The high frequency of phenylketonuria in Ireland and Western Scotland.

Abstract

Fewer than 5% of individuals with untreated typical ('classical') phenylketonuria (PKU) reproduce. Hence nearly two PKU genes are removed from the population pool per affected individual, which would be expected to lead to a decline in the incidence of the disease towards an equilibrium level maintained by mutation at between one in 60000 and one in 250000 live births (Woolf 1976). In Ireland and Western Scotland the incidence of PKU is about one in 6000 live births (Lindsay, 1970; Cahalane, 1976). The calculated rate of decline of the gene frequency makes it highly improbable (/9 < 0.00003 ~, where n = number of generations) that this high incidence results from chance (i.e. genetic drift) or the introduction of a fbw PKU genes into a population by, say, invaders or immigrants (Woolf, 1976). The known history of Ireland excludes the founder or bottleneck effect. Since mutation cannot account fbr the high PKU gene frequency, the sole remaining explanation is that the heterozygote must possess some advantage over the normal individual (Woolfet al., 1975); a 1.28% advantage (i.e. 1.28% more offspring) would maintain the present high gene frequency (Woolfand Goodwin, 1967). In a search for the nature of the heterozygote advantage, the obstetric-histories of 112 families with PKU children in Ireland and Western Scotland were compared with those of 863 matched control families. The average numbers of pregnancies were very nearly equal after correcting for incomplete ascertainment, an average of 3.09 pregnancies per PKU family and 3.1 pregnancies per control family (the uncorrected figure for PKU families would be 3.76). However in the PKU families 88.2% of pregnancies led to liveborn infants; in the control families there were 82.1% livebirths (both values corrected, Table 1). This difference is significant (p < 0.01). There appears to be some local hazard to fetal life against which maternal heterozygosity for PKU is protective. The only known biochemical or physiological difl);rence between PKU heterozygotes and normal individuals is a somewhat higher concentration ofphenylalanine in the blood and tissues in the former (Knox and Messinger, 1958), the fasting blood concentrations being 1.333_+0.23 and 0.818 _+ 0.19 mg/100 ml respectively (Woolf et al., 1967); this difference is a consequence of the smaller amount of phenylalanine hydroxylase in the liver of heterozygotes. We may hypothesize that the higher concentration of phenylalanine in the maternal circulation protects the fetus against some environmental hazard. It may be significant that Tocci and Beber (1973) found an association between congenital malformations in infants and relatively low blood phenylalanine concentrations in their mothers after a loading dose. The reduction of fetal mortality in PKU families constitutes a heterozygote advantage of magnitude 5.04°~, significantly greater than the 1.28% required for genetic equilibrium (p < 0.01). Hence the PKU gene frequency and the incidence of the disease may be rising in Ireland and Western Scotland; the rates of rise can be calculated from the equation in Table 2. I f the heterozygote advantage remains at 5.04%, the incidence will double to one in 3000 livebirths in from nine generations, say 240 years, i fPKU homozygotes are untreated, to seven generations, if all PKU homozygotes are treated and have the average number of offspring (Figure t). Halving the heterozygote advantage would extend the doubling period to about 20 generations; if the heterozygote advantage were reduced to 1.2% it would take about 90 generations (say 2500 years) to increase the