American researchers said that the laboratory tests had shown that the effects of the disease could be reversed.
But scientists said they need more cash to get the drug into clinical trials.
About 1 million Americans have type 1 diabetes, also known as juvenile diabetes because it is often diagnosed in children.
Cells destroyed
In type 1 diabetes, the body does not produce insulin, which is needed to convert sugar into fuel and is normally produced in the pancreas in cells called islet cells.
In people with type 1 diabetes, the islet cells are destroyed by the body's own misguided immune cells and sugar builds up dangerously in the blood.
Research, reported in the Journal of Clinical Investigation, showed it was possible to treat the mice for just 40 days and to re-train their immune systems so that they did not mistakenly attack the islet cells.
First, the researchers triggered a naturally occurring drug called TNF-alpha that killed the rogue immune cells.
Then they injected the mice with donor cells that re-educated their immune systems so that they would accept islet cells and not destroy them.
The insulin-producing cells appeared to re-grow in the diseased mice, providing enough so that their blood sugar came back to normal.
Study author Dr Denise Faustman, of Massachusetts, General Hospital said that up to 75% of the mice stayed healthy for 100 days or more, without any further treatment.
She said: "It's really a permanent reversal of an established autoimmune disease, and if that's not as good as it gets, the next thing that happens is the islets regenerate in the pancreas."
Funding
But Dr Faustman said the trials were being hampered by a lack of cash.
"If we had money to produce the drug and do the treatment, we'd be in clinical trials within a year."
Dr David Nathan, director of the diabetes centre at Massachusetts General Hospital, was cautiously optimistic.
He said: "This represents a very new discovery in an animal model of type 1 diabetes that has always been used to predict what happens in human diabetes.
"However until we explore more thoroughly whether the same phenomenon that occurs in mice occurs in human diabetes, we have to be cautious about the potential meaning of these findings ... whether this may represent the first insight in how to cure type 1 diabetes."
Mairi Benson, of Diabetes UK, said: "The research results arising from the work carried out at Massachusetts General Hospital are extremely interesting.
"There is still much to be understood, however, about the autoimmune process in human diabetes mellitus and Diabetes UK looks forward to seeing the results of this study being taken forward from work on mice to humans."