Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Jul;9(7):890-905.
doi: 10.15252/emmm.201606430.

Loss of Mpdz Impairs Ependymal Cell Integrity Leading to Perinatal-Onset Hydrocephalus in Mice

Free PMC article

Loss of Mpdz Impairs Ependymal Cell Integrity Leading to Perinatal-Onset Hydrocephalus in Mice

Anja Feldner et al. EMBO Mol Med. .
Free PMC article


Hydrocephalus is a common congenital anomaly. LCAM1 and MPDZ (MUPP1) are the only known human gene loci associated with non-syndromic hydrocephalus. To investigate functions of the tight junction-associated protein Mpdz, we generated mouse models. Global Mpdz gene deletion or conditional inactivation in Nestin-positive cells led to formation of supratentorial hydrocephalus in the early postnatal period. Blood vessels, epithelial cells of the choroid plexus, and cilia on ependymal cells, which line the ventricular system, remained morphologically intact in Mpdz-deficient brains. However, flow of cerebrospinal fluid through the cerebral aqueduct was blocked from postnatal day 3 onward. Silencing of Mpdz expression in cultured epithelial cells impaired barrier integrity, and loss of Mpdz in astrocytes increased RhoA activity. In Mpdz-deficient mice, ependymal cells had morphologically normal tight junctions, but expression of the interacting planar cell polarity protein Pals1 was diminished and barrier integrity got progressively lost. Ependymal denudation was accompanied by reactive astrogliosis leading to aqueductal stenosis. This work provides a relevant hydrocephalus mouse model and demonstrates that Mpdz is essential to maintain integrity of the ependyma.

Keywords: aqueductal stenosis; cerebrospinal fluid; ependymal cells; hydrocephalus; tight junction.


Figure 1
Figure 1. Generation of global and conditional Mpdz knockout mice

Schematic representation of the Mpdz gene locus and the translated protein. The numbered boxes represent exons 1–47 of the wild‐type allele. The relative position of the translational start and stop sites is indicated at exon 2 and exon 47, respectively. A gene‐trap cassette (β‐geo) was inserted into intron 11–12 leading to a stop signal that truncates the Mpdz protein after the third PDZ domain.

Western blotting to detect Mpdz protein in brain lysates of littermate wild‐type and global knockout Mpdz mice.

Schematic drawing of the conditional Mpdz gene targeting strategy. LoxP sites flanking exon 4 and a neomycin resistance cassette were inserted by homologous recombination. Transgenic mice were crossed with Flp deleter mice to remove the FRT‐flanked neomycin cassette. Cre recombinase removes exons 4 and 5 leading to a frameshift and a nonsense mutation. This truncates the Mpdz protein after the L27 domain.

Western blotting to detect Mpdz protein in brain lysates of Mpdz fl/fl and CMV‐Cre;Mpdz Δ/Δ mice.

Kaplan–Meier survival analysis of Mpdz −/− vs. Mpdz +/+ mice (n > 41 mice per genotype) and CMV‐Cre;Mpdz Δ/Δ vs. Mpdz fl/fl mice (n > 48 mice per genotype).

Representative images of Mpdz −/− and Mpdz +/+ mice at postnatal day 27 (P27) and CMV‐Cre;Mpdz Δ/Δ and Mpdz fl/fl mice at P24. Arrows indicate the enlarged and dome‐shaped skull and the enlarged brain hemispheres.

Source data are available online for this figure.
Figure 2
Figure 2. Mpdz‐deficient mice develop hydrocephalus

At postnatal day 27 (P27), Mpdz −/− and wild‐type littermates were subjected to a computed tomography. Three‐dimensional reconstruction shows macrocephaly and thinning of skull bones (arrows).

T2‐weighted coronal and sagittal magnetic resonance images of the head of Mpdz −/− vs. Mpdz +/+ mice at P27. CSF in the enlarged lateral ventricles appears hyperintense (asterisks).

Figure 3
Figure 3. Mpdz −/− mice develop postnatal hydrocephalus
H&E staining of brain sections from Mpdz −/− and littermate Mpdz +/+ mice at different developmental stages. At embryonic stage E14.5 and at birth (P0), coronal sections showed no overt alterations in Mpdz −/− mouse brains. At P3 and P7, enlarged lateral ventricles (*) were detected in Mpdz −/− mice. Horizontal sections demonstrate ventriculomegaly in Mpdz −/− mice at P3 and P7. Aq, cerebral aqueduct; Hi, hippocampus; LV, lateral ventricle; V, ventricle (3rd, 4th). Scale bars: E14.5, 500 μm; P0, P3, and P7, 1 mm.
Figure EV1
Figure EV1. Loss of Mpdz does not alter the brain microvasculature and endothelial‐restricted loss of Mpdz does not cause hydrocephalus

Brain sections from Mpdz −/− and littermate Mpdz +/+ mice were stained for the endothelial marker CD31. Scale bars, 50 μm. Graph showing relative changes in microvessel density. n = 3 animals per genotype; n.s., not significant (unpaired two‐sided Student's t‐test). Data are presented as mean ± SD.

Brain section stained for the tight junction proteins claudin‐5 and ZO‐1. No differences were observed between Mpdz −/− and littermate Mpdz +/+ mice at P7. Scale bar, 50 μm.

Transmission electron micrograph of cortical brain capillaries shows no difference in tight junction assembly between Mpdz −/− and Mpdz +/+ littermates. Boxes are highlighting endothelial cell junctions. One representative box shows the area as a zoom‐in. Scale bars, 1 μm and 100 nm. EC, endothelial cell; PC, pericyte.

Western blot to determine Mpdz protein expression in isolated lung endothelial cells from adult Tie2‐Cre;Mpdz ΔEC/ΔEC mice.

Endothelial cell‐specific Mpdz‐deficient mice (ΔEC/ΔEC) developed normally with no indication for hydrocephalus (n > 200 ΔEC/ΔEC observed for more than four months of age). H&E staining of brain sections from three‐month‐old mice.

Source data are available online for this figure.
Figure EV2
Figure EV2. Normal morphology of choroid plexus, liver, and kidney in Mpdz −/− mice

Scanning electron micrograph of choroid plexus epithelium of Mpdz −/− and Mpdz +/+ littermate mice at P7 shows no morphological abnormalities in cilia or microvilli. Scale bars, 3 μm.

Electron microscopic analysis of the choroid plexus from a lateral ventricle of Mpdz −/− and Mpdz +/+ littermates at postnatal day 3 and day 7. The plexus epithelial cells are typically polarized, carry apical microvilli (asterisks), and are interconnected by tight junctions (arrows). In the stroma, beneath the epithelium the fenestrated capillaries are seen (hashtags; for comparison, see Wolburg & Paulus, 2010). Scale bars indicated in images.

Kidney and liver of Mpdz −/− and Mpdz +/+ littermates at P7 were embedded in paraffin. H&E staining of sections shows no overt differences between the genotypes. Scale bar, 50 μm.

Figure 4
Figure 4. Impaired flow of CSF through the cerebral aqueduct after postnatal day 3

Flow of Evans blue dye injected into a lateral ventricle trough the ventricular system of Mpdz −/− and Mpdz +/+ littermates was analyzed at postnatal day 1 (P1), P3, and P5. Sagittal sections show the lateral (LV) and fourth ventricles (4th V, arrowhead). In mice at P1, flow of Evans blue from one lateral ventricle through the third into the contralateral and the fourth ventricles was unremarkable in Mpdz −/− and Mpdz +/+ mice. At P3 and P5, no flow of Evans blue through the cerebral aqueduct (Aq) into the fourth ventricle could be detected in Mpdz −/− mice.

Scanning electron micrograph of ependymal cells lining the roof of lateral ventricles at P7. No morphological alterations in motile cilia of Mpdz −/− compared to Mpdz +/+ mice. Scale bar, 20 μm.

Antibody staining against cleaved caspase‐3 (white color, arrows and arrowheads) to detect apoptotic cells. CP, choroid plexus; SVZ, subventricular zone. Scale bar, 100 μm.

Figure 5
Figure 5. Ependymal defects in the cerebral aqueduct of Mpdz −/− mice
H&E staining of horizontal brain sections from of Mpdz −/− and Mpdz +/+ littermates at postnatal days 0, 3, and 7 (P0, P3, and P7). A continuous single layer of ependymal cells lines the aqueduct in controls. Disruption of the ependymal cell layer was frequently observed in Mpdz‐deficient mice. The images at the right show areas of interest (boxed) in higher magnification. Arrows indicate regions where the ependymal layer is disturbed. Please note that images of Mpdz +/+ brains (P0, P3) are close‐ups of the images shown in Fig 3. Scale bars are indicated in images.
Figure EV3
Figure EV3. Mpdz inactivation using Nestin‐Cre mice causes hydrocephalus
Floxed Mpdz mice were crossed with Nestin‐Cre to delete Mpdz in glial and neuronal precursors.

Dome‐shaped skull (arrow) and the enlarged brain hemispheres (arrow) at postnatal day 7.

Lateral ventricles (LV) were enlarged (asterisk). Scale bar, 1,000 μm.

Figure 6
Figure 6. Loss of MPDZ increases epithelial cell permeability

Representative graph of transepithelial electrical resistance (TER; upper graph) and corresponding capacitance (Ccl; lower graph) of MCF‐7 monolayers for 5 days after silencing of MPDZ expression. The reduction of MPDZ expression led to a significant decrease of TER indicating increased permeability. The low corresponding capacitance (Ccl) values indicate formation of a dense cell monolayer. n = 3 technical replicates.

Mean TER values of five (siRNA) or three (shRNA) independent experiments with each three replicates at time point 72 h. **P = 0.008; *P = 0.0113 by two‐sided, unpaired Student's t‐test.

Representative graph of TER and corresponding capacitance (Ccl) of HIPCC cell monolayers for 5 days after silencing of MPDZ expression. n = 3 replicates.

Levels of active RhoA (RhoA‐GTP) and total RhoA levels in isolated astrocytes derived from brains of neonatal Mpdz −/− and Mpdz +/+ mice were detected by G‐LISA. Graph shows ratio of active RhoA vs. total RhoA. n = 3; P = 0.0171 by two‐sided, unpaired Student's t‐test).

Data information: All data are presented as mean ± SD.Source data are available online for this figure.
Figure 7
Figure 7. Diminished expression of Pals1 in ependymal cells of Mpdz −/− mice
Immunostaining for the planar cell polarity proteins Pals1 and Crb3 (white color) and the adherens junction protein E‐cadherin (brown color) in brains of Mpdz‐deficient mice at postnatal days 0, 3, and 7 (P0, P3, and P7).

Expression of Pals1 (white color, arrows) is absent (*) or diminished in Mpdz −/− mice. Scale bars: 10 μm for P0, 20 μm for P3, 10 μm for P7.

Expression of Crb3 is not altered in Mpdz −/− compared to control mice. Scale bars: 10 μm for P0, 20 μm for P3, 10 μm for P7.

The dynamic expression pattern of E‐cadherin is not altered in Mpdz −/− compared to control mice. Scale bars: 100 μm and for zoom‐ins 10 μm.

Data information: CP, choroid plexus; EP, ependymal cells; LV, lateral ventricle.
Figure 8
Figure 8. Hydrocephalus in Mpdz‐deficient mice is preceded by reactive gliosis
Immunostaining for glial fibrillary acidic protein (GFAP; white color, arrows).

Progressive astrogliosis in the subventricular zone (SVZ) of Mpdz −/− mice. CP, choroid plexus. Scale bars: 10 μm for P0, 40 μm for P3, 50 μm for P7.

Progressive astrogliosis (indicated by arrow) around the cerebral aqueduct of Mpdz −/− mice. Scale bars: 100 μm.

Progressive astrogliosis in the hippocampal region of Mpdz −/− mice. Scale bars: 100 μm for P0, 50 μm for P3 and P7.

Figure EV4
Figure EV4. Reactive astrogliosis in Mpdz‐deficient mice
Staining of glial fibrillary acidic protein (GFAP, DAB stain: brown color) at postnatal days 0, 3, and 7 (P0, P3, and P7) on paraffin‐embedded brain sections. GFAP expression is more intense and widespread in Mpdz‐deficient mice compared to wild‐type littermate controls. Boxed areas are shown in higher magnification at right row. Please note that the Mpdz −/− brain section (P3) is derived from the same specimen than in Fig 3. SVZ, subventricular zone; CP, choroid plexus; EP, ependymal cells. Scale bars: 500 μm left row, 200 μm middle row, 20 μm right row.

Similar articles

See all similar articles

Cited by 11 articles

  • Functional loss of Ccdc1 51 leads to hydrocephalus in a mouse model of primary ciliary dyskinesia.
    Chiani F, Orsini T, Gambadoro A, Pasquini M, Putti S, Cirilli M, Ermakova O, Tocchini-Valentini GP. Chiani F, et al. Dis Model Mech. 2019 Aug 2;12(8):dmm038489. doi: 10.1242/dmm.038489. Dis Model Mech. 2019. PMID: 31383820 Free PMC article.
  • DAPLE and MPDZ bind to each other and cooperate to promote apical cell constriction.
    Marivin A, Garcia-Marcos M. Marivin A, et al. Mol Biol Cell. 2019 Jul 22;30(16):1900-1910. doi: 10.1091/mbc.E19-02-0091. Epub 2019 Jul 3. Mol Biol Cell. 2019. PMID: 31268831 Free PMC article.
  • Linking Cell Polarity to Cortical Development and Malformations.
    Hakanen J, Ruiz-Reig N, Tissir F. Hakanen J, et al. Front Cell Neurosci. 2019 Jun 4;13:244. doi: 10.3389/fncel.2019.00244. eCollection 2019. Front Cell Neurosci. 2019. PMID: 31213986 Free PMC article. Review.
  • The RNA-Protein Interactome of Differentiated Kidney Tubular Epithelial Cells.
    Ignarski M, Rill C, Kaiser RWJ, Kaldirim M, Neuhaus R, Esmaillie R, Li X, Klein C, Bohl K, Petersen M, Frese CK, Höhne M, Atanassov I, Rinschen MM, Höpker K, Schermer B, Benzing T, Dieterich C, Fabretti F, Müller RU. Ignarski M, et al. J Am Soc Nephrol. 2019 Apr;30(4):564-576. doi: 10.1681/ASN.2018090914. Epub 2019 Mar 13. J Am Soc Nephrol. 2019. PMID: 30867249 Free PMC article.
  • Identification of genes required for eye development by high-throughput screening of mouse knockouts.
    Moore BA, Leonard BC, Sebbag L, Edwards SG, Cooper A, Imai DM, Straiton E, Santos L, Reilly C, Griffey SM, Bower L, Clary D, Mason J, Roux MJ, Meziane H, Herault Y; International Mouse Phenotyping Consortium, McKerlie C, Flenniken AM, Nutter LMJ, Berberovic Z, Owen C, Newbigging S, Adissu H, Eskandarian M, Hsu CW, Kalaga S, Udensi U, Asomugha C, Bohat R, Gallegos JJ, Seavitt JR, Heaney JD, Beaudet AL, Dickinson ME, Justice MJ, Philip V, Kumar V, Svenson KL, Braun RE, Wells S, Cater H, Stewart M, Clementson-Mobbs S, Joynson R, Gao X, Suzuki T, Wakana S, Smedley D, Seong JK, Tocchini-Valentini G, Moore M, Fletcher C, Karp N, Ramirez-Solis R, White JK, de Angelis MH, Wurst W, Thomasy SM, Flicek P, Parkinson H, Brown SDM, Meehan TF, Nishina PM, Murray SA, Krebs MP, Mallon AM, Lloyd KCK, Murphy CJ, Moshiri A. Moore BA, et al. Commun Biol. 2018 Dec 21;1:236. doi: 10.1038/s42003-018-0226-0. eCollection 2018. Commun Biol. 2018. PMID: 30588515 Free PMC article.
See all "Cited by" articles


    1. Abouhamed M, Grobe K, San IVLC, Thelen S, Honnert U, Balda MS, Matter K, Bähler M (2009) Myosin IXa regulates epithelial differentiation and its deficiency results in hydrocephalus. Mol Biol Cell 20: 5074–5085 - PMC - PubMed
    1. Adachi M, Hamazaki Y, Kobayashi Y, Itoh M, Tsukita S, Furuse M, Tsukita S (2009) Similar and distinct properties of MUPP1 and PATJ, two homologous PDZ domain‐containing tight‐junction proteins. Mol Cell Biol 29: 2372–2389 - PMC - PubMed
    1. Al‐Dosari MS, Al‐Owain M, Tulbah M, Kurdi W, Adly N, Al‐Hemidan A, Masoodi TA, Albash B, Alkuraya FS (2013) Mutation in MPDZ causes severe congenital hydrocephalus. J Med Genet 50: 54–58 - PubMed
    1. Assémat E, Bazellières E, Pallesi‐Pocachard E, Le Bivic A, Massey‐Harroche D (2008) Polarity complex proteins. Biochim Biophys Acta 1778: 614–630 - PubMed
    1. Assémat E, Crost E, Ponserre M, Wijnholds J, Le Bivic A, Massey‐Harroche D (2013) The multi‐PDZ domain protein‐1 (MUPP‐1) expression regulates cellular levels of the PALS‐1/PATJ polarity complex. Exp Cell Res 319: 2514–2525 - PubMed