Mystery Rays from Outer Space

“Adenovirus” (by Mapposity)

There are two aspects about virology that constantly amaze me: How much we know about viruses, and how little we know about viruses.

Adenovirus research offers examples of both. Adenoviruses are probably among the best-studied virus groups.¹ We really do know an amazing amount about them. But it was only last year that Linda Gooding’s group offered the most convincing demonstration yet that adenoviruses actually establish a truly latent infection — a really basic aspect of their lifestyle, ² and a new paper from her group³ is looking at some equally-basic implications of that finding. (I talked about Gooding’s earlier latency finding here.)

It’s been known pretty much since day 1 that adenoviruses persistently infect tonsils;⁴ that was why they were first isolated, when the virus grew out of apparently-normal tonsil tissue in culture. The critical distinction is between mere “persistence” and true “latency”. In a latent infection, the virus shuts down production of new viruses, and is maintained basically as DNA within the host cell. Persistence is cruder — the virus continues to replicate, but at a low level that balances its destruction. Simplistically, latency is a destruction-free process, while persistence can include viral and cellular destruction.

Adenoviruses establish their latency in tonsils, which of course have lots of lymphocytes, but we usually think of adenoviruses as infecting epithelial-type cells, or hepatocytes, or whatever. Clinically, these guys typically cause cold-type symptoms, which you tend to get from fairly superficial infections of the respiratory tract lining. We don’t tend to think of adenoviruses as effective infectors of lymphocytes, but it turned out that their latent infection was, in fact, in T lymphocytes.  It looks like adenoviruses have one cell type (epithelial-type cells) for a lytic infection that leads to shedding of infectious virus, and another cell type for latent infection, allowing the virus to remain in the host and potentially re-infect an epithelial type later on.

Accordingly, Gooding and her team set up infections of cultured T lymphocytes in vitro, to see what would happen. In particular, they wanted to know whether, and how, the viral replication cycle would be controlled; and whether and how the host cell would be affected by the infection. I will skip over most of their findings and and highlight a couple that surprised me:

(1) The “Occupied!” sign. To get into a cell, adenoviruses usually need to bind to their cellular receptor, the CAR receptor.⁵ But latently-infected cells almost permanently shut off this receptor. For hundreds of days after the initial infection, cells express little or no CAR. The latent virus doesn’t want any competition; it has found a congenial long-term environment, and it doesn’t want some interloper infecting its cozy cell and perhaps destroying it.

There seem to be several mechanisms for the shutdown, but at least part of it is that the virus apparently permanently modifies the host DNA:

CAR synthesis and expression remained repressed even after the viral genome was lost (Fig. 8 and data not shown), suggesting a virus-induced epigenetic change to the cells that does not require the continued presence of the virus.³

And in fact the CAR isn’t the only thing to be modified for this purpose:

Even when CAR levels were restored by transduction with a CAR-containing retrovirus, the previously infected cells could not be reinfected³

We don’t know how the latent viruses were blocking superinfection, but it’s clear that the latent viruses really don’t want company.

(2) Rearranging the furniture.  The latent virus doesn’t stop at hanging an “occupied” sign; it modifies its host cell in other ways as well, apparently again by long-term or even permanent epigenetic modification of the DNA. That means that even after the virus itself is altogether gone, not even latently present, there are modified cells hanging about:

Remembering that adenoviruses infect just about everyone, that may mean that we’re all walking around carrying cells that are tagged and functionally altered by these viruses.

There’s been speculation for many years that adenovirus infection may underlie some forms of human tumors. One argument against this has been that there’s no evidence of adenovirus DNA in tumors, for the most part.⁶ (One rule of thumb in determining if a virus is actually causing a tumor is if it’s actually present in the tumor.) But of course, if adenoviruses leave a permanent scar on cellular DNA that lasts longer than the virus itself, this may not be relevant:

One compelling reason to gain an understanding of this nonlytic infection is the likelihood that adenovirus gene products cause damage to the host cell genome. … While these functions are irrelevant to the lytic infection of epithelial cells where all infected cells die, they are of serious concern when infected lymphocytes have carried the viral genome and survived. … Despite this normal appearance, the cells display altered gene expression long after the virus is lost.³

1. There are over 40,000 papers on adenoviruses, or at least mentioning them, in PubMed.[↩]
2. To be fair, it’s been suspected for decades that they do go latent, but that was the first time it was actually proven.[↩]
3. Zhang, Y., Huang, W., Ornelles, D., & Gooding, L. (2010). Modeling Adenovirus Latency in Human Lymphocyte Cell Lines Journal of Virology, 84 (17), 8799-8810 DOI: 10.1128/JVI.00562-10[↩][↩][↩][↩]
4. I’m going to limit this discussion to the Group C adenoviruses — the latency concept may be true for other groups of adenoviruses but that hasn’t been directly shown.[↩]
5. “CAR” stands for the “Coxsackie B virus and Adenovirus Receptor”. Can anyone guess what other virus uses this receptor? Bueller? Anyone?[↩]
6. Also, the epidemiological links between tumors and adenoviruses are not very strong, at least in humans.[↩]

Dendritic cells in the skin (Langerhans cells) form a dense network of “sentinels” that act as first line of defense of the immune system.1

There’s a lot of interest in using dendritic cells as vaccines these days.  A paper in PLoS One² offers a cautionary note.

Dendritic cells (DC) are the main cell type that drive T cells from their normal naive state to an active state.  In the naive state, a T cell can recognize its target, but doesn’t do anything about it; in the active state, the T cell does something, ranging from spreading inflammation to killing infected cells, and so on.  The DC is needed to bridge these states.  DC do many things, but at the simplest level they connect  the presence of an antigen (a T cell target, in this case) with the presence of something dangerous or abnormal — a pathogen, or tissue damage.

There are some conditions where we’d like an immune response, where DC don’t detect one or the other of their components (i.e. antigen or danger).  For example, there may be a situation that we know is dangerous, but where there’s  little evidence of “danger” for the DC.  A vaccine, for example, doesn’t want to deliver a huge amount of tissue damage, but we’d still like to get a strong response to an antigen.  For a natural situation, cancers are often ignored by the immune system even though there may be lots of cancer antigens, and one reason (of many) for this ignorance is that the DC may not perceive a lot of danger in the context of the cancer.

So why not take the DC out of the system, alarm them with some danger information in the test tube, load them up with antigen, and then return them to the body? That’s called a dendritic cell vaccine, and there’s fairly intense interest in the approach.

There’s been some success using this approach, but perhaps less than you’d expect from the biology as we understand it.

Several clinical trials conducted over the past decade have demonstrated that DC vaccines can prime and boost antigen-specific CD8+ T cells in humans. However, their clinical efficacy remains to be definitively demonstrated [6], [19], [20], [21]. The lack of success has been variously attributed to several factors: administration of relatively low cell numbers of DCs, suboptimal route of administration, improper antigen dose, poor choice of antigenic targets, unsuitable maturation state of DCs, and inappropriate frequency of injections. However, understanding exactly which of these concerns represent true problems may be difficult because little is known regarding the fate and function of ex vivo generated DCs after they have been injected ²

Yewdall et al asked what happens to DC after they’re given this course and returned to the patient (mice, in this case).  Their surprising conclusion is that the DC don’t work to prime T cells directly.  Instead, they have to hand off their antigens to other cells in the body that have never left:

Contrary to previous assumptions, we show that DC vaccines have an insignificant role in directly priming CD8+ T cells, but instead function primarily as vehicles for transferring antigens to endogenous antigen presenting cells, which are responsible for the subsequent activation of T cells. … This reliance on endogenous immune cells may explain the limited success of current DC vaccines to treat cancer and offers new insight into how these therapies can be improved. Future approaches should focus on creating DC vaccines that are more effective at directly priming T cells, or abrogating the tumor induced suppression of endogenous DCs. ²

As always in science, a single paper needs to be confirmed by others, so we won’t get too distressed until we see if other groups replicate this, and if it’s a universal truth or something specific to the particular system these authors were looking at.  (And, of course, this doesn’t trump actual evidence of efficacy for DC vaccines.) My own suspicion is that the work is accurate but limited, and there’s something about this particular system which prevented the transferred DC from being good primers; but as I say, I’d like to see some followup from another group.

1. Tolerogenic dendritic cells and regulatory T cells: A two-way relationship. (2007) Karsten Mahnke, Theron S. Johnson, Sabine Ring and Alexander H. Enk. J of Derm Sci 46:159-167 doi:10.1016/j.jdermsci.2007.03.002 [↩]
2. Yewdall, A., Drutman, S., Jinwala, F., Bahjat, K., & Bhardwaj, N. (2010). CD8+ T Cell Priming by Dendritic Cell Vaccines Requires Antigen Transfer to Endogenous Antigen Presenting Cells PLoS ONE, 5 (6) DOI: 10.1371/journal.pone.0011144[↩][↩][↩]

The immune system is, by its nature, destructive. Its function is to eliminate problems. Because it’s so destructive, there are many layers of control that constantly check and limit the response. Equally, there are controls that try to ensure that the response doesn’t start until it’s needed.

How does the immune response know when it’s needed? It has to eliminate problems, which means it needs to detect problems. So, what’s a problem?

In general, the immune system perceives two conditions as “problems”. One is when microbes are detected, and another is when damage is detected. These conditions are both detected through specific sets of receptors, and both lead to similar cascades of events that culminate in the response we think of as classically “immune” – first an innate immune response, and then if appropriate, an adaptive immune response that’s triggered by the initial innate response.

I’ve talked before about these two forms of problem detection. To summarize and grossly oversimplify some of the history: Charlie Janeway predicted the first form, which we now call “Pathogen-associated molecular pattern” (PAMP) detection; Polly Matzinger predicted the latter form, which we can call “Danger-associated molecular pattern” (DAMP) detection. PAMPs include things that are unique to bacteria or viruses — cell wall components that are present in bacteria, but not in vertebrate cells, for example; or double-stranded RNA, which is found in lots of viruses but would be unusual in our own cells. “Danger” signals, on the other hand, are indications of cell death — internal components of a cell, for example, that have leaked out as the cell dies. ¹ For a while, it looked as if PAMPs were the major signal leading to innate and then adaptive immunity, but more recently it’s become clear that DAMPs are also very important.

One example of DAMP recognition would be tumor recognition. We know that tumors are recognized by the immune system — by T cells and B cells, which are adaptive immunity. We know that adaptive immunity is very inefficient without an innate response to set up the proper conditions. We also know that tumors aren’t pathogens as such, and so you wouldn’t expect them to trigger PAMP receptors. So what’s triggering the immune response to the tumor? The answer seems to be DAMPs. As tumor cells die, which they tend to do much more exuberantly than normal cells, they release internal components that the immune response registers as evidence of danger. ² It’s even been proposed that the massive tumor cell death caused by chemotherapy is the real reason chemo works: The cell death is detected by the immune system as evidence of massive danger, and it’s the resulting immune response that actually eliminates the tumor, not the chemo per se.

Neutrophil

So, historically, DAMPs are DAMPs and PAMPs are PAMPs, and never the twain shall meet. After all, internal cell components are quite different from microbes, right?

Well, except for the internal cell components that actually are microbes. Mitochondria, of course, are actually exceedingly symbiotic bacteria that live inside our cells, right? And it turns out that, yes, some DAMPs actually are PAMPs, because some of the danger responses are actually triggered by mitochondrial components that are really bacterial in origin. A lovely paper from Carl Hauser’s lab³ shows that mitochondrial components, released from cells after damage, trigger innate immune responses through pathways that are more traditionally associated with pathogen-specific patterns.⁴

As I say, immune responses can be very destructive, and Hauser’s interest in this arises from the destructive aspect.  Trauma that produces lots of tissue damage can lead to severe inflammation that looks a lot like sepsis, even though there are no bacteria involved, so he has been looking for triggers for this sterile systemic inflammatory response syndrome (SIRS):

Cellular disruption by trauma releases mitochondrial DAMPs with evolutionarily conserved similarities to bacterial PAMPs into the circulation. These signal through innate immune pathways identical to those activated in sepsis to create a sepsis-like state. The release of such mitochondrial ‘enemies within’ by cellular injury is a key link between trauma, inflammation and SIRS.³

I found it particularly interesting  that one of the mitochondrial DAMPs is formylated peptides. Formylation of peptides is typical of bacteria, not eukaryotes, so it’s a good way of detecting pathogens. Indeed, there are receptors for formyl peptides on neutrophils (FPR1), among other cells, and the mitochondrial DAMPs (including the formyl peptides) cause neutrophils to migrate toward the source (chemotaxis) — see the movie below:

(Compare to this other movie I posted a while ago, which shows neutrophils in a mouse’s ear, being attracted to areas of tissue damage.)

H-2M3 crystal structure5

In fact, formylated peptides have been long known to be a PAMP, but not just via the FPR1; they’re also presented by a mouse non-classical MHC class I molecule, H-2M3.  (I didn’t include a picture of H-2M3 in my Field Guide to the MHC earlier, so here’s a picture to the left. ⁵ Heavy chain in grey, beta₂-microglobulin⁶ in red, peptide in green with the formylated end of the peptide — see how neatly it tucks into the peptide-binding groove there? in magenta.) And again, H-2M3 presents formylated peptides, not just from bacterial pathogens, but also from mitochondria. ⁷

Most people probably don’t think of MHC-family molecules as innate immune detectors, but many of the non-classical MHC molecules are true PAMP receptors (pattern recognition receptors, PRRs). It’s even been argued — based on H-2M3 itself, in fact — that this broad pattern-recognition ability is the original function of MHC molecules, and the role of MHC molecules in adaptive immunity is the latecomer:

F. M. Burnet asserted that it was their polymorphism that made MHC genes biologically significant. Certainly this is true for I-a⁸ function, but modern PRR-like I-b molecules⁹ suggest an alternate model for MHC origins. … Because most genes are monomorphic or minimally oligomorphic, and most class I-like genes not linked to the MHC are monomorphic, parsimony suggests the ancestral MHC locus was also monomorphic. This primitive MHC molecule, functioning as a PRR, would have been preadapted for the evolution of polymorphic class I-a molecules in the evolving adaptive immune system. ¹⁰

1. In apoptosis, the programmed cell death that’s a normal part of tissue growth, internal cell components are carefully packaged up in such as way as to prevent this kind of response.[↩]
2. Which, of course, it is.[↩]
3. Zhang, Q., Raoof, M., Chen, Y., Sumi, Y., Sursal, T., Junger, W., Brohi, K., Itagaki, K., & Hauser, C. (2010). Circulating mitochondrial DAMPs cause inflammatory responses to injury Nature, 464 (7285), 104-107 DOI: 10.1038/nature08780[↩][↩][↩]
4. Thanks to Alex Ling, who wrote to me suggesting I talk about this paper.  I’d filed it in with the other 512 papers that I want to talk about here, some time, but her email made me take another look and appreciate how neat the work is.[↩]
5. Wang, C. R., Castano, A. R., Peterson, P. A., Slaughter, C., Lindahl, K. F., and Deisenhofer, J. (1995). Nonclassical binding of formylated peptide in crystal structure of the MHC class Ib molecule H2-M3. Cell 82, 655-664.[↩][↩]
6. Why doesn’t the beta symbol ? stick? No matter how often, or how, I enter the code, it changes to a ? in the published post.[↩]
7. Loveland, B., Wang, C. R., Yonekawa, H., Hermel, E., and Lindahl, K. F. (1990). Maternally transmitted histocompatibility antigen of mice: a hydrophobic peptide of a mitochondrially encoded protein. Cell 60, 971-980.
   Shawar, S. M., Vyas, J. M., Rodgers, J. R., Cook, R. G., and Rich, R. R. (1991). Specialized functions of major histocompatibility complex class I molecules. II. Hmt binds N-formylated peptides of mitochondrial and prokaryotic origin. J. Exp. Med. 174, 941-944.[↩]
8. I-a are the classical MHC class I molecules that present peptides to cytotoxic T lymphocytes[↩]
9. And, logically enough then, I-b are the non-classical MHC molecules that often present fairly conserved molecules.[↩]
10. Doyle, C. K., Davis, B. K., Cook, R. G., Rich, R. R., and Rodgers, J. R. (2003). Hyperconservation of the N-formyl peptide binding site of M3: evidence that M3 is an old eutherian molecule with conserved recognition of a pathogen-associated molecular pattern. J. Immunol. 171, 836-844.[↩]

Clonal evolution during in situ to invasive breast carcinoma progression1

What’s a tumor?

In some ways, that’s a bad question (never mind the answer) because it implies that a tumor is a single thing. But we know that’s not true. A tumor, by the time we can detect it, is a collection of many cells, at least billions of them, and those cells are not all the same. I’m not even talking about the normal cell types that are incorporated into a tumor (things like blood vessels and support cells). Even cells that are unambiguously cancerous are very different within a tumor. And of course, that’s important for the things we’re most interested in, prognosis and treatment, because it’s not the average tumor cell that we’re most concerned about, it’s the subset of tumor cells that are most resistant to treatment, or that are most aggressive.

The development of this variation is really fundamental to how we understand tumor formation and tumor growth. Cancerous cells don’t just appear, fully ready to metastasize and grow. What happens is that a normal cell mutates slightly and gains a little advantage. Most of its progeny stay like that, but one of them mutates again and changes a little more, and then one of that cell’s progeny mutates again, and so on. It probably takes at least a half-dozen mutations, over many cell generations, before a normal cell has progressed through to a detectably cancerous cell.  (I’ve talked about this before, here.)

Also, since truly normal cells simply don’t mutate that many times — there are too many checks and repair systems to allow a half-dozen mutations to accumulate in a single human’s² lifetime — one of the mutations is probably in the check/repair system, turning the cancerous pathway into a mutator pathway as well.

So we expect tumors to be made up of many different cell types, and this is indeed what we see:

With rare exceptions, human malignancies are thought to originate from a single cell, yet by the time of diagnosis, most tumors display startling heterogeneity in cell morphology, proliferation rates, angiogenic and metastatic potential, and expression of cell surface molecules. ¹

So how diverse are tumors?

That’s been a hard question to answer, because you’d need tools to look at individual cells, and you’d also need some way of expressing that diversity. A recent paper¹ looked at diversity in breast cancer using some individual-cell tools, which I’m not going to discuss, and took an interesting approach to describing the variability:

… we applied diversity measures from the ecology and evolution sciences to our copy number data. These diversity measures estimate the number and distribution of species in a certain geographical area or environmental niche. In our context, a species is a cancer cell population … Hence, a region of a tumor containing cancer cells with 3 different copy number ratios is interpreted to contain 3 distinct “species.” ¹

They suggest that this way of describing tumors could be a useful aid to prognosis and to predicting response to therapy, offering a quantitative description of tumor variability (which might correlate with the tumor’s potential for spread and escaping treatment).

I hadn’t thought of tumors as ecosystems before, but I wonder if the analogy could be taken further by considering, say,cytotoxic T lymphocytes as predators …

1. Park, S., Gönen, M., Kim, H., Michor, F., & Polyak, K. (2010). Cellular and genetic diversity in the progression of in situ human breast carcinomas to an invasive phenotype Journal of Clinical Investigation DOI: 10.1172/JCI40724[↩][↩][↩][↩]
2. let alone a mouse’s lifetime[↩]

Smallpox pustules (R. Carswell, 1831)

In contrast to the many viruses that block antigen presentation by MHC class I, only a handful appear to block presentation by MHC class II.  I don’t understand why any would try to block MHC class II in the first place, but another example of it has just been published.

A little background: Major histocompatibility complexes (MHC) are recognized by T cells. T cells come in several flavors, the best-understood of which are CD4 (T Helper) and CD8 (cytotoxic T lymphocyte; CTL) lymphocytes. CD8 T cells are fairly specialized to deal with cells infected with viruses;¹ they recognize MHC class I. CD4 T cells are at the top of the adaptive immune response; they coordinate subsequent responses, by calling in other cell types, driving antibody or CTL responses, and so on.

MHC class I is on the surface of most cells, as you’d expect, because most cells can be infected with viruses. MHC class I is, among other things, a way of directing the CTL attack to the appropriate, virus-infected, cell, and so they deal, fairly strictly, with what’s going on inside their own particular cell. They don’t take up proteins from outside the cell, because then the cell might get killed when it’s actually a neighbor that’s infected. ²

MHC class II, on the other hand, is a general alarm call that signals “Something’s invading the body, somewhere”. MHC class II is only on a limited number of cells, but those cells do take up protein from outside themselves and show it to CD4 T cells. Presentation on MHC class II does not mean that the particular cell is infected.

So it’s quite logical that viruses would be interested in blocking MHC class I, and as I say there are now many examples of viruses that do so. It’s also logical for viruses to want to block MHC class II, since doing so would reduce all the immune responses against them — antibodies, T cells, whatever.

But how would that work? Again: The cells that do MHC class II antigen presentation are not necessarily infected cells. If a virus is going to block MHC class II, it would have to go out of its way infect the MHC class II-presenting cells (known as professional antigen-presenting cells; APC). Not only that, it would probably have to infect a lot of them, to make a real impact on the overall CD4 T cell response, because even a few unaffected APC will drive a fairly significant immune response, making the suppressed ones irrelevant.

So even though viruses might “want” to block MHC class II, there are practical problems that make it hard to do. Nevertheless, there are a couple of viruses who have genes that can block MHC class II. Human cytomegalovirus is the clearest example, I think,³ and several groups have shown that vaccinia virus blocks MHC class II presentation in infected cells.⁴ Now a paper in Virology⁵ argues that the vaccinia gene catchily called “A35″ is responsible for this block. Since close relatives of A35 are present in many other poxviruses, MHC class II blockade may be widespread in this family.

Colocalization between A35 and RhoB in endosomes5

The data are reasonably convincing, though there are some complications. ⁶ But I’m still puzzled by how this is supposed to work. Vaccinia virus, and poxviruses in general, aren’t renowned for infecting dendritic cells and macrophages, which are the cell types they’d have to efficiently target if MHC class II blockade was to help them.

Removing A35 from vaccinia makes it much less virulent in mice:

A mutant A35 deletion virus (A35?) replicated normally in several tissue culture cell lines, but was highly attenuated (100–1000 fold) in the intranasal and intraperitoneal mouse challenge models⁷

And apparently this is associated with a reduced immune response to the virus:

Thus far our animal model data are consistent with this hypothesis, showing a reduction in both VV specific antibody and splenic T lymphocyte responses. ⁸

Which is consistent with a blockade of MHC class II, true, but if you have reduced viral replication for any reason you’d also expect reduced immune responses, because there would be less viral antigen to drive the response. That is, even though A35 blocks MHC class II, and A35 increases virulence, I’m not convinced that A35 increases virulence because it blocks MHC class II. Viral proteins are notoriously multifunctional, and I wonder if the MHC class II blockade is just one function of A35; or perhaps even if it’s just a side-effect of the “real” virulence function.

I’m open to the notion that A35 (and other viral proteins) are true MHC class II blockers, and that this is functionally important, but I’d like to see more data before I put it in the bank.

1. Also, intracellular bacteria, intracellular parasites, and tumor cells[↩]
2. There are exceptions to this rule, including an important phenomenon called “cross-priming” or “cross-presentation”, but that’s not relevant to this discussion now.[↩]
3. For example, Johnson DC, Hegde NR. Inhibition of the MHC class II antigen presentation pathway by human cytomegalovirus. Curr Top Microbiol Immunol. 2002;269:101-15.[↩]
4. For example, Li, P., Wang, N., Zhou, D., Yee, C.S., Chang, C.H., Brutkiewicz, R.R., Blum, J.S., 2005. Disruption of MHC class II-restricted antigen presentation by Vaccinia virus. J. Immunol. 175 (10), 6481–6488.[↩]
5. Rehm, K., Connor, R., Jones, G., Yimbu, K., & Roper, R. (2009). Vaccinia virus A35R inhibits MHC class II antigen presentation Virology DOI: 10.1016/j.virol.2009.11.008[↩][↩]
6. For example, it looks as if there may be other genes, besides A35, that also contribute to MHC class II blockade.[↩]
7. Roper, R.L., 2006. Characterization of the Vaccinia virus A35R protein and its role in virulence. J. Virol. 80 (1), 306–313.[↩]
8. Rehm, K.E., Jones, G.J.B., Tripp, A.A., Metcalf, M.W., and Roper, R.L., in press. The Poxvirus A35 Protein is an Immunoregulator. J. Virol.[↩]

TRegs infiltrate into a tumor

One of the reasons the immune system doesn’t destroy tumors is the presence of regulatory T cells (TRegs) that actively shut down the anti-tumor response.  For once, there’s a little bit of encouraging news on that front.

TRegs are normal parts of the immune system.  They actively prevent other T cells (and so on) from attacking their target. ¹  What’s more, TRegs are antigen-specific.  That is, they recognize a specific target, just as do other T cells, but instead of responding by, say, destroying the cells (like  cytotoxic T lymphocyte) or by releasing interferon (like a T helper cell) a TReg’s response to antigen is to prevent other T cells from doing anything in response to that antigen.  In other words, TRegs cause an antigen-specific inhibition of the conventional immune response. ²

Back to tumors.  We know that immune responses don’t routinely eliminate tumors by the time they’re detectable.  There is some evidence that lots of very small, proto-tumors, are in fact destroyed by the immune system very early on, before they’re clinically detectable, but those tumors that survive that attack seem to be pretty resistant to immune control.  At least part of that resistance is because TRegs get co-opted into the tumor’s control (see here, and references therein, for more on that).

So if TRegs are antigen-specific, and TRegs control immune responses to the tumor, what are the tumor antigens that are driving the TRegs?

I would have assumed that TRegs are looking at many, many tumor antigens, including both normal self antigens³ as well as classical tumor antigens.⁴  But a recent paper⁵ suggests, to my surprise, that this assumption is wrong.  Instead, “Tregs in tumor patients were highly specific for a distinct set of only a few tumor antigens“. ⁵ What’s more, eliminating TRegs cranked up the functional immune response, but only to those antigens TRegs recognized — as you’d expect, if the suppression is indeed antigen specific.

This is interesting for several reasons.  If TRegs can be specific for tumor antigens, then at least in theory ((In practice, we don’t quite have the tools yet, I think) it should be possible to turn off these TRegs while leaving the bulk of TRegs intact (and therefore not precipitating violent autoimmunity).  It also suggests that if the TRegs are only suppressing a subset of effector T cells, there’s something else preventing most effector T cells from, well, effecting.  Maybe those are antigen non-specific TRegs, or maybe there’s something else we need to know about.

I’d like to see this sort of study replicated, and a little more fine-tuning on identifying the TReg’s targets (the readout was intentionally fairly coarse here, in order to identify as many as possible).  Still, it’s an unexpected, and potentially very useful, observation.

1. It’s still not quite clear how they do this[↩]
2. There are also antigen-nonspecific TRegs, but we will ignore them for now.  They’re not as effective as the antigen-specific sort, anyway.[↩]
3. Because TRegs, unlike most immune cells, can be stimulated by normal self antigens[↩]
4. That is, antigens that are mutated, or dysregulated, and that therefore act as standard targets for immune cells[↩]
5. Bonertz, A., Weitz, J., Pietsch, D., Rahbari, N., Schlude, C., Ge, Y., Juenger, S., Vlodavsky, I., Khazaie, K., Jaeger, D., Reissfelder, C., Antolovic, D., Aigner, M., Koch, M., & Beckhove, P. (2009). Antigen-specific Tregs control T cell responses against a limited repertoire of tumor antigens in patients with colorectal carcinoma Journal of Clinical Investigation DOI: 10.1172/JCI39608[↩][↩]

TRegs in normal skin

Tumors are supposed to be destroyed by our immune system. So how come we still see tumors?

A big part of the answer is probably that our immune system is very good at destroying proto-tumors, but is not so good at handling those that manage to sneak through and grow to the point of detectability. That splits the first question into two questions: Why do some proto-tumors manage to sneak through, not being eliminated by the immune system? And why is it that detectable tumors are not effectively handled?

The first part, I think, may often be related to cell-intrinsic immune escape mutations. That is, pre-cancerous cells are constantly being attacked by the immune system; in turn (if they survive long enough) they constantly mutate, doing things like damaging the antigen-presentation pathway that makes them recognizable by the immune system. Eventually, they find some configuration that reduces the rate at which they’re killed. Once cancer cell replication is even fractionally greater than destruction,¹ a tumor can begin to grow.

So that’s probably the earliest stage of tumor growth. But once tumors reach a certain size, a second factor kicks in. Chronic immune responses are dangerous; after all, the whole point of the immune system is to kill things. The chronic immune response against the growing tumor is now shut down. This has been understood for quite a while — the immune system often becomes “tolerant” of a tumor. More recently, it’s become clear that it’s not merely “tolerance” (which implies that the immune system is simply benignly ignoring the tumor); the presence of a tumor actively forces the immune system to shut itself down, slamming on the brakes rather than just peacefully coasting by.

Brakes are a fundamental part of an active immune response. If you look at diagrams of normal immune responses, they show inverted “U” shaped curves (in here and here, for example), where the response is triggered, rapidly ramps up, hopefully does its thing, and then just as rapidly shuts down to near-background levels once again. There used to be a sort of general feeling that this was a rather passive thing — pathogen stimulates response, response destroys pathogen, no more stimulus, response goes away — but now we understand that the shut-down phase is just as active and dynamic as the upward curve. Just as with the upward phase, there are all kinds of different mechanisms to control the response; one of the most important is the “Regulatory T cell” (TReg).  And it’s pretty clear that TRegs are involved in controlling the immune response to tumors (I talked about that here, and links therein).

TRegs have been known for a while (I gave a brief history, including the I-J fiasco, here). The usual description of a TReg includes a number of markers;² one of the most basic is CD4. CD4 T cells used to be lumped together as “T Helper” cells, but now we have multiple sub-specialties in the CD4 category, and TRegs are one of those specialities.

More recently, TRegs — or at least cells that function the same way as TRegs — have been described in the CD8 population of T cells.³ CD8 T cells are traditionally called “Cytotoxic T lymphocytes” (CTL) (although it’s been increasingly clear that cytotoxicity is just one of many functions a CD8 T cell can offer), but it seems that these variants of CD8s can actively shut down an ongoing immune response, in a specific and targeted way. There seems to be a trend to calling these cells “suppressor cells” rather than “TRegs”. “Suppressor T cells” is an older term that was out of favor for a while, but it’s probably useful to bring it back and distinguish between the natural TRegs and some of the other cells that can do something similar but that have different sources and origins.

At least some of the CD8 suppressor T cells can arise from apparently-conventional CD8 T cells. That is, you can pull CD8 T cells out of a normal mouse’s spleen, and depending on what those cells see and are exposed to, they could progress to being conventional CTL — killing tumor cells, producing interferon and other cytokines, generally being a destructive force — or they could become suppressor CD8 T cells, and actively prevent that destruction from happening.

It turns out that one of the forces that can drive a CD8 T cell into being a suppressor T cell is a tumor. A recent paper from Arthur Hurwitz’s lab⁴ shows this quite clearly. They had shown previously that transferring specific CD8 T cells into a tumor-bearing mouse resulted in what they called “tolerance”.⁵ But now they demonstrate that it’s more than that; the transferred CD8s are converted into suppressor T cells that actively shut down immune responses.

Tumor-infiltrating TcR-I cells suppressed the in vitro proliferation of both melanoma Ag-specific CD8+ (37B7) T cells and OVA-specific CD4+ (OT-II) T cells. … Even at a ratio of one TcR-I cell to four responder T cells, we observed 30% suppression of proliferation. ⁴

This isn’t the only way that tumors escape immune recognition, but (at least for some tumors) it may be an important one. It’s clearly an important consideration for things like tumor vaccines and immune therapy, because it suggests that immunizing with tumor antigens (and thereby generating lots of tumor-specific CD8 T cells) may actually increase the suppressive effect of the tumor.

The conversion of CD8+ effector T cells into suppressor cells may be one mechanism by which tumors restrict the immune response from effectively controlling tumor growth. As subsequent effectors infiltrate the tumor, either following peripheral sensitization or as a result of adoptive transfer therapy, the induced regulatory cells may suppress these new effectors and reduce their ability to confer tumor immunity. This cyclic suppressive process may contribute to the profound loss of antitumor responses following adoptive immunotherapy. ⁴

(My emphasis.) On the other hand, if this is a common mechanism, then overriding it — which should be possible, using cytokines, specific T cell subsets, and/or targeted receptor ligands — may switch the suppressive population abruptly back into an effector group, turning the brainwashed traitors into resistance fighters.

1. Destruction would include far more than immune destruction, of course — it would include cells that become differentiated and no longer replicated, cells that outgrow their oxygen supply, cells that undergo apoptosis, and so on[↩]
2. FoxP3, CD25, and so on[↩]
3. I’m not sure who made the first identification; this looks as if it’s one of those fields where there were incremental advances, hinting more and more strongly at the presence of these cells, but with no single clearcut starting point. Papers in the early 2000s start to point at regulatory CD8s, and by 2004 a handful of relatively high-profile papers fairly solidly identified them. A 2004 review paper is
   Zimring, J., & Kapp, J. (2004). Identification and Characterization of CD8+ Suppressor T Cells Immunologic Research, 29 (1-3), 303-312 DOI: 10.1385/IR:29:1-3:303[↩]
4. Shafer-Weaver, K., Anderson, M., Stagliano, K., Malyguine, A., Greenberg, N., & Hurwitz, A. (2009). Cutting Edge: Tumor-Specific CD8+ T Cells Infiltrating Prostatic Tumors Are Induced to Become Suppressor Cells The Journal of Immunology, 183 (8), 4848-4852 DOI: 10.4049/jimmunol.0900848[↩][↩][↩]
5. Anderson MJ, Shafer-Weaver K, Greenberg NM, & Hurwitz AA (2007). Tolerization of tumor-specific T cells despite efficient initial priming in a primary murine model of prostate cancer. Journal of immunology (Baltimore, Md. : 1950), 178 (3), 1268-76 PMID: 17237372[↩]

A new technique called optical frequency domain imaging (OFDI) provides amazing images of tumors (especially their blood vessels) in situ.

“(a) OFDI images of representative control and treated tumors 5 d after initiation of antiangiogenic VEGFR-2. The lymphatic vascular networks are also presented (blue) for both tumors. (b) Quantification of tumor volume and vascular geometry and morphology in response to VEGFR-2 blockade.”

– Vakoc, B., Lanning, R., Tyrrell, J., Padera, T., Bartlett, L., Stylianopoulos, T., Munn, L., Tearney, G., Fukumura, D., Jain, R., & Bouma, B. (2009). Three-dimensional microscopy of the tumor microenvironment in vivo using optical frequency domain imaging Nature Medicine DOI: 10.1038/nm.1971

Pregnant woman (Ivory Coast, West Africa)

Recently I’ve mentioned a few cases of transmissible tumors — that is, cases where tumors actually spread from their original host, to other individuals. The two most dramatic transmissible tumors are Canine Transmissible Venereal Tumor (CTVT) and Tasmanian Devil Facial Tumor (TDFT), where the original tumor can spread widely throughout the entire species. (See this post, this one, and this one, for more detail.) There’s also at least one case of a tumor that accomplished a single transmission, from the original patient to the surgeon who operated on him. ¹

Tumors aren’t supposed to be able to spread in this way, because they’re essentially foreign transplants — they should be rapidly rejected, as if they were, say, a skin graft between two random people. In this post I talked about how CVTV and TDFT might have arisen. (There are also a number of cases of tumors that spread to immuno-suppressed individuals, such as after organ transplants, but those cases are easier to understand from an immunologic viewpoint.)

When checking up on references for the last post, I ran across a set of transmissible tumors I hadn’t known about: Vertically transmitted tumors, in which tumors spread from a pregnant mother to the fetus in utero.²

This also, I’m glad to say, seems to be very rare, as you’d expect. Even though mother and child are partially tissue-matched, and even though pregnancy is a very special situation, immunologically, the parent and her child are not genetically identical, and should reject grafts from each other pretty efficiently. (Transplants from parent to child still require immune suppressive treatment.) The review I ran across lists a total of 14 cases of vertical spread of tumors, from 1866³ to 2002.⁴ Although they do note that:

Given the lag time between birth and diagnosis in several of the infants, cases of maternal–fetal transmission may not be as rare as the literature would suggest, and the number of cases could be higher as the detection of metastatic tumor in the fetus may go undetected in cases of abortion or maternal–fetal demise. ²

Malignancy during pregnancy isn’t all that uncommon (0.1% of pregnancies, it says here), so the handful of cases with actual spread of the tumor to the fetus are “numerically inconsequential”. What was different about these 14 cases? We don’t really know, in general. Almost all of the described cases are earlier than 1965,⁵ predating the molecular era of medicine. Perhaps some, or many, of the infants were immune compromised, as the authors note:

Fetuses with a congenital immunodeficiency are likely to be at an even higher risk for the engraftment of such tumor cells.⁶ Other factors that may affect the likelihood of tumor cells entering the fetal circulation include maternal homozygosity for one of the fetal HLA haplotypes,⁷ metastatic potential of the maternal tumor, and a high maternal blood and/or placental tumor load. ²

The outcome of this transmission was very poor; only 3 of the 14 children survived the disease.

I don’t really have any lesson to draw from these cases. Without an extensive molecular workup that isn’t available for almost all of these cases, I don’t know that we can learn much about tumor transmission. Still, these stories are worth keeping in mind when thinking about mechanisms of tumor transmission.

1. Gartner HV, Seidl Ch, Luckenbach C, et al. Genetic analysis of a sarcoma accidentally transplanted from patient to a surgeon. N Engl J Med 1996;335:1494–1496.[↩]
2. Tolar J, & Neglia JP (2003). Transplacental and other routes of cancer transmission between individuals. Journal of pediatric hematology/oncology : official journal of the American Society of Pediatric Hematology/Oncology, 25 (6), 430-4 PMID: 12794519[↩][↩][↩]
3. Friedreich N. Beitrage zur pathologie des Krebses. Virchows Arch 1866; 36:465–477.[↩]
4. Tolar J, Coad JE, Neglia JP. Transplacental transfer of small cell carcinoma of the lung. N Engl J Med 2002; 346:1501–1502.[↩]
5. Not saying the molecular medicine abruptly switched on in 1965, it’s just a convenient cutoff[↩]
6. Pollack MS, Kirkpatrick D, Kapoor N, et al. Identification by HLA typing of intrauterine-derived maternal T cells in four patients with severe combined immunodeficiency. N Engl J Med 1982; 307:662–666.[↩]
7. Osada S, Horibe K, Oiwa K, et al. A case of infantile acute monocytic leukemia caused by vertical transmission of the mother’s leukemic cells. Cancer 1990; 65:1146–1149.[↩]

Canine Venereal Tumor phylogeny

Bayman commented, after reading this post:

So isn’t the real question why can’t all tumors be transmissible? If you believe the tumor immunologists, all tumors should be capable of avoiding T cell attack…no??

I don’t have answers, but I can speculate a little. ¹

Very quick background: In general, tumors are unique. They arise independently each time, and when their host dies, the tumor dies too. That’s in contrast to pathogens, whic may or may not kill their hosts, but which survive and are transmitted to a new host; pathogen infections are not unique, they have a long evolutionary history reaching back through many individual hosts. Tumors can’t do this, for the same reason that skin grafts are rejected by unrelated animals — tumors are essentially unrelated grafts, and should be very rapidly rejected by the new host.

But in very rare circumstances — there are two known instances, and a couple of other possible ones — tumors have arisen that can be transmitted from one host to another.  The two cases are canine transmissible venereal tumor, and Tasmanian Devil facial tumor.  There have been suggestions that these tumors are unique in some immunological way, but I am not convinced by those arguments: See this post and this one for more background.  That’s not to say that these tumors have no ways of evading the immune system; what I am saying, is that virtually all tumors have some way of evading the immune system, and the functions that have been convincingly described for the transmissible tumors don’t seem all that exceptional for tumors in general.

So, if these tumors can be transmitted, and they aren’t all that extraordinary immunologically, what does make them extraordinary?  As I say, I don’t know, but based on tumor immunology as I understand it, I can make some guesses.

The most important factor, I suspect, has nothing to do with immunology.  These tumors are unusual in that they have a built-in way of contacting new hosts. TDFT is spread through bites, CTVT is spread sexually.  There’s no similar way that, say, a liver tumor, or a brain tumor, could be spread.  So that immediately rules out the vast majority of tumors; even if they could survive after transmission, there’s no chance of a transmission chain. ²  But still, most tumors would be rejected even if they did manage to be transmitted.

The Three E’s of tumor immunity

What seems to happen with most tumors³ is that proto-tumors appear quite early, but are controlled by the immune system – perhaps for years — and never become detectable.  Many such proto-tumors are completely eliminated by the immune system, and we have no way of telling that they even existed.  Many more are controlled at the half-dozen cell stage, much too small to detect; they aren’t eliminated by the immune system, but they can’t escape and grow either.  A very small percentage of these equilibrium tumors, though, eventually find a way of at least partially escaping from immune control, and begin to grow. (Perhaps the immune system kills 99% of the new cells, but a 1.01% growth rate compounds itself fast enough to be eventually detectable.)  This is the “Three E’s” theory of tumor growth (discussed more here and here) — “Elimination, Equilibrium, Escape”.

Regulatory T cells and cancer

The Three E’s apply to the very small proto-tumors. But there is probably another factor that kicks in once the tumor becomes larger.  Tumors are themselves immunosuppressive — they shut down immunity throughout the entire body, to some extent, but they shut down immunity to themselves very powerfully.  The immune system has powerful safeguards that prevent it from attacking its own body; broadly speaking, tumors are their own body, and in many cases tumors probably also have been selected to massively amplify the normal protective signals. ⁴ (See this post, and this one, for more on that.)

So here⁵ is my speculation.  We suspect that the ability to be transmitted is present in several tumors, but they never get the opportunity to transmit.  Of those rarities that do get transmitted, most are rapidly rejected, as foreign grafts.  But a tiny minority of this minority may be able to survive because they have powerful immune suppression abilities on top of their common immune evasion abilities.

Why did the Tasmanian Devil cross the road?

Were CTVT and TDFT just lucky — just happened to have the right immune suppressive abilities?  I don’t think so.  I think they were in the right place at the right time.  They were tumors that had a mechanism for transmission, and that had some ability to immune suppress, but they would normally have been rejected as foreign grafts.  Except that both of these tumors, I think, arose at a time and place where their population was highly inbred.  CTVT arose, we speculate, as dogs were becoming domesticated; probably a small, inbred, closely-related population.  Tasmanian Devils in general may not be closely related, but I suspect there are sub-populations⁶ that were closely related and that would not have rapidly rejected skin grafts.  The early version of the respective tumors would not have been rapidly rejected by these closely-related new hosts, giving them a chance to establish their own immune suppressive regime.

Now we have the chance for natural selection of the tumors.  Variants with more powerful immune suppression could spread to a wider range of hosts; variants with standard immune suppression died out with their victims.  In dogs, this natural selection could occur over time; as dogs became gradually more variable, there would be continuous new selection for new tumors that could keep up with the dogs. ⁷ With the Devils, the selection would be over space: The tumors would be selected for their ability to spread within new sub-populations of the Devils, perhaps through gradually more distantly-related subgroups. Eventually, we see the tumors as being capable of transmission and growth throughout the entire population, but the original tumor might not have had this ability.

Rapid MHC diversity in Channel Island Foxes

This model suggests that humans are probably not at great risk of having a transmissible tumor spread in us; and the same is true for most species.  You need the combination of an inbred sub-population with a mechanism of tumor spread and the right kind of tumor. And inbred populations are usually a transient thing; MHC becomes diverse very rapidly, and then the window for tumor establishment is closed.

But this is just a guess, so don’t be too comforted.

1. And by the way, I disagree with Bayman’s suggestion here (“Tumor Immunology Is A Waste of Time”) that he “find[s] it impossible to believe that effective therapy will ever achieved by artificially stimulating the immune system to attack weak and largely self antigens.” But this post is already too long, so I’ll save my answer for another time.[↩]
2. There’s at least one case of a surgeon who apparently contracted a patient’s tumor after cutting himself during surgery — given the option, perhaps many more tumors could be transmissible, but don’t get the chance.[↩]
3. Not necessarily those induced artificially, with high doses of carcinogens or with powerful oncogenes, but with those that arise naturally, in older individuals[↩]
4. I suspect that tumors have many ways of achieving this localized immune suppression.  I also suspect that different tumors have different dependence on this localized immune suppression.  Those tumors that were highly successful as proto-tumors might already be very good at avoiding immunity — for example, they may secrete tons of TGF?, or otherwise have very powerful TReg-inducing abilities — and only need to shut down a little.  Those that barely squeaked by as proto-tumors, may have very potent immune suppression.  I don’t think the mechanisms for this tumor-based immune suppression are very well understood, though over the next couple years they probably will be. [↩]
5. Finally![↩]
6. Subpopulations that are now, probably, extinct, because of the tumors[↩]
7. I’m told that CTVT is eliminated faster or slower in different dogs.  It would be very interesting to correlate this with MHC types, to see if there’s still some effect of rejection even after 50,000 years of selection on these tumors.[↩]