Stretching effects at microscope. 2
A NEW METHOD TO ALIGN AND ORIENT HUMAN PATELLAR TENDON FIBROBLASTS DURING CYCLIC
*Wang, J H-C.; *Stone, D; +*Woo, S L-Y.
+*Musculoskeletal Research Center, Department of Orthopaedic Surgery, University of Pittsburgh
In vitro model systems have been developed to study the effects (e.g.,
cell proliferation [1,4]) of repetitive mechanical stretching on tendon
fibroblasts without controlling the cell alignment and shape. Therefore, the
cellular responses may not be similar to those in vivo, as the tendon
fibroblasts align with collagen fibers and are subjected to tension along the
tendon stretching direction. We have developed an improved model, in which
cells are grown on microgrooved surfaces, instead of commonly used smooth
surfaces. The purpose of this study was to examine the effects of
microgrooved surfaces on alignment and shape of tendon fibroblasts, and cell
proliferation in response to cyclic mechanical stretching.
MATERIALS AND METHODS
Six-well silicone dishes were custom-made. Each well (about 3 cm2) of
the dish contained microgrooved surfaces………..
Thegrooves in the dishes were oriented either parallel or perpendicular to the dish
long axis, along which cyclic stretching was applied.
Human patellar tendon (PT) fibroblasts were isolated from surgical
tendon samples using explant tissue culture techniques. The cells, passaged
between 6 and 9 times, were transferred to silicone microgrooves and smooth
surfaces, both having been coated with 10 μg/ml of ProNectin-F (BioSystem,
CA) to promote cell adhesion to silicone surfaces. ………..
the cell shape was quantified using shape index(SI),
which is defined to be SI = 4π·A/P2, where A is a cell area and P is its
Note that the smaller the SI, the more elongated the cell. A total of 12
fibroblasts from grooves and smooth surfaces, respectively, were used in
determining cell SI. In separate experiments, cyclic stretches of 0 (i.e., without
stretching) and 8%, at 0.5 Hz, were applied to the dishes, using the stretching
apparatus previously described . To determine cell proliferation, cells were
stretched for 2h, incubated in the stretching-conditioned medium for an
additional 24h, and directly counted with a hemocytometer.
Before being stretched, human PT fibroblasts aligned in silicone
microgrooves. After cyclic stretching, the cells remained aligned in the
grooves. This was true for cells in silicone microgrooves oriented both along
the stretching direction and perpendicular to the stretching direction. In
contrast, the same cells on smooth surfaces oriented away from the stretching
direction. Moreover, the cells in grooves had an elongated shape, whereas the
same cells on smooth surfaces had a well-spread shape (Fig 1).
The SI forcells in the grooves was 0.33 ± 0.10 (mean ± SD), whereas the SI for cells on
smooth surfaces was 0.60 ± 0.15. The two indexes were significantly
different (p < 0.0001).
Furthermore, cell numbers among the three conditions, that is, (a)
stretching cells in grooves oriented in the stretching direction, (b) stretching
cells in grooves oriented perpendicular to the stretching direction, and (c) no
stretching, were significantly different (p < 0.005). Specifically, cell number
under condition (a) was significantly higher than that under either conditions
(b) or (c) (p < 0.05), but cell number between (b) and (c) was not significantly
different (p > 0.05) (Fig 2).
The study showed that human PT fibroblasts can align in silicone
microgrooves with or without stretching. Further, the shape of cells in the
grooves became significantly elongated, compared with that of cells on
smooth surfaces. This elongated cell shape is similar to that of tendon
fibroblasts in vivo. Thus, these elongated cells in microgrooves may have a
metabolic state closer to that in vivo, since cell shape determines cell
Moreover, this study showed that human PT fibroblasts can proliferate
in silicone microgrooves, which was confirmed by assaying DNA synthesis
with BrdU incorporation. Furthermore, in response to cyclic mechanical
stretching, proliferation of the tendon fibroblasts was different when the cells
were stretched along their long axes or transverse axes, although stretching
magnitude in both cases appeared to be similar. This result suggests that other
cellular responses, for instance, stretching-induced inflammatory mediators
, may also depend on cells’ long axes with respect to stretching direction.