A substantial proportion of imprinted genes, i.e., genes expressed from only one parental chromosome, are involved in placental development and fetal growth in mammals.  In the mouse for example, Igf2 is expressed paternally in the placenta and fetus, while its receptor is expressed maternally.  Imprinted genes can act directly on the fetus by influencing cellular proliferation and apoptosis; they can also affect fetal growth by influencing placental structure and physiology and the supply of maternal nutrients.  Debate over the evolutionary significance of imprinting in mammals has led to the so-called genetic conflict hypothesis or theory of imprinting.  It predicts that paternally expressed genes act on the placenta to promote extraction of resources from the mother to enhance fetal growth while maternally expressed genes act to restrain fetal growth to conserve maternal resources for long-term reproductive fitness of the mother.  Testing this hypothesis has been difficult because the relevant genes are expressed in both placenta and fetus and their tissue-specific inactivation has not been achieved.

Recently, it has been shown that the mouse Igf2 has four promoters, one of which, designated P0, directs paternal expression of Igf2 in the labyrinthine trophoblasts of the placenta.  Deleting this promoter through gene targeting enabled Constância and colleagues to study the impact of paternally-directed placental IGF-II on fetal growth.  The P0 knockout for Igf2 was confirmed by in situ hybridization that revealed a marked reduction of Igf2 expression specifically in the labyrinthine trophoblasts.  Expression of Igf2 from its other promoters was normal in mutant placentas and fetal tissues as were levels of IGF-II in the fetal circulation.

Lack of the P0 Igf2 transcripts with paternal transmission primarily resulted in placental growth restriction, which was detected early in gestation at embryonic day 12 (E12) of the 19-day mouse gestation.  The impaired growth of the mutant placentas remained relatively constant throughout the remainder of the pregnancy (weight of mutant placentas 76%, 82%, 68%, 68% of normal at E12, E14, E16, E18, respectively) suggesting that the paternally-directed, labyrinthine trophoblast-specific Igf2 transcripts are required to sustain normal growth of the placenta.

In contrast to the early decrease in placenta size, the indirectly affected fetuses became growth restricted only toward the end of gestation.  Their weight was 96% of normal at E16, but dropped to about 70% at birth.  The ratio of fetal to placental weight increased as gestation proceeded and was significantly higher for mutant compared to normal pregnancies reflecting the small placenta size. 

To address the discrepancy between placental and fetal growth, the authors compared normal and mutant placentas structurally and functionally.  Other than size, no obvious differences in tissue organization or cell morphology were detected.  They next compared maternal-fetal transport of different radiolabelled compounds, one transferred by passive diffusion and the other by active transport.  Their results showed that passive diffusion declines proportionate to the relative reduction in placental size.  Active or system A transport, however, increases during mid gestation, apparently compensating for the loss of passive transfer until near the end of gestation when this compensation is insufficient to meet the needs of the fetus and fetal growth drops off.  Importantly, the system A transporter has been shown to be a determinant of fetal growth.

In summary, deletion of a placental-specific imprinted transcript results in fetal growth restriction, primarily through a decrease in total nutrient transfer across the placenta.  This example of a morphologically normal but small placenta affecting fetal growth supports the genetic conflict theory of imprinting, in which a placental-specific gene expressed from the paternal allele regulates the supply of nutritional resources to the fetus.  On the other hand, fetal demand for nutrients is genetically regulated by the level of growth factors such as IGF-I and IGF-II.  Increasing fetal size therefore requires a higher level of demand (for example, higher fetal IGF-II) as well as a higher level of supply (by increasing, for example, placental surface area).  Reduced fetal size can be the outcome of reduced supply (as in the P0 mutant described here) or of reduced demand (for example Igf1 knockout, which reduces fetal but not placental size).  The mouse Igf2 gene is remarkable in combining the control of both the supply and the genetic demand for maternal nutrients in a single gene.

First Editor’s Comment:  This work supports the genetic conflict theory of imprinting showing that placental-specific genes expressed from the paternal allele contribute substantially to the supply of nutrients a fetus receives from its mother.  It also shows that the placenta can partially compensate at least for the loss of this paternal effect.  It will be interesting to learn more about the nature of the compensation, which represents a potential mechanism to exploit in treating intrauterine growth retardation.  It is important to acknowledge, that the relationship between mother and fetus differs substantially between mice and humans, especially with regard to size and duration.

William A. Horton, MD

Second Editor’s Comment:  As a pediatric endocrinologist who has had a special interest in IUGR for many years, I found the reading of the original article most informative.  Not mentioned in the abstract or First Editorial comment was the following brief statement, “At birth, P0 mutant pups were 69% of normal birth weight.  This was followed by postnatal catch-up growth which was complete by three months of age.”  While, as Dr. Horton stated above that mice and humans (may) differ substantially, there is a corollary between the catch up growth in these IUGR mice and the catch up growth that is seen in most IUGR human neonates (primarily those without associated dysmorphology) in the first two years of life.  Subsequent studies dealing with the genetic conflict theory in humans should be very informative and intriguing.

Robert M. Blizzard, MD