Lock-and-mint: synthetic representation of a locked asset
The most common bridge design is lock-and-mint. A user sends an asset to a contract on the source chain, and that contract locks it, simply holding it in custody rather than destroying or moving it anywhere. Once the lock is confirmed, a corresponding amount of a new, wrapped token is minted on the destination chain and credited to the user. This wrapped token is not the original asset, it is an IOU-like claim that a bridge contract, somewhere, is holding the real asset on the user's behalf. To move value back, the user burns the wrapped token on the destination chain, which triggers the source-chain contract to unlock and release the original asset.
The appeal of lock-and-mint is that it works for any asset pair without requiring deep, pre-existing liquidity on the destination chain, since the destination-side supply is created on demand by minting. The cost is that it concentrates enormous value in the source-chain lock contract, which becomes a single, highly visible target holding the collateral backing every wrapped token in circulation. If that lock contract or its minting authorization logic is ever compromised, the wrapped tokens on the destination chain instantly become unbacked claims with nothing behind them, even though nothing was ever technically wrong with either underlying blockchain.