Specific substrate for CLN2 proteaseltripeptidylpeptidaseI assay

doi: 10.1053/ejpn.2000.0437 available online at http://www.idealibrary.com on I1[~!" European Journal of Paediatric Neurology 2001 ; 5(Suppl. A): 63-68

ORIGINAL

ARTICLE

Specific substrate for CLN2 protease/tripeptidylpeptidase I assay M O H A M M E D A JUNAID, 1 SUSAN SKLOWER BROOKS, 2 RAJU K PULLARKAT I I Departments of Developmental Biochemistry and 2Human Genetics, New York State Institute for Basic Research in Developmental Disabilities, New York, USA

The classic late infantile neuronal ceroid lipofuscinosis (LINCL, CLN2) is a fatal neurodegenerative disorder that results from mutations in a gene encoding a lysosomal proteinase, known as CLN2 protease (CLN2p) or tripeptidyl peptidase I (TPP-I). Three different substrates, fluorescein isothiocyanate-labelled haemoglobin, A-F-F-7-amino-4methylcoumarin (AAF-AMC) and G-F-F-L-7-amino-4-trifluoromethylcoumarin (GFFL-AFC) have been used for the CLN2p/TPP-I assay with varying degrees of residual activities in patients with LINCL. Further, conclusive identification of carriers are not possible with the first two substrates. An assay for the CLN2p/-rPP-I based on the cleavage of amino terminal tripeptide from G-F-F-L-AFC was applied to prenatal and postnatal diagnosis of LINCL patients and heterozygote carriers. In leukocytes, the CLN2p/TPP-I activities in controls and heterozygote carriers were 1995+ 154 (n--15) and 918-1-253 (n = 15) nmol/h/mg protein respectively. No CLN2p/TPP-I activity was detectable in all but two patients. These two patients had less than 2% residual activity, and had delayed clinical symptoms for LINCL. This shows that the G-F-F-L-AFC is a highly specific substrate for the CLN2p/TPP-I assay. The fact that with this substrate the enzyme cleaves a peptide bond between the two amino acids may be the reason for the high level of specificity.

Keywords:Lateinfantile neuronalceroid Iipofuscinosis.CLN2 protease.Tripeptidyl peptidaseI. Pepstatininsensitive.Acid protease.

Introduction The classic late infantile neuronal ceroid lipofuscinosis (LINCL, CLN2) is one of a group of closely related fatal neurodegenerative disorders that are characterized by accumulation of autofluorescent storage bodies in brain and other tissues. 1-3 Clinical symptoms include progressive visual failure, epileptic seizures, psychomotor deterioration, severe mental retardation, and result in premature death. Histopathologic lesions include profound loss of neurons and retinal cells. Identification of the gene defect in LINCL led to the discovery of a unique lysosomal acid protease (CLN2p) that was insensitive to the common

aspartic protease inhibitor-pepstatin A. 4 The CLN2p deficient in LINCL is now recognized as the tripeptidyl peptidase I (TPP-I) that had been purified earlier from rat spleen. 5,6 CLN2p/TPP-I is an acid protease present ubiquitously in all tissues, and cleaves N-terminal tripeptides from substrates that have free amino termini. 7,s Although, CLN2p/TPP-I shares significant amino acid sequence similarity with a group of bacterial pepstatin-insensitive carboxyl proteinases (BCP)4, CLN2p/TPP-I and BCP have entirely different substrate specificities. Unlike the CLN2/ TPP-I which cleaves tripeptides from amino termini, BCP cleave substrates towards the carboxyl termini. 9 Studies on the expression of CLN2p/TPPI have shown that the enzyme activity is elevated

Correspondence:MA Junaid PhD, Departmentof DevelopmentalBiochemistry, NYS Institutefor BasicResearch,1050ForestHill Road,StatenIsland, NY 10314, USA. e-mail: [email protected] 1090-3798/05/A063+6 $35.00

© 2001 European Paediatric Neurology Society

Original article: MA Junaid et al.

64

markedly in several disease conditions, including other forms of neuronal ceroid lipofuscinosis (NCL) and breast cancer, l°,n In view of extreme physiological importance of the CLN2p/TPP-I in several pathological conditions, various laboratories have developed assays for detecting the enzyme activity in biological tissues. 6,12,13 These enzyme assays are used for the diagnosis of LINCL patients and heterozygote carriers. Some of these assays have also been adapted for prenatal diagnosis of the LINCL. 14,~s In the present study, we demonstrate the extreme specificity and sensitivity of G-F-F-L-AFC as a substrate for the CLN2p/TPP-I assay for both prenatal and postnatal diagnosis of LINCL patients and heterozygote carriers.

Subjects and methods Heparinized blood samples (5 ml) were collected from normal and neurological control subjects (n=15), LINCL patients (n=14) and obligate heterozygote carriers (n = 15) upon informed consent by an institutional review board approved protocol. Leukocytes were prepared by dextran sedimentation. On several occasions, the lag between blood collection and leukocyte preparation extended over 48 hours, without any observed differences in the measured CLN2p/TPP-I activity by the present procedure. The leukocyte pellets were stored frozen at -20°C until analysed for measurement of the CLN2p/TPP-I activity. High performance silica gel HP-K plates were purchased from Whatman Laboratories, NJ, USA. The tetrapeptide substrate G-F-F-L-7-amino-4trifluoromethylcoumarin (G-F-F-L-AFC) was prepared from a commercially available peptide substrate and the CLN2p/TPP-I activity was measured in extracts prepared from leukocytes and chorionic villi as described earlier. 12 Protein concentrations in tissue extracts were determined by the modified Folin phenol method using bovine serum albmnin as the standard. 16 Between 1 and 10/~g total protein was used for each assay run in duplicate. The specific activity of CLN2p/TPP-I is expressed as nanomoles (nmol) of L-AFC formed per hour per milligram of protein. The values reported are average m e a n + s t a n d a r d deviation (SD). For analysis of the CLN2p/TPP-I activity by thin layer chromatography, 10 #1 of the assay mixture after 10 minutes incubation time was spotted on a high performance silica gel HP-K plate. The plate was developed in a solvent system comprising of chloroform:methanol:water that was mixed in the

ratio of 65:25:5. The substrate and the product bands were visualized and photographed under ultraviolet transilluminator.

Results and discussion In the present study, we have used an extremely sensitive and highly specific enzymatic assay for the CLN2p/TPP-I in leucocytes for the diagnosis of LINCL patients and heterozygote carriers. This assay has been successfully applied for the prenatal diagnosis of high risk pregnancies in families where both parents were obligate heterozygote carriers. Figure 1 shows CLN2p/TPP-I activities in leukocytes prepared from control subjects that also includes patients with other neurological diseases, LINCL patients and heterozygote carriers. The average mean + SD CLN2p/TPP-I activities in controls and heterozygote carriers were 1995_+ 154 (n = 15) and 918 _+253 (n = 15) n m o l / h / m g protein,

2500 =

2000

.?:, °°%,

i 1500

looo

,.

500 0 Controls

Patients

Carriers

Fig. 1. Scatter plot of the CLN2p/TPP-I activity in controls (n ----15), LINCL patients (n = 14) and heterozygote carriers (n-- 15). The substrate G-F-F-L-AFC (280/~M) was incubated with extracts (1-10/~g protein) in 50mM ammonium formate buffer pH3.5 containing 0.2mM pepstatin-A and 0.5 mM trans-epoxysuccinyI-L-leucylamido-(4-guanidino) butane (E-64) in a volume of 25 #1 at 37°C for 5 min. Reactions were terminated by adding icecold acetone followed by centrifugation at 12 000 x g for 2 min. Supernatants were evaporated to dryness, residues were dissolved in acetonitrile, and an aliquot was analysed by higher performance liquid chromatography on a reversed-phase C18 column. The substrate and the product were detected by a variable wavelength detector set at 340 nm. Amount of L-AFC formed was calculated from the area of the peak obtained by injecting known amounts of LAFC.

Original article: Specific CLN2p/-I-PP-I assay 1

2

65 3

4

5

6

L-AFC G-F-F-L-AFC

Origin Fig. 2. Thin layer chromatography analysis of the CLN2p/TPP-I activities in controls 1, 2, heterozygote carriers 3,4 and LINCL patients 5, 6. Extracts containing 6 #g protein were incubated with the substrate G-F-F-L-AFC for 10 min according to the conditions mentioned in the legend for Fig.l. Aliquots (10 #1) were applied on a high performance silica gel HP-K plate which was developed in a solvent system comprising of chloroform:methanol:water (65:25:5) and bands were visualized under ultraviolet light.

respectively. In agreement with a single copy of the defective allele for the CLN2 gene in the heterozygote carriers the CLN2p/TPP-I activity was roughly half that of normal and neurological controls. We have observed absolutely no overlap of CLN2p/TPP-I activities between controls and carriers. The assay is highly reproducible with coefficient of variance of 0.032 and 0.061 for two different samples each analysed four different times. Leukocytes from 12 LINCL patients analysed had undetectable CLN2p/TPP-I activity. Two other LINCL patients had less than 2% of the control activity. Both of these patients had delayed onset of clinical symptoms and this residual activity may have been a contributing factor. Earlier, two other atypical LINCL cases were reported with delayed onset of clinical symptoms, and both of these patients had extended lifespan of over 30 years. 17 Both these patients carried at least one allele which had a missense mutation of G - * A conversion that resulted in a hlstidine at position 447 instead of arginine. The residual enzyme activity associated with this mutant protein is

probably responsible for delayed onset which may be related to slower build up of peptides toxic to neurons and retinal cells. A simpler and rapid adaptation of the present assay was developed that circumvents the use of a high performance liquid chromatography system. After addition of the substrate, incubations were carried out for 10 minutes, and 10 #1 aliquots of the assay mixture were spotted on a high performance silica gel HP-K plate. The silica gel plate was developed in a solvent system comprising of chloroform:methanol:water (65:25:5), and the substrate and the product spots were visualized under ultraviolet light (Fig. 2). The leukocyte extracts from LINCL patients completely lack the product spot that corresponds to L-AFC. The intensity of the L-AFC spot for heterozygote carriers is roughly half that of controls and can even be visually inspected. This modification of the CLN2p/TPP-I assay is helpful in rapid high throughput screening for LINCL patients in particular. This enzymatic assay has also been used for prenatal diagnosis of LINCL in three high-risk

Original article: MA Junaid et al.

66

Prenataldiagnosis of LINCLin chorionic villi for pregnancies at risk

Table 1

Subjects

No.

CLN2p/TPP-I activity~ (nmol/h/ mg protein)

Controls Heterozygote carriers

4 3

1516 + 179 804 + 127b

aValues are expressed as mean+ SD; bp< 0.005. pregnancies by carrying out the CLN2p/TPP-I HPLC assay in chorionic villi. Table 1 shows the CLN2p/TPP-I activity in three high-risk pregnancies. The parents were obligate heterozygote carriers for LINCL and had affected children. The chorionic villi in all three pregnancies showed CLN2p/TPP-I activities that were about half that of normal controls. These observations were confirmed postnatally by analysing the CLN2p/TPP-I activities in leucocytes. These three subjects had CLN2p/TPP-I activities within the heterozygote carrier range. The molecular defect in LINCL has confirmed that lysosomal acid protease, CLN2p/TPP-I is the defective enzyme in LINCL.4 More than 24 different point mutations have so far been identified in LINCL patients. 17 Mutational analysis of the CLN2 gene or a sensitive and specific enzyme assay for measuring the CLN2p/TPP-I activity in biological tissues will greatly facilitate diagnosis and prevention of LINCL. Of over 24 different point mutations in the CLN2 gene identified to date, about 60% of cases have either a nucleotide 3556GC conversion at a splice junction site or a nucleotide 3670C---~T conversion resulting in a nonsense mutation (Arg208---~stop). 17,1s The rest of the mutations are scattered all over the CLN2 gene, which makes LINCL patient and carrier identification by mutational analysis very time consuming and expensive. In one study, a set of 13 primer pairs was used to sequence the complete CLN2 gene, 19 and demonstrates the technical difficulties assoTable 2

ciated with mutational analysis for LINCL patients and carriers. Whereas, a biochemical assay based on accurate measurement of the CLN2/TPP-I activity, as described in the present study, is a simple way to detect LINCL patients and carriers. LINCL is an autosomal recessive neurodegenerative disease that lacks therapy and proves fatal in early childhood. Carrier identification in the high risk population and prenatal diagnosis will provide effective preventive measures for such a neurogenetic disease. Before the identification of the genetic defect in LINCL, the only means of prenatal diagnosis was through ultrastructural features of the accumulated autofluorescent inclusion bodies. 2° This method did not always prove to be reliable in diagnosis. Three different substrates have so far been described for measuring the CLN2p/TPP-I activity in tissues by various laboratories (Table 2). Among these, the assay involving fluorescein isothiocyanate (FITC)-labelled haemoglobin is the least sensitive, and other laboratories have difficulties reproducing this procedure. Our laboratory,7 as well as another group, 8 has shown that the CLN2p/TPP-I cleaves tripeptides only from short peptide sequences, in view of this observation it is very unlikely that the CLN2p/TPP-I cleaves FITChaemoglobin by itself, rather, it cleaves fragments of peptides generated by the action of other proteases in the extracts. Since, the assay conditions include inhibition of other classes of proteolytic activities that will prevent formation of these peptide fragments, lower CLN2p/TPP-I activities are expected. Indeed the reported specific activities for CLN2p/TPP-I using FITC-haemoglobin were drastically lower 13,15 than those reported by other laboratories using more specific substrates. 6,12,14,21 Lower CLN2p/TPP-I activities will make carrier identification very difficult by this procedure. The CLN2p/TPP-I activity measured by FITChaemoglobin was indistinguishable from the patients in at least two carriers, and also there was overlap between control and carrier groups. 13

Comparison of CLN2p/TPP-I activity as measured with different substrates

Substrate

Incubation time

Specific activity in control leukocytesa

Residual activity in LINCL patients

Reference

FITC-haemoglobin AAF-AMC

2-4 h 1h 1/2-1 h Ih 5 min

0.29-0.794 Not reported b 147-431 42-339 1633-2133

4% 2-6% 5% Not reported Not detected c

13 6 21 14 Present study

GFFL-AFC

aValuesare range and expressed as nmol/h/mg protein, hc'aluesfor onlyculturedskin fibroblastswere reported, cTwo atypicalcases had less than 2% activity.

Original article: Specific CLN2p/TPP-I assay The substrate A-A-F-7-amino-4-methylcoumarin (AMC) which was initially developed by Vines and Warburton for measuring TPP-I activity,6 has become the substrate of choice for many laboratories due to its commercial availability. The CLN2p/TPP-! activity measured by this substrate is quite sensitive and has shown reproducibility in different laboratories. 6,]4,2] There are reports of another serine protease activity against this substrate at least in leukocytes that may pose difficulties while analysing patients and carriers. 21 This problem is reflected by highly variable and overlapping CLN2p/TPP-I activities in the two groups comprising of controls and heterozygote carriers. All the laboratories that have used A-A-FAMC substrate for measuring the CLN2p/TPP-I activity in leukocytes have reported mostly indistinguishable levels for controls and carriers. 14,21 These laboratories have also reported residual CLN2p/TPP-I activities both in leukocytes and cultured skin fibroblasts in LINCL patients. 6,14,21 Another shortcoming of using A-A-F-AMC substrate is the slow progress of the enzyme reaction. At least l h incubation is necessary to measure CLN2p/TPP-I activity even by spectrofluorimetric detection of AMC. These shortcomings are not encountered with the peptide G-F-F-L-AFC used as substrate in assay conditions described here. The CLN2p/TPP-I assay described in the present study is extremely rapid, simple and highly specific for measuring only the CLN2p/TPP-I activity in tissue extracts. There is generally no residual CLN2p/ TPP-I activity in LINCL patients (except for two atypical cases), nor overlap between control and carrier groups. Less than 5 #g protein is enough to complete the assay with an incubation time of 5 minutes. Complete analysis takes less than 1 hour to distinguish LINCL patients and carriers from controls. The lower CLN2p/TPP-I specific activity measured by A-A-F-AMC substrate compared with the G-F-F-L-AFC substrate may be related to the more favourable conformation around the cleavage site. In the tripeptide A-A-F-AMC substrate, an amide bond between F-AMC is cleaved while in the tetrapeptide G-F-F-L-AFC substrate, an amide bond between two amino acids F-L is cleaved. The stearic hindrance due to an unnatural molecule AMC adjacent to the cleavage site may be responsible for much slower catalysis by the CLN2p/TPP-I. A significantly lower Km of 50/~M for the substrate G-F-F-L-AFC for bovine brain enzyme, 7 while a Km of 730 #M for the substrate AA-F-AMC was reported for the rat spleen CLN2p/ TPP-I, s which shows that G-F-F-L-AFC is a better substrate for the enzyme. In conclusion, enzyme-

67 based assays for CLN2p/TPP-I by both A-A-FAMC and G-F-F-L-AFC substrates may detect LINCL patients, however, for identification of heterozygote carriers in the high risk population, G-F-F-L-AFC is the substrate of choice.

Acknowledgements Supported in part by NIH grant NS 30147 and funds from New York State Office of Mental Retardation and Developmental Disabilities and Children's Brain Diseases Foundation, San Francisco, California, USA.

References 1 Zeman W, Donahue S, Dyken P, Green J. Leukodystrophies and poliodystrophies. In: Vinken PJ, Bruyn GW (eds) Handbook of Clinical Neurology. Amsterdam: Elsevier Science, 1970; Vol. 10: 588-679. 2 Boustany R-M. Neurodystropies and neurolipidoses. In: Moser HW (ed) Handbook of Clinical Neurology. Amsterdam: Elsevier Science, 1996; Vol. 66: 671-700. 3 Mole S. Batten's disease: eight genes and still counting? Lancet 1999; 354: 443-445. 4 Sleat DE, Donnelly RJ, Lackland H et al. Association of mutations in a lysosomal protein with classical late-infantile neuronal ceroid lipofuscinosis. Science 1997; 277: 1802-1805. 5 Rawlings ND, Barrett AJ. Tripeptidyl-peptidase I is apparently the CLN2 protein absent in classical lateinfantile neuronal ceroid lipofuscinosis. Biochim Biophys Acta 1999; 1429: 496-500. 6 Vines DJ, Warburton MJ. Classical late infantile neuronal ceroid lipofuscinosis fibroblasts are deficient in lysosomal tripeptidyl peptidase I. FEBS Left 1999; 443: 131-135. 7 Junaid MA, Wu G, Pullarkat RK. Purification and characterization of bovine brain lysosomal pepstatin-insensitive proteinase, the gene product deficient in the human late-infantile neuronal ceroid lipofuscinosis. J Neurochem 2000; 74: 287-294. 8 Vines D, Warburton MJ. Purification and characterization of a tripeptidyl aminopeptidase I from rat spleen. Biochim Biophys Acta 1998; 1384: 233--242. 9 Ito M, Dunn BM, Oda K. Substrate specificities of pepstatin-insensitive carboxyl proteinases from Gram-negative bacteria. J Biochem 1996; 120: 845-850. 10 Junaid MA, Pullarkat RK. Increased brain lysosomal pepstatin-insensitive proteinase activity in patients with neurodegenerative diseases. Neurosci Lett 1999; 264: 157-160. 11 Junaid MA, Clark GM, Pullarkat RK. A lysosomal pepstatin-insensitive proteinase as a novel biomarker for breast carcinoma. Int J Biol Markers 2000; 15: 129-134.

Original article: MA Junaid et al.

68

12 Junaid MA, Sklower Brooks S, Wisniewski KE, Pullarkat RK. A novel assay for lysosomal pepstatin-insensitive proteinase and its application for the diagnosis of late-infantile neuronal ceroid lipofuscinosis. Clin Chim Acta 1999; 281: 169-176. 13 Sohar I, Sleat DE, Jadot M, Lobel P. Biochemical characterization of a lysosomal protease deficient in classical late infantile neuronal ceroid lipofuscinosis (LINCL) and development of an enzyme-based assay for diagnosis and exclusion of LINCL in human specimens and animal models. J Neurochem 1999; 73" 700-711. 14 Young EP, Winchester BG, Peter Logan W et al. Exclusion of late infantile neuronal ceroid lipofuscinosis (LINCL) in a fetus by assay of tripeptidyl peptidase I in chorionic villi. Prenat Diagn 2000; 20" 337-339. 15 Berry-Kravis E, Sleat DE, Sohar I et al. Prenatal testing for late infantile neuronal ceroid lipofuscinosis. Ann Neurol 2000; 47: 254-257. 16 Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem 1951; 193: 265-275.

17 Sleat DE, Gin RM, Sohar I et al. Mutational analysis of the defective protease in classic late-infantile neuronal ceroid lipofuscinosis, a neurodegenerative lysosomal storage disorder. Am J Hum Genet 1999; 64: 1511-1523. 18 Zhong N, Wisniewski KE, Hartikainen J e t al. Two common mutations in the CLN2 gene underlie late infantile neuronal ceroid lipofuscinosis. Clin Genet 1998; 54: 234-238. 19 Liu C-G, Sleat DE, Donnelly RJ, Lobel P. Structural organization and sequence of CLN2, the defective gene in classical late infantile neuronal ceroid lipofuscinosis. Genomics 1998; 50" 206-212. 20 MacLeod P, Dolman C, Nickel R et al. Prenatal diagnosis of neuronal ceroid lipofuscinoses. Am J Med Genet 1985; 22" 781-789. 21

Sohar I, Lin L and Lobel P. Enzyme-based diagnosis of classical late infantile neuronal ceroid lipofuscinosis- comparison of tripeptidyl peptidase I and pepstatin-insensitive protease assays. Clin Chem. 2000; 46" 1005-1008.