Feature: Immunity at Defcon B

This feature appeared in the November/December 2009 issue of Australian Life Scientist. To subscribe to the magazine, go here.

Robert Brink is an immunologist, and like most of his kind, he barracks for one of two teams: B cells or T cells. In Brink’s case it is B cells (and make sure you don’t accuse an immunologist of barracking for the wrong one), which are white blood cells charged by the immune system with producing antibodies against foreign antigens, such as those expressed by viruses and bacteria (and the immune system is not fussy whether the invaders are coming in by water or air). The ultimate job of these cell types is to produce antibodies that can eliminate the bad guys and either prevent or cure an infection.

Brink’s research group at the Garvan Institute of Medical Research in Sydney is particularly interested in the complex process by which B cells function to fight off an infection and in nutting out the decisions made by the immune system along the way, and he’s speaking at the Australasian Society of Immunology meeting on the Gold Coast in December.

A lot happens between the immune system detecting a foreigner on their shores and getting rid of it, and at ASI, Brink will map out the steps of this process – as observed in their in vivo model – and also talk about some exciting new findings regarding those B cell lifestyle decisions.

Flying under the radar

According to Brink, an age-old problem is actually ‘seeing’ this invasion and B cell defence process as it happens from beginning to end – mainly because of the way the immune system is set up.

“Essentially there are zillions of these immune cells floating around that are basically all the same except for the antigen receptors on the cell surface. The gene events that control the antigen specificity demonstrated by B cells in the first place are set from birth, and the result is an incredible diversity built into the antibody receptors so that our immune system can handle most of the antigens thrown at it over a lifetime.”

The flip side is that with such a spectrum of antigen-recognition specificities, the number of cells that can produce just the right antibody to recognise an invading molecule is small, and insufficient to mount an immediate and fully effective antibody-based defence against the threat. These cells need to be expanded, and when activated by the foreign antigen they go off and proliferate to produce an army of copies or ‘clones’ with the desired specificity.

---PB---

This basic process has been characterised for a long time with many aspects studied in genetic and molecular detail by like-minded B cell immunologists. The issue for Brink’s work was that all of the interesting stuff at the beginning happens at such a low frequency it is very hard to prove in vivo.

“Maybe after the B cells have expanded for a week or so we might be able to detect them in vivo by flow cytometry, but not in those really early stages when the decisions we were really interested in are made. It’s a black box up to that point.”

To address their issues, Brink’s team developed a system in mice whereby the antibody genes were engineered to make antibodies of a given specificity to a model antigen – in this case to hen egg lysosome – which is a robust and immunogenic protein to study without causing any harm to the mice.

“Now we are able to identify these low-frequency cells and so follow the very early processes as the cells proliferate, migrate and then differentiate into the different types of cells required to mount a full antibody response and fight off the infection, and try to work out how.”

Stages of response

When an infectious organism invades, the immune system works hard to make some antibodies fairly quickly, usually within three to four days. This is called the proliferation phase and is very important for first-line defence. However, although effective to a point, this initial response is still quite generalised. The secreted antibodies do recognise the pathogen and tag it for destruction, but they are of fairly low affinity and therefore not potent enough to hold off a strong invasion for long.

So, an extra system has to be induced to provide better-skilled reinforcements. While about half of the B cells are deployed (as plasma cells) to produce their low-affinity antibodies and eventually die, a few are held back. These migrate to a completely different location within the spleen and lymph nodes called the germinal centre.

Here, over the next 10 to 14 days, antibody production capabilities are improved and the highly specialised B cells that are needed to eliminate the infection are released. Brink likens them to the SAS being sent in.

‘Training’ of B cells in the germinal centre involves a process called somatic mutation, whereby random rearrangements are introduced into the B cells’ antibody genes.

---PB---

“Most of these mutations will not succeed or may even make it worse, but occasionally you will get a change in the antigen-binding region that increases the antibody affinity so that they bind better to the surface of the invading organisms,” says Brink. These mutated cells get selected by an as yet unknown process, and then are sent forth to multiply and conquer. Although much slower to appear after the initial insult, this second wave of antibodies will be much more potent: high-affinity and lethal weapons that will go on to provide long-term immunity against re-infection.

Decisions… decisions…

One of the main questions for Brink relates to the very earliest stages of the antibody response, around the end of the proliferation phase on day three or four. “How does the immune system decide to send some B cells off to keep making antibodies and others off to the germinal centre for reprogramming, particularly as these are equally important jobs,” he asks.

Using their mouse-based system, Brink’s team at the Garvan is starting to get some leads on this, and in the second part of his talk at ASI, Brink will describe some of their recent and very exciting findings, published earlier this year in Immunity.

“One thing we know already is that this differentiation decision happens very rapidly. Basically, we can see the B cells expanding in the mice and looking very similar for a few days, and then, between days three four, there is a sudden shift in direction for some of the cells – literally,” Brink says.

“At the same time, their gene expressions change very rapidly, and with that, their localisations. These cells move from being clonal proliferating B cells that all look the same into either plasma cells that keep making antibodies or into cells of the germinal centre. So they end up looking very different and going to very different places within the immune system.”

---PB---

Location… location…

What Brink and colleagues, such as Dr Dominique Gatto, found is one of the major determining factors of the sudden B cell choice: it is a cell surface receptor called EBI2. Named after the Ebstein-Barr virus, which infects B cells, this G-protein-coupled transmembrane receptor looked interesting from the start, even though no-one really knew what it does.

Firstly, it’s well-characterised sibling, EBI1, helps to regulate immune cell movement around the body. Secondly, Brink knew that EBI2 is regulated in B cells over a similar time scale to B cell decision time. “The expression on B cells floating around the body goes up when they see an invader, and later in the response it goes down, coinciding with the point at which the proliferating cells turn into either germinal-centre or plasma cells,” he says.

To test their hypothesis that EBI2 might be a player, Brink’s team made knockout mice for the EBI2 gene and ran a series of experiments. A crucial finding straight up was that without this gene being expressed, virtually all the activated B cells migrated into the germinal centre, rather than the normal 50 per cent. And in the opposite experiment, where EBI2 expression was turned back on and kept on in these mice, the B cells were nearly all pulled out of the germinal centre to where the plasma cells go.

“We were therefore able to show that movement of the responding B cells at this critical point when they are deciding which way to go is mediated by EBI2, and a really important feedback component of the system and the eventual B cell phenotype is the location.

“Our results suggested that where the cells move to, as mediated by EBI2, might actually feed back on their differentiation process, and this is one of the things we are looking at now,” says Brink.

EBI2 definitely induces the differentiating B cells to move locations, suggesting that moving into the germinal centre directly mediates the differentiation program. Alternatively, a signal at that location could reinforce a particular gene expression program, whereas moving out of the germinal centre does the opposite.

“Of course, we do not know any of this for sure yet. So, at the moment, we are trying new things to make these cells move to different places and see how that location or movement feeds back on the phenotype that the cells take on. It is still certainly an open question, but very interesting. It always comes down to what is regulating what and to what extent, and now where?”