Biochimica et Biophysica Acta. 1074(1991) 201-208 © 1991 ElsevierScience PublishersB.V. 0304-4165/91/$03.50 ADONIS 030d41659100177B
201
Tissue distribution and developmental profiles of immunoreactive aB crystallin in the rat determined with a sensiti~.e immunoassay system K a n e f u s a K a t o t, H a r u o S h i n o h a r a ~, N a o m i K u r o b e ~, Y u t a k a I n a g u m a 1, Kikuo Shimizu 2 and Kunihiro Ohshima 3 t Depart,lent of Biochemistry. Institutefor Deeelopmental Rexearch..qtchi Prefectural ('olot~v. Kasugai. Aichi (JapanL : Medwal and Biological Laboratories Co., Ltd.. Ina, Nagano (Japan1 and ¢ Technical Research Laboratories. Kurabo Industries l t d . Ne~'::',,;.z.va. Osaka ¢Japan)
(RL~.eived13 February 19911
Key words: Crystallin; lmmunoassay:Immunohistochemistry In order to determine the q'. ~:Jtitative distribution o f ~ crystailin (aB) in non-lentieular tissues, we have established a sensitive immunoassay system for specific measurement of aB. Antisera were raised in rabbits by injecting a B purified from bovine lenses, or C-terminal decapeptide (KPAVTAAPKK) of a B (aBp~p). The antibodies to a B and a l ~ p w e r e purified by the use of a aB-coupled Sepharose column. The F(ab') z fragments of antibody IgG to a B were immobilized on polystyrene balls and the Fab' fragments of antibody IgG to aBp~p were labeled with fl.D-ga!actosidase from Escherichia c o i l T h e sandwich-type enzyme immunoassay consisted of the abov~ two antibodies was sensitive, and the minimum detection limit of the assay was 10 pg a B without any crossreactivity with aA. By using the assay method, it is revealed that the a B was distributed in most of the tissues examined. Among the non-lentieular tissues, a B was present at high levels in the heart and striated muscles, especially in the soleus muscle, and kidney. High levels of a B in the muscle tissues were also seen in various animals. Developmental increases of a B in rat muscle tissues and kidney were observed from 16 days of gestational age to I or 5 weeks of postnatal age. In contrast, the a B in the brain kept a low level during the same period. After 5 weeks of age, a B concentrations in the brain increased sharply, reaching the adult levels at 9 weeks of age. lmmunohistochemieal staining with anti-aBp~p revealed that a B was positive not only in glial cells, in the central nervous tissues, but also in some neurons of spinal cord, brainstem, hippueampus, and olfactory bulb. Spermatocytes in the testis were also immunopositive for aB.
Introduction a Crystallin is a major structural protein of the vertebrate lens, and it is mainly composed of two acidic ( a A t and a A 2 ) and two basic (aBi and aB2) subunits [1]. It has been shown that aA2 and a ~ are the primary products of two distinct genes, which are located on different h u m a n chromosomes (on the long arm of chromosome 21 [2] and 11 [3], respectively), a A I
Abbreviations: aA and aB, the A and B subunit, respectively,of a crystallin" aB~p. C-terminal decapeptide (KPAVTAAPKK)of aB" SDS-PAGE. sodium dodecyl sulfate-polyacrylamidegel electrophoresis; HPLC, high performanceliquid chromatography. Correspondence: K. Kato. Department of Biochemistry,Institute for DevelopmentalResearch,Aichi PrefecturalColony. Kamiya, Kasugai. Aichi 480-03, Japan.
and aB~ are considered to be the phosphorylated forms of a A , and a ~ . respectively [4]. The molecular mass of each of these subunits is about 20 kDa [5,6]. However, a crystallin extracted from the lens has a molecular mass of approx. 800 kDa [7], indicating ~ crystallin is present as a polymeric form. Recently, it has been revealed that orb is also expressed in various non-lenticular tissues, including heart, skeletal muscle, kidney and nervous tissues [8-12], and it accumulates in the brain of patients with Alexander disease as a major component of Rosenthal fibers [9]. In order to clarify the physiological significance of aB in the non-lenticular tissues, we established a sensitive and specific enzyme immunoassay system by using rabbit antibodies raised against aB and the C-terminal decapeptide of a B , and determined the quantitative distribution of aB in rat tissues and developmental changes of aB levels in rat central nervous tissues, muscle tissues and kidney. Some
202 tion was repeated every 2 weeks, and the rabbits were bled after 2 months. The antibodies to aB and aBp~p were purified by the use of the aB-coupled Sepharose column with the same procedures as described previously [17]. The purified antibodies to aB then were passed through the aA-coupled Sepharose column to eliminate the crossreacting antibodies to aA. The yield of the antibody lgG in the above procedures was about 5 rag/100 ml serum. As shown in Fig. 1. the anti-aB and anti-aBp~p antibodies were specific to aB showing no reactivity with aA and other components of the crude extract of rat soleus muscle on the immunoblotting test, although the reactivity of anti-aB was significantly lower than that of anti-aBp~p (Fig. IC).
f!ndings on the immunohistochemical localization of aB were also described. Materials and Methods
Purification of aA and aB cr)'stallins Bovine a crystallin was purified from fresh lenses obtained at a local slaughter house as described by Spector et al. [13], and the aA and aB crystallins were isolated by the use of chromatofocusing column chromatography in the presence of 6 M urea as described by Bloemendal and Groenewoud [14]. The aA l and aB 2 crystallins were used as the aA and aB antigens, because each of the two fractions was homogeneous showing a single band on the polyncrylamide slab gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE) as shown in Fig. IA. The purified aA ~nd aB were stored at - 2 0 ° C in a 30% glycerol solution with 0.1% NAN3.
Enzyme immunoassay system for aB In order to increase the sensitivity of the assay and to p, otect the assay with the sample interference [18], the assay system was prepared with the antibody fragments without Fc portion. The purified antibody IgG was digested with pepsin, and the resultant F(ab')2 fragments were isolated by the use of a Sephadex G-150 column [19]. The antibody F(ab')2 fragments were immobilized noncovalently on polystyrene balls (3.2 mm in diameter, lmmuno Chemicals, Okayama, Japan) as described previously [20], and the balls were stored at 4 ° C in 10 mM sodium phosphate buffer (pH 7.0) containing 0.1 M NaCI, 1 mM MgCI 2, 0.1% bovine serum albumin, and 0.1% NaN 3 (buffer A) at least overnight before use for the immunoassay. The antibody F(ab')2 fragments were reduced with 2-mercaptoethylamine and the resultant Fab' fragments were coupled with fl-D-galactosidase from E. coil (Boehringer Mannheim, Mannheim, F.R.G.) by the use of N,N'-o-phenylenedimaleimide as described [19,21]. Amounts of labeled antibodies are expressed in units of
Peptide synthesis attd its conjugation to hemo~yanin The C-terminal decapeptide (KPAVTAAPKK) [6] of aB (aBpCp) was synthesized by solid-phase procedures [15] using a peptide synthesizer (PEPTICOU PLER 2200, Vega Biotechnologies, Tucson, AZ, U.S.A.) and purified with reverse-phase HPLC. The purity of the peptide was assessed by analytical HPLC and amino acid analysis. The peptide was conjugated with the equal amount of keyhole limpet hemocyanin (Sigma Chemicals, St. Louis, MO, U.S.A.) as described by Goldsmith et al. [161.
A ntisera and purification of antibodies Antisera were raised in rabbits by subcutaneous injection of the purified aB (about 0.5 mg/rabbit), or the same amounts of aBr,~pconjugated with the hemocyanin with Freund's complete adjuvant. The same immuniza-
m
A
B
143 30"---m
W
1
2
3
4
2
3
4
2
3
4
Fig. 1. SDS-PAGEof the purified aA and aB crystallinsand an extract of rat soleusmuscle(A), and immunoblotswith purified anti-aBpop(B) and anti-aB (C). Lane 1, standard proteinswith molecularmassin kDa; lane 2, 1 /~gof purified t~Al: lane 3, 1 #g of purified aB2.; lane 4, an extract of rat soleus musclecontaining about 50 .ug proteins. (A) Polyacrylamide 0570) gel stained with Coomassieblue: (B and C) nitrocellulose sheet stained with purified antibodies (0.02 p.g/ml) raised with aB~o, and aB, respectively. Peroxidase activity on the sheet was visualized with diaminobenzidine and H202.
203 galactosidase activity, and i unit of the activity is defined as that producing 1 ~tmol 4-methylumbelliferone/min under the conditions. Unless otherwise specified, a polystyrene ball with anti-aB antibodies was incubated in duplicate with shaking at 3 0 ° C for 5 h with a 10-/~1 aliquot of samples or standard aB crystallin in a final volume of 0.5 ml with 10 mM sodium phosphate buffer (pH 7.0) containing 0.3 M NaC1, 1 mM MgCI 2, 0.1% proteinase-treated gelatin [22], and 0.1% NaN3 (buffer G). Each ball was washed twice with 1 ml of cold buffer A, transferred into a fresh test tube with 0.2 ml of buffer A containing 1.5 milliunits of galactosidase-labeled anti-aBp~p antibodies, and incubated with shaking at 4 ° C overnight. The ball was washed as described above and the galactosidase activity bound on the ball was assayed with a fluorogenic substrate, 4-Methylumbelhferyl-/3-ogalactoside (Sigma), as described previously [20].
Preparation of tissue extract Wistar rats of various gestational and postnatal days were used. Tissues were sampled and kept frozen at - 8 0 ° C until analysis. The frozen tissues from postnatal rats were homogenized at 0 ° C with a Physcotron (NS-50, Niti-On, Chiba, Japan) in a 10- to 20-volume (v/w) of 50 m M Tris-HC1 (pH 7.5) containing 5 mM FDTA. The fetal rat tissues were homogenized with a Teflon-glass homogenizer and then each homogenate was sonicated at 0 ° C for 30 s. All ilomogenates were centrifuged at 4 ° C at 2 0 0 0 0 × g for 20 rain, and the supernatant fractions were used for analysis after dilution with buffer G. Bovine and porcine tissues were obtained at the slaughter house, and h u m a n tissues were obtained at autopsy. All samples were kept frozen at - 8 0 ° C and prepared as described above. Electrophoresis and immunoblotting SDS-PAGE was performed by the method of Laemmli [23]. lmmunoblots were carried out as described Towbin et al. [24] with modifications. In brief, the nitrocellulose sheets were incubated for 1 h with shaking in buffer G containing purified antibody lgG (0.02 p,g/ml), and then for 30 min with 1000-fold diluted goat (anti-rabbit IgG) lgG-labeled with horseradish peroxidase (Medical and Biological Laboratories, Nagoya, Japan). The peroxidase activity on the sheet was visualized with diaminobenzidine and H202 [25] or ECL Western blotting detection system (Amersham International, Amersham, U.K.). hnmunohistochemical staining Adult Wistar rats were deeply anesthetized with chloroform and perfused via the heart with physiological • saline, followed by a fixative solution composed of 1% glutaraldehyde, 4% paraformaldehyde, 0.2% picric acid.
and 2% sucrose in 0.1 M sodium acetate buffer (pH 6.0). The tissues were dissected out and fixed by soaking in the same buffer for 4 h. Tissues were rinsed and left standing in 50 mM Tris-HCI buffer (pH 7.4) overnight, then dehydrated, and embedded in paraffin. Sections (5-6 ~tm thicl:) were prepared and immunostained with purified antibody lgG to aBpep (2 /tg/ml) according to the indirect peroxidase-labeled antibody method [26]. using diaminobenzidine as the chromogen. For controls. sections were incubated with antibodies preabsorbed with aBp~p, which gave no positive staining.
Other methods Protein concentrations of tissue extracts were estimated with Bio-Rad protein assay (Bio-Rad, Richmond, CA, U.S.A.), which utilizes a principle of protein-dye binding [27]. Concentrations in Itg of the purified crystallins were also determined with the same method using bovine serum albumin as a standard. Results
Detection limit and specificio' of the #nmunoassay systems for a B crvstallin The three assay systems for aB were prepared by the combination of two antibody preparations, anti-aB and anti-aBo~p. As shown in Fig. 2, the assay system consisted of polystyrene balls with anti-aB and the galactosidase-labeled anti-aB~p was apparently more sensitive as compared with the system prepared with anti-aB or anti-aB~p alone, and the minimum detection limit, defined as the lowest concentration giving a galactosidase activity significantly different from that of zero standard at > 0.99% confidence, was less than 10 p g / a s s a y tube. Therefore. unless otherwise specified, the aB assay was conducted with the system consisted of two antibody preparations. The assay did not crossreact with aA. These results indicate that the aB assay system with two antibody preparations was sensitive and specific to the B subunit of a crystallin. Determinations of aB crystallin in various rat tissues In order to find the proportionality of the assay with tissue extracts and the optimum sample volume of each extract, several extracts of muscle tissues and kidney were diluted serially and the diluted extracts were subjected to the immunoassay for aB. As shown in Fig. 3, the values determined in each extract were p~oportional to the sample volume employed to the assay. The precision of aB assay was tested by assaying four samples of rat spleen extract ten times in one assay (within-assay) or in duplicate in seven consecutive assays (between-assay). The coefficients of variation in each assay were < 12%. The analytical recovery of
204 !000 ~
/,tj
TABLE I Concentrations of aB crystallin in t,ariolt~ rat tissues
i
The extracts of various adult male rat tissues were immunoassayed.
1
0
g
, 0
0.01
0.1 a
1
10
Crystallin (ng)
Fig. 2. Standard curves of the assay for aB crystallin and its crossreactivity with aA crystallin. Indicated amounts of aB crystallin (solid lines) or aA crystallin (broken line) were incubated in duplicate with polystyrene balls with anti-aB (e, t:], A) or anti-aBp¢p (o), and then the balls was incubated with the galactosidase-labeled anti-aB (ra) or anti-aBpe p (e. o. A) as described in the text. fl-D-Galactosidase activity bound on the ball is expressed as the fluorescence intensity of 4-methylumbelliferone produced in a 20-min reaction at 30°C with 0.1 mM 4-methylumhelliferyl-fl-D-galactoside. In the fluorescence intensity scale. 1000 equals 1 ,ttM 4-methylumbelliferone.
b o v i n e a B c r y s t a l l i n s (1 n g ) a d d e d to t h e i m m u n o a s s a y t o g e t h e r w i t h 5 /tl o f t h e s p l e e n e x t r a c t ( n = 5) w a s 101 _+ 7%. T h e s e r e s u l t s i n d i c a t e t h a t t h e c o n c e n t r a t i o n s 10
Tissues
aB crystallin ( n g / m g protein)
Cerebral cortex Cerebellum Brainstem Hippocampus Tongue Heart Esophagus Diaphragm Soleus muscle Rectus femoris muscle Kidney Bladder Testis Rectum Stomach Cecum Liver Spleen Thymus Lung Adrenal Xiphoid Adipose Blood plasma
6.3+ 2.1" 14.0_+ 2.1 36.5_+ 3.7 9.3_+ 1.7 371 _+113 2110 +400 1640 +250 2090 +420 19500 _+800 52.2+ 17.4 125 -+ 45 t 2.7_+ 2.9 1.2+ 0.2 14.6_+ 5.4 1.6_+ 0.6 4.2_+ 0.8 0.4 + 0.2 4.9 + 0.8 1.4+ 0.3 9.1 _+ 3.3 1.3+ 0.4 31.4+ 33.4 7.5 + 1.0 1.5 + 0.7 h
a Mean+_l S.D. of five rats. h ng/ml.
o f a B in t h e e x t r a c t c a n b e d e t e r m i n e d b y t h e p r e s e n t method. Table I shows the concentrations of immunoreactive a B in t h e s o l u b l e f r a c t i o n s o f v a r i o u s r a t t i s s u e s e x pressed as bovine aB equivalent ng/mg soluble protein. T h e a v e r a g e c o n c e n t r a t i o n o f a B in r a t lens w a s 247 p , g / m g p r o t e i n . I m m u n o r e a c t i v e orB c r y s t a l l i n w a s d e t e c t e d in t h e m o s t t i s s u e s e x a m i n e d , a n d it w a s p r e s e n t
TABLE II Concentrations of aB crystallin in human, bovine and porcine muscle tissues
Tissues
'O.1'
K/
O. 01 O. 1 1 10 Extract added ( id ) Fig. 3. Effect of sample dilution on the assay of aB crystallin in the crude extract. The soluble fraction of 1070 (w/v) homogenate of rat soleus muscle (o), heart (A), tongue in), or kidney (o, n) was diluted variously, and the samples containing the indicated volumes of original extract were subjected to the immunoassay. Each point is the mean of a duplicate assay.
Human heart pectoral muscle Bovine heart intercostal muscle Porcine heart intercostal muscle Mean±S.D.
No. of samples
aB crystallin ( n g / m g protein)
l 5
3080 4480+ 1070 a
5 5
4220+2280 4140 + 1540
5 5
956_+ 476 381 + 76
205 at high levels in tissues composed of striated muscle (such as leg muscles, diaphragm, esophagus, and tongue) and heart muscle. The soleus muscle, a slow twitch muscle, contained a much higher level (about 400-fold) of a B as compared witit the rectus femoris muscle, a fast twitch muscle, confirming that the a B is mainly localized in the type I muscle fibers as reported by lwaki et al. [12]. The kidney also contained a considerable a m o u n t of orB. Significant a m o u n t s of aB were present in the central nervous tissue, rectum, bladder. spleen, and xiphoid, a cartilaginous tissue. Rat blood p l a s m a contained about 1.5 ng a B / m l . High levels of immunoreactive a B were also determined in the bovine, porcine, and h u m a n muscle tissues (Table I1).
Immunoblotting of intmunoreactive aB in the crude extract On SDS-PAGE, the immunoreactive a B in rat soleus muscle (Fig. 1), rectus femoris muscle, kidney, and bladder (data not shown) migrated to the same position as that of bovine lens aB. The i m m u n o r e a c t i v e a B estimated in other muscle tissues of rat, human, bovine, and porcine also showed a single b a n d at the same position as that of bovine lens a B on the immunoblotting test (Fig. 4). These results indicated that the a B measured in each extract was very similar, if not identical, to lens a B as reported previously [8,9], and also suggested that the present assay specifically measured a B molecules in the crude extract.
Developmental profiles of aB in rat brain, muscle tissues. and kidney By using the present assay method, developmental changes of a B concentrations in rat central nervous
1
2
3
4
5
6
7
8
Fig. 4. Immunoblot of the extracts of muscle tissues of various animals with anti-aBpcp. The soluble fractions (10/~1) of 10% homogenates of rat heart (lane 2) and diaphragm (lane 3) human heart (lane 4l and pectoral muscle (lane 5). bovine heart (lane 6) and intercostal muscle (lane 7), and porcine heart (lane 8). together with 0.5 ag of purified aB (lane 1), were subjected to SDS-PAGE, followed by immunoblot. The peroxidase activity on the sheet was visualized with 3,3'-diaminobenzidine and H202.
6o
o m
o~
40
==
0-2
0
2
4 6 Weeks after birth
8
10
~' Fig. 5. Developmental profiles of aB crystallin in rat central ncr,'ous tissues. Concentrations (ng/mg protein) of aB in rat cerebral cortex (©). cerebellum (el. and brainstem ([:3)were determined from 15 days of gestational age to 9 weeks of postnatal age. tissues, heart, tongue, and kidney were determined from 15 days of gestational age to 9 weeks of postnatal age. As shown in Fig. 5, the a B was detected in the cerebral cortex, cerebellum, and brainstem of embryos, showing the concentration of < 5 n g / m g protein (1 to 3 n g / m g protein). The low levels of a B were seen until 4 to 5 weeks of age. Then the ~tB contents in the brain were increased sharply, reaching the adult levels at 9 weeks of postnatal age. C h a n g e s in the a B levels of kidney, tongue, and heart are shown in Fig. 6. The a B was detectable in the three organs of rat embryos. The a B in the kidney increased after birth and reached a plateau level at 5 weeks of age. The a B concentration in the embryonic heart was 1 0 - 5 0 n g / m g protein, but it increased sharply after birth, reaching the adult level of a b o u t 1500 n g / m g protein within 2 weeks of age. The a B in the tongue increased just after the birth and reached a peak value (about 800 n g / m g protein) o n 8 days of age, and then decreased to
i ~1~°
o2
0
2
4 6 Weeks after birth
8
10
Fig. 6. Developmental profiles of aB crystanin in rat heart, tongue, and kidney. Concentrations (ng/mg protein) of aB in rat heart (o). tongue (O), and kidney (O) were determined as described in Fig. 6.
206
D
¥
,/
~
t
,k !
.¢
A
3B
cA
Fig. 7. Immunohistochemical staining with anti-aB~p. The sections were stained with purified anti-aB~p IgG (2 .ug/ml) by the indirect peroxidase-labeled antibody method. (A) Immunoreactive neurons in the spinal cord. Arrow indicates the immunostained glial cell. Inset: immunoreactive glial cell, which seems to be an oligodendrocyte. (B) Mitral cells in the olfactory bulb are also immunoreaetive. (C) Scbwann cells in the peripheral nerves in the tongue are immunoreactive. On the right are immunoreactive muscle fibers of the tongue. (D) lmmunoreactive muscle fiber in the tongue. Muscle fibers are stained heterogeneously. (E and F) The spermatoeytes are immunoreactive but spermatogonia are not. Scale bar = 20 p.m. the adult level (about 100 n g / m g protein) by 4 weeks of age. Immunohistochemical localization of a B crystallin in the nert~ous and muscle tissues and testis
The i m m u n o s t a i n i n g was performed with antibodies to aBo~p, and the results are shown in Fig. 7. N e u r o n s in the spinal cord, brainstem, h i p p o c a m p u s were weakly
immunoreactive, while glial cells, most of which seem to be oligodendrocytes, were intensely i m m u n o r e a c t i v e (Fig. 7A). Mitral cells in the olfactory bulb (Fig. 7B) and Schwann cells in the peripheral nerves (Fig. 7C) were also immunoreactive. Skeletal muscle, heart muscle, and the muscle fibers in the tongue (Fig. 7D) were i m m u n o r e a c t i v e as described by lwaki et al. [12]. In the testis, spermatocytes were i m m u n o r e a c t i v e but sper-
207 matogonia, Sertoli cells and interstitial cells were not (Fig. 7 E and F).
Discussion Recently, much attention has been focusing on crystallin, a major component of vertebrate eye lens, because the expression and presence of aB in various non-lenticular tissues have been shown by the Northern and Western blotting tests, and immunohistochemical stainings [8-12]. However, the quantitative analysis of aB in non-lenticular tissues has not been reported. It is known that the amino acid sequence of t~ cD'stallin is highly homologous to the ubiquitous small heat-shock proteins [28,29]. However, the biological function of ctB crystallin is not known. In order to elucidate the physiological significance of orb crystallin, we established an immunoassay method for aB. The sandwich-type immunoassay described here is specific to aB and highly sensitive, and the minimum detection limit of aB is < 10 pg/tube. The competitive type radioimmunoassay for measurement of human [30], sheep [31], or mouse [32] a crystallins has been reported. However, these methods are not specific to aB crystallin, and the minimum detection limit was about 2 n g / m l or 200 pg/tube. The present sandwich-type immunoassay was prepared with use of the two antibody preparations, anti-orB and anti-aBp~p, because the anti-aBpcp has a high affinity to aB. Since only one epitope for anti-aBpcp is present per aB molecule, monomeric aB crystallin could not produce increase in the galactosidase activity in the ball in the sandwich immunoassay system consisted of anti-aBpep antibodies alone. However, when the assay was performed with polystyrene balls with anti-aBpcp, instead of the balls with anti-aB, a similar, but slightly less sensitive standard curve was obtained (Fig. 2). These results suggest that the aB antigens are present as the polymeric form in the immunoassay mixture. It is reported that the purified a crystallin subunits also has the ability to associate into a crystallin-like polymers [33]. However, because it is not known whether aB present in the tissues at a low concentration also forms the polymers, the assay system prepared with anti-aBp¢ 0 alone was not employed in the present study. The immunoreactive aB was measurable in most of the tissues determined, and the highest level of aB was seen in the soleus muscle (about 2% of the total soluble protein). The aB was distributed at a relatively high concentrations in the heart, diaphragm, and esophagus. However, aB concentrations were much lower in the rectus femoris muscle, a fast twiwh muscle, as compared with the above described muscle tissues. These results are in line with the previous report [12] that the immunoreactive aB is present predominantly in the slow twitch muscle fibers. Kidney also contains a rela-
tively high level of ctB, which is localized in the epithelial cells of proximal renal tubules, Henle's loop, and collecting duct [12]. The immunoreactive aB in these tissues is very similar (if not identical) to lens aB. because the immunoblotting test of several tissue extracts with anti-etBp~p antibodies indicated that the aB in the extracts displayed a single band at the same position as ctB purified from bovine lens. Developmental changes of aB concentrations in the nervous tissue and tongue are unique. The aB concentrations in the brain began to increase after 5 weeks of age, coincidently with the period when the maturation of central nervous system was almost completed. In contrast, the aB concentration in the tongue showed a maximum peak at about 1 week after birth. Although the physiological significance of these changes remains to be clarified, these results, together with the specific Iocalizauon of aB in the slow twitch fibers in the skeletal muscle, may suggest the function(s) of aB. The immunohistochemical staining of aB antigens in sections of rat tissues revealed that aB is present not only in oligodendrocytes and Schwann cells as reported by lwaki e t a l . [12] but also in some neurons in the nervous tissues, and in spermatocytes of the testis. The aB crystaUin is a major component of Rosenthal fibers, which accumulates abundantly in the brain (astrocytes) of patients with Alexander disease. If aB in the astrocytes leaks out to the cerebrospinal fluid of the patient, measuring aB with the present assay method might be useful for diagnosis of the patient with Alexander disease. Acknowledgment This work was supported in part by a Grant-in-Aid for Science Research on Priority Areas (Molecular Basis of Neural Connection), Ministry of Education, Science, and Culture of Japan.
References 1 Wistow,G.'i. and Piatigorsky,J. (1988) Annu. Rev. Biochem.57, 479-504. 2 Quax-Jeuken,Y., Quax, W., Van Rens, G., Meera Khan. P. and Bloemendal, H. (1985) Proc. Natl. Acad. Sci. USA 82, 5819-5823. 3 Brakenhoff,R.H., Geurts van Kessel,A., Oldenburg, M., Wijnen. J.T., Bloemendal, H., Meera Khan, P. and Shoenmarkers, J,G. (1990) Hum. Goner. 85, 237-240. 4 Spector, A., Chiesa, R., Sredy, J. and Garner, W. (1985) Proc. Natl. Acad. Sci. USA 82, 4712-4716. 5 Van dcr Ouderaa, F.J., DeJon8, W.W. and Bloemendal, H. (1973) Eur. J. Biochem.39, 207-222. 6 Van der Ouderaa, F.J., De Jong. W.W., Hilderink, A. and Bloemendal, H. (1974) Eur. J. Biochem.49, 157-168. 7 Van den Oetelaar, P.M., Clauwaert, J., Van Laethem, M. and Hoenders, H.T. (1985) J. Biol. Chem. 260. 14030-14034. 8 Bhat, S.P. and Nagineni. C.N. (1989) Biochem. Biophys. Res. Commun. 158, 319-325.
208 9 lwaki, T.. Kume-lwaki. A.. Liem. R.K.H. and Goldman. J.E. (19891 Cell 57, 71-78. 10 Nagineni. C.N. and Bhat. S.P. (1989) FEBS Lett. 249. 89-94. 11 Dubin. R.A.. Wawrousek. E.F. and Piatigorsky, J. (1989) Mol. Cell. Biol. 9. 1083-1091. 12 Iwaki. T.. Kume-lwaki. A. and Goldman. J.E. (1990) J. Histochem. Cytnchem 38, 31-39. 13 Spector, A., Li, L.-K.. Aagusteyn, R.C., Schneider. A. and Freund, T. (19711 Biochem. J. 124. 337-343. 14 Bloemendal. H. and Groenewoud. G. (19811 Anal. Biochem. 117. 327-329. 15 Merrifield, R.B. (1963) J. Am. Chem. Soc. 85, 2149-2154. 16 Goldsmith, P.. Gierschik, P.. Milligan. G., Unson, C.G.. Vinitsky. R.. Malech. H.L. and Spiegel. A.M. 0987) J. Biol. Chem. 262, 14683-14688. 17 Kato, K.. Suzuki, F. and Semba. R. (19811 J. Neurochem. 37, 998-1005. 18 Kato, K.. Umeda, Y., Suzuki. F. and Kosaka. A. (19791 J. Appl. Biochem. 1,479-488. 19 Kato. K.. Fukui, H.. Hamaguchi. Y. and Ishikawa. E. (19761 J. Immunol. 1 t6, 1554-1560.
20 Kato, K., Hamaguchi, Y., Okawa, S., Ishikawa. E., Kobayashi. K. and Katunuma. N. (19771 J. Biochem. 81, 1557-1566. 21 Kato. K. (19831 Methods Enzymol. 92. 345-359. 22 Kato. K., Umeda, Y., Suzuki, F. and kosaka, A. (1980) Clin. Chim. Acta 102, 262-265. 23 Laemmli, U.K. (19701 Nature 227, 680-685. 24 Towbin, H., Staehelin, T. and Gordon. J. (19791 Pro*.:. Natl. Acad. Sci. USA 76, 4350-4354. 25 Kurobe. N., Suzuki. F.. Okajima, K. and Kato. K. (19901 Clin. Chim. Acta 187, 11-20. 26 Nakane. P. (19751 Ann. NY Acad. Sci. 254, 203-21. 27 Bradford. M.M. (19761 Anal. Biochem. 72, 248-254. 28 Schlesinger, M.J.J. (I9861 Cell Biol. I03, 321-325. 29 Hickey, E.. Brandon. S.E., Potter, R., Stein, G., Stein. J. and Weber, L.A. (19861 Nucleic Acids Res. 14, 4127-4145. 30 Sandberg, H.O. and Closs, O. (19781 Exp. Eye Res. 27, 701-712. 31 Sandberg, H.O. and Closs. O. (19781 Exp. Eye Res. 27, 61-71. 32 Russel, P., Carper, D. and Kinoshita, J.H. (19781 Exp. Eye Res. 27. 673-680. 33 Augusteyn, R.C., Koretz, J.F. and Schurtenberger. P. (19891 Biochim. Biophys. Acta 999, 293-299.
201
Tissue distribution and developmental profiles of immunoreactive aB crystallin in the rat determined with a sensiti~.e immunoassay system K a n e f u s a K a t o t, H a r u o S h i n o h a r a ~, N a o m i K u r o b e ~, Y u t a k a I n a g u m a 1, Kikuo Shimizu 2 and Kunihiro Ohshima 3 t Depart,lent of Biochemistry. Institutefor Deeelopmental Rexearch..qtchi Prefectural ('olot~v. Kasugai. Aichi (JapanL : Medwal and Biological Laboratories Co., Ltd.. Ina, Nagano (Japan1 and ¢ Technical Research Laboratories. Kurabo Industries l t d . Ne~'::',,;.z.va. Osaka ¢Japan)
(RL~.eived13 February 19911
Key words: Crystallin; lmmunoassay:Immunohistochemistry In order to determine the q'. ~:Jtitative distribution o f ~ crystailin (aB) in non-lentieular tissues, we have established a sensitive immunoassay system for specific measurement of aB. Antisera were raised in rabbits by injecting a B purified from bovine lenses, or C-terminal decapeptide (KPAVTAAPKK) of a B (aBp~p). The antibodies to a B and a l ~ p w e r e purified by the use of a aB-coupled Sepharose column. The F(ab') z fragments of antibody IgG to a B were immobilized on polystyrene balls and the Fab' fragments of antibody IgG to aBp~p were labeled with fl.D-ga!actosidase from Escherichia c o i l T h e sandwich-type enzyme immunoassay consisted of the abov~ two antibodies was sensitive, and the minimum detection limit of the assay was 10 pg a B without any crossreactivity with aA. By using the assay method, it is revealed that the a B was distributed in most of the tissues examined. Among the non-lentieular tissues, a B was present at high levels in the heart and striated muscles, especially in the soleus muscle, and kidney. High levels of a B in the muscle tissues were also seen in various animals. Developmental increases of a B in rat muscle tissues and kidney were observed from 16 days of gestational age to I or 5 weeks of postnatal age. In contrast, the a B in the brain kept a low level during the same period. After 5 weeks of age, a B concentrations in the brain increased sharply, reaching the adult levels at 9 weeks of age. lmmunohistochemieal staining with anti-aBp~p revealed that a B was positive not only in glial cells, in the central nervous tissues, but also in some neurons of spinal cord, brainstem, hippueampus, and olfactory bulb. Spermatocytes in the testis were also immunopositive for aB.
Introduction a Crystallin is a major structural protein of the vertebrate lens, and it is mainly composed of two acidic ( a A t and a A 2 ) and two basic (aBi and aB2) subunits [1]. It has been shown that aA2 and a ~ are the primary products of two distinct genes, which are located on different h u m a n chromosomes (on the long arm of chromosome 21 [2] and 11 [3], respectively), a A I
Abbreviations: aA and aB, the A and B subunit, respectively,of a crystallin" aB~p. C-terminal decapeptide (KPAVTAAPKK)of aB" SDS-PAGE. sodium dodecyl sulfate-polyacrylamidegel electrophoresis; HPLC, high performanceliquid chromatography. Correspondence: K. Kato. Department of Biochemistry,Institute for DevelopmentalResearch,Aichi PrefecturalColony. Kamiya, Kasugai. Aichi 480-03, Japan.
and aB~ are considered to be the phosphorylated forms of a A , and a ~ . respectively [4]. The molecular mass of each of these subunits is about 20 kDa [5,6]. However, a crystallin extracted from the lens has a molecular mass of approx. 800 kDa [7], indicating ~ crystallin is present as a polymeric form. Recently, it has been revealed that orb is also expressed in various non-lenticular tissues, including heart, skeletal muscle, kidney and nervous tissues [8-12], and it accumulates in the brain of patients with Alexander disease as a major component of Rosenthal fibers [9]. In order to clarify the physiological significance of aB in the non-lenticular tissues, we established a sensitive and specific enzyme immunoassay system by using rabbit antibodies raised against aB and the C-terminal decapeptide of a B , and determined the quantitative distribution of aB in rat tissues and developmental changes of aB levels in rat central nervous tissues, muscle tissues and kidney. Some
202 tion was repeated every 2 weeks, and the rabbits were bled after 2 months. The antibodies to aB and aBp~p were purified by the use of the aB-coupled Sepharose column with the same procedures as described previously [17]. The purified antibodies to aB then were passed through the aA-coupled Sepharose column to eliminate the crossreacting antibodies to aA. The yield of the antibody lgG in the above procedures was about 5 rag/100 ml serum. As shown in Fig. 1. the anti-aB and anti-aBp~p antibodies were specific to aB showing no reactivity with aA and other components of the crude extract of rat soleus muscle on the immunoblotting test, although the reactivity of anti-aB was significantly lower than that of anti-aBp~p (Fig. IC).
f!ndings on the immunohistochemical localization of aB were also described. Materials and Methods
Purification of aA and aB cr)'stallins Bovine a crystallin was purified from fresh lenses obtained at a local slaughter house as described by Spector et al. [13], and the aA and aB crystallins were isolated by the use of chromatofocusing column chromatography in the presence of 6 M urea as described by Bloemendal and Groenewoud [14]. The aA l and aB 2 crystallins were used as the aA and aB antigens, because each of the two fractions was homogeneous showing a single band on the polyncrylamide slab gel electrophoresis in the presence of sodium dodecyl sulfate (SDS-PAGE) as shown in Fig. IA. The purified aA ~nd aB were stored at - 2 0 ° C in a 30% glycerol solution with 0.1% NAN3.
Enzyme immunoassay system for aB In order to increase the sensitivity of the assay and to p, otect the assay with the sample interference [18], the assay system was prepared with the antibody fragments without Fc portion. The purified antibody IgG was digested with pepsin, and the resultant F(ab')2 fragments were isolated by the use of a Sephadex G-150 column [19]. The antibody F(ab')2 fragments were immobilized noncovalently on polystyrene balls (3.2 mm in diameter, lmmuno Chemicals, Okayama, Japan) as described previously [20], and the balls were stored at 4 ° C in 10 mM sodium phosphate buffer (pH 7.0) containing 0.1 M NaCI, 1 mM MgCI 2, 0.1% bovine serum albumin, and 0.1% NaN 3 (buffer A) at least overnight before use for the immunoassay. The antibody F(ab')2 fragments were reduced with 2-mercaptoethylamine and the resultant Fab' fragments were coupled with fl-D-galactosidase from E. coil (Boehringer Mannheim, Mannheim, F.R.G.) by the use of N,N'-o-phenylenedimaleimide as described [19,21]. Amounts of labeled antibodies are expressed in units of
Peptide synthesis attd its conjugation to hemo~yanin The C-terminal decapeptide (KPAVTAAPKK) [6] of aB (aBpCp) was synthesized by solid-phase procedures [15] using a peptide synthesizer (PEPTICOU PLER 2200, Vega Biotechnologies, Tucson, AZ, U.S.A.) and purified with reverse-phase HPLC. The purity of the peptide was assessed by analytical HPLC and amino acid analysis. The peptide was conjugated with the equal amount of keyhole limpet hemocyanin (Sigma Chemicals, St. Louis, MO, U.S.A.) as described by Goldsmith et al. [161.
A ntisera and purification of antibodies Antisera were raised in rabbits by subcutaneous injection of the purified aB (about 0.5 mg/rabbit), or the same amounts of aBr,~pconjugated with the hemocyanin with Freund's complete adjuvant. The same immuniza-
m
A
B
143 30"---m
W
1
2
3
4
2
3
4
2
3
4
Fig. 1. SDS-PAGEof the purified aA and aB crystallinsand an extract of rat soleusmuscle(A), and immunoblotswith purified anti-aBpop(B) and anti-aB (C). Lane 1, standard proteinswith molecularmassin kDa; lane 2, 1 /~gof purified t~Al: lane 3, 1 #g of purified aB2.; lane 4, an extract of rat soleus musclecontaining about 50 .ug proteins. (A) Polyacrylamide 0570) gel stained with Coomassieblue: (B and C) nitrocellulose sheet stained with purified antibodies (0.02 p.g/ml) raised with aB~o, and aB, respectively. Peroxidase activity on the sheet was visualized with diaminobenzidine and H202.
203 galactosidase activity, and i unit of the activity is defined as that producing 1 ~tmol 4-methylumbelliferone/min under the conditions. Unless otherwise specified, a polystyrene ball with anti-aB antibodies was incubated in duplicate with shaking at 3 0 ° C for 5 h with a 10-/~1 aliquot of samples or standard aB crystallin in a final volume of 0.5 ml with 10 mM sodium phosphate buffer (pH 7.0) containing 0.3 M NaC1, 1 mM MgCI 2, 0.1% proteinase-treated gelatin [22], and 0.1% NaN3 (buffer G). Each ball was washed twice with 1 ml of cold buffer A, transferred into a fresh test tube with 0.2 ml of buffer A containing 1.5 milliunits of galactosidase-labeled anti-aBp~p antibodies, and incubated with shaking at 4 ° C overnight. The ball was washed as described above and the galactosidase activity bound on the ball was assayed with a fluorogenic substrate, 4-Methylumbelhferyl-/3-ogalactoside (Sigma), as described previously [20].
Preparation of tissue extract Wistar rats of various gestational and postnatal days were used. Tissues were sampled and kept frozen at - 8 0 ° C until analysis. The frozen tissues from postnatal rats were homogenized at 0 ° C with a Physcotron (NS-50, Niti-On, Chiba, Japan) in a 10- to 20-volume (v/w) of 50 m M Tris-HC1 (pH 7.5) containing 5 mM FDTA. The fetal rat tissues were homogenized with a Teflon-glass homogenizer and then each homogenate was sonicated at 0 ° C for 30 s. All ilomogenates were centrifuged at 4 ° C at 2 0 0 0 0 × g for 20 rain, and the supernatant fractions were used for analysis after dilution with buffer G. Bovine and porcine tissues were obtained at the slaughter house, and h u m a n tissues were obtained at autopsy. All samples were kept frozen at - 8 0 ° C and prepared as described above. Electrophoresis and immunoblotting SDS-PAGE was performed by the method of Laemmli [23]. lmmunoblots were carried out as described Towbin et al. [24] with modifications. In brief, the nitrocellulose sheets were incubated for 1 h with shaking in buffer G containing purified antibody lgG (0.02 p,g/ml), and then for 30 min with 1000-fold diluted goat (anti-rabbit IgG) lgG-labeled with horseradish peroxidase (Medical and Biological Laboratories, Nagoya, Japan). The peroxidase activity on the sheet was visualized with diaminobenzidine and H202 [25] or ECL Western blotting detection system (Amersham International, Amersham, U.K.). hnmunohistochemical staining Adult Wistar rats were deeply anesthetized with chloroform and perfused via the heart with physiological • saline, followed by a fixative solution composed of 1% glutaraldehyde, 4% paraformaldehyde, 0.2% picric acid.
and 2% sucrose in 0.1 M sodium acetate buffer (pH 6.0). The tissues were dissected out and fixed by soaking in the same buffer for 4 h. Tissues were rinsed and left standing in 50 mM Tris-HCI buffer (pH 7.4) overnight, then dehydrated, and embedded in paraffin. Sections (5-6 ~tm thicl:) were prepared and immunostained with purified antibody lgG to aBpep (2 /tg/ml) according to the indirect peroxidase-labeled antibody method [26]. using diaminobenzidine as the chromogen. For controls. sections were incubated with antibodies preabsorbed with aBp~p, which gave no positive staining.
Other methods Protein concentrations of tissue extracts were estimated with Bio-Rad protein assay (Bio-Rad, Richmond, CA, U.S.A.), which utilizes a principle of protein-dye binding [27]. Concentrations in Itg of the purified crystallins were also determined with the same method using bovine serum albumin as a standard. Results
Detection limit and specificio' of the #nmunoassay systems for a B crvstallin The three assay systems for aB were prepared by the combination of two antibody preparations, anti-aB and anti-aBo~p. As shown in Fig. 2, the assay system consisted of polystyrene balls with anti-aB and the galactosidase-labeled anti-aB~p was apparently more sensitive as compared with the system prepared with anti-aB or anti-aB~p alone, and the minimum detection limit, defined as the lowest concentration giving a galactosidase activity significantly different from that of zero standard at > 0.99% confidence, was less than 10 p g / a s s a y tube. Therefore. unless otherwise specified, the aB assay was conducted with the system consisted of two antibody preparations. The assay did not crossreact with aA. These results indicate that the aB assay system with two antibody preparations was sensitive and specific to the B subunit of a crystallin. Determinations of aB crystallin in various rat tissues In order to find the proportionality of the assay with tissue extracts and the optimum sample volume of each extract, several extracts of muscle tissues and kidney were diluted serially and the diluted extracts were subjected to the immunoassay for aB. As shown in Fig. 3, the values determined in each extract were p~oportional to the sample volume employed to the assay. The precision of aB assay was tested by assaying four samples of rat spleen extract ten times in one assay (within-assay) or in duplicate in seven consecutive assays (between-assay). The coefficients of variation in each assay were < 12%. The analytical recovery of
204 !000 ~
/,tj
TABLE I Concentrations of aB crystallin in t,ariolt~ rat tissues
i
The extracts of various adult male rat tissues were immunoassayed.
1
0
g
, 0
0.01
0.1 a
1
10
Crystallin (ng)
Fig. 2. Standard curves of the assay for aB crystallin and its crossreactivity with aA crystallin. Indicated amounts of aB crystallin (solid lines) or aA crystallin (broken line) were incubated in duplicate with polystyrene balls with anti-aB (e, t:], A) or anti-aBp¢p (o), and then the balls was incubated with the galactosidase-labeled anti-aB (ra) or anti-aBpe p (e. o. A) as described in the text. fl-D-Galactosidase activity bound on the ball is expressed as the fluorescence intensity of 4-methylumbelliferone produced in a 20-min reaction at 30°C with 0.1 mM 4-methylumhelliferyl-fl-D-galactoside. In the fluorescence intensity scale. 1000 equals 1 ,ttM 4-methylumbelliferone.
b o v i n e a B c r y s t a l l i n s (1 n g ) a d d e d to t h e i m m u n o a s s a y t o g e t h e r w i t h 5 /tl o f t h e s p l e e n e x t r a c t ( n = 5) w a s 101 _+ 7%. T h e s e r e s u l t s i n d i c a t e t h a t t h e c o n c e n t r a t i o n s 10
Tissues
aB crystallin ( n g / m g protein)
Cerebral cortex Cerebellum Brainstem Hippocampus Tongue Heart Esophagus Diaphragm Soleus muscle Rectus femoris muscle Kidney Bladder Testis Rectum Stomach Cecum Liver Spleen Thymus Lung Adrenal Xiphoid Adipose Blood plasma
6.3+ 2.1" 14.0_+ 2.1 36.5_+ 3.7 9.3_+ 1.7 371 _+113 2110 +400 1640 +250 2090 +420 19500 _+800 52.2+ 17.4 125 -+ 45 t 2.7_+ 2.9 1.2+ 0.2 14.6_+ 5.4 1.6_+ 0.6 4.2_+ 0.8 0.4 + 0.2 4.9 + 0.8 1.4+ 0.3 9.1 _+ 3.3 1.3+ 0.4 31.4+ 33.4 7.5 + 1.0 1.5 + 0.7 h
a Mean+_l S.D. of five rats. h ng/ml.
o f a B in t h e e x t r a c t c a n b e d e t e r m i n e d b y t h e p r e s e n t method. Table I shows the concentrations of immunoreactive a B in t h e s o l u b l e f r a c t i o n s o f v a r i o u s r a t t i s s u e s e x pressed as bovine aB equivalent ng/mg soluble protein. T h e a v e r a g e c o n c e n t r a t i o n o f a B in r a t lens w a s 247 p , g / m g p r o t e i n . I m m u n o r e a c t i v e orB c r y s t a l l i n w a s d e t e c t e d in t h e m o s t t i s s u e s e x a m i n e d , a n d it w a s p r e s e n t
TABLE II Concentrations of aB crystallin in human, bovine and porcine muscle tissues
Tissues
'O.1'
K/
O. 01 O. 1 1 10 Extract added ( id ) Fig. 3. Effect of sample dilution on the assay of aB crystallin in the crude extract. The soluble fraction of 1070 (w/v) homogenate of rat soleus muscle (o), heart (A), tongue in), or kidney (o, n) was diluted variously, and the samples containing the indicated volumes of original extract were subjected to the immunoassay. Each point is the mean of a duplicate assay.
Human heart pectoral muscle Bovine heart intercostal muscle Porcine heart intercostal muscle Mean±S.D.
No. of samples
aB crystallin ( n g / m g protein)
l 5
3080 4480+ 1070 a
5 5
4220+2280 4140 + 1540
5 5
956_+ 476 381 + 76
205 at high levels in tissues composed of striated muscle (such as leg muscles, diaphragm, esophagus, and tongue) and heart muscle. The soleus muscle, a slow twitch muscle, contained a much higher level (about 400-fold) of a B as compared witit the rectus femoris muscle, a fast twitch muscle, confirming that the a B is mainly localized in the type I muscle fibers as reported by lwaki et al. [12]. The kidney also contained a considerable a m o u n t of orB. Significant a m o u n t s of aB were present in the central nervous tissue, rectum, bladder. spleen, and xiphoid, a cartilaginous tissue. Rat blood p l a s m a contained about 1.5 ng a B / m l . High levels of immunoreactive a B were also determined in the bovine, porcine, and h u m a n muscle tissues (Table I1).
Immunoblotting of intmunoreactive aB in the crude extract On SDS-PAGE, the immunoreactive a B in rat soleus muscle (Fig. 1), rectus femoris muscle, kidney, and bladder (data not shown) migrated to the same position as that of bovine lens aB. The i m m u n o r e a c t i v e a B estimated in other muscle tissues of rat, human, bovine, and porcine also showed a single b a n d at the same position as that of bovine lens a B on the immunoblotting test (Fig. 4). These results indicated that the a B measured in each extract was very similar, if not identical, to lens a B as reported previously [8,9], and also suggested that the present assay specifically measured a B molecules in the crude extract.
Developmental profiles of aB in rat brain, muscle tissues. and kidney By using the present assay method, developmental changes of a B concentrations in rat central nervous
1
2
3
4
5
6
7
8
Fig. 4. Immunoblot of the extracts of muscle tissues of various animals with anti-aBpcp. The soluble fractions (10/~1) of 10% homogenates of rat heart (lane 2) and diaphragm (lane 3) human heart (lane 4l and pectoral muscle (lane 5). bovine heart (lane 6) and intercostal muscle (lane 7), and porcine heart (lane 8). together with 0.5 ag of purified aB (lane 1), were subjected to SDS-PAGE, followed by immunoblot. The peroxidase activity on the sheet was visualized with 3,3'-diaminobenzidine and H202.
6o
o m
o~
40
==
0-2
0
2
4 6 Weeks after birth
8
10
~' Fig. 5. Developmental profiles of aB crystallin in rat central ncr,'ous tissues. Concentrations (ng/mg protein) of aB in rat cerebral cortex (©). cerebellum (el. and brainstem ([:3)were determined from 15 days of gestational age to 9 weeks of postnatal age. tissues, heart, tongue, and kidney were determined from 15 days of gestational age to 9 weeks of postnatal age. As shown in Fig. 5, the a B was detected in the cerebral cortex, cerebellum, and brainstem of embryos, showing the concentration of < 5 n g / m g protein (1 to 3 n g / m g protein). The low levels of a B were seen until 4 to 5 weeks of age. Then the ~tB contents in the brain were increased sharply, reaching the adult levels at 9 weeks of postnatal age. C h a n g e s in the a B levels of kidney, tongue, and heart are shown in Fig. 6. The a B was detectable in the three organs of rat embryos. The a B in the kidney increased after birth and reached a plateau level at 5 weeks of age. The a B concentration in the embryonic heart was 1 0 - 5 0 n g / m g protein, but it increased sharply after birth, reaching the adult level of a b o u t 1500 n g / m g protein within 2 weeks of age. The a B in the tongue increased just after the birth and reached a peak value (about 800 n g / m g protein) o n 8 days of age, and then decreased to
i ~1~°
o2
0
2
4 6 Weeks after birth
8
10
Fig. 6. Developmental profiles of aB crystanin in rat heart, tongue, and kidney. Concentrations (ng/mg protein) of aB in rat heart (o). tongue (O), and kidney (O) were determined as described in Fig. 6.
206
D
¥
,/
~
t
,k !
.¢
A
3B
cA
Fig. 7. Immunohistochemical staining with anti-aB~p. The sections were stained with purified anti-aB~p IgG (2 .ug/ml) by the indirect peroxidase-labeled antibody method. (A) Immunoreactive neurons in the spinal cord. Arrow indicates the immunostained glial cell. Inset: immunoreactive glial cell, which seems to be an oligodendrocyte. (B) Mitral cells in the olfactory bulb are also immunoreaetive. (C) Scbwann cells in the peripheral nerves in the tongue are immunoreactive. On the right are immunoreactive muscle fibers of the tongue. (D) lmmunoreactive muscle fiber in the tongue. Muscle fibers are stained heterogeneously. (E and F) The spermatoeytes are immunoreactive but spermatogonia are not. Scale bar = 20 p.m. the adult level (about 100 n g / m g protein) by 4 weeks of age. Immunohistochemical localization of a B crystallin in the nert~ous and muscle tissues and testis
The i m m u n o s t a i n i n g was performed with antibodies to aBo~p, and the results are shown in Fig. 7. N e u r o n s in the spinal cord, brainstem, h i p p o c a m p u s were weakly
immunoreactive, while glial cells, most of which seem to be oligodendrocytes, were intensely i m m u n o r e a c t i v e (Fig. 7A). Mitral cells in the olfactory bulb (Fig. 7B) and Schwann cells in the peripheral nerves (Fig. 7C) were also immunoreactive. Skeletal muscle, heart muscle, and the muscle fibers in the tongue (Fig. 7D) were i m m u n o r e a c t i v e as described by lwaki et al. [12]. In the testis, spermatocytes were i m m u n o r e a c t i v e but sper-
207 matogonia, Sertoli cells and interstitial cells were not (Fig. 7 E and F).
Discussion Recently, much attention has been focusing on crystallin, a major component of vertebrate eye lens, because the expression and presence of aB in various non-lenticular tissues have been shown by the Northern and Western blotting tests, and immunohistochemical stainings [8-12]. However, the quantitative analysis of aB in non-lenticular tissues has not been reported. It is known that the amino acid sequence of t~ cD'stallin is highly homologous to the ubiquitous small heat-shock proteins [28,29]. However, the biological function of ctB crystallin is not known. In order to elucidate the physiological significance of orb crystallin, we established an immunoassay method for aB. The sandwich-type immunoassay described here is specific to aB and highly sensitive, and the minimum detection limit of aB is < 10 pg/tube. The competitive type radioimmunoassay for measurement of human [30], sheep [31], or mouse [32] a crystallins has been reported. However, these methods are not specific to aB crystallin, and the minimum detection limit was about 2 n g / m l or 200 pg/tube. The present sandwich-type immunoassay was prepared with use of the two antibody preparations, anti-orB and anti-aBp~p, because the anti-aBpcp has a high affinity to aB. Since only one epitope for anti-aBpcp is present per aB molecule, monomeric aB crystallin could not produce increase in the galactosidase activity in the ball in the sandwich immunoassay system consisted of anti-aBpep antibodies alone. However, when the assay was performed with polystyrene balls with anti-aBpcp, instead of the balls with anti-aB, a similar, but slightly less sensitive standard curve was obtained (Fig. 2). These results suggest that the aB antigens are present as the polymeric form in the immunoassay mixture. It is reported that the purified a crystallin subunits also has the ability to associate into a crystallin-like polymers [33]. However, because it is not known whether aB present in the tissues at a low concentration also forms the polymers, the assay system prepared with anti-aBp¢ 0 alone was not employed in the present study. The immunoreactive aB was measurable in most of the tissues determined, and the highest level of aB was seen in the soleus muscle (about 2% of the total soluble protein). The aB was distributed at a relatively high concentrations in the heart, diaphragm, and esophagus. However, aB concentrations were much lower in the rectus femoris muscle, a fast twiwh muscle, as compared with the above described muscle tissues. These results are in line with the previous report [12] that the immunoreactive aB is present predominantly in the slow twitch muscle fibers. Kidney also contains a rela-
tively high level of ctB, which is localized in the epithelial cells of proximal renal tubules, Henle's loop, and collecting duct [12]. The immunoreactive aB in these tissues is very similar (if not identical) to lens aB. because the immunoblotting test of several tissue extracts with anti-etBp~p antibodies indicated that the aB in the extracts displayed a single band at the same position as ctB purified from bovine lens. Developmental changes of aB concentrations in the nervous tissue and tongue are unique. The aB concentrations in the brain began to increase after 5 weeks of age, coincidently with the period when the maturation of central nervous system was almost completed. In contrast, the aB concentration in the tongue showed a maximum peak at about 1 week after birth. Although the physiological significance of these changes remains to be clarified, these results, together with the specific Iocalizauon of aB in the slow twitch fibers in the skeletal muscle, may suggest the function(s) of aB. The immunohistochemical staining of aB antigens in sections of rat tissues revealed that aB is present not only in oligodendrocytes and Schwann cells as reported by lwaki e t a l . [12] but also in some neurons in the nervous tissues, and in spermatocytes of the testis. The aB crystaUin is a major component of Rosenthal fibers, which accumulates abundantly in the brain (astrocytes) of patients with Alexander disease. If aB in the astrocytes leaks out to the cerebrospinal fluid of the patient, measuring aB with the present assay method might be useful for diagnosis of the patient with Alexander disease. Acknowledgment This work was supported in part by a Grant-in-Aid for Science Research on Priority Areas (Molecular Basis of Neural Connection), Ministry of Education, Science, and Culture of Japan.
References 1 Wistow,G.'i. and Piatigorsky,J. (1988) Annu. Rev. Biochem.57, 479-504. 2 Quax-Jeuken,Y., Quax, W., Van Rens, G., Meera Khan. P. and Bloemendal, H. (1985) Proc. Natl. Acad. Sci. USA 82, 5819-5823. 3 Brakenhoff,R.H., Geurts van Kessel,A., Oldenburg, M., Wijnen. J.T., Bloemendal, H., Meera Khan, P. and Shoenmarkers, J,G. (1990) Hum. Goner. 85, 237-240. 4 Spector, A., Chiesa, R., Sredy, J. and Garner, W. (1985) Proc. Natl. Acad. Sci. USA 82, 4712-4716. 5 Van dcr Ouderaa, F.J., DeJon8, W.W. and Bloemendal, H. (1973) Eur. J. Biochem.39, 207-222. 6 Van der Ouderaa, F.J., De Jong. W.W., Hilderink, A. and Bloemendal, H. (1974) Eur. J. Biochem.49, 157-168. 7 Van den Oetelaar, P.M., Clauwaert, J., Van Laethem, M. and Hoenders, H.T. (1985) J. Biol. Chem. 260. 14030-14034. 8 Bhat, S.P. and Nagineni. C.N. (1989) Biochem. Biophys. Res. Commun. 158, 319-325.
208 9 lwaki, T.. Kume-lwaki. A.. Liem. R.K.H. and Goldman. J.E. (19891 Cell 57, 71-78. 10 Nagineni. C.N. and Bhat. S.P. (1989) FEBS Lett. 249. 89-94. 11 Dubin. R.A.. Wawrousek. E.F. and Piatigorsky, J. (1989) Mol. Cell. Biol. 9. 1083-1091. 12 Iwaki. T.. Kume-lwaki. A. and Goldman. J.E. (1990) J. Histochem. Cytnchem 38, 31-39. 13 Spector, A., Li, L.-K.. Aagusteyn, R.C., Schneider. A. and Freund, T. (19711 Biochem. J. 124. 337-343. 14 Bloemendal. H. and Groenewoud. G. (19811 Anal. Biochem. 117. 327-329. 15 Merrifield, R.B. (1963) J. Am. Chem. Soc. 85, 2149-2154. 16 Goldsmith, P.. Gierschik, P.. Milligan. G., Unson, C.G.. Vinitsky. R.. Malech. H.L. and Spiegel. A.M. 0987) J. Biol. Chem. 262, 14683-14688. 17 Kato, K.. Suzuki, F. and Semba. R. (19811 J. Neurochem. 37, 998-1005. 18 Kato, K.. Umeda, Y., Suzuki. F. and Kosaka. A. (19791 J. Appl. Biochem. 1,479-488. 19 Kato. K.. Fukui, H.. Hamaguchi. Y. and Ishikawa. E. (19761 J. Immunol. 1 t6, 1554-1560.
20 Kato, K., Hamaguchi, Y., Okawa, S., Ishikawa. E., Kobayashi. K. and Katunuma. N. (19771 J. Biochem. 81, 1557-1566. 21 Kato. K. (19831 Methods Enzymol. 92. 345-359. 22 Kato. K., Umeda, Y., Suzuki, F. and kosaka, A. (1980) Clin. Chim. Acta 102, 262-265. 23 Laemmli, U.K. (19701 Nature 227, 680-685. 24 Towbin, H., Staehelin, T. and Gordon. J. (19791 Pro*.:. Natl. Acad. Sci. USA 76, 4350-4354. 25 Kurobe. N., Suzuki. F.. Okajima, K. and Kato. K. (19901 Clin. Chim. Acta 187, 11-20. 26 Nakane. P. (19751 Ann. NY Acad. Sci. 254, 203-21. 27 Bradford. M.M. (19761 Anal. Biochem. 72, 248-254. 28 Schlesinger, M.J.J. (I9861 Cell Biol. I03, 321-325. 29 Hickey, E.. Brandon. S.E., Potter, R., Stein, G., Stein. J. and Weber, L.A. (19861 Nucleic Acids Res. 14, 4127-4145. 30 Sandberg, H.O. and Closs, O. (19781 Exp. Eye Res. 27, 701-712. 31 Sandberg, H.O. and Closs. O. (19781 Exp. Eye Res. 27, 61-71. 32 Russel, P., Carper, D. and Kinoshita, J.H. (19781 Exp. Eye Res. 27. 673-680. 33 Augusteyn, R.C., Koretz, J.F. and Schurtenberger. P. (19891 Biochim. Biophys. Acta 999, 293-299.
