Such a programmed death would be extremely hard to do, risky and not necessary.
You could try to introduce a weakness to a particular poison (one that is not actually poisonous to anything you might want to introduce later), but that would be selected out against pretty fast. And genedrift would probably remove it even before you actually introduce the poison.
A vital dependency on some specific substance wouldn't work either. For one thing, you'd have to provide that substance Mars-wide, and also there would be a strong selection pressure to get around that dependency.
You could try to come up with some other timing methods. But genetical engineering is not magic.
>You could try to introduce a weakness to a particular poison (one that is not actually poisonous to anything you might want to introduce later), but that would be selected out against pretty fast. And genedrift would probably remove it even before you actually introduce the poison.
You could build the architecture of the genome such that any minor mutation to the "death gene" immediately kills the organism. You could probably get it to the point that it would require $n$ simultaneous point mutations to disable the "death gene" without killing the organism. What you'd really want to worry about is whole chunks of genome being deleted (either through a serious accident in reproduction or a virus). You'd want to design the genome such that each part checks the others parts (hell just use checksums SHA3 is available now), if any part comes up bad, the organism dies.
tl;dr It should be possible to freeze evolution (or just freeze particular gene's) using modern cryptographic techniques.
That's way ahead of current technology. And, those organisms would be quite unfit, and outcompeted fast.
The latter might not be a problem on Mars. But in such a harsh environment you do not only have to worry about competition, but about surviving at all.
You could try to introduce a weakness to a particular poison (one that is not actually poisonous to anything you might want to introduce later), but that would be selected out against pretty fast. And genedrift would probably remove it even before you actually introduce the poison.
A vital dependency on some specific substance wouldn't work either. For one thing, you'd have to provide that substance Mars-wide, and also there would be a strong selection pressure to get around that dependency.
You could try to come up with some other timing methods. But genetical engineering is not magic.