About this Author
DBL%20Hendrix%20small.png College chemistry, 1983

Derek Lowe The 2002 Model

Dbl%20new%20portrait%20B%26W.png After 10 years of blogging. . .

Derek Lowe, an Arkansan by birth, got his BA from Hendrix College and his PhD in organic chemistry from Duke before spending time in Germany on a Humboldt Fellowship on his post-doc. He's worked for several major pharmaceutical companies since 1989 on drug discovery projects against schizophrenia, Alzheimer's, diabetes, osteoporosis and other diseases. To contact Derek email him directly: Twitter: Dereklowe

Chemistry and Drug Data: Drugbank
Chempedia Lab
Synthetic Pages
Organic Chemistry Portal
Not Voodoo

Chemistry and Pharma Blogs:
Org Prep Daily
The Haystack
A New Merck, Reviewed
Liberal Arts Chemistry
Electron Pusher
All Things Metathesis
C&E News Blogs
Chemiotics II
Chemical Space
Noel O'Blog
In Vivo Blog
Terra Sigilatta
BBSRC/Douglas Kell
Realizations in Biostatistics
ChemSpider Blog
Organic Chem - Education & Industry
Pharma Strategy Blog
No Name No Slogan
Practical Fragments
The Curious Wavefunction
Natural Product Man
Fragment Literature
Chemistry World Blog
Synthetic Nature
Chemistry Blog
Synthesizing Ideas
Eye on FDA
Chemical Forums
Symyx Blog
Sceptical Chymist
Lamentations on Chemistry
Computational Organic Chemistry
Mining Drugs
Henry Rzepa

Science Blogs and News:
Bad Science
The Loom
Uncertain Principles
Fierce Biotech
Blogs for Industry
Omics! Omics!
Young Female Scientist
Notional Slurry
Nobel Intent
SciTech Daily
Science Blog
Gene Expression (I)
Gene Expression (II)
Adventures in Ethics and Science
Transterrestrial Musings
Slashdot Science
Cosmic Variance
Biology News Net

Medical Blogs
DB's Medical Rants
Science-Based Medicine
Respectful Insolence
Diabetes Mine

Economics and Business
Marginal Revolution
The Volokh Conspiracy
Knowledge Problem

Politics / Current Events
Virginia Postrel
Belmont Club
Mickey Kaus

Belles Lettres
Uncouth Reflections
Arts and Letters Daily
In the Pipeline: Don't miss Derek Lowe's excellent commentary on drug discovery and the pharma industry in general at In the Pipeline

In the Pipeline

« Ray Firestone's Take On Pharma's Plight | Main | Business Note: Random Promotion, Anyone? »

November 17, 2011

Brain Cells: Different From Each Other, But Similar to Something Else?

Email This Entry

Posted by Derek

Just how different is one brain cell from another? I mean, every cell in our body has the same genome, so the differences in type (various neurons, glial cells) must be due to expression during development. And the differences between individual members of a class must be all due to local environment and growth - right?

Maybe not. I wasn't aware of this myself, but there's a growing body of evidence that suggests that neurons might actually differ more at the genomic level than you'd imagine. A lot of this work has come from the McConnell lab at the Salk Institute, where they've been showing that mouse precursor cells can develop into neurons with various chromosomal changes along the way. And instead of a defect (or an experimental artifact), he's hypothesized that this is a normal feature that helps to form the huge neuronal diversity seen in brain tissue.

His latest work used induced pluripotent cells transformed into neurons. Taking these cells from two different people, he found that the resulting neurons had highly variable sequences, with all sorts of insertions, deletions, and transpositions. (The precursor cells had some, too, but different ones, suggesting that the neural cell changes happened along the way). And this recent paper suggests that neurons have an unusual number of transposons in their DNA, which fits right in with McConnell's results.

The implication is that human brains are mosaics of mosaics, at the cell and sequence levels. And that immediately makes you wonder if these processes are involved in disease states (hard to imagine how they wouldn't be). The problem is, it's not too easy to get ahold of well-matched and well-controlled human brain tissue samples to check these ideas. But that's the obvious next step - take several similar-looking neurons and sequence them all the way. Obvious, but very difficult: single-cell sequencing is not so easy, to start with, and how exactly do you grab those single neurons out of the tangle of nerve tissue to sequence them? Someone's going to do this, but it's going to be a chore. (Note: McConnell's group was able to do the pluripotent-cell-derived stuff a bit more easily, since those come out clonal and give you more to work with).

Now, the idea that neurons are taking advantage of chromosomal instability to this degree is a little unnerving. That's because when you think of chromosomal instability, you think of cancer cells (See also the link in that last paragraph. It's interesting, as an aside, to see that those last two are to posts from this blog in 2002 - next year will mark ten years of this stuff! And I also enjoy seeing my remark from back then about "With headlines like this, I can't think why I'm not pulling in thousands of hits a day", since these days I'm running close to 20K/day as it is).

So, on some level, are our brains akin to tumor tissue? You really wonder why brain cancer isn't more common than it is, if these theories are correct. There may well be ways to get "controlled chromosomal instability", though, as opposed to the wild-and-woolly kind, but even the controlled kind is a bit scary. And all this makes me think of a passage from an old science fiction story by James Blish, "This Earth of Hours". The Earthmen have encountered a bizarre civilization that seems to involve many of the star systems toward the interior of the galaxy, and a captured human has informed them that these aliens apparently have no brains per se:

"No brains," the man from the Assam Dragon insisted. "Just lots of ganglia. I gather that's the way all of the races of the Central Empire are organized, regardless of other physical differences. That's what they mean when they say we're all sick - hadn't you realized that?"

"No," 12-Upjohn said in slowly dawning horror. "You had better spell it out."

"Why, they say that's why we get cancer. They say that the brain is the ultimate source of all tumors, and is itself a tumor. They call it 'hostile symbiosis.' "


"In the long run. Races that develop them kill themselves off. Something to do with solar radiation; animals on planets of Population II stars develop them, Population I planets don't."

The things you pick up reading 1950s science fiction. Blish, by the way, was an odd sort. He had a biology degree, and a liking for James Joyce, Oswald Spengler, and Richard Strauss. All of these things worked their ways into his stories, which were often much better and more complex than they strictly needed to be. Here's a PDF of "This Earth of Hours", if you're interested - it's not a perfect transcription, though; you'll have to take my word for it that the original has no grammatical errors. It's a good illustration of Blish's style - what appears at first to be a pulpy space-war story turns out to have a lot of odd background dropped into it, along with speculations like the above. And for someone who didn't always write a lot of descriptive prose, preferring to let philosophical points drive his plots, I find Blish's stories strangely vivid, particularly the relatively actionless ones like "Beep" or "Common Time". He's pretty thoroughly out of print these days, but you can find the paperbacks used, and in many cases as e-books. Now if you're looking for someone who always lets philosophical points drive his stores, then you'll be wanting some Borges. (As it happens, I've had occasion to discuss that particular translation with an Argentine co-worker. But this is not a literary blog, not for the most part, so I'll stop there!)

Comments (30) + TrackBacks (0) | Category: Biological News | Book Recommendations | Cancer | The Central Nervous System


1. Morten G on November 17, 2011 9:54 AM writes...

No mention of antibodies?

Permalink to Comment

2. Imaging guy on November 17, 2011 10:18 AM writes...

"how exactly do you grab those single neurons out of the tangle of nerve tissue to sequence them?"

I should think a method called laser captured microdissection can be used to obtain DNA,RNA and proteins from the single cells of stained tissue section. I have seen people used this method to do single cells PCR and mass spectrometry.

Permalink to Comment

3. Wile E. Coyote on November 17, 2011 10:19 AM writes...

If the comparison is between pluripotent stems cells and thier derived neurons. I think caution needs to be exercised in interpreting the results. Cell cultures are notorious for picking up various genetic changes/chromosomal instablility. Question: did they also do the same comparison with pluripotent stem cells and other cell types induced. My bet is that these also had genetic changes. If not, then I think this is an interesting result.

Instead of doing the next step with human tissue, which they lament as difficult to procure, why not monkeys or some other lab animals? Seems that there ought to be the same processes going on in other species, not just humans.

Permalink to Comment

4. luysii on November 17, 2011 10:44 AM writes...

It's worth noting that tumors of central nervous system neurons are exceedingly rare. I don't think I ever saw one in years of clinical neurology. Brain tumors certainly do occur. Meningiomas are tumors of the covering tissues of the brain. The most common type of brain tumor is the glioma, derived from glia, supporting cells of the brain, not neurons. Then there are pituitary tumors, derived from neuroendocrine cells, ependymomas derived from non-neuronal cells lining the internal spaces of the brain containing CSF (e.g. the ventricles). None of these are neurons.

Tumors from neurons found out side the brain occur in children -- these are the neuroblastomas.

The evidence that we have stem cells in our brain is extremely poor, despite publicity to the contrary. It may happen in animals. Pasko Rakic of Yale agrees (personal communication) and will have a review out shortly.

I always thought that the inability of neurons to divide was the reason that we could form long term memories at all.

I certainly agree with #3 Wile E. Coyote -- iPSCs are the process of extremely nonphysiologic manipulations. They form with very low efficiency -- implying that most cells in the culture medium die from the treatment. Differentiating them to neurons is even more nonphysiologic.

Permalink to Comment

5. johnnyboy on November 17, 2011 10:45 AM writes...

"So, on some level, are our brains akin to tumor tissue? You really wonder why brain cancer isn't more common than it is, if these theories are correct."

Well if this instability is a true phenomenon rather than something specific to in vitro conditions these cells were in, it would only come into play during development, before neurons stop dividing. As you probably know, most (if not all) adult brain cancers do not involve neurons, but glial cells or other cell components.

Speaking of old sci-fi, you reminded me of a short story I read as an impressionable boy, in which a time-travelling assassin was coming from the future to kill all the prominent cancer researchers, because they were fighting against uncontrolled cell growth, which was the ultimate step in human evolution - to be able to change the body into any shape needed. Which you can agree would be a major life improvement when you have an itch you can't reach.

Permalink to Comment

6. Sam Weller on November 17, 2011 10:57 AM writes...

"You really wonder why brain cancer isn't more common than it is"

Two reasons that come to mind are that the genomic variations among brain cells are of certain types that are not akin to cancer variations, and that the brain is better insulated from many environmental factors (carcinogens, radiation etc) compared to the rest of our body.

Permalink to Comment

7. Paul on November 17, 2011 12:11 PM writes...

See also Peter Watts' story, "The Things" (which is "The Thing"/"Who Goes There?", but told from the point of view of the alien.)

Permalink to Comment

8. pete on November 17, 2011 12:41 PM writes...

What if elevated rates of transposition in neural genomes is more of a secondary outcome that results from having cells that are generally:
- post-mitotic
- very active transcribers
So, even if the effects of this process on neural genes isn't entirely neutral, maybe we shouldn't automatically infer too much evolutionary design.

OTOH, scattering homologous elements around the genome seems like a good way to increase recombination rates, doesn't it (?)

Permalink to Comment

9. PTM on November 17, 2011 12:54 PM writes...

Interesting but I find it somewhat unlikely that such rearrangements serve any functional role in neurons.

What would such instability accomplish? Genetic variation isn't a goal in itself here, so it would have to serve some purpose but what exactly?

And whatever the purpose genetic instability seems like a very costly way of achieving it since I would expect it to lead to all sorts of problems with random RNAs and proteins messing up the cell.

On the other hand, if those rearrangements are in parts of the genome that are not expressed in neurons, they could simply be a byproduct of differentiation and have little consequence for the way neurons operate. Or paradoxically they could even serve as a protection against cancer if they "shredded" the genetic code required for cells to divide, but that seems quite far-fetched as it would require a tight control of the whole process.

An artefact seems the most probable explanation at this point.

Permalink to Comment

10. Sigivald on November 17, 2011 1:20 PM writes...

That reminds me of the Peter Watts version of The Thing, told from the Thing's viewpoint.

The "brains as tumors" idea is the same - makes me wonder if Watts read Blish.

Permalink to Comment

11. Sigivald on November 17, 2011 1:22 PM writes...

I see Paul beat me to mentioning Watts' story.

I second his recommendation.

Permalink to Comment

12. patentgeek on November 17, 2011 1:25 PM writes...

I can't let a reference to Blish go by without putting in a plug for two other of his stories: "Sunken Universe" (1942) and the subsequent classic, "Surface Tension" (1952). I read the latter one golden afternoon over 45 years ago and still recall the rush of pleasure.

Permalink to Comment

13. Vader on November 17, 2011 1:28 PM writes...

"The problem is, it's not too easy to get ahold of well-matched and well-controlled human brain tissue samples to check these ideas."

Yeah. Seems like a lot of humans aren't fully in control of their brain tissue.

Permalink to Comment

14. Todd on November 17, 2011 2:05 PM writes...

#8, you got it right with the homologous elements. It does increase recombination rates a LOT. Also, they tend to hit up areas under a lot of selective pressure and which have low transcription rates. Considering the huge differentiation in the brain, you're almost guaranteed high recombination rates.

On top of that, this dovetails nicely with yesterday's post about transcription levels in cell lines. I think someone nailed it yesterday when they said that the low transcription levels mean a high "gain" on changes due to environmental pressure. This just seems to nail it.

(Oh, and if you want to learn more, check this out:

Permalink to Comment

15. Derek Lowe on November 17, 2011 2:53 PM writes...

Patentgeek, "Surface Tension" is indeed some good ol' classic SF. Reading "Earthman Come Home" when I was about 13 was a similar experience for me, in the "I can't believe how much I'm enjoying this book" way. If you hit this stuff at the right age, it stays with you.

Permalink to Comment

16. Henning Makholm on November 17, 2011 3:00 PM writes...

Sounds like a good evolutionary reason to disable mitosis in neurons.

Permalink to Comment

17. pete on November 17, 2011 3:00 PM writes...

@14 Todd
Interesting - but I'm not sure I follow as to why high phenotypic variation among brain nerve populations ensures high genomic recombination rates. How would that work?

ps: the idle speculation about increased "gain" -- that would be me

Permalink to Comment

18. Nick K on November 17, 2011 4:39 PM writes...

Talking of classic SF, who reads Theodore Sturgeon these days? A genuine master of the genre, and yet his books are all out of print.

Permalink to Comment

19. Anonymous on November 17, 2011 4:48 PM writes...

The cell selection during brain development is somewhat analogous to the cell selection of genetically divergent cells in the immune system, so a mechanism in the brain that takes advantage of enhanced neuronal diversity doesn't sound completely implausible.

Now just wait for someone to find a mechanism for the neuronal genetic selection to be passed on to offspring.....:P

Permalink to Comment

20. metaphysician on November 17, 2011 5:21 PM writes...

My first thought, when hearing this, is "memory storage medium." Maybe the genetic variation is because neurons use their DNA to to store information?

As for brain and classic sci-fi, let me recommend Brainwave, by Poul Anderson. A very neat book, especially in light of certain recent discoveries in physics. . .

Permalink to Comment

21. Norman Yarvin on November 17, 2011 5:29 PM writes...

Remember that "we found lots of variability" is not easily distinguishable from "our experimental procedure has a high noise level".

Permalink to Comment

22. LB on November 17, 2011 5:35 PM writes...

This has been known for some time (although without receiving much attention). Most of the aneuploid neural cells seem to be monosomic, whereas there seem to be much fewer trisomi cells. This correlates very well with systems studies in yeast where its has been demonstrated that most cellular systems work very well based on a haploid dosage level, indicating that diploidy has more of a redundancy aspect (to much dosage seems bad though, and sex chromosomes seem to be a special case). Its mainly sloppy thinking around what the genotype-phenotype concept actually constitutes and research overspecialization that makes people surprised by this. Only at the zygote stage ( = the genetic aspect of inheritance) do we actually have a genotype that is truly singular in its structure.

Permalink to Comment

23. Sam P on November 17, 2011 9:03 PM writes...

Nick K: Actually Theodore Sturgeon is mostly in print, just not in inexpensive editions. North Atlantic books has the (pricey) 13 volume series The Complete Stories of Theodore Sturgeon covering all the known published and previously unpublished short stories, and Vintage Books has most of the novels in trade paperbacks.

Permalink to Comment

24. cliffintokyo on November 18, 2011 2:23 AM writes...

RE: SF Classics - Reading the Asimov 'Foundation' trilogy was a similarly exhilarating experience for me as a teenager.
Its probably all that ultra imaginative 'thinking outside the classroom' stuff that does it.

I need to move to a Population I planet, so I can settle down to read some James Blish and Theodore Sturgeon.

Permalink to Comment

25. Nick K on November 18, 2011 12:04 PM writes...

#23 Sam P: Many thanks for the info - I'll check it out. I used to have several of his collections in paperback, but they have gone the way of all flesh, alas...

Permalink to Comment

26. Todd on November 18, 2011 2:28 PM writes...

@17 (aka Pete) you have it backwards. High genomic recominbation rates lead to higher phenotypic variability. In my unpublished data, this happens the strongest in the immune system genes (for obvious reasons) and there's good reason to assume that the same thing is going on in the brain.

Permalink to Comment

27. Vince on November 21, 2011 2:34 AM writes...

Todd said: and there's good reason to assume that the same thing is going on in the brain.

Hi Todd, I'm curious why you assume this? Or, restated, why would we want this?

I do not see the upside to a process which I'd imagine is functionally akin to attaching a random number generator to a neuron's genes and regulatory sequences.

Much to the contrary, I would assume that evolution, given the known mapping of function to neural structure/network topology for many tasks, would desire to narrow down the large set of potential gene variables to a smaller, more optimized set which broadly correlates with more optimal solutions in functional space. Especially given the hard limit set on network size and connectivity due to spatial and energetic bounds (see Simon Laughlin).

An example would be Parvalbumin-positive, fast-spiking neurons in the neocortex which are highly specialized for rapid and sustained firing. They are highly optimized, given the intrinsic trade-offs they face, for their niche. Wasting volume and energy on neurons that lack the metabolic capacity to function in the local network seems odd.

This idea seems to fly-in-the-face of not only my instinct, but much of the data I've seen. Granted there needs to be a source of variation, but to me this is more akin to a back-door way to substantiate a form of soft indeterminism and 'freedom from the chains of genes' than anything else.

Permalink to Comment

28. daedalus2u on December 2, 2011 6:15 PM writes...

PTM #9, no, it is a “feature”. Yes Vince #27 exactly right. When human tribes were evolving, the most successful tribes were those with a diversity of cognitive styles. Tribes with experts in diverse subjects (stone tools, wooden tools, animal tools, fire tools, food tools, etc) did better than tribes where everyone had the same cognitive abilities. Because tribes all shared the same genes, the diversity of cognitive styles couldn't come from diversity of genes.

What evolution could do was make neurodevelopment exquisitely sensitive to environmental influences via stochastic resonance. That is why no one can find “genes” for intelligence. There aren't any specific genes for intelligence. Intelligence is a property of a phenotype, not a genotype. The genotype makes a complicated brain with lots of plasticity and variation in order to produce cognitive styles with lots of plasticity and variation, so a to produce ideas and skill sets with lots of plasticity and variation.

That is also why there is no “gene” for autism, schizophrenia, or any of the other neuropsychiatric disorders. They are disorders of a phenotype, not a genotype. It is the differential influence of the environment on the myriad non-linear coupled pathways that comprise development. The butterfly effect writ large. Or rather the flagella effect writ large. A flip of a flagella one way modifies the diffusion of larger molecules and which transcription factor gets where first, which DNA fragment the transposon latches onto and copies at that precise moment of the cell cycle.

Permalink to Comment

29. Jolene on February 19, 2015 3:26 PM writes...

Wow, amazing weblog format! How lengthy have you ever been running a blog for?
you made blogging look easy. The overasll glance of your site is magnificent,
let alone tthe content!

Permalink to Comment

30. Jolene on February 19, 2015 3:27 PM writes...

Wow, amazing weblog format! How lengthy have you ever been running a blog for?
you made blogging look easy. The overasll glance of your site is magnificent,
let alone tthe content!

Permalink to Comment


Remember Me?


Email this entry to:

Your email address:

Message (optional):

The Last Post
The GSK Layoffs Continue, By Proxy
The Move is Nigh
Another Alzheimer's IPO
Cutbacks at C&E News
Sanofi Pays to Get Back Into Oncology
An Irresponsible Statement About Curing Cancer
Oliver Sacks on Turning Back to Chemistry