Generation and characterization of an Nxf7 knockout mouse to study NXF5 deficiency in a patient with intellectual disability

Abstract

Members of the Nuclear eXport Factor (NXF) family are involved in the export of mRNA from the nucleus to the cytoplasm, or hypothesized to play a role in transport of cytoplasmic mRNA. We previously reported on the loss of NXF5 in a male patient with a syndromic form of intellectual disability. To study the functional role of NXF5 we identified the mouse counterpart. Based on synteny, mouse Nxf2 is the ortholog of human NXF5. However, we provide several lines of evidence that mouse Nxf7 is the actual functional equivalent of NXF5. Both Nxf7 and NXF5 are predominantly expressed in the brain, show cytoplasmic localization, and present as granules in neuronal dendrites suggesting a role in cytoplasmic mRNA metabolism in neurons. Nxf7 was primarily detected in the pyramidal cells of the hippocampus and in layer V of the cortex. Similar to human NXF2, mouse Nxf2 is highly expressed in testis and shows a nuclear localization. Interestingly, these findings point to a different evolutionary path for both NXF genes in human and mouse. We thus generated and validated Nxf7 knockout mice, which were fertile and did not present any gross anatomical or morphological abnormalities. Expression profiling in the hippocampus and the cortex did not reveal significant changes between wild-type and Nxf7 knockout mice. However, impaired spatial memory was observed in these KO mice when evaluated in the Morris water maze test. In conclusion, our findings provide strong evidence that mouse Nxf7 is the functional counterpart of human NXF5, which might play a critical role in mRNA metabolism in the brain.

Figure 1. Relative expression analysis of NXF/Nxf in tissues.

(A) RT-PCR for mRNA level quantification of Nxf1, Nxf2, Nxf3 and Nxf7 in mouse tissues. ActB was used as the housekeeping gene. (B) RT-qPCR for relative expression analysis of human NXF5 and NXF2 normalized to GUSB and HPRT1. (C) RT-qPCR for relative expression of mouse Nxf2 and Nxf7, normalized to Gusb and Hprt1. Cx, cortex; Hc, hippocampus; Cb, cerebellum; Te, testis; Br, Brain; Mu, muscle. N = 3. Means ± standard deviations. *p<0.05.

Figure 2. In situ hybridization of mouse Nxf2 and Nxf7.

Lateral view of E7.5 and E8.5 embryos and sections of E12.5 testis, hybridized with the Nxf2 (A,B,C) or Nxf7 (D-I) antisense probes. Nxf7 mRNA is highly expressed in the allantois (D,E,F,H), chorion, yolk sac and giant cells of the trophoblast (F,G). al, allantois; ys, yolk sac; tb, trophoblast; ch, chorion; am, amnion; ec, ectoplacental cone; ed, extra-embryonal endoderm. No expression for Nxf2 could be detected in these tissues at any time point. Nxf2 is abundantly expressed in the adult testis (C) while much less staining was obtained for Nxf7 (I) in this tissue. (J) RT-qPCR for relative expression of mouse Nxf2 and Nxf7, normalized to Gusb and Hprt1. Al, allantois; Ys, yolk sac; N = 2. Means ± standard deviations. *p<0.05.

Figure 3. Expression of mouse Nxf7 in adult mouse brain as detected by in situ hybridization.

(A) Cortical Nxf7 signals (upper panel) and Dapi staining (middle panel). The lower panels show the Nxf7/Dapi-merged pictures of the yellow boxed region 1 (border between CC and CTX) and region 2 (layers V and VI). A schematic view of the brain is given in between both merged pictures with the brain region shown in the upper parts, indicated with a blue box. VL, lateral ventricle; CC, corpus callosum; CTX, Cerebral cortex; CP, Caudoputamen. (B) Merged Nxf7 (red) and Dapi (blue) staining of the pyramidal cell layer of the CA1 region of the hippocampus (Hc). Experiments were done in duplicate. Scale bar, 100 µm.

Figure 4. Generation and genotyping of Nxf7 knockout mice.

(A) Schematic representation of the targeted Nxf7 allele (upper part) and its Cre-mediated KO allele (middle part). (B) Schematic representation of the Nxf7 transcript with the PCR primers. Position of exon 3 is indicated in dark gray. Amplicon sizes are indicated in base pair (bp). The agarose gel shows the PCR data of the expression analyses for the indicated primer pairs performed on brain cDNA of KO and WT male mice. (C) Western blot on yolk sac from WT and Nxf7 KO male embryos with the polyclonal Nxf7-Nt Ab. (D) Western blot on protein lysates obtained from transfected HEK-293T cells with the wild type Nxf7-flag cDNA (lane 1) and the exon3-lacking Nxf7-flag cDNA (lane 2). Detection was performed with anti-flag Ab. The anti-Actb Ab was used as the loading control. Western blot for the HA-tag gave the same result (data not shown).

Figure 5. Immunofluorescence of Nxf7 on brain sections.

Staining with the polyclonal anti-Nxf7-Nt antibodies on hippocampus (Hc) CA1 (upper panels) and cortex (Cx) layer V (lower panels) sections of wild type (WT) and Nxf7 knockout (KO) mice. The assay was done in duplicate. Images were taken with a Leico microscope at 20×magnification. Scale bar, 100 µm.

Figure 6. Hippocampus-dependent spatial learning in the Morris water maze.

(A) Nxf7 KO animals (n = 21, open symbols) needed significantly longer to find the hidden platform than WT animals (n = 20, black symbols) during the 10 acquisition days. (B) During probe trials, only WT animals spent significant more time in target quadrant compared to chance (dotted line at 25%), while Nxf7 KO did not. Data are presented as means ± SEM. *p<0.05 compared to WT (Student’s t-test), # p<0.05 and ### p<0.001 compared to chance level (one sample t-test).