An immune system protein helps ovarian cancer survive

Scientists have discovered that a well-known protein, when produced by ovarian cancer cells, protects them from being killed by the immune system. This might apply to other kinds of cancer as well, opening doors to new immunotherapies and hopefully helping in the fight against cancer.

What makes us distinct from one another, from an immunological point of view, is a group of genes called HLA. They are varied across the human population and it is these genes that need to be tested and perfectly matched when looking for a transplant donor. One of them, called HLA-E, seems to also have a role in cancer, and this role is good for the cancer, bad for the patient. Marloes Gooden and colleagues made this discovery in ovarian cancers (1), which are usually diagnosed quite late and have poor prognosis, as only about one in two diagnosed women will live longer than 5 years.

Any cancer that arises in our body can be recognised and attacked by the immune system. In turn, the cancer cells can develop some strategies to evade the immune response and HLA-E production seems to be one of these strategies. Cancer masses are often infiltrated by immune system cells called cytotoxic T lymphocytes, or CTLs. They act a little bit like very aggressive policemen who chase and kill criminals. CTLs have the natural ability to recognise and kill any cell that doesn’t appear normal, which in this case are cervical cancer cells. But when a cancer cell produces this HLA-E protein, it seems to be functioning as an anti-CTL taser. The CTL is still there, in the tumour, and even though it might be able to “see” the nasty cancer cell, HLA-E makes it not do nothing about it.

The researchers analysed ovarian cancer samples from hundreds of patients, many of whom already died. When they looked at the amount of CTLs in the samples it became pretty evident that the more CTLs infiltrate the cancer (the more policemen chase the criminals), the longer the patient will live. But the picture didn’t seem quite complete at that stage. They decided to divide the cancer samples into two groups: one where the HLA-E levels were very low or even undetectable, and one where they were high or very high. In the first group they made a similar finding: the more CTLs infiltrated the cancer mass the longer the patient lived, and the difference in survival was even more evident. But when they looked at the patients with high HLA-E levels in the cancer cells, it turned out that the number of infiltrating CTLs did not make any difference to how long the patients would live.

The authors do mention that they stumbled upon more molecules that might be connected with this whole phenomenon, but just discovering this important function for HLA-E is an important piece of the puzzle. There are currently quite a lot of established and experimental therapies for cancer called monoclonal antibody therapies. They way they work, is by giving the patient antibodies against molecules that seem to facilitate cancer survival and aggressiveness. The study mentioned here might have identified one of these molecules and hopefully, in the future, new anti-HLA-E therapies will let women with ovarian cancer live much longer.

1. HLA-E expression by gynecological cancers restrains tumor-infiltrating CD8+ T lymphocytes.

Stepping over the stop

Imagine you have a robot that is baking a cake according to a precisely written recipe. And when it finishes making the cake base, the recipe says “stop here”, even though there are instructions for a creamy topic below. Cells do a similar thing: in an RNA molecule they have a recipe for a protein molecule, and they follow the instructions exactly, until they reach a ‘stop’ word. A recent finding by Karijolich and Yu (1) has shown that, at least in yeast cells, there are enzymes that can change the “stop” instruction and make the cells read it as continuation of the protein and keep reading the instructions further. It’s as if you erased the “stop here” instruction from the recipe and gave it back to the robot. After making the base it would go straight to making the topping.
Why would the cells do that? Sometimes they might need the protein to be bigger, because the extra bit can provide extra functions. For instance it can cope better in stressful conditions. In normal circumstances, introduction of such an extra bit would be a waste of energy. But when the environment changes from normal to stressful, the cells don’t need to produce a whole range of new protein-coding RNAs, they already have them, and all they need to do is change “stop” to “go on” and voila! A new protein that copes well with environmental stress. And it needn’t be stress, there can be plenty other things that can change inside or outside of the cell, and having such RNA molecules is quite a fast way of responding to these changes. It’s like you just found out that your friends are bringing kids to the dinner party and the kids just loooove creamy topping. You don’t need to prepare a whole new set of instruction for your robot to follow, you just remove the “end here” bit and a cake with creamy topping is on the way.
In the publication I’m mentioning here, the experiments were done in yeast. But it’s a well known fact that plenty of the processes that go on in yeast cell also happen in higher organisms – including humans. What implications has it got? In this publication, the cells were discovered to edit the “stop” instruction in the RNA molecule, but it is very likely that other bits of the instruction might be altered as well. This might lead to substantial change in how we understand the relation between instructions in RNA and what protein we expect to be made from these instructions.

1. Converting nonsense codons into sense codons by targeted pseudouridylation

How anti-genes can make diabetes worse.

Type 2 diabetes was recently declared an epidemic, as its incidence doubled from 1990 to 2005. Scientists have found several new genes that seem to be contributing to the condition and are now hoping this might be the new target of anti-diabetic drugs.

Sugar that we consume with our food gets from our gut to our bloodstream and then from our bloodstream to our body cells. The latter process is regulated by insulin. Sugar can enter the cells through a special door that are normally locked. Insulin is the key, and the lock on the cell’s door is called a receptor. Diabetes can result from either our body not producing insulin so that there’s no key to open the door (which happens in type 1 diabetes) or the locks in the doors get messed up in all cells so no matter how much keys you try – you can’t open the doors (which results in type 2 diabetes). Why the locks get messed up in our body cells is not exactly clear. The recent finding, however, sheds some light on how several genes might be contributing to this.

Scientists concentrated on two genes – miR-103 and miR-107. They aren’t regular genes that encode proteins, they are called micro RNA – small bits of RNA that can stop production of other proteins. They are – in a way – anti-protein genes. Trajkovski and colleagues(1) did two things with these genes in mice tissues – either over-activated them in healthy animals or silenced them in ones that suffer from type 2 diabetes. What seems pretty clear from their results is that the more active these genes are the more diabetes 2-like symptoms the animals showed. And like wise – silencing these genes in diabetic animals reduced the severity of their symptoms. The researchers even went a step further and asked why this might be. As I mentioned, the two genes they focused on do not code proteins but actually can stop the production of some of them. And that was the case here as well – the two genes prevented generation of caveolin 1, a protein which can be thought of as an important element of the lock in the door opened by insulin.

Scientists are obviously excited about this as this opens a whole new venue for treatment of type 2 diabetes. And what is also very exciting for me personally, is that when they set off to find new genes involved in diabetes, they found 5 of those non protein-coding micro RNAs. To keep things simple they only analysed two of them, but even with those two they got significant results. I’m looking forward to some more research on the other three.

1. MicroRNAs 103 and 107 regulate insulin sensitivity