The CNTF family of neuroregulatory cytokines and the signal transduction pathways activated by these factors

CNTF is a naturally occurring cytokine that is produced predominately by astrocytes within the central nervous system (CNS). CNTF belongs to a family of cytokines which include LIF, interleukin-6 (IL-6), cardiotrophin 1 (CD-1), and oncostatin M (OSM), which have been categorized as neuroregulatory cytokines because they display similar actions in the nervous system. Click to see a summary of these actions. Their actions include regulation of neurotransmitter phenotype, differentiation/maintenance of glial and neuronal precursor cells, and promotion of survival of differentiated neurons. The similar effects of the neuroregulatory cytokines occurs because they bind structurally similar receptor complexes containing the common signal-transducing receptor subunits glycoprotein 130 (gp130) and leukemia inhibitory factor receptor (LIFR), and activate a common complement of signal transduction pathways, including the JAK → STAT3, Ras → MEK → ERK, PI3K, PKC pathways.

Click here to enlarge

Although the neuroregulatory cytokines do not display amino acid sequence similarities, they do share a defining structural identity, which has been termed “the four-helix bundle.” Click to see 3D structure and alignment of CNTF, LIF and IL-6. This structural identity, along with an additional specificity-determining receptor subunit, allows the neuroregulatory cytokines to bind with high affinity to their receptor complexes at the cell membrane and initiate signal transduction. Click to see the different receptor subunit combinations. In the case of CNTF, the neuroregulatory cytokine binds a dimer composed of gp130 and the related signal-transducing receptor subunit leukemia inhibitory factor receptor (LIFR), only in the presence of the specificity-determining receptor subunit CNTF-alpha. The signal-transducing receptor subunits lack intrinsic tyrosine kinase activity, however, formation of the receptor complex results in activation of the receptor-associated JAK tyrosine kinases. JAK activation is followed by phosphorylation of tyrosine residues within the cytoplasmic domains of gp130 and LIFR. Gp130 contains five tyrosine residues in its cytoplasmic region (designated Y1-Y5) that when phosphorylated by a JAK family member provide binding sites for src-homology region 2 (SH2) domain-containing effector proteins. Tyr759 (Y1) is the docking site for the SH2 domain-containing protein tyrosine phosphatase (SHP2) and suppressor of cytokine signaling-3 (SOCS-3). Tyr767, Tyr814, Tyr905, and Tyr915 (Y2-Y5) are binding sites for STAT3. Docking of STAT3 to the receptor complex results in STAT3 tyrosine phosphorylation by the associated JAK, which promotes STAT3 dimerization. Following an additional phosphorylation event on a serine residue (the kinase that mediates this phosphorylation is discussed below), the STAT3 dimer translocates to the nucleus and binds specific DNA sequences, resulting in enhanced transcription of STAT-responsive genes, such as the c-fos and neuropeptide vasoactive intestinal peptide (VIP) genes. Studies from this laboratory have clarified the role of SHP2 in CNTF signal transduction and have shown that SHP2 acts as a negative regulator of the duration of activation of the JAK → STAT3 pathway, while in parallel acting as a positive regulator of another as yet undefined pathway that contributes to c-fos induction.

Other studies from this laboratory have further extended our knowledge of the signaling pathways activated by the shared receptor complex for the neuroregulatory cytokines, and have shown that the mammalian target of rapamycin (mTOR, also known as FRAP or RAFT) is a critical component in these signaling pathways. mTOR is a large molecular weight protein (≈300 kDa) containing a variety of functional motifs including a serine/threonine protein kinase domain and a phosphatidylinositide 3-kinase-like (PI3K) catalytic domain that is closely related to the catalytic domain of DNA damage repair protein kinases. In mammalian and yeast cells the TOR proteins regulate cell growth in response to nutrient availability or growth factor or cytokine stimulation by controlling a diverse set of growth-related processes. Click to see a table summarizing the documented functions of the TOR proteins. The mechanisms by which TOR controls these diverse processes are not known in all cases; however, the mechanisms have been elucidated for transcriptional and translational control. For transcriptional control, we have shown that mTOR regulates maximal activation of the transcription factor STAT3. Maximal transcriptional activation of STAT3 by CNTF family members requires phosphorylation on residues Tyr705 and Ser72. The JAK tyrosine kinases are responsible for phosphorylation on Tyr705 and mTOR is responsible for phosphorylation on Ser727. For translational control, biochemical studies have shown that mTOR regulates multiple steps of the apparatus responsible for translation of a subset of mRNAs via activation of the ribosomal S6 kinase (p70S6 kinase) and inhibition of the eukaryotic initiation factor 4E (eIF4E) inhibitor 4E-BP1. Click to see mTOR signaling pathway.

CNTF cytokines and MS

A role for CNTF or LIF has been shown in slowing the progression of demyelination in the experimental autoimmune encephalomyelitis (EAE) for MS. The mechanism for how CNTF promotes remyelination in these studies is unclear; however, the authors of these studies have proposed the mechanism may be in promoting OPC maturation and limiting oligodendrocyte death. Supporting these animal studies human genotyping studies have demonstrated an association between a null mutation in the human CNTF gene and early disease onset and more rapid progression of MS.

CNTF and Lipid Metabolism

We have found that CNTF stimulation of mouse precursor oligodendrocyte MOP cells induces a dramatic increase in intracellular lipid storage droplets and robust incorporation of these newly synthesized lipids and fatty acids into the membrane of MOP cells. These findings indicate that CNTF has a profound effect on lipid homeostasis and suggest that CNTF may control the activity of the sterol inducible serum response element (SRE) binding proteins (SREBPs), which have been shown to be key regulators of genes responsible for both endogenous lipid biosynthesis and extracellular uptake of low-density lipoprotein (LDL) via up-regulation of the LDL receptor (ldlr) gene Click for Brown and Goldstein Laboratory. Taken together, these data provide supportive evidence for our proposal that CNTF exerts its promyelinating effect in rodent models of MS by regulating endogenous and exogenous mechanisms for controlling lipid homeostasis.
The significance of up-regulation of lipid metabolism and the potential therapeutic value this has for demyelinating diseases such as MS is as follows: Decades of research, most significantly from the laboratory of Brown and Goldstein, have shown that the SREBPs are critical regulators of cholesterol and fatty acid metabolism. Numerous other laboratories have contributed to a remarkable understanding of the biochemical and biophysical properties of the myelin membrane. These studies are summarized as follows: 1) In the CNS, the myelin sheath is an extension of the oligodendrocyte membrane and its synthesis as well as turnover of the myelin sheath requires a constant source of myelin-specific lipids and proteins for its synthesis. Cholesterol, which is highly enriched in myelin membranes is a particularly important membrane constituent because it provides fluidity and contributes to formation of “Rafts”, which are regions of the plasma membrane enriched in receptor molecules such as the LDLR. The cerebrosides galactocerebroside and sulfatide are specific for myelin and constitute nearly 70% of the myelin membrane and it is well established that these myelin membrane lipid components are critical for synthesis and homeostasis of the myelin sheath. 2) Although cholesterol can be synthesized de novo using intracellular biosynthetic pathways, an alternative source of cholesterol as well as other myelin membrane lipids is through an extracellular uptake pathway where it is taken up by cells via endocytosis of LDL by the LDLR. The LDLR is a cell surface trans-membrane protein that mediates the uptake and lysosomal degradation of extracellular LDL. Similarly, sphingomyelin, which is the building block component of galactocerebroside and sulfatide, can be brought into the cell via LDLR-mediated uptake of LDL. Although it is unclear the factor or factors which promote loss of lipid homeostasis in oligodendrocytes, it is clear, at least from our studies, that CNTF can promote the up-regulation of lipid homeostasis.

Animal demyelination model and CNTF

The most commonly used rodent model to study demyelination is one in which animals are immunized with components of the myelin sheath which results in experimental autoimmune encephalomyelitis (EAE). However, because we have hypothesized that the demyelination observed in MS is initiated by a disorder in lipid homeostasis we have chosen to use a copper chelator (cuprizone) model, which causes profound demyelination in the CNS but does not induce a T- and B-cell autoimmunity response. Copper is an essential trace element for several metalloenzymes including copper/zinc-dependent superoxide dismutase (CuZnSOD1) and the copper-dependent ferroxidase ceruloplasmin, which play critical roles in metabolism, inflammation, angiogenesis, and protection against oxidative stress. Inclusion of 0.2 % cuprizone (bis-cyclohexanone oxaldihydrazone, a copper chelator) in the diet of 8 week-old mice produces massive demyelination in several regions of brain, most prominently in the cerebellar peduncles and corpus callosum. If cuprizone is withdrawn from the animal’s diet after demyelination, remyelination is initiated where by 6 weeks of recovery myelin levels return to near-normal levels. The cellular basis for this remyelination during and following the cuprizone-induced metabolic challenge may involve recruitment of oligodendrocyte progenitor cells because these cells have been shown to proliferate and invade the demyelinated areas and following cuprizone withdrawal.