Enzymic hydrolysis of myelin basic protein and other proteins in central nervous system and lymphoid tissues from normal and demyelinating rats

Release of myelin basic protein-degrading proteolytic activity from microglia and macrophages after infection with Theiler's murine encephalomyelitis virus: comparison between susceptible and resistant mice

Theiler's murine encephalomyelitis virus (TMEV) produces a chronic inflammatory demyelinating disease in its natural host, the mouse. A delayed-type hypersensitivity (DTH) response to viral antigens generally correlates with susceptibility to the disease and is thought to play an important role in the pathogenesis of demyelination in this model of human multiple sclerosis (MS). The hallmark of DTH responses is the recruitment by activated Th-1 cells of lymphoid cells and especially macrophages in infected areas. It is believed that soluble factors released by these cells would produce tissue damage, particularly myelin breakdown. In the present study, we compared TMEV-infected macrophages and microglia, isolated from both susceptible SJL/J and resistant C57BL/6 mice, for their ability to secrete proteolytic enzymes capable of degrading myelin basic protein. In addition, we studied whether supernatants from infected microglia/macrophages were also capable of killing oligodendrocytes in the same in vitro system. As detected by SDS-PAGE, MBP-degrading proteolytic activity was found only in supernatants from infected SJL/J microglia and macrophages, but not in supernatants collected from infected C57BL/6 microglia and macrophages, or in supernatants from mock-infected SJL/J and C57BL/6 cells. Similarly, incubation of E20.1 cells, an immortalized line of oligodendrocytes, with infected SJL/J, but not C57BL/6 supernatants, resulted in cytotoxic activity. When cells from resistant C57BL/6 mice were treated with LPS, they became susceptible to infection and also secreted proteolytic enzymes. The proteolytic activity released from infected microglia and macrophages was found to be dose-dependent, was inactivated by heat, and was inhibited by phenylmethylsulphonyl fluoride (PMSF). These results indicate that a serine protease is released from infected microglia and macrophages and suggest a role for proteases in TMEV-induced myelin injury.

Myelin basic protein purified on an ion-exchange continuous polymer bed in the presence of ethylene glycol and salt possesses activity against p-nitrophenyl acetate

In this paper we describe a fast and mild method based on the use of a unique cation exchanger and buffers containing ethylene glycol and salt for the purification of the myelin basic protein (MBP; MW 18.5 kDa). MBP thus purified hydrolyses catalytically p-nitrophenyl acetate. This esterase activity facilitates not only the purification of MBP but also indicates that probably it is in its native state, i.e. there is a good chance that the purified molecules are structurally and chemically identical. This is a prerequisite to obtain crystals appropriate for x-ray diffraction and other studies.

Cation-mediated release and proteolytic cleavage of basic protein of isolated human myelin at acid pH

Myelin from human brain was incubated at pH 4.4 with metal salts, including KCl, NaCl, CaCl2, and MgSO4, to elicit cation-dependent autoproteolysis of myelin proteins. Incubation of myelin resulted in soluble proteolytic breakdown products of Mr smaller than those of the three original myelin basic proteins (MBPs). Comparable polypeptides were essentially absent from residual myelin. Proteolysis was strongly stimulated by increasing millimolar concentrations of K+, Na+, and Mg2+ and only moderately by Ca2+. Breakdown products were traced to MBP by immunostaining. Their origin from MBP was also indicated by identical electrophoretic cleavage patterns from endogenous myelin protein and exogenous MBP. All four metal salts, in addition to activating endogenous proteolysis, also caused a biphasic extraction of MBP. Electrophoresis of myelin revealed a quick initial and a slow further loss of protein, eventually leading to the removal of up to 78% of original MBP. The results are consistent with a concurrent extraction of MBP and activation of latent-bound acid protease activity by metal cations. It is therefore suggested that, in particular disease states, unfavorable changes in electrolytes and pH of white matter may cause a selective loss and proteolytic cleavage of MBP.

Cleavage of myelin basic protein by neutral protease activity of human white matter and myelin

Polypeptides arising from neutral in vitro proteolysis of myelin basic protein (MBP) of human brain were evaluated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. At pH 7 a marked breakdown of MBP resulted in the formation of 8-12 polypeptides ranging from 6 to 17 kd in molecular weight. As neutral proteolytic activity was not eliminated by either gel filtration or cation-exchange chromatography acid-soluble protease(s) involved probably have a size and electric charge similar to that of MBP. The enzymatic nature of neutral proteolysis was ascertained by heat inactivation and inhibition by alpha 2-macroglobulin. Incomplete inhibition of proteolysis and the failure of small peptides (less than 6 kd) to show up on electrophoresis seem to suggest that MBP was degraded by exopeptic proteases as well. Acid extracts of purified myelin yielded polypeptides similar to those of MBP of delipidated white matter. The results are consistent with a sequential limited proteolysis of MBP by neutral proteases probably associated with myelin and possibly related to the in situ catabolism of MBP in man.

Medullasin activity in granulocytes of patients with multiple sclerosis

Medullasin activity in mature granulocytes was measured in the blood of 22 patients with multiple sclerosis (definite type, 16; probable type, 6). The activity was elevated in every patient in relapse; it decreased to nearly normal levels at the beginning of improvement and further decreased to normal levels with remission. Serial determinations of the level of medullasin activity in 3 patients revealed that activity increased several days before the onset of acute exacerbation. Medullasin activity level in mature granulocytes obtained from patients with neurological diseases other than multiple sclerosis was largely within the normal range, except in 2 patients with spinocerebellar degeneration. Measurement of medullasin activity in mature granulocytes may become useful in both diagnosis and evaluation of multiple sclerosis.

Degradation of myelin proteins by brain endogenous neutral protease

A proteolytic enzyme isolated from calf brain cytosol, degraded purified myelin basic protein in the presence of Ca2+ at pH 7.0 (Singh, I. and Singh, A.K., Trans. Amer. Soc. Neurochem., 13 (1982) 119). This proteolytic enzyme also degrades basic proteins when incubated with intact myelin in the presence of Ca2+ at a neutral pH. Three hour treatment of purified myelin with this protease resulted in degradation of large basic protein and small basic protein by 73 and 89%, respectively. This proteolytic activity was inhibited by EDTA, leupeptin and low pH (pH 4.0), but was not affected by phenylmethylsulfonylfluoride, p-nitrophenylguanidinobenzoate and pepstatin A. Purified myelin preparations also contain small amounts of Ca2+-activated proteolytic activity. The fact that this is a neutral protease, endogenous to the brain, suggests that it may play a role in the degradation of myelin under physiological and pathological conditions.

Observations on the effects of protease inhibitors on the suppression of experimental allergic encephalomyelitis

Inhibitors of proteolytic enzymes were tested for their ability to suppress the clinical signs and CNS lesions produced by injection of purified myelin in complete Freund's adjuvant into Lewis rats. Pepstatin or a series of neutral protease inhibitors including aprotinin, soybean trypsin inhibitor, leupeptin, antipain, trans-aminomethyl cyclohexane carboxylic acid (AMCA), epsilon-amino caproic acid (EACA) nitrophenyl guanidino benzoate (NPGB), D- and L-polylysine, or a new commercial protease inhibitor, dipropionyl Rhein (DPR) were injected daily beginning on day 7 after immunization of rats with myelin. Aprotinin and soybean trypsin inhibitor exacerbated the symptoms and lesions of experimental allergic encephalomyelitis (EAE), leupeptin and antipain had no effect, and the plasminogen activators AMCA, EACA, NPGB, as well as poly-L- and poly-D-lysine and DPR suppressed various aspects of EAE. The measurement of acid protease as a biochemical method for quantitation of the degree of cellular infiltration into the CNS is proposed, and the results with the various treatments presented. AMCA and NPGB may exert their effects at the site of entrance of the lymphoid cells into the CNS.

The renal degradation of myelin basic protein peptide 43-88 by two enzymes in different subcellular fractions

Previous studies have demonstrated that the kidney is the major site for clearance and catabolism of a peptide (residues 43-88) of encephalitogenic or basic protein (BP) derived from central-nervous-system myelin. In the present investigation rat renal tissue was shown to be capable of degrading human BP peptide 43-88 over the pH range 4-11.5 with peaks of activity at pH5 and pH9. The enzymic activity at pH5 was localized mainly to the 5900 g pellet (crude mitochondrial fraction) and, on the basis of its sulphydryl features, was inferred to be cathepsin B. The enzyme activity at pH9 was greatly enriched in the 100 000 g pellet (microsomal brush-border fraction), and its sensitivity to inhibitors suggested that it was a metalloproteinase. The activity at alkaline pH in the 100 000 g pellet was stimulated 3-fold by non-ionic detergents and 20-fold by ATP and polyphosphates. Through a series of experiments the ATP stimulation of the alkaline proteinase activity was concluded to be the result of a reversal of inhibition imposed by the presence of another cationic protein, methylated bovine serum albumin. Inhibition by certain bivalent cations, the irregular effects of chelators and the effects of poly-L-lysine supported this conclusion. These studies indicate the availability of renal enzymes of different types and in different cellular compartments that are capable of degrading BP peptide 43-88. In particular, the relative amounts of bivalent cations, anions and charged proteins and peptides are likely to be major influences on the activity of the alkaline proteinase in vivo. The control of this degradation as well as the features of the smaller fragments of the peptide formed may determine biological and immune events subsequent to the release of this potentially autoantigenic material.

Degradation of myelin basic protein by cerebrospinal fluid: preservation of antigenic determinants under physiological conditions

Cerebrospinal fluid contains several proteolytic enzymes that can degrade myelin basic protein (BP) under physiological conditions into peptide fragments of various sizes which still contain antigenic determinants capable of binding antibodies to BP. These enzymes are optimally active in either acid (pH 4) or nuetral (pH 7 to 8) conditions and can be characterized by the nature of the BP peptide fragments produced. Proteinases resembling cathepsin D, thrombin, plasmin (fibrinolysin), or kallikrein are present in variable amounts in CSF. No relationship to any particular disease has yet been established.

Isolation of a product from the trypsin-digested glycoprotein of sciatic nerve myelin

When purified rabbit sciatic nerve myelin, whether lyophilized or not, is treated with low amounts of trypsin (25 microgram/ml) for 0.5, 3, or 24 h the resulting protein patterns viewed on sodium dodecyl sulfate (SDS) gel electrophoresis are similar. The most striking feature of the trypsinized myelin is the accumulation of a heavy band at the basic protein position, molecular weight 19 000, which is accounted for as a degradation product of the PO protein, referred to as the TPO protein. The PO protein, the major glycoprotein of sciatic nerve myelin, as well as the 23K and P2 proteins and albumin, an absorbed component, are all partially degraded; most high molecular weight bands are lost. The TPO protein, isolated by gel filtration in 2% SDS on an agarose column, like the PO protein, is highly insoluble in aqueous solvents. It is a glycoprotein (8% carbohydrate), staining with periodic acid-Schiff reagent; containing 3 mannose, 1 galactose, 3 N-acetylglucosamine, 1 sialic acid, and 1 fucose residues and is identical to the nonasaccharide of the parent PO protein. The amino acid composition of the TPO protein, is similar to the PO protein, but has a much higher content of hydrophobic residues and begins with NH2-methionine. This suggests that the PO protein is an amphipathic membrane protein in which its more polar character is confined to the first third of its NH2-terminus. This polar domain is probably positioned above the lipid leaflet where it is accessible to trypsin which cleaves a sensitive lysinyl (or argininyl)-methionine linkage. The more hydrophobic domain (the TPO protein) is buried in the myelin bilayer where it is protected from further tryptic attack. Thus trypsin can serve as a useful probe of myelin structure.

Degradation of basic protein in myelin by neutral proteases secreted by stimulated macrophages: a possible mechanism of inflammatory demyelination

In inflammatory demyelinating diseases such as multiple sclerosis and experimental allergic encephalomyelitis, myelin destruction occurs in the vicinity of infiltrating mononuclear cells. The observations that myelin can be altered prior to phagocytosis and in areas not contiguous with inflammatory cells suggests a common mechanism for the initial stages of demyelination. Because stimulated macrophages secrete several neutral proteases, including plasminogen activator, we have investigated the possibility that myelinolysis could be mediated directly or indirectly by these enzymes. Isolated myelin was incubated with conditioned media from cultures of thioglycollate-stimulated mouse peritoneal macrophages in the presence and absence of plasminogen. Myelin appeared to be vulnerable to attack by at least two proteolytic activities secreted by the macrophages, a plasminogen-dependent and a plasminogen-independent activity; of the major proteins in myelin, the basic protein was most susceptible. The direct myelinolytic activity of macrophage-conditioned media was abolished by EDTA, and the plasminogen-dependent hydrolysis was abolished by p-nitrophenylguanidinobenzoate, an inhibitor of plasminogen activator and plasmin. These results suggest that the plasminogen activator released by the stimulated macrophages generated plasmin which hydrolyzed basic protein in intact myelin. This interpretation was confirmed by the observation that urokinase, a plasminogen activator, in the presence of plasminogen brought about marked degradation of basic protein in myelin. We propose that the release of neutral proteases by stimulated macrophages involved in cell-mediated reactions, and its amplification by the plasminogen-plasmin system, may play a significant role in the demyelination observed in several inflammatory demyelinating diseases.

The action of trypsin on central and peripheral nerve myelin

In contrast to other studies, our results demonstrate that low concentration of trypsin degrades a high proportion of proteolipid from CNS myelin. The Wolfgram protein and BP are vulnerable and completely lost on trypsinolysis, perhaps accounting for some of the peptides retained by the myelin. In PNS myelin, the major PO protein, a hydrophobic glycoprotein, is readily degraded to a stable 18,000--19,000 molecular weight unit, referred to as TPO protein, still retaining the carbohydrate unit which probably exists as a nonasaccharide grouping. Production of the TPO glycoprotein results from cleavage of a lysinyl-methionine or arginyl-methionine linkage probably found approximately 80--100 residues from the NH2-terminal isoleucine of the PO molecule. This linkage must be especially accessible to trypsin since the TPO protein is also generated in high yield when isolated PO protein is treated with trypsin in solution for 0.5 hours. Further incubation for 24 hours fully degrades the TPO protein to over 20 tryptic peptides, shown by peptide mapping, unlike the situation in myelin where the TPO unit is stable and resists further proteolysis. The TPO unit is also produced when PO protein is treated with BrCN. The PO protein contains 3 methionine residues but presumably the methionine residue in the trypsin-sensitive region is crucial; cleavage leads to the same TPO unit minus NH2-terminal methionine. Another methionine residue also exists in the TPO protein but it may be resistant to BrCN cleavage or else occupy a near-end position. Other proteins were also identified on PAGE of trypsinized PNS myelin: albumin, P2 protein, and PO protein. Albumin and P2 protein were identified in the acidic extract by reaction with specific antibody. The PO protein was isolated; it moved similarly to standard protein on SDS-PAGE and gave the appropriate amino acid analysis. However, it cannot be determined at this time whether a portion of these proteins remains because they are partially inaccessible to trypsin, or else are slightly attacked and thus represent early stages of trypsinolysis. The P2 protein of trypsinized myelin appears to migrate slightly faster than standard P2 protein on PAGE. Further work should clarify this point. Amino acid analysis and sequence data show that the PO protein is particularly hydrophobic, very likely existing in PNS myelin as an amphipathic molecule which penetrates the bilayer but which has a hydrophilic portion exposed. It is this hydrophilic region that contains much lysine, particularly the crucial lysinyl-methionine linkage, that is so trypsin-sensitive. Determination of the amino acid sequence of terminal portions of the isolated PO and TPO proteins serves to firmly establish the PO protein as a unique entity probably exclusive to PNS myelin. It can be concluded that the study of trypsin activity toward PNS myelin has made possible a new understanding of how proteins are positioned in the membrane, and provided valuable insight into the PO protein.

Neutral proteinases secreted by macrophages degrade basic protein: a possible mechanism of inflammatory demyelination

In the inflammatory demyelinating diseases, such as multiple sclerosis, Landry-Guillain-Barré syndrome and experimental allergic encephalomyelitis, demyelination occurs in the vicinity of infiltrating mononuclear cells. Although the histopathology is characteristic of each disease, the general observation that myelin destruction in inflammatory lesions begins prior to phagocytosis suggests a common mechanism for myelinolysis in these diseases. Recent studies show that stimulated macrophages secrete several neutral proteinases, including plasminogen (Plg) activator. We have tested the possibility that these proteinases could, directly or indirectly, initiate myelin destruction. Isolated brain myelin was incubated with supernatant media from cultures of stimulated mouse peritoneal macrophages in the presence and absence of Plg. Cell supernatants alone caused some degradation of basic protein (BP) in myelin. The amount degraded was considerably enhanced in the presence of Plg. The other myelin proteins remained essentially intact. While the Plg-independent proteolytic activity in the supernatants was abolished by EDTA, known to inhibit the neutral proteinases, the Plg-dependent hydrolysis was inhibited by p-nitrophenylguanidinobenzoate, an inhibitor of Plg activator and plasmin. These results suggested that the Plg activator secreted by the macrophages generated plasmin, which selectively degraded BP. This interpretation was confirmed by the observation that urokinase, a Plg activator, plus Plg was effective in degrading BP in myelin. We propose that the action of neutral proteinases released by stimulated macrophages, and its amplification by the Plg-plasmin system, may play a significant role in several inflammatory demyelinating diseases; and that the relative specificity of these reactions for myelin lies in the extreme susceptibility of BP to proteolysis.

Turnover of myelin proteins in mouse brain in vivo

The incorporation of tyrosine into proteins was measured after the subcutaneous implantation of a pellet of [14C]tyrosine in mice. This method keeps the specific radioactivity of free tyrosine fairly constant and makes it possible to follow incorporation up to a 10-day period. At the end of 10 days most of the protein-bound tyrosine was replaced (i.e. most protein turned over) in lung, liver, heart, kidney and spleen; about half was replaced in brain, one-quarter in muscle. The rate of protein turnover in myelin was approx. 40% of that of whole brain proteins; at 10 days one-fifth of the myelin proteins were replaced. All protein components of myelin measured were in a dynamic state; incorporation decreased in the following order, Wolfgram greater than DM-20 greater than basic greater than proteolipid proteins. The incorporation of tyrosine into each protein fraction was greater in the 0-5-day than in the 5-10-day period, indicating heterogeneity of metabolic rates. The results show that after myelination at least a portion of each protein component of myelin is undergoing significant metabolic turnover. In the adult, myelin components are not stable, but turnover is heterogeneous, and each protein may be compartmentalized. Turnover can be influenced by a variety of factors.

Changes in CNS myelin proteins and glycoproteins after in situ autolysis

The effects of postmortem autolysis in situ on myelin proteins and glycoproteins were studied in 25- and 125-day-old mouse brain and in adult bovine brainstem. In bovine myelin a loss of the major myelin glycoprotein was the only difference observed when the tissue was left at 19 degrees C for 24 hours compared to immediately frozen material. In the autolysed mouse brain, the myelin major glycoprotein was the most affected component with a 55% decrease. Both myelin basic protein components were degraded with a 35% loss. The other myelin proteins did not change under the conditions used for this study. There was also no change in the specific activity of 2',3'-cyclic nucleotide 3'-phosphohydrolase, a myelin-associated enzyme. Using the double labelling technique with [3H]fucose and [3 5S] sulfate as precursors injected intracranially, a shift of the major myelin glycoprotein labelled with radioactive sulfate towards a smaller apparent molecular size was observed as a result of the autolysis whereas the electrophoretic mobility of the fucose labelled major peak was unaffected.

A lymph node neutral proteinase acting on myelin basic protein

A neutral protease present in inguinal and popliteal lymph nodes of rats with acute experimental allergic encephalomyelitis (EAE), rats injected with Freund's adjuvant, and rats that are normal has been found to hydrolyze basic protein present in purified brain and spinal cord myelin. The enzyme has been enriched by ammonium sulfate precipitation, and its properties have been studied. The protease activity toward different substrates was very specific and decreased in the following order: Protamine sulfate = polylysine (MW 183,000) > myelin basic protein > histone > polylysine (MW 2000) > polyarginine > cytochrome c. Other proteins including casein, freshly denatured hemoglobin, egg albumin, bovine serum albumin, and ribonuclease were ineffective as substrates. The pH curve showed a peak at pH7 for rat myelin, isolated beef basic protein, and histone. A possible role for this enzyme in demyelination in acute experimental allergic encephalomyelitis is suggested.