Fbxo45 forms a novel ubiquitin ligase complex and is required for neuronal development

Abstract

Fbxo45 is an F-box protein that is restricted to the nervous system. Unlike other F-box proteins, Fbxo45 was found not to form an SCF complex as a result of an amino acid substitution in the consensus sequence for Cul1 binding. Proteomics analysis revealed that Fbxo45 specifically associates with PAM (protein associated with Myc), a RING finger-type ubiquitin ligase. Mice deficient in Fbxo45 were generated and found to die soon after birth as a result of respiratory distress. Fbxo45(-)(/)(-) embryos show abnormal innervation of the diaphragm, impaired synapse formation at neuromuscular junctions, and aberrant development of axon fiber tracts in the brain. Similar defects are also observed in mice lacking Phr1 (mouse ortholog of PAM), suggesting that Fbxo45 and Phr1 function in the same pathway. In addition, neuronal migration was impaired in Fbxo45(-)(/)(-) mice. These results suggest that Fbxo45 forms a novel Fbxo45-PAM ubiquitin ligase complex that plays an important role in neural development.

FIG. 1.

Fbxo45 does not form a canonical SCF complex due to an amino acid substitution in the F-box domain. (A) Lysates of HEK293T cells expressing 3×FLAG (3F)-tagged Fbxw2, Skp2, or Fbxo45 were subjected to immunoprecipitation (IP) with anti-FLAG, and the resulting precipitates, as well as the original cell lysates (Input), were subjected to immunoblot analysis (IB) with antibodies to the indicated proteins. (B) Clustal W alignment of the F-box domains of various human (h) or mouse (m) F-box proteins. Conserved residues are shaded, and residues thought to be important for interaction with Cul1 are boxed. (C) Lysates of HEK293T cells expressing 3×FLAG-tagged Skp2, Skp2(E115R), Fbxo45, or Fbxo45(R42E) were subjected to immunoprecipitation with anti-FLAG, and the resulting precipitates and original cell lysates were subjected to immunoblot analysis with antibodies to the indicated proteins.

FIG. 2.

Identification of PAM as a protein that interacts with Fbxo45. (A) Lysates of HEK293T cells expressing 3×FLAG-tagged Fbxo45 were subjected to immunoprecipitation with anti-FLAG, and the resulting precipitates were subjected to SDS-PAGE and staining with Coomassie blue. Proteins identified by LC-MS/MS are indicated. (B) Schematic representation of deletion mutants of human PAM. FL, full length. (C) Lysates of HEK293T cells expressing HA-tagged Fbxo45 and the 3×FLAG-tagged deletion mutants of PAM shown in panel B were subjected to immunoprecipitation (IP) with anti-FLAG, and the resulting precipitates, as well as the original cell lysates, were subjected to immunoblot (IB) analysis with antibodies to HA and to FLAG. (D) Lysates of HEK293T cells expressing HA-tagged Fbxw2, Skp2, or Fbxo45 together with the 3×FLAG- tagged R7 deletion mutant of PAM were subjected to immunoprecipitation with anti-FLAG, and the resulting precipitates, as well as the original cell lysates, were subjected to immunoblot analysis with antibodies to HA or to FLAG. (E) Lysates of HEK293T cells expressing the HA-tagged deletion mutants of mouse Fbxo45 indicated on the left and the 3×FLAG-tagged R7 deletion mutant of PAM were subjected to immunoprecipitation with anti-FLAG, and the resulting precipitates, as well as the original cell lysates, were subjected to immunoblot analysis with antibodies to HA or to FLAG. (F) HEK293T cells expressing HA-tagged Fbxo45 (or HA-Skp2 as a negative control) with or without the 3×FLAG (3F)-tagged R7 deletion mutant of PAM (residues 2413 to 2712) were treated with cycloheximide for the indicated times, after which cell lysates were subjected to immunoblot analysis with antibodies to HA, to FLAG, or to Hsp70 (loading control). The band intensities of HA-tagged Fbxo45 or Skp2 in the upper panels were quantified relative to the corresponding values for time zero (bottom panel). (G) Model for formation of an Fbxo45-PAM complex. In typical F-box proteins, the F-box domain interacts not only with Skp1 but also with Cul1, resulting in the formation of an SCF complex. In contrast, an amino acid substitution in the F-box domain of Fbxo45 interferes with the binding to Cul1 and thereby prevents formation of an SCF complex. Instead, the SPRY domain of Fbxo45 binds directly to PAM, resulting in the formation of an Fbxo45-PAM complex. Ub, ubiquitin.

FIG. 3.

Targeted disruption of Fbxo45. (A) Schematic representation of the wild-type Fbxo45 locus, the targeting vector, and the mutant allele after homologous recombination. A 1.0-kb genomic fragment including exon 1 of Fbxo45, which encodes the F-box domain, was replaced by IRES-lacZ and PGK-neo-poly(A)-loxP cassettes. Exon (Ex.) 2 and the coding portions of exons 1 and 3 are depicted by filled boxes, with the open boxes indicating the noncoding portions of exons 1 and 3. A genomic fragment used as a probe for Southern blot analysis is shown as a solid bar. Restriction sites: B2, BglII; S, SpeI. DT-A, diphtheria toxin A cassette. (B) Southern blot analysis with the probe shown in panel A of genomic DNA isolated from the tail of adult mice and digested with BglII and SpeI. The 7.6- and 5.6-kb bands corresponding to the wild-type and mutant alleles, respectively, are indicated. The Fbxo45 genotypes of the analyzed mice are shown above each lane. (C and D) Immunoblot analysis with anti-Fbxo45 and anti-Hsp70 (loading control) of lysates of the brain (C) or the indicated tissues (D) from mice of the indicated Fbxo45 genotypes at postnatal day 1. (E and F) Gross appearance of newborn littermates (E) and the skeletons of E18.5 littermates (F) of the indicated Fbxo45 genotypes. Arrowheads indicate lordotic body posture specific to Fbxo45^−/− pups. (G and H) Lung sections of newborn wild-type (G) or Fbxo45^−/− (H) littermates were stained with hematoxylin-eosin. The pulmonary alveoli were expanded with air only in the wild-type animal. Scale bar, 200 μm.

FIG. 4.

Innervation defects in Fbxo45^−/− mice. The diaphragm of wild-type (A and C) or Fbxo45^−/− (B and D) embryos at E14.5 (A and B) or E15.5 (C and D) was subjected to whole-mount immunofluorescence staining with antibodies to neurofilaments and to synaptophysin. The boxed regions in panels A and B are shown at higher magnification in panels a through h. In wild-type embryos, the phrenic nerve reaches the diaphragm, branches, and projects both dorsally and ventrally (A). In Fbxo45^−/− embryos, however, branching and projection of the phrenic nerve are impaired (B). Furthermore, the phrenic nerve sometimes branches toward the inside in Fbxo45^−/− mice (arrows in panel f), whereas it branches toward the outside of the diaphragm in wild-type embryos (a, c, e, and g). Arrows in panel D indicate excessive growth and entanglement of the right and left phrenic nerves in the mutant embryo. Scale bars, 800 μm (A to D) or 250 μm (a to h).

FIG. 5.

Neuromuscular synapses in mice lacking Fbxo45. The diaphragm of wild-type (A, B, C, H, I, and J) or Fbxo45^−/− (D, E, F, K, L, and M) embryos at E16.5 or of wild-type (N to P) or Fbxo45^−/− (Q to S) embryos at E18.5 was subjected to whole-mount fluorescence staining with antisynaptophysin (for presynaptic nerve terminals) and with α-bungarotoxin (for AChRs), as indicated. Synaptophysin is colocalized with AChRs in the diaphragm of both wild-type and Fbxo45^−/− embryos, indicating that NMJs are correctly connected in the mutant; however, the number of synapses is decreased in the diaphragm of Fbxo45^−/− embryos (A to F). Quantification of the mean endplate bandwidth is shown in panel G. Data are means ± SEM for five embryos of each genotype. The P value was determined by Student's t test. **, P < 0.01. Higher-magnification views of the left portion of the pars sternalis in panel H through M reveal the formation of synapses between muscle and the left phrenic nerve, with muscle fibers having only one endplate band in their central region, in wild-type embryos. In contrast, synapses form not only between muscle and the left phrenic nerve (arrows) but also between muscle and the right phrenic nerve extending across the ventral midline (arrowheads) in Fbxo45^−/− embryos. Higher-magnification views of the diaphragm also reveal an aneural region in the Fbxo45^−/− embryo at E18.5 (Q to S) and the corresponding region of a wild-type embryo (N to P); arrows indicate postsynaptic differentiation in this region of the mutant. Scale bars: 200 μm (A, B, C, D, E, F, H, I, J, K, L, and M) or 50 μm (N to S).

FIG. 6.

Defective development of fiber tracts in Fbxo45 mutant embryos. Coronal sections of the head portion of wild-type (A, C, E, and G) or Fbxo45^−/− (B, D, F, and H) mice at E18.5 were stained with cresyl violet. Higher-magnification views of the boxed regions in panels A, B, E, and F are shown in panels C, D, G, and H, respectively. The anterior commissure (arrows in panel A) and internal capsule are apparent in the wild-type but not the Fbxo45^−/− brain. In the Fbxo45^−/− brain, the internal capsule is replaced by an aberrant bundle (arrows in panel H). Abbreviations: Ac, anterior commissure; Ctx, cortex; Hp, hippocampus; Lv, lateral ventricle; Th, thalamus; V3, third ventricle. Scale bars, 500 μm.

FIG. 7.

Impaired neuronal migration in Fbxo45^−/− forebrain. (A and B) Coronal sections of wild-type (A) or mutant (B) neocortex beneath bregma at postnatal day 1 were stained with hematoxylin-eosin. (C and D) BrdU was administered to pregnant mice at E14.5, the brain was isolated from wild-type (C) or Fbxo45^−/− (D) embryos at E18.5, and coronal sections of the neocortex were subjected to immunofluorescence staining with anti-BrdU. (E) To quantify the distribution of BrdU-positive nuclei, the cortical plate and intermediate zone were divided into four equal bins from the pial side to the ventricular side. The number of nuclei in each bin was determined, and these are shown as percentages of the total numbers of nuclei in the cortical plate and intermediate zone. Data are means ± SEM for three embryos of each genotype. The P value was determined by Student's t test. *, P < 0.05; **, P < 0.01. (F to J) Coronal sections of the brain tissue of wild-type (F) or Fbxo45^−/− (G) embryos at E18.5 were subjected to immunofluorescence staining with anticalretinin. Arrows indicate accumulation of calretinin-positive cells in the basal telencephalon of the mutant embryo. The boxed region in panel G is shown at higher magnification in panel H; staining of the same region with Hoechst 33258 is shown in panel I, and the merged image in panel J reveals colocalization of calretinin and Hoechst 33258 staining. Abbreviations: CP, cortical plate; IZ, intermediate zone. All scale bars, 200 μm.

FIG. 8.

Impaired neuronal migration in Fbxo45^−/− spinal cord. (A to D) Structure of the spinal cord of wild-type (A and C) or Fbxo45^−/− (B and D) embryos at E18.5. Transverse sections of the neck region were stained with hematoxylin-eosin. The images in panels A and B are shown at higher magnification in panels C and D, respectively. The gray matter is enlarged, distended, and flattened laterally, resulting in the loss of its invagination around the lateral column (arrowheads), in the Fbxo45^−/− spinal cord. Abbreviations: DRG, dorsal root ganglion; Gm, gray matter; Wm, white matter. (E to H) Immunofluorescence staining of transverse sections of the brachial portion of wild-type (E and G) or Fbxo45^−/− (F and H) embryos at E12.5 with anti-choline acetyltransferase (E and F) or with anti-Isl1 (G and H). The populations of motor neurons positive for choline acetyltransferase (ChAT) or for Isl1 form the LMC (arrows) and the MMC (arrowheads) in the wild-type spinal cord. The LMC is poorly formed, and the MMC is aberrantly distributed (asterisks) in the Fbxo45^−/− spinal cord. All scale bars, 200 μm.