While we all know that multiple sclersosis is an immunologically mediated disease, it is interesting to learn about some of the non-immunological mechanisms that are associated with this condition. 

For example, we generally think of T cells as being the main effectors of demyelination and damage to the central nervous system because agents that suppress T cell infiltration into the central nervous system or their activation seem to be useful in the treatment of multiple sclerosis.  However other indirect mechanisms are also involved.  For example, microglial cells which secrete inflammatory cytokines also release glutamate, which causes excitotoxicity to neurons.

Astrocytes are the major “glial” component of the central nervous system.  The primary function of astrocytes is to support neurons in their activities.  Astrocytes do this in many ways, including; a) providing control of blood flow in neuronal areas; b) maintaining a proper nutrient environment locally, for example, astrocytes produce lactate and other nutrients that are needed for neuronal function; c) telling the oligodendrocytes when to start stimulating production of myelin; d) acting in repair of damaged tissue; and e) cleaning up left over neurotransmitters.

Based on the numerous functions of astrocytes, the authors of the paper we will be discussing (Tilleux et al. Selective up-regulation of GLT-1 in cultured astrocytes exposed to soluble mediators released by activated microglia.  Neurochem Int 2009 Jul-Aug;55(1-3):35-40), sought to determine whether astrocytes may uptake excess glutamate, and how astrocyte uptake of glutamate may be initiated.

As a model of injury, rat microglial cells were stimulated with LPS in vitro.  LPS is the component of Gram Negative Bacteria that causes endotoxic shock and is one of the most potent activators of cells of the myeloid lineage.  Specifically, LPS is known to induce production of many cytokines from macrophages and microglial cells including TNF-alpha, IL-12, and IL-18. 

Conditioned media from the LPS stimulated microglial culture was added to cultures of rat astrocytes and expression of the type 1 glutamate transporter was assessed.  The Type 1 glutamate transporter is an important protein found in glial and neurons that sucks up the glutamate from the synaptic cleft in order to prevent excitotoxicity.  The conditioned media, which represented numerous inflammatory stimuli was capable of upregulating expression of the type 1 glutamate transporter. 

Upregulation of astrocyte expression of the type 1 glutamate transporter could also be achieved by direct addition of TNF-alpha to the astrocytes.

Treatment of astrocytes with dibutyryl cAMP, which activates astrocytes, was also capable of upregulating expression of the type 1 glutamate transporter. 

These data suggest that astrocytes are an important protective mechanisms against glutamate toxicity and that astrocytes can actually “feel” when inflammation will occur and respond accordingly by upregulating glutamate transporters.

Microglia, the macrophages that reside in the brain, are believed to be involved in the process of neuronal degeneration in multiple sclerosis and animal models of the disease.  What are the mechanisms by which microglia may be pathological?  A recent report (Shijie et al. Blockade of glutamate release from microglia attenuates experimental autoimmune encephalomyelitis in mice. Tohuko J Exp Med 2009 Feb;217(2):87-92) suggests that microglial production of glutamate may be a cause of toxicity.

How can glutamate kill neurons?  Glutamate is used by neurons to communicate with each other.  The concentration of glutamate in the brain is tightly controlled by the blood brain barrier, which has specific glutamate transporters to only allow as much glutamate as is needed.  Additionally, when neurons communicate with each other, mechanisms exist to clear up the glutamate very rapidly after the signal is transmitted.  Too much glutamate causes what is called excitotoxicity, that is, death of the neurons from over stimulation.

In the current study the investigators demonstrated that in vitro activated microglial cells produced high concentrations of glutamate, which induced killing of neurons.  Microglial cells were demonstrated to have high concentrations of glutaminase, which generates glutamate.  When glutaminase was inhibited in vitro by addition of a small molecule inhibitor, the ability of the activated microglial cells to make glutamate, and subsequently to kill neurons was decreased. 

Furthermore, it was found that administration of the glutaminase inhibitor to mice suffering from experimental allergic encephalomyelitis (mouse multiple sclerosis) resulted in functional improvement.

These data suggest that microglial toxicity of neurons may be not only related to immunological means, but also through direct production of mediators that kill neurons.

It should be noted that Riluzole, the first drug approved for treatment of ALS works in part through suppressing interacton of glumate with its receptor, as well as upregulating activity of glutamate transporters that clear glutamate.  In fact, Riluzole has actually been demonstrated to inhibit MS-like disease in the EAE model (Gilgun-Sherki et al. Riluzole suppresses experimental autoimmune encephalomyelitis: implications for the treatment of multiple sclerosis. Brain Res 2003 Nov 7;989(2):196-204).

Macrophages, known as the “Big Eaters” are populations of immune cells that protect the body against various pathogens by engulfing them, by secreting various chemicals to let the other components of the immune system know that the body is under attack, and also by causing inflammation.  The macrophages inside the central nervous system are called “microglia”.  Glia are cells in the central nervous system that are not neurons but are involved in maintaining the function of neurons.  Specifically, glial cells are involved in providing a proper chemical environment for the neurons so that they can communicate properly.  For example, glial cells provide nutrition, physical support, and contribute to production of myelin.  Microglia comprise approximately 20% of the glial cell population.  Other glial cells include astrocytes (that amongst other things physically connect neurons to their blood supply) and oligodendrocytes (which make myelin).

While microglia are normally involved in protecting the brain from infections, they also are believed to play a pathological role in multiple sclerosis.  For example, microglial production of inflammatory cytokines, free radicals, and matrix metalloproteases is believed to be one of the reasons why multiple sclerosis damages the central nervous system (Benveniste et al. Role of macrophages/microglia in multiple sclerosis and experimental allergic encephalomyelitis. J Mol Med 1997 Mar;75(3):165-73).  This makes sense in some ways since T cells producing the cytokine interleukin-17 (IL-17), called Th17 cells, have been found both in mice and humans with multiple sclerosis.  Interestingly, microglia express receptors for IL-17, and the level of receptor expression increases during disease induction (Das Sarma et al. Functional interleukin-17 receptor A is expressed in central nervous system glia and upregulated in experimental autoimmune encephalomyelitis. J Neuroinflammation 2009 Apr 28;6:14).  On the flip side of the coin, microglial cells have been postulated to also be involved in repair of the brain during multiple sclerosis (Napoli and Neumann.  Protective Effects of Microglia in Multiple Sclerosis. Exp Neurol. 2009 May 3).

Macrophages are interesting cells.  As we discussed previously in the same why that T cells may be effector T cells and T regulatory cells, broadly speaking, so too macrophages may be proinflammatory (called M1) and antiinflammatory (called M2).  In fact, one of the reasons why using your own fat stem cells may be therapeutically beneficial in multiple sclerosis is because there are high concentrations of antiinflammatory M2 macrophages found in the fat.  The classical distinction between these two types of macrophages has been that M1 macrophages produce nitric oxide when they are activated and that M2 macrophages produce products of the enzyme arginase.  Mesenchymal stem cells, which have demonstrated activity against the mouse model of multiple sclerosis, have been demonstrated to be able to induce an M2-like phenotype in macrophages by stimulating them to produce the antiinflammatory cytokine interleukin-10.  Before we continue, we should state that macrophages are derived from monocytes, and there is some work suggesting that monocytes in special situations may act as stem cells.

Now if mesenchymal stem cells have the ability to induce production of interleukin-10 in macrophages, what if someone discovered a chemical that could induce production of this antiinflammatory cytokine (which theoretically would be beneficial for multiple sclerosis) simply by administration of the chemical?  While this would not induce the regenerative or growth factor effects of mesenchymal stem cells, it may be useful for inhibiting the destruction cause by activated macrophages. 

A recent paper (Correa et al. Anandamide enhances IL-10 production in activated microglia by targeting CB(2) receptors: Roles of ERK1/2, JNK, and NF-kappaB. Glia 2009 Jun 29) demonstrated that the endocannabinoid anandamide may actually have this property.

Anandamide a neurotransmitter of the cannabinoid family that occurs naturally in the human central nervous system.  It activates the CB1 receptor in the brain and the CB2 receptor in the periphery.  It is known that activation of cannabinoid receptors is associated with decreased symptoms in animal models of multiple sclerosis (Cabranes et al. Decreased endocannabinoid levels in the brain and beneficial effects of agents activating cannabinoid and/or vanilloid receptors in a rat model of multiple sclerosis. Neurobiol Dis 2005 Nov;20(2):207-17).  It is also known that deletion of the enzyme Fatty Acid Amide Hydrolase (FAAH), whose role is to metabolize anandamide, causes regression of symptoms of multiple sclerosis in animals, presumably by increasing the amount of endogenous anandamide (Webb et al. Genetic deletion of Fatty Acid Amide Hydrolase results in improved long-term outcome in chronic autoimmune encephalitis. Neurosci Lett. 2008 Jul 4;439(1):106-10).  Therefore in the paper that we are discussing, the investigators wanted to see if anandamide may alter production of cytokines and behavious of microglial cells.

The authors demonstrated that administration of anandamide results in upregulation of interleukin-10 production from microglial cells that have been activated.  The induction of interleukin-10 seemed to be associated with suppression of the transcription factor NF-kB, which is involved in numerous inflammatory conditions.  Furthermore, the investigators demonstrated that microglial cells treated with anandamide had suppressed ability to produce the inflammatory cytokines IL-12 and IL-23, which was correlated to inhibition of Th1 and Th17 generation.

These data suggest a very important point, which is that numerous neurotransmitters may have effects on immunological cells.  Theoretically, this should not be that surprising.  The immune system and the nervous system are the only systems of the body that share the features of: a) Distinguishing between what is self and what is non-self; b) Ability for specificity; c) Memory; and d) Innate reactivity.  At a practical level, the major immunological organs such as they thymus and spleen are highly innervated.  Additionally, most immunological cells, including dendritic, T cells, B cells, and NK cells all have expression of receptors for neurotransmitters. 

Therefore we at Stemnow.com believe that this area of “neuroimmunology” will be an exciting one to watch in the upcoming years.

Stem cells may be attracted to the injured central nervous system of patients with multiple sclerosis by virtue of the molecule known as SDF-1, which is expressed at the onset of disease.  Indeed, we do know that stem cells selectively home to the central nervous system, at least from animal studies, in which adult mesenchymal stem cells are selectively found associated with areas of injury.  But could there be other injury signals that attract stem cells? 

We discussed previously that receptors associated with pain-related peptides, such as the kinin receptor B1, have the ability to make the disease worse or better depending on inhibition or activation, respectively.  An interesting molecule called Substance P, is a peptide neurotransmitter that is released in various situations of tissue injury.  We will discuss a recent paper (Hong et al. A new role of substance P as an injury-inducible messenger for mobilization of CD29(+) stromal-like cells. Nat Med. 2009 Apr;15(4):425-35)  demonstrating that Substance P is associated with homing of stem cells.

The investigators describe a model system in which injury induces mobilization of a mesenchymal stem cell-like population that expresses CD29 and is involved in acceleration of wound healing.

They  demonstrate that administration of Substance P in absence of injury in either mice, rabbits, or rats, induces mobilization of the CD29 cells from out of the bone marrow and into systemic circulation.  

To demonstrate that the mobilized CD29 cells actually had regenerative activity, they harvested CD29 cells that were mobilized, and injected mobilized cells, together with substance P in a rabbit wound model, in which the wound is induced by alkaline injury. Engraftment of the transplanted cells, as well as acceleration of healing, was observed. 

In order to make the case for clinical relevance of these observations, the investigators performed a series of experiments using human bone marrow mesenchymal stem cells as a model system for in vitro study.  It was observed that Substance P augmented the rate of transmigration, induced nuclear translocation of beta-catenin, triggered cell proliferation, and stimulated the activation of ERK1 and ERK2 pathways.

The authors conclude with the statement that: “This finding highlights a previously undescribed function of substance P as a systemically acting messenger of injury and a mobilizer of CD29(+) stromal-like cells to participate in wound healing”

If indeed new stem cell mobilizers can be identified in addition to G-CSF and the Anormed compound, perhaps one day it may be possible to simply redistribute your stem cells between body compartments so as to not need to take stem cells from outside of the body.

When I started my training, if you wanted to know whether one gene goes up or down, it would take at least a week to figure out for every gene.  Now, in one afternoon a scientist in one shot take a look at level of expression of all known genes (about 30,000) of the human body!  This revolution in science has allowed for the discovery of new molecular pathways without having to know what one is looking for.  So in conditions such as multiple sclerosis, many scientists basically “go fishing” to try to find new genes whose expression correlates with disease.  People have even taken it further by being able to assess not only all genes, but also proteins made by the genes (called proteomics), and more recently, like my friend Gabriela Cezar at Stemina does, look for all small molecules (called metabolomics).

These very powerful techniques are beginning to bear their fruits.  A recent paper (Schulze-Topphoff et al.  Activation of kinin receptor B1 limits encephalitogenic T lymphocyte recruitment tothe central nervous system.  Nature Medicine.  June 28, 2009) identified that patients with multiple sclerosis, as well as in animals bearing a disease similar to multiple sclerosis (experimental allergic encephalomyelitis) have increased expression of the kinin receptor B1.  This receptor is activated by components of the kinin-kallikrein system, which are a group of proteins involved in pain, inflammation, and coagulation of blood.

The investigators found that giving mice developing experimental allergic encephalomyelitis (mouse model of multiple sclerosis) activators of the kinin receptor resulted in less disease, whereas administration of inhibitors of this receptor resulted in acceleration of disease onset.  This was demonstrated when the compounds were given before disease onset, but in other experiments even after disease onset. 

Manipulation of receptors using small molecules can be tricky business.  In other words, it may be that the small molecule receptor activator/inhibitors may have been working through other biological pathways to alter disease course.  Therefore, in order to know conclusively whether the kinin receptor B1 is responsible or not for alteration in disease process, the investigators used mice lacking the kinin receptor B1.  These mice suffered from accelerated disease, thus suggesting that the receptor is normally involved in controlling the disease. 

Expression of the receptor had to be on the T cells in order to mediate protection from disease.  It was demonstrated that activation of the kinin receptor B1 selectively suppressed the infiltration of Th17 cells into the central nervous system.  Most interestingly suppression of infiltration was limited to Th17, with Th1 cells still infiltrating.  One intersting question is whether there is selective expression of the kinin receptor B1 associated with various TCR clonotypes that are expanded in multiple sclerosis progression, as seen in this video describing a paper by Eli Sercarz.

These data suggest a brand new molecule that can be targetted in multiple sclerosis.  It also illustrates the power of using “discovery based” approaches.  If indeed selective inhibition of Th17 entry into the CNS can be achieved in humans, this may be useful as a synergistic agent with bone marrow or fat stem cell approaches to multiple sclerosis.