Spreading of P301S Aggregated Tau Investigated in Organotypic Mouse Brain Slice Cultures

Abstract

Tau pathology extends throughout the brain in a prion-like fashion through connected brain regions. However, the details of the underlying mechanisms are incompletely understood. The present study aims to examine the spreading of P301S aggregated tau, a mutation that is implicated in tauopathies, using organotypic slice cultures. Coronal hippocampal organotypic brain slices (170 µm) were prepared from postnatal (day 8-10) C57BL6 wild-type mice. Collagen hydrogels loaded with P301S aggregated tau were applied to slices and the spread of tau was assessed by immunohistochemistry after 8 weeks in culture. Collagen hydrogels prove to be an effective protein delivery system subject to natural degradation in 14 days and they release tau proteins up to 8 weeks. Slices with un- and hyperphosphorylated P301S aggregated tau demonstrate significant spreading to the ventral parts of the hippocampal slices compared to empty collagen hydrogels after 8 weeks. Moreover, the spread of P301S aggregated tau occurs in a time-dependent manner, which was interrupted when the neuroanatomical pathways are lesioned. We illustrate that the spreading of tau can be investigated in organotypic slice cultures using collagen hydrogels to achieve a localized application and slow release of tau proteins. P301S aggregated tau significantly spreads to the ventral areas of the slices, suggesting that the disease-relevant aggregated tau form possesses spreading potential. Thus, the results offer a novel experimental approach to investigate tau pathology.

Keywords: Alzheimer; collagen hydrogels; organotypic brain slices; spreading; tau; tauopathy.

Figure 1

Characterization of collagen hydrogels. (A) A representative image showing coronal hippocampal organotypic brain slices (170 µm thick) on top of extra-membranes inside cell culture inserts. (B) A schematic demonstrating the placement of collagen hydrogels on the top of the slice and above the hippocampus region (green circle). The numbers indicate the three areas that were delineated to extract optical density after the incubation of brain slices with P301S aggTau-loaded collagen hydrogels. Area 1 is the location of the collagen hydrogel; Area 2 is the cortex and Area 3 is the ventral region of the slice. (C) Collagen hydrogels are approximately 2 mm in size and could be visualized after loading with a green fluorescent AlexaFluor-488 antibody. (D) P301S aggTau was detectable with the Tau5 antibody when it was loaded into collagen hydrogels. (E) Collagen hydrogels are visible after 4 days on organotypic brain slices in a slightly fragmented form. (F) No green fluorescent AlexaFluor-488 signal was observed after the collagen hydrogels were in culture 14 days, suggesting a complete degradation and release of tau proteins. The * denotes the location of the collagen hydrogel. Scale bar in F = 8890 µm in (A); 230 µm in (C–F).

Figure 2

Characterization of the P301S aggTau using Western blots. (A) Immunostaining with the Tau5 antibody detects full-length (FL) tau in the range of 20–500 ng and the bands were observed at the expected size of 60 kDa. (B) FL tau aggregated (agg) was evident by the smear of proteins with a molecular weight >60 kDa. (C) P301S aggregated tau (P301S aggTau) appeared as a smear of large proteins >50 kDa. (D) GSK-3β hyperphosphorylated the tau proteins, which were detectable with the phospho-tau-396 antibody. P301S aggTau was detected as a smear of proteins in 50-160 kDa range. (E) P301S aggTau loaded in collagen hydrogels displays positive detection of tau at the expected 60 kDa mark, which was also observed when protease inhibitors were added. Hyperphosphorylated P301S aggTau is detected at the 60 kDa mark with the phospho-tau-396 antibody. (F) When P301S aggTau protein was loaded into collagen hydrogels and placed onto brain slices for 4 or 8 weeks, a slightly weaker band was visible at 60 kDa. (G) The tau monoclonal HT7 antibody detected P301 aggTau (500 ng/lane), but it did not detect P301S aggTau loaded into collagen hydrogels. However, when the P301 aggTau-loaded collagen hydrogels were applied on brain slices for 4 and 8 weeks, clear but weaker bands were seen at the 60 kDa mark. Additionally, a smaller tau fragment was observed at the 30 kDa size.

Figure 3

Release of tau proteins from collagen hydrogels. Full-length (FL) tau (red circles) or P301S aggTau (green diamonds) were loaded into collagen hydrogels and placed on small pieces of Parafilm face-down in culture media from 0 to 8 weeks. Empty collagen hydrogels (blue squares) were also incubated as a negative control. The media were collected weekly, and the level of tau was determined by a tau-specific ELISA. P301S aggTau was released more from the collagen hydrogels than the FL tau but it does not reach maximal release. No tau was detected from the empty hydrogels. The Lumipulse measurements were repeated from the same samples for Coll (P301S aggTau) for 4, 5-, 6-, 7- and 8-week time points. Values are reported as ng/mL.

Figure 4

Effects of P301S aggTau on slice viability. (A) Slices were incubated with collagen hydrogels loaded with no load [CollH (−)] or P301S aggTau [CollH (P301S aggTau)] for 8 weeks. Subsequently, the slices were pooled together and probed for various cellular markers through a Western blot. Actin served as the loading control. There is no qualitative difference in the expression levels of neuronal NF, oligodendroglial MOG, microglial CD11b and astroglial GFAP between the CollH (−) and CollH (P301S aggTau) groups. Catalase expression is weak and variable in both groups. (B) Quantification of the band intensity reported no significant difference between CollH (−) and CollH (P301S aggTau) groups for any cellular marker. (C) Levels of tau protein were quantified through Lumipulse technology in slices from the CollH (−) and CollH (P301S aggTau) groups, which report a significantly higher tau level in slices incubated with P301S aggTau−loaded hydrogels (p = 0.012). (D) LDH release into the culture media was quantified and there was no significant difference between the slices from the CollH (−) and CollH (P301S aggTau) groups. Values are reported as mean ± SD and the number in parentheses represents the number of animals analyzed per group. A student’s t-test with equal variance was employed to compare mean values from both groups. Each lane in the Western blots represents an individual mouse sample. Values in parentheses indicate the number of analyzed animals.

Figure 5

Spreading of P301S aggTau in organotypic brain slices. (A–E) Composite images of brain slices displaying the spread of P301S aggTau as detected by Tau5 antibody starting from inside of the collagen hydrogels at 0 week to the ventral parts of the slices at 8 weeks. (F–J) Magnified representative images of tau immunostainings of the black boxes indicated in (A–E). (F) A clear boundary of the P301S-loaded hydrogel can be seen from slices that were immediately post-fixed and immunostained after the application of collagen hydrogels. (G,H) Tau-specific positive immunostaining is observed in cells along with what appears to be neuronal connections after 1 and 2 weeks in culture. (I,J) After 4 and 8 weeks in culture, tau immunostaining appears in the ventral regions of the slices. (K–O) Composite images of brain slices of the spread of P301S aggTau as detected by the HT7 antibody, which is specific for the human tau isoform. The pattern of spread is similar to the one observed with Tau5 antibody. (P–T) Magnified representative images of the black boxes from the (K–O) images showing tau immunoreactivity spreading from the collagen hydrogels to the ventral areas. Scale bar in T = 930 µm in (A–E,K–O); 100 µm in (F–J,P–T).

Figure 6

Spread of P301S aggTau along neuroanatomically connected pathways. (A,B) Schematic representations of the slices that were incubated with P301S aggTau-loaded collagen hydrogels, one each on the left and right hemisphere regions (green ovals). (A) Slices without a lesion and (B) slices that were subject to a lesion created by a cell scraper one week after the application of the collagen hydrogels (red rectangle). Subsequently, spreading of P301S aggTau to the ventral areas of the slices (Area 3) was analyzed (black boxes). (C) Quantification of the extracted optical density, which serves as a proxy for Tau5 positive immunostaining in Area 3. Slices with a lesion demonstrate significant decreased spreading of P301S aggTau to Area 3 compared to the control slices. The two groups were compared using a student’s t-test with equal variance (*** p < 0.001). Values in parentheses represent the number of animals analyzed. Data are presented as mean ± SD. Values in parentheses indicate the number of analyzed animals. Please note that we consider an optical density (OD) value of 21 to be unspecific background.

Figure 7

Localization of P301S aggTau in organotypic brain slices. (A) Representative image of P301S aggTau-specific positive fluorescent staining appeared in the ventral parts of the slices (green; AlexaFluor-488). (B) Little background staining was visible in the red channel. (C) Slices were counterstained for the nuclear dye, DAPI (blue fluorescence). (D) A merged image was generated, which reveals tau immunostaining in the cell bodies and connections between these cells. (E,F) Representative images of the 3-D reconstruction of confocal microscopy z-stacks stained for P301S aggTau (green) and the nuclei are shown in blue. (G–J) Slices were stained for tau (green; AlexaFluor-488), neurofilament as a neuronal marker (red; AlexaFluor-546) and nuclei (blue; DAPI). The merged image indicates that P301S aggTau colocalizes with neurofilament. Scale bar in B = 40 µm in (A–D); 6 µm in (E,F); 60 µm in (G–J).