As early as 1963, Howard Temin had shown that replication of retroviruses was sensitive to not only to the transcription inhibitors Actinomycin D and alpha-amanitin, but more curiously, nucleoside analogues like 5-bromodeoxyuridine and cytosine arabinoside, which were known to inhibit DNA replication. In 1970, the famous "provirus" hypothesis was made, and shortly afterwards Howard Temin and David Baltimore separately isolated and described reverse transcriptase (for which they shared a Nobel Prize).

The above scheme for retrovirus replication is fairly well worked out although details of the second transfer are unclear.

Note that the 3' U3 becomes the 5'U3 for the next generation of virus. Similarly, the 5'U5 serves as template for the the 3' U5 for the integrated provirus.

The integrated form (proviral) of all retroviruses contain transcription regulatory sequences primarily in Long Terminal Repeats (LTR). LTR sequences are derived from sequences unique to the 5' end of viral RNA (U5), from sequences unique to the 3' end of viral RNA (U3), and from sequences repeated at both ends of the viral RNA (R). The integrated provirus is larger than the viral genome but its complexity is the same because of duplication of U3 and U5 during its synthesis.