877. Evidence for Dominant-Negative Interference in the Pahenu2 Mouse Model of PKU

Abstract

Top of pageAbstract Phenylketonuria (PKU) is a metabolic disease caused by the deficiency of the enzyme phenylalanine hydroxylase (PAH). PAH is the obligate enzyme in the conversion of phenylalanine (phe) to tyrosine (tyr). Even though a successful treatment modality has been in place since the early 1970s, its success has led to an increase in maternal PKU syndrome. The syndrome results in a serious increase in risks of birth defects to children born of PKU women, especially when the mothers' blood phe levels do not reach a normal range before or very early in pregnancy. A mouse model for PKU was created in 1993 by ENU mutagenesis of male germ cells. The induced missense mutation, F263S, is located in exon 7 of the gene and renders the enzyme inactive. The mouse has “classic” PKU with elevated blood phe levels, cognitive deficiencies, and it exhibits maternal PKU syndrome. We have successfully treated these mice using liver-directed gene therapy with AAV2 containing the mouse PAH gene driven by a Chicken-β-actin hybrid promoter. However unusually high vector doses were needed to achieve normal blood phe levels (see abstract by Laipis et. al.). Message levels were measured in the mice by Northern blot: while PAH is very abundant in the liver, no differences were detected between wild type, heterozygote and null mice. At the protein level, wild type mice have two times as much protein as heterozygote and null mice. When considering that the PAH enzyme is a tetramer, or a dimer of dimers, it is possible to explain the heterozygote protein levels by assuming simple unstability of the missense protein. However, null mice show significant protein levels, about equivalent to heterozygotes, thus the F263S protein cannot be completely unstable. We hypothesize that the missense protein could tetramerize with the introduced protein during gene therapy in a dominant-negative fashion, reducing the total potential activity and thus the effectiveness of the treatment. To investigate this hypothesis, we performed calcium phosphate transfections on HEK-293 cells using plasmid DNA under the CBA promoter containing either the normal mPAH or mPAH-F263S gene. When mixing a one to one ratio of mPAH to mPAH-F263S, the PAH specific activity in the cell lysate drops to one half that observed with only mPAH. The reduction in PAH activity strongly suggests that the monomers can interact. A ribozyme was previously designed to cleave the mouse PAH mRNA. The ribozyme was tested in vitro and then cloned into an AAV vector. Simultaneously a ribozyme-resistant form of mPAH was cloned with strategically placed silent mutations (named mPAH-HdRz). In cell transfection experiments, the one to one combination of mPAH-HdRz and mPAH-F263S also drops to approximately one half the activity seen in mPAH-HdRz alone. When the ribozyme is added to the same combination, PAH activity is restored to near normal levels. Since the ribozyme has been shown not to cleave the mPAH-HdRz mRNA, we conclude that the combination of mPAH and mPAH-F263S proteins cause the observed reductions in total PAH activity. Dominant-negative interference is thus the most likely cause of the high vector doses needed to cure the Pahenu2 mice. Further work is underway to confirm the hypothesis in vivo.