| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,1
* Experimental Genetics Group, Department Human Genetics, K.U.Leuven, Leuven, Belgium;
Section Neurobiology, Swammerdam Institute of Life Sciences, Amsterdam, The Netherlands;
Department of Anatomy and Cell Biology, Rheinland-Westfalen Technischen Hochschule University of Aachen, Aachen, Germany;
Department of Neuroscience, Medical School, Osaka, Japan; and
|| Netherlands Institute for Brain Research, Amsterdam, The Netherlands
2Correspondence: Experimental Genetics Group, Department Human Genetics, KULeuven - Campus Gasthuisberg ON1–06.602, B-3000 Leuven, Belgium. E-mail: fredvl{at}med.kuleuven.be
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: transgenic mice • neurogenesis • cognition • primary cultures
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2
3)
). The isoforms differ by having none, one or two N-terminal inserts and either three or four C-terminal microtubule binding repeat domains. The three-repeat (3R) tau isoform predominates in early developmental stages in both humans and rodents. But whereas tau-3R and tau-4R levels are about equal in adult human brain, a practically complete switch to tau-4R expression occurs in rodent brain after the second postnatal week (4
5
6)
.
The biological significance of differential tau isoform expression may be inferred from their respective regulation during development and from the role of tau-4R in the neurodegenerative disorders known as tauopathies. Of particular interest are the familial forms of frontotemporal dementia (FTD), caused by intronic tau mutations that result in perturbed tau isoform ratios rather than expression of mutant tau. FTD and many different neurodegenerative diseases that manifest late in life are characterized by neurofibrillar inclusions of highly phosphorylated protein tau (2
, 7)
.
Transgenic mouse strains expressing human tau have previously been generated to study the function of wild-type and mutant tau and their contributions toward tauopathy and neurodegeneration (7)
. Expression of human tau (htau) by a genomic DNA construct on a null tau murine background resulted in progressive hyperphosphorylation and tau aggregation in aged mouse brain (6)
. On the other hand, expression of human tau-4R (htau-4R) or htau-4R(P301L) on a murine tau background have been shown to cause mutually exclusive pathologies in a dose-dependent manner (8
, 9)
. Both gene dosage and tau isoform ratios may thus influence pathology.
In this study, we describe a simplified mouse model, wherein the endogenous murine tau gene has been inactivated and replaced by a single copy of the htau-4R/2N isoform (knockin/knockout). The mice express htau-4R at physiological levels, allowing us to study isoform specific functions of tau-4R in neurodevelopment and disease.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, 9)
. The excised insert was ligated in the BamHI site of the pBluescript vector, followed by ligation into a unique SmaI site of a 1.9 kb NotI fragment encoding the hygromycin B phosphotransferase gene driven by the phosphoglycerate kinase promoter (10)
. The total insert of this construct was excised to remove vector sequences and ligated in the unique NcoI site in exon1 of the mouse tau gene, partially subcloned from the 129 mouse strain (11)
. The targeting vector was linearized by NotI restriction and electroporated into embryonic stem (ES) cells that were cultured and genotyped as in (12)
. ES cells were microinjected into 4 d old blastocysts from C57/Bl6 mice. Reimplantation into pseudo-pregnant CD1 females resulted in coat-color chimeras, proving germline transmission. Chimeric offspring were crossed with FVB/N mice and genotyped.
Stereological analysis
Total numbers of cells in specified brain regions were quantified in 30 µm saggital cryostat sections stained with cresyl violet. Every 10th section was analyzed in a series of systematically sampled hippocampal sections that totaled 10–15 sections per brain. For each genotype, 7 mice at age 6 mo were analyzed.
Neuronal number quantification was performed by estimating the volume of the region through the point-counting method and Cavalieri’s principle. Neurons were counted by registering cells within three-dimensional optical dissectors systematically and randomly spaced throughout the region. Unbiased neuron number estimates were obtained from the cell counts using the fractionator principle (13)
. Stereological analysis was performed on a stereology system (CAST-Grid; Olympus, Glostrup, Denmark) consisting of a microscope, motorized specimen stage, microcator, CCD video camera, computer, and stereology software (14)
. Postprocessing section thickness was measured at each dissector location. Cells with typical neuronal morphology, including clearly delineated nucleolus, were counted within the optical dissector frame at x100. For hippocampal pyramidal cells, borders were evident between subiculum and dentate gyrus cells. Hippocampal granular cells were clearly delineated in and from the molecular and hilar regions of the dentate gyrus. Statistical analyses were carried out with GraphPad Prism 3.0 (GraphPad Software, San Diego, CA, USA).
Behavioral testing
At 5 and 9 wk of age, mice were subjected to behavioral tests. General motor ability was assessed by rotarod testing, using a revolving, horizontal rod of 3.2 cm diameter (Med Associates, Georgia, Vermont, USA). After one (5-wk-old mice) or two (9-wk-old mice) training sessions of 5 min at 16 revolutions per minute (rpm), the mice were placed on the rod, and the speed was increased from 4 to 40 rpm over 3 min. The time that the mice stayed on the rod was recorded.
Exploratory and motor activities were determined in an open field setting. Each mouse was placed for 5 min on an elevated, white opaque plastic surface of 52 x 52 cm without bordering walls. The travel path was recorded and analyzed using dedicated software (Ethovision-Noldus, The Netherlands). The center of the open field was defined as an inner square of 40 x 40.
Hippocampus-dependent learning and memory were analyzed using the novel object recognition test (NORT), as described previously (12)
. Individual mice were habituated on day 1 for 2 x 5 min in a box (52x52x40 cm), with black walls and a white opaque floor illuminated from below. On day 2, mice were familiarized for 10 min with two identical objects placed in adjacent quadrants of the box. Travel paths and the time the mice spent exploring an object with its snout directed toward the object within nostril reach were recorded. In the memory retention trial of 10-min duration, the mice were tested with one familiar and one novel object after a delay of either 1 or 3.5 h. The novel object and its position were randomized to avoid preferences not based on novelty. The level of discrimination (d2) was defined as d2 = (b–a)/(b+a), wherein a and b are the exploration times spent on old and novel objects. Statistical analyses were performed with 2-way ANOVA.
Electrophysiological analysis
Mice were decapitated, brains were removed, and kept at 4°C in artificial cerebrospinal fluid (ACSF) as described by Boekhoorn et al. (15)
. For Schaffer collateral recordings, 400 µm transversal hippocampal sections were cut with a tissue chopper; for perforant path recordings, 400-µm horizontal forebrain sections were cut with a vibroslicer. After 1 h of incubation at room temperature in oxygenated recording ACSF, slices were transferred to a recording chamber and perfused with oxygenated ACSF at 31.5°C. Bipolar stimulating electrodes isolated with stainless steel were placed in the Schaffer collaterals or the perforant path to record field excitatory postsynaptic potentials (fEPSPs) using a glass microelectrode (2–5 M
filled with ACSF) placed in the stratum radiatum of the CA1 or the middle third of the molecular layer of the DG. To evoke robust LTP in the DG, GABA-mediated activity was blocked with 10-µM bicuculline methiodidine (Sigma, St. Louis, MO, USA) in the ACSF. Before baseline recording, the maximal fEPSP amplitude and slope were determined by increasing the stimulus intensity until the response was saturated. The relationship between the stimulus intensity and the evoked response was fit to a Boltzmann equation: R(i) = Rmax/(1 + exp(I– ih)/(S)), wherein R(i) is the response at stimulus intensity (i), Rmax is the maximal response, ih is the intensity at which the half-maximal response is observed and slope factor S is the index that describes the slope of the stimulus-response curve. ih was used to record baseline responses for at least 20 min. Paired-pulses with interstimulus intervals of 50, 100, 200, and 500 ms were tested in the DG to make sure the medial perforant path was stimulated. LTP was evoked in the CA1 and the DG, using a theta burst protocol consisting of two trains of four pulses at 100 Hz, 200 ms apart. The procedure was repeated five times with an interval of 30 s. After theta burst stimulation, LTP was recorded for 60 min. Statistical analyses were performed with 2-way ANOVA.
Antibodies
Antibody HT-7 (Innogenetics, Gent, Belgium) specifically recognizes human tau, whereas antibody JN-RF.5 (generous gift from M. Mercken) reacts only with murine tau. Antibodies R2 (5)
and RD4 (U.S. Biologicals, Swampscott, MA, USA) are specific for the second repeat domain and thereby for tau-4R isoforms. Antibody tau-5 (BD, Brussels, Belgium) detects all tau species. The BrdU antibody (Abcam, Cambridge, UK) detects cells that have passed through the DNA synthesis phase of the cell cycle. Antibodies against nestin (Abcam) and NeuN (Chemicon, Leuven, Belgium) detect nonproliferating precursors and differentiated neurons, respectively. The doublecortin antibody (C-18; Santa Cruz, USA) detects young, migratory neurons. SMI-312 is a mixture of monoclonal antibodies (Sternberger, Lutherville, MD, USA) that stains medium and high molecular weight neurofilaments in axons.
Immunohistochemistry
All mice were anesthetized and transcardially perfused with ice-cold saline. Brains were immersion fixed in 4% paraformaldehyde for 24 h and then sectioned on vibratome (40 µm) or embedded in paraffin and sectioned serially (8 or 10 µm).
At P30, mice were injected intraperitoneally with 5 mg/ml BrdU in 0.9% saline at 50 mg/kg for seven consecutive days. Four weeks later, they were sacrificed, and BrdU-labeled nuclei were visualized as described by (16)
. After quenching, formic acid and HCl treatment, free-floating sections were stained with the BrdU antibody (1: 3000, Roche, Switzerland) in phosphate buffer with 0.1% bovine serum albumin (BSA), and 0.3% Triton X-100. Sections were incubated with biotinylated secondary antibody and avidin-biotin complex (ABC) (Vector Laboratories, Amsterdam, The Netherlands). The peroxidase reaction was visualized with DAB/H2O2.
For doublecortin staining, sections were incubated with primary antibody, 1:1500 in 0.25% gelatin, 0.5% triton-X-100, 0.1 M TBS (Supermix), then with biotinylated secondary antibody (1:200). The avidin-biotin reaction was enhanced by biotinylated tyramide and, amplified with avidin-biotin peroxidase complex. The peroxidase reaction was visualized with DAB/H2O2. TUNEL staining was performed on 10-µm paraffin sections according to (17)
. At least 4 individual mice were used for each age group and at least 3 or 4 separate sections from each. Statistical analyses were carried out with a 2-way ANOVA.
Golgi silver impregnation
Golgi staining was performed as described in (15)
. Briefly, mice were decapitated, brains removed and placed in Golgi-Cox solution. After washing and dehydration in graded alcohols and ether, brains were saturated in celloidine solutions and cut into 200-µm coronal sections. Staining was visualized with 16% ammonia and 1% sodium thiosulfate. Dentate granule cells were selected from the middle third of the inner pyramidal blade at bregma –2.54 mm. Pyramidal cells were selected from the same level of the CA1 area opposite to the middle third of the suprapyramidal blade of the DG. If neurons were completely stained and horizontally orientated within the section, 99 Z-stacks of 1 µm were recorded with Image-Pro Plus, (Media Cybernetics, Silver Spring, MD, USA), using an Axioplan 2 (Zeiss) microscope, equipped with an Evolution QEi FAST camera (Media Cybernetics), at x40, then analyzed with the drawing tool NeuroDraw (Image-Pro Plus, Netherlands Institute for Brain Research, Amsterdam, The Netherlands) (18)
. Statistical analyses were performed with the Mann-Whitney U test.
Cell culture
Embryonic hippocampal neuronal cultures were prepared from E17.5 pups according to standard procedure (19
, 20)
. After dissection and tissue dissociation, the cells were seeded on glass coverslips (Menzel-Glaser, Germany) coated with 1 mg/ml poly-L-lysine (Sigma) in 0.1 M borate buffer, pH 8.5, at a density of 100,000 cells/coverslip. Cells were cultured on an astroglial feeder layer in Neurobasal medium (Invitrogen, Merelbeke, Belgium). Some cultures were pulsed with 10 µM BrdU (Sigma, USA) for 6 h.
cDNA constructs and transfection
Tau-3R and tau-4R DNA constructs described in (21)
were subcloned into the pcDNA3 vector and transfected into hippocampal cells by electroporation. After tissue dissociation, 106 cells were mixed with 10 µg DNA in 0.5 ml HBSS (Invitrogen) and electroporated (0.2 kV, 960 µF). Cells were plated on glass coverslips coated with 1 mg/ml poly-L-lysine and 10 µg/ml laminin (Invitrogen). Transfection rates were quantified by tau immunostaining and ranged between 15–25%.
Immunocytochemistry
Cells were fixed with 4% paraformaldehyde in 0.1 M phosphate buffered saline (pH 7.4) (PBS). For BrdU staining, cells were treated with 2N HCl. Nonspecific binding was blocked with 10% fetal calf serum in 0.1% Triton-X100 PBS, and cells were stained with primary antibody. Dilutions as follows: HT-7 (1:2,500), JN-RF.5 (1:2,500), BrdU (1:100), nestin (1:1,000), NeuN (1:1,000), SMI-312 (1:10,000), ßIII-tubulin (1:1000), doublecortin (1:5,000), and MAP-2 (1:1,000). Cells were incubated with biotinylated secondary antibody (1:500), then with avidin-biotinylated peroxidase complex (Vector Laboratories, Burlingame, CA, USA). The peroxidase reaction was developed with DAB/H2O2. For double labeling, cells were quenched with 3% peroxide, then stained with the second primary antibody with Vector SG or VectaRed as chromogen. TUNEL staining (Promega, Leiden, The Netherlands) was performed according manufacturer’s protocol. Microscopy was performed with an Axioplan 2 (Zeiss) microscope. All experiments were repeated on three or more independent cell preparations and cultures, scoring at least 1500 cells per group. Group comparisons were made with the Kruskal-Wallis method and subsequent post hoc tests with the Mann-Whitney U test.
Neurite outgrowth assay
Neurite outgrowth was assayed by culturing the cells in a filter-based compartmentalized culture system, according to manufacturer’s instruction (Chemicon). Hippocampal neurons were seeded on laminin coated membrane inserts at a density of 100,000 cells per well in 24-well plate and cultured in Neurobasal medium. The membrane inserts were removed, and the neurites at the opposite side of the membrane were quantified and visualized. Group comparisons were made with the Kruskal-Wallis method, and subsequent post hoc tests with the Mann-Whitney U test.
Western blot analysis
Snap-frozen brain tissue was homogenized in 2 ml of 0.1 M MES buffer (pH 6.4), 0.5 mM MgCl2, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, 0.2 mM PMSF, 20 mM NaF, 0.2 mM Na3VO4, 1 µM okadaic acid, 5 µg/ml leupeptin, 5 µg/ml pepstatin, 5 µg/ml soybean trypsin inhibitor, 1% sodium deoxycholate, 1% Triton-X-100, and 0.1% SDS. After centrifugation, aliquots of the supernatant brain extract were diluted in Laemmli sample buffer and boiled for 10 min. Cells were lysed in PBS containing 1% Triton X-100, 0.05 mM orthovanadate and complete protease inhibitor mixture (Roche, Belgium). Proteins were separated by SDS-PAGE on 8% Tris-glycine gels (Novex, Zandhoven, Belgium) and Western blotting performed according to standard procedures.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. A single copy of the construct was targeted to exon 1 of the murine tau gene (Fig. 1
A) (8
, 9)
. Recombination in ES cells supplied six correctly recombined clones, one of which yielded appreciable coat color chimeras. One male offspring was mated to FVB/N females, and heterozygous breeding was performed for eight generations until homozygous tau-KOKI mice were obtained. Resulting offspring were genotyped by Southern blot analusis and later identified routinely by PCR reactions detecting either the thy-1-tau-4R construct or the unbroken exon 1 of the murine tau gene.
|
The desired htau-4R expression and complete deficiency of murine tau in tau-KOKI mice were demonstrated by immunohistochemistry and Western blot analysis (Fig. 1B-C
). At no time was tau-3R expressed in the tau-KOKI mice. Htau-4R expression was shown to initiate around P12 and reach robust stable levels around P21 (Fig. 1C
). The single copy number generated lower expression levels of htau-4R than in the overexpressing tau-4R mice (8)
, thereby preventing axonopathy and motoric problems (9)
.
Nontransgenic mice express tau-3R in brain before birth and until around P12, after which tau-4R isoforms become the dominant tau species. Within the hippocampus, tau-3R was found in both cornu ammonis (CA) and dentate gyrus (DG) at P7, but only in the DG at P15 and not at all in brain of older mice (not shown). At P15, tau-4R was expressed in the cortex, basal ganglia, and brain stem of tau-KOKI and nontransgenic mice. The highest levels were found in the hilus and CA3 stratum radiatum (Fig. 1D
). The tau-KOKI mice generally expressed lower tau-4R levels than the nontransgenic ones, except for the DG outer molecular layer. Tau expression was very low in the CA1 subfield for both tau-KOKI and nontransgenic mice. Some small cells in DG, CA3, and basal ganglia expressed high somal content of tau-4R in nontransgenic mice; this was not seen in the tau-KOKI mice.
Increased volume and neuronal cell number in hippocampus of tau-KOKI mice
Up to six months of age, tau-KOKI mice were phenotypically indistinguishable from nontransgenic mice with respect to body weight, behavior, in the home-cage, rearing, grooming, fertility, and litter size. Despite minor differences in tau phosphorylation compared to nontransgenic mice, they developed neither tau pathology nor suffered memory impairment (9)
, and only late in life did they display some minor motoric and behavioral defects. Consequently, analysis of brain of young and adult tau-KOKI mice by a wide range of histochemical and immunohistochemical staining revealed few differences relative to nontransgenic mice matched for age, gender, and genetic background.
Nissl staining and volumetric analysis did, however, show the hippocampal volume to be increased in adult tau-KOKI mice compared to nontransgenic ones. Unbiased stereological analysis showed that both the volume and neuronal number in the hippocampus of tau-KOKI mice were significantly increased by 
20% (Fig. 2
A, B). The increased volume in CA1 (P<0.005) was paralleled by increased numbers of pyramidal cells in CA1 (P<0.001) and of granular cells in the DG (P<0.003). No such volume increases were found in the cerebellum, a brain region with little or no expression of the transgene (Fig. 2C
), either in the granular cell layer, the molecular cell layer, or in white matter. Neither did volumetric MRI-analysis performed as in (22)
reveal any differences in neocortex or cerebellum size (not shown). From this, we conclude that the neuronal cell number in hippocampal subregions, but not in neocortex or cerebellum, depends on tau-4R expression.
|
Increased cell proliferation in hippocampus of tau-KOKI mice
Since increases in hippocampal size and cell number, though not in cell density, were found in the tau-KOKI mice already at 2 mo of age (Fig. 3
A–C), the rest of the analyses was performed on mice of that age.
|
Hippocampal neurogenesis and survival were further assessed by stereological analysis after injection of S-phase marker BrdU and by doublecortin immunostaining for young migrating neurons. BrdU was administered for seven consecutive days from P30 onward, and labeled cells were quantified 4 wk later. The BrdU labeling index (LI) was significantly higher in the subgranular zone (SGZ) of tau-KOKI mice compared to nontransgenic mice (P=0.0005), implying increased survival of nascent cells in the tau-KOKI mice (Fig. 3D, F
).
Increased numbers of doublecortin positive cells in the SGZ and the granular cell layer (GCL) of tau-KOKI mice (P=0.0077 and P=0.024, respectively) demonstrated the presence of a young cell population, confirming the increase in proliferation (Fig. 3E, F
). Also, TUNEL staining of brain sections of P15 mice demonstrated less programmed cell death after onset of htau-4R expression in tau-KOKI mice (not shown).
Morphological analysis of dendritic extensions by Golgi silver impregnation revealed a significant decrease in dendritic arborization in the DG of tau-KOKI mice (P=0.001), resulting in less overall space occupied by individual cells (P=0.001) (Fig. 4
A–C, G). The number and segment lengths of the dendrites were otherwise similar in tau-KOKI and in nontransgenic mice, only the number of filopodia was decreased in tau-KOKI mice (P=0.014) (Fig. 4D-F
, G). None of these parameters were altered in the CA1 area, in which tau-4R expression was similar in nontransgenic and tau-KOKI mice (not shown).
|
These data imply that tau-4R contributes to a mechanism that prolongs the proliferative phase of precursor cells, enhances their survival, and delays their differentiation into the neuronal phenotype. The end result is the establishment of a younger, less differentiated, but more numerous neuronal population in the hippocampi of tau-KOKI mice.
Improved hippocampal memory retention
Since neurogenesis in the dentate gyrus is an important determinant in hippocampal learning and memory (23
24
25)
, we investigated whether the increased hippocampal size would be paralleled by altered cognitive functions in the tau-KOKI mice. We selected the novel object recognition test (NORT) as a typical and robust hippocampus-dependent task of memory retention in transgenic mice (12, 15, and references therein). NORT depends on short-range visual and tactile stimuli, and the length of delay determines actual hippocampal involvement.
Tau-KOKI mice performed similarly to nontransgenic mice in the 1-h delay NORT. High d2 values indicate that both tau-KOKI and nontransgenic mice remembered the familiar object (Fig. 5
A). In the 3.5 h delay NORT, tau-KOKI mice still recognized the familiar object, with d2 values similar to those after 1 h and significantly better than nontransgenic mice (P=0.021) (Fig. 5B
). This superior memory retention of tau-KOKI mice was observed in two independent sets of experiments in two different cohorts of nontransgenic and tau-KOKI mice at age 9 wk (Fig. 5)
and 5 wk (not shown).
|
Prior to the NORT, the same cohorts of mice were tested for basal sensorimotor parameters by rotarod and open field tests. Tau-KOKI and nontransgenic mice performed equally well in the rotarod test (Fig. 5C
). In the open field test, tau-KOKI mice traveled equal distances and spent equal time exploring the objects as nontransgenic mice during the same time interval. Although tau-KOKI mice spent significantly more time in the center, indicating less anxiety (P=0.040), (Fig. 5D-E
) locomotion, and explorative behavior were not affected.
Normal LTP in CA1 and DG
To determine cellular and molecular contributions to the improved memory retention, LTP was recorded in the CA1 and DG areas of brain sections from the same cohorts of mice used for immunohistochemistry. Three major determinants, i.e., basal transmission, maximal amplitude or slope, half maximal stimulation intensity and slope factor were not significantly different in either the CA1 or DG of tau-KOKI mice compared to nontransgenic mice (Table 1
). No differences were observed in LTP at any time during 1 h of recordings (Fig. 6
A–B), indicating that changes in local hippocampal circuit or field properties do not contribute to the improved memory of tau-KOKI mice.
|
|
Tau-4R in primary hippocampal cultures
The contribution of human protein tau to the apparent proliferation and/or neuronal differentiation in the hippocampi of tau-KOKI mice was further examined in primary embryonic hippocampal cultures (E17.5). No htau-4R expression was detectable in tau-KOKI neurons at 4 days in culture (4 DIC) (Fig. 7
A, B). Between 4 DIC and 10 DIC, expression of htau-4R increased, reaching stable levels around 10 DIC. The neuronal cultures thus paralleled the timeline of expression of protein tau in vivo, allowing us to investigate the consequences of 1) absence of tau-3R, 2) initial absence of all tau during early neurodevelopment and 3) expression of human tau-4R in a cell system without confounding effects of other isoforms or endogenous tau.
|
Neuronal proliferation is suppressed and neuronal differentiation promoted by tau-4R
To define proliferation and neuronal differentiation in the absence and presence of tau, primary cultures were labeled and scored for S-phase marker BrdU and neuronal marker NeuN. At 4 DIC, the BrdU labeling index (LI) was significantly higher (P=0.00005) and the ratio of NeuN positive (NeuN+) cells lower (P=0.0018) in tau-KOKI cultures compared to nontransgenic ones (Fig. 7C-D
, F). In the absence of protein tau, the population of proliferating precursors (BrdU+ cells) increased and their differentiation into a neuronal phenotype (NeuN+ cells) was delayed.
At 10 DIC, the BrdU LI in the tau-KOKI cultures was reduced to nontransgenic levels, while the ratio of NeuN+ cells was increased (P=0.0209). These changes coincided with onset of htau-4R expression, indicating that htau-4R promoted neuronal differentiation. Moreover, staining for precursor marker nestin showed a significantly increased ratio of nonproliferating precursors at 4 DIC (P=0.005) and a significantly decreased ratio at 10 DIC (P=0.0018) in the tau-KOKI cultures (Fig. 7E, F
). TUNEL staining showed cells undergoing programmed cell death in 4 DIC cultures to be less than 10% and in 10 DIC to 12–20%, with no difference between tau-KOKI and nontransgenic cultures (not shown).
These data provide further evidence that absence of tau delayed the neuronal differentiation in a large proportion of precursor cells, allowing the pool of proliferating precursors to expand. Expression of human tau-4R efficiently restored the neuronal differentiation and directed excess precursors toward a neuronal phenotype and fate. The net result is an effective reduction of proliferating precursors to control values, paralleled by an increased number of neurons.
Tau-4R promotes axonal and neurite outgrowth
The delayed neuronal maturation and differentiation in tau-KOKI cultures was quantitated in a filter-based compartmentalized culture system. Tau-KOKI primary cells showed significantly less neurite outgrowth at 4 DIC than nontransgenic cultures (P=0.0495). Neurite outgrowth of 10 DIC tau-KOKI cells expressing htau-4R did not differ from nontransgenic cells (Fig. 7G
). Immunostaining cells with SMI312, specific for axonal phosphorylated neurofilaments, also showed markedly decreased axonal development at 4 DIC, consistent with delayed differentiation in the absence of tau (7H). Again, expression of htau-4R at 10 DIC restored axonal outgrowth. No differences between tau-KOKI and nontransgenic primary cultures were observed for the dendritic marker MAP-2, early neuronal marker doublecortin or ßIII-tubulin (not shown).
The effects of tau-4R are isoform specific
Next, we investigated whether the defective phenotype of tau-KOKI primary cells could be restored to a nontransgenic phenotype of neuronal differentiation at an earlier time point in neuronal development and/or in an isoform-specific mode. To do this, we transfected tau-KOKI primary cells with either htau-3R/2N or htau-4R/2N. Transfected cultures contained comparable number of transfected cells with similar levels of either htau-3R or htau-4R (Fig. 8
A).
|
Tau-KOKI primary cultures transfected with htau-4R, but not with htau-3R, showed a reduced BrdU LI (P = 0.0495), similar to that of nontransgenic cultures at 4 DIC (Fig. 8B
). Similarly, expression of htau-4R, but not htau-3R, restored the ratio of NeuN+ cells in tau-KOKI cultures to that of nontransgenic cultures (P=0.0495) (Fig. 8C
). Either of the tau isoforms significantly increased neurite outgrowth at 4 DIC (P=0.0495) (Fig. 8D
), implying that both isoforms overlap largely in their capacity to promote neurite extension and/or stabilization, though the htau-4R generated a three-fold larger increase in neurite protein levels than the htau-3R.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The tau-KOKI mouse was generated as a simplified model expressing only human tau-4R in postnatal neurons in a murine tau-null background. Consistent with the idea of tau being redundant for viability and fertility, tau-KOKI mice evolved normally with only very minor motor and behavior problems at an advanced age (18–24 mo) (9)
, confirming that mouse brain can develop without tau and undermining an important role for tau-3R isoforms during embryonal and postnatal development. This may be explained by functional overlap with other MAPs (11
, 20)
or considerable developmental plasticity.
The hippocampus is one of few brain regions to retain the ability to generate neurons throughout life. Neurogenesis in the hippocampus includes proliferation, survival, and differentiation of dentate granule progenitors in the SGZ, which migrate into the GCL, and integrate within existing neuronal circuits. Understanding the developmental profile of newborn granule cells in the neonate and adult is extremely important since the dentate gyrus is a recipient of the perforant pathway and, as such, highly responsive to plastic and pathological changes.
In mice, the generation of hippocampal neurons and formation of the hippocampus are situated in a postnatal window coinciding with the switch in expression from tau-3R to tau-4R (26
27
28)
. In tau-KOKI mice, this window also coincides with the onset of expression of the thy1-tau-4R transgene (9
, 28)
. The unexpected observation of a larger hippocampal volume in tau-KOKI mice relative to gender, age, and genetic background-matched nontransgenic mice was shown to be due to a larger number of neurons in CA and DG. Our data demonstrate that survival of BrdU-positive progenitors was significantly higher in tau-KOKI mice after onset of htau-4R expression, while programmed cell death decreased. The corresponding increase in doublecortin positive cells in the SGZ and GCL further confirmed the increased neurogenesis. Morphologically, neuronal Golgi staining showed less branching of neurites in the DG, indicating a younger, less mature cell population.
The enlarged hippocampal size of tau-KOKI mice was reflected in behavioral tasks, as they performed better than nontransgenic mice in the novel object recognition test. The task depends on hippocampal cognitive capacities, including short-range visuals and tactile stimuli. We conclude that increased neurogenesis in the tau-KOKI mice due to the absence of tau is a time-discreet event that nonetheless resulted in enlarged hippocampal volume and improved cognitive capacity.
Using criteria of axonal sprouting and process extension, an important role for protein tau has been suggested in establishing neuronal polarity and formation of contacts and circuits by inhibition with antisense mRNA in primary rat cultures (29)
or laser inactivation of labeled tau in chick dorsal root ganglion cultures (30)
. Primary hippocampal neurons lacking tau were delayed in development of neuronal polarity along with reduced elongation of axons and dendrites, while human tau expression restored sprouting and growth (20)
. Conversely, hippocampal neurons derived from tau-deficient mice did not show any defects in neuronal maturation, implying that other MAPs compensated for the absence of tau (11
, 31)
. These issues were examined in primary hippocampal cell cultures from tau-KOKI mice in the absence and presence of human tau-4R. We show that absence of tau delays neuronal differentiation of a large proportion of precursor cells, expanding the pool of proliferating precursors in tau-KOKI cultures compared to nontransgenic cultures. The htau-4R isoform, but not the tau-3R isoform, effectively restored the neuronal differentiation process and subsequently directed the excess precursors toward the neuronal phenotype and fate. Furthermore, the htau-4R isoform is able to rescue the impaired/retarded axonal outgrowth of tau-KOKI-derived primary neuronal cultures. The net result is an effective reduction of the levels of proliferating precursors to control values, concomitant with an increased number of neurons. These observations in cell cultures recapitulate the in vivo situation and explain the differential role of tau-3R and tau-4R isoforms.
It has been shown that tau-3R and tau-4R isoforms distinctly affect various parameters of microtubule nucleation and assembly in vitro (32
, 33)
. Our results imply additional in vivo differences between tau-3R and tau-4R isoforms during development. While tau-3R does not influence cell fate, its assisting of microtubule assembly may allow the generation of a precursor pool. In contrast, tau-4R inhibits proliferation and promotes neuronal differentiation and axonal outgrowth, likely by stabilizing the microtubular network. The postnatal isoform switch to tau-4R may thus be taken as a commitment of precursor cells to a neuronal fate. We suggest that the tau-4R isoform essentially contributes to hippocampal development by controlling proliferation and differentiation of neuronal precursors.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Received for publication December 13, 2005. Accepted for publication February 1, 2007.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. Deshpande, K. M. Win, and J. Busciglio Tau isoform expression and regulation in human cortical neurons FASEB J, July 1, 2008; 22(7): 2357 - 2367. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Terwel, D. Muyllaert, I. Dewachter, P. Borghgraef, S. Croes, H. Devijver, and F. Van Leuven Amyloid Activates GSK-3{beta} to Aggravate Neuronal Tauopathy in Bigenic Mice Am. J. Pathol., March 1, 2008; 172(3): 786 - 798. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Muyllaert, D. Terwel, A. Kremer, K. Sennvik, P. Borghgraef, H. Devijver, I. Dewachter, and F. Van Leuven Neurodegeneration and Neuroinflammation in cdk5/p25-Inducible Mice: A Model for Hippocampal Sclerosis and Neocortical Degeneration Am. J. Pathol., February 1, 2008; 172(2): 470 - 485. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||
0.05
