Tolerance to rat monoclonal antibodies. Implications for serotherapy