Vitamin D deficiency during various stages of pregnancy in the rat; its impact on development and behaviour in adult offspring

Summary

Developmental vitamin D (DVD) deficiency alters brain development and behaviour in the rat. The aim of this study was to vary levels of vitamin D deficiency during gestation and examine the effects on developmental milestones and behaviour in adult offspring. By manipulating the withdrawal and reintroduction of vitamin D in the diet of female Sprague–Dawley rats, their offspring were subjected to four different prenatal vitamin D conditions: (a) control (normal vitamin D throughout gestation); (b) early-DVD deficiency; (c) late-DVD deficiency; and (d) full-DVD deficiency. We show that the standard measure for vitamin D status, 25(OH)D₃, can be significantly manipulated within 7 days by dietary intervention. We also show that levels of the active form of this vitamin, 1,25(OH)₂D₃, replete within the same time frame as 25(OH)D₃ but are slower to deplete. Developmental milestones remained normal across all four dietary groups. Concerning the adult behavioural phenotype, both full- and late-DVD deficiency were associated with MK-801-induced hyperlocomotion. Overall, these data suggest that vitamin D deficiency restricted to late gestation only is sufficient to disrupt adult brain functioning in the rat. These findings suggest there may be a therapeutic window for maternal dietary intervention in the DVD model of psychosis.

Introduction

There is increasing evidence that low prenatal vitamin D may be associated with a range of adverse health outcomes (Hollis and Wagner, 2006a, Hollis and Wagner, 2006b). While the links between prenatal vitamin D and later growth and bone health are in keeping with the traditional role of vitamin D (Javaid et al., 2006; Morley et al., 2006), there is now a growing body of evidence linking low prenatal vitamin D to adverse neuropsychiatric outcomes. For example, epidemiological evidence indicates that prenatal hypovitaminosis D is a candidate risk factor for schizophrenia (McGrath, 1999), certain types of brain tumour (Ko et al., 2005), and multiple sclerosis (Willer et al., 2005). Because hypovitaminosis D is more prevalent in winter and spring (Holick, 1995), it has been hypothesised that low prenatal vitamin D may be the biological mechanism behind many of the disorders that have a season of birth effect (McGrath, 1999, McGrath, 2001).

In an attempt to establish the biological plausibility of this hypothesis a model of developmental vitamin D (DVD) deficiency has been developed in rodents. Changes in DVD deficient rats include subtle alterations in brain morphology (e.g., significantly enlarged lateral ventricles), increased cell proliferation, reduced expression of nerve growth factor (NGF), glial derived neurotrophic factor (GDNF) and the low-affinity neurotrophin receptor p75^NTR (Eyles et al., 2003). Furthermore as adults, rats exposed to DVD deficiency have shown persisting brain changes, including larger lateral ventricles, reduced NGF protein content and abnormal expression of genes involved in cytoskeleton maintenance (MAP2, NF-L) and neurotransmission (GABA-Aα4) (Feron et al., 2005).

DVD deficiency results in long lasting changes in the behaviour of adult rats. For example, DVD rats show altered locomotion in the open field (Burne et al., 2004, Burne et al., 2006; Eyles et al., 2006) and disrupted latent inhibition (Becker et al., 2005). In particular, DVD deficient animals are particularly sensitive to the locomotor inducing effects of the N-methyl-D-aspartate receptor (NMDA-R) antagonist, MK-801 (Kesby et al., 2006). MK-801-induced hyperlocomotion is a key feature in several different animal models of schizophrenia (Al-Amin et al., 2000; Daenen et al., 2003; Tiedtke et al., 1990). The psychotomimetic effects of MK-801 in non-psychotic individuals closely resembles symptoms of schizophrenia and enhanced sensitivity to NMDA antagonists suggests dysregulation of (at least) dopamingeric and glutaminergic pathways (Olney et al., 1999). Thus, MK-801-induced hyperlocomotion can be used as a ‘biobehavioural marker’ of one aspect of how low prenatal vitamin D alters brain development.

Whilst the animal evidence linking low prenatal vitamin D to altered brain development is now convincing, it is not clear how findings from a profoundly vitamin D deficient maternal rat relate to human pregnancy. For instance we have no idea what prenatal developmental period is most vulnerable to low levels of this vitamin. Also increased foetal requirements for vitamin D (related to increased calcium transfer during the third trimester) lead to a fall in maternal vitamin D levels especially in winter months (MacLennan et al., 1980; Markestad et al., 1984). Therefore, the final trimester may be the most relevant for vitamin D's actions in the developing brain. Based on the evidence linking winter/spring birth with an increased risk of various adverse neuropsychiatric diseases (Davies et al., 2003; Torrey et al., 2000), and the known association between maternal vitamin D levels and neonatal growth (Javaid et al., 2006; Morley et al., 2006) we speculated that the last trimester may be the period in humans most vulnerable to low prenatal vitamin D.

Gestation in the rat (21 days) is short. Also the functional half-life of the major form of this vitamin, 25(OH)D₃ although unknown in rats is about three weeks in non-pregnant humans (Davies et al., 1997). These factors combine to provide limited options for rapid manipulation of vitamin D levels through the diet. However, several patterns of gestational hypovitaminosis D can be readily modelled by manipulating the maternal diet. For example, the addition of vitamin D to vitamin D deplete dams at conception should result in low levels of 25(OH)D₃ in the early phases of gestation, rising across gestation. Offspring in this group should only be exposed to early-gestational hypovitaminosis D. Conversely, the removal of vitamin D from the maternal diet at conception should result in falling levels of 25(OH)D₃ across gestation. Offspring of this group should be exposed to normal vitamin D levels during early gestation, which fall across the gestational period, leaving the foetus exposed to late-gestational hypovitaminosis D.

The overall aim of this study was to examine the impact of different maternal vitamin D dietary manipulations on selected biochemical, developmental and behavioural outcomes in the DVD rodent model. First, we wished to characterize the influence of manipulating maternal vitamin D status on blood concentrations of both the major circulatory form of this vitamin, 25(OH)D₃; the biologically active form of this vitamin_, 1,25(OH)₂D₃; parathyroid hormone (PTH); calcium and phosphate levels in serum during gestation and in pups at birth. Because prenatal exposures can be associated with transient developmental delays (i.e. early neurological signs that normalise across later development), we also wished to examine development milestones in the offspring at post-natal day (P) 7, 14 and 21. Finally, whilst it is known that full-DVD deficiency is associated with increased behavioural sensitivity to the psychomimetic agent MK-801 in the adult offspring (Kesby et al., 2006) we wished to establish more precisely the gestational period which is most relevant to the onset of this behaviour. We anticipate that these experiments will clarify whether any such critical gestational window for the adverse effects of DVD deficiency exists.