Well I wouldn't say the multicellular system is reverting to a single older cell phase per say. Rather it needs NAD+ so it optimizes for that. The cells also

Well I wouldn't say the multicellular system is reverting to a single older cell phase per say. Rather it needs NAD+ so it optimizes for that. The cells also lose adhesion and then migrate so clearly there's some "programming exception" not being caught. I view the cell as a bunch of copy paste code that's had some degree of refactoring applied so there are different programs and control systems in operation at any given time. Sometimes these get activated or not repressed erroneously, sometimes the source code gets scrambled.