[Neurosciences/BBB] Alpha-Synuclein pre-formed fibrils impair tight junction protein expression without affecting cerebral endothelial function

Hi everyone, today I am experimenting a journal club on a blog, sharing my thought on some recent publications in the BBB field.
The paper I will be discussing today is a study recently published by Dr. Roger A Barker (University of Cambridge, Cambridge, UK) in the journal Experimental Neurology (IF=4.5) titled “Alpha-Synuclein pre-formed fibrils impair tight junction protein expression without affecting cerebral endothelial function” (http://www.sciencedirect.com/science/article/pii/S0014488616302710).
Why did I choose this paper? Because this paper is investigating the interaction between alpha-synuclein (aSyn) and the blood-brain barrier. I have recently developed interests to see how the BBB behave in neurodegenerative disorders such as Alzheimer’s, Parkinson’s and Huntington’s disease and in particular how the BBB let peptides involved in such diseases go in and out. In this study, they have used the hCMEC/D3 immortalized human brain endothelial cell line (Weksler et al., FASEB 2005) as well as primary human neurons and astrocytes co-cultures. For the hypothesis, they have used monomeric aSyn as well as what they referred as preformed fibrils (pff).
One of the caveat of this study is the use of hCMEC/D3 cells that are notoriously known for their poor barrier properties. This poor barrier properties was indeed displayed in the paper as the authors reported values of only 15 Ohms*cm2 that is very low. I was also surprised that the authors reported the use of FITC-dextran of 10kDa size as this compound is big enough to poorly cross the BBB. But the use of such tracer also make sense as it can also a comparison to aSyn. aSyn (as pff) did not have much effect on the barrier function, we can even observe an increase in the TEER and decrease in the permeability suggesting a possible induction of the barrier. Another piece of data presented purposively show a immunostaining for ZO1, a protein adapter for tight junction complexes. The immunolocalization of ZO1 was suboptimal, making difficult the interpretation of the negative effect of pff on ZO1. There is a increase  in immunoreactivity following pff treatment. Astrocytes co-cultures improved the barrier function and again no effect of pff was noted on the barrier function. The authors also showed no effects of pff on astrocytes GFAP expression.
The interesting but also the data that raised some skepticism is the experiments involving hCMEC/D3 cells co-cultured with primary neurons. The TEER is lower than the monocultures and astrocyte co-cultures (~8 Ohms.cm2) yet they display permeability values for FITC-dextran 10kDa that are 200X lower than the monoculture. Aside from this issue (that should have been noted by the peer-reviewers), there were also disprecancies between TEER and permeability. If we consider the relative permeability to untreated group accurate, we can note a 50% increase in permeability following treatment with pff or aSyn monomers. Again the immunostaining was pointless as the staining for ZO1 looked poor but also the representative pictures are displaying different cell densities (as noted by  DAPI-cell nuclei density per field).
To better understand the impact of pff on neurons, the authors treated some fetal cortical cells with aSyn monomers or pff using a TUNEL assay (a common technique to observe cell death in vitro) but this was using an immunohistochemistry approach (HRP with DAB stain) instead of the classical immunofluorescence.
Finally, the authors showed some quantitative protein expression analysis from the in vitro cultures and from post-mortem tissues obtained from PD patients. An interesting feature observed was the increase in ZO1, claudin-5 expression following neuronal co-culture, but also a very strong regulation of tricellulin and MarvelD3 proteins (their detection was weak if not negative in monocultures). Surprisingly, the pff treatment decreased occludin and ZO-1 expression at protein levels. The PD Western blots were showing much inter-individuals variability that makes hard to translate from in vitro to in vivo.
In conclusion, the title was very attracting and interesting but I felt the data was poorly supporting the claims. The hCMEC/D3 model is not the best model for modeling the BBB in vitro, especially considering the barrier values reported were below what are commonly reported. I was left on my hunger, it had some interesting data but also some data with lesser quality and foremost did not address if aSyn, as its monomeric or pff form can cross or not the BBB.



Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s