Tuesday, March 10, 2009

Citations for Dr Belmonte

I posted a comment at The Autism Crisis which referred to a recent talk given by Dr Matthew Belmonte. Ms Dawson felt the comments were off topic and so I have moved the discussion to my blog.

I am still in the process of trying to understand the comment that Dr Belmonte left, in particular the concept of the veridical percept. When I do understand it I will post a reply.

List of citations for Dr Belmonte. They are listed in approximate order of relevance. Also posted at


I have added some commentary. I have tried to use as few citations as possible. This is more to show that NO physiology is sufficiently complex to support the many complex physiological effects observed in autism and ASDs.

Thomas DD, Ridnour LA, Isenberg JS, Flores-Santana W, Switzer CH, Donzelli S, Hussain P, Vecoli C, Paolocci N, Ambs S, Colton CA, Harris CC, Roberts DD, Wink DA. The chemical biology of nitric oxide: implications in cellular signaling. Free Radic Biol Med. 2008 Jul 1;45(1):18-31. PubMed link

This is quite a good review paper on signaling by NO. Dave Wink is a senior researcher in the field, he just chaired the Gordon Conference on NO last week. Most of the higher concentration effects are relevant for metabolic physiology (and his specialty cancer), not so much neurological effects. The levels that are most relevant for neuronal effects are in the 1-30 nM/L range (0.03 to 0.9 ppb by weight). That is the basal level is around 1 nM/L up to a few times the EC50 of sGC in cells (~10 nM/L). There are no techniques to measure those levels in vivo on the time and length scales that are known to be important. We know that levels can't be that high (10 nM/L) long term in the endothelium because if they were, there would be systemic hypotension (as in septic shock). This is the first review that really puts together all the different concentration scales that NO is important at.

Because NO sources and sinks are small, and NO is highly diffusible, there are large gradients in NO concentration. Because NO is ~10x more soluble in isotropic lipid, lipid membranes have a large effect on the diffusion. Many lipid membranes are not isotropic. What effect that has on NO diffusion has not been investigated.

Sources are important and sinks are important. Superoxide is an important sink (depending on its concentration). I have a slight quibble (actually very slight). They mention that SOD levels are high enough that superoxide shouldn't affect NO levels in the nM/L level. If everything were uniform and isotropic I would agree, but SOD levels are not isotropic, and superoxide is usually generated vectorally to the inside of vesicles, mitochondria and microsomes. Superoxide is confined by lipid membranes so levels can (in theory) get high inside of those vesicles. Superoxide is an anion, so it will be affected by local electric fields and be held up against the inside of the lipid membrane that is confining it. The SOD would be floating around in the bulk. NO is ~10x more soluble in the lipid membrane, so depending on the precise locations of the NO source, the superoxide source, the SOD, the local electric field and the lipid volumes in the vicinity the effects could be complex.

The major take home message is that because NO can diffuse everywhere, and each NO sensor only senses the sum of NO from all sources, all sources and all sinks matter, including the basal level. The basal level is more important the lower the levels are, the lowest levels mediate things acutely through sGC, as in the brain. There can be effects from lower NO levels which don't activate sGC, for example when NO combines with superoxide and forms peroxynitrite and nitrates proteins. These proteins can accumulate NOx over long periods and integrate a NO/ROS signal over that time. I think this is important in regulating mitochondria number in neurons where mitochondria biogenesis (regulated by NO) must be matched to mitochondria need for the entire neuron (which can vary by 3 or more orders of magnitude depending on axon length). My hypothesis is that mitochondria accumulate nitrated proteins over their metabolic lifetime and that integrates the metabolic load on those mitochondria. When those mitochondria are recycled via autophagy, the NOx is released and produces a NO signal that triggers the biogenesis of the appropriate number of mitochondria to support the needed metabolic load.

Garthwaite J. Concepts of neural nitric oxide-mediated transmission. Eur J Neurosci. 2008 Jun;27(11):2783-802. PubMed link

This is a good review on neuronal signaling via NO, pointing out that NO does cause both LTP and LTD and that the quantities needed for activity are very small in the nM/L range. That is the same level that activates sGC and causes vasodilation.

Many neuropeptides and growth factors, when they are endocytosed by their receptor, activate nitric oxide synthase, and the receptor/NOS complex is conveyed back to the cell body while it is producing NO. I suspect that the NO that the endocytosed receptor complex produces is important in the LTP produced by many of these growth factors as that NO from inside the axon adds to the local NO on the outside. Myelin is transparent to NO, but the alternating layers of lipid and aqueous would certainly modulate its diffusion in complex ways.

NOS binds to the serotonin transporter which couples NO release to serotonin uptake.

Wang S, Paton JF, Kasparov S. Differential sensitivity of excitatory and inhibitory synaptic transmission to modulation by nitric oxide in rat nucleus tractus solitarii. Exp Physiol. 2007 Mar;92(2):371-82. PubMed link

This paper shows changes in membrane potential following exposure to ~ nM/L levels of NO. My conceptualization is that the NO release that causes the vasodilation observed in fMRI BOLD also changes the sensitivity of neurons in that affected volume element to be triggered.

It has been shown (Goense & Logothetis, 2008) that there can be vasodilation without neuronal activation. But we already knew that because even in a volume element that is observed to be highly activated, not every neuron is in a state of discharge. The NO primes the neurons to discharge, but which ones do discharge depends on which ones the action potentials propagate into.

Krause DN, Duckles SP, Pelligrino DA. Influence of sex steroid hormones on cerebrovascular function. J Appl Physiol. 2006 Oct;101(4):1252-61. PubMed link

This paper is quite interesting because it relates to the "extreme male brain" hypothesis of ASDs. Testosterone decreases NO levels and estrogen increases NO levels. This paper relates that primarily to blood flow in adults, I think the effects on fetal neurodevelopment in utero would be important. NO inhibits Leydig cell production of testosterone, so low NO causes high testosterone levels. High testosterone levels cause a more hirsute phenotype, which expands the niche for the bacteria I am studying, causing increased NO/NOx production by them, exerting feedback control on androgen synthesis. My hypothesis is that the association of in utero testosterone levels with ASDs and ASD-like behaviors is real, but that both of them are secondary to low NO status. Many conditions associated with hyperandrogenic effects in adults are also associated with low NO, as in polycystic ovarian syndrome. Stress causes low NO and stress tends to cause high androgen levels.

I think a better characterization of the autism would be as the "extremely stressed brain" because most of the differences are not gender specific.

In utero levels of testosterone and pathways by which that can affect neurodevelopment lead into the next papers, on epigenetic programming of adult physiology.

Racasan S, Braam B, Koomans HA, Joles JA. Programming blood pressure in adult SHR by shifting perinatal balance of NO and reactive oxygen species toward NO: the inverted Barker phenomenon. Am J Physiol Renal Physiol. 2005 Apr;288(4):F626-36. Epub 2004 Nov 16. Review. PubMed link

This paper relates to the epigenetic programming of adult physiology by NO/ROS balance in utero. The mechanism(s) for epigenetic programming by NO/ROS balance are not fully understood. The mechanisms for epigenetic modification (changes to DNA methylation for example) are mechanisms that could pertain in essentially every cell type and in essentially every tissue compartment. If NO/ROS mediated epigenetic programming modified adult function in some organs it would be surprising if that same mechanism(s) did not pertain in many organs in that the substrate (genomic DNA) is the same.

Stress in utero in animals is known to change adult brain size (some types of stress increase it) and adult behaviors. Stress in utero is also known to increase the incidence of ASDs. People with ASDs are exquisitely sensitive to stress. Chronic stress does increase acute sensitivity to stress even after the chronic stress is removed, even in adults.

MeCP2 deletion is known to cause autism-like symptoms (Rett Syndrome). MeCP2 codes for the gene that allows for differential expression of methylated DNA, DNA that is methylated as a consequence of the epigenetic programming of those cells. Presumably if aberrant readout of methylated DNA can "cause" autism-like symptoms, then sufficient changes to methylation of DNA at any step along the way could conceivably do similar things.

I don't think of Rett Syndrome as "autism", rather I see it as "autism-like". I see "autism" and the ASDs as being due to neuroanatomy and epigenetic programming of that neuroanatomy most of which occurs in utero (due to low NO (my hypothesis)). I see "autism-like" as being due to acute decreases in NO brought about by a variety of mechanisms. An important mechanism is epigenetic programming of the brain in utero to be in a lower NO state. Social isolation does this also, so can metabolic stress as in mitochondrial disorders (which generate lots of superoxide) and things like Rett Syndrome where there is metabolic stress due to tissue compartments being mosaic (with different MeCP2 status) resulting in cells not working "in sync". I think that raising NO levels can help "autism-like" symptoms, changing the neuroanatomy is (I think) not possible. I think that raising NO levels soon enough in neurodevelopment can switch an individual from development on an ASD trajectory to an NT trajectory (to some extent).

I think the acute resolution of autism symptoms with fever (Zimmerman 2007) is due to increased NO from iNOS. I have an extensive blog about what I see as the physiology behind it.

Godfrey KM, Barker DJ. Fetal nutrition and adult disease. Am J Clin Nutr. 2000 May;71(5 Suppl):1344S-52S. Review. PubMed link

See table 1, a list of 10 tissue compartments where there is evidence of in utero programming in humans. The physiology of many adult human organs is known to be epigenetically programmed in utero. It would be beyond surprising if the most important organ, the brain, was not.