Viruses replicate inside cells, which shields them from some components of the immune system. In particular, antibodies can’t penetrate inside a cell1 to bind to a virus there, so antibodies are not much use for eliminating a viral infection.2 For some viruses that have to exit the cell to spread to a new target cell, antibody may help limit spread, but many viruses can spread directly from one cell to its neighbor without ever being exposed to antibodies. So once a virus has entered a host’s cells, you probably want mostly cell-mediated immune responses, such as T helper cells and cytotoxic T lymphocytes, to get rid of the virus.
That’s for eliminating viral infections. What antibodies are often extremely good at is blocking infections–stopping the virus from ever getting a foothold. A virus that enters your body has to be exposed to extracellular components at least briefly before it can burrow into its protective cell. During that phase the virus is vulnerable to antibody-mediated inhibition. Antibodies therefore may be relatively unhelpful for getting rid of an ongoing infection, but they can be very good at protecting against new infections.
Not surprisingly, then, most3 antiviral vaccines depend on inducing a strong and specific antibody response. That also means you can often get away with killed virus vaccines like the Salk polio vaccine, or even subunit vaccines like Hepatitis B vaccine; these are very poor at inducing cytotoxic T lymphocyte (CTL) responses, but they don’t have to. Killed vaccines are, in principle, intrinsically safer than the attenuated viruses, or even vector-based recombinant vaccines.
Why are researchers looking for alternatives?
So why is there so much interest in developing vaccines that stimulate CTL? Why are so many groups working on vector-based vaccines or attenuating viruses? One reason is that these vaccines are (again, in principle) intrinsically more immunogenic than killed vaccines. If you can give one dose of vaccine, and then have your antigens stick around for a couple of weeks, or even amplify themselves as they replicate in situ, then you may not need to give a second (booster) dose of the vaccine. That’s a moderate advantage in the first world, and potentially a huge advantage in the third world, where you may only have one chance to visit your patients.
Another reason is that to a large extent we’ve already nailed the simple problems. If a killed vaccine can protect against a major pathogen, we probably already have that vaccine up and running. We’re left with those virus diseases that, for one reason or another, are not easily prevented by antibody-type responses, and so cellular CTL-type responses are the most promising next step.
What keeps a virus from being blocked by antibodies? There are a number of reasons, but the most obvious is that the virus offers a moving target to antibodies. HIV is probably the most famous example of this approach. The HIV surface is dominated by glycoproteins that are enormously variable; an antibody that blocks one particular HIV strain does nothing against a different strain. Hepatitis C virus (HCV) is another virus with highly variable surface proteins. Malaria, a parasite rather than a virus, has enough room in its genome to take this strategy even further. As well as using strain variation, individual parasites can dynamically change their surface proteins, stepping methodically through some 60 variants.4
Surface antigens in these pathogens have probably evolved to be variable; there’s been selective pressure for a pathogen to be different from the majority, since that way they’re less likely to infect an immune host.5 Internal antigens–those that are not exposed to antibodies-tend to be more highly conserved. Internal antigens aren’t exposed to antibodies, but they’re perfectly good targets for CTL. This is one reason for the interest in developing vaccines that induce good CTL responses.
Back to the future: Workarounds for antibody-based vaccines
There’s another approach, though. We have a lot of experience with antibody-based vaccines. It would be nice if there was a way to use them against HIV and HCV. Are there sections of the virus surface that are not variable? If so, then designing a vaccine that raises antibodies against these regions might be effective against many different strains. That’s been a hot topic for quite a while, and in fact there have been some steps forward on this front for HIV.6 More recently, in the latest issue of Nature Medicine7 there’s an article suggesting that some antibodies may be able to neutralize many hepatitis C strains.
The results provide evidence that broadly neutralizing antibodies to HCV protect against heterologous viral infection and suggest that a prophylactic vaccine against HCV may be achievable.
- Yes, I know that some forms of antibody routinely penetrate cells as they’re pumped into the gut, for example, but let’s stay relevant.[↩]
- Perhaps antibodies are important in some cases for triggering antibody-dependent cell-mediated cytotoxicity (ADCC) by NK cells, but again let’s not get sidetracked.[↩]
- If not all. I can’t think of a counterexample offhand[↩]
- Developmental selection of var gene expression in Plasmodium falciparum. Qijun Chen, Victor Fernandez, Annika Sundstram, Martha Schlichtherle, Santanu Datta, Per Hagblom & Mats Wahlgren. Nature 394, 392-395 (23 July 1998) [↩]
- That being said, I don’t know that this has been formally shown for any of these agents; and in fact the only study I know of off the top of my head specifically did not find evidence for frequency-dependent selection in malaria surface protein alleles: Sequence Variation in the T-Cell Epitopes of the Plasmodium falciparum Circumsporozoite Protein among Field Isolates Is Temporally Stable: a 5-Year Longitudinal Study in Southern Vietnam. Amadu Jalloh, Huynh van Thien, Marcelo U. Ferreira, Jun Ohashi, Hiroyuki Matsuoka, Toshio Kanbe, Akihiko Kikuchi, and Fumihiko Kawamoto. Journal of Clinical Microbiology, April 2006, p. 1229-1235, Vol. 44, No. 4Â [↩]
- For example, Structural definition of a conserved neutralization epitope on HIV-1 gp120. Tongqing Zhou, Ling Xu, Barna Dey, Ann J. Hessell, Donald Van Ryk, Shi-Hua Xiang, Xinzhen Yang, Mei-Yun Zhang, Michael B. Zwick, James Arthos, Dennis R. Burton, Dimiter S. Dimitrov, Joseph Sodroski, Richard Wyatt, Gary J. Nabel & Peter D. Kwong. Nature 445, 732-737 (15 February 2007)–the source of the image at top here[↩]
- Broadly neutralizing antibodies protect against hepatitis C virus quasispecies challenge. Mansun Law, Toshiaki Maruyama, Jamie Lewis, Erick Giang, Alexander W Tarr, Zania Stamataki, Pablo Gastaminza, Francis V Chisari, Ian M Jones, Robert I Fox, Jonathan K Ball, Jane A McKeating, Norman M Kneteman & Dennis R Burton. Nature Medicine Published online: 6 December 2007 [↩]
> Are there sections of the virus surface that are not variable? If so, then designing a vaccine that raises antibodies against these regions might be effective against many different strains. That’s been a hot topic for quite a while,
That’s a fascinating idea – and to me (though I’m not learned in vaccinology or HIV) it has that ring of something that really stands a chance of working out in practice. PLoS Pathogens recently had one on HIV superinfection; I read only the abstract:
“[S]tudies to determine the incidence of HIV-1 superinfection have yielded conflicting results. Furthermore, few studies have attempted to identify superinfection cases occurring more than a year after initial infection [...] Among 36 [infected, I think] individuals, we detected seven cases of superinfection, including cases in which both viruses belonged to the same HIV-1 subtype, subtype A. In five of these cases, the superinfecting strain was detected in only one of the two genome regions examined, suggesting that recombination frequently occurs following HIV-1 superinfection. In addition, we found that superinfection occurred throughout the course of the first infection: during acute infection in two cases, between 1–2 y after infection in three cases, and as late as 5 y after infection in two cases. Our results indicate that superinfection commonly occurs after the immune response against the initial infection has had time to develop and mature.”
http://pathogens.plosjournals.org/perlserv/?request=get-document&doi=10.1371/journal.ppat.0030177
While I guess there is already pessimism in some or many quarters about prospect for a “traditional” HIV vaccine, I hadn’t heard that widespread superinfection was a possible phenomenon. Doesn’t seem to bode well for any “traditional,” well-precedented kind of vaccine.
> Yes, I know that some forms of antibody routinely penetrate cells as they’re pumped into the gut, for example, but let’s stay relevant.
I assume that trans-epithelial transcytosis of antibodies is solely endosomal (though I’ve had a hard time googling up a platinum ref on that, especially since I’m off-campus).
So, I am pretty sure that natural “cell-penetrating antibodies” as they’re called, which seem to access the cytosol and nucleus of the intact cell, are a separate phenomenon. I was shocked when I first learned about them, and thought they might bear mentioning despite being off-topic:
“An alternative hypothesis for the renal pathogenicity of anti-dsDNA antibodies was proposed more than 20 years ago, namely that certain autoantibodies could penetrate into living cells and thus induce damage. Work from several laboratories has recently provided firm support for this iconoclastic theory, which contradicted prevailing immunologic dogma that cell interiors are inaccessible to antibodies. Here, we review the evidence [...]”
Autoimmun Rev. 2004 Feb;3(2):7-11.
New approaches to the renal pathogenicity of anti-DNA antibodies in systemic lupus erythematosus.
Putterman C. PMID: 15003182
This subject isn’t a huge concern for me, so I’ve never plowed through the experimental papers to see what I think of said evidence. So far the idea seems not to have been fully baptized by the mightiest journals, though it has been mentioned briefly by Putterman (or perhaps it was Yaakov Naparstek) in the pages of the authoritative and excellent _Dubois’ Lupus Erythematosus_. Research seems to be ongoing, though not exploding in pace. A lot of the stuff one gets when searching is about engineered cell-penetrating antibodies, rather than natural ones.
I assume that trans-epithelial transcytosis of antibodies is solely endosomal
That was one of my thoughts; I thought I had seen something saying that transcytosing antibody went through the cytosol, but I was weaseling because I couldn’t remember for sure. Janeway says
So you’re right, it’s not free in the cytoplasm. (Fortunately, I don’t teach the antibody section of our immunology course.)
I also remembered reading about the cell-penetrating antibodies, but again I couldn’t remember details, and it was off topic enough I thought I could let it slide.
Your superinfection point is a good one, and I was thinking about commenting on the PLOS paper some time; it came as a surprise to me, because I had thought that previous studies had established that superinfection was rare or non-existent, implying that resistance could happen.
However, it doesn’t really change the feasibility of an antibody-based attack on invariant regions of the virus surface. For reasons that I don’t understand, even though there do seem to be reasonably constant regions on HIV they seem to be very poor inducers of antibodies, and the vast majority of the antibody response to the virus is to the variable regions. Teleologically, this makes sense (the virus would “want” to vary regions that are most immunogenic, and then having them variable it would be advantageous to make them more immunogenic 1) but I’m not sure mechanistically how that works. It does mean that the constant regions get identified by screening large numbers of clones of monoclonals.
1. I think it would, anyway, but I’m not sure how much individual and how much population benefit are involved.
> I also remembered reading about the cell-penetrating antibodies, but again I couldn’t remember details, and it was off topic enough I thought I could let it slide.
Oh, sure, it’s quite extraneous, and their very existence may not be 100% certain… I only mentioned it as an interesting whatever. A rare chance for me to add something interesting to your page (which I always read engrossedly)… I sure didn’t get very far with the MHC diversity/nondiversity enigma, despite much contemplation…
I wonder if these cell-penetrating Ig, assuming they are for real, will turn out to have any anti-infective relevence. I certainly haven’t heard of them doing anything except maybe contribute to lupus.
> For reasons that I don’t understand, even though there do seem to be reasonably constant regions on HIV they seem to be very poor inducers of antibodies, and the vast majority of the antibody response to the virus is to the variable regions. Teleologically, this makes sense (the virus would “want†to vary regions that are most immunogenic, and then having them variable it would be advantageous to make them more immunogenic) but I’m not sure mechanistically how that works.
Given that I’m not going very far into the mechanism, perhaps the following amounts to merely repeating what you have just said – and maybe you’ve dealt with this idea already in some of your immunodominance posts. (Perhaps, at least, the refs I’ll cite will be of interest, in any case.) But anyway here goes: how about “epitope competition” – the idea of immunodominance being a function not only of certain peptides loading well or poorly per se (and other things like that), but also being a function of competition between epitopes for immunodominance. The high immunogenicity of the variable domains may actually lower the immunogenicity of the constant domains. As these guys put it:
“Immunodominant epitopes are known to suppress a primary immune response to other antigenic determinants by a number of mechanisms. Many pathogens have used this strategy to subvert the immune response and may be a mechanism responsible for limited vaccine efficacies.”
I was confused for a minute about why they mentioned the “primary” immune response in particular. I think it’s to help contrast what they are talking about against the concept of original antigenic sin, since the latter concerns memory responses specifically. Anyway, it goes on…
“HIV-1 vaccine efficacy appears to be complicated similarly by a limited, immunodominant, isolate-restricted immune response generally directed toward determinants in the third variable domain (V3) of the major envelope glycoprotein, gp120. To overcome this problem, we have investigated an approach based on masking the V3 domain through addition of N-linked carbohydrate and reduction in net positive charge. N-linked modified gp120s were expressed by recombinant vaccinia virus and used to immunize guinea pigs by infection and protein boosting. This modification resulted in variable site-specific glycosylation and antigenic dampening, without loss of gp120/CD4 binding or virus neutralization. Most importantly, V3 epitope dampening shifted the dominant type-specific neutralizing Ab response away from V3 to an epitope in the first variable domain (V1) of gp120. Interestingly, in the presence of V3 dampening V1 changes from an immunodominant non- neutralizing epitope to a primary neutralizing epitope with broader neutralizing properties. In addition, Ab responses were also observed to conserved domains in C1 and C5.”
So at minimum, masking V3 redirected some Ig firepower against V1 – and possibly against some conserved domains as well, though the abstract above seems mildly ambiguous on that point and I don’t feel like reading the whole paper (namely, RR Garrity et al, “Refocusing neutralizing antibody response by targeted dampening of an immunodominant epitope”).
I wonder why they whipped out the N-linked carb masking tape, rather than just cloning the conserved domains alone, adding some adjuvant, and messing around with that.
Anyway it sounds like the variable domains may be significantly effective as a distraction display. But – and maybe I misunderstood you, and the following enigma was actually your whole point – how can a variable region remain hyper-immunogenic? Is the variation, perhaps, low and slow enough that the sequence can retain some agreement with binding motifs of host MHC alleles – or do whatever it is that it does to stay hyperimmunogenic? Not being a viro jock, I had to google for whatever might seem like an answer:
“A major goal of HIV and SIV research is to understand the selective forces driving variation in each of the variable domains of env. It is not the replication error rates that are remarkable for HIV and SIV, but rather the capacity of their envelope proteins to tolerate changes (Coffin 1986). Although rates of [envelope protein] sequence change can be reasonably measured over the course of years, the numbers are too high to permit extrapolation to an evolutionary timescale.”
http://books.google.com/books?id=Awcmfx8EaOAC&pg=PA603&lpg=PA603&dq=hiv+variable+domains+recombination&source=web&ots=MgWnodpcfo&sig=6MnZ9XBakZXFJr16utUORlb-s7Q
Eric, I haven’t forgotten your comment. I don’t have time for a little while, but I want to look into the principles of antibody-based immunodominance (I know a reasonable amound about T-cell-based immunodominance, but not so much about the humoral system, and it’s something I want to pick up anyway). When I’ve looked at it some I’ll probably write a more detailed thing about it.
Oh, sure… please don’t feel obliged to address my comments in your post, of course. Mild chronic illness has me out of biomed academia for at least the near future – but I’ve spent a little time there and know how crazy the life is.
Anyway, as usual, I wrote out my little notes largely for the intrinsic benefits – it powerfully fixes things in memory, and forces me to formalize my thoughts the way I know I should (but don’t want to). A lot of notes I just file away, but it’s always nice to put them out there where they may be of some interest.
Anyway, as usual, I wrote out my little notes largely for the intrinsic benefits – it powerfully fixes things in memory, and forces me to formalize my thoughts the way I know I should (but don’t want to).
Of course, this is exactly the reason I write Mystery Rays.
[...] My research is focused on T cell responses to viruses, so I don’t tend to talk about antibodies all that much here. For that matter, I personally don’t find antibodies very interesting, research-wise. But I don’t want to dis antibodies as clinical entities, and a few recent papers emphasize how useful they can be. (See also my previous post, Antibody-based vaccines) [...]
[...] My research is focused on T cell responses to viruses, so I don’t tend to talk about antibodies all that much here. For that matter, I personally don’t find antibodies very interesting, research-wise. But I don’t want to dis antibodies as clinical entities, and a few recent papers emphasize how useful they can be. (See also my previous post, Antibody-based vaccines) [...]