Jg. Yodh et al., Evidence for nonrandom behavior in 208-12 subsaturated nucleosomal array populations analyzed by AFM, BIOCHEM, 38(48), 1999, pp. 15756-15763
Atomic force microscopy was used to determine the population distributions
in reconstituted, subsaturated 208-12 nucleosomal arrays. The features foun
d in these distributions vary with the average nucleosome loading per templ
ate molecule (n(av)): at n(av) < 4, the distributions show a single peak wh
ose breadth is equal to that expected for a random loading process; at n(av
) = 4-8, the distributions are broader than random distributions and are co
mplex; i.e., they contain multiple peaks and/or shoulders. Moreover, the pe
aks/shoulders typically occur at two nucleosome intervals, i.e., 2, 4, 6 or
3, 5, 7 nucleosomes. This two-nucleosome periodicity is statistically sign
ificant. The precise cause for such discrete features within the distributi
ons is unknown, but at least these features would seem to indicate some pai
rwise preference in nucleosome occupation at these loading levels. In these
intermediate-level (n(av) = 4-8) distributions, the major peak contains a
larger fraction of the total templates than a random nucleosome loading pro
cess would produce. This feature indicates that at these intermediate popul
ation levels there is some tendency for correlated nucleosome loading among
the templates. Hyperacetylated nucleosomal arrays show only subtle differe
nces in their population distributions compared to nonacetylated arrays and
demonstrate the above features. AFM allows one to study unfixed chromatin
arrays; we find that nucleosomes on the 208-12 template demonstrate signifi
cant lability when they are not glutaraldehyde-fixed.