EXPERIMENTAL
NEUROLOGY
110,
121-126 (1990)
Proteinase Inhibitor a,-Antichymotrypsin Has Different Expression in Various Forms of Neuronal Ceroid Lipofuscinosis KRYSTYNA New York State in Developmental
E. WISNIEWSKI
Ofice of Mental Retardation Disabilities, Department
AND ELIZABETH
KIDA
and Developmental Disabilities, Institute for Basic Research of Pathological Neurobiology, Staten Island, New York 10314
Defective proteolytic degradation is most widely maintained as the major pathogenetic factor in neuronal ceroid lipofuscinosis (NCL). The goal of the present study was to examine the expression in NCL brain tissue of one of the serine proteinase inhibitors, al-antichymotrypsin. Our study was based on previous llndings of a,-antichymotrypsin association with CNS amyloidoses related to amyloid /3 protein deposits and our previous findings suggesting abnormal processing of amyloid B-protein precursor (ABPP) in NCL brains. Immunocytochemical study was performed on formalinfixed brain tissues collected from 15 NCL cases representing four different forms of the disorder and from 16 control cases comprising age-matched controls, older nondemented individuals, and Alzheimer disease cases. Our present study has shown that the expression of qantichymotrypsin is generally higher in NCL cases than in control cases; however, it manifests in distinct variations of intensity and proportions of immunostained cells. The strongest immunoreactivity was found in the infantile form of NCL, which is characterized by a rapid clinical course and widespread tissue damage. We found no evidence of direct involvement of a,-antichymotrypsin in either the ceroid lipopigment accumulation or the abnormal processing of ABPP in NCL. However, our findings may reflect the heterogeneity of the pathomechanism underlying this group of disorders and suggest that, similarly to blood circulation, al-antichymotrypsin can also represent an acute-phase protein in brain tissue. @ isso Academic PWS, k.
INTRODUCTION The neuronal ceroid lipofuscinoses (NCL) represent hereditary disorders manifesting lysosomal lipopigment storage material within cells of the central nervous system (CNS) and other organs. On the basis of the age of onset, clinical presentation, and pathological features the following four main forms of the disease are distinguished: infantile (INCL), late infantile (LINCL), juvenile (JNCL), and adult (ANCL) variants (29). However,
in addition, some less frequently encountered subtypes have also been described (7). There is general agreement that NCL is caused by a lysosomal dysfunction, but the proper biochemical defect remains unknown. Recent studies supplied data indicating that the well-known clinicopathological differences in forms of NCL might result from their genetic heterogeneity. In JNCL, the site of disease locus was located on the long arm of chromosome 16 (8). Cathepsin H deficiency, whose gene was localized to chromosome 15, was found in some LINCL cases (6, 27), whereas in individual INCL cases, lysosomal phospholipase A, deficiency, whose gene has not yet been cloned, was described (6). In lipopigments isolated from LINCL and JNCL cases and from sheep model of the disease, the storage of mitochondrial ATP synthase subunit C was disclosed (14). The common postulation of defective lysosomal proteolysis, as the basic lesion in NCL is supported by experimental findings disclosing an accumulation of lipofuscin-like substance after application of cysteine proteinase inhibitors (10). Recently, the question was also raised of the participation of proteinase inhibitors in the abnormal processing of amyloid P-protein precursor (ABPP) in Alzheimer disease (AD) cases (1, 12, 17, 21, 28). It is of interest in this regard that dense concentrations of ABPP were found in secondary lysosomes (5) and, moreover, that neuronal aging lipofuscin shows immunostaining with antibodies to fragment of amyloid P-protein (3). Our previous studies have disclosed strong neuronal immunoreactivity in the NCL brain tissues that accumulated ceroid lipopigment with antibodies generated against various domains of ABPP, including amyloid ,&protein’s epitopes (13,24,25). These findings suggest abnormal processing of ABPP also in the NCL brain tissue, as is postulated for AD. Thus, on the basis of the data pointing to the putative role of proteinase inhibitors in both ABPP processing and lipopigment storage, we have undertaken the present study to examine the immunoreactivity of the NCL brain tissue with antiserum to serine proteinase inhibitor, ai-antichymotrypsin (ACT). Our goal is to provide further in-
121 All
Copyright 0 1990 rights of reproduction
0014-4886/90 $3.00 by Academic Press, Inc. in any form reserved.
122
WISNIEWSKI
sight into the presumed tion in the NCL brain. MATERIAL
abnormal
AND
proteolytic
degrada-
METHODS
Immunocytochemistry was performed on formalinfixed, paraffin-embedded, 5-pm-thick tissue sections collected from 15 NCL cases comprising four different forms of the disease [(4 INCL (20 months-5 years), 2 LINCL (6-12 years), 8 JNCL (lo-25 years), and 1 ANCL (56 years)]. Clinicopathological features of some of the NCL cases have been previously reported (26). Additional material included autopsied brain tissues from 10 normal controls, age 12 months to 51 years; 3 nondemented older controls, age 65 to 85 years; and 3 proven AD cases, age 51 to 85 years. The avidin-biotin-peroxidase complex method was used as the staining procedure. Hydrated tissue sections were incubated for 30 min in 0.2% hydrogen peroxide in methanol to block endogenous peroxidase, pretreated with 0.05% trypsin/CaCl, in Tris buffer, pH 8 (Sigma Chemicals, St. Louis) for 30 min at 37°C and then incubated with 10% normal goat serum in PBS for 30 min at room temperature. Specific rabbit antiserum to human cu,-antichymotrypsin was purchased from Accurate Chemical and Scientific Corp. The sections were incubated with the primary antiserum (diluted 1:500) overnight at 4°C. Procedure controls involved the substitution of the primary antiserum by PBS and nonimmune species-specific serum and the use of GFAP-antibodies (1:500) (Dako Corp., Carpinteria, CA). Biotinylated goat anti-rabbit antibodies (1:200) followed by avidinbiotin-peroxidase complex (Vector Laboratories, Burlingame) were applied to the sections, each for 30 min. The reaction product was visualized with diaminobenzidine (Sigma Chemicals, St. Louis) in the presence of 0.01% hydrogen peroxide. The sections were counterstained with hematoxylin, dehydrated, and mounted with Permount. RESULTS In the NCL brain tissue, the antiserum to ACT stained various types of CNS cells, including both the perikarya and nuclei of numerous neurons and astrocytes. However, generally, the intensity of the staining and the percentage of immunoreactive neuronal and glial cells varied distinctly among not only the different forms of NCL studied but also the individual cases within the same form of the disease. The strongest immunoreactivity presented nerve cells located in the cerebral cortex of two INCL cases. Typically in INCL, a prominent neuronal loss in the cortical mantle is observed: however, in these two cases, virtually all surviving neurons showed large, coarse granular material in their distended cytoplasm (Fig. la). Immunostaining of
AND
KIDA
neurons in the other brain areas was somewhat weaker and was confined to the presence of a finely granular reaction product. Neuronal nuclei in INCL cases were rarely stained as compared with the cases of the other forms of the disease examined. In LINCL, JNCL, and ANCL, the reaction product in neuronal perikarya was represented solely by finely granular deposits. The intensity of the neuronal staining in these forms of NCL varied, and often, even in close proximity to strongly labeled neurons, unstained nerve cells could be found (Fig. lb). Some of the affected neurons that were overloaded with lipopigment material presented finely granular material not only in the perikarya but also in thin cellular processes (Fig. lc) and sporadically in enlarged proximal axons. Numerous neurons with cytoplasm distended by accumulated lipopigment showed, in addition to perikaryal staining, heavily stained nuclei (Fig. Id). Nuclear staining was observed in some of the neurons manifesting weak or no immunoreactivity (Fig. le). Some neurons overloaded with stored lipopigment were not immunoreactive (Fig. If). Generally, the proportion of immunostained neurons was the greatest in INCL and only slightly smaller in ANCL. In JNCL, variations in both the intensity of immunostaining and the percentage of labeled neurons between particular cases were the most prominent. In this form of NCL, apart from the cases with a large number of moderately stained neurons, the cases with only a few immunoreactive neurons, usually confined to the hippocampus, some neocortical areas, the basal ganglia, and the granular layers of the cerebellar cortex, could be found. Astroglial cells in all the NCL cases studied showed immunostaining of the nuclei or perikarya, or both. The strongest immunoreactivity of the astrocytes was seen in the grey matter in INCL cases. In these cases, the reaction product presented large, coarse granules in the cytoplasm of numerous astrocytes (Fig. 2a): less frequently, such strong immunoreactivity was seen in the astrocytes located in the white matter, where light immunostaining of the perikarya of the astroglial cells predominated. In LINCL, JNCL, and ANCL, immunostaining of the perikarya of astrocytes in the grey matter was rarely observed, however, their nuclei often showed immunoreactivity (Fig. 2b). In some NCL cases however, numerous reactive astrocytes with moderately stained perikarya were observed in the hippocampal region. There was no close correspondence between the intensity and the number of immunostained astrocytes in LINCL and JNCL and the degree of tissue destruction. Even in cases with advanced neuronal degeneration, only a few reactive astrocytes with immunoreactive cytoplasm could be found in affected areas. Moreover, in the above-mentioned NCL forms, larger numbers of immunostained astrocytes were observed in the white rather than in the grey matter (Fig. 2~). In AD cases, numerous neurons in cerebral cortex
a,-ANTICHYMOTRYPSIN
IN
NEURONAL
CEROID
LIPOFUSCINOSIS
123
FIG. 1. ACT immunostaining of nerve cells in different NCL forms. (a) Coarse granular material in the perikaryon of distended neuron in cerebral cortex. INCL case, x750. (b) Large pyramidal neurons in the cerebral cortex strongly labeled by finely granular reaction product, among them unstained neurons. ANCL case, x300. (c) Finely granular reaction product in the perikarya and processes of cortical neuron overloaded with lipopigment. ANCL case, x750. (d) Heavily immunostained nucleus in a neuron with moderate perikaryal staining. ANCL case, x300. (e) Neurons overloaded with lipopigment from hippocampal CA1 area showing weak perikaryal immunoreactivity and nuclear staining. JNCL, x200. (f) Unstained neurons manifesting stored lipopigment in inferior olivary nucleus. JNCL, X200.
showed moderate immunostaining, including both neuronal perikarya and their nuclei. In three AD cases, sporadically in ANCL cases, and in one case of older controls, moderate immunostaining of neuritic plaques was seen (Fig. 2d). Moderate immunostaining of some blood vessels was observed in AD cases. The percentage of astrocytes with immunostained perikarya was still greater in white than in grey matter; however, that discrepancy was smaller in AD cases than in NCL cases. In all the cases studied, including NCL, normal controls, and AD cases, heavy immunoreactivity of the epithelium of the choroid plexus, when available, was present (Fig. 2e). In normal control individuals, the pattern and intensity of immunostaining varied between particular cases. Generally, however, both neurons and astrocytes showed immunoreactivity less frequently than in NCL and AD cases. Immunostaining, usually of weak or moderate intensity, was seen, mostly in large neurons of the cerebral cortex, mainly in the hippocampus (Fig. 2f ), in the medulla, and in the granular layer of cerebellar cor-
tex and less frequently in the Purkinje cells. Nuclei and perikarya of astrocytes were clearly less frequently stained in both grey and white matter. The pattern of immunostaining of the ependymal cells and of pericytes varied between particular cases, regardless of the group of cases studied. Semiquantitative analysis of neuronal and glial cells immunolabeling is summarized in Fig. 3. DISCUSSION
Our present study has shown that the same types of CNS cells in brains of NCL and control cases were immunostained with antiserum to ACT. Proportions and intensity of immunostaining varied depending on the particular brain area and the case studied. The strongest immunoreactivity in all cases examined presented epithelial cells of the choroid plexus. Immunostaining of neuronal perikarya in both NCL and control cases was most often seen in the cerebral cortex, especially in the hippocampus, in the medulla, and in the cerebellar cortex. Nuclear staining of both neurons and glial cells was
124
WISNIEWSKI
AND
KIDA
FIG. 2. ACT immunostaining in NCL and control cases. (a) Heavily immunostained astroglial perikarya in cerebral cortex in X750. (b) Numerous astrocytes with stained nuclei, but unstained perikarya in the cerebral cortex in ANCL case. x200. (c) manifesting nuclear and perikaryal staining in the cerebral white matter of JNCL. x300. (d) Immunoreactivity of neuritic plaques cortex in AD case. X300. (e) Strong immunostaining of choroid plexus epithelium. Control case, X300. (f) Various intensities immunostaining in hippocampal end-plate. Control case, X200.
also found, resulting from ACT binding to DNA and nucleosomes (6,11,16). A similar pattern of ACT immunostaining distribution was described previously in normal human CNS (11). However, our present study disclosed a generally higher percentage and intensity of perikaryal neuronal and astroglial immunostaining in NCL cases than in controls. The proper physiological function of ACT in CNS still remains to be clarified. ACT is one of the human plasma serine proteinase inhibitors, which are involved in regulation of such processes as coagulation, fibrinolysis, or inflammation, and which by inhibiting the activity of their target proteinases may prevent tissue damage. ACT represents a major acute-phase protein whose plasma concentration increases rapidly in various pathological conditions (23). This serine proteinase inhibitor forms equimolar complexes with its target enzyme: neutrophil cathepsin G, with chymotrypsin and mast cells “chymases” (4, 19). However, the question was raised whether the true substrate for ACT has really been found, because cathepsin G was considered an inefficient proteinase of slow hydrolytic activity (22). There are data indicating that apart from the liver,
INCL case. Astrocytes in cerebral of neuronal
which represents a major source of ACT in the circulation and some extrahepatic sites, ACT can be synthesized in situ in the CNS. ACT messenger RNA was demonstrated and found to be greatly elevated in AD grey matter compared with controls (1) and produced mostly by astrocytes (15). Therefore, it was suggested that if only astrocytes were ACT producers, those neurons that manifested ACT immunoreactivity might take up ACT either from the extracellular fluid or from the blood, and that the observed increase in immunoreactivity of astrocytes in the grey matter might reflect a general, nonspecific response of the brain to different disease processes that result in neurodegeneration (2). Our present study has shown that NCL brain tissue immunoreactivity is distinctly varied, strong in INCL cases and showing the greatest variations in JNCL, where in some individuals only slight immunoreactivity confined to only a few neurons and reactive glial cells was observed. These results contrast with our previous observations disclosing the strongest immunoreactivity of JNCL and ANCL brain tissue with antibodies to amyloid P-protein (25). Moreover, we could not find in our material any close correlation between the degree of
or-ANTICHYMOTRYPSIN
IN
NEURONAL
CEROID
125
LIPOFUSCINOSIS
circulation exist and, therefore, that also in the CNS ACT represents an acute-phase protein particularly, strongly elevated in rapidly evolving pathological processes with abnormal proteolytic activity. Greatly increased ACT immunoreactivity in INCL might be related to the mechanisms operating specifically only in this form of NCL, differing from the other NCL forms, and might reflect an etiopathogenetical heterogeneity for these disorders.
1
2
3
4
5
6
ACKNOWLEDGMENTS
7
GROUP FIG. 3. Semiquantitative analysis of neuronal and glial cells’ immunoreactivity. Results show the average numbers of immunostained neurons (black columns) and of glial cells (white columns) per 1 mm2 counted in 10 nonoverlapping cortical areas from INCL (column I), LINCL (column 2), JNCL (column 3), ANCL (column 4), age-matched and old controls (columns 5 and 6, respectively), and AD (column 7). The small number of immunostained neurons in column 1 results from severe cortical atrophy found in these cases.
lipopigment accumulation and ACT immunoreactivity as even neurons severely overloaded with lipopigment showed various intensities of staining or were not immunoreactive. Therefore, the direct participation of ACT in the pathogenesis of lipopigment storage and abnormal processing of ABPP in NCL is very unlikely. ACT was previously found in amyloid deposits in amyloid /3protein-related CNS amyloidoses (1, 16). Other serine/ cysteine proteinase inhibitors are able to induce neurodegenerative changes in CNS and accumulation of lipofuscin-like substance (10, 20). Nevertheless, it is well documented that some serine protease inhibitors can produce a protective effect on various tissues due to the inhibition of uncontrolled proteolysis and that their deficiency causes tissue damage as is evidenced in the case of a,-protease inhibitor (9). Moreover, they are supposed to be responsible for efficiency of proteolytic processing and normal sorting of lysosomal enzyme precursors to lysosomes (18). INCL is characterized by a rapid clinical course and the most severe tissue destruction. Strong ACT immunostaining, as was disclosed in INCL, with heavy astroglial immunoreactivity, suggests increased ACT synthesis. It may reflect a secondary and unspecific response of the CNS to severe and widespread tissue injury, followed by the activation of protective mechanisms. These mechanisms, however inefficient, might be directed to prevent uncontrolled proteolysis and subsequent progression of tissue destruction. It is of interest in this regard that in LINCL and JNCL even in cases in which prominent tissue degeneration and neuronal loss were found, ACT expression was distinctly weaker than in INCL cases. These discrepancies may indicate that in the CNS, mechanisms of ACT synthesis similar to those found in blood
We thank Ms. Madeline Tinney for her secretarial assistance, Maureen Stoddard Marlow for critically reading the manuscript, Mr. Lawrence Black for bibliographical editing assistance. This was supported by NIH Grant NS23717.
Ms. and work
REFERENCES 1.
2.
ABRAHAM, C. R., D. J. SELKOE, AND H. POTTER. 1988. Immunochemical identification of the serine protease inhibitor cY,-antichymotrypsin in the brain amyloid deposits of Alzheimer’s disease. Cell 62: 487-501. ABRAHAM, C. R., T. SHIRAHAMA, AND H. POTTER. 1990. cY,-antichymotrypsin is associated solely with amyloid deposits containing the beta-protein. Amyloid and cell localization of ol,-antichymotrypsin. Neurobiol. Aging 11: 123-129.
3. BANCHER,
C., I. GRUNDKE-IQBAL, K. IQBAL, K. S. KIM, AND H. M. WISNIEWSKI. 1989. Immunoreactivity of neuronal lipofustin with monoclonal antibodies to the amyloid P-protein. Neurobiol. Aging 10: 125-132.
4. BEATTY,
K., J. BIETH, AND J. TRAVIS. 1980. Kinetics of association of serine proteinases with native and oxidized cu,-proteinase inhibitor and cu,-antichymotrypsin. J. Biol. Chem. 255: 39313934.
5. BENOWITZ,
L. I., W. RODRIQUEZ, P. PASKEVICH, E. J. MUFSON, D. SCHENK, AND R. L. NEVE. 1989. The amyloid precursor protein is concentrated in neuronal lysosomes in normal and Alzheimer disease subjects. Exp. Neurol. 106: 237-250.
6. DAWSON, heterogeneity
279. 7. DYKEN, Neurol.
G., S. A. DAWSON, AND S. J. CHAN. 1989. Molecular in Batten disease. Trans. Am. Sot. Neurosci. 20:
P. R. 1989. The 4: 165-174.
neuronal
ceroid
lipofuscinoses.
J. Child
8. EIBERG,
H., R. M. GARDINER, AND J. MOHR. 1989. Batten disease (Spielmeyer-Sjegren disease) and haptoglobins (HP): Indication of linkage and assignment to chr. 16. Clin. Genet. 36: 217-
218. 9. HUTCHISON, tor deficiency.
D. C. S. 1988. Natural history of a,-protease Am. J. Med. 84(Suppl. 6A):3-12.
inhibi-
10.
Iw, G. O., F. SCHOTTLER, J. WENZEL, M. BAUDRY, AND G. LYNCH. 1984. Inhibitors of lysosomal enzymes: Accumulation of lipofuscin-like dense bodies in the brain. Science 226: 985-987.
11.
JUSTICE, D. L., R. H. RHODES, AND Z. A. TBKBs. 1987. Immunohistochemical demonstration of proteinase inhibitor ai-antichymotrypsin in normal human central nervous system. J. Cell. Biothem. 34: 227-238.
12.
KITAGUCHI, N., Y. TAKAHASHI, Y. TOKUSHIMA, H. ITO. 1988. Novel precursor of Alzheimer’s protein shows protease inhibitory activity.
331:530-532.
S. SHIOJIRI, AND disease amyloid Nature (London)
126 13.
14.
15.
16.
17.
18.
19.
20.
WISNIEWSKI KITAGUCHI, T., K. E. WISNIEWSKI, S. MASLINSKI, D. MASLINSKA, AND T. M. WISNIEWSKI. 1990. @-protein immunoreactivity in brains of neuronal ceroid lipofuscinosis: Ultrastructural and biochemical demonstration. Neurosci. Lett. 112: 155-160. PALMER, D. N., R. D. MARTINUS, S. M. COOPER, G. G. MIDWINTER, J. C. REID, AND R. D. JOLLY. 1989. Ovine ceroid lipofuscinosis. The major lipopigment protein and the lipid-binding subunit of mitochondrial ATP synthase have the same NHZ-terminal sequence. J. Biol. Chem. 10: 5736-5740. PASTJXRNACK, J. M., C. R. ABRAHAM, B. J. VAN DYKE, H. POTTER, AND S. G. YOUNKIN. 1989. Astrocytes in Alzheimer’s disease grey matter express a,-antichymotrypsin mRNA. Am. J. Pathol. 135: 827-843. PICKEN, M. M., M. LARRONDO-LILLO, M. L. SHELANSKI, AND B. FRANGIONE. protease inhibitor or-antichymotrypsin amyloid. J. Neuropathol. Exp. Neurol.
F. CORIA, G. R. GALLO, 1990. Distribution of the in cerebral and systemic 49: 41-48.
PONTE, P., P. GONZALEZ-DEWHITT, J. SCHILLING, J. MILLER, D. Hsu, B. GREENBERG, K. DAVIS, W. WALLACE, I. LIEBERBURG, F. FULLER, AND B. CORDELL. 1988. A new A4 amyloid mRNA contains a domain homologous to serine protease inhibitors. Nature (London) 331: 525-527. RICHARDSON, J. M., N. A. WOYCHIK, D. L. EBERT, R. L. DIMOND, AND J. A. CARDELLI. 1988. Inhibition of early but not late proteolytic processing events leads to the missorting and oversecretion of precursor forms of lysosomal enzymes in Dictyostelium discoideum. J. Cell Biol. 10’7: 2097-2107. SCHECHTER, N. M., J. L. SPROWS, 0. L. SCHOENBERG, G. S. LAZARUS, B. S. COOPERMAN, AND H. RUBIN. 1989. Reaction of human skin chymotrypsin-like proteinase chymase with plasma proteinase inhibitors. J. Biol. Chem. 264: 21,308-21,315. TAKAUCHI, S., AND K. MIYOSHI. 1989. Degeneration of neuronal
AND
21.
22. 23.
KIDA processes in rats induced by a protease inhibitor, leupeptin. Acta Neuropathol. (Berlin) 78: 380-387. TANZI, R. E., A. I. MCCLATCHEY, E. D. LAMPERTI, L. VILLA-K• MAROFF, J. F. GUSELLA, AND R. L. NEVE. 1988. Protease inhibitor domain encoded by an amyloid protein precursor mRNA associated with Alzheimer’s disease. Nature (London) 331: 528530. TRAVIS, J. 1988. Structure, function and control of neutrophil proteases. Am. J. Med. 84(Suppl. 6A):37-42. TRAVIS, J., AND G. S. SALVESEN. 1983. Human plasma proteinase inhibitors. Annu. Rev. Biochem. 62: 655-709.
24.
WISNIEWSKI, K. E., E. KIDA, TOH. 1990. Altered amyloid neuronal ceroid lipofuscinosis tion.
25.
WISNIE~SKI, K. E., AND D. MASLINSKA. of ceroid lipofuscin storage pigment monoclonal antibodies to the amyloid Med. 320: 256-257. [Letter].
26.
WISNIEWSKI, K. E., I. RAPIN, AND J. HEANEY-KIEXAS. 1988. Clinico-pathological variability in the childhood neuronal ceroid lipofuscinoses and new observations on glycoprotein abnormalities. Am. J. Med. Genet. S(Supp1.): 27-47. WOLFE, L. S., AND V. APPANNA. 1989. Deficient activity of cathepsin H in Batten disease. Trans. Am. Sot. Neurochem. 20: 145.
27. 28.
29.
W. GORDON-MAJSZAK, AND T. SAI/3 protein precursor processing in brain. Neurosci. Lett., in prepara1989. Immunoreactivity in Batten disease with beta protein. N. Engl. J.
VAN NOSTRAND, W. E., S. L. WAGNER, M. SUZUKI, B. H. CHOI, J. S. FARROW, J. W. GEDDES, C. W. COTMAN, AND D. D. CmNINGHAM. 1989. Protease nexin II, a potent antichymotrypsin, shows identity to amyloid P-protein precursor. Nature (London) 341: 546-549. ZEMAN, W. 1976. The neuronal ceroid lipofuscinoses. Prog. Neuropathol. 3: 203-223.
NEUROLOGY
110,
121-126 (1990)
Proteinase Inhibitor a,-Antichymotrypsin Has Different Expression in Various Forms of Neuronal Ceroid Lipofuscinosis KRYSTYNA New York State in Developmental
E. WISNIEWSKI
Ofice of Mental Retardation Disabilities, Department
AND ELIZABETH
KIDA
and Developmental Disabilities, Institute for Basic Research of Pathological Neurobiology, Staten Island, New York 10314
Defective proteolytic degradation is most widely maintained as the major pathogenetic factor in neuronal ceroid lipofuscinosis (NCL). The goal of the present study was to examine the expression in NCL brain tissue of one of the serine proteinase inhibitors, al-antichymotrypsin. Our study was based on previous llndings of a,-antichymotrypsin association with CNS amyloidoses related to amyloid /3 protein deposits and our previous findings suggesting abnormal processing of amyloid B-protein precursor (ABPP) in NCL brains. Immunocytochemical study was performed on formalinfixed brain tissues collected from 15 NCL cases representing four different forms of the disorder and from 16 control cases comprising age-matched controls, older nondemented individuals, and Alzheimer disease cases. Our present study has shown that the expression of qantichymotrypsin is generally higher in NCL cases than in control cases; however, it manifests in distinct variations of intensity and proportions of immunostained cells. The strongest immunoreactivity was found in the infantile form of NCL, which is characterized by a rapid clinical course and widespread tissue damage. We found no evidence of direct involvement of a,-antichymotrypsin in either the ceroid lipopigment accumulation or the abnormal processing of ABPP in NCL. However, our findings may reflect the heterogeneity of the pathomechanism underlying this group of disorders and suggest that, similarly to blood circulation, al-antichymotrypsin can also represent an acute-phase protein in brain tissue. @ isso Academic PWS, k.
INTRODUCTION The neuronal ceroid lipofuscinoses (NCL) represent hereditary disorders manifesting lysosomal lipopigment storage material within cells of the central nervous system (CNS) and other organs. On the basis of the age of onset, clinical presentation, and pathological features the following four main forms of the disease are distinguished: infantile (INCL), late infantile (LINCL), juvenile (JNCL), and adult (ANCL) variants (29). However,
in addition, some less frequently encountered subtypes have also been described (7). There is general agreement that NCL is caused by a lysosomal dysfunction, but the proper biochemical defect remains unknown. Recent studies supplied data indicating that the well-known clinicopathological differences in forms of NCL might result from their genetic heterogeneity. In JNCL, the site of disease locus was located on the long arm of chromosome 16 (8). Cathepsin H deficiency, whose gene was localized to chromosome 15, was found in some LINCL cases (6, 27), whereas in individual INCL cases, lysosomal phospholipase A, deficiency, whose gene has not yet been cloned, was described (6). In lipopigments isolated from LINCL and JNCL cases and from sheep model of the disease, the storage of mitochondrial ATP synthase subunit C was disclosed (14). The common postulation of defective lysosomal proteolysis, as the basic lesion in NCL is supported by experimental findings disclosing an accumulation of lipofuscin-like substance after application of cysteine proteinase inhibitors (10). Recently, the question was also raised of the participation of proteinase inhibitors in the abnormal processing of amyloid P-protein precursor (ABPP) in Alzheimer disease (AD) cases (1, 12, 17, 21, 28). It is of interest in this regard that dense concentrations of ABPP were found in secondary lysosomes (5) and, moreover, that neuronal aging lipofuscin shows immunostaining with antibodies to fragment of amyloid P-protein (3). Our previous studies have disclosed strong neuronal immunoreactivity in the NCL brain tissues that accumulated ceroid lipopigment with antibodies generated against various domains of ABPP, including amyloid ,&protein’s epitopes (13,24,25). These findings suggest abnormal processing of ABPP also in the NCL brain tissue, as is postulated for AD. Thus, on the basis of the data pointing to the putative role of proteinase inhibitors in both ABPP processing and lipopigment storage, we have undertaken the present study to examine the immunoreactivity of the NCL brain tissue with antiserum to serine proteinase inhibitor, ai-antichymotrypsin (ACT). Our goal is to provide further in-
121 All
Copyright 0 1990 rights of reproduction
0014-4886/90 $3.00 by Academic Press, Inc. in any form reserved.
122
WISNIEWSKI
sight into the presumed tion in the NCL brain. MATERIAL
abnormal
AND
proteolytic
degrada-
METHODS
Immunocytochemistry was performed on formalinfixed, paraffin-embedded, 5-pm-thick tissue sections collected from 15 NCL cases comprising four different forms of the disease [(4 INCL (20 months-5 years), 2 LINCL (6-12 years), 8 JNCL (lo-25 years), and 1 ANCL (56 years)]. Clinicopathological features of some of the NCL cases have been previously reported (26). Additional material included autopsied brain tissues from 10 normal controls, age 12 months to 51 years; 3 nondemented older controls, age 65 to 85 years; and 3 proven AD cases, age 51 to 85 years. The avidin-biotin-peroxidase complex method was used as the staining procedure. Hydrated tissue sections were incubated for 30 min in 0.2% hydrogen peroxide in methanol to block endogenous peroxidase, pretreated with 0.05% trypsin/CaCl, in Tris buffer, pH 8 (Sigma Chemicals, St. Louis) for 30 min at 37°C and then incubated with 10% normal goat serum in PBS for 30 min at room temperature. Specific rabbit antiserum to human cu,-antichymotrypsin was purchased from Accurate Chemical and Scientific Corp. The sections were incubated with the primary antiserum (diluted 1:500) overnight at 4°C. Procedure controls involved the substitution of the primary antiserum by PBS and nonimmune species-specific serum and the use of GFAP-antibodies (1:500) (Dako Corp., Carpinteria, CA). Biotinylated goat anti-rabbit antibodies (1:200) followed by avidinbiotin-peroxidase complex (Vector Laboratories, Burlingame) were applied to the sections, each for 30 min. The reaction product was visualized with diaminobenzidine (Sigma Chemicals, St. Louis) in the presence of 0.01% hydrogen peroxide. The sections were counterstained with hematoxylin, dehydrated, and mounted with Permount. RESULTS In the NCL brain tissue, the antiserum to ACT stained various types of CNS cells, including both the perikarya and nuclei of numerous neurons and astrocytes. However, generally, the intensity of the staining and the percentage of immunoreactive neuronal and glial cells varied distinctly among not only the different forms of NCL studied but also the individual cases within the same form of the disease. The strongest immunoreactivity presented nerve cells located in the cerebral cortex of two INCL cases. Typically in INCL, a prominent neuronal loss in the cortical mantle is observed: however, in these two cases, virtually all surviving neurons showed large, coarse granular material in their distended cytoplasm (Fig. la). Immunostaining of
AND
KIDA
neurons in the other brain areas was somewhat weaker and was confined to the presence of a finely granular reaction product. Neuronal nuclei in INCL cases were rarely stained as compared with the cases of the other forms of the disease examined. In LINCL, JNCL, and ANCL, the reaction product in neuronal perikarya was represented solely by finely granular deposits. The intensity of the neuronal staining in these forms of NCL varied, and often, even in close proximity to strongly labeled neurons, unstained nerve cells could be found (Fig. lb). Some of the affected neurons that were overloaded with lipopigment material presented finely granular material not only in the perikarya but also in thin cellular processes (Fig. lc) and sporadically in enlarged proximal axons. Numerous neurons with cytoplasm distended by accumulated lipopigment showed, in addition to perikaryal staining, heavily stained nuclei (Fig. Id). Nuclear staining was observed in some of the neurons manifesting weak or no immunoreactivity (Fig. le). Some neurons overloaded with stored lipopigment were not immunoreactive (Fig. If). Generally, the proportion of immunostained neurons was the greatest in INCL and only slightly smaller in ANCL. In JNCL, variations in both the intensity of immunostaining and the percentage of labeled neurons between particular cases were the most prominent. In this form of NCL, apart from the cases with a large number of moderately stained neurons, the cases with only a few immunoreactive neurons, usually confined to the hippocampus, some neocortical areas, the basal ganglia, and the granular layers of the cerebellar cortex, could be found. Astroglial cells in all the NCL cases studied showed immunostaining of the nuclei or perikarya, or both. The strongest immunoreactivity of the astrocytes was seen in the grey matter in INCL cases. In these cases, the reaction product presented large, coarse granules in the cytoplasm of numerous astrocytes (Fig. 2a): less frequently, such strong immunoreactivity was seen in the astrocytes located in the white matter, where light immunostaining of the perikarya of the astroglial cells predominated. In LINCL, JNCL, and ANCL, immunostaining of the perikarya of astrocytes in the grey matter was rarely observed, however, their nuclei often showed immunoreactivity (Fig. 2b). In some NCL cases however, numerous reactive astrocytes with moderately stained perikarya were observed in the hippocampal region. There was no close correspondence between the intensity and the number of immunostained astrocytes in LINCL and JNCL and the degree of tissue destruction. Even in cases with advanced neuronal degeneration, only a few reactive astrocytes with immunoreactive cytoplasm could be found in affected areas. Moreover, in the above-mentioned NCL forms, larger numbers of immunostained astrocytes were observed in the white rather than in the grey matter (Fig. 2~). In AD cases, numerous neurons in cerebral cortex
a,-ANTICHYMOTRYPSIN
IN
NEURONAL
CEROID
LIPOFUSCINOSIS
123
FIG. 1. ACT immunostaining of nerve cells in different NCL forms. (a) Coarse granular material in the perikaryon of distended neuron in cerebral cortex. INCL case, x750. (b) Large pyramidal neurons in the cerebral cortex strongly labeled by finely granular reaction product, among them unstained neurons. ANCL case, x300. (c) Finely granular reaction product in the perikarya and processes of cortical neuron overloaded with lipopigment. ANCL case, x750. (d) Heavily immunostained nucleus in a neuron with moderate perikaryal staining. ANCL case, x300. (e) Neurons overloaded with lipopigment from hippocampal CA1 area showing weak perikaryal immunoreactivity and nuclear staining. JNCL, x200. (f) Unstained neurons manifesting stored lipopigment in inferior olivary nucleus. JNCL, X200.
showed moderate immunostaining, including both neuronal perikarya and their nuclei. In three AD cases, sporadically in ANCL cases, and in one case of older controls, moderate immunostaining of neuritic plaques was seen (Fig. 2d). Moderate immunostaining of some blood vessels was observed in AD cases. The percentage of astrocytes with immunostained perikarya was still greater in white than in grey matter; however, that discrepancy was smaller in AD cases than in NCL cases. In all the cases studied, including NCL, normal controls, and AD cases, heavy immunoreactivity of the epithelium of the choroid plexus, when available, was present (Fig. 2e). In normal control individuals, the pattern and intensity of immunostaining varied between particular cases. Generally, however, both neurons and astrocytes showed immunoreactivity less frequently than in NCL and AD cases. Immunostaining, usually of weak or moderate intensity, was seen, mostly in large neurons of the cerebral cortex, mainly in the hippocampus (Fig. 2f ), in the medulla, and in the granular layer of cerebellar cor-
tex and less frequently in the Purkinje cells. Nuclei and perikarya of astrocytes were clearly less frequently stained in both grey and white matter. The pattern of immunostaining of the ependymal cells and of pericytes varied between particular cases, regardless of the group of cases studied. Semiquantitative analysis of neuronal and glial cells immunolabeling is summarized in Fig. 3. DISCUSSION
Our present study has shown that the same types of CNS cells in brains of NCL and control cases were immunostained with antiserum to ACT. Proportions and intensity of immunostaining varied depending on the particular brain area and the case studied. The strongest immunoreactivity in all cases examined presented epithelial cells of the choroid plexus. Immunostaining of neuronal perikarya in both NCL and control cases was most often seen in the cerebral cortex, especially in the hippocampus, in the medulla, and in the cerebellar cortex. Nuclear staining of both neurons and glial cells was
124
WISNIEWSKI
AND
KIDA
FIG. 2. ACT immunostaining in NCL and control cases. (a) Heavily immunostained astroglial perikarya in cerebral cortex in X750. (b) Numerous astrocytes with stained nuclei, but unstained perikarya in the cerebral cortex in ANCL case. x200. (c) manifesting nuclear and perikaryal staining in the cerebral white matter of JNCL. x300. (d) Immunoreactivity of neuritic plaques cortex in AD case. X300. (e) Strong immunostaining of choroid plexus epithelium. Control case, X300. (f) Various intensities immunostaining in hippocampal end-plate. Control case, X200.
also found, resulting from ACT binding to DNA and nucleosomes (6,11,16). A similar pattern of ACT immunostaining distribution was described previously in normal human CNS (11). However, our present study disclosed a generally higher percentage and intensity of perikaryal neuronal and astroglial immunostaining in NCL cases than in controls. The proper physiological function of ACT in CNS still remains to be clarified. ACT is one of the human plasma serine proteinase inhibitors, which are involved in regulation of such processes as coagulation, fibrinolysis, or inflammation, and which by inhibiting the activity of their target proteinases may prevent tissue damage. ACT represents a major acute-phase protein whose plasma concentration increases rapidly in various pathological conditions (23). This serine proteinase inhibitor forms equimolar complexes with its target enzyme: neutrophil cathepsin G, with chymotrypsin and mast cells “chymases” (4, 19). However, the question was raised whether the true substrate for ACT has really been found, because cathepsin G was considered an inefficient proteinase of slow hydrolytic activity (22). There are data indicating that apart from the liver,
INCL case. Astrocytes in cerebral of neuronal
which represents a major source of ACT in the circulation and some extrahepatic sites, ACT can be synthesized in situ in the CNS. ACT messenger RNA was demonstrated and found to be greatly elevated in AD grey matter compared with controls (1) and produced mostly by astrocytes (15). Therefore, it was suggested that if only astrocytes were ACT producers, those neurons that manifested ACT immunoreactivity might take up ACT either from the extracellular fluid or from the blood, and that the observed increase in immunoreactivity of astrocytes in the grey matter might reflect a general, nonspecific response of the brain to different disease processes that result in neurodegeneration (2). Our present study has shown that NCL brain tissue immunoreactivity is distinctly varied, strong in INCL cases and showing the greatest variations in JNCL, where in some individuals only slight immunoreactivity confined to only a few neurons and reactive glial cells was observed. These results contrast with our previous observations disclosing the strongest immunoreactivity of JNCL and ANCL brain tissue with antibodies to amyloid P-protein (25). Moreover, we could not find in our material any close correlation between the degree of
or-ANTICHYMOTRYPSIN
IN
NEURONAL
CEROID
125
LIPOFUSCINOSIS
circulation exist and, therefore, that also in the CNS ACT represents an acute-phase protein particularly, strongly elevated in rapidly evolving pathological processes with abnormal proteolytic activity. Greatly increased ACT immunoreactivity in INCL might be related to the mechanisms operating specifically only in this form of NCL, differing from the other NCL forms, and might reflect an etiopathogenetical heterogeneity for these disorders.
1
2
3
4
5
6
ACKNOWLEDGMENTS
7
GROUP FIG. 3. Semiquantitative analysis of neuronal and glial cells’ immunoreactivity. Results show the average numbers of immunostained neurons (black columns) and of glial cells (white columns) per 1 mm2 counted in 10 nonoverlapping cortical areas from INCL (column I), LINCL (column 2), JNCL (column 3), ANCL (column 4), age-matched and old controls (columns 5 and 6, respectively), and AD (column 7). The small number of immunostained neurons in column 1 results from severe cortical atrophy found in these cases.
lipopigment accumulation and ACT immunoreactivity as even neurons severely overloaded with lipopigment showed various intensities of staining or were not immunoreactive. Therefore, the direct participation of ACT in the pathogenesis of lipopigment storage and abnormal processing of ABPP in NCL is very unlikely. ACT was previously found in amyloid deposits in amyloid /3protein-related CNS amyloidoses (1, 16). Other serine/ cysteine proteinase inhibitors are able to induce neurodegenerative changes in CNS and accumulation of lipofuscin-like substance (10, 20). Nevertheless, it is well documented that some serine protease inhibitors can produce a protective effect on various tissues due to the inhibition of uncontrolled proteolysis and that their deficiency causes tissue damage as is evidenced in the case of a,-protease inhibitor (9). Moreover, they are supposed to be responsible for efficiency of proteolytic processing and normal sorting of lysosomal enzyme precursors to lysosomes (18). INCL is characterized by a rapid clinical course and the most severe tissue destruction. Strong ACT immunostaining, as was disclosed in INCL, with heavy astroglial immunoreactivity, suggests increased ACT synthesis. It may reflect a secondary and unspecific response of the CNS to severe and widespread tissue injury, followed by the activation of protective mechanisms. These mechanisms, however inefficient, might be directed to prevent uncontrolled proteolysis and subsequent progression of tissue destruction. It is of interest in this regard that in LINCL and JNCL even in cases in which prominent tissue degeneration and neuronal loss were found, ACT expression was distinctly weaker than in INCL cases. These discrepancies may indicate that in the CNS, mechanisms of ACT synthesis similar to those found in blood
We thank Ms. Madeline Tinney for her secretarial assistance, Maureen Stoddard Marlow for critically reading the manuscript, Mr. Lawrence Black for bibliographical editing assistance. This was supported by NIH Grant NS23717.
Ms. and work
REFERENCES 1.
2.
ABRAHAM, C. R., D. J. SELKOE, AND H. POTTER. 1988. Immunochemical identification of the serine protease inhibitor cY,-antichymotrypsin in the brain amyloid deposits of Alzheimer’s disease. Cell 62: 487-501. ABRAHAM, C. R., T. SHIRAHAMA, AND H. POTTER. 1990. cY,-antichymotrypsin is associated solely with amyloid deposits containing the beta-protein. Amyloid and cell localization of ol,-antichymotrypsin. Neurobiol. Aging 11: 123-129.
3. BANCHER,
C., I. GRUNDKE-IQBAL, K. IQBAL, K. S. KIM, AND H. M. WISNIEWSKI. 1989. Immunoreactivity of neuronal lipofustin with monoclonal antibodies to the amyloid P-protein. Neurobiol. Aging 10: 125-132.
4. BEATTY,
K., J. BIETH, AND J. TRAVIS. 1980. Kinetics of association of serine proteinases with native and oxidized cu,-proteinase inhibitor and cu,-antichymotrypsin. J. Biol. Chem. 255: 39313934.
5. BENOWITZ,
L. I., W. RODRIQUEZ, P. PASKEVICH, E. J. MUFSON, D. SCHENK, AND R. L. NEVE. 1989. The amyloid precursor protein is concentrated in neuronal lysosomes in normal and Alzheimer disease subjects. Exp. Neurol. 106: 237-250.
6. DAWSON, heterogeneity
279. 7. DYKEN, Neurol.
G., S. A. DAWSON, AND S. J. CHAN. 1989. Molecular in Batten disease. Trans. Am. Sot. Neurosci. 20:
P. R. 1989. The 4: 165-174.
neuronal
ceroid
lipofuscinoses.
J. Child
8. EIBERG,
H., R. M. GARDINER, AND J. MOHR. 1989. Batten disease (Spielmeyer-Sjegren disease) and haptoglobins (HP): Indication of linkage and assignment to chr. 16. Clin. Genet. 36: 217-
218. 9. HUTCHISON, tor deficiency.
D. C. S. 1988. Natural history of a,-protease Am. J. Med. 84(Suppl. 6A):3-12.
inhibi-
10.
Iw, G. O., F. SCHOTTLER, J. WENZEL, M. BAUDRY, AND G. LYNCH. 1984. Inhibitors of lysosomal enzymes: Accumulation of lipofuscin-like dense bodies in the brain. Science 226: 985-987.
11.
JUSTICE, D. L., R. H. RHODES, AND Z. A. TBKBs. 1987. Immunohistochemical demonstration of proteinase inhibitor ai-antichymotrypsin in normal human central nervous system. J. Cell. Biothem. 34: 227-238.
12.
KITAGUCHI, N., Y. TAKAHASHI, Y. TOKUSHIMA, H. ITO. 1988. Novel precursor of Alzheimer’s protein shows protease inhibitory activity.
331:530-532.
S. SHIOJIRI, AND disease amyloid Nature (London)
126 13.
14.
15.
16.
17.
18.
19.
20.
WISNIEWSKI KITAGUCHI, T., K. E. WISNIEWSKI, S. MASLINSKI, D. MASLINSKA, AND T. M. WISNIEWSKI. 1990. @-protein immunoreactivity in brains of neuronal ceroid lipofuscinosis: Ultrastructural and biochemical demonstration. Neurosci. Lett. 112: 155-160. PALMER, D. N., R. D. MARTINUS, S. M. COOPER, G. G. MIDWINTER, J. C. REID, AND R. D. JOLLY. 1989. Ovine ceroid lipofuscinosis. The major lipopigment protein and the lipid-binding subunit of mitochondrial ATP synthase have the same NHZ-terminal sequence. J. Biol. Chem. 10: 5736-5740. PASTJXRNACK, J. M., C. R. ABRAHAM, B. J. VAN DYKE, H. POTTER, AND S. G. YOUNKIN. 1989. Astrocytes in Alzheimer’s disease grey matter express a,-antichymotrypsin mRNA. Am. J. Pathol. 135: 827-843. PICKEN, M. M., M. LARRONDO-LILLO, M. L. SHELANSKI, AND B. FRANGIONE. protease inhibitor or-antichymotrypsin amyloid. J. Neuropathol. Exp. Neurol.
F. CORIA, G. R. GALLO, 1990. Distribution of the in cerebral and systemic 49: 41-48.
PONTE, P., P. GONZALEZ-DEWHITT, J. SCHILLING, J. MILLER, D. Hsu, B. GREENBERG, K. DAVIS, W. WALLACE, I. LIEBERBURG, F. FULLER, AND B. CORDELL. 1988. A new A4 amyloid mRNA contains a domain homologous to serine protease inhibitors. Nature (London) 331: 525-527. RICHARDSON, J. M., N. A. WOYCHIK, D. L. EBERT, R. L. DIMOND, AND J. A. CARDELLI. 1988. Inhibition of early but not late proteolytic processing events leads to the missorting and oversecretion of precursor forms of lysosomal enzymes in Dictyostelium discoideum. J. Cell Biol. 10’7: 2097-2107. SCHECHTER, N. M., J. L. SPROWS, 0. L. SCHOENBERG, G. S. LAZARUS, B. S. COOPERMAN, AND H. RUBIN. 1989. Reaction of human skin chymotrypsin-like proteinase chymase with plasma proteinase inhibitors. J. Biol. Chem. 264: 21,308-21,315. TAKAUCHI, S., AND K. MIYOSHI. 1989. Degeneration of neuronal
AND
21.
22. 23.
KIDA processes in rats induced by a protease inhibitor, leupeptin. Acta Neuropathol. (Berlin) 78: 380-387. TANZI, R. E., A. I. MCCLATCHEY, E. D. LAMPERTI, L. VILLA-K• MAROFF, J. F. GUSELLA, AND R. L. NEVE. 1988. Protease inhibitor domain encoded by an amyloid protein precursor mRNA associated with Alzheimer’s disease. Nature (London) 331: 528530. TRAVIS, J. 1988. Structure, function and control of neutrophil proteases. Am. J. Med. 84(Suppl. 6A):37-42. TRAVIS, J., AND G. S. SALVESEN. 1983. Human plasma proteinase inhibitors. Annu. Rev. Biochem. 62: 655-709.
24.
WISNIEWSKI, K. E., E. KIDA, TOH. 1990. Altered amyloid neuronal ceroid lipofuscinosis tion.
25.
WISNIE~SKI, K. E., AND D. MASLINSKA. of ceroid lipofuscin storage pigment monoclonal antibodies to the amyloid Med. 320: 256-257. [Letter].
26.
WISNIEWSKI, K. E., I. RAPIN, AND J. HEANEY-KIEXAS. 1988. Clinico-pathological variability in the childhood neuronal ceroid lipofuscinoses and new observations on glycoprotein abnormalities. Am. J. Med. Genet. S(Supp1.): 27-47. WOLFE, L. S., AND V. APPANNA. 1989. Deficient activity of cathepsin H in Batten disease. Trans. Am. Sot. Neurochem. 20: 145.
27. 28.
29.
W. GORDON-MAJSZAK, AND T. SAI/3 protein precursor processing in brain. Neurosci. Lett., in prepara1989. Immunoreactivity in Batten disease with beta protein. N. Engl. J.
VAN NOSTRAND, W. E., S. L. WAGNER, M. SUZUKI, B. H. CHOI, J. S. FARROW, J. W. GEDDES, C. W. COTMAN, AND D. D. CmNINGHAM. 1989. Protease nexin II, a potent antichymotrypsin, shows identity to amyloid P-protein precursor. Nature (London) 341: 546-549. ZEMAN, W. 1976. The neuronal ceroid lipofuscinoses. Prog. Neuropathol. 3: 203-223.
