ELSEVIER Forensic Science International

87 (1997) 145-1.54

Forensic Science

International

A sandwich enzyme immunoassay for brain S-100 protein and its forensic application

Yasuhisa Seo, Eiji Kakizaki, Keiichi Takahama”

Department of Legal Medicine, Miyazaki Medical College, Kiyotake, Miyazaki 889-16, Japan

Received 10 September 1996; revised 5 March 1997; accepted 18 March 1997

Abstract

A sensitive sandwich enzyme immunoassay for identification of brain S-100 protein in blood or bloodstains containing brain tissue is described. A polystyrene ball coated with rabbit anti-S-100 protein IgG was incubated with human S-100 protein, and then with anti-S-100 Fab’-peroxidase conjugate. Peroxidase activity bound to the polystyrene ball was assayed by fluorometry using 3-(4-hydroxyphenyl)propionic acid as the hydrogen donor. The detection limit of human S-100 protein was 0.6 pg (30 amol) per assay tube. The cross-reaction of this sandwich enzyme immunoassay to other organs was approximately l/100 or less. Antigenic activity of S-100 protein in bloodstains containing brain extracts was detectable after storage for 36 days at room temperature. The ratio of S-100 protein to total protein (ng/mg) in bloodstains when brain tissue was mixed with normal human blood at concentrations of 5-500 mglml was approximately loo-fold those of other samples (liver, heart, intestine, and skeletal muscle). These results indicated that bloodstains mixed with brain tissue were clearly distinguishable from others. Thus, in forensic practice, measurement of S-100 protein or the ratio of S-100 protein to total protein is useful to identify blood and bloodstains containing brain tissue. 0 1997 Elsevier Science Ireland Ltd.

Keywords: Brain; S- 100 protein; Enzyme immunoassay; Tissue identification

1. Introduction

S-100 protein, an acidic calcium-binding protein, was originally isolated from the central nervous system by Moore [ 11, and is thought to be one of the proteins specific to

*Corresponding author. Fax: +81 985 856406

0379-0738/97/$17.00 0 1997 Elsevier Science Ireland Ltd. All rights reserved PII SO379-0738(97)00049-2

146 Y. Sea et al. I Forensic Science International 87 (1997) 145-154

the nervous system [2]. In addition, it is a species-nonspecific antigen showing a close immunological relationship among various vertebrates [2]. S-100 protein is a dimer of two subunits, (Y and p, which have homologous amino acid sequences (58% identity) [3], and consists of three forms: (YOL (S-lOOa,), arp (S-lOOa), and l3p (S-100b) [4]. S-100a and S-100b are dominant in brain tissue, and S- lOOa, is a minor component (3-4% of the total S-100 protein) which is mainly distributed in the heart and skeletal muscle [4,5]. S-100 protein in cerebrospinal fluid (CSF) is utilized as a biochemical marker for evaluation of brain damage [6,7] and, histologically, it is used for diagnosis of melanocytic tumors [8].

In forensic practice, traumatic brain damage is frequently observed in traffic accidents, fall injuries, and firearm wounds. If multiple bullets or vehicles were found at the scene of a head injury, it would be important to determine which of them had damaged the brain.

A specific or especially abundant substance in brain tissue would be a useful marker for discriminating blood or bloodstains containing brain tissue. Kimura et al have shown that the brain-type myosin heavy chain isoform is an excellent marker for identification of brain tissue by using enzyme-linked immunosorbent assay [9].

Recently we described the liver-specific antigen (LSA), originally isolated from the human liver, which would be a useful marker for detecting liver injury [ 10,111. Yukawa et al reported that the detection of sucrase-isomaltase (SI), localized specifically in the small intestine, was available for identification of the stains on a knife which had penetrated the small intestine, and developed a sandwich enzyme immunoassay for SI 1121.

In the present study, we describe a sensitive sandwich enzyme immunoassay for brain S-100 protein and its application for detecting bloodstains mixed with brain tissue.

2. Materials and methods

2.1. Specimens

Human tissue samples (brain, lung, heart, liver, kidney, spleen, pancreas, intestine, and skeletal muscle) were obtained from cadavers at postmortem intervals of 12 to 24 hours, and were stored at -80°C until use. To prepare water-soluble protein fraction of various tissues, tissues were homogenized (50% w/v) in 0.25 M sucrose, pH 8.0, and centrifuged at 100,000 g for 60 min. The protein concentration of water-soluble protein fractions was calculated by the method of Lowry, using bovine serum albumin (BSA) as a standard [13].

Normal human sera were obtained from ten healthy males and ten healthy females. Normal human blood was collected in heparinized Vacutainer tubes.

2.2. Antigen and antibody

Standard human brain S-100 protein was purchased from Chemicon International Inc. (Temecula, CA, USA). Lyophilized human S-100 protein was reconstituted to a

Y. Sea et al. I Forensic Science International 87 (1997) 145-154 147

concentration of 1 mg/ml using 10 mM sodium phosphate (Nap) buffer, pH 7.0, containing 0.1 M NaCl.

Rabbit anti-bovine S-100 protein IgG was commercially obtained from DAKO- Immunoglobulins A/S (Copenhagen, Denmark). F(ab’), was prepared by digestion of IgG with pepsin, and Fab’ was prepared by reduction of F(ab’), with 2-mercap- toethylamine [ 141.

2.3. SDS-polyacrilamide gel electrophoresis (SDS-PAGE) and immunoblotting

SDS-PAGE was carried out with 15% homogeneous gel. The water-soluble protein fractions from various tissues were treated with 5% 2-mercaptoethanol and 1% SDS at 100°C for 5 min. The proteins separated by SDS-PAGE was electroblotted onto a PVDF membrane (Millipore, MA, USA), and was incubated with anti-S-100 protein IgG (1500) and then with anti-rabbit IgG (G+L) HRP conjugate (1:2000) (Biosource Inc., CA, USA). The peroxidase activity was visualized using a commercial immunostain kit for HRP (Konica Corp., Tokyo, Japan).

2.4. Sandwich enzyme immunoassay for human S-100 protein

Anti-S- 100 protein Fab’ was conjugated to horseradish peroxidase (Boehringer Mannheim GmbH, Mannheim, Germany) using N-succinimidyl-6-maleimidohexanoate (Dojindo Laboratory, Kumamoto, Japan) [15]. The amount of the conjugate was calculated from the peroxidase activity, which was assayed at 30°C by fluorometry using 3-(4-hydroxyphenyl)propionic acid as a hydrogen donor [ 141. The fluorescence intensity was measured relative to 1 mg/l quinine in 50 mM H,SO, at an excitation wavelength of 320 nm and an emission wavelength of 405 nm using a Hitachi 650-60 spec- trofluorophotometer (Hitachi, Ltd., Tokyo, Japan).

Polystyrene balls (3.2 mm in diameter, Immuno Chemical Inc., Okayama, Japan) were coated with rabbit anti-S-100 protein IgG (0.1 mg/ml) by physical adsorption [16].

An anti-S-100 protein IgG coated polystyrene ball was incubated with human S-100 protein in a total volume of 0.15 ml at 30°C for 4 h and 4°C for 16 h. Standard human S-100 protein was diluted from 2 ng (100 fmol) to 0.6 pg (30 amol) per tube (0.15 ml) with 10 mM NaP buffer, pH 7.0, containing 0.5 M NaCl, 0.1% BSA, and 0.1% NaN,. The polystyrene ball was washed twice with 10 mM NaP buffer, pH 7.0, containing 0.1 M NaCl, and then incubated with 50 ng of anti-S-100 protein Fab’-peroxidase conjugate at 20°C for 3 h. The polystyrene ball was washed twice and then transferred to a clean tube. Bound peroxidase activity was assayed as described above.

2.5. Assay procedure

Water-soluble protein fractions from various organs were serially diluted to protein concentrations of 10 pg to 100 rig/ml, and 0.15 ml of each sample was subjected to the sandwich enzyme immunoassay.

A 5 pl of serum sample from each of ten healthy males and females was diluted to a total volume of 0.15 ml and assayed as described. The recoveries of S-100 protein added

148 Y. Seo et al. I Forensic Science International 87 (1997) 145-154

to 1 pl or 5pl of human serum were estimated. To each serum sample 1.2 ng and 0.06 ng of S-100 protein was added, followed by dilution to a total volume of 0.15 ml.

For detection of S-100 protein in bloodstains, blood was placed on filter paper (Phenylketonuria-filter for mass-screening, Toyo Roshi Co. Ltd., Tokyo, Japan) and stored at room temperature for 1 to 36 days. Blood samples were prepared from normal human blood and blood mixed with 20 kg/ml of brain extracts. Filter paper discs with bloodstains were punched out to a standard diameter of 3 mm. Two or eight pieces of these filter paper discs were then immersed in 0.81 ml of 10 mM NaP buffer, pH 7.0, containing 0.5 M NaCl and 0.1% NaN,, and the blood was extracted at 30°C for 2 h. The volume of blood per disc was estimated to be 2.7 pl [17]. The protein concentration of these samples was calculated, and 0.15 ml of these extracts were assayed.

To obtain practical samples, brain tissues were added in proportions of 5 kg/k1 to 500 p,g/p,l to normal human blood, homogenized, and the blood was air dried at room temperature for 1 to 9 days. After one, three, or nine days, the bloodstains mixed with brain tissues were extracted to a final concentration of 1 mg/tube (0.15 ml) at 30°C for 2 h. The total protein concentrations were calculated and 0.15 ml of these extracts were assayed. Bloodstains mixed with heart, liver, intestine, and skeletal muscle tissues, respectively, were prepared and assayed as described.

3. Results and discussion

3.1. Detection of S-100 protein in human brain

The water-soluble protein fraction extracted from human brain was separated by 15% SDS-PAGE and was electroblotted. An anti-bovine S-100 protein IgG showed a single clear band corresponding to the molecular mass of about 10 kDa, as detected by immunostaining (Fig. 1). In addition, no cross-reaction against various organs (lung, heart, liver, kidney, spleen, pancreas, intestine, and skeletal muscle) was observed (data not shown).

3.2. Detection limit of human S-100 protein

The dose-response curve of the sandwich enzyme immunoassay for human S-100 protein is shown in Fig. 2. The detection limit of human S-100 protein was 0.6 pg (30 amol) per assay (0.15 ml), which was l/5 that achieved by other ELISA methods for S-100 protein previously reported [ 18,191.

S-100 protein is a species-nonspecific protein [2] and the anti-bovine S-100 protein antibody reacted closely with human S-100 protein [7]. The amino acid sequences in human and bovine S-100 protein are different at only one residue of the (Y subunit [3] and three residues of the B subunit [20].

3.3. Organ-specijicity of the sandwich enzyme immunoassay

Proteins extracted from human brain showed the strongest response with a maximum fluorescence intensity of 1,502 on average, whereas those of proteins from intestine and

Y. Seo et al. I Forensic Science International 87 (1997) 145-154 149

kOa

. B8 43lO

Fig. 1. SDS-PAGE and immunoblotting of water-soluble protein fraction (20 p,g) extracted from human brain. A: Lane 1, Molecular weight marker. Lane 2, Water-soluble protein fraction of human brain. B: Water-soluble protein fraction of human brain separated on 15% homogeneous gel and immunostaining (arrow shows the band corresponding to S-100 protein).

./

5

! I p"B

%

10

l . - . . . ..- A . . . ..- A . 0 0.1 1 10 100 loo0

Human S-l 00 Protein @/tube)

Fig. 2. Dose-response curve of human S-100 protein by sandwich enzyme immunoassay.

150 Y. Seo et al. I Forensic Science International 87 (1997) 145-154

0 0.01 0.1 1 10 loo

Protein Concentration @/ml)

Fig. 3. The sandwich enzyme immunoassay for water-soluble protein fractions from various tissues including brain (O), lung (U), heart (0). liver (O), kidney (4), spleen (A), pancreas (0), intestine (A), and skeletal muscle (X).

heart were 16.1 and 6.1, when 100 rig/ml of proteins were assayed (Fig. 3). Cross- reaction of the sandwich enzyme immunoassay to intestine and heart were about l/ 100 and l/300, respectively. Kato et al reported that the homodimer of the CY subunit was present at six and three times higher in heart and skeletal muscle, respectively, than in brain, and the content of the homodimer of the p subunit in intestine was estimated to be about l/30 of brain [5]. In brain tissue, S-100 protein consists mainly of a heterodimer and a homodimer of the p subunit, whereas the homodimer of the (Y subunit represents only 3 to 4% of the total S-100 protein [21].

3.4. Serum levels of S-100 protein and recovery of S-100 protein added to serum

The concentration of S-100 protein in normal human serum showed values below the detection limit of this sandwich enzyme immunoassay, so the level of S-100 protein in normal human serum was calculated to be less than 0.12 rig/ml.

The recovery rates of 1.2 ng or 0.06 ng S-100 protein added to 5 ul of normal human serum were determined to be 98.6+3.O(S.D.)% and 85.5+9.8(S.D.)%, respectively (Table 1). When 1 ~1 of serum was used, the recoveries were calculated at 99.4?1.7(S.D.)% and 83.7+8.O(S.D.)%, respectively. No differences between serum samples from male and female subjects were observed.

Y. Sea et al. I Forensic Science International 87 (1997) 145-154 151

Table 1 Recovery of S-100 protein added to serum

Sample” S- 100 protein Volume of sernm added (ng / tube) used (~1)

1 1.2 5

2 1.2 1 3 0.06 5

4 0.06 1

* The assays were performed on 10 male and 10 female subjects.

Recovery % (mean5S.D.)

98.623.0

99.421.7

85.529.8

83.7k8.0

3.5. Stability of S-100 protein in bloodstains

S-100 protein in normal human blood was not detectable, but the immunoreactive S-100 protein in the bloodstains mixed with brain extract was recovered well after one day of storage (Fig. 4). The level of immunoreactive S-100 protein, after 36 days of storage at room temperature, gradually decreased, and 35.3% and 37.9% of the activity was lost, when 2 or 8 discs of filter paper, respectively were used. The volume of blood extracted from 2 and 8 discs were estimated to be 1 ~1 and 4 kl/tube (0.15 ml), respectively [ 171.

The ratio of S- 100 protein to total protein also declined, but the total protein

02 -

I

==-e====*

x li

O&.~.‘.=..‘~..-‘....‘...=‘...-‘....’= 6 10 16 20 26 50 E%

Duratknofswagebyd

Fig. 4. Time course of changes of human S-100 protein and the ratio of S-100 protein to total protein in bloodstains; (0): bloodstain from normal subjects (n=3); (M): S-100 protein levels (mean and S.D.) from bloodstains mixed with 20 pg/ml of brain extract in two pieces of filter paper per tube (n=3); (A): S-100 protein levels from bloodstains mixed with brain extract in eight pieces of filter paper per tube (n=3); (0): ratio of S-100 protein to total protein from two pieces of filter paper per tube; (A): ratio from eight pieces of filter paper per tube.

152 Y. Seo et al. I Forensic Science International 87 (1997) 145-154

concentrations in each sample extracted from bloodstains did not significantly change (Fig. 4).

3.6. Detection of S-100 protein in bloodstains

The detection threshold of the sample mixed with brain tissue was estimated at 5kg tissue/p1 blood in which case 1.2kO.36 (S.D.) ng of S-100 protein could be detected (data not shown). The protein concentration of one mg of bloodstains corresponded to that of four ~1 of blood. Total protein concentrations and the S-100 protein levels in each sample, which were extracted one, three, and nine days later, were scarcely changed. The ratio of S-100 protein to total protein in terms of ng/mg was approximate- ly loo-fold in bloodstains mixed with brain tissue than that of other tissues (Fig. 5). Although the ratio of S-100 protein to total protein in bloodstains mixed with 500 p,g of various tissues (heart, liver, intestine, and skeletal muscle) were less than 0.5, the ratio of bloodstains mixed with 5 p,g of brain tissue was reverse of other samples which assayed 500 Fg of tissues (Fig. 5). These results suggested that blood mixed with brain tissue was clearly distinguishable from the other blood, when the ratio of S-100 protein to total

loo

Fig. 5. Detection of human S-100 protein from bloodstains mixed with various tissues. The sandwich enzyme immunoassays were performed after one, three, or nine days storage at room temperature; (0): ratio of S-100 protein to total protein in bloodstains mixed with brain tissue (n=6; mean and standard error); (0): heart (n=3; mean); (0): liver: (0): intestine; (A): skeletal muscle.

Y. Sea et al. I Forensic Science International 87 (1997) 145-154 153

protein was higher than 0.5, and it was possible to detect at least 10 pg of brain tissue mixed and dried with 1 ~1 of blood.

References

[I] B.W. Moore, A soluble protein characteristic of the nervous system. Biochem. Biophys. Res. Commun. 19 (1965) 739-744.

[2] E. Bock, Nervous system specific proteins. J. Neurochem. 30 (1978) 7-14. [3] T. Isobe, T. Okuyama, The amino-acid sequence of the (Y subunit in bovine brain S-1OOa protein. Eur. J.

Biochem. 116 (1981) 79-86. [4] T. Isobe, T. Okuyama, The amino-acid sequence of S- 100 protein (PAP I-b protein) and its relation to the

calcium-binding proteins. Eur. J. Biochem. 89 (1978) 379-388. [5] K. Kato, S. Kimura, SltXla, (OILY) protein is mainly located in the heart and striated muscles. B&him.

Biophys. Acta. 842 (1985) 146-150. [6] A. Kruse, K.G. Cesarini, F.W. Bach, L. Persson, Increases of neuron-specific enolase, S-100 protein,

creatine kinase and creatine kinase BB isoenzyme in CSF following intraventricular catheter implanta- tion. Acta Neurochir. 110 (1991) 106-109.

[7] L. Persson, H.G. Hbrdemark, M.D.J. Gustafasson, G. Rundstrom, I. Mendel-Hattvig, T. Esscher, S. Pahlman, S-100 protein and neuron-specific enolase in cerebrospinal fluid and serum: markers of cell damage in human central nervous system. Stroke 18 (1987) 911-918.

[8] L.G. Kindblom, P. Lodding, L. Rosengren, J. Baudier, K. Haglid, S-100 protein in melanocytic tumors. Acta Path. Microbial. Immunol. Stand. Sect., A 92 (1984) 219-230.

[9] A. Kimura, H. Ikeda, S. Yasuda, K. Yamaguchi, T. Tsuji, Brain tissue identification based on myosin heavy chain isoforms. Int. J. Leg. Med. 107 (1995) 193-196.

[IO] Y. Seo, K. Takahama, A sandwich enzyme immunoassay for liver-specific antigen and its forensic evaluation. Jpn. J. Legal Med. 46 (1992) 169-176.

[ 1 I] Y. Seo, K. Takahama, A highly sensitive sandwich enzyme immunoassay for human liver-specific antigen (LSA) and its forensic application. Jpn. J. Legal Med. 48 (1994) 150-155.

[ 121 N. Yukawa, K. Takahama, Detection of sucrase-isomaltase in stains on a knife by sandwich enzyme immunoassay to determine penetration of the small intestine in a pig model. Jpn. J. Legal Med. 49 (1995) 222-228.

[13] O.H. Lowry, N.J. Rosebrough, A.L. Farr, R.J. Randall, Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193 (1951) 265-275.

[I41 E. Ishikawa, M. Imagawa, S. Hashida, S. Yoshitake, Y. Hamaguchi, T. Ueno, Enzyme-labeling of antibodies and their fragments for enzyme immunoassay and immunohistochemical staining. J. Immuno- assay. 4 (1983) 209-327.

[15] S. Hashida, M. Imagawa, S. Inoue, K-H. Ruan and E. Ishikawa, More useful maleimide compounds for the conjugation of Fab’ to horseradish peroxidase through thiol groups in the hinge. J. Appl. Biochem. 6 (1984) 56-63.

[16] E. Ishikawa, K. Kato, Ultrasensitive enzyme immunoassay. Stand. J. Immunol. 8, Suppl. 7 (1978) 43-55.

[17] S. Hashida, K. Tanaka, T. Kohno, E. Ishikawa, H. Umehashi, T. Hayashida, T. Mori, H. Imura, H. Ogawa, A highly sensitive sandwich enzyme immunoassay of human thyroid-stimulating hormone in dried blood on filter paper discs for mass-screening of neonatal hypothyroidism. J. Clin. Lab. Anal. 1 (1987) 198-204.

[ 181 F. Suzuki, T. Nakajima, K. Kato, Peripheral distribution of nervous system-specific S-100 protein in rat. J. Biochem. 92 (1982) 835-838.

[19] M. Takayasu, M. Shibuya, M. Kanamori, Y. Suzuki, K. Ogura, N. Kageyama, H. Umekawa, H. Hidaka, S-100 protein and calmodulin levels in cerebrospinal fluid after subarachnoid hemorrhage. J. Neurosurg. 63 (1985) 417-420.

1.54 Y. Sea et al. I Forensic Science International 87 (1997) 145-154

[20] T. Isobe, A. Tsugita, T. Okuyama, The amino acid sequence and the subunit structure of bovine brain S-100 protein (PAP I-b). J. Neurochem. 30 (1978) 921-923.

[21] T. Isobe, N. Ishioka, T. Masuda, Y. Takahashi, S. Ganno, T. Okuyama, A rapid separation of SlOO subunits by high performance liquid chromatography: the subunit composition of SlOO proteins. Biochem. Int. 6 (1983) 419-426.