5' and 3' strand directionality
/ five-prime / three-prime /
A strand of DNA is not a symmetrical noodle you can read either way; it has a head and a tail, like a sentence that only makes sense left to right. Those two ends are named 5' (five-prime) and 3' (three-prime), and the direction from one to the other gives the strand its built-in sense of which way is forward.
The names come from the numbered carbons of the sugar. One end of the strand has a free 5' carbon (with its phosphate) sticking out, unjoined to anything; the other end has a free 3' carbon (with its hydroxyl). Because every phosphodiester bond links a 3' carbon to the next 5' carbon, the whole strand has a consistent polarity, conventionally read and written 5'-to-3'. So a sequence like 5'-ATGC-3' means something specific; written backwards it would describe a different molecule.
Directionality is not a bookkeeping convention — the cell genuinely cares. DNA and RNA polymerases can only add new nucleotides to a 3' end, so strands are always built 5'-to-3', never the reverse. Ribosomes read messenger RNA 5'-to-3'. Knowing which end is which is essential for designing the short primers used in PCR and for reading any sequence correctly; flip the direction and you have the wrong molecule.
When ordering a PCR primer, you write its sequence 5'-to-3' by convention; order it backwards and it will anneal to the wrong place, or not at all — a classic, costly beginner mistake.
5' is the head, 3' is the tail — and the cell reads only one way.
The prime marks (5', 3') refer to the sugar's carbons, not the bases — the base carbons are numbered without primes. The prime is what keeps the sugar's '5' distinct from any '5' on the base.