transformed and Transformed Cells. Biochim. Biophys. Acta 1985,780, 21-65. Wechezak, A. R.; Viggers, R. F.; Sauvage, L. R. Fibronectin and F-actin Redistribution in Cultured Endothetial Cells Exposed to Shear Stress. Lab. Invest. 1985,53,639-647. Ziegler, T.; Girard, P. R.; Nerem, R. M. Division of Cultured Endothelial Cells in the Presenceof flow. 82nd AnnualMeeting of the American Institute of Chemical Engineers, Chicago, IL, Nov. 11-16, 1990,Abstract 7f. Accepted August 24, 1992.
