The plural of anecdote is NOT data (in medicine at least)

The previous post ( showed that collecting a bunch of small studies (anecdotes) was extremely helpful in seeing the larger picture.

In medicine exactly the opposite occurs. The only way to find out if something works is to do a controlled study. [ Science vol. 297 p 325 ’02 ] There were over 50 observational studies showing benefits for hormone replacement in menopausal women.. Observational studies are basically anecdotes. During the planning study for the Women’s Health Initiative (WHI), some argued that it was unethical to deny some women hormones and give them a placebo. The reason HERS (Heart and Estrogen/Progesterone Replacement Study) was even done was that Wyeth couldn’t get the FDA to approve hormone replacement therapy as a treatment to prevent cardiovascular disease, so they funded HERS to prove their case. Most readers of this have probably read all sorts of bitching about the slowness of the FDA in approving drugs but in this case they did the female populace a huge favor.

As you probably know, the results of hormone replacement in both studies were a disaster (the HERS trial was stopped at 5.2 years after because of increased breast cancer in the treated group). There was also an increased risk of coronary heart disease by 30%, stroke by 41%. At least hip fracture was reduced. Fortunately, even though these were bad outcomes, they were infrequent,(but more frequent in the treated group).

These weren’t lab animals, but someone’s wife and/or mother.

How could they have been so far off? Before all this started, estrogen users were different from nonUsers in several respects — first they were doing something about their health, and clearly had more medical supervision. In addition they were better educated, smoked less and of a higher social class, all of which tend to diminish morbidity and mortality.

Something very similar happened in my field of neurology (not that vascular disease doesn’t severely impact the nervous system). There was a very logical operation to improve cerebral circulation — the pulse just in front of your ear is the superficial temporal artery, a branch of the common carotid after it splits in the internal carotid which goes into the skull and supplies blood to the brain, and the external carotid. If the internal carotid is blocked and the common carotid artery is open, then open the skull and hook (anastomose) the superficial temporal artery to a vessel on the surface of the brain, bypassing the blockage. If you want to know how it is done see —

There was all sorts of anecdotal evidence of miraculous recovery from stroke. The neurosurgeons and vascular surgeons mounted a wonderful controlled study of the surgery even though many thought it was unnecessary — so 1377 patients were prospectively randomized to have the surgery or medical management. The surgery wasn’t better than medical management N Engl J Med 1985; 313:1191-1200November 7, 1985DOI: 10.1056/NEJM198511073131904, so the procedure was abandoned.

The plural of anecdote IS data

Five years ago I wrote a post on the perils of implicating a gene as the cause of a disease because one or two people with the disease had a mutation there (see the bottom). That is now back in spades with a new report from the Exome Aggregation Consortium (ExAC) [ Nature vol. 536 pp. 249, 277 – 278, 285 – 291 ’16 ].

What they did was to aggregate sequence data from 60,704 people on the parts of their genomes coding for the amino acids making up proteins (the exome — The paper has 80+ authors. The data is publicly available and is planed to grow to 120,000 exomes and 20,000 whole genomes in the next year. Both are orders of magnitude larger than any individual exome study so far. So study enough anecdotes (small studies) and pretty soon you have real data

The articles state that over a million people have now had either their exomes or their whole genomes sequenced ! ! !

The amount of variation in the human genome is simply incredible. Some 7,404,909 variants in the exome were described, of which 54% had never been seen before. These account for 1/8 of all the sites in all our exomes, implying that the exome comprises 60 megaBases of the 3200 megaBase human genome (1.8%). Most of the variants were single amino acid changes due changes in a single nucleotide, but there were 317,381 insertions or deletions (95% shorter than 6 nucleotides).

99% of all variants had a frequency of under 1% (e.g. not found in in more than 607 people), with half being found only once in the 60,704. 8% of the sites with variation contain more than one (consistent with what you’d expect of a Poisson distribution).

What is so remarkable is that the average participant has 54 variants previously classified as responsible for a genetic disorder. Not only that 183/192 variants thought to cause a rare hereditary disease were found in many healthy people, implying that they were incidental findings (anecdotes) rather than causal. It shows you what happens when you have adequate data.

They are pretty sure that their work will stand, because the exomes were sequenced many times over (deeply sequenced in the lingo) more than 10x in over 80% of the cohort.

I’d also written earlier about how full of errors our genomes are — see

A lot of the variants produced termination codons in the body of the exome, so a full-length protein couldn’t be produced from the gene (these are called truncation variants) — some 179,774 in the 7,404,909. Most occurred just once. Even so this means that most of the cohort had at least one or two. Even this rather negative knowledge was useful — since we have about 20,000 protein coding genes, they found 3,230 in which truncation variants NEVER occurred, implying that the protein is crucial to survival.


We’ve found the mutation causing your disease — not so fast, says this paper (posted 17 July 2011)

This post takes a while to get to the main points, but hang in there, the results are striking (and disturbing).

First: a bit of history. In the bad old days (any time over about 30 years ago) there was basically only one way to look for a disc in the spinal canal pressing on a nerve producing symptoms (usually pain, followed by numbness and weakness). It was the myelogram, where a spinal tap was done, an oily substance (containing iodine which Xrays don’t penetrate well) was injected into the spinal canal, and Xrays taken. The disc showed up as a defect in the column of dye (not really a dye as any chemist can see). This usually led to surgery if a disc was found, even if it was one or two spinal levels from where clinicians thought it should be based on their examination and other tests such as electromyography (EMG). This was usually put down to anatomic variability. Results were less than perfect.

Myelography was a rather stressful procedure, and I usually brought patients into the hospital the night before, got a cardiogram (to make sure their heart could take it, and that they hadn’t had a silent heart attack). Then the myelography itself, which wasn’t painful as the radiologist put the needle in under fluoroscopy so they could see exactly where to go. However many people got severe post-spinal headaches (invariably doctor’s wives), sometimes requiring a blood patch to plug the hole where the (large) needle used to inject the ‘dye’ went — it had to be large because the ‘dye’ was rather oily (viscous). The bottom line was that you didn’t subject a patient to a myelogram unless they were having a significant problem. Only very symptomatic people had the test, and usually when nonsurgical therapy had been tried and failed.

Fast forward to the MRI (Magnetic Resonance Imaging) era (nuclear magnetic resonance to the chemist, but radiologists were smart enough to get the word nuclear removed so patients would submit to the test). A painless technique, but stressful for some because of the close quarters in the MRI machine. You could look at the whole spinal canal, and see far more anatomic detail, because you actually see the disc (rather than its impression on a column of dye) and the surrounding bones, ligaments etc. etc.

What did we find? There were tons of people with discs where they shouldn’t be (e.g. herniated discs) who were having no problems at all. This led to a lot more careful assessment of patients, with far better correlation of anatomic defect and clinical symptoms.

What in the world does this all have to do with the genetics of disease? Patience; you’re about to find out.

There’s an interesting interview with Eric Lander (of Human Genome Project fame) in the current PNAS (p. 11319). He notes that in 1990 sequencing a single genome cost $3,000,000,000. He thinks that at some time in the next 5 years we’ll be able to do this for $1,000, a 3 million-fold improvement in cost. The genome has around 3,000,000,000 positions to sequence. As things stand now, it’s literally nothing to determine the sequence of a few million positions in DNA.

On to Cell vol. 145 pp. 1036 – 1048 ’11 which sequenced some 9,000,000 positions of DNA. This didn’t make a big splash (but its implications might). Just a single paper, buried in the middle of the 24 June ’11 Cell — it didn’t even rate an editorial. Now, as chemists, if you’re a bit shaky on what follows, all the background you need can be found in the series of articles found here –

As a neurologist, I treated a lot of patients with epilepsy (recurrent convulsions, recurrent seizures). 2% of children and 1% of adults have it (meaning that half of the kids with it will outgrow it, as did the wife of an old friend I saw this afternoon). Some forms of epilepsy run in families with strict inheritance (like sickle cell anemia or cystic fibrosis). 20 such forms have been tied down to single nucleotide polymorphisms (SNPs) in 20 different genes coding for protein (there are other kinds of genes) — all is explained in the background material above). 17/20 of these SNPs are in a type of protein known as an ion channel. These channels are present in all our cells, but in neurons they are responsible for the maintenance of a membrane potential across the membrane, which has the ability change abruptly causing an nerve cell to fire an impulse. In a very simplistic way, one can regard a convulsion (epileptic seizure) as nerve cells gone wild, firing impulses without cease, until the exhausted neurons shut down and the seizure ends.

However, the known strictly hereditary forms of epilepsy account for at most 1 – 2% of all people with epilepsy. The 9,000,000 determinations of DNA sequence were performed on 237 ion channel genes, but just those parts of the genes actually coding for amino acids (these are the exons). They studied 152 people with nonhereditary epilepsy (also known as idiopathic epilepsy) and, most importantly, they looked at the same channels in 139 healthy normal people with no epilepsy at all.

Looking at the 17/237 ion channels known to cause strictly hereditary epilepsy they found that 96% of cases of nonhereditary (idiopathic) epilepsy had one or more missense mutations (an amino acid at a given position different than the one that should be there). Amazingly, 70% of normal people also had missense mutations in the 17. Looking at the broader picture of all 237 channels, they found 300 different mutations in the 139 normals, of which 23 were in the 17. Overall they found 989 SNPs in all the channels in the whole group, of which 415 were nonsynonumous.

Well what about mutational load? Suppose you have more than one mutation in the 17 genes. 77% the cases with idiopathic epilepsy had 2 or more mutations in the 17, but so did 30% of the people without epilepsy at all.

The relation between myelography and early genetic work on disease should be clear. Back then, a lot was taken as abnormal as only the severely afflicted could be studied, due to time, money and technological constraints. As the authors note “causality cannot be assigned to any particular variant”. Many potentially pathogenic genetic variants in known dominant channel genes are present in normals.

What was not clear to me from reading the paper is whether any of the previously described mutations in the 17 are thought to be causative of strictly hereditary epilepsy were present in the 139 normals.

A very interesting point is how genetically diverse the human population actually is (and they only studied Caucasians and Hispanics — apparently no Blacks). No individual was free of SNPs. No two individuals (in the 139 + 152) had the same set of SNPs. Since they found 989 SNPs in the combined group, even in this small sample of proteins (17 of 20,000) this averages out to more than 3 per individual. Well, are there ‘good’ SNPs in the asymptomatic group, and ‘bad’ SNPs in the patients with idiopathic epilepsy? Not really, the majority of the SNPs were present in both groups.

I leave it to your imagination what this means for ‘personalized medicine’. We’re literally just beginning to find out what’s out there. This is the genetic analog of the asymptomatic disc. We may not know all we thought we knew about genetics and disease. Heisenberg must be smiling, wherever he is.

What reading the literature is like when things are barely understood

There is a very exciting paper to be described in a post to appear shortly. I ran a muscular dystrophy clinic for 15 years, and saw lots of Amyotrophic Lateral Sclerosis (ALS) — even though, strictly speaking it is not a muscular dystrophy. The muscular Dystrophy Association was founded by parents of weak children, before we could actually separate motor neuron disease from myopathy. In retirement, I’ve kept up an interest in ALS (particularly since all I could do for patients as a doc was — (drumroll) — basically nothing).

The fact that a fair amount of even sporadic ALS has a problem with a protein called C9ORF72 was particularly fascinating. All this came out less than five years ago (October 2011). Everything is far from clearcut even now.

That being the case, it might be of interest to look at the notes I accumulated as scientists began to explore what was wrong with C9ORF72, how the protein normally does whatever it does (we still don’t know really) and how the mutated product of the gene causes trouble (there are 3 main theories).

What you’ll see in what follows is the heat of scientific battle (warts and all), where things are far from clear. Enjoy. This is basically what used to be called a core-dump (back in the day when computer memory was made of metallic cores). Things are far from cut and dried even now so it might be of interest to see the many angles of attack on the problem, the confusion, the conflicting theories, as things became a bit more clear. It’s the scientific enterprise in action against a very horrible disease (trust me).

I’ll try and clear up the typos. I’ll also try to put the notes on the papers in semi-chronological order, but I make no guarantees. The notes may be incomprehensible, as they include only what I didn’t know rather than all the background needed to understand what’s in them .

First a bit of background — FTD stands for FrontoTemporal Dementia.

The #9p21 chromosomal region is another locus for ALS/FTD. It contains something called C9orf72, which contains a GGGGCC hexnucleotide repeat in the intron between noncoding exons 1a and 1b. Normal alleles contain less than 24 repeats (range 2 – 23). Those with ALS + FTD contain over 30 (actually they think the repeat length is much higher — 700 to 1,600 ! ! !). ORF probably stands for open reading frame.

The expansion is present in 12% of familial FTD and 22.5% of familial ALS — making it the most common genetic abnormality in both conditions. More importantly it is found in 21% of sporadic ALS and 29% of FTD in the Finnish population. Later they say it is the most common genetic cause of sporadic ALS (but only in 4%).

There are 3 possible mechanisms of toxicity
l. The RNA transcribed from the repeat acts as an RNA sponge, binding all sorts of RNAs it shouldn’t
2. Repeat Assoaicted Non-ATG translation (RAN translation) see later
3. Decreased expression of the mRNA for C9ORF72.

[ Science vol. 338 pp. 1282 – 1283 ’12 ] Now 40% of familial ALS, 21% of familial frontotemporal dementia, and 8% of sporadic ALS, 5% of sporadic frontotemporal dementia have expansions in C9orf72.

Not much is known about C9orf72 — it is conserved across species. It contains no previously known protein domains. The expansion leads to loss of one alternatively spliced C9ORF72 isoform (normally 3 isoforms are expressed), and to the formation of nuclear RNA foci (which appear to be composed mostly of the expansion). [ Neuron vol. 79 pp. 416 – 438 ’13 ] The function of C9ORF72 is unknown (8/13).

The current (12/12) thinking is that the repeats produce a glob of RNA which traps RNA binding proteins which have better things to do. The best analogy is myotonic dystrophy in which an expanded 3 nucleotide repeat sequesters muscleblind, an RNA binding protein involved in splicing.

The expansion is present in 46% of familial ALS in Finland and 21% of sporadic ALS there. But Finns are somewhat different genetically. The expansion is found in 1/3 of European ancestry familial ALS.

Interestingly some of the patients with FTD presented with nonfluent progressive aphasia.

[ Cell vol. 152 pp. 691 – 698 ’13, Neuron vol. 77 pp. 639 – 646 ’13 ] The protein aggregates of C9orf72 mutants contain TDP43 inclusions. But they also show additional p62 and ubiquilin positive pathology (with no TDP43 present). The abnormal proteins are due to translation of the expanded GGGGCC repeats (which should be nonCoding as they are in introns). This is an example of Repeat Associated Non-ATG translation (RAN). This was first shown for expanded CAG repeats, which can be translated in all 3 reading frames giving polyGlutamine, polyLysine and polySerine . A minimum of 58 CAG repeats was required for translation.

This work looked for translation of GGGGCC in all 3 reading frames (poly glycine-proline, poly glycine-alanine, polyglycine-arginine. They found that poly glycine-proline was found and in the protein inclusions which were p62 positive and TDP43 negative. Similar inclusions weren’t present in other neurodegenerative diseases, known to have nucleotide inclusions.

[ Proc. Natl. Acad. Sci. vol. 110 pp 7533 – 7534, 7778 – 7783 ’13 ] The expanded C9orf72 repeat is enough to cause neurodegeneration (mammalian neurons, and D. melanogaster). They placed either 3 or 30 copies of GGGCC into an epidermal growth factor vector between the start of transcription and the first ATG codon. The repeat can sequester the RNA binding protein Pur alpha (and other Pur family members). Interestingly, TDP43 didn’t bind to the repeat RNA, nor did hnRNP A2/B1 which binds to fragile X CGG repeat containing RNA. Overexpression of of Pur alpha is able to abort the neurogeneration in the mammalian neuonal cell line (Neuro-2a). So probably the excessive repeat number is acting as an RNA sponge.

Pur alpha is evolutionarily conserved. It controlls the cell cycle and differentiation. It is also a pomonent of the RNA transport granule. It interacts with Pur beta.

30 was as many repeats as they could manipulate experimentally — normals have 2 – 8 repeats, but patients with disease have from 100s to 1,000s of repeats, so the pathogenesis might be different.

[ Neuron vol. 80 pp. 257 – 258, 415 – 428 ’13 ] Expression of C9orf72’s mRNA in frontotemporal dementia/als (FTD/ALS) patients is reduced by 50%, and the expanded repeat and neighboring CgP islands are hypermethylated consistent with transcriptional silencing. Also the cytoplasmic aggregates staining positively for P62 appear to result from protein translation through the hexanucleotide repeat.

This work used induced pluripotent stem cells (iPSCs) derived from C9ALS/FTD patients. They show decreased C9orf72 mRNA, nuclear and cytoplasmic GGGGCC RNA foci, and expression of one RAN product (Gly Pro dipeptide). Neurons derived from the iPSCs also show enhanced sensitvity to glutamic acid excitotoxicity, and a transcriptional profile that ‘partially’ overlaps with transcriptional changes seen in iPSC neurons derived from mutant SOD1 ALS patients.

In addition, some 19 proteins were found which associate with the GGGGCC repeats in vitro. ADARB2 does this and participates in RNA editing.

ASOs (AntiSense OIigonucleotides ??) were used to suppress C9orf72 RNA expression. This led to reversal in many of the phenotypes of the iPSC neurons (suppression of glutamic acid toxicity, reduction in RNA foci formation). This implies that the GGGGCC repeats trigger toxicity through a gain of function mechanism. [ Proc. Natl. Acad. Sci. vol. 110 pp. E4530 – E4539 ’13 ] Nuclear RNA foci containing GGGCC in patient cells (wbc’s fibroblasts, glia, neurons) were ssen in patients with repeat expansion. The Foci weren’t present in sporadic ALS or ALS/FTD caused by other mutations (SOD1, TDP43, tau), Parkinsonism, or nonNeurological controls. Antisense oligonucleotides reduced the GGGGCC containing nuclear foci without alteraling overall C9orf72 RNA levels. SiNRAS didn’t work.

The Rx was applied to living mice and it was well tolerated.

[ Proc. Natl. Acad. Sci. vol. 110 pp E4968 – E4977 ’13 ] C9orf72 antisense transcripts are elevated in the brains of those with the expansion. Repeat expansion GGCCCC RNAs accumulate in nuclear foci in the brain. Sense and antisense foci accumulate in the blood and are potential biomarkers. RAN translation occurs in BOTH sense and antisense expansion transcripts — so all 6 proteins described above are made. The proteins accumulate in cytoplasmic aggregates in affected brain regions (e.g. frontal and motor cortex, spinal cord neurons).

[ Nature vol. 507 pp. 175 – 177, 195 – 200 ’14 ] C9orf72 has repeated hexanucleotide units (GGGGCC). Two or more G quartets stacked on top of one another form a G-quadruplex. In the expanded repeats of C9orf72 in ALS and frontotemporal dementia, stable quadruplexes form in DNA as well as the RNA transcribed from it.

Sequences which can form G-quadruplexes are conserved during evolution, so they presumably are doing something useful. They are found in transcriptional start sites. This work shows that G-quadruplex assembly in DNA increases transcriptional pauses in the expanded repeat (unsurprising). Also the G-quadruplexes in C9orf72 DNA promote the formation of stable R-loops — triple stranded structures that assemble when a newly form RNA transcript exiting RNA polymerase II invades the double helix and binds to one DNA strand, displacing the other. If the R-loops aren’t resolved, they can halt transcriptional elongation.

Not only that, but abortive GGGGCC containing RNAs accumulate in the spinal cord and motor cortex of patients with the expanded repeats. The RNAs are truncated in the GGGGCC region, and the amount is linearly proportional to the length of the hexanucleotide repeat. This explains how they could accumulate along with decreased level of full length C9orf72 mRNA (and presumably the protein made from it).

A ‘few dozen’ proteins binding the GGGGCC repeats have been found. One of them is nucleolin, involved in the formation of the ribosome within the nucleolus It is mislocalized to RNA foci in neurons of the motor cortex of patients with C9orf72 related disease. The lack of mature ribosomes results in the buildup of untranslated mRNA in the cytoplasm.

[ Science vol. 345 pp. 1118 – 1119, 1139 – 1145, 1192 – 1194 ’14 ] Normally the number of GGGGCC repeats in C9orf72 ranges from 2 to 23, with hundreds or even thousands of copies in the disease range. Possibilities
l. Interference with C9orf72 expression — e. g. loss of function
2. Sponging up RNA binding proteins by the transcript
3. Repeat associated non-ATG translation (RAN translation) in all reading frames (sense and antisense).

A series of stop codons in both the sense and antisense RNAs was engineered every 12 repeats, stopping formation of the dipeptide repeat proteins. The new RNAs still formed the G-quadruplexes, and both RNAs formed RNA foci when expressed in cultured neurons.

Putting them into Drosophila showed that the pure repeats able to form dipeptides causing degeneration in the fly eye, while the interrupted constructs (producing RNA only) did not. The same was true when expressed in the nervous systems of adult flies. Blocking translation of the RNA partially suppressed the phenotype.

There are 5 possible dipeptide products of RAN of GGGGCC (GA, GP, PA, GR, PR — G == Glycine, P == Proline, A == Alanine, R = Arginine). Then RNAs using alternate codons for the dipeptides were used (so GGGGCC wasn’t present). Expressing Glycine Arginine (GR) or Proline Arginine (PR) was toxic, Glycine Alanine showing ‘some’ toxicity later in life.

Some RNA binding proteins containing low complexity sequences (aka prion-like domains) — these are FUS, EWSR1, TAF14, hnRNPA2 — form polymeric assemblies, which incorporate into hydrogels in vitro. The assemblies are similar to RNA granules. Many of the RNA binding proteins associating with hydrogels hare serine arginine (SR) sequences. The SR domain proteins are regulated by phosphorylation on serine, also controlling the association with hydrogels. It is hypothesized that the GR and PR transcripts associate with hydrogels (or similar assemblies such as RNA granules), but are impervious to the regulatory action of the kinases (no serine to phosphorylate), so they might clog up the trafficking of SR domain containing RNA binding proteins moving in an out of the granules to transfer information throughout the cell.

[ Neuron vol 84 pp. 1213 – 1225 ’14 ] Proline Arginine dipeptides are neurotoxic. They form aggregates in nucleoli in experimental systems. Nuclear aggregates were also found in postmortem spinal cord from C9ORF72 ALS and ALS/FTD patients. Intronic GGGGCC transcripts are also toxic. Repeat associated non-ATG translation (RAN translation) is thought to depend on RNA hairpin structures using GC pairing.

[ Cell vol. 158 pp. 967 ’14 (abstract of something to appear in Science) ] Peptide translated from GGGGCC expansions containng arginines (Gly Arg and Pro Arg) are harmful — 3 other dipeptide repeats are harmless. The peptides bind to nucleoi and impede RNA biogenesis. Interestingly Ser-Arg repeats proteins (SR proteins) are important in RNA splicing. The GlyARG and PROARG repeat peptides alter splicing of the amino acid transporter EAAT2, similar to that seen in ALS. Interestingly, the peptides are readily taken up by cells in culture, translocating to the nucleus.

Also a small molecule has been developed which targets GGGGCC RNA expansions. It inhibits translation of the dipeptide repeat proteins from the expansions (see Science vol. 353 pp. 64 ****

GlyPro in CSF is a biomarker of ALS patients with the C9orf7s expansion.

The normal function of C9orf72 isn’t known. It is structurally related to DENN (Differentially Expressed in Normal and Neoplastic cells) proteins, which are GDP/GTP exchange factors for Rab GTPases.

At this point it isn’t known if the proteins generated by RAN are toxic. The protein inclusions are present in unaffected areas of the brain (lateral geniculate) as well as the vulnerable areas (cortex, hippocampus).

The initiation of RNA translation is thought to depend on RNA hairpin structures which use C:G complementary pairing. CAG (but not CAA) repeats undergo RAN translation. Protein aggregates occured only in brain intestes despite the fact that C9orf72 is expressed all over the body (but expression is highest in brain).

It is possible that antisense RNA could be formed from the opposite strand (e.g. CCCCGG) giving poly pro-ala, poly pro-gly and poly pro-arg.

[ Science vol. 1106 – 1112 ’15 ] Just expressing 66 GGGGCC repeats without an ATG start codon using an AdenoAssociated Virus (AAV) vector in mice was enough to produce neurodegeneration with RNA foci, inclusoins of poly QP, GA and GR and TDP43 pathology. There was cortical neuron and cerebellar Purkinje cell loss and gliosis.

[ Nature vol. 525 pp. 36 – 37, 56 – 61, 129 – 133 ’15 ] (GGGGCC)30 was expressed in the Drosophila eye. This leads to the rough eye trait and is easily scored, allowing you to look at the effect of other genes on it. Mutations activating RanGAP suppressed rough eyes. RanGAP binds to GGGGCC on the cytoplasmic face of the nuclear pore. Enhancing nuclear import or suppressing nuclear export of proteins also suppressed neurodegeneration. RanGAP physically interacts with the GGGGCC Hexanucleotide Repeat Expansion resulting in its mislocalization. The mislocalization is found in neurons derived from iPSCs from a patient with C9orf72 type ALS, and also in brain tissue from other patients with C9orf72 ALS.

Nuclear import is impaired due to HRE expression (fly and iPSC derived neurons). The defects can be ‘rescued’ by small molecules and antisense oligonucleotides targeting the HRE G-quadruplexes. This may actually be a way to Rx ALS ! ! ! !

Another paper crossed (GGGGCC)58 flies with missing chromosomal segments. They found a variety of nuclear import factors whose inactivation worsened rough eye.

Expression of constructs of in GGGGCC)8, 28 and 58 lacking an AUG start codon in Drosophila was done. The constructs could only produce Repeat Associated NonAUG translation products (e.g. dipeptides). The dipeptides disrupt nuclear import of fluorescent test substrates and of normal nuclear proteins (notably TDP43). In addition RNA export from the nucleus is also compromised. The deleterious effects could be modified by 18 genetic regions (found by large scale unbiased genetic screening). THey coded for components of the nuclear pore complex, nuclear RNA export machinery and nuclear import.

Dipeptides produced from GGGGCC and GGGGCCn’s disrupt the nucleolus, so this may be an additional cause of repeat toxicity.

[ Neuron vol. 88 pp. 892 – 901 ’15 ] A mouse model containng the full human C9orf72 repeat which was either normal (15 repeats) or expanded (100 – 1,000 repeats) — using bacterial artificial chromosomes (BACs) — thes mice are called C9-BACexpanded. They show widespread RNA foci and RAN translated dipeptides. Nucleolin distribution was altered. However the mice showed normal behavior and there was no neurodegenration. This is surprising.

[ Nature vol. 535 p. 327’16 (abstr. of Sci. Transl. Med ’16) ] Mice with mutations diminishing or eliminating the function of C9ORF72 (unknown as of 8/13) developed autoimmune disease.

[ Science vol. 351 pp. 1324 – 1329 ’16 ] Two independent mouse lines lacking the ortholog of C9orf72 (3110043021Rik) in all tissues developed normally and aged without any motor neuron disease. Instead they developed progressive splenomegaly and lymphadenopathy with accumulation of engorged macrophagelike cells. There was age related neuroInflammation similar to C9orf72 ALS but not sporadic ALS. There was no evidence of neurodegeneration however.

[ Neuron vol. 90 pp. 427 – 430, 531 -534, 535 – 550 ’16 ] BAC transgenic mice using patient derived gene constructs expressing (some of? all of?) C9ORF72 are reported.

A germline knockout develops blood abnormalities (splenomegaly, lymphadenopathy and premature death). The data conflict on which of the 5 products of RAN (Repeat Associated NonATG) translation are the most toxic (GP, GA, GR, PA, PA, PR).

In this study, mice with increased levels of repeats (up to 450) showed no evidence of motor neuron disease, and the brain was normal. They at least did have some trouble with cognition.

THe second study put in the full C9 gene with 5′ and 3′ flanking sequences. 4 lines of transgenics with repeats ranging from 37 to 500 were characterized. These mice did have peirpheral and central neurodegeneration, with motor deficits. There was a decrease in cortical neurons, Purkinje cells. This is the first time any transgenic has shown neurodegeneration. The deficits are reversible with antisense oligonucleotides. There was a disparity in disease expression between male and female mice.

RNA foci and DPR (DiPeptide Repeat) proteins don’t accumulate in the most affected brain regions.

[ Science vol. 353 pp. 647 – 648, 708 – 712 ’16 ] Spt4 is a highly conserved transcription elongation factor which regulates RNA polymerase II processivity (along with its binding partner Spt5). Spt4 is required to transcribe long trinucleotide repeats found in open reading frames, or in non protein coding regions of DNA templates (in S. cerevisiae). Mutations of Spt4 decrease synthesis of (and restored enzymatic activity to) expanded polyQ proteins (in yeast) without affecting genes lacking the excessive CAG repeats. It might also work in nonCAG repeats.

Targeting Spt4 (with antiSense oligonucleotides) reduces production of the C9orf72 expansion associated RNA and protein, and helps neurodegeneration in model systems. Repeat expansions are transcribed in both the sense and antisense directions. Yeast Spt4 (human homolog SUPT4H) is a small evolutionarily conserved zinc finger protein which forms a complex with Spt5, which then binds to RNA polymerase II regulating transcription elongation (pol II processivity).

DRB is a RNA polymerase II inhibitor. The complex of Spt4 and Spt5 homologs in man (SUPT4H, SUPT5H) is called DSIF (DRB Sensitivity Inducing Factor)

Depletion of Spt4 or its binding partner (Spt5 ) decreases the number of both sense and antisense repeat transcripts and RNA foci. One of the 6 RAN translation products (polyGlyPro) is substantially reduced by Spt4 depletion.

The study was in human c9ALS fibroblasts. However, side effects are certainly possible — in addition to decreasing the expression of C9ORF72, 95% depletion of SUPT4H1 altered (how?) the expression of another 300 genes. In mice deletion of both copies of SUPT4 is embryonic lethal, but deleting one produced no effects up to 18 months of age.

Time for drug chemists to go to the Multiplex

30 – 40% of all the drugs currently in clinical use are thought to target G Protein Coupled Receptors (GPCRs). Just how many GPCRs inhabit our genome isn’t clear. The latest estimate is 850 which is 4.2% of the 20,077 annotated protein genes we have. That being the case, it behooves drug chemists to know everything about them and how they work.

A recent paper [ Cell vol. 166 pp. 907 – 919 ’16 ] shows that a lot of the old thinking about GPCRs is wrong. Binding of a ligand to a GCPR results in a conformational change in its 7 transmembrane segments, so that the parts inside the cell bind to a heterotrimer of proteins which bind (and hydrolyze) GTP — this is the G protein. So far so good. The trimer splits up into its 3 constituents, unimaginatively called alpha, beta and gamma, each of which can act as a messenger that a ligand from outside the cell has landed on a GPCR, binding to other proteins causing all sorts of effects (e.g. can act as a second messenger)

All good things must end, and termination of GPCR signaling was thought to involve phosphorylation of the intracellular segment of the GPCR, binding of another protein (betaArrestin), removal from the cell membrane (so it can no longer bind its extracellular ligand) and then either destruction or recycling back to the cell membrane. So the old paradigm was betaArrestin binding equals the end of signaling.

It was thought that betaArrestin and the G protein competed for binding to the same intracellular amino acids of the GPCR. Not so says this paper. For some GPCRs both can bind, and signaling can continue, even though the complex of GPCR, G protein and betaArrestin is now inside the cell in an endosome. The complex is called the Multiplex. The examples given are GPCRs for parathyroid hormone (PTH) and Thyroid Stimulating Hormone (TSH). Blurry pictures are given of the complex. GPCRs have been divided into several classes and GPCRs for TSH and PTH are class B GPCRs — which contain a long phosphorylatable tail in the cytoplasm. The G protein binds to these GPCRs by its core region, while betaArrestin binds to the tail. Signaling continues apace.

From the newspaper of record

Sorry, nothing earthshaking scientifically to write about, so here are 3 leads from today’s New York Times for your enjoyment

l. Page 1 top left — “The Failing Inside Mission to stop Hillary from Lying”
2. Page 1 top right — “How Trump plans to use Obama’s Embrace of Executive power”
3. Sunday review –Page 1 bottom “Hillary is making America more comfortable with pay to play”

Fair and balanced as Fox would say if they wrote them.

But they didn’t. It was the Times masquerading two opinion pieces as news on page 1, and an actual opinion piece in the Review.

Well, the titles were a bit different.

1. Page 1 top left — “The Failing Inside Mission to Tame Trump’s Tongue”

2. Page 1 top right — “How the President Came to Embrace Executive Power”

3. Sunday Review — page 1 bottom “Trump is Making America Meaner”

#2 is particularly interesting, as it is basically an excuse for ruling by decree, the dream of the left. The apologia comes in the third paragraph — “Blocked for most of his presidency by Congress, Mr. Obama has sought to act however he could.” So much for the constitution.

Ruling by decree has always been a goal of left utopians and pragmatists. Go back to the great Serge Eisenstein movie about Ivan the Terrible (part I 1944). It was commissioned by Stalin, and it’s all wonderful propaganda for the leader to do what he wishes unimpeded. Smash the Boyars. Let Ivan be Ivan. Only he truly loves the country. The second part wasn’t released until 1958 as Stalin didn’t like it. It’s really handy to rule by decree.

Well Maduro in Venezuela is currently ruling by decree, as has Fidel for years.


Nothing exciting enough scientifically to post about this week, so enjoy this. “We have enough neurologists in the Air Force, please resign your commission as you will be assigned to the Army.” This was followed in May 1968 by an Army preference for assignment form so out of date, that if I extended my two year tour to four I could take my family with me to Vietnam. This at a time when we had 500,000 troops over there.

What to do? What I did was residency by day, and Scotch and Faulkner by night, while looking at my widow to be and two orphans to be.

If the world has you down, and if you think your life is hard, it’s time to read about Yoknapatawpha county, and life when it was really hard. It’s practically biblical. I’d start with “The Hamlet”, and continue through the trilogy. Well over half of Faulkner’s work takes place here, so the rest will make sense.

I don’t read novels anymore having seen far more of life as a doc than some pup half my age. Faulkner is different. It has the ring of truth.

He had eyes the color of stagnant water, and other such delights await you.

Trigger warning — the N word appears prominently. Even so after reading him, you’ll never be upset by microaggressions again.

Here’s how Faulkner introduces the place.

“The people .. came from the northeast, through the Tennessee mountains by stages marked by the bearing and raising of a generation of children. They came from the Atlantic seaboard and before that from England and the Scottish and Welsh Marches … They brought no slaves and no Phyfe and Chippendale highboys; indeed, what they did bring most of them could (and did) carry in their hands. They took up land and built one- and two-room cabins and never painted them, and married on another and produced children and added other rooms one by one to the original cabins and did not paint them either, but that was all. Their descendants still planted cotton in the bottom land and corn along the edge of the hills and in the secret coves in the hills made whiskey of the corn and sold what they did not drink. Federal officers went into the country and vanished. … County officers did not bother them at all save in the heel of election years. They supported their own churches and schools, they married committed infrequent adulteries and more frequent homicides among themselves and were their own courts, judges and executioners. They were Protestants and Democrats and prolific; there was not one Negro landowner in the entire section. Strange Negroes would absolutely refuse to pass through it after dark.”

Exactly the way I felt that September driving through Meridian Mississippi with Pennsylvania plates on my car on my way into the Air Force and I’m not Black.

What is a hormone? What is an endocrine organ?

We all knew what hormones were back in the day. They were chemicals released by an endocrine gland into the blood where they went everywhere and affected distant organs. The classic example were the sex hormones (estrogen, progesterone, testosterone) eleased by the gonads affecting the reproductive organs, and not least the brain.

Things have changed mightily, and just about every tissue in the body does this now. There are at least 20 adipokines released by fat — examples are adiponectin, adipsin, and of course leptin. Muscle may be also getting into the act with irisin (although that is controversial). Other muscle produced hormones (myokines)  include atrial natriuretic peptide released by the heart and skeletal muscle releases at least 8 more.

There is even more stuff released into local tissue fluids which don’t get into the blood so they aren’t hormones. You can regard all neurotransmission this way. Paracrines are compounds which act only on cells close to them (because they don’t get into the blood). Examples include the huge class of prostaglandins and polypeptide growth factors such as the 22 member fibroblast growth factor family.

What to make of [ Cell vol. 166 pp. 424 – 435 ’16 ] which describes PM20D1 (Peptidase M20 Domain containing 1) which is secreted by fat cells. It’s an enzyme which builds compounds from substances already in the blood. The chemistry is simplicity itself — it takes a long chain fatty acid and an amino acid and forms the fatty acid amide — or an N-acyl amino acid.

What does the product do? It causes uncoupling of oxidative phosphorylation by mitochondria, so it just produces heat (something useful to an animal in the cold). Administration of N-acyl amino acids to mice increases energy expenditure and improves glucose metabolism. It’s possible that they could be used therapeutically.

Another example of how little we knew about what is going on inside us.

Yes it’s hot, but

A few years ago, before things calmed down, hurricanes were predicted to become more frequent and more severe. So although global warming fans predicted higher temperatures, they also predicted this. Here’s an example
and another —

So far this year (and the major part of the hurricane season is about to begin) the Atlantic hurricane season is the quietest it’s ever been. There have only been 4 tropical storms this year and no hurricanes at all not even one. One of the 4 storms occurred in January, which is rather bizarre.

Scientific theories when faced with a falsified prediction are usually modified or abandoned.  Not so with global warming.  It’s just been rebranded as climate change.

There is a rather imperfect measure of the amount of power produced by the storms of the season, called Accumulated Cyclonic Energy (ACE). The average per year is an ACE of 110. This year (so far) it’s only 6.

ACE is calculated as the square of the wind speed every 6 hours, and is then scaled by a factor of 10,000 for usability. The ACE of a season is the sum of the ACE for each storm and takes into account the number, strength, and duration of all the tropical storms in the season. The caveat to using ACE as a measure of the activity of a season is that it does not take the physical size of the hurricane or tropical storm into account which is why it’s imperfect.

If you know any physics, ACE is velocity squared * time — which is not the dimension of energy (it’s acceleration). I wonder if satellite radar is good enough to give us ground windspeed over small areas, which could be summed over the hurricane area if the division was fine enough. This would allow us to tell big storms from smaller ones. Unfortunately, there would be no way to compare this new measure to ones in the past such as ACE.

You are alive because the lipid bilayer of your plasma membrane is asymmetric

You are an organism with trillions of cells. A mosquito bit you depositing millions of viruses in your tissues. The virus can reproduce only within one of your cells and it has exploited all sorts of protein protein chemistry to get in. Antibodies (if you are fortunate enough to have them) can get rid of the extracellular critters. However, 500,000 have made into the same number of your cells, and are merrily trying to reproduce.

How does the asymmetry of the lipid bilayer of your plasma membrane help you survive. If each virus infected cell killed itself before the virus reproduced, you’d survive. Although 500,000 is a large number is is less than 1 millionth of your cell total.

Well you do have intracellular defenses against viruses, called the innate immune system. One of them is a protein with the ugly name of gasdermin D. The activated innate immune system (in the form of inflammatory caspases) cleaves gasdermin. This breaks up the inhibition of the amino terminal part of gasdermin by the carboxy terminal part giving a fragment which binds to one particular membrane component (phosphatidyl serine) which makes up 20% of the inner leaflet of the cell membrane. It then forms a large diameter (to a cell 140 Angstroms is quite large) pore in the cell membrane. No cell can survive this, so it dies, releasing cellular contents (probably some viral components but not fully formed one). For details see [ Nature vol. 535 pp 111 – 116, 153 – 158 ’16 ]

Wait a minute. The toxic gasdermin fragment is also released. So how come it doesn’t kill everything in sight? Because our cellular membranes keep phosphatidyl serine confined to the inner membrane, normal cells don’t show it on their exterior, so they can be bathed in gasdermin with no ill effect. What is responsible for this asymmetry — believe it or not an ATP consuming enzyme called flippase (about this more later) which takes any phosphatidyl serine it finds on the outer leaflet and schleps it back inside the cell.

There is all sorts of elegant chemistry which explains just how gasdermin binds to phosphatidyl serine and none of the many other phospholipids found on the inner leaflet. There is more elegant chemistry explaining how flippase works (see later).

What chemistry cannot explain, is why organisms would ‘want’ an asymmetric membrane. As soon as you get into the function of a particular compound in an organism, chemistry is powerless to tell you why. Nothing else can explain how a given molecule does what it does on the molecular level but that is not enough for a satisfying explanation.

One further explanation before some hard core cellular biochemistry follows (after ***). Our cells are dying all the time. The lining of your gut is replaced every 5 days. Even the longest lasting element of your blood is gone after half a year, and most other elements are turned over at least once a month. When these cells die, they must be cleaned up, without undue fuss (such as inflammation). The cleaners are cells called macrophages. A dying cell releases chemical signals, actually called ‘eat me’, one of which is phosphatidyl serine found on the membrane fragments of a dead cell. The fact that flippases keep it on the inner leaflet means that macrophages won’t attack a normal cell.

Slick isn’t it?


Flippase is a MgATPdependent aminophospholipid translocase. It localizes phosphatidylserine and phosphatidylethanolamine to the inner membrane leaflet by rapidly translocating them from the outer to the inner leaflet against an electrochemical gradient. The stoichiometry between amino phospholipid translocation and ATP hydrolysis is close to one (how will the cell have enough ATP to do anything else?). The flippase is inhibited by high calcium, and by pseudosubstrates such as vanadate, acetylphosphate and para-nitrophenyl phosphate, and by SH reactive reagents such as N-ethylmaleimide and pyridyldithioethylamine (PDA) a specific inhibitor of phospholipid translocation

[ Proc. Natl. Acad. Sci. vol. 109 pp. 1449 – 1454 ’12 ] P4-ATPases are a subfamily of P-type ATPases. They transport aminophospholipids from the exoplasmic to the cytoplasmic leaflet (and are known as flippases). Man has 14 P4-ATPases, expressed in various cell types. They are thought to be similar to the catalytic subunits of the Ca++ ATPase, and the Na, K ATPase, consisting of cytoplasmic, N, P and A domains and a membrane domain made of 10 transmembrane helices (M1 – M10).

[ Proc. Natl. Acad. Sci. vol. 111 pp. E1334 – E1343 ’14 ] The P4-ATPases are thought to resemble the classic P-type ATPase cation pumps — a transmembrane domain of 10 helices and 3 cytoplasmic domains (P for phosphorylation, N for nucleotide binding and A for actuator). ATP8A2 forms an intermediate phosphorylated on aspartic acid (E2P)and undergoes a catalytic cycle similar to the sodium pump (Na+, K+ ATPase). Dephosphorylation of E2P is activated by the transported substrates phosphatidyl serine (PS) and phosphatidyl ethanolamine (PE), similar to the K+ activation of dephosphorylation in the sodium pump.

PE and PS are 10x as large as the cations transported by the sodium pump. This is known as the giant substrate problem. This work shows that isoleucine #364 (mutated in — patients with the ataxia, retardation and dysequilibrium syndrome Eur. J. Hum. Genet. vol. 21 pp. 281 – 285 ’13 aka CAMRQ syndrome ) forms a hydrophobic gate separating the entry and exit sites of PS. I364 likely directs the sequential formation and annihilation of water filled cavities (as shown by molecular dynamics simulations) allowing transport of the hydrophilic phospholipid head group, in a groove outlined by TMs 1, 2, 4 and 6, with the hydrocarbon chains following passively, still in the membrane lipid phase (and presumably outside the channel) — this must disrupt the hell out of the protein as it passes. They call this the credit card model — only the interaction with part of the molecule is important — just as the magnetic stripe is the only important thing about the credit card.

Functional MRI research is a scientific sewer

First a primer about the science underlying functional Magnetic Resonance Imaging (fMRI). Chemists use MRI all the time, but they call it Nuclear Magnetic Resonance. Docs and researchers quickly changed the name to MRI because no one would put their head in something with Nuclear in the name.

There are now noninvasive methods to study brain activity in man. The most prominent one is called BOLD (Blood Oxygen Level Dependent), and is based on the fact that blood flow increases way past what is needed with increased brain activity. This was actually noted by Wilder Penfield operating on the brain for epilepsy in the 1930s. When a patient had a seizure on the operating table (they could keep things under control by partially paralyzing the patient with curare) the veins in the area producing the seizure turned red. Recall that oxygenated blood is red while the deoxygenated blood in veins is darker and somewhat blue. This implied that more blood was getting to the convulsing area than it could use.

BOLD depends on slight differences in the way oxygenated hemoglobin and deoxygenated hemoglobin interact with the magnetic field used in magnetic resonance imaging (MRI). The technique has had a rather checkered history, because very small differences must be measured, and there is lots of manipulation of the raw data (never seen in papers) to be done. 10 years ago functional magnetic imaging (fMRI) was called pseudocolor phrenology.

Some sort of task or sensory stimulus is given and the parts of the brain showing increased hemoglobin + oxygen are mapped out. As a neurologist as far back as the 90s, I was naturally interested in this work. Very quickly, I smelled a rat. The authors of all the papers always seemed to confirm their initial hunch about which areas of the brain were involved in whatever they were studying. Science just isn’t like that. Look at any issue of Nature or Science and see how many results were unexpected. Results were largely unreproducible. It got so bad that an article in Science 2 August ’02 p. 749 stated that neuroimaging (e.g. functional MRI) has a reputation for producing “pretty pictures” but not replicable data. It has been characterized as pseudocolor phrenology (or words to that effect). Keep reading you’re about to find out just why this was.

What was going on? The data was never actually shown, just the authors’ manipulation of it. Acquiring the data is quite tricky — the slightest head movement alters the MRI pattern. Also the difference in NMR signal between hemoglobin without oxygen and hemoglobin with oxygen is small (only 1 – 2%). Since the technique involves subtracting two data sets for the same brain region, this doubles the error.

Under two years ago, it was shown that 70% of people having functional MRIs (fMRIs) were asleep during the test, and that until then fMRI researchers hadn’t checked for it. For details please see You don’t have to go to med school, to know that the brain functions quite differently in wake and sleep.

Recent work shows that functional MRI work is even worse. A devastating report in [ Proc. Natl. Acad. Sci. vol. 113 pp. 7699 – 7600, 7900 – 7905 ’16 ] showed that certain common settings in 3 software pacakages (SPM, FSL, AFNI) used to analyze fMRI data gave false positive results ‘up to’ 70% of the time. Some 3,500 of the 40,000 fMRI studies in the literature over the past 20 years used these settings. The paper also notes that a bug (now corrected after being used for 15 years) in one of them also led to false positive results.

Here’s a bit more detail on what they did. It turns out that analyzing one voxel (essentially a single MRI pixel) at a time produces valid results. The problem comes when multiple voxels (clusters) are analyzed together. Clusterwise inference considers both the strength of activity at spots throughout the brain as well as the size of the spots. When a parameter called the cluster defining threshold (CDT) is set too low, the analysis is more likely to be false positive. This was true for all 3 packages tested. Parametric statistical methods produce the problem (not for voxels but for clusters). It relies on Gaussian Random Field Theory (RFT) for clusters , which depends on two other assumptions (1) the spatial autocorrelation function has a squared exponential shape — e.g. Gaussian (2) the spatial smoothness of the fMRI signal is constant over the brain. Neither of these assumptions is correct. Those of you who’ve read Nassim Nicholas Taleb about the stock market know about ‘fat tails’. It turns out that the spatial correlation function has them. Here’s what a fat tail is all about. Human height goes fall quite nicely into a Gaussian distribution. There are 7 and 8 footers about but they are rare. If the human height distribution wasn’t Gaussian but had a fat tail, we’d see 12 and 15 footers.

If that wasn’t bad enough,the following is even worse (in my opinion). 40% of 241 recent fMRI studies didn’t report using well known methods for correcting for multiple testing. They may have done so, but every biomedical paper, and drug study says so explicitly. Not only that but drug studies are required to explicitly state the hypothesis (or hypotheses) they are testing.

This is probably why in the early days, fMRI researchers always confirmed their original hypothesis. They could test the massive fMRI for statistical rarity, and since the data was so large, find it and post hoc propter hoc publish it. Possibly they did so out of ignorance, but even so this is inexcusable