[Histonet] microglia/philosophy OT

Sun Jun 24 20:47:18 CDT 2007

Is a very long answer to Carl as well as some philosophy so just hit delete right now if you have other matters or interests...

Hi again Carl,
So this is the ultimate demonstration of how e-mail and the internet cannot be used readily for true scientific discussion.  The inexactitude of any statement anyone can make in an e-mail simply cannot stand up to the scrutiny of logic and reason in a final sense.  People would have you almost believe that we are all so sophisticated now with e-mail questions and answers that something like a theory or thesis or journal article can just be submitted by e-mail and that is somehow the final and lasting story.  This instrument, no matter what computer geeks say, was, is and will never meet the requirement of logical expression during dynamic interaction.  Not meant for debate, disagreement or thorough discussion.

I minored in math as an undergrad and actually I enjoy the certitude and precision of math sometimes more than I do science.  Thus it is my lifelong hobby.  To me, a complicated problem you intellectually solve in trigonometry is simply very relaxing and mind settling.  If we were to discuss  different opinions regarding the steps of integration of the complicated integrand in an integral calculus problem, back and forth by e-mail, we'd never reach a conclusion.  It just can't be done.

Besides e-mail not being able to solve such discrepancies, there is another problem.  I assume John Kiernan is out there lurking somewhere although I don't remember seeing him for a while and I certainly don't put my brain anywhere near the same intellectual category as his.  But I believe it was he who has cautioned over and over about believing what is said on the Histonet, simply because someone wrote it.  I guess that is true when you think about it.  After all, it is certainly possible that in reality I am a disgruntled moonshiner, deep in the Ozark Mountains of the "Show Me" state and doing my best to stay hidden from the revenooers.  My shack has no electricity and I run my laptop off an old solar panel when I'm not guarding the still with my 12 gauge shotgun.  I might have never been in a histo/patho/immunology lab in my life so my answers and thoughts might be pure hogwash.

In any event, I'll try to clarify a bit more what I believe I said and follow-up on some of your questions.  To pin this down with more precision, we'd have to wait until I make it to the UK and we can sit for an afternoon with some pitchers of ale to discuss it.  Or you can get over here, and we will sit for the day while sharing some of my 180 proof hootch.  My place is 5 miles just over the ridge from Jed Clampetts old shack.

1)  Yes F4/80 is an antigen but indeed there are F4/80 antibodies (that are not BM8 clones).  Whether rightly or wrongly called, there are antibodies called F4/80.  They exist and because of my thesis project I commonly used an antibody called F4/80.  I knew what it was but catalogs are full of such a designation.
2) the term "slightly".  Use F4/80 antibody and BM8 antibody on flow in spleen, lymph node and other secondary lymphoid organs.  Gate out populations and see "slight" differences in percentages and when you back-gate.  Do single color histograms, overlay, and see unexplained "shoulders" or spread out peaks.  Sort them on cell sorters onto microscope slide and see very similar but "slightly" different cells.  So that is what I mean by "slightly".  By pure histology and IHC, visual recognition sensitivity might be below detectable limits but I do not believe the populations to be "the same".  So "slightly different".  Sometimes oxymorons convey the most precise definition.
(3) I agree with the picture again that as far as my resolution goes, it looks commendable for the non p-tyr pictures.  I asked about rat-on-rat for as far as I can remember, people have always struggled with the concept of an antibody raised in a host being placed on the same species tissue.  Looking at that antibody then with detection against the antibody brings on the problem of picking up endogenous tissue Ig.  Everyone I know frets with the problem of using mouse hosted antibodies on mouse tissues, rat hosted antibodies on rat tissues and human backbone configured antibodies on human tissue.  If you very easily and readily use a rat antibody (like BM8) on rat tissue, or mouse Ab on mouse or human Ab on human, I'd sure like to know the details and I'm sure the entire IHC world would to.  
(4) iba1 = ionized calcium binding adapter molecule.  It doesn't have an alternative name.  It has about 15 alternative names, interferon gamma responsive transcript (irt1), allograft inflammatory factor 1 (AIF1).  You can find a host of others I can't possibly remember on a gene search or antibody search or some data sheets.
(5)No need to send pictures.  I absolutely believe what you say is there and you see staining.  I take it as axiom.  RCA-1 is ricinus communis agglutinin-1, a lectin.  Can get unconjugated, biotin, perox and I usually go to Vector Labs.  At first I was going to say I made an error and that RCA-1 could only be used on human microglia but a few papers I see uses RCA on mouse microglia.  And if you are in rat there are the famous OX series OX-41,42,11 that stain microglia but others know much more about them than I do.  See Bertolotto, JHC vol 41#4 481-487 1993, see Mason, J AJP 164 #5 May 2004 on Oligodendrocytes and precursers, see Hematopoietic Prostaglandin D Synthetase is expressed in microglia in mouse brain, Mohri,I in GLIA 42:263-274 (2003) for frozen, paraffin, EM of microglia staining with "other" antibodies.  I found them in 4-5 minutes in PubMed and I HATE, HATE, HATE computers (except for Power Point, word processing and data entry to Excel).  All the information is out t
here with picture s and alternative markers in abundance.
(6)What is the anti p-tyr staining due to?  Haven't the slightest idea.  Background, false signal, smudge, true signal on inappropriate epitope, 50 other things.  One possibility is that it is exactly what you think it is.  True anti p-Tyr staining you are looking for.  All I'm saying is this.  When I look for a membrane bound or housekeeping, constiuitive target (CD4 on T-helper cells, cytokeratins in epithelial cells, etc, etc) that have been characterized around the world in thousands of labs over and over and by myself, when I see the stain and knowing where it should be and shouldn't be in absolute terms, my confidence level is as absolute as is possible that I am seeing true staining.  But the whole concept of IHC staining for phosphoylated targets is very new and very tricky.  In grad school and in research, I have had opportunity to parse apart several different signal transduction pathways and looking at the phosphorylated intermediates.   Used westerns, lysates, cell line
s, stimulated cell lines, flow, tissue IHC (by perfusion/paraffin, frozen and dunking into formalin and paraffin) and trying to make concordance with all the data to what I saw under a microscope.  While I absolutely agree I found antibodies that MIGHT work as well in IHC as in more standardized molecular techniques, I am still very, very wary of what I see on phospho-protein targeting antibodies on a tissue section.  To me, this does not have so much to do with affinity as with specificity.  I've looked at signaling transduction proteins in the cascade that were phosphorylated say at sites tyr82, tyr99, tyr118, tyr173 and tyr176.  Sometimes individually or alone, or in combinations or all at once and for only a brief few minutes before down-regulating.  That is pretty intricate and tough detection science to put on an antibody used in IHC, no matter what a commercial company says or what is the avidity or affinity of the antibody.   In short, your staining maybe absolutely real fo
r p-tyr.  I just can't hang my hat on it without further proof than that something is turning dark brown.
(7) The data sheet I accessed from the information uses that p-tyr on Westerns.  And they themselves state to boil (as in with heated water or as in soap as in SDS-PAGE) to reduce the protein.  That in itself says to me that this target could possibly be a buried, short epitope that needs to be linearized as opposed to an exposed conformational epitope that you can see more naturally (like in a tissue section).  If you get hold of them and they tell me you can detect using their antibody on Westerns but can run the gel in a "non-reduced" state, I'm more likely to believe it to be good for standard IHC.  Not totally convinced but would ease my mind a bit.
(8) Ki-67 does have a short half-life but that cell is not going to be in G1,S,G2 and M very long anyway.  Absent from G0, in which that cell might have been for extraordinarily long periods, once the molecules and machinery get the go signal to enter G1, that proliferation goes quickly but assuredly.  Entering into anaphase and telophase is toward the end of that proliferation.  So yes it is there for a short period, less than an hour, but then it is actually there.  Even if you take a half life as 1/2 hour, half is still left after 1/2 hour.  And that is a lot.  And remember that Ki-67 is in nucleus.  Take a "little" of something and concentrate it in a small dot (nucleus) and it might be easy to visualize.  Take a "little" of something and spread it out over the whole volume of a cytoplasmic process and it might not be so easy to see.  I am not sure I agree with you that at any point there must be positivity.  Cells were set up, at least in this instance to respond to small amou
nts of phophorylation.  So the entire protein milieu does not have to be phosphorylated.  Just a little.  And if that phosphorylation lasts 1 or 2 minutes before de-phosphorylation, I just don't think it is enough to reliably detect.  That is except, as I said before, in the case of tumors, which can have phosphorylation cascades permanently turned on by definition.  Notwithstanding Dr Rosencrantz and Dr. Guilensterns ideas on a game of spinning coins --as others have said "the law of probability has no jurisdiction here".  And their learned conversations have been noted to be "quite often a series of non-sequitors".

Remember, this is just one ordinary Histo-grunts opinion so I think I'm done now and besides, I believe I hear a durn revenooer nosin down by where the still is.  Good luck with your project.

Ray Koelling
Phenopath Labs
Seattle, WA