betaig-h3 is an extracellular matrix protein and its expression is highly induced by TGF-beta and it has also been suggested to play important roles in skin wound healing. In this paper, we demonstrate that betaig-h3 is present in the papillary layer of dermis and synthesized in the basal keratinocytes in vivo and its expression is induced by TGF-beta in normal human keratinocytes (NHEK) and HaCaT cells. betaig-h3 mediates not only adhesion and spreading of keratinocytes but also supports migration and proliferation. These activities are mediated through interacting with alpha3beta1 integrin. Previously identified two alpha3beta1 integrin-interacting motifs of betaig-h3, EPDIM, and NKDIL, are responsible for these activities. The results suggest that betaig-h3 may regulate keratinocyte functions in normal skin and potentially during wound-healing process.

When endothelial cells isolated from human umbilical veins were cultured for 6 days using 96-well microplates, the final cell density in each well was found not to be the same although the medium composition of each well was exactly the same. Cell growth in the wells located at the periphery of a microplate was low, while that in the central area of the plate was high. A possible cause for different rate of growth was proposed as the uneven concentration of oxygen in the culture medium of each well.

Vascular endothelial cells (ECs) are usually difficult to culture in a large scale because of their complicated requirements for cell growth. As the vascular endothelial growth factor (VEGF) is a key growth factor in the EC culture, we transfected human umbilical vein endothelial cells (HUVEC) using a plasmid containing VEGF gene and let them grow in a culture medium eliminated an important supplement, endothelail cell growth supplement (ECGS). The expression of VEGF by HUVEC tansfected with VEGF gene was not enough to stimulate the growth of HUVEC, only 40% of maximum cell density obtainable in the presence of ECGS., However, when the culture medium was supplied with 2.5 ng/mL of basic fibroblast growth factor (bFGF), a synergistic effect of VEGF and bFGF was observed. In this case, the final cell density was recovered up to about 78% of maxium value.

We tested the hypothesis that division of inputs between the hemispheres could prevent interword letter migrations in the form of illusory conjunctions. The task was to decide whether a centrally-presented consonant-vowel-consonant (CVC) target word matched one of four CVC words presented to a single hemisphere or divided between the hemispheres in a subsequent test display. During half of the target-absent trials, known as conjunction trials, letters from two separate words (e.g., "tag" and "cop") in the test display could be mistaken for a target word (e.g., "top"). For the other half of the target-absent trails, the test display did not match any target consonants (Experiment 1, N = 16) or it matched one target consonant (Experiment 2, N = 29), the latter constituting true "feature" trials. Bi- as compared to unihemispheric presentation significantly reduced the number of conjunction, but not feature, errors. Illusory conjunctions did not occur when the words were presented to separate hemispheres.

We demonstrate that division of inputs between the hemispheres does not just generally improve performance but specifically reduces inter-item perceptual interference relative to single hemisphere presentation. Subjects verified the presence of colored letter targets in briefly presented letter pairs presented to the same (unihemispheric condition) or different (bihemispheric condition) sides of space. Four experiments showed consistently that although illusory conjunctions (i.e. perceptual miscombinations of stimulus features) occurred commonly during unihemispheric presentations, they never occurred during bihemispheric presentations, even though accuracy of detecting individual stimulus features was held constant across the conditions. Increasing inter-item similarity resulted in a general interference effect based in discrimination difficulty that was sufficient to eliminate illusory conjunctions even in the unihemispheric condition. However, when inter-item interference was manipulated using perceptual grouping by movement the bihemispheric illusory conjunction advantage increased with increasing interference. Our findings suggest that the two cerebral hemispheres can operate as relatively independent work stations during visual perceptual processing, thereby maximizing the efficiency of information processing.