Saturday, October 29, 2016

Better Than Nanites: Custom T-cells

Rather startling developments in the use of our internal maintenance cells to target cancer or other problems.

I am a watching a very nice science fiction series, about a motley crew in space who try to be kick-ass and all, but deep down are just ... very nice people. Because they are Canadian, of course! Every show seems to steal another plot from past classics, like the Bourne Identity, Star Trek Deep Space 9, and even one featuring Zombies.

One crew member is an android, (named "Android"), but is touched with a bit of schizophrenia, a la Commander Data or Seven-of-Nine or Spock, about the virtues of humanity and being humanely idiosyncratic. She also features nanites- apparently tiny machines in her high-tech body that run around and repair things when she takes a hit for the ship.

Android to android: another android, shooting the ship's Android. Repair will now commence.

Such nanites are quite a stretch, current technology having nothing remotely similar, and Android's body being rather inhospitable to anything running around among all the wires, metal, electricity, and whatnot. Such nanites would have to have some kind of master plan for guidance, which would be pretty difficult to fit into a nano package.

Yet our own bodies do have nanites, called the cells of the immune system. This system as a whole is an organ that has no fixed location or shape, but travels around the body in the blood stream, lymph and elsewhere between cells- anywhere where damage occurs. These cells have a highly complex communication system that finds damage, detects what type, cleans out the damage, attracts other helper cells as needed, reads the local developmental and tissue patterns to help local cells do the fix correctly, and gradually turns itself off when finished.

One of the central actors of this system are helper T-cells, which intermediate between the damage signals, which come from normal tissue as well as specialized cells that roam around looking for damage, and the inflammatory and damage repair system, such as cells that create antibodies (B-cells), or that phagocytose and kill infected or damaged cells directly (CTL cells, macrophages). Some T-cells activate immune system actions, and other T-cells dampen them, and they do this over the whole time course of the damage reaction. HIV is an infection mostly of T-cells, killing them and leading to the collapse of the whole immune system.

One of the magic properties of T-cells is specificity. Like the antibody system of B-cells, T-cells use genetic/genomic trickery to generate a galaxy of specific receptors, called, as a family, the T-cell receptor, which can recognize specific molecules, such as proteins from viruses and bacteria. Each T-cell generates and shows one such variant on its surface, and thus the right individual T-cell has to go to the right place to initiate its response, part of which is rapid growth and replication into an army of T-cell clones (do that, nanite!). There is also a process, carried out mostly in the thymus, which deletes all the newly-born T-cells whose specificity is against proteins from its own body rather than against foreign entities.

Given all this, it has been interesting to learn that the immune system often acts against cancers as well. While composed of the body's own DNA and cells, cancers can express various altered proteins due to their mutations and deranged regulation, and also may express stress molecules that tip off parts of the immune system that those cells should be killed. On the other hand, cancers can also, though natural selection, cleverly express other signal molecules that turn the immune system off, thus shielding themselves from destruction. That is a serious problem, obviously.

So many researchers have been casting about for ways to get the immune system to overcome such barriers and attack cancers in a more robust way, especially in resistant cases. And after a lot of false starts, these approaches are starting to bear remarkable fruit. Some are drug-based approaches, but more direct are methods that re-engineer those cells to do what we want.

Since they are travelling cells, T-cells can be taken out of the patient. This allows new genes to be introduced, mutations made, etc., especially using the new CRISPER technologies. One approach is to add a receptor specific to the patient's cancer, such that the refreshed T-cells target it directly, and get activated by the tumor environment, and start to resolve the tumor. This approach has been quite successful, to the point that some patients undergo tumor lysis syndrome- a somewhat dangerous consequence of the tumor getting destroyed too quickly for the body to handle the resulting trash.

A recent paper elaborated this re-engineering approach to make it far more broad. Researchers introduce not only a new receptor to direct the T-cells to particular targets, but a multi-gene system to perform any additional function desired in response to targeting, such as pumping out a toxin, or a regulator / activator of nearby cells. This promises to supercharge the T-cell therapy approach, beyond the native scope of action of normal T-cells, however well-targeted.

For example, in a demonstration experiment, mice were given tumors on two sides of their bodies, one of which contained an additional genetic marker- the fluorescent protein GFP expressed on its surface. This is not a mammalian protein at all, but from an obscure bacterium, and would have no effect, if the experimenters had not also engineered a batch of that mouse's T-cells to express a combination of new genes.

One was a version of the common protein receptor Notch, which had its cell-external receptor portion replaced by a receptor for GFP, and its cell-interior portion replaced with the transcription factor Gal4. When the exterior portion of Notch proteins are activated, the internal portion gets cleaved off and typically travels to the nucleus to do its thing- activate a set of responsive genes. The other engineered gene was a Gal4-responsive gene expressing a cancer-fighting drug called Blinatumomab. This is an antibody specific to a B-cell antigen, which is appropriate since the introduced tumor is B-cell derived.

Demonstration of tumor targeting with engineered T-cells; description in the text.

The synthetic receptor is shown in green (synNotch), exposing a GFP receptor on the outside and a cleavable transcription regulator on the inside. Upon encountering the GFP-expressing tumor (green), it activates transcription of an antitumor drug (custom antibody) abbreviated BiTE, which attacks cells expressing the cell surface receptor CD19, which these tumors do. The green tumor regresses within two weeks, while the control tumor does not.

The demonstration shows that this engineered treatment can address practically any target that can be specifically distinguished from normal cells (indeed, one can imagine multiple engineered receptors being used in combination), and generate any gene product to treat it.

It also shows the increasingly expensive direction of medical care. Not only is the expressed gene product one of those recently-developed, highly expensive cancer drugs, but the T-cell extraction, reprogramming, and re-introduction has to be done on a custom basis for each patient, which is likely to be even more expensive.


  • The NRA has a screw loose ... arm in arm with Wayne LaPierre!
  • Guess which constitutional amendment is the most important?
  • Smoking still at fault for 30% of cancer deaths ... after all this time.
  • We are in deep CO2.
  • Financial regulation works.
  • The disorder has a name.
  • And a bitter end is in sight.