Retroelements (these are also transposable elements) are those nucleic acid sequences, which either originate partly or fully, or propagate, through reverse transcription (RNA→DNA). These retroelements are classified into (i) viral retroelements and (ii) non-viral retroelements. Both these classes are further subdivided on the basis of structural and biological criteria (Table 15.6). Since viral retroelements also have an extracellular phase and are infectious, they may not be very important for a discussion on extranuclear accessory genomic elements, included in this section. The non-viral retroelements include retrotransposons, which resemble transposons, except that they have RNA origin. They have (i) LTR (long terminal repeat) sequences needed for transposition, (ii) primer binding site (PBS), (iii) a promoter, and (iv) processing/ polyadenylation signal. These attributes are missing in retroposons. Retrons include ms DNA of mycobacteria and E. coli, but may include additional sequences from other prokaryotes. They are characterized by an unconventional mechanism of reverse transcription. Retrosequences include cDNA genes or pseudogenes, which end in poly (A) tails and are flanked by short repeat sequences.

Myxobacteria, a soil bacterium (Stigmatella aurantiaca)- and 10% of all clinical strains of E. coli tested have been shown to produce peculiar RNA-DNA complexes consisting of (i) a DNA component called multicopy single stranded DNA (ms DNA) and (ii) an RNA component called multicopy single stranded DNA like RNA (msd RNA). The genomic DNA consists of (i) sequences coding for reverse transcriptase, (ii) a msd sequence coding for ms DNA and (iii) a msr sequence coding for msd RNA. DNA molecule is attached at 5' end to a G-residue of RNA molecule through a phosphodiester bond, and on the 3' end it attaches non-covalently via a few residues. Both DNA and RNA have characteristic stem-loop structures (Fig. 15.20). For more details the readers should consult Benjamin Lewin's Genes IV (1990).