The key to finding TCR sequences in RNA-seq data

I had previously written a short blog post touching on how I'd tried to mine some PBMC RNA-seq data (from the ENCODE project) for rearranged T-cell receptor genes, to try and open up this huge resource for TCR repertoire analysis. However, I hadn't gotten very far, on account of finding very few TCR sequences per file.

That sets the background for an extremely pleasant surprise this morning, when I found that Scott Brown, Lisa Raeburn and Robert Holt from Vancouver (the latter of whom being notable for producing one of the very earliest high-throughput sequencing TCR repertoire papers) had published a very nice paper doing just that!

This is a lovely example of different groups seeing the same problem and coming up with different takes. I saw an extremely low rate of return when TCR-mining in RNA-seq data from heterogeneous cell types, and gave up on it as a search for needles in a haystack. The Holt group saw the same problem, and simply searched more haystacks!

This paper tidily exemplifies the re-purposing of biological datasets to allow us to ask new biological questions (something that I consider a practical and moral necessity, given the complexity of such data and the time and costs involved in their generation).

Moreover, they do some really nice tricks, like estimating TCR transcript proportions in other data sets based on constant region usage, investigate TCR diversity relative to CD3 expression, testing on simulated RNA-seq data sets as a control, looked for public or known-specificity receptors and inferred possible alpha-beta pairs by checking all each sample's possible combinations for their presence in at least one other sample (somewhat akin to Harlan Robins' pairSEQ approach).

All in all, a very nice paper indeed, and I hope we see more of this kind of data re-purposing in the field at large. Such approaches could certainly be adapted for immunoglobulin genes. I also wonder if, given whole-genome sequencing data from mixed blood cell populations, we might even be able to do a similar analysis on rearranged variable antigen receptors from gDNA.

Translating TCR sequences addendum: not as easy as FGXG

I recently wrote a blog post about the strategies used to translate T-cell receptor nucleotides en masse and extract (what can arguable be considered) the useful bit: the CDR3.

In that talk I touched on the IMGT-definition of the CDR3: it runs from the second conserved cysteine in the V region to the conserved FGXG motif in the J. Nice and easy, but we have to remember that it's the conserved bit that's key here: there are other cysteines to factor in, and there are a few germline J genes that don't use the typical FGXG motif.

However even that paints too simple a picture, so here's a quick follow up point:

These are human-imposed definitions, based more on convenience for human-understanding than biological necessity. The fact is that we might well produce a number of TCRs that don't make use of these motifs at all, but that are still able to function perfectly well; assuming the C/FGXG motifs have function, it's possible alternative motifs might compensate for these.

I have examples in my own sequence data that appear to clearly show these motifs having been deleted into, and then replaced with different nucleotides encoding the same. Alternative residues must certainly be introduced on occasion, and I'd be surprised if none of these make it through selection; we just don't see these because we aren't able to generate rules to computationally look for these.

I actually even recently found such an example with verified biological activity: this paper sequenced tetramer-sorted HIV-reactive T-cells, revealing one that contained an alpha chain using the CDR3 'CAVNIGFGNVLHCGSG'.*

For the majority of analyses, looking for rare exceptions to rules probably won't make much difference. However as we increase the resolution and throughput of our experiments, we're going to find more and more examples of things which don't fit the tidy rules we made up when we weren't looking so deeply. If we're going to get the most out of our 'big data', we need to be ready for them

* I was looking through the literature harvesting CDR3s, which reminds me of another point I want to make. Can I just ask, from the bottom of my heart, for people to put their CDR3s in sensible formats so that others can make use of them? Ideally, give me the nucleotide sequence. Bare minimum,  give me the CDR3 sequence as well as which V and J were used (and while I stick to IMGT standards, I won't judge you if you don't - but do say which standards you are using!). Most of all, and I can't stress this enough, please please PLEASE make all DNA/amino acid sequences copyable.**

** Although spending valuable time copying out or removing PDF-copying errors from hundreds of sequences drives me ever so slightly nearer to a breakdown, it does allow me to play that excellent game of "what's the longest actual word I can find in biological sequences". For CDR3s, I'm up to a sixer with 'CASSIS'.

T-Cell Receptors, Part Five: A Receptor For All Cells?

Back to Part 4: More Odd TCR Transcripts

This last installment in this week of blogging covers the more recent, and potentially more controversial stories I've found of wandering TCRs.

The claims made in the following papers are big, and could have far reaching connotations if they bear out. All of these papers report whole, functional TCRs, being expressed and used by white blood cells. Just the wrong ones.

It all starts in 2006, with a sentence. A tantalising, if near-painfully vague sentence:

"A series of control experiments prompted us to test the hypothesis that human neutrophils express components of the TCR machinery."

A series of control experiments ey, that old chestnut.Wait, what, neutrophils?

These cells are one of the major components of the innate immune system, and account for the majority of circulating white blood cells at any one given time, despite their rapid turnover.

Their job is to turn up early to places where the body is in trouble (such as inflamed or infected areas) and do some damage control, which largely consists of eating any pathogens they can find, and making it harder for any they can't find to survive or spread.

Previously they'd been thought to act completely via innate immune receptors. The presence of one of the hallmark receptors of adaptive immunity on them is a little surprising to say the least.

Using antibodies directed against the alpha and beta constant regions, they report that around five to eight percent of freshly isolated neutrophils express αβ-TCR, seemingly in a manner comparable to typical TCR expression.

They seem to express a varied repertoire, when looking for different Vα and Vβ transcription by RT-PCR, that are revealed to be rearranged as per normal. There's CD3 components, CD28, all seemingly upregulated by TCR agonists, as well as a number of proteins required for TCR signalling.

So what are these TCRs supposed to be doing in these neutrophils? Upon TCR stimulation (with anti-CD3 and anti-CD28 antibodies, which is thought to represent a fairly physiological level of stimulation), and watched what the neutrophils did next.

What they did next was live long and prosper; it seemed stimulating the TCRs inhibited neutrophil apoptosis and increased secretion of IL-8, the chemokine responsible for recruiting more neutrophils to the danger zone.

So, the theory goes that some neutrophils express TCR, which presumably helps them recognise either specific pathogens, or a broader swathe of pathogens, which can then find bugs quicker, helping to recruit more neutrophils to the threat.

The same group (Wolfgang's Kaminski's group from Heidelberg) has had a couple of follow up papers on this, during which time it gets re-dubbed the TCRL_n, as it's TCR-'like', and in neutrophils, which somewhat solves the quandry of the 'TC' in TCR.

One of these papers is really just a long observation that the repertoire of different TCRs expressed in neutrophils starts off broad, and shrinks with increasing age.

The next is just a case study of a patient with a nasty autoimmune condition (where their red blood cells get targetted and destroyed by their own antibodies), whose number of TCR positive neutrophils had jumped from 5% to 80%.

There's also one final neutrophil paper from a dental group, who noted that oral neutrophils also seem to have a higher expression of TCR than do their circulating cousins, which ties in with their previous work showing that oral neutrophils have a different phenotype.

It takes three years after the start of the neutrophil story before the next cell joins the party. Eosinophils, a kind of granulocyte (mostly responsible for killing parasitic worms) enter the fray, wielding not αβ, but γδ-TCRs.

It's another frustrating start; this time they were on the hunt for γδ TCRs in eosinophils because of the "surprising similarities" between them. Either this is fantastically lucky fishing, or there's a few experiments they're not telling us about.

Either way, they find the γδ-TCR on the eosinophils by flow, along with CD3. Interestingly, they don't find much γδ in lymphocytes (1.4%, which is a few percent shy of usual), nor do they find αβ expressed on neutrophils, putting them at odds with the previously discussed papers.

It seems that while possessing all the bits and bobs needed for TCR recognition, these eosinophils don't produce nearly as much TCR message as T-cells, nor as diverse a range of TCRs. However, activation with TCR agonists caused eosinophils to do exactly what they're supposed to do when activated normally, with degranulation and the release of cytotoxic proteins and reactive oxygen species (ROS).

There's even a couple of figures showing exploring possible functions for the γδ receptor. The presence of γδ-blocking antibodies inhibits the ability of eosinophils to produce ROS in response to mycobacterium, or to induce apoptosis in a colorectal cancer cell line.

The final additions to the TCR club is reported by the same Heidelberg group, giving the impression they probably went on a TCR-testing rampage.

Macrophages are key cells responsible for maintaining immunity in the tissues, by phagocytosing and killing pathogens and presenting their antigens to T-cells. They differentiate from monocytes, which circulate in the body looking for signs for infection or inflammation.

These are classical examples of innate immune cells that bridge the gap to adaptive immunity, through their antigen presentation. These recent papers suggest that they might go one further, with some subpopulations expressing either the αβ or γδ TCRs.

The story unfolds much like the others. There's a small percentage of monocytes and macrophages expressing a limited repertoire of 'TCRL_m' (5% for αβ, 3% for γδ), along with other TCR signalling components.

The obligatory search for function touches on some big topics.

For the TCRL_mαβ, they investigate the intereaction with mycobacteria, which it seems upregulate the expression of the TCR. They go further, and stain lung sections from tuberculous patients, showing that the contacting edge of cells around the granulomas are enriched with TCR expressing macrophages.

TCRL_mγδ on the other hand was investigated during (murine) bacterial meningitis, during which the presence of γδ-macrophages was enriched in the CSF. They also found γδ-MΦs in atherosclerotic plaques, indicating a pretty broad range of possible interactions.

There's a lot to muddy the water in these papers. There's a couple of not-completely-convincing figures, and the obvious matter of the contradiction between the papers. They do all however go a long way to refute the possibility they're just seeing T-cell contamination (either by immunohistochemistry or FISH).

These papers seem to ask more questions than they answer. How does this happen. Do they undergo any selection, and if not, how do they avoid auto-reactivity? If so many other cells can make use of the TCR, why do we even need T-cells in the first place?

If true, these are hugely interesting findings. Here we have terminally differentiated myeloid cells, seemingly expressing one of the classical lymphoid, adaptive immune receptors. Having presented some of these at journal clubs I've seen first hand a bit of resistance to accepting these possibilities outright.

Personally, my view is that biology is a massively confusing thing, and the closer we look at it the more wierd, unexpected stuff we're going to find. Once biology has evolved a system such as the TCR, there's no reason why other cells within the same organism shouldn't make use of it. As technology increases the throughput and sensitivity at  which we can operate, more and more of our accepted models are going to gather inconvenient aberrations like these.

There we have it; the case of the wandering TCR. Whether or not you believe these stories (or more importantly, whether or not you think they have any biological signnificance), I hope you found them as interesting as I did.

While this week of blogging ballooned into a much bigger project than intended, I've enjoyed doing it. Over the coming months I hope to do some other key aspects of TCR biology, just maybe not all at once next time.

T-Cell Receptors, Part Four: More Odd TCR Transcripts

Back to Part 3: TCRs Where They Shouldn't be 

After the discovery of TARP, things were a quiet for a little while; it looked like TCRs seemed happy to remain in the T-cells. However, there were a couple of blips on the radar, around the same time, that never really developed into fully-fledged stories.

The first of these two papers was in 2002, working off the back of an interesting observation from a previous study; when analysing the expression of TCR expression in T-cells that adhere to stromal bone marrow cells, they found that the control group of stromal cells alone seemed to express TCR as well as the T-cells (NB, in mouse) .

Closer investigation revealed that mesenchymal cells expressed αβ TCR (both primary cells and cell lines) mRNA, as well as CD3, revealed by RT-PCR. However, neither of the RAG proteins are expressed, and what little TCR there is expressed (by T-cell standards) isn't re-arranged, which you'd expect without the necessary recombinases present.

Once again we see that the transcripts are made of (different) J regions correctly spliced onto constant regions, akin to what we saw with TARP. As the whole constant region was present, they were afforded the luxury of easily looking for protein on cells, as constant region antibodies exist; lo and behold MEFs (as a model mesenchyme system) stain positive for TRBC, while MEFs from TCRβ-/- mice don't (although no T-cell staining as a positive control? Seriously?).

Curiously, mesenchyme cell lines that expressed more of the TCR (at RNA level) seemed to have different growth properties to those that didn't, both proliferating more and causing a greater incidence of cancer when injected intradermally into nude mice.

The next case we see of wandering TCRs takes us to the relatively young field of neuroimmunology.

While we all know and appreciate the role MHC proteins and CD3 play in immunology, it turns out they're required for a number of neurological roles, such as correct synaptic formation and hypothalamus development.

This, and similar observations, prompted Harvard researchers to wonder if perhaps there might be some TCR lurking in the CNS, to act as the ligand for these seemingly important neuronal MHC proteins.

A series of in situ hybridisation experiments shows that there was indeed TCRβ expressed in brain segments, and that it is indeed localised to the neurons. Examination of the message revealed a curious thing about the transcripts (and I think you've guessed it); they're non-recombined, consisting of a (specific) J region spliced to the constant region.

They find no TCRα mRNA, and detect no constant region protein by immunostaining. In addition, TCRβ knock out mice fail to show a similar neurological phenotype to MHC or CD3 knock out mice, which seems to put paid to the idea that this neuronal TCRβ expression provides the ligand for MHC in the CNS.

The authors offer a few suggestions for why the CNS might be expressing this TCR; maybe the RNA serves a purpose, maybe the protein or activity is just very low and undetectable, or maybe the expression is just part of some vestigal transcriptal program, playing no role but doing no real harm. Not the most earth-breaking of papers, but another curious case of TCR message getting around and seeing the sights.

(As a brief aside, this reminds me of an interesting neuroimmunology paper I saw a little while ago, where a genome wide investigation into narcolepsy revealed the major associative polymorphisms mapped to the TCRα locus. This was only an association study, but it does speak to possible non-immune roles for the TCR.)

Both of these stories are a little reminiscent of the B-cell TCR expression we saw in part 3, expression of funny TCR transcripts (spliced but lacking Vs), in cells that we wouldn't expect them in. Interestingly, there's even a link to increased cell proliferation, which echoes the TARP story somewhat.

However, there's very little (at present) to convince me these transcripts are doing much, if anything. Personally, from my own experience sequencing TCR repertoires, I know in T-cells we get a bunch of odd transcripts popping up, the type of which have been known about for a long time. It's always hard to say whether these things biologically important or just interesting looking enough to give postgrads false hope.

Chances are good they just represent the noise in the machine, little ripples of expression as a stone is dropped nearby in the transcriptional pond, unintended byproducts of another pathway. Either way, these papers raise some interesting possibilties, and pretty convincingly show TCR sequence popping up where it's not supposed to, so they've got a welcome entry in this series.

The next (and final) entry into this week of TCR blogging brings us up to the present crop of non-T-cell TCRs, including the papers that actually set off my interest in the topic. Having presented some of the papers to my own journal club I won't be surprised by a mixed response; who wouldn't want to double check the presence of TCRs in no fewer than three other non-lymphoid white blood cell types?

On to part 5, where we finish exploring the landscape of where TCRs emerge

T-Cell Receptors, Part Three: TCRs Where They Shoudn't Be

Back to Part 2: A History of TCR Discovery

Wait, this isn't the right lymphocyte!

As the name implies, TCRs are receptors that you find on T cells. They're pretty well named in that regard. Seek ye TCRs, find ye T cells. Or so the story goes.

Biology is a busy, messy place, and our cells and the proteins inside them don't always behave as the textbooks might want them too. This series started out by me wanting to share some of the interesting stories I've stumbled across in my research, of TCRs popping up were perhaps we might not have expected them...

The first hints we have that perhaps the 'TC' in TCR isn't 100% accurate came not longer after the discovery of the delta chain, and in a very similar cell type. Several papers in a row show transcription of truncated, non-rearranged TCR gene transcription in B-cells, both alpha and beta chains, in normal and cancerous B-cell lines.

This perhaps isn't that surprising. B-cells are the other branch of the adaptive immune system that also produces variable antigen receptors (via V(D)J recombination). They're closely related lymphocytes, with related functions, that share an ancestry in the common lymphoid progenitor cell.

What's more, due to the shared developmental origin, somatic recombination and related purposes of the TCR and BCR loci, it's possible (if not likely) the various loci would share certain aspects of transcriptional control - the recombination machinery is certainly conserved among them.

Bearing in mind that the individual gene segments have their own promoters, and are capable of individual, non-rearranged transcription (as per the accessibility model of V(D)J recombination, where such transcription is required), it's not a huge leap to imagine that maybe B-cells just have some factors left over that drive a bit of TCR expression, of little physiological significance. The fact this story sputtered out not long after its inception might lend weight to that possibility.

The odd lymphocyte out?

I should probably mention that black sheep of the lymphocyte family, the TCR-bearing NKT cell. Now we're talking proper TCR here; rearranged genes, functional αβ heterodimers signalling on the cell surface, the whole shebang.

Discovered in 1987, these chimeric cells typically express a semi-invariant TCR, although there is a rarer, less studied type that expresses a more variable repertoire.

However, NKTs don't really have a place in this story; they're really just T-cells that went to NK cell school, so it's no surprise at all they have TCRs.

From here on in we're in much murkier territory.

The prostate examination

A little time passes before we find TCR expression popping up in new places. In 1999, while searching for highly expressed prostate genes, NIH researchers - headed by Ira Pastan - noticed an odd transcript popping up; none other than the TCRγ gene.

Oddly, it looked like the γ chain going solo; there was no δ, nor CD3, which you would have expected if the transcript was being carried in by infiltrating γδ T-cells. In situ hybridisation experiments revealed that the TCRγ transcript was being produced by acinar epithial cells.

A closer look at the transcript itself revealed that it wasn't even a full, recombined receptor gene being expressed. Instead, the sequence started with a particular TRGJ region*, which was correctly spliced onto the γ constant region.

Having done some in vitro translation on these odd γJC transcripts, they knew what size proteins they might make, if they were indeed translated in vivo. Well, fast-forward a few years and the same group is back with a vengeance. I mean antibodies, they're back with antibodies.

In their follow up paper, they show that the smaller of the two proteins encoded by the transcript (in an alternate reading frame to the typical γ transcripts we know and love) is not only expressed in the prostate, but in prostate cancer, and several breast cancer cell lines to boot. In welcoming it to the world of the translated, this protein gets a name; TCRγ alternate reading frame protein (TARP).

Over the coming few years, TARP enjoys a small but regular input of papers, showing that while involved in regulating growth, it itself is up-regulated by testosterone, via an androgen-receptor binding site in its promoter, and ends up expressed on the mitochondria.

This is pretty interesting stuff; we have (part of!) an immune receptor, being expressed in epithelial cells, under different regulation, with a different subcellular location, for presumably different purposes, but all off the same locus (which is itself within/part of the locus for another antigen receptor chain).

However, there is a whole other story that's come to dominate the TARP field in recent years; that of cancer therapeutics.

Being expressed seemingly only in healthy prostate tissues, or prostate and breast cancers, TARP makes for a pretty attractive drug target. 

One possibility is to use portions of the TARP promoter to design gene therapy vectors that will be specifically expressed in prostate tissues, allowing you to throw in a for something that (for example) inhibits cancer progression.

Closer to my own heart is the burgeoning field of TARP immunotherapeutics. Two groups have isolated CD8+ cytotoxic T-cells (CTLs) targetted against TARP peptides (HLA-A2 restricted), while another group has found some TARP-specific CD4+ T helper cells. The idea is this information could be used to possbily generate anti-TARP cancer vaccines, or T-cell therapies against established tumours.

What's more, one of the TCRs from TARP-specific CTLs has been engineered back into peripheral blood T-cells; T-cells with this TCR genetically engineered (or transduced) into them were shown to be active against HLA-A2+, TARP+ breast and prostate cancer cells, which could make for a promising treatment.

Of all the cases of wandering-TCR we see, TARP is perhaps the best appreciated and studied (I guess being a possible cure for cancer will do that). In the next two instalments in the series, we'll see some cases where the evidence is a little slimmer, or the findings perhaps a little bit more controversial, but hopefully still just as interesting.

*A note on TCR gene segment nomeclature (as defined by IMGT, whose system I use throughout my blog, along with their numbering). The first two characters represent the gene type (TR for TCR, IG for immunoglobulin), the next letter is the chain (alpha, beta, gamma, delta), and the final letter is the type of region (V, D, J or C).

On to Part 4, where we almost see TCRs where they shouldn't be 

T-Cell Receptors, Part Two: A History of TCR Discovery

Back to Part 1: The immtroduction
 
Nowadays we know quite a lot about TCRs. We know the general mechanics of how they're made, what they respond to, and how they respond to those things (note the 'general' - all three of these are vastly complicated processes, which are still topics of intense research). However, as is the way with all knowledge, this was not always the case.

The first appearance TCRs make into the literature (that I'm aware of) occurred back in the tail end of 1982. Researchers from Texas (among them, apparently, a member of the botany department, which is nice) were looking to generate antibodies against specific mouse T-cell lymphoma clones.

They constructed a library of B-cell hybridomas and immunised them with a model lymphoma line from a different mouse strain, before screening for antibodies that would recognise just the lymphoma cells. They found one; their antibody only reacted with the intended lymphoma; other lymphomas, T-cells, or spleen cells weren't targetted.

Pulling down the target by radioimmunoprecipitation (which is such an old and risky immunoprecipitation technique, it's not even listed on the wikipedia page) revealed a disulphide-linked glycoprotein heterodimer.

More importantly, examination of purified T-cells (but not B-cells) showed that other T-cells had similar structures on their cell-surfaces, but these weren't reactive against their new antibody. Sounds like we have a clonal marker of T-cells here!

The next year saw a flurry of papers: the finding was repeated in humans, and evidence piled up to show that this was the variable, clonotypic, T-cell specific receptor in both humans and mice. It wasn't called it yet, but the αβ TCR had been found.

Next came the laying of the groundwork for my project (which I assume is why they did it); the search for the genes that encode these receptors. The majority of the TCR genes were cloned through a nucleic acid subtraction strategy, which was based upon a few assumptions about the nature of the TCR genes, such as:

• they will be expressed by T-cells, but not B-cells
• they will likely be rearranged, as B-cell antigen receptors were already known to be
• in addition to their variable region (for antigen binding), they should have a constant region (to mediate downstream effects of antigen binding).

With these concepts in mind, the substraction strategy proved to be a pretty cunning choice. Total RNA was extracted from antigen specific, T-cell hybridomas, and reverse transcribed into first strand cDNA, radiolabelled with ³²P. Then second strand cDNA (i.e. the same orientation as the sequence encoded in the RNA) is prepared from another closely related, but non-identical source (some papers used B-cell hybridomas, while others opted for different T-cell hybridoma clones), which is then incubated with the labelled T-cell cDNA.

Any sequence that is shared by both cell types will hybridise together (as the prepared cDNA for each was of the opposing strand). Any non-shared sequence remains single stranded, and can be selected out by hydroxyapatite fractionation; the shared transcripts get subtracted from the T-cell specific ones.

Using this technique, the β chain genewas cloned first. The next gene cloned was assumed to be the pair of the β, which would make it what we now call the α chain. However, partial protein sequences from the binding partner of β didn't match the translated sequence of this second gene.

What'd they'd actually cloned was the γ chain; it was the third gene cloned that turned out to be the partner of the original, and thus the α chain.

(For those that are interested, there's a great contemporaneous review written by Frank Fitch that sums up pretty much all of what was known about TCRs at the time.)

This left us with a little mystery; if the first TCR gene (β) paired with the third one found (α), what's the second one (γ) doing?

It was a couple of years before the answer came to light, with the discovery of γδ T-cells; non-αβ T-cells which expressed CD3 (or T3, in old money) associated with the product of the TCRγ gene along with the previously unknown binding partner for γ: δ had been δiscovered.

It took a different trick to discover the gene for TCRδ, along with a bit of serenpidity. These researchers were using pulsed-field gel electrophoresis (a technique for the separation of very large DNA fragments) to investigate recombination in the TCRα locus.

In running these huge gels, they noticed that a region ~90kb upstream of the TRAC gene was undergoing rearrangement. Upon closer inspection there were homologous VDJ gene segments, getting recombined together and spliced onto a novel constant region.

It's small wonder the δ locus was so hard to find; it was hiding inside the α locus, making use of members of the same V region array. Plus, it gets completely deleted by any successful α recombination - which incidentally makes for a nice failsafe to ensure an αβ-T-cell can't express any γδ-TCR by mistake.

There we have it; a brief(ish) account of the discovery of all four chains of the TCR (discounting the funny ones that some other animals have). 

So, now we know how and where they're supposed to be expressed, we can start to appreciate the intrigue about when they pop up elsewhere...

On to Part 3, where the TCRs start to roam

T-Cell Receptors, Part One: The Immtroduction

Part 0: The introduction to the introduction

For those that might not have much background immunology knowledge, here's a quick whistlestop tour of the relevant basics, in order to better appreciate the rest of the series. If you already have a decent immunology or TCR background, you might just want to skip to the next instalment.

Our immune system exists to keep foreign pathogens (viruses, bacteria, fungi, worms, amoebi and so on) out of our body, or under control, to stop them causing us disease.

It works on a simple basis; discriminate between 'self' and 'non-self' (or 'host' vs. 'pathogen', 'us' vs. 'them', 'cops' vs. 'robbers' etc.). The immune system itself can then be divided into two broad categories; innate and adaptive.

Innate immunity is a fast acting defence that's hard-coded in our DNA, protecting us from anything that looks like it might be bad. There are specialised immune cells which are always on the lookout for the usual suspects that we know (from evolutionary experience) would like to infect us. In addition to these special cells, all the other cells in the body contains a number of factors to prevent infection, or to alert the immune system to get them to take care of things.

Adaptive immunity (the focus of this series) is a bit slower, but has the advantage of being, well, adaptable. It's not just about reacting to things, but about trying to antipate what we might be infected with, and remembering what has infected us in the past. Instead of inheriting our adaptive immune systems along with our innate systems, we just inherit the tools to make ours up as we go along.

There are two main white blood cell types responsible for adaptive immunity - T-cells and B-cells - which do different roles, working together to keep invaders out. Due to the nature of how I spent my childhood, I tend to describe them with a gaming analogy; B-cells are your ranged attackers, lurking in the lymph nodes attacking pathogens from afar, while T-cells are your melee troops, patrolling the tissues, looking for any cell that's showing signs of disease.

Both cell types use a very special kind of receptor to look for signs of infection or disease, but do different things once they've found one. An important thing to appreciate though is that all the different T- and B-cells are looking for different signs of infection.

In this series I'm not going to go into exactly how they stop pathogens, focusing instead on how they find them.

Briefly, B-cells secrete antibodies (or immunoglobulins), which circulate through the body and bind specifically to non-self things, such as viruses and bacteria. Binding of antibodies to a pathogen can either directly neutralise them, or can act as a flag to other immune cells, as if to say "hey guys, we found this critter trying to creep in, let's get him!".

T-cells come in more flavours, each with slightly different modes of action, but all work by getting up close and personal with the cells they're surveying. The action of T-cells relies on the fact that all cells in the body display bits of whatever's going on inside of them on their cell surface.

The most important system by which cells reveal what's going inside is the MHC presentation pathway. In this system, cells chop up proteins inside of them, and put the resulting peptide fragments into a special groove in an MHC molecule, before putting the whole thing on the outside of their membrane.

This means that passing T-cells just have to check out the MHCs of a given cell to find out what's going on under the surface, and thus tell if cells are infected (by looking for the presence of non-self peptides on the other cells' MHCs). If it finds a sick cell, a T-cell can either kill that cell directly, or get other cells (including B-cells) involved to sort out the problem.

MHC presentation is so important, there is a cell called the dendritic cell whose main job is to take in antigens from its surroundings, just to display their peptides to T-cells. There are a few other cells who do this, particularly macrophages and B-cells; together they're known as professional antigen presenting cells (or pAPC).

It's a pretty well-oiled machine, with (loosely) B-cells taking care of bad stuff floating around outside of our cells, and T-cells taking care of bad stuff floating around inside of our cells (or just bad cells themselves, like cancer cells).

But how do T- and B-cells recognise these sick cells? The trick is that they make use of very special proteins, called variable antigen receptors (where an antigen is the part of a pathogen or molecule recognised by the adaptive immune system).

The basic principle of these variable antigen receptors is simple, and is based upon the fact that there are more pathogens out there than we could easily encode a response to in our genome.

To explain that a little clearer: there are very many bugs and nasty pathogens out there that would like to infect us. We don't necessarily want them infecting us, so we need ways of detecting (and eliminating) them.

Theoretically we could just lots of genes in our genomes, each encoding a receptor for every different antigen. Sadly, as there are effectively near-infinite potential pathogens, this wouldn't work (as we can't produce a near-infinite genome, due to the lack of nearly-infinitely sized cells).

Instead, we've evolved a fiendishly clever system that uses a relatively small of amount of DNA (a fraction of our genome) to create thousands of times more genes then the rest of our genome could hope to hold. These genes encode the T-cell receptor (TCR) and B-cell receptor (BCR) (whose secreted, soluble forms are antibodies). I'll let you guess which is on which cell type.

Effectively, every different T- or B-cell has a different TCR or BCR, allowing each to recognise a different antigen. Our bodies can make millions (if not gazillions) of TCRs and BCRs, thus enabling us to protect ourselves against a world full of untold nasties.

The antigen receptors are surface expressed heterodimers; that is, they're made up of two different polypeptide chains linked together, and appear on the outside of their respective cells. There are two kinds of TCR, made up of different polypeptides: αβ and γδ TCRs, which are found on the two main different kinds of T-cell (αβ T-cells and γδ T-cells respectively; see, sometimes scientists make nice nomenclature).

Each chain of each receptor is encoded at a different location in the genome (with one noteable exception, that will be discussed in a later blogpost).

At each location, instead of just having one standard, continuous gene (ignoring introns), the DNA that codes for the antigen receptors is instead broken up into an arrays of gene segments, which can be joined in different combinations to produce different proteins, from the same original germline DNA.

These gene segments can be divided up into one of three types: Variable (V), Diversity (D) and Joining (J) regions, hence the name this process is given: V(D)J recombination. In order to bring 'selected' gene regions together, the RAG recombinase complex loops out the DNA between two gene segments and chops the loop out, joining the two cut ends together.

Every receptor chain has a V and a J region, while within a given receptor only one of the two chains contains a D region. In the case of TCRs, it is the β and δ chains that have a D, while the α and γ chains go without.

So, in α and γ chains one of the Js is joined to one of the Vs, while in β and δ there's an extra step; a J must first recombine with a D, then this DJ can join to a V. Recombined variable regions are then spliced onto a constant region in the mRNA, before translation and dimerisation with their partner chain. The region of the protein formed at the intersection between the V(D)J regions is known as the CDR3 region, and is the major factor in determining what antigen is recognised by that TCR.

There are a lot of Vs and Js to choose from (along with a couple of Ds in the chains that have them); you can see how many over at IMGT (here are the α/δ, β and γ loci). This means there's a lot of potential combinations, meaning a lot of different possible TCR. This is aided by the fact that the receptors are heterodimers, so the diversity is compounded.

However, the number of possible TCRs gets really ridiculous when you consider a further two mechanisms used to generate diversity.

When the DNA gets cut in order to be recombined, it's not always entirely accurate; there can be a bit of exonucleolytic 'nibbling' away of the ends. After the nibbling, a different amount of any of the four bases (A, C, T or G) may be added by the enzyme Tdt, before the segments are finally sealed together by DNA repair enzymes.

It might sound fiddly, but it really is a beautifully elegant system for keeping us going in the face of a dangerous microbial world. We can see how important adaptive immunity is when looking at the severe problems faced by people who have a defect in a vital component of the system.

I've only covered a few things here, and even then in a fraction of detail they deserve. I've not even touched on a lot of other hugely important aspects of immunology, such as, I don't know, the role that all the other immune cells play. Even on topic, I've only gone over the bare minimum; maybe in a future series (or later in this one if I have time) I'll cover topics such as how we ensure our TCRs onlyrecognise 'bad' peptide-MHC complexes, or why we need both αβ andγδ T-cells.

For now, this little introduction should be enough to understand most of what I'll be covering in the rest of the series.

On to Part 2: A history of TCR discovery