Molecular analysis of the heterogeneity region of the human ribosomal spacer.

The human ribosomal non-transcribed spacers are 30 X 10(3) base-pairs (or 30 kb) in length with a limited length heterogeneity localized in a specific region downstream from the 3' end of the transcribed region. Total DNA digested with EcoRI and BamHI and hybridized with a probe containing the 3' end of the 28 S ribosomal RNA coding region shows four major bands of 3.9 kb, 4.6 kb, 5.4 kb and 6.2 kb. The 5.4 kb band is the most abundant in every individual, followed by the 4.6 kb band. The longest and the shortest size classes are less well-represented and may even be absent. Every individual shows his own pattern of relative abundance of non-transcribed spacer length classes that can be followed through generations. We decided to investigate the molecular structure of the heterogeneity region, in order to cast light onto the mechanisms underlying the origin and maintenance of this length heterogeneity. Pertinent spacer regions of eight ribosomal clones from two human genomic libraries were subcloned and analyzed by restriction mapping and nucleotide sequencing. In the minimal length class, there is a sequence of 700 base-pairs that appears to be tandemly duplicated once, twice or three times in the other length classes. This repeated DNA module contains a region consisting of repetitions of simple pyrimidine groups like C-T, C-T-T-T or C-C-C-T. DNA module repeats may differ by the length of this pyrimidine-rich region. However, these length variations are not continuous, as revealed by Southern transfer analysis of several individuals and different cloned gene units: instead, the repeated modules fall into two discrete length classes of about 700 base-pairs and 800 base-pairs. An imperfect duplication of a short sequence of 86/89 base-pairs is present at the boundary between the heterogeneity region and the upstream flanking region, representing a very ancient duplication event.