The Cell–Surface Interaction - Tissue Engineering: Cell-Surface Interactions for Tissue Culture - page 18

(a) (b)

(c)

(d)

Fig. 5 How a cell may respond to the roughness spectrum. Cells must be able to perceive the
surface roughness for them to respond in a surface-dependent manner. a When a surface is
smooth, osteoblast cells will adopt a fibroblast-like morphology, becoming very flat and well
spread. b On a wavy surface where the distance between peaks is more than the average cell size,
cells will perceive the surface as smooth and will behave similarly to the behaviour shown in
a. c If, however, a surface has frequent surface irregularities, producing a microrough surface, the
cells are unable to spread and adopt typical osteoblast morphology. d On surfaces with mixed
topographies, cell behaviour will reflect the average of rough and smooth microtopographies

convincingly shown that cells (MSCs, osteoblasts) can detect, interact and respond
to nanotopographical features in vitro (see [ 85 ] for a review). In fact, this sensi-
tivity has been described as far as the 15-nm range [ 86 ]. However, can features of
this size be detected in vivo? And if so, how much actual influence do they have on
determining the cell/tissue-material interaction? Recent studies tentatively indi-
cate that cell and tissue interaction can be determined by nanotopography. For
instance, Bjursten et al. [ 87 ] have recently shown that titanium dioxide nanotubes
significantly enhance bone bonding, as measured by torque removal and per-
centage of bone contact, in an in vivo rabbit tibial model compared with grit-
blasted titanium. However, it is difficult from the data presented to differentiate if
the effect of microtopography was fully negated since the surface morphology of
the nanotube surface also appeared to have microscale morphology. Furthermore,
surface roughness measurements were made using scanning electron microscopy
alone and did not include any validated quantitative methods.
Meirelles et al. [ 88 ] have also shown that nano-titania and nano-hydroxyapatite
surfaces support bone on-growth in a rabbit model. However, it is worth noting
that the 'nano' surfaces had S a (mean arithmetic height measurement) of 121 nm
for titanium and 170 nm for hydroxyapatite surfaces compared with 225 nm for
the polished control. Although Meirelles et al. removed microstructures via
grinding, again the surface morphology did appear to have a level of micror-
oughness, which is supported by the fact that the height measurements were
reflective of a microtopography rather than a true nanotopography, i.e. tens rather
than hundreds of nanometres. Several methods such as photolithography exist for
producing nanometric surfaces, so perhaps in time more convincing evidence will
emerge that supports the theory that nanotopographical surface features can
determine tissue-implant interaction.