Rh System and Intrauterine Growth. Interaction with Season of Birth

Based on the hypothesis that maternal-fetal genetic differences in membrane transport and signal transduction may influence intrauterine development, the recent acquisition on transport function of Rh protein prompted us to study the relationship between joint maternal-fetal Rh phenotype and birth weight. Considering that metabolic effect of maternal-fetal competition could be amplified by environmental conditions, we have investigated possible seasonal effects on such relationship. We have studied 5291 infants born in Sardinia in the period January 1993–December 1996 and 984 infants born in Rome during 1996. In Rh(–) mothers there is a significant association between season of birth and birth weight that shows the highest mean value in infants born in autumn (i.e. conceived in winter). The association is much more evident in male than in female infants. In male infants from Rh(–) mothers, the association between birth weight and season is significant in Rh(+) male newborns only. Recent observations by our group in NIDDM suggest that glucose transport in RBC may be related to D protein, thus we propose an interpretation of the present observation in terms of transport function. When the density of D protein in the infant is greater than in the mother, the balance is in favour of the infant who may attain a significant developmental advantage when conceived in the cold season.


Introduction
Recent studies have shown that Rh locus is composed of two genes: RhD, which encodes the major D antigen and is present only in Rh-positive genomes, and RhCE, which encodes both Cc and Ee polipeptides, most likely by alternative splicing events. Several membrane components including the Rh proteins and other glycoproteins recently characterized are probably different subunits of an oligomeric complex with transport function in the erythrocyte [3,9,13]. Rh may play an important role in the organization of membrane phospholipids and/or membrane expression of various glycoproteins [1]. Variations of lipid bilayer, on the other hand, could influence the function of various proteins involved in substrate transport and signal transduction [5]. Indeed recent observations by our group suggest that genetic variability of Rh proteins may influence glucose transport in the erythrocyte [6].
Maternal-fetal genetic differences within the normal variability in membrane transport and signal transduction could contribute significantly to the normal variability of intrauterine developmental parameters [2]. Moreover, effects of maternal-fetal competition connected with differences in relative efficiency of substrate transport and/or signal transduction could be amplified by environmental conditions including seasonal variation [7,8].
In the present note we have searched for possible effects of joint mother-newborn Rh phenotype on intrauterine growth in two samples of consecutive infants from healthy puerperae.  Rh(D) blood group was determined according to standard laboratory method. Variance analysis was performed by SPSS programs [12]. The procedure was carried out on both log 10 of birth weight and on raw data.

Subjects and methods
Among Rh(−) mothers a direct Coombs test was present in 6 newborns in Sassari (Sardinia) and in 2 newborns in Rome. Excluding these subjects from analyses, the results in Table 1 and in Table 2 remain practically unchanged.
In few infants Rh determination has not been performed: this explains the fact that the number of male newborns from Rh(−) mothers is greater in Table 1 than in Table 2. Table 1 shows the weight of newborns in relation to season of birth and to Rh phenotype of the mother. In male infants from Rh(−) mothers there is a significant association between birth weight and season of birth. The association is significant after Bonferroni correction for multiple comparison. Variance analysis evaluates the effects of the two independent variables: season of birth (four categories) and population (two categories: Sassary and Rome) on birth weight (dependent variable). Overall (main effect) there is a significant effect on birth weight in males from Rh(−) mothers. This effect is due to season while the variable population has no significant effect and there is no significant interaction between the two independent variables, sea-son and population. In brief, the analysis shows that the effect of season on birth weight in male newborns from Rh(−) mothers is the same in Rome and Sassari.

Results
Tukey test has shown that birth weight is significantly lower in infants born in winter than in infants born in autumn. Such difference is 246 g in Sassari and 588 g in Rome. The mean value of difference (weighted for the different number of newborns in the two samples) is 327 g (p < 0.003; after correction for multiple comparison p < 0.02).
The strength of association among birth weight, Rh and season of birth seems slightly variable in different years. In the population of Sassari a higher birth weight in male infants born in autumn from Rh(−) mother has been observed in 1993, 1994 and 1995 but not in 1996 (data not shown).
An analysis of gestational duration in relation to season of birth and Rh phenotype of the mother has shown no significant association.
An analysis of variance and covariance has also been performed considering as dependent birth weight and as independent season of birth, population and gestational duration (covariate). Controlling for the effect of gestational duration, the association between birth weight and season of birth remains significant. Table 2 shows the weight of male newborns in relation to season of birth and Rh phenotype. The association between birth weight and season is significant in Rh(+) male newborns from Rh(−) mothers only.
Rh(D) distribution in mothers and infants (all infants and males and females separately) has not shown significant association with season of birth. Possible Table 2 Birth weight in relation to season of birth and Rh phenotype in male newborns. Values marked with asterisk refer to analysis on log 10 of birth weight  distortion of the joint mother-infant Rh distribution in relation to season of birth has been examined without significant results.

Discussion
Because of the clinical relevance of hemolytic disease of the newborn due to maternal-fetal incompatibility in Rh blood groups, studies on this system have been focused on immunological aspects. These studies eventually relieved the clinical burden but shadowed the biological problem of maintenance of a polymorphism showing a strong heterozygote disantvantage. It is likely that the hemolytic disease of the newborn is an accidental phenomenon largely compensated by a still unknown advantage connected with some important function of Rh system in the maternal-fetal relationship.
Recently, it has been shown that Rh is a transport protein. In addition, we have recently shown that in NIDDM the mean level of glycosilated Hb (HbA1c) is higher in Rh(+) than in Rh(−) suggesting that glucose transport in red blood cells may be related to Rh protein [6]. This prompted us to search for a possible relationship between Rh system and intrauterine development in a perspective of maternal-fetal difference in transport-function.
As pointed by Sing et al. [11], the first step in the analysis of genetic architecture of multifactorial variables including developmental parameters and common diseases is the estimate of the joint distribution of genetic and environmental causal agents. Moreover, when studying the effect of genetic and environmental factor on intrauterine development an obvious subdivision is that between males and females [11]. In fact is well known that these two classes show differences in the pattern of intrauterine development, endocrine functions and metabolism.
The present data from two unrelated populations show that in newborns from Rh(−) mothers the birth weight is higher in those conceived in the winter (born in autumn) than in those conceived in other seasons. The phenomenon is very clear in Rh(+) newborns conceived by Rh(−) mothers, suggesting an involvement of Rh system in the maternal-fetal metabolic relationship with selective effects opposite to that of Rh hemolytic disease.
Assuming that maternal-fetal genetic differences within the normal variability of receptors, membrane transport and signal tranduction could contribute significantly to the normal variability of intrauterine development [2,4], we propose a metabolic interpretation centered on maternal-fetal differences in transport rate depending on the joint Rh phenotype. If the density of D protein is greater in the infant than in the mother, the balance would be in favour of the infant who may take advantage for its development when he is conceived in the cold season (and born in autumn). Such effects are much more evident in males than in females.
At present, however, the possibility that the association is causally due to some gene near Rh and in linkage disequilibrium with it cannot be excluded. In fact, a number of genes with important metabolic functions are clustered on the short arm of chromosome 1. Starting from telomere there are the following loci: Glucose Dehydrogenase, 6-Phosphogluconate Dehydrogenase, (Rh cluster), UDP Galactose-6-Epimerase, Glucose Transporter Protein 1, Glucose Transporter Protein 5, Phosphoglucomutase-1, Amylase loci (AMY1A -salivary; AMY1B -salivary; AMY2A -pancreatic; AMY2B -pancreatic) [10].