Just For Fun--Blood Bank Quiz Game!

[Deny Morlino]

1) Bhende YM, Deshpande CK, Bhatia HM, Sanger R, Race RR, Morgan WT, Watkins WM. (May 1952). "A "new" blood group character related to the ABO system". Lancet. 1 (6714): 903–4.

[Dr. Pepper]

Malcolm and Deny, you are correct that Bhende et al first described the phenotype. They investigated three individuals: one was hurt in a railroad accident, the second admitted for a stab wound, and the third a first time donor found in a deliberate search of 160 donors. (Since the incidence of Oh in India is 1:7600, I want that tech in my lab screening units for my next problem patient with multiple antibodies. He/she had the golden touch!). They postulated a new homozygous and recessive allele at the ABO locus, and called the patients' antibodies "the most potent examples of human anti-H so far reported".

Malcolm, you are correct that Levine et al's seminal paper in Blood in 1955 described the relationship between the H, h, A and B genes. They studied a "remarkable family", three members of which were of this unusual phenotype apparently identical to those investigated earlier. Of particular interest was how an apparent group O mother and a group A father could produce a group AB daughter. They correctly concluded that the mother carried a B gene but could not express it herself. They suggested that a homozygous dose of a "x" gene could supress ABO phenotype and secretion of ABH substances. Xx were replaced by Hh, and a hugely important cornerstone in our understanding of blood group serology was laid. And now it's time for my part of the story.

In 1973 I was a fledgling blood banker just weeks out of school. In those days we typed every patient admitted to the hospital. We used A1, A2, B and O cells for the backtype. One of our routine admission patients front typed an unremarkable O but agglutinated the O cells on the backtype. An antibody screen was positive as well, and she agglutinated all panel cells 4+ in all phases but not her own cells. I actually did consider Bombay, and I believe typed her with anti-H which was outdated and didn't have a package insert! I seem to remeber getting a reaction and dismissing Bombay. (I didn't say we were good blood bankers in those days!) We called the floor and found that she was not going to need blood and sent her off to the reference lab at Pfizer. We were thrilled when the report came back saying she was indeed a Bombay. I shared the news with a worker at another hospital, who said they had found one back in the 50s. It soon became obvious that we had been beaten to the serological punch: our patient was the propositus in the Levine study. And Malcolm, you did remember that she was an American of Italian ancestry, but didn't remember that she was from Providence, Rhode Island.

We saw her many times over the years. She had kidney issues and arthritis and had several total knee replacements and redos. She was not a great candidate for autologous donation, having poor viens, a chronic 10 gram hemoglobin and a rather crusty attitude (Levine told her that her blood was so rare that people should pay her every time her blood was drawn.) Eventually she figured out that Bombay blood does not grow on trees and we got a few units frozen away at our local blood center. The routine became only transfuse if the post-op hgb got too low, otherwise give her erythropoetin. The last time we went to tranfuse her, we got the deglyced unit from the blood center and cut off the segment to do the retype (ha!) on the bag and a crossmatch. When the tech returned to the unit she discovered, to her horror, that the last seal on the tubing was faulty and the bag had filled with air! We clamped it off with a hemostat. After some discussion with her surgeon and our medical director, it was decided not to give her the unit.

I am fascinated by the history of science, and in particular our own field of immunohematology. Pretty much all of the trailblazing articles and publications are readily available through your hospital's library and their connected sources, and make for fascinating reading. I have presented several times in seminars on Bombay and other serological oddities, and promise to convert my powerpoint and write out the whole Bombay story, historical and personal, when I have that mythical free moment. There are still some serological irregularities in the Levine study that have yet to be explained (and probably never will).

It was stated earlier in the thread that classic Bombays are Le(a-b-). They are hh and sese. Unless I am very mistaken, I was under the impression that the Lewis transferase can still function and add its L-fucose to the subterminal sugar of type 1 chains and create Lea antigen. Classic Bombays cannot be Le(b+), though, as that requires a Se gene. Both Levine and I found our Bombay patient to be Le(a+b-).

Our profession, sadly but inevitably, has become increasingly depersonalized. Techs don't go up on the floors any more to draw blood. Faces and names are being replaced by bar codes and cup numbers on an analyzer. Once in a while, however, we should pause and reflect that most advances in our science have a human story behind them, often not a happy one. To have met, talked with, tested, and helped our Bombay patient has been a great honor and priviledge.

She passed away of sepsis at our hospital a few days ago at the age of 81.

[Dr. Pepper]

Oops, I forgot to pass on the question baton. Deny or Malcolm?

[Malcolm Needs ☆]

Oops, I forgot to pass on the question baton. Deny or Malcolm?

That was a quite exceptional answer Dr. Pepper, providing much food for thought and for which I thank you very much indeed.

As I say, I do not want to monopolise this thread, so over to you Deny!!!!!!!!!!!!!!!!!!!

:D:D:D:D

[Deny Morlino]

Awesome story!! I have to agree that even in a smaller hospital as we focus on crossing our "t's" and dotting our "i's" we lose focus on the patients. Thanks for sharing a fascinating story here.

I vote for Malcolm to post the next question.

I see Malcolm beat me to the punch. I will post a question shortly.

Edited June 16, 2010 by Deny Morlino
Beaten to the punch

[L106]

Thank you so much for sharing that story with us, Dr. Pepper. Isn't it incredible when you think of all the historic discoveries and investigations that have taken place involving "real live patients" and "normal blood bank personnel" during our early career years (and just a few years prior to that)? Hard to believe that your propositus just recently passed away; the typical laboratory student of today will probably assume that the original individual from the Levine study likely passed away 40 or 50 years ago! I still find this field exciting and fascinating, after all these years!

Until a few years ago, our Blood Bank personnel issued and delivered the donor units right to the patients' bedsides. We were all thrilled when we discontinued the practice, considering how much time running all over the hospital took us away from the "bench". Don't tell my staff that I said this, but I do miss that patient contact. Even though our encounters were rather brief, you had a face that went with the name, you got to see them through good and not-so-good times, and sometimes developed a relationship with some of their family members.

Since she lived to the age of 81, she obviously received good care at your facility all these years, Dr. Pepper.

[Dr. Pepper]

I would hope that she received good care, although after one knee redo she fell and messed it up and it had to be redone again She was convinced that we had screwed it up and went to a different hospital. I'm not sure that we were sorry to see her go there. All of her surgeries seemed to be balancing acts and it was sometimes difficult to explain to the doctors that if we deglyced her units to have them ready and they didn't need them we would be wasting priceless, irreplacable blood. Once her surgeon told me they'd be extra special careful and wouldn't need blood. This disturbed me as one would hope that they would always be careful; did this mean they might do a crappy job on my knee because I'm a run-of-the-mill O Pos? Anyhow, we know the moral of this story: Never get sick, because then you might get treated!

[LisaM]

So did we answer the latest question? If so, here's the next one:

Of the following antibodies, which one or ones do not occur naturally, as in they don't appear without any known stimulus: M, Lewis-a (Lea), P1, Duffy-B (Fyb), N and Kpa ?

[Deny Morlino]

Fyb and Kpa do not occur naturally. The remainder listed do occur naturally although some are much less likely to be found naturally occurring than others. (I think)

What is the suspected reason Anti-D does not activate compliment?

[Malcolm Needs ☆]

Fyb and Kpa do not occur naturally. The remainder listed do occur naturally although some are much less likely to be found naturally occurring than others. (I think)

What is the suspected reason Anti-D does not activate compliment?

Hmm, I'm not sure that I totally agree with you Deny (sorry)! The first ever anti-Kpa described, that found in Mrs. Penney, was said to be "naturally occurring" (Allen FH, Lewis SJ. Kpa [Penney], a new antigen in the Kell blood group system. Vox Sang 1957; 2: 81-87), and I've certainly seen quite a few where there was no known stimulus (untransfused males and the like), but hey, I'm quibbling (again).

Guess what?! When I was extremely young and working at the BGRL in London, before it became the IBGRL and moved to Filton in Bristol, they told me of an anti-D (Ripley) that was complement binding; but I take your point!

:redface::redface:

Edited June 16, 2010 by Malcolm Needs
Was I ever extremely young - if so, I can't remember it!

[L106]

What is the suspected reason Anti-D does not activate compliment?

I believe it is because the spacing arrangement of the Rh antigens on the red cell surface is not condusive to activating complement. (Correct me if I'm wrong.)

Next question: Which low incident antigen is much higher in the Mongolian population than in any other population?

[Eagle Eye]

what is Dia? (Diago a)

What BGS antigens are sensitive to all enzymes(Ficin, Papain, Bromelin, Trypsin & a-Chymotrypsin?

[Han Chau]

I believe it is because the spacing arrangement of the Rh antigens on the red cell surface is not condusive to activating complement. (Correct me if I'm wrong.)

Next question: Which low incident antigen is much higher in the Mongolian population than in any other population?

I agree with you. Two RhD antigens on RBC surface are not close enough to be linked by C1q, so complement can't be activated.

Cheers

Han

[Malcolm Needs ☆]

I agree with you. Two RhD antigens on RBC surface are not close enough to be linked by C1q, so complement can't be activated.

Cheers

Han

I agree with you both that this is the "received" reason, but I do wonder if that is all there is to it. If this were completely the answer, one might expect more human anti-D reagents to cause complement haemolysis when tested against samples of blood from individuals with an exulted D phenotype, such as -D-/-D-, but they don't (even though such red cells are agglutinated by some IgG anti-D reagents, without potentiators, proving that the D antigens are very close to one another on such red cells).

I don't really understand what is this other "unknown" though.

There was one theory that the complement pathway is activated, but that the antigens are too far from the red cell membrane, so that when the activated complement comes off the antibody, and falls onto the membrane, it decays before it has time to activate the next part of the pathway. I'm don't know whether or not this is true.

:confused::confused:

Edited June 17, 2010 by Malcolm Needs
Forgot a bit.

[Han Chau]

Thanks Malcolm for your question since that the thing I've never thought about when studying about Rhesus blood group system :tongue:

However, from brainstorm here is my possible explanation:

R2R2 (I'm talking about exceptional -D- later) has the largest number of D antigen site on RBC (approx 16,000 to 33,000) compared to other Rhesus phenotypes but it still can't activate complement while Jk with lower number of antigen sites on RBC (approx 14,000) but CAN activate complement due to the distribution and arrangement of antigen. Jk antigens appears as cluster, not randomly appearing as RhD. Therefore, I thought the way the Ags go together on the RBC surface is more important than the number of Ag site (in some aspect)

However, with -D-, it has the largest number of antigen site (110,000 - 202,000) but it can't either (as you said). In -D-, there is an increase of D expression which makes 2 IgG very close for anti-IgG to crosslink and cause agglutination but possibly because of its random distribution, it is still too far for C1q to cross link. I think in this case it relates to the distance between 2 IgG and the size of C1q.

This is just my thought without references except the given numbers . However, I'll do a search on that after my exam time is finished

And for the theory you mentioned, I don't think it's reasonable. From what I was taught, complement bind very strongly to RBC by covalent bond once it's activated (and that's why in CAD, patient's RBC have DAT positive for complement only). The only complement component that can flow off the RBC is C5b, and C5b is formed once complement cascade was already activated (this means the related RBC was already attacked by complement). In Rhesus, complement hasn't been activated, how is C5b formed to flow off the RBC? That's why I thought it's not the answer. However, my knowledge is still limited, and possibly my explanation here is not....correct .

I will do a search on that after my exam. At the meantime, if any of you have the answer, please let me know so that I can learn from that. Thanks

And once again, thank Malcolm for you question

Cheers

Han

Edited June 17, 2010 by Han Chau
some typing mistakes

[Deny Morlino]

Malcolm,

Never a need to apologize for disagreeing with me or quibbling! I learn invaluable information from you frequently and hope you and others here will "set me straight" in the quest to become a better blood banker.

I agree with the approach that the D antigen explaination is incomplete but currently the excepted theory we have. I was hoping a discussion would ensue (as has happened) regarding this topic. Thanks for the opinions and theories / arguments. This will be an interesting subject to watch for developments.

L106 gave the answer I was looking for and has posted a question. Aakupaku also posted a pair of questions. Keep this going please as the information has been challenging and stimulating!!

Thanks,

D

[Deny Morlino]

OK Aakupaku threw me off by answering a question Jeopardy style lol. Sorry to appear dense .

[Malcolm Needs ☆]

=Han Chau;27956]Thanks Malcolm for your question since that the thing I've never thought about when studying about Rhesus blood group system :tongue:

However, from brainstorm here is my possible explanation:

R2R2 (I'm talking about exceptional -D- later) has the largest number of D antigen site on RBC (approx 16,000 to 33,000) compared to other Rhesus phenotypes but it still can't activate complement while Jk with lower number of antigen sites on RBC (approx 14,000) but CAN activate complement due to the distribution and arrangement of antigen. Jk antigens appears as cluster, not randomly appearing as RhD. Therefore, I thought the way the Ags go together on the RBC surface is more important than the number of Ag site (in some aspect)

However, with -D-, it has the largest number of antigen site (110,000 - 202,000) but it can't either (as you said). In -D-, there is an increase of D expression which makes 2 IgG very close for anti-IgG to crosslink and cause agglutination but possibly because of its random distribution, it is still too far for C1q to cross link. I think in this case it relates to the distance between 2 IgG and the size of C1q.

This is just my thought without references except the given numbers . However, I'll do a search on that after my exam time is finished

And for the theory you mentioned, I don't think it's reasonable. From what I was taught, complement bind very strongly to RBC by covalent bond once it's activated (and that's why in CAD, patient's RBC have DAT positive for complement only). The only complement component that can flow off the RBC is C5b, and C5b is formed once complement cascade was already activated (this means the related RBC was already attacked by complement). In Rhesus, complement hasn't been activated, how is C5b formed to flow off the RBC? That's why I thought it's not the answer. However, my knowledge is still limited, and possibly my explanation here is not....correct .

I will do a search on that after my exam. At the meantime, if any of you have the answer, please let me know so that I can learn from that. Thanks

And once again, thank Malcolm for you question

Cheers

Han

Gosh, I had to back to my books for this one (particularly as complement is not my strongest [or favourite] subject). My brain hurts now!!!!!!!

There is no doubt that the Kidd multipass proteins are clustered on the red cell surface, and you are most certainly correct in saying that this is why they nearly all are able to initiate the classical complement cascade, but, just as importantly, many Kidd antibodies are a mixture of IgG and IgM, and, of course, a single IgM antibody molecules can initiate the classical complement cascade.

You may be correct that the D polypeptide is distributed far and wide on the red cell surface, and this prevents initiation of the classical complement cascade, BUT, it only takes about 1000 IgG molecules bound to antigens to initiate this cascade. In the case of an exaulted D, I am certain that there simply must be at least this number of antibodies bound to antigen, and in close enough proximity for, at least, the theoretical initiation of the cascade. I could be wrong; I am just thinking aloud!

From re-reading my books (in particular Issitt and Anstee), it is C4b that binds to the red cell membrane after activation of C4 by C1qrs (I've seen a lecture with animations that showed this molecule, sort of floating down from the antibody - but I have to admit to two things about this; 1] it was a long time ago and 2] I had been "networking" with a few colleagues the night before, and was very badly hung-over when I heard the lecture), so I could be completely wrong.

On the subject of the theory, however, I was just putting it forward as just one theoretical explanation that I had heard/read about, and I am none too sure that it is correct (as I say, complement is NOT my favourite subject).

Others, with a greater knowledge in this area, PLEASE COME TO MY ASSISTANCE!!!!!!!!!!!!!!!!!!!!

:eek::eek::eek:

[LisaM]

what is Dia? (Diago a)

What BGS antigens are sensitive to all enzymes(Ficin, Papain, Bromelin, Trypsin & a-Chymotrypsin?

Did anyone answer this one yet? If not, I'll venture a guess: Rh, Kidd and Lewis systems?

Next question: In cases of TRALI, what does a person make antibodies against?

[Han Chau]

TRALI caused by anti-HLA in donor's plasma, so anti-HLA is the answer, isn't it?

My question is: What may be other cause of TRALI?

[Yanxia]

And anti-HNA.

[Yanxia]

QUESTION: What is the cause of lea+b+?

[Malcolm Needs ☆]

Did anyone answer this one yet? If not, I'll venture a guess: Rh, Kidd and Lewis systems?

Next question: In cases of TRALI, what does a person make antibodies against?

I'm afraid not Lisa; the antigen/antibody reactions of all of these Blood Group Systems are enhanced by the action of enzymes. Those that are sensitive are Chido/Rodgers (although some very rare, really strong examples of anti-Ch will react with papain-treated red cells), the Indian BGS, JMH and Xg. The low incidence antigen in the 701 series, Bp(a), is also senstivie to these enzymes.

:redface::redface:

[Han Chau]

=Han Chau;27956]

Gosh, I had to back to my books for this one (particularly as complement is not my strongest [or favourite] subject). My brain hurts now!!!!!!!

There is no doubt that the Kidd multipass proteins are clustered on the red cell surface, and you are most certainly correct in saying that this is why they nearly all are able to initiate the classical complement cascade, but, just as importantly, many Kidd antibodies are a mixture of IgG and IgM, and, of course, a single IgM antibody molecules can initiate the classical complement cascade.

You may be correct that the D polypeptide is distributed far and wide on the red cell surface, and this prevents initiation of the classical complement cascade, BUT, it only takes about 1000 IgG molecules bound to antigens to initiate this cascade. In the case of an exaulted D, I am certain that there simply must be at least this number of antibodies bound to antigen, and in close enough proximity for, at least, the theoretical initiation of the cascade. I could be wrong; I am just thinking aloud!

From re-reading my books (in particular Issitt and Anstee), it is C4b that binds to the red cell membrane after activation of C4 by C1qrs (I've seen a lecture with animations that showed this molecule, sort of floating down from the antibody - but I have to admit to two things about this; 1] it was a long time ago and 2] I had been "networking" with a few colleagues the night before, and was very badly hung-over when I heard the lecture), so I could be completely wrong.

On the subject of the theory, however, I was just putting it forward as just one theoretical explanation that I had heard/read about, and I am none too sure that it is correct (as I say, complement is NOT my favourite subject).

Others, with a greater knowledge in this area, PLEASE COME TO MY ASSISTANCE!!!!!!!!!!!!!!!!!!!!

:eek::eek::eek:

Malcolm, you're completely right (your memory serves you ) when you said that C4b floated off from the ANTIBODY(IES) after complement was activated (mentioned in the book of Issitt and Anstee). You talked about floating off from ANTIBODY, and I talked about floating off from RBC MEMBRANE , hehe

Also from the book, I feel like many authors have tried to figure out why Rh blood group Ab doesn't fix complement and the question is still opened. However, the book is from 1999, I'm not sure if there is any new research done after that.

However, after reading the book, personally, the explanation of 2 IgG are too far for complement to bind to is still....not satisfying me (although it's clear that RhD antigen widely spread on RBC membrane) since the IgM of Rh blood group system doesn't activate complement either. So, personally there is possibly some thing on the Rh blood group Ag or Ab which prevents the complement from being activated. If it's the problem of the distance, IgM can activate irrespective of that, indeed it's not. However, I'm just wondering since it may be hard to find the final answer.

And...about the hypothesis you mentioned (it's Rosenfield's explanation) , the book has one part that mentions about it. However, at the end of the paragraph, the authors said that it's possibly not because Rh antigens have seemed to be 'embedded in or located very closed to RBC membrane', so it's less likely that C4 decays before it reaches the membrane. However, the authors didn't completely deny the hypothesis (I think) since they mentioned it somewhere in another chapter in the book as many mentioned explanations for why Rh Ab doesn't fix complement.

If we both are not satisfied with the information from the book, let's do some more search about recent findings about it to ensure that the problem is really pending there .

Edited June 20, 2010 by Han Chau
some correction

[Malcolm Needs ☆]

I'll give it a go Han Chau, but I do maintain that complement is NOT my favourite subject (so don't hold your breath)!

:):)