Abstract
Nephronophthisis-related ciliopathies (NPHP-RC) are a group of inherited genetic disorders that share a defect in the formation, maintenance or functioning of the primary cilium complex, causing progressive kidney failure and other clinical manifestations. Mutations in centrosomal protein 164 kDa (CEP164), also known as NPHP15, have been identified as a cause of NPHP-RC. Here we have utilised the MRC-Wellcome Trust Human Developmental Biology Resource (HDBR) to perform immunohistochemistry studies on human embryonic and foetal tissues to determine the expression patterns of CEP164 during development. Notably expression is widespread, yet defined, in multiple organs including the kidney, retina and cerebellum. Murine studies demonstrated an almost identical Cep164 expression pattern. Taken together, this data supports conserved roles for CEP164 throughout the development of numerous organs, which we suggest accounts for the multi-system disease phenotype of CEP164 mediated NPHP-RC.
Introduction
Nephronophthisis-related ciliopathies (NPHP-RC) are a collection of inherited genetic disorders, grouped together by a core defect in the formation, maintenance or functioning of the primary cilium complex [1, 2]. NPHP-RC patients typically present with nephronophthisis, a fibrotic cortico-medullary cystic kidney phenotype, which frequently leads to end stage-renal disease (ESRD) [3, 4]. In some NPHP-RC cases, including Senior-Loken syndrome (SLSN), Alstrom syndrome (AS), Bardet Biedl syndrome (BBS) and Joubert syndrome (JBTS), patients have retinal dysplasia and degeneration phenotypes, such as Leber congenital amaurosis, which can deteriorate to blindness [5-7]. Neurological abnormalities are often present; JBTS patients have midbrain cerebellar vermis hypoplasia, characterised by the “molar tooth” sign on MRI analysis [5, 8]. This can cause numerous problems including ataxia, hypotonia and breathing abnormalities. Intellectual disability and developmental delay can also be present, which is demonstrated throughout the spectrum of NPHP-RC. Additionally, BBS patients are often diagnosed with hypogonadism and/or obesity [9]. Severe NPHP-RC phenotypes, including Jeune syndrome and Meckel Gruber syndrome, can also present with skeletal dysplasia, and lethal occipital encephalocele. Consistent with ciliopathy syndromes, polydactyly, liver fibrosis, facial dysmorphism, cardiac abnormalities, hearing loss and type 2 diabetes can present as secondary symptoms [9-11]. Although some disease management therapies are available, notably, there is no cure for NPHP-RC.
There are at least 25 NPHP-RC causative genes currently identified [12], accounting for a molecular genetic diagnosis in around 60% of NPHP-RC patients. Mutations in CEP164, also known as NPHP15, have been identified as a cause of NPHP-RC [1, 9]. These patients have a largely heterogenous clinical presentation. The majority suffer with NPHP and a retinal phenotype, which is some patients cause blindness prior to two years of age. One patient was diagnosed with JBTS, and two others have BBS-like manifestations with an absence of renal problems [1, 9]. It has been proposed that homozygous truncating mutations of CEP164, cause a severe JBTS phenotype, whereas hypomorphic mutations may cause a less severe phenotype [1]. Patients with identical CEP164 mutations can display different clinical manifestations, which makes it difficult to understand the pathology and progression of the disease.
CEP164 was first identified and cloned from an adult foetal brain cDNA library [13] and is located on chromosome 11, q23.3. Its largest, most commonly cited isoform is 5,629 bp, encoding the 1,460 amino acid protein (CEP164) (NM_014956) [1, 14]. There is an alternative isoform of CEP164 (NM_001271933), with a 1,455 amino acid product [1, 13, 14]. CEP164 is a centrosomal protein that localises to each of the nine distal appendages of the primary cilia mature centriole, in a microtubule-independent manner [1, 2, 15-22]. This has been demonstrated by super-resolution microscopy techniques in multiple human and murine cell lines. Cell-cycle dependent recruitment of CEP164 to the basal body distal appendages is hierarchical, within a network of other distal appendage proteins. These include CEP83 (mutations in CEP83 cause NPHP18), CEP89, SCLT1 (variants in SCLT1 may be associated with OFD IX), and FBF1. CEP164 is recruited last, defining the proximal end of the transition zone [2, 15, 18, 22-26].
Proteomics analysis has identified that CEP164 has an N-terminal tryptophan-tryptophan (WW) domain, an area of lysine-rich repeats (LR), and multiple serine/glutamine (SQ/TQ) potential phosphorylation sites [2, 14]. There are at least 3 predicted coiled-coil domains [1, 2, 14, 15] and the C-terminal domain is currently undefined. (S1 Figure A). Multiple fusion protein studies have demonstrated that the CEP164 C-terminal domain is required for localisation of CEP164 to the distal appendages [1, 15]. Additionally, CEP164 has been localised to the nucleus [1, 14, 17, 27].
Numerous CEP164 knockdown studies (siRNA/CRISPR) have demonstrated that CEP164 loss causes aberration of ciliogenesis, with disruption of primary cilia production prior to transition zone formation; notably centriolar structure is not disrupted [15, 23, 25, 28, 29]. Upon initiation of ciliogenesis Rabin 11 imports the GTPase Rabin 8 to the basal body. CEP164 interacts with Rabin 8, facilitated by Chibby 1, to recruit Rab8 positive vesicles to the centrosome [15, 17, 19, 24, 30]. Vesicle docking is required for subsequent basal body anchoring to the plasma membrane and primary cilia development. Additionally, CEP164 forms a complex with tau tubulin kinase 2 (TTBK2) via its N-terminal WW domain. CEP164 recruits TTBK2 to the primary cilia basal body, which allows removal of the centriolar capping protein, CP110, potentially via phosphorylation, initiation of intraflagellar transport recruitment and subsequent axonemal extension [1, 15, 29, 31, 32]. The CEP164/Rabin 8 and CEP164/TTBK2 pathways work independently of each other [29]. There are other predicted interactors of CEP164 including dishevelled, NPHP3, NPHP4 and ARL13B, indicating that CEP164 is likely to have other ciliary roles [1, 15, 24, 32].
Several studies have indicated a potential role for CEP164 in ATM/ATR-mediated DNA damage response (DDR) and UV-induced nucleotide excision repair pathways, however results are conflicting [14, 17, 33]. Likewise, data supporting a role for CEP164 in cell-cycle regulation is inconsistent [1, 2, 14, 15, 25]; both of these CEP164 roles need to be further validated.
CEP164 has numerous orthologs, including M.muscularis, D.melanogaster, C.reinhardtii and D.rerio (S1 Figure B,C,D). The murine ortholog of CEP164 is a 30-exon gene, located on chromosome 9, qA5.2, which shares 77% identity to the full-length human CEP164. It encodes a 1333 amino acid protein, which shares 58% identity to the human CEP164 1,460 amino acid protein [34, 35]. Like human CEP164, murine Cep164 may have multiple isoforms.
Given CEP164 mutations cause NPHP-RC, we sought to determine the expression of CEP164 throughout human and murine development, particularly focusing on the cerebellar-retinal-renal phenotype. Collaboration with the Medical Research Council (MRC) Human Developmental Biology Resource (HDBR) allowed the procurement of human embryonic and foetal samples, which was compared to expression data from the murine 129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+ gene trap model.
Materials and Methods
Study approval
The study was conducted with full ethical approval and consent. Ethical approval was obtained from the National Research Ethics Service Committee North East – Newcastle & North Tyneside 1 (08/H0906/21+5). Human embryonic and foetal tissue samples were collected with appropriate consent and ethical approval, via the Medical Research Council (MRC) Wellcome trust-funded Human Developmental Biology Resource (HDBR). Animal experiments were performed under Home Office Licences (United Kingdom) in accordance with the guidelines and regulations for the care and use of laboratory animals outlined by the Animals (Scientific Procedures) Act 1986. Protocols conducted were approved by the Animal Ethics Committee of Newcastle University and the Home Office, United Kingdom.
Mouse Genetics
C57BL/6NTac-Cep164tm1a(EUCOMM)Wtsi/+ mice, which were generated for the International Phenotyping Consortium Initiative, were obtained from MRC Harwell [36]. These were backcrossed onto a 129/Ola-Hsd background, forming mice heterozygous for the gene trap, 129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+. 129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+ heterozygous (HET) mice were mated with wild type (WT) 129/OlaHsd-Cep164+/+ mice, gaining pups HET:WT 1:1. In the Cep164tm1a allele, upon pre-mRNA processing, exon 3 of Cep164 splices into the LacZ/Neomycin cassette, within intron 3, which introduces a frame-shift and subsequent nonsense mutations (S2 Figure). This generates a Cep164 null allele by hypothesised nonsense mediated decay. The LacZ cassette contains an internal ribosomal entry site (IRES), allowing translation of a beta-galactosidase fusion gene upon activation of the native Cep164 promoter, thus acting as a Cep164 reporter gene. Upon additional of X-Gal substrate, beta galactosidase hydrolyses X-Gal forming 5-bromo-4-chloro-3-hydroxyindole-1, which oxidises to a blue precipitate, 5,5’-dibromo-4,4’-dichloro-indigo-2. HET 129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+ mice were used for X-Gal Cep164 expression studies, as they are phenotypically normal, WT mice were used as littermate controls.
Cep164 genotyping
Ear or tail biopsies (embryos) were lysed for 1 h, at 95°C, in 50 μl alkaline lysis buffer (25 mM NaOH, 0.2 mM EDTA, pH ≈ 12) and then neutralised in 50 μl neutralising buffer (40 mM trizma hydrochloride, pH 5-5.5). A PCR reaction using GoTaq G2 DNA Polymerase (Promega) for Cep164tm1a(EUCOMM)Wtsi/+, was completed using the following primers flanking the second loxp site; F1 5’ CTC CCA CAG TGA CAA ATG CC 3’, R1 5’ GGT AGT TGT TAC TTC TGT CAG 3’ (Eurofins Genomics). Expected amplicon sizes are as follows; WT (141 bp), homozygous (Hom) (163 bp), HET (141 bp and 163 bp). PCR products were run on a 1.5% agarose gel, with GelRed Nucleic Acid GelStain (1:10,000) (Biotium) at 150V for 45 min. To confirm correct genotyping, representative samples were Sanger-sequenced (GATC-BIOTECH).
Murine Tissue Collection, Fixation, Sectioning and Staining
Murine tissue (kidney, brain, eye, heart, lung, liver, testes) was collected at P0.5/P1.5, P15.5 and P29.5/P30.5, using a standard dissection procedure. Tissues were fixed in 0.2% glutaraldehyde fix (0.2% glutaraldehyde, 2mM MgCl2, 5mM EGTA in PBS) for 90 min at 4°C, washed in PBS and then stored in 15% sucrose/PBS overnight at 4°C.Tissues were transferred to 30% sucrose/PBS and incubated at 4°C overnight or until samples sunk. Tissues were frozen in OCT compound, and stored at −80°C.
Sections of tissues were cut (10-12μm) using a cryostat and mounted on charged glass slides. These were incubated in 0.2% glutaraldehyde fix (0.2% glutaraldehyde, 2mM MgCl2, 5mM EGTA in PBS) for 5 min, washed in PBS for 10 min, and then washed in X-Gal wash (2 mM MgCl2, 0.01% Sodium Deoxycholate, 0.02% NP-40 in PBS) for 10 min. Slides were incubated with X-Gal stain (1mg/ml of X-Gal DMSO in 2 mM MgCl2, 0.01% Sodium Deoxycholate, 0.02% NP-40, 5mM Potassium Ferricyanide, 5mM Potassium Ferrocyanide in PBS) at 37°C in the dark, until the blue precipitate stain intensity did not further increase or WT littermate controls showed endogenous beta galactosidase staining. Slides were washed in PBS and then dehydrated to 100% ethanol before clearing in Histoclear II (National Diagnostics) and mounting in DPX mounting medium (Sigma-Aldrich), slides were imaged using the SCN400 Side Scanner (Leica).
Murine Embryo Collection, Wholemount Fixation and Staining
Murine embryos were collected at E9.5, E10.5 and E12.5 using a standard dissection procedure. Embryos were fixed in 0.2% glutaraldehyde solution (0.2% glutaraldehyde, 2mM MgCl2, 5mM EGTA in PBS) for 1 hour on ice and washed with X-Gal wash (2 mM MgCl2, 0.01% Sodium Deoxycholate, 0.02% NP-40 in PBS) prior to overnight storage at 4°C. Embryos were incubated in X-Gal stain, consisting of 25mg/ml of X-Gal/DMSO solution in a 1 in 25 dilution of X-Gal staining buffer (2 mM MgCl2, 0.01% Sodium Deoxycholate, 0.02% NP-40, 5mM Potassium Ferricyanide, 5mM Potassium Ferrocyanide in PBS) at 37°C in the dark; incubation was completed once the blue precipitate stain intensity did not further increase or WT littermate controls started to show endogenous beta galactosidase staining. Embryos were washed in 1x phosphate buffered saline solution (PBS) (Gibco, pH7.45) and dehydrated in 70% ethanol, prior to imaging using a Leica Man Stereomicroscope and Axiovision software.
Human Tissue collection, fixation and processing
Human embryonic and foetal tissues were obtained from the MRC Wellcome Trust-funded Human Developmental Biology Resource. CS23 whole embryos were fixed in 4% paraformaldehyde (PFA), for 72 hours, with incision in the skull down the sagittal plane and a slit in the abdomen at the umbilicus. Tissues were transferred to methacarn until further processed. This same protocol was used for processing of foetal eye, brain and kidney; the embryonic and foetal kidney were cut sagittally, to allow the fix to penetrate. Human embryonic and foetal tissues were washed in increasing concentrations of ethanol, and then incubated in 3 changes of xylene before embedding in paraffin wax. Sections, on positively charged glassed slides, were de-waxed in two washes of xylene, and then rinsed in two changes of absolute ethanol. Slides were incubated in methanol peroxide solution (0.5% H2O2) for 10 min to block endogenous peroxidase. Slides were rinsed in tap water and then antigen retrieval was completed using citrate buffer. After rinsing in Tris-buffered saline (TBS), slides were incubated with 10% goat serum (Vector) for 10 min at room temperature and then incubated with the primary antibody in goat serum with TBS overnight at 4 °C (S1 Table). After washing with TBS, slides were incubated with the goat anti rabbit BA-1000 biotinylated secondary antibody (Vector Laboratories) diluted in goat serum with TBS (1/500), for 30 min at room temperature. The secondary was washed with TBS and then slides were incubated with VECTASTAIN Elite ABC kit PK6100 tertiary complex (Vector Laboratories), for 30 min at room temperature. After TBS washes, the stain was developed for 10 min at room temperature using ImmPACT DAB peroxidase substrate (SK-6100) solution (Vector Laboratories). The slides were washed thoroughly in water, and counterstained with haematoxylin, dehydrated, cleared and mounted.
An indirect, two-step, method was utilised for the foetal kidney sections. Samples were incubated with a HRP goat anti-rabbit IgG peroxidase secondary (Vector) (1/500). After TBS washes, the stain was developed using DAB peroxidase substrate solution, as described previously. No primary controls sections were utilised as negative controls, anti-PAX6 was utilised as a positive control for the brain and cerebellum (S1 Table).
Image analysis
Images of human and murine tissues were analysed using the SCN400 Slide scanner software (Leica). Figures 1-4, and S3-S5 Figures were generated using Adobe Photoshop CS3 Extended.
Data Availability
The authors declare that all data supporting the results are presented with the article, supporting data can be gained from the corresponding author upon reasonable request.
Results
CEP164 expression during human and murine renal development
In the human embryonic and foetal kidney, immunohistochemical staining demonstrates that CEP164 is present in the metanephric epithelial-derived renal vesicles, comma-shaped bodies and subsequent s-shaped bodies, during all time points analysed (8PCW-18PCW) (Figure. 1 A.I.II.III.VII B.I.II C.I.II). This expression pattern is maintained throughout nephrogenesis, with CEP164 present in the primitive nephron tubule (8PCW), but also in more defined proximal and distal nephron segments, as well as cells of the loop of Henle, as seen from 14 PCW (Figure 1A.I.IV.VII, B.I.IV.VII, C.I.IV.VII). Specifically, CEP164 expression is defined to the apical membrane of the epithelial cells lining the developing nephron lumen, during all the developmental time points analysed (8PCW-18PCW). Notably, CEP164 expression is stronger in the premature nephron renal vesicles, comma-shaped bodies, s-shaped bodies and immature renal tubules, which is then reduced in the more developed cortical and medullary tubular segments (8PCW-18PCW). CEP164 does not seem to be expressed, or is expressed very weakly, in the metanephric cap mesenchyme (Figure 1A.III, B.III).
Cap mesenchyme (CM), collecting duct (CD), distal tubule (DT), loop of henle (LH), postnatal day (P), post conception weeks (PCW), proximal tubule (PT), renal corpuscle (RC), renal tubule (RT), renal vesicle (RV), s-shaped body (SSB), uteric bud (UB).
In the developing human kidney, CEP164 is expressed in the glomerulus of the renal corpuscle at 8 PCW-18 PCW, indicating podocyte CEP164 expression (Figure 1A.V.VI, B.V.VI, C.V.VI). The data suggests that CEP164 expression is reduced as the glomeruli matures; this is most clearly seen at 18 PCW.
The human embryonic/foetal renal uteric bud, which forms the collecting duct, is derived from the metanephrogenic diverticulum. Both the uteric bud and collecting duct also show CEP164 expression in the apical membrane of the epithelial cells lining the tubular lumen (8PCW-18PCW) (Figure 1A.III.VII, B.III.VII, C.III.VII). CEP164 is also expressed at the basolateral membrane of the collecting duct tubule (8PCW-18PCW). CEP164 expression is not present in the human renal interstitium (Figure 1A-C). Negative control staining indicating that CEP164 expression in the human kidney was specific (S4 Figure A,B).
In the murine kidney (129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+), the LacZ reporter assay demonstrates that Cep164 expression correlates with human kidney CEP164 expression. Correspondingly, Cep164 is expressed in the developing murine renal vesicles, s-shaped bodies (P1.5) (Figure 1D.I.II) and the subsequent renal tubules, including nephron segments in both the cortex and medulla (P29.5) (Figure 1D.IV, E.I.III.IV). The staining indicates that there is potentially cell-specific expression within each of the tubular segments. Cep164 is also expressed in the glomeruli of the renal corpuscle at P1.5, which is not present in mature glomeruli at P29.5 (Figure 1D.V.VI, E.II). Likewise, Cep164 is expressed in the developing mesonephric uteric bud and collecting ducts (Figure 1D.II). Controls demonstrate that there is no endogenous LacZ expression in the kidney (S5 Figure B).
CEP164 expression during human and murine retinal development
In the developing human retina, CEP164 expression is widespread yet defined (Figure 2A). In early retinal development (11 PCW) CEP164 is expressed weakly in the developing nerve fibre layer (NFL) (fibrous extensions from the optic nerve) and nerve fibres in the differentiating ganglion cell layer (GCL)/inner plexiform layer (IPL) (Figure 2A.I). The basally located layer of cone precursors, which has differentiated from the outer neuroblastic cell layer (ONBL), also demonstrates strong CEP164 expression (Figure 2A.I). By 14 PCW, CEP164 expression is defined to the NFL, the IPL and the photoreceptor layer which now has both developing rods and cones (Figure 2A.II). Later in development, 19 PCW, once all of the primitive retinal layers have formed, CEP164 expression is maintained in the nerve fibres of the NFL, the ganglionic and inner nuclear nerve fibril synapses of the IPL, and the inner segments (IS) and outer segments (OS) of the photoreceptor cell layer (Figure 2A.III). At 19 PCW, CEP164 expression is also present in the developing outer plexiform layer (OPL) which contains nerve fibril synapses between the inner nuclear layer (INL) and the outer nuclear layer (ONL). CEP164 expression in the retinal pigment epithelial layer (RPE) cannot be defined due to its natural dark colouring.
Ganglion cell layer (GCL), inner nuclear layer (INL), inner plexiform layer (IPL), inner segment (IS), nerve fibre layer (NFL), outer neuroblastic cell layer (ONBL), outer nuclear layer (ONL), outer plexiform layer (OPL), outer segment (OS), photoreceptor segment layer (PS), retinal pigment epithelium (RPE).
Murine Cep164 retinal expression in (129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+) largely corresponds to the human CEP164 expression patterns, with expression maintained in the developing IPL, OPL and PS layer (P1.5-P29.5) (Figure 2B). However, there are some clear differences. Cep164 is expressed in retinal precursor cells throughout the murine ONBL, which in the developing human retina is restricted to the cone precursor cells. Additionally, strong Cep164 expression is present and maintained throughout the GCL (P1.5-P29.5) (Figure 2B). Cep164 is also expressed sparsely in some cells of the INL and ONL, however these tend to be at the boundaries of the plexiform layers (P15.5-P29.5) (Figure 2B.II.III). Notably, at later stages of development, Cep164 expression is clearly defined to the IS of the photoreceptor cell layer (P15.5-P29.5) (Figure 2B.II.III). Additionally, the retinal pigment epithelial layer demonstrates strong, cell-specific Cep164 expression throughout development (P1.5-P29.5) (Figure 2B). WT controls demonstrate no endogenous X-Gal staining (S5 Figure A).
CEP164 expression during human and murine neuronal and cerebellar development
CEP164 is expressed widely throughout human brain development (8PCW-18PCW) (Figure 3A-D). At 8PCW CEP164 is strongly expressed in the neuroepithelium of the developing telencephalon (Tel), which is defined in cells lining the lateral ventricle (LV), including the ventricular layer (VL), subventricular layer (SVL), cortical plate (CorP) and marginal layer (MaL), (Figure 3A.I.VI, B.I.VI). At 8 PCW, CEP164 is also present in the developing neuroepithelium of the diencephalon (Di), mesencephalon (Mes), metencephalon (Met) and myencephalon (Mye), with expression strongest in the apical epithelium lining the brain ventricles (Figure 3A.I.V.VI.VII, 3B.I.VI.VII.VIII). At 8PCW there is strong, well-defined CEP164 expression in the midline of the myencephalon (Figure 3B.IV.V), as well as cells of the nasal epithelium (Figure 3B.IX). Throughout human brain development (8PCW-18PCW) the choroid plexus demonstrates strong CEP164 expression, specifically in the ependymal cells, and more weakly in the choroid plexus pia matter (Figure 3A.II.III.IV, B.II.III, C.V, D.V). The developing human cerebellar folds show defined CEP164 expression in the migrating molecular cell layer at 16 PCW, but this is reduced by 18PCW (Figure 3C.I.II, D.I.II). CEP164 is expressed at the apical membrane of the ventricular surface of the pons and cerebellum, as well as the myencephalon, specifically in the axon tracts of the medulla oblongata (Figure 3A.VII.VIII, C.III.IV.VI, D.III.IV.VI). This expression pattern can be seen from 8PCW to 18 PCW, however CEP164 expression levels are much weaker by 18 PCW. At 8PCW CEP164 is expressed in the white matter, but this is lost with maturation of the cerebellum (Figure 3A.VII, C.VII, D.VII). Notably the human foetal brain negative controls demonstrate no positive staining (S4 Figure D-F). A PAX6 antibody was used as a positive control (S4 Figure E,F).
Developing murine hindbrain (E-H), P0.5 (E-F), P15.5 (G) and P30.5 (H). Cep164 expression is seen in the developing P0.5 murine brain (E, G.I, H.I) with expression in the cortex, striatum and cerebrum of the maturing cerebral cortex (G.V, H.V), thalamus (G.VI) and midbrain (G.VII, H.VI) at P15.5 and P30.5. Cep164 expression is strong and defined in the ventricular neuroepithelium (G.VII.VIII). The murine choroid plexus demonstrates strong Cep164 expression, however it is not defined to a specific cell type. (E.III, F.IV). In the murine model, Cep164 is expressed in the migrating molecular layer (FI.II), however this expression is maintained throughout development with expression also in the molecular layer, ganglion cell layer and Purkinje cell layer (G.III,H.III). Cep164 expression is seen in the murine cerebellum (F.III,G.III), in the murine pons (F.V, H.VI), and weakly in the cerebellar white matter (F.VI), which is lost with maturation (H.IV).
Aquaduct (Aq), axon tract (AT), cerebellum (Ce), cerebral cortex (CC), cerebral hemisphere (CH), cerebrum (Cr), choroid plexus (CP), cortex (Cor), cortical plate (CorP), diencephalon (Di), ependymal cells (Ep), fourth ventricle (FV), ganglion cell layer (GCL), inner plexiform layer (IPL), intermediate zone (IL), lateral ventricle (LV), marginal layer (MaL), medulla oblongata (MO), mesencephalon (Mes), metencephalon (Met), midbrain (MB), midline (Mi), molecular cell layer (MCL), myencephalon (Mye), nasal epithelium (Na), outer neuroblastic layer (ONBL), pia (P), pons (PO), purkinje cell layer (PCL), retinal pigment epithelium (RPE), striatum (St), sub-ventricular layer (SVL), telencephalon (Tel), thalamus (Th), third ventricle (TV), ventricular layer (VL), ventricular surface (VS), white matter (WM).
Corresponding with the developing human brain, Cep164 is expressed in the neuroepithelium of the murine telencephalon (129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+) (cerebral hemisphere), diencephalon, mesencephalon (midbrain), metencephalon (pons) and myencephalon (medulla oblongata) at P0.5 (Figure 3E.I.VI.VII.VIII, F.V). At later stages of development (P15.5-P30.5), clear Cep164 expression can be seen throughout the brain including the cerebrum and striatum of the cerebral cortex, the thalamus of the diencephalon, the midbrain and hindbrain pons and medulla oblongata (Figure 3G.I.V.VI.VII.VIII, H.I.V.VI.VII). Additionally, correlating with human foetal expression, Cep164 is expressed strongly in neuroepithelium lining the brain ventricles, including the third ventricle, fourth ventricle and cerebral aquaduct (Figure 3G.VI.VII).
At P0.5 to P15.5 Cep164 is expressed in the migrating molecular cell layer and Purkinje layer of the cerebellum (Figure 3F.I.II, G.I.II.III). Later in development, Cep164 is also expressed in the ganglion cell layer (P30.5) (Figure 3H.I.II.III), potentially this might be seen later on in human foetal development. Cep164 is also expressed at the apical membrane of the ventricular surface of the cerebellum at P0.5 and P15.5 (Figure 3F.III, G.III). Unlike the human cerebellar expression, Cep164 is expressed in the white matter of the cerebellum which is present, although reduced at P30.5 (Figure 3E.IV, F.VI, G.IV, H.IV). Cep164 is also expressed in the choroid plexus, ependymal cells, which is demonstrated in P0.5 sections (Figure 3E.I.II.III, F.IV). Murine controls demonstrate that there is no endogenous LacZ expression in the murine brain (S5 Figure F-H).
CEP164 expression during human and murine development in other tissues
CEP164 is expressed widely throughout the developing human embryo at 8PCW (Figure 4A.I.). In the 8PCW lung, CEP164 is expressed most strongly in the epithelial cells lining the lumen of the bronchi and bronchioles (Figure 4A.II.III). Weaker CEP164 expression is also present in the bronchiole smooth muscle, and the alveoli primordia (Figure 4A.II.III). In the 8PCW developing heart, CEP164 is expressed in cardiomyocytes (Figure 4A.IV.V). The gastrointestinal tract also shows strong defined CEP164 expression in the inner mucosa squamous epithelium cell layer, muscularis mucosae cell layer and the external muscularis externa layer at 8PCW (Figure 4A.VI). At 8PCW in the developing gonads, CEP164 is expressed weakly in the germline epithelium, but strongly in the seminiferous cord tubules (Figure 4A.VII.VIII). Additionally, CEP164 is expressed strongly in the dorsal root ganglia of the spinal cord, with weak expression in the developing vertebrae (Figure 4A.X). CEP164 is also expressed in developing bone primordia of the limbs at 8 PCW (Figure 4A.X).
Aveoli (Av), alveoli primordia (Av Pri), bile duct (BD), bone primordia (BP), bronchiole (Br), cardiomyocyte (CM), cartilage (Ca), connective tissue (CT), costal cartilage (CC), dorsal root ganglia (DRG), epithelial (Ep), gastrointestinal tract (GI), germline epithelium (GE), gonad (Go), heart (Ht), hepatocytes (Hep), intercostal muscle (IC), kidney (Ki), lamina propria (LM), liver (Li), lung (Lu), muscularis externa (ME), muscularis mucosae (MM), respiratory bronchiole (R), seminiferous cord (SeC), seminiferous tubule (ST), smooth muscle (SM), spermatids (Sp), spermatocytes (SC), spermatogonia (Spg), squamous cell mucosae (Mc-sq), submucosae (SB), trachea (Tr), terminal bronchi (TB), vertebrae (Vt).
Murine postnatal tissues also demonstrate widespread Cep164 expression. At P0.5 Cep164 is expressed in the developing costal cartilage primordia and intercostal muscles (Figure 4B.I.II.III). Correlating with the human lung, in the developing murine P15.5 and P30.5 lung, Cep164 is present in the epithelial cells lining the trachea, bronchioles and terminal/respiratory bronchioles, as well as the smooth muscle and cartilage of the trachea and bronchi. (Figure 4B.IV.V.VII.VIII, C.I.II.III.IV.V.VI).Weak Cep164 expression is seen in the alveoli at P15.5, which is reduced by P29.5 (Figure 4B.IX, C.VII). Like human embryonic tissues, Cep164 is present in cardiomyocytes at P15.5 and P30.5 (Figure 4B.X.XI, C.VIII.IX). In postnatal P15.5 murine testes, Cep164 is expressed in the connecting tubules and Leydig cells (Figure 4B.XII.XIII.XIV). In the seminiferous tubules, Cep164 is expressed in the smooth muscle cells and weakly in developing spermatogonia and spermatocytes. Strong Cep164 expression is present in the spermatid tails. By P30.5 Cep164 expression is defined to developing spermatogonia, spermatocytes, and spermatids (Figure 4C.X.XI.XII). At P15.5 and P30.5 Cep164 is expressed strongly in cells lining the ciliated cholangiocyte cells of the bile duct, however hepatocytes do not show any expression (Figure 4C.XV.XVI.XVII, D.XIII.XIV). Wholemount Cep164 expression studies (E9.5, E10.5, E12.5), demonstrate that Cep164 expression is widespread throughout murine embryonic development, including the bronchial arches and organ primordia, as well as the developing CNS, heart and limbs (S3 Figure). Taken together this indicates that CEP164 expression is widespread throughout human and murine development in tissues beyond the cerebello-retinal-renal structures associated with typical disease phenotypes.
Discussion
In this study we have described the expression of CEP164 in the developing human embryo and foetus, utilising immunohistochemistry, focusing upon the kidney, retina and cerebellum. Conservation of CEP164 expression was explored using a LacZ gene trap assay to characterise Cep164 expression in corresponding murine tissues (129/OlaHsd-Cep164tm1a(EUCOMM)Wtsi/+). Notably, during murine development, the kidney, retina and cerebellum continue to develop postnatally. This reflects the chosen murine postnatal timepoints in this study, corresponding to estimated human embryonic and fetal developmental timepoints (S2 Table, S3 Table, S4 Table).
Our results demonstrate that during human and murine development CEP164 is expressed widely, in multiple organs. Notably, CEP164 expression is clearly defined within tissues. In the developing human kidney, CEP164 is expressed in the apical epithelium membrane of metanephric-derived renal vesicles and subsequent nephron tubules (8PCW-18PCW), correlating with the presence of primary cilia [37]. Primary cilia are vital in the kidney for mechanosensation and chemical signal transduction, which is required for orientated cellular divisions during development and kidney maintenance. As CEP164 expression seems to be low/not present in the cap mesenchyme, it could be speculated that CEP164 expression is switched on during mesenchymal-epithelial transition, which coincides with the formation of the primary cilium during establishment of apical-basal polarity, cell junctions and lumen formation [37]. CEP164 is expressed within the glomerulus of the renal corpuscle, with a reduction in expression with maturity. Interestingly, this correlates with the loss of primary cilium in podocytes seen with glomeruli maturity in rats [38]. It could be postulated that in the developing human kidney, primary cilium, and thus CEP164 expression, is lost in maturing glomeruli [38]. This could protect against overstimulation of calcium-mediated signalling due to an increase in glomerular filtration rate [38]. CEP164 is also expressed in the uteric bud, which develops into the primary ciliated collecting duct network (8PCW-18PCW), consistent with numerous IMCD3 expression studies [1, 19, 20, 25, 32]. The human CEP164 expression pattern is conserved in the murine kidney (P0.5-P29.5); there may be cell-specific Cep164 expression within nephron tubular segments, however this would need to be further studied.
Although our data is focused upon development, the human protein atlas indicates that the CEP164 expression is maintained in the adult kidney [39]. A combination of P30.5 expression data and previous wholemount Cep164 expression studies also indicate that Cep164 expression is maintained in the adult mouse kidney [36]. Together this suggests that CEP164 may have roles in both normal renal development and maintenance of kidney function in the human and mouse. Aberrant CEP164 function in the kidney could contribute to abnormal primary cilium functioning, causing dysfunctions in cell divisions and cell signalling, leading to cystogenesis. This is consistent with the NPHP-RC phenotype present in most patients with CEP164 mutations [1].
In the human retina, CEP164 is expressed in the NFL, OPL, IPL and photoreceptor layer between 8PCW-18PCW. CEP164 expression in the GCL is lost prior to 14 PCW, whereas it is maintained in the murine retina (P0.5-P29.5). The murine retina also shows Cep164 expression in the ONBL primordial layer, with punctuate expression at the edges of the INL and ONL. Multiple studies have suggested that primary cilia are present and functional in neurones [40]. It could be speculated they are also present in the neurones seen in the NFL, OPL, IPL, GCL, ONBL, IBL and OBL, which would link with CEP164 expression. The human and murine retina have shown some differences in gene and protein expression in previous studies, potentially this is the case with CEP164 in the retina [41].
Retinal photoreceptor cells have a specialised connecting primary cilium between the inner and outer segment of the photoreceptors. It therefore seems reasonable that CEP164, a distal appendage centrosomal protein, is expressed in photoreceptor segment in the developing human and murine retina. CEP164 expression has been previously identified in the murine connecting cilium, further confirming our results [1, 42].
In the murine retina, Cep164 expression is defined to the inner segment of the photoreceptor layer by P15.5. This correlates with previous preliminary Cep164 In Situ hybridisation studies [43]. It can be hypothesised that the Cep164 is transcribed and translated in the inner segment but then the CEP164 protein is transported to the outer segment of the photoreceptor cells; this would not be detected by the X-Gal assay but could be defined in the human assay. Interestingly, the murine Cep164 expression pattern is maintained in the adult mouse. It is likely that human CEP164 expression is also maintained in the human adult retina, potentially with a role in retinal development and tissue functioning.
It is also well established that CEP164 is present at the basal body of retinal pigment epithelial cells, clarifying the results from our study [1, 2, 15, 17, 23]. It can be hypothesised that with abnormal functioning/loss of CEP164, there could be atypical outer segment formation, which could lead to the accumulation of phototransduction proteins that could trigger cell loss, leading to a retinal degeneration phenotype, as seen in some CEP164 NPHP-RC patients [1, 9, 44].
CEP164 is expressed throughout the developing human brain (8PCW-18PCW) and the developing murine brain (P0.5-P29.5), including the telencephalon, diencephalon, mesencephalon, metencephalon and myencephalon. In early human development (8 PCW) CEP164 is expressed widely throughout the cerebellum, including the white matter. By 16 PCW this expression is focused to the migrating molecular cell layer, however CEP164 expression is still strong in the axon tracts of the medulla oblongata and the apical neuroepithelium membrane surrounding the pons and cerebellum. At 18 PCW CEP164 staining in the molecular layer is considerably weaker. Human protein atlas studies indicate that CEP164 is present at low levels within the cerebral cortex, hippocampus and caudate, but is not expressed within the adult cerebellum, however these studies have not been validated [39].
Primary cilia are vital in the developing brain for Shh signalling, which is needed for proliferation of neuronal granular cell precursors in vertebrates, being a major driver of cerebellar precursors [45, 46]. Shh is also required for the cell-specific expansion of postnatal progenitors [40]. Wnt signalling, transduced by the primary cilia, is also required for neuronal patterning, cell proliferation and neuronal migration [46-48]. In the adult nervous system, primary cilia are thought to be required in stem cell regulation and tissue regeneration, however the full extent of primary cilia function in the adult central nervous system is still to be determined [40, 45]. It could be suggested that CEP164 has a functional role in human brain development. Potentially, aberrant function of CEP164, may lead to brain dysplasia via abnormal primary cilia functioning in neuronal precursors. Some CEP164 NPHP-RC patients show neurological phenotypes, including abnormal developmental delay, intellectual disability and in one patient, cerebellar vermis aplasia, an archetypal feature of Joubert syndrome. Another patient also experiences seizures [1, 9].
In the murine cerebellum, Cep164 expression is widespread including the migrating molecular, ganglion and Purkinje cell layers, as well as the white matter (P0.5-P29.5). Wholemount expression studies from the international phenotyping consortium have shown that Cep164 is expressed strongly throughout the adult murine brain [36]. It may be that Cep164 RNA expression does not directly correlate with CEP164 protein expression. Although there is believed to be high conservation between the human and murine transcriptomic networks, CEP164 may be part of the divergent pathway in the human-murine brain transcriptomics [46].
Leptin receptors are situated in the primary cilia on the choroid plexus, and in the hypothalamus [47]. Leptin signalling is vital in the hypothalamic satiety pathway. Patients with mutations in centrosomal BBS proteins demonstrate hyperleptinemia due to leptin resistance, which causes hyperphagia and subsequent weight gain. It could be hypothesised that CEP164, which is expressed in the developing choroid plexus of the human (8 PCW, 16 PCW, 18PCW) and murine brain (P0.5), may not only play a role in gross neurological development, but may also be involved indirectly in leptin receptor localisation or functioning within the primary cilium. This could contribute to the obesity phenotype seen in some CEP164 NPHP-RC patients [1, 9].
Ependymal cells of the choroid plexus, which contain motile cilia, are required for the production and regulation of cerebral spinal fluid (CSF). Additionally, the lateral ventricle epithelium contains motile cilia which is important for the movement of CSF throughout the brain ventricles. Previous studies have established a ciliogenesis role for CEP164 in the motile cilium and demonstrated CSF defects including hydrocephalus in zebrafish and murine models lacking a functioning CEP164 [24]. This further validates the strong CEP164 expression in ependymal cells and the neuroepithelium lining the brain ventricles.
CEP164 is also expressed in tissues secondary to the cerebellar-retinal-renal phenotype, some of which have motile cilium. Taking together results from a previous murine CEP164 motile cilia study [24], it is plausible that CEP164 is expressed in the ciliated epithelial cells of the trachea, bronchi and bronchioles in the human (8 PCW) and murine (P15.5-P29.5) lung. Previous studies indicate that this is likely to be maintained in the adult human and murine lung [36, 39]. Cep164 is also expressed in the murine spermatids, which have a flagellum, effectively a modified motile cilium [48]. Studies have demonstrated that CEP164 is expressed in the capitulum and striated columns of the human sperm neck [49]. Using information from a recent FOXJ1 mediated knockout of Cep164, it could be hypothesised that CEP164 is required for the formation, maintenance and functioning of motile cilia within the respiratory epithelium, reproductive tissues and the sperm flagella [24]. Mutations and subsequent aberrant functioning or loss of CEP164 may contribute to lung phenotypes such as pulmonary bronchiectasis, as seen in some patients with CEP164 mutations [1]. This may also contribute to aberrant spermatid function, which may cause infertility problems as seen typically in BBS models [9].
CEP164 is also expressed in other developing tissues that contain a primary cilium, (human 8 PCW, mouse P15.5-30.5); including cardiomyocytes of the heart, cholangiocytes of the liver bile ducts, smooth muscle cells, cartilage, spermatogonia and spermatocytes, bone primordia and the epithelium lining and smooth muscles of the GI tract [50-57]. Accordingly, CEP164 is not present in the hepatocytes, which do not contain primary cilium [58].
CEP164 is expressed in tissues involved in the CEP164 NPHP-RC phenotype, but is also expressed in tissues not associated with this phenotype. It could be that CEP164 has cell-specific functions, not all tissues may rely on the primary cilium for development or have other pathways that can compensate for the abnormal functioning or loss of CEP164. Remarkably, human CEP164 expression patterns demonstrated here correlate with previously reported expression patterns of the JBTS genes AHI1 and CEP290 [59]. This could indicate a universal mechanism that underlies NPHP-RC.
In summary, CEP164 demonstrates widespread yet defined expression throughout human and murine development, which is predominantly maintained into adult life. Human and murine data largely correlate and CEP164 function is likely to be conserved between the two species. CEP164 is expressed in tissues affected in CEP164-NPHP-RC patients, however clinical heterogeneity, commonly seen in ciliopathies, needs to be further investigated.
Supporting Information Captions
S1 Table. Working dilutions of primary antibodies used for immunohistochemistry of human tissues
S2 Table. Comparison of human and murine kidney developmental timeline
S3 Table. Comparison of human and murine retina developmental timeline
S4 Table. Comparison of human and murine cerebellar development timeline
S1 Figure. Conservation and protein domains of human CEP164
(A) Predicted human CEP164 protein domains; tryptophan-tryptophan (WW) domain conserved with two Tryptophan (W) residues, Lysine-rich repeat (LR) and predicted coiled-coil (CC) domains. Values marked are amino acid number. (B) Sequence alignment of human CEP164 and its orthologs in M.musculus, D.melanogaster, C. reinhardtii and D rerio. (C) Unrooted phylogenetic tree of CEP164 orthologs. (D) Taxonomic profiles of CEP164 derived from EggNOG4.5.1 showing absence of CEP164 in nematoda taxa.
S2 Figure. Diagram of the 129/OlaHsd-Cep164tm1a allele
Upon pre-mRNA splicing of Cep164tm1a, exon 3 splices into the splice acceptor (SA) of the LacZ cassette, causing a frameshift and subsequent formation of a premature termination codon, this forms the tm1b allele. The LacZ has an internal ribosomal entry site (IRES), and thus the LacZ fusion gene acts as a reporter gene for Cep164.
Internal ribosomal entry site (IRES), splice acceptor (SA), polyadenylation site (pA).
S3 Figure. CEP164 wholemount expression throughout murine embryonic development and corresponding WT littermate controls
At E9.5 (A), Cep164 widespread expression is seen, including the branchial arches, developing forebrain, midbrain and hindbrain (A.I). There is also Cep164 expression seen in the developing neural tube, including the neuroepithelium surrounding the anterior neuropore (A.II). There is Cep164 expression seen in the developing heart including the central ventricle, bulbous cordis and outflow tract (A.III), as well as the developing limb buds (A.IV). This Cep164 expression pattern is maintained at E10.5 (B), Cep164 expression is seen in the optic cup and olfactory pit (B). At E12.5 Cep164 expression is widespread but is defined in the spinal cord, vertebrae, cerebral cortex, midbrain, hindbrain, pons and medulla oblongata, otocyst and heart (C.I.II.III). Cep164 expression is seen in the retina (C.IV) and at the tips of the developing digits, where the apical ectodermal ridge is present (C.V.VI.VII.VIII). WT littermates do not show endogenous beta galactosidase expression at E9.5 (D.I), E10.5 (D.II), and E12.5 (F).
Anterior nucleopore (ANP), apical ectodermal ridge (AER), branchial arch (BA), bulbous cordis (BC), cerebral cortex (CC), common atria (CA), common ventricle (CV), forebrain (FB), forelimb bud (FL), heart (Ht), hindbrain (HB), hindlimb bud (HL), medulla oblongata (MO), midbrain (MB), neural tube (NT), olfactory pit (OF), optic cup (OC), organ primordia (Pri), otocyst (OT), outflow tract (OFT), pons (PO), somites (SM), spinal cord (SC), vertebral column (VC).
S4 Figure. Human CEP164 expression controls
No primary antibody controls in the 8 PCW Kidney (A) and 18 PCW (B) kidney. Renal vesicles, comma-shaped vesicles, S-Shaped body, uteric bud and cap mesenchyme (A.I.II.III, B.I.II.III). Renal tubule and the renal corpuscle (A.IV.V.VII, B.IV.V.VI), Loop of Henle and collecting ducts (B.VII). Lung, including the bronchioles and alveoli (C.I.II.III), cardiomyocytes of the heart (C.I.IV,V), gastrointestinal tract (C.VI) and gonads (C.VII.VIII). Dorsal root ganglia (C.VIX) and bone primordia (C.VX), liver (C.IX). Developing brain (D.I, E.I) including the telencephalon (C.VI), metencephalon (C.V) and myencephalon (C.VIII). Midline (D.II), and neuroepithelium surrounding the brain ventricles (C.VIII, D.III.IV). Choroid plexus ependymal cells (C.II.III.IV) and choroid plexus pia matter (C.III.IV). Molecular cell layer, Purkinje and ganglion cell layers of cerebellum (F.I.II). Ventricular surface (F.III.IV), medulla oblongata (F.VI), and white matter (F.VII). Choroid plexus at 16 PCW (F.V). PAX9 positive control antibody (E.V, F.IX).
Alveoli primordia (Av Pri), bone primordia (BP), bronchiole (Br), cap mesenchyme (CM), cardiomyocyte (CM), cerebellum (Ce), choroid plexus (CP), collecting duct (CD), cortical plate (CorP), diencephalon (Di), dorsal root ganglia (DRG), ependymal cells (Ep), ganglion cell layer (GCL), gastrointestinal tract (GI), germline epithelium (GE), gonad (Go), intermediate Zone (IL), heart (Ht), hepatocytes (Hep), fourth ventricle (FV), kidney (Ki), lateral ventricle (LV), lamina propria (LM), liver (Li), loop of henle (LH), lung (Lu), marginal Layer (MaL), medulla oblongata (MO), mesencephalon (Mes), metencephalon (Met), midline (Mi), molecular Cell Layer (MCL), muscularis externa (ME), muscularis mucosae (MM), myencephalon (Mye), nasal epithelium (Na), pia (P), post conception weeks (PCW), purkinje cell layer (PCL), renal corpuscle (RC), renal tubules (RT), renal vesicle (RV), seminiferous cord (SC), s-shaped body (SSB), submucosae (SB), sub-ventricular layer (SVL), squamous epithelium mucosae (Mc-sq), telencephalon (Tel), uteric bud (UB), ventricular layer (VL), ventricular surface (VS), white matter (WM).
S5 Figure. Littermate WT controls for murine Cep164 expression
(A) murine retina. (B-C) murine kidney (D) murine secondary tissues, (E-F) murine cerebellar tissues. Retina at P1.5 (A.I), P15.5 (A.II), and P29.5 (A.III). Renal vesicles (B.I.II.III), uteric bud (B.III), renal tubules (B.IV) and renal corpuscle (B.V.VI). Renal tissue at P30.5 has low endogenous beta galactosidase expression (C.I) in the renal tubules (C.III.IV), but not in the renal corpuscle (C.II). Developing brain (D.I), costal cartilage (D.II) and intercostal muscle (D.III). The P0.5 lung (E.I.II.III.IV), and alveoli (E.V), P30.5 murine lung trachea, bronchioles or alveoli (E.I.II.III.IV.V.VI). Murine heart, cardiomyocytes (D.VI.VII, E.VI.VII), developing testes, spermatogonia, spermatocytes, spermatids and smooth muscle cells in the seminiferous tubules (D.VIII.IX.X, E.VIII.IX.X). Hepatocytes of the liver, hepatic portal vein (D.XI.XII.XIII, E.XI.XII.XIII). Murine brain cerebral hemisphere (F.I.V, G.I,H.I), the midbrain (F.VI,G.VII,,H.V), the striatum (G.V,) and thalamus (G.VI),H.VI).The choroid plexus shows some weak endogenous beta galactosidase expression (F.IV). Ganglion cell layer, molecular cell layer, Purkinje cell layer and white matter of the cerebellum (F.II.III.IV, G.II.III.IV, H.II.III).
Alveoli (Av), bronchiole (Br), cardiomyocyte (CM), cartilage (Ca), cerebellum (Ce), cerebral hemisphere (Ch), cerebrum (Cr), choroid plexus (CP), connective tissue (CT), costal cartilage (CC), epithelial (Ep), ganglion cell layer (GCL), heart (Ht), hepatocytes (Hep), inner nuclear layer (INL), inner plexiform layer (IPL), intercostal muscle (IC), liver (Li), lung (Lu), marginal layer (MaL), medulla oblongata (MO), midbrain (MB), molecular cell layer (MCL), outer neuroblastic layer (ONBL), outer nuclear layer (ONL), outer plexiform layer (OPL), photoreceptor segment layer (PS), pons (PO), postnatal day (P), purkinje cell layer (PCL), renal corpuscle (RC), renal tubule (RT), renal vesicle (RV), respiratory bronchiole (R), retinal pigment epithelium (RPE), seminiferous tubule (ST), smooth muscle (SM), spermatids (Sp), spermatocytes (SC), spermatogonia (Spg), striatum (St), terminal bronchi (TB), thalamus (Th), trachea (Tr), uteric bud (UB), white matter (WM).
Acknowledgements
L.A.D is funded by the Medical Research Council Discovery Medicine North Doctoral Training Partnership, and The Northern Counties Kidney Research Fund.