This is actually REALLY unlikely since the British had also broken JN-25, using an entirely different method than the one used by the U.S. and considering the fact that the U.S. had been breaking Japanese codes since the 1920s.
A POD is easy.
Pick the date your want a POD. Japan (or Germany if late enough) gets a spy into the US intelligence situation. Japan gets indisputable evidence that their codes are broken, so they change them and otherwise improve radio procedures. These changes also counter the British code breaking method. Many a brilliant espionage setup has been destroyed by a double agent or just another spy that brings back enough information for the bad guys to figure out what happens. It only takes one spy, one stolen sheet of paper, or one giveaway communication to change history. Some idea from POD would include:
1) Some source claimed that the US Navy believed the phone lines to Pearl Harbor might be compromised. Maybe its true, maybe not. But if somewhere between the Department of the Navy and the Pearl there is a spy or wire tap. The Japanese Navy is getting the messages going to Pearl. There may or may not be enough in OTL for the Japanese to figure it out, but say add some very specific warning such as the time the Japanese would deliver the message to the Secretary of State, and the game is up.
2) We had code breakers in the Philippines and Hawaii. Someone use a female agent to turn one of the few hundred people working in the program. Or maybe one of the Philippines intel guys is capture and breaks under torture.
3) We do something too obvious. For example, imagine the Japanese had a good source of intel on just the troops on Midway. As the battle date approaches, Yamamoto receives information of all the units being moved to Midway, and guess correctly that the codes are broken. He then makes changes to the sensitive information, but uses the old codes to setup a trap for the USA. When the USA sends ships into the trap, he knows it is his command codes.
4) Someone on a ship who knows is capture. How many people in the US Navy knew that the codes were broken or had a strong suspicion? Were any of these individuals ever in a war zone on land or ship? Even the most brave admirals can break under torture. I am not familiar enough with this item to give possible examples, but it would seem likely that someone such as the admiral at the Coral Sea knew he was working on broken Japanese codes. Or if never told, at least suspected that is where his hot tips came from. I have read accounts that submarines were give information that was "always correct" by a special code group that told them it was "always" correct. Only the captain knew what these letters translated to, and he was not told the sources. But some claimed after the war they had a pretty good guess it was broken codes. A capture submarine captain is real easy to imagine, and combined that with say a captured log book, a little torture, and the Japanese might easily figure out that the only way the sub could have been sent to point X to sink capital ship Y was broken codes, specifically their daily position report.
The Russians got the A-bomb secret with multiple agents, so USA could make intel mistakes.
Now like most of these, when did the Japanese fix their codes, the exact date changes the outcome, but it is easy to say a few battles go worse for the USA, maybe even much worse. If we say January 1, 1942, then the coral sea does not happen. SW Pacific fights the Japanese in/from Australia.
It is hard to say, but I think the Midway battle goes the way the Japanese plan. The achieve surprise, and likely take Midway. Then the US fleet reacts, and their is a 4 on 6 carrier battle, that either side could win. There is really a lot of luck in carrier battles of this time frame, and in OTL, something as simple as a different search pattern by the Japanese would greatly change the battle.