Jerry Silver and Other Discussion from ChinaSCINet Update

[paolocipolla]

Paolo,

A transected spinal cord is extremely rare and not something that anybody can easily do a clinical trial of, except in very special centers. Why are you so concerned about transected spinal cords? Do you happen to know of somebody with a transected cord? Are you interested because of In Vivo Therapeutics? If they are planning a therapy that can only be done on people with transected spinal cords, that is not a very good business model. If they are planning to transect or remove part of the spinal cord of chronic patients to transplant some scaffolding, I have already indicated repeatedly that this would not be justifiable.

Dr. Henreich Cheng in Taiwan is the only doctor that I know who has a treatment that he has applied to transected spinal cords of patients with spinal cord injury. As people here may recall, he published an important paper in 1995, reporting that peripheral nerve grafts between white and gray matter allows axons to grow across the gap of a transected cord. For perhaps 15 years, he has been doing searching for patients with transected spinal cords to do his bridging therapy on. I am not sure how many he has found but I think that he has found only several patients to date who have transected spinal cords.

Please stop saying that we have selected less severely injured patients. It is not true. You don't know what you are talking about. All the patients that we studied have severe spinal cord injuries. They all had ASIA A complete injuries. The only patient that we excluded from the study is one patient who had a DTI scan showing more than 3 segment gap of his white matter. All the patients had a clear gap in their white matter at the injury site. None of the patients that were screened for the trial had a transected spinal cord.

Have you read the paper by Liu, et al. (2011)? I attach it and suggest that you read it, if you have not. Yes, there is extensive sprouting in the rostral spinal cord but axonal growth continue for many weeks in hemisected mice. Why aren't the additional corticospinal axons blocked, not only from growing across the rostral but across the caudal side of the lesion? In animals where PTEN had not been deleted (they were injected with AAV-GFP rather than AAV-Cre), the growing axons stop right at the rostral lesion edge. Why do they stop? Jerry Silver and others have suggested that they stop because of the glial scar and the presence of chondroitin-6-sulfate-proteoglycan (CSPG). The deletion of PTEN from the cortex of these mice does not eliminate CSPG or gliosis at the lesion edge and yet the axons grow through the so-called glial scar.

By the way, if you read the papers, note that axons don't like to grow where astrocytes are absent. For example, Liu, et al. wrote in their discussion that robustly axons failed to penetrate into GFAP-negative areas of the lesion site. Interesting that the axons don't stop where the glial cells are.


Absence of astrocytes in fact deters growth of axons. Axons prefer to grow on glia. Incidentally, this is similar to what Michael Sofroniew also observed in his review of the role of reactive astrocytes in brain and spinal cord repair. Axons grow readily through thickets of reactive glia. He concluded that astrocytes protect tissue and preserve function rather than block regeneration and repair. Sofroniew pointed out that there are opposing views concerning whether reactive astrocytes are friend or foe.

Our study of contused spinal cords show that a loose tissue matrix of mostly astrocytes accumulate at the contusion site. Many axons enter into this tissue matrix and some may even exit, accounting for why there may be delayed recovery in some animals after spinal cord injury. If there were a tight barrier that prevents axon growth at the contusion site, why are there many axons entering the injury site? There are many reasons why axons entering the injury site may not grow into the distal cord, including the presence of CSPG, Nogo, and other growth inhibitors at the caudal edge. Probably some axons do grow out of the contusion site. In any case, it doesn't matter. The fact that the axons are growing into the injury site is a strong argument against an important role of the glial scar preventing axonal growth into the injury site.

Wise.

Wise,

did you see the second presentation of Jerry Silver at W2W 2012?

Do you have any comment?

Paolo

[paolocipolla]

Paolo, you don't know what the BBB score means. A BBB score of 8 is a non-walking score. Most of the movements that occur in BBB scores from 1-8 are probably reflexive. At a score of 10 or less, the rat is unable to stand or support its weight with its legs. The 25-mm weight drop contusions that we are studying in the rats is equivalent to ASIA A injury in human.

Wise.

Wise,

if you test with electrophisiology a rat who has 25 mm drop SCI do you see any signal crossing the injury site?
According to the MASCIS study (page 457) after a 25 mm drop the average BBB score is 10.6 which means that the rats can support their weight, which makes them ASIA D.

Paolo

[paolocipolla]

Perhaps some discussion will help clear things up. There are few, if any, absolutes in biology. All cell-to-cell interactions involve balances between positive and negative influences. Astrocyte/axon interactions are no exception. Astrocytes are highly maleable cells. They respond to a variety of factors in their environment in different ways and they differ in their responses depending on their state of maturity and upon what they encounter. Astrocytes produce both growth promoting as well as inhibitory molecules that they deposit in various ratios upon their surfaces again depending upon what they interact with. When astrocytes encounter vigorously growing axons, even when the astrocytes are mature, they can provide support and guidance for axonal growth and they produce less inhibition. They also align themselves and wrap around the axons. Unfortunately, in the adult, just after SCI astrocytes do not encounter robustly growing axons, unless pTEN is deleted long before the injury. Instead, they encounter mostly inflammatory cells and dystrophic axons and myelin debris. Here their job is to build a wall around the injury which is both physically obstructive because the cells arrange themselves perpendicular to the lesion and they also hypertrophy and form lots of tight junctions between themselves. In addition, they produce far more inhibitory molecules. Everybody but Wise refers to this process as "scarring". In the mouse after a surgical or narrow crush lesion, and where pTEN is deleted, the lesion remains relatively small (compared to rat or human) and axons are in a robust growth state from the moment they are lesioned. Thus, the balance shifts, at least partially, to a growth supportive astrocyte rather than just a reactive scarring astrocyte. Actually, it is likely that there would be a mixture of both scarring and bridge building cells. Even in the pTEN deletion animals most axons still get hung up at the rostral end of the lesion. The majority of pTEN deleted axons that don't cross the injury site do so where they lack a bridge but they also likely see more inhibitory astrocytes and lakes of macrophages. Unfortunately, we have yet another cell in the lesion core that is very destructive to axon regeneration and that is the macrophage (who is a really bad guy especially at early stages after injury). Thus, things are far more complicated and fluid than the black and white picture that Wise likes to present. It helps to make his case that there is no scar because he is claiming to have promoted an unprecedented long distance regeneration in chronically injured humans, who would have well established scar, without doing anything especially potent to remove or overcome it. He has not deleted pTEN or made any attempt remove scar. I would love to see even in vitro evidence that UMBCs form highly growth growth supportive substrates that can overcome inhibitory molecules. Those of you who would believe as a matter of faith that whatever Wise tells you is absolutely true then that is your right. However, I have have a multitude of questions that need answering before I accept an unprecedented conclusion:

Truly regenerating axons NEVER grow in tightly aligned bundles. Regeneration in the adult is not like the fasciculated axon growth that occurs in the embryo. Indeed, the presence of aligned bundles of axons passing in the vicinity of a cord lesion is always suggestive of spared axons.

Axons can regenerate in relative alignment if they are given an aligned cellular environment such as that which could be induced by a biomatrix or tube. Given that there was no attempt to align the UMBCs it is very unlikely that axons could regenerate as a single straight bundle.

The idea that a lengthy, truly regenerated bundle of axons could degenerate and then magically regenerate yet again in the chronically injured human cord is absurd. Just stop and think of the myriad of barriers that would be thrown up after that.

Without specific axonal labels how does one conclude definitively using DTI only that a bundled structure is axonal?

Without the use of multiple intermediate images to document a growing front of the bundle how does one conclude that it is growing but has not yet reached its target. It is incredibly premature to conclude this.

How can adult human neurons grow so rapidly (1mm per day according to Wise) without any modification of their intrinsic growth potential?

Thanks Dr. Silver for the expalination and for the critical considerations.

Paolo

[Wise Young]

Wise,

if you test with electrophisiology a rat who has 25 mm drop SCI do you see any signal crossing the injury site?
According to the MASCIS study (page 457) after a 25 mm drop the average BBB score is 10.6 which means that the rats can support their weight, which makes them ASIA D.

Paolo

This should be an average score. What MASCIS study are you referring to? Are you referring to the original BBB score paper? A 10.6 is not equivalent to an ASIA D. A lot of people can do weight-bearing and they are not ASIA D.

A 25 mm weight drop eliminates somatosensory evoked potentials in 90% of rats.

Wise.

[paolocipolla]

Dr. Young and Dr. Silver,

I ask this with the utmost respect and no intention of stirring stife, but speaking of the existence of this scar tissue, how can two scientists come away with different opinions on the existence of something that should be quite evident?

Todd

That may suggest you that there is something wrong in SCI research, that is why it is necessary to learn and question as much as possible rather than having blind faith and just rise money for SCI research. Too much money going into SCI research are wasted. In some cases it is impossible to tell in advace which one are wasted (that's research), but sometimes there are elements to determine what worth founding and what it doesn't.

Paolo

[paolocipolla]

I have been studying this structure for most of my scientific career. I have had a long standing NIH grant entitled "Regeneration beyond the glial scar" for nearly 30 years. If one simply reads the literature , uses the proper techniques and does the right experiments there is no doubt that scar exists. Indeed, the scar is well known by neurosurgeons who attempt to aspirate away non-invasive brain tumors that are surrounded by scar. They, aspirate until they push up against a tough tissue border that is the surrounding scar. By the way, contrary to what Wise suggests, there is no need for fibroblasts to create the tough scar. If you'd like, read my 2004 Nature Reviews Neuroscience review. Regeneration beyond the glial scar : Article : Nature Reviews ...
www.nature.com/nrn/journal/v5/n2/full/nrn1326.html - Similar
Review. Nature Reviews Neuroscience 5, 146-156 (February 2004) | doi : 10.1038/nrn1326 ... the glial scar. Jerry Silver1 & Jared H. Miller1 About the authors ...

Here is another reference:


Glial scar
From Wikipedia, the free encyclopedia
Glial scar formation (gliosis) is a reactive cellular process involving astrogliosis that occurs after injury to the Central Nervous System. As with scarring in other organs and tissues, the glial scar is the body's mechanism to protect and begin the healing process in the nervous system. In the context of neurodegeneration, formation of the glial scar has been shown to have both beneficial and detrimental effects. Particularly, many neuro-developmental inhibitor molecules are secreted by the cells within the scar that prevent complete physical and functional recovery of the central nervous system after injury or disease. On the other hand, absence of the glial scar has been associated with impairments in the repair of the blood brain barrier.[1]

Contents [hide]
1 Scar components
1.1 Reactive astrocytes
1.2 Microglia
1.3 Endothelial cells and fibroblasts
1.4 Basal membrane
2 Beneficial effects of the scar
3 Detrimental effects of the scar
4 Primary scar molecular inducers
4.1 Transforming growth factor β (TGF-β)
4.2 Interleukins
4.3 Cytokines
4.4 Ciliary neurotrophic factor (CNTF)
4.5 Upregulation of nestin intermediate filament protein
5 Suppression of glial scar formation
5.1 Olomoucine
5.2 Inhibition of Phosphodiesterase 4 (PDE4)
5.3 Ribavirin
5.4 Antisense GFAP retrovirus
5.5 Recombinant monoclonal antibody to transforming growth factor-β2
5.6 Recombinant monoclonal antibody to interleukin-6 Receptor
6 References

More very interesting things to learn for me.

Paolo

[Marcus]

Dr Silver seems to fear that the research of Dr Wise can overthrow his beliefs about spinal cord scar be a invencible barrier without removal surgery.

[paolocipolla]

Dr Silver seems to fear that the research of Dr Wise can overthrow his beliefs about spinal cord scar be a invencible barrier without removal surgery.

Marcus,

if you listen at minute 31 of the W2W presenattion of Dr, Silver you will learn that Dr. Silver saies that there might be other ways to deal with the scar.. so he is open to a less invasive solutions

http://vimeo.com/56641033

Paolo

[ay2012]

lol .. I think it would be much easier to 'espionage' in person than on an internet message board .. but what do I know. lol

They want to take lithium? Be my guest! It was used to treat schizophrenia - have fun!

If a course of lithium helps regenerate your spinal cord you wouldn't take it? Its toxicity level is well established: http://en.wikipedia.org/wiki/Lithium...y#Side_effects
Just because its used to treat manic depression doesn't mean you'll become a manic depressive if you take it...

[Marcus]

lol .. I think it would be much easier to 'espionage' in person than on an internet message board .. but what do I know. lol

They want to take lithium? Be my guest! It was used to treat schizophrenia - have fun!

Nobody is been asked to take any drug before the final results.

[Moe]

Just because its used to treat manic depression doesn't mean you'll become a manic depressive if you take it...

That is true.

[paolocipolla]

Oh, but you are wrong here. He does deny that in the absence of fibroblasts in the CNS there is scar. Thus, he denies that gliosis (astrocyte only behavior) can form a barrier. There are no fibroblasts in the brain or spinal cord normally, so in the absence of a penetrating injury that freely allows them into the CNS compartment from the meninges, it has been thought that few fibroblasts enter the CNS. So the astrocytes have taken over the job in the CNS of walling off inflammation. When fibroblasts do enter CNS in large numbers then the scar that is made by astrocytes becomes even more impenetrable because fibroblats produce a myriad of additional inhibitory molecules and the reactive astrocytes wall them off as well with a membranous structure called basal lamina. If you or Wise wish to call what the astrocytes do something other than "scar' that is no problem but you will not be able to communicate with the rest of the world where use of the word scar is more flexible. In the end, the important point is that reactive astrocytes contribute to regeneration failure in the CNS in a big way and they have to be overcome, removed or altered somehow to get regeneration to occur especially at chronic stages. Until data is presented, there is no evidence that UMBCs or lithium are capable of overcoming scar or the inhibitory molecules associated with it.

I think the above is very informative and open mind at the same time.
All researchers and neurosurgeons I have had the opportunity to talk with about the scar so far are on the same line of thinking based on the scientific evidence so far available.
If Wise can prove that the scar is no problem I'll be super happy as we'll have an easier way out of chair. Still waiting after many years hearing Wise's theory.

Paolo

[paolocipolla]

This should be an average score. What MASCIS study are you referring to? Are you referring to the original BBB score paper? A 10.6 is not equivalent to an ASIA D. A lot of people can do weight-bearing and they are not ASIA D.

A 25 mm weight drop eliminates somatosensory evoked potentials in 90% of rats.

Wise.

Wise,
see the study that you uploaded for me here: https://www.carecure.net/forum/showp...3&postcount=85

As I know a rat with a BBB score of 10 or above can walk well enogh that can be examined with a catwalk and get usefull data while below 10 cawalk is not very usefull because the belly touches the flor and owerlap the foot prints of the rear paws: http://www.noldus.com/animal-behavio...oducts/catwalk

Paolo

[crabbyshark]

No I am not wrong.

Yes, you are. Axons are regrowing in your body right now, very slowly.

Understanding the answers to these questions require an understanding of the slow rate of axonal growth, the long distance of axonal growth, and the concept of neuron survival above and below the injury site. Let me first go through these three concepts and then answer your question.

1. Rate of axon growth. Axons grow slowly. In the peripheral nerve, regeneration occurs at 1 mm/day or less. So, if you crush your ulnar nerve at the elbow and lose sensation in the fourth and fifth fingers of your hand, the axons in the ulnar nerve will regenerate but the distance (at least for my arm) is 38-40 cm. At 1 mm/day, it may take more than a year before one regains sensation. Nobody really knows how fast spinal axons grow in human but I have suggested that the rate of growth spinal axons is no faster than peripheral nerve. It may well be slower than 1 mm/day, particularly if you are older. The other comparison is the rate of hair growth. Most of us grow hair at the rate of about a mm per day when we are young and slower when we are older.

2. Distance and barriers to axon growth. Axons usually die back to their first branch point before the injury site. This statement has multiple implications. First, most axons connect to many neurons. For example, a corticospinal neuron has branches that connects to the midbrain, the brainstem, the cervical spinal cord, and the lumbar spinal cord. So, an injury at the thoracic level will disconnect the axon from the lumbar spinal cord and the axon may die back to its first branch point in the cervical spinal cord. So, the regenerating axons must grow from the cervical spine down to the lumbar cord, in order to reconnect. On the other hand, in animal models, it is clear that most of the spinal axons will die back and then can and will grow from the first branch point to the edge of the injury site. This was one of the reasons by Jerry Silver (a very good scientist) once told me that this was why he did not believe in the Nogo theory of axon growth inhibition. After all, the axons that have died back and then regrow to the injury edge must have been growth amongst myelinated tracts and should have been inhibited. Jerry Silver showed and hypothesized that chondroitin-6-sulfate proteoglycans at the edge of the injury site stopped axonal growth. In experiments with treatments such as olfactory ensheathing glia, once axons cross the injury site, they seem to be able to keep growing, again suggesting that myelin inhibitors are not as important as originally thought. It is possible that axons grow slower in the presence of myelin growth inhibitors rather than stop altogether. Very recently, a former graduate student (Kai Liu) showed that if he shuts off a gene called PTEN in neurons, he could induce massive regeneration of the corticospinal tract in rats. This was the first time that anybody had demonstrated that the obstacles to axonal growth are not just axonal inhibitors but the presence of genes that hold back or prevent axonal growth. Finally, decades after spinal cord injury, one can see axons forming dystrophic terminals at the edge of the injury site. These terminals were once thought to be degenerating ends of terminals but is now thought of as "frustrated" growth comes. This indicates that axons are continuing to growing to the injury site and stopping at the edge, possibly due to chondroitin-6-sulfate-proteoglycans surrounding the injury site.

3. Neuronal survival above and below the injury site. Trauma damages not only axons but neurons in the spinal cord. When the neurons are damaged, this means that axonal regeneration may not restore function. Thankfully, most of us have more neurons that we need and we continue to be able to function even if we lose 20-50% of the neurons in a given segment. Until recently, although axonal regeneration is considered to be possible, neuronal replacement was considered to be in the neverneverland of scientific fantasy. However, the discovery of neural stem cells in the late 1990's and the demonstration that they can create neurons that incorporate into the neural circuity of the brain changed this pessimism. Today, most scientists believe that neuronal replacement is also possible, although there have not been many clear examples of neuronal replacements leading to functional recovery.

Given the above, you can perhaps understand why most neuroscientists are optimistic about the possibility of spinal cord regeneration and restoration of function. Please understand how new these are ideas are. In 1997, when I started Spinewire, we did not know the second two. In 2001, when we converted Spinewire to CareCure, we did not know the third point. IN fact, if you go back to my writings in the early days of CareCure, you will find that I was very pessimistic about neuronal replacement.

So, back to your question concerning what happens to the places where axons have degenerated. Do the degenerated pathways "collapse"? Do regenerating axons have to make new pathways for growth? I don't know the answers to these questions but can provide some educated guesses based on what I have seen. First, degenerated tracts often remain for months after injury. Called "Wallerian" degeneration, after Waller who noted that degenerating tracts can be identified with silver stains for months or even years after injury, these tracts are very likely to remain available to axons to regrow into them. This may or may not be true years after injury but I suspect that the problem may not be loss of physical space for growth but the loss signals to tell the axon that it is growing on the right path and to keep on going.

In the mid-1990's, I likened the journey of axons in the spinal cord to the journey of Odysseus back home. First, the Odyssesus is a long ways from home. Second, after he crosses the injury site where he had to combat a variety of monsters, including cyclops (macrophages), he has to sail past the syrens (neurons) along the way. Third, as he gets closer to home, he may stop at a place that is not his home (Circe's Island). Finally, when he gets home, he will find that many suitors have taken his place and he has to fight them off in order to get Penelope back.

I think that we have tools to start Odysseus on his journey. Getting him home is another matter. On the other hand, much evidence suggests that the nervous system is much more plastic than we had ever thought and that intensive exercise helps strengthen desirable connections and eliminates undesirable connections. Therefore, if we get a lot of sailors (axons) growing home and they connect, they can do the job but it will require a lot of learning and practice before things work again.

Wise.

If UCBMC injection is effective, recovery is likely to take a long time.

Neurons aren't known to regrow/regenerate like axons. That's one part of why the current UCBMC treatments being tested, if shown to work, are not the end-all-be-all silver bullet "cure" for SCI.

I'm not going to repeat myself on this. You can re-read my earlier posts. 20+ references of UCB studies and all on acute. None showing efficacy in chronic. That should tell you something on which is harder to fix.

It might tell me something about which one is harder to study. It tells me nothing about which one is harder to fix.

For instance:
According to this study, the scar resulting from a SCI is fully formed in a dog at 4 weeks.
This study (click HTML, go to the discussion at the end), researchers divided dogs into four groups of three. One group was control, one group injected with cMSC (canine UCBC) 12 hours after injury, one group injected 1 week after injury (when scar was forming), and one group injected 2 weeks after injury.
They determined injection at 1 week after injury was most effective, 2 weeks after injury was second most, and 12 hours after was the worst.
If this is true, isn't it at least possible that an injection at 4 weeks (or 5 weeks or 10 weeks or 100 weeks) might be more effective than an injection given at 12 hours?
Not saying you think it's likely, but you must allow for the possibility.

That is one scientist's opinion. I don't know if that is true or not. That seems like it is an easy thing to abuse.

Do people normally live to be 80? Do golden retrievers usually live to be 12?

Why not just do animal studies 2-3 months after injury to be sure it is a chronic model?

Because it's expensive to maintain chronic SCI animal and they often die before their part of the experiment starts.

c) I don't see them crossing all the way across the injury site, let alone going past it.

4.7A! They look like ants marching across the injury site (the black area). It's beautiful.

d)Is Wise's trial using USSC cells or Mesenchymal cells or both in his human trial?

The current trial is not using mesenchymal cells. USSC make up a portion of UCBMC.

Let me first define the cells that we are transplanting. We are using HLA-matched umbilical cord blood mononuclear cells (UCMBC). Umbilical cord blood contains four major categories of cells: red blood cells (RBC), platelets, neutrophils (polynuclear cells), and mononuclear cells. UCMBC are usually isolated by density centrifugation into what is called the "buffy-coat" layer, which excludes RBC, platelets, and neutrophils. Mononuclear cells include monocytes (40%), lymphocytes (30%), and other cells (macrophages, basophils, mast cells, etc.).

Mononuclear cells contain several populations of known stem cells and progenitor cells in cord blood.
• CD34+ cells. These are usually endothelial progenitor cells but some hematological and pluripotent stem cells may express CD34+. These are the cells that are most commonly counted in umbilical cord blood as a marker of the number of stem cells. They are usually about 0.5% of the UCBMC.
• CD133+ cells. These are pluripotent stem cells, many of which co-express CD34+. They are generally fewer in number than CD34+ cells.
• VSEL cells. These "very small embryonic-like" cells are believed to be pluripotent. They are usually less than CD133+ and, because of their size, often lost during the density centrifugation procedure.

b) I wouldn't consider the amounts of axons to be robust. Looking at figure 4.8 on pg. 67, I only see a few axons.

That picture was taken three weeks after injection. Compared to the control, axon proliferation in injury site is robust. The DTIs Dr. Wise saw were taken a year after injection. Either way, we may not need to regenerate very many axons to get better.

In animal studies, we and many others have shown that less than 10% of the spinal cord is sufficient to restore unassisted walking in animals and probably in humans. I have been in the operating room with patients who have had tumors removed from their spinal cord and they walk out of the hospital even though they have no more than 10% of their spinal cords.

So, the goal of regenerating the spinal cord is to add enough axons that reconnect to the correct places to restore function. One does not have to regenerate more than 10% of the spinal cord to restore function. If there are already 8% of the spinal tracts present, perhaps it is sufficient to add 2% to the mix.

P.S. I'm not mad at you if that's the vibe you're getting. Rubber Soul is a great album.

[crabbyshark]

Oh, but you are wrong here. He does deny that in the absence of fibroblasts in the CNS there is scar. Thus, he denies that gliosis (astrocyte only behavior) can form a barrier. There are no fibroblasts in the brain or spinal cord normally, so in the absence of a penetrating injury that freely allows them into the CNS compartment from the meninges, it has been thought that few fibroblasts enter the CNS. So the astrocytes have taken over the job in the CNS of walling off inflammation. When fibroblasts do enter CNS in large numbers then the scar that is made by astrocytes becomes even more impenetrable because fibroblats produce a myriad of additional inhibitory molecules and the reactive astrocytes wall them off as well with a membranous structure called basal lamina. If you or Wise wish to call what the astrocytes do something other than "scar' that is no problem but you will not be able to communicate with the rest of the world where use of the word scar is more flexible. In the end, the important point is that reactive astrocytes contribute to regeneration failure in the CNS in a big way and they have to be overcome, removed or altered somehow to get regeneration to occur especially at chronic stages. Until data is presented, there is no evidence that UMBCs or lithium are capable of overcoming scar or the inhibitory molecules associated with it.

Our results suggest that transplanted cMSCs promote the formation of neuronal network through regulating the intracellular pathways of the actin cytoskeleton to overcome repulsive forces which result from scar formation and severance of axon.

SOURCE