Maternal heterozygosity of Slc6a19 causes metabolic perturbation and congenital NAD deficiency disorder in mice

ABSTRACT Nicotinamide adenine dinucleotide (NAD) is a key metabolite synthesised from vitamin B3 or tryptophan. Disruption of genes encoding NAD synthesis enzymes reduces NAD levels and causes congenital NAD deficiency disorder (CNDD), characterised by multiple congenital malformations. SLC6A19 (encoding B0AT1, a neutral amino acid transporter), represents the main transporter for free tryptophan in the intestine and kidney. Here, we tested whether Slc6a19 heterozygosity in mice limits the tryptophan available for NAD synthesis during pregnancy and causes adverse pregnancy outcomes. Pregnant Slc6a19+/− mice were fed diets depleted of vitamin B3, so that tryptophan was the source of NAD during gestation. This perturbed the NAD metabolome in pregnant Slc6a19+/− females, resulting in reduced NAD levels and increased rates of embryo loss. Surviving embryos were small and exhibited specific combinations of CNDD-associated malformations. Our results show that genes not directly involved in NAD synthesis can affect NAD metabolism and cause CNDD. They also suggest that human female carriers of a SLC6A19 loss-of-function allele might be susceptible to adverse pregnancy outcomes unless sufficient NAD precursor amounts are available during gestation. This article has an associated First Person interview with the first author of the paper.

. The placentas of affected embryos at E18.5 have normal structure and vascularisation. (A) Summary of embryo phenotypic outcomes at E18.5 from wild-type x wild-type and Slc6a19 +/− x Slc6a19 +/− timed mating and dietary treatments for placental assessments. The maternal diet given throughout gestation was either not restricted in NAD precursors and contained 30.0 mg/kg of vitamin B3 and 1.8 g/kg of tryptophan, corresponding to 80% of the NAD precursor content of the standard diet (SF22-012; Specialty Feeds, Glen Forrest, Australia) or restricted in NAD precursors (NTF+TW600, Disease Models & Mechanisms: doi:10.1242/dmm.049647: Supplementary information see Table S1). Female wild-type mice were given SF22-012 diet for at least 3 weeks prior to the start of gestation. Because embryos of Slc6a19 +/− mothers pretreated with SF22-012 diet and given NTF +TW600 throughout gestation were not viable, due to their heightened sensitivity towards NAD precursor restriction, another set of Slc6a19 +/− females was given the breeding diet (see Table S1) prior to gestation. NTF, NAD precursor vitamin-depleted and tryptophan-free feed; TW, tryptophansupplemented water. M, maternal Slc6a19 genotype; P, paternal Slc6a19 genotype. Small embryo size was defined as weighing less that an average E17.5 embryo from a wild-type mother on the breeding diet. (B) Graphical representation of the embryo survival and malformation incidences. (C) Weights of live E18.5 embryos summarised in (A) and (B). Dissected embryos were weighed prior to assessment for congenital malformations. Bars indicate mean ± s.d.. Significance of differences relative to the wild-type SF22-012 diet control group (first column) was determined by one-way ANOVA with Tukey's multiple comparisons test with ****P<0.0001 and **P<0.01. Dotted lines indicate average weights of embryos from wild-type mothers fed the breeding diet and collected at E16.5, E17.5, and E18.5, respectively. (D-K) Assessment of placentas from E18.5 embryos. Multiple sections of collected placentas were processed by haematoxylin/eosin (H&E) (O'Reilly et al., 2014) and Isolectin GS-IB4 (IB4) staining. For IB4 staining, the sections were deparaffined in the same way as for H&E, and incubated 2× 5 min in PBS+0.1% TWEEN, blocked for 1 h in 1%BSA+10% horse serum at room temperature, then incubated with Isolectin GS-IB4, Alexa Fluor TM 488 Conjugate (ThermoFisher Cat #I21411, 1:250 dilution) at 4°C overnight. The IB4 reagent has been validated previously (Moreau et al., 2014). Images of IB4-stained placental sections were acquired by fluorescence microscopy at 2.5× magnification using excitation/emission parameters for GFP. H&Estained placental sections were assessed by light microscopy and images acquired at 2.5× magnification. The placentas of neither malformed embryos (H, I) nor small embryos (J, K) exhibited any obvious difference in gross structure or vascularisation beyond normal variability seen in placentas of normal embryos from wild-type x wild-type matings (D, E, F) or Slc6a19 +/− x Slc6a19 +/− matings (G).

Fig. S5. Maternal dietary NAD restriction during pregnancy causes varying degrees of
developmental delay that is observable at E9.5. Embryo outcomes at E9.5 with (A) a Slc6a19 +/x Slc6a19 +/mating scheme or (B) a wild-type x wild-type mating scheme and different maternal dietary conditions. Maternal diets from the start of pregnancy are indicated on the left. Each horizontal bar represents a litter and length of the bars indicates the number of embryos per litter.
Total counts and percentages of embryos within each treatment group are summarized on the right. The 'delayed' group was defined as embryos phenotypically E9 or younger. WT, wild-type. This dataset is related to Table 1. Part of this dataset is from a previous study (Cuny et al., 2020).  group includes embryos with visible malformations such as short, curly, or kinked tails (caudal agenesis), and underdeveloped eyes. WT, wild-type. This dataset is related to Table 1. Part of this dataset is from a previous study (Cuny et al., 2020).  Significance of differences relative to the pregnant wild-type standard diet group (first column) was assessed by one-way ANOVA with Tukey's multiple comparisons test with *P<0.05 and ns = not significant. See Table S6 for numerical values and statistics.  Plasma metabolite concentrations at E9.5 (A, C) and at E11.5 (B, D). Colours indicate the litter survival status per mother at dissection, with green: all embryos alive, orange: one or more embryos dead, black: all embryos dead. The maternal Slc6a19 genotype and diet are indicated underneath each graph. Trends between litter survival and maternal metabolite levels were assessed by twoway ANOVA, comparing the 3 different litter outcomes across all 4 different maternal diet/genotype conditions. P values are indicated within the graphs. This dataset is related to Table 4 Table 4 for numerical values and additional statistics.  Table S7 for numerical values and additional statistics.   Table S7 for numerical values and additional statistics.  maternal Slc6a19 genotype and diet are indicated underneath each graph. Significance of differences in metabolite concentration relative to the pregnant wild-type standard diet group (first column) was assessed by one-way ANOVA with Tukey's multiple comparisons test with ****P<0.0001 and ns = not significant. See Table S7 for numerical values and additional statistics     Amounts of NAD precursors (=niacin equivalents) and tryptophan per day are based on an average consumption of 3.9 g food and 6.2 mL water per day and on the approximation that 60 mg of dietary tryptophan is equivalent to 1 mg of vitamin B3 for the conversion to NAD (Bachmanov et al., 2002;Goldsmith, 1958). 2 The percentages refer to the NAD precursor and tryptophan supply, respectively, relative to the standard diet (first row). TRP, tryptophan; NA, nicotinic acid; NTF, NAD precursor vitamin-depleted and tryptophan-free feed; TW, tryptophan-supplemented water; NW, nicotinic acid-supplemented water. Lowercase letters indicate which rows were compared. The rows also match the rows in Table 1 in the main manuscript. P values were calculated by two-sided Fisher's exact test (with Freeman-Halton extension) when comparing 2 or 3 groups, or χ 2 test when comparing multiple groups.
2 Part of this dataset is from a previous study (Cuny et al., 2020). M, maternal Slc6a19 genotype; P, paternal Slc6a19 genotype. For the numbers and percentages of embryos for each dietary condition, see Table 1.   Polydactyly -6 (7%) 33 (51%) ---2 (12%) - For representative images of several commonly occurring malformations, see Fig. S1. Uppercase letters M and P designate maternal and paternal Slc6a19 genotype, respectively. Percentages summarise defects observed in live embryos only. Malformation "Total" refers to the total number of affected embryos for the given "Location", irrespective of the cumulative number of malformations.

1
Dataset is from a previous study (Cuny et al., 2020). 2 C1 and C2 were occasionally removed during dissection. Therefore, isolated defects in the vertebrae C1 and/or C2 were not counted. Ossification point abnormalities (e.g. flattened shape, less compaction, dumbbell shape, smaller than usual, two separate ossification points) were not counted, as they can result from delayed ossification and are not necessarily leading to a skeletal defect (Carney and Kimmel, 2007). Vertebral abnormalities observed and counted as congenital malformations included vertebral fusions, butterfly vertebrae, and hemivertebrae. genotypes from Slc6a19 +/x Slc6a19 +/matings. 2 Part of this dataset is from a previous study (Cuny et al., 2020).
For the embryo numbers and NAD levels for each dietary condition, see Table 2.