The transcription factor RFX3 directs nodal cilium development and left-right asymmetry specification

Abstract

There are five members of the RFX family of transcription factors in mammals. While RFX5 plays a well-defined role in the immune system, the functions of RFX1 to RFX4 remain largely unknown. We have generated mice with a deletion of the Rfx3 gene. RFX3-deficient mice exhibit frequent left-right (LR) asymmetry defects leading to a high rate of embryonic lethality and situs inversus in surviving adults. In vertebrates, specification of the LR body axis is controlled by monocilia in the embryonic node, and defects in nodal cilia consequently result in abnormal LR patterning. Consistent with this, Rfx3 is expressed in ciliated cells of the node and RFX3-deficient mice exhibit a pronounced defect in nodal cilia. In contrast to the case for wild-type embryos, for which we document for the first time a twofold increase in the length of nodal cilia during development, the cilia are present but remain markedly stunted in mutant embryos. Finally, we show that RFX3 regulates the expression of D2lic, the mouse orthologue of a Caenorhabditis elegans gene that is implicated in intraflagellar transport, a process required for the assembly and maintenance of cilia. In conclusion, RFX3 is essential for the differentiation of nodal monocilia and hence for LR body axis determination.

FIG. 1.

Generation of Rfx3-deficient mice. (A) Schematic map of the mouse Rfx3 gene, the targeting vector, and the different targeted alleles. FRT and Lox-P sites are represented by gray and black triangles, respectively. The orientation of the neoR gene is indicated by an arrow. HindIII (H), BamHI (B) and PstI (P) sites, the 5′ and internal probes used for Southern blot analysis, expected fragment sizes for the B-P and H digests, and PCR primers used for genotyping (*) are indicated. A restriction fragment length polymorphism (#) was observed for the B-P digests between the wild-type alleles from the 129 (12 kb) and C57BL/6 (17 kb) backgrounds. (B) Southern blot analysis of the Rfx3^flox,neoR and Rfx3^neoR alleles; B-P-digested genomic DNA was hybridized with the 5′ probe. Southern blot analysis of the Rfx3⁻ and Rfx3^flox alleles is not shown. (C) Southern blot analysis of the Rfx3^flox,neoR and Rfx3^neoR alleles; H-digested genomic DNA was hybridized with the internal probe. Southern blot analysis of the Rfx3⁻ and Rfx3^flox alleles is not shown. (D) Genotyping by PCR of genomic DNAs isolated from tail biopsies. The results show the amplification products obtained for mice carrying the wild-type (500 bp), Rfx3^flox,neoR (2 kb), and Rfx3^neoR (1.5 kb) alleles. Results for mice carrying the Rfx3^flox (700 bp) and Rfx3⁻ alleles (200 bp) are not shown.

FIG. 2.

Rfx3-deficient mice exhibit growth retardation. (A) Body weights of homozygous Rfx3-deficient (Rfx3^−/−) and control (Rfx3^+/− and Rfx3^+/+) littermates compared for males or females at different ages, ranging from newborn pups to adults. At least five individuals of each genotype were weighed at each stage except for the point for mutant males at day 30, for which only one male remained alive. Representative Rfx3 deficient (−/−) and control (+/−) newborn pups (B) and adults (C) are shown to illustrate the differences in size.

FIG. 3.

Rfx3-deficient mice show characteristic LR asymmetry defects. (A and B) Dissections of 3-week-old control (A) and Rfx3-deficient (B) mice. The mutant mouse had a completely reversed orientation of the visceral organs (situs inversus). (C and D) Dissections of 18-day-old control (C) and Rfx3-deficient (D) embryos. The mutant mouse had a left pulmonary isomerism. (E to H) Transverse histological sections of 18-day-old control (E and G) and Rfx3-deficient (F and H) embryos showing various asymmetry defects, including left pulmonary isomerism, dextrocardia, and a central stomach. (I and J) Orientation of cardiac looping and embryo turning in 10.5-day-old control (left side view) (J) and mutant (right side view) (I) embryos. The mutant embryo had an inversion of cardiac looping and embryonic turning. The tail bud was removed to improve visualization of the heart. (K and L) In situ hybridization of Rfx3 knockout embryos with nodal and lefty probes. The latter probe detects both lefta and leftb expression, in the prospective floor plate and lateral plate mesoderm, respectively. Abnormal bilateral expression of nodal and lefta/b was observed in the Rfx3-deficient embryos (dorsal view). R1 to -4, right pulmonary lobes 1 to 4; L1 to -4, left pulmonary lobes 1 to 4; DC, dextrocardia; LC, levocardia; S, stomach; Li, liver; Ki, kidney; L, left; R, right; lv, left ventricle; mlv, morphological left ventricle.

FIG. 4.

Rfx3 is expressed in the embryonic node and regulates ciliary growth. The localization of Rfx3 mRNA (A) and protein (B) was examined in the embryonic node at E7.5. (C) Changes in the lengths of node monocilia between 7 and 8 days of embryonic development in control and mutant embryos. Theiler stage 10c, late streak, early bud; Theiler stage 11a, early neural plate; Theiler stage 11b, late neural plate; Theiler stage 11c/11d, early to late head fold; Theiler stage 12a, one to three somites. (D to O) Scanning electron micrographs of the embryonic nodes of wild-type (D, E, H, I, L, and M) and Rfx3-deficient (F, G, J, K, N, and O) embryos at three different stages. For each embryo, two magnifications are presented to show the position of the node and the nodal monociliated cells. At late streak stages (D to G), the node monocilia were of similar sizes (average, 1.6 μm) in control (D and E [n = 6]) and Rfx3-deficient (F and G [n = 4]) embryos. By the late neural plate stage (H to K), the cilia were longer in control embryos (average, 2.7 μm; n = 3) (H and I) but had not grown in Rfx3-deficient embryos (average size, 1.6 μm; n = 3) (J and K). This difference was even greater at stage 11c/11d, when cilia in control embryos had an average size of 3.3 μm (L and M [n = 3]), whereas cilia in the Rfx3-deficient embryos still had an average length of only 1.5 μm (N and O [n = 2]). The apparent morphological alteration of the mutant node in panel N is not significant as it was not observed for all mutant embryos. At late somite stages, the nodal monocilia in the mutants started to grow slightly but remained half as long as in the controls (data not shown). Bars, 100 μm (D, F, H, J, L, and N) and 2 μm (E, G, I, K, M, and O). n, node; nc, notochord; ps, primitive streak; R, rostral; C, caudal.

FIG. 5.

Rfx3 regulates D2liC expression in ciliated node cells and the adult brain. (A) D2liC and Tg737 mRNA expression was quantified by real-time RT-PCR for wild-type and Rfx3-deficient embryos at E7.5. Quantification was normalized with respect to two different control mRNAs, TBP and glyceraldehyde-3-phosphate dehydrogenase. The results are presented as the expression levels in mutant versus wild-type embryos. D2liC expression was significantly reduced in Rfx3-deficient embryos. Tg737 expression did not differ significantly between the wild-type and Rfx3-deficient embryos. The means and standard deviations from experiments performed with three to five independent embryos of each genotype are shown. (B) D2liC mRNA expression was quantified by real-time RT-PCR in brain RNAs prepared from wild-type and Rfx3-deficient adults. Quantification was normalized with respect to TBP mRNA. The results are presented as the expression levels in mutant versus wild-type embryos. The means and standard deviations from three experiments are shown. (C) D2LIC protein is expressed (green) in the monociliated embryonic node cells in wild-type and mutant embryos. Cilia are labeled (red) with antibodies against acetylated tubulin. Bottom panels, merge of D2LIC and acetylated tubulin staining.