Antibodies
I'm going to try and explain some important facts about antibodies.
Remember that a protein is a big molecule made up of strings of amino acids. They are present everywhere inside and outside of our bodies. There are lots of them in a steak, but much few of them in butter (which is made up of strings of lipids that we call "fat") or celery (which is made up of strings of sugars that we call "starch").
There are these things called antigens can get into your body. The word "antigen" comes from ANTIbody GENerating, and that's sort-of what they do: generate antibodies. Antigens are simply molecules that are almost always proteins (strings of amino acids). When you eat that steak, the proteins from that steak go from your intestines into your bloodstream, and there these proteins (large molecules, remember) come into contact with a specific B cell. This B cell then "wakes up" and starts producing a single type of antibody which reacts with that antigen. When I write "reacts", I mean that the antibody sticks to the antigen.
Let's talk about how this "sticking to" happens. An antigen is a large molecule, and it's really "bumpy". The individual atoms line up in certain ways to make particular bumpy structures. Think of these bumpy structures like a formation of Lego blocks. This structure is called the antigenic determinant, and the antibody for this antigen has the inverse of this structure built into the structure of the antibody. Thus, the antibody mindlessly (it's just a protein molecule after all) has its business end stuck to the antigenic determinant of the antigen. So far so good?
Here's where the nice theory starts to rot. We like to think in terms of "antigen A corresponds with antibody A, whereas antigen B corresponds with antibody B." It doesn't work that way.
An antigen is, after all, a large protein molecule. It can, and does, have more than one antigenic determinant on it, and these antigenic determinants can, and do, have different configurations. The scientific term for this is "polyclonal". Antigen A can have antibody A AND antibody B stick to it. In other words, the antibodies correspond to antigenic determinants, not to antigents.
This means it gets even worse. Suppose you have two different antigens, antigen A and antigen B. They both have several different antigenic determinants in multiple configurations. This means that antibody A can stick to BOTH antigen A AND antigen B. The scientific term for this is "promiscuous" and it is said that antibodies "cross-react". Here is a real-world example of this fact in practice. The Epstien-Barr virus shares an antigenic determinant with the red blood cells of horses. If it is suspected that you have an Epstien-Barr virus infection, then your blood will be mixed with horse blood. If there is a reaction, then that is used as evidence that you are infected with Epstein-Barr virus (not horse blood).
Epstien-Barr virus also shares an antigenic determinant with the red blood cells of sheep.
And it gets even worse from here. An antibody is a protein molecule. Remember when I wrote that antigens are usually other proteins? This means that an antibody can itself be an antigen! When your body makes antibodies, it also makes antibodies for those antibodies.
With all of that information in mind, think about the HIV antibody test. There are actually two different HIV antibody tests, and they work under the same premise: if the antibodies in your blood (and there are always antibodies present in your blood) react with a set of proteins that are assumed to be proties from the HIV retrovirus, then it is said that you are "HIV positive". Put bluntly: "antibody reaction means HIV infection".
The question that jumps to the lips on an "HIV positive" test is this: How do I know that my antibodies were NOT cross-reacting? In other words, how do we know that these proteins that you tested for don't share an antigenic determinant with some other protein, and that is why there was an antibody reaction?
And there has never been an answer to that question that wasn't of the form, "You are not a scientist. Trust us."
Do not, under any circumstances, have an "HIV test".
The information from this was gleaned from this fantastic interview with Dr. Valender Turner of the Perth Group. I have exchanged e-mail with him as well. In it, he also describes what HIV researchers actually do to be able to claim that the "HIV test" is "99.999% sensitive and specific". (Hint: unscientific)
DO ANTIBODY TESTS PROVE HIV INFECTION?
Remember that a protein is a big molecule made up of strings of amino acids. They are present everywhere inside and outside of our bodies. There are lots of them in a steak, but much few of them in butter (which is made up of strings of lipids that we call "fat") or celery (which is made up of strings of sugars that we call "starch").
There are these things called antigens can get into your body. The word "antigen" comes from ANTIbody GENerating, and that's sort-of what they do: generate antibodies. Antigens are simply molecules that are almost always proteins (strings of amino acids). When you eat that steak, the proteins from that steak go from your intestines into your bloodstream, and there these proteins (large molecules, remember) come into contact with a specific B cell. This B cell then "wakes up" and starts producing a single type of antibody which reacts with that antigen. When I write "reacts", I mean that the antibody sticks to the antigen.
Let's talk about how this "sticking to" happens. An antigen is a large molecule, and it's really "bumpy". The individual atoms line up in certain ways to make particular bumpy structures. Think of these bumpy structures like a formation of Lego blocks. This structure is called the antigenic determinant, and the antibody for this antigen has the inverse of this structure built into the structure of the antibody. Thus, the antibody mindlessly (it's just a protein molecule after all) has its business end stuck to the antigenic determinant of the antigen. So far so good?
Here's where the nice theory starts to rot. We like to think in terms of "antigen A corresponds with antibody A, whereas antigen B corresponds with antibody B." It doesn't work that way.
An antigen is, after all, a large protein molecule. It can, and does, have more than one antigenic determinant on it, and these antigenic determinants can, and do, have different configurations. The scientific term for this is "polyclonal". Antigen A can have antibody A AND antibody B stick to it. In other words, the antibodies correspond to antigenic determinants, not to antigents.
This means it gets even worse. Suppose you have two different antigens, antigen A and antigen B. They both have several different antigenic determinants in multiple configurations. This means that antibody A can stick to BOTH antigen A AND antigen B. The scientific term for this is "promiscuous" and it is said that antibodies "cross-react". Here is a real-world example of this fact in practice. The Epstien-Barr virus shares an antigenic determinant with the red blood cells of horses. If it is suspected that you have an Epstien-Barr virus infection, then your blood will be mixed with horse blood. If there is a reaction, then that is used as evidence that you are infected with Epstein-Barr virus (not horse blood).
Epstien-Barr virus also shares an antigenic determinant with the red blood cells of sheep.
And it gets even worse from here. An antibody is a protein molecule. Remember when I wrote that antigens are usually other proteins? This means that an antibody can itself be an antigen! When your body makes antibodies, it also makes antibodies for those antibodies.
With all of that information in mind, think about the HIV antibody test. There are actually two different HIV antibody tests, and they work under the same premise: if the antibodies in your blood (and there are always antibodies present in your blood) react with a set of proteins that are assumed to be proties from the HIV retrovirus, then it is said that you are "HIV positive". Put bluntly: "antibody reaction means HIV infection".
The question that jumps to the lips on an "HIV positive" test is this: How do I know that my antibodies were NOT cross-reacting? In other words, how do we know that these proteins that you tested for don't share an antigenic determinant with some other protein, and that is why there was an antibody reaction?
And there has never been an answer to that question that wasn't of the form, "You are not a scientist. Trust us."
Do not, under any circumstances, have an "HIV test".
The information from this was gleaned from this fantastic interview with Dr. Valender Turner of the Perth Group. I have exchanged e-mail with him as well. In it, he also describes what HIV researchers actually do to be able to claim that the "HIV test" is "99.999% sensitive and specific". (Hint: unscientific)
DO ANTIBODY TESTS PROVE HIV INFECTION?
posted by Jimmy Gatt at 9:13 AM
1 Comments:
Makes sense....
Post a Comment
<< Home