Wednesday, 10 February 2016

How to Read an EAE paper


Gibson-Corley KN, Boyden AW, Leidinger MR, Lambertz AM, Ofori-Amanfo G, Naumann PW, Goeken JA, Karandikar NJ. A method for histopathological study of the multifocal nature of spinal cord lesions in murine experimental autoimmune encephalomyelitis. PeerJ. 2016 ;4:e1600. eCollection 2016.

Experimental autoimmune encephalomyelitis (EAE) is a well-established mouse model for multiple sclerosis and is characterized by infiltration of mononuclear cells and demyelination within the central nervous system along with the clinical symptoms of paralysis. EAE is a multifocal and random disease, which sometimes makes histopathologic analysis of lesions difficult as it may not be possible to predict where lesions will occur, especially when evaluating cross sections of spinal cord. Consequently, lesions may be easily missed due to limited sampling in traditional approaches. To evaluate the entire length of the spinal cord while maintaining anatomic integrity, we have developed a method to section the cord within the decalcified spinal column, which allows for the study of the multifocal nature of this disease and also minimizes handling artifact. HE and Luxol fast blue staining of these spinal cord sections revealed a paucity of lesions in some areas, while others showed marked inflammation and demyelination. The percentage of spinal cord affected by EAE was evaluated at four separate areas of longitudinally sectioned cord and it varied greatly within each animal. Immunohistochemical staining of in situ spinal cords which had undergone decalcification was successful for key immuno-markers used in EAE research including CD3 for T cells, B220 for B cells and F4/80 for murine macrophages. This method will allow investigators to look at the entire spinal cord on a single slide and evaluate the spinal cord with and without classic EAE lesions
Recently when we wrote a grant and we said that we would assess neurodegeneration using a quantitative biochemical method , which we have validatd it against conventional histology, which assesses the total nerve content within the nervous system using a technology, where you can test hundreds of samples in one day.
However we were chastised and told to go away and come back with a proposal that used histology to show nerve lost!

However, I said this was not properly qualitative because with histology you would usually only count nerves in a few places and not the whole nerve content in a spinal cord section and maybe only look in one area in the whole of the spinal cord and maybe only give is a +, ++ scoring which is not really quantitative. Furthermore you can select areas to show what you like. Next it may take a few weeks to process the tissues, section then and stain them. 


However, there was a blinkered response and you just have to waist your time and do it to keep the i dotters and t crossers happy.


So this paper needs to be read by these people and if you read animal study papers then you need to think about this also because you can show what you like withhistology by being selective. What you need to do is to show pictures that are representative.


How many times have you seen a picture of drug X showing no infiltrates compared to the controls that are jam packed with infiltrates yet when we look at the clinical scores, the drug treatment only drops the severity of disease by a small fraction. 


If there is clinical disease then you must expect to see infiltrates. If there are none and there is clinical disease the pictures are selective and not representative.


So in this paper they get the whole cord processed within the spinal column. They treat the bone to make it soft, they then section and stain the spinal cord.


So here you see if you look for infiltrates (clusters of the blue dots, which are the nuclei of cells and are circled) you can see in some areas there are none and in other areas there is alot. So if you cut at B there is nothing and cut at C there is alot, yet they ar occuring at the same time. Likewise if you looked at C maybe a week or two later the inflitrates may have gone as the clinical signs abate and so waiting until the end of the experiment may give you meaningless results.


No comments:

Post a Comment

Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.