Daniela Veber

Experimental and clinical evidence of the role of cytokines and growth factors in the pathogenesis of acquired cobalamin-deficient leukoneuropathy

Our experimental and clinical studies have highlighted the non-coenzyme functions of cobalamin (Cbl; vitamin B12). The neuropathy of the rat central nervous system (CNS) due to Cbl deficiency is associated with increases in CNS tissue and/or cerebrospinal fluid (CSF) levels of some neurotoxic molecules, and decreases in local and/or CSF levels of some neurotrophic molecules. The increased molecules are nerve growth factor (NGF), tumor necrosis factor (TNF)-alpha, and the soluble (s)CD40:sCD40 Ligand dyad; the decreased molecules are epidermal growth factor (EGF) and interleukin-6. The morphological lesions of the CNS white matter in Cbl-deficient (Cbl-Df) rats can be prevented to the same extent by treatments replacing Cbl or the deficient neurotrophic molecules, or treatment with agents that antagonize the excess neurotoxic molecules. Patients with neurological and/or hematological manifestations of severe Cbl deficiency also have high TNF-alpha levels and low EGF levels in CSF and serum. Cbl replacement treatment corrects cytokine and growth factor abnormalities in Cbl-Df patients and Cbl-Df rats, and so Cbl-Df CNS neuropathy is also due to an imbalance in local cytokine/growth factor networks. TNF-alpha and NGF levels are also increased in Cbl-Df rat liver, which is morphologically unaffected by Cbl deficiency. The increases in TNF-alpha and NGF levels increase nuclear factor-kappaB activity levels in both the CNS and liver, and this indirect regulation supports the idea that Cbl may modulate the expression of some cytokine/growth factor genes in rat CNS and other tissues. Finally, we have tried to harmonize our pathogenetic theory of cytokine and growth factor dysregulation with the biochemical interpretation.

New basis of the neurotrophic action of vitamin B12.

Abstract Over the last few years we have reproduced all of the key morphological and biochemical features of human subacute combined degeneration in the central nervous system and peripheral nervous system of rats made cobalamin-deficient by means of total gastrectomy or a chronic cobalamin-deficient diet. We have also recently clarified the pathogenesis of experimental subacute combined degeneration induced in the rat by cobalamin deprivation. The results of our studies strongly support the notion that cobalamin plays a pivotal role in regulating the balance of the network of cytokines and growth factors in the central nervous system of the rat. We have demonstrated that cobalamin tightly regulates the central nervous system synthesis and/or the cerebrospinal fluid level of two cytokines, tumor necrosis factor-αand interleukin-6, and a growth factor, epidermal growth factor. Of these neuroactive agents, one, tumor necrosis factor-α, is neurotoxic, whereas the others are neurotrophic. Therefore, it becomes clear that cobalamin-deficient central neuropathy is caused not by the withdrawal of the vitamin, but reflects a locally increased production of neurotoxic agents, combined with the locally decreased production of neurotrophic agents.

Cobalamin (vitamin B12) positively regulates interleukin-6 levels in rat cerebrospinal fluid

We have previously demonstrated that the repeated intracerebroventricular (i.c.v.) microinjection of interleukin-6 (IL-6) prevented the myelinolytic lesions of cobalamin-deficient (Cbl-D) central neuropathy [or subacute combined degeneration (SCD)] in totally gastrectomized (TGX) rats. We therefore hypothesized that cobalamin (Cbl) may actually regulate IL-6 levels in rat cerebrospinal fluid (CSF). We measured IL-6 levels in the CSF of rats made Cbl-D by means of total gastrectomy (TG) or chronic feeding with a Cbl-D diet and killed at different times from the beginning of the experiment, and found that IL-6 levels significantly and progressively decreased over time. Chronic 2-month Cbl administration started 1 week after surgery prevented the decrease in IL-6 levels and, when it was started 2 months after surgery, it significantly increased IL-6 levels, but not to presurgical values. We also investigated whether IL-6 decrease might be ultimately due to the Cbl-deficiency-linked decrease in epidermal growth factor (EGF) synthesis. Repeated i.c.v. administrations of EGF to TGX rats did not modify CSF IL-6 levels. These results, together with those of a previous study showing the preventive effect of IL-6 treatment on SCD lesions, demonstrate that: (i) Cbl selectively regulates CSF IL-6 levels; and (ii) decreased IL-6 availability plays a role in the pathogenesis of the experimental SCD, in which no evidence of inflammatory and/or immunological reaction has been observed.

Increased spinal cord NGF levels in rats with cobalamin (vitamin B12) deficiency.

We have recently demonstrated that the neuropathological morphological alterations caused by cobalamin (Cbl) deficiency in the rat central nervous system are related to the vitamins inability to modulate the synthesis of some neurotoxic and neurotrophic agents in opposite directions. In the present study, we measured nerve growth factor (NGF) levels in the spinal cord (SC) and cerebrospinal fluid (CSF) of rats made Cbl-deficient (Cbl-D) by means of total gastrectomy (TG) or a Cbl-D diet. In both cases, Cbl deficiency increased SC and CSF NGF levels after the appearance of myelinolytic lesions in the SC white matter (SCWM) (i.e. after the second post-TG month), and these changes were normalised by Cbl treatment in the 4-month-totally-gastrectomised (TGX) rats. Intracerebroventricular (i.c.v.) anti-NGF-antibody treatment prevented the onset of the myelinolytic SCWM lesions in the 2-month-TGX rats (i.e. when SC and CSF NGF levels are still normal) and normalised the ultrastructure of the SCWM in the 4-month-TGX rats, which was however worsened by the i.c.v. administration of NGF. These findings demonstrate that: (i) Cbl deficiency increases SC and CSF NGF levels; and (ii) endogenous NGF seems to play a noxious role in the progression of rat Cbl-D central neuropathy.

Indirect down‐regulation of nuclear NF‐κB levels by cobalamin in the spinal cord and liver of the rat

We used electrophoretic mobility shift assays to investigate the effects of cobalamin (Cbl) deficiency on the levels of activated nuclear factor‐kappa B (NF‐κB) in the spinal cords (SCs) and livers of rats made Cbl‐deficient (Cbl‐D) by total gastrectomy or a Cbl‐D diet. We chose the SC and liver because they are severely or scarcely affected, respectively, by Cbl deficiency in terms of histological damage. We found permanently increased NF‐κB levels (particularly the p50 and p65 subunits) in the SCs and livers of both types of Cbl‐D rats, and Western blot analysis demonstrated increased p65 levels. NF‐κB and p65 protein levels normalized when the totally gastrectomized (TGX) rats were treated with Cbl replacement. As we have previously demonstrated that Cbl deficiency increases tumor necrosis factor (TNF)–α and nerve growth factor (NGF) levels in the SC (each of which is a known NF‐κB activator), we redetermined NF‐κB levels in the SCs and livers of TGX rats treated with anti‐TNF‐α or anti‐NGF antibodies and found that NF‐κB levels normalized in both tissues after either treatment. These results demonstrate that: (1) Cbl physiologically and indirectly down‐regulates NF‐κB levels in rat SC and liver, and (2) NF‐κB is an important signaling molecule after Cbl deficiency injury.

Cobalamin (vitamin B12)-deficiency-induced changes in the proteome of rat cerebrospinal fluid.

We studied the changes in the proteome of CSF (cerebrospinal fluid) in two animal models of Cbl (cobalamin) deficiency: TGX (totally gastrectomized) rats and rats fed a Cbl-D (Cbl-deficient) diet. Two-dimensional PAGE was used to detect qualitative and quantitative variations in proteins in the CSF samples. The peak increase in total CSF protein concentration was observed 4 months after TG (total gastrectomy) and after 6 months of eating a Cbl-D diet. There is a specific increase 4 months after TG in the spots corresponding to alpha1-antitrypsin and the de novo presence of thiostatin and haptoglobin beta. Cbl-replacement treatment in 4-month-TGX rats corrected these alterations in the CSF proteome. However, most of the CSF proteome alterations attenuated in Cbl-untreated 8-month-TGX rats and in rats fed a Cbl-D diet for 16 months. Transthyretin concentration varied slightly in the CSF of both types of Cbl-D rat, whereas the relative abundance of prostaglandin D synthase rose sharply in the CSF of the rats fed a Cbl-D diet for 16 months. We have demonstrated previously that the histological and ultrastructural CNS (central nervous system) damage in both types of Cbl-D rat appears within 2-3 months of Cbl deficiency, and thus appears to precede the alterations in the CSF proteome. The CSF proteome patterns of rats in which phlogosis was induced in or outside the CNS are quite different from those of the CSF of Cbl-D rats. All these findings demonstrate that the alterations in the CSF proteome of Cbl-D rats are specifically linked to Cbl deficiency.

Decreased GFAP-mRNA expression in spinal cord of cobalamin-deficient rats

We have demonstrated previously that chronic vitamin B12 [cobalamin (Cbl)] deficiency preferentially affects glial cells in the rat central nervous system (CNS) and severely affects peripheral glial cells independently of and concomitantly with the central neuropathy. In this study, we determined the mRNA levels for myelin basic protein (MBP) and glial fibrillary acidic protein (GFAP) in different CNS areas of rats made Cbl‐deficient by total gastrectomy, as well as the mRNA levels for glycoprotein Po and peripheral myelin protein (PMP)22 in the sciatic nerve. GFAP‐mRNA levels were significantly decreased in the spinal cord (SC) and hypothalamus, but not in the cortex, hippocampus, or striatum of totally gastrectomized (TGX) rats. No differences in GFAP protein levels were found in the SC and hypothalamus of the TGX rats treated or not with Cbl. MBP‐mRNA levels were significantly decreased only in the hypothalamus, and the levels of mRNA for both glial markers returned to normal with Cbl replacement therapy. The levels of mRNA for the various myelin proteins in the sciatic nerve were not modified by Cbl deficiency. These results demonstrate that: a) the neurotrophic action of Cbl in rat CNS occurs in a zonal manner; and b) Cbl deficiency does not affect myelin synthesis (with the sole exception of the hypothalamus).

Loss of epidermal growth factor regulation by cobalamin in multiple sclerosis.

We investigated whether the physiological regulation of cerebrospinal fluid (CSF) levels of tumor necrosis factor (TNF)-alpha, epidermal growth factor (EGF), and nerve growth factor (NGF) by cobalamin (Cbl) that is observed in rat and human central nervous system (CNS) is retained in the CSF of patients with multiple sclerosis (MS). The study involved 158 MS patients grouped on the basis of the different clinical courses (relapsing-remitting (RR), secondary-progressive (SP), and primary-progressive (PP)), and 76 gender- and age-matched control patients with other non-inflammatory and non-neoplastic neurological diseases. The MS patients were therapy-free at the time of lumbar puncture. CSF Cbl and EGF were blindly measured by means of radioimmunoassays, and CSF TNF-alpha, and NGF by means of highly sensitive enzyme-linked immunosorbent assays. Serum EGF was also measured in 38 of the MS patients and 20 healthy controls. CSF Cbl levels were significantly higher (RR patients 27.9+/-9.7 pg/ml, p<0.0001 vs. C; SP patients 25.4+/-8 pg/ml, p<0.02 vs. C), and CSF TNF-alpha and EGF levels significantly lower in the patients with the RR (TNF-alpha 28.3+/-23.4 x 10(-3) pg/ml, p<0.0001 vs. C; EGF 129.9+/-44.8 pg/ml, p<0.02 vs. C) or SP (TNF-alpha 20.5+/-20.5 x 10(-3) pg/ml, p<0.001 vs. C; EGF 116.5+/-24.8 pg/ml, p<0.05 vs. C) clinical course than in controls (Cbl 21+/-4.6 pg/ml; TNF-alpha 75.6+/-34.7 x 10(-3) pg/ml; EGF 170.2+/-54.8 pg/ml). There were no differences in CSF NGF or serum EGF levels between any of the MS clinical courses and controls. Our results indicate that: (a) the positive Cbl-mediated regulation of myelino- and oligodendrocyte-trophic EGF is lost in the CSF of RR- or SP-MS patients; (b) the decrease in EGF levels in the CSF may be one factor impeding CNS remyelination in MS; and (c) the PP clinical course may have different pathogenetic mechanism(s) also on the basis of the molecules investigated in this study.

Increased levels of the CD40:CD40 ligand dyad in the cerebrospinal fluid of rats with vitamin B12(cobalamin)-deficient central neuropathy

The levels of the soluble (s) CD40:sCD40 ligand (L) dyad, which belongs to the tumor necrosis factor (TNF)-alpha:TNF-alpha-receptor superfamily, are significantly increased in the cerebrospinal fluid (CSF), but not the serum of cobalamin (Cbl)-deficient (Cbl-D) rats. They were normalized or significantly reduced after treatment with Cbl, transforming growth factor-beta1 or S-adenosyl-L-methionine, and the normal myelin ultrastructure of the spinal cord was concomitantly restored. The concomitance of the two beneficial effects of these treatments strongly suggests that the increases in CSF sCD40:sCD40L levels may participate in the pathogenesis of purely myelinolytic Cbl-D central neuropathy in the rat. In keeping with this, an anti-CD40 treatment prevented myelin lesions.

Cobalamin deficiency-induced down-regulation of p75-immunoreactive cell levels in rat central nervous system

We investigated immunoreactivity for p75 neurotrophin receptor (NTR) in the spinal cord white matter and septum of rats made cobalamin-deficient (Cbl-D) by means of total gastrectomy or a Cbl-D diet. Cbl deficiency down-regulates p75NTR-immunoreactive cell levels in spinal cord white matter and septum with different time courses. On the whole, the spinal cord white matter seems to be more affected in terms of p75NTR-immunoreactive cells, most of which are astrocytes. The p75NTR-immunoreactive cell levels in the spinal cord white matter and septum normalized in rats treated with Cbl (scheme b) and killed 4 months after total gastrectomy. However, Western blot analysis of p75NTR in the spinal cords of Cbl-D rats shows increased p75NTR protein levels, which are resistant to Cbl replacement. These findings demonstrate that a neurotrophic vitamin (Cbl) positively regulates the levels of a neurotrophic receptor (p75NTR) (at least in terms of immunohistochemistry) in rat central nervous system, although the underlying mechanism(s) are still unknown.